donald r. woods - skills4success.cosaharapcc.skills4success.co/spcc/er/successful trouble shooting...

Donald R. Woods

Successful Touble Shootingfor Process Engineers

A Complete Course in Case Studies

Author

Prof. Donald R. WoodsChemical Engineering DepartmentMcMaster UniversityHamiltonOntarioCanada, L8S 4L7

& All books published by Wiley-VCH arecarefully produced. Nevertheless, authors,editors, and publisher do not warrant theinformation contained in these books,including this book, to be free of errors.Readers are advised to keep in mind thatstatements, data, illustrations, proceduraldetails or other items may inadvertentlybe inaccurate.

Library of Congress Card No.: applied for

British Library Cataloguing-in-Publication DataA catalogue record for this book is availablefrom the British Library.

Bibliographic information published byDie Deutsche BibliothekDie Deutsche Bibliothek lists this publicationin the Deutsche Nationalbibliografie; detailedbibliographic data is available in the Internet at<http://dnb.ddb.de>.

� 2006 WILEY-VCH Verlag GmbH & Co. KGaA,Weinheim

All rights reserved (including those oftranslation into other languages).No part of this book may be reproducedin any form – nor transmitted or translatedinto machine language without writtenpermission from the publishers. Registerednames, trademarks, etc. used in this book,even when not specifically marked as such,are not to be considered unprotected by law.

Printed in the Federal Republic of Germany.Printed on acid-free paper.

Typesetting K�hn & Weyh, Satz und Medien,FreiburgPrinting betz-druck GmbH, DarmstadtBookbinding J. Sch�ffer GmbH, Gr�nstadt

ISBN-13: 978-3-527-31163-7ISBN-10: 3-527-31163-7

V

Preface XIII

1 What is Trouble Shooting? 11.1 Characteristics of a Trouble-Shooting Problem 21.1.1 Similarities among TS Problems 21.1.2 Differences between TS Problems 31.2 Characteristics of the Process Used to Solve Trouble-Shooting

Problems 31.2.1 How the Type of Problem Guides the TS Process or Strategy 31.2.2 Five Key Elements Common to the TS Process 41.3 Self-Test and Reflections 51.4 Overview of the Book 91.5 Summary 91.6 Cases to Consider 9

2 The Mental Problem-Solving Process used in Trouble Shooting 172.1 Problem Solving 192.2 Trouble Shooting 232.2.1 Considerations when Applying the Strategy to Solve Trouble-Shooting

Problems 232.2.2 Problem-Solving Processes Used by Skilled Trouble Shooters 242.2.3 Data Collection and Analysis: Approaches Used to Test Hypotheses 252.3 Overall Summary of Major Skills and a Worksheet 252.3.1 Getting Organized: the Use of a Trouble-Shooter’s Worksheet 252.3.2 Feedback about your Trouble Shooting 292.4 Example Use of the Trouble-Shooter’s Worksheet 352.5 Summary 402.6 Cases to Consider 40

3 Rules of Thumb for Trouble Shooting 433.1 Overall 433.1.1 General Rules of Thumb and Typical Causes 433.1.2 Corrosion as a Cause 45

Contents

Successful Trouble Shooting for Process Engineers. Don WoodsCopyright � 2006 WILEY-VCH Verlag GmbH & Co. KGaA, WeinheimISBN: 3-527-31163-7

VI

3.1.3 Instruments, Valves and Controllers 463.1.4 Rules of Thumb for People 473.1.5 Trouble-Shooting Teams 483.2 Transportation Problems 513.2.1 Gas Moving: Pressure Service 523.2.2 Gas Moving: Vacuum Service 533.2.3 Liquid 543.2.4 Solids 563.2.5 Steam 583.3 Energy Exchange 583.3.1 Drives 583.3.2 Thermal Energy: Furnaces 603.3.3 Thermal Energy: Fluid Heat Exchangers, Condensers and Boilers 613.3.4 Thermal Energy: Refrigeration 653.3.5 Thermal Energy: Steam Generation 663.3.6 High-Temperature Heat-Transfer Fluids 663.4 Homogeneous Separation 673.4.1 Evaporation 673.4.2 Distillation 693.4.3 Solution Crystallization 723.4.4 Gas Absorption 733.4.5 Gas Desorption/Stripping 753.4.6 Solvent Extraction, SX 763.4.7 Adsorption: Gas 773.4.8 Adsorption: Liquid 773.4.9 Ion Exchange 773.4.10 Membranes: Reverse Osmosis, RO 793.4.11 Membranes: Nanofiltration 793.4.12 Membranes: Ultrafiltration, UF, and Microfiltration 793.5 Heterogeneous Separations 793.5.1 Gas–Liquid 803.5.2 Gas–Solid 813.5.3 Liquid–Liquid 823.5.4 Gas–Liquid–Liquid Separators 843.5.5 Dryer for GS Separation 853.5.6 Screens for Liquid Solid Separation or Dewatering 853.5.7 Settlers for LS Separation 863.5.8 Hydrocyclones for LS Separation 863.5.9 Thickener for LS Separation 863.5.10 Sedimentation Centrifuges 873.5.11 Filtering Centrifuge 873.5.12 Filter for LS Separation 883.5.13 Screens for Solid–Solid Separation 883.6 Reactor Problems 883.6.1 PFTR: Multitube Fixed-Bed Catalyst, Nonadiabatic 89

Contents

VII

3.6.2 PFTR: Fixed-Bed Catalyst in Vessel: Adiabatic 913.6.3 PFTR: Bubble Reactors, Tray Column Reactors 933.6.4 PFTR: Packed Reactors 943.6.5 PFTR: Trickle Bed 943.6.6 PFTR: Thin Film 963.6.7 STR: Batch (Backmix) 963.6.8 STR: Semibatch 983.6.9 CSTR: Mechanical Mixer (Backmix) 993.6.10 STR: Fluidized Bed (Backmix) 1013.6.11 Mix of CSTR, PFTR with Recycle 1063.6.12 Reactive Extrusion 1063.7 Mixing Problems 1073.7.1 Mechanical Agitation of Liquid 1073.7.2 Mechanical Mixing of Liquid–Solid 1083.7.3 Solids Blending 1083.8 Size-Decrease Problems 1093.8.1 Gas Breakup in Liquid: Bubble Columns 1093.8.2 Gas Breakup in Liquid: Packed Columns 1093.8.3 Gas Breakup in Liquid: Agitated Tanks: 1103.9 Size Enlargement 1103.9.1 Size Enlargement: Liquid–Gas: Demisters 1103.9.2 Size Enlargement: Liquid–Liquid: Coalescers 1103.9.3 Size Enlargement: Solid in Liquid: Coagulation/Flocculation 1113.9.4 Size Enlargement: Solids: Tabletting 1113.9.5 Size Enlargement: Solids: Pelleting 1113.9.6 Solids: Modify Size and Shape: Injection Molding and Extruders 1123.9.7 Coating 1263.10 Vessels, Bins, Hoppers and Storage Tanks 1263.11 “Systems” Thinking 1273.12 Health, Fire and Stability 1303.12.1 Individual Species 1303.12.2 Combinations 131

4 Trouble Shooting in Action: Examples 1334.1 Case’3: The Case of the Cycling Column 1334.2 Case’4: Platformer Fires 1384.3 Case’5: The Sulfuric Acid Pump 1414.4 Case’6: The Case of the Utility Dryer 1444.5 Case’7: The Case of the Reluctant Crystallizer 1574.6 Reflections about these Examples 162

5 Polishing Your Skills: Problem-Solving Process 1655.1 Developing Awareness of the Problem-Solving Process 1655.1.1 Some Target Skills 1665.1.2 The TAPPS Roles: Talker and Listener 166

Contents

5.1.3 Activity 5.1: (35 minutes) 1685.1.4 Feedback, Self-Assessment 1725.2 Strategies 1735.2.1 Some Target Skills 1745.2.2 The Extended TAPPS Roles: Talker+ and Listener+ 1755.2.3 Activity 5.2: (time 35 minutes) 1765.2.4 Feedback, Self-assessment 1795.3 Exploring the “Context”: what is the Real Problem? 1805.3.1 Example 1805.3.2 Activity 5-3 1815.4 Creativity 1835.4.1 Some Target Skills 1835.4.2 Example: Case’10: To dry or not to dry! (based on Krishnaswamy

and Parker, 1984) 1865.4.3 Activity 5-4 1905.4.4 Feedback, Self-Assessment 1915.5 Self-Assessment 1915.5.1 Some Target Skills 1925.5.2 Activity for Growth in Self-Assessment 1925.5.3 Feedback About Assessment 1935.6 Summary and Self-Rating 194

6 Polishing Your Skills: Gathering Data andthe Critical-Thinking Process 195

6.1 Thinking Skills: How to Select Valid Diagnostic Actions 1966.1.1 How to Select a Diagnostic Action 1966.1.2 Select from among a Range of Diagnostic Actions 1966.1.3 More on Gathering and Interpreting Data 2006.1.4 Summary 2096.2 Thinking Skill: Consistency: Definitions, Cause–Effect and

Fundamentals 2096.2.1 Consistent Use of Definitions 2106.2.2 Consistent with How Equipment Works: Cause fi Effects:

Root Cause-Symptoms 2126.2.3 Consistent with Fundamental Rules of Mathematics and English 2176.2.4 Consistent with Fundamental Principles Of Science:

Conservation of Mass, Energy, High to Low Pressure,Properties of Materials 218

6.2.5 Consistent with Experience 2186.2.6 Summary 2196.3 Thinking Skills: Classification 2196.3.1 Classify the Starting Information 2196.3.2 Classifying Ideas from Brainstorming 2206.4 Thinking Skills: Recognizing Patterns 2216.4.1 Patterns in the Symptoms 221

ContentsVIII

6.4.2 Patterns in the Evidence 2236.5 Thinking Skill: Reasoning 2236.5.1 Step 1: Classify the Information 2246.5.2 Step 2: Write the Conclusion 2256.5.3 Step 3: Identify the Context 2256.5.4 Step 4: Clarify the Meaning of the Terminology 2266.5.5 Step 5: Consider the Evidence 2276.5.5a Identify the Evidence 2276.5.5b Check for Consistency 2276.5.5c Which Evidence is Pertinent? 2286.5.5d Diagram the Argument 2296.5.6 Step 6: Formulate the Assumptions 2306.5.7 Step 7: Assess the Quality of the Reasoning 2306.5.8 Step 8: Assess the Strengths of the Counterarguments 2326.5.9 Step 9: Evaluate the Consequences and Implications 2326.5.10 Activity 6-14 2326.6 Feedback and Self-Assessment 2336.7 Summary 2336.8 Exercises 234

7 Polishing Your Skills: Interpersonal Skills and FactorsAffecting Personal Performance 237

7.1 Interpersonal Skills 2377.1.1 Communication 2377.1.2 Listening 2387.1.3 Fundamentals of Interaction 2397.1.4 Trust 2407.1.5 Building on Another’s Personal Uniqueness 2437.2 Factors that Affect Personal Performance 2447.2.1 Pride and Unwillingness to Admit Error 2447.2.2 Stress: Low and High Stress Errors 2457.2.3 Alienation and Lack of Motivation 2497.2.4 “I Know Best!” Attitude 2497.2.5 Tendency to Interpret 2497.3 The Environment 2537.4 Summary 2537.5 Exercises and Activities 253

8 Prescription for Improvement: Put it all Together 2598.1 Approaches to Polish Your Skill 2598.1.1 Triad Activity 2598.1.2 Individual Activity 2628.2 Cases to Help you Polish Your Skill 2638.2.1 Guidelines for Selecting a Case 263

Contents IX

8.2.2 The Cases and Understanding the Choice of Diagnostic Actions for eachCase 263

8.3 Summary 396

9 What Next? 3979.1 Summary of Highlights 3979.2 Reflection and Self-Assessment are Vital for the Development of

Confidence 4019.3 Going Beyond this Book: Setting Goals for Improvement 4029.3.1 Prepare Yourself for Success 4029.3.2 Use Reflection and Self-Assessment Effectively 4039.4 Going Beyond this Book: Updating your Rules of Thumb and

Symptom ‹ Cause Data for Process Equipment 4039.5 Beyond this Book: Sources of Other Cases 403

Literature References 405

Index I 1

CD Contents

Appendix AFeedback about Experience with Process Equipment 411

Appendix BImproving “Systems Thinking” 415

Appendix CFeedback on the Cases in Chapters 1, 2 and 7 423

Appendix DCoded Answers for the Questions Posed to Solve the Cases 435

Appendix EDebrief for the Trouble-Shooting Cases 537

Appendix FOther Tasks for the Skill-Development Activities in Chapter 5 565

Appendix GSelected Responses to the Activities in Chapters 6 and 7 569

Appendix HData about “Causes” for Selected Process Equipment 573

ContentsX

Appendix IFeedback about Symptoms for Selected Causes 579

Appendix JGuide for Students: How You Can Get the Most from this Book 581J-1 Getting Started: Get the Big Picture 581J-2 Try a Trouble-Shooting Case where the Problem is Reasonably Well

Defined 582J-3 See How Others Handle a Case 591J-4 Pause, Reflect on the Pretest, and Invest Time Polishing Specific

Skills 591J-5 Work your First Cases Starting with Case’19 591J-6 Trouble Shooting on the Job 591J-7 Summary 592

Contents XI

XIII

My McMaster University colleague Tom Marlin describes trouble shooting as “thebread and butter” activity of engineering. Indeed, the financial health of a processunit depends so much on the skill of the engineers to trouble shoot problemspromptly, safely and effectively.

Training in trouble shooting should be part of every undergraduate engineer’s ed-ucation. Yet, it rarely is, even though the introduction of trouble-shooting examplesreceives a warm welcome by the students. As Scott Lynn of UC Berkeley reports,“Our experience was that most students really got into the spirit of the thing andtrouble shooting was one of the most popular parts of the course.” Perhaps some ofthe reasons why the development of trouble-shooting skill is not introduced are theneed for excellent problem-solving skills, the lack of a variety of industrial problemsand, perhaps most significantly, the student’s lack of a rich set of practical experi-ence and understanding of equipment. There may also be a lack of the faculty’s con-fidence in using such open-ended experiences. Whatever the reason, I have de-signed this book to overcome these shortcomings. I hope that trouble-shooting skilldevelopment becomes part of every undergraduate experience.

Training in trouble shooting in industry tends to occur from the school of hardknocks, by trial and error and gradually from the experience of solving problems asthey occur, with no well-designed program of instruction. This is relatively ineffi-cient and it does little to develop confidence. This book is designed to improve skilland confidence of process engineers and engineering students.

This book is based on my experience developing trouble-shooting skills in under-graduate engineering programs, in short courses in industry and in courses pre-sented at conferences.

This book is designed to help develop your skill and confidence. This book is tai-lored to help you improve your skill no matter where you are in your journey tobecome an outstanding trouble shooter.

A number of excellent books have been published about trouble shooting. Liber-man (“Trouble-shooting process Operations”) describes a wide range of problemsthat he encountered, the fault that he discovered and the corrective action. His per-sonal approach to trouble shooting is illustrated. Saletan (“Creative Trouble Shoot-ing in the Chemical Process Industries”) provides interesting examples to illustratedifferent components in the trouble-shooting process. However, no specific educa-

Preface


Preface

tional plan is apparent. No activities, with feedback, are provided for skill develop-ment. Branan’s “Rules of Thumb for Chemical Engineers” is mainly an excellentcollection of rules of thumb. In addition he has chapters on trouble shooting andplant startup. He includes material from a range of topics and resources but he doesnot presented a synthesis of this material. The focus in these books tends to be apersonalized approach of how the author solved trouble-shooting problems. Noteveryone will or should follow Lieberman’s (1985), Saletan’s (1994), Gans’s (1983),Kister’s (1979) or my style in trouble shooting. The key is to identify your style anddevelop confidence in using it.

Developing your style and skill requires that we draw on the extensive researchabout the trouble-shooting process. A skill development program should give you achance to solve a wide range of trouble-shooting problems, to think about how yousolved them and to set goals for improvement. The central core of this book is 52trouble-shooting cases that are presented in a unique format that allows you to selectthe process you will use to solve the problem. Feedback is given to help you assessyour approach. Target skills used by successful trouble shooters are given; structuredactivities provided, and feedback is supplied. This book is unique in its coverage,ease in use, focus on skill development using proven methods, self-selection andinclusion of activities that are challenging but fun to do. Included are a range of self-assessment tools.

Here are the details:Chapter 1 outlines four types of trouble-shooting problems and summarizes the

five key skill areas needed in trouble shooting: skill in problem solving, practicalknowledge about a range of process equipment, specific knowledge about safety,hazards, systems thinking and people skills. A self-test is included to help identifywhich of the five key skill areas might be of most interest to you. Five trouble-shoot-ing cases are posed from a variety of industries and unit operations: distillation, heatexchange, pumps, adsorption and crystallization and that pose a range of difficulty.The results of the self-tests can be used to guide you as to how best to use theremaining chapters and appendix material.

The focus of this book is on developing skill in the mental process used to solvetrouble-shooting problems. Chapter 2 summarizes the research evidence of whatskilled trouble shooters do, provides a Trouble-Shooter’s Worksheet (and illustrates itsapplication) and a feedback form to help focus attention on the problem-solving,synthesis, data-handling and decision-making activities used. This gives you achance to compare the processes used with those used by skilled trouble shooters,and hence improve your skill and confidence in trouble shooting.

To illustrate the application of these skills five scripts are provided in Chapter 4 oftrouble shooters tackling the trouble-shooting problems Cases ’3–7. These are realproblems taken from industrial experience; only the names of the trouble shootershave been changed. Each of the five scripts consists of about three parts with eachpart concluding with a few questions for you to consider. This reflective break wasintroduced to give you a chance to reflect on how you would have handled the case,and to decide what you should do next. I recommend that, as you read each script,you play the game. An assessment is given of the problem-solving processes used by

XIV

Preface

each of the trouble shooters. Other examples of the process are given in Appendix Cand scattered as activities throughout most of the Chapters. Case ’3 in Chapter 6;Case ’6, in Chapter 6; Case ’8, in Chapters 2 and 6; Cases ’9 and 10, in Chapters5 and 6; Case’11 in Chapter 6; Cases ’12–18 in Chapter 7.

The central activity of the book is in Chapter 8. Here, trouble-shooting problemsare posed so as to help you develop your skill. The activity asks you to select anaction or question to take for each selected case (from about 30 possible actions). Acoded answer to each action is given in Appendix D. By posing a series of actions youwill gather evidence until you have “solved the problem”. Feedback about the process isgiven in Appendix E where an answer is given and key elements of the process usedby an experienced trouble shooter are listed. These problems are sequenced andclassified so that you can start with easy and familiar Cases and build up your con-fidence gradually. The classification notes the degree of difficulty, the type of equip-ment involved, and the chemicals/process technology involved. Some of the Casesrelate to similar processes. For example, six cases relate to the depropanizer-debuta-nizer system. Two, to the ethylene process; five, to the ammonia-reformer.

Since the trouble-shooting cases require the use of the five key skills, the rest ofthe book provides skill-development activities for each of these five skills.

Skill ’1: problem solving. The development of problem-solving skills is the theme ofChapters 5 and 6. In Chapter 5 the focus is on awareness, strategies, exploring theproblem, creativity and self-assessment. Target skills are given, activities are intro-duced, a range of tasks are given (in Chapter 5 and Appendix F) and feedback isprovided. Chapter 6 provides activities to develop skill in gathering data, checkinghypotheses and critical thinking. For the various skills being developed, the processis illustrated (in the context of a trouble-shooting case), and tasks are given. Thisactivity-based, workshop-style approach has been proven to be extremely effective;the proof is given in the award-winning paper” Developing problem-solving skills:the McMaster Problem Solving program, “Journal of Engineering Education”, April,vol 86, no 2, pp. 75–91, 1997. A wide range of tasks are provided from which you canselect those most pertinent to your experience with feedback available in Appendix G.

Skill ’2: knowledge of process equipment. Chapter 3 provides a convenient summaryof the practical aspects about equipment needed for trouble shooters of over 50 differ-ent types of process equipment. For most types of process equipment the followinginformation is given: overall fundamentals, guidelines for good operation, and trou-ble shooting. For trouble shooting, typical symptoms are given together with a prior-itized list of typical causes. Some will want to keep this text handy for just this sum-mary of practical know-how. More details are given in Appendices A, H and I.

Skill ’3: process safety and properties of materials. Chapter 3 also gives some succinctrules of thumb related to safety and hazard identification in Section 3.12.

Skill ’4: systems thinking. Guides to and activities to help develop “Asystems think-ing” are given in Chapter 3 in Sections 3.1 and 3.11 and Appendix B.

XV

Preface

Skill ’5: people skills. Chapter 7 addresses interpersonal skills and looks at the factorsinfluencing personal performance. More is given in Appendices C, F and G.

Chapter 9 offers ideas of what to do next.This book would not have been possible without the help of many. In the seven

companies for whom I worked before coming to McMaster University, I was fortu-nate to have worked with a variety of excellent trouble shooters who patiently helpedme polish my skill, Don Ormston and Ted Tyler of Distiller’s Company Ltd, Saltend,UK; Stan Chodkiewicz, Polysar, Sarnia, J. Mike F. Drake, British Geon Ltd, Barry,South Wales.

I thank Tom Marlin, Adam Warren, Iryna Bilovous, the late R.B. Anderson,Archie Hamielec, Terry Hoffman, Cam Crowe, John Vlachopoulos, Raja Ghosh,Douglas Dick, Dave Cowden and Lisa Crossley, McMaster University; Peter Silves-ton, University of Waterloo; Jud King and Scott Lynn, University of California, Ber-keley; Ian Doig, University of New South Wales; Frank Bajc; Pierre Cote, ZenonEnvironmental, Douglas R. Winter and Robert French, Universal Gravo-plast, inc,Toronto, my students, my alumni who sent back problems (and answers), partici-pants in the industrial workshops and in the conference workshops and Esso Chem-icals, Nova Corporation, Prices of Bromborough, Unilever who generously providedme with problems and gave me permission to use them.

McMaster Alumni who sent me Cases (and answers) include Bill Taylor (B Eng ’66),Ian Shaw (B. Eng. ’67), John Gates (B. Eng. ’68), Don Fox (B. Eng. ’73), R.J. Farrell(B. Eng. ’74), Jim Sweetman (B. Eng. ’77), Mike Dudzic (B Eng ’80), Mark Argentino(B. Eng. ’81), Vic Stanilawczik (B. Eng. ’83), Gary Mitchell (B. Eng. ’83), David Goad(B Eng and Mgt, ’91), Kyle Bouchard (B Eng ’93), Doug Coene (B. Eng. ’97) andJonathan Yip (B. Eng. ’97).

I have learned much from the cases solved and the approaches taken by NormanLieberman, David Saletan and Henry Kister that they published in their books andarticles.

I thank Tom Marlin, and Brendan J. Hyland (B Eng and Society, ’97). With finan-cial support from McMaster University Instructional Development program, theyproduced detailed versions of over 40 cases, some of which were used in this book.

I am especially indebted to Luis J. Rodriguez, Downstream Oil Company, Water-down; Douglas C. Pearson, Technical Support Consultant, Parry Sound and TomMarlin, McMaster University, who gave me feedback and detailed suggestions onthe case problems.

Many colleagues supplied me with interesting trouble-shooting cases and infor-mation about the cause and perhaps some details about the TS process used to solvethe problem. However, in writing up the interactive cases, I had to provide addi-tional information to flesh out the case, provide some red herrings and address abroad range of possible hypotheses so that the fault is not immediately obvious. Ihave done my best, and any errors in this elaboration are mine.

Waterdown, September 2005 Don Woods

XVI

1

Process plants operate about 28 days of the month to cover costs. The remainingdays in the month they operate to make a profit. If the process is down for five days,then the company cannot cover costs and no profit has been made. Engineers mustquickly and successfully solve any troublesome problems that occur. Sometimes theproblems occur during startup; sometimes, just after a maintenance turn-around;and sometimes unexpectedly during usual operation.

A trouble-shooting (TS) problem is one where something occurs that is unex-pected to such an extent that it is perceived that some corrective action may beneeded. The trouble occurs somewhere in a system that consists of various pieces ofinteracting equipment run by people. The TS “corrective” action required may be:

. to initiate emergency shut-down procedures,

. to forget the situation; it will eventually correct itself,

. to return the situation to “safe-park” and identify and correct the cause andtry to prevent a reoccurrence,

. to identify and correct the cause while the process continues to operate undercurrent conditions.

Here are two example TS problems.

Example Case’1:“During the startup of the ammonia synthesis reactors, the inlet and outlet valves to the

startup heater were opened. The pressure in the synthesis loop was equalized. The valves tothe high-pressure stage of the synthesis gas compressor were opened and the firing on thestart-up heater was increased. However, we experienced difficulty getting the fuel-gas pres-sure greater than 75 kPa; indeed a rumbling noise is heard if we try to increase the pres-sure. The process gas temperature is only 65 �C. What do you do?”

Example Case’2:“The pipe on the exit line from our ammonia storage tank burst between the vessel and

the valve. An uncontrolled jet of –33 �C ammonia is streaming out onto the ground. Whatdo you do?”

1

What is Trouble Shooting?


1 What is Trouble Shooting?

Trouble shooting is the process used to diagnose the fault safely and efficiently,decide on corrective action and prevent the fault from reoccurring. In this chapterwe summarize the characteristics of a trouble-shooting problem, give an overview ofthe trouble-shooting process and “systems” thinking used to correct the fault andpresent an overview of this book.

1.1Characteristics of a Trouble-Shooting Problem

TS problems share four common characteristics; TS problems differ in their serious-ness and when they occur. Here are the details for each.

1.1.1Similarities among TS Problems

TS problems share the following four characteristics: a) exhibit symptoms of devia-tions from the expected, b) have tight time constraints, c) are constrained by thephysical plant layout and d) involve people.

a) Trouble-shooting situations present symptoms. The symptoms may suggestfaults on the plant or they might be caused by trouble upstream or down-stream. The symptoms may be false and misleading because they result fromfaulty instruments or incorrect sampling. The symptoms might not reflectthe real problem. For example, in Example Case’1 the cause is not that thefuel-gas pressure is too low. Instead, the suction pressure of the synthesis gascompressor was lower than normal, the alarms on the cold bypass “low flow”meter had been disarmed and the real problem was that there was insuffi-cient process gas flow through the heater.

b) The time constraints relate to safety and to economics. Is the symptom indic-ative of a potential explosion or leak of toxic gas? Should we initiate immedi-ate shutdown and emergency procedures? The release of ammonia, in Exam-ple Case’2, causes an immediate safety hazard. Time is also an economicconstraint. Profit is lost for every minute when off-specification or no prod-uct is made.

c) The process configuration constrains a trouble shooter. The process is fabri-cated in a given way. The valves, lines and instruments are in fixed locations.We may want to measure or sample, but no easy way is available. We have towork within the existing process system.

d) Sometimes the cause of the problem is people. Someone may not have fol-lowed the expected procedure and was unwilling to admit error. Someonemay have opened the bypass valve in the belief that “the process operates bet-ter that way.” As in Case’1, the alarm may have been turned off. The orificeplate may have been put in backwards. Someone may have left his lunch inthe line during the construction. Instructions may have been misinterpreted.

2

1.2 Characteristics of the Process Used to Solve Trouble-Shooting Problems

1.1.2Differences between TS Problems

Here are four ways that TS problems differ. Some TS problems pose a) safety andhealth hazards. TS problems can arise b) during startup, c) after a shutdown formaintenance or after a change has been made and d) during usual operations.

1.2Characteristics of the Process Used to Solve Trouble-Shooting Problems

The TS process or strategy used differs depending on the type of TS problem. Yet,the TS process has five common key elements.

1.2.1How the Type of Problem Guides the TS Process or Strategy

The four different types of TS problems (described in Section 1.1.2) call for differentTS strategies.

. Handling trouble that poses a hazard

At the design stage engineers should anticipate causes of potentially unsafe anddangerous operation (through such analyses as HAZOP and fault tree) and preventhazardous conditions from ever occurring. They should include the four elementsof control: the usual control, alarms, system interlock shutdown, SIS, and shut-down/relief. However, despite best efforts trouble can occur – such as in ExampleCase’2.

The TS strategy is to recognize unsafe conditions and initiate emergency mea-sures or, where possible, to return the operation to “safe-park” conditions whereoperation is safe until the trouble is solved.

. Handling trouble during the startup of a new process

When we start up a process or new approach for the first time, we may encountertrouble-shooting problems. However, because these are “first-day” problems theyhave characteristics that differ from the usual trouble that can occur on an existingprocess. Hence, a different set of information or experience, and sometimesapproach, can be useful. In particular, four events could cause trouble:

1. garbage or stuff left in the lines or equipment,2. incorrect installation, for example, a pipe hooked up to the wrong vessel,3. during startup, there are often many people around to get things going cor-

rectly – this can interfere with the lines of communication,4. residual water or air left in process vessels and lines.

3


Furthermore, although we have theory and often computer simulations to provideideas about how the plant or process should be operating; we have no actual data.Example Case’1 is a startup problem.

The TS strategy is to focus on the basic underlying principles and create hypoth-eses about how the process and operations should function.

The financial penalty is usually higher for delays during startup. The penaltiesinclude penalties written into the contract for delays, insurance costs and govern-ment regulation costs.

. Handling trouble that occurs after a maintenance turnaround or a change.

Changes that can cause faulty operation include

1. equipment is taken apart for maintenance,2. processing conditions change because, for example, the feedstock is changed,3. there is a change in operating personnel.

In these examples, we have information about performance before and after thechange.

The TS strategy is to identify the change that seems to have triggered the fault.

. Handling trouble that occurs during usual operation or when conditionschange gradually.

Sometimes we encounter trouble when the process is operating “normally” orwhen we gradually increase the production rate.

The TS strategy is to focus on the basic underlying fundamentals of how the pro-cess works, create hypotheses that are consistent with the evidence and use tests toconfirm the hypothesis.

1.2.2Five Key Elements Common to the TS Process

Skill in trouble shooting depends on five key elements: 1) skill in problem solving,2) knowledge about a range of process equipment, 3) knowledge about the proper-ties, safety and unique characteristics of the specific chemicals and process condi-tions where the trouble occurs, 4) “system” thinking and 5) people skills. Here aresome details about each.

For general problem solving, one of the most important skills is in identifyingwhich evidence is significant and how the evidence relates to appropriate hypothesesand conclusions.

Concerning the importance of knowledge about process equipment, the differencesbetween skilled and unskilled trouble shooters are more in their repertory of theirexperiences than in differences in general problem-solving skills. In other words, itis the knowledge about process equipment, common faults, typical symptoms andtheir frequency that is of vital importance. A trouble-shooter’s effectiveness dependsprimarily on the quality of the knowledge that relates i) symptom to cause; and ii)the relative frequencies of the symptoms and the likelihood of causes.

4

1.3 Self-Test and Reflections

Specific knowledge about the chemicals and equipment configuration must beknown to handle safety and emergencies. For example, if knowledge of the hazardsof ammonia is not known, then Example Case’2 is not treated with the urgencyrequired.

Trouble occurs in a process “system” even though it might initially appear asthough it is in an isolated piece of equipment. Equipment interacts; people interactwith the equipment. Viewing the trouble-shooting problem in the context of a “sys-tem” is vital.

Interpersonal skills are needed. The interpersonal skills needed between the troubleshooter and the people with whom he/she must interact include good communica-tion and listening skills, building and maintaining trust and understanding howbiases, prejudice, and preferences lead to interpersonal differences in style.

1.3Self-Test and Reflections

Reflect on your trouble-shooting skills based on the five common key elementsdescribed in Section 1.2.2. Rate yourself-on the five or six elements in each categoryand then set goals to improve. A rating of 0 means that nothing is known. The max-imum scale is 10. Descriptions are given for ratings of 1, 5 and 10.

(1) Problem-solving skill as applied to trouble shooting

– Monitoring, being organized and focusing on accuracy: rate: _____

1 = aware that it’s important when problem solving. 5 =monitor about onceper 5 minutes, use a personal “strategy”, tend to let time pressures dominate.10 =monitor about once per minute, use an evidence-based strategy flexiblyand effectively, focus on accuracy, check and double check frequently.

– Data handling, collecting, evaluating and drawingconclusions: rate: _____

1 = think of a variety of data to be collected. 5 = systematically collect data thatseem to test the hypotheses, unclear of accuracy of data, unaware of commonfaults in reasoning, emphasis on opinions. 10 = systematically decide on datato collect and correctly identifies its usefulness; aware of the errors in mea-surements; use valid reasoning, focus on facts, aware of own biases in collect-ing data.

– Synthesis: creating and working with hypotheses asto the cause: rate: _____

1 = aware that should have a hypothesis. 5 = can identify several workinghypotheses that seem technically reasonable. 10 = can generate 5 to 7 techni-cally reasonable hypotheses for any situation; willing to change hypothesesin the light of new data.

5


– Decision making: rate: _____

1 = use intuitive criteria. 5 = systematic, consider many options, unaware ofany biases. 10 = use measurable must and want criteria explicitly, prioritizedecisions and aware of personal biases and try to overcome these.

(2) Experience with process equipment

– Centrifugal pumps: rate: _____

1 = flow capacity and head, location of inlet and exit, principle of operation.5 =NPSH and problems related to this, impact of reverse leads on the motor,correct location of the pressure gauge on the exit and the implications ofshutting the exit valve, pumps operate on the head-capacity curve and theimplications. 10 = implications of worn volute tongue and worn wear rings,lubrication, seals and glands.

– Shell and tube heat exchangers: rate: _____

1 = size area. 5 = size on area and Dp, baffle-window orientation, correction toMTD for multipass system, some options for control. 10 = tube vibrations,steam traps, nucleate versus film boiling and conditions, different causes offouling, maldistribution issues and can use a variety of control options.

– Distillation columns: rate: _____

1 = estimate the number of trays, know impact of feed conditions, reflux ratioand bottoms and overhead composition. 5 = familiar with a variety of inter-nals and can size/select, size downcomers, issues related to sealing downco-mers, familiar with some control options, can describe the interaction be-tween condenser and reboiler. 10 = jet versus downcomer flooding, surfacetension positive vs negative, pump arounds, vapor recompression, wide vari-ety of control options.

(3) Knowledge about safety and properties of material on the processes withwhich I workI can list the conditions and species that pose:

– Flammable risk rate: _____

1 = can identify individual species and conditions for five chemicals thatmight produce “flammable risk”. 5 = can identify individual and combina-tions of species and conditions for over 30 chemicals and the process faultsor failures that might produce “flammable risk”. 10 = can identify individualand combinations of species and conditions for over 100 chemicals and theprocess faults or failures that might produce “flammable risk”.

– Health risk rate: _____

1 = can identify individual species and conditions for five chemicals thatmight produce “health risk”. 5 = can identify individual and combinations of

6

1.3 Self-Test and Reflections

species and conditions for over 30 chemicals and the process faults or failuresthat might produce “health risk”. 10 = can identify individual and combina-tions of species and conditions for over 100 chemicals and the process faultsor failures that might produce “health risk”.

– Explosive risk rate: _____

1 = can identify individual species and conditions that might produce “explo-sive risk” for five chemicals. 5 = for over 30 chemicals and can identify indi-vidual and combinations of species and conditions and the process faults orfailures that might produce “explosive risk”. 10 = for over 100 chemicals andcan identify individual and combinations of species and conditions and theprocess faults or failures that might produce “explosive risk”.

– Mechanical risk rate: _____

1 = can identify pressure and moving equipment risk for about five types ofequipment. 5 = can identify overpressure, thermal and moving equipmentrisk for a P&ID with 20 pieces of Main Plant Items, MPI. 10 = can identifyoverpressure, thermal, corrosive and moving equipment risk for a P&ID with50 MPI.

– Unique physical and thermal properties: rate: _____

1 = can identify chemicals and conditions that have “unique properties” forone chemical. 5 = for 10 chemicals. 10 = for 30 chemicals.

(4) “Systems” thinking.

– Faulty operation of and carryover from/to upstream/downstream equip-ment: rate: _____

Can estimate/predict the effects of pulses, cycling, contamination on down-stream equipment. Can predict potential sources of pulses, cycling and con-tamination from upstream equipment. 1 = for one piece of equipment. 5 = fora P&ID with 10 MPI. 10 = for a P&ID with 40 MPI.

– Impact of environmental conditions rate: _____

1 = can estimate the environmental impact for the atmosphere from about 10main plant items. 5 = for about 20 MPI and atmospheric, aqueous and solidimpact. 10 = for about 50 MPI and atmospheric, aqueous and solid impact.

– Pressure profile: rate: _____

1 = can calculate a pressure profile for one pipe from detailed calculations.5 = can use rules of thumb to estimate the pressure profile for about five pip-ing configurations. 10 = can estimate pressure profiles for a P&ID with inter-connecting piping with 50 MPI.

7


– Process control: rate: _____

0 =Unable to identify and rationalize a process control system. 5 = For aP&ID with 10 MPI, can identify good and bad process control; can identifythe presence and absence of four levels of process control (control, alarm,SIS, relief and shutdown). 10 = For a P&ID with 40 pieces of equipment, canidentify good and bad process control; can identify on the P&ID the presenceof and absence of four levels of process control (control, alarm, SIS, reliefand shutdown).

(5) People skills

– Communication skills: rate: _____

1 =write or speak to tell them what you know, use acceptable grammar andfollow expected format. 5 = correctly identifies single audience, answersneeds and questions; includes some evidence related to conclusions, reason-ably well organized with summary, coherent and interesting, defines jargonor unfamiliar words, grammatically correct and follows the expected formatand style. Some misunderstanding occurs in some verbal or written instruc-tions. 10 = correctly identify multiple audiences, answer their needs and ques-tions; include evidence to support conclusions, well organized with summaryand advanced organizers, coherent and interesting, defines jargon or unfami-liar words, grammatically correct and follows the expected format and style.Verbal and written instructions are carried out correctly.

– Listening skills: rate: _____

1 = listen intuitively. 5 = aware of some elements of listening and usually candemonstrate attending. 10 = aware of the characteristics and foibles of listen-ing, skilled at opening conversations, attending, following and reflecting.

– Fundamentals of relationships: rate: _____

1 = handles relationships intuitively. 5 = aware of most of the fundamentalsand unacceptable behavior. 10 = claims and respects fundamental rights andavoids using contempt, criticism, withdrawal and defensiveness.

– Developing and building trust: rate: _____

1 = knows a few principles for developing trust; 5 = understands how todevelop trust. 10 = can develop mutual trust naturally.

– Building on another’s personal preferences: rate: _____

1 = intuitively aware of own preferences and that others are different.5 = explicitly aware of own preferred style and aware of uniqueness of othersbut not very effective in exploiting the differences positively. 10 = familiarwith my uniqueness and those of my colleagues and use the differences toimprove our work instead of promoting conflict.

8

1.6 Cases to Consider

Total your scores. Identify the areas with the lowest scores and set goals for your-self. For problem solving, see Chapters 2, 5 and 6. For experience with processequipment, see Chapter 3 and Appendix A. For knowledge about safety, see Chapter3. For “systems thinking”, see Chapter 3 and Appendix B. For people skills, seeChapter 7. If you have high scores in all areas, Congratulations. Go directly to Chap-ter 8 and enjoy!

1.4Overview of the Book

This book is about improving your approach to trouble shooting. This book has basi-cally five parts. Chapters 2 and 3 provide details about the mental process and prac-tical knowledge of common symptoms and causes for a variety of process equip-ment. Chapter 4 gives some examples of trouble shooters in action as they workthrough a variety of problems. This is included to give you a chance to reflect onyour approach. Chapters 5, 6 and 7 provide example training opportunities to polishyour skill in trouble shooting in the areas of problem solving, critical thinking andtesting hypotheses and interpersonal skills, respectively. Chapter 8 gives cases thatyou, the reader, can use to polish your skill. The final chapter suggests the next levelof considerations to polish your skill further.

1.5Summary

Trouble-shooting situations present symptoms, symptoms that may not reflect thereal problem. Trouble shooters are constrained by time and the existing equipmentlayout. Trouble-shooting situations inevitably include people.

Solving a trouble-shooting problem uses the five elements: skill in problem sol-ving, knowledge about equipment and about hazards, skill in systems thinking andpeople skills.

Problems occur that pose a hazard, when the process is started up for the firsttime, when the process is started up after change or maintenance or during usualoperations or when we are trying to increase the capacity of the process. Slightly dif-ferent TS strategies are used for the different types of TS problem.

1.6Cases to Consider

Here are five cases. Consider each and write out the approach you would take tostart each. For example, you might ask What is the problem? What questions might Iask? What are the possible causes? What tests might I do? What samples might be takenfor analysis?

9


Case’3: The Case of the cycling columnThe shut-down and annual maintenance on the iC4 column has just been com-pleted. When the operators begin to bring the column back on-stream the level inthe bottom of the column cycles madly, that is, the level rises slowly about 0.6 mabove the normal operating level and then quickly drops to about 0.6 m below nor-mal. The process then repeats. You have been called in as chief trouble shooter tocorrect this fault. It costs our company about $500/h when this plant is off-stream.Get this column working satisfactorily. The system is given in Figure 1-1.

Temp

1

SteamCondensate

Column

Trap

Figure 1-1 A distillation column for Case’3.

Case’4: The case of the platformer firesHeavy naphtha is converted into high octane gasoline in “Platforming”. Byproductsof the reaction include low-pressure gas and hydrogen-rich gas containing 60–80%hydrogen. The products from the platformer reactor (at 4.8 MPa g and 500 �C) areheat exchanged with the feed naphtha to preheat the reactor feed. Figure 1-2 illus-trates the layout. In the past three weeks since startup we have had four flash firesalong the flanges of the stainless steel, shell and tube heat exchanger. The plantmanager claims that because of the differential thermal expansion within the heatexchanger, because of the diameter of the exchanger (1 m), and because it’s hydro-gen, we’re bound to have these flash fires. The board of directors and the factorymanager, however, refuse to risk losing the $90 million plant. Although the loss indowntime is $10,000/h, they will not let the plant run under this flash-fire hazardcondition. “Fix it!” says the technical manager. Maintenance have already broken sixbolts trying to get the flange tighter, but they just can’t get the flanges tight enough.

10


Naphtha feed

Platformer

Figure 1-2 The platformer for Case’4.

Case’5: The sulfuric acid pump problemDilute sulfuric acid is stored in a horizontal, cylindrical tank in the basement, as isshown in Figure 1-3. The tank diameter is 1.8 m; the length, 3.6 m. An exit line goesfrom the bottom of the tank and rises 3.6 m up through the ground floor to a centri-fugal transfer pump that pumps the acid to a reservoir 7.5 m above the ground level.

acid return lines

basement

1.8 m

3.6 m

vent

siteguage

to reservoir atelevation +7.5 m

ground level

acid storage tank

Figure 1-3 The configuration for Case’5.

11


Acid is recycled to the tank from different parts of the process at such a rate thatabout every two hours the pump is activated to transfer the acid to the elevated reser-voir. However, each time the transfer pump operates, the level gauge at the side ofthe tank shows that there is still about 0.7 m of acid in the bottom of the tank whenthe transfer pump makes a “crackling” noise that the operator says “sounds like cavi-tation”. At this time the operator stops the pump. This means that the transferpump has to operate more frequently than need be and that cavitation may be erod-ing the impeller. What do you suggest that I do to fix the problem?

Case’6: The case of the utility dryer (courtesy of C.J. King, University of California,Berkeley)Our plant has a utility air drying unit, which dries all of the utility air used for thepneumatic instrument lines and other purposes. The air is compressed to about 550kPa g and then passes through the drying unit, the flow diagram of which isattached. For this unit, two beds are hooked in series with the first bed being regen-erated and the second bed drying. The first bed experiences two hours of regenera-tion with hot air followed by a one hour flow of cold incoming air to cool down theregenerated bed. The second bed drys the air for 3 hours. After 3 hours, the flowsswitch so that the regenerated bed becomes the drying bed and vice versa. The plantoperators are following the vendor’s instructions in setting the timer dials on thevarious valves: all the valves (the four way valves, V2 and V3, and the three-wayvalve, V1) are thrown every three hours. The 3-way valve, V1, is also thrown twohours after a cycle change to send fresh air to cool the regenerated bed. The hot airused to regenerate the bed is heated in a steam heater with the TRC-1 set at 175 �C.The present utility air flowrate through the dryer is 4000 Ndm3/s or about 1�2 designflowrate. The proportionating valve is governed by pressure P3. At present, full pres-sure is kept on the valve; the valve is shut so that no air goes directly to the dryerbed. The diagram shows the valve settings for Bed A being regenerated and Bed B,drying. All the air flow goes, via the 3-way valve V1 to the steam heater for the regen-eration phase of Bed A. The adsorbent is activated alumina with typically 0.14–0.22kg water adsorbed/kg dry solid. Each bed contains 5000 kg of activated alumina.The available sample valves are labeled “S”.

Now that it is winter we have been experiencing much colder nights, and we haveencountered several instances where the instrument air lines have been freezing.This has been traced to the air coming out of the drying unit being too wet, on aver-age. We estimate that this problem will cost us about $8,000 per day until we get itfixed. The job is yours – fix it.

12


Figure 1-4 The utility dryer for Case’6.

Case’7: The case of the reluctant crystallizer (the case is supplied by W.K. Taylor,B Eng. McMaster, 1966 and used with permission)Process solution, at 55 �C, enters the vacuum crystallizer (VC) where it is concen-trated and cooled to cause precipitation of the product.

Normally, the first and second stage ejectors are used to start syphoning feed so-lution into the VC until it is two-thirds to three-quarters full. The first hour of opera-tion is done at 6.5 kPa absolute supplied by the first- and second-stage ejectors withcity water to the interstage condenser. When the batch cools to 40 �C the boosterejector and the barometric condenser are turned on to give an absolute pressure of2.5 kPa abs. The batch time is 8 hours during this time the liquid level in the VCslowly drops about 40 to 50 cm. The city water is much colder than the bay waterand so to ensure that the temperatures in the barometric leg is less than 26 �C, citywater can be used to supplement the bay water. If the booster ejector is turned ontoo soon, it will not hold but rather kicks out. This happens when the steam goesdirectly into the VC instead of through the ejector nozzle. This phenomena makes arecognizable sound.

Today, the plant operator phones, “The booster does not hold! After about half toone hour of operation it kicks out.”

13


While the booster was holding, the liquid level dropped at such a “fantastic rate”that you could actually watch the level drop, whereas it would normally drop 40 to50 cm over an 8-hour period.

Pressure gauge P7 indicated a “wildly fluctuating pressure”. The needle jumpedback and forth from 140 to 550 kPa g while the booster was “holding”.

All the other pressures and temperatures were normal. Here is a summary:

Pressure, kPa g

Steam Water

P1 P2 P3 P4 P8 P5 P6 P7

Normal readings 550 550 550 725 550 0–35 310 205Today 550 550 550 725 550 0–35 310 140–550

Temperature, �C

Barometric legs

T1 T2

Normal readings < 27 < 27Today < 27 < 27

Figure 1-5 illustrates the system.

14


Figure 1-5 The vacuum crystallizer for Case’7.

Feedback for these cases is provided in Chapter 4.

15

17

A trouble-shooting problem occurs. As the trouble shooter processes the evidence,he/she mentally scans past experience to see if he/she has successfully solved any-thing like it before. If that past experience is limited; if there are no past examplesthat bear any relationship to the current troubling problem, then this is a “problemto solve”. The process used will be called “problem solving.” Problem solving is illus-trated in Figure 2-1. Data are gathered and an internal, mental representation is cre-ated of the problem situation. That mental representation is compared with pastexperience to see if a problem similar to this has been solved successfully in thepast. If not, then “it’s a problem!”. We systematically draw on our problem-solvingskills, scan our bank of pertinent knowledge and combine our skills and knowledgeto “solve the problem”. We then elaborate and take time to encode and store thatsuccessful solution to the problem in our mental experience bank. In the future,when we encounter a similar problem we don’t have to agonize through all of theproblem-solving process. We simply recall a solution from our experience in a pro-cess we call “exercise solving”.

“Exercise solving” is illustrated in Figure 2-2. Data are gathered and an internal,mental representation is created of the problem situation. That mental representa-tion is compared with past experience to see if a problem similar to this has beensolved successfully in the past. If yes, then “it’s an exercise!”. The past solution isrecalled from experience and modified to solve the current problem.

Beginning trouble shooters with limited experience start their journey as problemsolvers and gradually build up experience. Experienced trouble shooters are primarilyexercise solvers who draw on their knowledge and experience. Research evidence sug-gests that for experienced trouble shooters 95% of the situations they encounter willbe “exercises”. They still need problem-solving skill for 5% of the situations.

In Section 2.1 we summarize research about problem solving, in general. Thesecharacteristics of skilled problem solvers are used in solving any type of problemsuch as setting goals, making decisions, making a purchase and trouble shooting.

In Section 2.2 additional research evidence into the process of solving trouble-shooting problems is given. In Section 2.3 is given a worksheet or template for sol-ving trouble-shooting problems. Also given is an assessment form to provide feed-back about one’s performance as a trouble shooter. An example use of the Work-sheet is given in Section 2.4 for Case’8.

2

The Mental Problem-Solving Process used in Trouble Shooting


2 The Mental Problem-Solving Process used in Trouble Shooting

Figure 2-1 Problem solving.

18

2.1 Problem Solving

Figure 2-2 Exercise solving.

2.1Problem Solving

Here are eighteen characteristics of skilled problem solvers. The first eight could becalled “the problem-solving process or how”; the second set of characteristics arecalled “synthesis”. The third class is called “decision making”. The other importantcharacteristics, “data and analysis”, are given in Section 2.2.3. Research has shownthat attitudes and other related skills are also important.

19


The “problem-solving process, how”:

1. Be able to describe your thought processes as you solve problems.2. Be organized and systematic. An evidence-based strategy for solving problems,

in general, is given in Figure 2-3. The six stages are as follows. 1. Engagewith the problem or dilemma, listen, read carefully and manage your distresswell. Say “I want to and I can!” 2. Analyze the data available and classify it:the “goal, the givens, the system, the constraints and the criteria”. 3. Explore:build up a rich visual/mental picture of the problem and its environment;through simplifying assumptions explore the problem to see what is reallyimportant; identify the real problem. 4. Plan your approach to solving theproblem. 5. Carry out the plan and 6. Check the accuracy and pertinence ofyour answer. Did it answer the problem? satisfy the criteria? Reflect on theproblem-solving process used to discover new insights about problem sol-ving. Elaborate on the answer and the problem situation to discover answersto other problems, to extend the solution to other situations and relate thisproblem experience to other technical problems you have solved in the past.Cue this experience into memory. This systematic approach is not sequential.Skilled problem solvers bounce back and forth between the stages. A typicalapproach would be engage, analyze, engage, explore, engage, explore, ana-lyze, engage, explore, plan, engage and so on.

3. Focus on accuracy instead of speed.4. Actively write things down. Make charts, draw diagrams, write down goals, list

measurable criteria and record ideas from brainstorming.5. Monitor and reflect. Mentally keep track of the problem-solving process and

monitor about once per minute. Typical monitoring thoughts are “Have I fin-ished this stage? What have I discovered so far? Why am I doing this: if I calculatethis, what will this tell me? What do I do next? What seems to be the problem? Isthis the real problem? Should I recheck the criteria?” Typical reflections that lookback on the process and attitudes used are: “This didn’t work, so what have Ilearned? Am I focusing on accuracy or am I letting the time pressures push me tomake mistakes? Am I managing my stress? I can do this! Am I monitoring theprocess? “

6. Explore the “real” problem by creating a rich perspective of the problem. Duringthe explore stage, see it from many different points of view. Be willing tospend at least half the total available time defining the problem. Ask manywhat if questions. Try to bound the problem space. “Swim with the data” tosee how it responds. Identify the real problem, by asking a series of Why?questions to generalize the situation and to see the problem in the context ofa “system”. This activity of identifying the real problem was called the Explorestage and is the heart of the problem-solving process.

7. Identify the subcomponents of the problem, yet keep the problem in perspective.8. Are skilled at creative and critical thinking.

20

2.1 Problem Solving

Figure 2-3 The MPS Strategy for problem solving.

These first eight items we could call “Problem-solving process, how”. Table 2-1lists detracting and enriching behaviors. Activities to help develop these skills aregiven in Chapter 5.

The next two items are related to “Synthesis” with detracting and enriching behav-iors listed in Table 2-1.

21


9. Be flexible.10. Keep at least five hypotheses active. Do not quickly close on one hypothesis.

Issues related to “decision making” are next.

11. Spend time where it benefits you the most. Use Pareto’s principle (80% of theresults can be found from 20% of the effort). Find the key 20%.

12. Be an effective decision maker. Express the goal as results to be achieved ratherthan as actions to be taken. Make decisions based on criteria that are explicitand measurable. Distinguish between must criteria (the process must havean internal rate of return of 35%) and want criteria (the process might havethe potential to be licensed). Reject options that do not meet the must criteria.Use a rating system to score the want criteria.

The remaining research evidence relates to “attitude toward problem solving” andsome related skills.

13. See challenges and failure as opportunities for new perspectives.14. Be willing to risk.15. Manage stress well. Solving problems is stressful. When we initially encounter

a problem we experience distress because of the uncertainty. Such stresstends to immobilize us. When we successfully solve a problem we experiencethe joy and exhilaration of stress (that distracts us from checking and doublechecking that our answer is the best). A certain level of stress motivates us.Excessive stress makes us make mistakes. Data suggest that operators withconfidence and training working under high stress make 1 mistake in 10actions. Operators with confidence and training who receive feedback abouttheir actions and are under low stress make 1 mistake in 1000 actions. Al-though these data refer to plant operators, the same trends can be extendedto suggest how stress, lack of reflection and feedback might interfere withengineering practice. High stress would be a rating of over 450 on theHolmes–Rahe scale (Holmes and Rahe, 1967).Ten suggested approaches to managing stress include: worry only about

things over which you have control, include physical exercise as part of yourroutine, have hobbies and destimulating activities in which you can lose your-self, plan ahead, avoid negative self-talk, rename the events that are stressfulto you, build a support system, be decisive, put the situation into perspectiveand use role models of others who have succeeded.

16. Manage your time well. Covey (1990) offers excellent suggestions on timemanagement. Identify problems and decisions according to their importanceand urgency. Shift the important situations to being non-urgent. Learn to say“No”.

17. Understand your strengths, limitations and preferred style. See Section 6.1.3, part c.18. For problems involving people, use the 85/15 rule. 85% of the problems occur

because of rules and regulations; 15% of the problems are because of people.

22

2.2 Trouble Shooting

2.2Trouble Shooting

The general problem-solving characteristics listed in Section 2.1 apply to trouble-shooting problems. However, unique to this type of problem are other research find-ings. Here we summarize the use of the strategy, in Section 2.2.1, general considera-tions, in Section 2.2.2 and hypothesis testing in Section 2.2.3.

2.2.1Considerations when Applying the Strategy to Solve Trouble-Shooting Problems

Two different types of ideas help us focus on our use of the general strategy: prob-lems where there is an apparent change and where there is no apparent change.

a. Using this Strategy for “Change” ProblemsThe overall strategy, described in Section 2.1.1, is applied to identify the change thatoccurred to cause to trouble. The hypothesis is that the symptoms arise because ofsome change made to the system. Therefore the plan is to identify the change. Thebasis of the approach is to learn to ask the right questions. Kepner–Tregoe (1985)illustrate the application of this approach. The questions that usually are most help-ful are those that help identify an obvious change.

. what is happening and what should be happening but is not, and “is thisdifference significant?”

. where is and where is not,

. who is and who is not,

. when is and when is not.

This approach is usually most helpful for “people problems”, for problems thatoccur just after maintenance and for processes that have worked well in the past andnow seem to be malfunctioning after a change has been made – in raw material, inoperating procedure, in operators, in weather or in season.

b. Using this Strategy for “Basics” ProblemsThe same overall strategy, described in Section 2.1.1, is applied when a change isnot obvious. The emphasis is different in that we focus on the basics instead of on achange. The conditions when this apply could be because 1) we are starting up aprocess for the first time, and we have no practical data of what should be happen-ing or 2) something internal to the process changes and we have no simple way toidentify that change. No one ordered the raw materials from a new supplier. No onerepaired a pump. No one changed the temperature setting on the heater. Instead,inside the equipment a hunk of corroded metal fell into the liquid; or a truss weak-ened and gave way inside the vessel. The catalyst bed collapsed. We cannot easilyidentify the change because there is nothing to “see” from the outside. For this situ-ation, we rely on our fundamental principles and knowledge of how the process andequipment should operate, we create hypotheses, check for consistency between the

23


hypotheses and the evidence and test the hypothesis. The questions that tend to besuccessful are as follows.

. What basically is going on in this process?

. What fundamentals are important?

. What are the key operating principles that guide the operation of the equip-ment?

. How are the fundamentals reflected in the observed data?

2.2.2Problem-Solving Processes Used by Skilled Trouble Shooters

Here are the characteristics of skilled trouble shooters especially when we are tryingto diagnose the cause when using a “basics” strategy.

1. Generate hypotheses early based on limited cues. Consider the most com-mon hypothesis first.

2. Be systematic and organized: each piece of information requested shouldrelate to an organized plan of attack.

3. Make hypotheses consistent with the evidence: no hypothesis should bemore specific or more general than the evidence justifies.

4. Keep two to five competing hypotheses under consideration at any one time.5. Explore the option of multiple causes especially when evidence suggests a

single rare cause. Do not neglect the possibility of two common causes. Iftwo or more common causes would produce disastrous results, and you can-not confirm or refute these causes; act as if both are the cause.

6. Whenever a new or revised hypothesis is generated, check the implications ofprevious cues.

7. Prioritize test procedures; use the simple, inexpensive ones first beforeexploring the high-cost option.

8. Do not guess. Use a systematic TS process.9. Find the root cause; do not correct the symptoms.

Some example data (from the Medical literature, Elstein et al., 1978) are:

. total different bits of information sought/given: 200 per case,

. total bits accumulated before the first hypothesis was generated: 20,

. number of active hypotheses: 3,

. total number of different hypotheses throughout whole case: 6,

. number of cues acquired: 50

. number of critical cues acquired: 50

Table 2-1 describes detracting and enriching behaviors for these activities underthe topic synthesis.

24

2.3 Overall Summary of Major Skills and a Worksheet

2.2.3Data Collection and Analysis: Approaches Used to Test Hypotheses

Successful trouble shooters

. assign a weighting of +++, ++, +, 0, –, – –, – – – to the cues/evidence andthen select the hypothesis that maximizes the positive cues or that has themaximum difference between positive and negative cues.

. use Bayes’ theorem if the probabilities of various causes are known.

. are sensitive to and try to overcome personal biases (related to premature clo-sure and anchoring).

. consider the evidence with respect to all hypotheses (to overcome the mostcommonly encountered bias of pseudodiagnosticity or overinterpretation).

. gather data to disconfirm a hypothesis and are willing to discard a “favored”hypothesis (to overcome confirmation bias).

. consistently use fundamentals when analyzing the evidence-hypothesis link(to overcome representativeness bias).

. use diagrams, trees and tables to systematically chart hypotheses, cause andevidence (to overcome omitting cues and overcome the limitations of Short-Term Memory).

. restrain from creating a new hypothesis for each new clue and thereby gener-ate excessive data and have trouble with closure.

Table 2-1 lists detracting and enriching behaviors for these activities under thetwo topics of data analysis and decision making. Ideas on how to improve this skillare given in Chapters 5 and 6.

2.3Overall Summary of Major Skills and a Worksheet

The research summarized in Sections 2.1 and 2.2 can be converted into a Trouble-Shooter’s Worksheet to guide our approach and an assessment form, to give feed-back for growth. These are discussed in turn.

2.3.1Getting Organized: the Use of a Trouble-Shooter’s Worksheet

To help us to be systematic, we use the McMaster 6-step strategy in Figure 2-3 andconvert this into a worksheet. A succinct version highlighting the key features isgiven in Trouble shooters Worksheet 2-1.

25

2 The Mental Problem-Solving Process used in Trouble Shooting26

Trouble-Shooter’s Worksheet 2-1: Succinct summary(� copyright 2003 D. R. Woods and T. E. Marlin)

1. Engage: Gather initial information.

. Establish if emergency priority: safety? damage? shut down or safe-parkor continue?

. Describe what’s going on.

. Manage panic: “I want to and I can.”

. Monitor: Have you finished this stage? Can you check? What next?

2. Define the stated problem: based on given information. If the information isnot known at the stage, gather it later.

IS IS NOT

WHAT __________________________________________

(should be happening but it is not)_________________________________

WHEN ?& _________________ ?& ______________________________WHERE ?& _________________ ?& ______________________________WHO ?& _________________ ?& ______________________________

Identify situation as 1) startup new process; 2) startup after maintenance orchange, 3) usual operation. Monitor: Have you finished this stage? Can youcheck? What next?

3. Explore: Exercise? or problem? Strategy for change or basics? Useful tobroaden withWhy? Why? Why?

Gather information. Perspectives: customers? suppliers? weather? changedeconomics? politics? environment?

. Prioritize: product quality, production rate or profit?

. Goal: safe-park? short term? long term? SMARTS$

. Data consistency? Pertinent fundamentals? Likelihood of problem type.

. Explore with What if?

. List changes made and/or list trouble-shooting experience: root causesbased on symptom. (Chapter 3)

. Brainstorm hypotheses

. Hypotheses and evidence of symptoms:

Evidence of symptoms: a. ________________ b. ________________________

c. ___________________ d. ___________________ e. ___________________

2.3 Overall Summary of Major Skills and a Worksheet 27

Working Hypotheses Initial Evidence Diagnostic Actions

a b c d e A B C D

1

2

3

4

5

S = supports; D = disproves, N = neutral

Diagnostic actions: A. ______________________ B. _____________________

C. __________________________ D. _________________________________

4. Plan5. Do it6. Look back

1. Engage: Take in the evidence. Listen carefully to the phone call. Sense the evi-dence. Establish priority: Safety? Hazard containment? Equipment damage? If yes,then invoke emergency measures or alter conditions to safe-park the process in theinterim. For example, a distillation column might be “parked” by isolating it and continu-ing to operate under full reflux. If emergency or safe-park options are not needed, con-tinue. Take time to really understand the physical process. We find it useful to writedown a word description of what happens in the process. If a diagram is available,trace around the lines and describe what is flowing in each line, how it is controlledand what should be happening. Once you feel that you have some understanding ofthe process, manage your distress by saying “I want to and I can. I have a strategythat works. Let’s systematically follow it.”

2. Define the stated problem Understand the given information. Classify andchart the information using IS and IS NOT for WHAT, WHERE, WHEN and WHO.List the symptoms as given. Don’t guess! Don’t overinterpret! Don’t infer! Just sys-tematically classify the given information. Some may not be known, so identify thisas information to be gathered. Note whether this is startup of a new plant, startup aftermaintenance or change or usual operations. If this is not known, then identify thisas information to be gathered.

3. Explore: Build a rich description of the situation. Gather information to be gath-ered noted in Define. See the situation from many different perspectives. Decide ifthis is an exercise or a problem. Decide if a more effective strategy might be to focuson change or basics.


Gain a rich perspective about the situation by considering the problem situationfrom the following points of view: a) viewpoint of fundamentals and practical oper-ability, b) viewpoint of trouble-shooting rules of thumb: likelihood specific to type ofsituation or conditions, example startup (from Section 3.1), likelihood specific toequipment (Chapter 3 plus experience) c) viewpoint of controlling factors via Whatif? plus order-of-magnitude estimates to bound the behavior and identify keyassumptions and variables, d) viewpoint of isolated equipment, equipment in thecontext of a subsystem, in the context of a mini-system, in the context of the plantsite including utilities and in the context of the corporation; e) viewpoint in contextof the weather, political, environmental, legal and economic environment; f) view-point of trends and time changes, g) perhaps broaden the context of the situation byasking Why? Why? Why? h) viewpoint of stakeholders: customers, plant manager,operators, vendors, process control, auditors, statisticians, instrument and controlspecialists, or unit operation specialist.

Then select the “real” goal. Decide priority based on product quality, productionrate and profit. Describe the goal and criteria with the aid of the acronymSMARTS$, introduced by colleague Tom Marlin.

The Goal should describe results – not actions – and be expressed in Specific andMeasurable terms. The Goal should be Attainable. The Goal should produce a Reli-able and stable result. Use as Criteria: Timely: the problem should be solved quickly.Promptness is critical for emergency priorities in safety, containment of hazardsand damage to equipment. This was considered in the Engage stage. The Goalshould produce a Safe operation, safe product, safe startup and shutdown. Use asCriteria: $ The downtime costs, the testing costs and the corrective and preventativecosts should be minimized.

Chart the given evidence and symptoms.Brainstorm hypotheses as to the root cause. Defer judgement. Don’t be afraid to

list crazy ideas and to spend an extra five minutes dreaming up and building onwild ideas. Remember, often the unique ideas are found in the midst of completelyuseless ideas.

List five to seven working hypotheses as to the cause. These should be expressedas potential root cause rather than symptom cause.

Systematically chart and check, based on experience and fundamentals, as towhether the evidence Supports, Disproves or is Neutral toward each hypothesis.Look for obvious flaws and inconsistencies. For example, the measured data on arefrigeration unit are not consistent with refrigerant data on a pressure–enthalpy diagram.or On a pipe, the pressure gauge downstream reads higher than the upstream gauge.

4. PlanUse the criteria to select a sequence of diagnostic actions. Sometimes several

actions can be combined. However, usually it is best to wait for the results from thefirst action before we recycle back to the Explore stage and relook at our hypotheses.The criteria in selecting an action include: will the action provide background infor-mation to ensure the problem is understood in context or is the action to test ahypothesis? will the action produce results that give the accuracy needed? is the

28


action simple? inexpensive? safe? Stopping production to “inspect” or “changeequipment” is usually very costly.

5. Do itCarry out the first action in the plan. Check. Monitor.6. Look backCompare the results obtained with the hypotheses. Look back at the process used.

Self-assess. Return to previous stages of Engage, Explore, Plan and continue.An example of the use of this Trouble-Shooter’s Worksheet is given in Section

2.4.

2.3.2Feedback about your Trouble Shooting

Based on the evidence presented in Sections 2.1 and 2.2, the four things to look forin an effective trouble shooter are: the overall approach to problem solving, datahandling, synthesis, and decision making. Table 2-1 summarizes the detracting andenriching behaviors for each. Worksheet 2-2 provides a summary worksheet thatcan be used by you to self-assess your TS process or can be completed by an observerto give you feedback about your TS process.

Table 2-1 Summary of detracting and enriching behaviors for trouble shooters.

Problem solving in general: How you did it

Theme Detracting behaviors Enriching behaviors

Monitors the thoughtprocess

No assessment of potentialgain from a question or action.

Asks “What will this get me?”

Unclear of type and purposeof question asked; just askswhat pops into mind.

Knows clearly the purpose: ask fish-ing or shooting questions; whethercreating hypotheses or checking forchange or gathering information forclarification.

Does not monitor or askquestions as to Why?or implications.

Asks “Am I through?”, “Am I finishedwith this task?”, “Where is this lead-ing me?”, “This should tell me ...”

Checks and doublechecks

Assumes everything is OK.Does not check instruments,diagrams, hardware, procedures.

Checks and double checks instru-ments; checks if the equipment andlines are as on diagrams. Calibratesand recalibrates instruments.

Is systematic Jumps all around, confused,and no apparent plan.

Identifies plan and follows it system-atically yet flexibly. Uses tables orcharts to keep track of idea flow.

29


Subproblems andperspective

Keeps whole problem anddoes not identify subproblems.No identification of a strategy.

Breaks overall task into ones of situa-tion clarification, or hypothesis test-ing and/or identify the change; intoemergency action, cause identifica-tion, fault correction and futureproblem prevention.

Confuses issues, factors,fault detection, solutions.

Identifies phases clearly and worksthrough systematically.

Solves a minor fault whilethe process explodes.

Keeps situation in perspective, doesnot get lost in a subproblem.

Data handling: What you did


Data resolution Gathers data but does notknow what it tells him/her.

Correctly identifies the usefulness ofthe data collected.

Asks any old question. Matches hypotheses with the observedevidence to see if the hypothesis is con-sistent with the evidence.

Believes all he/she seesand hears; unclear of errorsin information.

Explicitly states limitations of theinstruments, measurements andchecks these systematically.

No data gathered explicitly.Jumps in making correctiveaction without statingpossible hypothesis or cause.

Gathers data for problem clarificationand hypothesis testing/or changerather than jumping in with correc-tive action without any data.

Gathers data expensively,Takes process apart foreverything. Overlookssimple ways of gatheringinformation.

Gathers data easily through simplechanges in operating procedure, putscontrollers on manual.

Asks for samples, but assumesthat sample locations andprocedures are as usual.

Is present when samples are taken,bottles labelled.

Gives imprecise instructions:“Check out the instrument”;“Open up the exchanger”.

Gives precise instructions.

Actions based onfundamentals

Based on intuition. Based on fundamentals; estimates be-havior based on fundamentals.Does mass and energy balances withat least two independent measure-ments.

Does pressure profiles through units.

30


Reasoning Jumps to invalid conclusions. Draws valid conclusions; tests bothpositive and negative: what is; what isnot; if it does happen; if it does not.

Error in inference:Confirmational bias.

Completeness Uses only part of the information.Doesn’t check the designcalculations, or data fromstartup or data from initial, cleanfluid; didn’t think of human error.

Uses all resources.

Synthesis: How you put it all together


Hypotheses Becomes fixed, thinks of onlyor selects one hypotheses;selects one at the start andcannot become unfixed.

Keeps at least four working hypoth-eses; keeps options open as data aregathered.

Makes everything complex. Keeps it simple, especially if there is a“big failure”.

One view. Many viewpoints: operators, design,human error, instruments, corrosion.

Critical of ideas; limitedbrainstorming.

Defers judgement when appropriate.

Flexibility Considers only a “basics”strategy or a “change” strategyand sticks with it regardless ofthe evidence.

Selects either a “basics” strategy or a“change” strategy. Shifts strategywhen its warranted.

Considers steady state only;considers only the facts

Considers unsteady state as well; con-siders the people too (the stress theymight be under; the environment thatallows open discussion; turf fights).

Overall synthesis Cannot put all the ideastogether into a reasonablestory. Becomes fixed on onecause even when evidencepoints otherwise.

Can put the ideas together into a plau-sible explanation.

31


Decision making: How you put it all together


Priorities No priorities for hypotheses;just start anywhere; indeed,may not even create a hypothesis!

Sets and uses priorities.Keeps track and moves from toppriority to second. At least four work-ing one hypothesis; keeps optionsopen as data are gathered.

No priorities for gatheringevidence; just collects something

Prioritizes; gathers the easy andcheap tests first. Visits the site.

No priorities about the urgencyof the situation; diddlesaround while the plantexplodes; keeps customerswaiting until problemcompletely solved.

Prioritizes urgency; willing to use acontingency plan to get things goingsafely and later corrects the real fault.

Bias Biased, stacks the deck so thefavorite fault will be selectedeven when the evidence refutes it.

Unbiased. Selects either a “basics”strategy or a “change” strategy. Shiftsstrategy when it is warranted.

Biased: tests for only positiveelements.

Tests for both positive and negative.Proves that hypothesis is correct andthe other options are not correct.

Overall process No criteria used, or if they are,they are not measurable.

Uses measurable criteria to makedecisions.

32

2.3 Overall Summary of Major Skills and a Worksheet 33

Worksheet 2-2: Summary observation form for feedback about Trouble Shooting.

TS name _____________________ Case _____ Initials ES _____ Obs ___Rough work area:Process: how Data/analysis: whatMonitoring _____________________ Data resolution _______________Checking _______________________ Fundamentals? _______________Systematic ______________________ Reasoning ___________________Subs and perspective _____________ Completeness ________________

Decision making: how Synthesis: whatPriorities ______________________ Hypotheses __________________Bias ___________________________ Flexibility ____________________

Rating and Feedback

Clarity of Communication

MostSomeNone All

Process used:

MostSomeNone All

Data collections and analysis:

MostSomeNone All

Synthesis:

MostSomeNone All

Decision-making:

MostSomeNone All

Five Strengths:________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Two areas for improvement________________________________________________________________________________________________________________________________


Figu

re2-4

P&ID

forthede

prop

anizer

andde

butanizer.

2.4 Example Use of the Trouble-Shooter’s Worksheet

2.4Example Use of the Trouble-Shooter’s Worksheet

Case’8 Depropanizer: The temperatures go crazy (used courtesy of T.E. Marlin,McMaster University, Hamilton, Canada)The process shown in Figure 2-4 is being started up for the first time. It has beenrunning well for several shifts. Just an hour ago, a new operator came on duty andthe operator changed the pressure at which the Depropanizer, C-8, is operated,raising the pressure by 0.1 MPa. About 10 minutes after the pressure was increased,the tray temperatures began to go crazy and the bottoms level started to decrease.Figure 2.4 shows the P and ID for this process.

35

Trouble-Shooter’s Worksheet 2-3: Case’8: The depropanizer: the temperaturesgo crazy (� copyright 2003 Donald R. Woods and Thomas E. Marlin)

1. Engage: Write down what is said; what you sense, smell, hear. If someone istelling you, then use skilled reflective statements to ensure you accuratelyobtain the information.

. Emergency priority: Safety? Hazard? Equipment damage?shut down &; safe park &. If not &, then:

. Draw a sketch of the process and mark on values. Provide a descriptionin words of what is going on. This is a simple distillation column but allthe piping and instrumentation details make it look complex. On theLHS, feed from upstream processing enters drum V29. This feed ispumped (via either a steam driven or motor driven centrifugal pump,F25, 26) through a preheater, E24, and into the depropanizer at tray 18.As the name suggests, the purpose of this column is to take overhead“propane and all lighter species”. Let’s follow the overhead. The overheadis condensed in two condensers in series, E25, collected in overheaddrum V-30 with the non-condensibles (such as methane and hydrogen)removed from the drum and vented to the fuel-gas system. The pressureon the column, C8, is controlled by the valve on the vent system, PV 10.Condensed propane is pumped, F-27, from the drum V-30 forward asproduct, through product cooler E-26, and returned to the column asreflux. The reflux is flow controlled. Following the bottoms: a thermosy-phon reboiler is steam heated. The bottoms flows forward to the nextcolumn, the debutanizer. No pump is needed because of the pressuredifference between the depropanizer, 1.7 MPa, and the debutanizer, 0.48MPa. I’m not sure at this stage if this is a control “problem” so I won’telaborate further on the system at this time. I also will focus on thedepropanizer, and not explore the debutanizer at this time.


. Manage any panic you might feel by saying “I want to and I can. I have astrategy that works. Let’s systematically follow it.”

. Monitor: Have you finished this stage? Can you check? What next? I’vesystematically followed my way around the flow diagram. I think Iunderstand enough for now.

2. Define the stated problem: Systematically classify the given information usingIS and IS NOT. If the information is not known at the stage check ?& toremind you to gather this information

IS IS NOT

WHAT tray temperatures “go crazy”;bottoms level decreases.

(should be happening but it’s not)tray temperatures steady; bottoms level steady.

WHEN ? & 10 minutes after thepressure in column C8increased by 0.1 MPa.

? & before the pressure increase; running wellfor several shifts. First plant startup.

WHERE ? & depropanizer, C8. ? &� maybe upstream; no information aboutdownstream debutanizer, yet!

WHO ? & new operator. ? & not with previous operators.

– startup new process &� suggest use Basics– startup after maintenance or change & suggest use Change– usual operation but changes made in operation but not in equipment& suggest use Basics

– usual operation & suggest use Basics.

. Monitor: Have you finished this stage? Can you check? What next? Yes. Ithink I’ve finished.

3. Explore: Gather information to be gathered ? & in Define stage &. Perhapsquestions about downstream effects. The decrease in liquid level could bebecause the flow has increased to the debutanizer (because the Dp) hasincreased.

Exercise? & or a problem? &�. I haven’t seen anything like this beforeStrategy: change & or basics &� .Perspectives. Why? Why? Why? no information at this time that suggests

this might be useful.____________________________________________________________

Why? ›____________________________________________________________

Why? ›____________________________________________________________

2.4 Example Use of the Trouble-Shooter’s Worksheet 37

Why? ›____________________________________________________________

Why? ›____________________________________________________________

Why? ›

Start fi _________________________________________________________

. Prioritize: product quality &�; production rate &; profit &

. Goal: safe-park? &�: short term now with long term later &�;long term now &

Action to be achieved: Specific terms and Measurable:level out the temperatures and the bottoms level

Attainable? total reflux is a start but I hope it’s attainableReliable? depends on my short-term solutionTimely? will work on solving it quicklySafe? cannot think of major hazard now$

. Check consistency of data/symptoms: inter-data consistency?OK & no &data consistent with fundamentals?OK & no &

. Type of problem: startup new process &� maybe mechanical electricalfailureusual operation &�: ambient temp? &maybe fluids problems;high temperature? & then maybe materials problemsSystem? failure of heat exchanger & > rotating equipment & >vessels & > towers &

. Identify key andWhat if?

What if? temperatures “going crazy” = temperature cyclingthen focus on cycling symptoms / causes

What if? only temperature “cycling” and no decrease in levelthen bottoms and tops temperature and pressures should be

cycling tooWhat if? only bottoms level drop and no temperature “going crazy”

then root cause related to bottoms level dropWhat if? column pressure increases

then condensation temperature at top increases; DT condenserincreases and condensation should be easier; boiling tem-perature at bottoms increases; DT reboiler decreases somight shift from film to nucleate boiling giving higher heatflux, causing increased boilup or if nucleate to start withthen insufficient area and boilup decreases.


. List changes made & and/or trouble-shooting causes based on symp-tom &� .

“Crazy temperatures” and decreasing bottoms level sounds like a control prob-lem.

From Chapter 3, no symptoms listed for bottoms level dropping, but symp-toms related to “oversized condenser” and “undersized reboiler” are:

“Insufficient boilup”: [ fouling on process side]*/ condensate flooding, seesteam trap malfunction, Section 3.5 including higher pressure in the conden-sate header/ inadequate heat supply, steam valve closed, superheated steam/boiling point elevation of the bottoms/ inert blanketing/ film boiling/ increasein pressure for the process side/ feed richer in the higher boiling components/undersized reboiler/ control system fault/ for distillation, overdesigned conden-ser.

But if there was “insufficient boilup”, then the bottoms level should beincreasing and not decreasing. This doesn’t make sense??

For:“Cycling of column temperatures:” controller fault/ [ jet flooding]*/ [downcomer

flooding]*/ [ foaming]*/ [dry trays]* with each of the []* items listed as separatesymptoms with their own root causes.

[ jet flooding]*: excess loading/ fouled trays/ plugged holes in tray/ restrictedtransfer area/ poor vapor distribution/ wrong introduction of feed fluid/ [ foam-ing]*/ feed temperature too low/ high boilup/ entrainment of liquid because ofexcessive vapor velocity through the trays/water in a hydrocarbon column.

[downcomer flooding]*: excessive liquid load/ restrictions/ inward leaking ofvapor into downcomer/ wrong feed introduction/ poor design of downcomerson bottom trays/ unsealed downcomers/ [ foaming]*

[ foaming]*: surfactants present/ surface tension positive system/ operatingtoo close to the critical temperature and pressure of the species/ dirt and corro-sion solids.

[Dry trays]*: flooded above/ insufficient reflux/ low feedrate/ high boilup /feed temperature too high.

. Brainstorm root causes: summary of major ideas generated:

change in feed, too much overheads in feed, not enough feed, tray collapsedin stripping section, too much vaporized feed, pump F26 failure, pump F26cavitates, increased pressure and DT increases, dry tray, flooded in rectification,insufficient reflux, low feedrate, feed temperature too high, boilup too high, toomuch feed to debutanizer, leak in bottoms, vaporizer flashes 90% (instead of67%), failure of check-valve on idle pump outlet, boilup controller fault.

Some of these are symptoms and not root cause, e.g. “not enough feed” “dry trays”

. Hypotheses: list in Chart; Symptoms: code and list in chart; Analyzewith S supports; D disproves and N neutral or can’t tell.

2.4 Example Use of the Trouble-Shooter’s Worksheet 39

Symptom a. 10 min after column pressure increased, column temperatures gocrazy

b. 10 min after column pressure increased, bottoms level decreasesc.d.e.


a b c d e A B C D

1. tray collapsed stripping section S S 4

2. too much bottoms fed to debutanizer N S 4

3. too much overheads in feed N S 4

4. feed valve FV1 stuck S S 4

5. pump F-26 not working S S 4

6. check valve on idle pump allows backflow S S 4

7.

Diagnostic actions:

A. readings of instruments on columnB. visit site and listen to pump for cavitationC. visit site and see location of valve stem on FV-1D. shut isolation valves on idle pump

4. PlanSelect “read instruments” as the first task because it is inexpensive and shouldhelp test many of the hypotheses. Many of the key variables are displayed in thecontrol room.

5. Do itGo to the control room, notebook in hand.

6. Look back

Comment: This example illustrates the approach one might take in being system-atic.


2.5Summary

Research on problem solving and on trouble shooting provide key informationabout the trouble-shooting process. Based on this research the four key features ofthe process are:

. the problem-solving process: monitoring, checking and double checking,being organized and systematic and keeping the problem in perspective.

. data handling and critical thinking: data gathering and resolution, based onfundamentals, with valid reasoning and being complete.

. synthesis: having five to seven working hypotheses, being flexible and put-ting it all together well.

. decision-making: based on criteria, priorities and avoiding bias.

A feedback form is given in Figure 2-3. The Trouble-Shooter’s Worksheet was cre-ated to aid the process. Its use was illustrated.

2.6Cases to Consider

For each of the five cases given in Section 1.6 complete the Trouble-Shooter’s Work-sheet 2-4.

40

Trouble-Shooter’s Worksheet 2-4:(� copyright 2003 Donald R. Woods and Thomas E. Marlin)


. Emergency priority: Safety? Hazard? Equipment damage?shut down &; safe park &. If not, & then:

. Draw a sketch of the process and mark on values. Provide a descriptionin words of what is going on.


. Monitor:Have you finished this stage? Can you check? What next?


2.6 Cases to Consider 41

IS IS NOT

WHAT __________________________________________

(should be happening but it is not)_________________________________

WHEN ?& _________________ ?& ______________________________WHERE ?& _________________ ?& ______________________________WHO ?& _________________ ?& ______________________________

– startup new process & suggest use Basics– startup after maintenance or change & suggest use Change– usual operation but changes made in operation but not in equipment& suggest use Basics



3. Explore: Gather information to be gathered ? & in Define stage &Exercise? & or a problem? &.Strategy: change & or basics &.Perspectives. Why? Why? Why?

____________________________________________________________Why? ›

____________________________________________________________Why? ›

____________________________________________________________Why? ›

____________________________________________________________Why? ›

____________________________________________________________Why? ›

Start fi _________________________________________________________

. Prioritize: product quality & ; production rate &; profit &

. Goal: safe-park? &: short term now with long term later &;long term now &

Action to be achieved: Specific terms and Measurable: __________________Attainable? ______________________________________________________Reliable? ________________________________________________________Timely? _________________________________________________________Safe? ___________________________________________________________$ _______________________________________________________________



. Likelihood of problem: startup new process &maybe mechanical electrical failureusual operation &: ambient temp? &maybe fluids problems;high temperature? & then maybe materials problemsSystem? failure of heat exchanger &> rotating equipment & >vessels & > towers &


What if? _______________________ then _____________________________What if? _______________________ then _____________________________What if? _______________________ then _____________________________

. List changes made & and/or trouble-shooting causes based on symp-tom &.

. Brainstorm root causes:


Symptom a.b.c.d.e.


a b c d e A B C D

1

2

3

4

5

6

7

Diagnostic actions:

A.B.C.D.

43

In Section 3.1 we consider the general rules of thumb for processes and differenttypes of problems, instruments and people and the environment. In Sections 3.2 to3.10 we consider different types of equipment: transportation, energy exchange,homogeneous phase separations, heterogeneous phase separations, reactions, mix-ing, size reduction, size enlargement and bins. Sections 3.11 and 3.12 consider “sys-tems” and hazards. Guidelines and trouble-shooting rules of thumb are not availablefor many pieces of equipment. Often, guidelines for good practice are given sincetrouble can often results from “poor practice”. The style used for presenting infor-mation about trouble shooting is as follows. The “symptom” is shown in italics inquotes. This is followed by the root causes, separated by a slash, / root cause/ rootcause. The causes are listed with the most-likely cause first, next-likely cause secondand so on. Some causes are not root causes. Such causes are shown as [cause]*.Those causes are listed in square brackets with an *, for example [corrosion]* mightbe listed as a cause. But what is the root cause of the corrosion? Such “cause–symp-toms” are listed separately with their root causes. For example, [Corrosion]*: inade-quate stress relief for metals/ wrong metals chosen/ liquid flows at velocities > criti-cal value.

3.1Overall

Consider general rules of thumb and typical causes, rules of thumb about corrosion,for instrumentation and for people, respectively.

3.1.1General Rules of Thumb and Typical Causes

Gans et al. (1983) suggests that big failures usually have simple causes, such as acompressor that will not start. On the other hand, small failures (or deviations fromthe norm) often are caused by complex causes, such as the product does not quitemeet specifications because of a buildup of contaminants.

3

Rules of Thumb for Trouble Shooting


3 Rules of Thumb for Trouble Shooting

The general most likely causes for failure differ depending upon whether this isthe

. startup of a new process or

. startup after a shutdown and maintenance or

. fault that develops for an on-going, operating process.

a) Common Faults for First Time Startup.

The faults encountered are:

75%Mechanical/electrical failures leaks, broken agitators, plugged lines,frozen lines, air leaks in seals.

20%Faulty design or poor fabrication unexpected corrosion, overloadedmotors, excessive pressure drop in heatexchangers, flooded towers.

5% Faulty/inadequate initial data often chosen to be the scapegoat byinexperienced trouble shooters.

b) Startup after Maintenance

Ask questions about what specifically was changed, repaired, or modified.

c) Trouble for On-going Processes

For ambient temperature operations, about 80% of the problems experienced arefluid dynamical.

For high-temperature operations, about 70% of the problems experienced arematerials failure.

Frequency of failures based on type of equipment:

17% heat exchangers.16% rotating equipment: pumps, compressors, mixers.14% vessels.12% towers.10% piping.8% tanks.8% reactors.7% furnaces.

Another approach is to consult data for Mean Time Between Failures, MTBF.Some example MTBF, in years, are: reciprocating compressors (nonlubricated), 2.8years; gas turbines, 3.9; centrifugal pumps, 4; screw compressors, 4; reciprocatingcompressors (lubricated), 4.1; motors, 11.4; large induction motors, 16. See, forexample, H.P. Bloch, “Looking for RCTA databases. Consider Failure statistics”,Hydrocarbon Processing, Jan 2002, p 36.

44

3.1 Overall

3.1.2Corrosion as a Cause

Corrosion could cause trouble. Here are some general ideas about corrosion andsome details about trouble shooting.

General ideas:

1. The strength of materials depends totally on the environment in which thematerials function and not on the handbook values.

2. All engineering solids are reactive chemicals – they corrode.3. The usual eight forms of material failure are: 1) uniform corrosion: uniform

deterioration of the material (32%); 2) stress corrosion: simultaneous pres-ence of stress and corrosive media (24%); 3) pitting: stagnant areas with highhalide concentration (16%); 4) intergranular corrosion: most often found instainless steels in heated areas (14%); 5) erosion: sensitive to high flowrates,local turbulence with particles or entrained gas bubbles. For flowing gas-sol-ids systems the rate of erosion increases linearly with velocity and dependson abrasiveness of particles (9%); 6) crevice corrosion: concentration cellsoccur in stagnant areas (2%); 7) selective leaching or dealloying: removal ofone species from a metallic alloy (1%) and 8) galvanic corrosion: dissimilarmetals coupled in the presence of a solution with electrolyte (negligible).

4. Stress corrosion (the second most significant form of corrosion) can startfrom perfectly smooth surfaces, in dilute environments in material withstresses well below the yield stress.

5. >70% of stress corrosion cracking is related to residual – not applied – stress-es.

6. The penetration of stress corrosion cracking as a function of time depends onthe alloy composition, structure, pH, environmental species present, stress,electrochemical potential and temperature.

Trouble shooting:

“High concentration of metals (Fe, Cr, Ni, Cu) in solution”: [corrosion]*/ contaminantsfrom upstream processing.

“Ultrasonic monitoring shows thin walls for pipes, internals or vessels”: faulty ultrason-ic instrument/ [corrosion]*/ faulty design. “Failure of supports, internals, vessels”: [cor-rosion]*/ faulty design/ unexpected stress or load.

“Leaks”: [corrosion]*/ faulty installation/ faulty gasket/ faulty alignment.[Cavitation in pumps]*: pump rpm too high/ suction resistance too high/ clogged

suction line/ suction pressure too low/ liquid flowrate higher than design/ entrainedgas.

[Corrosion]*: [corrosive environment]*/ inadequate stress relief for metals/ wrongmetals chosen/ liquid flows at velocities > critical velocity for the system; for aminecircuits: > 1 m/s for carbon steel and > 2.5 m/s for stainless steel/ large step changesin diameter of pipes/ short radii of curvature/ flange or gasket material projects intothe pipe/ [cavitation in pumps]*/ improper location of control valves.

45


[Corrosive environment]*: temperature too high; for amine solution: > 125 �C/ highdissolved oxygen content in liquid/ the liquid concentration differs from design; forsteam: trace amounts of condensate or condensate level in condensers > expected; for316 stainless steel: trace amount of sodium chloride; for sulfuric acid: trace amounts ofwater diluting concentrated acid; for amine absorption: total acid gas loadings > 0.35mols acid gas/mol MEA, > 0.40 mols acid gas/mol DEA, > 0.45 mols acid gas/molMDEA; makeup water exceeds specifications; for amine absorption units exceeds:100 ppm TDS, 50 ppm total hardness as calcium ion, 2 ppm chloride, 3 ppmsodium, 3 ppm potassium and 10 ppm dissolved iron; for sour-water scrubbers: cya-nides present/ pH change/acid carryover from upstream units/ high concentrationof halide or electrolyte/ presence of heat stable salts/ bubbles present/ particulatespresent/ invert soluble precipitates with resulting underlying corrosion/ sequenceof alternating oxidation-reduction conditions.

3.1.3Instruments, Valves and Controllers

Trouble-shooting sensors: Most sensor faults are because of improper selection,incorrect installation or adverse environmental conditions. “Wrong signal”: fouled orabraded sensors/ bubbles or solid in fluid/ sensing lines plugged or dry/ electricalinterference or grounding/ sensor deformed/ process fluid flow <design or laminarinstead of turbulent flow/ contamination via leaky gaskets or O-rings/ wrong materi-als of construction/ unwanted moisture interferes with measurement or signal/high connection or wiring resistance/ nozzle flappers plugged or fouled/ incorrectcalibration/ orifice plate in backwards/ orifice plate designed incorrectly/ sensor bro-ken/ sensor location faulty/ sensor corroded/ plugged instrument taps: for sour-water strippers: water or steam purge of taps malfunctioning or local tempera-tures < 82 �C at which ammonium polysulfides form. “Wrong input”: sensor at wronglocation/ insufficient upstream straight pipe for velocity measurement/ feedbacklinkages shift or have excessive play/ variations in pressure, temperature or compo-sition of the process fluid. “Fluctuating signal”: bubbles in the liquid/ flashingbecause Dp across an orifice plate > design or fluid too close to the boiling pointcausing cavitation.

Trouble-shooting control valves: The signal to the valve should be midrange;otherwise the signal depends whether the valve is fail open or fail close. “Leaks”: ero-sion/ corrosion/ gaskets, packing or bolts at temperatures, pressures and fluids thatdiffer from design. “Can’t control low flowrate”: miscalibration/ buildup of rust, scale,dirt/[ faulty design]*. “Can’t stop flow”: miscalibration/ damaged seat or plug. “Exces-sive flow”: excessive Dp. “Slow response”: restricted air to actuator/ dirty air filters.“Noise”: cavitation/ compressible flow. “Poor valve action:” dirt in instrument air/sticky valve stem/ packing gland too tight/ faulty valve positioner. “Cycling”: stiction.[Faulty design]*: valve stem at design flowrate is not at midrange. [Stiction]* is thesticking and friction related to valve movement and measured as the difference be-tween the driving values needed to overcome static friction upscale and downscale.Likely cause of small amplitude, continuous cycling: gland too tight/ insufficient

46

3.1 Overall

driving force/antistiction coating damaged/ faulty valve positioner/ poorly tunedcontrol system/ incorrect valve/ incorrect actuator.

Trouble-shooting transmitters: “Erratic or fluctuating output”: vibrations/ improperorientation/ loose connections.

Trouble-shooting block valves: Check that the arrow on the valve is in the samedirection as the flow through the valve. Test via “turn and seal” to check movementand closure. “Reduced flow”: valve not fully open/ plugged with dirt. “Poor control bythe control system”: block valve on bypass partially open/ block valves upstream ordownstream of the control valve partially closed.

Trouble-shooting check valves: “Noisy”: backpressure too high. “Reduced flow”:backpressure too high.

Trouble-shooting control systems: “Oscillation”: feedforward/ poorly tuned/ valvesticks or has excessive hysteresis. “Returns with offset”: proportional control only.“Two related variables start to deviate”: lack of ratio controllers/ failure to relate analy-ses to flows.

3.1.4Rules of Thumb for People

Nine suggestions are given.

1. Become aware of your own uniqueness and personal style, and how youmight differ from the style of others.

2. Honor the seven fundamental rights of individuals, RIGHTS. R, to beRespected; I, Inform or to have an opinion and express it; G, have Goals andneeds; H, have feelings and express them; T, trouble and make mistakes andbe forgiven; S, select your response to others expectations and claim theserights and honor these in others.

3. Avoid the four behaviors that destroy relationships: Contempt, Criticism,Defensiveness and Withdrawal/ stonewalling.

4. Trust is the glue that holds relationships together. Three elements of trust arebenevolence, integrity and competence.

5. Build trust by benevolence through loyalty to others, especially when they arenot present and by not doing anything that would embarrass or hurt them.Build trust by integrity by keeping commitments to yourself-and others; clari-fying expectations that you have of yourself-and of others; showing personalintegrity, and honesty; apologizing promptly and sincerely when you knowyou are wrong; honoring the fundamental RIGHTS listed above and avoidingthe killers; listening and understanding another’s perspective; being truthful;and accepting others “warts and all”. Build trust by being competent in yourarea of expertise.

6. Destroy trust by the reverse of the Builders of trust listed above, and by selec-tively listening, reading and using material out of context; not accepting theexperience of others as being valid; making changes that affect others with-out consultation; blind-siding by playing the broken record until you’ve even-

47


tually worn them out or subtly make changes in the context/issues/wordinggradually so that they are unaware of what is happening until it is too late.

7. The 12:1 rule applies to rebuild relationships. 12 positive experiences areneeded to overcome 1 negative experience.

8. To improve and grow we need feedback about performance. Give feedback toothers to encourage and help them; not for you to get your kicks and putthem down. Focus on five strengths for every two areas to improve on.

9. Be skilled at responding assertively. “When you... I feel .. adjust by...” .

3.1.5Trouble-Shooting Teams

Use Worksheet 3-1 after each meeting, set goals and celebrate achievement. Use theframework developed by Francis and Young (1979) for growth; consult Fisher et al.(1995) for more short-term ideas.

Trouble-shooting team meetings

These are organized by symptom with possible corrective responses suggested forchair {C} and member {M}.

Problems with Purpose and Chairperson

“No apparent purpose for the meeting”: {C} don’t have a meeting. {M} question thepurpose of the meeting. See also Agenda and timing problems.

Agenda and Timing Problems

“No agenda”: {M}: phone {C} and ask for agenda. Invoke “no agenda, not atten-dance.”/at meeting: “Perhaps the first thing we should do is to create an agenda.” /After 5 minutes, “We seem to be lost. Could we draw up an agenda and followthat?”

“Meeting drags on and on”: {C} should have circulated an agenda with times foreach item and used the 20 minute rule/ {M} “Perhaps we can follow the agenda.”/{M} indicate to {C} ahead of time the amount of time you have available for themeeting and then leave at that time.

“Get off the track”: {M} seek direction, purpose, summary of progress. see also Be-havior problems: “Subgroups interrupting and talking”.

“Group gets bogged down”: state problem/ summarize/seek agenda clarification/invoke 20 min rule.

“Decisions made just at the end of the meeting”: state frustration/ suggest tabling/suggest future corrective way to handle in future. See also Agenda and chairpersonproblems.

Behavior and Participation Problems

“People come into a meeting cold:” {C or M} suggest reconvene meeting when all areprepared.

48

3.1 Overall

“Late arrivals”: {C} start meeting on time and continue with the agenda throughthe disruption of the retardee/ {M} “I realize that not everyone is here but I suggestthat we start. It looks like a long agenda to get through.”

“Some people do all the talking and some remain silent”: wrong membership/ encou-rage quiet ones to contribute/ ask each, in turn, to summarize his/her point ofview/ ask a “safe” question of the silent ones/ privately check with the silent onesand reevaluate whether they need to attend/ ask open ended questions/ use nominalgroup.

“Sub groups interrupting and talking”: identify problem/ suggest discussing oneissue at a time and add subgroup’s issues to agenda/ be silent until the side conver-sation stops. “Thank you.” / Interrupt the side conversation.

“Indecisive members, continual question asker”: ask for their ideas early/ redirectquestions he/she asks back to him/her.

Conflict or Apparent Conflict

“Conflict because of differing views:” restate the importance and value of everyone’sopinions, restate the RIGHTS/ attempt to bring conflict into the open/ summarizedifferent views/ focus on different performance or opinions and not personalities/remind of fundamental RIGHTS.

“Conflict over facts:” stop the argument, identify problem as you see it and checkthat that is a problem/ identify facts we need clarified and probable expert.

“Conflict over values, goals, criteria, process or norms:” stop discussion, identify prob-lem as you see it and check that that is a problem/ use problem solving.

“Resistance to new ideas, we tried that before, it won’t work, over my dead body, wedon’t have the resources”: surface the resistance/ honor the resistance/ invoke conse-quence of no decision or of repeating what we’ve always done before/ use consensusbuilding techniques/ reflect on the home turf of the objector and the impact thedecision might have on them; explore if this might be brought to the group as anissue to address/ root cause of most resistance is fear of change, apathy, vested inter-ests, not invented here, negativism, overwhelmed by enormity of proposal.

49

3 Rules of Thumb for Trouble Shooting50

Worksheet 3-1: Rating form for teams

Assessment of your team and team meeting Name: _________________Date: __________________

Purpose of team: ______________________________________ unclear &Purpose of this meeting ________________________________ unclear &

Agenda for this meeting: detailed, clear and circulated ahead of time &bare minimum circulated ahead of time &none &

Three-minute team task to seek consensus about the rating of the Task andMorale:

. Teamwork: Task all members clear about and committed to goals; allassume roles willingly; all influence the decisions; know when to dis-band for individual activity; all provide their unique skills; share informa-tion openly; the team is open in seeking input; frank; reflection andbuilding on each other’s information; team believe they can do theimpossible; all are seen as pulling their fair share of the load.

The degree to which these descriptors describe your team’s performance (assubstantiated by evidence: meetings, engineering journal, interim report, pre-sentations).

None of Few of these Most features All ofthese behaviors but demonstrated thesebehaviors major omissions behaviors

& & & & & & &1 2 3 4 5 6 7

. Teamwork: Morale: Trust high, written communication about any indi-vidual difficulties in meeting commitments; cohesive group; pride inmembership; high esprit de corps; team welcomes conflict and uses meth-odology to resolve conflicts and disagreements; able to flexibly relievetension; sense of pride; we attitude; mutual respect for the seven funda-mental rights of all team members; Absence of contempt, criticism,defensiveness and withdrawal.


3.2 Transportation Problems 51


& & & & & & &1 2 3 4 5 6 7

Each, in turn, gives a 30-second summary of his/her perception of his/hercontribution. This is presented without discussion.

Individual, 30 second reporting of his/her contribution to this meeting:____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Four-minute team task to reach consensus about the five strengths and the twoareas for growth.

Strengths of your team Areas to work on for growth___________________________ ______________________________________________________________ ____________________________________________________________________________________________________________________

D.R. Woods (2005)

This form should be completed after each meeting and copies used as evidence ofgrowth.

3.2Transportation Problems

Fundamentals of why fluids move: Fluids move from high pressure to low pressure,vertically because of gravity force, dragged along by a moving boundary or belt orbecause of density differences; won’t flow out of a sealed vessel or vacuum unlessthere is a vent break. These are expressed, on the macroscopic level, as Bernoulli’sequation. Most trouble-shooting problems encountered are fluid-dynamical prob-lems. Centrifugal pumps are often selected to pump liquids; such pumps operateon their head-capacity curve showing decreasing head with increasing capacity. Forpumping liquids, cavitation usually occurs when pumping hot liquids near theirboiling temperature or when sucking liquids out of a sump. Whenever plantsstartup for the first time or after a shutdown, wood, sandwiches, bolts and other


crud could have been left unintentionally in the lines. Symptoms and possiblecauses for specific pieces of equipment are presented as follows: gas-moving equip-ment for pressure, Section 3.2.1 and for vacuum service, Section 3.2.2. Pumpingliquids are considered in Section 3.2.3; pumping solids, Section 3.2.4. Considera-tions for steam are given in Section 3.2.5.

3.2.1Gas Moving: Pressure Service

Fans, blowers and centrifugal and reciprocating compressors.

Fans: Trouble shooting: “Noise”: vortex, flow separation/ loose bearings. “Dis-charge pressure low”: instrument error/ fans in series rotating in the same direction/operating below the stall point/ density increase. “Low flowrate”: instrument error/flow separation/ pitch angle of blades too shallow/ speed slow/ required system dis-charge high.

Blowers: For rotary lobe: when used for pressure pneumatic conveying install acheck valve in the blower discharge. Trouble shooting: “Discharge pressure high”:instrument error/ restriction in downstream line/ check valve jammed in closedposition/ dirty intake filter. “Discharge pressure low”: instrument error/ slippage ofthe drive belts/ relief valve stuck open/ increasing air loss at the rotary valve due tolarger clearance opening from wear/ loss of air caused by larger lobe clearance inthe blower due to wear/ a leak, such as a ruptured hose, in a vacuum system/ a rup-tured bag in the downstream bag house.

Centrifugal compressors: Good practice: allow safety margins of design speed5%, design head 10% and design power 15%. The sonic velocity decreases with anincrease in gas molar mass. Trouble shooting: “Surging”: insufficient flow/ increaseddischarge pressure required by the system/ deposit buildup in diffuser. “Dischargepressure low”: instrument error/ compressor not up to speed/ excessive inlet temper-ature/ leak in discharge system. Provide separate anti-surge system for compressorsoperating in parallel; need careful design of suction piping for double flow compres-sors.

Reciprocating-piston compressors: Good practice: design velocity throughvalves < 40 m/s. Trouble shooting: major faults: valves and piston rings. “Knocking”:frame lubrication inadequate/ head clearance too small/ crosshead clearance toohigh; “Vibration”: pipe support inadequate/ loose flywheel or pulley/ valve LPunloading system defective. “Discharge pressure high”: instrument error/ valve LPunloading system defective / required system discharge high. “Discharge pressurelow”: instrument error/valve LP unloading system defective/ LP valve worn/ systemleakage. “Discharge temperature high”: instrument error/ LP valve worn/ valve LPunloading system defective/required system discharge pressure high. “Cooling-watertemperature high”: instrument error/ water flowrate low/fouled area/LP valve worn.“Valve temperature high”: instrument error/required system discharge pressure high/run unloaded too long/LP valve worn. “Cylinder temperature high”: instrument error/required system discharge pressure high/LP valve worn/ wrong speed. “Flow low”:

52

3.2 Transportation Problems

instrument error/LP valve worn/valve LP unloading system defective/dirty suctionfilter.

3.2.2Gas Moving: Vacuum Service

Liquid piston pump, dry vacuum pump and steam ejectors.

Liquid piston pump: Good practice: cool the seals with 0.03 L/s of clean coolingwater at pressure at least 35 kPa greater than discharge pressure of pump. Troubleshooting: “Noisy”: service liquid level too high/coupling misaligned. “Capacity low”:suction leakage/ service liquid temperature too high/ speed too low/ seal water flow-rate < design. “Power excessive”: service liquid level too high/ coupling misaligned.“Service liquid temperature high”: clogged strainer/ partially closed valve/ fouled heatexchanger.

Dry vacuum pump: Good practice: size for usual discharge pressure 20–35 kPagauge to allow for downstream discharge. Vacuum pumps run hot: 50–70 �C. Allow30-min warmup period before putting on-line. Allow 60 min purge before shut-down. Try not to have the pump discharge into a common header. Multistagepumps tend to run cooler than single stage. Install a check valve on the discharge. Ifthe discharge pressure is > 35 kPa, add a positive displacement blower (designed for6 Ndm3/ s at design conditions for the vacuum pump) with a bypass that is open forstartup.

Trouble shooting: “Loss of vacuum”: condensation in the suction line/condensationof species from other units connected to a common exhaust header/ increase in dis-charge pressure from restriction in downstream processing or pressure blowout inother units connected via common discharge header. “Excessive corrosion”: for sys-tems handling acid gas or connected to such systems via common discharge header:warmup period too short/ shutdown purge too short. “Overheating”: low cooling-water flow/ fouled cooling system/ inlet gas temperature > 70 �C. “High amps.”:buildup of polymer caused by operating temperature too high/ polymerizable spe-cies gain access via common discharge header.

Steam ejectors: Good practice: operability of steam ejectors is very sensitive to thestability in the motive fluid (steam) pressure. Prefer vacuum pumps to steam ejec-tors. Keep diameter of pipes= diameter of inlet and discharge flanges of ejectors. Fordistillation columns, as the column overhead mass flowrate increases above design,so will the column overhead pressure and vice versa. Compression ratios per ejector:6:1 to 15:1. If the inlet gas temperature < 0 �C or below the triple point of water (0.61Pa) then add steam jacketing to cope with ice formation. Seal for the hot well: sub-merge > 30 cm. The volume in the hotwell between the pipe and the overflow weirshould be 1.5 times the volume in the down spout sealed. Replace any nozzles ordiffusers where the area is >7% larger than design.

Trouble shooting: check the last stage first and then move upstream. “Unstableoperation or loss of vacuum”: steam pressure < 95% or > 120% of design/ steam super-heated > 25 �C/ wet steam/ inlet cooling-water temperature hot/ cooling-water flow-

53


rate low/ condenser flooded/ heat-exchange surface fouled/ 20–30% higher flow ofnon-condensibles (light end gases, air leaks or leaks from fired furnaces) / seal loston barometric condenser/ entrained air in condenser water/ required dischargepressure requirement high/ fluctuating water pressure. “Water coming out of dis-charge”: upstream condenser flooded.

3.2.3Liquid

The types of pumps include centrifugal, peripheral, reciprocating, rotary, gear androtary screw.

Centrifugal: Good practice: head-capacity curve should not be too flat if pump ca-pacity is controlled by valve positioner. Select pump such that a larger diameterimpeller could be installed later. An increase in flowrate causes an increase in re-quired NPSH and a decrease in available NPSH.

Trouble shooting: “No liquid delivery”: instrument error/not primed/ [cavitation]*/supply tank empty. “Liquid flowrate low”: instrument error/ [cavitation]*/ non-con-densibles in liquid/ inlet strainer clogged. “Intermittent operation”: [cavitation]*/ notprimed/ non-condensibles in liquid. “Discharge pressure low”: instrument error/ non-condensibles in liquid/ speed too low/ wrong direction of rotation (or impeller inbackwards if double suction). “Power demand excessive”: speed too high/ density liq-uid high/ required system head lower than expected/ viscosity high.

Peripheral: Trouble shooting: “No liquid delivery”: instrument error/ pump suctionproblems/ suction valve closed/ impeller plugged. “Liquid flowrate low”: instrumenterror/ speed too low/ incorrect impeller trim/ loose impeller. “Discharge pressurelow”: instrument error/ speed too low/ incorrect impeller trim/ loose impeller.“Power demand excessive”: speed too high/ improper impeller adjustment/ impellertrim error.

Reciprocating: Trouble shooting: “No liquid delivery”: instrument error/excessivesuction lift / [cavitation]*/ non-condensibles in liquid. “Liquid flowrate low”: instru-ment error/ excessive suction lift/ [cavitation]*/ non-condensibles in liquid.

Rotary: sometimes NPSH is expressed as Net Inlet Pressure Required, NIPR, (oravailable NIPA), expressed as kPa absolute (kPa abs). Trouble shooting: “No flow”:instrument error/ [pump not turning]*/ [pump not primed]*/ relief valve notadjusted correctly or dirt keeping the relief valve open/ wrong direction of rotation/[cavitation]*/ excessive suction lift. “Flow < design”: instrument error/ rpm too low/air leak via bad seals or faulty pipe connections/ [ flow going elsewhere]*/ [highslip]*/ suction line clogged/ insufficient liquid supply/ [air or gas in liquid]*. “Startsbut loses prime”: air leakage/ liquid vaporizing in suction line/ insufficient liquid sup-ply. “Noisy operation”: [cavitation]*/ [air or gas in liquid]*/ [mechanical noise relatedto pump]*/ relief valve chatter/ drive-component noise. “Power > design”: higher vis-cous losses than expected/ pressure > design/ fluid viscosity > expected/ fluid “setsup” or solidifies in the line or pump during shut down/ fluid builds up on pump

54


surfaces/ rotating elements bind. “Short pump service life”: [corrosion]*/ abrasivespresent/ speed and pressures > design/ lack of lubrication/ misalignment.

[Air or gas in liquid]*: [ fluid vaporizes]*/ air bleed missing/ fluid gasifies underoperating conditions/ leaks in pumps or piping.

[Air lock]*: [ fluid vaporizes]*/ air bleed missing/ fluid gasifies under operatingconditions.

[Cavitation]*: [ fluid vaporizes]*[Flow goes elsewhere]*: relief valve faulty or jammed open/ discharge flow diverted

to wrong branch line.[Fluid vaporizes]*: [NPSH supplied too small]*/ fluid viscosity > design/ fluid tem-

perature > design/ vapor pressure of fluid too high.[High slip]*: clearance between rotors > specs/ worn pump/ pressure > design.[NPSH supplied or NIPA supplied too small]*: strainer clogged/ temperature too

high/ inlet line clogged/ inlet line diameter too small or length too long/ atmospher-ic pressure < design.

[Mechanical noise related to pump]*: wrong assembly/ pump distortion because ofwrong piping installation/ pressure > rating/ worn bearings/ worn gears/ loosegears/ twisted shaft/ sheared keys/ worn splines.

[Pump not primed]*: valve on inlet line closed/ inlet line clogged/ air leaks/ pumprpm too low/ liquid drains or siphons out during off-periods/ check-valve missingor faulty/ [air lock]*/ worn rotors.

[Pump not turning]*: drive motor stopped/ key sheared or missing/ belt drive bro-ken/ pump shaft broken.

Gear: Good practice: the higher the viscosity, the lower the rated rpm. On the dis-charge install a check valve and an expansion chamber or pulsation dampener onthe discharge, the latter to reduce noise. For infrequent operation, operating pres-sure should be 20–30%< rated pressure. For continual operation, operating pressure<< rated pressure and rpm< rated rpm. Never allow them to run dry. For startup,idle off-line for about an hour.

Trouble shooting: usually performance does not break down suddenly; insteadthere is a gradual decrease in performance. Gear pumps are particularly susceptibleto cavitation and erosion. “Low discharge pressure”: instrument error/ leakage/ lowdrive power/ faulty relief-valve setting/ [internal leakage]*/ [abrasion]*. “No liquiddelivery”: instrument error/ suction line clogged/ drive not turning shaft/ check-valvefault. “Low liquid delivery”: instrument error/ drive power low/ [internal leakage]*/[abrasion]*/ [cavitation]*. “Noisy”: entrained air in liquid/ liquid doesn’t drain fromgrooves/ misaligned drive and pump shafts/ faulty bearings/ loose mountings/ res-onance because mating frequency of gears= natural frequency of gear train / rpmtoo high/ worn parts/ [cavitation]*. “Overheating”: liquid viscosity higher thanexpected/ liquid feed temperature too low/ faults in drive system such as misaligneddrive and pump shafts. “Shaft won’t rotate”: drive system not working/ material inpump not melted/ temperature too low/ seized pump. “Significant oscillation inpump suction pressure”: instrument wrong/ faulty control system/ suction pressuresetpoint too low for the process. “Pump discharge pressure oscillates”: instrumentwrong/ starved feed to pump/ change in viscosity of feed/ damaged pump internals.

55


“Inadequate volumetric efficiency”: decrease in viscosity/ pressure in upstream processincrease/ worn pump.

[Abrasion]*: grit in liquid/[cavitation]*/ pH different from design. [Cavitation]*:pump rpm too high/ suction resistance too high/ clogged suction line/ suction pres-sure too low/ liquid flowrate higher than design/ entrained gas. [Internal leakage]*:excessive clearance between the gear and sides or end plates/ [abrasion]*.

Rotary screw (mono): Good practice: stator has one more lobe or screw than therotor; Reduce rpm for abrasives; starting torque= 4 � initial torque. If particles arepresent, try to minimize the abrasion by using viscosity > 5000 mPa s. Backflow or“slip” is reduced as viscosity increases. NPSH problems are usually not importantexcept for suction lift, pumping from a vacuum and fluid vapor pressures of > 15kPa. Trouble shooting: “No liquid delivery”: instrument error/ wrong direction of rota-tion/ insufficient suction lift/ clogged inlet/air leaks on suction/faulty pressurerelief valve/ worn pump. “Rapid wear”: discharge pressure too high/ pump runsdry/ incorrect materials of construction/ speed too high for abrasives/ viscosity toolow for abrasives. “Noisy”: insufficient feed flowrate/ air leak in suction/gas in feedliquid/ speed too high/ poor alignment. “Excessive power”: rpm too high/ liquid vis-cosity > design/ operating pressure > design/discharge line plugged/ stator expandedor swollen. “Failure of the stator”: bond failure (pH> 10 or local hot spot)/ tempera-ture > design. “Initially OK but gradual increase in power needed”: swelling of elasto-meric stator coating because of chemical attack.

3.2.4Solids

Johanson’s1), 2) definitions of terms used to characterize solid particles are given inSection 3.7.3. The important terms are AI, RI, HI, FRI, FDI, BDI, CI, RAS and SBI.

Bucket elevators: Trouble shooting: major difficulties are unloading and loading:jamming of materials between the buckets and the side of the boot.

Rotary/star valve: Trouble shooting: use amps as guide to solids throughput. Keepair velocity high enough to prevent plugging of the air-vent line.

Pneumatic conveying: dilute phase: for vacuum: Trouble shooting: air leakageand powder arching in hopper are the major threats. “No flow or flow < design”: airleaks/ powder arching in feed hopper, see Section 3.10/ low solids flow because ofincreased air loss in rotary valve/ wrong type of rotary valve used/ insufficient air/line too long/ vacuum pump problems, see Section 3.2.2 . “Pressure (vacuum) at suc-tion to blower > design (vacuum< design)”: air leaks/ failure of discharge valve to sealon the receiver. “Erratic pressure readings”: irregular feed. “Explosion”: moisture toolow/ lines not grounded. “Does not sound “tinny” when listening with stethoscope”:material accumulated inside pipe at this location.

56

1) J.R. Johanson, 2002, “Troubleshooting bins,hoppers and feeders,” Chem. Eng. Prog. Aprilpp. 24–36.

2) J.R. Johanson, 2000, “Smooth out solidsblending problems,” Chem. Eng. Prog. April,p. 21.


Pneumatic conveying: dilute phase: for pressure: Trouble shooting: use pressureat the outlet of the blower as prime indicator. “Dp across blower > design or 2:1 ratio”:restriction in downstream conveying line/ check valve jammed closed/ dirty intakefilter/ plugged discharge silencer/ increase in feed to the system/ length of pipe >design. “Dp across blower < design”: slipping v-belts/ air loss at the rotary valve. “Noflow”: [plugged line]* “No flow or flow < design”: overfed fan system/ insufficient air/insufficient solids/ line too long/ inlet air pressure too low. “Erratic pressure read-ings”: irregular feed. “Amps on rotary valve < usual”: solids flow< design/ air lossthrough the rotary valve/ increased clearances. “Does not sound “tinny” when listeningwith stethoscope”: material accumulated inside pipe at this location. “Gradual decreasein performance”: wear on the blower caused by dusty air.

[Plugged line]*: within the first couple of metres of the beginning of the system:material feed problems/air supply problems. [Plugged line]* after the first couple ofmetres: air leak with the plug occurring about 10 m downstream of leak/ erosion ofrotary valve causing increase in air leakage.

Pneumatic conveying: dense phase: Trouble shooting: “No flow or flow < design”:plugged line/ malfunction of line boosters because of stuck check valve/ highhumidity. “Solids fed to conveying line < design”: ratio of air to fluidize in the blow tankrelative to convey is too small/ fault in control system. “Solids fed to conveying line >design”: ratio of air to fluidize in the blow tank relative to convey is too large/ fault incontrol system.”Solids flow= 0”: top discharge and the ratio of air to fluidize to conveyis too small. “Solids flow gradually decreases”: restriction in the discharge pipe/ blind-ing of the fluidizing membrane.

Feeder: volumetric for extruder: Trouble shooting: “Does not run”: no power/jammed. “Stalls”: material jam/ current limit set too low. “Erratic speed control”: con-troller poorly tuned/ sensor malfunction/ material jam. “Feed rate variable”: particlesarching in the hopper/ moisture level too high/ overheated polymer (prematurelyfused) feed polymer.

Feeder: screw conveyor: Trouble shooting: “Shear pins on feeder drive break”: screwdiameter < exit hole from bin. “Motor overload on feeder drive”: screw conveyor diame-ter < exit hole from hopper. “Screw feeder initially OK then motor overloads”: screwflight spacing in the direction of sold flow decreases markedly/ difference betweenFDI and BDI < 5% suggests a moderately incompressible solid whose flow is verysensitive to screw flight spacing.

Feeder from bottom of hopper: Trouble shooting: “Feeder motor overloads immedi-ately:” wrong wiring/foreign material in feeder/ hopper is full and solids give exces-sive solids pressure because of particle characterization and hopper design / FDIlarge and large HI. “Feeder exit flowrate suddenly < expected”: blockage in hopper out-let/ lumps of particles forming in hopper/ large RI and small HI possibly caused bytemperature cycles. “Feeder exit flowrate gradually < design:” solids builup in the fee-der/ large CI, large AI and RI/ wrong materials of construction in feeder. (Oftenhappens with vibrating feeder.)

Feeder: belt feeder from the bottom of a hopper: Trouble shooting: “Belt feeder ini-tially starts but suddenly stops with motor overload”: gap between the belt and hopper

57


interface edge is too small/ belt sags between pulleys/ large FDI and small% differ-ence between FDI and BDI.

3.2.5Steam

Good practice: Take steam off the top of the steam header; put condensate into thetop of the condensate return header.

3.3Energy Exchange

The fundamentals for thermal energy exchange are that heat flows from a high tem-perature to a low temperature. Thermal forms of energy are not always available todo work. Overall energy is conserved; often we write expressions for the mechanicalenergy balance (on the macroscopic level this is Bernoulli’s equation) and the ther-mal energy balance (on the macroscopic level this is q=UA LMTD). When troubleshooting heat exchangers, usually the fault is fluid dynamical: liquids don’t drain;baffles are placed so that liquid can’t go where you expect; vents are missing thatprevent us from bleeding off trapped gases. Fluids to watch are water and hydrogen;both have extremes in thermal properties. Thermal expansion will occur whenexchangers are brought up to temperature. This may cause a leak at the head-to-tube sheet joint if the difference between the temperature on the tubeside less thetemperature of the bolts > 50 �C. For systems involving steam, scrutinize the steamtrapping system: ensure that traps are not flooded, that the appropriate trap hasbeen installed, that the bypass is not left open, and that thermodynamic traps arenot fed to a common header. Steam should come from a nozzle on the top of thesteam main; condensate should be discharged into the top of the condensate header.

More specifically, we list symptoms and possible causes for the following equip-ment: In this chapter we consider first providing mechanical drives, in Section3.3.1. Furnaces are considered in Section 3.3.2. Heat exchangers, condensers andreboilers are listed in Section 3.3.3. Sections 3.3.4, 3.3.5 and 3.3.6 consider refrigera-tion, steam generation and high-temperature heat-transfer fluids, respectively.

3.3.1Drives

Engines: Good practice: use high efficiency motors when replacing or repairingexisting installations. Trouble shooting: “Hammering/knocking”: loose parts/ seizedparts. “Pre-ignition”: fuel with unstable hydrocarbons/ incorrect timing. “Detonation”:wet fuel/ incorrect timing/ intake air too hot/ glowing carbon on the piston/ leakingvalve stem/ worn valve guides. “Misfiring”: incorrect timing/ faulty ignition ele-ments/ wrong gap in the spark plugs/ wet fuel/ spark-plug gap coated or filled withcarbon or oil. “Overheat”: lubrication failure/ inadequate cooling/ poor quality fuel/

58

3.3 Energy Exchange

fuel to air ratio too lean. “Sooty exhaust”: incorrect fuel/ air- fuel ratio too rich/ inade-quate cooling/ wrong valve adjustment. “Valve leaking”: inadequate cooling/ valveangle incorrect/ wrong metallurgy. “Piston blow-by:” over lubrication. “Worn bear-ings”: misaligned crankshaft.

Electric motor: Trouble shooting: “Won’t start”: overload trip/ loose connection/grounded winding/ grounded stator. “Runs backwards”: reversed phase sequence.“Excessive noise”: 3 phase machine single phased/ unbalanced load between phases.“Synchronous motor fails to come up to speed”: faulty power supply or overload trip/windings grounded. “Overheat”: unbalanced load between phases/ wrong line volt-age/ short circuit in stator winding. DC Motors: “Won’t start”: weak field/ low arma-ture voltage/ open or short circuit in armature or field. “Runs too slow”: low armaturevoltage/ overload/ brushes ahead of neutral. “Runs too fast”: high armature voltage/weak field/ brushes behind neutral. “Brushes sparking”: brushes worn/ brushespoorly seated/ incorrect brush pressure/ dirty, rough or eccentric commutator/brushes off neutral/ short-circuited commutator/ overload/ excessive vibration.“Brush chatter”: incorrect brush pressure/ high mica/ incorrect brush size. “Bearingshot”: belt too tight/ misalignment/ shaft bent/ damaged bearings/ wrong type ofbearings.

Steam turbine: Good practice: consider extracting energy via a steam turbine forany pressure reduction in steam service. Use high-pressure steam for energy; low-pressure steam for heating. Don’t operate with wet steam. Trouble shooting: “Tur-bine fails to start:” too many hand valves closed/ nozzles plugged or eroded/ dirtunder carbon rings. “Slow startup”: throttle-valve travel restricted/ steam strainerplugged/ load > rating. “Insufficient power”: throttle-valve travel restricted/ too manyhand valves closed/ oil relay governor set too low. “Speed increases as load decreases”:throttle-valve travel restricted/ throttle assembly friction/ valve packing friction.“Governor not operating/ excessive speed variation”: governor droop adjustmentneeded/ governor lubrication problem/ throttle-valve travel restricted. “Overspeed tripon load changes”: trip valve set too close to operating speed/ throttle-valve travelrestricted/ throttle assembly friction. “Overspeed trip on normal speed”: excessivevibration/ dirty trip valve/ trip valve set too close to operating speed. “Leaking glands”:dirt under carbon rings/ worm or broken carbon rings/ scored shaft.

Steam turbine used for the generation of electricity: Trouble shooting: “Turbineoverspeeding”: [load disconnection suddenly]*/ [Trip Throttle Valve stuck]*/ controlvalve fault// [extraction valve fault]*. “Bearings damaged”: [turbine overspeeding]*/[lube oil]*/ excessive vibration/ no lube oil/ bearing temperature too high/ flow ofparasitic currents/ [clogging]*/ [electronic pin clogging]*.

[Clogging]*: [lube oil]*/ long time without operating.[Electronic pin clogging]*: [lube oil]*/ long time without operating.[Extraction valve fault]*: wear on valve bearing/ loss of hermetic seal.[Load disconnection suddenly]*: operator error/ automatic bus bar protection

because of downstream changes in electric system.[Lube oil]*: low pressure/ oil temperature too high/ oil too old/ oxidation/ water

contaminates oil.

59


[Solenoid valve malfunction]*: [electronic pin clogging]*/ [clogging]*/ solenoidshorted coil/ faulty control signal/ sensor error.

[Trip Throttle Valve stuck]*: [clogging]*/ [solenoid valve malfunction]*Gas turbine: consists of a compressor, combustor and turbine sections. Trouble

shooting: “Combustion noise:” fouled or clogged combustor/ loose or cracked liningin combustor. “Vibration”: bearing failure in compressor or turbine/ blade damagein compressor or turbine/ surging compressor/ fouled turbine. “Exhaust tempera-ture > design”: combustor fouling. “Exhaust temperature < design”: combustor clogged.“Thermal efficiency < design”: fouled turbine/ turbine blade damage/ turbine nozzledistortion. “Mass flow < design”: compressor fouling/ compressor filter clogged/ com-pressor blades damaged.

3.3.2Thermal Energy: Furnaces

Multi-use including heating, boiling, reactions. Related topics distillation, Section3.4.2.

For fired furnaces: monitor CO and excess air to reduce rejected energy andimprove efficiency, consider the installation of economizers and air preheaters torecover additional heat from the flue gas.

For steam generation: preheat boiler feed water with available low-temperatureprocess streams, maximize the use of heat-transfer surfaces by optimizing soot-blowing frequency and decoking of tubes, flash blowdown to produce low-pressuresteam if required.

Trouble shooting: “Gas temperature > design”: instrument wrong/ insufficientexcess air/ process side coking of tubes/ leak of combustible material from processside/ overfiring because of high fuel-gas pressure. “Gas temperature < design”: instru-ment fault/ fouling/ too much excess air/ insufficient area/ fuel-gas pressure < de-sign. For convection furnace: “Exit process gas temperature < design”: excess air/decrease in flame temperature/ damper has failed closed. “Pressure inside furnace >design”: instrument wrong/ fouling on the outside of the tubes in the convection sec-tion/ exhaust fan failure. “Faint blue-gray smoke rising from top of furnace”: foulingoutside tubes in the convection section/ pressure in furnace > atmospheric. “Puffing,rhythmic explosions”: burners short of air for short period causing minor over-firing/wind action/ start up too fast. “Tube failure”: localized overheating/ burning acidgases as fuel/ free caustic in water and dryout/ dry out and attack by acid chloridecarried over from water demineralization/ breakthrough of acid into water fromdemineralizer. “High fuel-gas pressure”: failure of pressure regulator. “Tube dryout”:tubeside velocity too low. “Low furnace efficiency”: high combustion air flow/ air leakinto the firebox/ high stack temperature/ heat leaks into the system. “Equipment sud-denly begins to underperform”: fouling/ bypass open. “Temperature-control problems”:missing or damaged insulation/ poor tuning of controller/ furnace not designed fortransient state/ unexpected heat of reaction effects/ contaminated fuel/ design error.

60

3.3 Energy Exchange

3.3.3Thermal Energy: Fluid Heat Exchangers, Condensers and Boilers

For not truly countercurrent, if the correction factor for the LMTD drops below 0.75we run the risk of temperature crossover. Provide pressure relief to allow for systemswhere block valves could isolate trapped fluids. Include impingement baffles at shellinlet nozzles to prevent erosion of tubes and flow-induced vibration. Account for thelarger heat exchange that occurs for clean tubes/surfaces; the design was based onreduced heat-transfer coefficients that accounts for ultimate dirty film resistance.Ensure the air is vented. Liquids being heated should leave at the top of the exchan-ger to prevent the buildup of gases coming out of solution and vice versa for liquidswith suspended solids or viscous fluids. Orient baffle windows to facilitate drainage.Slope condensers to remove the condensed phase. Maximum cooling-water temper-ature is 45 �C. Prefer water or other nonflammable heat-transfer media. For flam-mable heat transfer fluids, select operating temperature below its atmospheric boil-ing temperature. If refrigeration is required, prefer less hazardous refrigerant evenif this means operating at higher pressures. Use pinch analysis, identify inefficientexchanges and retrofit heat-exchanger networks to maximize heat recovery. Opti-mize cleaning schedule. Consider “on-line” mechanical cleaning where fouling is aproblem. Use turbulence promoters in laminar flow and gas services and whereturndown has significantly reduced the heat-transfer coefficient. For air-cooled sys-tems include a trim cooler with water as coolant.

Shell and tube heat exchangers: Good practice: to provide lower inventory andintensify, prefer plate exchangers to shell and tube exchangers with the highest sur-face compactness. Trouble shooting: “Thermal underperformance on both streams(coolant exit temperature < design; hot exit > design temperature):” instrument fault/ notenough area/ thermal load reduced via flowrate or change in thermal properties (eg,less hydrogen than design)/ inerts blinding tubes/ [ fouling]* more than expected/tube flooded with condensate (see faulty steam trap, Section 3.5.1) or trap in back-wards or insulated inverted bucket steam trap. “Equipment suddenly begins to under-perform”: fouling/ bypass open. Temperature-control problems”: missing or damagedinsulation/ poor tuning of controller/ not designed for transient state/ unexpectedheat of reaction effects/ contaminated feeds/ design error/ unexpected heat of solu-tion effects/ changes in properties of the fluids. “Heat transfer to shell side fluid < de-sign and Dp < design”: instrument/ increase in viscosity/ fluid bypasses baffles (bafflecut > 20%, no sealing strips, excessive baffle clearance, shell side nozzles too farfrom tube bundle)/ stratification/ faulty location of exit nozzles/ faulty baffling/inlet maldistribution. “Heat transfer to tubeside < design and uneven (and uneven tube-end erosion at inlet)”: maldistribution to the tubes (axial nozzle entry velocity > tubevelocity, for radial nozzle entry velocity > 1.9 tube velocity). “Heat transfer to onefluid < design and Dp= design”: instrument fault/ oil contamination of water. “Thermaloverperformance both fluids, and usually Dp > design on hot fluid side, perhaps charring ofcold stream and freezing of hot stream:” instrument fault/ cocurrent piped incorrectlyas countercurrent/ area too large/ hydrogen concentration in gas stream>design/clean tubes but design area selected on dirty service. “Thermal overperformance one

61


stream: cold exit temperature > expected”: instrument/ plugged tubes/ inlet velocity< design, fouled screen on pump suction/ pump problems, see Section 3.2.3/increased heat load. “Poor control of outlet temperatures (–5 �C)”: poor tuning of con-trol/ instrument fault/ oversized area combined with multipass with local changesin effective MTD with fluid velocity. “Rapid tube failure or glass or karbate tubes break”:inlet gas velocity too high and directed onto tubes/ gas velocity > 5 m/s causing tubevibration/ surges in cooling water/ surges cause by syphon without vent break.“Higher Dp when operating at design flows and temperatures”: underdesign/ design for2-phase stratified flow but slug flow occurs/ gas service but the operating pressur-e < design.

“Leaks”: erosion/ corrosion/ vibration/ improper tube finishing/ cavitation/ lackof support for tube bundle/ tube end fatigue. “Leaks from the gasket at the tube sheetjoint”: sudden upsets that cause the DT between the flange and the bolt to be > 50 �C.“Noise/ vibration”: excessive clearance between baffles and tubes/ inlet gas velocitytoo high and directed onto tubes/ gas velocity > 5 m/s causing tube vibration/ surgesin cooling water/ surges cause by syphon without vent break. “Gradual reduction inheat transfer and increase in Dp”: small tube leaks.

[Fouling]*: change in pH/ water temperature high and invertly soluble com-pounds precipitate/ water temperature high and algae and fungi form/ corrosionproducts/ sublimation/ process condensate freezes/ coolant fouling/ silt deposits/aggregation and destablization of colloids causing wax and asphaltenes to depositfrom hydrocarbons. For more on [ fouling] see Section 3.11.

Shell and tube condensers: Trouble shooting: “Condensation duty < design; exitvapor temperature > design, high flowrate of vapor out vent”: instrument fault/ under-sized condensers/ change in process gas pressure/ inward leakage of non-condensi-bles/ change in feed composition/ [ fouling]* on the process side/ vapor binding/vapor pockets/ inert blanketing (usually near the condensate outlet for condensersoperated flooded for pressure control)/ condensate flooding, see steam traps, Sec-tion 3.5.1/ baffle orientation horizontal not vertical/ excessive entrainment in vaporfeed/ baffle window>45%/ drain line too small/ leakage between the tubesheet andbaffles/ bowed tubesheet/ condenser designed for horizontal service installed verti-cally. “Condensation duty > design:” excess condenser area/ clean tubes/ condenser de-signed for vertical service installed horizontally/ liquid entrainment in feed. “Con-densation duty < design and Dp process > design and excessive flow out vapor vent”: under-sized condenser. “Coolant water temperature > design”: instrument fault/ low coolantflowrate/ high coolant inlet temperature/ cooling tower fault/ excess condenserarea. “Cooling water exit temperature > design and higher steam usage in distillation col-umn reboiler and uneven column operation:” excess condenser area via overdesign orclean surfaces. Heat transfer drops off > rate than expected and Dp increases faster thanexpected”: [ fouling]* because of oversized kettle reboiler on distillation column orchange in pH or flow regime laminar when design was turbulent or higher level ofcontamination in fluids or crud carryover from upstream equipment (e.g. silicafrom catalyst in upstream reactor) or compensation for oversize by reduced coolantflowrate. “Loss of volatile vapor out vent, high vent-gas temperature, degree of subcool-ing < design and unusual temperature profile between vapor inlet and condensate outlet:”

62

3.3 Energy Exchange

instrument error/ underdesign. “Loss of volatile vapor out vent, apparent undersizedarea for condensation of immiscible liquids”: lack of subcooling of condensate/ conden-ser installed horizontally instead of vertically. “Fog formation”: high DT with noncon-densibles present/ high DT with wide range of molar mass of the vapor. “Equipmentsuddenly begins to underperform”: [ fouling]*/ bypass open. Temperature-control prob-lems”: missing or damaged insulation/ poor tuning of controller/ not designed fortransient state/ unexpected heat of reaction effects/ contaminated feeds/ designerror/ unexpected heat of solution effects/ changes in properties of the fluids.

[Fouling]*: change in pH/ water temperature high and invertly soluble com-pounds precipitate/ water temperature high and algae and fungi form/ corrosionproducts/ sublimation/ process condensate freezes/ coolant fouling/ silt deposits.For more see Section 3.11.

Shell and tube reboilers: Good practice: to provide lower inventory and intensify,prefer thermosyphon reboilers to kettle reboilers. If DT> 25 �C, probably the boilingmechanism is film boiling. If DT< 25 �C, usually the boiling mechanism is nucleateboiling.

Trouble shooting, general:”Insufficient boilup”: [ fouling on process side]*/ conden-sate flooding, see steam trap malfunction, Section 3.5.1 including higher pressurein the condensate header/ inadequate heat supply, steam valve closed, superheatedsteam/ boiling point elevation of the bottoms/ inert blanketing/ film boiling/increase in pressure for the process side/ feed richer in the higher boiling compo-nents/ undersized reboiler/ control system fault/ for distillation, overdesigned con-denser. “Equipment suddenly begins to underperform”: [ fouling]*/ bypass open. Temper-ature-control problems”: missing or damaged insulation/ poor tuning of controller/not designed for transient state/ unexpected heat of reaction effects/ contaminatedfeeds/ design error/ unexpected heat of solution effects/ changes in properties ofthe fluids.

“Insufficient boilup and gradual increase in steam pressure to maintain boilup:” [ foul-ing]*/ inerts in steam. “Insufficient boilup and gradual decrease in steam pressure tomaintain boilup:” steam blowing, see steam trap malfunction, Section 3.5.1. “Watercontamination”: leak. “Cycling (30 s–several minutes duration) steam flow, cycling pres-sure on the process side and, for columns, cycling Dp and cycling level in bottoms”: instru-ment fault/ condensate in instrument sensing lines/ surging/ [ foaming]* in kettleand thermosyphon/ liquid maldistribution/ steam-trap problems, see Section 3.5.1,with orifice Dp across trap < design/ temperature sensor at the feed zone in a distil-lation column/ collapsed tray in a distillation column. “Level high in reboiler”: instru-ment/ inlet or exit pipe nozzle too small/ wrong nozzle orientation/ steam trapfault, see Section 3.5.1/ steam trap is above the reboiler. “Breathing: puffs of vaporand entrained liquid leave reboiler:” overdesign/ clean tubes when designed for fouledconditions.

[Inadequate heat supply]*: wet steam/ too great a Dp across steam valve gives wire-drawing and superheat/ steam valve closed/ control system fault.

Kettle: Good practice: rarely underdesigned and should not be used for foams.Trouble shooting: general plus the following symptoms and causes unique to thistype of reboiler:

63


“Surging”: poor liquid distribution/ [ fouling]*. “Low boilup rate and gradualincrease in steam”: film instead of nucleate boiling/ too high a DT/ clean tubes/ con-servative overdesign/ [ fouling]*/ flooding with condensate because of steam-trapproblems, see Section 3.5.1/ bottom temperature elevation/ increase in columnpressure/ feed concentration of light components < design/ not enough heating me-dium. “Low boilup rate and decrease in steam pressure”: steam trap blowing, see Sec-tion 3.5.1. “Low boilup rate, pressure increase in reboiler and surges”: [ foaming]*/inerts/ leaks/ undersized reboiler/ diameter of vent line too small/ top tubes notcovered with liquid/ high liquid level that floods the vapor disengagement space/inlet feed maldistribution/ inadequate vapor disengagement.

[Foaming]*: see Section 3.11.[Fouling on the process side]*: low liquid level causing vapor-induced fouling/ solids

in feed that are trapped by the overflow baffle. For more on fouling see Section 3.11.Thermosyphon: Good practice: vertical thermosyphon reboilers are usually not

used for vacuum or extremely high pressure service. Trouble shooting: general plusthe following symptoms and causes unique to this type of reboiler:

“Insufficient boilup”: [ fouling on the process side]*/ insufficient steam flow/ con-densate flooding/ low liquid level in distillation column gives low liquid circulationand increased fouling/ high liquid level in distillation column (static head > design)or higher density of feed liquid gives higher boiling temperature and circulation andinsufficient vaporization for vertical thermosyphon/ pipe lengths < design/ pipe dia-meter > design/ process fluid in vertical thermosyphon drops below 30–40% of thetube length. For horizontal thermosyphon, appears to be undersized but the causeis liquid feed maldistribution. “Surges in boilup”: process fluid circulation rate toolow/ [ fouling on the process side]*/ wide boiling range/ overdesign. “Cycling (30 s–several minutes duration) steam flow, cycling pressure on the process side and, for columns,cycling Dp and cycling level in bottoms”: in addition to general, all natural circulationsystems are prone to surging/ feed contains high w/w% of high boilers/ vaporiza-tion-induced [ fouling]*/ constriction in the vapor line to the distillation column. Forhorizontal thermosyphon: maldistribution of fluid temperature and liquid.

[Fouling]*: insufficient static head/ excess friction in the pipes/ on the tubesidethe outlet nozzle area < total tube area/ on tubeside the inlet nozzle area < 0.5 totaltube area/ rate of vaporization > 25% of circulation rate/ mass rate of vaporization >mass rate of circulation/ natural circulation rate < 3� vaporization rate/ vaporizationinduced solids. For more see Section 3.11.

Forced circulation: operates with sensible heat mode in the tubes. Trouble shoot-ing: general plus the following symptoms and causes unique to this type of reboiler:“Unstable”: insufficient NPSH in pump, see Section 3.2.3. “Insufficient boilup andrapid fouling”: insufficient circulation/ pump fault, see Section 3.2.3/ plugged circu-lation lines. “Insufficient boilup”: [ fouling]*/ circulation rate low/ pump problems,see Section 3.2.3/ no vortex breaker/ excessive circulation and a wide spread in boil-ing temperatures in bottoms. “Excessive vapor in flash chamber, unstable distillationcolumn operation and apparent underdesign of overhead condenser”: overdesign.

64

3.3 Energy Exchange

Vertical falling-film evaporator: see also Absorbers, Section 3.4.8 and Evaporators,Section 3.4.1 and falling-film reactors, Section 3.6.6. Good practice: always check theliquid feed distribution with water before putting on line.

Trouble shooting: “Boilup < design”: [liquid maldistribution]*.[Liquid maldistribution]*: tubes not vertical/ inadequate calming of feed/ varia-

tions in weir height.Spiral plate exchanger: Trouble shooting: “Heat transfer < design”: stratification

caused by faulty inlet and exit nozzle location/ baffling/ maldistribution.Plate exchanger: Good practice: put regulating and control valves on the inlet

lines, never on the outlet lines, to minimize pressure in the exchanger. Never allowthe exchanger to be under a vacuum. Keep temperature < 120 �C; pressure< 2.5MPa. Trouble shooting: “Leaking gaskets”: temperature too high/ temperaturespike/ pressure too high/ cold fluid stopped but hot fluid continues/ superheatedsteam/ under vacuum.

Air cooled: Good practice: induced draft preferred to forced draft to minimizehot-gas recirculation. Include a water-cooled “trim cooler”. Ensure the exit tubes are“flooded” so that the vapor doesn’t bypass condenser. If extreme cold conditions areexpected, allow for fan to operate in reverse to counteract the overcooling by the nat-ural circulation of cold air. Trouble shooting: “Insufficient condensation”: instrumentfault/ maldistribution along either feed or exit headers/ buildup of non-condensiblesin bottom tube rows/insufficient area/ ambient temperature too high/ fan not work-ing/ blades wrong pitch/ baffles stuck/ [ fouled tubes]*/ hot-gas recirculation/ tubesnot sealed. “Cycling”: control system/ vent for syphon-break is missing on exit mani-fold. “Outlet temperature on tube-side is high”: undersized/ tube [ fouling]* on inside oroutside/ flow maldistribution on process or air side/ hot air recirculation/ air flow-rate too low. “Dp on process side high”: [ fouled]* tube side/ increased liquid viscosity/overcooling/ vapor not condensed. “Exit air temperature > expected”: low air flowrate/flow maldistribution on tube side/ ambient air temperature > expected/ unexpectedhot air recirculation. “Exit air temperature < expected”: high air flowrate/ flow maldis-tribution on tube side/ ambient air temperature < expected. “Sluggish control”: theuse of fan pitch variation as the control variable.

[Fouling]*: see Section 3.11.

3.3.4Thermal Energy: Refrigeration

Trouble shooting: use the p-H diagram for the refrigerant as a basis for troubleshooting. “Compressor discharge pressure < design”: turbine drive problem, power lim-ited/ overloaded centrifugal compressor or valve problem for reciprocating compres-sor/ wrong composition for the speed/ not enough refrigerant/ compressor fault,see Section 3.2.1. “Compressor discharge pressure > design:” fouled condenser/ insuffi-cient air to the cooling tower/ low flowrate of water to the condenser/ air in therefrigerant/ too much refrigerant, level too high. “Compressor discharge pressure > de-sign and condensing temperature normal”: poor drainage from the condenser/ noncondensibles in the refrigerant/ refrigerant letdown valve plugged. “Suction pressure

65


< design”: process feedrate < design/ low circulation of refrigerant/ not enough refrig-erant in chiller, level too low/ leak causing a loss of refrigerant/ compressor prob-lems, see Section 3.2.1. “Suction pressure > design”: process coolant load > design/throttle valve incorrectly adjusted/ low level refrigerant/ compressor problems, seeSection 3.2.1. “Process exit temperature > design, refrigerant temperature from the chil-ler > design and the approach temperature= design”: chiller pressure > design/ notenough refrigerant/ heavy ends impurities in refrigerant/ level of refrigerant in chil-ler < design/ expansion valve plugged/ restriction in refrigerant suction line/ unittoo small/ process fluid velocity too slow/ too much refrigerant in chiller causingflashing in the compressor suction. “Process exit temperature > design, refrigerant tem-perature in chiller= design and approach temperature > design”: fouling on refrigerantside/ fouling on the process side/ chiller unit too small. “Condenser temperature > de-sign”: [ fouled]* condenser

[Fouling]*: see Section 3.11.

3.3.5Thermal Energy: Steam Generation

See thermal energy furnaces/ boilers, Section 3.3.2. See Section 3.2.5 for steam dis-tribution

Trouble shooting steam generation: “Tube failure”: feed water contains impurities/for forced circulation: water circulation rate too low/ dry spots in tubes/ vibrationinduced tube failure. “Wet steam”: rate of steam generation > design causing in-adequate demisting in the steam drum. “Steam production < design”: fuel-gas pres-sure < design/ soot in flue-gas passages/ thermostat incorrect and burners cut outtoo soon/ wrong type of fuel-gas burner. “Stack temperature > design”: not enough air/overfiring.

3.3.6High-Temperature Heat-Transfer Fluids

Good practice: usually a portion of the liquid is purged and replaced with freshmakeup.

Trouble shooting: “Rapid cycling of the furnace or heating elements”: [ fluid velocitylow]*. “Vapor pressure increased”: [thermal cracking of fluid]*. “Noisy pump”: contami-nants such as water/ [ fluid velocity low]*/ [thermal cracking of fluid]*. “Pump dis-charge pressure fluctuates”: contaminants such as water/ [ fluid velocity low]*/ [ther-mal cracking of fluid]*. “Startup of cold unit takes longer than usual”: [oxidation offluid]*. “Heater cannot achieve setpoint”: [oxidation]*/ [thermal cracking]*. “Poor con-trol”: [control valve plugged]*/ [heater cannot achieve setpoint]*/ control designfaulty/ controller not well tuned.

[Control valve plugged]*: [thermal cracking]*/ [oxidation]*/ filter plugged/ filtermissing/ filter not working.

[Fluid velocity low]*: pump problems, see Section 3.2.3/ filter plugged/ controllernot well tuned/ wrong location for filter/ crud left in the lines during maintenance.

66

3.4 Homogeneous Separation

[Oxidation]*: temperature of air in expansion tank > 60 �C/ for higher tempera-tures in expansion tank, dry inert gas blanket not used in the expansion tank.

[Thermal cracking]*: fluid velocity in the furnace or heater < design.See also trouble-shooting suggestions related to gas-liquid separators, Section

3.5.1, furnaces, Section 3.3.2, and pumps, Section 3.2.3.

3.4Homogeneous Separation

The fundamentals upon which most of these processes are based include: mass isconserved; mass transfers because of bulk movement and diffusion. The rate ofmass transfer is proportional to the concentration driving force of the target speciesand the surface area across which the transfer occurs. Phase equilibrium is a usefulstarting approximation but usually it is the rate at which the system moves towardequilibrium that is important. Surface phenomena effects, especially foaming andfouling, wetting and dispersed phase stability are issues to consider.

In this section we consider the separation of species contained in a homogeneousphase, such as a liquid or gas. The separation is based on exploiting a fundamentaldifference that exists between the species. Methods that exploit differences in vaporpressures are evaporation, in Section 3.4.1 and distillation, in Section 3.4.2. Methodsexploiting solubility are solution crystallization, Section 3.4.3; absorption, Section3.4.4, and desorption, Section 3.4.5. Solvent extraction, Section 3.4.6, exploits differ-ences in partition coefficient.

Methods based on exchange equilibrium and molecular geometry include adsorp-tion of species from a gas, Section 3.4.7, and of species from a liquid, Section 3.4.8.Ion exchange, Section 3.4.9, exploits differences in surface activity and exchangeequilibrium.

Membrane separations described include reverse osmosis, Section 3.4.10; nanofil-tration, Section 3.4.11; and ultra and micro-filtration, Section 3.4.12. Separation oflarger sized species are considered “heterogeneous systems” and are considered inSection 3.5.

3.4.1Evaporation

Good practice: keep the pressure drop between the last effect and the inlet to thevacuum device < 3 kPa. Consider vapor recompression for conventional low DT eva-porators such as falling film, forced circulation and horizontal tube falling film.Vapor recompression is rarely used on high DT systems such as rising film, calan-dria and submerged tubes. Trouble shooting: “Product contamination”: leakingvalves/ crud left in storage tanks/ crud left in dead legs in piping/ [corrosion]* prod-ucts/ unexpected chemical reactions/ sampling fault/ analysis fault/ unexpected sol-ubility effects. [Corrosion]*: see Section 3.1.2.

67


Vapor recompression evaporators: “Evaporation rate < design”: [ fouled]* heat-trans-fer surface/ uneven movement of liquid over heat-transfer surface/ feed propertychanges/ excessive noncondensibles from leaks or present in feed/ flooded conden-sate, trap malfunction, Section 3.5.1/ feed temperature < design/ water leakage intothe system/ lower compressor suction pressure, see also Section 3.2.1.”Steam econ-omy low”: instrument fault/ excessive venting especially the first, second and thirdeffects/ vapor exiting through condensate, trap problems, Section 3.5.1 / vapor blow-ing into product flash tank through the liquor lines/ [ foaming]*/ internal afterhea-ters leaking/ afterheater scaled so that liquor from the colder effect is not correctlypreheated for the next effect/ [entrainment]*/ excessive vacuum/ [ fouling]*. “Recov-ery-boiler efficiency low”: [ fouling]*. “Vibration”: vapor velocity high through the firstrow of tubes. “Vacuum problems”: see vacuum, Section 3.2.2.

[Corrosion]*: see Section 3.1.2.[Entrainment]*: poor design of deflector/ liquid level above the tubes/ [ foaming]*.[Foaming]*: see Section 3.11.[Fouling]*: sodium suflate precipitates especially in the first effect/ lignin precipi-

tates especially in the first and second effect/ vapor sulfurization and condensationin third and fourth effects/ velocity too small.

Falling-film evaporator: Trouble shooting. “Evaporation rate < design”: [liquid mal-distribution]*/ steam trap malfunction, see Section 3.5.1/ steam flowrate too small/[ foaming]*/ [ fouling]*.

[Foaming]*: see Section 3.11.[Fouling]*: tubular velocity too small: for 5-cm diameter tubes, recommended velo-

cities are: for viscous liquids use 3 m/s; for the finishing effect, 2–2.7 m/s; for theintermediate effects, 1.5–1.8 m/s; for the initial effects, 1.2–1.5 m/s/ pump prob-lems, see Section 3.2.3.

[Liquid maldistribution]*: not vertical/ inadequate calming of feed/ variations inweir height.

Forced-circulation evaporator: Trouble shooting: usual problems are fouling/scal-ing and high liquid viscosity.

[Fouling]*: tubular velocity too small: for 5-cm diameter tubes, recommended velo-cities are: for viscous liquids use 3 m/s; for the finishing effect, 2–2.7 m/s; for theintermediate effects, 1.5–1.8 m/s; for the initial effects, 1.2–1.5 m/s/ pump prob-lems, see Section 3.2.3. For a more general consideration of fouling see Section3.11.

[Liquid maldistribution]*: not vertical/ inadequate calming of feed/ variations inweir height.

Multiple-effect evaporator: Good practice: capacity of one or more effects in seriesis proportional to (condensing temperature of the steam supplied – temperature ofthe liquid boiling in the last effect) and the overall heat-transfer coefficient. If foam-ing occurs, reduce the liquid level in the effect. Trouble shooting: “Reduced flowratefrom last stage to maintain target strength”: water temperature to contact condenser toohigh/ insufficient condensing area/ [decreased UA]*/ [ foaming]*. “DT higher thanusual before stage “x” and DT lower than usual after stage “x””: [decreased UA in stage“x”/ [ foaming]*. “Steam usage higher than normal”: steam leak into an effect/ bleed

68


rate too high/ poor trap performance, see 3.5.1. “Cycling (30 s–several minutes dura-tion) steam flow, cycling pressure on the process side and, for columns, cycling Dp andcycling level in bottoms”: instrument fault/ condensate in instrument sensing lines/surging/ [ foaming]* in kettle and thermosyphon/ liquid maldistribution/ steam-trap problems, see 3.5.1, with orifice Dp across trap < design/ temperature sensor atthe feed zone in a distillation column/ collapsed tray in a distillation column/unsteady vacuum see Section 3.3.2.

[Decreased UA]*: inadequate condensate removal/ liquid level too low in theeffect/ [ fouling]*/ inadequate removal of non-condensible gas.

[Foaming]*: natural occurring surfactants/ pH far from the zpc/ naturally occur-ring polymers/ solids particles/ corrosion particles/ mechanical foam breaker notrotating/ baffle foam breaker incorrectly designed or damaged/ antifoam ineffective(wrong type or incorrect rate of addition)/ gas velocity too high/ rate of evaporationtoo fast/ overhead disengaging space insufficient height/ liquid downflow over foamtoo low, see also Section 3.11.

[Fouling]*: tubular velocity too small: for 5-cm diameter tubes, recommended velo-cities are: for viscous liquids use 3 m/s; for the finishing effect, 2–2.7 m/s; for theintermediate effects, 1.5–1.8 m/s; for the initial effects, 1.2–1.5 m/s.

3.4.2Distillation

Good practice: for trays, add 10% more trays or two trays to improve operability.Weir height: 5 cm with length 75% of the tray diameter to provide a liquid weir over-flow rate > 5 and < 20 L/s m of weir into the downcomer. Usually use 15 L/s m. Forlower flows use a picket weir. Overall downcomer area should be > 5% total trayarea. For foaming liquids increase downcomer area by 50%. The downcomer exitshould be at least 1.2 cm below the top edge of the outlet weir. Include four, 6-mmdiameter weep holes in each tray for shutdown drainage.

For packing, water test the liquid distributor for good liquid distribution beforestartup.

[Surface tension negative]*: If the surface tension of the distillate > surface tensionof the bottoms (surface tension negative) prefer the use of trays to packings to mini-mize potential for liquid film breakup.

[Surface tension positive]*: If the surface tension of the distillate < surface tensionof the bottoms (surface tension positive), the foam above trays might be unexpect-edly stable.

Trouble shooting: The relationship between the symptom and the causes partlydepends on the control system used. Check the auxiliaries to see if they are at fault:reboilers and condensers, see Section 3.3.3; vacuum, see Section 3.2.2; pumps, seeSection 3.2.3. For packed towers, 80% of the causes are liquid maldistribution.

“Dp across the column » design (> 1�2 the column height), reflux flowrate » usual; DTacross column< design, overhead composition contains heavies > design; surges in the liq-uid overhead, bottoms level low or fluctuates, bottoms pressure > design, higher columnpressure and higher temperature profile below the flooded portion of the column the temper-

69


ature profile > design and all trays below the flood are dry and bottoms composition offspec”: [ jet flooding]*. “Dp across the column » design, reflux flowrate gradually increas-ing; DT across column < design, overhead composition contains heavies > design; bottomslevel low or fluctuates, bottoms pressure > design, and higher temperature profile below theflooded portion of the column the temperature profile > design and “all trays below the floodare dry””: [downcomer flooding]*. “Dp across the column> design”: instrument fault/high boilup rate/ steam flow to reboiler > design. “Dp across the column< design”:instrument fault/ [low boilup rate]*, see Section 3.3/ dry trays/ low feedrate/ feedtemperature too high.

“Feed flowrate < design”: instrument fault/ pump problems, see Section 3.2.3/ filterplugged/ column pressure > design/ feed location higher than design.

“Temperature of feed > design”: instrument fault/ preheater fouled/ feed flowratelow/ heating medium temperature < design, see heaters, Section 3.3.3. “Temperatureof bottoms < design”: instrument fault/ [low boilup]* see Section 3.3.3/ loss of heatingmedium/ steam trap plugged, see 3.5.1/ feedrate to column>design/ feed concen-tration of low boilers (overheads) > design/ feed distributor fouled. “Temperature ofbottoms > design”:instrument fault/ [column pressure > design]*/ high boilup/ over-head condenser vent plugged/ insufficient condensing, see Section 3.3.3. “Tempera-ture at top > design”: instrument fault/ bottom temperature > design/ reflux too low/distillate feed forward too high/ column pressure high/ [ flooding]*. “Temperature attop > design and overhead composition contaminated with too many heavies”: vaporbypassing caused by excessive vapor velocities (high boilup) or not enough liquid ontray or packing, or downcomers not sealed, or sieve holes corroded larger thandesign and tray weeps/ reflux too low/ feed contains excessive heavies. “Temperatureat top < design”: instrument fault/ control temperature too low/ [low boilup]* see Sec-tion 3.3.3. “All temperatures falling simultaneously”: [low boilup]*. “All temperatures ris-ing simultaneously:” pressure rising.

“Overhead off spec”: poor tray or packing efficiency/ [maldistribution]*/ notenough trays or packing/ loss of efficiency/ high concentration of non-condensibles/missing tray/ collapsed tray/ liquid entrainment/ liquid bypass and weeping/ liquidor gas maldistribution. “Overhead contaminated with heavies and excessive reflux rateand high boilup rate”: inadequate gas-liquid contact/ insufficient liquid disengage-ment from vapor/ presence of non-condensibles in feed. “Overhead and bottoms offspec and decreases across column in both DT and Dp “: [dry trays]*. “Overhead and bot-toms off spec”: bypass open on reflux control valve. “Overhead and bottoms off spec,decrease in DT across column and perhaps Dp increase and cycling of liquid in the bot-toms”: [damaged tray]*. “Distillation overhead off spec”: excessive inerts fromupstream/buildup of trace/ purge not sufficient from recycle.

“Separation performance of column decreases”: trace amounts of water/ traceamounts of water trapped in column/ [bumping resulting in plate damage]*. “Levelof bottoms > design”: bottoms pump failure, see Section 3.2.3/ bottoms line plugged.and see implications for reboiler, Section 3.3.3. “Level in bottoms > design” and [columnpressure > design]*/ high boilup/ overhead condenser vent plugged. “Level in bot-toms > design and pressure increase in kettle reboiler and surges”: [ foaming]*, inerts/leaks in kettle reboiler/ undersized reboiler. See also Section 3.3.3. “Bottoms off spec”:

70


loss of tray efficiency/ contamination of bottoms from pump, (from light oil lubri-cant in bottoms pump or forced circulation reboiler)/ transient vapor puff from hor-izontal thermosyphon reboiler, see Section 3.3.3. “Distillate flow too low”: feedratelow/ feed composition of overhead species low/ [low boilup]*/ reflux too high/ over-head control temperature too low. “Distillate flow too high:” feedrate high/ feed com-position of overhead species high/ reflux ratio too low.

“Bottoms and overhead flowrates < design”: [ flooding]*/ excessive entrainment/[ foaming]*/ excessive Dp but not flooded/ plugging and fouling/ [maldistribution]*.“Water hammer in column”: process fluid above the tube sheet of a thermosyphonreboiler. “Cycling of column temperatures:” controller fault. “Product contamination”:leaking valves/ crud left in storage tanks/ crud left in dead legs in piping/ corrosionproducts/ unexpected chemical reactions/ sampling fault/ analysis fault/ unex-pected solubility effects. “Cycling (30 s–several minutes duration) steam flow, cyclingpressure on the process side and, for columns, cycling Dp and cycling level in bottoms”:instrument fault/ condensate in instrument sensing lines/ surging/ [ foaming]* inkettle and thermosyphon/ liquid maldistribution/ steam-trap problems, see Section3.5.1, with orifice Dp across trap < design/ temperature sensor at the feed zone in adistillation column/ collapsed tray in a distillation column.

[Bumping resulting in plate damage]*: trace amounts of water.[Column pressure > design]*: [high boilup]*/ overhead condenser vent plugged.[Damaged trays]*: leak of water into high molar mass process fluid/ large slugs of

water from leaking condensers or steam reboilers/ startup with level in bottoms> design/ attempt to overcome flooding by pumping out bottoms at high rate/ toorapid a depressurization of column/ unexpected change in phase.

[Downcomer flooding]*: excessive liquid load/ restrictions/ inward leaking of vaporinto downcomer/ wrong feed introduction/ poor design of downcomers on bottomtrays/ unsealed downcomers/ [ foaming]*.

[Dry trays}*: flooded above/ insufficient reflux/ low feedrate/ high boilup / feedtemperature too high.

[Foaming]*: surfactants present/ surface tension positive system/ operating tooclose to the critical temperature and pressure of the species/ dirt and corrosion sol-ids/ natural occurring surfactants/ pH far from the zpc/ naturally occurring poly-mers/ solids particles/ corrosion particles/ antifoam ineffective (wrong type or incor-rect rate of addition/ gas velocity too high/ vapor velocity too high/ tray spacing toosmall/ asphaltenes present. A more generic listing of the causes of foaming is givenin Section 3.11.

[Jet flooding]*: excess loading/ fouled trays/ plugged holes in tray/ restricted trans-fer area/ poor vapor distribution/ wrong introduction of feed fluid/ [ foaming]*/ feedtemperature too low/ high boilup/ entrainment of liquid because of excessive vaporvelocity through the trays/water in a hydrocarbon column.

[High boilup]*: see Section 3.3.3.[Low boilup]*: see Section 3.3.3.[Maldistribution]*: weirs not level/ low liquid load/ backmixing/ faulty design.[Premature flooding]*: internal damage/[ fouling]*/ change in feed composition or

temperature/ unexpected entrainment/ [ foaming]*/ incorrect design for downco-

71


mers/ unstable control system/ level control problems/ instrument error/ secondliquid phase in the column.

3.4.3Solution Crystallization

Good practice: to prevent plugging, avoid having natural sumps for suspension typecrystallizers. Control the degree of supersaturation. For crystallizers operated withcooling or evaporative crystallization, the supersaturation occurs near the heatexhange surface. For antisolvent or reaction crystallizers, the key control of super-saturation is local (often the mixing). Differentiate among the different types ofproduct or impurities to solve problems: surface contamination, agglomerationtraps impurities, inclusions, polymorphism. Check that impurities are soluble forthe end point conditions of crystal growth= condition of separation.

Trouble shooting: base approach on mass and energy balances, population ornumber balances. Follow the population density of number versus size. Must knowthe type of crystals and the mode of operation.

“Yield < design”: initial concentration < design. “Impure product because of surfacecontamination”: poor solid-liquid separation/ poor washing, see Sections 3.5.12,3.5.13, 3.5.14. “Impure product because of agglomeration trapped impurities”: wrongpH/ wrong magma electrolyte concentration/ wrong mixing. “Impure productbecause of inclusions”: supersaturation driving force too large. “Impure product becauseof polymorphism”: change in crystal habit during the crystallization process/ isoelec-tric point/ mixing problem. “Crystal habit (shape and aspect ratio) differs from specs”:wrong temperature during growth/ impurities especially surfactants/ supersatura-tion level too high. “Size distribution > design”: supersaturation too close to metastablelimit. “Filtration rate slow”: crystals too small/ large size distribution/ fault with filtra-tion, see Sections 3.5.12 and 3.5.13. “Incrustation, fouling, deposits”: cold spots/ miss-ing insulation/ low suspension density/ protrusions and rough areas on the processsurface/ local supersaturation too high/ cooling surfaces too cold. “Product contami-nation”: leaking valves/ crud left in storage tanks/ crud left in dead legs in piping/[corrosion]* products/ unexpected chemical reactions/ sampling fault/ analysisfault/ unexpected solubility effects. [Corrosion]* see Section 3.1.2.

Vacuum and circulating systems: Trouble shooting: “Crystal size too small”: lowsuspension density/ high circulation rate/ solids in feed causing nucleation sites/feed flowrate > design/ excessive turbulence/ local cold spots/ subsurface boiling/supersaturation too high or too close to the metastable limit. “Insufficient vacuum”:see Section 3.2.2/ obstruction in vapor system/ insufficient cooling water to conden-ser/ temperature of the cooling water to the condensers > design/ air leaks. Forsteam ejectors: steam pressure < design. For mechanical vacuum pumps: seal waterflowrate < design/ rpm<design. “Liquid level in crystallizer fluctuates wildly”: checkout the vacuum system, see Section 3.2.2/ low steam pressure to the steam ejectors/fluctuation in the flow of cooling water to the condensers. “Circulation rate differsfrom design”: see pumps, Section 3.2.3. [Foaming]*: air leaks in pump packing/ air in

72


feed/ air leak in flanges or valve stems. A generic listing of the causes of foaming isgiven in Section 3.11.

3.4.4Gas Absorption

Good practice: the more selective the absorbent the more difficult it is to regeneratethe absorbent. Prefer the use of low holdup internals. Select materials of construc-tion to promote wetting: select critical surface tension of the solid is > the surfacetension of the liquid. If the surface tension of the feed liquid > 2 mN/m larger thanthe surface tension of the bottom exit liquid or the absorption of the solute lowersthe surface tension (surface tension negative) prefer the use of trays to packings tominimize potential for liquid film breakup. If the surface tension of the feed liq-uid > 2 mN/m smaller than surface tension of the bottom exit liquid (surface tensionpositive), the foam above trays might be unexpectedly stable; stable films on pack-ing.

Trouble shooting: for multitube cocurrent falling film absorber: “Concentration ofproduct acid < design, inadequate absorption”: liquid maldistribution/ gas maldistribu-tion. “Low heat-transfer coefficient”: liquid or gas maldistribution. “Hydraulic instabil-ity”: no vent break on the syphon. “Product contamination”: leaking valves/ crud leftin storage tanks/ crud left in dead legs in piping/ corrosion products/ unexpectedchemical reactions/ sampling fault/ analysis fault/ unexpected solubility effects.

For amine absorption of sour gas, Good practice: keep inlet amine solvent tempera-ture at least 5 �C hotter than inlet gas temperature to minimize condensation of vol-atile hydrocarbons in the inlet gas stream. Trouble shooting: “Insufficient absorptionor off-specification for exit scrubbed gas”: feed gas concentration off spec/ feed gas tem-perature or pressure outside operating window: for amine absorbers: > 50 �C for H2Sand < 24 �C for CO2/ feed gas pressure has decreased/ [solvent flowrate too low]*; forglycol dehydration: 12.5 to 25 L TEG per kg water removed / [solvent incorrect]* /incorrect feed try location/ [column operation faulty]*/ absorber operating condi-tions differ from design/ [absorber malfunction]*.

“Dp across absorber > design”: gas flowrate > design/ pressure < design/ [ foaming]*/plugged trays/ plugged demister pads/ collapsed tray or packing. “Dp on column fluc-tuating”: [ foaming]*. �Solvent carryover from the top of the column”: [ foaming]*. “Liquidlevel in vessels fluctuates”: [ foaming in column]*. “Change in absorption rate”: for amineabsorption: decrease in removal of H2S and increase in removal of CO2/ [ foaming]*.“Overloaded liquid in downstream gaseous processing equipment”: [ foaming in absorb-er]*. “Solvent losses high”: [physical losses]*/ [entrainment]*/ [solubility]*/ [vaporiza-tion]*/ [degradation]*/ [loss elsewhere]*/ for glycol dehydration typical losses= 0.015mL/ m3 gas treated.

[Amine concentration too high or too low]*: if too high, lack of equilibrium drivingforce/ if too low, insufficient moles of amine for the feed concentrations.

[Column operation faulty]*: plugged tray or packing/ poor distribution for pack-ing/liquid flowrate <minimum required for loading/ [gas velocity too high]*/ col-

73


lapsed trays or packing/ plugged or broken distributors/ [ foaming]*/ solvent–strip-per overhead temperature too low. see also Section 3.4.2

[Corrosion]*: see Section 3.1.2.[Degradation]*: chemical reaction; for amine: reacts with CO2 and O2; forms stable

salts: for glycol: reacts with O2/ thermal decomposition; for amine: surface tempera-tures > 175 �C ; for glycol: surface temperatures > 205 �C.

[Entrainment: GL]*: demister plugged, missing, collapsed, incorrectly designed/[ flooding]*/ [ foaming]*/ inlet liquid line or distributor undersized of plugged/ poordistribution for packing/liquid flowrate <minimum required for loading/ [gas veloc-ity too high]*/ solvent feed temperature > specifications/ [column operation faulty]*/tray spacing < design. see also GL separators Section 3.5.1

[Entrainment: L-L]*: fluid velocity too high; example > 10 L/s m2/ liquid distributororifice velocity > design; for amine: for amine > 0.8 m/s; for hydrocarbon > 0.4 m/s/faulty location of exit nozzles/ interface level wrong location/ faulty control of inter-face/ no vortex breaker/ exit fluid velocities > design/ insufficient residence time/[stable emulsion formation]*. see also decanters, Section 3.5.3.

[Foaming]*: [ foam-promoting contaminants]*/ [gas velocity too high]*/ [liquid res-idence time too low in GL separator]*/ antifoam addition faulty/ faulty mechanicalfoam breaker/ [liquid environment wrong]*. A generic listing of causes for foamingis given in Section 3.11.

[Foam promoting contaminants: soluble]*: naturally occurring or synthetic poly-mers/ naturally occurring or synthetic organics >C10; example lube oils/ naturallyoccurring or synthetic surfactants; for amine systems: the surface active contaminantsinclude condensed hydrocarbons, organic acids, water contaminants, amine-degra-dation products/ faulty cleaning before startup; surfactants left in vessels.

[Foam promoting contaminants: solid]*: [corrosion products]*; for amine systems:iron sulfides; amine salts formed from organic acids + hydrocarbons/ faulty cleanupbefore startup; rust left in vessel/ dust/ dirt/ particulates.

[Flooding]*: see Section 3.4.2.[Gas velocity too high]*: vessel diameter too small for gas flow/ column pressure

< design/ trays or packing damaged or plugged giving excessive vapor velocity/ tem-perature too high/ upstream flash separator passing liquids: feed contaminated withexcessive volatile species/ stripping gas fed to column too high/ flowmeter error/design error.

[Liquid environment wrong]*: pH far from the zpc/ electrolyte concentration toolow.

[Physical losses]*: leak to atmosphere/ purges for sampling/ sampling/ heatexchanger leak/ pump seal flushes/ filter changes/ piping, fitting, valve stems, gas-kets, pumps.

[Solubility losses]*: liquid-liquid systems: system pressure < design/ for amine: con-centrations > 40% w/w/ system temperatures too high.

[Solvent contaminated]*: carryover from upstream equipment; example oil fromcompressor; brines, corrosion inhibitors, sand, [corrosion products]* / oxygen leaksinto storage tank/ inadequate corrosion control, example low pH causing corrosion/degradation via overheating, ex-hot spots in reboiler tubes or fire tubes/ ineffective

74


filters/ ineffective cleaning before startup/ for amine absorbers: corrosion products/FeS/ chemicals used to treat well

[Solvent feed temperature too high]*: fouled exchanger/ undersized heat exchanger/ambient temperature too high.

[Solvent flowrate too low]*: flowmeter or sensor error/ absorber pressure > design/plugged strainer, lines of filters/ low liquid level in pump feed tank/ [cavitation]*/air-locked pump and see Section 3.2.3 for trouble shooting pumps.

[Solvent incorrect]*: incorrect concentration of active ingredient: for amine absor-bers: [amine concentration too high or too low]*; for glycol dehydration: solvent con-centration TEG< specifications / [solvent stripping inadequate]*/ [solvent feed tem-perature too high]*/ [solvent contaminated]*.

[Solvent loss elsewhere]*: upstream units, for example for glycol dehydration: glycoldumped with hydrocarbons separated in upstream flash drum/ loss in downstreamsolvent stripper.

[Solvent stripping inadequate]*: not enough steam in stripper/ incorrect pressure instripper/ [ foaming]*/ [contaminated solvent]*/ contaminated feed: for amine strip-pers: other sulfur species causing high partial pressure/ leak in the feed preheatercontaminating feed with stripped solvent.

[Vaporization losses]*: system pressure < design/ for amine: concentrations > 40%w/w/ system temperatures too high.

3.4.5Gas Desorption/Stripping

Trouble shooting: “Solvent or stripped liquid concentration > design”: boilup rate orsteam stripping rate too low/ feed concentration > expected/ feed contamination; forsour-water stripper: acid in feed may be chemically bonded with NH3 and prevent ade-quate stripping of NH3/ [ foaming]*/leak in preheater exchanger/ [column malfunc-tion]*.

“Overhead from stripper < specifications”: insufficient flowrate of stripping gas/ forglycol dehydration: reboiler temperature < 175–200 �C or reboiler too small for re-quired duty or fouling of reboiler tubes/ [ foaming]*/ dirty or broken packing orplates/ [ fouled or scaled internals]*/ [ flooding]*/ top pressure > design/ leak in pre-heater exchanger/ [ feed concentration off specification]*. “Overhead temperature onstripper > design”: reflux flowrate too low/ [ flooded]*/ [ foaming]*/ feed contaminatedwith light hydrocarbons. “For sour-water strippers or glycol dehydration: Pressure atreboiler > design”: instrument error/ top pressure > design/ [Dp across column>de-sign]*/ overhead line plugged/ [ flooding]*/ for stripper for glycol dehydration: slug ofhydrocarbon in feed is flash vaporized at reboiler and blows liquid out of stripper.“For sour-water strippers: odor or H2S problems at the storage tank”: 0.6 to 1 m layer ofoil on top of water missing/ oil layer exceeds 0.6 to 1 m depth/ faulty inert gas opera-tion. “Plugging of overhead system”: top temperature not within the operating window;for sour-water strippers: temperature < 82 �C at which ammonium polysulfides formbut temperatures too high give excessive water in overhead vapor causing problemsfor downstream operation / overhead lines not insulated/ insufficient steam tracing

75


on overhead vapor lines. “Feed flowrate and composition to the stripper varies”: [instru-ment error]*/ sampling error/ analysis error/ [ faulty separation in flash drum]*/[ foaming in upstream absorber]*/ no intermediate storage tank between the flashdrum and the stripper/ storage tank faulty operation or design: for SWS: residencetime < 3 to 5 days; stratification occurs, bypassing occurs, insufficient mixing intank/ oil layer on top of water in storage tank exceeds 0.6 to 1 m depth.

[Column malfunction]*: [ feed concentration off specification]*/ excessive strippinggas or steam velocity/ too much cooling or condensation/ top temperature > design/insufficient reflux cooling/ packing broken, damaged/ [ fouled or scaled internals]*/[ foaming]*/ [ flooding]*, see also Section 3.4.2.

[Feed concentration off specification]*: [ foaming in upstream absorber]*/ for glycoldehydration: upstream flash separator passing water; for oil or hydrocarbon in feedto SWS”: residence time for sour water in flash drum is < 20 min.

[Flooding]*: see Section 3.4.2.[Foaming]*: see Section 3.11.[Fouling]*: see Section 3.11[Fouled or plugged internals]*: for SWS: cooling water leak/ pH of feed water too

basic/ calcium ion concentration too high causing precipitation when temperaturesin stripper exceed 122 �C/ temperature < 82 �C at which ammonium polysulfidesform/ overhead lines not insulated.

[Instrument error]*: calibration fault/ sensor broken/ sensor location faulty/ sensorcorroded/ plugged instrument taps: for sour-water strippers: water or steam purge oftaps malfunctioning or local temperatures < 82 �C at which ammonium polysulfidesform.

[Dp across column> design]* see Section 3.4.2.

3.4.6Solvent Extraction, SX

Good practice: the dispersed phase should not preferentially wet the materials ofconstruction. If unexpected rapid coalescence occurs, suspect [Marangoni effects]*and change the dispersed phase. Treat the buildup of the “rag” at the interfacesbased on the cause: corrosion products or stabilizing particulates, surfactants, oramphoteric precipitates of aluminum or iron. Consider adjusting the pH. Solid par-ticles tend to accumulate at the liquid–liquid interface.

Trouble shooting: “Poor separation”: level control fault/ phase velocities too high/contaminant gives stable dispersion/ smaller drop size than design/ rag formation/temperature change/ pH change/ decrease in electrolyte concentration. [Rapid coa-lescence]*: wrong phase is the continuous phase/ [Marangoni instabilities]*/ pH atthe zpc/ high electrolyte concentration in the continuous phase.

[Marangoni effects]*: non-equilibrated phases/ local mass transfer leads to localchanges in surface tension and stability analysis yields stable interfacial movement.

For column extractors: “Decrease in extraction efficiency”: agitator speed to fast/excessive backmixing/ flooding.

76


[Flooding]*: agitator speed too fast/ feed sparging velocity too high/ drop diametersmaller than design.

For centrifugal extractors: for bioprocessing/proteins: Good practice: the partitioncoefficient sensitive to pH, electrolyte type and concentration.

“Product contamination”: leaking valves/ crud left in storage tanks/ crud left indead legs in piping/ corrosion products/ unexpected chemical reactions/ samplingfault/ analysis fault/ unexpected solubility effects.

Other suggestions for trouble-shooting decanters are given in Section 3.5.3. Moreabout stable emulsion formation is given in Section 3.11.

3.4.7Adsorption: Gas

Trouble shooting: ”Wet gas”: steam leak/ leaky valves/ inadequate regeneration/wrong adsorbent/ adsorbent damaged by excessive regeneration temperature/adsorption cycle too long/ [early breakthrough]*. “Dp high”: fine particulates in feed/breakdown of adsorbent/ high gas feedrate. “Product contamination”: leaking valves/crud left in storage tanks/ crud left in dead legs in piping/ [corrosion]* products/unexpected chemical reactions/ sampling fault/ analysis fault/ unexpected solubilityeffects.

[Corrosion]*: see Section 3.1.2. [Early breakthrough]*: gas short-circuiting bed/faulty regeneration/ increased concentration in feed/ other contaminants in feed.

3.4.8Adsorption: Liquid

Good practice: carbon regeneration by multiple hearth furnaces. For edible oils pre-vent contact with air.

Trouble shooting: “Early breakthrough”: liquid short-circuiting bed/ faulty carbonregeneration/ increased concentration in feed/ other contaminants in feed. “Pressuredrop high”: fine particulates in feed/ breakdown of carbon/ high liquid feedrate.

3.4.9Ion Exchange

Good practice: use an upstream degasser to remove carbonic acid.Trouble shooting: usual sources of trouble are change in ions in the feed, the mul-

tiport valves improperly seat so that feed or regenerant bypass into the effluent,clogged liquid distributors, clogged underdrains; degradation of the resin and faultybackwash. Organic fouling mainly affects anionic exchangers. “Throughput capaci-ty < design”: instrument error/ increase in feed concentration/ less resin volume thandesign/ regenerant concentration < design/ regenerant volume <design, 0.5–3.5 L/sL of resin/ regenerant flowrate < design/ wrong regeneration ion/ contamination ofregenerant with high valence ions. “In the spring, reduced flowrate through the unit

77


demineralizing river water”: high concentration of particulates, clay in spring riverwater.

“In the summer for a unit demineralizing water, throughput of the cationic exchangerdecreeses, exchange capacity for calcium and magnesium decreases but the anionicexchanger is unaffected”: suspect ferric or high-valency cation present in the feed. “Inthe summer for a unit demineralizing river water, throughput of the anionic exchangerdecreases, exchange capacity decreases but the cationic exchanger is unaffected”: suspectfertilizer runoff with phosphate, carbonic acid and high sulfate anions as contami-nation. “Contamination in exit liquid > design”: instrument error/ sampling error/ on-line too long/ faulty regeneration/ [ fouled]*/ [poisoned]*/ high feed concentrationof target ions/ feed concentration of high valence co-ions. “Dp > design”: dirt in feed/water from river in springtime/ instrument error/ temperature/ resin void volumechanges/ inlet distribution system blocked/ [resin degradation]* and backwashesinto inlet/ backwash rate too high/ underbed blocked.

[Fouling of the resin]*: iron and high-valence ions/ oil/ mud/ polyelectrolyte/ cal-cium sulfate precipitate/ silica/ barium sulfate/ carbonic acid/ sulfate or phosphate/organics/ algae and bacterial fouling.

[Poison resin]*: cobalticyanide/ polythionate/ ferricyanides/ complex humic acid/color bodies in sugar juices.

[Resin degradation]*: ingress of oxidants/ free chlorine in feed/ temperatureincrease/ [ fouled]*/ [poisoned]*/ corrosion products/ [resin fines]*.

[Resin fines]*: thermal or physical shock/ freeze/thaw.WAC: “Alkalinity leakage during exhaustion cycle”: inadequate regeneration. “Hard-

ness leakage during exhaustion cycle”: regeneration fault with calcium sulfate precipi-tation (if sulfuric acid is the regenerant).

SAC: “Sodium leakage”: inadequate regeneration: wrong concentration, wrongflowrate, wrong length of time. “Hardness leakage during exhaustion cycle”: regenera-tion fault with calcium sulfate precipitation (if sulfuric acid is the regenerant).

WBA: “Mineral acid leakage”: under regenerated/ upstream SAC malfunctioning.“Sodium leakage, high pH and high conductivity”: SAC resins contaminated the bed.“Silica problems”: series regeneration with SBA with pH falling below the isoelectricpoint of silica in the resin bed.

SBA: “Increase in silica leakage”: [resin degradation]*. “Leakage of target ions”:organic fouling. “Low pH and high conductivity”: organic fouling. “Increase rinse quan-tities”: organic fouling. “Low pH (5.5), increased conductivity, increase silica leakage,increase rinse volumes, loss of throughput capacity:” organic fouling. “Product contami-nation”: leaking valves/ crud left in storage tanks/ crud left in dead legs in piping/corrosion products/ unexpected chemical reactions/ sampling fault/ analysis fault/unexpected solubility effects.

78

3.5 Heterogeneous Separations

3.4.10Membranes: Reverse Osmosis, RO

Good practice: consider pretreating hydrophobic membranes for aqueous use.Trouble shooting: “Permeate flow < design:” physical fouling (incorrect/incomplete

pretreatment/ scaling/ biofouling)/ chemical fouling: (pH shift/ incorrect anti-sca-lant dosage). “Permeate quality degradation:” failure of mechanical seal/ chemicalattack of membrane by pH, chlorine or biodegradation / concentration polarization/ post-contamination.

3.4.11Membranes: Nanofiltration

Trouble shooting: “Permeate flow < design:” physical fouling (incorrect/incompletepretreatment/ scaling/ biofouling)/ chemical fouling: (pH shift/ incorrect anti-sca-lant dosage). “Permeate quality degradation:” failure of mechanical seal/ chemicalattack of membrane by pH, chlorine or biodegradation / concentration polarization/ post-contamination.

3.4.12Membranes: Ultrafiltration, UF, and Microfiltration

Good practice: for membranes that are not hydrophobic; check the isoelectric orzero point of charge point of the species in solution compared with the charge onthe membrane and consider changing the pH of operation so that the surfacecharges are the same. For hydrophobic membranes treating aqueous feeds, considerpretreating the membrane to make the membrane surfaces hydrophilic.

Trouble shooting: “Permeate flux < design:” physical clogging (inadequate pre-screening/ backwash problems/ aeration/ recirculation/ increase in influent solidsloading); chemical fouling (change in water quality/ inadequate cleaning). “Permeatequality < design:” failure in mechanical seal, breakage of the membrane or hollowfibres / post contamination via regrowth/ degradation of membrane by pH or chlo-rine.

3.5Heterogeneous Separations

In heterogeneous phase separation we start with at least two phases. Sections 3.5.1to 3.5.4 address the separation of gas from liquid, gas from solid, liquid from liquidand gas from liquid, respectively. Liquid-solid separators include drying, Section3.5.5; screens, Section 3.5.6; settlers, Section 3.5.7; hydrocyclones, Section 3.5.8;thickeners, Section 3.5.9; sedimentation centrifuges, Section 3.5.10; filtering centri-fuges, Section 3.5.11; and filters, Section 3.5.12. Trouble shooting screens to sepa-rate solids is discussed in Section 3.5.13.

79


3.5.1Gas–Liquid

Good practice: install a demister.In this section knockout pots and steam traps are considered.

Knockout pots: Trouble shooting: “Poor separation”: [ foaming]*/ insufficient resi-dence time/ feed and exit nozzles at wrong location/ faulty design. [Foaming]*: sur-factants present/ dirt and corrosion solids/ natural occurring surfactants/ pH farfrom the zpc/ naturally occurring polymers/ insufficient disengaging space abovethe liquid/ antifoam ineffective (wrong type or incorrect rate of addition)/residencetime insufficient/ designed for a vertical vessel but a horizontal vessel installed/vapor velocity too high/ mechanical foam breaker not rotating/ baffle foam breakerincorrectly designed or damaged/ asphaltenes present/ liquid downflow velocitythrough the foam is too low.3) See also Section 3.11.

Steam traps: Good practice: install trap below condensate exit (or with a waterseal if the trap is elevated), use a strainer before most traps, use a check valve forbucket traps. Slant pipes to the trap. Use a downstream check valve for each trapdischarging to a common header. Pipe diameter should be greater than or equal tothe trap inlet pipe diameter. Prefer to install auxiliary trap in parallel instead of abypass. Do not group trap thermodynamic traps because of their sensitivity to down-stream conditions.

Float and thermostatic: usually discharges continuously, low pitched bubblingnoise. High pitch noise suggests live steam is blowing.

Balanced thermostatic: leave about 0.6 m of uninsulated pipe upstream of trap.Diagnostics: when bellows placed in boiling water the expansion should be 3 mm.

Inverted bucket: use initial prime to prevent steam blowing. Diagnostics sounds:when it is functioning well: loud initially, then lower pitch bubbling and thensilence. Discontinuous discharge. When steam is blowing through the trap, thesound is a steady bubbling if primed with a light load or constant rattling; or contin-uous high pitched whistling. Diagnostic for loss of prime: close outlet valve for sev-eral minutes, then open valve slowly and operation should return to normal. If thisfails then check seat and valve.

Thermodynamic: about 6 cycles/minute.Trouble shooting: the major faults are wrong trap, dirt, steam locking in the trap,

group trapping, air binding and water hammer. Too large a trap gives sluggishresponse and wastes steam. Too small a trap gives poor drainage, backup of conden-sate. There is a DT across all traps. “No condensate discharge”: strainer or lineplugged/ steam off/ valves plugged/ no water or steam to the trap/ trap clogged/wrong trap selected/ worn orifice/ steam pressure too high (inverted bucket)/ orificeenlarged by erosion (bucket trap)/ incorrect Dp across the orifice (inverted bucket)/air vent clogged (inverted bucket or thermostatic air vent on float trap)/ valve seatchoked (inverted bucket)/ flabby or elongated bellows (thermostatic)/ superheated

80

3) Turner, J. et al., 1999, HP June p. 119.


steam caused burst joints or scale (thermostatic). “Cold trap + no condensate dis-charge”: strainer or line plugged/ steam off/ valves plugged/ no water or steam to thetrap/ trap clogged. “Hot trap + no condensate discharge”: bypass open or leaking/ trapinstalled at high elevation/ broken syphon/ vacuum in heater coils/ pressure toohigh (inverted bucket)/ orifice too large (inverted bucket)/ vent hole plugged(inverted bucket)/ defective trap parts (inverted bucket)/ clogged orifice (thermody-namic). “Live steam blowing, and inlet and exit temperatures are equal”: bypass open orleaking/ worn trap components/ scale in orifice/ valve fails to seat/ trap lost prime(inverted bucket)/ sudden drops in pressure/ [backpressure too high]* (thermody-namic)/ faulty air release (float)/ trap too large (thermodynamic). “Continuous dis-charge when it should be discontinuous”: trap too small/ dirt in trap/ high-pressuretrap installed incorrectly for low pressure service (bucket trap)/ valve seat cloggedwith dirt/ excessive water in the steam/ bellow overstressed (thermostatic)/ one trapserves > one unit/ strainer clogged. “OK when discharging to the atmosphere but notwhen to a backpressure condensate header”: condensate line diameter too small/ wrongorifice/ interaction with other traps connected to a common header/ condensateline partially plugged/ [backpressure too high]*. “Slow and uneven heating ofupstream equipment”: trap too small/ insufficient air handling capacity/ short-circuit-ing when units are group trapped. “Inverted bucket trap loses prime:” sudden drop inpressure/ faulty seat/ faulty valve. “Upstream process cycling”: defective float/ multiplesources of condensate to a single trap/ trap flooded from condensate header/ con-densate discharged into the bottom of the condensate header/ Dp across the orificeis incorrect for the orifice (inverted bucket).

[Backpressure too high and trap is hot]*: return line too small/ other traps blowingsteam / obstruction in return line/ bypass open/ pressure in header too high.

[Backpressure too high and trap is cold]*: obstruction in return line/ excess vacuumin return line.

3.5.2Gas–Solid

Bag filters and dry cyclones are discussed in this section.

Bag filters: Good practice: replace a complete set of bag filters annually. Install abypass. Limit the number of parallel rows of bags on either side of the walkway to3–4 rows for 20-cm diameter bags and 2–3 rows for 30-cm diameter bags. For clean-ing, use 0.5–0.7 kPa clean, dry air with an air: cloth ratio of 2: 1 for reverse jet and2.5: 1 for shaking. Trouble shooting: “Excessive particle emissions”: cleaning too often/pressure used to clean it too high/ bag breaks/ gas temperature too high and parti-cles crust on movable blowrings and tear bag. “Dp across bags > design”: faulty clean-ing/ improper bag tension/ excessive moisture causing blinding/ poor air distribu-tion/ hopper plugged, see Section 3.10.3/ gas velocity > design. “Short bag life”: exces-sive cleaning/ high inlet gas velocity/ fines > design/ blinding because of condensa-tion, improper cleaning, excessive dust load or high cake density.

81


Dry cyclone: Trouble shooting: “Increase in catalyst losses”: [poor separation incyclone]*. “Opaque flue gas from the vessel”: [poor separation in cyclone]*. “Particulatecarryover that affects operation of downstream equipment”: [poor separation incyclone]*. “Temperature hot spots in upstream reactor”: [maldistribution]*/ localexothermic reactions.

[Attrition of the particles]*: local velocities upstream of cyclone > 60 m/s/ particletoo fragile.

[Change in size of particles in the feed]*: [generation of fines]*/ [coarse particles]*.[Coarse particles (diameter > design)]*: agglomeration of catalyst/ [sintered parti-

cles]*/ wrong specifications for catalyst.[Dipleg unsealed]*: solids level does not cover end of dipleg/ Dp indicator for cata-

lyst level faulty/ Dp indicator for catalyst level OK but bed density incorrect.[Generation of fines]*: [attrition of the catalyst]*/ fines in the new catalyst.[Maldistribution]*: feed distributor poorly designed/ feed distributor plugged.[Plugged dipleg]*: spalled refractory plug/ level of catalyst in bed too high / Dp indi-

cator for catalyst level faulty/ Dp indicator for catalyst level OK but bed density incor-rect. air out periods with a lot of water of steam in vessel.

[Plugged grid holes]*: foreign debris entering with fresh catalyst/ faulty grid design.[Poor separation in cyclone]*: [stuck or failed trickle valve]*/ [plugged dipleg]*/ [dip-

leg unsealed]*/ gas velocity into cyclone too low or too high/ faulty design ofcyclone/ solids concentration in feed too high/ cyclone volute plugged/ hole incyclone body/ pressure surges/ [change in size of particles in feed]*.

[Sintered particles]*: high temperature upstream/ [temperature hot spots in theupstream reactor]*.

[Stuck or failed trickle valve]*: binding of hinge rings/ angle incorrect/ wrong mate-rial/ hinged flapper plate stuck open/ flapper plate missing.

3.5.3Liquid–Liquid

Decanters and hydrocyclones are discussed in this section.

Decanter: Good practice: contamination can interfere with the operation. Tradi-tionally this contamination is surfactants, or particulates. The particulates can becorrosion products, amphoteric precipitates of aluminum or iron. Try changing thepH of the water to alter the surface charge on the dispersed drops. The separationcapacity of a settler/decanter doubles for every 20 �C increase in temperature. Cau-tion, if, to ease this separation, the temperature is increased, such an increase intemperature will increase the bulk-phase contamination because of the increasedcross-contamination by the mutual solubility.

Trouble shooting: “Entrained droplets in liquid effluent”: sensor error/samplingerror (immiscible drops are not being entrained)/ faulty design of separator/ impro-per cleaning of vessel after shutdown, e.g, rust left in vessel/ pressure fluctuation/pressure too low causing flashing/[inaccurate sensing of interface]*/ [drop doesn’tsettle]*/ [drop settles and coalesces but is re-entrained]*/ [drop settles but doesn’t

82


coalesce]*/ [stable emulsion formation]*. “Fluctuation in liquid level”: no vacuum breakon syphon line for bottoms/ level sensor error/ poorly tuned controller/ surges in feed.

[Coalescer pads ineffective]*: temperature too high/ pH incorrect/ fibers have thesame charge as the droplets/ surface tension negative system/ wetting properties offibers changed/ fibers “weathered” and need to be replaced/ flowrate too slowthrough fibers/ wrong mix of fibers/ prefiltering ineffective/ surface tension< 1mN/m for fluoropolymer fibers or < 20 mN/m for usual fibers/ wrong design/included in decanter but should be separate horizontal coalescer promoter unit/faulty design. see Section 3.9.2.

[Density difference decrease]*: dilution of the dense phase/ reactions that dilute thedense phase; for sulfuric acid alkylation: if acid strength < 85% w/w the olefins poly-merize with subsequent oxidation of the polymers by sulfuric acid. As a self-perpe-tuating continuing decrease in acid strength. Alkylate-acid separation is extremelydifficult when acid concentration is 40% w/w.

[Drop doesn’t settle]*: [density difference decrease]*/ [viscosity of the continuousphase increases]*/ [drop size decreases]*/ [residence time for settling too short]*/[phase inversion or wrong liquid is the continuous phase]*/ pressure too low caus-ing flashing and bubble formation.

[Drop settles and coalesces but is re-entrained]*: faulty location of exit nozzles for liq-uid phases/ distance between exit nozzle and interface is < 0.2 m/ overflow bafflecorroded and failure/ interface level at the wrong location/ faulty control of inter-face/ liquid exit velocities too high/ vortex breaker missing or faulty on underflowline/ no syphon break on underflow line/ liquid exit velocities too high.

[Drop settles but doesn’t coalesce]*: [phase inversion]*/ pH far from zpc/ surfactants,particulates or polymers present/ electrolyte concentration in the continuous pha-se < expected/ [coalescer pads ineffective]*/ [drop size decrease]*/ [secondary hazeforms]*/ [stable emulsion formation]*/ [interfacial tension too low]*/ [Marangonieffect]*.

[Drop size decrease]*: feed distributor plugged/ feed velocity > expected/ feed flowspuncture interface/ local turbulence/ distributor orifice velocity > design; for amineunits: for amine > 0.8 m/s; for hydrocarbon > 0.4 m/s/ [Marangoni effects]*/upstream pump generates small drops/ [secondary haze forms]*/ poor design offeed distributor.

[Inaccurate sensing of the interface]*: instrument fault/ plugged sight glass.[Interfacial tension too small]*: temperature too high/ [surfactants present]* at

interface.[Marangoni effects]*: non-equilibrated phases/ local mass transfer leads to local

changes in surface tension and stability analysis yields stable interfacial movement.[Phase inversion]*: faulty startup/ walls and internals preferentially wetted by the

dispersed phase.[Rag buildup]*: collection of material at the interface: [surfactants present]* / parti-

culates: example, products of [corrosion see Section 3.1.3]*, amphoteric precipitatesof aluminum/ naturally occurring or synthetic polymers.

[Residence time for settling too short]*: interface height of the continuous phasedecreases/ [inaccurate sensing of interface]*/ turbulence in the continuous phase/

83


flowrate in continuous phase > expected; for example > 3 L/s m2 / sludge settles andreduces effective height of continuous phase/ [phase inversion]*/ inlet conditionsfaulty.

[Secondary haze forms]*: small secondary drops are left behind when larger dropcoalesces, need coalescer promoter, see Section 3.9.2.

[Stable emulsion formation]*: [surfactants present]* / contamination by particu-lates: example, products of [corrosion products. see Section 3.1.3]*, amphoteric pre-cipitates of aluminum or iron/ pH far from the zpc/ contamination by polymers/temperature change/ decrease in electrolyte concentration/ the dispersed phasedoes not preferentially wet the materials of construction/ coalescence-promoter mal-functioning/ improper cleaning during shutdown/ [rag buildup]*.

[Surfactants present]*: formed by reactions/ enter with feed, example oils, hydro-carbons >C10, asphaltenes/ left over from shutdown, example soaps and detergents/enter with the water, example natural biological species, trace detergents.

[Viscosity of the continuous phase increases]*: temperature too low, for alkylate-acidseparation, temperature < 4.4 �C/ [phase inversion]*/ contamination in the continu-ous phase/ unexpected reaction in the continuous phase causing viscosity increase.

Hydrocyclones: Good practice: control on pressure drop. May be operated as openor flooded underflow. Trouble shooting: “Incorrect separation”: faulty design/ inletpressure too low/ wrong Dp from feed to overflow/ interfacial tension < 10 mN/m/wrong volume split/ feed drop size too small.

3.5.4Gas–Liquid–Liquid Separators

Horizontal drum: Good practice: separates gas, oil and water; as for example as anearly separation of natural gas upstream of drying or to handle sour water. Typically,a relatively small load of hydrocarbon. Often called a “flash drum”. Often follow theflash drum with a storage tank to allow further separation of water and hydrocarbon.Contamination from naturally occurring or synthetic surfactants or polymers, orcorrosion products from upstream processing can cause stable foam or emulsionformation.

Trouble shooting: “Entrained liquid in overhead gas”: sensor error/ [entrainment:GL]*. “Incomplete separation of oil from water”: faulty design of separator/ residencetime of liquid phases too short/ liquid velocity in the decant phases too fast/ [Maran-goni instabilities]*/ liquid feed velocity too high/ poor distribution of liquid feeds/faulty location of exit nozzles for liquid phases/ overflow baffle corroded and failure/interface level at the wrong location/ faulty control of interface/ no vortex breaker atwater and heavy oil exit nozzles/ liquid exit velocities too high/ [emulsification]*/contaminant gives stable dispersion/ smaller drop size than design/ rag formation/temperature change/ pH change/ decrease in electrolyte concentration. See Sections3.5.1 and 3.5.3 for more details.

[Entrainment: GL]*: vessel diameter too small for gas flow/ no demister or demis-ter malfunctioning/ vessel pressure < design/ [ foaming]*/ inlet liquid line or distri-butor undersized or plugged.

84


[Entrainment: L-L]*: liquid velocity too high; example > 10 L/s m2/ liquid distribu-tor orifice velocity > design; for amine: for amine > 0.8 m/s; for hydrocarbon> 0.4m/s / faulty location of exit nozzles/ interface level wrong location/ faulty con-trol of interface/ no vortex breaker/ exit fluid velocities > design/ insufficient resi-dence time/ [stable emulsion formation]*.

[Foaming]*: see Section 3.11 for generic causes.[Marangoni instabilities]*: non-equilibrated phases/ local mass transfer leads to

local changes in surface tension and stability analysis yields stable interfacial move-ment.

[Stable emulsion formation]*: see Section 3.11 for generic causes; Section 3.5.3 formore specific causes. The dispersed phase should not preferentially wet the materi-als of construction. If unexpected rapid coalescence occurs, suspect Marangonieffects and change the dispersed phase. Treat the buildup of the “rag” at the inter-faces based on the cause: corrosion products or stabilizing particulates, surfactants,or amphoteric precipitates of aluminum or iron. Consider adjusting the pH. Solidparticles tend to accumulate at the liquid–liquid interface.

3.5.5Dryer for GS Separation

Trouble shooting: work with an overall mass and energy balance. For continuousrotary steam-tube dryer: “Product moisture content high”: insufficient steam flow/ rota-tional speed too fast/ insufficient area/ particles clump/ moisture content of thefeed too high/ faulty design of dryer/ flights damaged/ incorrect angle of inclina-tion/ upstream batch centrifuge gives periodic wet cake. For fluidized-bed dryer:“Product moisture content high”: solids buildup on gas sparger in fluidized bed(caused because inlet gas temperature too high). For spray dryer: “product wet andclumps form inside spray dryer”/ insufficient gas flow/ inlet gas temperature too low,instrument fault/ feed solids concentration lower than design/ liquid drops largerthan design. For fixed bed-hopper to dry polymer feed for extruder (hot gas < 120 �C):“Feed material not dry”: incorrect drying temperature/ solids throughput > design/instrument error/ input air too moist/ ambient air leaking into drying air circuit/adsorbent for drying air incorrectly regenerated/ air dryer (gas adsorber) fault. Seerelated unit adsorption: gas, Section 3.4.4.

Hopper dryer for polymer feed to extruder. “Polymer pellets leaving hopper are notdry”: incorrect drying temperature/solids throughput > design/ instrument error/input drying air too moist/ ambient air leaking into drying air circuit/ adsorbent fordrying air incorrectly regenerated/ air dryer (adsorber) fault see Adsorption: gas,Section 3.4.7.

3.5.6Screens for Liquid Solid Separation or Dewatering

Batch Screen pack downstream of extruder: Good practice: install standby screenpack with diverter valve to bring standby on line when on–line filter blinds. Trouble

85


shooting: “Solid contamination of product”:mesh size too large/ contamination down-stream of screen pack. “Gels in final product”: gels form in extruder/ Dp across screenpack excessive/ screen area too small/ size and type of screen cannot retain gel/ gelsform downstream of screen pack/ downstream temperature promotes gel forma-tion. “Dp excessive”: filter media too fine/ screen area too small/ screen temperaturetoo low/ gel formation in extruder. “Filter media pushes through back support plate inscreen pack”: screens on the downstream side of the pack not coarse enough or notrigid enough/ support plate holes too large.

3.5.7Settlers for LS Separation

Grit chamber: Trouble shooting: “Floating sludge”: sludge decomposing and buoyedto the surface/ infrequent sludge removal/ sludge not removed from the hoppers.“Excessive sedimentation at the inlet”: fluid velocity too slow. “Intermittent surging”:intermittent pumping rates/ liquid maldistribution of feed. “Sludge hard to removefrom hopper”: high feed concentration of grit, clay/ low velocity in the sludge with-drawal lines.

3.5.8Hydrocyclones for LS Separation

Good practice: control on pressure drop.Trouble shooting: “Underflow too dilute, underflow appears as smooth inverted cone”:

inlet velocity low/ inlet feed pressure low. “Underflow appears as slow, vertical rope ofcoarse solids”: underflow opening too small/ feed concentration of solids higher thandesign. “No discharge from the underflow”: plugged inlet/ plugged underflow. “Under-flow unsteady and variable inlet pressure”: air-gas in the feed.

3.5.9Thickener for LS Separation

Good practice: consider the use of flocculants or use a deep cone. Flocculant dosageshould be related to feed inlet concentration, see also Section 3.9.3. Include high-pressure water purge lines for both forward and reverse flow at > 1 m/s. Raise andlower the rake once per shift. For startup, pump feed into the empty tank andrecycle underflow until the design underflow densities are achieved. Trouble shoot-ing: “Stalled rake”: uneven central feed distribution/ excessive flocculant causingisland formation/ underflow concentration > design/ unpumpable underflow/ try-ing to maintain the underflow concentration when the feed contains fines > design/storing too many solids in the thickener/ “sanding out”/ particle shape differs fromdesign. “Plugged underflow lines”: insufficient fines/ targetting underflow concentra-tion > design/ temperature change/ pump problems, see Section 2.2.3 / suction ve-locity < 0.6–2.5 m/s. “Underflow concentration of solids too low”: removal of too muchunderflow/ flocculation problems that give islands that lead to the feed concentra-

86


tion ratholing directly to underflow. “Supernatant cloudy”: feed velocity excessivecausing breakup of colloidal flocs/ changes in pH or electrolyte concentration caus-ing floc breakup/ excessive feed turbulence/ excessive vertical drops of feed/ insuffi-cient flocculant added or flocculant feedrate constant instead of proportional to sol-ids concentration in the feed. “Sanding out”: too high an underflow concentration/feed concentration of dense particles > 200 mm is > design/ power failure. “Tor-que > expected:” feed concentration suspended solids > expected.

3.5.10Sedimentation Centrifuges

Horizontal scroll discharge decanter. Trouble shooting: “Centrifuge won’t start:” vibra-tion switch triggered/ no power/ motor or starter failure/ overheated motor/ [brokenshear pin]* / lubrication oil flowswitch tripped. “Centrifuge shuts down”: blown fuse/overload relays tripped/ motor overheated/ [broken shear pin]* / lube oil flowswitchtripped. “Excessive vibration”: broken isolators/ motor on flexible mounts/ motorbolts loose/ flexible piping not used/ misalignment/ bearing failure or damaged/loss of plows/, damaged conveyor hub/ solid product buildup in conveyor hub/ con-veyor or bowl not balanced/ conveyor flights worn or portion of blade missing/ tru-nions cracked or broken/ conveyor bowl cracked or broken/ leaking effluent weirs/plugged solids in the effluent hopper/ not level. “High moisture in exit solids”: liquiddams not set alike or set incorrectly/ feed temperature too low/ feedrate too high/effluent hopper plugged or not vented/ conveyor flights worn. “High solids in liquideffluent”: feedrate excessive/ effluent dams set wrong/ no strip installed/ incorrectfeed temperature.

[Shear pins breaks]*: feedrate too high/ solids concentration too high/ foreignmaterial stuck in bowl.

[Solid and screen bowl shear pins break]*: plugged discharge hopper/ conveyorblades bent or rough/ worn bowl strips/ loose or broken trunion bolts/ bowl inade-quately washed/ clearance too large for blade tip to bowl wall/ bowl inside rough/wrong size shear pin.

3.5.11Filtering Centrifuge

Good practice: monitor pH and temperature.

Pusher: Trouble shooting: “Machine floods”: feed concentration < design/ feed-rate > design/ irregular feedrate/ change in size distribution or particle diameter.“Unstable cake formation”: feedrate > design.

87


3.5.12Filter for LS Separation

Good practice: Precoat: 0.75 kg/m2 to give a precoat thickness of 1.6 mm. Rate forprecoat: concentration between 0.3 and 5% w/w and at a rate of 0.7–1.4 L/s. m2.This should give a Dp= 14 kPa. For leaf or rotary filters, maintain consistent pres-sure differential across cake once the cake is formed. Consider adding body feedcontinuously when filtering gelatinous species.

Trouble shooting: “Poor clarity”: leak/ cracks in cake/ partially blinded septa/cakewashing too fast/ flashing of filtrate/ air in filter of feed liquid/ changes in liquidproperties/ incorrect filter aid/ change in temperature of pH/ small diameter parti-cles in feed than design/ process upset. “Short cycle/ high pressure/ low flow”: flowlines too small/ obstruction in outlet line/ pump sucking air/ pressure differentialtoo low/ wide fluctuations in feedrate/ air trapped in filter/ too high a filtration rate.

3.5.13Screens for Solid–Solid Separation

Screen, vibrating: Good practice: if damp or sticky, predry or use heater above thescreen to reduce moisture to < 3%. Avoid resonance frequencies. Usual angle ofoperation is 12 to 18� ; for wet, inclined vibrating screen to 7 to 110. Capacitydecreases if the angle of inclination is too high. Blinding is mainly caused by mate-rial that is 1 to 1.5 times the hole size. Feed thickness should not exceed 4� aperturesize for 1.6 Mg/m3; and not exceed 2 to 3� aperture size for 0.8 Mg/m3. Troubleshooting: “Capacity decreases”: angle of inclination too high/ blinding.

3.6Reactor Problems

Temperature is usually a key variable. Increasing the temperature by 10 �C, doublesthe rate of reaction.

Operating temperature should be at least 25 �C less than the maximum tempera-ture for a catalyst.

For PFTR: increasing the temperature may break down the catalyst into a powderthat causing dusting/ contamination/ plugging problems downstream and increasethe pressure drop in the reactor; may cause the catalyst to agglomerate and deacti-vate with a drop off in conversion and increase in pressure drop; may lead to tubefailure; may promote coking. Increasing the velocity of reactants through the reactordecreases the time available for reaction and exit concentration of product shoulddecrease, pressure drop should increase.

Here are typical trouble shooting symptoms and causes for plug flow tubularreactors, stirred tank reactors and reactive extrusion. For PFTR, we consider multi-tube fixed bed catalyst, nonadiabatic, Section 3.6.1; fixed bed, adiabatic, Section3.6.2; bubble reactors, Section 3.6.3; packed column reactors, Section 3.6.4; trickling

88

3.6 Reactor Problems

bed, Section 3.6.5 and thin film, Section 3.6.6. For STR, we considered batch STR,Section 3.6.7; semibatch, Section 3.6.8; CSTR, Section 3.6.9; fluidized bed, Section3.6.10. In Section 3.6.11 we consider a mix of CSTR, PFTR with recycle. Finally reac-tive extrusion is considered in Section 3.6.12.

3.6.1PFTR: Multitube Fixed-Bed Catalyst, Nonadiabatic

Trouble shooting: “Pressure surge”: possible shutdown?/[runaway reactor]*. “Dpincreases dramatically, top of tubes hot, less conversion than expected”: possible shut-down?/ contamination in feed / [poisoned catalyst]*.

“Rapid decline in conversion”: unfavorable shift in equilibrium at operating temper-ature, for exothermic reactions/ [sintering]*/ [agglomeration]*/ poison in new feed .“Gradual decline in conversion”: sample error/ analysis error/ temperature sensorerror/ [catalyst activity lost]*/ [maldistribution]*/ [unacceptable temperature pro-files]*/ [inadequate heat transfer]*/ wrong locations of feed, discharge or recyclelines/ faulty design of feed and discharge ports/ wrong internal baffles and inter-nals/ faulty bed-voidage profiles. “Gradual decline in conversion and axial temperatureconstant with depth of region increasing with time”: [poisoned catalyst]*. “Gradualdecline in conversion and axial temperatures < usual”: [poisoned catalyst]*. “Gas exitconcentration of reactants high”: sample error/ analysis error/ catalyst selectivity low/[catalyst activity lost]*. “Exit concentration of product higher than design”: reactor leak-ing. “Change in product distribution”: [maldistribution]* / [poisoned catalyst]*/ feedcontaminants/ change in feed/ change in temperature settings.

“Temperature runaways”: [temperature hot spots]*/ [reactor instability]*. “Pressureand bed temperature and reactor unsteady”: water in feed/ [maldistribution]*. “Localhigh temperature/hot spot with T > 100 �C above normal”: [maldistribution of gasflow]*/ instrument error/ extraneous feed component that reacts exothermically.“Local low temperature within the bed”: [maldistribution of gas flow]*/ instrumenterror/ extraneous feed component that reacts endothermically. “Exit gas temperaturetoo high”: instrument error/ control-system malfunction/ fouled reactor coolanttubes. “Temperature varies axially across bed”: [maldistribution]*. “Soon after startup,temperature of tubewall near top > usual and increasing and perhaps Dp increase and lessconversion than expected or operating temperatures > usual to obtain expected conversion”:inadequate catalyst regeneration/ contamination in feed; for steam reforming sulfurconcentration > specifications/ wrong feed composition; for steam reforming:steam/CH4 < 7 to 10. “Soon after startup, temperatures over full length of some tubes >usual and perhaps Dp > or < usual and may increase with time”: faulty loading of thecatalyst/ [maldistribution]*. “Hot bands or stripes; perhaps Dp increase”: low ratio ofsteam to methane/ [carbon formation; whisker type]*/ wrong feed composition: forsteam reforming steam/methane < 7 to 10: 1. “Hot bands or stripes near top and per-haps over all tube and rapidly increasing Dp and conversion < specifications”: [deactivatedcatalyst by pyrolytic coke formation]*/ feed concentration wrong: for steam reform-ing high concentration of heavier hydrocarbons/ steam to hydrocarbon ratio low/[catalyst poisoned]* by sulfur. “Temperature at inlet high and high Dp”: [ for steam

89


reforming: steam contaminated with inorganic solids]*. “Hot bands in top 1/3 oftubes and methane > usual in exit gas and perhaps Dp increase”: contamination in feed/ [poisoned catalyst]*.

“Dp higher than design”: catalyst degradation/ instrument error/ high gas flow/sudden coking/ crud left in from construction or revamp. “Dp increasing graduallyyet flowrate constant”: [coke formation]*/ [dust or corrosive products from upstreamprocesses]*.

“Startup after catalyst regeneration, conversions < standard”: [regeneration faulty]*.“Startup after catalyst replacement, poor selectivity”: bad batch of catalyst/ precondition-ing of catalyst faulty/ temperature and pressures incorrectly set/ instrument errorfor pressure or temperature. “Startup after catalyst replacement, Dp < expected and con-version < standard”: [maldistribution]* and axial variation in temperature/ larger sizecatalyst. “Startup after catalyst replacement, conversion < standard and Dp increasing”:[maldistribution and axial temperatures different]*/ feed precursors present for po-lymerization or coking. “Startup after catalyst replacement, Dp for this batch of cata-lyst > previous batch”: catalyst fines produced during loading/ poor loading. “Startupafter catalyst replacement, conversion < specifications per unit mass of catalyst and moreside reactions”: [maldistribution]*/ faulty inlet distributor/ faulty exit distributor.“Startup after catalyst replacement, poor selectivity”: bad batch of catalyst/ precondition-ing of catalyst faulty/ [tube walls not passified]*/ temperature and pressures incor-rectly set/ instrument error for pressure or temperature. “Startup after catalyst re-placement, increased side reactions and conversion < specification”: catalyst loading notthe same in all tubes.

[Active species volatized]*: [regeneration faulty]*/ faulty catalyst design for typicalreaction temperature/ [hot spots]*.

[Agglomeration of packing or catalyst particles]*: [temperature hot spots]*.[Attrition of the catalyst]*: flowrates > expected/ catalyst too fragile.[Carbon buildup]*: [inadequate regeneration]*/ [excessive carbon formed]*.[Catalyst selectivity changes]*: [poisoned catalyst]*/ feed contaminants/ change in

feed/ change in temperature settings.[Catalyst activity lost]*: [carbon buildup]*/[regeneration faulty]*/ [sintered cata-

lyst]*/ excessive regeneration temperature/ [poisoned catalyst]*/ [loss of surfacearea]*/ [agglomeration]*/ [active species volatized]*.

[Excessive carbon formed]*: operating intensity above usual/ feed changes/ temper-ature hot spots.

[Dust or corrosive products from upstream processes]*: in-line filters not working ornot installed/ dust in the atmosphere brought in with air/ air filters not working ornot installed.

[Loss of surface area]*: [sintered catalyst]*/ [carbon buildup]*/ [agglomeration]*.[Maldistribution]*: faulty flow-distributor design/ plugging of flow distributors

with fine solids, sticky byproducts or trace polymers/ [sintered catalyst particles]*/[agglomeration of packing or catalyst particles]*/ fluid feed velocity too high/ faultyloading of catalyst bed/ incorrect flow collector at outlet.

[Poisoned catalyst]*: poisons in feed/ flowrate of “counterpoison” insufficient/ poi-son formed from unwanted reactions.

90


[Poisons in feed]*: depends on reaction/ contamination in feed/ upstream processor equipment upsets/ changes in feed. Poisons for platforming include high sulfurin feed and high feed end point with upstream equipment failure being compressorfailure/ water upset/ chloride upset. Poisons for steam reforming: sulfur, arsenicand alkali metals in the hydrocarbon or steam.

[Reactor instability]*: control fault/ poor controller tuning/ wrong type of control/insufficient heat transfer area/ feed temperature exceeds threshold/ coolant temper-ature exceeds threshold/ coolant flowrate < threshold/ tube diameter too large.

[Regeneration doesn’t remove all carbon from the catalyst]*: regeneration temperaturenot hot enough/ regeneration time not long enough/ [maldistribution]*.

[Regeneration faulty]*: temperatures too high/ oxygen concentration < standard/oxygen concentration > standard causing too rapid a burn/ incorrect temperatureand time so that coke left on catalyst. [regeneration doesn’t remove all carbon fromthe catalyst]*/ excessive temperature during regeneration.

[Runaway reactor]*: feed temperature too high/ [temperature hot spot]*/ coolingwater too hot/ feed temperature too high.

[Sintered catalyst]*: temperature sensor error/ [temperature hot spots]*/ [maldistri-bution]*/ temperature in reactor too high/ regeneration temperature too high.

[Temperature hot spots]*: bed too deep/ [maldistribution]*/ flowrate < design/instrument error/ extraneous feed component that reacts exothermically.

[Tube walls not passified]*: walls activated unwanted side reactions and faulty passi-vation treatment/ wrong passivation treatment/ no passivation treatment.

3.6.2PFTR: Fixed-Bed Catalyst in Vessel: Adiabatic

Trouble shooting:4) gas-catalytic reactions. Temperature and pressure drops acrossbed are usually key variables. When a hot spot develops, it usually develops at thefront end of the bed and gradually moves through the bed. It may take three to fourweeks to travel through the full bed. If the hot spot is 100–200 �C above normal,then usually carbon is deposited and the catalyst is irrevocably damaged. Tempera-ture control is critical for exothermic reactions. “Dp rapidly increases”: emergencyshutdown? “Pressure surge”: possible shutdown?/[runaway reactor]*.

“Rapid decline in conversion”: unfavorable shift in equilibrium at operating temper-ature, for exothermic reactions/ [sintering]*/ [agglomeration]*/ poison in new feed .“Gradual decline in conversion”: sample error/ analysis error/ temperature sensorerror/ [catalyst activity lost]*/ [maldistribution]*/ [unacceptable temperature pro-files]*/ wrong locations of feed, discharge or recycle lines/ faulty design of feed anddischarge ports/ wrong internal baffles and internals/ faulty bed-voidage profiles.“Gradual decline in conversion and axial temperature constant with depth of regionincreasing with time”: [poisoned catalyst]*. “Gradual decline in conversion and axial

91

4) Based on R.B. Anderson, person communica-tion; H.F. Rase “Fixed bed reactor design anddiagnostics”, 1990, Wiley and Dutta, S. and

R. Gauly, Hydrocarbon Processing, 1999,Sept, 43–50.


temperatures < usual”: [poisoned catalyst]*. “Gas exit concentration of reactants high”:sample error/ analysis error/ catalyst selectivity low/ [catalyst activity lost]*. “Exitconcentration of product higher than design”: reactor leaking. “Change in product distri-bution”: [maldistribution]* / [poisoned catalyst]*/ feed contaminants/ change infeed/ change in temperature settings.

“Temperature runaways”: [temperature hot spots]*/ [reactor instability]*. “Pressureand bed temperature and reactor unsteady”: water in feed/ [maldistribution]*. “Localhigh temperature/hot spot with T > 100 �C above normal”: [maldistribution of gasflow]*/ instrument error/ extraneous feed component that reacts exothermically.“Local low temperature within the bed”: [maldistribution of gas flow]*/ instrumenterror/ extraneous feed component that reacts endothermically. “Exit gas temperaturetoo high”: instrument error/ control-system malfunction. “Temperature varies axiallyacross bed”: [maldistribution]*.

“Dp higher than design”: catalyst degradation/ instrument error/ high gas flow/sudden coking/ crud left in from construction or revamp. “Dp increasing graduallyyet flowrate constant”: [coke formation]*/ [dust or corrosive products from upstreamprocesses]*.

“Startup after catalyst regeneration, conversions < standard”: [regeneration faulty]*.“Startup after catalyst replacement, poor selectivity”: bad batch of catalyst/ precondition-ing of catalyst faulty/ temperature and pressures incorrectly set/ instrument errorfor pressure or temperature. “Startup after catalyst replacement, Dp < expected and con-version < standard”: [maldistribution]* and axial variation in temperature/ larger sizecatalyst. “Startup after catalyst replacement, conversion < standard and Dp increasing”:[maldistribution and axial temperatures different]*/ feed precursors present for po-lymerization or coking. “Startup after catalyst replacement, Dp for this batch of cata-lyst > previous batch”: catalyst fines produced during loading/ poor loading. “Startupafter catalyst replacement, conversion < specifications per unit mass of catalyst and moreside reactions”: [maldistribution]*/ faulty inlet distributor/ faulty exit distributor.

[Active species volatized]*: [regeneration faulty]*/ faulty catalyst design for typicalreaction temperature/ [hot spots]*.

[Agglomeration of packing or catalyst particles]*: [temperature hot spots]*.[Attrition of the catalyst]*: flowrates > expected/ catalyst too fragile.[Carbon buildup]*: [inadequate regeneration]*/ [excessive carbon formed]*.[Catalyst selectivity changes]*: [poisoned catalyst]*/ feed contaminants/ change in

feed/ change in temperature settings.[Catalyst activity lost]*: [carbon buildup]*/[regeneration faulty]*/ [sintered cata-

lyst]*/ excessive regeneration temperature/ [poisoned catalyst]*/ [loss of surfacearea]*/ [agglomeration]*/ [active species volatized]*.

[Excessive carbon formed]*: operating intensity above usual/ feed changes/ temper-ature hot spots.

[Dust or corrosive products from upstream processes]*: in-line filters not working ornot installed/ dust in the atmosphere brought in with air/ air filters not working ornot installed.

[Loss of surface area]*: [sintered catalyst]*/ [carbon buildup]*/ [agglomeration]*.

92


[Maldistribution]*: faulty flow-distributor design/ plugging of flow distributorswith fine solids, sticky byproducts or trace polymers/ [sintered catalyst particles]*/[agglomeration of packing or catalyst particles]*/ fluid feed velocity too high/ faultyloading of catalyst bed/ incorrect flow collector at outlet.


[Poisons in feed]*: depends on reaction/ contamination in feed/ upstream processor equipment upsets/ changes in feed. Poisons for platforming include high sulfurin feed and high feed end point with upstream equipment failure being compressorfailure/ water upset/ chloride upset.

[Reactor instability]*: control fault/ poor controller tuning/ wrong type of control/feed temperature exceeds threshold.

[Regeneration doesn’t remove all carbon from the catalyst]*: regeneration temperaturenot hot enough/ regeneration time not long enough/ [maldistribution]*.

[Regeneration faulty]*: temperatures too high/ oxygen concentration < standard/oxygen concentration > standard causing too rapid a burn/ incorrect temperatureand time so that coke left on catalyst. [regeneration doesn’t remove all carbon fromthe catalyst]*/ excessive temperature during regeneration.

[Runaway reactor]*: feed temperature too high/ [temperature hot spot]*.[Sintered catalyst]*: temperature sensor error/ [temperature hot spots]*/ [maldistri-

bution]*/ temperature in reactor too high/ regeneration temperature too high.[Temperature hot spots]*: bed too deep/ [maldistribution]*/ flowrate < design/

instrument error/ extraneous feed component that reacts exothermically.

3.6.3PFTR: Bubble Reactors, Tray Column Reactors

Wide variety of configurations ranging from tube loop, jet loop, air lift loop and spar-ger “bubble reactor”. Related reactors include CSTR, Section 3.6.9; STR, Section3.6.7.

Trouble shooting: Carryover”: [ foaming]*.[Foaming]*: surfactants present/ dirt and corrosion solids/ natural occurring sur-

factants/ pH far from the zpc/ naturally occurring polymers/ insufficient disenga-ging space above the liquid/ antifoam ineffective (wrong type or incorrect rate ofaddition)/ bubble rate too high/ mechanical foam breaker not rotating/ baffle foambreaker incorrectly designed or damaged/ asphaltenes present/ liquid downflow ve-locity through the foam is too low. See Section 3.11 for generic causes of [ foam-ing]*.

See Trouble shooting: section STR, Section 3.6.7 for more on trouble shootingaerobic bioreactors.

93


3.6.4PFTR: Packed Reactors

These include trickling filters and gas-liquid-solid packed-column bioreactor.

Trickling filter: Trouble shooting: “Plugged: interstitial voids become filled with biolog-ical growth”: packing too small/ packing of variable diameter/ organic to liquid loa-ding > design. “Ice formation on top filter surface”: liquid maldistribution/ feed liquidtemperature too low/ air temperature too low. “Odors”: loss of aerobic conditions/accumulation of sludge and biological growth/ lack of chlorine in influent/ highorganic loadings in feed especially from milk processing and canneries. [Foaming]*:surfactants present/ dirt and corrosion solids/ natural occurring surfactants/ pH farfrom the zpc/ naturally occurring polymers/ insufficient disengaging space abovethe liquid/ antifoam ineffective (wrong type or incorrect rate of addition)/ vapor ve-locity too high/ mechanical foam breaker not rotating/ baffle foam breaker incor-rectly designed or damaged/ asphaltenes present/ liquid downflow velocity throughthe foam is too low.

Gas-liquid-solid packed-column bioreactor: Trouble shooting: Carryover”: [ foam-ing]*.

[Foaming]*: surfactants present/ dirt and corrosion solids/ natural occurring sur-factants/ pH far from the zpc/ naturally occurring polymers/ insufficient disenga-ging space above the liquid/ antifoam ineffective (wrong type or incorrect rate ofaddition)/ bubble rate too high/ mechanical foam breaker not rotating/ baffle foambreaker incorrectly designed or damaged/ asphaltenes present/ liquid downflow ve-locity through the foam is too low. See Section 3.11 for generic causes of [ foam-ing]*.

See Trouble shooting: section STR, Section 3.6.7 for more on trouble shootingaerobic bioreactors.

3.6.5PFTR: Trickle Bed

Good practice: gas-liquid flow cocurrently down through a packed bed of catalyst.Porosity 0.38–0.42. Ensure operation in the correct flow regime. The effectiveness ofthe solid catalyst and of the gas-liquid mass transfer decreases if solid catalyst isnon-wet. For good wetting of the solid keep the surface tension of the solid > surfacetension of the liquid. Prevent foaming. The efficiency depends on the skill ininitially distributing the gas and the liquid. Use liquid distribution plate similar todesign used for packed towers. The liquid distribution plate should have at least50 holes/m2 of catalyst bed.

94


Trouble shooting:5) For trickle bed reactors with specific applications to hydro-treating.

“Low conversion”: feed composition change/ wrong catalyst for feed/ sample error/flowrate error/ feedrate higher but reactor temperature not increased/ temperatureprofile wrong/ thermocouple fault/ controller fault/ feed bypassing reactor throughleak in heat exchanger/ [channeling]*/ [catalyst]*/ [ foaming]*/ For hydrotreating:[hydrogen starvation]*/ catalyst not presulfided/ [incomplete presulfiding of cata-lyst]*. “Sudden loss of activity of catalyst”: heat exchanger leak/ change in feed compo-sition/ For hydrotreating: [hydrogen starvation]*.

“Dp across the catalyst bed > design”; [channeling]*/ cracked hydrocarbon feedstored without effective nitrogen blanket/ solids in feed/ corrosion products fromupstream operations/ bypass on feed filter open/ feed distributor fault/ top catalystsupport tray has holes that are too small/ bottom catalyst bed support tray holes aretoo large/ pugged or partially plugged outlet/ crush strength of catalyst exceededand fines plug bed/ excessive recycle compressor surge causing breakdown of toplayer of catalyst.

For hydrotreating: “Rapid breakthrough of H2S during catalyst sulfiding”: [channel-ing]*.

“Nonuniform bed temperatures across the diameter during sulfiding”: [channeling]*.“Color > specifications”: composition change in feed/ catalyst aged..[Catalyst]*: regeneration failed to remove carbon from catalyst/ excessive regen-

eration temperature > 540 �C causing sintering, > 760 �C molybdenum sublima-tion, > 820 �C reduction in crush strength and change in alumina/ poisons in feed/aged catalyst.

[Channeling]*: nonuniform catalyst bed density/ low superficial flowrate< 1.4 kg/ sm2/ offset, tilted or faulty feed distributor/ thermal shock to upstreampipes or equipment causes scale to dislodge and buildup on bed/ internal vesselobstructions such as thermowells or supports.

[Foaming]*: surfactants present/ dirt and corrosion solids/ natural occurring sur-factants/ pH far from the zpc/ naturally occurring polymers/ insufficient disenga-ging space above the liquid/ antifoam ineffective (wrong type or incorrect rate ofaddition)/residence time insufficient/ designed for a vertical vessel but a horizontalvessel installed/ vapor velocity too high/ mechanical foam breaker not rotating/ baf-fle foam breaker incorrectly designed or damaged/ asphaltenes present/ liquiddownflow velocity through the foam is too low/ operating in the wrong flow regime.

[Hydrogen starvation]*: change in feed composition without corresponding changein hydrogen/ leaks/ dissolution of hydrogen in liquid product/ lower concentrationof hydrogen in treat gas/ flowrate of treat gas < expected because of recycle compres-sor fault.

95

5) Based on Koros, R.M. “Engineering Aspectsof Trickle Bed Reactors”, pp. 579–630 in“Chemical Reactor Design and Technology”H. de Lasa (Ed), Martinus Nijhoff Publishers,

1986 and M.D. Edgar, D.A. Johnson, J.T. Pis-torius and T. Varadi “Trouble Shooting Madeeasy” HP May 1984 p. 65.


[Incomplete presulfiding of catalyst]*: contact with hydrogen at high temperature fortoo long a time/ maximum temperature of 150–175 �C exceeded/ use of crackedfeed/ excessive addition of presulfiding agents.

[Rapid coking of catalyst]*: [hydrogen starvation]*/ temperatures too high.

3.6.6PFTR: Thin Film

Related topics evaporation, Section 3.4.1 for gravity and agitated falling films, absor-bers, Section 3.4.8, and shell and tube heat exchangers, Section 3.3.3.See Section3.3.3 for trouble shooting Vertical falling-film evaporator.

3.6.7STR: Batch (Backmix)

Trouble shooting: Batch STR used for polymerization and, to a lesser extent, nitra-tion, sulfonation, hydrolysis, neutralization and, to a much lesser extent, dehydro-genation, oxidation and esterification can pose potentially unsafe operation. Keyindicators of such potential hazards include “Sudden increase in pressure”, “Unex-plained increase in temperature”, “Failure of the mixer”, “Power failure”, and “Loss ofcooling water”. For any of these conditions our first question should be: emergencyshut down? Our knowledge of the MSDS information for the species and their inter-action with each other and with the environment is critical.

Aerobic bioreactors: Trouble shooting: “inoculation cannot be used for the reactor/fermenter”: [contamination]*. “product formation is inhibited”: [contamination]*. “targetproduct cannot be separated from contaminating species”: [contamination]*. “fermenta-tion broth cannot be filtered”: [contamination]* .”steam out of air filter yields dark brownliquid”: media blowback. “reduction in cell volume, no further product production, nooxygen uptake, no heat production”: [contamination by bacteriophage]* . “foaming”: airleaks through gaskets, coils, jacket, hatch/ pH shifted away from zpc/ particles pres-ent.

[contamination in the first 24 hours]*: contaminated inoculum/ poor sterilization oftank accessories and content/ unsterile air.

[contamination comes in after 24 hours]*: air supply/ nutrient recharges/ antifoamfeed/ loss of pressure during the run/ lumps in the media/ media blowbacks.

[contamination]*: [stock culture contaminated]*/ [raw materials contaminated]*/[inoculation tank contaminated]*/ [ fermenter contaminated]* / [incorrect procedures]*/[ faulty maintenance]*/ [contamination by bacteriophage]*

[stock culture contaminated]*: foreign microorganisms in culture stock/ contami-nated inoculation flash/wrong sterilization procedure/ temperature and pressureinstruments wrong/ air left in sterilization chamber/ sterile area contaminated/ cot-ton plugs contaminated/ [ faulty sterilization]*/ raw material contaminated withspores combined with inadequate germination-sterilization.

[raw materials contaminated]*: dry materials not finely ground/ lumps notremoved/ insoluble solids not suspended in solution well because of lumping or

96


inadequate mixing/ lumps too big to be sterilized in the time–temperature avail-able/ mixing inadequate to keep particles suspended/ particles enter air sparger dur-ing filling operation/ starches or proteins not prehydrolyzed with enzymes.

[inoculation tank contaminated]*: wrong procedures/ tank dirty/ air leak/tempera-ture and pressure instrument fault/ sample line, inoculation fitting dirty/dirty deadspots, debris, corrosion/ media blow back into air filter/ unsterilized air filter/faultyantifoam or pH additive lines.

[ fermenter contaminated]*: [inoculum tank contaminated]* / inoculum line contami-nated/ procedure wrong/ tank dirty/ air leak/ leak from the coil or jacket/faulty sen-sors/ antifoam is not sterile/ dirty gaskets, bottom valve, sample line and valve, ventline valve, vacuum breaker/ nutrient feed tank or line not sterile/ all lines were notup to sterilization temperature/ steam condensate left in lines/ the humidity of thefermenter air upstream of the “sterile filter” is > 90%/ pH and DO probes were notcleaned between runs/ probe holders were not brushed and cleaned with a hypo-chlorite or formaldehyde solution/ for a previously contaminated vessel the valvesand gaskets were not replaced, instrument sensors were not removed and cleaned;high boiling germicide, such as sodium carbonate or sodium phosphate was notused.

[ faulty maintenance]*: braided packing (on agitator shafts for sterile vessels) notreceiving enough germicide/ mechanical seals (on agitator shafts for sterile vessels)not lubricated with sterilizing liquid/ instruments faulty/ bolts on flanges not tigh-tened after heat up to 120 �C/ packed bed air filters not packed to correct density of200–250 kg/m3.

[ faulty sterilization]*: particles too coarse and dry/ particles not wetted/ particlesnot suspended/raw material contaminated with spores plus inadequate germina-tion-sterilization.

[contamination by bacteriophage]*: source usually difficult to trace/ substitute animmune strain/ develop strain resistant to the phage.

“Carryover”: [ foaming]*[Foaming]*: surfactants present/ dirt and corrosion solids/ naturally occurring

surfactants/ pH far from the zpc/ naturally occurring polymers/ insufficient disen-gaging space above the liquid/ antifoam ineffective (wrong type or incorrect rate ofaddition)/ bubble rate too high/ mechanical foam breaker not rotating/ baffle foambreaker incorrectly designed or damaged/ asphaltenes present/ liquid downflow ve-locity through the foam is too low. See Section 3.11 for generic causes of [ foam-ing]*.

Anaerobic digesters: Trouble shooting: “Sludge temperature fluctuates”: instrumentfault/ fluctuating feedrate. “Poor heat transfer with the hot water coils, exit water tem-perature < design: “ sludge solids adhere to heat-transfer surface. “Temperature con-stant but production of methane gas < design”: increased accumulation of scum or grit/excessive acid production with lower pH and volatile acid > 500 mg/L/ organic over-load/ toxic metals in feed/ highly acidic feed/ overdigested sludge. “Foaming”:incomplete digestion/feedrate > design/ inadequate mixing/ temperature too low/withdrawal of too much product (digested sludge)/ rate of reaction > design/ large

97


quantities of organics in feed/ insufficient reaction volume for the high organic fee-drate/ pH shift away from zpc.

3.6.8STR: Semibatch

Trouble shooting: Semi-batch STR used for polymerization and, to a lesser extent,nitration, sulfonation, hydrolysis, neutralization and, to a much lesser extent, dehy-drogenation, oxidation and esterification can pose potentially unsafe operation. Keyindicators of such potential hazards include “Sudden increase in pressure”, “Unex-plained increase in temperature”, “Failure of the mixer”, “Power failure”, and “Loss ofcooling water”. For any of these conditions our first question should be: emergencyshut down? Our knowledge of the MSDS information for the species and their inter-action with each other and with the environment is critical.

Polymerizer: Trouble shooting: “Temperature increases suddenly: “ emergency shut-down?/ mixer stopped/ fouling of heat exchanger/ “gel effect” in polymerizationreaction/ coagulation and product fouls the walls of the reactor. “Particle productsize < design”: too many nucleation sites/ lower level of oxygen than design/ toomuch emulsifier/ too much initiator. “Particle product size > design”: coagulation/ toofew initial nucleation sites/ too much oxygen in the feed/ too little emulsifier/ toolittle initiator/ emulsifier post feed is too late. “Temperature increases > design”: emer-gency shutdown?/ coagulation/ emulsifier post feed too late. “Batch times < design”:too many nucleation sites/ lower level of oxygen than design/ too much emulsifier/too much initiator. “Batch times > design”: too few initial nucleation sites/ too muchoxygen in the feed/ too little emulsifier/ too little initiator.

Agitated bubble reactors: Trouble shooting. Consider increasing the impeller di-ameter or using a disk turbine to increase mass transfer. Trouble shooting: “foam-ing”: mixer tip speed too high/ linear gas velocity too high/ surfactant contaminants/decrease in electrolyte concentration in the liquid/ change in pH/ use of turbineimpeller/ lack of a gas sparger/ mechanical foam breaker not rotating/ disengage-ment space not high enough/ mechanical baffle foam breakers faulty/ antifoamineffective.

“flooded impeller”: too small a diameter impeller/ speed too slow.Gas-liquid-solid bioreactor: Trouble shooting: Carryover”: [ foaming]*.[Foaming]*: surfactants present/ dirt and corrosion solids/ natural occurring sur-

factants/ pH far from the zpc/ naturally occurring polymers/ insufficient disenga-ging space above the liquid/ antifoam ineffective (wrong type or incorrect rate ofaddition)/ bubble rate too high/ mechanical foam breaker not rotating/ baffle foambreaker incorrectly designed or damaged/ asphaltenes present/ liquid downflow ve-locity through the foam is too low. See Section 3.11 for generic causes of [ foam-ing]*.

See trouble shooting: section STR, Section 3.6.7 for more on trouble shootingbioreactors.

98


3.6.9CSTR: Mechanical Mixer (Backmix)

Consider complications because of catalyst deposition and erosion. Trouble shoot-ing: CSTR used for polymerization and, to a lesser extent, nitration, sulfonation,hydrolysis, neutralization and, to a much lesser extent, dehydrogenation, oxidationand esterification can pose potentially unsafe operation. Key indicators of suchpotential hazards include “Sudden increase in pressure”, “Unexplained increase in tem-perature”, “Failure of the mixer”, “Power failure”, and “Loss of cooling water”. For any ofthese conditions our first question should be: emergency shut down? Our knowl-edge of the MSDS information for the species and their interaction with each otherand with the environment is critical. See Semibatch and STR Sections 3.6.8 and3.6.7 for more.

Liquid–liquid: Typically reactor is a CSTR followed by a decanter to separate thephases and recycle the “catalyst” phase to the reactor. Trouble shooting: see decanter:Section 3.5.3a. “Alkylate is purple”: [stable emulsion formation]*/[density differencedecrease]*/ [drops don’t settle]*/ [acid runaway]*. “Dp across the alkylate cooler> design”: [stable emulsion formation]*/ [density difference decrease]*/ [drops don’tsettle]*/ [acid runaway]*/ acid recirculation rate too fast.

[Acid runaway]*: excessive contaminants in feed to reactor/ feedrate too fast/ poorcontact or mixing between isobutane, olefin and acid/ fresh acid makeup feedratestopped/ faulty control/ faulty meter/ ratio of acid: hydrocarbon outside range 45–60% v/v/ ratio of isobutane: olefin < 8: 1/ initial reactor temperature too highor > 18 �C/ poor mechanical design for fresh acid addition. “temperature of the recycledacid is > 1.7 �C hotter than feed entering the reactor”: [acid runaway]*/ alkyl sulfatespolymerize in the decanter/ acid recirulation rate too fast.

[Density difference decrease]*: dilution of the dense phase/ reactions that dilute thedense phase; for sulfuric acid alkylation: if acid strength < 85% w/w the olefins poly-merize with subsequent oxidation of the polymers by sulfuric acid. as a self-perpetu-ating continuing decrease in acid strength. Alkylate-acid separation is extremely dif-ficult when acid concentration is 40% w/w.

[Drop doesn’t settle]*: [density difference decrease]*/ [viscosity of the continuousphase increases]*/ [drop size decreases]*/ [residence time for settling too short]*/[phase inversion or wrong liquid is the continuous phase]*/ pressure too low caus-ing flashing and bubble formation.

[Drop settles and coalesces but is re-entrained]*: faulty location of exit nozzles for liq-uid phases/ distance between exit nozzle and interface is < 0.2 m/ overflow bafflecorroded and failure/ interface level at the wrong location/ faulty control of inter-face/ liquid exit velocities too high/ vortex breaker missing or faulty on underflowline/ no syphon break on underflow line/ liquid exit velocities too high.

[Drop settles but doesn’t coalesce]*: [phase inversion]*/ pH far from zpc/ surfactants,particulates or polymers present/ electrolyte concentration in the continuous pha-se < expected/ [coalescer pads ineffective]*/ [drop size decrease]*/ [secondary hazeforms]*/ [stable emulsion formation]*/ [interfacial tension too low]*/ [Marangonieffect]*.

99


[Drop size decrease]*: feed distributor plugged/ feed velocity > expected/ feed flowspuncture interface/ local turbulence/ distributor orifice velocity > design; for amineunits: for amine > 0.8 m/s; for hydrocarbon > 0.4 m/s/ [Marangoni effects]*/upstream pump generates small drops/ [secondary haze forms]*/ poor design offeed distributor.

[Inaccurate sensing of the interface]*: instrument fault/ plugged sight glass.[Interfacial tension too small]*: temperature too high/ [surfactants present]* at

interface.[Marangoni effects]*: non-equilibrated phases/ local mass transfer leads to local

changes in surface tension and stability analysis yields stable interfacial movement.[Phase inversion]*: faulty startup/ walls and internals preferentially wetted by the

dispersed phase.[Rag buildup]*: collection of material at the interface: [surfactants present]* / parti-

culates: example, products of [corrosion see Section 3.1.2]*, amphoteric precipitatesof aluminum/ naturally occurring or synthetic polymers.

[Residence time for settling too short]*: interface height of the continuous phasedecreases/ [inaccurate sensing of interface]*/ turbulence in the continuous phase/flowrate in continuous phase > expected; for example > 3 L/s m2 / sludge settles andreduces effective height of continuous phase/ [phase inversion]*/ inlet conditionsfaulty.

[Secondary haze forms]*: small secondary drops are left behind when larger dropcoalesces, need coalescer promoter, see Section 3.9.2.

[Stable emulsion formation]*: [surfactants present]* / contamination by particu-lates: example, products of [corrosion products, see Section 3.1.2]*, amphoteric pre-cipitates of aluminum or iron/ pH far from the zpc/ contamination by polymers/temperature change/ decrease in electrolyte concentration/ the dispersed phasedoes not preferentially wet the materials of construction/ coalescence–promotermalfunctioning/ improper cleaning during shutdown/ [rag buildup]*.

[Surfactants present]*: formed by reactions/ enter with feed, example oils, hydro-carbons >C10, asphaltenes/ left over from shutdown, example soaps and detergents/enter with the water, example natural biological species, trace detergents.

[Viscosity of the continuous phase increases]*: temperature too low, for alkylate-acidseparation, temperature < 4.4 �C/ [phase inversion]*/ contamination in the continu-ous phase/ unexpected reaction in the continuous phase causing viscosity increase.

100


3.6.10STR: Fluidized Bed (Backmix)

Trouble shooting: First we consider fluidized-bed reactors in general, then fluidizedcombustors or regenerators and then provide specifics for a fluid catalyst crackingunit, FCCU, which consists of a riser or fluidized-bed reactor, cyclone separator,steam stripper, spend catalyst transport, air-oxidizing regenerator, cyclone separatorand a regenerated catalyst return.6)

General fluidized-bed reactor: “Gradual change in yield”: [carbon buildup]*. “Pooryield”: [Loss of catalyst activity]*/ [maldistribution]*/ [unacceptable temperature pro-files]*/ [inadequate heat transfer]*/ wrong locations of feed, discharge or recyclelines/ faulty design of feed and discharge ports/ [inadequate mixing]*/ [excessivebackmixing]*/ wrong internal baffles and internals/ [poor bubbling hydrody-namics]*/ [inadequate solids circulation rates in reactor]*. “Change in product distri-bution”: [maldistribution]* / poisoned catalyst/ feed contaminants/ change in feed/change in temperature settings.”Temperature hot spots”: [maldistribution]*/ localexothermic reactions. “Temperature runaways”: temperature hot spots. “Pressure andbed temperature and reactor unsteady”: water in feed/ reactor grid hole erosion/ [mal-distribution]*/ for FCCU: surging regenerator holdup/ unsteady reactor-regeneratordifferential pressure controller operation/ rough circulation/ incorrect aeration ofU-bend/ incorrect aeration of standpipe/ sticky stack slide valves/ sensor control per-formance for stack slide valve unsatisfactory.”Particulate carryover that affects opera-tion of downstream equipment”: [poor separation in cyclone]*. “Shifts in yield distribu-tion”: [Feed contaminated with light hydrocarbons]*/ [sintered catalyst]*/ coarse par-ticles. “Dp increase across the grid”: [plugged grid holes]*/ fluid flow >usual. “Dpacross grid < expected”: air flowrate < design/ [eroded grid holes]*/ for FCCU: [Failureof internal seals in regenerator]*. “Erratic or cycling pressures”: [surging of the catalystbed]*. “Catalyst losses increase”: [poor separation in cyclone]*/ insufficient head spaceabove bed/ fluidization velocity too high/ increase in volume of product throughunexpected side reactions/ change in feed flowrate/ flowrate instrument error/ ve-locity through reactor too high/ pressure surges/ attrition of catalyst.

[Attrition of the catalyst]*: steam flowrate > expected/ air flowrate > expected/ localvelocities into the dense phase > 60 m/s/catalyst too fragile.

[Carbon buildup]*: [inadequate regeneration]*/ [excessive carbon formed]*.[Coarse particles (diameter > design)]*: [generation of fines]*/[loss of catalyst

fines]*/ [poor separation in cyclone]*/ agglomeration of catalyst/ [sintered catalyst]*/wrong specifications for catalyst.

[Eroded grid holes]*: hole velocity too high / materials of construction/ contami-nants in fluid. [Excessive backmixing]*: [maldistribution]*/ [poor bubbling hydrody-namics]*.

101

6) Based on Luckenbach, E.C. et al. “EncyclopediaProcessing andDesign”,Marcel Dekker 1981p. 89; Dutta, S. and R. Gualy, “Overhaul pro-

cess reactors”, HP 1999 Sept pp. 43–50, Lieber-man, N. P. “Troubleshooting Process Opera-tions”, 2nd ed. 1985 Pennwell Books.


[Excessive carbon formed in cracker]*: cracker operating intensity above usual; forFCCU excess aromatics in feed / changes in feed/ poor catalyst stripping/ heavierrecycle/ leakage of fractionator bottoms into the feed/ [sintered catalyst]*/ [ feed con-taminated with metals]*/ [ feed contaminated with heavy hydrocarbons, especiallyaromatics]*.

[Failure of internal seals in regenerator for FCCU]*: pressure bump during startup/regenerator pressure too high/ velocity through the grid too low/ low flow of air tothe grid/ stresses too high/ erosion/ abnormal conditions with the auxiliary burneron startup.

[Gas bubbles too big]*: particles heavier than design/ particles larger than design/sintered particles/ single fluidized bed too deep.

[Gas bypassing in fluidized bed]*: particles heavier than design/ particles largerthan design/ agglomerated particles/ single fluidized bed too deep instead of multi-ple beds in series.

[Gas velocity too high]*: [increase in production of light ends in reactor]*.[Generation of fines]*: [attrition of the catalyst]*/ fines in the new catalyst.[Inadequate heat transfer]*: [maldistribution]*/ insufficient heat exchanger area/

design error/ fouled exchanger. see Section 3.3.8.[Inadequate mixing]*: [maldistribution]*/ [poor bubbling hydrodynamics]*. see

Section 3.7.1.[Inadequate regeneration]*: [regenerator doesn’t remove all carbon from the cata-

lyst]*/ excessive temperature during regeneration/ coarse particles.[Loss of catalyst activity]*: [carbon buildup]*/[inadequate regeneration]*/[sintered

catalyst]*/ excessive regeneration temperature/ [poisoned catalyst]*/ [loss of surfacearea]*.

[Loss of catalyst fines]*: insufficient disengaging space above the top of the bed/agglomeration of catalyst/ [poor separation in cyclone]*/ Dp indicator for catalystlevel faulty/ Dp indicator for catalyst level OK but bed density incorrect.

[Loss of surface area]*: [sintered catalyst]*/ [carbon buildup]*.[Maldistribution]*: faulty feed-distributor design/ plugging of fluid distributors

with fine solids, sticky byproducts or trace polymers/ [temperature hot spots]*/ [sin-tered catalyst particles]*/ [poor bubbling hydrodynamics]* / [poor circulation]*.

[Plugged dipleg]*: spalled refractory plug/ level of catalyst in bed too high / Dp indi-cator for catalyst level faulty/ Dp indicator for catalyst level OK but bed density incor-rect. air out periods with a lot of water or steam in vessel.

[Plugged grid holes]*/ foreign debris entering with fresh catalyst/ faulty griddesign/ lumps of coke or refractory in catalyst/ failure of grid hole inserts/ [sinteredcatalyst]*/ bits of refractory.


[Poisons in feed]*: depends on reaction: for FCCU poisons in the feed include ni-ckel, vanadium and sodium; the counterpoison is a solution of antimony.

[Poor bubbling hydrodynamics]*: [Gas bypassing in fluidized bed]*/ [gas bubblestoo big]* particles heavier than design/ [particles larger than design]*/ [sintered cat-alyst]*/ fluid feed velocity too high/ too deep a bed of catalyst/ [maldistribution]*.

102


[Poor circulation]*: coarse particles/ [maldistribution]*.[Poor separation in cyclone]*: [stuck or failed trickle valve]*/ [plugged dipleg]*/ dip-

leg unsealed/ solids level does not cover end of dipleg/ gas velocity into cyclone toolow or too high/ faulty design of cyclone/ solids concentration in feed too high/cyclone volute plugged/ hole in cyclone body/ Dp indicator for catalyst level faulty/Dp indicator for catalyst level OK but bed density incorrect/ pressure surges.

“Dp > design”: fines in packed beds/ fines in distributors/ fines in exit nozzles/crud left in from construction or revamp.

[Reactor instability]*: control fault/ poor controller tuning/ wrong type of control/insufficient heat transfer area.

[Regenerator doesn’t remove all carbon from the catalyst]*: damaged air grid/ insuffi-cient air/ excessive regenerator velocity/ poor spent catalyst initial distribution/coarse particles.

[Sintered catalyst]*: local high temperatures/ [maldistribution]*/ for FCCU [after-burn in regenerator]*/ [Feed contaminated]*/ high temperature in the regenerator/[temperature hot spots in the reactor]*.

[Solids conveying lines flow capacity < design]*: sticky fines buildup in lines/ wrongDp across line.

[Stuck or failed trickle valve]*: binding of hinge rings/ angle incorrect/ wrong mate-rial/ hinged flapper plate stuck open/ flapper plate missing.

[Surging of the catalyst bed]*: water in the feed/ [plugged grid holes]*/ faulty griddesign/ [grid holes eroded]* /[ for FCCU: failure of internal seals in regenerator]*/for FCCU: [seal failures]*/ hole in the overflow well/ [reactor instability]*/ controlfault in Dp between cracker and regenerator.

[Unacceptable temperature profiles]*: fluctuating temperature/ unsteady bed tem-peratures.

Specific for a fluidized bed combustion/catalyst regenerator: “Increase in catalystlosses”: [poor separation in cyclone]*/ [ failure in regenerator plenum]*/ for FCCU[ failure of internal seals in regenerator]*.

[Failure in regenerator plenum]*: faulty cyclone design/ catalyst feed too high/regenerator velocity too high/ faulty spray nozzles causing impingement of plenumsprays/ temperatures too high causing failure in plenum.

Specific for fluid cat cracker unit – including regenerator system: “Overloaded wetgas compressor”: for FCCU high hydrogen production/ increase in production of lightends. “Gas compressor flow reversal”: [poisoned catalyst]*. “Gas compressor surge”: [poi-soned catalyst (that causes production of lower MM species)]*. “Gas compressor flowreversal”: [poisoned catalyst]*. “Wet gas compressor surge”: [poisoned catalyst (thatcauses production of lower MM species)]*.

“Hydrogen concentration in wet gas increases”: [poisoned catalyst, especially with ni-ckel and vanadium]*/ [ feed contaminated with metals, especially nickel and vana-dium]*/ [loss of catalyst activity]*/ feed concentration high in hydrogen/ loss of anti-mony solution addition. “Increase in the production of light ends”: for FCCU [ feed con-taminated with metals]*/ feed concentration high in light ends.

“Erratic or cycling instrument records on holdup, density and the overflow well”: [sur-ging of the catalyst bed]*. “Opaque flue gas from the regenerator”: [poor separation in

103


cyclone in regenerator]/ fluidization velocity too high/ increase in volume of productthrough unexpected side reactions/ change in feed flowrate/ flowrate instrumenterror. “Vibration in the preheat system”: [ feed contaminated with water]*. “Dp increasebetween reactor and fractionator inlet”: [coking in overhead lines]*. “Dp lower on theregenerator slide valve”: [poisoned catalyst]*. “Dp between cracker and regenerator incor-rect”: fault with the input air blower/ fault with the flue-gas slide valve on the regen-erator/ fault with the regenerated catalyst slide valve/ fault with the spent-catalystslide valve/ fault with the wet gas compressor/ fault downstream of the wet-gas com-pressor, such as plugged fractionator overhead condensers (with ammonium chlo-ride salts)/ changes in environment air conditions. Regenerator should be about20 kPa higher than the cracker for Dp across the regenerated catalyst slide valve.“Dp between cracker and regenerator fluctuating”: fluctuating temperature in cracker/fluctuating pressure in regenerator/ fluctuating catalyst circulation rate/ fluctuatinglevel in the overflow well/ shift in catalyst between cracker and regenerator/ incor-rect aeration of U-bend/ incorrect aeration of standpipe/ sticky stack slide valves/sensor control performance for stack slide valve unsatisfactory/ moisture in aerationmedium/ unsteady control of air/ U-bend vibration. “Dp across cylcone > expected”:steam flowrate > expected/ air flowrate > expected. “Pressure fluctuating in regenera-tor”: incorrect aeration of U-bend/ incorrect aeration of standpipe/ sticky stack slidevalves/ sensor control performance for stack slide valve unsatisfactory. “Pluggedpump on the bottoms of the fractionator”: [poor separation in cyclone]*/ velocitythrough reactor too high/ faulty cyclone design. “Overflow well level low”: [eroded gridholes]*. “Overflow well level high”: [plugged grid holes]*. “Overflow well level fluctuat-ing”: incorrect aeration of U-bend/ incorrect aeration of standpipe/ sticky stack slidevalves/ sensor control performance for stack slide valve unsatisfactory/ hole in theoverflow well.

“Catalyst loss from the regenerator increased”: [plugged grid holes]*/ [eroded gridholes]*/ foreign debris entering with fresh catalyst/ faulty grid design/[poor separa-tion in cyclone]*/ steam flowrate > design/ air flowrate > design. “Catalyst circulationfluctuates”: [Dp between cracker and regenerator fluctuating]*/ fluctuating tempera-ture in cracker/ fluctuating temperature in regenerator/ fluctuating level in the over-flow well/ shift in catalyst between cracker and regenerator/ incorrect aeration ofU-bend/ incorrect aeration of standpipe/ sensor control performance for air systemunsatisfactory/ moisture in aeration medium/ unsteady control of air/ U-bend vibra-tion/ coarse particles/ hole in the overflow well/ incorrect aeration of U-bend/ incor-rect aeration of standpipe/ sticky stack slide valves/ sensor control performance forstack slide valve unsatisfactory/ [surging of the catalyst bed]*. “Catalyst becomeslighter in regenerator gradually”: [afterburn in regenerator]*. “Catalyst in fractionatorbottoms”: [poor separation in cyclone]*/ velocity through reactor too high/ faultycyclone design. “Catalyst has salt and pepper appearance after regeneration”: air griddeficiency/ [Failure of internal seals in regenerator]*. “Reduced rates of spent catalystwithdrawal”: [poor separation in cyclone in regenerator]*.

“Temperature difference between bed and cyclone inlet in regenerator”: [ failure of inter-nal seals in regenerator]*/ [afterburn in regenerator]* and [inadequate regenera-tion]*. “Temperatures of bed and cyclone are uneven in the regenerator”: hole in the over-

104


flow well/ [plugged grid holes]*/ foreign debris entering with the fresh catalyst/faulty grid design. “Temperature on regenerator shell or U-bend high”: damaged refrac-tory. “Temperature increase in the dilute phase relative to the dense phase”: [afterburn inregenerator]*. “Temperature of the regenerator cannot be lowered”: [low catalyst circula-tion rate]*. “Temperatures of regenerator too high > 750 �C”: excessive heat release.

“Temperature in dilute phase decreases relative to temperature of the dense bed in theregenerator”: [regenerator doesn’t remove all carbon from the catalyst]*. “Feed preheatrequirements > usual”: [low catalyst circulation rate]*. “Unexplained increase in coke”:[poor catalyst stripping]*. “High bottom sediment and water levels in the slurry oil prod-uct”: [poor separation in cyclone in the cracker]*. “Higher H/C ratio”: [poor catalyststripping]*. “Excess oxygen in regenerator high”: [afterburn in regenerator]*/ [pluggedgrid holes]*/ [eroded grid holes]*/ faulty grid design. “Ratio of carbon dioxide to car-bon monoxide is higher than usual”: [afterburn in regenerator]*. “Uneven oxygen distri-bution in the dilute phase”: [Failure of internal seals in regenerator]*. “Unsteady heatbalance”: [surging of the catalyst bed]*. “Stripping steam flowrate > expected”: flow-meter error/ steam traps faulty/ partially opened valves/ missing restrictive orifice.“Air flowrate > expected”: flowmeter error/ partially opened valves/ missing restrictiveorifice. “Flow reversal with feed going incorrectly to the regenerator”: [Dp across theregenerator slide valve is < design]*.

[Afterburn in regenerator]*: for FCCU [ failure of internal seals in regenerator]*/too much excess air/ oxygen recorder reading incorrect/ meter error for feed andrecycle flowmeters/ meter error for cyclone flowmeter/ [insufficient carbon produc-tion on catalyst during cracking]*/ air flowrate to regenerator too high/ [pluggedgrid holes]*/ [eroded grid holes]*/ faulty grid design causing localized air-distribu-tion problem.

[Coking in overhead lines]*: insulation missing or damaged on transfer line/ ex-tremely cold/ increase in heavies and condensibles in reactor products.

[Control air flowrate too low]*: controller for air faulty or poorly tuned.[Failure in regenerator plenum]*: faulty cyclone design/ catalyst feed too high/

regenerator velocity too high/ faulty spray nozzles causing impingement of plenumsprays/ temperatures too high causing failure in plenum.

[Feed contaminated with metals]*: abnormal operation in the upstream atmospher-ic and vacuum units.

[Feed contaminated with heavy hydrocarbons]*: leak in heat exchangers/ partly openvalves. [Feed contaminated with light hydrocarbons]*: leak in heat exchangers/ partlyopen valves.

[Feed contaminated with sodium]*: seawater leak in upstream equipment/ treatedboiler feedwater leaks into feed/ upset in upstream caustic unit.

[Feed contaminated with water]*: water in feed tanks/ leaks from steam-out connec-tions/ steam leaks in tank heaters/ water not cleaned out of the lines at startup/moist air not removed from lines at startup.

[Higher reactor velocities]*: [ feed contaminated with metals]*.[Higher regenerator holdup]*: hole in the overflow well.

105


[Increased air requirements in regenerator for the same conversion in the cracker]*:[ feed contaminated with heavy hydrocarbons]*/ [inadequate regeneration]*/ [cokeon catalyst > usual].

[Insufficient coke production on catalyst during cracking]*: cracking operation inten-sity is lower than usual/ higher quality of feed to the cracker than usual for FCCUfewer aromatics in feed.

[Low catalyst circulation rate]*: partial blockage of the U-bends/ excessive strippingsteam/ insufficient aeration/ [control air flowrate too low]*/ differential pressure be-tween cracker and regenerator set incorrectly or fluctuating.

[Poor catalyst stripping]*: insufficient steam stripping flowrate/ faulty flow control-ler on steam flow/ faulty design of stripper/ reactor temperature too low/ faulty con-tacting between steam and catalyst/ circulation rate too high/ coarse particles.

[Dp across the regenerator slide valve is < design]*/ sudden drop in regenerator pres-sure/ regenerator slide valve sticking partly open/ compressor surge (see Section3.3.2).

[Regenerator doesn’t remove all carbon from the catalyst]*: [excessive coke formed incracker]*/ low excess oxygen/ oxygen sensor error/ flowmeter error for air/ [poor airdistribution]*/ flowmeter error for feed and recycle/ air flowrate too small.

[Sodium on catalyst]*: carryover of sodium from upstream units (caustic)/ treatedboiler feedwater used in regenerator sprays/ [ feed contaminated with sodium]*.

[Uneven oxygen distribution in the regenerator]*: hole in the overflow well/ [pluggedgrid holes]*/ foreign debris entering with fresh catalyst/ faulty grid design.

[Unstable catalyst bed]*: airflow too low/ grid holes eroded/ faulty grid design.

3.6.11Mix of CSTR, PFTR with Recycle

Conventional activated sludge: Trouble shooting: “Increase in sludge volume index”:high-density inerts in feed and usually caused few operating problems. “Decrease insludge volume index and “bulking”: high concentration of dissolved organics in feed.“Sludge rises”: excessive nitration. “Frothing”: decrease in aeration suspended solids/increase in surfactants in feed/ aeration > design/ increase in temperature.

3.6.12Reactive Extrusion

Trouble shooting: “Inadequate mixing of liquid reactant with polymer”: liquid flowratetoo high/ screw channel under injection not full of polymer/ faulty screw design.“Residuals in final polymer > design”: vent temperature too low/ screw speed too low/polymer feedrate too high/ screw design does not provide enough shear/ vent pres-sure too high. “Polymer has crosslinked or degraded”: screw rpm too high/ degree offill too low/ feedrate too low/ heat-zone temperatures set too high/ screw-designfault giving excessive shear/ [screw tip pressure too high]*. “Extruder torque exces-sive”: throughput too high/ screw speed too low/ heat-zone temperatures set toolow/ faulty screw design. “Unable to melt material”: throughput too high/ screw

106

3.7 Mixing Problems

speed too low/ faulty screw design/ material too slippery. “Residence time too short”:throughput too high/ [degree of fill too high]* / faulty screw design. “Residence timetoo long”: throughput too low/ [degree of fill too low]* / faulty screw design. “Gels orcrosslinked materials”: localized initiator concentration too high/ [melt temperaturetoo high]*.

[Degradation of melt in extruder]*: [RTD too wide]*/ barrel temperature too high/screw speed too high (causing overheating and shear damage)/ oxygen present/[oxi-dation]*/ nitrogen purge ineffective/ wrong stabilizer/ wrong screw/ flows notstreamlined/ stagnation areas present/ extruder stopped when tempera-tures > 200 �C/ copolymer not purged with homopolymer before shutdown/[resi-dence time too long]*.

[Degree of fill too high]*: feedrate too high/ screw speed too slow.[Melt temperature too high]*: screw speed too high/exit barrel zone temperatures

too high/ screw tip pressure too high/ degree of fill too low/ [shear intensity toohigh]*/ heat-zone temperatures set too high/ [screw tip pressure too high]*.

[RTD too narrow]*: [degree of fill too high]*[Screw tip pressure too high]*: screens plugged/ die or adapter or breaker plates too

restrictive and give too much Dp/ [polymer viscosity too high]*/ temperatures in dieassembly too low/ barrel temperature too low/ screw speed too high/ [shear inten-sity too low]*/ lubricant needed/ flow restriction/ throughput too high/ die land tooshort/ cold start/ [degradation of melt in extruder]*.

[Shear intensity too low]*: screw speed too low/ faulty screw design.For other symptoms see Section 3.9.6.

3.7Mixing Problems

Here we consider mechanical agitation of liquids and liquid-solid systems. Operat-ing information is given for solids blenders, in Section 3.7.3.

3.7.1Mechanical Agitation of Liquid

Good practice: prefer motionless mixers to intensify. For systems where the viscosityincreases with time (ex. polymer reactors) prefer turbines to propellers because tur-bines are power self-limiting. Check shaft wobble to ensure that impeller will not hitvessel walls if turned on in an empty tank. Consider a foot bearing.

Trouble shooting propeller/ impeller mixers: “Shaft wobble/vibration”: impellerspeed too close to the 1st critical speed/ shaft runout at the impeller and impellereccentricity too large/ insufficient support.

“Excessive gear-reducer maintenance”: excessive load/ high shock loads/ excessiveshaft bending.

107


“Excessive packing wear”: insufficient lubrication/ excessive shaft wobble/ shaft isout of round.

“Failure of the mechanical seal”: dirty lubricant/ not enough lubricant/ excessiveshaft wobble.

3.7.2Mechanical Mixing of Liquid–Solid

Trouble shooting stirred tanks: “Solids floating on the surface”: solids not wetted byliquid/ insufficient vortex.

3.7.3Solids Blending

Use the Johanson indices to characterize particles: (see also related topic storagebins, Section 3.10)

Arching index [m], AI, = diameter of the circular exit hole from a hopper that willensure that an arch collapses in a conical bin or circular mixer, values range 0–1.2 m;

Ratholing index [m], RI, = diameter of the circular exit hole from a hopper thatwill ensure rathole failure and cleanout in a funnel-flow bin or mixer, values rangefrom 0–9 m. (If RI > 3 then likely “lumps”.)

Hopper index [degrees], HI = the recommended conical half-angle (measuredfrom the vertical) to ensure flow at the walls. Usually add 3� to account for variabili-ty. Values range 14–33� with 304 s/s.

Flow ratio index [kg/s], FRI =maximum solids flowrate expected after deaearationof a powder in a bin. (measures consistency: small FRI for fine, highly compressibleparticles; Large FRI for particles > 400 mm, incompressible, very permeable.) Valuesrange 0–90 kg/s.

Bin density index [Mg/m3], BDI = bulk specific mass expected in a container fullof solids or, in a mixer, when mixer stops and solid is allowed to deaerate. Values0.3–1.6 Mg/m3.

Feed density index [Mg/m3], FDI = bulk specific mass at the conical hopper ormixer’s discharge outlet. Values are 1–60%<BDI.

Chute index [degrees], CI, = recommended chute angle (with the horizontal) atpoints of solids impact. Values= angle of slide. values= 20–90�. High values suggestparticles stick to sides of mixer or bins.

Rough wall angle of slide [degrees], RAS= angle (relative to the horizontal) thatcauses continual sliding on a solid on an 80-grit sandpaper surface with a pressureof 140 kPa gauge. Values 20–35�. Approximately equal to the angle of repose.

Adhesion angle index [degrees], AAI = difference between angle of slide (with hor-izontal) after an impact pressure of 7 MPa or (CI–10�) and the angle of slide withoutimpact pressure.

Spring back index [%], SBI = percentage of solids that spring back after consolida-tion.

108

3.8 Size-Decrease Problems

Good practice: usually 300 s blend time. We can invest much effort into mixingsolids, but we must prevent demixing after the blends are mixed. The four mecha-nisms of demixing are 1) sifting, 2) angle of repose, 3) fluidization and 4) air cur-rent. Here are the details: 1. demixing via sifting: occurs if the particles are free flow-ing with mix of particle size with one size > 3� the diameter of the other. 2) demixingvia angle of repose: this occurs with moderately free-flowing particles (AI < 0.18 m)with different angles of repose or two different RAS. For particles characterized inthis way, the only blender that seems to prevent demixing is the air-pulse blender.3) Demixing by fluidization: tends to occur if the blend contains > 20% fluidizingfines characterized by AI < 0.18 m; RI < 1.5 m; FRI < 0.76 kg/s plus coarser materialwith AI < 0.012; RI < 0.6 m and FRI > 7.6 kg/s. This type of demixing tends to occurif the action of the mixer induces air. 4) Demixing by air current: air carriers super-fine, easy-flow, non-agglomerating particles into voids. This is a problem if AI < 0.18m; RI < 1.5 m; FRI < 0.4 kg/s. Again try to avoid mixers whose action induces air.

Trouble shooting for polymer blenders of feedstock for extruder: “Material does notflow”: bridging/ see also hoppers, Section 3.10. “Components do not feed”: jammedvalve or auger/ solids blockage or bridging/ power fault in feeder. “Inconsistent flow-rate”: bridge or block in blender/ jammed discharge mechanism/ inconsistent feed-rates to blender. “Wrong blend compositions:” calibration error in feeder.

3.8Size-Decrease Problems

The focus here is on gas liquid systems that include bubble columns, Section 3.8.1,in packed columns, Section 3.8.2, and in agitated tanks.

3.8.1Gas Breakup in Liquid: Bubble Columns

Good practice: electrolytes in the liquid alter the bubble diameter, the holdup, theinterfacial area per unit volume in mechanically agitated devices and affects the kL afor bubble columns.

3.8.2Gas Breakup in Liquid: Packed Columns

Good practice: the critical surface tension of the solid packing should be greaterthan the surface tension of the liquid to ensure that the liquid film remains intact ina packed contactor.

109


3.8.3Gas Breakup in Liquid: Agitated Tanks:

Good practice: consider increasing the impeller diameter or using a disk turbine toincrease mass transfer. Trouble shooting: “Foaming”: mixer tip speed too high/ line-ar gas velocity too high/ surfactant contaminants/ decrease in electrolyte concentra-tion in the liquid/ change in pH/ use of turbine impeller/ lack of a gas sparger.“Flooded impeller”: too small a diameter impeller/ speed too slow.

3.9Size Enlargement

The fundamentals used to solve troubles for size-increase operations are surfacephenomena. Surfaces are attracted to each other by van der Waals forces; surfacesare repelled by the electrochemical double layer or by steric hindrance. Surface ener-gies, contact angles, and wetting are important. Specific symptoms and causes forequipment to coalesce drops in gas are listed in Section 3.9.1; for coalesce liquiddrops in a liquid environment, Section 3.9.2, and create solid aggregates or flocs in aliquid environment, Section 3.9.3. Then we consider the creation of larger-size parti-cle clusters by tabletting, Section 3.9.4; and pelleting, Section 3.9.5. The last twoprocesses considered in this section focus on change in shape by injection moldingand extrusion, Section 3.9.6; and coating, Section 3.9.7.

3.9.1Size Enlargement: Liquid–Gas: Demisters

Good practice: consider using mesh pads upstream of impact filter beds to reducethe load on the filter bed. Consider installing mesh pads vertically to facilitate drai-nage and minimize re-entrainment. For non-corrosive and non-fouling, considerinstalling vane separators downstream of mesh pads to collect larger drops shearedoff from the mesh pad. Cannot be used for up to 25% turnup capacity; avoid the useof inertial devices for up to 25% turndown capacity. Trouble shooting: “Demistersineffective”: temperature too high/ fibers have the same charge as the droplets/ wet-ting properties of fibers changed/ fibers “weathered” and need to be replaced/ flow-rate too slow through fibers/ wrong mix of fibers/ prefiltering ineffective/ [ foam-ing]*/ wrong design/ re-entrainment. [Foaming]*: see Section 3.11.

3.9.2Size Enlargement: Liquid–Liquid: Coalescers

Good practice: consider decreasing the temperature to decrease the solubility andincrease the surface tension. Adjust pH for water flowing through fibrous and meshbeds so that drop and fiber have the opposite surface charge. Promote coalescencein solvent-extraction systems by using surface tension positive configurations. Trou-

110

3.9 Size Enlargement

ble shooting: “Coalescer pads ineffective”: temperature too high/ pH incorrect/ fibershave the same charge as the droplets/ surface tension negative system/ wettingproperties of fibers changed/ fibers “weathered” and need to be replaced/ flowratetoo slow through fibers/ wrong mix of fibers/ prefiltering ineffective/ surface ten-sion < 1 mN/m for fluoropolymer fibers or < 20 mN/m for usual fibers/ wrongdesign/ included in decanter but should be separate horizontal coalescer promoterunit/ faulty design/ [stable emulsion formed]*. [Stable emulsion formed]*: see Section3.11.

3.9.3Size Enlargement: Solid in Liquid: Coagulation/Flocculation

Coagulation and flocculation in general: Trouble shooting: “Supernatant not clear”:[coagulation doesn’t occur]*/ flocculation doesn’t occur]*/ [ floc doesn’t settle out]*/[ floc forms but breaks up]*.

[Coagulation doesn’t occur]*: wrong dosage of coagulant-flocculant/ wrong counter-ion/ pH different from expectations/ pH far from zpc/ faulty mixing in the rapidmix/ valence on the counterion too small/ charge on the dispersed particles or dropsreversed from expectations.

[Flocculation doesn’t occur]*: faulty fluid dynamics into the basin/ reel at wrongrpm/ residence time too short/ mixing not tapered/ unexpected turbulence/ tooshort a residence time between coagulant and subsequent flocculant dosage.

[Floc doesn’t settle out]*: floc formed is too loose/ settler fault.[Floc forms but breaks up]*: local turbulence > shear strength of floc.For water treatment: Trouble shooting: “Coagulation-flocculation ineffective, super-

natant murky”: pH>10/ wrong dosage of alum or coagulant / pH<4/ increase inconcentration of particles in feed/ rpm of reels in flocculation basin too slow/ feedtemperature < 12 �C/rpm of reels in flocculation basin too fast.

For latex: Trouble shooting: “Exit crumb too small”: brine concentration too high/temperature too low/ power input too high/ wrong pH. “Excessive amount of fines insupernatant”: brine concentration too high/ wrong pH/ temperature too low.“Strength of the resulting crumb < specifications”: pH too high and brine concentrationtoo high.

3.9.4Size Enlargement: Solids: Tabletting

Trouble shooting: “Product tablet weight > design”: sample error/ lab error/excessivefines.

3.9.5Size Enlargement: Solids: Pelleting

Trouble shooting: For strand pelletizer for polymer resin. “Pellet diameter too small”:hole too small for the desired throughput/ extruder output too low. “Pellet diameter

111


too large”: output too high/ speed too low/ feedroll speed too low/ output too highfor die-size. “Pellet too short or too long”: mismatch ratio of feedroll speed versus rotorteeth speed. “Strands dropping”: feedroll pressure too small/ throughput too low/die-plate has too many holes. “Pellet cuts are angled”: feed not perpendicular to strands/strands overlapping. “Pellet oval shaped”: feedroll pressure too high/ inadequate cool-ing before cutting. “Pellet has tails”: incorrect clearance between rotor and cutters.For water-ring pelletizer for polymer resin: “Pellet diameter too small”: hole too smallfor the desired throughput/ throughput too large. “Pellet diameter too large”: outputtoo low. “Pellet too short or too long”: mismatch throughput versus cutter speed.“Blocked holes”: nonuniform pressure on the die face/ throughput too low/die-platehas too many holes. “Pellet oval shaped”: cutter speed too high/ inadequate cooling.“Pellet has tails”: incorrect clearance between die and cutters/ worn cutter blades.

3.9.6Solids: Modify Size and Shape: Injection Molding and Extruders

Consider first injection molding machines, then extruders.

3.9.6.1 Injection molding machinesGood practice: resin should be dried < 0.02%. Do not use resin that has been out ofthe dryer for > 20 min. Cold molds are difficult to fill and require higher injectionpressures. Hot molds, generally, give better finish and less molded-in stress. Melttemperature is very sensitive to very small changes in rpm or backpressure despitesensor or controller set point. Measure with hand-held pyrometer or laser sensor.Need slower fill rate for sprue-gated parts to prevent blush, splash or jetting. If thewalls are > 5 mm, then slow fill helps reduce sinks and voids.

Backpressures of 0.35 to 0.7 MPa help ensure homogeneous melt and consistentshot size. As backpressure increases, melt temperature increases. Holding or back-pressures that are 0.4 to 0.8 of injection pressures are typical. To purge a machine,acrylic is recommended.

Trouble shooting: injection molding: Basically the cause can be with the material,the machine, the operator, the operating conditions, the mold or the part design. Tocheck on the material, try material from another supplier; to check the machine, usesame material, conditions, mold on another machine; if the trouble is random, thenit is probably the machine; try a different operator on the machine; trouble appearsin same location in the product, then flow conditions and look for problems fromthe front of the piston to the gate. The symptom-cause information is presented asissues related to appearance (color, surface finish and transparency), strength andshape defects and operation and symptoms.

. Appearance (Color, surface finish, transparency)

Color: “Discoloration” (typically appears before burn marks appear; location appears atthe weld line or where air is trapped in the mold): [contamination in heating cylinder]*/sensor error/ control error/ [degradation, mechanical]*/ [degradation, thermal]*/[melt too hot]*/ [melt not homogeneous]*/ overall cycle too long/ [contamination in

112


hopper and feed zones]*/ incorrect cooling of ram and feed zone/ [venting in moldinsufficient]*/ [residence time too long]*/ cooling time too short/ dryer residencetime too long / excessive clearance between screw and barrel / clamp pressure toohigh/ injection forward time too long/ gate too small/ runner-sprue-nozzle toosmall. “Black specs inside transparent product”: faulty cleanout of machine from pre-vious molding operations/ failure to purge when not running for extended times/nozzle too hot/ barrel temperature in the feed area is too low combined with highscrew speed or high backpressure/ sensor error/ sensor located too far from heaterbands/ hangup in nozzle tip/ nozzle adapter and end-cap. “Brown streaks/burning”:wet feed/ [melt too hot]*/ [shear heating in the nozzle]*/ [degradation, mechani-cal]*/ loose nozzle/ wrong nozzle/ dead spots in hot manifold/ mold should be coldrunner system/ gate or runner too small/ [contamination]*/ injection speed toofast/ booster time too long/ injection pressure too high/ mold design lacks vents atburn location/ gate size too small or at wrong location/ plunger has insufficient tol-erance to allow air to escape back around the plunger/ poor part design/[venting ofmold insufficient]*/ [residence time too long]*. “Brown streaks at the weld lines or atthe end of flow paths, black or charred marks”: [air trapped in mold]*. “Brown streaks atthe same location”: nozzle loose, wrong, too hot/ [shear heating (at gate, runner, cavityrestrictions)]*. “Brown streaks dispersed throughout”: material fault at the hopper: wetmaterial. “Weld burns”: [melt too hot]*/ injection speed too fast/ [mold too cold]*/injection hold time too long/ injection pressure too high/ faulty nozzle heatingbands/ [air trapped in mold]*.

Surface finish: “Sink marks” (difficult to remove by changing processing conditions):[cooling insufficient before removal from mold]*/ [short shot]*/ [melt too hot, caus-ing excessive shrinkage]*/ [solidification at mold wall too slow]*/ wrong location forgate/ holding pressure too low/ injection speed to fast/ backpressure too high/ gatestoo small/ faulty runners/ booster time too low/ fault with nozzle, sprue or runners/[mold temperature non-uniform]*/ moist feed/ [thick sections continue to shrinkafter the melt path is frozen]*/ hold time too short/ holding pressure too low/ [back-flow from mold]*/ lubricant insufficient/ volatiles in feed/ [solidification at moldwall delayed]*/ [viscosity too high]*/ cooling-water temperature too low /excessivecushion in front of ram/ size of nozzle, sprue and runner too small/ [air trapped]*.“Fine ridges running perpendicular to the flow front”: [melt too cold]*/ [short shot]*.“Flow lines” see also “jetting”: feed moist/ injection pressure too low/ [melt too cold]*/screw not rotating during injection/ injection speed too slow/ backpressure too low/nozzle orifice too small/ [mold too cold]*/ gates too small/ [venting of mold insuffi-cient]*/ feedrate too small/injection rate too low/ injection hold time too low/ boos-ter time too low/ relocate gates/ faults with nozzle, runner, sprue or gate/ clamppressure too high/ unequal filling rates between cavities/ core position incorrect/gates too small/ mold design fault with non-uniform thickness of sections or exces-sive heavy bosses or ribs. “Low gloss, dull or rough surface”: moist feed/ injection pres-sure too low/ [mold too cold]*/ [melt too cold]*/ injection rate too slow/ relocategates/ mold cooling time too short/ surface of sprue, runner, or cavity rough/ [con-tamination]*/ [venting of mold insufficient]*/ screw rpm too low/ injection madewithout screw rotation/ injection speed too fast/ backpressure too low/ nozzle ori-

113


fice too small/ increase or decrease mold temperature/ gate size too small/ [melt toohot]*/ diameter or depth of cold slug is too small/ wrong location of gate/ wronglocation for water channels/ particle size not uniform/ too many fines in feed/wrong type of lubricant. “Streaks on part”: stock temperature too high or too low/screw rpm too high/ nozzle or shutoff valve no tight/ injection speed too fast/ back-pressure too low/ cooling and mold-open time too short. “Splay marks: coarse lines orlumps”: [degradation of melt thermally]*/ injection rate too fast/ increase or decreasethe mold temperature/ screw decompression too long/ overall cycle too long/ [con-tamination, fluid]*/ [shot size too large]*/ [drooling]*/ screw decompression is miss-ing from the molding cycle/ gates too small/ fault in the hot-runner system/ nozzleorifice too small/ sprue and runner size too small/ gate not perpendicular to runner.“Splay marks: fine lines”: wet feed/ residual non-aqueous volatiles in feed. “Blush atthe gate, dull spot in the part at the gate”: moist feed/ [melt fracture at the gate]*/[mold too cold]*/ injection pressure too low/ [melt too hot]*/ injection speed toofast/ injection hold time too short/ nozzle diameter too small/ gate land area toolarge/ diameter of sprue, runner and /or nozzle too small/ depth or diameter of coldslug too small/ wrong location of gate/ [venting of mold insufficient]*. “Silverstreaks”: moist feed/ [nozzle or cylinder too hot]*/ plasticizing capacity, in kg/s, ofmachine is exceeded/ variation in temperature of feed in hopper/ plastic tempera-ture is too high/ injection pressure too high/ air trapped between granules in thecold end of the machine/ [mold too cold]*/ injection speed too fast/ lack of or exces-sive external lubrication/ feed is mixture of course and fine particles as withreground/ rear cylinder temperature too high/ [venting of mold insufficient]*/ gatesnot balanced or at wrong location/ insufficient addition of zinc stearate when usingreground/ [air trapped in melt]*/ [degradation of melt, thermally]*/ [air trapped inmold]*/ [cold slugs at the nozzle or hot tip]*/ [contamination]*/ faulty mold designwith too many sharp corners or edges. “Drag marks”: rough surface of mold/ injec-tion pressure too high/ injection hold time too long. “Worn tracks on part”: [melt toocold]*/ [nozzle too cold]*/ screw rpm too low/ injection speed too fast/ backpressuretoo low/ nozzle orifice too small/ gates too small/ cold slug well too small. “Jetting”dull spots and disturbances that look like a jet: moist feed/ [melt too cold]* / [mold toocold]*/ injection rate too fast/ nozzle diameter too small/ depth or diameter of coldslug too small/ diameter of sprue, and runner too small/ wrong location of gate,incorrectly at a thick section / [nozzle too cold]*/ gate too small/ gate land length toolong. “Wave marks”: feedrate too small/ injection pressure too low/ [melt too hot]*/[mold too hot]*/ clamp pressure too low/ injection hold time too short/ cycle timetoo short/ wrong location of water channels/ stock temperature either too hot or toocold/ injection speed too fast or too slow/ nozzle diameter too small/ moist feed.“Flashing”: the flow of material into unwanted areas; if at the end of the flow paths thenits cause is usually [shot size too big]; flash in the runner system may indicate continuedholding pressure after the gates freeze off: [melt too hot]*/ injection pressure too high/injection hold time too long/ injection speed too fast/ clamp pressure too low/[mold too hot]*/ rework the mold design/ vents too deep/ damaged mold/ misa-ligned platen/ wet feed/ [shot size too big]*/ [ feedrate into mold too high]* / erraticfeed/ [design of part faulty]*/ erratic cycle time. “Weld lines, knit lines”: injection rate

114


too small/ injection pressure too low/ injection hold time too short/ [mold toocold]*/ [melt too cold]*/ vent missing in location of weld/ overflow well missingnext to the weld area/ wrong gate location/ too much filler.

Transparency: “Cloudiness or haze for clear plastics”: [contamination]*/ moist feed /[melt too cold]*/ faulty adjustment of barrel temperature profile/ injection pressuretoo low/ backpressure too low/ [mold too cold]*. “Bubbles in clear plastics”: moistfeed/ [melt too hot]*/ injection pressure too low/ injection rate too fast/ injectionhold time too short/ booster time too low/ [mold too cold]*/ mold cooling time tooshort/ [cooled too fast]*.

. Strength or Shape Defects

“Voids”: [short shot]*/ [mold: external surfaces solidify and shrinkage occursinternally]*/ [thick sections continue to shrink after the melt path is frozen]*/ injec-tion rate too fast/ [melt too hot]*/ booster time too short/ molding cooling time tooshort/ [cooled too fast]*/ feed moist/ insufficient blowing agent. “Blisters”: feedmoist/ injection pressure too low/ backpressure too low. “Lamination, peeling”: moistfeed/ [mold too cold]*/ [melt too cold]*/ injection speed too fast/ nozzle diametertoo small/ gate land area too large/ depth or diameter of the cold slug is too small/diameter of sprue, runners and or nozzle too small/ [contamination]*/ backpressuretoo low/ injection made without screw rotation/ screw rpm too low/ [nozzle toocold]* / injection rate too low. “Warpage, part distortion”(usually caused by non-uniformshrinkage as the molded part cools from ejection temperature to room temperature): incor-rect differential in mold temperatures to account for geometry or mold design/incorrect handling after ejection/ injection hold time too short and stopped beforegate freezes/ cooling time too short/ injection pressure too high or too low/ [moldtoo cold]*/ shrink fixtures and jigs to promote uniform cooling are missing/ wronggate locations and too few/ gates too small/ faulty part design/ uneven cooling sys-tem on molds/ injection pressure too high/ [melt too cold]*/ holding pressure andtime too long/ screw not rotating with injection done/ injection speed too fast/ back-pressure too low/ mold temperature either too hot or too low/ time for cooling andmold-open are too short. “Weld weak”: [mold too cold]*/ injection speed too slow/[melt too cold]*/ injection pressure too low/ nozzle opening too small/ gate landarea too large/ sprue, runner or gate size too small. “Brittle”: feed moist/ [mold toocold]*/ injection rate too fast/ [melt too cold]*/ injection pressure too high/ gate di-ameter too small/ nozzle orifice too small/ not enough gates or gates at wrong loca-tion/ injection pressure too low/ gate land area too large/ sprue, runner or gate sizetoo small/ gates too small/ cold slug well too small/ holding pressure and time toolong/ screw rpm too low/ screw should rotate during injection/ injection speed toofast/ backpressure too low/ mold temperature either too high or too low/ sensorerror/ [degradation material]*/ [stress high in part]*/ faulty mold design withnotches causing local stress. “Dimensional variation”: faulty feedrate/ [melt toocold]*/ injection pressure too low/ [mold too cold]*/ injection hold time too short/injection speed too fast/ cycle time too short/ nozzle diameter too small/ faulty gatelocation/ gate land area too large/ diameter of sprue, runner and or nozzle toosmall/ incorrect location of water channels/ [control of machine faulty]*/ [mold con-

115


ditions wrong]*/ poor part design/ moist feed/ irregular particle size/ batch to batchvariation in feed. “Cracking”: [mold too cold]*/ wrong mold design/ ejection pinspoorly located and give unbalanced push/ [shot size too large]*. “Low heat distortiontemperature”: [variation in section thickness]*/ [mold too cold]*/ mismatch betweencylinder and mold temperatures/ feedrate too high/ pressure too high/ plungerdwell too long/ excessive temperature variation between front and back of mold/freezing in the gate because gate orifice too large.

. Operation

“Sticking in cavity”: [mold too hot]*/ [melt too hot]*/ injection pressure too high/injection hold time too long/ injection hold time too short/ gate land area too large/diameter of sprue, runner or gate too small/ mold surface is rough/ injection speedtoo high/ faulty mold design/ incorrect radius of nozzle and sprue bushing/ moldrelease not used/ air was not provided for ejection/ hold pressure too high/ feed notadjusted to provide a constant cushion/ cooling time too long or too short/ cavity orcore temperatures do not have the < 7 �C differential between mold halves/ nozzletoo hot/ mold has undercuts and insufficient drafts. “Sticking parts”: [mold toohot]*/ injection pressure too high/ rough surface on mold/ holding pressure toohigh/ wet feed/ cooling time too short/ faulty design of ejector/ highly polished orchrome-plated mold surfaces. “Sticking in sprue bushing”: injection pressure toohigh/ injection hold time too long/ booster time too long/ cooling time too short/mold is too hot at the sprue bushing/ nozzle pulled back from mold/ [nozzle toocold]*/ incorrect seat between the sprue and mold/ nozzle orifice is not 7.5 mmsmaller in diameter than OD of sprue/ rough surface on sprue/ sprue puller ineffec-tive/ sprue does not have sufficient draft angle for easy release/ screw decompres-sion too low or missing. “Runner breaks”: holding pressure and time too long/ [moldtoo hot]*/ sprue, runners and gates are rough/ incorrect radius in the nozzle andsprue bushing/ time for cooling and open time too short. “Discoloration of sprue”:[melt too hot]*/ nozzle or shutoff valve not tightened/ injection speed too high/ noz-zle orifice diameter too small/ [mold too cold]*/ cold slug well too small/ gate diam-eter too small. “Drooling at nozzle”: shutoff valve dirty or clogged/ injection toosoon/ wrong nozzle pressure/ poor radius of nozzle and sprue bushing/ [nozzle toohot]*. “Screw does not return”: screw rpm too low/ backpressure too high/ wet feed/hopper out of feed/ obstruction/ temperature in the rear zone too high. “Ejection ofpart poor”: rough mold walls/ [shot size too large]*/ knockout system inadequate/insufficient taper. “Cycle erratic”: operator/ [pressure erratic]*/ [ feedrate erratic]*/[cylinder temperature cycles]*. “Cycle too long”: [cooling cycle too long]*/ [heatingcycle too long]*/ [operator issues]*/ material should be more heat resistant.

. Symptoms

[Air trapped in melt]*: screw decompression/ backpressure too low.[Air trapped in mold]*: [venting of mold insufficient]*/ gate diameter too small/

[mold too hot]*.[Backflow from the mold]*: suckback/ faulty non-return valve.[Backpressure too high]*: injection rate too fast.

116


[Barrel too hot]*: melt temperature > 271 �C/ sensor error/ faulty barrel heater con-trol system/ worn or incorrectly fitted screw and barrel configuration.

[Contamination]*: dirty machine/ dirty hopper/ moist feed/ too many volatiles infeed/ [degradation]*/ lubricant or oil on mold/ incorrect mold lubricant/ feed con-taminated during material handling/ faulty raw material from supplier/ poor shut-down procedures.

[Contamination, fluid]*: water or oil leaking into mold cavity.[Control of machine faulty]*: incorrect screw stop action/ inconsistent screw speed/

malfunction of non-return valve/ worn non-return valve/ uneven control of back-pressure/ faulty temperature sensor/ heater band faulty/ control system fault orpoorly tuned/ machine has inadequate plasticizing capacity/ inconsistent control ofcycle.

[Cooling cycle too long]*: [melt too hot]*/ [mold too hot]*/ inadequate cooling inlocal heavy sections.

[Cooling insufficient before removal from mold]*: faulty mold design especially forrib design/ injection speed too slow/ injection hold time too short/ injection pres-sure too low/ melt too hot/ mold too hot/ [venting of mold insufficient]*/ sprue andrunners too small diameter/ gate too small/ gate land length too long/ gate not closeto thicker areas/ core missing from heavy section.

[Cooled too fast]*: [mold too cold]*/ mold cooling time too short.[Cylinder overheated]*: nozzle too hot/ cylinder temperature too high.[Cylinder temperature cycles]*: controller fault/ sensor error/ incorrect line voltage/

power factor problems/ heater bands faulty/ variation in feed temperature.[Degradation, mechanical]*: barrel temperature in the feed area is too low com-

bined with high screw speed or high backpressure/ short transition section inscrew/ radius between the screw root and the flighter is too small/ small tolerancebetween the plunger and the wall/ fine material trapped between the plunger andthe wall/ excessive reground/ rear cylinder temperature too low/ plunger off-center.

[Degradation in the extruder of melt thermally]*: temperature sensor error/ [melt toohot]*/ temperature controller fault/ improperly designed or defective non-returnvalve.

[Design of part faulty]*: incorrect mold dimensions/ unequalized filling rate in cav-ity/ mold not sealing because of flash between surfaces/ [venting of mold insuffi-cient]*/ vents too large/ gate land area too large/ runner, sprue and gate dimensionsincorrect.

[Drooling, introduces solid material into part giving defects]*: wet feed/ [melt toohot]*/ suckback pressure too low/ injection pressure too high/ injection forwardtime too long/ injection boost time too long/ shutoff valve dirty or clogged/ injectiontoo soon/ poor radius of nozzle and sprue bushing/ [nozzle too hot]*.

[Feed rate erratic]*: feeding mechanism/ bridging in hopper/ hopper design fault.[Feed rate into mold too high]*: injection feedrate too fast/ feed setting too high/

sensor error.[Flow of polymer into the cavity uneven during high velocity flow into an open area]*:

injection rate too fast/ faulty gate location/ gate too small.

117


[Granules not melted]*: plastic temperature too low/ cycle too short for cylinder ca-pacity/ nozzle diameter too large.

[Heating cycle too long]*: insufficient heating capacity.[Injection too slow]*: screw rpm too high/ backpressure too high/ injection speed

too slow/ injection pressure too low/ injection forward time too short/ booster timetoo short/ cycle too short.

[Insufficient plastic in mold]*: thick sections, bosses, ribs/ not enough feed/ injec-tion pressure too low/ plunger forward time too short/ unbalanced gates/ pieceejected too hot/ variation in mold open time/ no cushion in front of injection ramwith volumetric feed.

[Melt not homogeneous]*: backpressure too low.[Melt too cold]*: sensor error/ control system error/ lack temperature confirmation

via hand-held pyrometer or laser sensor/ cylinder too cold/ screw rpm too low/ back-pressure too low/ insufficient plasticizing capacity of machine/ [nozzle too cold]*/heating band fault/ excessive flow length in mold.

[Melt too cold at the nozzle or hot tip]*: nozzle too cold/ temperature sensor error/too few heater bands/ heater bands too far from nozzle tip/ hot tip heat source toofar from orifice or faulty/ sharp corners near the gate.

[Melt too hot]*: sensor error/ control system error/ lack temperature confirmationvia hand-held pyrometer or laser sensor/ cylinder too hot/ screw rpm too high/back-pressure too high/ [mold too hot]*/ [nozzle too hot]*/ injection rate too slow/ gatetoo large/ gate land too short/ resin too hot/ holding pressure and time too long/moist feed/ cooling and mold-open time too long/ [residence time too long]*.

[Melt too hot, localized overheating]*: [barrel too hot]*/ faulty barrel heater controlsystem/ [nozzle too hot]*/ sensor error/ faulty or incorrectly designed check valve/worn or incorrectly fitted screw and barrel configuration.

[Melt fracture at the gate]*: [melt too cold]*/ temperature sensor error/ injectionrate too fast/ gate too small/ sharp edge in gate area/ cold slug well in the runnertoo small.

[Mold conditions wrong]*: [mold temperature non-uniform]*/ injection pressurelow/ injection forward time too short/ injection boost time too short/ [barrel toohot]*/ [nozzle too hot]*/ inconsistent control of cycle.

[Mold temperature non-uniform or erratic]*: poorly designed water or coolant lines/[venting of mold insufficient]*/ coolant supply fault.

[Mold: external surfaces solidify and shrinkage occurs internally]*: [mold too cold]*/mold includes sections that are “too thick”/ [melt too cold]*.

[Mold too cold]*: cooling lines in wrong location/ coolant too cold/ coolant flowratetoo high/ sensor error.

[Mold too hot]*: cooling lines in wrong location/ coolant too hot/ coolant flowratetoo low/ sensor error.

[Non-uniform shrinkage as the molded part cools from ejection temperature to roomtemperature]*: wrong packing times/ wrong packing pressures/ wrong gate location/cooling system fault/ temperature sensor fault/ need separate temperature adjust-ment for mold halves.

118


[Nozzle too hot]*: sensor error/ control error/ temperature setpoint at nozzle toohot/ localized heater bands on the nozzle instead of being spread along the nozzle.

[Operator issues]*: slow setup of mold/ need to trim “flashing”/ poor monitoringof cycle times/ excessive machine dead time.

[Premature gate freeze-off ]*: gate size too small.[Pressure too low]*: injection pressure too low/ loss of injection pressure during

the cycle/ feed control set too high causing lower injection pressure.[Pressure too high]*: injection pressure too high/ injection time too long/ boost

time too long.[Pressure erratic]*: sensor error/ control system tuning fault/ leaks in the hydrau-

lics.[Residence time too long]*: machine provides shot size that is too large/ dead spots

in hot manifold/ temperature too high/ poorly designed manifold system/ [contam-ination]*.

[Resin feedrate too low]*: no material in the hopper/ hopper throat partiallyblocked/ feed control set too low/ faulty control of feed system/ bridging in the hop-per/ faulty hopper design.

[Shear heating of melt]*: injection rate too fast/ injection pressure too high/ gatestoo small/ nozzle orifice too small < 0.8 of sprue bushing/ nozzle dirty/ sharp cor-ners/ injection rate too fast/ shutoff nozzle used instead of a general purpose noz-zle/ improperly designed or defective non-return valve.

[Short shot]*: [resin feedrate too low]*/ injection pressure too low/ [mold toocold]*/ injection speed too low/ [melt too cold]*/ injection hold time too short/ cycletime too short/ diameter of gate, sprue, and runner too small/ nozzle orifice toosmall/ gate land length too long/ incorrect gate location/ [venting of mold insuffi-cient]*/ [nozzle too cold]*/ nozzle dirty/ shutoff valve dirty/ inject with screw not-rotating/ machine undersized for the shot required/ cycling from wet to dry resin/excessive flow length in mold/ excessive feed buildup in cylinder/ [mold tempera-ture non-uniform]*/ [air trapped in mold]*/ not enough external lubricant/ poor bal-ance of plastic flow into multiple cavity mold/ holding pressure too low.

[Shot size too large]*: resin feedrate too high/ injection pressure too high/ machineshot size much larger than mold requirement.

[Shrinkage excessive]*: [melt too hot]*.[Solidification at the mold wall delayed]*: [mold too hot]*.[Stress high in part]*: [mold too cold]*/ [melt too cold]*/ injection pressure too

high/ faulty post-mold conditioning/ faulty mold design.[Thick sections continue to shrink after the melt path is frozen]*: faulty mold design,

with too much variation in part cross section/ [premature gate freeze-off ]*.[Venting of mold insufficient]*: injection rate too fast/ booster time too long/ injec-

tion pressure too high/ vents plugged/ not enough vents/ clamp pressure too high/wrong location of gates relative to vents/ [melt too hot]*/ [mold too hot]*.

[Viscosity of melt too high]*: [melt too cold]*/ wrong resin.

119


3.9.6.2 Extruders for PolymersGood practice: for startup, the barrel heaters are critical because screw is not rotat-ing. Major concerns about cold start. Rear-barrel temperature usually remainsimportant because it affects the “bite” or rate of solids conveyed in the feed. Barreltemp. must be set appropriately for polymer. Head and die temperatures= desiredmelt temp (except where want gloss, flow distribution or pressure control).

Screw speed is changed by reducing the motor speed by one of three options:1) 10 to 20 in two stages: either pair of gears or pulley but second stage is alwaysgears with the screw set in the middle of the last big “bull” gear. For very slow-mov-ing extruders (e.g. twins for rigid PVC) there are usually three stages of reduction toget to < 30 rpm. Most extruder drives are constant torque with max power only avail-able at top screw speed with the reduction ratio sometimes mismatched to the job.

For maximum solids conveying “Stick to the barrel and slip on the screw”. Mostplastics normally slip on the root of the screw as long as the feed temperature <melttemperature with those that are most likely to stick being highly plasticized PVC,amorphous PET and certain polyolefin copolymers. For amorphous PET covert thefeed to less-sticky semicrystalline form by heating to high temperature for at leastan hour in an agitated hopper.

Particles must stick to the barrel; trouble occurs with a “slippery feed” such asHDPE and fluoroplastics.

Material is the biggest cost (usually 80%) so reuse as much trim and scrap as pos-sible and keep close thickness tolerances so as not to have excessive thickness.

Shear rate is important because this affects viscosity. All common plastics are“shear thinning” eg. PVC flow is 10� faster if double the push but LLDPE flowincreases 3 to 4 times for double the push.

Single screw: typically operated 100% filled. Usually flood feed.Twin screw: typically operate 20–100% filled. Cannot be flood fed if running at

high speeds.Twin screw with vent: melt seal is about 1 L/D upstream of the vent; feed screw

section under the vent operate < 0.5 full. During startup increase the vacuum gradu-ally. Use low degree of fill.

Coating wire and cable: preheat the wire to about 120 (for HDPE) to 175 �C (cellu-lar PE) to minimize shrinkage.

Trouble shooting: Extruders: the approach usually is to 1) adjust the temperatureprofile, 2) check the hardware such as the thermocouples, controllers, speed, 3) alterthe processing conditions or 4) change the resin or the screw and barrel design. Thesymptom-cause information is presented as issues related to production, off-specthickness or shape, off-spec strength, off-spec surface features, and usual symptoms.

. Production: “Throughput < design”: [low bulk density of feed]*/ wrong screwdesign/ worn screw or barrel elements/ screw rpm too low/ wrong tempera-ture set points/ caking on the feed screw/ caking on the feed port. “Slow andsteady reduction in throughput”: buildup of contaminants on screen pack. “Tor-que > design”: feedrate too high/ [degree of fill too high]*/ screw speed too

120


low/ heat zone set points too low/faulty screw design.”Machine stalls above acertain speed or with certain materials”: constant-torque drive (magneticclutch)/ AC-DC drive system with constant-drive and constant torque combi-nation. “Feed from hopper not feeding smoothly”: material too light and fluffyfor gravity feed/ material damp/ bridging/ screw channels in the feed zoneare not deep enough/ too much external lubricant. “Drive amps > design”: poly-mer viscosity too high/ screw pumping too high/ screw speed too high/ bar-rel temperatures set too low. “Amps high for melt pump drive or pump won’trotate”: shear pin/ degraded polymer caught in gears. “Variation in driveamps”: [solids conveying instabilities]*/% regrind too high/ feed bulk densitywrong. “Cycling motor amps”: [surging]*. “Extruder noisy”: loss of feed/ foreignor metal contaminant in feed/ bent screw/ bent barrel/ 1�2 heater burnedout. “Local temperature fluctuations with cycles < 5 min”: instrument circuitryfault/ inconsistent melt/ poor heater contact/ thermocouples poorly seated/sensor error/ poor sensor location/heating element fault/ controller fault/[solid conveying instabilities]*. “Barrel temperatures differ from the set tempera-tures”: controller fault/ burnt-out heater/ blue screw syndrome where the rearend bites off more than the front end can pump. “Real wall temperature > theset point”: the rear end bites off more than the front end can pump. “Varia-tions in melt pressure”: drift or cycle time variation > 1 min: [low feeding effi-ciency]*/ low friction characteristics/ [low bulk density of feed]*/[melting toosoon]*/ adequate early barrel pressure but [melting unstable]*/first barrelheating too high/screw tip pressure too low. “Screw tip pressure too low”: noresin in feed hopper/ bridging in feed hopper/ temperature too high in extru-der entrance zone/ polymer wrapped around screw. Related topic [Screw tippressure too high]*. “Unstable melt pressure”: screw speed too high/ degree offill too low/ screw design gives mixing of melt inadequate or low shear/cycling control on heat zones/ feeder problems. “Unstable pumping”: forvented extruders: first stage [surge]*/ poor screw balance between stages.“Material flow out vent”: for vented extruders: poor vent-diverter design/ firststage pumping rate too fast/ screw tip pressure too high/ flowrate > design/temperature set points for last barrels too low or heaters faulty/ screw designgives localized pressure under the vent/vacuum too high/ if adding liquids,then poor mixing.

. Product thickness or shape does not meet specifications. “Small size varia-tion”: variation in drive speed/ wrong screw design/ variation in puller. “Largesize variation”: [surging]*. Here are more specific details: “Variation in thick-ness in transverse direction and always in the same place”: see “Variation in localtemperature”. “Variation in thickness in transverse direction and floating acrossor around the product”: [mixing of melt inadequate]*/ die temperature settingwrong/ dirty die/[screw tip pressure too low]*/backpressure on extruder< 20MPa/ wrong design or die or screw/ [degradation of melt in extruder]*/for blown film: air ring not centered or level/ thermocouple error/ bubblesubjected to hot or cold air/ polymer feed has > 6 �C variation in melt temper-ature. “Waviness or ridges around the circumference”: [surging]*/ nonuniform

121


water cascade/ uneven take-off speed/ vibration in the take-off equipment.“Variation in thickness in the direction of extrusion”: [surging]*/ puller slip orincorrect control of tension/ drawdown too much or too fast/ poor align-ment/variation in take-up reels/ erratic variation in feed materials/ hot-lipscontroller cycling/ untuned controller/ faulty controller/ temperature varia-tion in die/ variation in motor load/ variation in melt pressure/ damaged ori-fices in die or feedblock/ holes in die too large/ incorrect barrel temperatureprofile/ faulty adjustment of die/ faulty screw design/ plugged screen pack/temperature sensor fault in barrel/ hopper bridging/ throughput too high/gels/ for blown film: inconsistent nip roll speed control/ frost line too low/polymer melt temperatures too low/ bubble-cooling control fault/ variationin air flow from blower/ cooling air nonuniform/ gap opening too large/cooling air flowrate too low. “Periodic variation in thickness in direction of extru-sion”: spinneret temperature too low/ orifice wrong diameter/ wrong draw-down ratio. “Cyclical variation in thickness in the direction of extrusion”: [drawresistance instability]*.

“Filament breaks”: [surging]*/ some die holes blocked/ temperature variation inhead or die/ melt temperature too low or too hot/ drawdown too great/ holes in dietoo large/ moisture/ [contamination]*/ [melt too hot]*/ gap between bath and dietoo large. “Wrong filament shape: correct cross section but too large”: too little pull/ drawdistance from die to take-off is too short/ take-off speed too slow/ die-land length tooshort/ melt temperature too low. “Wrong filament shape: distorted cross section”:unequal die temperatures/ die incorrect shape. “Wrong filament shape: size OK butwarped”: cooling too intense / linear take-off speed too fast. “Filament oval cross sec-tion”: filaments too hot while passing over rolls/ rolls too hot/ die holes oval/ temper-ature gradients in die/ tension too high in take-up rolls.

“Holes in blown film or coating”: moisture in resin/ die lip gap too large/ air gaptoo small/ vacuum too high; for coating: moisture/ substrate too rough/ coatingthickness too thin/ contamination/ decomposition/ compound temperature toohigh/ see also “Gels”.

. Product does not meet strength specifications. “Product strength < specs for allsamples”: faults with the feed/ [degradation of melt in extruder]*. “Productstrength < specs for some samples”: [contamination]*/faults with the feed/ [deg-radation of melt in extruder]*”. Pipe strength < specs”: melt temperature toolow/ throughput too fast/land length too short/ air gap too short/ excessivedrawdown at cold temperatures/ too much scrap in feed/ moisture in resin/dirty metal surfaces/ material sticking on extruder parts/ short die-landlength/ high internal angular discontinuities into the die-land section/ linearextrusion speeds excessive/ uneven water coolant cascade/ misalignedsleeve/[mixing of melt inadequate]*. “Stiffness < design”: for cast or sheet: chillroll temperature too low/ low resin density. For coatings: “Poor adhesion”: avariety of apparently contrary causes related to polymer viscosity [polymer vis-cosity too high]*, degradation especially [oxidation]*, tackiness, temperature:

122


melt temperature too low or high/ air gap too small/ chill roll temperaturetoo low or hot/ line speed too fast/ poor match between coating and sub-strate/ substrate problems. “Low tenacity”: ratio of roll speeds too small/ [deg-radation of melt in extruder]*/ wrong resin/ nicks in die. “For wire and cable:covering separates from wire/adhesion”: wire not preheated hot enough/ melttemperature too low/ dirty or moist wire/ [degradation of melt in extruder]*/cooled too fast/ air-cooling gap too short/ air trapped between wire and coat-ing [trapped air]*. “Low modulus of elasticity”: melt temperature too low/ airgap distance too short. “Interfacial instability for coextruded film”: excessiveshear stress at the die gap > 0.06 MPa/ throughput too high/ die gap too nar-row, melt temperature too low/ polymer viscosity too high/ the relative velo-cities where polymer flows combine differ by > 4: 1.

. Appearance: gloss, fisheyes, shark skin, pits, holes, clarity. Some are surfaceeffects, such as shark skin, regular, ridged, surface deformity with ridges per-pendicular to extrusion direction. Others are defects of the whole body ofextrudate, caused by [melt fracture]*. These include spiral, bamboo, regularripple. “Rough surface or dullness”: [contamination]*/ moisture/ linear speedtoo fast or screw speed too fast/ die holes too small/ die temperature too low/die land too short/ [mixing of melt inadequate]*/ [melt fracture]*/ no ventsused/ hopper vacuum inadequate/ [screw tip pressure too low]*/ discontinu-ity in the melt flowlines / low melt temperature/ dirty metal surfaces/ mate-rial sticking on extruder parts/ uneven water coolant cascade/ misalignedsleeve/faulty screw design/ screw too hot/ extrudate too hot in the coolantbath causing boiling. “Pits on surface”: [contamination]*/ moisture/ watersprays onto extrudate just after exiting the die/ water bath too hot. “Fisheyesin film”: moisture/ damp polymer/ too many volatiles in polymer. “Gels”: [con-tamination]*/[degradation of melt in extruder]*/ [shear intensity too low]*/[screw tip pressure too low]*/ number and density of the screen pack toolow/ moisture too high/ screw speed too low/ incompatible blend/ [residencetime too long]*/ lack of streamlines in extruder/ incorrect startup proce-dures/ [melting inadequate]*/ [melt too hot]*/ for reactive: localized initiatorconcentration too high. “Shark skin”: [melt fracture]*/ die temperature toolow at the land end/ linear extrusion speed too high/ throughput too high/viscosity of polymer too high/ MWD of polymer too narrow/ lubricant addi-tive missing/ [shear intensity too high]*/ die gap too small.

“Polymer buildup on die”: melt temperature too low/ throughput too high/ die gaptoo small/ wrong screw design/ low level of antioxidants.

“Porous or bubbles in product”: poor melt quality at vent/ plugged vent opening/insufficient vent vacuum/ excessive volatiles in feed/ screw speed too high/ vacuumvent needed. “Spotted, warped or pocked surface”: [mixing of melt inadequate]*/ mois-ture/ roll too cold/ contamination/ screen size too large/ dirty die/ [trapped air]*/dirt on rolls/ drafty air/ wrong tension/ boiling on extrudate in cooling bath. “Lineson the product”: surface scratches on tip or die/ local buildup/ [die swell too high]*/throughput too high/ polymer adhesion on channels, tip or die/ incorrect contact in

123


the quench tank/ melt temperature too low/ throughput too fast/land length tooshort. “Indented pock marks on pipe after water cooling”: coolant water spray velocitytoo high. “Raised pock marks on pipe”: water drops on surface in the air-drying zone.“Discolored material”: temperatures too high/ wrong formulation/ discontinuities in-side extruder.

. Symptoms

[Contamination]*: contaminated feed/ contaminated additives/ dirty die/ polymeron die lips.

[Degradation of melt in extruder]*: [RTD too wide]*/ barrel temperature too high/screw speed too high (causing overheating and shear damage)/ oxygen present/[oxi-dation]*/ nitrogen purge ineffective/ wrong stabilizer/ wrong screw/ flows notstreamlined/ stagnation areas present/ extruder stopped when tempera-tures > 200 �C/ copolymer not purged with homopolymer before shutdown/[resi-dence time too long]*.

[Degree of fill too high]*: feedrate too high/ screw speed too slow.[Die swell too high]*: tip too short/abrupt change in flow near tip or die/ melt tem-

peratures too low in die assembly.[Draw resistance instability]*: for blown film, fiber spinning, blow molding: draw

ratio too high.[Extrusion instabilities]*: screw speed too high/ screw temperature too high/ barrel

temperature at delivery end too high/ channel depth too high in the metering sec-tion/ the length of the compression section too short/ read barrel end temperaturestoo low/ diehead pressure is too low.

[Feedrate too high]*: screw speed too fast/ feed from hopper too fast.[Gels]*: [contamination]*/[degradation of melt in extruder]*/ [shear intensity too

low]*/[screw tip pressure too low]*/ number and density of the screen pack too low/moisture too high/ screw speed too low/ incompatible blend/ [residence time toolong]*/ lack of streamlines in extruder/ incorrect startup procedures/ [meltinginadequate]*/ [melt too hot]*/ for reactive: localized initiator concentration too high.

[Low bulk density of feed]*:% regrind too high/ grind too coarse.[Low feeding efficiency]*: low friction characteristics.[“Melt fracture” where the critical shear stress of polymer (about 0.1 to 0.4 MPa)

exceeded in the die; excessive shear stress at the wall > 0.1 MPa]*: exit speed at the die istoo fast/ melt too cold/ throughput excessive/ die land too short/ die opening toosmall/ entrance to die not sufficiently streamlined/ screw speed too high/ MM andmelt viscosity too high/ cross-sectional area in exit flow channel too small/ externallubricant additive missing.

[Melt too hot]*: screw speed too high/exit barrel zone temperatures too high/degree of fill too low/ [shear intensity too high]*/ heat-zone temperatures set toohigh/ [screw tip pressure too high]*.

[Melting inadequate]*: barrel or die temperature too low/ screw speed too fast/screw design gives insufficient mixing/ [shear intensity too low]*/ feedrate too high/material too “slippery”/[degree of fill too high]*/ additional component has too low amelting point/[residence time too short]*.

124


[Melting too soon]*: wrong bulk density of feed.[Melting unstable]*: especially for screws with high compression ratio and short

compression length: insufficient melt capacity/ too large a channel depth in themetering section/ temperature in the metering end of the screw too high/ wrongscrew design.

[Mixing of melt inadequate]*: [screw tip pressure too low]*/ [ feedrate too high]*/screw speed too high and [residence time too short]*/ screw speed to low and [shearintensity too low]*/ [degree of fill too high]*/ faulty screw design for mixing/ tem-perature set points incorrect/ instrument error in temperature sensors/ tempera-tures too high/ loading excessive for one component/ no static mixer included/ forreactive extrusion: liquid flowrate too high/ screw channel under injection not fullof polymer.

[Oxidation]*: temperature too high/ screw speed too low/ [residence time toolong]*/ oxygen present/ nitrogen purge ineffective/ antioxident stabilizer ineffec-tive/ [trapped.air]*/ hopper vacuum inadequate.

[Polymer viscosity too high]*: temperature too low/ wrong blend/ [shear intensitytoo low]*.

[Residence time too short]*: screw speed too high/ too much feed/ [degree of fill toohigh]*/ poor screw design.

[Residence time distribution, RTD, too wide]*: [degree of fill too low]*/ feedrate toosmall/ screws speed too fast.

[Screw tip pressure too high]*: screens plugged/ die or adapter or breaker plates toorestrictive and give too much Dp/ [polymer viscosity too high]*/ temperatures in dieassembly too low/ barrel temperature too low/ screw speed too high/ [shear inten-sity too low]*/ lubricant needed/ flow restriction/ throughput too high/ die land tooshort/ cold start/ [degradation of melt in extruder]*.

[Shear intensity too low]*: screw speed too low/ faulty screw design.[Solids conveying instability]*: feed hopper fault/ internal deformation of the solid

bed in the screw channel/ insufficient friction against the barrel surface.[Surging]*: 30–90 s: feed particles are not sufficiently softened (usually at the

beginning of the second compression zone) / too rapid compression screws/ [lowfeeding efficiency]*/ low friction characteristics/ low bulk density of feed/[meltingtoo soon]*/ adequate early barrel pressure but [melting unstable]*/ first barrel heat-ing too high/ screw speed too fast/ faulty screw design/ additional compound slip-pery/ bridging in resin feed hopper/ feed-zone temperature too high/ [screw tippressure too low]*/ compound temperature too high/screw too high/ nonuniformtake-off speed/ take-off speed too high/ throughput too fast/ controller fault/ feedresin not mixed well/ melt temperature too low.

[Trapped air in extruder]*: unvented extruder/ wrong screw design/ pressure toolow/ rear-barrel temperature too high/ screw speed too high/ vacuum too low infeed hopper/ powder feed instead of pellets.

125


3.9.7Coating

Trouble shooting: “Poor adhesion”: a variety of apparently contrary causes related topolymer viscosity, degradation, oxidation, tackiness, temperature: melt temperaturetoo low or high/ air gap too small/ chill roll temperature too low or hot/ line speedtoo fast/ poor match between coating and substrate. “Rough wavy surface (apple-sauce)”: wrong resin/ temperature too low or high. “Edge tear”: draw ratio too high/die end temperature too low/ temperature too low or high. “Oxidation”: temperaturetoo high/ screw speed too low/ flows not streamlined/ extruder stopped when tem-peratures > 200 �C/ copolymer not purged with homopolymer before shutdown.“Pinholes in coating”: substrate too rough/ coating thickness too thin. “Surging”:bridging in resin feed hopper/ feed-zone temperature too high/ wrong screw design.“Voids”: moisture/ leaks in resin handling system/ inadequate drying and storage/[thermal degradation]*/ [gels]*. “Die lines”: nicks in die/ dirty lips/ particles in die.

“Pin holes and breaks”: coating too thin/ contamination/ decomposition/ com-pound temperature too high/ moisture.

“Web tears”: compound temperature too low/ too much drawdown/ die lip open-ing too large. “Poor adhesion”: compound temperature too low/ substrate problems.“Excessive neck-in”: die-to-roll gap too large/ material temperature too high/ die-lipopening too large/ throughput too low/ use resin with lower Melt Index/die-landlength too long.

See Symptoms for Section 3.9.6 for the following [Contamination]*; [Degradationof melt in extruder]*; [Screw tip pressure too low]*; and [Shear intensity too low]*.

[Gels]*: [contamination]*/[degradation of melt in extruder]*/ [shear intensity toolow]*/[screw tip pressure too low]*/ number and density of the screen pack too low/moisture too high/ screw speed too low/ incompatible blend/ [residence time toolong]*/ lack of streamlines in extruder/ incorrect startup procedures/ [meltinginadequate]*/ [melt too hot]*/ for reactive: localized initiator concentration too high.

[Surging]*: screw speed too high/ take-off speed too high/ backpressure too low/compound temperature too high.

[Thermal degradation/crosslinking]*: polymer temperature too high/ screw speedtoo low.

3.10Vessels, Bins, Hoppers and Storage Tanks

Bins and hoppers: In general cohesive strength of powders increases with consolida-tion pressure. Trouble shooting: “No flow”: [arching]*/ [rat holing]*. “Erratic flow”:obstructions alternating between arching and ratholing. “Flooding or flushing” when arathole collapses it entrains air, becomes fluidized and the material floods through the out-let uncontrollably: “ fine powders such as pigments, additives and precipitates thattend to rathole/ insufficient residence time in hopper for deaeration. “Flow-rate lim-itation”: fine particles where movement of the interstitial air causes an adverse Dp.

126

3.11 “Systems” Thinking

“Limited live capacity”: [ratholing]*. “Product degradation”: [ratholing]*. “Incomplete ornon-uniform processing”: [ratholing]*.

[Arching]*: particle diameter large compared to outlet/ cohesive particles probablycaused by moisture or compaction.

[Ratholing]*: cohesive particles probably caused by increased moisture or by com-paction (fine powders < 100 mm such as pigments, additives and precipitates).

3.11“Systems” Thinking

Processes are complex systems in which the performance of one piece of equipmentinteracts with and influences another. Typically when we trouble shoot we tend tofocus on individual pieces of equipment and forget that perhaps something away upstream or even downstream could be causing the local difficulty.

To some extent we are already familiar with systems and system interactions whenwe consider a distillation column. This is not one piece of equipment. Rather it is acomplex collection of a column with a series of trays linked by vapor and liquid flowwith condenser, reboiler, pumps, controllers all interacting. Thus, troubles on thecondenser are reflected in the performance of the reboiler and everything in be-tween. We are skilled at being able to make those connections for a distillation “sys-tem”. However, when we have a crystallizer, screen, centrifuge, dryer combinationwe may not be able to easily visualize how the performance of one can dramaticallyaffect the others.

In “systems thinking” the focus can be on how performance from one unit istransmitted to units elsewhere in the system. Since equipment is linked by pipesand conveyors carrying fluids and solids, the interaction is through the temperature,flowrates, pressures, compositions and cycling in the streams. Another perspectiveto “systems thinking” is startup and how the system can move from cold, air-filledconditions to the on-line temperatures, pressures and compositions. A third per-spective is how the environment system affects the process system.

A useful approach it to reflect on the principle that anything that is created insidethe system must go somewhere; where does it go and how does it get out of thesystem? For example, mass is conserved.

1. When you start a plant, all the equipment is filled with air; where does the airgo?

2. If a dissolved trace metal gets into the system, where does it go?3. If a particle breaks down where do the particles go?4. If recycle is used then any trapped species will build up unless there is a

bleed or purge.

In trouble shooting “systems”, some common issues include: solvent lossessomewhere in the system; fouling; foaming or stable emulsion formation that causeequipment malfunction and carryover; corrosion; recycle causing a buildup of spe-cies that may not be removed from the system without adequate bleeds or blow-

127


down. Although many of these are considered for specific pieces of equipment, weinclude a generic consideration of some of these here. In this listing, the concept orsymptom is shown in parentheses and italics, for example, “Foaming”, followed bypossible causes separated by /. If the cause is not a root cause, then it is representedin square bracket plus an *, [ foam-promoting systems]*. These intermediate causesare then listed alphabetically .

. “Fouling”: velocity too slow/ [particulate fouling]* for example, rust, corrosionproducts from upstream, scale from upstream units, oil, grease, mud or silt/[precipitation fouling]* for example sodium sulfate, calcium sulfate, lignin /[biological fouling]* species present such as algae and fungi/ [chemical reac-tion fouling]*, example coke formation and polymerization fouling/ [ floccu-lation fouling]* or destabilization of colloids, for example asphaltenes orwaxes from hydrocarbons/ corrosion products for this unit, see Section 3.1.2/[solidification fouling]* or incrustation such as the freezing on a solid layeron the surface or crystallization/ [condensation fouling]* such as vaporizationof sulfur.

[Biological fouling]*: temperature, pH and nutrients promote growth of algae andfungi/ biomaterials present.

[Chemical reaction fouling]*: polymerizable species in the feed/ high temperaturecausing cracking/ high wall temperatures/ stagnant regions near the wall or velocitytoo slow< 1 m/s/ reactant droplets preferentially wet the solid surface/ addition of“fouling suppressant” insufficient, for PVC polymerization oxalic acid or its salt orammonium or alkali metal borate/ pH change.

[Condensation fouling]*: wall temperature too low/ contamination in the vapor.[Flocculation fouling]*: pH at the zpc/ low concentration of electrolyte/ increase in

humic acid concentration in water in the fall and spring/ colloids present.[Particulate fouling]*: filter not working or not present/ contaminant in feed/

upset upstream/ erosion/ increase in silt and clays in water in the spring.[Precipitation fouling giving scale or sludge]*: soluble species present in feed/ tem-

perature high for invertly soluble/ temperature too low for incrustation or crystalformation.

[Solidification fouling]*: wall temperature too low/ missing insulation/ cold spotson wall/ sublimation.

. “Foaming”: [ foam-promoting systems]*/ [ foam-promoting contaminants]*/[gas velocity too high]*/ [liquid residence time too low in GL separator]*/antifoam addition faulty/ faulty mechanical foam breaker/ [liquid environ-ment wrong]*.

[Foam promoting contaminants: soluble]*: naturally occurring or synthetic poly-mers/ naturally occurring or synthetic organics >C10; example lube oils, asphal-tenes/ naturally occurring or synthetic surfactants; for amine systems: the surfaceactive contaminants include condensed hydrocarbons, organic acids, water contami-nants, amine degradation products/ faulty cleaning before startup; surfactants leftin vessels.

128

3.11 “Systems” Thinking

[Foam promoting contaminants: solid]*: [corrosion products, see Section 3.1.2]*; foramine systems: iron sulfides/ faulty cleanup before startup; rust left in vessel/ dust/particulates.

[Foam promoting systems]*: those that foam naturally: methyl ethyl ketone, aerobicfermentation, textile dyeing foam more readily than amine and glycol absorptionsystems and latex strippping > amine, glycol and Sulfolane strippers > slightly foampromoting: fluorine systems such as freon, BF3/ systems operating close to thecritical temperature and pressure/ surface tension positive system/ [Marangonieffects]*.

[Gas velocity too high]*: temperature too high/design error/ [ foaming]*/ vessel di-ameter too small for gas flow/ column pressure < design/ trays or packing damagedor plugged giving excessive vapor velocity/ temperature too high/ upstream flashseparator passing liquids: feed contaminated with excessive volatile species/ strip-ping gas fed to column too high/ input stripping gas flowmeter error/ design error.

[Inaccurate sensing of the interface]*: instrument fault/ plugged sight glass.[Liquid environment wrong]*: pH far from the zpc/ electrolyte concentration too

low.[Liquid residence time too low in gas liquid separator]*: interface height decreases/

[inaccurate sensing of interface]*/ turbulence in the liquid phase/ flowrate > ex-pected/ sludge settles and reduces effective height of phase/ inlet conditions faulty.

. “Corrosion”: see Section 3.1.2.

. “Stable emulsion formation”: contamination by naturally occurring or syntheticsurfactants: example, lubricating oils/ contamination by particulates: exam-ple, products of [corrosion, see Section 3.1.2]*, amphoteric precipitates of alu-minum or iron/ pH far from the zpc/ contamination by polymers/ tempera-ture change/ decrease in electrolyte concentration/ the dispersed phase doesnot preferentially wet the materials of construction/ coalescence–promotermalfunctioning/ improper cleaning during shutdown/ [rag buildup]*.

[Marangoni effects]*: non-equilibrated phases/ local mass transfer leads to localchanges in surface tension and hence stable interfacial movement

[Rag buildup]*: collection of material at the interface: naturally occurring or syn-thetic surfactants: example, lubricating oils/ particulates: example, products of [cor-rosion, see Section 3.1.2]*, amphoteric precipitates of aluminum/ naturally occur-ring or synthetic polymers.

. Other possible causes of trouble for “systems” include:

“Bumping resulting in plate damage”: trace amounts of water.“Catalyst contamination”: trace amounts of water.“Conversion less than expected”: temperature spike causing catalyst decomposition/

temperature sensor reading low/ [catalyst contamination]*.[Cycling]*: two vessels in series on level control/batch processes in sequence but

out of synchronization/ no intermediate storage.“Distillation overhead off spec”: excessive inerts from upstream/buildup of trace/

purge not sufficient from recycle.

129


[High Dp across bed of catalyst/particles/resin]*: temperature spike/ temperaturesensor reading low.

“Leaks”: temperature spike/ pressure spike/ temperature sensor reading low/glands not tight enough around rotating shafts/ valve stem leaks/ thermal expan-sion of the different metals and parts not correctly accounted for, example, flangesand gaskets on high-temperature heat exchangers.

“Particle agglomeration in pneumatic conveying”: increased humidity in air/ tracecontaminants in air.

“Separation performance of column decreases”: trace amounts of water/ traceamounts of water trapped in column/ [bumping resulting in plate damage]*.

“Temperature runaway in reactor”: operator error: overcharge reactor/ trace water/[poor temperature control]*.

3.12Health, Fire and Stability

In general, we should all be aware of the hazards before we encounter trouble onthe plant. Our response should be instantaneous in identifying the degree to whicha trouble on the plant or process poses a hazard.

3.12.1Individual Species

Many indicators are available to guide us about the hazard posed by individual spe-cies. The hazards are usually considered to be health hazard (toxic or causing death;causing genetic defects; causing long term disability; causing specific illnesses suchas asthma, cancer; impacting on the environment); flammable hazard and explosivehazard.

A simple guideline is the NFPA ratings. These are a scale 0 to 4 with the highernumber indicating the extreme in hazard (see Woods, 1994, Data for ProcessDesign and Engineering Practice, Prentice Hall). MSDS documentation is availablefor most chemicals. Some sources include http://www.msdssearch.net orhttp://www.msdsxchange.com although these direct you to manufacturer’s MSDSinformation. The quality of the documentation varies from company to company.Also consult http://toxnet.nlm.nih.gov.

Health

Vinyl species, benzene ring compounds. Carcinogenic to humans: asbestos, ben-zene, benzidine, b-naphthylamine, vinyl chloride. Proven carcinogenic in animaltrials: acrylonitrile, butadience, N-nitrosamines

Flammability

Species with low Ignition temperatures, say in the range 85–100 �C. Recall that theignition temperature for combustion in air could be lower if in the presence of pure

130

3.12 Health, Fire and Stability

oxygen or chlorine. For example, for toluene the ignition temperature in air of535 �C is reduced in chlorine to 175 �C.

Stability/explosiveness

Guidelines from structure: azide, perchlorates, nitro compounds, peroxides, vinylspecies.

For reactivity or potential for explosion or thermal runaway: if heat of decomposi-tion > 0.2–0.3 MJ/kg; those > 0.5 MJ/kg may be explosive and those > 0.8 MJ/kg maybe detonable.

Hazardous reactants and types of reactions

Rosenmund reduction, oxidation of nitrous acids; oxidation of low molar mass per-acids; alkylation of alkali acetylides, alkylation via Arndt–Eistert; alkylation of dia-zoalkane and aldehyde; alkylation of diazoalkane; condensation of carbon disulfidewith aminoacetamide; esterification of carboxylic acid and diazomethane; esterifica-tion of acetylene and carboxylic acid-vinyl ester; reactions involving concentrated per-oxides and peracids.

Heats of reaction

Extremely exothermic: direct oxidation of hydrocarbons with air; chlorinations, poly-merizations of polyethylene without diluent. > 3 MJ/kg (or > 150 MJ/kmol: examples:combustion –900 MJ/kmol; hydrogenation of nitroaromatics, –560 MJ/kmol; nitro-decomposition, –400 MJ/kmol; diazo-decomposition, –140 MJ/kmol).

Strongly exothermic: nitrations, polymerization of propylene, styrene butadience.1.2–3 MJ/kg (nitration, –130 MJ/lmol; amination –120 MJ/kmol; sulfonation or neu-tralization with sulfuric acid, –105 MJ/kmol; epoxidation, –96 MJ/kmol; diazotiza-tion).

Moderately exothermic: condensations or polymerization reactions of species withmolar mass 60–200, 0.6–1.2 MJ/kg.

Exothermic heat of reaction gives an adiabatic temperature rise > 100–200 �C.Impact sensitivity of < 60 J for solids and < 10 J for liquids.For powders: explosive usually if diameter is < 200 mm with highest probability of

dust explosion if diameter < 65 mm. Note upper and lower concentration for explo-sive mixture.

3.12.2Combinations

With water, with air with other chemicals:Chemicals that react violently with water include: acetyl chloride, aluminum

alkyds, aluminum alkyl halides, aluminum chloride, calcium hydride, calcium ox-ide, ethyl aluminum dichloride, fluorine, lithium hydride, phosphorous pentoxide,phosphorous trichloride, potassium, silicon tetrachloride, sodium, sulfur trioxide,sulfuric acid, thionyl chloride, titanium tetrachloride, zinc alkyls. Moderately: aceticanhydride, activated alumina, aluminum phosphide, calcium, calcium carbide, cal-

131


cium phosphide, lithium, activated molecular sieves, potassium hydroxide, activatedsilica, sodium hydroxide, sodium peroxide.

132

133

In this chapter we meet five engineers as they trouble shoot the five cases intro-duced in Chapter 1. While you are reading these cases focus on the trouble-shootingprocess used. Compare the approaches taken here with the approaches you took inaddressing the problems posed at the end of Chapter 1. The engineers have a rangeof practical experience. In Case ’3, Michelle has three years experience. In Case’4, Pierre has 12 years experience. For Case ’5, Dave has relatively no practicalexperience. For Case ’6, Saadia has 10 years experience. For Case ’7, Frank has 25years of experience. These cases have been carefully selected to provide a range ofapproaches, degrees of difficulty and to illustrate the variety of trouble shooting ap-proaches taken.

Each of the five scripts consists of about three parts with each part concludingwith a few questions for you to consider. This reflective break was introduced to giveyou a chance to reflect on how you would have handled the case, and to decide whatyou should do next. I recommend that as you read each script, that you play thegame. At the end of each case an assessment is given of the problem-solving pro-cesses used by each of the trouble shooters. Their scores range from F to A+; therichness is in the detail of the feedback. Not everyone will follow Lieberman’s(1985), Gans’s (1983), Kister’s (1979) or my style in trouble shooting. The key is toidentify your style and develop confidence in using it.

The cases do not have to be addressed in any order. Select what you would prefer.Enjoy!

4.1Case ’3: The Case of the Cycling Column

Michelle1) graduated three years ago. She is on her journey to become an excellenttrouble shooter but she lacks much practical experience. “Cycling column? Called inas chief trouble shooter!” exclaimed Michelle. Well, I’ll do my best, she thought.Michelle recalled the Trouble-Shooter’s Worksheet. She carefully noted the key infor-

4

Trouble Shooting in Action: Examples


1) In all these cases, the names of the engineershave been changed.

4 Trouble Shooting in Action: Examples

mation: startup after maintenance, iC4, and cycling level in the bottom of a distilla-tion column. From the diagram, it looks like a thermosyphon reboiler and aninverted bucket steam trap. Well, I want to and I can! I have an organized strategy toapply.

What are the strengths and weaknesses of Michelle’s approach so far?What would you have done differently?Would you head out to the plant directly or are some quick checks to be done from yourfiles first?

Let’s check, thought Michelle. This isn’t an emergency. Until I learn more I can-not think of a “safe-park” condition I should impose. The key information is thatthese symptoms are occurring just after shutdown so I really need to find out whatwas changed and what was worked on. She cautioned that she should change herthinking from “cycling level in the bottom of the column” “to apparent cycling in thelevel”. Michelle quickly scanned the P&IDs for the iC4 unit and confirmed that thediagrams showed a thermosyphon reboiler and an inverted bucket trap with abypass. She checked in Chapter 3 for the cycling and reboilers and then checkedsteam traps:

Reboilers in general:

“Cycling (30 s – several minutes duration) steam flow, cycling pressure on the process sideand, for columns, cycling Dp and cycling level in bottoms”: instrument fault/ controllerfault/ condensate in instrument sensing lines/ surging/ foaming in kettle and ther-mosyphon/ liquid maldistribution/ steam-trap problems, see Section 3.5, with ori-fice Dp across trap < design/ temperature sensor at the feed zone in a distillationcolumn/ collapsed tray in a distillation column.

For thermosyphon:“Cycling (30 s – several minutes duration) steam flow, cycling pressure on the process

side and, for columns, cycling Dp and cycling level in bottoms”: in addition to general, allnatural circulation systems are prone to surging/ feed contains high w/w% of highboilers/ vaporization-induced fouling/ constriction in the vapor line to the distilla-tion column. For horizontal thermosyphon: maldistribution of fluid temperatureand liquid.

Section 3.5 steam traps:

. Good practice: Install a demister.

Steam traps: install trap below condensate exit (or with a water seal if the trap iselevated), use a strainer before most traps, use a check valve for bucket traps. Slantpipes to the trap. Use a downstream check valve for each trap discharging to a com-mon header. Pipe diameter 3 trap inlet pipe diameter. Prefer to install auxiliary trapin parallel instead of a bypass. Do not group trap thermodynamic traps because oftheir sensitivity to downstream conditions.

Float and thermostatic: usually discharges continuously, low pitched bubblingnoise. High-pitch noise suggests live steam is blowing.

Balanced thermostatic: leave about 0.6 m of uninsulated pipe upstream of trap.

134

4.1 Case’3: The Case of the Cycling Column

Inverted bucket: use initial prime to prevent steam blowing.Thermodynamic: about 6 cycles/minuteTrouble shooting: the major faults are wrong trap, dirt, steam locking in the trap,

group trapping, air binding and water hammer. Too large a trap gives sluggishresponse and wastes steam. Too small a trap gives poor drainage, backup of conden-sate. “Cold trap + no condensate discharge”; steam pressure too high/ no water orsteam to the trap/ plugged line or strainer/ orifice enlarged by erosion (bucket trap)/incorrect Dp across the orifice (inverted bucket)/ bucket vent clogged (invertedbucket)/ current operating pressure > design/ trap clogged. “Hot trap + no condensatedischarge”: bypass open or leaking/ trap installed at high elevation/ broken syphon/vacuum in heater coils. “Live steam blowing”: bypass open or leaking/ worn trap com-ponents/ scale in orifice/ valve fails to seat/ trap lost prime (inverted bucket)/ sud-den drops in pressure/ backpressure too high (thermodynamic). “Continuous dis-charge when it should be discontinuous”: trap too small/ dirt in trap/ high-pressuretrap installed incorrectly for low pressure service (bucket trap)/ valve seat cloggedwith dirt/ excessive water in the steam/ bellow overstressed (thermostatic). “OKwhen discharging to the atmosphere but not when to a backpressure condensate header”:condensate line diameter too small/ wrong orifice/ interaction with other traps con-nected to a common header/ condensate line partially plugged/ [backpressure toohigh]*. “Slow and uneven heating of upstream equipment”: trap too small/ insufficientair handling capacity/ short-circuiting when units are group trapped.

[Backpressure too high]*: return line too small/ other traps blowing steam/ obstruc-tion in return line/ excess vacuum in return line.

Michelle grabbed her notebook and headed out for the unit. She mentally wentover some hypotheses.

What are the strengths and weaknesses of Michelle’s approach so far?What would you have done differently?What is the problem?What are your hypotheses?

Wait a minute. Here I am thinking of hypotheses when I should slow down. Firstlet’s do an IS and IS NOT, define the problem and focus on what was done duringthe shutdown. She joined a group of operators at the base of the column. You couldhear the cycling in the steam control valve. Her temptation was to immediately putthe control system on manual and see what happens but she methodically gatheredthe following information:

During the shutdown the condensate from this inverted bucket trap, that pre-viously discharged to atmosphere, was repiped to discharge into a 200 kPa g conden-sate header. The condensate discharged into the top of the header. Steam to thereboiler was “saturated” at 1.7 MPa g. For this unit the only other maintenance wasinstrument checks and visual inspection of the trays. The visual inspection involvedopening the access holes. No faults were found in the instruments and no changeswere made to the settings of the controllers. Michelle then wrote the following inher notebook:

135


What IS: cycling in the level in the sight glass. IS NOT: changes in feedrateor composition; no other apparent cycling upstream or in the feed.

Where IS: on this unit.When IS: used to work OK before the shutdown.Who IS: this is the first shift.

What is the problem? Problem SMARTS$

Specific: stop the level in the sight glass from cycling. Measure via level in glassAttainable? Should be. Depends on the root cause.Reliable? Depends on the root cause.Timely? Several simple tests should be able to resolve this.Safe? Doesn’t seem to be an issue here.$ yes, we are losing money every minute.

She confirmed that the steam entered the top of the reboiler. However, shethought “all natural circulation systems are subject to surging”.

She quickly listed some hypotheses:

. collapsed tray.

. change in downstream pressure affecting bucket trap.

. instruments wrong; liquid level is not cycling.

. control system.

. restriction in the vapor line.

. change in concentration of high boilers in the feed.

. temperature sensor incorrectly located in the feed zone.

She discarded “instrument wrong” because she could see the level rising and fall-ing with her own eyes. It may lag or lead the actual level in the column but “some-thing is cycling in there.” Collapsed tray is an unlikely possibility because no onewent down through the column checking each plate. They “visually inspected” bylooking in the access holes. She reflected on the type of tests she could do and prior-itized them. 1. Checking the control system was relatively inexpensive and fast. 2.Opening the bypass on the trap and/or changing the condensate to discharge to at-mosphere temporarily are other options to check out while she rechecks the specsand sizing used on the bucket trap.

1) When the control system was put on manual the level gradually increased.“Let’s think about this”, she said, “steam enters the thermosyphon reboiler,condenses and boils a given amount of bottoms. This causes a pressure dif-ferential and fresh bottoms cycles into the tubes. However, if the trap doesnot remove the condensate then the condensate builds up, decreases thearea, decreases the heat transferred and hence the boilup. The level increasesbecause feed continues to the column but the boil up rate is decreased.”When the set point was increased manually, the valve stem on the steammoves up, the liquid level appears in the sight glass and continues to drop

136

4.1 Case’3: The Case of the Cycling Column

but shortly thereafter the level appears in the glass and is rising. I concludethat the control system is not at fault.

2) Open the bypass and manually try to change the bypass setting to “level out”the level. Frustrating as this is to try to adjust, this seems to provide somelevel of control of the cycling.

Michelle concludes that the plant could operate by having the bypass partly openor by unhooking the condensate from the main and discharging to atmosphere as ithad operated before. She elects to do the former while she returns to her office tocheck on the sizing and selection of the trap.

What are the strengths and weaknesses of Michelle’s approach so far?What would you have done differently?Did she address the symptoms or the root cause?What tests and questions would you have posed?What corrective action would you have taken?

When Michelle consulted the design file, she found that the trap had beenselected to handle a design condensate flowrate of 0.3 kg/s for an inlet pressure of1.7 MPa and atmospheric discharge. Over the years the production rate hadincreased so that, from her energy balance calculations, the actual condensate flow-rate was about 0.6 kg/s. However, throughout these changes the trap had not beenchanged and was now operating at its maximum condensate capacity. When the dif-ferential pressure across the trap was reduced, the condensate handling capacitywas reduced by about 10% so that the trap was now undersized. To correct thisMichelle selected a new trap, or a different orifice in the existing trap, to be installedto handle 1.2 kg/s condensate under the steam pressure of 1.7 MPa and the down-stream pressure of 200 kPa. This followed the general guideline of selecting invertedbucket traps based on double the normal capacity. She also recommended that acheck valve and strainer be installed upstream of the trap.

DiscussionOverall degree of difficulty of this problem is 4/10, relatively easy. Michelle took halfa day to solve this problem. Let’s reflect on Michelle’s approach following the guide-lines in the feedback form. Overall, Michelle recognized her inexperience and triedto follow the Worksheet and to draw on the suggestions from Chapter 3. Michelletreated this as a “problem” and not an exercise right from the beginning.

Problem Solving

She was systematic, organized, used the IS and IS NOT approach and wrote thingsdown. Not much verbal monitoring was apparent and limited checking and doublechecking. She became a little too hasty in the hypothesis checking and prioritizationstages. She discarded some hypotheses without explicitly noting this. This worked OKfor her in this case but she should be developing good habits. Overall a rating of B.

137


Data Handling

Actions were based on fundamentals and she astutely enriched her experience bychecking files and trouble shooting symptoms/cases (as given, for example, inChapter 3). Her reasoning is OK for what she described. However, it is difficult toassess because she mentally discarded hypotheses and didn’t systematically checkthe evidence with the hypotheses. This worked for her this time because there wereonly two pieces of evidence: changes were made to the condensate system in theturnaround and the level in the sight glass cycles. Overall rating B.

Synthesis

She listed a variety of hypotheses. Considered many viewpoints initially. Narrowedinto the two hypotheses that were easy to check. Overall rating B+.

Decision Making

She seemed to continually try to prioritize. No obvious bias was apparent. Sheshould have discussed her decision to “operate on bypass” with the operatorsbecause they have to buy in to this change. The criteria she used were not apparent.Overall rating C.

Strengths: systematic, based on fundamentals, wrote things down, variety ofhypotheses, used resources.

Areas to work on: more explicitly aware of the process used, think about the opera-tors.

4.2Case ’4: Platformer Fires

Pierre is frustrated. Another fire. “That hydrogen just leaks out of every crevice. Andthe high temperatures we are talking about, 500 �C. Heh, how can we hope toratchet the flange tight at room temperature and hope that it will remain tight whenwe heat it up!” He pulled out the book on flange design and starting doing calcula-tions of the relative thermal expansion of metals for 1 1/4 Cr–1�2 Mo. And we hydro-statically tested the tube bundle based on the pressure differential of 2–2.7 MPa totest for leaks between the tube sheet and the shell. We were careful to use the differ-ential, instead of the absolute pressure of 4.8 MPa.

How would you characterize Pierre’s approach so far?What are the strengths and weaknesses of his approach so far?What would you have done differently?What is the problem?What are your hypotheses?

Pierre tossed the pencil down on his desk. Let’s go back to basics. I have a gas athigh pressure and temperature that is leaking through a crevice, igniting and flam-ing. We have used all our best mechanical engineering brains to design the flange toprevent leaking. Different gaskets, different tightening approaches. These were de-

138

4.2 Case ’4: Platformer Fires

signed so that the temperature differential on the bolts was negligible even thoughthe tubesheets were very thick. They’ve done all the thermal expansion stuff already.All the bolts were supposedly torqued the same. I’ve got to think out of the box!Let’s pull out the Trouble-Shooter’s Worksheet I used to use.

Is it a safety hazard? Yes siree! It is a fire hazard. Hydrogen + oxygen + spark.What’s going on is pretty straightforward, except that it’s not clear whether thenaphtha feed is on the tube or shell side. OK, that completes Engage. Define thestated problem is next.

What IS and IS NOT: fires are on the effluent exchanger. IS NOTelsewherenor on other units.

When IS and IS NOT: ever since we started upWhere IS and IS NOT: IS around the flanges on the shell and tube. IS NOTon

the exchanger

This is a fundamentals problem. OK. I think I have written down the facts so far.Explore. This is a problem! But, I need to bring out the files and clarify informa-

tion. From the files, Pierre found that:

a) the naphtha is on the tube side and at the higher pressure; the hydrogen-richgas is on the shell side.

b) the platformer temperature is kept < 540 �C to prevent thermal degradation ofthe catalyst. The hydrogen-rich gas is 500 �C and used to preheat thenaphtha.

c) from the diagrams of the effluent exchanger, the most likely hydrogen leakseems to be between the flange from the shell, the head and the bonnet.

OK, Let’s put this into perspective via theWhy? Why? The problem as I see it is to“prevent hydrogen-rich gas from leaking out the flange” so I’ll put that at the start.

Perspectives. Why? Why? Why?

6. so that I can be paidWhy? ›

5. so we can sell platformate and make profitWhy? ›

4. so the whole process can operateWhy? ›

3. make it safe; prevent flames to other parts of the processWhy? ›

2. prevent fires on the effluent exchangerWhy? ›

Start fi 1. prevent hydrogen-rich gas from leaking out the flange

That was pretty useful. Maybe I have been working on the wrong problem! MaybeI should just let the hydrogen leak out and focus on how to prevent fires on theeffluent exchanger.

139


What are the strengths and weaknesses of Pierre’s approach so far?What would you have done differently?What would you do next?

How to prevent fires? A fire is hydrogen + oxygen + spark; I said that right at thestart. OK, if I have the hydrogen, then I can remove the oxygen or the spark or both.Let’s brainstorm to remove the oxygen: nitrogen blanket, steam blanket,... I’m goingto stop there. I’ll just install a circular sparge ring whose circumference is about5 cm beyond the flange, drill holes on the inside and sparge low-pressure steam atthe flange. The heat from the steam will tend to minimize the temperature differen-tial between the inside and the outside of the flange and the steam should displacethe oxygen. Viola, no fire. Perhaps in the future they will develop Platforming pro-cesses that operate at lower temperatures and pressures, said Pierre wishfully.

DiscussionThe overall degree of difficulty for this problem is about 3/10, this is relatively easy.However, it wasn’t easy at the start for our frustrated Pierre. His frustration showsthrough in his approach. Fortunately, he got out of the box by using the Why? Why?Why? technique. Consider now some feedback for Pierre.

Problem Solving

Pierre started with more frustration and lack of confidence. He forgot that he hadan organized strategy that had been helped him in the past. He used very little mon-itoring, checking and double checking and even when he started to use the Work-sheet he used it rather superficially. He did use the IS and IS NOT effectively andtried theWhy?Why?Why? Fortunately, he used the latter effectively. Overall ratingC–.

Data Handling

He gathered information from the files, and tried to reformulate the problem basedon fundamentals. However, he really didn’t get to hypothesis generation. His rea-soning was OK based on what he told us explicitly. He didn’t formally draw on pastexperience. Overall rating D.

Synthesis

He failed to explicitly list hypotheses. Considered very few viewpoints. Overall ratingD to F.

Decision Making

He seemed to place the blame on the mechanical engineers and the inability todesign a flange that would keep the gas from escaping. He focused initially on pre-venting the gas from leaking. No apparent criteria were given for the decisions.Overall rating D.

Strengths: got an answer, could think outside the box, was somewhat systematiconce he started to use the Worksheet, used resources, based on fundamentals.

Areas to work on: self-confidence, be more systematic.

140

4.3 Case’5: The Sulfuric Acid Pump

4.3Case ’5: The Sulfuric Acid Pump

Dave is a new graduate engineer. Dave notes that the evidence points to cavitation.However, being relatively new on the trouble shooting scene he starts to write downa description of the process. Sulfuric acid is stored below the level of the pump. Thepump has to “lift” the acid up to the intake. The storage tank is open to the atmo-sphere. Ok, let me check. Have I understood this correctly? Dave runs his penaround the diagram and checks that he has included all that he thinks is pertinentabout the situation. He underlines the evidence and inserts “?” when he has some-thing he is concerned about: the vertical dimensions, two hour cycle of operation,receives “acid”? via return lines from all over the plant; dilute acid (?density andcorrosiveness); when the site gauge reads 0.7 m left in the tank, the pump makes a“crackling” noise that the operator says “sounds like cavitation”. Concern about ero-sion of the impeller (? Why, because of acid? Because of cavitation?). OK, “I want toand I can!” says Dave as he takes out his trusty Trouble-Shooter’s Worksheet. “Ok,what’s my next step? I think I am finished with Engage. But wait a minute. Whatabout Safety? Hazard? Safe-hold operation? Other than the fact that this is acid, Idon’t see this as an emergency priority so I can proceed with Define the stated prob-lem.

He quickly wrote down the IS and IS NOT information. Dave noted that he prob-ably had enough information here to “solve the problem” without going out on thesite. He recalled the cardinal rule of trouble shooting “Go out on the site and see it!”Dave acknowledged this but felt he should do some detailed calculations beforegoing out. His focus would be on the “Basics” around NPSH and suction lift sincethis trouble occurred from startup.

What were the strengths and weaknesses of Dave’s approach so far?What would you have done?Would you go directly to the plant or would you do calculations first?What calculations would Dave make?Is this really an NPSH problem?Is this a suction-lift problem?

Before Dave started into the calculations he looked out and saw the rain beatingdown. It was a good time to stay in the office! He thought. Then, he rememberedthat when it’s raining the atmospheric pressure is usually low. Since this process isopen to the atmosphere, he noted that he should include a possible lowering of theatmospheric pressure as a potential cause of the lack of NPSH. Oh, Oh! I did itagain, I said lack of NPSH instead of saying apparent lack of NPSH. By this timeDave realized that he had skipped into the Explore stage without doing his usualchecking and double checking. He perused what he had done, said a few positive“yeps” and then started focusing on the SMARTS$. His goal is to stop the “crack-ling” noise when the level drops to 0.7 m. In his mind Dave realized that he hadchanged the problem from a “crackling” problem to a “cavitation” problem to a lackof NPSH problem. Hmmmm.

141


Dave decided to continue his analysis of the NPSH and suction lift. He hauledout some texts and posed a number of What if? questions: what if the acid returnliquid is more dense? Less dense? What if the atmospheric pressure really drops?What if the vent to atmosphere becomes plugged? What if the site gauge is readingwrong? What if gas or air is dissolved in the acid? Dave stopped his ruminations andconsulted his files on cavitation:

[cavitation]*: design fault/ liquid too hot/ non-condensibles in liquid/ vortexentraining gas/ decrease in density of the liquid/ blockage or excessive Dp onsuction/ suction velocity too high/ increase in rpm without increase inNPSH/ increase flowrate increases NPSH demand. For suction-lift situa-tions: suction lift too high/ low atmospheric pressure (for open systems)/ airleakage into suction line.

Dave reflected on this information and decided that the key hypothesis to checkwas “design error”. He realized he should list a variety of hypotheses about the rootcause of “cavitation” but he wanted to do the simple checks first. If the design esti-mations looked OK, then he would list a range of hypotheses. However, the designcalculations seemed too fundamental to ignore. He would: 1) estimate the NSPHsupplied; 2) check the files and see the NPSH required from the vendor’s informationand 3) compare. He systematically did each in turn.

1) Estimate the NPSH supplied. The design files showed that the typical “diluteacid” had a density of 1.2 Mg/m3 (whereas 98% acid had a density of 1.84Mg/m3) and at the temperature of operation the vapor pressure of water isabout 5 kPa; that of dilute sulfuric acid is about 3 kPa. It was raining and theatmospheric pressure was low, he estimated it to be 90 kPa abs. He pluggedvalues into the equation for NPSH supplied for an atmospheric pressure of90 kPa, and an estimated friction loss of 0.7 m= (Atmospheric pressure–vapor pressure of fluid at operating temperature) converted to head–head oflift–head loss in friction.7.4 m–3.6 m–0.7 m= 3.1 m of acid.He rechecked his calculations. OK.

2) From the vendor files he found that the NPSH required was 1.4 m “of water at21 �C” for this pump at 1800 rpm and the design flowrate of 15 L/s. Since theNPSH required is for a pressure drop in the horizontal plane, NPSH requiredis relatively independent of the density and temperature of the fluid (al-though he remembered that some corrections have been published for theNPSH required for hydrocarbons that show the NPSH required as being about0.6 times the NPSH required for water. However, for conservative purposesuse NPSH required for cold water.) OK, so NPSH required = 1.4 m.

3) Compare. Recommendations about the relationship between the NPSH sup-plied and required vary. Some suggest that NPSH supplied should be 20%higher than that required; or 0.5 m of water higher than that required or theratio of NPSH supplied/NPSH required > 1.4. For this problem NPSH suppliedis 160% higher; larger by 3.1–1.4 = 1.7 m and the ratio is 2.6. Let’s check.

142

4.3 Case’5: The Sulfuric Acid Pump

Dave went over his calculations. I’d better return to my earlier What ifs?What if the acid density is 1.4 and not 1.2? Dave quickly rechecked his valuesand came to about 2.1 m acid. This still meets the requirements. What if thepump flowrate is 20 L/s? From the vendor’s data Dave noted that the NPSHrequired increased to about 2.5 m water. OK, thought Dave; it doesn’t look asthough it’s a design error. I’ll now go back and formally list some hypothesesconsistent with a “crackling noise” that an operator interprets as being cavita-tion. Let’s see, and Dave checked over the list for “cavitation” and selected– liquid too hot– non-condensibles in liquid– air leakage into suction line– vortex entraining gas– decrease in density of the liquid– blockage or excessive Dp on suction– suction velocity too high– increase in rpm

Wait a minute! I’m depending solely on a list from some book. Let me think frombasics about this.

What were the strengths and weaknesses of Dave’s approach so far?How well is Dave following the Trouble-Shooter’s Worksheet?What would you have done?Do the calculations prove that there should be sufficient NPSH supplied?What are your hypotheses?What would you do next?

Dave put on his raincoat and headed out to the plant. By now he had zeroed in onthe hypothesis that a vortex was forming when the level reached 0.7 m. After hetalked with the operator and looked around he decided to test his hypothesis by re-ducing the flowrate from the pump. This should lower the velocity in the suctionline, and lower the friction loss in the suction line. Under these conditions, the liq-uid level should drop lower in the tank before it starts cavitating. His discussionswith the operator and his view of the layout confirmed what he had thought. Herealized, however, that if he had come out to the plant earlier he could have hookedup the utility air line to the vent, increased the pressure in the vessel and checked tosee if it was an NPSH required being insufficient quite easily.

When the operator reduced the flowrate the liquid level dropped to 0.58 m beforecavitation. Ahah, cavitation is caused by vortex formation.

Dave shifted gears now to think about how to correct this temporarily and in thelong run. The key goal is to prevent cavitation (and the resulting erosion of the work-ing parts of the pump) and to lengthen the pump cycle. To prevent the vortex forma-tion he could float, temporarily, a wooden egg-carton construction on the surface toserve as a vortex breaker. Although Dave thought of floating a double layer of pingpong balls on the surface he realized that if a vortex did form and the balls weresucked into the pump there would be real trouble! This wooden floating vortex

143


breaker should allow pumpage to lower levels in the tank. Another option would beto reduce the flowrate as the level drops. This would mean a signal would have to betransmitted to a location near where the pump exit flowrate could be controlled. Atthe next shutdown, a well-designed vortex breaker could be attached over the exitpipe; or a control valve could be installed on the pump discharge that graduallyclosed as the level in the tank dropped.

Dave reflected on how he handled this, one of his first real trouble-shooting prob-lems. He made notes about what he would do the same and what he would do dif-ferently on his next trouble-shooting assignment.

DiscussionOverall degree of difficulty of this problem is 5/10, relatively easy. Dave took a day tosolve this problem. Overall, Dave tried to follow the Worksheet and to draw on thesuggestions from Chapter 3. This was a “problem” and not an exercise for Dave, andhe handled it that way.

Problem Solving

Dave tried to be systematic, organized, used the IS and IS NOT approach and wrotethings down. He included much verbal monitoring and checking and double check-ing. Overall a rating of A.

Data Handling

Actions were based on fundamentals and he used books, files and trouble shootingsymptoms/causes (as given, for example, in Chapter 3). He should have visited thesite earlier. He systematically checked the evidence with the single hypothesis. Over-all rating B–.

Synthesis

He focused early on the process operator’s judgment about the symptom. He usedthe resources in Chapter 3 to create a list of hypotheses but then prioritized theseand checked first the design error. Many hypotheses should be kept active concur-rently. Overall rating C.

Decision Making

He prioritized. No obvious bias was apparent. Overall rating B.Strengths: use of resources, monitoring, checking and double checking, being sys-

tematic, based on fundamentals.Areas to work on: visit the site, keep more hypotheses active.

4.4Case ’6: The Case of the Utility Dryer

Saadia perused the trouble-shooting case that appeared on her desk. Over the pastten years she had very successfully solved a range of problems with dryers, but

144

4.4 Case’6: The Case of the Utility Dryer

usually the adsorbers were hooked in parallel with one unit on-line and the otherbeing regenerated. For this case, the units are in series with the regeneration occur-ring before the adsorption. “Very interesting,” she murmured, “but I’m going totreat this case as a problem and not an exercise.” She retrieved the Trouble-Shooter’sWorksheet and checked off the items. This is not an emergency; except that we mayneed dry instrument air from a skid mount on these cold days. Let’s be sure that Iunderstand this particular process. She traced the path of the feed air on the dia-gram and noted that all traps discharged below ground and were inaccessible forsampling. “Fortunately, there seems to be enough sampling ports,” she noted as shechecked off S1 to S4. “Often the problems in the past have been leakage of steamfrom the regenerator into the air, leaky valves, inadequate regeneration, wrongadsorbent or adsorbent damaged by excessive regeneration temperature, or adsorp-tion cycle too long or change in concentration in feed.” “Hold on a minute,” Saadiasaid to herself, “you said you were going to treat this as a problem and not relive yourpast achievements! Make sure you understand this flow diagram and then move todefine the stated problem.” Saadia looked again at the flow diagram and traced theflows for Bed B being regenerated, then cooled. She wrote down succinctly: P1 reads550 kPa ? Cycle set on V1 for 2 h? Then 1 h ? And valves V2 and V3 for 3 h? TRC1= 175 �C? Flowrate about 1�2 design? Proportioning valve actually shut? Is the airreally wet leaving our unit? Is this the first time this is reported? Did the vendor giveus a performance check on this unit? Current weather? OK, I think I have finishedwith the Engage part. Now to Define the stated problem.

What IS and IS NOT: instrument air lines freezing; claims of “wet” air leavingthe drying bed.

When IS and IS NOT: on colder nights; not reported at other timesWhere IS and IS NOT: on instrument lines. No reports for air supplied else-

where in the plant.Who IS and IS NOT: plant operators claim following vendor’s instructions.

“I’ll focus on the basics. I think I’m ready to really get into this problem. Let’sExplore” said Saadia.

What were the strengths and weaknesses of Saadia’s approach so far?What would you have done?What would you do next?

Before I define the real problem, I want to gather three forms of baseline data:1) I’ll pull the vendor’s specs on this unit; 2) I’ll do a simple spot-check of perfor-mance data, with some samples and 3) do some simple rule-of thumb checks.

. Vendor’s specs. The design conditions guaranteed by vendor’s file said:– Adsorbent: 5000 kg of activated alumina H-151 per bed:– Drying cycle: 3 hours– Regeneration cycle: 2 hours– Cooling cycle: 1 hour

145


– Dry air product, standard cubic decimetres per second, Ndm3/s, FRI 1:7300

– Minimum air flow for regeneration, Ndm3/s: 2360– Proportional valve: part closed but closed enough to ensure minimum air

flow for regeneration.– Inlet air temperature, �C, T1: 32.2– Inlet air pressure, kPa g, P1: 550– Moisture content of inlet air, saturation temperature at the design pres-

sure, �C: 32.2– Final moisture content in “dry air”, �C, S4:–42.7 at pressure– Pressure drop, kPa, P1–P4 = 13.8– Steam pressure to the heater, MPa g, P2: 1.0 min–5.5 max.

. Simple spot-Test I. Readings should be made of all the instruments 1 hourinto the cycle. Use an open cup dew point apparatus attached to the samplelines S1 (inlet dew point), S3 (after the separator) and S4 (dry air dew point).I realize this is the dew point at atmospheric pressure and I have to convert itto pressure conditions if I want to compare. The results were:– Pressure, Inlet air, P1, kPa: 689– Temperature, Inlet air, T1, �C: 22.2– Dew point, Inlet (atmos) S1, �C:–6.7– Dew point, Inlet (press), S1, need to calculate.– Air flow to regeneration, all because pressure P3 is full on to close the

valve.– Pressure, Steam P2: 5.2 MPa– Temperature, Steam T2: 277 �C– Temperature, TRC 3 exit gas from heater: set 177 �C– Temperature after regenerator, T4, �C: 104– Temperature after cooler, T5, �C: 24– Dew point, (atmos) exit of Separator, S3, �C: unable to measure because it

fogged up immediately at ambient temperature and pressure.– Pressure, Effluent air, P4, kPa g: 662– Air flow, FR 1, Ndm3/s: 4000– Exit (atmos) Dew point, S4, �C:–48.3– Dp, kPa, calculated P1–P4, 27

. Simple checks: “Lots of numbers, but I want to select the key ones,” saidSaadia, astutely.– Anything unexpected with the feed, the steam, the regenerator, the

separator? No, temperatures and conditions pretty consistent with designconditions from the vendor: except flowrate= about 1�2 design.

– Dew-point conversion between atmospheric pressure and pressure condi-tions of about 762 kPa abs is a) determine the partial pressure, pp, of thewater vapor at the dew point, atmospheric; multiply by the ratio of theprocess pressure absolute to atmospheric to obtain the partial pressure ofwater vapor under the process conditions; from humidity tables/charts

146


find the dew point comparable to this partial pressure of water. Alterna-tively, I could just express everything in ppm water. Since Saadia realizedshe would be doing this conversion often she created a simple programto speed her calculations. For the test run, the Exit (press) Dew Point,corresponding to sample location S4 = –29.5 �C or a moisture content ofabout 50 ppm or more than 3� greater than expected of 15 ppm.

– According to vendor’s specs of 32.2 �C saturation at 650 kPa abs, theincoming moisture content is 7375 ppm. For the test conditions, theincoming moisture content was 3430 ppm. Hence, the exhaust specs of15 ppm should be easy to achieve if everything is working as expected.

– Interesting: at sample location S3, exit from the separator, since today’stemperature is 21 �C the moisture content is > 24,000 ppm or 2.4% v/vmoisture. “I’ll call this a surprise because I have no other data as to whatI should expect here. This suggests a carryover of mist from the separa-tor.”

– Surprises: exit amount of moisture in the air is= 3 times greater than thetarget value of 15 ppm and pressure drop= double expected. Before sheproceeded further Saadia took several minutes to recheck her calculations.Let’s see, Dew point (atmospheric) of –48.3 �C corresponds to aDew point (atpressure conditions) of about –30 �C (which is consistent with the dewpoint under pressure conditions not meeting the specs of –42.7 �C).

Saadia realized that this simple test showed that the problem was the exit mois-ture content (from the sample at S4) in the exhaust, “dried” air is about 3 timeshigher than expected. This is consistent with “freezing” lines when the outside tem-peratures are cold. The other symptom is the pressure drop is double the expectedvalue even when the flow rate is 1�2 the design value.

“It’s time for hypotheses! ... and tests” declared Saadia, whereupon she added fol-lowing list to her chart:

Symptom a. exit air “wet”: 3� higher than specificationsb. pressure drop double expected value even when the flowrate

1�2 design

c.

d.

e.

147



a b c d e A B C D

1. Steam leak S N

2. Excessive moisture carryover from theseparator

S N

3. Valve S2 leaking S N

4. Adsorbent lacking adsorption capacity. S N

5. Absorber on-line too long: breakthrough S

6. Not enough regeneration time S

7. Condenser not cooling sufficient S

8. Instruments wrong, pressure N S

9. Absorbent broken down ? S

10. Temperature TRC, T3 wrong S S

For the Diagnostic actions or tests, Saadia thought that the root cause of a highDp could be “damaged alumina adsorbent” causing the Dp across the bed to be toohigh. One factor that could cause such damage would be excessive temperature inthe regeneration. This might be caused by an incorrect TRC T3 sensor that reads177 �C but it really is 245 �C, for example. Such damaged adsorbent would also haveless adsorbent capacity.

Test A. Check/calibrate T3.Test ?: Another test would be to open up the adsorbers and sample the adsorbent,

checking the adsorption loadings and checking for decomposition. However, this isexpensive and takes the unit off-line. She preferred to do other simple on-line tests first.

Test II and III: Gather a set of data that can be used for four other tests, called B,C, D and E. Since this is a batch-cyclical process, most of tests require data that aregathered every 15 minutes over the whole cycle and with bed A being both adsorberand regeneration mode. Since the steam trap and the separator traps dischargebelow grade, data about the steam/condensate flowrate are not easily accessible.However, the trap on the separator can be isolated and condensate collected fromS2.

In Test II, 6-hour cycle with the proportional valve closed, measure the exit mois-ture concentration in the air at S4, collect condensate at S2, measure temperatureT4, plus all other usual readings.

In Test III, a 6-hour test with the proportional valve open and a correspondingsmall flow of air through the regenerator heater and the condenser-separator, mea-sure the exit moisture concentration in the air at S4, measure temperature T4, plusall other usual readings.

Such information could then be used to do four other tests:Test B. Mass balance on moisture. Excessive water coming out might suggest a

steam leak.Test C. Compare adsorption specs on alumina with moisture adsorbed in one

cycle. Low adsorption/ mass of adsorbent means the cycle time is not long enoughto load the adsorbent or the alumina is lacking the adsorption capacity.

148


Test D. Compare time plots of temperature T4, liquid collected S2 and Dew Pointsfrom S1, S4. She sketched the time plots she expected. These are shown in Figures4-1 and 4-2.

Figure 4-1 Prediction of the adsorption cycle.

Figure 4-2 Prediction of the regeneration.

149


Figure 4-1 shows a typical loading with the exit concentration below the target 15ppm. Near the end of the cycle, the concentration may increase as the start of thebreakthrough curve appears. That’s what I would expect to see for all the tests.

Figure 4-2 shows what I would expect for the regeneration and condensation ofthe water from the separator. During the hot regeneration period, the temperatureon T4 should increase and level off on a plateau around 105 �C and then rise to theinlet temperature, 177 �C when the bed is almost regenerated. If regeneration wasstarted before the bed was fully loaded, then the alumina near the end of the bedwill be dry, then wet and then dry as the desorption band sweeps through the bed. Ifthe bed loading is small (or the gas velocity high) then the length of the temperatureplateau is short. The condensate collected should spike before the cool-down period.For the cool down, the effluent temperature T4 should continue to rise but will dropafter the high-temperature wave from the heated, regenerated adsorbent passesthrough the bed.

Test E. Compare pressures on P1–P4 for four different conditions: a) the heatingcycle for regeneration versus b) the cooling cycle (to see Dp over heater) and d)when the proportional valve is shut versus e) partially open to see the Dp in theadsorbing bed versus including the Dp for regenerating bed and cooler-separator.These tests should be done with both A and B beds being adsorbers.

Before I order the tests, let me check. She reflected on how she had handled theprogram so far, her hypotheses and the proposed tests. Satisfied, she planned thetests.

What were the strengths and weaknesses of Saadia’s approach so far?What would you have done?Should Saadia have just replaced valve V2?Should Saadia have called in the vendor?Should she have opened both beds and tested the alumina?What would you do next?

Saadia was pleased that she had resisted the urge to jump in and change Valve 2,test the alumina and to call in the vendor. The 2 days of tests and analysis of datashould clarify many issues and dispel incorrect hypotheses before such major actionis taken. Here are the results.

Test A: TRC 1 responds to change and calibrates OK. Conclude: the regenerationair temperature is 177 �C. Hypothesis 10 that the regenerated air temperature is hot-ter than 177 �C is disproved.

Test B: Mass balance on moisture. Test II, for the 6-hour cycle with the propor-tional valve “closed”; exit gas flowrate, FR 1 = 2830 Ndm3/s; loading bed B for thefirst 3 hours and collecting the condensate from the regeneration of B during thesecond 3 hours:

79.7 kg in=? 66.1 kg out.For bed A:79.7 kg in=? 74.3 kg out.

The balance isn’t within the 10% that I like to see, but this suggests that steam isnot a source of water coming into the system. The condensate collected is less than

150


the apparent bed loading. Reject hypothesis 1 that there is a steam leak and perhapsconsider poor separation in the separator with resulting carryover of entrained waterto the adsorber.

Test C: Check adsorption loading of water/ mass of dry adsorbent.From my files for activated alumina: 0.14–0.22 kg water/kg dry adsorbent.Calculated from the vendor specs: 0.089 kg water/kg dry adsorbent for air flow

7300 Ndm3/sCalculated from Test II for the 6-hour cycle with proportional valve “closed”; exit

gas flowrate, FR 1 = 2830 Ndm3/s; loading bed B with subsequent regeneration:0.016 kg/kg.

Notes: for the Test II conditions the flowrate is below vendor design specs so themoisture coming to the unit is less. If this incoming moisture is increased in pro-portion to the flowrates, however, the loading on the adsorbent becomes 0.016 �7300/2830 Ndm3/s or 0.0412 kg/kg. This suggests that the adsorbent is not adsorb-ing the amount of water expected from the vendor design specs (0.09 versus actual0.04 or about 1�2) nor that expected based on file data about activated alumina (0.14versus actual 0.04). Conclude thatHypothesis 4might be true.

Test D: Plots of data are given in Figure 4-3 for Test II, the 6-hour cycle with theproportional valve closed. The data are the exit concentration S4 from the adsorbingbed for beds A and B respectively, in Figures 4-3 a and b. What a surprise! For bothbeds A and B on the adsorption cycle, there is a spike of water in the middle of thecycle. A comparison with the water evolution from the concurrently occurringregeneration, in Figure 4-4 a and b, shows consistency in the spike of water. Thissuggests leakage across valve V2. Saadia estimated the amount of gas leakingthrough the valve based on the times of the peak condensate collection to be about1.5–3 dm3 / s. I conclude that Hypothesis 3 is correct.

Figure 4-3a Test II, bed A adsorb with proportional valve closed.

151


Figure 4-3b Test II, bed B adsorb with proportional valve closed.

Figure 4-3c Test III, bed A adsorb with proportional valve open.

152


Figure 4-3d Test III, bed B adsorb with proportional valve open.

Figure 4-4a Test II, bed B regenerate (while bed A is adsorbing) with proportional valve closed.

153


Figure 4-4b Test II, bed A regenerate (while bed B is adsorbing) with proportional valve closed.

Figure 4-4c Test III, bed B regenerate (while bed A is adsorbing) with proportional valve open.

154


Figure 4-4d Test III, bed A regenerate (while bed B is adsorbing) with proportional valve open.

Test III, a 6-hour test with the proportional valve open and a corresponding smallflow of air through the regenerator heater and the condenser-separator. The adsorp-tion results are given in Figures 4-3c and d. Bed A seems to show the appropriateadsorption behavior. Bed B exhibits breakthrough of the water front. No condensatefrom S2 was gathered from this run so a mass balance could not be done. Since theflow through the cooler is relatively small, I am beginning to think that the separatormay not be operating well, based on high concentrations of water found at S3of > 24 000 ppm. The temperature profiles during regeneration are given in Figures4-4 c and d. These profiles are not what I expected even though Bed B adsorbed onlya small amount of water.

Test E. Pressure drop profile: P1–P4. Vendor: = 14 kPa. Simple spot test I with pro-portional valve closed= 27 kPa. Test II: proportional valve closed: regeneration of bedB and loading of bed A: Dp= 62 kPa and remained the same during the coolingcycle. I conclude that the pressure drop across the heater is negligible because therewas negligible change in Dp when the heater was bypassed via the 3-way valve V1.For the regeneration of bed A and the loading of bed B: Dp= 125 kPa with negligiblechange when the heater was bypassed. Test III: proportional valve is open with mostof the air going directly to the adsorbent bed: for bed A adsorbent Dp= 34 kPa; withbed B adsorbent Dp= 62 kPa. These results suggest that there is a significant Dpacross the cooler-separator and that the adsorbent in bed B, with the larger Dp andearly breakthrough behavior, should be tested.

155


Saadia used the data to perform an energy balance over the regeneration opera-tion and found, to her surprise that the energy would balance only if the flowthrough the “closed proportional valve” was 1270–1360 Ndm3/s. The balance couldalso be closed if 60% of the water entering the separator was entrained.

Saadia arranged for the 4-way valve, V2, to be changed, and she contacted the ven-dor. For the new valve she recommended the replacement of the iron core with astainless steel core. She also was concerned that the temperature of the gas leavingregeneration exceeded the plateau temperature of about 110 �C.

DiscussionOverall rating of the difficulty of this problem 9/10, relatively difficult because manyfaults are present. Saadia took several days to identify the fault in the valve and todefine a series of concerns about this unit. Overall, Saadia drew on her experiencewith adsorption, especially in designing the tests and predicting what she expectedto see before the test results were obtained. She astutely chose to follow the Work-sheet because this setup was different from her experience. This was a “problem”and not an exercise for Saadia and she handled it that way.

Problem Solving

Monitored, especially by predicting the shape of the adsorption and regenerationcurves before doing the tests. She was organized and systematic. She checked anddouble checked. Her dominant P style showed through in that she delayed changingthe valve and calling in the vendor until she had more data. Overall a rating of A.

Data Handling

Actions were based on fundamentals. She related the test to the hypothesis. Goodreasoning displayed. Overall rating B+.

Synthesis

She listed a variety of hypotheses. Considered many viewpoints. She had largemasses of data to consider; she managed this well. Overall rating A.

Decision Making

She seemed to continually try to prioritize. No obvious bias was apparent. Criteriawere present for some of the decisions, but for others, the criteria were not notedexplicitly. Overall rating B+.

Strengths: Recognized the need to problem solve, used fundamentals, systematic,managed large volumes of data well, rich set of hypotheses, did not get discouraged.

Areas to work on: provide more explicit details for decisions.

156

4.5 Case ’7: The Case of the Reluctant Crystallizer

4.5Case ’7: The Case of the Reluctant Crystallizer

Frank has 25 years of experience. He is a process engineer responsible for severalprocesses on site including the crystallization unit. The phone call comes in fromPhil, the operator in the control room. “Something wierd is happening on the vac-cum crystallizer, Frank. We’ve been operating the VC for about two hours now. Thefirst hour with only two ejectors turned on, the operation was OK. We turned on thebooster ejector and it “held” for the first 1/2 h or so and then it “kicked out”. Whilethe booster ejector was “holding” the liquid level in the crystallizer drops at a fantas-tic rate. What’s going on here? We just can’t produce quality crystals under theseconditions.”

“I agree, it sounds wierd. I’ll be right out,” responds Frank.Frank has developed a good working relationship with all the operators so it

doesn’t surprise him to receive the call and for Phil’s succinct summary. He tookout the diagram for the process and mentally reviewed his thoughts. The vacuumcrystallizer, VC, is the core of this operation. Pregnant liquor feed enters the vacuumcrystallizer at 55 �C where it is concentrated and cooled to precipitate the product.This is a batch process. To start, we pull a vacuum on the VC (using the two-stagesteam ejectors with city water to the interstage condenser), open the valves from thefeed tank and syphon feed into the VC until the unit is 2/3 to 3/4 full, as seen onthe sight glass. The feed valves are then shut and the eight-hour batch processbegins. While operating with the two-stage ejector we estimate the absolute pressureto be about 6.5 kPa abs. After about an hour, the temperature of the liquor hasdecreased to 40 �C and the liquid level has dropped by about 10 cm. At that time, thebooster ejector and the second barometric condenser are turned on to decrease thepressure further to about 2.5 kPa abs. At the end of the 8 hours the liquid level hasdropped about 40 to 50 cm. The temperatures of the water in both barometric con-densers are monitored to ensure that the water temperature does not exceed 26 �C.The operators have some control over this temperature by judiciously mixing coldcity water with warmer water from the bay. Experience has shown that if the boosterejector is turned on too soon, it will not “hold” but rather “kicks out”. “Kicking out”produces a very distinctive sound. This jargon has been used ever since Frankstarted on the unit. “Kicking out” has been interpreted as being when the steam con-trolled by valve S3 goes directly into the VC instead of through the ejector nozzleand thus does not pull the required vacuum.

Frank headed out to the control room. Here are some of his thoughts. Steam ejec-tors ... . they are so sensitive to upstream steam pressure. Frank mulled over hisexperience: “unstable operation or loss of vacuum”: steam pressure < 95% or > 20% ofdesign/ steam superheated > 25 �C/ wet steam/ inlet cooling-water temperature hot/cooling-water flowrate low/ condenser flooded/ heat-exchange surface fouled/ 20–30% higher flow of non-condensibles (light end gases, air leaks or leaks from firedfurnaces)/ seal lost on barometric condenser/ entrained air in condenser water/ re-quired discharge pressure requirement high/ fluctuating water pressure. Hmm.. Itwas a hot summer day.. so the bay water is likely hotter than usual; the maintenance

157


turnaround was only a month ago and the barometric condenser was inspected andlooked OK. Everything had operated fine since the turnaround. We usually operatethe steam to the booster ejector with the valve half open. Hmm, lots of possibilities.By this time Frank had zeroed in on five hypotheses:

. steam to the booster ejector was at too low a pressure,

. steam superheated or wet,

. cooling-water inlet temperature too high,

. loss of vacuum caused maybe by loss of seal in the barometric leg or leaks inVC,

. pressure in the barometric condenser higher than usual.

In a moment he would be in the control room. He formulated some questions hewanted to ask Phil before he went out on the plant.

What were the strengths and weaknesses of Frank’s approach so far?How would you have handled the call from Phil?Would you go directly to the plant before seeing Phil so that you are better prepared to

ask questions?Would you have phoned the boiler house first to see if they had upsets?What questions would you ask Phil?What are your hypotheses?What is the evidence so far?

Frank realized that he worked better when he wrote things down but he wantedto do that with Phil’s help. Furthermore, one of the reasons Frank got along so wellwith plant operators is that he always went to the control room first, before ever ven-turing out on the plant, and he always respected their experience and ideas.

After greeting a worried Phil, Frank and Phil sat down to put their thoughts onpaper, the way they usually did in situations like this. Phil knew that Frank liked touse the IS and IS Not approach to summarizing the evidence. “Check me if I’mwrong,” said Frank, “but this is what we have so far.” He wrote:

IS: what and when: operating OK during startup, feed intake and first hour ofoperation before the booster ejector started. The booster ejector operates on “hold”without “kicking out” for an hour.

IS NOT: what and when: is not operating as usual when booster operating as“hold” with evidence that the liquid level (on the sight glass) drops “at a fantasticrate”.

Phil looked at the list and said “Looks OK.” and added “When the liquid levelstarted to drop I went up and listened to the booster ejector and it sounded fine. Theonly thing I noticed was the pressure gauge on the bay water line to the booster con-denser was fluctuating wildly. I also knew that you attribute most vacuum-systemmalfunctioning to steam pressure to the ejector. I checked the steam pressuregauges on P1, P2, P3 and P8. They are all 550 kPa g and rock steady. Gauge P4 was725 kPa g, as usual, and steady.”

Frank said, “Good thinking. I need some clarification. What did you observewhen the liquid level drops at a fantastic rate?”

158


“The level in the sight glass dropped about 1 cm every two seconds. You couldvisibly watch it go down.”

“Thanks. When you checked the pressure gauge P8, did you tap the gauge?”“Yes, I tapped it and it seems to be working OK. The pressure on all steam gauges

was 80 psi.”Frank mentally converted this to 650 kPa abs, that was normal steam pressure.

“The pressure on the bay water line usually reads 205 kPa g; today it fluctuatedwildly. How extensive was the variation, Phil ?”

“We don’t normally check that pressure. It isn’t shown in the control room. Outon the plant that gauge needle was jumping back and forth between 140 and 550kPa g. Wild!”

Thinking about today’s hot August temperature Frank asked “What about thetemperatures on the barometric legs?”

“The temperatures on both legs were usual, certainly less than 27 �C; T2 wasabout 24 �C and T1, about 22 �C.”

“Just for the record, what were the readings on the instruments you checked onthe plant and how do these compare with the usual valves?”

“The steam pressures I talked about earlier; pressure on the city water P6 was theusual: 310 kPa g (45 psi) before the valve and P5 was 0 to 35 kPa g (5 psi) after thevalve. I didn’t note the temperature of the liquor in the VC. The level had disap-peared below the sight glass. That’s when I called you.”

Frank added that evidence to his charts and thought about his hypotheses now inthe light of the new evidence:

. Steam to the booster ejector was at too low a pressure. But all the pressuregauges read the same usual values, and they are steady! Perhaps the pressuregauge P8 is faulty.

. Steam superheated or wet. Can’t tell, but if it was then the other ejectorsshould malfunction as well. They don’t; so this is unlikely.

. Cooling water inlet temperature too high. Still possible but the exit tempera-tures in the barometric legs suggest this is not a cause.

. Loss of vacuum caused maybe by loss of seal in the barometric leg or leaks inVC. Perhaps; relatively easy to check, so why not?

. Pressure in the barometric condenser higher than usual. No gauge is avail-able on the condenser or downstream of valve W5. Not easy to tell but per-haps this is connected to the wild variation in pressure in the bay water com-ing in. Perhaps valve W5 has a worn valve stem that is vibrating and causingthe oscillations in pressure. Slugging flow of water? Is the bay water pumpbehaving itself? Should it be checked?

So to sum up, Frank is down to three hypotheses: 1) steam pressure to the boosteris low or fluctuating but the pressure gauge is broken. Check the pressure gauge. 2)changes in the vacuum with leaks; questionable but let’s check. Frank can’t see howthe fluctuating pressure gauge on the bay water is connected to rapid evaporation inthe VC. However, he decides to change his focus to be “Why is gauge P7 fluctuat-ing?” He hypothesizes that P7 is fluctuating because of 3) fluctuations in the pump

159


exit pressure from the bay water pump or oscillations of the valve stem in valve W5.What’s going on here?

What were the strengths and weaknesses of Frank’s approach so far?What would you have asked Phil?What are your hypotheses now?What has Frank missed?Is Frank exhibiting pseudodiagnosticity or fixation when he keeps looking for a steam

fault to the booster ejector?Should Frank have considered the evidence of “kicking out” of the booster ejector?What would you do next?

Frank makes a list of his next steps.

1. While the plant is still operating, visually inspect the VC for a leak. Attach agauge to the VC via the valve and nozzle at the top and read the vacuum/pressure noting it every 10 minutes for two hours. Hypothesis, if there is aleak, the pressure will gradually increase. If the pressure remains constant,then there is no leak.

2. Shut down the process, and remove and visually check valve W5; while weare at it check valve W4. Frank realizes that this is not going to confirm anyof his hypotheses, except perhaps his wild idea of a vibrating valve stem inhypothesis 3. However, he still can’t see any connection between the oscillat-ing pressure and the rapid evaporation. However, since the change is occuringnear the valve, let’s check out the valve.

3. To test hypothesis ’1, either replace the pressure gauge at P8 or recalibratethe existing pressure gauge P8.

Frank hoped that the results of these tests would resolve the mystery. The resultswere:

. Hypothesis ’1) steam pressure to the booster is low or fluctuating. Evidenceis that gauge P8 is working and calibration is correct. Hypothesis denied.

. Hypothesis ’2) changes in the vacuum with leaks. Evidence is that a visualinspection and the vacuum test could be interpreted as being no leaks.Hypothesis denied.

. Hypothesis ’3) fluctuations in the pump exit pressure from the bay waterpump or oscillations of the valve stem in valve W5. Frank received telephoneconfirmation from Utilities that the exit pressure on the bay water pump isconstant, steady and the usual value. Valve W5 (and valve W4) were removed,visually checked as OK. A frustrated Frank decided to replace W5 with a newvalve anyway. The process is started up. A number of other engineers havegathered in the control room to see if Frank has solved the mystery. The sys-tem behaves the same as it did before the changes: the booster ejector “held”for the first 1/2 h or so and then it “kicked out”. While the booster ejector was“holding” the liquid level in the crystallizer drops at a fantastic rate.

160


Frank revisits his starting hypotheses and notes that perhaps the water tempera-ture is too high in the barometric condenser. After all it’s a really hot day. Frank fol-lows his dominant J behavior and decides to take action and “correct the fault” ratherthan gather more data. Frank repiped the system with a temporary hose that sentcity water into the bay water feed line so that only cold city water went to the conden-ser. The use of the hotter bay water was eliminated. Unfortunately this repipingmakes no improvement!

Frank returns to his office. “It’s back to basics! I’ve never seen anything like thisbefore. I can do this!” What can cause the liquid level in the crystallizer to drop at such arate? Brainstorm: liquid is leaking out, the change is in the sight glass and not inthe vessel; something is sequestering the liquid, the vacuum is much higher thanexpected and the liquid is flashing off, the temperature is higher than expected, theabsolute pressure is much lower than expected; steam flow into the booster is higherthan expected; the steam flow is oscillating but its oscillations are so fast that thechanges aren’t picked up on the steam gauge P8 but are picked up on the water loadand reflected on P7; the fluctuating water flow into the condenser is causing a vacu-um that is pulling off more vapor than expected; liquid is going back into the feedtank; the discharge line is open, liquid leaks around the mixer shafts; the tempera-ture gauge on the VC is broken so the temperature is much higher than it reads;liquor from the VC is entrained and not evaporated; liquor goes out through thebooster ejector.

Now let’s refocus and brainstorm on the causes of wildly fluctuating gauge P7 whenthe bay water pump exit pressure is stated as being steady and the usual value. Frankwrote down the following: change in water pressure, upstream oscillations, oscilla-tions in the steam, oscillations in the liquid level, oscillations in condenser, citywater hitting the bay water in condenser, oscillating liquid level in barometric leg,vibrating baffles in condenser, vibrating valve stem on water, water bypassing peri-odically directly into the barometric leg, corrosion particle in the pressure tap,unstable bourdon tube to pressure gauge, bay water pressure is not steady.

“I recall the case of the filter press operation was affected by the upstream oscilla-tions of a centrifuge”, murmured Frank to himself. “An idea that links the rapidevaporation and the fluctuating pressure gauge could be oscillations in the steamflowrate going into the booster ejector.” Frank hasn’t quite given up on the idea thatthe steam into the ejector is the key. Frank headed back out to the control room.Frank and Phil went to scrutinize the steam line into the booster ejector. The valvewas a globe valve that was operated partly open. When Phil carefully adjusted thevalve he felt some vibration. “Maybe that’s it,” exclaimed Frank. They shut down thetroubled plant, removed the steam valve and found that the valve stem had beenseverely eroded. They replaced the steam valve S3 with a tapered plug valve. The re-placement of the valve solved the problem. Case dismissed.

DiscussionThe overall degree of difficulty of this problem is 10/10, very difficult. Frank took aweek to solve this problem. The costs in loss of product and loss of pregnant liquorcarryover into the hot well was extremely high. Let’s reflect on Frank’s approach

161


following the guidelines in the feedback form. Frank started this problem using anexercise approach although even here he showed excellent problem-solving charac-teristics: IS and IS NOT, systematic, patient.

Problem Solving

Systematic and organized even though he was very frustrated. Patient and takestime to understand the system (even though he had worked many years on thisplant). Some monitoring. Some checking. He’s active and wrote things down. Over-all rating B+.

Data Handling

Based on fundamentals–especially after he recognized it as a problem and not anexercise. Very systematic in planning and gathering evidence. Used simple tests.Overall rating A–.

Synthesis

Used a variety of hypotheses. Flexible. Kept an open mind even when nothingseemed to be working out. Overall rating A–.

Decision Making

Prioritized. His bias about the low steam pressure actually paid off. Decisionstended to be made intuitively with few explicit criteria. Overall rating B.

Strengths: excellent interaction with the operators, very systematic, didn’t panic,flexible, time well spent exploring and understanding the problem, patient.

Areas to work on: more monitoring and more explicit thought process especiallywith criteria for decisions.

4.6Reflections about these Examples

In each of the cases, the engineer “solved the problem.” Each used a different style.The less- experienced engineers, Michelle and Dave, used the Trouble-Shooter’sWorksheet with varying degrees of comfort. The more-experienced engineers,Pierre, Saadia and Frank, tended to resort to the Trouble-Shooter’s Worksheet asneeded. Pierre used it to remind himself to try the Why? Why? Why? approach. Hemight have tried this without resorting to the Worksheet. Indeed, the problem hewas addressing seemed to be one of the few where that technique was useful. Saadiarecognized the uniqueness of her problem early on and, although she had a lot ofexperience with adsorption, she wisely disciplined herself to work slowly and sys-tematically through the Worksheet. Of the five engineers she did the best job ofusing the “hypothesis-symptoms-actions” chart. Frank treated the problem like anexercise until he discovered it really had more unusual components than he hadusually encountered. Before it was too late he started to use the Worksheet.

162

4.6 Reflections about these Examples

What components of the Trouble-Shooter’s Worksheet seemed to be used moreeffectively?

All tried to slow down the process and work carefully in the Engage process, al-though Pierre’s efforts were not the greatest. This was astute of the trouble shootersbecause most mistakes are made right at the start. Dave almost got caught here.Pierre was stymied here for a frustrating few moments.

The Define the stated problem stage was handled better when all five engineersused the IS and IS NOT tactic effectively.

For the Explore stage, many used a variety of approaches and, indeed, were selec-tive in the elements they included. Dave used the What if? elements effectively. Theless experienced engineers drew on the symptom-cause information in Chapter 3 tohelp them overcome their limited practical experience. Fortunately, for the problemsthey were working on some data were available. For some equipment, the publishedinformation is lacking.

The “hypotheses-symptoms-action” chart was only used well by Saadia. Theothers, especially Frank, tended to use the ideas intuitively, but such an intuitiveapproach has pitfalls. Saadia also excelled at using her fundamentals to predict,ahead of time, what she expected to see from the results of the tasks.

Frank seemed to be the only one who was comfortable working with and drawingon the experience of the operators. This didn’t come without patience in developingtrust and a good working relationship with Phil. Michelle, and Dave, should polishtheir skills in this area.

163

165

Five components are useful in developing your skills: becoming comfortable talkingaloud about your thought processes, identifying a strategy and monitoring thestages you use, defining the problem, creativity and self-assessment. Activities todevelop your awareness, skill and confidence in these skills are given1). To gain themost, select the areas you want to work on and complete the activities. Developingskill is not a spectator sport. Participate.

5.1Developing Awareness of the Problem-Solving Process

Awareness is the ability to describe what goes on in our mind as we solve problemsand make decisions. Such awareness is a prerequisite for the triad improvementactivities in Chapter 8.

Such awareness also helps us improve our problem solving – and trouble shoot-ing – skill because:

. we can explicitly identify “where we are in a problem”,

. we can compare “how we do it” with how “others do it”,

. we can identify our strengths and areas to work on to improve our problemsolving,

. we can get ourselves “unblocked” if we cannot seem to solve a problem,

. we need to be able to describe our thoughts for team problem solving,

. we can describe our thought processes to another so that we can improve,

. we build on this awareness to develop skills with strategies, in Section 5.2.

One of the more successful techniques to quickly acquire confidence and skill in“awareness” is the Whimbey pairs or Talk Aloud Pairs Problem Solving, TAPPS,approach. Two people are needed. One plays the role of a “talker” or problem solver;the other plays the role of the “listener”.

5

Polishing Your Skills: Problem-Solving Process


1) The materials in this chapter are from the MPS program, copyrightDonald R. Woods 1982 ff and used with permission.

5 Polishing Your Skills: Problem-Solving Process

In this section we list the target skills that are developed by doing the activities inthis section, describe the roles, give the activities and the feedback forms.

5.1.1Some Target Skills

Research in problem solving has shown that successful problem solvers exhibit thefollowing characteristics (related to awareness and the thought processes and theactivities in this section): They

. are aware of their thought processes and use that awareness to identifywhere they are in the process and get themselves unstuck,

. are skilled in describing aloud their thoughts as they solve problems,

. pause and reflect about the process and about what they are doing,

. accept that their particular style works for them; others may have a differentpreferred style,

. are active as opposed to passively trying to remember stuff. They are active bywriting things down, making charts and diagrams. Being active helps over-come the space limitations of Short-Term Memory (the portion of the brainwhere problem-solving tasks are done. This portion can only “hold” five tosevens bits of information at a time.)

. focus on accuracy and not on speed,

. accept that problem solving is a social process; we interact with others,

. know that self-assessment is about performance and not about them as a per-son,

. know that self-assessment is based on evidence and not on gut feelings orwishful thinking, and so to assess progress in developing skills they will usefeedback forms and written evidence and reflections.

In addition, this activity provides an opportunity to

1. Acquire some skill at listening,2. Acquire some skill in self-assessment,3. Acquire some skill in giving and receiving feedback,4. Through self-awareness, begin to improve self-confidence.

5.1.2The TAPPS Roles: Talker and Listener

Talker/ problem solver: In pairs, with a partner as a listener, the talker reads theproblem statement aloud, and talks aloud as he/she works on/solves the problem.The goal is not to get the “right” answer to the problem. The goal is to talk aloudcontinuously about the process. The goal is to have, in a ten-minute period of talk-ing, fewer than two silent periods of more than 10 s duration. The goal is to focuson accuracy. The goal is to be active and write things down.

166

5.1 Developing Awareness of the Problem-Solving Process

Your listener will not give you any hints about how to solve the problem. Yourlistener will help you to talk about what you are thinking so that the listener canfollow and understand what you are saying. More specifically, here is what you do inthis role:

1. Sit side by side; have paper and pencils available.2. The talker starts by reading the problem statement aloud.3. Then start to solve the problem on your own. You are solving the problem.

Your partner is only listening to you. He or she is not solving the problemwith you or for you.

4. Talking and thinking at the same time is not easy. At first you might find ithard to think of the right words to use. Do not worry. Say whatever comes toyour mind. No one is testing you or marking you. You are playing the role.

5. Go back and repeat any part of the problem you wish. Use such words as “Iam stuck! I do not know what to do! Maybe I should read the problem state-ment again.”

6. Try to solve the problem no matter how easy it is. You are learning to talkabout your thinking methods. We use simpler problems to help make thisactivity as easy as possible.

Listener: You have an important and difficult role to play. You are to help the talkertalk. You are not to solve the problem nor to give hints to the talker about how youwould solve the problem. This is the problem talker’s turn. You will get a turn laterto be talker.

Encourage them to talk aloud. At the same time, you are to monitor their think-ing. Can you understand what they are saying? Can you follow the path that theirmind is following? Could you describe what they are thinking to others? You are tohelp them to talk about the mental processes they are using–no matter how silly orincorrect they might be. You must not laugh at them. You must not criticize themand tell them that they are wrong. If you think they made a mistake, then say “Canyou check that?” or “Are you sure?” Do not tell them what they should be doing. Donot tell them what you think is the “correct” answer.

1. Help the talker to see that you are not a “critic”. Instead, you are a helper-for-talking. You might say “ Please keep talking.” or “I was not able to understandor follow what you just said; would you please explain.” “Can you tell mewhat you are thinking now?” “Do not worry about how it sounds – just say anidea about what you are thinking.” “Can you check?” “Are you sure?” “OK”“Ahmmm”. (For more suggestions about listening, see Section 7.1.2.)

2. You might tell the talker that your role is to:– Remind them to keep talking,– Help them improve the accuracy of their talking about their thinking by

saying “Can you check?” “Are you sure?”– Be able to understand and follow each step of the talker’s thinking.

167


3. Do not turn with your back to the talker and try to solve the problem on yourown. Do not solve the problem on your own and then tell the talker what theyshould do.

4. Do not let the talker continue if:– you do not understand what they have done. Say “I didn’t quite under-

stand that, could you please elaborate on your thoughts, Thanks.”– you think that a mistake has been made. Do not say, “You have made a

mistake.” You might say “Are you sure?” or “Do you want to check that?”

Your goal, when you receive feedback on how well you played the listener role, isthat you should be within two scale ratings of “about what I wanted” on both the“degree of interaction” and the “tone of the interaction.”

5.1.3Activity 5.1: (35 minutes)

This activity will take 35 minutes. 3 minutes getting set up; 10 minutes talker A,5 minutes reflection, complete feedback forms; switch roles, 2 minutes getting setup, 10 minutes talking, 5 minutes reflection, complete feedback forms.

. Getting set up. (3 min) Find a partner, flat table, two chairs side by side, pen-cil and paper. One person starts as talker, the other is the listener. Read overyour role (Section 5.1.2).Talker selects an “exercise” (or perhaps a “problem”) from among Tasks

5.1A to E.. First 10-minute talk (10 min): talker talks for 10 minutes. If the talker com-

pletes the selected task early, then the talker selects another task. Do notchange roles! The talker talks for 10 minutes.

. Reflections, discuss and feedback forms (5 min). As individuals, write downreflections about the activity, use Worksheet 5-1. (1.5 min). In pairs, discusswhat you experienced. Don’t continue working on a Task. (1.5 min). Listenercompletes feedback form in Worksheet 5-2 and signs it; gives it to the talkeras evidence. Talker completes feedback form in Worksheet 5-3 and signs it;gives it to the listener as evidence. Worksheets 5-1 to 5-3 are given in Section5.1.4.

. Switch roles and get ready (2 min) Read the instructions about your new role(Section 5.1.2). Talker select a Task from among Tasks 5.1.A to E.

. Second 10-minute talk (10 min). See instructions above.

. Reflections, discuss and feedback forms (5 min) as above.

Task 5.1.A: Content-generalA.1 (based on Lochhead and Clement, Univ. of Massachusetts) At Mindy’s restau-rant, for every four people who ordered cheesecake, there are five who ordered stru-del. If C represents the number of cheesecakes ordered and S represents the num-

168


ber of strudels ordered and M represents Mindy’s restaurant, write an equationusing these variables to represent this situation:

a) M: 4C= 5Sb) 4C � M=5S � Mc) 4C = 5Sd) Total= 4C +5Se) (5/9) S =Cf) M: 5C= 4Sg) 5C � M=4S � Mh) 5C= 4Si) S = (4/9) Cj) other

A.2 (reprinted courtesy of Art Whimbey) There are two clocks, A and B. Clock Akeeps perfect time but clock B runs fast. When clock A says 4 minutes have passed,clock B says that 6 minutes have passed. Both clocks are set correctly at 5 a.m. Whatis the correct time when clock B shows 9 p.m.?

Task 5.1.B Tasks related to science and engineeringB.1 (based on Lochhead and Clement, University of Massachusetts)

A pillar supports a floor. Is the pillar doing any work?:

a) Yes, of course it is doingwork; otherwise the floor would come crashing down.b) No.c) Yes, a definition of work is a force acting over a distance; the force is the force

pushing up on the floor above and the distance is the height of the pillard) Yes, we know this because of energy considerations. The potential energy is

the mass of the slab or floor times the height of the pillar that is supportingit. If the pillar wasn’t there, the potential energy would turn into kinetic ener-gy and would eventually be released in the form of heat and sound as itcrashed down.

e) Other

Task 5.1.C General tasks more related to trouble shooting.C.1 A researcher predicted that if part of a leaf is in the shade and the other part ofthe leaf is in the sun, then equal amounts of starch will be found in both parts ofthe leaf. Which of the following hypotheses is the researcher most likely assumingis true?

1. Chlorophyll is present in both the shaded and sunny parts of the leaf.2. In the shaded part of the leaf photosynthesis is increased.3. Carbon dioxide and water can enter the leaf cells in both the sunny and the

shaded parts of the leaf.4. By shading part of the leaf, photosynthesis is increased in the sunny part of

the leaf.5. Starch can move to different parts of the leaf.

169


Task 5.1.D Engineering relatedD.1 Oversized condenser

The overhead condenser on the distillation column is oversized by 38%. That is, ithas the correct number of tubes to promote turbulent flow inside the tubes. How-ever, the length of the tubes has been increased. The baffle spacing on the shell sideand the baffle window has been sized for the design overhead rate. This is in addi-tion to the usual allowances for fouling. The condenser is horizontal. Which of thefollowing observations are consistent with this situation when the condenser is firstput into service:

a) The exit cooling-water temperature will be colder than expected.b) The pressure drop on the shell side will be 1.382 more than we expected.c) The reboiler will act as though it were undersized and the column operation

will be unstable with vapor blanketing in the reboiler because film boilingnow occurs.

d) Nothing unexpected; the controller action will account for the overdesign.e) Increased power required on the cooling-water circulation pumps.f) The feed location should be changed to be closer to the reboiler.h) Other

LIC

3

LIC

1

Cooling water

TRC

FRC

1Feed

Steam

Figure 5-1 A column.

170


Task 5.1.E Technical tasks related to methane-steam reformingE.1 In the methane-steam reformer shown in Figure 5-2, the total feedrate to thereformer is 25% higher than the flow instrument reads. The other instrumentsusually read reformer outlet temperature T1 = 454 �C (850 �F) with the exit methaneconcentration of 9.5 mol %. The tubewall temperature is usually 960 �C (1765 �F).The design Dp= 345 kPa (50 psig). The reformer works on outlet temperature con-trol. Which of the following observations accurately describe the situation:

a) Nothing different; everything works fine because the controllers adjust toyield the same exit temperature.

b) The tube wall temperature will be about 30 �C higher because the controllersadjusted to yield the same exit temperature; the exit methane concentrationis 7.5 mol %.

c) The tube wall temperature will be about 23.5 �C lower than usual because ofthe cooling effect of the larger mass of gas flowing through.

d) The inlet gas temperature will increase by 38.2 �C because of the controlleraction.

e) With controller action, the methane at the exit will increase to 10.5 mol %and the tube wall temperature will remain 960 �C.

f) With controller action, the Dp will be about 500 kPa.g) With controller action, the exit product gas temperature will be 20 �C lower

than design.h) Other.

171

Figure 5-2 A steam reformer.


5.1.4Feedback, Self-Assessment

Self-assessment is based on written evidence. Throughout this activity there are sev-eral times when you are asked to write reflections and complete forms. This is anecessary part of the skill development process.

The evidence can include:

. the problem statement you used when being the talker and the marks, under-lines and notations you made directly on this.

. the paper you worked on.

Other, more structured evidence includes:

. your reflections you made after you did the task, Worksheet 5-1;

. the feedback from the listener about the task, Worksheet 5-2.

. the feedback from the talker to the listener about the role playing, Worksheet5-3.

172

Worksheet 5-1: Reflections: A place for you to record your ideas about the TAPPSMethod:

Being the talker: What did you enjoy most? What was most difficult about thetask? What did you discover by being the talker? What did you discover frominteracting with a listener? What were your strengths? Focus on accuracy? Veryfew silent periods? Being active? Good communication?

________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Being the listener: What did you enjoy most? What was most difficult aboutthe task? What did you discover by being the listener? What did you discoverabout problem solving by comparing the talker’s approach to yours? What wereyour strengths as listener? Quality of your prompts and degree of interaction?Tone of interaction? Good communication? Non-intrusiveness?

________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

5.2 Strategies 173

Worksheet 5-2: Feedback from the listener to the talker about the process used.

Awareness ____________________________ ___________________________problem listener

Number of silent periods 0 1 2 3 4 5 >5Number of checks, double checks >5 5 4 3 2 1 0Amount of writing/ charting >5 5 4 3 2 1 0Comments:______________________________________________________________________________________________________________________________________

Validated by: ________________________________________________________

Worksheet 5-3: Feedback to the listener:

problem _______________________ listener __________________________

I found the listener:

. The quality of the comments:

–10 –8 –6 –4 –2 J –2 –4 –6 –8 –10

too passive little too passive about right a little interruptive too interruptive

. The attitude displayed:

–10 –8 –6 –4 –2 J –2 –4 –6 –8 –10


. The listener’s emphasis was listening to me &; helping me verbalize &;helping me solve the problem &; solving the problem for me &.

validated by talker __________________________________________________

5.2Strategies

A strategy is an organized approach used in solving a problem. Such an organizedapproach identifies steps or stages for different parts of the process. Based on a sur-vey of the cognitive literature and a critique of over 150 published strategies, theMPS 6-stage strategy, given in Figure 2-3, is a good working strategy to use. A strat-egy is important because:

. we all usually use one,

. a strategy helps us to be organized and systematic,


. having a strategy helps calm us down if we become anxious when we aregiven a very difficult problem to solve,

. having a strategy helps us to “monitor” our mental processes.

One of the more effective ways to build our skill and confidence with the use ofstrategies is an extension of the Whimbey TAPPS approach used in Section 5.1.This is an extension of that experience in that the talker is also asked to move a mar-ker to indicate the stage in the strategy in which he/she is working and asked toexplicitly say monitoring statements frequently. The listener’s role is expanded toinclude recording the minutes the talker spends working in each of the stages in thestrategy and recording monitoring statements made by the talker.

In this section we list the target skills, describe the roles, give the activities andthe feedback forms.


Research in problem solving has uncovered the following behaviors of skilled prob-lem solvers as they relate to the use of a strategy:

. Spend time reading the problem statement (up to three times longer thanunsuccessful problem solvers).

. Define the problem well; do not solve the wrong problem. Are willing tospend up to half the available time defining the problem. Most mistakesmade by unsuccessful problem solvers are made in the define stages.

. The problem that they solve is their mental image of the problem; such amental image is called the internal representation of the problem.

. Differentiate between exercise solving and problem solving (that were illus-trated in Figures 2-1 and 2-2.)

. Use an organized strategy that focuses on defining the real problem, whereasunsuccessful problem solvers tend to search for an equation that uses up allof the given variables or given information (regardless of whether it appliesto the situation or not).

. Define the real problem with a focus on key fundamentals whereas unsuc-cessful problem solvers tend to memorize and try to recall equations and so-lutions that match the description of the situation.

. Break “Defining the problem” into three separate activities to avoid errors.These are 1) listen, read, get initial information and manage both distressand panic (called Engage), 2) classify the initial information into the “statedgoal or task to do”, “constraints and criteria”, and “description of the situa-tion”, without trying to “define the real problem” (called Define the statedproblem), and 3) create a rich, internal image of the problem as seen frommany different perspectives and evolve a definition of the real problem(called Explore).

. Use a strategy so as to be systematic and organized, whereas unsuccessfulproblem solvers tend to take a trial and error approach.

174

5.2 Strategies

. Are aware that a strategy consists of a series of about six stages with eachstage using different thinking and feelings. This strategy is not used serially(following rigidly one step after another). Rather it is used flexibly; appliedmany times while solving a single problem with frequent recycling from onestage to another.

. Problem-solving skill draws on subject knowledge (needed to solve the prob-lem) and with the sample solutions (from past-solved similar problems) aswas illustrated in Figures 2-1 and 2-2.

. Monitor their thought processes about once per minute while solving prob-lems.

In summary, the goal of this activity is to develop your skill and confidence in

1. Extending and reinforcing the skills addressed in Section 5.1 on Awareness.2. Recognizing patterns in the problem-solving process.3. Realizing that a “strategy” is not applied linearly and sequentially; that it is

used flexibly.4. Recognizing the difference between problems and exercises.5. Understanding the relationship between subject knowledge, past solutions to

problems and problem solving.6. Acknowledging the importance of defining problems and to recognize this as

a three-step process.7. Acknowledging the importance of reading the problem statement.8. Realizing that problem solving is not “doing some calculations.” Conversely, to

correct the misconception that if you are not “doing some calculations” youare not solving problems.

9. Acquiring skill in explicitly monitoring the process.

5.2.2The Extended TAPPS Roles: Talker+ and Listener+

Talker/problem solver+: This is an extension of the pairs activity with roles describedin Section 5.1.2. Do all of the activities described in Section 5.1.2 and now the talkeralso

1) moves a marker on a strategy board, please use Figure 2-3, p. 21, to indicatewhich of the 6 stages you think is being addressed. By considering the loca-tion of the marker, the listener should agree 80% of the time that the activitiesyou describe are consistent with the stage represented by the marker. Youshould need prompting no more than three times in a ten minute period.

2) tries to say frequently such monitoring statements as “Where am I?” “Have Ifinished this?” “If I calculate ..., what will that tell me?” “If I ask this question ...,what will that tell me?”” Where do I go next?” “Can I check this?” If a hypothesisis shown to be wrong or if you calculate a “strange answer.” then ask “OK, Whatdid I learn from that?” You should exhibit four verbal management statementsduring a ten minute period of problem solving.

175


This is difficult to do. Be patient with yourself. You may not completely under-stand the meanings of the stages yet. You may use different stages than the ones onthe Strategy Board. Please, do your best. The listener will not move the marker foryou. The listener will not tell you what stage you are in. The listener might ask you“Are you still in the “Explore” stage?” Remember to keep talking, to be active, to usepencil and paper, and to check and check again. Before you start, go over the mean-ings of the six different stages in the McMaster-6-Step strategy with the listener.Agree on the meanings of the words.

1. Sit side by side; have paper and pencils available, have the Strategy Board andthe marker.

2. The talker moves the marker to the Engage part of the Strategy Board andstarts by reading the problem statement aloud.

3. Then move the marker to whatever stage you are going to work on next andstart to solve the problem on you own. Keep talking aloud. You are solving theproblem. Your partner is only listening to you. He or she is not solving theproblem with you or for you.

4. It is not easy to talk, think and move the marker at the same time. You mightforget to move the marker. That is OK. Do the best you can.

5. Go back and repeat any stage of the strategy you wish.

Listener+: This is an extension of the listener role described in Section 5.1.2.Here you are to record the amount of time the talker spends in each of the stages

on the Strategy Board. Do not correct them; do not argue with them about whichstage the talker is working on. Do not move the marker for them. You may have toask “Are you still in the ... stage?”

In all that you do, your interventions will be judged by the talker to be helpful, andnot judged to be disruptive. The worksheet to be used to record the evidence for thetalker is given in Worksheet 5-4; some example data are given in Figure 5-3. Add a .to the chart whenever a monitoring statement is said.

5.2.3Activity 5.2: (time 35 minutes)

Allow the same timing as used in Activity 5.1, in Section 5.1.2. The physical arrange-ment for the ten- minute talk period is illustrated in Figure 5-4.

The Tasks may be selected from Tasks 5.1.A to E or from Task 5.2.A includingoptions from Appendix F, Sections 5.1 and 5.2.

In pairs, one be a talker, the other be a listener. The talker plays the role for 10 min-utes and “solves” problems during all the allotted time. Do not change roles.

Use Worksheet 5-1 for reflections and Worksheet 5-5 for feedback about listening.The Worksheet 5-4 will be validated by the listener and given to the talker as evidence.

176

5.2 Strategies

Figure 5-3 (top) Some example data on the Worksheet.

Figure 5-4 (bottom) The physical arrangement: the listener isshown on the left; the talker, on the right.

177

5 Polishing Your Skills: Problem-Solving Process178

Worksheet 5-4: Record of the talker’s strategy with . for monitoring statements.

Talker __________________ Case _________ Listener ____________________

Stage

Engage: “I want to and I can!”

Define-the-stated problem: Sort the givenproblem statement

Explore the problem to discover what theproblem really is

Plan

Do it

Look back: elaborate, check

0 2 4 6 8 10 12

Time, minutes

Task 5.2.A: Terry Sleuth in the Poly RoomTerry Sleuth ventured into the polymerizer room on the way to the R&D lab for anappointment with Bill Wright. Seeing Terry, Charlie, the poly-room engineer, called“Hi Terry, please come over here and help us sort out this mess. Look at these reac-tors. We’re continually losing quality product in reactors two and three but not onreactor one.”

“Tell me more,” encouraged Terry.“The problem is that just near the end of the run, the motors driving the mixers

overload and cut out. They stop! Then the whole batch is ruined because the heattransfer is insufficient,” explained Charlie.

“Does it happen only on reactors two and three and not on reactor one?” askedTerry.

“Yes. Reactors two and three have calandria coolers and marine propeller mixers.Reactor one has an internal coil cooler and a turbine mixer. The mixing flow pat-terns are chosen specifically for the heat-exchanger configuration: the propellermoves the liquid up through the tubes with the coolant on the shell side of the inter-nal calandria; the turbine shoots the liquid out onto the tubes with the coolant insidethe tubes. Perhaps the calandria tubes are plugged with polymer,” continues Char-lie.

Terry thought for a moment and then asked “Are both cooling systems about thesame surface area and serviced by the same cooling water?”

“Yes.”

5.2 Strategies

“Has this problem ever occurred before?”“This is the first time we have processed this product on any of these reactors; we

have processed other products for years on all three reactors and with great success,”beamed Charlie.

Terry smiled. Terry had all the information needed to identify the cause. What didTerry say?

Task 5.2-B: Terry Sleuth and the Case of the Delinquent DecanterRing ... The incessant ringing of the telephone disrupted the steady hum of the engi-neering office. Betty answered, looked perplexed, looked across at Terry Sleuth’sdesk and knit her brow as she carried on an animated conversation. Terry couldalmost guess from Betty’s reactions that the call was about the hexane-water decan-ter on the soybean-miscella still. The Ministry of Environment had been after us forthe last week to get the concentration of hexane in the wastewater from the decanterdown to acceptable levels. Unfortunately, the decanter wasn’t working well. Bettyhung up, donned her hard hat and headed toward Terry’s desk. “It’s the decanteragain, isn’t it!” said Terry. “You’ve said it”, replied Betty disheartenly. “Before wehead out to see the beast, let’s review what we know”, suggested Terry encoura-gingly. “OK”, Betty said and went on “the decanter is a vertical cylinder with a con-ical bottom; the feed enters about midpoint with exit lines at the top for the hexaneand at the bottom for water. The feed is a mixture of about 60% hexane and 40% hotwater. The foam or droplet layer fills the center band of the decanter; the drops coa-lesce to form pure layers of hexane (that rises to the top) and a pure layer of water.There is a cover on the decanter that has a vent. The last time we talked we realizedthat some hexane dissolvers in the water and some droplets of hexane may go outwith the water because the drops haven’t coalesced. How’s that?” Terry replied,warmly, “Very good. However, we were unsure as to whether the feed is hexanedrops in water or water drops in hexane. Since there is more hexane, it is probablethat the feed is hexane containing droplets of water. So it is the water drops that arecoalescing.” “Yes, but I don’t see that that makes any difference whether it is hexanedrops coalescing or water drops coalescing. What I have done, since we last talked,is a lot of research about decanters. Mizrahi and Barnea report in a refereed articlethat for every 10 �C increase in temperature, the coalescence time will be faster by afactor of two. I think I’ll rig up a heater on the feed line so that the incoming tem-perature of the feed is 70 �C instead of 60 �C. That should fix this baby!” gloatedBetty. Terry paused, checked the Handbook of Chemistry and Physics and said,“Don’t be too sure.” What did Terry look up and why did Terry cast doubt on Betty’sidea?

5.2.4Feedback, Self-assessment

Self-assessment is based on written evidence, not on intuitive feelings. Throughoutthis activity evidence has been gathered. The evidence includes:

179


. the problem statement, with all your underlining and notations,

. your paper you worked on,

. the time-stage-monitoring evidence of Figure 5-4,

. the reflections, similar to Worksheet 5-1 but with prompts related to thisactivity,

. feedback about your role as listener, Worksheet 5-5. You might elect to provideadditional feedback to the talker from Worksheet 5-2 about the process used.

180

Worksheet 5-5: Evidence for listening: Feedback to listener & the listener willencourage verbalization, an emphasis on accuracy, active thinking and encou-rage the problem solver to move the marker correctly on the strategy board. Yourinterventions will be judged by the problem solver to be helpful, and not judgedto be disruptive.

Activity 1: Talker ____________ Case ___________ Listener _____________

encourage verbalization: not needed interruptive OK really helpedencourage emphasis on accuracy: not needed interruptive OK really helpedencourage active thinking not needed interruptive OK really helpedinterventions: not needed interruptive OK really helped

Comments:______________________________________________________________________________________________________________________________________

5.3Exploring the “Context”: what is the Real Problem?

During the Explore stage, we may wish to place the problem in a larger context. Avery useful technique is Basadur’s Why? Why? Why? technique (Basadur, 1995). Inthis technique, we start with the initial “problem”, ask Why? and redefine the prob-lem in a broader context. The process is repeated. After we have completed severallevels of context, we then can address “which level of generality is best, given thecurrent constraints and contexts.”

This approach was used very successfully by Pierre in the Case’4, the case of thePlatformer Fires, Section 4.3. Consider first an example and then an activity.

5.3.1Example

Here we illustrate the application of the approach for Case ’7, the Case of the reluc-tant crystallizer, problem 1.5 in Chapter 1 and considered in Frank’s approach inSection 4-5.

Beginning symptom: “the liquid level in the crystallizer is dropping at a fantasticrate.”

5.3 Exploring the “Context”: what is the Real Problem?

Rephrase as Why do I want to stop the liquid level in the crystallizer from drop-ping at a fantastic rate. OK. Perhaps your answer is “so that I can produce qualitycrystals from the crystallizer.” OK. Now ask Why? Now your answer might be “Sothat I have quality crystals to sell to my customers.” This process continues. It isconvenient to summarize this are follows:

Why? so that have Happiness and Bliss›

Why? so that have me a rich and productive life;›

Why? so that have challenging and exciting employment;›

Why? so that keep the company profitable;›

Why? so that keep sales healthy,›

Why? so that have quality crystals to sell to my customers;›

Why? so that produce quality crystals from the crystallizer›

Start: fi stop the liquid level in the crystallizer from dropping at a fantastic rate.Why?

In this case, perhaps the most important problem to address is “to have qualitycrystals to sell to my customers.” That being the case, Frank should have addressedhis efforts toward obtaining the quality and quantify of crystals, even if it meansbuying them from a competitor. Thus, this technique helps us to see the problem ina bigger context and prompts us to ask “what is the real problem?”.

5.3.2Activity 5-3

For Case’9 The bleaching plant, do aWhy? Why? Why? analysis.

Case ’9: The bleaching plant.Our process makesmargarine. One step in the process is to remove the odors and colorbodies from the edible oils used to make margarine. This removal is done in the blea-cher illustrated in Figure 5-5. The light volatiles are removed by subjecting the oil tohigh vacuum. The color is removed by adsorbing the color species on powdered adsor-bent that is subsequently filtered from the oil. The bleacher is shown in the diagram.The procedure, as set out in the startup manual, is to charge the vessel with edible oil,isolate the bleaching vessel (by turning the appropriate valves), draw a vacuum bymeans of the steam ejector system until the absolute pressure is 20 kPa abs on the ves-sel, open the valve connecting the bleach vessel to the hopper filled with the adsorbingpowder, Fuller’s earth, and suck (or pneumatically convey) the powder into the bleachvessel. The physical arrangement, including the approximate elevation, is shown in thediagram. This is a batch process.

181


PI

1

Oil feed

Hopper of

Fuller's earth

Cooling coils

Light

To filter press to remove

Fuller's earth

Steam

View

Figure 5-5 The bleacher for Case’9.

The gauge on the vessel read 13 kPa abs, but when we opened the valve to conveythe powder into the bleacher, nothing happened. That is, we expected to see,through the view port, the powder dumping into the liquid in the bleacher. This isthe first time this plant has ever started up. The product has already been sold andbecause of previous delays in startup we are now losing $15 000 per day.

182

________________________________________________________________Why? ›

________________________________________________________________Why? ›

________________________________________________________________Why? ›

________________________________________________________________Why? ›

________________________________________________________________Why? ›

Start fi _____________________________________________________________

Reflect on this approach. Was it easy to do? What insights did you gather? Whatmight you address as the real problem? When might you use this approach?________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

5.4 Creativity

5.4Creativity

Whether solving exercises or problems we need creativity to generate differentpoints of view (for the creation of a rich internal representation), to create hypoth-eses about change and to create hypotheses based on the basics.


Skilled creative thinkers:

1. Defer judgment;2. Are succinct;3. Can list 50 ideas in 5 minutes;4. Create a risk-free environment;5. Encourage free and forced association of ideas;6. Can piggy back on previous ideas;7. Use triggers, such as those listed in Table 5-1, to maintain the flow of ideas;8. Aren’t discouraged. In the last two minutes of a five-minute brainstorming

session, over 85% of the ideas are not practical. But, they spend time to iden-tify the treasures among the 15%;

9. Are positive;10. Manage stress well. Manage any negative self-talk;11. Use impractical and ridiculous ideas as “stepping stones” to innovative, prac-

tical options.

After they have generated a large list of ideas, skilled creative thinkers then listmeasurable criteria and select at least five technically feasible hypotheses.

Table 5-1 Checklist of triggers for brainstorming.

Name of trigger to change point of view Elaboration of how it is tobe used

Comments

To be used forobjects

“How to improveion-exchange resin”

Function What is the function ofthis object? How else canwe achieve this function?What function cannot it do?

Objects for engineers are“products” and hardware.

Physical uses What are its physical prop-erties and characteristics?What are they not? Howelse might we obtain oruse these physical proper-ties?

Probably the most usefulperspective; use first. Thenegative view is often ex-tremely illuminating.

183



Comments

Chemical uses What are its chemicalproperties? How can thesebe exploited? What arethey not?

Easy for us to use; usesecond.

Personal uses What personal uses canwe make of this object?

These three give a uniqueperspective that is oftenoverlooked.

Interpersonaluses

What interpersonal usescan be made?

Aesthetic uses How can we createpictures? music? artisticcreations with the object?sculpt? weave?

Mathematicalor symbolicproperties

What are the mathematicalor symbolic uses? Whatcan they not be used for?What are they often con-fused with?

Only applicable for certainobjects or ideas.

Situations

“I need to design anion-exchange unit”“The reformer is notfunctioning; fix it.”

Checklist Of the many checklistspublished SCAMPER isprobably the mosteffective:S: substitute who? What?Other processes? Otherplaces?C: combine purposes?Ideas? Appeals? Uses?A: adapt what else is likethis, new ways?M: modify, maximize,minimize?P: put to other uses, otherlocationsE: eliminate,R: reverse, rearrange

Effective for objects andsome situations. Try eachviewpoint. Easy to do.

184

Table 5-1 Continued.

5.4 Creativity


Comments

Wildest fantasy Think of the craziest ideaor use.

Need to establish self-confi-dence and group confi-dence before really outland-ish ideas are presented.Build confidence in themerit of this approach bylater bridging to uniqueideas. Brings laughter andtension relief.

How naturedoes it

Identify the situation andthen think of how naturefulfills this function.Bridge to engineeringreality.

Great potential; depends onthe situation. Successful indesigning new bridges;new bog machines.

What if? In theextremes

Extrapolate from theunfamiliar to a simplifiedversion

Extension of problem-solving skill in definingsimple problems. Relativelyeasy to apply.

Boundaryexploration

Identify the constraintsand remove them

Much easier for some to dothan for others.

Functionalanalogy

How else is the functionachieved?

Appearanceanalogy

How else might we get theappearance of this situa-tion?

Morphology Break the problem situa-tion into a series of parts.For each part list 10+options. Then, systemati-cally combine one optionfrom each part and askwhy not?

More mechanical; easy tocomputerize; most of thework is in setting up theparts and the options. Funand surprising to see someof the results.

Symbolicreplacement

Replace the original prob-lem by an interesting ideagenerated in the session;refocus.

Occurs naturally in manybrainstorming sessions.Add this trigger explicitly ifneeded.

Juxtaposition Bring in three completelyrandom words and bridgeto current situation.Example “refrigerator,light switch, clock.”

Very effective. Don’t fake itby using words “you thinkwill help”. Use randomwords.

185




Comments

Personal analogy Imagine yourself-as partof the situation. Describeyour feelings and whatyou are experiencing. Beimaginative.

Tricky. Works for some peo-ple; not for others. Effectivewith fluid-dynamic prob-lems.

Reversal Do the reverse. More challenging than onethinks. Focus systematicallyon the reverse of differentelements of the situation ata time.

Book title Create a title for a “best-selling” novel. The titleshould sum up the currentsituation. “Will exchangingbring happiness to MaryLou?”

Interesting. Worth a try for1 minute.

Letter, word,sentence.

Focus on three differentlevels of detail correspond-ing to the big view (asentence), an intermediateview (a word in the sen-tence), and a letter (in aword). In ion exchange thismight be “the ions, theresin, the packed bed, theseparation”.

Very effective for most ofour problems. Consider a5-minute brainstorm ateach level.

Visual image Look at three or fourfamous paintings.Describe aloud whatyou see and makeconnections.

At first glance this soundstoo bizarre. However, forsome this is one of themore effective ways of see-ing the situation from anew perspective. Difficultto do out on the plant.

5.4.2Example: Case ’10: To dry or not to dry! (based on Krishnaswamy and Parker, 1984)

Our product is dry crystals of calcium nitrate. The crystals are precipitated in a continu-ous crystallizer. The exit flow from the crystallizer – consisting of a slurry of crystals andmother liquor – goes to a fixed, parabolic screen (DSM) to separate the large, productcrystals from the undersized crystals that are recycled. The undersized are pumped con-

186


5.4 Creativity

tinuously through a hydrocyclone and returned to the continuous crystallizer. As screen-ing progresses the screen gradually blinds because the finer material that gets stuck inthe screen cloth causes toomany fines to be carried over with the large product crystals.Hence the screen is operated batchwise. The feed to the screen is stopped; the screen iswashed, and then the screen is put back in service.

The large crystals from the screen go to a scraper-discharge centrifuge. The opera-tion of the centrifuge is continuous but the moist crystals from the centrifuge aredischarged batchwise because the centrifuge goes through a filter-wash-peel/dis-charge cycle. The exit from the centrifuge goes directly to a continuous, steam-tuberotary dryer via a chute and a conveyor screw. In the past, the crystal washing in thecentrifuge was not very efficient. They modified the cycle to provide more washwater in the washing cycle of the centrifuge. The system is shown in Figure 5-6.

“The crystal product from the rotary dryer has a moisture content of 4.5%whereas the design value is 1.5%. Cake seems to be building up in the dryer feedchute, in the feed screw and on the steam tubes at the feed end of the rotary dryer.”Fix the problem. Get the dryer working again so that the exit crystal moisture con-tent is always below 1.5%.

Steam

dry solids

Continuous

steam-tube

dryer

Centrifuge

Parabolic

screen

Feed

cyclone

Recycle

fines

Overflow

supernatant

Liquid

Figure 5-6 The system used to dry the crystals; Case’10.

Example results of brainstorming:Too much water in wash cycle for dryer to handle, cake buildup on steam tubesdecreases heat transfer, cake dries up too early causing buildup while the rest of thecake is very wet, not enough steam supplied, wrong centrifuge, wash water carryoverfrom the centrifuge, steam leak into dryer, it’s raining, cycle wrong in centrifuge,cycle from screen not coordinated with cycle from centrifuge, crystals too wet fromthe screen, washing from the screen contaminates the feed to the centrifuge, feed

187


from the screen too wet, fines carryover to the centrifuge causing blinding in centri-fuge, peel cycle too short, peel cycle too long, filter cycle too short, filter cycle toolong, cycles now out of synchronization between the screening and filtering, feederfrom screen to centrifuge broken, feed screw to dryer rotating too fast, rotating tooslow, the rotational speed of the dryer has increased, the residence time in the dryerhas decreased, the steam pressure has changed, the quality of steam from the boilerhouse has changed, the condensate trapping on the dryer is backing condensate upand reducing the heating area, air is blanketing the heat-transfer area in dryerbecause air was not bled off before restarting dryer, pitch of the dryer has changed,operating procedure has changed, pump from the screen bottoms malfunctioningand water spilling back into feed to centrifuge, cycle on the parabolic screen hasbeen increased, more fines into peeler centrifuge causing poor filter cycle.

Try trigger “craziest”: operators not reading instruction correctly, sandwich droppedin line and plugging flow of crystals, wash down water used to clean the floors issplashing onto feed screw, the dryer is turning backwards, the peeler cycle isreversed to peel-wash-filter, the feed to the centrifuge is going directly to the dryer,feed from the crystallizer is going directly to the dryer, for this phase of the moonthe vampires are out (is that crazy enough?).

Try trigger “reversal”: focus on the product from the dryer is too dry: possible reasonsare: long residence time, higher steam pressure/temperature, excess drying area,feed to the dryer from the centrifuge is dryer than expected, centrifuge rpm fasterthan usual, washing of the cake in the centrifuge is done with new solvent that eva-porates faster, feed to the centrifuge from the screen is dryer than usual. OK, for thecurrent problem take the opposite of all these.

Try trigger “juxtaposition” using random words elastic bands, stapler and coffee cup.The process I will use is to take characteristics of this idea and find links to possible feasi-ble ideas: I’ll show that as arrows fi and keep going until I get an idea that I think isfeasible.

Elastic bands: stretches fi returns to original shape fi screen becomes com-pressed and later pops back

Elastic bandsfi made of rubberfi gaskets leaking?Elastic bandsfi long and narrowfi crystal size changesStapler: batchwise operation fi dryer should be run batchwise tooStapler: runs out of staples, need to refill periodically fi centrifuge operates but

has run out of feedCoffee cup: holds liquid fi liquid condenses near feed end of dryer and drops

onto feed; recycle of wet damp gas in dryer.Coffee cup: handle fi holds fi liquid retained in crystals by interstitial forcesCoffee cup: handles hot liquids fi wash water is too hot; hot condensate clogs

dryer tubeCoffee cup: status symbolwith logos embossed on itfiwordsfi instructionswrong.Coffee cup with metal: can’t put in the microwave fi contamination from rust,

plugging linesCoffee: brown liquid fi dirty wash waterCoffee: tastyfimicro-organisms growing in centrifuge, inwashwater, in crystallizer

188

5.4 Creativity

Coffee: justification for breaks fi operators don’t pay attention when they should;breaks in the cycles between the screen and the centrifuge.

Coffee: social eventfi water clings to crystal, not water anymoreSelect the craziest as a “stepping stone” or “piggy back” to feasible idea.”: Crazy idea:

“for this phase of the moon the vampires are out” The process I will use is to take char-acteristics of this idea and find links to possible feasible ideas: I’ll show that as arrows fiand keep going until I get an idea that I think is feasible.

Phase fi sinusoidal motion fi cycles fi periodicity fi the cycle of the screeninterferes with the cycle of the centrifuge fi periodic pulses of wet stuff to dryer.

Moon fi night fi cool at night fi is there a temperature effect on the feed? thefeed to the dryer is too cold; is the steam too cold? is the wash water in the centrifugetoo cold?

Moon fi pale light in the dark fi hard to see fi difficult to see the moisture con-tent in the feed because it changes so fast fi things are happening on this plant thatare hard to see fi slow cycles and feedrate down to each section of the plant in turnand try to “see better”

Moon fi changes shape from crescent to circle fi the crystals change shape sofiltering is different

Moon fi changes shape from crescent to circle fi crescent fi curve fi the curveof the parabolic screen is too steepfi decrease in capacity

Moonfi changes shape from crescent to circlefi changes fi same ideas as listedabove for Phase

Moon fi land on the moon fi long distance away fi something a long distanceaway from the dryer is causing problemsfi utilities? steam? wash water?

Moon fi land on the moon fi long distance away fi something a long distanceaway from the dryer is causing problems fi feed to crystallizer? pregnant liquor?crystallizer? screen? hydrocyclone? pump to hydrocyclone? screw feeder to dryer?centrifuge?

Moon fi land on the moon fi long distance away fi something a long distanceaway from the dryer is causing problems fi head office giving new policies andstressing out employees fi spouses stressing out the operators? fi events stressingout the operators fi operators making mistakes

Moon fi land on the moon fi long distance away fi something a long distanceaway from the dryer is causing problems fi vendors supplied faulty equipment?lousy wash system?

Moonfi land on the moonfi different environment on the moon, lack of oxygenfi oxygen in wash water affecting crystals? contaminant in wash water causing poorwashing? is it the washing cycle? of the drying cycle in the centrifuge?

Vampire fi person fi operators error?Vampire fi person fi operators fi operating instruction error?Vampire fi teeth fi sharp objects breaking up crystal size fi slower filtering and

slower drainage rate in centrifuge.Vampire fi drinks blood fi blood fi operator injured? water is not water, it’s con-

taminated with low vapor pressure.

189


Comment: This session was characterized by:

. mix of symptoms and roots causes;

. lots of duplication;

. most feasible ideas in the first 32 ideas;

. a lot of impractical ideas, especially later in the session but there were somenew feasible ideas amongst these.

Total ideas: 98 with 32 from the initial burst, trigger “craziest” = 8; trigger “rever-sal” = 8; juxtaposition= 20; stepping stone= 30. The stepping stone seemed to workwell for me on this problem.

From this are selected the following possible hypotheses:

1. not enough steam,2. wash water carryover from the centrifuge,3. cycle from screen not coordinated with cycle from centrifuge,4. feed crystals too wet from the screen,5. rotational speed to the dryer has increased,6. more fines into peeler centrifuge causing filter cycle to be too long; fines car-

ryover to the centrifuge causing blinding in centrifuge,7. centrifuge rpm faster than usual,8. crystal size change; crystals change shape so filtering is different,9. centrifuge operates but has run out of feed,10. dryer feed too cold,11. vendor supplied faulty equipment.

– slow down the cycles and feedrate down to each section of the plant, inturn, and try to see better what is happening.

– “slower filtering and slower drainage rate in centrifuge” symptom! use aspotential brainstorming idea.

– “periodic pulses of wet stuff to dryer” symptom! not a root cause butmight be profitable as symptom.

5.4.3Activity 5-4

For Case ’9, the bleacher (described in Section 5.3) brainstorm 50 possible causesand write these down. Suggested time frame: initial write 6 min.

. Say to yourself-“I can do this.” reread the problem statement: 2 min.

. Try trigger “wildest fantasy” 1 minute

. Try trigger “what if? in the extremes. 1 minute

. Try trigger “juxtaposition” using the words pencil, hat, and perfume. 1 minuteeach word.

. Try trigger “reversal” 1 minute.

Stretch.

190

5.5 Self-Assessment

Look over your list and select the “craziest one”. Use this idea as a “steppingstone” to a technically feasible one. By this we mean, take the properties of the crazyidea and use these in a feasible idea.

Reflect on what happened. Complete the feedback form given in Figure 5-7.

5.4.4Feedback, Self-Assessment

Self-assessment is based on written evidence, not on intuitive feelings. Throughoutthe brainstorming evidence has been gathered. The evidence includes the numberof ideas produced and the succinctness of the ideas.

In addition we can write reflections, using a form similar to Worksheet 5-1. Fig-ure 5-7 gives a convenient summary form specific for brainstorming.

Figure 5-7 Feedback form for brainstorming.

5.5Self-Assessment

Your skills and confidence as a trouble shooter will improve the most if you create aformal approach to self-assess your current skill, set goals and gather evidence aboutyour progress to achieving the goals. Skilled self-assessment develops objectiveawareness of how you perform a task and develops self-confidence. In this section

191


we define the target skills for self-assessment, in Section 5.5.1, and then proposeactivities for growth in Section 5.5.2.


Skilled assessors realize that assessment is:

1. about performance; it is not about personal worth.2. based on evidence; it is not based on wishful thinking or gut feelings.3. essential for growth.4. not possible without published unambiguous goals and measurable criteria.5. based on a wide variety of different types and forms of evidence.

5.5.2Activity for Growth in Self-Assessment

Attitude and skill – these are the key elements to work on.Assessment involves a change in attitude. Many think of exams, performance

reviews and evaluations as being stressful and something to be avoided. We need torealize that self-assessment helps develop growth and is one of the most positiveactivities we could do to improve trouble-shooting skills. Assessment is about perfor-mance! Assessment is based on evidence and is a judgment done in the context ofpublished goals and with measurable criteria.

The three skills needed relate to creating goals, measurable criteria and designingforms of evidence

a) Ability to identify and create observable and unambiguous goals. Exampleshave been given in each section. In general unambiguous means that we canobserve the performance. Such words as “know”, “create” are unacceptablebecause we cannot observe someone demonstrating that they “know this”.Words like “list”, “write out” are an improvement.

b) Ability to identify and create measurable criteria related to the goals. Thistask is difficult, boring and tedious – but necessary. For example, in Section5.4 the goal was to list ideas. To make that measurable we need a numberand a time: 50 ideas in 5 minutes.

c) Ability to write out, gather and evaluate evidence as it relates to the goals. Asdemonstrated previously in this book, evidence in the form of reflections, theworksheets and feedback forms are all useful. Usually, once the goals andcriteria are created, it is relatively easy to create the forms of evidence.

Activity 5-5 Self-assessmentBased on the self-assessment in Section 1.3 create goals for growth, measurable cri-teria and pertinent forms of evidence. Begin gathering a collection of evidence andsystematically go over the evidence and self-assess your progress.

192

5.5 Self-Assessment

5.5.3Feedback About Assessment

Figure 5-8 provides a feedback form related to the assessment process.

Figure 5-8 Feedback about the assessment process.

193


5.6Summary and Self-Rating

In this chapter we identified five skills related to problem solving. These were aware-ness or the ability to describe your problem-solving processes; strategies or the abil-ity to see patterns in the process; exploring the context using the Why? Why? Why?process; being creative and being skilled in self-assessment. For each target skillswere listed, some examples were provided, activities to develop the skills weredescribed and forms of evidence were given.

Reflect on what you have experienced in this chapter:________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Rate I thinkI amskilled

I have someevidence andhave someskill

I am confident.I know the goalsthe criteria andhave a varietyof evidence

Notsurethis isfor me

Awareness: can describe, focus on accuracy,am active & & & &Strategies: see patterns, monitor & & & &Explore the context using Why? Why? Why? & & & &Am creative: 50 in 5 min, use triggers, & & & &can step from crazy ideas to feasible ones & & & &Self-assess:

attitude about performance; & & & &needed for growth, evidence-basedskill: create observable goals; & & & &create measurable criteria; & & & &identify pertinent evidence; & & & &valid judgment in this context; & & & &

194

195

Critical-thinking skills are needed by trouble shooters. These skills include selectingand designing tests, checking for consistency, classifying sets of ideas or data, recog-nizing patterns and reasoning and drawing valid conclusions.

Here are some examples of where some of these critical-thinking skills are usedin the task of trouble shooting. We:

. See/receive information that suggests something it wrong: compare condi-tions with unsafe conditions and decide from among a) emergency shut-down, b) change to “safe-park” or c) continue current conditions. Select thesymptoms. Separate fact from opinion. The thinking skills needed are consis-tency and classification.

. Gather more information about the current situation and compare and con-trast this with the “expected” performance to obtain a rich mental image ofthe situation. The thinking skill needed is how to compare and identify differ-ences.

. See if there are any patterns in the data. The thinking skill needed is patternrecognition.

. Brainstorm possible faults or causes that could cause the symptoms. Alterthose ideas that are “symptoms” to be “root causes”. Classify the list andapply criteria to prioritize the list of hypotheses/causes that are consistentwith practical experience. The thinking skills needed are classification andconsistency.

. Select hypotheses/faults that are consistent with the symptoms. The thinkingskills are using cause–effect information and classification.

. Create tests; the tests must be pertinent to the hypothesis, and not sufferfrom confirmation bias. The thinking skill needed is how to select valid diag-nostic actions.

. See if there are any patterns or trends in the evidence collected. The thinkingskill needed is pattern recognition.

. Check that the results of the tests confirm and disprove the hypothesis. Thethinking skill needed is reasoning.

In this chapter we consider first, in Section 6.1, how to select valid diagnosticactions. In Sections 6.2 to 6.5, we look at the critical-thinking skills of checking for

6

Polishing Your Skills: Gathering Dataand the Critical-Thinking Process


6 Polishing Your Skills: Gathering Data and the Critical-Thinking Process

consistency, classification, seeing patterns and reasoning. Each is considered in turnand illustrated through examples.

6.1Thinking Skills: How to Select Valid Diagnostic Actions

First, we offer criteria for selecting diagnostic actions; then we provide a structuredlist of possible actions. Then we consider how to perform and interpret the actions.In Section 6.1.3 we explore how to become sensitive to personal preferences, styleand biases that might interfere with the selection and use of diagnostic actions. Con-sider each in turn

6.1.1How to Select a Diagnostic Action

In selecting an action to take, the following criteria are useful:

. Safety first!

. Keep it simple. Pertinent, easy-to-gather information should be gatheredfirst.

. The results should provide the accuracy needed.

. Safety and time are critical.

. Select actions that will prove or disprove hypotheses.

. There is an expense associated with any action or lack of action.

. Stopping production to “inspect” or “change equipment” is usually verycostly.

6.1.2Select from among a Range of Diagnostic Actions

The actions taken differ widely from immediate emergency response, to gatheringinformation to better understand the problem, to testing hypotheses, to immediatecorrection of the suspected cause. In general, the actions selected depend on thetype of TS problem, the TS strategy you elect to follow and your personal style. Astructured list of options includes: a) is it an emergency? b) put the situation in thecontext of recent events, c) understand what should be happening, d) check what ishappening now, e) test hypotheses and, perhaps, f) take immediate corrective action.Each is considered in turn.

a. Safety first! Is it an emergency? What’s the general background information

about this situation?

. Invoke emergency shutdown? Although trouble shooters should know thehazards related to the process, information about hazard, safety, MSDS

196

6.1 Thinking Skills: How to Select Valid Diagnostic Actions

sheets can be obtained from Chapter 3, Section 3.12, the MSDS sheets on fileor from two web sites for MSDS data http://www.msdsxchange.com orhttp://www.msdssearch.net.

. Is “safe-park” a required option?

. Tabulate IS and IS NOTdata.

. Note details about the weather and offsite services: air temperature andhumidity, rain-snow, ice-fog? bay water temperature, river temperature, citywater properties/temperature, tower water, steam and air plant; wastewatertreatment plant.

Comment: Start here and put the process into perspective.

b. Should I be using a TS strategy best suited for “change”? What has happened recently

that might affect the process?

What and when did changes occur:

. changes made to operating instructions, to feed, to specifications, to flow-rates,

. work done on the plant,

. what was done during the maintenance turnaround,

. what was done in routine maintenance.

Comment: Easy to do. You may know this already but check out the details ofexactly what was done. This provides the basis for selecting your TS strategy andsubsequent diagnostic actions.

c. What should be happening: expected, “usual” performance?

This information is usually obtained in the office from files and records, from sim-ple calculations and from simple, astuteWhat if? exploration.

Files and records: consult key files and records and predictions of performance:

. design files and simulation: design basis, fouling factors, assumptions made,specifications and predicted conditions throughout the plant.

. vendor files for the equipment: performance expected, sometimes trouble-shooting information* (* often available in both the engineering office andthe control room).

. operating procedures*.

. commissioning data.

. recent P&ID. Often not available, or if available, not up-to-date.

. recent tests and internal reports.

Data from handbooks and texts in your office.Trouble-shooting files: data relating symptom to root cause with the probabilities

or likelihood, similar to that given in Chapter 3.

197


Calculations or estimations that can be done before special tests are done.

. calculated/estimated pressure profiles: these can be done from simulationsor from order-of-magnitude estimates. Note the pressure difference acrossbarriers to determine the direction of flow if a leak occurs.

. mass balance: predicted by simulation or estimated.

. energy balance: heat in= heat out across exchangers, furnaces, condensers,reboilers. Steam usage with 5 kg organic evaporated by1 kg steam, flametemperature,% excess air.

. thermodynamics: equilibrium conversion, vapor-liquid equilibrium, energychanges for flow across turbines, ejectors and valves.

. rate: use the temperature difference in the reboiler to estimate whether boil-ing is in the nucleate, film or transition regime.

Equipment performance calculations: check that the sizing and general operabil-ity conditions are met. Will the installed equipment do the job expected?

For process control, check the location and type of sensor, type of control used.Exploration from simple “What if?” questions.Sometimes we can quickly eliminate extraneous tests by asking such questions as

“What if you could actually look at the flow patterns inside the vessel?” “What if youcould look at the catalyst surface?” “What if there was no insulation? “What if therewas infinite insulation?”

Comment: This should be a valuable, third level of data acquisition. However, theusefulness of these actions depends on the quality of the documentation available(often it is poor), and of your files, on your skill with order-of-magnitude estimates;and on the information available in the case-problem statement.

d. What is currently happening?

We start with information in the control room and from the operators.In the control room, here are some options:

. read the displayed information and check past records of same variables.Look for trends. Look for interconnects (two or more events that seem to behappening together: temperature increase and conversion decrease).

. check the operating procedures*.

. scan the vendor files including performance data and trouble shooting*.

. talk to the operators about changes, about this shift versus previous shifts,about their hypotheses, about the facts.

Comment: This is a vital source. Meet the operators; find out the details; read theinstruments displayed. This is a must visit before venturing out onto the plant.

e. What is currently happening and gathering information to test hypotheses?

We learn more about what is currently happening i) by putting the process in con-text; ii) by using our senses to gather information from out on the plant, iii) by gath-ering data for calculations, iv) by checking out the sensors and controllers, v) from

198


quick checks with specialists, vi) from more ambitious tests and vii) shut-down activ-ities.

i) Put the process in context. Check with the operators of the utilities and ofupstream and downstream plants.

ii) On the plant visit. Out on the plant we use our senses, do simple tests, checkfor consistency, and look for trends.

Use your senses. Read the instruments, listen, smell, look. Note the position of thevalve stem, look for steam leaks, look at the discharge from steam traps, read themotor amps, check the condition of the insulation, look for rust around a flange,leaks around a gland, look at flows (if they are visible). Hot smells? Listen for noise:indicating flow, cavitation or liquid level in vessels. Use your intuitive feelings.

Simple on-site tests: Shift to a more quantitative perspective: estimate/measure:

. the temperature: use the glove test or a laser or surface pyrometer.

. the humidity or dew point.

. the response of sensors: check the response to a change in set point.

. the signals to controllers/valves: inlet pressure/signal; pressure/signal to thevalve.

. diameter of insulation/ of pipes.

. the liquid flows in accessible drains.

. the exit pressure for zero flow for centrifugal pumps.

. if the bypass valve is open or shut.

. condition of the valves: turn and seal.

Comment: Your senses and simple on-site tests are easy to use and often provideexcellent, key information.

Check for consistency: do multiple sensors agree? are the lab data consistent withthe results from on-stream analyzers? temperature–pressure–composition agree-ment? phase-rule agreement? mass-balance agreement? make-sense agreement(fluids flow from high to low pressure? thermal energy flows from high temperatureto low?) physical-thermal data agree with measurements (pressure enthalpy data forrefrigerant)?

Check for trends: have the temperatures, pressures or yields been changing gradu-ally over the weeks? Is there a trend every 5 minutes? every 15 minutes? every hour?What is the frequency and amplitude of cycling?

Check that the P&ID agrees with the actual process configuration.

iii) Gather data for calculations. Fundamentals underpin the process. Check amass and energy balance. Gather data to estimate performance of equip-ment.

iv) Sensors and controllers. Put the controller on manual. Is there evidence thatthe sensors are working and are accurate? Use temporary instruments tocheck measurements. Request specialists to calibrate the sensors or tune thecontrollers.

199


v) Quick checks with specialists. Consultants: call vendors, original designer,consultants, licensee of process and/or suppliers of raw materials, adsor-bents, catalysts.

Comment: the operative word is “quick”. Although it’s better to talk by phone thane-mail, the key people may not be available to answer your call.

vi) More ambitious tests. Samples of gas, liquids or solids for analyses: obtain thesamples: Are the samples representative? Valid? Is there a time dependency?Are all the samples correctly labelled? Can the sampling be done safely? Typeof analysis: often chemical or particle-size analysis.

Specialists: Should a velocity profile be measured? gamma scan? tracer study?Perhaps a well-designed set of experiments should be run and the results ana-

lyzed statistically.Comment: although the process can continue to operate, these tests take time and

may be expensive.

vii) Shut-down-type activities. Open and inspect: shut the process down, safely iso-late the piece of equipment, and clear of the process fluids and vapors. Knowwhat you are looking for, but be prepared for surprises.

Comment: this requires that the process be stopped. This is expensive. Try to leavethis action as the last resort. Much can be learned from the previous activities.

f. Possible corrective action

Some trouble shooters, especially those who prefer action to patiently gathering evi-dence (dominant J behavior as described in Section 6.1.3c), take “corrective” actionprematurely. Others might astutely have identified the cause early; others, are lucky.In general, be cautious in taking corrective action before simple tests of hypothesescan be made. Usually “Take corrective action” requires the expensive shutting downof the process and making alterations.

6.1.3More on Gathering and Interpreting Data

Here we consider the guidelines for designing experiments to test hypotheses, andlist the resources needed for gathering data from simple tests. We also explore theimplications of your personal style in planning and selecting tests.

6.1.3.1 Guidelines for Selecting and Designing Experiments to Test HypothesesThe tests should be simple, inexpensive and should provide positive and negativevalidations of a hypothesis. Usually we try to:

. avoid the temptation to “open the equipment and see”. Usually some verysimple tests can keep the process going and provide the insight needed toidentify the fault.

. check one hypothesis/variable at a time.

200


. isolate variables.

. test the “most likely” fault based on probabilities of past failures. Table 6-1,for example, lists some data related to instrument failure, relative to “equip-ment failure”.

Table 6-1 Failures per annum for instruments and equipment (from Woods, Process Design andEngineering Practice, originally published by Prentice Hall, 1995, � Donald R. Woods).

Instruments Failures Equipment

Analyzer: GLC 20.9 failures/annumAnalyzer: CO2 10.5Analyzer O2 7Analyzer: general 6.2Analyzer: pH 4.3Flow: Dp transducer 1.9Liquid level Dp transducer 1.8Liquid level float type transducer 1.6Liquid level general 1.6Flow: general 1.1Purge system 1Pressure: general 1Temperature: transducer 0.9

0.9 turbine0.6 centrifugal pump

Temperature sensor 0.30.15 distillation column, reactor0.1 exchanger

The tests should be based on a good understanding of the equipment and the sys-tem.

Example 6-1:For cavitation of a centrifugal pump, we could a) listen for the crackling noise typicalof cavitation, and b) reduce the flow through the pump by partially closing the lineon the discharge line. This should cause crackling noise to subside. Fundamentally,at lower flowrates, less NSPH is required and the friction loss on the suction side isreduced by the (velocity)2.

Activity 6-1: Selecting testsWe hypothesize that there is an obstruction in a 10-m length of pipe, that has nu-merous bends and fittings, through which liquid is flowing. Andre suggests the fol-lowing tests:

a. stop the operation, open the pipe at the various fittings and look.b. increase the flowrate and note the difference in pressure drop.c. decrease the flowrate and note the difference in pressure drop.

201


d. estimate the pressure drop and compare with measured values.e. stop the operation, open one end and send a plumber’s worm through the

line.f. and maybe there might be some others.

What would you select and why?

Activity 6-2: Selecting testsIn a refrigeration cycle we hypothesize that impurities got into the refrigerant. Com-ment on the appropriateness, limitations and assumptions of each of the followingtests:

1) check the pressure difference between the refrigerant side and the processside.

2) sample the refrigerant and analyze for impurities using GLC.3) read the temperatures and pressures and compare these with the data on

pressure–enthalpy charts for the refrigerant.4) read the pressure gauge on the compressor suction.

Activity 6-3: Case’6For Case’6, Saadia completed the following chart on the Trouble-Shooter’s Work-sheet for the case of the utility dryer (details are given in Section 4.4). Complete thechart below by matching the actions to the hypothesis. Comment on the appropri-ateness of each action.

Symptom a. exit air “wet”: 3 � higher than specificationsb. pressure drop double expected value

202

Activity 6-3: Hypothesis chart for worksheet.


a b c d e A B C D

1. Steam leak S N

2. Excessive moisture carryover from the separator S N

3. Valve S2 leaking S N

4. Adsorbent lacking adsorption capacity. S N

5. Absorber on-line too long: breakthrough S

6. Not enough regeneration time S

7. Condenser not cooling sufficient S

8. Instruments wrong, pressure N S

9. Absorbent broken down ? S

10. Temperature TRC, T3 reads low S S

6.1 Thinking Skills: How to Select Valid Diagnostic Actions 203

Diagnostic actions:

A. Test calibration of temperature T3B. Test calibration of pressure gauges P1 and P4C. Calculate a mass balance on moistureD. Calculate the expected removal of adsorbent from adsorption/kg adsor-

bent dataE. Sample adsorbent to determine if damagedF. Read pressure drops across different parts of the system and when differ-

ent parts are “on-line”G. Predict water removed via regeneration, temperatures and moisture in

exit gas profiles and compare with data taken over several cyclesH. Vary the regeneration time allowed

6.1.3.2 Resources for Gathering DataHere we consider the types of equipment to take, and the time required to do sometests.

a) Simple equipment and stuff to take with you on-site

Here are some of the things I used to take with me on a trouble-shooting mis-sion:

. notebook and calculator: keep accurate records.

. leather gloves: especially important to allow me to sense the temperature oneither side of a steam trap.

. stethoscope: very useful to magnify the sounds inside a vessel, steam trap orvalve. This helps to identify, for example, the typical flow through a thermo-dynamic steam trap; to listen for vibrations.

. string: in trying to estimate the diameter of a pipe or insulated pipe I found itmuch easier to measure the length of string around its circumference andthen calculate the diameter.

. tape measure.

. clamp-on ammeter.

. flashlight.

. stopwatch.

. knife.

. pens and marker, tape, labels, sample bags.

. tachometer.

. magnifying glass.

. brick: to help me estimate a mass or energy balance I often needed to knowthe flow of water in the drains. On some of the plants these were easily acces-sible. A brick could be placed in the rectangular drain to create a dam or“weir” and then from a measure of the height of liquid above the “weir” theflow could be estimated.


. pail and stop watch: measure other types of flows.

. long needle: allowed me to estimate the thickness of insulation.

. spray bottle of soap solution and tape: to identify leaks around valve stems,spray soap solution. For flanges, tape around the flange and then spray on asingle hole you make in the tape.

. a surface pyrometer or laser pyrometer may be a useful addition, especially ifsteam and steam traps are causing problems.

. camera.

In your journey as a trouble shooter, you too will collect your favorite collection ofsimple things to make it easy for you to uncover the secrets of the process.

b) Time it takes for gathering data

Here are some examples of the time it takes to gather certain types of data and tomake alterations to the process.

. Laboratory analysis: routine analysis: 2 h; special test for impurities 8 h.

. Instruments: install orifice plate 10 h; rotameter, 5 h, level gauge 8 h.

. Agitator: hook up, 40 mh; blower, hook up 25 mh; pump, hook up 50 mh.

6.1.3.3 Personal biases and style in collecting evidence and reaching conclusionsTrouble shooters create hypotheses and then select cues, evidence and tests to con-firm or disprove the hypotheses. They check that the evidence/ data/ cues actuallysupport their hypothesis. Sounds straightforward. However, most mistakes occurhere.

Each person has a personal style. Furthermore, mistakes and biases affect howdata are collected and conclusions reached. Consider each in turn.

a) Personal style in trading off data gathering versus taking action

Each of us has a preferred style of making decisions. Some prefer to be active, tomake choices even though they might be wrong. Others, want to gather data andreally understand the situation before action is taken. The P-J dimension of the Jun-gian typology, or Myers Briggs Type Indicator (MBTI) may provide you with insightas to your preferences. For example, a dominant P style is characterized by wantingto collect detailed information and data before making a decision; a person with adominant J style wants action and may use an approach of changing a suspectedcause before doing simple tests to check whether that really is the cause.

Example 6-2:For the Case’10, to dry and not to dry, and described in Section 5.4, the hypothesesselected by Jason and Heather are 1) wash water carryover from the centrifuge, 2)cycle from screen not coordinated with cycle from centrifuge, 3) feed crystals toowet from the screen, 4) condensate trap on the dryer is malfunctioning causing thecondensate to back up in the tubes and reduce the heat transfer area, 5) more finesinto the centrifuge causing the filter cycle to be too long; fines carryover to the cen-

204


trifuge causing blinding in centrifuge, and 6) crystal size change; crystals changeshape so that filtering is different.

Jason, after consulting with the vendor of the dryer, wants to replace the steamtrap on the dryer (since the vendor suggests that this is a likely cause and it’s rela-tively easy to change). “I’m convinced it is hypothesis’4. Let’s act.” Heather, on theother hand, wants to sample the crystals going into and out of the centrifuge every30 s for three cycles of operation. “The samples can be analyzed for water contentand for particle-size distribution. This is relatively easy to do and this will help ustest hypotheses’1, 3, 5 and 6. On the other hand, Jason, if you really feel stronglythat it is the steam trap, let’s go out and check out the trap first.”

Comment:Here we have apparent disagreement. Jason is showing predominant J behavior;Heather, predominant P. Provided Jason and Heather see their disagreement as“Hurrah! we balance out our different styles,” then a better result will occur. If Jasonand Heather had both been dominant J, then the steam trap would probably havebeen changed, only to discover that the steam trap was not the fault.

b) Common biases in collecting evidence

The four major types of error are pseudodiagnosticity, confirmation, availabilityand representative biases.

. pseudodiagnosticity or overinterpretation: actively seek worthless data andchange opinion based on irrelevant data; treat noncontributory cues as rele-vant. This happens about 30% of the time and is the most common bias.

. confirmation bias: actively support a favored hypothesis even though all theevidence points elsewhere. Seek confirming information and ignore discon-firming cues.

. availability bias: prefer to use data that are more readily available.

. representativeness bias: see similarity that doesn’t exist between two events.Even when given the correct underlying knowledge trouble shooters consis-tently disregard that knowledge in favor of stereotypes about how “represen-tative” these particular data/characteristics are.

c) Common biases in reaching conclusions

These biases tend to relate to personal preference and the amount of training andexperience.

Personal foibles include: premature closure and anchoring may occur despiteexperience and training.

. premature closure: the conclusion is not justified by existing data. Tend to be aJungian typology “dominant J” (discussed in part i).

. anchoring: adhere to a preconceived belief even though the evidence refutesit.

205


Three other biases that occur usually with inexperienced trouble shooters wholack the training and experience with the process equipment are inadequate synthe-sis, underinterpretation and misinterpretation.

. inadequate synthesis: unjustified conclusions are drawn. The trouble shooterfails to use knowledge effectively in interpreting and making inferences fromthe data.

. omission of clue or underinterpretation: an important clue is ignored. Thisoccurs about 2% of the time.

. wrong synthesis or misinterpretation: the available data contradict the conclu-sion. This occurs about 6% of the time.

Activity 6-4: Confirmation biasConsider the task of testing the hypothesis that “Every card that has a vowel willhave an even number on the other side”. The four cards shown below in Figure 6-1are available to test this hypothesis. (From Johnson-Laird and Watson, 1970). Whichcard or cards do you need to turn over in order to test the validity of the hypothesis?

U F 2 9

Figure 6-1 Four cards.

Activity 6-5: Feedback about your styleTo give yourself-some feedback about your style, consider the following case.

Assume that the information given in the scenario is factually correct. A code let-ter is given at the end of separate bits of evidence in the account of the murder, forexample “Tom Dayton had many enemies (a).” where the code letter is (a).

206

Worksheet 6-1: The Tom Dayton murder (adapted from Sherlock Holmes).

Tom Dayton had many enemies. (a) He was a scalawag and a prankster whonever passed up an opportunity to embarrass someone through a practical joke(b). It was Tom who invented the joy buzzer and the whoopee cushion, andsome even credit him with having originated fake vomit.

It is well known that Tom’s favorite target was his old headmaster, StanleyBosworth, (c) at Bromley School. Stanley Bosworth was the victim of some ofTom’s most elaborate pranks. Tom’s eventual marriage to Bosworth’s daughter,Melissa, was considered by many to be Tom’s ultimate joke (d) on the respectedheadmaster.

Among the more prominent victims of Tom Dayton’s past pranks were JudgeWalter Brighton (e), Lord and Lady Morton (f) of Westchester, banker MortimerFawcett (g), Doctor Fabian Peerpoint (h), and tobacco merchant Dawes Flescher(i).

6.1 Thinking Skills: How to Select Valid Diagnostic Actions 207

All of the above were present at the dinner party held on the Bosworth yachtin honor of Stanley Bosworth’s 60th birthday (j).

Following an uneventful dinner, most of the guests retired to their state-rooms to freshen up. The clock in the dining room struck 10 pm (k) when ashot rang out (l). Most later claimed they heard a second shot (m). All aboardthe yacht, including the yacht’s captain, Jonas Fenton, (n) and cook, the curvac-eous Mildred Weekson (o), arrived at Tom Dayton’s stateroom to find him dead– shot in the forehead (p).

A smoking revolver lay near the doorway (q); Tom’s body lay on the flooracross the room (r), just below an open porthole (s). Scrawled in the dust nearDayton’s body were the initials SB (t). In the corner of the room was a suitcasefilled with $500,000 (u). No bullets were found in the walls, ceiling or floor ofTom’s room (v). Who killed Dayton?

During the investigation the following factual evidence was produced:

w. Although just about everyone claim they heard two shots, Tom Daytonhad one bullet in his head.

x. Lady Morton was being blackmailed.y. The clock in the dining room was 15 minutes slow.z. The $500 000 in the suitcase was counterfeit and was accompanied by a

withdrawal slip for $1 000 000 from Mortimer Fawcett’s bank.aa. Bosworth angrily stated at dinner that he felt Tom was mistreating his

daughter.bb. The actual murder weapon was found under water below the porthole.cc. Dr Fabian Peerpoint advocates mercy killing.dd. The smoking revolver in the room belonged to Stanley Bosworth.ee. Tobacco merchant Dawes Flescher is a talented mountain climber.ff. Jonas Fenton, the yacht’s captain, went to school with Tom Dayton.gg. Dr Fabian Peerpoint revealed that Tom Dayton was terminally ill with

only a few months to live.hh. The bullet in Tom’s head did not come from the smoking revolver on the

floor in Tom’s room.ii. Melissa says that she was visiting her father in his stateroom from 9:45

pm until she heard the shot.jj. Mildred Weekson was having a secret affair with Tom Dayton for the

past two years; and with Lord Morton.kk. The shot that killed Tom Dayton was fired from outside the porthole.

Directly below the porthole is water.

Based on the evidence so far would you:

1. accuse ____________ of murder based on evidence (list the letters of theevidence supporting your conclusion) ____________________________.

2. accuse Dr. Peerpoint of mercy killing based on evidence (list the letters)_____________________________________________________________

6 Polishing Your Skills: Gathering Data and the Critical-Thinking Process208

3. conclude that Tom died from __________ based on evidence.(list the letters) ________________________________________________

or... .4. require that the following information is needed before any conclusion

can be drawn:ll) what pranks had Tom played on _________________________mm) check for poison in Tom’s bodynn) other _______________________________________________

Feedback about your style is given in Appendix G.

d. Cautions about interpreting data

All data have errors. We should have an idea of

a. What is an acceptable error in all the target or specification conditions.b. What is an acceptable error for all the measurements.c. When is there a significant enough deviation that we recognize that some-

thing is wrong.

The evidence may be relayed to us as fact, opinion or opinionated fact. Actually itis difficult for people to relate just the facts; they want to infer and give their owninterpretation to the information. As trouble shooters we need to separate carefullyfact from opinion. We need to concentrate on getting the facts and the focus on relat-ing the facts to the cause.

Maybe time variation occurs. Perhaps the data have to be collected over a timecycle.

Normal instrument error or a fault? All instruments have errors. Judgment isneeded to tell whether the following set of measurements are about what we expectfrom the instrument? Or that there is trouble on the process?

Example 6-3:Here are records from the operator’s log book of the digital printout for the tempera-ture at the top of the reactor. 932.56; 938.64, 930.28, 935.67, 932.19, 937.52.

If the expected temperature is 936 are the values within acceptable error or issomething wrong?

Evidence that trouble exists usually comes with error associated with it. We haveto distinguish whether the variation in the data represent “expected error” or a fault.

Example 6-4:The laboratory reports that the analysis of the recycle gas shows 3.4% methane. Itshould be 0.3%. All the other instruments on the process read normal.

Follow-up action: Another sample was taken; the laboratory reported 3.5%. Allother instruments still read normal.

6.2 Thinking Skill: Consistency: Definitions, Cause–Effect and Fundamentals

Next: No action was taken by the process operators. They continued to operate theplant as usual. Each day, the laboratory reported values of methane in the range 3.5to 3.65%.

Later: Two weeks later the head chemist returned from holidays and notes thatwhile she was away, all the analyses had been taken on the “calibration mode”.When the instrument is set on its correct settings, the concentration of methane inthe recycle line was 0.25%.

Too often we hope that the data are applicable. A colleague, in designing a petro-chemical plant was unable to locate the physical properties of the organics. Hedecided to assume they were the same as water and hope that they would work out.Just a short time spent in a critical assessment of this assumption would have savedsix months of wasted work.

Too often we accept data from the published literature; yet about 8% of data pub-lished are mistakes. “The temperature into the hydrodealkylation reactor is>1150 �C” states one reference. This should read >1150 �F. Another example is thata major handbook published an incorrect value of the heat of vaporization throughseveral editions. Check the data coming from computer programs and simulations.Check the physical property package estimates.

6.1.4Summary

Trouble shooters gather information to solve the problem. Information can be gath-ered for six different purposes: a) safety check plus provide background informationabout the situation, b) put the situation in the context of recent events, c) understandwhat should be happening, d) check what is happening now, e) gather data to testhypotheses and, perhaps, f) take immediate corrective action. Criteria are given forhow to select which type of information to gather. In general, start simply, isolatevariables, test the “most likely” faults first and have a purpose for gathering eachparticular piece of information. We need to realize that each of us has a preferredstyle (and perhaps bias) in how we gather and interpret data. Two inventories wereused to help identify preferences and biases. These ideas were illustrated by a rangeof examples and activities including revisiting Cases’6 (the utility dryer) and ’10(to dry and not to dry).

6.2Thinking Skill: Consistency: Definitions, Cause–Effect and Fundamentals

Scriven (1976) emphasizes that the main criterion used in critical thinking is consis-tency. Indeed, we trouble shoot in a world defined by consistent terms, fundamentalsand concepts. We work with facts. A clear distinction needs to be made between factsand opinions. The equipment with which we work has clearly defined symptomsthat are generated by each fault. The behavior of fluids and materials follows funda-mental principles. We communicate in English, with its defined rules; we work with

209


mathematics that follows rules. In this section we remind ourselves of those rulesand framework in which we check for consistency.

6.2.1Consistent Use of Definitions

Probably the most crucial concept is Facts. To effectively separate facts from opinionwe use definitions based on Obform coding developed by Johnson of the JournalismDepartment at the University of Wisconsin.

a. Facts.

Johnson defines three sources of facts as factual data, conclusions and backgroundinformation. Factual data may be accepted as fact if we can attribute an observer withbeing able to hear, feel, smell, taste or see the observations recorded or stated. Forexample, “the gas evolved from the anode is oxygen and that from the cathode is hydro-gen.” The facts we can observe are that colorless and odorless gases are evolved fromthe anode and the cathode. We can infer something about the nature of the gasesfrom further tests. However, the facts are not that “oxygen is evolved from theanode”, etc.

Conclusions from factual data may be taken as facts if the reasoning is correct:1) the validity of each step in the sequence of reasoning is proven, 2) there are suffi-cient steps and 3) the nine steps for valid reasoning (given in Section 6.5) are fol-lowed.

Background information is factual if references are given for the direct quotes andif statements made by people are direct quotations. For the latter, the message thatthe speaker said might be factually incorrect, but it is a fact that the person saidthose words.

b. Opinion.

This is information other than Facts.

c. Opinionated facts.

This is factual information that contains opinion. For example, The temperature is ashigh as 45 �C. The fact is the temperature is 45 �C. The opinion is that this tempera-ture is “high”. Put the two together and we have opinionated fact.

210


Example 6-5: Examples of Facts and Opinions

Example statement Factual statementby people

Factual data aboutoperation

John said “The gauge reads 50 kPa.” Yes YesJohn said “The pressure is 50 kPa.” Yes NoJohn said that the pressure gauge reads 50 kPa. No YesJohn said the pressure gauge reading is too high. No Opinion via words

“too”John said that the pressure is too high. No OpinionJohn said “The pressure gauge reads 50 kPa, the gaugewas calibrated yesterday, the sampling line is clear.If I increase the pressure slightly, the gauge readingincreases slightly. I infer that the pressure is 50 kPa.”

Yes Yes

John said “The pressure gauge reads 50 kPa, the gaugewas calibrated yesterday, the sampling line is clear. If Iincrease the pressure slightly, the gauge reading increasesslightly. I infer that the pressure is 50 kPa. I conclude thatpressure is too high.”

Yes Yes + Opinion

Activity 6-6: Facts and opinionsAnalyze the following passage and classify it according to facts, opinion and opinio-nated facts.

Here is an accurate account of what happened.The telephone rang! “Trouble out on the ethylbenzene unit,” said Bill. Harry said

that he would be right out as he slammed down the phone. As Harry approachedthe unit Bill came out to meet him and said, “I’m sure that the heat transfer is insuf-ficient in the reboiler to the product column; I’ll show you what I mean.”

Harry glanced at the rotameter and saw that the flow to the column was the usualamount of 3000 gpm; the pressure gauge read 150 psig and the bottoms tempera-ture was 140 �C. Rounding the column, he saw that the liquid level in the bottomslevel gauge was rising at a rate of about 3 cm/min. The liquid level disappeared outof the top of the level gauge. After about two minutes the level reappeared in thesight glass and disappeared out of the bottom of the sight glass within several min-utes. “See,” said the operator, “we have lost all the bottoms out of the column justlike that!” “It has gone off to the storage tank,” offered John. “No, it has gonethrough the reboiler and straight up the column. You can see by the instabilities inthe pressure gauges that occur just after the level disappears out the bottom of thesight glass,” said Bill.

Activity 6-7: Based on the account given in Activity 6-6, which of the following state-ments are True (T), false (F) or can’t tell (?).

1. Bill said that there was trouble out on the ethylbenzene unit. T F ?2. Harry said “I’ll be out immediately.” T F ?3. The heat transfer is insufficient. T F ?

211


4. The trouble is in the reboiler. T F ?5. The flow to the column is 3000 gpm. T F ?6. The pressure was 150 psig. T F ?7. The bottoms temperature was 140 �C. T F ?8. The level in the bottom of the column

was building up and then suddenly dropping. T F ?9. John said, “ The bottoms have gone off to the storage”. T F ?10. When the level in the bottoms of the column

drops the pressure gauges show instabilities. T F ?

Example 6-6: Case’3:Consider the account of Michelle working on Case’3. The case of the cycling col-umn (presented in Chapter 4, Section 4.1). At one stage, Michelle is trying to decideif the control system is at fault. Her diagnostic action was to “put the control systemon manual.”

When the set point was increased manually, the valve stem on the steam moves up, theliquid level appears in the sight glass and continues to drop but shortly thereafter the levelappears in the glass and is rising. I conclude that the control system is not at fault.

For Michelle’s thinking reproduced above, has Michelle focused on facts? opin-ion? opinionated facts?

a. When the set point was increased manually, the valve stem on the steam moves up.

These are both observable, and therefore called “facts”.

b. When the set point was increased manually, the liquid level appears in the sightglass and continues to drop but shortly thereafter the level appears in the glass andis rising. These three are observable and therefore are “facts”.

Comment:Michelle did a good job here. She might have incorrectly said, “When the set pointwas increased manually, the steam flow increased.” Since Michelle could not have seenthe steam flow, saying the steam flow increased would have been an opinion.

In summary, both facts and opinions are used to trouble shoot. However, we needto know which are facts and which are opinions. A definition of facts is given. Exam-ples illustrate how to use that definition consistently. We revisited Case’3, thecycling column.

6.2.2Consistent with How Equipment Works: Cause fi Effects: Root Cause-Symptoms

Equipment is fabricated and works according the fundamental principles of scienceand engineering. Therefore, if a fault occurs in the equipment, certain symptomswill appear. We define a symptom as something that can be observed, heard or feltrelated to the performance of equipment or a system of equipment. The degree towhich a symptom can be explicitly observed by the trouble shooter depends on the

212


sensors and configuration for the equipment layout. For example, for the usual con-figuration of a pump the symptoms we might be able to observe include the flowrate(if there is a flowmeter), the exit pressure/head (if there is a gauge), the power onthe drive (if the amperage is measured), the “crackling noise” (if we can get closeenough to “hear”) and the temperature of the motor/drive and of the suction line (ifwe can “feel” or use a surface pyrometer to measure the surface temperature). It isimportant for us to document cause–effects or cause–symptoms for different typesof equipment in terms of the usual types of instrumentation that is available.

a. Cause fi symptom consistency

An important task is to ensure that we are familiar with cause fi symptom data fordifferent pieces of equipment.

Example 6-7: Causes (and symptoms)Three of the common faults with centrifugal pumps (and the symptoms) are: oper-ates at very low capacity (vibration and noise, and pump overheats and/or seizes),rotor not balanced (short bearing life, vibration and noise, short mechanical seal life,pump overheats and/or seizes and stuffing box leaks excessively) and impeller par-tially clogged with solids (either no liquid delivered or flow lower than expected,power demands higher than expected).

Activity 6-8: Listing symptoms for causes or faults.For the depropanizer shown in Case’8, for the following faults/causes,

a. list the symptoms that would be observed, heard or smelt.b. estimate the order of magnitudes of the deviations (or the extent of the symp-

toms).

1. The vortex breaker in the overhead drum V-30 is welded such that the cross-sectional area for flow has been reduced to 15% of the pipe internal cross-sectional area.

2. The vortex breaker in the overhead drum V-30 has corroded away. The exitpipe is fully open.

3. In the overhead condenser, E-25, the inerts have been inadequately ventedfrom the shell side (from the tube side) before startup.

4. In the reboiler, E-27, the inerts have been inadequately vented from the shellside (from the tube side) before startup.

5. For the depropanizer, the trays are bent so that the downcomer clearance is1�2 what it was supposed to be.

6. For the depropanizer, the trays are bent so that the downcomer clearance isdouble what it was supposed to be.

7. For the depropanizer, corrosion as increased the diameter of the holes in thesieve trays by 10%.

8. For the depropanizer, tray 5 has collapsed because of inadequate support.9. For the depropanizer, trays 5, 13, 20 and 25 are not level. They are 30� to the

horizontal.

213


More examples of typical faults/causes are given in Appendix H. Appendix I liststhe symptoms for some of these.

We can refer to the cause fi symptom statements as If ... then statements andinvoke the rules of logic. The characteristics of If ... then statements are (we includethe terms antecedent and consequent from the reasoning literature):

. If the “cause” (antecedent) is positive, then the “symptom” (consequent) ispositive.

. If the “symptom” (consequent) is negative, the “cause” (antecedent) is negative.

. If the “cause” (antecedent) is negative, we cannot conclude anything aboutthe “symptoms” (consequent). For example, If the pump impeller is not turn-ing backwards, then we cannot conclude anything about the flowrate andhead.

. If the “symptom” (consequent) is positive, then we cannot conclude anythingabout the “cause” (antecedent). For example, If the flowrate and head areabnormally low, then we cannot say the impeller is turning backwards. Otherreasons can cause this.

This last characteristic causes frustration for the trouble shooter because typicallywe are not given the cause, rather we are given the symptom and expected to deducea cause. In other words, symptom ‹ cause. Furthermore, trouble shooting is alsoconfounded because there may be more than one cause that could be contributingthe symptoms. “Symptoms” are not necessarily caused by one and only “root cause”.Nevertheless, documenting cause fi symptom information is a good starting pointfor trouble shooting from which we can develop information about symptom ‹cause.

b. Symptom ‹ cause consistency

The starting point for trouble shooting is a list of symptoms. The challenge is tocreate hypotheses as to the probable cause that are consistent with the symptoms.This is the reverse of the data discussed in section a. Some example symptom ‹cause information is given in Chapter 3 for different equipment. The challenge is,that unlike cause-effect data that are “true”, the symptom–cause data are “not neces-sarily true”.

Activity 6-9: Symptoms and causesConsider Case’11, the symptoms and the hypotheses/causes. Are they consistent?

Case’11: The Lazy Twin (courtesy of W. K. Taylor, B. Eng. McMaster, 1966)The situation: Pump A is usually running. Pump B, identical to pump A, is thespare pump. The operators of the process notice that the flow meter, FRC-100,shows that not enough flow is going to the process. So they switch over from pumpA to pump B by shutting off the power to the drive motor for pump A and turningon the power to the drive motor for pump B. Now the flow to the process comesback to where it should be. What is wrong? The pump configuration is shown inFigure 6-2. Data about the flow and setting of the valves are given in Figure 6-3.

214


T200

V200

A

B

V206

V202

V201

V205

V100

PI

220

PI

210

FRC

100

Figure 6-2 Operator’s sketch of the equipment for Case’11.

F100

CO

NT

RO

LL

ER

OU

TP

UT

F100

100%

0%

ONLY

PUMP A

ON

S. P.

TIME

ONLY

PUMP B

ON

VALUE = 100%

Figure 6-3 Flow and valve-setting data for Case’11.

215


For this Case’11 we might summarize the symptoms as:

a. when pump A is running, the flow meter shows that not enough flow is goingto the process

b. when pump B is running, the flow meter shows the correct flow is going to theprocess

In the following table, column “a” represents symptom “a”, “b”, symptom “b”.Consider each hypothesis and in columns “a” and “b” place a code S supports; D dis-proves and N neutral or can’t tell this hypothesis or fault.

Table 6-2 Worksheet for hypotheses.

Working Hypotheses or cause Initial Evidence:symptoms

Diagnostic Actions

a b c d e A B C D

1. Flowmeter is reading wrong.

2. Pump A has a sandwich stuck in the impeller.

3. Pump B is much bigger capacity than we need

4. Full electrical current does not go to pump A.

5. A fuse is blown in the circuit to pump A.

6. Liquid is vaporizing in the line to pump A and sopump A cannot produce the expected flow.

7. The operator cannot read the flowmeter correctly.

8. Some corrosion products are clogging the flow-meter.

9. There is not enough Net Positive Suction Head forthe Pump A.

10. There is not enough liquid in the tank to the pumps.

11. The signal is not getting from the controller to thevalve.

12. The flow-control valve is stuck partway shut.

c. Symptom or “root cause” dilemma

In brainstorming possible causes (that are consistent with the symptoms) some-times the idea is another symptom, instead of a “root cause”.

Example 6-8:Chapter 3 suggests that for distillation if all temperatures are falling simultaneouslythen the cause might be low boilup. Is “low boilup” a root cause? No. The root causeis something that causes low boilup, which (for a thermosyphon reboiler) could besuch causes as condensate flooding/ inadequate steam supply/ steam valve closed/superheated steam/ boiling-point elevation of the bottoms/ inert blanketing/ filminstead of nucleate boiling/ increase in pressure on the process side/ undersizedreboiler/ control system fault/ fouling on the process side/ low liquid level/ high

216


liquid level/ heavies in the feed/ pipe lengths< design/ pipe diameter > design/ pro-cess fluid level < 30–40% of tube length.

To test if we have a symptom or a root cause, test the postulated “cause” by asking“What would cause this?” If there is no new answer, then we probably have the rootcause.

Activity 6-10: Root causeFor Case’11, The Lazy Twin, and the possible causes listed, which of these causesare root causes and which are symptoms?

In summary, this is perhaps the key section in this chapter. Consistency betweencause and symptom; symptom and cause; hypotheses and symptom are key to trou-ble shooting. It is simplest to start with typical causes/faults and listing symptomsthat are produced. Then we move to the trickiest part, namely the reverse: identify-ing possible causes from a given set of symptoms. A new Case’11, the lazy twin,was introduced. The challenge of working with “root causes” was emphasized.

6.2.3Consistent with Fundamental Rules of Mathematics and English

Our reasoning and actions need to be consistent with the rules of mathematics andEnglish.

For Mathematics consider the following:

Example 6-9:Please check the reasoning in the following:ffiffiffiffiffiffiffi

�11

r=

ffiffiffiffiffiffiffi�11

r

ffiffiffiffiffiffiffi1�1

r=


r

ffiffiffi1

pffiffiffiffiffiffiffi�1

p =


pffiffiffi1

p

1 =ffiffiffiffiffiffiffi�1

p ffiffiffiffiffiffiffi�1

p

1 = –1

What went wrong? Here we encounter an incorrect answer so we know that wewere inconsistent with the rules of mathematics.

But what happens when we don’t have a surprise? We need to condition ourselvesto check for mathematical consistency in all the calculations we do as we troubleshoot. We do have guidance in Sections 6.2.4 and 6.2.5 because our answers shouldbe mathematical answers that are consistent with fundamental laws and withexperience. To some extent Saadia followed these principles in Case’6. She calcu-

217


lated the expected results ahead of time. Then when she saw the results she wouldknow if there were surprises.

For English, we have the rules of grammar and the meanings of words. Wordsonly have meaning in people. Therefore, we need to use words that are “understood”by all. Consider the following examples:

Example 6-10:An open window, a gust of air, glass on the floor, water on the floor, Mary’s dead.Why?

Example 6-11:A man is walking along. He tears his sleeve on a sharp corner. Within minutes he isdead. Why?

In these two examples, if we realize that Mary is a goldfish and that “walkingalong” means walking on the bottom of the ocean, then these are not puzzles andwe all understand.

When we are trouble shooting, some of the words that are ambiguous and thatinterfere with communication include:

“The instrument looks OK!”“I followed the usual practice.”

Activity 6-11: Ambiguous wordsMake a list of ambiguous words that you encounter when you trouble shoot. Thenthink of ways you can overcome this ambiguity.

In summary, critical thinking requires that we are consistent in our use of Mathe-matics and English.

6.2.4Consistent with Fundamental Principles Of Science: Conservation of Mass, Energy,High to Low Pressure, Properties of Materials

Mass and energy balances, pressure profiles and an understanding of the uniqueproperties of materials are probably the most useful source of information in troubleshooting. Yet, trouble shooters often neglect these.

6.2.5Consistent with Experience

Having a wide range of “rules of thumb” and memorized experience factors help usto validate and check answers and ideas for consistency. Sources include Brannan(1998), Walas (1988), Woods (1995) and Woods (2001).

218

6.3 Thinking Skills: Classification

6.2.6Summary

Consistency is the key to critical thinking. Here our focus was on consistency with

. definitions: to sort facts from opinions;

. cause fi effect information for equipment: to guide in the creation and test-ing of hypotheses;

. the rules of mathematics and English: to help us focus on accuracy;

. the fundamentals: to remind us of the basics to check;

. experience: to validate our thinking, calculations and assumptions.

Most of the effort was spent on the first two topics: facts versus opinions and sort-ing out cause fi effect relationships.

6.3Thinking Skills: Classification

Classification is dividing the whole into parts such that there is a meaningful rela-tionship among the parts. The classification is done for a purpose, and for each levelof classification there must be one and only one criterion. Each level of classificationshould be complete. There should be no single subclass. The classification shouldbe consistent in the amount of detail given and the classification should haveneither faulty coordination nor faulty subordination. These are the general charac-teristics of a good classification.

Skill in classification is needed to:

. classify the starting information during the stage of Define the stated prob-lem, discussed in Chapter 2.

. classify the possible causes generated during the brainstorming session aspart of the Explore stage, discussed in Chapter 2 with the brainstorming ses-sion being illustrated in Chapter 5.

6.3.1Classify the Starting Information

For the purpose of understanding the problem, the starting information should beclassified using the criterion “what are the key parts to the trouble-shooting problemstatement”. Usually the key parts are:

. the situation or system,

. the symptoms that suggest a fault,

. the triggering event,

. the criteria for success (either stated or inferred),

. the constraints.

219


The focus here is to identify the “symptoms”. The symptoms are defined as “evi-dence stated by the operators, those displayed by sirens or alarms, reports from thelaboratory, or the analyzer showing non-specification behavior, or complaints fromcustomers describing unexpected or unusual behavior.” Sometimes the symptom isaccompanied by a triggering event. The triggering event is often expressed as“When ...”. Sometimes there is no apparent triggering event. Sometimes the trigger-ing event is not related to the symptom; it is coincidental. However, when we starttrouble shooting the problem we may not realize the coincidence. Some examples oftriggering events include:

When we increased the flowrate to new higher capacities ...When we started up for the first time ...When we started up for the first time after maintenance ...When the operator increased the pressure in the column above the usual value ...

The mental TS process is to scan all the given, starting information and apply thedefinition of “symptom” to identify this portion of information.

Example 6-12: Identifying symptoms.For Case’9, the symptom is stated ambiguously as “nothing happened.” Before wecan proceed, we should identify why this is unexpected behavior. From the context,we can rephrase this as, the vacuum inside the deodorizer should have been suffi-cient to suck the Fuller’s earth into the deodorizer when the connecting valve wasopened. However, the Fuller’s earth was not sucked into the deodorizer because wecould not see powder dumping in when we looked through the view port.

Example 6-13: Identifying pertinent triggering events:Several When’s occur in this situation in Case’9: “When the valve was opened;”“when the vacuum had stabilized”; “when we started up this plant for the first time”.Are any of these triggering events? The most likely triggering event is when westarted up this plant for the first time.

6.3.2Classifying Ideas from Brainstorming

As illustrated in Chapter 5, usually we obtain 50 to 100 ideas in a brainstormingsession. Of these often 80% are non-sensible. That’s OK because it is worth a fewminutes to generate such ideas since hidden among the crazy ideas are some realwinners. After the brainstorming session, however, we need to classify for the pur-pose of listing at least six feasible hypotheses. The first level of classification wouldbe “technically feasible” versus “non-sensible” or it could include an additional classof “interesting”. The next level of classification would be to sort the technically feasi-ble ideas according to underlying type of cause.

220

6.4 Thinking Skills: Recognizing Patterns

6.4Thinking Skills: Recognizing Patterns

Patterns in the information could include increasing or decreasing trends, cyclingdata, series or unexpected systematic external changes that affect performance.

Here are three examples.

Example 6-14:One process operator might be experiencing great personal stress at home. When-ever this operator is on shift the process does not run smoothly. If we were unawarethat the operator was under stress it would be hard to identify the cause.

Example 6-15:Whenever there is a severe thunderstorm some of the instruments on the plant mal-function.

Example 6-16:Everything seemed to be working OK although gradually the performance was mov-ing off specification. The cause was a minor leak in a heat exchanger.

In these three examples, the trick was to be able to see a pattern between stressand poor performance; thunderstorms and malfunctioning instruments; and gradu-al changes in performance and cross contaminant because of a leak.

The general guidelines and criteria for identifying patterns are not very helpful.Keep an open mind. Plot (and be sensitive to) time variation in data. Note the possi-ble interaction between different systems and especially the interaction betweencycling processes.

Consider nowmore about the patterns in the symptoms and in howwe collect data.

6.4.1Patterns in the Symptoms

The unexpected and outside factors that can affect performance include the weather:

. showers and rain; extreme cold; hot sunny weather affects cooling towersand air-cooled condensers.

. electrical storms affect sensitive electronic/electrical instrumentation.

. atmospheric pressure affects pumps that pump liquid from tanks that areopen to the atmosphere.

. extreme cold can freeze vents shut; freeze bucket steam traps.

. hot weather plus steam tracing can cause vaporization and vapor binding.

Cycling processes, that might interfere with other cycling processes include:

. batch distillation and batch reactors,

. adsorption and ion exchange,

. centrifuges and filters.

221


When cyclical processes are used, the data should be reported for different partsof the cycle and not reported just as average values. Averages mask the trends andpatterns.

When cyclical processes occur, the frequency of the cycle should be determined.For example, cycling steam flow or column pressure for a distillation (30 s to severalminutes) coincides with a specific set of probable causes. On the other hand, ther-modynamic steam traps should have about six cycles/minute.

Example 6-17: The cycling level in the bottom of the column.Engineer Ted Tyler used the cycle to help him predict that the collapsed tray, causingthe cycling, was tray 13 to 20 from the bottom.

Example 6-18: Another cycling level in the bottom of the column.In another cycling column, the coordination between the cycle of the level and thecycle of the steam trap, as determined by listening with a stethoscope to the trap,helped pinpoint the steam trap as the fault.

Example 6-19: The bombastic bagging machine.PVC powder was pneumatically conveyed to a hopper located above the baggingmachine. The dust was filtered from the conveying air. The problem was that peri-odically there would be loud thumping in the hopper that sounded like an avalancheof powder suddenly arrived in the hopper. Yet all the evidence suggested that thepowder was being fed to the hopper smoothly and continually. This periodic appear-ance of slugs of powder caused problems with the continual bagging operation.What was discovered was that the bags in the dust filter were being periodicallycleaned. The design fault was that the powder-laden air flowed up the inside of thecylindrical, vertical bags so that the blow-ring cleaners could travel on the outside ofthe bag. As the blow ring cleaned a bag it blew the powder into the center of thecylindrical vertical bag and the powder was held in the up-flowing conveying air.The powder became fluidized in some of the bags. When the mass of fluidized pow-der became excessive in the central core of one of the bags, the whole mass from thefluidized bed would avalanche down into the hopper. Which bags became fluidizedbeds and when the bed became unstable seemed to be completely random. The cor-rective action was to replace the configuration so that the feed air flowed down thecentral core of the bag. In this example the cyclic operation was the cleaning of thefilters.

The gradual buildup of solids, of corrosion products or of trace contaminants cancause trends in performance. The cause might be that the solids are too wet, theamount of purge is insufficient or a leak. Gradual changes that cause trouble areoften the most difficult to detect. Good records are needed to note the trends.

Unexpected pockets of water or condensate in low-lying sections of pipe can causepatterns in performance.

222

6.5 Thinking Skill: Reasoning

Activity 6-12: Case’7: The reluctant vacuum crystallizerThe data given in this case are cyclical! Analyze the approach Frank took, asdescribed in Section 4.5 of Chapter 4, and look for patterns in the data that Frankshould have spotted.

Activity 6-13: Case’10: To dry and not to dryThis case is described in Section 5.4. The operation consists of a continuous crystal-lizer, a batch screen, a batch centrifuge and a continuous rotary dryer. The symp-toms are “The crystal product from the rotary dryer has a moisture content of 4.5%whereas the design value is 1.5%. Cake seems to be building up in the dryer feedchute, in the feed screw and on the steam tubes at the feed end of the rotary dryer.”Critique these symptoms.

6.4.2Patterns in the Evidence

When we suspect a pattern, then we should collect data that will make the patterneasy to spot.

Example 6-20: In Case’6: The utility dryerSaadia sampled every 15 minutes for the 120-minute regeneration cycle. She alsoastutely sampled such that she could have data for bed A adsorbing and then beingregenerated (and for B adsorbing and being regenerated).

Activity 6-14: Case’10: To dry and not to dryBefore the change in washing, the centrifuge and screen cycle were 10 minutes on-line and 3 minutes water wash. What sampling would you recommend to try toidentify a pattern?

6.5Thinking Skill: Reasoning

In evaluating evidence and making decisions about the hypotheses, the reasoningthat we use should be sound. In this section we outline an organized nine-stepapproach to evaluate the reasoning.

In general, an overall nine-step approach for critical thinking is:

1. Classify all the given information into the key parts.2. Write the conclusion.3. Identify the context. What are the stated and the inferred contexts? Are there

other pertinent contexts?4. Check the definitions; identify and clarify ambiguous terminology.4a. Change the argument to show the relationship between the conclusion and

the evidence.

223


5. Consider the evidence. What is the quality of the evidence? Is it sufficient?Relevant? acceptable? What are the facts? What are the opinions? Is the cor-rect type of data gathered for the purpose? Usually the purpose is to test ahypothesis. Check the evidence. The evidence is the basis of an argument.

6. Formulate the assumptions.7. Assess the quality of the reasoning.8. Assess the strengths of the counterarguments.9. Evaluate the consequences and implications.

To illustrate this process consider some of Michelle’s reasoning as presented inChapter 4, Section 4.1 as she tries to resolve Case’3, the case of the cycling column.At one stage, Michelle is trying to decide if the control system is at fault. Her diag-nostic action was to “put the control system on manual.”

When the set point was increased manually, the valve stem on the steam moves up, theliquid level appears in the sight glass and continues to drop but shortly thereafter the levelappears in the glass and is rising. I conclude that the control system is not at fault.

6.5.1Step 1: Classify the Information

The given information includes the context, the conclusions, the evidence related tothe conclusion, the stated counterarguments and the stated assumptions. Later wewill add inferred assumptions and counterarguments.

The context: the situation. Supply WHO, WHAT, WHERE, WHEN IS and ISNOTdetails plus the equipment and system.

The conclusion: that is usually expressed as either what is the cause or what is notthe cause.

The evidence: collected as described in Section 6.1.The stated counterarguments: evidence that disproves a statement or hypothesis

that you are trying to prove.The assumptions: a statement for which no proof or evidence is offered.The qualifier: a constraint or restriction or limiting condition on the conclusion.

Example 6-21: Classify the information given in the short segment of Case’3, thecase of the cycling column:Context: a level in the sight gauge at the bottom of a distillation column is cycling.The frequency is about once per 2 minutes. More details will be given in Step 3. Thecontrol system was “put on manual.”

Conclusion: I conclude that the control system is not at fault.Evidence: When the set point was increased manually, the valve stem on the steam

moves up, the liquid level appears in the sight glass and continues to drop butshortly thereafter the level appears in the glass and is rising.

Stated counterarguments: none in this passage.Stated assumptions: none in this passage.

224


Inferred assumption: the control system actually was on manual; the level in thesight glass is related to the level in the bottom of the column.

Stated qualifier: none.

6.5.2Step 2: Write the Conclusion

It is important that the focus of the reasoning is stated clearly. To locate the statedconclusion look for words like “therefore,” “because”, “I conclude”, “so” and “thus”.Usually the conclusion is near the end of dialogue. There may be several conclu-sions each building on another. Identify the main conclusion. Write it down so thatwe can focus on the conclusion. The conclusion can be in many different forms.

. We could be identifying the context (as in Section 6.5.3) and our conclusionis “Therefore, we have correctly identified the context.” But have we?

. We could be monitoring our trouble-shooting process (as was recommendedin Chapter 2) and we conclude “Thus, we have completed the hypothesis gen-eration stage, now let’s move to the data collection stage.” But is this correct?

. We may conclude that “In summary, we have selected a reasonable set ofhypotheses from the brainstorming activity in Chapter 5.” But have we? Arethe causes root causes or are some of them symptoms?

. We might have listed the symptoms and the hypotheses and concluded that“Therefore, hypothesis’1 is consistent with the evidence.” But is this a validconclusion?

. We might have gathered some evidence and concluded that a hypothesis hasbeen confirmed, denied and can’t tell.

Example 6-22: Michelle’s main conclusion is “I conclude that the control system is notat fault.” The question is, Is her conclusion valid?Label each conclusion.

6.5.3Step 3: Identify the Context

What are the stated and the inferred contexts? Are there other pertinent contexts?The context depends on the conclusion. I find it helpful if I sketch a cross section ofthe equipment to remind me of the internal possibilities. I draw a line around it todefine the system showing the ins and outs. Label specific contexts.

Example 6-23: in Case’3 the context is that the level in the sight glass cycles. Thedistillation column is as shown in the Case and, for this portion of the evidence, thecontrol system was placed on manual. Whether this is included as a qualifier or con-text depends on how detailed you want to present the argument. A sketch of the con-text is illustrated in Figure 6-4.

225


Figure 6-4 Sketch of the system.

This sketch includes more detail around the steam control valve; it includes thebottoms draw-off and uses the dotted line to indicate the system.

6.5.4Step 4: Clarify the Meaning of the Terminology

Identify and clarify ambiguous terminology. Write out your interpretation of anyambiguous words and any unstated but intended implications. Check that youunderstand all the words and terminology. Often, on process plants the personnelcreate jargon words for operations and pieces of equipment. For example, on oneplant the condensate receiver was called “the pig”. If you are unfamiliar with theterminology you need to ask for clarification. This takes courage. Use the principlesof consistency outlined in Section 6.2, especially Sections 6.2.1 and 6.2.3.

Example 6-24: in Case’3 the words that Michelle should know include “the setpoint”, “control system” and “on manual”. On manual = signal from the sensor nolonger alters the steam-valve setting; manually change the set point and the valveshould respond.

226


6.5.5Step 5: Consider the Evidence

In trying to decide on the validity of Michelle’s conclusion the factors to considerinclude: What is the quality of the evidence? Is it sufficient? relevant? acceptable?What are the facts? What are the opinions? Are the correct types of data gathered forthe purpose? Usually the purpose is to test a hypothesis. Check the evidence. A dia-gram usually helps.

6.5.5aIdentify the Evidence

Label each assertion. Sometimes several assertions appear in the same sentence.Separate each one. Don’t give a separate number to the same assertion even thoughthe same assertion might be stated several times. Give numerals to conclusions,counterarguments and assumptions.

Example 6-25: For Case’3

[With the control system on manual, 2].[When the set point was increased manually, 3][the valve stem on the steam moves up, 4],[the liquid level appears in the sight glass and continues to drop but shortly thereafter

the level appears in the glass and is rising. 5][I conclude that the control system is not at fault, 1]

6.5.5bCheck for Consistency

Our basis is consistency. Systematically scan over the different elements for consis-tency in Section 6.2. 1) Label facts, opinions and remove the opinion from opinio-nated facts. 2) Check for causefi effect consistency. 3) Check that definitions didnot change in different parts of the argument; that symbols are not defined in oneway and used in another; that a variable is not treated as a constant; that a principleis not applied beyond its range of applicability, that an assumption is not made andthen violated, that coupled variables are not treated as independent.

Activity 6-15: AdiabaticityAn adiabatic change is one in which no heat enters or leaves the system. A primitiveJoule experiment in which a lead shot is shaken in a cardboard tube is approxi-mately adiabatic. The lead shot gets hotter. Does it then have more heat in it? If so,how does the heat get there? Is the experiment adiabatic after all? Check for consis-tency.

227


Example 6-26: for Case’3 and Michelle’s reasoning:

[With the control system on manual, 2].[When the set point was increased manually, 3][the valve stem on the steam moves up, 4],is consistent with behavior of a control system.Inference: [if the set point is kept constant, 6], [the valve-stem position is constant, 7],

[the steam flow is constant, 8].

6.5.5cWhich Evidence is Pertinent?

The overall context is to find the cause of the cycling level in the bottoms of thecolumn. Therefore any evidence gathered should relate to what does and does notaffect the level of liquid in the bottoms of the column. Here we focus on the hypoth-eses-symptom-action chart from the Trouble-Shooter’s Worksheet. Alternatively weconsider the symptom‹ cause data given in Chapter 3, subject to the concernsexpressed in Section 6.2.2b. Some might prefer to use a diagram to illustrate thesymptom with all the related possible causes/hypotheses.

Example 6-27: for Case’3, Figure 6-5 is an example of symptom‹ cause diagram.

Figure 6-5 Symptom‹ cause chart for Case’3.

228


6.5.5dDiagram the Argument

Scriven and Halpern (1996) take different approaches in diagramming an argu-ment. Scriven places the evidence at the top of the page and moves down the pageto a conclusion. Halpern places the conclusion at the top and uses different evidenceto “support” the conclusion. Here are some suggestions:

. Use the numbers for the evidence and conclusions that were used in Section6.5.5a.

. Connect evidence to conclusions with an arrow with separate arrows for eachdifferent form of evidence.

. Rate the quality of the support that the evidence lends to the conclusion:weak, moderate, strong and write this rating on the arrow connecting theevidence to the conclusion.

. Include assumptions. I use a dotted line if the assumption is inferred.

. Include the counter-arguments with a wriggly line.

Example 6-28: Case’3 Figure 6-6 shows a diagram of the arguments related toMichelle’s thinking during the example scenario.

Figure 6-6 Diagram of the arguments.

229


6.5.6Step 6: Formulate the Assumptions

An assumption is a statement for which no proof or evidence is offered. Considereach bit of evidence and the resulting conclusion and identify the inferred assump-tions. This should include the assumptions made by a) you, b) the designer, c) theprocess operator, d) the laboratory analysis team, e) the instrument technicians, f)the maintenance personnel and g) vendor. Show the assumptions on the diagram.

Example 6-29: Case’3:Michelle has assumed:

. valve-stem movement= valve opening; the seat is connected to the valve stem[9]

. the steam pressure from the utilities plant (and at the Battery Limits) is con-stant [10]

These are shown in Figure 6-6 as dotted lines

6.5.7Step 7: Assess the Quality of the Reasoning

Consider the reasoning related to: i) the symptoms; ii) the symptom ‹ cause rela-tionships; iii) the data gathered to test the hypothesis or probable causes; iv) theassumptions and v) the conclusion.

Focus on the assumptions and on consistency.Use a chart of the arguments to guide the analysis. Consider each of the compo-

nents in turn.

i) The symptoms. Were the symptoms correctly identified in Step 1? In theCase’3, we have no reason now to alter the symptom. There is an assump-tion related to the symptom, and shown on Figure 6-5 as a dotted line: cyclingin the level in the sight glass= cycling in the level in the bottoms =? cycling inthe level in the reboiler. This was noted but not addressed.

ii) The symptom‹ cause relationships. Revisit Chapter 3 and see if any perti-nent causes have been omitted. Also check that the diagram is consistent,based on the principles of Section 6.2.2. Recall that symptom‹ cause evi-dence is where many mistakes are made because, although the causefisymptom connection is valid, the reverse (symptom‹ cause) is not alwaystrue since multiple causes can display the same symptoms.

Example 6-30: Case’3: reconsider Section 3.3, reboilers; Section 3.4 distillation andSection 3.5 steam traps.Section 3.3 reboilers: general:

“Cycling (30 s–several minutes duration) steam flow, cycling pressure on the processside and, for columns, cycling Dp and cycling level in bottoms”: instrument fault/ con-densate in instrument sensing lines/ surging/ [ foaming]* in kettle and thermosy-

230


phon/ liquid maldistribution/ steam-trap problems, see Section 3.5, with orifice Dpacross trap < design/ temperature sensor at the feed zone in a distillation column/collapsed tray in a distillation column.

[ foaming]*: surfactants present/ surface tension positive system/ operating tooclose to the critical temperature and pressure of the species/ dirt and corrosion sol-ids.

Section 3.3 reboiler: thermosyphon:“Cycling (30 s–several minutes duration) steam flow, cycling pressure on the process

side and, for columns, cycling Dp and cycling level in bottoms”: in addition to general, allnatural circulation systems are prone to surging/ feed contains high w/w% of highboilers/ vaporization-induced fouling/ constriction in the vapor line to the distilla-tion column. For horizontal thermosyphon: maldistribution of fluid temperatureand liquid.

Section 3.4 distillation: Only lists cycling temperatures:“Cycling of column temperatures: “ controller fault/ collapsed tray/[damaged trays]*/

[ jet flooding]*/ [ foaming]*/ [downcomer flooding]*/ [dry trays]* [ jet flooding]*:excess loading/ fouled trays/ plugged holes in tray/ restricted transfer area/ poorvapor distribution/ wrong introduction of feed fluid/ [ foaming]*/ feed temperaturetoo low/ high boilup/ entrainment of liquid because of excessive vapor velocitythrough the trays/water in a hydrocarbon column; [downcomer flooding]*: excessiveliquid load/ restrictions/ inward leaking of vapor into downcomer/ wrong feed intro-duction/ poor design of downcomers on bottom trays/ unsealed downcomers/[ foaming]*; [ foaming]*: surfactants present/ surface tension positive system/ operat-ing too close to the critical temperature and pressure of the species/ dirt and corro-sion solids. [Dry trays]*: flooded above/ insufficient reflux/ low feedrate/ high boilup/ feed temperature too high. [Damaged trays]*: leak of water into high molar massprocess fluid/ large slugs of water from leaking condensers or steam reboilers/startup with level in bottoms > design/ attempt to overcome flooding by pumpingout bottoms at high rate/ too rapid a depressurization of column/ unexpectedchange in phase.

Section 3.5 steam traps.No specific entries for “cycling”“No condensate discharge”: bypass open or leaking/ scale in the orifice/ plugged

strainer/ inlet pressure too high/ for inverted bucket trap, bucket vent clogged,incorrect Dp across the orifice.

Comment:The focus in Figure 6-5 was on the causes related primarily with the control sys-

tem. The bold items listed above do not seem to be explicitly included in the chart.Scrutiny of these suggests that no major issues have been left out consideringMichelle’s purpose in creating this chart.

iii) The data gathered to test the hypothesis or probable causes. Check the evi-dence from the questions and tests that were gathered. The general optionsand the biases and interpretation issues were considered in Section 6.1 andanalyzed in Step 5, Section 6.5.5.

231


Example 6-31: Case’3From Figure 6-6 we note that Michelle focused on evidence [2], [3] and [4]. Sheinferred [6] and [7]. She should have explicitly checked this by setting the flow= con-stant; noting if the valve stem was constant and then checked for cycling [5]. BecauseMichelle did not do this explicitly, conclusion [8] is not proven. What Michelleshould have concluded from evidence [3] and [4] is that the signal to the steam valveworks.

If she had shown that the steam flow is constant and yet the level cycles [5], thenshe could conclude that the steam control system is not at fault [1].

iv) The assumptions. Focus on the assumptions and on consistency.

Example 6-32: Case’3The two assumptions Michelle noted were [9] and [10]. Assumption [9] would havebeen easy to check.

v) The conclusion. Strong evidence should support the conclusion; or a largecollection of weak to moderate evidence. There should be no strong counter-arguments nor strong assumptions.

6.5.8Step 8: Assess the Strengths of the Counterarguments

Use What if? counterexamples to help clarify the reasoning. These can be thereverse of the assumptions.

6.5.9Step 9: Evaluate the Consequences and Implications

Once we have accepted a valid conclusion, explore the So Whats? In most trouble-shooting situations this means identify another feasible cause or, once the cause hasbeen identified, to suggest remedial or corrective action.

6.5.10Activity 6-14

Critique the arguments described in one of the Cases presented in Chapter 4.

232

6.7 Summary

6.6Feedback and Self-Assessment

Reflections:

From this chapter, what have you discovered about trouble shooting? about planningand performing tests? about asking questions? about reasoning and drawing conclu-sions?________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Rate yourself on the following already dothis

might workfor me

not for me

Diagnostic actions– systematic and structured pattern to obtain information

& & &

– use criteria to select tests & & &– unique style and potential biases & & &Consistency– separate fact versus opinion

& & &

– cause–symptom data & & &– symptom–cause relationships & & &– identify root cause versus “symptom” & & &– rules for English, mathematics, science fundamentals & & &Classification– use single basis/criteria per level

& & &

– apply classification principles to identify symptoms & & &Patterns– methods to identify patterns

& & &

Reasoning– systematically use a process similar to the 9-step

& & &

– draw/construct symptom‹ cause diagrams & & &– diagram an argument and use this to critique & & &– well-developed methods to identify assumptions & & &

6.7Summary

A variety of critical-thinking skills are used when we trouble shoot. The first critical-thinking skill relates to how we gather information about the process. As we selectdiagnostic actions to gather information and evidence about the system we usuallystart with general background information (although if we are familiar with the pro-cess we know this already). Then we build up our mental image of the system andcontext by determining what happened recently, reminding ourselves of the detailsof what should be happening and then exploring what really is happening.

233


In testing hypotheses/causes we should start simply and test the most-likely causeearly. Above all we should refrain from the expensive tactic of shutting down, open-ing and inspecting. Much crucial information can be gained from the instruments,simple calculations and sound reasoning. A range of biases and mistakes made ingathering and interpreting data are listed: confirmation bias, over-interpretation,under-interpretation and mis-interpretation, availability bias, premature closure,anchoring. Jungian typology (or MBTI) dimension P-J provides a useful indicator ofone characteristic of your personal style for gathering and interpreting data.

A second critical-thinking skill relates to the importance of being consistent in ouruse of words, and our use of knowledge about processes and process equipment.Specifically the focus was on identifying fact versus opinion, gathering accuratecause–symptom data and astutely reversing the connection to link symptom–cause.Our usage should be consistent with the rules of English, mathematics, the funda-mentals of science and engineering and practical experience.

A third critical-thinking skill is classification; the process of dividing large sets ofinformation into meaningful parts. In making the division we should use a singlebasis/criteria per level, use no single entries and avoid faulty coordination or subor-dination. This was illustrated for the task of classifying the starting information into“symptoms” and “triggering” events.

A fourth critical-thinking skill is to be able to identify patterns.A final critical-thinking skill discussed here was reasoning. A systematic 9-step

process is suggested and illustrated in the context of part of Michelle’s reasoning inCase’3.

6.8Exercises

1. Harry lives on the eleventh floor of an apartment building. Each morning herides the automatic elevator down to the first floor and goes to his specializedjob in the cramped quarters assembling components in the airplane fuselage.Each evening he rides the elevator up to the seventh floor and then walks upto his own floor. Why?

2. Mr Tom Jones goes to the doctor’s office twice a week to pick up his pills.This particular Tuesday his wife decides to go with him so that she can con-tinue on and do the shopping afterwards. Tom parks the car in front of thedoctor’s office, leaves his wife in the car and goes in to get the pills. While heis talking with the doctor, he hears a terrific crash, rushes to the window andupon seeing his car completely demolished exclaims “My wife’s been killed!”whereupon the doctor reaches into the desk drawer, pull out a gun andshoots Tom. Why?

234

6.8 Exercises

3. Check out the mathematics in the following:

n2 ¼ nð 1þ 1þ 1þ 1þ . . . . . . þ 1|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}

nÞ

n2 ¼ ð nþ nþ nþ nþ . . . . . . þ n|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}

nÞ

dn2

dn¼ d

dnð nþ nþ nþ nþ n . . . . . . þ n|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}

nÞ

2n ¼ ð 1þ 1þ 1þ 1þ . . . . . .þ 1|fflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl{zfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflfflffl}n

Þ

2n= n

2 = 1

235

237

In any problem, people are involved. You as trouble shooter are the major person wemight focus on. Yet there are others: the operators, the maintenance team, otherteam members, the designers, your supervisors and your colleagues. Interpersonalskills are needed as you work with others. These include skill in communication,listening, building and maintaining trust and building on personal uniqueness.

Furthermore, some factors affect performance: our performance and that ofothers. These factors include pride and willingness to risk admitting a mistake,stress and distress, and the environment.

Consider interpersonal skills and then factors that affect performance.

7.1Interpersonal Skills

Included here are skills in communication, listening, applying the fundamentals,building trust and accounting for personal uniqueness.

7.1.1Communication

Communication is speaking and writing that a) correctly identifies multiple audi-ences, answers their needs and questions; b) has content that includes evidence tosupport conclusions, c) is well organized with summary and advanced organizers,d) uses a style that is coherent and interesting, defines jargon or unfamiliar words,and e) includes a format that is grammatically correct and follows the expected for-mat and style. These five elements should characterize every communication. Inaddition, in TS situations requests for changes in operation and for tests should bein writing and should consider the safety of the operator.

Activity 7-1: CommunicationCritique the following request from the point of view of audience, content, organiza-tion, style and form. Then, if pertinent, rewrite the communication.

7

Polishing Your Skills: Interpersonal Skills and Factors AffectingPersonal Performance


7 Polishing Your Skills: Interpersonal Skills and Factors Affecting Personal Performance

a. A verbal request from engineer Jose to process operator, Marco. The contextis that Jose is trouble shooting the process and believes that the cycling ob-served in the controller on the reflux relates to the changes in composition inthe overhead. Jose is looking for a time variation.“Take about a dozen samples of the overhead and send them to the lab for a full

analysis. Tell them it’s urgent.”b. In March, hourly data have been collected on the day shift over three weeks

so that you can do a detailed analysis of the process operation. In April theprocess will be shut down for routine maintenance. Andre realized it wasimportant to ensure that the meter readings collected over the test periodwere accurate. The last week of March Andre wrote a memo, not an e-mail,to the Pulp Mill engineer:“Would you please request that the instrument shop determine the calibrations

on the six flowmeters during the April shutdown. Thanks.”c. “Record the temperatures at the top of the distillation column once every

shift.” Fact. There is only a temperature indicator at the top of a 30-tray, self-standing tower. Question: how many operators do you think actually climbthe tower in the rain to read the instrument?

7.1.2Listening

Listening includes focusing attention on the talker, avoiding distracting behaviors,showing respect and frequently acknowledging through appropriate body languageand “ahums” and reflecting statements. The process can be modeled as Sensing,Interpreting, Evaluating and Responding or SIER. That is, we sense the message;we internally interpret what is being communicated; we evaluate the message in thecontext of the situation, our feelings, needs and goals; and we select how to respond.

Here’s what we know about listening:

. sensing the message is complex because about 55% of the message is com-municated by body language, 38% by tone and 7% by the words;

. listening is about four times slower than thinking;

. about 80% of our waking hours are spent in verbal communication; withabout 1�2 of that spent listening;

. untrained listeners understand and retain between 25–50% of a conversa-tion;

. only about 5% self-assess themselves as being highly skilled listeners.

In TS, we usually need to gather key information from others. We ask them ques-tions; we listen to their answers. Trouble shooters need to encourage the person tocommunicate clearly, to listen carefully and to interpret the answer correctly. Threelistening skills to aid this process are: attending, following/tracking and comprehen-sion checking/reflecting. Here are the details.

238

7.1 Interpersonal Skills

Attending: posture is inclined forward and open, facing squarely approximately1 m apart; no distracting behavior and eye contact is called “soft focus” (contrastedwith looking away or staring).

Tracking/following: provides minimal encouragement (for example, “Tell me more”,“sure..” “Oh..”, “ Then ..”) and infrequent questions (for example, prefer “What?”questions to “Why?”) and attentive silence.

Reflecting is responding with a concise restatement of the content and feelingsexpressed in the listener’s own words. That is, include the content and feelings ofwhat was said, express it in the listener’s own words, do not add new ideas, do notleave out ideas. Some example approaches include saying “As I understand it..” or“Are you saying that ..” Reflecting is usually used when someone is very emotional,or when you see differences developing between you and the other person, whenthere is disagreement, when the talker seems to be confused or when the talkerneeds encouragement that his/her contribution is valuable.

Activity 7-2: ListeningAssess the quality of the listening by the engineer in the following situation. Notethe five strengths and the two areas to work on.

Engineer Ahmed goes out to the control room for Case’7, the case of the reluctant crys-tallizer. Upon entering the control room Engineer Ahmed says, “Hi Phil. I’m new here. Iunderstand from Frank that you’re a great plant operator. I hear that we’ve got trouble onthis VC. Let’s see, that gauge is vibrating all over the place, so is that one. Hey, this is azoo. Don’t worry I’ll solve it soon.”

Activity 7-3: ReflectingFor Case’7, operator Phil said “When the liquid level started to drop I went up andlistened to the booster ejector and it sounded fine. The only thing I noticed was thatthe pressure gauge on the bay water line to the booster condenser was fluctuatingwildly.”

Which of the following might you use to show Phil that you are listening to him?

a. “Ahum”b. “Ok, please continue.”c. “As I understand it, the booster ejector sounded as you expected when the

liquid level started to drop and the pressure gauge on the bay water line reallyfluctuated. Is that correct?”

d. “Why is that noteworthy, Phil?”e. “I’m listening”f. Other

7.1.3Fundamentals of Interaction

The fundamentals of interaction are summed up in the seven RIGHTS and the fourdestroyers. Consider each in turn.

239


. Claim and honor the seven fundamental rights of individuals, RIGHTS(Woods, 1994)R the right to RespectI the right to Inform or to have an opinion and express itG the right to have Goals and needs and express these.H the right to Have feelings and express them.T the right to have had Trouble and make mistakes and be forgiven.S the right to Select your response to other’s expectations.and the right to Claim these rights and honor these in others.

. Avoid the four destroyers of relationships (Woods, 1994):ContemptCriticismWithdrawal and stonewallingDefensiveness.

I remember these by recalling the phrase “Did the contemptuous critter sit on defence or the stone wall?”

Activity 7-4: Fundamental rights and destroyersAnalyze the conversation and identify claiming of rights, honoring rights in othersand evidence of the four destroyers.

Two engineers Tonya and Marcos are trouble shooting Case’8: the depropanizer:the temperatures go crazy. Let’s listen in on their conversation. The parts of the con-versation are coded to guide discussion.

Tanya: “OK, the six hypotheses are (i) 1. tray collapsed in the stripping section, 2.too much bottoms fed to the debutanizer, 3. too much overheads in the feed, 4. feedvalve FV1 stuck, 5. pump F-26 not working, 6. not enough feed to the column (ii).What do you think? (iii)

Marcos: “It’s got to be a collapsed tray. (iv) I encountered something like that onthe S256 plant last year. Same evidence. It’s got to be a tray. (v)”

Tanya: “Hey! (vi) You’ve done it. (vii) You’ve zeroed in on one hypothesis when weneed to keep an open mind and do some simple checks first. (viii)”

Marcos: “Don’t get huffy about it. (ix) I’m just trying to get this solved fast. (x) Myexperience tells me it’s the tray! So what great insight are you bringing to this prob-lem? (xi) Ms. Smarty. (xii) Besides your hypothesis’6 is a symptom, and not a rootcause. (xiii)”

Tanya: “Now you resort to criticizing my hypotheses. (xiv) That’s not fair (xv)”

7.1.4Trust

Build and maintain trust. Trust glues relationships together. Trust is based on integ-rity, competency and benevolence (not doing anything that will hurt the other pur-pose).

240


. We build trust by such acts of integrity as:– keeping commitments to yourself and others.– clarifying expectations that you have of yourself and of others.– showing personal integrity, honesty and loyalty to others.– promptly and sincerely apologizing when you know you are wrong.– honoring the fundamental RIGHTS listed above and avoiding the

destroyers.– taking time to see things from the perspectives of others.– accepting others “warts and all.”and by such benevolent acts as:– not saying ill of the person behind his/her back or when they are not

present.

. We destroy trust by– the reverse of the Builders of trust listed above, and– not meeting commitments.– selectively listening, reading and using material out of context.– not accepting the experience of others as being valid.– asking others to give up their fundamental RIGHTS.and such non-benevolent acts as:– making changes that affect others without consultation.– playing the broken record until you�ve eventually worn them out.– subtly making changes in the context/issues/wording gradually so that

they are unaware of what is happening until it is too late. They were side-swiped.

Activity 7-5: TrustSomeone requests “Would you please be chair of the upcoming conference. It won’ttake much work and you are the ideal person to do it”. Your situation is that youhave been chair of a similar conference before; this would take the equivalent of atleast 2 months of concerted effort. You have promised your family to spend moretime with them. You are just barely managing to meet your commitments now. Theconference will draw many from abroad and being chair would bring you a lot ofpersonal satisfaction as well as increase your visibility and reputation. How do yourespond?

Activity 7-6: Self-assess trustComplete the inventory about trust given in Worksheet 7-1.

241

7 Polishing Your Skills: Interpersonal Skills and Factors Affecting Personal Performance242

Worksheet 7-1: Trust.

Trust is having the confidence that you can mutually reveal aspects of yourself-and your work without fear of reprisals, embarrassment or publicity.

Trust works both ways: you trust them and they trust you. Trust is not devel-oped overnight, trust takes time to develop. Trust can be destroyed by one incor-rect act.

Check your current status

Building your trustworthiness getting them totrust you

alreadydo this

needssome work

need lotsof work

unsureif this isfor me

1. Do what you say you will do. & & & &2. Be willing to self-disclose: don’t hide

your shortcomings; share yourself-honestly.& & & &

3. Listen carefully to others and reflect tovalidate your interpretation.

& & & &

4. Understand what really matters to others;do your best to look out for their bestinterests.

& & & &

5. Ask for feedback. & & & &6. Don’t push others to trust you more than

you trust them.& & & &

7. Don’t confuse “Being a buddy” with trust-worthiness.

& & & &

8. Tell the truth. & & & &9. Keep confidences. & & & &10. Honor and claim the 7 RIGHTS. & & & &11. Don’t embarrass them. & & & &

Checking your trustworthiness do they trust you? always mosttimes

some-times

don’t thinkapplies

1. Do they disclose confidential informationtrusting that you will keep it confidential?

& & & &

2. Do they assign you challenging tasks to dowithout frequently checking up on you?

& & & &

3. Do they honor your RIGHTS? & & & &4. Do they seem to look out for your best

interests?& & & &

5. Honest and forthright. & & & &6. Do not leave you feeling that they haven’t

told you everything about the situation;they seem to be holding back.

& & & &

copyright 1999 �, Donald R. Woods


7.1.5Building on Another’s Personal Uniqueness

. The Unique You and the Unique Them. Each of us has our biases, prejudicesand preferences or style. A variety of questionnaires and inventories can beused to help understand preferred styles for managing conflict, making deci-sions, applying creativity, differing style of conversing, validating ideas, gath-ering and using data, accounting for facts versus feelings and consideringdetails versus the big picture. Inventories of Johnson and Johnson (1986),Kirton (KAI: 1976), and Jung (MBTI: 1984) are examples of such inventories.Your style is unique; it will differ from others. Accept, respect and improvethe quality of your interaction through these differences. Do not let these dif-ferences lead to conflict. Some applications of this are given in Section6.1.3.3.

Activity 7-7: The unique youYou are on a trouble-shooting team with the following people whose personal styleare given in the following table. Write in your scores.

Johnson style for conflict Jungian Kirton

Withd Accom Force Comp Negotiate IE TF PJ SN

You

Marie –1 2 –3.5 3.4 9 I T J S 82Phil –6.1 1.3 –5.8 0.6 7.5 E F P S 98Jean –1.1 3.7 –3.1 2.4 6.7 E T J S 87Terry –4.1 3.5 –6.4 1.2 7.8 E F P N 83

1. Where are your blind spots? Describe this in actions. For example, if you area dominant S then your blind spot might be seeing the big picture, focusing toomuch on the details.

2. Does the team have any blind spots?3. With whom might you have minimum differences? What are those differ-

ences?4. With whom will you have maximum differences? What are those differences?

5. How can you build on this to trouble shoot efficiently and well.

We should also recognize the personal tendency or bias to prefer to report inter-pretation and inferences, instead of “just the facts”. This is discussed in Section7.2.5.

243


7.2Factors that Affect Personal Performance

If an operator makes a mistake, the type of mistake is likely to be:

90% no action taken (when some kind of action was needed),5% took corrective action but moved the correct and appropriate variable in the

wrong direction. Thus, he/she knew that the temperature should bechanged but increased it instead of decreased it.

5% took corrective action on the wrong variable. Thus, he/she should havechanged the temperature but, instead, changed the composition to thereactor.

Why do operators tend to take no action when a clearly defined action is needed?Such factors as inability to admit error, stress, alienation and lack of motivation, ten-dencies to infer and an “I know best” attitude all affect anyone’s ability to perform atask. These factors can affect the operators and people with whom we must interact,and these factors can affect us.

7.2.1Pride and Unwillingness to Admit Error

Each person has his/her own pride and self-esteem to keep intact. We do not wantto admit:

. that we are wrong;

. that we made a mistake;

. that we are guilty of wrongdoing.

People want to behave so that they “look good” in the eyes of those who matter...their spouse, their colleagues, their supervisors, themselves.

Not all evidence may be presented because some mistakes may have been madeor we are embarrassed and do not want to show we have made a mistake.

Example 7-1:Marg, the design engineer, is now part of the startup team. She is convinced thatshe designed the reactor correctly. It may be difficult for her to work with data thatsuggests the design is wrong.

Example 7-2:An operator, while making a preliminary inspection, was to observe and not touch.However, he opens and closes a valve. However, the valve does not go as far shut ashe thinks it should.

Example 7-3:For a working process, an operator accidentally drops a mercury thermometer intothe receiving and blend tank. The mercury will be dispersed throughout the feed

244

7.2 Factors that Affect Personal Performance

but “the concentration will be very low” and the “glass should be all broken up any-way.” In a couple of weeks the catalyst activity seems to fall off.

Example 7-4:The new regulation for energy conservation requires that all bypass valves on steamtraps be closed. Bill strongly disagrees with this policy and cracks open the bypassvalve. He thinks the process works better that way. Later it is found that the temper-ature control on the reboiler seems to be inadequate.

Example 7-5:Engineers may visit the plant site, but they may not touch or adjust things. Peter, ona recent visit to the site, notices a small leak around a valve stem. Believing thatsometimes a leak can be stopped by turning the stem slightly and then returning itto its former position, he tries this. However, the valve jams after he turns it fromhalf open to quarter open. He cannot get it to return to half open. He leaves andreturns to his office. A few minutes later he is called out to the plant because thereis trouble on the plant.

7.2.2Stress: Low and High Stress Errors

We encounter stress as an accumulation of events from home, work and play. Thework environment also provides daily stress: through the amount and type of inter-ruptions, the noise and cleanliness of the environment and the complexity of thetasks being done. The amount of stress one has experienced affects our perfor-mance. If there is not much stress or there is too much stress in our life, we areprone to make mistakes.

Powers and Lapp (1983) and Kletz (1986) summarize the probability of differenttypes of operator error to occur. For tasks involving sensing – interpreting – acting,if the person is well trained, and motivated and with no stress then,

. he/she will make about 1 error in 1000 trials if feedback is given to the per-son after they have made the action;

. he/she will make about 1 error in 100 trials when there is no feedback to theperson for the action taken.

if the person is under distress – not because of high stress levels cumulatingthroughout the year – but because of the situation, then

. he/she will make about 1 error in 10 trials. This might happen in a busyoperating center where other alarms are sounding, the telephone is ringingand people are asking for information about a part of the process.

. he/she will make about 1 error in 2 trials if, for example, many complexactions are required and the implications if an error is made are frightening.

245


The distress comes from poor training, confusion because of poor training, con-flicting data; from the need for fast action; or from a large penalty if a mistake ismade, or from extensive confusing and contradictory types of demands.

The training should be competency based and oriented to develop skill, and notjust to develop knowledge. Training is particularly important when new technologyis introduced, for infrequent events and for new people.

When we TS, it is wise to monitor our own stress and to be sensitive to the stressexperienced by others.

Activity 7-8: Stress and controlOne common cause of stress is that we worry about – and get angry over – thingsover which we have no control. Research evidence suggests that we have controlover only 1 out of 10 things we might worry over.

Trouble shooting can be particularly stressful. You are under a lot of pressure tofind the root cause and correct it safely and quickly. You are trouble-shootingcase’12 given in the list below. Which of the following do you have control over?

246

Worksheet 7-2: The problem given in Case’12 arises at 8:30 am. The followingbackground and resources are available. Identify which ones you think aredirectly under your control. If you think you have control over the item, circle Y;if you do not have control over it, circle N.

1. The safety officer, Hack, is overbearing, not liked andgets carried away about simple things. Y N

2. The laboratory can analyze liquid samples with their equipmentbut gaseous samples cannot be analyzed because theirinstrument is broken. Y N

3. The lab schedule is busy with top priority analyses.Samples could not be analyzed until after 3:00 pm. Y N

4. The union prevents you from analyzing any samples;if you do, there will be a strike. Y N

5. The upstream styrene plant is operating at 80% capacity. Y N6. The upstream ethylene cracking furnace is operating

at 95% capacity. Y N7. The upstream propylene plant is shut down. Y N8. The operator on the ethylene plant is cooperative. Y N9. Samples can be taken at any of the sewer gates within

the Battery Limits and at A, B, and C. Y N10. No blueprints are available for the sewer system. Y N11. Your performance review to establish your salary is being

done next Thursday at 4:30 pm. Y N12. You have to prepare your “record of progress” record

for the performance review on Thursday. Y N


Activity 7-9: Stress and self-talkA common contributor to your stress is your self-talk. Too many people say to them-selves “You are stupid” “You can’t do this”. Make a list of any negative self-talk com-ments you say to yourself-during a week. Create an action plan to minimize yournegative self-talk and to maximize your positive self-talk.

Case’12 Safety on the ethylene plant (courtesy of John Gates, B. Eng. 1968,McMaster University)The part of the ethylene plant that relates to this problem concerns the drying sec-tion to remove moisture from the feed gas and the distillation train to separate thegas stream into the desired component section. Drying section: Three aluminadryers are installed. One dryer is regenerated while two are hooked in series onstream. For example, dryer V106 is being regenerated with V107 and 108 removingthe moisture in the process stream to less than 4 ppm. Then V107 will be regener-ated with V108 and 106 in series and so on. The cycle lasts 12 hours. The dried gasgoes to a knockout pot (to remove any entrained material) and then is chilled, inexchangers E107 and 108, before entering the separation towers.

During regeneration of the dryers, “fuel gas”, heated with 2.8 MPa steam, flowsthrough the dryer in the direction reverse to normal flow. Once it is through thedryer the regeneration effluent gas is cooled and returned to the fuel-gas system.During regeneration the dryer temperature rises to 190 �C and is maintained at thistemperature for one hour. Then the fuel gas bypasses the heater and is sent directlyto the dryer to cool it. For this plant, the “fuel gas” or regeneration gas is the non-condensable overhead from column T 101, the demethanizer.

The dryers are all appropriately manifolded and valved so that any dryer can beregenerated, bypassed or used. The regenerating dryer is separated from the linedryers by gate valves.

The separation is performed in a train of three distillation columns operating atabout 3.2 MPa. These are a demethanizer, de-ethanizer and C2 splitter, T101, T102and T103, respectively.

The process is illustrated in Figure 7-1. The overhead from Tower T101 is con-densed with ethylene as refrigerant. The overhead from Tower T102 is condensedwith propane as refrigerant. The overhead from Tower T103 is condensed with pro-pylene as the refrigerant.

The safety inspector telephones to say that this morning’s gas samples from thesewer drop boxes within our battery limits (for ethylene distillation units) are explo-sive. These sewers serve other plants: styrene, ethylene furnaces and propyleneplants. The plot plan of the unit is shown in Figure 7-2. Clear up this matter imme-diately; this hazard cannot be permitted!

247


Figure 7-1 The ethylene plant for Case’12.

T101 T102 T103

V108V107V106

E130

E131

A B C

Battery limits

To disposal

from Styrene,

ethylene &propylene

Sewer gates

Figure 7-2 Plot plan of the ethylene plant for Case’12.

248


7.2.3Alienation and Lack of Motivation

The working environment should motivate employees. Employees should have aclear idea of the expectations, and rewards should be forthcoming for achievingthose expectations. People are demoralized if people considered to be “stars” arealways given the challenging work, if information sessions about the company arenot given and if the promotion and salary adjustments are unclear or seem to bebased on who you know and not on performance. People are alienated when a) taskslack imagination or are to be performed under strained conditions, b) where theworking conditions are sloppy maintenance, extremes in temperature, odors anddust and hazards, c) where the emphasis is on machines ... and not on people, andd) when management decisions lack consistency, predictability and transparency.

7.2.4“I Know Best!” Attitude

In other situations, sometimes, people deliberately fail to follow instructionsbecause “they know better”. An operator opens the bypass valve on the steam trap oroverrides a control system because “those silly systems don’t work! I know best.”This attitude is often related to increased alienation and lack of motivation.

7.2.5Tendency to Interpret

Instead of saying “the gauge reads ...” we tend to say “the temperature is ...” It isvital that a trouble shooter listens carefully to what is said and how he/she interpretsthe information.

Communicating just the facts is boring. We want to describe what we think iswrong.

Example 7-6:The designer might say “the rogue data point for the heat capacity of acetic acid vapor,shown in Figure 7-3, is probably caused because the instruments or the researcher made amistake" and proceed to design the preheater assuming the heat capacity at 200 �C isabout 0.45 CHU/lb �C.

Comment: the rogue point is correct.

249


Figure 7-3 Heat capacity of acetic acid vapor.

Example 7-7:The operator might say that the valve on the bypass is misbehaving again; see theinstabilities in the flow control of the reactor feed. Is it really the valve? Is there really aninstability?

Example 7-8:The engineer might suggest that the reason the low yield for the reformer, comparedwith the expected yield, is caused by low catalyst activity.

Example 7-9:The engineer says that the pressure in the line is 50 kPa; the temperature is 450 �C.Comment: the engineer should have said the pressure gauge reads 50 kPa; the ther-mocouple reads 450 �C.

Activity 7-10: Triad talkingIn groups of three, one is the “talker”, one the “fact-summarizer” and the other is the“inferrer”. The talker describes, for three minutes, his/her greatest frustration, favor-ite activity or hobby. The fact-summarizer takes one minute to summarize the factsgiven by the talker. No personal opinions or inferences should be given. The inferrerdescribes in one minute what inferences and messages came through from theideas presented, by the body language and by what was not said. Rotate responsibil-ities so that each has a chance to play all three roles. This activity takes 15 minutestotal. At the end, share your experiences.

250


Activity 7-11: Pair talking and “just the facts”In pairs, with one being the talker and the other the listener. The talker reads overCase’13 “The Lousy Control System” to himself/herself-for about five minutes.Then, referring only to the scenario for details the talker has three minutes todescribe the facts in the Case to the listener. Do not simply read the scenario to thelistener. You may show your listener the sketch of the process but you cannot showthe written material. You have to describe the facts.

Change roles and repeat with Case’14, “The condenser that was just too big.”After each role, take two minutes and write out your reflections about the activity.

Discuss with your partner.

Case’13 The Lousy Control System (courtesy Esso Chemicals)All the texts on process control say that controlling the overhead condensate exittemperature by varying the fan pitch is a slow and clumsy method of control. Yet,that is the method used on column T6. Today it is raining, the temperature of thecondensate is 3 �C subcooled. Yesterday it was hot and sunny. The fans were run-ning flat out but still the gas exit valve controlling the pressure was open half-waymost of the day. All the uncondensed gas went to flare. The boss storms in “You’vegot trouble on this plant; too much stuff went to flare yesterday. What’s wrong?”You’re sure that the control system is lousy. The system is shown in Figure 7-4.

PC100

TC200

LC202

TO

FLARE

PSV1

PI201

Figure 7-4 The control system for Case’13.

251


Case’14 The Condenser That was Just too BigFatty acids are solid at room temperature. Fatty acids are purified in a distillationcolumn operating under a very high vacuum. The “coolant” in the overhead conden-ser cannot be water because the fatty acids would solidify in the condenser. Insteadthe coolant is boiling water with the “coolant” temperature controlled by the pres-sure at which the water boils. This plant is started up for the first time. A commonfault in startup situations is that the condensers are over-designed for “clean condi-tions” because fouling factors have been used in the selection of the overall heat-transfer coefficient. This would mean that at startup the condensing fatty acidswould be subcooled. When you go out on the plant one hour after startup, the wor-ried operators say “Look at the overhead pressure gauge – it’s much too high. Itreads 2.7 kPa (20 mm Hg absolute) when it should be reading 0.4–0.7 kPa (3–5 mmHg).” Just as you expected! the fatty acids are being subcooled in the condensers.Solid is starting to build up in the tubes. The excessive pressure drop across thetubes means that the vacuum system can’t get rid of the air leaks fast enough andthe pressure builds up. You pull out the drawings showing the configuration of theplant (Figure 7-5). This problem needs to be solved fast!

Figure 7-5 Fatty-acid distillation column for Case’14.

252

TI

TI

TI

TI

PI

PI

PI

Cooling

Water

reciprocatingpump

60ºC

BackupCondenser

BoilingCondenser

240 kPa

to/fromDowtherm

system

Fatty acid

Feed

240ºCto

260ºC

Cooling water exit

Cooling waterCooling water

return

0.4 kPa

Booster

steam

wellwater

barometriccondenser

wetvacuumpump

120ºC

Coil

in tank

7.5 Exercises and Activities

7.3The Environment

The environment affects performance. If making a mistake is considered the worstpossible thing to do; then no one will admit to making a mistake. If mistakes areaccepted, understood and no serious repercussions occur, then people will be moreready to admit their mistakes.

Activity 7-12: Rate the environmentUse Worksheet 7-3 to help reflect on the environment and its impact this mighthave on the people with whom we interact.

7.4Summary

We interact with people when we trouble shoot. Our skill in trouble shooting oftendepends on our skill in communicating, listening, applying the basic fundamentalprinciples of interpersonal relationships, building trust and understanding our ownuniqueness and the uniqueness of others. These ideas were illustrated and activitiesprovided to give you a chance to work with these skills.

We also need to be aware of the factors that affect our performance and the perfor-mance of others. These include unwillingness to admit having made a mistake, andthe impact that stress has on our ability to work effectively. Sometimes people arefrustrated and unmotivated. Sometimes people follow their own approach eventhough it may result in incorrect and even unsafe operation of the process. We alsoprefer to infer and interpret what we see or hear. Reporting the facts is usually bor-ing. So we read a gauge and say “The pressure is 1.4 MPa” when in reality we shouldhave said “The pressure gauge reads 1.4 MPa.”

A questionnaire is included to give us a chance to evaluate the environment inwhich we trouble shoot.

7.5Exercises and Activities

1. Consider ten cases in Kletz’s text “What went wrong?” From Kletz’s descrip-tion of the case, classify the cause of the equipment malfunction, humandesign mistake, human maintenance mistake, human operator mistake orother.

2. Typical feedback from industrial workshops on trouble shooting.

For Activity 7-10: Fact-summarizing was either easy or difficult depending on the qual-ity of the talker’s presentation. Taking notes helped. Inferring was most enjoyable. Com-pare your experience with this.

253


For Activity 7-11: Giving facts was difficult. Take my time. Repeating was difficult.Write everything down. Compare your experience with this.

3. Cases to consider. For Cases ’15, 16, 17 and 18 create hypotheses as to thecause.

254

Worksheet 7-3: Feedback about your environment.

To what extent do you agree with the following descriptors of your environmentwhere you usually “trouble shoot”.

People are willing to admit error: “The people that I work with are veryunwilling to admit errors; they blame others, they pass the buck and, if neces-sary, would purposely mislead me rather than to admit error.”

Strongly Moderately Slightly Slightly Moderately StronglyDisagree Disagree Disagree Agree Agree Agree1 2 3 4 5 6________________________________________________________________

Encourage risk taking: “Risking is rewarded. We are expected to take risksabout 10 times a day. Risks should be wisely, not indiscriminately selected. But,nevertheless, we are not only encouraged but we are rewarded for risk taking.”


General stress at work they are under: “Their environment is very stressful.People have many deadlines and interruptions. The consequences of makingmistakes is very high. The issues are complex. The environment changes oftenand includes a lot of uncertainty.”


7.5 Exercises and Activities 255

General stress at work you are under: “My environment is not stressful. I donot have many deadlines or interruptions. The consequences of making mis-takes are low. The issues are straightforward. The environment is safe, stableand secure.”


People’s listening and responding: “The people are open, communicate well,can clearly identify facts, will offer opinion when asked for, and are very compe-tent but are not aggressive “know-it-alls””.


Case’15 The flooded boot (problem supplied by W.K. Taylor, B. Eng. McMaster1966)As part of our energy conservation program we recycle condensate to our boilerhouse. This condensate comes from large, steam-driven turbines whose exhaustconditions are full vacuum. These turbines drive the feed gas, air, refrigeration andsynthesis gas compressors on the reformer section of the ammonia plant. Theexhaust steam goes to the shell side of a surface condenser. Chilled water is on thetube side. Vacuum is maintained by a two-stage steam ejector with inter and aftercondensers. These are located downstream of the surface condenser and pull a vacu-um on the condenser itself.

Lately we have been increasing the load on the compressors and hence the steamconsumption has increased. However, for these new conditions we can’t seem to getrid of the condensate from the boot at the bottom of the surface condenser. Wesometimes have to run the spare pump 122.JA (electric drive) because the turbine-driven pump 122.J cannot handle all the flow. “These high levels of condensate andthe unusual operation with our spare pump are keeping my fertilizer productiondown. It’s got to stop! My production rate is barely reaching 120 Mg/d and I thinkwe can easily produce 132 Mg/d. That’s $5000 a day. Fix this bottleneck!”. The sys-tem is given in Figure 7-6.


Figure 7-6 The configuration of the flooded boot for Case’17.

Case’16 The case of the dirty vacuum gas oil (problem supplied by R.J. Farrell,B. Eng. McMaster University, 1974)Word spread quickly. “There’s trouble on the vacuum tower!” Normally the sideproduct, vacuum gas oil, is water clear. Today the product has been off-specificationin color. Indeed “It looks brown” explained a frustrated supervisor. The feed to thevacuum tower is crude oil. The crude oil is heated by a series of three exchangersand the furnace. The first two exchangers extract heat from the vacuum gas oil (API27.8), and from the mid-side-draw cut that has an API gravity of 25.4. The thirdexchanger extracts heat from the tower bottoms (of API gravity 11.5). The columnhas pump-around draw-off for the bottoms, to recycle to crude, to mid-side, and tovacuum oil-overhead condensates. The mid-side and vacuum gas oil are blendedand cooled to yield a product of API 26.2. The blend is sampled after the cooler; theblend is off-specification. Fix the problem.

Case’17 Is it hot or is it not? (case supplied by Jonathan Yip, B. Eng. 1998,McMaster)The crude fatty-acid high-vacuum fractionation column is having trouble. The bot-toms recirculation and transfer pumps need maintenance immediately. To copewith this the engineer placed the column on safe-park. The feed was stopped andthe sump and overheads recycled back to the crude fatty-acid tank. The steam ejec-tors were turned off and all the valves were shut to keep the column under vacuum.

256

SURFACE CONDENSER CHILLEDWATER

6"

122-J

4"

122JA

LC-2

LC

-2A

LIC

PI

2"

1/4"

3"

IONEXCHNG

4"

FI

PI

PI

Bypass

6"

3/4

"

2"

6"

3/4"SEALING

WATER

2"

BOOT

6"

TO VACUUMSYSTEM

REFRIGERANTCOMPRESSOR

GASCOMPRESSOR

EXHAUST FROMVARIOUS TURBINE

DRIVES

7.5 Exercises and Activities

The pump was dismantled for repair and an hour into the repair job the plantoperator noticed that the temperature gauge at the top of the column was showingan increase in temperature. Indeed, the temperature alarms started screaming.

“That’s crazy!” said the operator. “There’s nothing going on in the column. Thecolumn is isolated. The column is not running. That temperature should be decreas-ing instead of increasing. That temperature gauge and those alarms have been tem-peramental for a long time.” He manually suppressed the nitrogen purge valve andcut off the alarms so that he could hear himself-think.

Case’18 The streptomycin dilemmaIn our process for the production of streptomycin we evaporate the eluate in tallrising film evaporators. These are fixed tube sheet with a shell expansion bellows toallow for the thermal expansion. Our plant operates on a single, eight-hour shift sothat after each shift the evaporator is shut down. However, because we process phar-maceuticals we must clean and sterilize all units after each shift. The procedure isto open the end of the bundle, to brush the inside of all tubes, rinse, fill the tubeswith water and then apply steam to the shell to boil and thereby sterilize the tubesin preparation for the next day’s operation. The plant has operated without a hitchfor the past four years. However, over the past months the tubes are starting to fail;leaks are developing at the tube/tube sheet and the tubes are buckling. The patternseems to be random. Get this fixed.

257

259

This is the time to put all your skills together and work on some trouble-shootingcases. The purpose is to polish your skill. In this chapter we suggest several ap-proaches you could take: work in triads or as an individual. Then an outline is givenof the different cases included in this book. Then details are given for the cases.Enjoy.

8.1Approaches to Polish Your Skill

You can work in triads or as an individual.

8.1.1Triad Activity

One of the most effective ways to develop skill is to use a triad activity in which eachperson, in turn, plays the role of expert system, trouble shooter and observer. In a2-hour period each can play each role once. Here are the three roles.

a. The Expert System

The expert system poses a trouble-shooting Case and then provides responses“from the system” to any request made by the trouble shooter.

In preparation for this role, 1) select a case; read it over carefully, 2) then trackthrough all of the data requests given at the end of the case; 3) locate the answersand then read Appendix E for a debrief and answer. Understand the process ex-tremely well. You might wish to complete a Trouble-Shooter’s Worksheet (fromChapter 2) for yourself. Think about how “the fault” will affect all of the process vari-ables. Try to anticipate the kinds of actions that the trouble shooter might take. Whatwould the fault do to the system under those conditions? Give the results of experi-ments. Do not give explanations. Give correct information but do not be generous.If, for example, the fault occurs periodically, and you are asked to give the lab analy-sis for one sample taken, then assume Murphy’s law applies and give them theresult when the system was operating normally. Insist that they write out all

8

Prescription for Improvement: Put it all Together


8 Prescription for Improvement: Put it all Together

requests; write down the results opposite. Do not talk. . . . . just acknowledge thatthey are working on it by saying “Ahemmm, mmmmm,”

Insist on their instructions or requests be written precisely. If they write “Inspectthe instrument” then respond “It’s OK”. If they ask what you did, then say “I wentout and looked at it.” Be tough. Do not offer more information than they asked for.

b. The Trouble Shooter

You have a challenging role to play.

. you are to talk aloud so that the Observer can track what you are doing. Thinkabout the process you will be using: are you searching for change? or for thebasics?; are you clarifying the situation or testing an idea? You may feel fru-strated because the Expert System is going to supply written responses to yourrequests for tasks to be done; the Expert System will not discuss things withyou. He/she may say “Hmmm” or “mmmmmm” to you occasionally so thatyou do not feel as though you are talking to a wall, but the Expert System isthere to provide an instant system written response to your requests.

. you are to display all the good problem-solving skills we have developed. Dothis by verbally monitoring your progress, being active with pencil and paperto keep track of the route you are following. Indeed, you might wish to pre-pare a Trouble-Shooter’s Worksheet. Recall, from Chapter 2 (and Chapters 5and 6), the performance characteristics of successful trouble shooters. Try todisplay those performance characteristics.

. you are to write out your requests for information from the Expert System.These should be written out precisely and unambiguously.

c. Observer for trouble shooting

Your worksheet is the Feedback form given in Chapter 2, Worksheet 2-2. As thetrouble shooter is tackling the problem, your task is to assess how well the problem-solving components are handled. This is challenging because the skills are difficultto identify – let alone observe and assess. To some extent your role is similar to thelistener role in TAPPS, described in Section 5. 1. The feedback form is made to helpyou look at the mental process used by the trouble shooter. Try to focus more on the“actions taken”, and on the “talk-aloud description of the thought process”. Look atthe organizational pattern used; listen for the monitoring of the process. Consider,“Does he/she confuse activities unknowingly?”

Let the Expert System focus on “how well the trouble shooter wrote out the ques-tions and tasks to be done”.

Activity 8-1

Prepare for the activity ahead of time with each person in the triad selecting a Caseand preparing for the role of Expert System. As expert system, make a copy of the Caseto give to the trouble shooter. When it’s convenient, the triad members meet for atleast 2 hours.

260

8.1 Approaches to Polish Your Skill

1) The person with surname first in the alphabet starts as trouble shooter. Next inthe alphabet, as expert system and last in the alphabet as observer. Seat your-selves approximately as illustrated in Figure 8-1, although the barrier be-tween the trouble shooter and expert system is imaginary.

Figure 8-1 The triad activity.

2) Refresh your memory as to how to play each role.3) Set the timer for 20 minutes, start with the expert system handing the Case to

the trouble shooter. The trouble shooter reads the case aloud and then, by talk-ing aloud proceeds to “solve the case”. He/she gathers information by writingout actions to be taken. These actions should be written one at a time with a

261


written response being given by the expert system to each task. The expert sys-tem responds in writing and so the role-playing activities continue. Focus onthe process and not on rushing to trying to solve the problem in the availabletime. In the time available, the trouble shooter may only complete the Trouble-Shooter’s Worksheet and never even ask a question of the expert system.That’s OK.

4) When the 20 minutes has elapsed, the expert system reveals the root causeand possible solution. This is not a discussion. This is simply the expert systemsharing the root cause about the case. Time= 2 minutes.

5) The observer completes the feedback form and gently shares his/her feedbackwith the trouble shooter about what he/she observed. 4 minutes.

6) The trouble shooter collects the evidence about the process: the case-problemstatement, any worksheets, perhaps the Trouble-Shooter’s Worksheet, andthe action-request form with responses that were written between the troubleshooter and the expert system.

7) All three people write their reflections of what they learned from the activity.5 minutes.

8) Discuss this briefly with the people in your triad. 3 minutes.9) Rotate roles and repeat until everyone has had a chance to play all three roles

and gather evidence about how they played the role of trouble shooter.

As an aside, although the primary goal of this activity is to improve your trouble-shooting skill through the role as trouble shooter, most participants report that theylearned the most from playing the role of expert system. Interesting.

8.1.2Individual Activity

As an individual, read over the cases given in Section 8.2. The cases are gradedaccording to degree of difficulty; some cases are written in a general context; theycould occur in any process. Others are process specific. For the case you selected,you might start by completing the Trouble-Shooter’s Worksheet given in Chapter 2.Then, scan the list of diagnostic actions given, prioritize, select your choices insequence and obtain the answer in sequence from the code in Appendix D. For exam-ple, in Case’8 you might decide that your first activity is “put on safe-hold”. The code is1880. From Appendix D, the result of this activity is 1880: “Not needed at the beginninguntil after we have collected data. Indeed if put on safe-hold we won’t be able to collectdata to figure out what is wrong. $3000” Keep a record of the codes you use. Continueuntil you believe you have identified the root cause or have corrected the fault.Check your answer with the fault encountered in industry, recorded in Appendix E.Reflect on the problem-solving process used and compare the sequence of actionswith those reported for the case in Appendix E. Total up the cost. You might alsogive yourself-feedback using the form given in Worksheet 2-2.

262

8.2 Cases to Help you Polish Your Skill

8.2Cases to Help you Polish Your Skill

First, the cases are listed and guidelines provided for each. Then the cases are giventogether with a set of diagnostic tasks from which you can select. These tasks havebeen listed from the choices made by previous trouble shooters. I hope that I haveincluded all the options you like.

8.2.1Guidelines for Selecting a Case

The actual process problems have been selected carefully from our files of severalhundred. These represent varying levels of difficulty, different types of situations(from startup to usual operations) and different varieties of equipment. Table 8-1summarizes these options. In selecting the sequence of cases to work on, I recom-mend that you start simply and build up your confidence and skill. The criteria touse in selecting the case are:

. Degree of difficulty. Although this is a subjective rating, I think it is useful.The basis for the ratings is given in Appendix E. Start looking at cases atLevel 3 and 4. The lower ratings start with Case’19.

. Type of equipment involved. The key pieces of equipment (plus people) thatare in the process are listed. For example, Case’19 relates to filtration andpumps. If your experience with process equipment is being developed for thistype of equipment, then you may wish to return to this case later. Alterna-tively, you can consult Chapter 3 and trouble shooting and suggestions forgood practice early in your consideration of the case. As you scan the diagnos-tic actions listed for each case, you might find the Vendor files and “call to thevendor” activities helpful to do early.

. Type of process. Some cases apply to any process. Others include characteris-tics that are unique to the industry. As a start, you may want to select casesthat are “general”. Alternatively, some information unique to certain pro-cesses may be obtained by consulting the diagnostic actions: MSDS, processdescription, andHandbook.

. Continuity. Some cases all relate to similar processes. For example, fourcases relate to the depropanizer-debutanizer system. Two, to the ethylene pro-cess; five, to the ammonia-reformer. The interconnections are given in col-umn six in Table 8-1. Once you are on a roll with one type of process youmight want to continue with problems for the same type of process.

8.2.2The Cases and Understanding the Choice of Diagnostic Actions for each Case

The cases are listed in Table 8-1. Each case has two parts: the case statement (anddiagram) and the list of coded diagnostic actions from which you can select. Here

263


are some more details about how I have designed the diagnostic actions (and thefeedback provided for each from Appendix D).

There are no diagnostic tests listed for the case that tells you the fault. The diagnos-tic actions provide you with sufficient evidence that you can identify the fault. Iden-tifying the fault, correcting the fault and preventing the fault from reoccurring areleft to you. But you have to select the most pertinent diagnostic actions to give youthe key evidence.

You are free to select any diagnostic action you wish from among those listed foreach case. Every action you request incurs a cost. Unless otherwise specified, I haveused $600/h as a cost for loss in production. To this needs to be added the cost ofthe people and equipment needed. For example, the diagnostic action of determin-ing “what’s the weather today and in the past?” costs $50 (representing about a10-minute activity). The diagnostic action of “sampling and analysis” would cost $3700to $15 000/sample depending on the type of analysis involved. (This represents a6–24-h activity). For each case I tried to assign reasonable costs for each action.

The diagnostic actions fall into four classes. The first actions relate to safety! Thesecond set of actions help you understand the background. The third set are factualgathering of information about the process as it operates now. The fourth set arevarious tests, calculations and activities to test hypotheses, and provide evidence asto the fault. As mentioned at the start of this section, the activities do not necessarilytell you the fault. Even the actions called Take Corrective Action may not correct thefault; they may, but they may not! Do not look for “the answer” to be among theoptions called Take “corrective” action.

Make your selections based on logic combined with safety and economics.Here are more details of the four sets of activities that you may select.

1. The first two activities relate to safety

The first priority is safety: recognize hazards (via MSDS) and take action.

. MSDS

Usually all engineers working on any site should know this information already.If you don’t for the particular case, MSDS information is available: this usually isexpressed as the three National Fire Protection Association, NFPA, ratings forhealth, fire and spontaneous reaction/explosion for individual chemicals. The rat-ings range from 0 to 4 with 0 meaning negligible and 4 meaning extreme. Thus anNFPA rating of 0, 4, 0 would mean that the species is not an issue for health, it isextremely flammable and it is stable. I tried to include information about how onechemical might react with another when this is important.

. Immediate action for safety and hazard elimination. The first decision relatesto safety. The initial evidence might suggest a health, a fire or an explosivehazard. Act now! There are four options:– Continue with the trouble shooting without implementing activities related to

safety: there is no hazard.

264


– Put on safe-park: this keeps the process going but under conditions thatare safe. This could mean isolating a distillation column and keeping iton total reflux; or reducing the throughput to conditions that previouslydid not pose a hazard.

– Safety interlock shut down, SIS: this should happen automatically if thecontrol system has been designed with the four levels expected. However,sometimes this has to be actively initiated. This gives you a chance toreflect on the situation and decide if this should have happened.

– SIS plus evacuation: the SIS should happen automatically if the controlsystem has been designed with the four levels expected. Now, because ofthe hazard posed we should add evacuation.

2. The second set of activities: background that you probably know

You probably know most of this information, except the weather and informationrelated to maintenance. However, as you are developing your skill this section pro-vides easy access to background information related to the case.

. More about the process. This is to provide some reassurance about what theprocess is about; this additional information might help. I have not includedkey information that is needed to solve the case in this activity.

. IS and IS NOT: this is based on given problem statement. This might help asfeedback to you about how you have done this task.

. Why? Why? Why?: may be helpful to put some of the cases in the larger con-text. You can do this activity on your own and use this question to give youfeedback.

. Weather Today and past. All cases refer to weather conditions in Ontario,Canada, where there are four seasons with snow and freezing weather forDecember through March; hot humid summers June to August. Many of thecases are sensitive to the weather.

. Maintenance: turnaround. Three conditions might apply: new plant startup,startup of an existing plant that has just been through its annual turnaroundand operation after some maintenance has been done.– That the problem is with a first time or startup of a new process is usually

given in the statement of the problem. Therefore, no separate question isposed related to this.

– The Maintenance: turnaround activity relates to startups after the annualturnaround. During the turnaround the minimum is usually– the inspection of most pieces of equipment,– the replacement of worn parts,– the installation of changes to the process, repiping, and changing the

operation to implement ideas to optimize or improve operation,– the calibration of sensors,– cleaning exchangers, and– changing the catalyst.

It is important to know “When and what done?”

265


. Maintenance: routine. During routine maintenance something will havechanged. It is important to know what and the extent. It could be that all theisolation valves were not correctly opened after the maintenance was com-pleted; that the key was not correctly fitted into the drive shaft; that the pumpwas not primed. The answers I give to this question will not admit to suchmistakes. However, this will open your mind to possible things to check andlook at.

. What should be happening based on office records and files: the resourcesinclude the simulation/design computer background; records of the design;the collection of information from the equipment vendors and any internalreports on past tests, or trials done on the plant. Handbook data, mainlyproperties, for the conditions and species in the case are available. Trouble-shooting suggestions are also available as are simple calculations you mightdo based on the given information in the case.

Design and simulation files (allowances made for fouling, overdesign and uncer-tainties). Here I tried to list all the possible pieces of equipment involved in the case.All equipment is designed according to the Codes and Standards; the informationgiven here are decisions within a designer’s judgement.

Vendor files: practical information and some specifications from the suppliers ofheat exchangers, steam ejectors, pumps and equipment that would be purchasedfrom a vendor. The information may be slightly different from the information inthe design files.

Commissioning data, P&ID, internal reports. Often new plants are constructedunder contract. The contract may include penalty clauses stating a financial penaltythat is paid if the process does not perform up to specification within a stated periodof time. For insurance and governmental regulations, certain performance stan-dards have to be met. Usually records are kept of the performance trials. However,don’t look for much information here for the startup of a new plant.

Handbook: physical and thermal properties of the chemicals in the case: vaporpressure–temperatures, steam tables, and thermal properties may be given.

Trouble-shooting files. Here I refer to specific sections of Chapter 3 that pertain tothe equipment in the case.

. Calculations and estimations that can be done in the office based on theinformation given in the case statement and the diagram (if available). Thegiven information varies from Case to Case. Nevertheless, some simplechecks and calculations can provide neat insight. These include:– Pressure profile: in most cases we can obtain the pressures on either side

of a barrier and then state the direction of flow that would incur if thereis a leak in the barrier.

– Mass balance: nice idea but you usually can’t do this at this early stage inthe problem because the key data are not given in the case statement.

– Energy balance: sink= source; the heat lost by one fluid= heat gained bythe other. Often we are given enough information in the case statementto allow us to do this check.

266


– Thermodynamics: use the principles of thermodynamics to predicttrends, in general.

– Rate: use DTs to estimate whether nucleate or film boiling might predo-minate. Similarly, we might be able to estimate rates of mass and heattransfer.

– Equipment performance: we might be able to estimate the performanceof equipment; usually, however, we need more information than thatgiven in the case statement.

3. The third set are factual gathering of information about the process as it oper-ates now.

. What is the current operation

Unless you arranged over the phone to meet the operator at the piece of equip-ment, you will go to the control room first and check in. You might scan the datareported in the control room; discuss the operating activities with the operator andexplore the operating procedures.

Visit control room: control-room data: values now and from past records.Process operators: you can ask for information about what happened (from their

perspective) and they might offer ideas as to what is the fault. The operators areusually a valuable source of information.

Operating procedures: knowledge of the usual operating procedure to be used forthis condition might help.

. Check with colleagues about hypotheses

The purpose of this book is to develop your confidence and trouble-shooting skill.To provide feedback about reasonable hypotheses (that you will have generated bythis time in your journey through the case), I give this diagnostic check, but only forthe Cases up to Case’29. You may wish to create a symptom ‹ cause diagram simi-lar to the one illustrated in Figure 6-5, in Section 6.5.5c but I give no feedback aboutthis for the cases.

4. The fourth set are various tests and activities selected to test hypotheses/orcorrect

This is the central action for each case. You may choose the diagnostic actions inany order. For example, you might suspect that the sensors are incorrect. One test isto gather data that checks for consistency. In the format of this text, I rarely includea section called “consistency tests”. I expect you to identify neat ways to check forconsistency. Methods might include:

1) to compare two sensors at the same or close-by location,2) to look for agreement between composition, temperature and pressure (say

at the top or bottom of a distillation column),3) to check for agreement between temperature and pressure on a pressure–

enthalpy diagram for pure refrigerant,

267


4) to check that the conditions on a stream are the same at two locations. Thislatter type of information we often obtain by contacting the operators of theutilities or of the plants upstream or downstream from our process calledContact with on-site specialists.

The diagnostic activities from which you can select are:

. visit the site, go out on the plant, see, smell, feel and listen; and read gaugeswhose values are not shown in the control room,

. check that the diagram given with the case agrees with reality, Check diagramand P&ID versus what’s out on the plant,

. do on-site simple tests, such as “turn and seal” test on valves; and check fortrends in the data.

. gather data needed to do key calculations,

. check the sensors. This includes checking the sensor’s response to change,using temporary instruments, and calibrating the sensors;

. consider the control system: put the system on manual, retune, call in instru-ment specialists.

. get information from vendors,

. gather samples and have these analyzed,

. domore complicated tests, such as gamma scans, tracer studies.

. open and inspect, and

. take corrective action.

Cautions must be given about the last two diagnostic options. Open and inspectis expensive and usually a last resort. Use this only when you have narrowed theproblem down to a specific fault and piece of equipment. As mentioned at thebeginning, the cause is not usually found in the list of Take “corrective” action.Sometimes this does corrects the fault. Sometimes it doesn’t.

When you have completed your set of diagnostic actions, then you should have allthe evidence needed for you to write down the fault, think about how to correct itand how to prevent its reoccurrence. Enjoy!

Now here, in Table 8-1, is the list of cases that I have selected to help you developyour skill in trouble shooting.

Table 8-1

Case number and name Degree ofdifficulty

Equipment involved Type ofprocess

Continuity:related to Case’

1 Ammonia startup 6 reactor, startup heaterand compressor

ammonia ’29, 15, 36, 42,50–52

2 Leak 4 pipe and storage tank ammonia ’29, 15, 36, 42,50–52

3 Cycling column 4 distillation column plusauxiliaries

general

268





4 Platformer fires 4 heat exchanger,platformer reactor

refinery

5 Acid pump 5 pump, storage tank general6 Air dryer 8 adsorption, regeneration general7 The reluctant

crystallizer10 vacuum, crystallizer general

8 Depropanizer: thetemperatures go crazy

8 distillation column plusauxiliaries

depropani-zer

’32, 38, 41,43,45, 48

9 Bleaching plant 7 conveying powder foradsorption in vacuumdeoderizer

foods

10 Dry and not to dry 8 crystallizer, screen,hydrocyclone, centrifugeand steam dryer

foods

11 Lazy twin 5 pumps general12 Drop boxes 3 distillation, adsorption,

regeneration dryers,exchangers, pumps

ethylene Case’24

13 The lousy controlsystem

4 distillation column,overhead condenser

general ’30

14 Condenser that wasjust too big

3 distillation column, vacu-um, overhead condenser

fatty acids,food

15 The flooded boot 9 vacuum, condenser andpumps

general,reformer,ammonia

see Appendix C

16 The case of the dirtyvacuum gas oil

5 distillation plusauxiliaries

refinery

17 Is it hot or is it not? 4 vacuum distillation foods, phar-maceutical

18 The streptomycindilemma

7 evaporator food, phar-maceutical

19 The belt filter 2 filter, screen, pump,sedimentation tank

wastewater ’37

20 The fussy flocculatorpump

3 pump, storage tanks,flocculation

general,wastewater

21 The flashy flare 3 refinery, flare system,compressor

refinery

22 pH pump 4 pump, mixing general,wastewater

23 The hot TDI 4 polymerizer, mixer,cooling system

polymer

24 Low production onthe ethylene plant

4 distillation, adsorption,regeneration dryers,exchangers

ethylene Related to’12

269






25 The case of the delin-quent exchangers

4 reformer reactor, furnace,exchangers, pump

refinery

26 The droopingtemperature

5 furnace, pump general

27 IPA column 5 distillation column plusauxiliary equipment

petro-chemical,refinery

28 The boiler feed heater 5 shell and tube exchangerto heat boiler feedwater

general

29 Reluctant reactor 5 reactor, compressor,separator, condensers,refrigeration unit

ammonia ’36, 15, 42,50–52

30 The case of thereluctant reflux

6 distillation column plusauxiliaries

general ’13

31 Ethylene productvaporizer

6 heat exchangers, boiling ethylene.

32 What does the alarmmean?

6 sequence of distillationcolumns plus auxiliaries

petro-chemical

’8, 32, 38, 41,43, 45, 48

33 Chlorine feedregulation

6 slurry, pump, controlsystem, storage tank

mineralsprocessing

34 The cement plantconveyor

6 solids conveyer, bagging,dust filters, cyclones

general,ceramic

35 The cycling multipleeffect evaporator

6 long tube evaporators,vacuum system

glycerine

36 The really hot case 7 reactor and heat exchan-ger, steam generation

reformer;ammonia

’29, 15, 42,50–52

37 Mill clarifier 7 thickener, sludge pumps,control

pulp andpaper

’19

38 More trouble on thedeprop


petro-chemical,refinery,

’8, 32, 41, 43,45, 48

39 The case of the lumpysunglass display

7 feed bin, moldingmachine, mold andmold design

polymer,injectionmolding ofthermo-plastics

40 The cool refrigerant 7 refrigeration system,compressor, turbine,control, knockout pot

general

41 Ever increasingcolumn pressure

7 distillation columnplus auxiliaries

general ’8, 32, 38, 43,45, 48

42 The weak AN 7 storage tank, pump,reactor, cooling coil

ammonia ’29, 15, 36,50–52

270






43 High pressure in thedebut!



’8, 32, 38, 41,45,48

44 Reactant storage 8 storage tank, heatingcoil, steam traps

general

45 The deprop bottomsand the ISO dilemma



’8, 32, 38, 41 43,48

46 Not so cool chiller 8 pumps, exchangers,refrigeration cycle

polymeriza-tion

47 The fluctuatingproduction of aceticanhydride

8 feed vaporizer, vacuumpump, absorber,condensers, reactor

petro-chemical,acetic acid

48 The column that justwouldn’t work



’8, 32, 38, 41,43, 45

49 The case of the faultystretcher pedal

8 feed bin, moldingmachine, mold and molddesign.

injectionmolding ofthermo-plastics

50 The cleanup column 9 vacuum distillationcolumn plus auxiliaries

general,ammonia

’29, 15, 36, 42,50–52

51 More trouble on thecleanup column

9 vacuum distillationcolumn plus auxiliaries

general,ammonia

52 Swinging loops 9 reactor, compressor,separator, condensers,refrigeration unit

ammonia

Case’8: Depropanizer: the temperatures go crazy [8, distillation column plus auxili-aries, depropanizer]The problem statement is given in Chapter 2, Section 2.4.

. MSDS, 1495

. Immediate action for safety and hazard elimination, Put on safe-park, 1881Safety interlock shut down, 1531SIS plus evacuation, 732

. More about the process, 1010

. IS and IS NOT, Where? upstream, 537Where? downstream in the debutanizer, 979

. Why? Why? Why?, Best goal? To keep bottom level steady and stop tray tem-perature cycling, 1335

271



. Weather, Today and past, 23

. Maintenance: turnaround, When and what done?, 421

Maintenance: routine, When and what done?, 2

. What should be happening,

. Design and simulation files (allowances made for fouling, overdesign anduncertainties)Condenser, E-25, 591Distillation column, C8, 515Thermosyphon reboiler, E-27, 380Feed pump, F-25; F-26, 16Turbine drive, 1193Motor drive, 1072Reflux pump, F-27, 286Feed preheater, E-24, 1989Feed drum, V-29, 1478Overhead drum, V-30, 1124

Vendor files: Condenser, reboiler and preheater, 10Steam traps, 481

Commissioning data, P&ID, internal reports, 1245Handbook, Cox charts, 803Trouble-shooting files, 1430

. Calculations and estimations (that can be done in the office before specialtests are done and based on rules of thumb and information given in thecase)Energy balance: sink= source, Estimate the steam flow to the reboiler basedon the reflux rate and the fact that each kg steam boils 5 kg typical organic,821

. What is current operation

Visit control room: control-room data, Feed to the C8, depropanizer. FC/1, 563Reflux flowrate, FIC/4, 155Pressure drop Dp I/1, 78Level bottoms LIC/2, 1436Temperature bottoms TI/4, 1765Temperature mid-column TIC/5, 1502Temperature top, TI/3, 1947Pressure on overhead drum, PIC/10, 1610

Process operators, This shift, 334Previous shift, 33

Operating procedures,Column pressure, 659Feed to the column, 1834

272


. Check with colleagues about hypotheses, 2395

Call to others on-site,Call operators of process supplying the feed, V29, about possible upsets anddetails of flowrate and composition of feed, 2371Call operators of downstream units receiving the propane, 2757Call operators of downstream unit receiving butane, 769

Visit site, read present values, observe and sense.Column pressure, PI-4, 303Look at the flare; same size as usual?, 799Pressure relief to flare PSV-1, 1177Temperature mid-column TI- 8, 1025Valve position for column feed, FV- 1, 1951Valve position for steam to preheater E-24, 208Valve-stem position on PV-10, 614Level in feed drum, V-29; LI- 1, 885Pressure on exit of pump F-26, PI-3, and compare with head-capacity curve atfeed flowrate, 691Listen to the check valve on turbine-driven pump, downstream of PI-2, 1273Observe whether shaft is rotating for the feed pump F-26. Is the pumpnoisy?, 1906

Check diagram and P&ID versus what’s really out on the plant, 2769On-site simple tests:

Shut isolation valves around turbine-driven pump F-25, 1585Put TIC/-5, FIC/-1 and -4 and LIC/-2 controllers on manual and try to steadyout the column, 1737Shut exit valve on discharge of pump F-26 and read pressure, 1316

. Gather data for key calculations,

Pressure profile,Drum V-29 to pumps, F-25, 26, 1792Dp across pump, converted to head, 1666Pump F-25–26 exit to feed location, 63Drum V-30, pump F-27 and reflux into column, 917Pump F-27, 1868Vapor from top of column to vapor space in V-30, 1171Thermosyphon reboiler process fluid side, 133

Mass balance, over column, 442Perform more complicated tests, Gamma scan over the stripping section to locate

collapsed tray, 255

. Take “corrective” action,Put column on “safe-park”, 1390

273


Case’9: The bleaching plant [7, conveying powders, adsorption, vacuum deodorizer,foods]The problem statement is given in Chapter 5, Section 5.3.2.

. MSDS, 1480

. Immediate action for safety and hazard elimination,Put on safe-park, 1196Safety interlock shut down, 1345SIS plus evacuation, 2400

. IS and IS NOT, What, 1416When, 1896Who, 2838Where, 2245

. Why? Why? Why?, Best goal? to suck Fullers earth into deodorizer, 1008


. What should be happening

Design and simulation files (allowances made for fouling, overdesign and uncer-tainties)

For the diameter and length of conveying line, the design flowrate of Fuller’searth and the design conveying velocity, what vacuum is needed in the blea-cher?, 2690Hopper design, 2562Piping for pneumatic conveying, 347Vacuum system, 59

Handbook, Physical properties of Fuller’s earth, 938Trouble-shooting files, 535

. Calculations and estimations

Pressure profile,Is the pressure difference between atmospheric and vacuum (Dp= 80 kPa)sufficient to convey Fuller’s earth?, 1368


Visit control room: control-room data, none of the data are shown in the controlroom; all are on the unit,

Process operators, When you say “nothing happened” and “you couldn’t see pow-der dumping into the liquid”, what did you hear and see? 1097

Operating procedures, 1925

274



Call to others on-site,Utilities: any upsets in the steam? Check that the steam pressure at the boilerhouse is approximately the steam pressure at the ejectors, 2431

Visit site, read present values, observe and sense.Pressure gauge, 193Check that there is Fuller’s earth in the hopper, 2920Is the level of liquid in the bleacher so high that it covers the inlet line for theFuller’s earth into the bleacher, 2630

Check diagram and P&ID versus what’s out on the plant, 1206On-site simple tests:

Check for leaks into the bleacher from around the agitator shaft, 2988Rap the pipe and the side of the hopper to try to dislodge any bridging thatmight be occurring in the humid weather, 2767Use “turn and seat” to check the valve on the conveying line and leave open,2334

Sensors: check response to change, Pressure gauge on the bleacher, 2150Sensors: calibrate, Replace/calibrate the pressure gauge, 240Contact vendor supplier, Did other customers receive Fullers earth similar to

batch number 4853 that we received, and if so, have they had any comments orqueries?, 366

Are there any particular precautions we should be taking?, 831Samples and measurements, Sample Fuller’s earth, 457More ambitious tests

Remove the Fuller’s earth from the hopper, and crack open the valve. Listenfor air flowing into the bleacher and observe vacuum gauge, 2456Insert porous tubes into the bed of Fuller’s earth and position them such thatcompressed air is blown in to try to fluidize the powder near the inlet of theconveying line, 1856

Open and inspect, Open conveying line and check for plugs, 1442

. Take “corrective” action, Replace the valve, 2437Relocate the inlet to the conveying line. Instead of using a pipe stuck into thebed, attach the inlet to the bottom of the conical hopper, 2750

Case’10: To dry or not to dry [8, crystallizer, screen, hydrocyclone, centrifuge, steamdryer, foods]The problem statement is given in Chapter 5, Section 5.4.2.

. MSDS, 52



275


. IS and IS NOT, What, 2209Where, 2279When, 2642

. Why? Why? Why?, Best goal? To get the exit crystal moisture content < 1.5%,2779






Crystallizer, 20Hydrocyclone, 427Pump, 45Surge vessel installed between the screen and the centrifuge?, 4Surge vessel installed between the centrifuge and the dryer?, 405

Vendor files:Centrifuge, 816Hydrocyclone, 510Pump, 992Rotary dryer, 1490Feeder screw, 1224Parabolic screen, 711

Commissioning data, P&ID, internal reports, 878Trouble-shooting files, 533


Visit control room: control-room data, all data on the plant.Process operators, This shift; elaborate on the changes you have made, 1411Operating procedures,

What is the new cycle time on the centrifuge and how does this comparewith before?, 1009Cycle time for screening and washing for the screen, before and now 1041


Call to others on-site,Utilities: steam plant: any upsets, pressure delivered to our site, 2219Stores: any available spare equipment: centrifuges, screens, 1804Utilities: wash water for screen and centrifuge: change in quality, upsets,2786

Visit site, read present values, observe and sense.Records of analysis of moisture content in feed to dryer. Samples taken at thebeginning and end of a shift, 71

276



Reduce throughput for the unit and thus reduce feedrate to dryer, 360Steam condensate trap on dryer working?, 1971Estimate feedrate to the screen over the screening cycle and compare withfeedrate before the change, 1514Observe feed condition from the screen to the centrifuge during the washcycle of the screen, 1114Estimate feedrate from the screen to the centrifuge over the filter cycle andcompare with the feedrate before the change, 1458Estimate the feedrate to the centrifuge during the screen wash cycle and com-pare with the feedrate before the change, 561Estimate the feedrate from the centrifuge to the dryer during the filter cycleand compare with the feedrate before the change, 909Estimate the feedrate to the dryer during the centrifuge wash cycle and com-pare with the feedrate before the change, 545Stop the centrifuge when the screen is in its wash cycle. Test the degree ofcompaction of the residual crystals under the peeler, 54

Gather data for key calculationsPressure profile,

Recycle through hydrocyclone, 1671Energy balance: sink= source, Measure amount of condensate by submerging exit

line in bucket with preweighed amount of cold water. Does energy lost from conden-sation of steam= energy gained by evaporation?, 603

Equipment performance,Rating dryer, 1860Screen, 1566Centrifuge, 1977

Call to vendors, licensee, Rotary dryer, 482Samples and measurements,

Sample the crystal discharge from the screen every 30 s for three cycles ofscreen operation; measure % screen overs as a % of feed. Compare withresults for old operation, 1968Sample the discharge from dryer every 20 s for three cycles of screen opera-tion; measure moisture content. Compare with data for previous operation,1520Sample the liquid-fine underflow from the screen every 30 s for the screeningcycle. Compare with data from previous operation, 38Draw a composite sample the wash water exit from the screen. Analyze forparticle-size distribution. Compare with previous operation, 318Draw a composite sample of the wash water from the centrifuge. Analyze forparticle-size distribution. Compare with previous operation, 1911Sample the feed to the dryer every 20 s for three cycles of screen operation;measure moisture content. Compare with previous operation, 196

277


Return to previous operating conditions,Sample the discharge from dryer every minute for three cycles of screenoperation; measure moisture content, 1705Sample the feed to the dryer every minute for three cycles of screen opera-tion; measure moisture content, 1032Sample the crystal discharge from the screen every 30 s for three cycles ofscreen operation; measure % screen overs as a% of feed, 558

Open and inspect,Centrifuge, 913Dryer, 689

Take “corrective” action,Revamp the condensate removal system: install a thermodynamic trap,slanted exhaust pipe to the trap and minimized the distance between the trapand heating tube, 795Operate the centrifuge on former operation, 1194

Case’11: The lazy twin [5, pumps, general]The problem statement is given in Chapter 6, Section 6.2.2.

. MSDS, 376




. Why? Why? Why?, Best goal? get the flow up to design rate when pump A isrunning, 1667






Centrifugal pump A, 1595Centrifugal pump B, 1510Heat exchanger, 166Check valves, 395

Vendor files: Centrifugal pumps A and B, identical and from the same vendor,930

Heat exchanger, 540

278



. Calculations and estimations. None can be done based on the given informa-tion


Visit control room: control-room dataWhat is the temperature after the heat exchanger when pump A is on-line?,386What are the levels in the downstream equipment when pump A is on-line?,470Is FRC/-100 local or remote?, 506“What is the pressure in the storage tank T-200?”, 973

Process operators, Current shift, 1470Operating procedures, About use of pumps, 1006


Call to others on-site, Contact downstream cat-cracking unit. Everything workingas expected? is the behavior consistent with the flowrate signalled from FRC/-100?,648

. Visit site, read present values, observe and sense.Do either pumps A or B sound as if they are cavitating?, 1998Look and see that valves V200, V201, V202, V205 and V206 are all open, asexpected, 518Read the controller output to valve V100 when pump A is running, 2325Read the controller output to valve V100 when pump B is running, 2441Observe the valve-stem position on F100 when pump A is pumping, 2086Observe the valve-stem position on F100 when pump B is pumping, 2845Observe whether the arrow on each valve V200, V201, V202, V205, V206 is inthe direction of flow through the valve, 2935Observe whether the direction of flow through both check valves agrees withvalve installation, 1587Check the tab on the orifice plate FRC-100 that it is the correct diameter andfacing the correct direction, 1160Pump A running hot?, 1617

. On-site simple tests:Shut exit valve on pump B; read pressure on gauge PI-220 when pump isrunning. Convert to head of fluid, 1030Shut exit valve on pump A; read pressure on gauge PI-210 when pump run-ning. Convert to head of fluid, 1465Put pump A on-line and check that the motor started. Repeat for pump B,602Test valves V200, V201, V202, V205 and V206 with “turn and seal” to ensurethey are working properly, 1461

279


What is the pressure at P210 and at P220 when pump A is on and pump B isoff? Express this is head (so I can compare with the head-capacity curve), 550What is the pressure at P210 and at P220 when pump A is off and pump B ison? Express this is head (so I can compare with the head-capacity curve), 170Stop pump B and close valve V205. Start pump A. What is the flow on FRC/-100?, 102Stop pump A and close valve V201. Start pump B. What is the flow on FRC/-100?, 1482

Check diagram and P&ID versus what’s really out on the plant, 2372Gather data for key calculationsPressure profile,

From Tank T-200 through pump A to cat cracker, 1432From Tank T-200 through pump B to cat cracker, 7

Energy balance: sink= source,Read clamp-on ammeter, voltmeter and power-factor meters, assume densi-ty= design density, compare power for motor to drive pumps A and B, 140

Sensors: check response to changeChange set point on FRC/-100; does the valve V100 respond?, 1081

Sensors: calibrate, Flowmeter, FRC/-100, 1552Open and inspect,

Pump B and ask maintenance to inspect and look for reasons why the pumpis not functioning well, 1886Pump A and ask maintenance to inspect and look for reasons why the pumpis not functioning well, 213Stop the process. Open and inspect valves V201 and V202, 453

. Take “corrective” action,Shut down the process. Replace valves V201 and V202, 276Shut valve V205 whenever pump A is running, 2027Stop the process. Replace check valves on the exit lines from both pumps,678

Case’12: Drop boxes [3, distillation, adsorption, regeneration dryers, evaporators,ethylene]The problem statement is given in Chapter 7, Section 7.2.2.

. MSDS, 2667


. More about the process, About the dryers and dryer cycle, 1715About the distillation, 264About the sewer systems, 420About the low-temperature condensation, 499

280


. IS and IS NOT, What, 827When, 2001Where, 1571

. Why? why? why?, Goal: “clear up the hazard as stated by the safety inspector”,637






Exchanger E131: cool regeneration gas to the dryer during last portion ofregeneration: water tube side; regeneration fuel-gas shell side, 1773Heater E130: heat regenerative fuel gas to dryer, 256Exchanger E107and E108: precool feed from 38 �C to –29 �C, 444Demethanizer overhead condenser: process fluid shell side; ethylene refrig-eration on the tube side, 462Demethanizer reboiler: process stream shell side; propylene on tube side,987Steam trap on regenerative gas heater, 2997Knockout pot, 2990

Vendor files:Heat exchanger, 2518Dryer system, 2022Steam trap, 2533

Commissioning data, P&ID, internal reports, P&ID, 437Sewer plot plan indicating which streams go to each “sewer gate” and whichgo to drop box B and which to drop box C, 1101

Trouble-shooting files, 2223

. Calculations and estimations. no calculations can be done based on the lim-ited information in the problem statement


Process operators, Any changes in operation?, 2330


Call to others on-site,Safety inspector Where were the samples taken and what was the composi-tion?, 1699Operators of upstream styrene, ethylene and propylene plants: any upsets,any calls from the safety inspector, 2882

. Visit site, read present values, observe and sense,Look at the flare, 2674

281



Vent valves closed on process (or shell side) of exchangers E107, E108, con-denser E114, E131?, 2477Vent valves closed on utility side (or tube side) of exchangers E107, E108, con-denser E114, E131?, 2364Vent valves closed on shell and tube side of heater E130?, 2876Drain valve on knockout pot shut?, 2617For exchanger E131 cooler; block off the in and out water lines; open vent ontube side and note fluid leaking out, 1211For the vent valves on the process (or shell side) of exchangers E107, E108,condenser E114, E131, test by “turn and seal”, 2433For the vent valves on the utility side (or tube side) of exchangers E107, E108,condenser E114, E131, test by “turn and seal”, 2363Test the vent valves on shell and tube side of heater E130 by “turn and seal”,2812Test the drain valve on knockout pot by “turn and seal”, 2557Retest drop box B and C for explosive mixture at half-hour intervals for 2hours, 554Retest drop box A for explosive mixture at half-hour intervals for 2 hours, 119


Exchanger process fluid pressure relative to utility pressure: direction of leak,2067Exchanger utility pressure relative to atmospheric; direction of leak, 2472Exchanger process fluid relative to atmospheric; direction of leak, 2814

Mass balance, Over demethanizer, 2698Samples and measurements, Sample cooling water leaving cooler E131, 1488

Gas sample from drop boxes B and C using an evacuated bomb. Lab analysisfor the overall concentration of light hydrocarbons and a breakdown of thehydrocarbon portion, 2434

Open and inspect,Cooler E131 and look for leaks in tubes, 2576

. Take “corrective” action,Replace the vent valves on the shell side of E107, 108, 130, 131, 2923Replace drain valve on KO pot, 168

Case’13: The Lousy Control System [4, distillation, overhead condenser, general]The problem statement is given in Chapter 7, Section 7.2.5, Activity 7-9.

. MSDS, 2266


282



. IS and IS NOT,What, 2317When, 1770Where, 621

. Why? why? why?, Goal: “to prevent too much stuff from going to the flare”,232

. Maintenance at turnaround, When and what, 1106




Fan: control system on the air flow to the condenser, 324Condenser, air-cooled, 739Distillation column, 1996Reflux drum, 2070Reflux pump, 2458

Vendor files: Condenser, air cooled, 2662Reflux pump, 2910



Equipment, Rate condenser: heat load, 411


Visit control room: control-room dataControl-room values now and from past records, Temperature at the top ofthe column, 2502Temperature on liquid-gas exit from the condenser, 2685Pressure at the top of the tower, 2722Pressure on the overhead receiver, 778Level in the overhead receiver, 1521



Call to others on-site, Operators of the upstream process, 2185Operators of the downstream process, 1613

Visit site, read present values, observe and sense.Hot exhaust air recirculation to the intake of the air-cooled condenser, 220Inspect the hydraulic configuration on the exit header from the condenser,89Look at the flare, 456

283


Sounds around the condenser, 325Check diagram and P&ID versus what’s really out on the plant, 1462On-site simple tests:

Temperatures and humidities of inlet and exit air for the air-cooled conden-ser, 1866Lower the elevation of the exit from the header to decrease the amount offlooding, 2879


On process pipe from the top of the column to the overhead receiver, 2346Across condenser, 2062

Equipment performance,Rating of air-cooled condenser, 2844Fans, 1631Reflux pump, 1876Column, 1670

Sensors: check response to changeTemperature sensor exit of condenser, 1087

Sensors: use of temporary instruments,Surface temperature near thermowell at exit of condenser measured by a con-tact pyrometer, 192

Sensors: calibrate,Temperature sensor on exit of condenser, 684

Control system,Put on manual; change the set point and note response of fan system, 487

Sampling and measurementsConcentration: vent from the overhead receiver; concentration of heavies,2587Concentration: vent from the overhead receiver; concentration heavies, tensamples taken at one-hour intervals, 2152

Open and inspect,Air-cooled condenser: visual inspection plus air and water pressure tests,1254Fan and fan blades, 2293

. Take “corrective” action,Operate on manual, 2793Install hose and let spray of water fall over tubes, 367Stop operation; open the headers and water-wash with high pressure hoses toremove scale, 1230

284


Case’14: The Condenser that was just too big [3, distillation, vacuum, overhead con-denser, fatty acids, food]The problem statement is given in Chapter 7, Section 7.2.5, Activity 7-9.

. MSDS, 51






Condenser, 1660Backup condenser, 2446Booster ejector, 3000Ejector, 2524Wet vacuum pump, 1781Barometric condenser, 21Coil cooler in tank, 966Reciprocating pump, 1271

Vendor files: Reciprocating pump. F1400, 1528Ejectors and booster ejector, 1961Wet vacuum pump F1401, 2163

Commissioning data, P&ID, internal reports, Any files?, 2004Handbook, Vapor pressure of fatty acids, 2490Trouble-shooting files, 2969


Pressure profile, Estimate the pressure in C1400; vapor side of E1400; vapor sideof E1401, 2511

Mass balance, Mass balance on the overhead pumped by F1400 compared withexpectations based on feed composition and rate, 2781

Energy balance: Check coolant temperature for E1400, 2580


Visit control room: control-room data,P1, 2273T4, 2377T3, 390

Process operators, Flow pumped from F1400, 486Trends in pressures and temperatures, 188Liquid level in bottom of E1400 as feed to V1400, 871Anything strange happening?, 1317Was the vacuum pump started up according to operating procedures?, 1020

285


Visit site, read present values, observe and sense.Cooling-water flow into E1403, 1322Cooling-water flowrate and temperature to E1404, 1091Read P3 and look for oscillations in pressure, 1851

Is the water level in the hot well above the exit from the downcomer from thebarometric condenser? Water temperature?, 1594

Note whether steam valves to booster and regular ejector are full or partiallyopen, 2201Does the booster ejector sound as though it is “kicking out”?, 2094

Is the level in the boiling condenser E1400 dropping? Test by trying to add moreliquid via the top up funnel at the top of E1400, 2655

Listen for cavitation in pump F1400, 326On-site simple tests: Visually and audibly check for any leaks of air into the sys-

tem, 716Sensors: check response to change

T4: 933T3: 522

Sensors: use of temporary instruments,Put surface temperature sensor on suction line to F1400 and check value, 1155Call to vendors, licensee, Reciprocating pump: what could cause the knocking its

head off?, 1142Sensors: calibrate, T4, 1920

T3, 1614Samples and measurements,Feed: usual composition? and does this contain more volatiles than expected?,

1719Open and inspect, Suction line from bottom of E1400 through to F1400, 2859

. Take “corrective” action, Flow cold water over the outside casing of pump F1400, 2939

Case’19: The case of the reluctant belt filter (supplied by Mike Dudzic, B. Eng. 82,McMaster University) [2, filter, screen pump, deep thickeners, wastewater treatment]The following diagram shows a section of the sludge concentration and dewateringsection of our wastewater treatment process. Water containing sludge is concen-trated in a deep settling tank and then pumped to a continuous-belt filter. Polymeris added just upstream of the filter to improve dewatering. The process is shown inFigure 8-2.

For the past three weeks the plant has been operating smoothly although latelythe operator thinks that the flow to the filter has been a bit lower than normal. Theoperator is relatively new on this plant. He has been on the unit for three months.Today, the operator notes that the sludge has stopped flowing to the filter. The pumpis running. If the fault is not corrected quickly, the sludge will build up in the set-tlers and shortly the whole process will be shut down.

286


Air at 550 kPaFlush water

East

thickener

10 cmdiam line

15 m

Heavy dewatered

sludge

water

Filtered water

Polymer

addition

7.5 cmline

2.5 cm

10 cm

Continuous belt filter

Closed

Westthickener

(idle)

Figure 8-2 The dewatering system for Case’19.

Case’19: Reluctant belt filter




. Why? why? why?, Goal: “Get the sludge flowing to the filter”, 596


. Maintenance: turnaround, When and what done? 722

Maintenance: routine, When and what done? 2660



Thickener, 707Centrifugal pump for handling sludge, 765Strainer, 997Continuous-belt filter, 2143

Vendor files, Centrifugal pump, 1518Trouble-shooting files, 223

287


. Calculations and estimations that can be done in the office based on theinformation given in the problem statement

Pressure profile,Calculate to see if 550 kPa g pressure is sufficient for air to actually backflowthrough the thickener, 1635


Process operators, When did the flow appear to decrease? 1330How long had the east thickener been idle before you started using it? 1757Have you ever seen anything like this before? and what did you do?, 2329

Operating instructions, Standard procedure: if there is low sludge flow; flush theexit lines from the thickeners with high-pressure water and high-pressure air, 2052

Check with colleagues about hypotheses, 1805Call to others on-site, High pressure air; have there been any interruptions in ser-

vice? what is the pressure of the air being delivered to site? 1258Visit site, read present values, observe and sense.

Is the shaft of the pump rotating? 322Is the block valve V101, to the idle thickener closed? 731Are the isolation block valves around the strainer and around the pumpopen? 294


Flush the lines with high-pressure water and air for five minutes, 793Test by “opening and closing” the block valves V100, V101, 2672Test by “opening and closing” the block valves on the flush out lines V102,V103, 2216

Sensors: use of temporary instruments,Use a clamp-on ammeter to measure the amps to the pump motor and com-pare with design or usual value, 1855Use a clamp-on ammeter, portable voltmeter and power-factor meters andcalculate the power drawn by the pump motor and compare with design orusual value, 1648

Call to vendors, suppliers or licensee, Pump supplier. what might be going onhere?, 1199

Open and inspect,The line between the thickener exit tee and the strainer; is it clear?, 1289Strainer, 1576Isolate, flush and when safe, open and inspect pump, 1923

. Take “corrective” action,Shut-off the polymer to the filter, 798Shut down. Block off and drain. Clean out the crud in the line from the bot-tom of the thickener to the screen, 249Replace the block valve V101 on the bottom from the idle thickener, 2300Install a recording ammeter on the pump motor, 2800

288


Case’20: The case of the fussy flocculator pump (problem supplied by Jonathan Yip,B. Eng. McMaster University, 1997) [3, pump, storage tanks, flocculation, general]Pump 40P002 is a centrifugal pump that transfers wastewater from the buffer tankto the flocculation tank. The liquid overflows into the Dissolved Air Flotation, DAF,unit. Figure 8-3 shows the system. One day the operator noticed that the level in theflocculation tank was lower than normal and the resulting overflow to the DAF wasless than expected. Pump 40 P002 just wasn’t performing as it should! Get it fixed.The manual ball valve on the exit line is wide open.

Figure 8-3 The flocculation system for Case’20.

Case’20: The case of the fussy flocculator pump


. IS and IS NOT, What, 195Where, 1892When, 2744

. Why? why? why?, Goal: “Get the overflow to DAF to expected valve”, 2596





Design and simulation files (allowances made for fouling, overdesign and uncer-tainties), Centrifugal pump, 1297

Vendor files: Centrifugal pump, 2525Trouble-shooting files, 2798


Process operators, Anything change in this portion of the plant?, 604Operating procedures, For pumping, 898


289


Call to others on-site, Operators of DAF: is the flow less than expected?, 779Visit site, read present values, observe and sense.

PI gauge, 2322Listen to the pump for sounds of cavitation, 2343Is drive motor on the pump running hot? Touch with gloved hand, 2420Is the pump running hot? Touch with gloved hand, 2411Does the handle on the ball valve, on the exit line, move easily and smoothly?,2243Does the handle on the ball valve, on the entrance into the flocculation tank,move easily and smoothly?, 2460Current level in the buffer tank; usual value?, 2029

Check diagram and P&ID versus what’s out on the plant, 1342On-site simple tests: Close the discharge valve on the pumpexit and readPI, 2764

Backflush the pigtail on the pressure gauge with city water to clear any block-age, 1099Tachometer reading on drive shaft rpm for the pump, 2927Stethoscope on the check valve, 2877Measure motor amps with clamp-on ammeter, calculate power assumingvolts, power factor and density, 2506

Gather data for key calculationsEquipment performance, Pump, 641Sensors: check response to change

Pressure gauge PI response to partial closure of valve on exit line, 112Sensors: calibrate, Pressure gauge PI, 413Samples and measurements, Sample solution in buffer tank. Compare the den-

sity and composition with specifications, 676Sample solution in the flocculation tank. Compare the density and compositionwith specifications and with specifications for solution in buffer tank, 932

Open and inspect,Shut down system, isolate. Open and inspect ball valve, 953Shut down system, isolate. Open and inspect centrifugal pump: clearance be-tween impeller and volute tongue; status of the wear rings; erosion of theimpeller, key between the shaft and the impeller. Pluggage?, 1473Shut down system, isolate. Open and inspect line to the pressure gauge fordirt and pluggage causing sluggish response, 1630Shut down system, isolate. Open and inspect exit line from the valve to theflocculation tank, 2190Shut down system, isolate. Open and inspect check valve, 2894

. Take “corrective” action,Change the pressure gauge, 2569Replace the ball valve on pump exit, 2852Replace the check valve, 2965Replace the impeller in the pump, 1727Realign drive motor and pump shaft, 1644

290


Case’21: The case of the flashy flare (problem supplied by Mark Argentino, B. Eng.McMaster University, 1981) [3, refinery, flare system, compressor, refinery]The flare system at your refinery works as follows. When there is a high-pressurebuild-up in a vessel, tower, or exchanger a relief valve opens and the high pressurevapors, and maybe some liquid, flow into a pipe network that eventually ties into acommon header that flows into a large knockout drum where the liquid is removedand the vapors are drawn overhead. Other sources of vapor in the flare system couldbe off-specification products sent to flare, hazardous vapors educted from pumps, orany hydrocarbon or non-hydrocarbon sources in the refinery that are not of anyproduct value but cannot be vented to atmosphere. The typical composition of theflare gases is as follows

CH4 20% C4 s 3%C2H6 15% C5

+ 1%C3H8 5% N2 50%C3H7 5% H2S 1%

The vapors then pass through a seal pot that is a vertical cylindrical drum with asmaller cylinder in the center. The vapors flow down through the central cylinder,out the bottom and into the annulus where they bubble through a liquid and are fedto the burner, flare, where they are burned. To visualize the operation of the seal,consider an analogous situation of a large drinking straw in a glass of water. Whenno air is blown down the straw, the water in the glass is at the same level as thewater in the straw. As one blows down the straw the water level drops in the strawand rises in the water glass. This liquid differential in the system forms a certainliquid head that must be overcome for one to blow air out of the straw and bubble itthrough the water. As with the drinking-glass analogy, the seal pot has a certain levelof liquid in it so that pressure must be applied on the inlet line (the straw in ouranalogy) for any vapors to pass through the seal and be burned. In this way it servesas a seal. The process is given in Figure 8-4.

Last turnaround a flare-gas compressor was added to the system. The compressortakes suction just downstreamof the knockout drum. The function of the compressor isto recompress the flare gases from 106.5 kPa abs to 480 kPa abs so that the vapor can berecovered as fuel, which along with purchased natural gas, is used in the boilers andfired heaters. The compressor has a spill back or kick back valve and line, which is asmall pipe that reroutes some of the compressed vapors in the discharge line, via a con-trol valve on pressure control, back to the compressor suction, and hence to the flare lineto keep the suction pressure at 106.5 kPa abs. The compressor will shut down on highsuction pressure= 112 kPa abs. In the cold Canadian winter kerosene is used as thesealant liquid. Nevertheless, there are freeze ups in lines and equipment failures; hencefrequent flaringwhen there is a high release of vapors to the flare linewhen the pressurebuilds up in the flare line. The pressure builds up in the flare line because the com-pressor will only pump a fixed maximum amount of vapors. If there is a higher flowof flare gases then the restriction of flow at the seal pot causes a pressure increaseup to the blowpoint. The maximum blow pressure is 107.8 kPa. That is the level ofliquid sealing the dip tube in the seal pot is equivalent to a 6.8 kPa differential.

291


6.8 kPa

6.8 kPa

To flareTo flare

When pressure

exceeds 6.8 kPa

Flare

Flare gas compressor

Kickback

loop

PIC

1

PIC

2

PIC

3

KO pot

Liquid

Pressure relief

from equipment &

other sources to

the flare system

Flare gas

5.5 kPa-

gague

Recompressed

vapors for

reprocessing

Seal pot

Figure 8-4 The flare system for Case’21.

Today it is cold, dry and windy. The pressure gauge on the flare line reads 112 kPaabs; the compressor shuts down. The operator explains that “when the compressorshuts down, the accompanying surge in the flare gas flow is so great that the kero-sene is all blown into the flare burner along with the usual flare vapors. The result isthat the flare flashes and smokes like a giant fire.” All the neighbors are phoning inwith smoke complaints. With the flare seal blown, all the flare gas goes to the flare.When the compressor is down we are losing $1000/h in non-recovered vapors. Fix it.

Case’21: Flashy flare

. MSDS, 215,

. Immediate action for safety and hazard elimination,Put on safe-park, 84Safety interlock shut down, 2341,SIS plus evacuation, 2900

. IS and IS NOT: (based on given problem statement),What, 2076When, 2641Who, 1992Where, 1730

. Why? why? why?, Goal: “Prevent the compressor from shutting down”, 597



292





Compressor, 75Control, 471Seal pot, 9Knockout drum, 500

Vendor files: Compressor, 852Commissioning data, P&ID, internal reports, 576Handbook, Density of kerosene and color, 957Trouble-shooting files, 625

. Calculations and estimations (that can be done in the office before specialtests are done)

Pressure differential,Height of kerosene in the seal pot= 6.8 kPa?, 1127

Equipment performance, Compressor, 1497


Visit control room: control-room data: values now and from past records,Motors amps on the compressor, 1511

Process operators, 1974


Call to others on-site,Call operators on other units to see if excessive pressure in the flare linemight be originating on units, 1842

Visit site, read present values, observe and sense.Pressure gauge on flare line, 2382Pressure gauge by seal pot= pressure gauge on kick-back control= 112 kPa,2610

Check diagram and P&ID versus what’s out on the plant, 70On-site simple tests: Atmospheric pressure, 2883

Drain off the KO pot to verify that the liquid level is not > expected, 694Sensors: check response to change

Pressure gauge on flare line, 141Sensors: use of temporary instruments,

Surface sensor for temperature of suction gas to compressor, 410 Use clamp-on ammeter to measure amps when running under usual conditions, 2487

Sensors: calibrate, Pressure gauge on flare line, 1034Control system, Control-valve stem moves when suction pressure < 106. 5 kPa

abs?, 1284Call to vendors, licensee, Compressor, 742

293


Samples and measurements, Flare gas: composition, 1240Liquid density of liquid from the knock-out drum, 1070Kerosene in seal pot, density, 1590Kerosene in seal pot, color, 1447

Open and inspect,Open seal pot and check for blockage, 2119Open and inspect KO pot for frozen ice in the demister, 729

. Take “corrective” action,

Isolate the seal pot. Attach compressed air to the feed line to the seal pot andblow out line through the flare, 2465

Case’22: The pH control unit (used courtesy of Scott Lynn, University of California,Berkeley, CA) [4, pumps, control, storage tank, acid-base wastewater treatment]Concentrated hydrochloric acid is used to neutralize caustic wastes being fed to anewly built effluent treatment plant. The process is illustrated in Figure 8-5. Thevolumetric flowrate of wastes is approximately constant at 12.6 L/s but due to thenature of their source, the concentration varies from 1 to 10 g/L equivalent NaOH.The average is about 5 g/L. Control is usually good, but at times it becomes erraticand occasionally the acid flow stops altogether. Turning off and restarting the acidfeed control system usually serves to get the acid flow going again, but this malfunc-tion threatens to shut down the entire plant. Find the bug and get rid of it!

pH IC

1

Mixing tee

Caustic waste

12.6 L/s

6 m vertical

5 cm diam

pipe

To effluent treatment

Storage tank for

concentrated HCl

7560 L

1.25 cm

diam PVC

2.4 m

Figure 8-5 Feed system to the effluent treatment process for Case’22.

294


Case’22: pH pump

. MSDS, 2250

. Immediate action for safety and hazard eliminationPut on safe-park, 2167Safety interlock shut down, 2035SIS plus evacuation, 1740

. IS and IS NOT,What, 2940When, 2700Who, 2081Where, 2358

. Why? why? why?, Goal: “To prevent erratic control of pH”, 2579






Feed pump, 468Vendor files:

Feed pump, 592Commissioning data, P&ID, internal reports, 998Handbook, Density of 30% HCl, 530Trouble-shooting files, 1326

. Calculations and estimations based on given informationControl system, Feed-forward control, 2909Valve stiction and hysteresis, 426Fluid dynamics, Calculate the residence time between the acid injection andpH sensor, 1425Calculate if the flow is turbulent in the caustic waste line near the mixing tee,1191Velocity of waste in the line, 1403

. What is happening

Visit control room:Process operators, This shift, 2509

Did the motor overload and trip off, 2677


Call to others on-site,Effluent treatment: variability in flowrate or pH, 1290

Visit site, read present values, observe and sense.

295


Control valve: does the control-valve stem move in response to a signal fromthe controller, 2002Check that the block valves on the control valve are open, 1068Check that the valve on the bypass around the control valve is shut; if notshut it, 1878Is there a vent on the acid storage tank, 1835Note the characteristics of the response: cycling? amplitude and frequency?,1682Sounds near the pump, 1150


Check valve for stiction: remove backlash by doing a bump of 2 to 3%; thenmove controller output slowly via a slow ramp or bumps of 0.1%; observecontroller output: pressure to the actuator, the valve stem and the pH output.Repeat ramping up and ramping down, 2467Is period of oscillation in pH close to the time delay, 2799Manually start with a high acid flowrate demand gradually decrease the flowdemand, 700


For the feed acid from the storage tank to the injection location, 1003Energy balance:

Estimate power required for density of acid, 1499Equipment performance,

Rating pump, 2850Sensors: check response to change

pH, 184Sensors: use temporary instruments,

Use clamp-on ammeter to measure amps to pump, compare with expected,896

Sensors: calibrate, pH sensor, 850Control system, Put on manual, 498Samples and measurements, Sample acid and check that density is 1.48, 1075Open and inspect,

Pump discharge line from pump to injection point, 1700Pump suction line, 2100

. Take “corrective” action,Replace pH sensor, 2500Relocate acid injection line so that the acid enters the top of the caustic wasteline, instead of the bottom, 2600Install a check valve in the discharge of the acid pump, 378To improve mixing, install a static mixer just after the mixing tee in the caus-tic waste line, 125

296


Relocate the pH sensor to 3.2 m from the injection point to give a 2-s resi-dence time, 1218Run cold water over the pump, 1300

Case’23: The hot TDI (based on Barton and Rogers, 1997) [4, polymerizer, mixer,cooling system, polymer]In the manufacture of polyurethane prepolymer, toluene diisocyanate (TDI), atroom temperature, 25 �C, was charged to an 8-tonne reactor overnight. It was rain-ing cats and dogs this morning. At the start of the 8 am shift, 1.7 Mg of polyol wasadded gradually over a 20-minute period as spelled out in the operating procedures.The mixer operated continuously. The temperature was monitored carefully. It roseto 127 �C as expected. After 35 minutes the operator in the control room noted thatthe reactor temperature read 170 �C or 43 �C hotter than expected. The red light indi-cated that the stirrer stopped. The operator couldn’t get the stirrer restarted. Shortlythereafter the temperature sensor on the reactor read 200 �C.

Case’23: The hot TDI

. MSDS, 2252



. Why? why? why?, Goal: “get the reactor temperature to 127 �C and the mixergoing”, 1596






Custom designed reactor with coolant coils and stirrer, 2468Commissioning data, P&ID, internal reports, P&ID, 991Handbook, TDI, 531Trouble-shooting files, 1325

. What is happening

Visit control room: control-room sensors and historical data,TI in reactor and TRC in reactor, 1195

297


Pressure in reactor, 2289Pressure relief on top, 2893Indication light that shows stirrer is turning; green means yes; red meansstopped, 291

Process operators, Was the TDI charged correctly?, 1201Is the feed cooling water cold?, 1422Is the cooling-water flowing to the cooling coils?, 1618Is there a power failure that might cause the mixer to fail?, 1986Please describe the procedure you used to add the polyol, 2188


Call to others on-site,Purchasing: did you change suppliers of the TDI or of polyol?, 1743Utilities: any upsets or changes in the cooling water supplied to our site,2274

Visit site, read present values, observe and sense.Anything obvious interfering with the mixer shaft preventing it from turn-ing?, 2791TRC valve position for the cooling-water valve controlling water to the jacket,2824Emergency block valve on vent (that bypasses the PCV in case the PCV failsto open under SIS), 723Signal to the TRC control valve for the cooling water, 1261

Check diagram and P&ID versus what’s out on the plant,No diagram supplied; only verbal description, 1837

On-site simple tests: Glove test on temperature of reactor, 2996Sensors: check response to change

Temperature on reactor, 306Temperature on cooling water, 49

Control system,Put on manual to control temperature, 651

Sensors: calibrate,Temperature sensors on reactor, 977Temperature sensors on cooling water, 2736

Samples and measurements,Sample the polyol. Is it within specs?, 2522Sample the TDI. Is it within specs?, 709

Open and inspect,Isolate, drain, vent, make safe to enter. See if the cooling coils are fouled, 543Isolate, drain, vent, make safe to enter. See if there is anything preventingthe mixer from turning, 1769

. Take “corrective” action,Replace motor on mixer with motor with double the kW, 1503

298


Case’24: Low production on the ethylene plant (courtesy of John Gates, B. Eng. 1968,McMaster University) [4, distillation, adsorption, regeneration dryers, exchangers,ethylene]The part of the ethylene plant that relates to this problem concerns the drying sec-tion to remove moisture from the feed gas and the distillation train to separate thegas stream into the desired component section. Drying section: Three aluminadryers are installed. One dryer is regenerated while two are hooked in series onstream. For example, dryer V106 is being regenerated with V107 and 108 removingthe moisture in the process stream to less than 4 ppm. Then V107 will be regener-ated with V108 and 106 in series and so on. The cycle lasts 12 hours. The dried gasgoes to a knockout pot (to remove any entrained material) and then is chilled, inexchangers E107 and 108, before entering the separation towers.

During regeneration of the dryers, “fuel gas”, heated with 2.8 MPa steam, flowsthrough the dryer in the direction reverse to normal flow. Once it is through thedryer the regeneration effluent gas is cooled and returned to the fuel-gas system.During regeneration the dryer temperature rises to 190 �C and is maintained at thistemperature for one hour. Then the fuel gas bypasses the heater and is sent directlyto the dryer to cool it. For this plant, the “fuel gas”or “town gas” is purchased froman off-site, independent utility supply pipeline. This gas is primarily methane withsome hydrogen, is supplied at a pressure of 1.1 MPa and with a specified moisturecontent of < 6 ppm. This company has recently been expanding its facilities andpipelines.

The dryers are all appropriately manifolded and valved so that any dryer can beregenerated, bypassed or used. The regenerating dryer is separated from the linedryers by a gate valve.

The separation is performed in a train of three distillation columns operating atabout 3.2 MPa. These are a demethanizer, de-ethanizer and C2 splitter, T101, T102and T103 respectively.

The process is illustrated in Figure 7-1 (Case’12). The overhead from TowerT101 is condensed with ethylene as refrigerant. The overhead from Tower T102 iscondensed with propane as refrigerant. The overhead from Tower T103 is con-densed with propylene as the refrigerant.

The current situation. At low flowrates of 70 Mg/d of ethylene, we encounter nodifficulties with production. Recently, we have had excessive pressure drops acrossthe third column when all conditions were the same except that the production ratehad increased to 150 Mg/d. At these higher rates of production, the pressure dropwas so large that we could not operate satisfactorily. The lost capacity is worth $20000/day. We cannot shut the plant down because the rest of the site uses ethylene asa raw material and our current inventories are very low. The usual operating condi-tions are given in Figure 8-6. Get our inventories up so that the rest of the site canfunction properly.

299

8 Prescription for Improvement: Put it all Together300

Figu

re8-6

P&ID

fortheethylene

plan

tforCase’24

.


Case’24: Low ethylene production

. MSDS, 320


. More about the process,About the dryers and dryer cycle, 2888About the low-temperature condensation, 2514


. Why? why? why?, Goal: “remove the high Dp across the trays in the C2 split-ter, T103 at high throughput”, 488






Exchanger E131: cool regeneration gas to the dryer during last portion ofregeneration: water-tube side; regeneration fuel-gas shell side, 1773Heater E130: heat regenerative fuel gas to dryer, 257Exchanger E107and E108: precool feed from 38 �C to –29 �C, 451Reboiler E115 on bottoms of De-ethanizer, 2130Demethanizer overhead condenser: process fluid shell side; ethylene refrig-eration on the tube side, 2885Demethanizer reboiler: process stream shell side; propylene on tube side,2999Steam trap on regenerative gas heater, 2632Distillation columns T101, 102 and 103, 2501Knockout pot, 1777

Vendor files:Heat exchanger, 1569Dryer system, 280Steam trap, 824Utilities: town gas supplier, 501

Commissioning data, P&ID, internal reports, 982Handbook or Google, Gas hydrates, 1493Trouble-shooting files, 2992

301


. Calculations and estimations based on information given in the problemstatement

Pressure profile,Exchanger Process fluid pressure relative to utility pressure for E 131, 107,108, 113, 114 to show direction of leak, 2726Exchanger Process fluid pressure relative to utility pressure for E 130 to showdirection of leak, 159Exchanger utility pressure relative to atmospheric to show direction of leak,466Exchanger process fluid relative to atmospheric, 2429Column T101 Dp estimate across trays at 150 Mg/d, 2936Column T102 Dp estimate across trays at 150 Mg/d, 2116Column T103 Dp estimate across trays at 150 Mg/d, 2112


Visit control room: read instruments and past records,For 150 Mg/d; Temperatures: top and bottom for demethanizer T101, 1214For 150 Mg/d; Temperatures: top and bottom for de-ethanizer T102, 435For 150 Mg/d; Temperatures: top and bottom for C2 splitter T103, 1377For 150 Mg/d, measured Dp across demethanizer column T101, 2432For 150 Mg/d, measured Dp across de-ethanizer column T102, 2012For 150 Mg/d, measured Dp across C2 splitter T103, 2840

Process operators, Any changes in operation other than increase in flowrate?,2043


Call to others on-site,Operators of upstream feed gas facilities: upsets? any moisture in feed? anychanges in composition at the higher feedrates?, 559Operators of fuel-gas system; any changes in the fuel gas we are sending youfrom the regeneration process for our adsorbers?, 675Operators of the refrigeration units for the ethylene and propylene refrigera-tion loops: any changes or upsets?, 1129Utilities: any changes in the 2.8 MPa steam supplied to site, 1286

. Visit site, read present values, observe and sense,Look at the flare, 1912


Drain knockout pot after adsorbers, 2171Sensors: check response to change

Dp and pressure gauges on demethanizer T101, 2540Dp and pressure gauges on C2 splitter T103, 2232

302


Sensors: calibrate,Calibrate the pressure gauges at the top and bottom of the demethanizer col-umn T101, 2834Calibrate the pressure gauges at the top and bottom of the de-ethanizer col-umn T102, 2507Calibrate the pressure gauges at the top and bottom of the C2 splitter T103,371

Call to vendors, licensee, supplier,Call utility supplying town gas about changes, 2032Call utility supplying town gas about specifications, 2050

Samples and measurements,Sample town gas entering battery limits. Analyze for water content. Samplesevery 30 min for 2 hours, 892Sample town gas leaving heater E130 before entering the dryer for regenera-tion. Analyze for water content. Samples every 30 min for 2 hours, 1000Sample town gas entering battery limits. Analyze for water content, 527Sample town gas leaving heater E130 before entering dryer for regeneration.Analyze for water content, 1874Sample process gas leaving KO pot as feed to tower T101. Analyze for watercontent, 2387Sample process gas entering dryers. Analyze for water content, 2880Sample process gas leaving the last dryer in the series. Analyze for water con-tent, 2545

Perform more complicated tests,Gamma scan near the top of the demethanizer and of the C2 splitter to locatecollapsed tray, 2790

Open and inspect,Shut down tower, vent to safety, open access hole near top and look for plugsin the downcomers or on the trays. Demethanizer column T101, 2370Shut down tower, vent to safety, open access hole near top and look for plugsin the downcomers or on the trays. C2 splitter T103, 2972Shut down column T102, isolate the reboiler E115 on the de-ethanizer, onceconditions are safe, pull bundle, hydraulically pressure test to identify leaksin the tubes or between the tubesheet and the tubes, 1190Heater E130; isolate, once conditions are safe, pull bundle, hydraulically pres-sure test to identify leaks in the tubes or between the tubesheet and tubes,348

. Take “corrective” action,Replace the tube bundle on reboiler E115 on the De-ethanizer. T102, 818

303


RE

AC

TO

R

FU

RN

AC

E

TO

OT

HE

R U

NIT

S

FR

C

10

0

PI

10

0

NA

PH

TH

A

FE

ED

EX

CE

SS

HY

DR

OG

EN

TO

TO

WE

RS

CO

OL

ING

WA

TE

R F

RO

M

CO

OL

ING

TO

WE

R

FR

C

30

0

RE

CY

CL

E H

YD

RO

GE

N

LO

W P

RE

SS

UR

E

SE

PA

RA

TO

RLC

E204

V2

00

TI

10

1

TI

10

2

TI

20

1T

I

20

2

TI

20

3T

IT

I

E202

E203

E201

V300

304

Figu

re8-7

Thena

phthaprocessforCase’25

.


Case’25: The case of the delinquent exchangers [4, reformer, furnace, exchangers,pumps; refinery]In the reforming process of naphtha, naphthenes are dehydrogenated (cyclohexaneto benzene and hydrogen) and paraffins are isomerized (n-hexane to 2.methylpen-tane) and dehydrocyclized (n-hexane to benzene and hydrogen). In the reactor, a rel-atively high hydrogen:hydrocarbon feed ratio is maintained to minimize coking.The exit gas from the reformer is a very useful source of heat at a temperature ofabout 415 �C and is used to heat several streams for other units. Figure 8-7 showsthe system. To keep up-to-date, we changed the catalyst so that the feed hydrogen:hydrocarbon flowrate could be reduced to about half its original value. Not only doesthis allow a reduction in the amount of hydrogen recycled, but this allows anincrease in the naphtha flowrate to keep the space velocity in the current reactor thesame as it was before we changed to the new catalyst.

We have just started up the unit after the new catalyst has been installed. Immedi-ately the operators of the other units phone to say that their process streams are nolonger getting the amount of heating (via E202 and E 203) they used to get from thereformer before the shutdown. The exchangers are delinquent! Fix the problem.

Case’25: The case of the delinquent exchangers on the naphtha reformer

. MSDS, 473



. Why? why? why? Goal: “to provide the usual amount of heat to the processstreams of other units” 738






Exchangers: E201, 202, 203, 1036Naphtha pump, 1474Furnace, 1235Reformer, 1526

305


Vendor files:Heat exchangers E201, 202, 203, 1636Pump, 1975

Commissioning data, P&ID, internal reports, 2941Handbook, Approximate thermal properties of hydrogen and hydrocarbon vapors

in the stream. Approx. Prandtl numbers, 2551Trouble-shooting files, 2998


Equipment performance, Exchangers E201, 202, 203, 2451


Visit control room: control-room data,Into reformer, TI 203, 2111Ex reformer, TI 101, 2404Ex exchanger E203, TI 102, 356Naphtha flow, FRC 100, 65

Process operators,On other units: Has anything else changed for you because of the shutdown,922On the reformer unit: anything changed with the new catalyst? 503

Operating procedures, Please walk me through the startup procedure you used,874


Visit site, read present values, observe and sense.TI 201, 686TI 202, 2418PI 100, 2122Ton exit stream to other units from E202, 2816Ton exit stream to other units from E203, 2503Visual check around exchanger E202 and E203, 2383Look at the flare, 2645


Vent exchanger E202 to check if inert gas trapped in shell side, use gas snifferdesigned to test for expected gas, 2059Vent exchanger E203 to check if inert gas trapped in shell side, use gas snifferdesigned to test for expected gas, 2212Vent exchanger E201 to check if inert gas trapped in shell side, use gas snifferdesigned to test for expected gas, 1751


Reformer exit to low-pressure separator, 2951Exchanger: tube side, 2016

306


Naphtha from pump to reformer, 2316Mass balance,

Space velocity and mass flowrate through the reformer and the downstreamexchangers, 2294

Energy balance, Overall energy balance on the system, 1086Equipment performance,

Check heat exchanged on exchanger E202 based on new conditions, 1807Check heat exchanged on exchanger E203 based on new conditions, 1550

Sensors: check response to changeTI 202, 1901TI 203, 11

Sensors: use of temporary instruments,Surface or laser thermometer on the line between E202 and E203, 313

Sensors: calibrate,TI 202, 600TI 201, 668Ton exit stream to other units from E202, 1276Ton exit stream to other units from E203, 1489

Control system,Controls related to the reformer; put on manual and check that working cor-rectly, 851FRC 100: put on manual and check that it is working well, 701

Call to vendors, licensee,Phone call to catalyst vendor: any unexpected behavior difference betweennew catalyst and old catalyst given the conditions we are using, 392

Samples and measurements,Reformer exit gas and check for hydrogen content, 1956

Open and inspect,For exchanger E202: pull the bundle and measure baffle spacing; check seal-ing strips, check that baffles are not loose. Look for fouling; check that ventworks. Look for condensed liquid, 1555For exchanger E201: pull the bundle and measure baffle spacing; check seal-ing strips, check that baffles are not loose. Look for fouling; check that ventworks, 2866For exchanger E203: pull the bundle and measure baffle spacing; check seal-ing strips, check that baffles are not loose. Look for fouling; check that ventworks, 2586

. Take “corrective” action,Open exchanger E202 and decrease the baffle spacing to increased the vaporflow across the bundle by 20%, 2711

307


Feed tank

FC 1

P3

V300

TC 3T4

Fuel

oil

F 2T

7 Pro

duct

tank

C.W

.

F 7

Air

Inta

ke

FC 5

PR

OC

ES

S F

LU

ID

T 10

308

Figu

re8-8

Thefired

heater

processforCase’26

.


Case’26: The drooping temperatures (used with permission from T. E. Marlin)[5, furnace, pump; general]The process shown in Figure 8-8 consists of a fired heater, which raises the tempera-ture of a hydrocarbon stream via convective and radiative heat transfer, and apacked-bed reactor. The process has been working well for over a year. Recently, themarket for the product is growing, and the plant would like to maximize the produc-tion rate. Therefore, the operators have been slowly increasing the feedrate.

You happened to be in the control room one morning to collect some data whenan operator asks for your assistance. She shows you the trend of selected variablesseen in Figure 8-9. She is quite concerned; you better solve this problem fast!

T3

F2

FC1

time

1 minute

Figure 8-9 Trend plot of some process variables for Case’26.

Case’26: The case of the drooping temperature

. MSDS, 283



. Why? Why? Why? Goal: to operate plant safely and efficiently, 1444


309






Pump, 1999Blower, 1782Heat exchanger, furnace/ fired heater, 1532

Vendor files:Blower, 1005Pump, 1381

Commissioning data, P&ID, internal reports,Packed-bed reactor: data about the capacity: can it handle the increased fee-drate? 1401



Visit control room: control-room data,Flow: Process liquid in. FC-1, 955Flow: process liquid ex reactor F-7, 2851Flow: Fuel gas F-2, 2542Flow: Air to furnace FC-5, 1225Temperature: exit process liquid TC-3, 1092


Operating procedures, Please guide me through the procedures you use, 1894


Visit site, read present values, observe and sense.Temperature: process liquid out. T4, 1623Pressure: on furnace P3, 2191Stack gas: Look at quality of flue gas out of stack, 2101Pump: process liquid: sound like cavitation? 2046Valve: control: fuel oil, 2881Valve: control air to furnace, 381Temperature of the process fluid entering fired heater, T-10, 2428Block valves around the fan fully open? 2140Readings of the draft gauges upstream and downstream of the damper, 2878Position of the damper; compare with expected 1/3 closed, 1279Inspect orifice tabs the three flow meters to ensure that the plates wereinstalled with sharp edge upstream, 1139

Check diagram and P&ID versus what’s really out on the plant, 171

310


On-site simple tests,Use laser temperature sensor or pyrometer to measure temperature of tubesin the radiant section, 443Use laser temperature sensor or pyrometer to measure temperature of thetubes in the convection section using the observation port in that section, 820Shut the valve on the process pump discharge and compare the measuredpressure with the head-capacity curve for that pump, 690Change set point on air flow, FC-5, and note response, 553

. Gather data for key calculations

Pressure profile,On process pipe through the furnace, reactor and product tank, 1902On fuel-oil line bringing fuel into the furnace, 1601On the combustion air line from the intake to the burner, 1801Direction of leak of process fluid versus furnace, 661Direction of leak of air: into the furnace or out of furnace? check pressure P3,897

Mass balance, On process liquid, 2301Energy balance: On combustion: temperatures, 2003

On combustion: excess air, 101Equipment performance,

Heat exchanger, 307Air blower, 150Process liquid pump, 801

Sensors: calibrate,Temperature: process liquid out. T4, 287Pressure: on furnace P3, 469

Control system,Temperature: Controller: Signal output from controller TC/3, 2574Put the process fluid temperature controller, TC-3, on manual control to con-trol under the high flowrate conditions, 2740Air-flow controller: signal output from controller FC/5, 2077

Samples and measurements,Oxygen in flue gas, 929

Open and inspect,Furnace, 1431Blower, 1189

. Take “corrective” action,Put TC3 on manual. Reduce output from valve V300, 1938Install sensor in the flue gas to measure percentage oxygen, 2379

311


Case’27: The IPA column [5, distillation column plus auxiliary equipment;petrochemical or refinery]Recently, the water-cooled condenser on the IPA column was replaced by an air-cooled condenser. Variable-pitch fans underneath the horizontal bank of finnedtubes direct air upwards across the tubes. Since the installation, the ratio of IPA outto IPA fed to the column is 0.70. Previously we could account of 0.995. Where is theIPA going? This costs us $4,000/h.

Case’27: The IPA column: where is it going?

. MSDS, 2529



. IS and IS NOT:What, 398When, 1758Who, 1002Where, 1708

. Why? why? why? Goal: “to prevent the apparent loss of IPA.”, 2162






Fan: control system on the air flow to the condenser, 284Condenser, air-cooled, 126Distillation column, 2000Lines insulated? 485

Vendor files: Condenser, air cooled, 2181Commissioning data, P&ID, internal reports, Commissioning report of tests

done before startup, 2993Handbook, Pertinent properties of IPA, 1166Trouble-shooting files, 1017

. Calculations and estimations: more data are needed before calculations canbe made.


Visit control room: control-room data: values now and from past records,Temperature at the top of the column, 1501Temperature on liquid exit from the condenser, 1685

312


Pressure at the top of the tower, 1722Pressure on the overhead receiver, 777Level in the overhead receiver, 521Reflux flowrate, 1385Feedrate change? 2394Bottoms feedrate change? 2846

Process operatorsThis shift, 1179Previous shift, 1356


Call to others on-siteOf upstream suppliers of feed to IPA unit, 86Of downstream vent scrubber unit, 1291Of downstream processors of IPA, 377

Visit site, read present values, observe and sense.Hot exhaust air recirculation to the intake of the air-cooled condenser, 18Sounds around the condenser, 338Look at the flare, 239

Check diagram and P&ID versus what’s really out on the plant, Same informationas given in more about the process, 2134

On-site simple tests:Temperatures and humidities of inlet and exit air for the air-cooled conden-ser, 106Tape accessible flanges on the line to the top of the condenser and after thecondenser. Test for leaks with soap solution. Test valve stems with soap solu-tion, 1879Record the vertical dimensions around the exit of the condenser, 1135


On process pipe from the top of the column to the overhead receiver, 926Across condenser, 1062

Mass balance, On process liquid, IPA, 1508Energy balance: Over the condenser: heat extracted in air= heat of condensation

for all IPA? 1026Equipment performance,

Rating of air-cooled condenser, 844Fans, 632Column, 670

Sensors: check response to changeTemperature at top of column, 50Temperature of the IPA leaving the condenser, 200

Sensors: use of temporary instruments,Temperature and humidity of inlet and exhaust air for the condenser, 1275

313


Air velocity to the face of the condenser for the maximum and minimumblade pitch, 1423

Sensors: calibrate,Temperature at top of the column, 351Temperature of IPA leaving condenser, 1808

Control system,For fan, put on manual and increase blade pitch to maximum, 1069Ask control specialist about the quality of this type of control, 311Use clamp-on ammeters, voltmeter and power-factor meters to obtain data toestimate power to the fan, 475

Call to vendors, licensee, Air-cooled condenser. Describe symptoms to the vendor,1517

Samples and measurementsSample bottoms of column; measure concentration of IPA and compare withpast data, 1658Sample feed to column; measure concentration of IPA and compare withpast data, 1172Concentration: vent from the overhead receiver; concentration IPA, singlesample, 222Concentration: vent from the overhead receiver; concentration IPA, ten sam-ples taken at one-minute intervals, 152Concentration: liquid from bottoms; IPA concentration. Three samples, at 2-min intervals, 461Concentration: feed to the column; IPA concentration. Three samples, at 2-min intervals, 956

Open and inspect,Air-cooled condenser: visual inspection plus air and water pressure tests,1652Fan and fan blades, 251

. Take “corrective” action,Run cold water over the tubes for an 8-hour shift, 1916Tip the tube bank so that it is no longer level “so that the condensate can runout easily”, 495Repipe exit line so that the syphon is removed. “The condensate flowsdirectly from the level of the bottom tubes to the overhead receiver”, 1943

Case’28: The boiler feed heater (adapted from case of P. L. Silveston, University ofWaterloo) [5, shell and tube exchanger, steam heated, condensate traps; general]Waste flash steam from the ethyl acetate plant is saturated at slightly above atmo-spheric pressure. It is sent to the shell side of a shell and tube exchanger to preheatboiler feed water to 70 �C for a nearby boiler house. Condensate is withdrawnthrough a thermodynamic steam trap at the bottom of the shell. The water flowsthrough 1.9-cm nominal tubes. There are 100 tubes. See Figure 8-10. “When the

314


system was put into operation 3 hours ago everything worked fine,” says the super-visor. “Now, however, the exit boiler feed water is 42 �C instead of the design value.What do we do? This difficulty is costing us extra fuel to vaporize the water in theboiler. “ Fix it.

STEAM BOILER

FEED

WATER

CONDENSATE

RETURN TO

HEADER

CLOSED

TEMP

DRAINVALVE

CLOSED

TOBOILER

PI

100

TI

100

Figure 8-10 The boiler feedwater heater for Case’28.

Case’28: The boiler water preheater

. MSDS, 349


. IS and IS NOT: (based on given problem statement),What, 2861When, 1320Where, 2391

. Why? why? why?, Goal: “to heat the boiler water to 70 �C”, 254




Exchanger, 2765Float steam trap, 995Piping: steam, 1990Piping: water, 1270

Handbook, Approximate thermal properties of steam and ethyl acetate. Approx.Prandtl numbers. Saturation temperature for assumed 200 kPa g steam, 1504


315


. Calculations and estimations (that can be done in the office before specialtests are done and based on information in the problem statement and rulesof thumb)Flowrate, Estimate flow of boiler feed water based on tube area and estimatedvelocity in the tubes, 1237

Energy,Difference in heat load, assuming inlet water temperature= 18 �C, 1788Heat transfer rates at new and design conditions based on assumed steampressure of 200 kPa g, 2199Estimate steam condensation, 2699


Visit control room: control-room data: values now and from past records. Dataonly available on site

Process operators, This shift, 1329


Visit site, read present values, observe and sense.Inlet steam pressure, 323Inlet steam temperature, 177


Open bypass on the steam trap, 477Open air vent on top of the exchanger; leave the vent open for 10 min. Readthe temperature gauge on the exit boiler feed, 364After the vent has been opened for 10 min, close the vent and read the boilerfeed exit temperature after 3 h of operation, 876At the end furthest from the steam inlet, remove a vertical section of theinsulation about 10–30 cm wide. Tap the side with metal moving up theexchanger and listen for a change in sound indicating a liquid-vapor inter-face, 617

Collect data for key calculationsPressure profile,

Exchanger: shell side, 1067Exchanger: tube side, 66Steam: from ethyl acetate plant to header to boiler feed heater, 465

Mass balance,On steam-condensate; measure the condensate by measuring the volume col-lected over a timed period, 2005

Energy balance: sink= source, Amount of steam= amount of heat picked up bywater, 108

Equipment performance,Rating exchanger, 144Type of steam trap, 1632Upstream Ystrainer on condensate line upstream of the trap, 1930

316


Sensors: check response to changeSteam temp TI 100, 3Exit water temperature, 210

Sensors: use of temporary instruments, Water temp in, 502Sensors: calibrate, Steam temp TI 100, 368

Exit water temp, 635Call to vendors, licensee, Heat exchanger, 1519

Steam traps, 1533Samples and measurements,

Measure amount of condensate to get a measure of the steam. Compare theenergy loss from the steam to the energy gained by boiler water, 388

Open and inspect,Exchanger: visually inspect. Water and air tests for leaks, 1577Exchanger: pull the bundle and measure baffle spacing; check sealing strips,check that baffles are not loose, 57

. Take “corrective” action,Replace the float trap with an inverted bucket trap with an air vent, 1924Operate with the bypass around the trap partially open, 480Stop operation; pull the bundle and clean tubeside and shell side, 1944

Case’29: The Reluctant reactor (courtesy of W. K. Taylor, B. Eng. McMaster, 1966)[5, reactor, compressor, separator; ammonia]Ammonia is produced on two interconnected reactor loops as given in Figure 8-11.Feed gas consists of hydrogen and nitrogen in the proper 3:1 ratio with about 1%methane as an inert. In this ammonia-synthesis reaction about 10% conversionoccurs per pass through the reactor. Feed gas is compressed to 34.5 MPa abs and fedto a common header that feeds two reactor loops. Liquid product is condensed andremoved from the system; gas is recycled back to the loop via the recycle stage com-pression. The reactor operates at 500 �C. There is an internal gas-gas heat-exchangerwithin the reactor. The DTdue to exothermic reaction is about 50 �C.

Each compressor is a multistage reciprocating constant speed machine rated atabout 3000 kW. Bypass valve B is operated to control the Dp across the recycle stagethat must not exceed 3.5 MPa. Opening valve B lowers the Dp and the flow of recyclegas to the loop. The recycle flow is about five times the flow of fresh feed.

Bypass valve A is operated to trim the loop: closing this valve forces more gas overto the reactor; opening the valve causes gas to bypass the loop. Valve A is used tocontrol the reaction temperature. If too much gas is fed to the reactor and the cata-lyst is inactive, the high flow might extinguish the reaction. Similarly if the flow tothe reactor is too low, the reaction will go further because of the longer reactiontime; the reactor will overheat because there is not enough flow to carry away theheat of reaction. Normally valve A is open slightly during plant operation.

Methane is an inert coming in with the feed. The methane concentration is keptabout 15% in the loop gas to the reactors by maintaining a small purge.

317


H2, A

R

1

LR

CL

ICF

R 1

Pu

rge

Re

frig

era

nt

Co

oli

ng

wate

r

Liq

uid

am

mo

nia

Man

ual In

dic

ato

r

co

ntr

oll

er,

A

Ele

ctr

ic s

tart

up

heate

r

Reacto

r

H2, A

R

1

LR

CL

ICF

R 1

Pu

rge

Re

frig

era

nt

Co

oli

ng

wate

r

Liq

uid

am

mo

nia

Man

ual In

dic

ato

r

co

ntr

oll

er,

A

Reacto

r

Feed

gas

Feed

gas

3 s

tages o

f

com

pre

sson

3 s

tages o

f

com

pre

ssio

n

Kic

kback

BB

No

rth

co

mp

res

so

r

Kic

k

back

So

uth

co

mp

resso

r

So

uth

lo

op

No

rth

lo

op

Recycle

com

pre

ssor

318

Figu

re8-11

TheAmmon

iasynthesisreactors.


The pressures, levels in the separators and the temperature profile in the reactorare shown in the control room.

The design provides operating flexibility. If one compressor breaks down the othermachine can feed both loops thus keeping the reactors at operating temperatures whilerepairs are done. This avoids costly startup expense. Isolation block valves on thecompressors are not shown on Figure 8-11. Furthermore, both loops are equalized inpressure thus evening out any slight variations introduced by the compressors.

The problem:The plants are in the final phase of startup after a turnaround shutdown. The com-pressors are sending feed gas and recycled gas to the reactors. Startup pressure is 7MPa. The electric cal rod heaters, used to heat up the reactor catalyst bed duringstartup, are on. Heat-up normally proceeds at 50 �C/h and at 8 am the reactors wereup to 300 �C. By noon the reactors should be at 500 �C and in steady production andthe plant in normal operation with all the gas vents closed. It is now 3 pm and theloops have only heated up 25 �C to 325 �C and have had no increase in temperatureover the past hour. “Get these plants producing. This is costing us $20 000/h. “

Case’29: The reluctant reactor

. MSDS, 352




. Why? why? why?, Goal: “to heat up the reactors to 500 �C”, 446





Reactor, 2452Internal heat exchanger, 2010Reciprocating compressor, 1832Refrigeration system, 1507Condensers: refrigerant, 1733Condensers: water, 1960Gas-liquid separators, 1600Valves A and B, 1011

319


Vendor files:Reciprocating compressors, 178Refrigeration system, 484

Commissioning data, P&ID, internal reports, 452Handbook, Thermal properties of hydrogen, nitrogen and ammonia, 385Trouble-shooting files, 751


Mass balance, Purge rate and methane buildup, 624Energy balance:

Energy transfer from Cal Rod heaters for startup (from experience), 532


Visit control room: control-room data: values now and from past records,Pressure at exit of compressors, in the loop, 1459Hydrogen concentration in the North loop, 1412Hydrogen concentration in the South loop, 1117Temperature in the catalyst bed, North reactor, 1048Temperature in the catalyst bed, South reactor, 1546Temperature leaving catalyst bed, North reactor, 1815Temperature leaving catalyst bed, South reactor, 1946Dp across the reactor, North loop, 1664Dp across the reactor, South loop, 2415Flow of liquid ammonia from North loop, 2897Flow of liquid ammonia from South loop, 2959

Process operators, Anything surprising other than the temperatures?, 2558Refrigeration condensers colder than usual?, 2226Cooling-water condensers colder than usual?, 1720

Operating procedures, Please tell me about the startup procedures you used, 1969

. Check with colleagues about hypotheses,

Visit site, read present values, observe and sense.Amps to cal rod heaters, 1564Check that sample valves are closed, 1789

Check diagram and the P&ID versus what’s actually out on the plant, 1216On-site simple tests: Increase the pressure in the loop to about 10 MPa, 1362Control system,

Over ride indicator control on board; go to site and manually open valve Afully. North loop. Check temperature after one hour, 1073Over ride indicator control on board; go to site and manually open valve Afully. South loop. Check temperature after one hour, 912

Sensors: calibrate, Temperature sensors, reactor, North loop, 577Temperature sensors reactor, South loop, 2836

320


Samples and measurements,Cooling water, North loop condenser. Analyze for ammonia, 2601Feed gas. Analyze for hydrogen concentration, 2327Continuity check on cal rod for reactor in North loop, 333Continuity check on cal rod for reactor in North loop, 73

Open and inspect,Open and inspect the reactor and especially the top with the cal rod heaters.North loop, 277Open and inspect the reactor and especially the top with the cal rod heaters.North loop, 1281

. Take “corrective” action,Open the kickback to reduce the pressure in the loop, 1085Open the purge line to the maximum, 660Shut down one compressor and provide loop gas to both North and Southloops from one compressor. Allow one hour of operation, 640

Case’30: The case of the reluctant reflux (courtesy Esso Chemicals) [6, distillationcolumn plus auxiliaries; general]The 20-tray column, operating at 520 kPa g, has just started up for the first time.The control configuration is illustrated below. The pump can deliver only 2/3 of thedesign value of the reflux, and although the reflux valve is fully open, the level in theaccumulator continues to increase. Do something quickly before the reflux drum isfull and sort out the problem. The financial penalty is high because of the clauses inthe construction contract for producing specification product within the commis-sioning period, and because of insurance issues and government regulations. Figure8-12 illustrates the system.

321

PC

100

TC

200

LC

202

To

Flare

PSV

1

PI

201

Figure 8-12 The overhead system for the column in Case’30.


Case’30: Reluctant reflux

. MSDS, 272


. IS and IS NOT: (based on given problem statement), What, 2930When, 2621Where, 2013





Reflux pump, 1058Condenser, 1450Reflux drum, 1364

Vendor files: Reflux pump, 1137Commissioning data, P&ID, internal reports, 1823Trouble-shooting files, 1828


Pressure profile,Recheck NPSH supplied, 993From reflux drum to column via reflux pump, 825

Equipment performance, reflux pump, 26Design of pressure-control system, 288Design of level-control system, 138


Visit control room: control-room data: values now and from past records,Overhead temperature, 460Overhead composition from routine lab analyses, 752Level in reflux drum, 1864

Process operators, This shift. What’s happening? 1606Visit site, read present values, observe and sense.

Control valve on the reflux pump discharge line: valve position and other fea-tures of valve; for example, put in backwards, 1513Note direction of rotation of the shaft and confirm that this corresponds withthe arrow on the casing, 1985Ask maintenance if the impeller might have been put in backwards, 2484Difference in pressure between PC 100 and PI 200 and compare with usualDp across condenser, 2071

322


Level indication on controller 200, 2320Listen for sounds of cavitation around the reflux pump, 2213


Shut the valve on the discharge side of the reflux pump and read the pres-sure; compare with head-capacity curve, 2499Listen with a stethoscope to check valve. Note if the direction through thevalve is the same as the flow direction, 2960Turn the valve stems on suction and discharge block valves on the pump,2664

Sensors: check response to changeLC202, 2582PI 201, 2186TC 200, 2409

Sensors: calibrate,LC 200 on reflux drum, 681PI 200; pressure on the reflux drum, 610

Control system,Put pressure-control system on manual, 1875Put level-control system on manual, 1675Put temperature control on manual, 1169

Samples and measurements,Feed to column and analyze for the amount of light components, 961Overhead vapor from column and analyze for heavies, 900

Open and inspect,Control valve, 595Reflux pump, 29Suction line from reflux drum to tower, 292Nozzle where line attaches to column, 407Line from pump to column, 1375

. Take “corrective” action,Install larger-size impeller in the pump, 1775Replace motor on air-cooled condenser, 2339Replace the level sensor on the reflux drum, 2733Reduce feed to the column, 2611Replace the check valve on the discharge line of the pump, 2550Install vent break line on condenser, 1266

Case’31: Ethylene product vaporizer (courtesy of C. J. King, Chemical EngineeringDept., University of California, Berkeley) [6, heat exchangers, boiling, steam; general]Our New Jersey petrochemical complex includes an ethylene plant that supplies 6.8Mg/h of ethylene through a pipeline to various consumers. It is important that wemaintain a steady flow of ethylene to our users, and, as a result, our plant contains alarge storage sphere of liquid ethylene. The ethylene must be a vapor, however,

323


when it enters the pipeline and must be at a temperature close to ground tempera-ture (slightly more than 1.6 �C) so as to avoid thermal stresses on the pipeline. Forthese reasons we have installed an ethylene vaporizer between the sphere and thepipeline. The ethylene is vaporized by condensing n-butane, which in turn is vapor-ized by steam. The cascade vaporization system is required so as to avoid unduethermal stresses across the heat-exchange surfaces.

Under normal operation, a small amount of ethylene, about 1.6 Mg/h is sentthrough the vaporizer, but the vaporizer frequently is called upon to provide more,or all, of the total ethylene supply. Before the vaporizer, the liquid ethylene ispumped up to 4.4 MPa gauge and metered through a flow-control valve; the ethyl-ene pressure in the vaporizer is roughly the pipeline pressure of 3. 9 MPa gauge.

The butane pressure controller set point, PIC, can respond to anywhere from 0.58to 0. 97 MPa. A set point of 0.8 MPa has been used successfully at all ethylene flowrates during the past year, although the outlet ethylene temperature has been slowto recover following a change in the ethylene flow rate.

In the past few months we have found it necessary to increase the PIC set point.Even so, we found yesterday when the ethylene unit came down that the vaporizercould not handle the full ethylene flow without tripping the low-temperature shut-off switch at the pipeline entry, which is set at 1.6 �C. This situation will cost us$6000/h plus inestimable customer good will if we stop the flow, or else it may wellnecessitate expensive and time-consuming pipeline repair if we continue. Figure8-13 illustrates the layout.

Figure 8-13 The ethylene vaporizer in Case’31.

324


Case’31: Ethylene product vaporizer

. MSDS, 34









Butane vaporizer, 609Ethylene vaporizer, 120

Vendor files: Butane-steam system, 414Commissioning data, P&ID, internal reports, 293Handbook, Temperature-vapor pressure data for ethylene and butane, 989Trouble-shooting files, 967


Pressure profile,If there is a leak in the ethylene-butane system, the direction of the leakwould be, 672If there is a leak in the steam-butane system, the direction of the leak wouldbe, 1365

Rate,Butane-steam system: Is the boiling nucleate or film boiling?, 1147Ethylene-butane system: Is the boiling nucleate or film boiling?, 1717


Visit control room: control-room data: values now and from past records, Pastrecords, 2427

Process operators, Please tell me what happened before the system shut down,2248

Tell me a bit about operation over the past few months, 2832Call to others on-site,

Steam utilities; any changes?, 2952

325


Waste water treatment, 2528Visit site, read present values, observe and sense.

Steam pressure, 2603Ethylene pressure in pipeline, 2807Ethylene temperature in pipeline, 2549PIC setting, 2109Level in the butane, 2407Valve position on condensate line and size of control valve relative to linesize, 2498Steam goes into the top channel of the butane evaporator, 2040Ethylene feed goes into the bottom channel of the ethylene evaporator, 1987Inspect insulation and look for cracks or holes in the external surface, 1776


Open vent on butane evaporator for several minutes; close and observe opera-tion, 1549Open vent on ethylene evaporator for several minutes; close and observeoperation, 1745Use stethoscope and listen to the sound of the fluid flowing through the con-densate valve, 1138Increase the level of butane in the butane evaporator as shown on the levelindicator, 1257Open the bypass on the condensate control valve, 656

Sensors: check response to changePIC, 699Level indicator on butane, 105PI on ethylene line, 353TI on ethylene line, 2390Temperature shut-off sensor switch, 2469

Sensors: use of temporary instruments,Use a contact pyrometer to measure temperature of the ethylene pipeline,2919

Control system,Reduce ethylene flowrate. Put PIC on manual and note response whenincrease the PIC setting, 2572Retune the PIC control system, 2020

Sensors: calibrate,PIC, 2154PI ethylene line, 1553TI ethylene line, 1963Level indicator on butane, 1238Temperature shut-off sensor switch, 1104

Samples and measurements,Butane from the shell side of the evaporator and analyze for ethylene andwater contamination, 881

326


Condensate and analyze for butane contamination, 958Ethylene and analyze for butane contamination, 55Butane from storage and analyze for contamination; compare with specs, 252

More ambitious tests,Drain all the butane and replace with fresh butane, 1914

Open and inspect,Ethylene evaporator for fouling, 1812Ethylene evaporator, pressure test and look for leaks, 1541Butane evaporator, pressure test and look for leaks, 2869Butane evaporator for fouling, 2055

. Take “corrective” action,Replace the pressure gauge PIC, 2048Install new control valve on condensate line; the size is same as line size,1153

Case’32: The alarming alarm (courtesy of T. E. Marlin, Chemical Engineering,McMaster University) [6, sequence of distillation columns with auxiliaries;depropanizer-debutanizer]The process is the depropanizer-debutanizer described in Case’8, Chapter 2. AP&ID is given in Figure 2.4 accompanying Case’8.

The operation of the upstream process is being modified to accommodate a newcatalyst and modified feed composition. The upstream units have been on-line andoperating smoothly for nearly a shift. Suddenly the high-pressure alarm on thedebutanizer, column C-9, rudely disrupts the quiet. And you thought everything wasgoing smoothly!

Case’32: The alarming alarm

. MSDS, 1495




. Why? Why? Why?, Best goal? To find out why the high-pressure alarm issounding, 2335





327



Condenser, E-28, 2006Distillation column, C9: 2367Thermosyphon reboiler, E-30, 1382Reflux pump, F-29, 1112Overhead drum, V-31, 1455

Vendor files: Condenser, reboiler, 1448Steam traps, 1244



Energy balance: sink= source, Estimate the steam flow to the reboiler based onthe reflux rate and the fact that each kg steam boils 5 kg typical organic, 2132


Visit control room: control-room data,Feed to C9, debutanizer, FI/2, 2227Feed to the C8, depropanizer. FC/1, 2304Overhead liquid product butane, flowrate FIC/7, 2011Reflux flowrate, FIC/6, 2461Pressure drop Dp I/2, 1724Level bottoms LIC/4, 1957Level in feed drum, V-31; LIC-5, 1506Temperature bottoms TI/12, 1870Temperature top, TI/11, 1152Analyzer A-1, 1357Pressure on overhead, PIC/19, 1052

Process operators, This shift, 1408Call to others on-site, Utilities: any change in the cooling-tower operation that

might affect our site, 1323Visit site, read present values, observe and sense.

Column pressure, PI-12, 1173Pressure relief to flare PSV-3, 717Temperature of the feed to the column TI-10, 619Cooling-water temperature in to the condenser, TI-13, 910Cooling-water temperature out of the condenser, TI-14, 802Valve position for column feed, FV-2, 258Valve position on PIC/19; downflow from the condenser to the reflux drum, 13Signal to PIC/19, 359Pressure on exit of reflux pump F-29, PI-20, and compare with head-capacitycurve at feed flowrate, 135Listen to the reflux pump F29 for sounds of cavitation, 455

328


Observe whether shaft is rotating for the reflux pump F-29. Check that thedirection of rotation is consistent with arrow on the housing, 767


Open bleed valves to fuel on the shell of the condenser for 10 minutes, thenshut, 606Tap the side of the condenser along a vertical line to listen for change insound associated with liquid level in the condenser, 858

Gather data for key calculations,Pressure profile,

Drum V-31 to pumps, F-29, 2863Dp across reflux pump, converted to head, 2984Drum V-31, pump F-29 and reflux into column, 2723Pump F-29, 2594Thermosyphon reboiler process fluid side, 2531

Sensors: check response to change, Temperature sensor at the top of column, TI-11, 963

Pressure sensor at top of column PI-12, 586Pressure sensor on overhead line for PIC/19; PT-19, 1307Pressure sensor on reflux pump exit line, PI-20, 1013

Sensors: use of temporary instruments,Measure the surface temperature on the outside of the condenser, 1252

Sensors: calibrate,Temperature sensor at the top of column, TI-11, 1802Pressure sensor at top of column PI-12, 1997Pressure sensor on overhead line for PIC/19; PT-19, 1785Pressure sensor on reflux pump exit line, PI-20, 2202

Control system,Put overhead pressure control, PIC/19; reflux control FIC-6 controllers onmanual and try to steady out the column, 2159

Sample and analyze,Feed to the upstream depropanizer for the amount of C4 and C5s, 2103Concentration of effluent from upstream reactor where catalyst was changed.Analyze for C4 and C5 and compare with previous, 2491

. Open and inspect,Condenser, 2901Pressure control valve PV-19, 2752Reflux pump, F-29, 2602Line from the top of the column to the condenser, E-28, 2742Sensor tap for pressure on the overhead line, PT-19, 2818

. Take “corrective” actionReplace the pressure control valve PV-19, 2260Replace the pressure gauge PT-19, 2139

329


Direct water from the fire hose onto the outside shell of the overhead conden-ser, 401Reduce the flowrate to C8 and thus reduce the flowrate to C9, 2054

Case’33: Chlorine feed regulation (courtesy of Scott Lynn, Chemical EngineeringDepartment, University of California, Berkeley) [6, slurry, pump, control system,storage tank; minerals processing]A copper mine is treating its crushed ore with a dilute solution (5%) of sodiumhypochlorite to improve the recovery of molybdenum disulfide by flotation. Sodiumhydroxide solution of the appropriate strength is reacted with chlorine gas in a 5.5-cm diameter pipe that serves as the reactor. Flow is continuous and relatively con-stant at 3.15 L/s. The pipe carries the bleach solution to the slurry tank. An oxida-tion-potential probe at the pipe outlet is used to regulate the flow of chlorine. Thesystem has just been installed, as shown in Figure 8-14 and serious trouble has beenencountered with the chlorine feed regulation. The flow of gas can be readily con-trolled manually but fluctuates wildly when put on automatic. A time recording ofthe oxidation potential, OPRC, is shown in Figure 8-15. Please correct this problemquickly!

1.5 m 7.6 m

PFC OPRC

FRC

Water

50% NaOH

Liquid chlorine

Slurry

treatment

tank

Figure 8-14 The chlorine feed system for Case’33.

330


100 mV

1 min

Manual operation

Automatic operation

Figure 8-15 Sample recording chart for the OPRC for Case’33.

Case’33: Chlorine feed regulation

. MSDS, 110








Orifice plate on sodium hydroxide line, 1027Orifice plate in water feed line, 697Control valve on chlorine feed line, 542Control valve on caustic feed line, 145Control valve on water feed line, 2794Caustic feed pump, 2813Pipe-reactor, 2943Source of water, 2546Chlorine injection line, 2079

Commissioning data, P&ID, internal reports, 1621Handbook, Density and viscosity of 50% sodium hydroxide, 1915Trouble-shooting files, 1787

. Calculations and estimations (that can be done in the office before specialtests are done)Implications of cycling, observed frequency of fluctuation, 1167

331


Observed amplitude of fluctuation, 1379Mass balance,

Flowrates of caustic and chlorine for reaction, 1096Fluid mechanics, mixing and residence time,

Estimate the Reynolds number in the “reactor”, 1019Estimate velocity in reactor, 781Estimate residence time before chlorine addition, 981Estimate residence time between chlorine injection and OPRC sensor, 2889

Rate,Of reaction to form sodium hypochlorite, 2922

Equipment performance,Control system: check the degrees of freedom, 2682Control system: and stability, 225


Visit control room: control-room data: values now and from past records,Flowrate of water, 99Flowrate of caustic, 2848Check that the output from OPRC fluctuates rapidly and significantly, 2609

Process operators, 345Call to others on-site, Operators of upstream units providing water: any variation

in the composition, temperature or flowrate?, 176Visit site, read present values, observe and sense.

Valve movement of chlorine when operating on automatic, 1349Valve movement of caustic feed control valve, 1446Valve movement of the water, 1249Check that the bypass valves on all control valves are shut and that the blockvalves are fully open, 2299Sounds of cavitation in the caustic pump, 2099


Check valve for stiction: remove backlash by doing a bump of 2 to 3%; thenmove controller output slowly via a slow ramp or bumps of 0.1%; observecontroller output: pressure to the actuator, the valve stem and the pH output.Repeat ramping up and ramping down, 1561

Gather data for key calculationsTemperature of the water such that the water may be flashing in the orificemeter sensor in the PFC loop, 1547Check that the oscillation is close to the time delay for the concentration toflow from the mixing point to the analyzer, 1931

Sensors: check response to changeOPRC, 616

Sensors: use of temporary instruments,Use a contact pyrometer to measure the temperature variation in feed water,945

332


Use a contact pyrometer to measure the temperature of the chlorine storagetank to ensure it is < 50 �C, 593Use clamp-on ammeter to measure amps to pump, compare with expected,62

Sensors: calibrate, OPRC, 439Control system, Retune control system, 267

Place controller on manual; observe the magnitude of the noise on the mea-surement, 2261

Call to vendors, licensee, or suppliers, Suppliers of chlorine, 2457Suppliers of caustic, 2449

Samples and measurements,Caustic and check against the specs from the supplier, 2198Chlorine and check against the specs from the supplier, 2656Water and check for high levels of silt or humic acid from spring runoff,1774

Open and inspect,Check that orifice plates are not installed backwards in the two locations,1909Open reactor pipe and check that it is clear (free of obstructions) and that thechlorine injection line is centered, 1948

. Take “corrective” action,Shutdown the process. Insert “static mixer” just downstream of chlorine-injection point, 1598Replace the water controller with FFC (flow fraction control, a ratio) insteadof PFC, 1209Shutdown the process. Change control loop on water line to be feed-forwardcontrol, 663Shut down the process. Add a support structure to the chlorine feed line toremove the vibration of the feed line, 899Redesign the caustic-water mixing area to provide better mixing. Shut downthe process and install the improved system, 749Insert “static mixer” just downstream of caustic injection point, 295

Case’34: The cement plant conveyor [6, solids conveyor, bagging, dust filters,blowers, fans, cyclone: ceramic]This plant produces dry mixes of mainly cements and coarser aggregates. The rawmaterials are added – one batch at a time – to a mixing hopper, transported by ahigh-pressure, batchwise, conveying system to a solids blender or mixing vessel.The mixed material is conveyed from the mixer to a packaging hopper feeding abagging machine. The finished, mixed product is packaged in 25-kg bags. The Qual-ity Control Department checks regularly on the finished product to ensure that thematerial has adequate strength by forming casts and determining the break strengthof the casts. Two or three days ago, the QC Department found that the strengths ofthe cast samples had decreased substantially. What’s going on?

333


Case’34: The cement plant conveyor







Specs for blend, 771Specs on mixing time, 1143

Handbook, Pertinent properties of cement blends, 1917Trouble-shooting files, 1626


Equipment performance,Dust collection, 212Fans, 441Conveying from mixer to packaging hopper, 25Mixer/ blender, 569Packaging, 1372Materials hoppers and star valve feeder, 1083Blower and batch-conveying system from the bins to the mixers, 1001


Visit control room: control-room data: values now and from past records. No sen-sors or data in control room.

Process operators,Mixing: What did you do differently three days ago?, 137Packaging: What did you do differently three days ago?, 728This shift on mixers, 538This shift on packaging, 504Previous shift on mixers, 941Previous shift on packaging, 1417

Check diagram and P&ID versus what’s out on the plant, 1260Control system,

For dust collector, 865For packaging, 585

Call to vendors, licensee, Packaging unit, 336Solids mixing unit, 432

334


Baghouse, 391Pressurized conveying system from the feed hopper to the solids mixer, 812

Samples and measurements,Monitor sampling of raw material, QC. Is the particle-size distribution withinspecs? QC perform tests, 1312Monitor sampling. after the mixer. Sample to QC: Measure strength of casts ;particle-size distribution; composition of components, 1078Monitor sampling. From the package Sample to QC: Measure strength ofcasts ; particle-size distribution; composition of components, 60Sample air leaving bag filter and measure concentration and size of particu-lates. Compare with previous, 2886

Open and inspect,Feed hopper, 508Solids mixer, 790Conveyor hopper, blower and conveying line, 115Dust collector on top of packaging hopper, 31Packaging unit, 382

. Take corrective action,Bang side of the hopper to the bagging machine to break any bridging, 2075Repipe the bagging machine so that the dusty air flows down through thebags instead of up through the bags. The goal is to prevent fluidization of thefines in some of the bags, 2496

Case’35: The cycling triple-effect evaporator [6, long tube evaporators, vacuumsystem, steam; glycerine]We are starting up a new plant, and it is like a zoo out there. We have three, multi-ple-effect evaporators to concentrate glycerine. However, we can’t seem to get thesystem to behave. It will not steady out. The flowrates and pressures in the threeevaporators all seem to cycle. Clear up the problem. All pressures are absolute pres-sures. Figure 8-16 shows the process. Figure 8-17 illustrates the ejector system usedto pull the vacuum.

335


PI

to high

vacuum

cooling water

to/from the

tower

barometricleg

0.43 kg/s

7.9 kPa

0.43 kg/s0.54 kg/s0.61 kg/s

TI

TI

TI TI FR

TI

FR

Steam

0.64 kg/s

PI

228 kPa

PI

164 kPa

98ºC

1.19

kg/s

109ºC93.5ºC 70ºC

1.80 kg/s

to product

0.24 kg/s

80ºC

0.65

kg/s

PI

115 kPa

drainsteam

trap

Figure 8-16 Three multiple effect evaporators for Case’35.

steam

from thesystem

vacuum system

water steam water steam

barometric legs

Figure 8-17 The high vacuum system for the multiple effect evaporators for Case’35.

336


Case’35: Cycling triple effect

. MSDS, 1871








Vacuum system, 2746Three evaporators, 2489Two preheaters, 2344Steam trap, 2147Feed pump, 2097

Vendor files: Triple effect vacuum unit, 2041Commissioning data, P&ID, internal reports, 2255Handbook, Steam tables, 2450Trouble-shooting files, 1818


Pressure profile,If there is a leak: stages 1, 2 and 3; which direction would it leak?, 1759If there is a leak in the preheaters, which direction would it leak?, 1853

Mass balance,Steam: condensate per stage, 1028Process liquid out from the third stage, 1635

Energy balance: sink= source, estimate heat loads based on steady state informa-tion given, 1079

Rate, boiling range: nucleate or film, 1128Equipment performance, three stages, 1178


Visit control room: control-room data: values now and from past records. Sensorsout on the plant.

Process operators, Please tell me about the startup, 1228

337


Contact on-site personnel, Steam utilities: steam conditions to ejector system,1278

Cooling water to the barometric condensers; is the cooling water hotter thanusual, 1328

Visit site, read present values, observe and sense.Glycerine feedrate, FR, 1378Pressure PI stage 1; 228 kPa, abs, 1427Pressure PI stage 2; 164 kPa, abs, 1476Pressure PI stage 3; 115 kPa abs, 814Pressure PI inlet to vacuum system. 7.9 kPa abs, 611Temperature TI inlet to preheater, 962Temperature TI exit stage 1.98 �C, 915Temperature TI exit stage 2.80 �C, 864Temperature TI into stage 1; 109 �C, 866Temperature TI exit preheater 1: 93.5 �C, 571What is the cycle time, 509Steam flowrate to stage 1, 44Steam pressure to the ejectors, 139


Knock on the side of the vertical third effect and listen for a change in soundthat corresponds to the cycle, 281Gloved hand to test the temperature of trap and upstream and downstreamof the steam trap. Do this over the full cycle, 431

Sensors: use of temporary instruments,Use a stethoscope to listen to the steam going to each of the three ejectors inthe vacuum system. Listen for “vibration”, 472

Call to vendors, licensee,Triple-effect evaporator system, 983Steam-trap vendor, 770

Control system, Put control system on manual, 860Samples and measurements,

Sample glycerine feed before pump and just as it enters the first stage; ana-lyze for glycerine content and compare with specifications, 2265Immerse steam-trap discharge line in weighed amount of cold water andmeasure condensate flow over three successive one-minute periods, 2108

Open and inspect,Triple effects and pressure test for leaks, check for blockages, evidence ofexcessive foaming, and fouling of inside or outside of tubes, 2280Preheaters: and pressure test for leaks, check for blockages, and fouling ofinside or outside of tubes, 2207Centrifugal feed pump, 2931Barometric condensers; look for blockage that could have flooded the conden-ser, 2678

338


. Take “corrective” action,Raise the level of the overflow baffle on the hot well to give better seal of thedowncomer, 2904Retune the control system, 2136Replace the valve on the steam line to the first ejector after the booster ejec-tor, 1703

Case’36: The really hot case (courtesy W. K. Taylor, B. Eng. 1966, McMasterUniversity) [7, reactor, heat exchanger, steam drum; reformer, ammonia]The heat from exit gas from the secondary reformer is used to generate steamaccording to the scheme shown in Figure 8-18. The temperatures and pressures formaking 100 kg/s (as measured by F4) are as follows: the location, the design valueand the current value for temperature and pressure, respectively.

PC

6

TIC

3

TI

2

STEAM DRUM

PI

1

TI

1

TI

4

PI

4

LC

1

A

TI

TI

TI

TI

TI

TI

5

WASTE HEAT

BOILER

BOILER FEED WATER

STEAM

SUPERHEATER

METHANE

STEAM

AIR

HP STEAM

HP AIR

COOLING WATER

BYPASS VALVE

CONTROLLED

BY TIC-3

FI

4

PI

3

NATURAL

CIRCULATION

AAA

BYPASS AND

ALTERNATIVE

COOLING

Figure 8-18 Steam generation by the secondary reformer exit gas for Case’35.

339


Location Design T, �C; Reads, �C Design P, MPa Reads, MPa

Reactor exit T1 = 1000 1000 P1 = 4.5 4.55Inside boiler T2 = 750 820 ? ?Boiler exit T3 = 600 730 P3 = ? 4.5Superheat feed T4 = 325 (sat) 325 P4 = 12.0 12.0Superheat exit T5 = 353 443 ? ?

We cannot seem to control the exit gas temperature TI-3; the TI-3 controller out-put to bypass is 100% closed. The steam is much too superheated and could damagethe turbine where it is used. Correct the fault.

Case’36: The really hot case

. MSDS, 113



. IS and IS NOT: (based on given problem statement),

What, 1018When, 2408Where, 2170






Reformer, 2712Steam drum, 2853Waste-heat boiler, 2953Steam superheater, 2651

Vendor files: Waste-heat boiler, 2253Steam superheater, 2453Reformer catalyst, 2056

Commissioning data, P&ID, internal reports, 2017Handbook, Steam tables, 2435Trouble-shooting files, 555


340


Pressure profile,Direction of leak: waste-heat boiler, 902Direction of leak: steam superheater, 520Dp for process gas through the boiler and compare with rule-of-thumb, 37

Energy balance: sink= source,For waste-heat boiler: Heat loss in gas= steam generated, 428For waste-heat boiler: heat load in first section vs second, 1742For superheater: heat loss in gas= superheat, 1958

Rate, Boiling regime (film or nucleate), 1778Equipment performance,

Estimate performance of boiler based on given data, 1538Estimate performance of superheater based on given data, 2967


Visit control room: control-room data: values now and from past records,Temperature TI-1, 2526

Process operators, What has been done? 2688Call to others on-site, Downstream users of superheated steam, 341

Upstream feed to the reformer: any change, 1449Visit site, read present values, observe and sense.

FI-4: flow of steam, 2946Output signal from TIC-3 to baffle, 2599


Are the thermocouple temperatures in the reformer bed consistent with theexit temperature TI-1?, 2120

Sensors: check response to changeTemperature TI-5; superheated steam exit superheater, 2352Temperature TI-2; process gas inside boiler, 2047Temperature TI-3; process gas boiler exit, 1677Pressure PI-1, 1984Pressure PI-3, 53Pressure PI-4, 354Flowmeter FI-4, 804

Sensors: use of temporary instruments,Use contact or laser pyrometer to measure TI-5, 1413Use contact or laser pyrometer to measure TI-1, 1861

Sensors: calibrate,Temperature TI-5; superheated steam exit superheater, 854Temperature TI-2; process gas inside boiler, 40Temperature TI-3; process gas boiler exit, 459Pressure PI-1, 2470Pressure PI-3, 2350Pressure PI-4, 2950Flowmeter FI-4, 2634

341


Control system,Put TIC-3 on manual and open the bypass valve. Note changes is TIC-3 andTI-5, 2495Retune the TIC-3 control-baffle system, 2976Put PC-6 on manual. Open the valve and watch steam drum pressure and PI-4, 584

Call to vendors, licensee,Catalyst in the secondary reformer, 2360

Samples and measurements,Analyze the material found, 2083

Open and inspect,Orifice plate for FI-4 to check if the plate was in backwards. Tab is not clearlymarked, 1556Isolate, drain the superheater. Pull the bundle and inspect for fouling; inspectthe inside of the tubes for fouling, 1904Isolate, drain the superheater. Pressure test the tubes with water and look forleaks. Plug any leaking tubes, 627Isolate, drain the boiler. Pressure test the tubes with water and look for leaks.Plug any leaking tubes, 128Isolate and drain the boiler. Pull the bundle and check for fouling on the out-side of the tubes, 2810Isolate and drain the boiler. Check the location of the bypass valve. Inspectthe inside of the tubes for fouling, 2647

. Take “corrective” action,Correct the linkage between the TIC-3 and baffle, 2149Replace the sensor TIC-3, 1105

Case’37: The mill clarifier (courtesy D. F. Fox, B. Eng. 1973, McMaster University)[7, thickener, sludge pumps, control; pulp and paper]The mill clarifier is 45 m in diameter and designed for 90% reduction in suspendedsolids from an inflow of 0.53 m3/s. The retention time is 2 hour. Parshall flumesmeasure the total inlet and outlet flows. There are 24-hour composite samplers onthe influent and overflow effluent lines. The system is shown in Figure 8-19.

Our target is to have the clarifier effluent < 50 ppm. The operator cannot under-stand why the clarifier effluent has been around 200 ppm for the last 20 days. Thesludge pump has been “wide open”, yet all that occurs is flooding at the belt filter.The feed concentration is about 400 ppm. That’s the problem. Get it fixed pronto.

342


Figure 8-19 The mill clarifier for Case’37.

Case’37: The mill clarifier








Clarifier, 1203

343


Sludge pump, 1456Belt filter, 104524-h composite sampler, 636Parshall flumes, 954

Vendor files: Parshall flumes, 507Commissioning data, P&ID, internal reports, 61Trouble-shooting files, 317


Visit control room: control-room data: values now and from past records,Torque, 493Records of 24 h sampler on feed over the past week, 47Records of 24 h sampler on effluent over the past week, 204Influent flowrate based on Parshall flume, 403Bottoms flowrate from sludge pump to belt filter, 703Overflow flowrate based on Parshall flume, 579

Process operators, What was done this shift? 861Previous shift, 1310

Call to others on-site, Operators of upstream units about possible upsets, 1154Visit site, read present values, observe and sense.

Sludge pump: cavitation? 1402Sludge pump exit valve wide open or mid-range? 1133Rake and skimmer turning at expected rpm, 1056Belt filter: flooding? 1752


Clear any stuff out of the sampler lines on 24 h composite sampler of feedand resample via 24-h composite, 1580

Sensors: use of temporary instruments,Use point gauge on both Parshall flumes and use this information to deter-mine total flowrate, 1516

Sensors: calibrate,Recheck zero setting for the Parshall flumes, 2204

Samples and measurements,Grab samples of feed whenever torque starts to increase. Five successively at20-min intervals. Analyze for suspended solids and compare with 400 ppmreading, 2107Feed to the belt filter. Grab sample when torque increases and when increasepump output. Analyze for suspended solids and compare with 3%, 2353

Open and inspect,Sampler lines; look for plugs in sampler line, 2463Stop process; drain clarifier and inspect rake, 2961

344


. Take “corrective” action,Turn off sludge pump and increase rake speed until torque just starts toincrease; then start sludge pump, 2622Decrease rake speed; keep sludge pump at usual design flowrate, 2534Reduce the sludge-pump flowrate to 5 L/s, 1819Increase vacuum on the belt filter, 1615

Case’38: More trouble on the deprop! (courtesy of T. E. Marlin, Chemical Engineer-ing, McMaster University) [7, sequence of distillation columns with auxiliaries;depropanizer-debutanizer]The process is the depropanizer-debutanizer described in Case’8, Chapter 2. AP&ID is given in Figure 2-4 accompanying Case’8.

There has been a lot of trouble on this unit. So last January, we did a detailedstudy, and simulation, of the deprop and debut units. Our conclusion was that thisunit was operating on spec. for the usual range of feedstocks with all equipmentcomponents operating very close to the design values.

It’s a hot steamy day in August; you are glad that you are in your air-conditionedoffice. The phone rings! The laboratory analysis of the vapor product on the depropindicates a loss of propane to the fuel system that is 21�2 times the design value.This loss of valuable product to fuel gas is costing lots of money. Fix it fast.

Case’38: More trouble on the depropanizer

. MSDS, 1495



. IS and IS NOT, What, 2978When, 2627Where, 2336

. Why? Why? Why?, Best goal? To reduce the propane loss to fuel gas, 1360







345



Commissioning data, P&ID, internal reports, 1043Handbook and Google, Cox charts, 2024Trouble-shooting files, 85


Visit control room: control-room data,Feed to the C8, depropanizer. FC/1, 2831Reflux flowrate, FIC/4, 2036Overhead product flowrate of propane, FIC/5, 1633Pressure drop Dp I/1, 1926Level bottoms LIC/2, 387Temperature bottoms TI/4, 173Temperature mid-column TIC/5, 980Temperature top, TI/3, 789Pressure on overhead drum, PIC/10, 513

Process operators, This shift, 552Previous midnight shift, 1352

Operating procedures, Column pressure, 1185Call to others on-site,

Utilities: what is the temperature of the water leaving the cooling tower; whatis the flowrate to our unit, 1908Operators on downstream propane unit: amount and quality of propanereceived, 872

Visit site, read present values, observe and sense.Column pressure, PI- 4, 2405Pressure relief to flare PSV-1, 2384Look at the flare, 2994Temperature mid-column TI- 8, 2123Valve position for column feed, FV- 1, 2701Valve position for steam to preheater E-24, 2606Valve-stem position on PV-10, 2504Level in feed drum, V-29; LI- 1, 2942Isolation valves around the condenser, 2516Feel temperature of water line going into condensers, 2008

Check diagram and P&ID versus what’s out on the plant, 2769On-site simple tests: Open vents to fuel gas on each condenser for 10 min; then

close, 1512Gather data for key calculations,Pressure profile,

Drum V-29 to pumps, F-25, 26, 959Dp across pump, converted to head, 605Pump F-25–26 exit to feed location, 857Drum V-30, pump F-27 and reflux into column, 549

346


Pump F-27, 1464Vapor from top of column to vapor space in V-30, 1966Thermosyphon reboiler process fluid side, 1676

Mass balance, Over column, 1558Energy balance: Heat load for condenser. Heat load condensed= heat load picked

up in cooling water, 2911Sensors: check response to change, Temperature at the top of column C8: TI 3,

1955Pressure at the top of column C8, PI 4, 1175

Sensors: use of temporary instruments,Use a surface temperature sensor to measure the temperature of coolingwater going into the condensers, 1012

Sensors: calibrate,Temperature at the top of column C8: TI 3, 1263Pressure at the top of column C8, PI 4, 327

Samples and measurements, Sample feed and analyze for the amount of pro-pane, 93

Sample overhead concentration in gas to flare from top of drum every 10minutes for 1 hour. Analyze for propane, 2442Sample overhead concentration from the tower every 10 minutes for 1 hour.Analyze for propane, 2871

Open and inspect,Condensers, 2554

. Take “corrective” action,Direct water from the fire hose onto the outside shell of the overhead conden-sers, 2174Increase column pressure to 1.78 MPa, 424Decrease feed to the column to 80% usual flowrate, 774

Case’39: The case of the lumpy sunglass display (from D. R Winter, UniversalGravo-plast, Toronto, 2004) [Difficulty 7; involves feed bin, molding machine,mold and mold design. Context: injection molding of thermoplastics]The customer wants a display stand for sunglasses. The display required several con-toured plates (35 cm by 18 cm) molded of polypropylene. These are mounted verti-cally and held in place at the edges. The display is to be installed in windows ofdrugstores so that the sunglasses on display are illuminated by sunshine during theday and by fluorescent tubes behind the molded plate during darker hours. Thecustomer requested a special pearlized pigment be added so that the display hadpizzaz.

The feed was a mixture of polypropylene resin, a UV stablizer and the pearlescentpigment. The feed was dried by contacting it with hot air at 93 �C for three hoursand was used within 20 minutes of drying. The injection-molding machine was areciprocating screw; L:D of 20:1; compression ratio 2.5:1 and the shot size for thisjob was 70% of the machine capacity. The processing temperature was 230 �C. The

347


mold pressure was 4.2 MPa, consistent with the part width and the flow length. Amold was created of aluminum with appropriate gate and vent locations and coolingarrangements for the mold. The nozzle, sprue and cold runner system were care-fully designed. The molding cycle was established, a molded sample approved bythe customer and production started.

The first 2000 parts had been shipped when the customer called and complainedthat all the parts looked fine in reflected light but in transmitted light, lumps wereevident in the panels. Production was stopped. Your job is to eliminate the lumpsthat appear when the panels are viewed with transmitted light.

. MSDS, 1572



. More about the product and the mold, 2238






Injection-molding machine, 1983Mold, 1445Feed for this product, 1004

Commissioning data, P&ID, internal reports,Prototype activity. Check if the prototype had lumps when viewed by trans-mitted light, 911

Handbook, Data sheets for resin, 667Trouble-shooting files, 511


Visit control room: control-room data: values now and from past records,Fill cycle, 136Cool cycle, 2738Open cycle, 2521Melt feed temperature into mold, 2310Injection pressure, 1934Injection rate, 1545Extruder: rear-barrel temperature, 1857

348


Extruder: head temperature, 1183Backpressure, 1084Mold pressure, 730Hold pressure, 594Mold temperature, 92

Process operators, 425Operating procedures, Shutdown procedures used, 279

Were the feed materials from the same batches of resin, UV stabilizer andpearlized pigment as was used for the prototype?, 1945


On-site simple tests:Clean possible dirty machine by running a charge of acrylic through theextruder; then try again, 2087Use another molding press; run the process under the prototype conditions,2724Redry the resin for three hours at 93 �C and dry additives and use immedi-ately, 2575Reduce the screw rpm by 5% and use a pyrometer to measure melt tempera-ture, 2771Change melt temperature, increase by 20 �C to 251 �C and keep the coolingtime the same, 2187Change melt temperature, increase by 20 �C to 251 �C and increase the cool-ing time, 2009Use the standard conditions but try an old batch of dried polypropylene froma different supplier together with the UV stabilizer and pearlized pigment,1662Use the standard conditions, use the polypropylene we have been using andthe pearlized pigment but delete the UV stabilizer, 1718Use the standard conditions, use the polypropylene we have been using andthe UV stabilizer but delete the pearlized pigment, 1624Clean the hopper; add a cover over the hopper to prevent atmospheric dirtfrom falling into the feed, 1285Reduce the injection pressure to 7.5 MPa, 1367

Redesign of product and mold,Increase the diameter of the gates by 10%, 631

Sensors: check response to changeHot-melt temperature sensor at nozzle, 1337Exit cooling-water temperature on top mold, 304Exit cooling-water temperature on the bottom mold, 241Pressure sensor at nozzle, 1841

Sensors: use of temporary instruments,Use a pyrometer or laser sensor to measure melt temperature and comparewith temperature sensor, 2663

Sensors: calibrate, Hot-melt temperature sensor, 2635

349


Call to vendors, licensee, or suppliers,Properties of polypropylene, 77Properties of UV stabilizer, 183Properties of pearlized pigment, 2359

Samples and measurements,Sample resin and analyze for moisture, 2206

More complicated tests,Change mold from a QC-7 aluminum mold to a mold P-20, 2873

Open and inspect, Check the shutoff valve for dirt or contamination, 2592

. Take “corrective” action, Replace the shutoff valve, 2849

Case’40: The cool refrigerant (courtesy T. E. Marlin, Chemical Engineering Depart-ment, McMaster University) [7, turbine, compressor, KO pot, refrigeration system;general]The propylene refrigeration system, given in Figure 8-20, was operated successfullyfor several years. Since the steam turbine and refrigerant compressor had spare ca-pacity, they modified the process by adding an additional heat exchanger, E101, tocool the process stream in E101, as shown in Figure 8-21. When the process wasstarted up, the design values could not be obtained for either of the streams to becooled. The temperatures, given in �C, are:

For E100: T5: actual 100, design 101; T6, actual 10; design 5.For E101: T7: actual 70; design 68; T8, actual 4; design 10.Into the compressor: T2: actual –5; design –4.

350

C.W.

PC

1

H.P. Steam

periodic flow

F

1

T

2

L

1

L

2

LC

3

T

5

T

6

chilledprocess

E 100

K.O. Drum

compressorsteamturbine

Figure 8-20 The original propene refrigeration system for Case’40.


Because T6 is too high and T8 is too low, the plant cannot produce saleable mate-rial. In checking over the design calculations, you realize that the system shouldwork. What’s going on here?

C.W.

PC

1

H.P. Steam

periodic flow

F

1

T

2

L

1

L

2

LC

3

T

5

T

6

chilledprocess

E 100

K.O. Drum

compressorsteamturbine

LC

4

E 101

T

7

T

8

chilled

warm

Figure 8-21 The modified refrigeration system for Case’40.

Case’40: The cool refrigerant

. MSDS, 914





351





E100 chiller, 1031E101 chiller, 567

Vendor files: E100 chiller, 883E101 chiller, 2485

Commissioning data, P&ID, internal reports, 2215Handbook, Propylene temperature–pressure data and latent heat, 2913Trouble-shooting files, 2544


Pressure profile, Direction of leak: unable to tell because the pressures of the pro-cess streams are not known.

Direction of leak from propylene–cooling water in condensers, 2658Energy balance,

Heat load for E100; actual versus design, 2708Heat load for E101; actual versus design, 2157Propylene flow: design versus actual calculated from heat loads on E100 andE101, 1734

Rate,Does propylene boil as nucleate or film form, 2208

Equipment performance,E100: does heat flow in the correct direction?, 1707E101: does heat flow in correct direction?, 1582UA calculations for E100 actual versus design, 190UA calculations for E101 actual versus design, 740


Visit control room: control-room data: values now and from past records,Temp and pressure of propylene: T2 and P1; are these on pH diagram forsaturated propylene gas; if not might it be sensors? or contamination? 1140Level: E100, LC/ 3, 69Level: E101, LC/4, 362Level: KO pot, L/1, 449Level: liquid propylene drum, L/2, 944Flow of propylene, F/1, 643Propylene pressure at compressor suction, PC/1, 713

Process operators, Please tell me what has happened to far, 1369Operating procedures, 1125

352


Contact with on-site specialists,Operators of unit that provides and receives the process stream from E100:flows and temperatures constant and consistent with what’s on this unit?,1046Operators of unit that provides and receives the process stream from E100:pressure in the lines and have you detected any contamination in your pro-cess streams, 1809Operators of unit that provides and receives the process stream from E101:pressure in the lines and have you detected any contamination in your pro-cess streams, 1932Operators of unit that provides and receives the process stream from E101:flows and temperatures constant and consistent with what’s on this unit?,2338Utilities: steam pressure, flow and degree of superheat, 2144Utilities: cooling-water temperature, flow to propylene condensers, 2414

Visit site, read present values, observe and sense.Steam leave off the top of the steam header?, 2125Valve-stem position on valve LC/3, 2783Valve-stem position on valve LC/4, 2843


Observe movement of the valve stem on LC/3 when the set point is changed,1180Observe movement of the valve stem on LC/4 when the set point is changed,615Open vent bleed on the top of E100 for 5 minutes, close and observe change,842Open vent bleed on the top of E101 for 5 minutes, close and observe change,688Open vent bleed on the top of propylene drum for 5 minutes, close andobserve change, 124

Consistency check,Temp and pressure of propylene: T2 and P1; are these on pH diagram forsaturated propylene gas; if not might it be sensors? or contamination?, 1140

Sensors: check response to changeLC/3, 394LC/4, 247

Control system,Put LC/3 on manual and adjust level to control temperature T/6, 94Put LC/4 on manual and adjust level to control temperature T/8, 2389Retune controller LC/3, 2890Retune controller LC/4, 2683

353


Sensors: calibrate, LC/3, 2597LC/4, 2072

Samples and measurements,Sample propylene and analyze for contaminants, 657

More complicated tests,Reduce the propylene pressure such that the target range of cooling occurswhen neither bundle is completely covered with liquid; adjust the levels inE100 and E101 separately until target temperatures are achieved, 130

Open and inspect,Propylene drum to check the condition of the vortex breaker at the bottomexit nozzle, 265E100 and look for fouling inside or outside tubes, 1649E100; isolate and pressure test for leaks, 2263E101 and look for fouling inside or outside tubes, 2789E101; isolate and pressure test for leaks, 2019Check for plugs, obstructions or junk in the line from the liquid propylenereservoir and E100, 1638

. Take “corrective” action,Replace LC/3 on E100, 1967Replace LC/4 on E101, 1094Replace control valve on propylene to E100, 775

Case’41: The ever-increasing column pressure (courtesy T. E. Marlin, Chemical Engi-neering Department, McMaster University) [7, distillation plus auxiliaries; general]The column for the new process was started up for the first time one month ago.From the beginning, the pressure control was not very good; the pressure seemed todeviate from its set point more than in other columns in the plant. However, theproduct purities seemed to be within specifications, so you didn’t worry about it.

Today, the plant operator calls, “We’ve got a problem on that new column! Thepressure controller is not working! The pressure in the column is higher than theset point yet the controller output is 0%! Yes, the production rate and the specifica-tions on all lines are OK. But this pressure is really worrying me.” The column isshown in Figure 8-22.

354


Figure 8-22 The column discussed in Case’41.

Case’41: Ever-increasing pressure

. MSDS, 379








355


Column, 2706Reboiler, 2559Control system, 2956Condenser, 2867

Commissioning data, P&ID, internal reports, 2695Handbook, Options for controlling overhead pressure based on Chin’s classic arti-

cle from 1979Hydrocarbon Process, 2184Type of hydrocarbon based on overhead pressure and temperature, 2410



Pressure profile,Direction of flow if there is a leak: condenser, 2069Direction of flow if there is a leak: reboiler, 129Flow from top of column to the reflux drum, 328

Equipment performance,Control system, 445Condenser, 205


Visit control room: control-room data: values now and from past records,Current overhead temperature, 886History of overhead temperature, 736History of overhead pressure, 1434Current overhead pressure; PC/1, 630

Process operators, When you first started up two months ago, was the controlvalve on the CW almost closed?, 534

Contact with on-site specialists,Utilities: cooling water: temperature and flowrate, 1935Operators of upstream plant supplying the feed to the column: any change inlight ends in composition or change in flowrate, 2385

Visit site, read present values, observe and sense.Check for consistency between PC/1 and the pressure on the top of the col-umn, 2205Valve-stem position on the cooling water, 1109Look at the flare, 1247Amount of steam flowing to the reboiler, 1313Bottom pressure, 1299Read the controller output signal to the cooling-water valve, 1619Is the control valve on the cooling water fail open or fail closed? does thedirection arrow on the valve agree with the direction of flow?, 2938Bottom temperature, 1454

Check diagram and P&ID versus what’s out on the plant, 1119

356


On-site simple tests:Open vent on the condenser for 2 min to bleed off any accumulated inerts,close and check performance, 1753Does the valve stem on the cooling water respond to increase in set point,1543Increase the resistance in the vent break line to prevent uncondensed gasfrom bypassing the condenser, 1888Tap the side of the condenser in a vertical line and listen for the change insound suggesting the location of the vapor-condensate interface, 2631

Sensors: check response to changeDoes PC/1 respond to change in set point, 2160

Sensors: use of temporary instruments,Use a contact surface temperature probe to measure inlet and outlet temper-atures of the cooling water to the condenser, 2063

Sensors: calibrate,Pressure PC/1, 2720

Control system,Retune the PC/1 control system, 2921

Samples and measurements,Sample the feed; analyze for light ends and non-condensibles and comparewith previous records, 2539

More complicated tests,Stop the process; crack open the flange on the vent break line and insert aresistance disk with a smaller diameter hole to increase the resistance in the“bypass” line, 2652

Open and inspect,Control valve on the cooling water; inspect for quality of trim, plugged, 1559Condenser and inspect for fouling inside the tubes, 1123

. Take “corrective” action,Reduce the feedrate to the column, 1371Direct fire hose water over the condenser, 231

Case’42: The weak AN (courtesy W. K. Taylor, B. Eng. 1966, McMaster University)[7, storage tank, reactor, pump, heat exchanger; ammonia]Ammonium nitrate is formed by the exothermic neutralization of nitric acid withammonia. The nitric acid is produced as a 56% solution and is pumped from stor-age tanks to the reaction leg of the primary neutralizer. In the reaction leg, the acidis mixed with superheated ammonia vapor and also with an ammonia-bearing off-gas stream from the urea plant. The heat of reaction produces a large amount ofsteam, and the liquid, which is removed to a secondary neutralizer is an 85% solu-tion of ammonium nitrate. Usually about 50% of the ammonia needed to neutralizethe acid is provided by the off-gas stream from the urea plant. The usual tempera-tures in the system are: nitric acid in storage: 40–50 �C; superheated ammonia vaporentering neutralizer: 90–100 �C; the urea plant off-gas: 100–110 �C; the AN product

357


solution: 130–140 �C; the temperature inside the neutralizer: 135 �C. The process isshown in Figure 8-23.

Primary

Ammonium

Nitrate

Neutralizer

Nitric

Acid

Storage

Tank

steam

TI

TI

Flow

meter

FCV

cooling

water

PI

Acid

Transfer

Pump

NH

vapor

3

urea plant

off gas

C.W.

return

A.N.

product

to secondary

neutralizer

Figure 8-23 The process for AN for Case’42.

The urgent phone call poses today’s problem. The temperature inside the neutra-lizer-reactor has been falling. The most recent analysis of the AN shows a concentra-tion of about 79%. “This weak AN will cause problems in the downstream process-ing units where the solution is concentrated in a falling film evaporator to about99% and then prilled. Let’s get this fixed quickly. The evaporator can’t concentratethe solution enough for prilling and the whole process will shut down. “

Case’42: Weak AN

. MSDS, 2716









358


Storage tank, 906Check the sensors on the off-gas and ammonia line to ensure that there is nomercury used in the sensor, 396Design calculations for the orifice plate measuring the nitric acid flowrate,1940Transfer pump, 1796Neutralizer, 2826Cooling coil, 2628

Vendor files: Transfer pump, 2957Commissioning data, P&ID, internal reports, 1338Handbook, Heat of formation of ammonium nitrate, 1186Trouble-shooting files, 1116


Pressure profile,Direction of leak: cooling-coil reactant, 743Direction of leak: steam heating coil to acid in storage tank, 645Acid flow from the storage tank into the neutralizer, 142

Rate,Of neutralization reaction, 2448Of cooling, 2492

Equipment performance,Neutralizer and sources of water, 2937


Visit control room: control-room data: values now and from past records,Temperature on acid storage tank, 2749Pressure at exit of pump, 2638Acid flowrate, 2105Temperature in neutralizer, 2306Temperature of AN leaving the neutralizer, 1306Cooling-water temperature out, 1149Steam generation from overhead of neutralizer, 1050

Process operators, Please describe what has happened so far, 750Has this ever happened before and what did you do, 570

Contact with on-site specialists, Utilities: cooling water, 199Operators of downstream plant receiving our AN: check on temperature andconcentration of AN received, 448Supplier: urea plant off-gas: change in flow, temperature or concentration,263Supplier: ammonia plant supplier: change in flow, temperature or concentra-tion, 1829Purchasing; have we changed suppliers for the nitric acid, 1735

Visit site, read present values, observe and sense.

359


Check that the location of valve stem on acid flow-control valve is mid-range,1599Level of liquid in the neutralizer from the level gauge, 1731Do the tabs on the orifice plate, used to measure the acid flowrate, indicatethat the orifice is facing the correct direction, 2290What sounds do you hear by the neutralizer, 2795Are there sounds of cavitation from the transfer pump?, 2532Are there any water/steam lines hooked up to any of the process equipmentor lines feeding the neutralizer, 1250Are there any water/steam lines hooked up to the neutralizer, 1350Does the steam off the top of the neutralizer go into the top or bottom of thesteam header, 1100


Check that valve on bypass of acid flow-control valve is shut and block valvesfully open, 849Shut the valve on exit of transfer pump and compare pressure (head) withhead on head-capacity curve for this impeller rpm, 948

Consistency checks,Does composition of the off-gas add up to 100%, 250Does temperature and composition of AN leaving neutralizer agree with con-ditions downstream, 296

Trend checks, Do both the temperature and the concentration of AN decrease inthe neutralizer?, 1838

Gather data for key calculationsMass balance,

Based on the measured flows of reactants, products, 82Based on the measured flows of off-gas and pure ammonia, calculate the frac-tion of the required amount of the ammonia that comes from the off-gas andcompare with design, 422

Energy balance: sink= source,Heat of reaction= heat removed in cooling water, 2347

Sensors: check response to changeTemperature of AN in the neutralizer, 1697Flow-control valve of nitric acid feed, 1548

Sensors: use of temporary instruments,Contact or laser sensor of temperature of the cooling water entering and leav-ing the cooling coil, 1877Contact or laser sensor of temperature of the contents of the neutralizer,2841Contact or laser sensor of temperature of urea feed to the neutralizer, 2264Contact or laser sensor of temperature of the ammonia vapor feed to the neu-tralizer, 2090Contact or laser sensor of temperature of the AN leaving the neutralizer,1849

360


Contact or laser sensor of temperature of the acid in the storage tank, 1749Sensors: calibrate,

Temperature sensor on the AN neutralizer, 346Temperature sensor on the AN at the exit nozzle from the neutralizer, 118

Control system,Retune the flow control on the nitric acid, 1200

Call to vendors, licensee, or suppliers,Supplier of nitric acid: concentration and additives, 1293

Samples and measurements,Product AN at exit nozzle from the neutralizer; analyze concentration of AN,811Nitric acid in storage tank; analyze concentration, 2233Nitric acid at nozzle at the entry into the neutralizer; analyze concentration,2466Nitric acid at sampler before acid-transfer pump; analyze concentration, 2775Nitric acid at flow-control valve; analyze concentration, 2581Ammonia vapor at nozzle into neutralizer; analyze concentration of ammo-nia, 2321Urea off-gas at nozzle into neutralizer: analyze concentration, 1485

Open and inspect,Cooling coil; shut down; pressure test for leak in the cooling coil in the neu-tralizer by installing a gauge; pressurizing with air, isolating and notedecrease in pressure with time, 838Cooling coil; shut down; drain neutralizer, pressure test for leak in the cool-ing coil in the neutralizer by isolating and hydraulically pressurize coil; lookfor water leaks, 791Cooling-coil fouling: Shut down; drain, make safe for entry. Visually inspect.Look for fouling on outside of coil and at entry and exit on the inside, 355Trombone, steam heating coil in acid storage tank. Isolate and statically pres-sure test with air, gauge and observe pressure decrease with time, 936Condition of gas sparge rings on both off-gas and ammonia. Shut down;drain, make safe for entry. Visually inspect, 1472Acid transfer pump; isolate, drain, inspect, 2155

. Take “corrective” action,Interchange the off-gas and ammonia feeds to nozzles on the neutralizer sothat the off-gas is at the lowest nozzle, 2703Relocate the ammonia entry nozzle and sparge so that this is below the cool-ing coils; install two layers of static mixers above the gas sparge to improvemixing, 2945Install a central draft tube in the neutralizer to improve mixing in the neutra-lizer, 1399

361


Case’43: High pressure in the debut! (courtesy of T. E. Marlin, Chemical Engineering,McMaster University) [7, sequence of distillation columns with auxiliaries;depropanizer-debutanizer]The process is the depropanizer-debutanizer described in Case’8, Chapter 2. AP&ID is given in Figure 2-4 accompanying Case’8.

This is the startup of the unit after the annual turnaround. During the turn-around several valves were replaced, pumps were dismantled and reassembled, thecolumn internals were inspected, the heat exchangers, condensers and reboilerscleaned and the sensors calibrated. This was done for both the debutanizer, C-9, andthe depropanizer, C-8.

Shortly after the unit starts up there is an upset on the depropanizer and we havejust resolved that in Case’41. This Case’41 had symptoms that the depropanizeroverhead product has far too much C4 and the debutanizer overhead has far toomuch C3. (You don’t have to work Case’41 before you work on this one.) Now theoperator panics and calls you again. This time she thinks that the pressure is sohigh in the debutanizer that the safety relief valve has popped! That means butaneis spewing out; the flare should be flashing! Now what?

Case’43: Debutanizer pressure relief

. MSDS, 1495




. Why? Why? Why?, Best goal? To manage the purported high pressure on thedebutanizer, 1321






Vendor files: Condenser, reboiler, 1448Steam traps, 1244

Commissioning data, P&ID, internal reports, 2051

362


Handbook and Google, Cox charts, 2114Trouble-shooting files, 2809


Visit control room: control-room data,Feed to C9, debutanizer, FI/2, 365Feed to the C8, depropanizer. FC/1, 601Overhead liquid product butane, flowrate FIC/7, 994Reflux flowrate, FIC/6, 1491Pressure drop Dp I/2, 1102Level bottoms LIC/4, 1701Level in feed drum, V-31; LIC-5, 1551Temperature bottoms TI/12, 2401Temperature top, TI/11, 2902Analyzer A-1, 2570Alarm on PIC/19, 2505Pressure on overhead, PIC/19, 2443

Process operators, This shift, 278Call to others on-site, Utilities: any change in the cooling tower operation that

might affect our site, 22Visit site, read present values, observe and sense.

Column pressure, PI-12, 458Pressure relief to flare PSV-3, 202Observe flare, 417Temperature of the feed to the column TI-10, 505Cooling-water temperature in to the condenser, TI-13, 960Cooling-water temperature out of the condenser, TI-14, 735Valve position for column feed, FV-2, 599Valve position on PIC/19; downflow from the condenser to the reflux drum,853Signal to PIC/19, 1301Pressure on exit of reflux pump F-29, PI-20, and compare with head-capacitycurve at feed flowrate, 1253Listen to the reflux pump F29 for sounds of cavitation, 1016Observe whether shaft is rotating for the reflux pump F-29. Check that thedirection of rotation is consistent with arrow on the housing, 1991


Open bleed valves to fuel on the shell of the condenser for 10 minutes, thenshut, 1605Tap the side of the condenser along a vertical line to listen for change insound associated with liquid level in the condenser, 1509


Drum V-31 to pumps, F-29, 2863

363


Dp across reflux pump, converted to head, 2984Drum V-31, pump F-29 and reflux into column, 2723Pump F-29, 2594Thermosyphon reboiler process fluid side, 2531

Energy balance: sink= source,Estimate the steam flow to the reboiler based on the reflux rate and the factthat each kg steam boils 5 kg typical organic, 104

Sensors: check response to change,Temperature sensor at the top of column, TI-11, 1634Pressure sensor at top of column PI-12, 2151Pressure sensor on overhead line for PIC/19; PT-19, 2319Pressure sensor on reflux pump exit line, PI-20, 2031

Sensors: use of temporary instruments,Measure the surface temperature on the outside of the condenser, 329

Sensors: calibrate,Temperature sensor at the top of column, TI-11, 464Pressure sensor at top of column PI-12, 28Pressure sensor on overhead line for PIC/19; PT-19, 702Pressure sensor on reflux pump exit line, PI-20, 607

Control system,Check if alarm on PIC/19 is faulty or if signal to control room is faulty, 813Put overhead pressure control, PIC/19; reflux control FIC-6 controllers onmanual and try to steady out the column, 901

Sample and analyze, Current feed to the debutanizer for the amount of C3, 2488

. Open and inspect,Condenser, 2916Pressure control valve PV-19, 2552Reflux pump, F-29, 2513Line from the top of the column to the condenser, E-28, 2825Sensor tap for pressure on the overhead line, PT-19, 1219

. Take “corrective” actionReplace the pressure control valve PV-19, 1460Replace the pressure gauge PT-19, 1055Replace high-pressure alarm on PIC/19, 1347Direct water from the fire hose onto the outside shell of the overhead conden-ser, E-28, 786Reduce the reboiler duty, 2983

Case’44: Reactant storage [8, steam coil, storage tank, controller; general]Reactant must be stored at 50– 1 �C. To achieve this a steam-heated coil is placedinside the open, 2.25-m3 vertical cylindrical storage tank as illustrated in Figure8-24. The instrument engineers designed a control system that satisfies the temper-ature requirement. Although the sensor, valve and controller are the fanciest on themarket, the best it can do is keep the temperature 50 –10 �C according to the control-

364


ler chart. The control engineers have worked on tuning the control loop for a weekbut find nothing wrong. This reactant ties up $2,000 worth of production per hour.Solve the problem.

10030

Sample Chart

Effluent

Condensate toBoiler House

BucketSteam

Trap

Feed

Steammain

Condensatemain

T

Time

40

60

Figure 8-24 The reactant storage vessel for Case’44.

Case’44: Reactant storage







365



Controller, 2412Steam control valve, 2718Heating coil, 2811Steam trap, 2696

Vendor files: Steam traps, 2547Commissioning data, P&ID, internal reports, 2124Handbook, Steam tables: Temperature for saturated steam at 0. 205 MPa g and

1.5 MPs-g, 2014Trouble-shooting files, 2254

. Calculations and estimations (that can be done in the office before specialtests are done) not sufficient data given.


Visit control room: control-room data: values now and from past records,Steam pressure in main, 1220Feed flowrate into the tank, 1409

Process operators, General comments about the situation, 1492Temperature variation in the feed entering, 1037

Call to others on-site, Utilities: steam pressure and quality of steam, 1014Visit site, read present values, observe and sense.

Pressure in the condensate header, 727Check if insulated and quality of the insulation, 1810Check that the bypass valve on the condensate trap is closed and that theblock valves are open, 935

Check diagram and P&ID versus what’s out on the plant, 2287On-site simple tests: Open bypass on the steam control valve, 517

Open bypass on condensate trap, 2421Open bypass on steam control valve; block off steam valve; open drain, 612Give the bucket trap a sharp hit to dislodge any crud that might interferewith the mechanism, 402Shut valve on exit of trap for several minutes and then open slowly to “reseal”the trap, 79Note position of valve stem on steam valve over “two cycles” of the tempera-ture variation, 2917


Pressure drop across the control valve on inlet side; estimate, 2462Estimate the pressure to push the condensate up vertically to the condensateheader, 2092

Mass balance, estimate the mass balance on the steam, 1668Energy balance: sink= source, steam required= amount of heat up the reactant,

1903

366


Sensors: use of temporary instruments,Use contact or laser pyrometer to measure the temperatures upstream anddownstream of the trap; compare with expected 200 and 134 �C, 2777Use contact or laser pyrometer to measure feed temperature into the vesselover two cycles, 2577Use the stethoscope to listen to the bucket trap over “two cycles” of the tem-perature variation and compare with expected “loud initially, followed bylower-pitch bubbling and then no noise” for each cycle, 2979Use two contact or laser pyrometers to measure the temperatures upstreamand downstream of the trap over “two cycles” of the temperature variationand compare with expected 200 and 134 �C, 1702

Sensors: calibrate, Temperature sensor in vessel, 1207Pressure sensor on steam main, 705

Control system, Put controller on manual, 2687More complex tests, Add visible tracer to observe mixing patterns in vessel, 1560Open and inspect,

Block off steam trap; operate on bypass; open strainer and check for clogging.If clogged, clean and restart, 588Steam heating coil, 156Bucket steam trap; then clean if necessary and restart, 408

. Take “corrective” action,Insulate vessel, 666Repipe steam connections to coil so that steam comes in the top and conden-sate to trap is out the bottom of the coil, 1093Add a portable, top-entry propeller mixer, 1494Relocate the feed pipe entry under the surface, 2898Replace bucket trap with a float trap, 2530Install a check valve on the line to the condensate header, 1763

Case’45: The deprop bottoms and the ISO dilemma (courtesy of T. E. Marlin, ChemicalEngineering Dept., McMaster University) [8, sequence of distillation columns plusauxiliaries; petrochemical, refinery]The process is the depropanizer-debutanizer described in Case’8, Chapter 2. AP&ID is given in Figure 2.4 accompanying Case’8.

To be able to sell your products, your plant must obtain ISO certification. As aresult, you have established a routine analysis of various streams in the depropani-zer-debutanizer system. The initial laboratory analyses indicate too much variabilityin the mole fraction of propane in the bottoms of the depropanizer, C-8. For the lastday, the mole fraction has been about 0.04, while the target is 0.015. Before the newprocedures, we never knew that we were operating the plant so poorly, so no onecared! Now everyone is frustrated that you have found this fault.

If you cannot obtain ISO certification, the company will not be able to sell prod-ucts to key customers. Resolve the problem.

367


Case’45: The deprop bottoms and the ISO dilemma

. MSDS, 1495



. IS and IS NOT, What?, 1477When?, 2237Where?, 2537

. Why? Why? Why?, Best goal? To reduce the propane concentration in the bot-toms to within specifications, 2211






Condenser, E-25, 591Distillation column, C8: 515Thermosyphon reboiler, E-27, 2380Feed pump, F-25; F-26, 16Turbine drive, 1193Motor drive, 1072Reflux pump, F-27, 286Feed preheater, E-24, 1989Feed drum, V-29, 1478Overhead drum, V-30, 1124




Equipment performance, Nucleate or film boiling in reboiler? check DT, 642Energy balance: Estimate the steam flow to the reboiler based on the reflux rate

and the fact that each kg steam boils 5 kg typical organic, 2821


Visit control room: control-room data,Control temperature on the C8, depropanizer. TC/5, 1563Bottoms temperature. TI/4, 1156

368


Steam flow to reboiler. FIC/3, 2078Debutanizer overhead analyzer for C3: AC/1, 436Level in bottoms, LC/2, 103Flow of bottoms to the debut, FI/2, 2251Feed flowrate to column. FIC/1, 766

Process operators, This shift; variation in AC/1?, 1334Previous shift: variation in AC/1, 433

Call to others on site,Operators of facility receiving the butane as overheads from the debutanizer:propane contamination in product, 219

Visit site, read present values, observe and sense,Column temperature on tray’9, TI-8, 1303Observe the flare, 2815Valve-stem position on LV/2, bottoms flowrate from the deprop, C-8, 2989Valve-stem position on FV-3; steam to reboiler, 2614


Put control of steam to reboiler E-27 on manual and control based on TC-5,2585

Trend check, do trends in the lab sample results follow the same trends in thereadings from AC/1?, 652

Control system,Retune the control system for the bottoms of the deprop, C-8. TC/5, FC/3and LC/2, 894Operator adjust the set point on TC-5 based on the measurement of A-1, 201

Sensors calibrate,Calibrate sensor on the top of the debut. A/1, 454Calibrate temperature sensor, TC/5 on tray’9 of the column, 952

Samples and measurements,Sample feed to C-8, the deprop and check for > usual concentration of C3,1202Sample the bottoms from C-8 once per hour for three hours. Flush sampleline well before collecting each sample. Analyze in lab for C3, 1107Draw three samples of the bottoms from C-8; send one sample to our lab andothers to two independent labs. Analyze for C3, 1453Are the lab sample results consistent with the readings from AC/1?, 2102

Open and inspect,Reboiler: pull bundle and check for fouling on both inside the tubes and out-side, 1952Column at tray’9 and check that TI-8 and TC-5 are “well situated” and thewells are not corroded; that the downcomer is sealed and the tray is level,1651Column at bottoms and check that vortex breaker is present and that thermo-well for temperature sensor TI-4 is “well situated” and the well is not cor-roded. Level sensor is OK?, 1051

369


. Take “corrective” actionInstall a C3 analyzer on the feed into the column, 2751Replace the thermocouple at TC-5; retune the controller, 1391Replace the steam trap on the condensate leaving reboiler, E-27, 1007

Case’46: The not so cool chiller (courtesy of Scott Lynn, Chemical EngineeringDepartment, University of California, Berkeley) [8, pump, exchanger, ethylenerefrigeration; polymerization]The system below is designed to prepare butene in methyl chloride solvent for poly-merization. This reactor feed is to be dried, condensed and cooled to–29 �C withammonia refrigerant, and finally chilled to –96 �C with boiling C2H4 in a thermosy-phon chiller. The ethylene tube-and-shell chiller will cool the reactor feed to –96 �Cat only half the design flowrate or to only –60 �C at the design flowrate. This reducespolymer production by 0.45 Mg/h at an out-of-pocket loss in profit of $1/kg. Get thisgoing correctly. The process is shown in Figure 8-25.

Liquid ethylene

LRC

1

TI

4

TI

2

TI

3

TI

1

Chiller

Feed pump

Ethylene

head tank

To Reactor

Alumina

dryer

FRC

2

FRC

1

Feed at

design

rate, 450

kPa g;

25C

PRCV

set at 22

kPa g-101C

-101C

-29C

Ammonia

Thermosyphon

Figure 8-25 The chiller for the reactant feed for Case’46.

370


Case’46: Not so cool chiller

. MSDS, 423









Chiller, 1616Feed pump, 1223Drier, 1061Ammonia condenser, 647Feed drum, 823Ethylene head tank, 312FRC controller, 1937

Vendor files: Feed pump, 2924Commissioning data, P&ID, internal reports, 2680Handbook, Methyl chloride: vp at 25 �C and compare with pressure on vapor of

450 kPa g, 2555Heat capacities of butene, water, methyl chloride and ammonia, 2640



Pressure profile,From inlet to drier at 450 kPa g to feed drum, 2598Direction of leak: ammonia condenser, 343Direction of leak: chiller, 134From the vapor from the heads tank to the ethylene compressor, 792From feed drum to reactor, 613

Energy balance: sink= source, vaporisation load: design load vs; current full loadand half-load, 907

Rate, In chiller, nucleate vs film boiling on shell side for full versus half-load, 863Equipment performance,

Chiller: rating comparison of UA for full load vs half-load, 746

371


Chiller: tube side and shell side transfer coefficient, 1429Compressor-condenser in the refrigeration loop, 1187


Visit control room: control-room data: values now and from past records,Flowrate of vapor to the driers, FRC, 1095Feed temperature to chiller, TI, 1302Temperature of the process fluid exit of chiller, 1397Temperature of the ethylene vapor off the chiller, 1780Flow of ethylene off top of head tank, FR, 1942Setting on PRCV on ethylene head tank, 2222

Process operators, Please tell me about the operation so far, 2471Contact with on-site specialists,

Contact operators of reactor; temperature and flowrate to reactor consistentwith the temperatures and flowrates from chiller unit, 2193Unit supplying and receiving the ethylene, 2788

Visit site, read present values, observe and sense.Feed pump: sounds like cavitation? 2584Valve position for vapor feed at FRC/1, 2697Tab on orifice: indicating that the plate has been put in correctly, 2914Check that the block valve on bypass around the FRC on the vapor feed to thedryers is CLOSED, 2217Liquid level in ethylene head tank, 2084Valve position for liquid ethylene feed, 2042Temperature on vapor line off top of ethylene head tank, 1625Check that the two valves on the vapor bypass line are fully closed. Checkthat the “flow direction” through the valve is correct, 1817Check that the valve on the ethylene thermosyphon line is fully open andthat the “direction of flow” through the valve is correct, 714Check that the bypass valve on the feed pump is fully closed and that the“direction of flow” through the valve is correct, 875Check that the valve on the pump discharge is fully open. Check that “flowdirection” on the valve is correct, 1869


Adjust the PRCV to a slightly higher pressure and note temperatures, 1693Complete a “turn and seal” test on two valves on the vapor bypass line.Ensure the valves are left shut, 1592Complete a “turn and seal” test on the valve on the ethylene thermosyphonline. Ensure the valve is fully open, 1929Complete a “turn and seal” test on the bypass around the pump. Ensure thevalve is fully closed, 2739Tap side of the ethylene head tank to try to detect liquid level. Compareheight with level gauge reading, 2616Tap side of the feed drum to try to detect liquid level, 2173

372


Tap side of the chiller to try to detect the liquid level. Compare with the levelof exchanger tubes. Are the tubes completely covered by the ethylene, 2340

Consistency checks, Check that temperature and pressure measured at top ofchiller are consistent for “pure” ethylene, 1821

Temperature sensors: TI/3 on ethylene chiller agree with TI/4 on top of headtank, 1688Process temperature from chiller exit agree with downstream unit, 1529

Sensors: check response to changeTemperature sensor of the ethylene vapor TI/3, 1071Temperature sensor of the process liquid out of chiller; TI/2, 1213

Sensors: use of temporary instruments,Surface or laser temperature sensor for the overhead ethylene from the chil-ler, 127Use clamp-on ammeter to measure the amps on the motor driving the feedpump; compare with specs, 397

Sensors: calibrate,Temperature sensor of the ethylene vapor TI/3, 756Temperature sensor of the process liquid out of chiller; TI/2, 1333

Control system,Retune the FRC controller on the feed to the drier, 1042Retune the LRC controller on the ethylene heads tank, 2298

Samples and measurements,Sample ethylene and analyze for contamination, 2440Sample process liquid from the feed drum. Analyze for water and ammonia,2875

More complex tests,Steam regenerator for the drier system. Isolate, once conditions are safe,hydraulically pressure test to identify leaks in the tubes or between the tube-sheet and the tubes. Any leak of steam into the drier system might possiblyload the adsorbent with water, 2538Gamma scan to determine the interface location in the chiller, 2947

Open and inspect,Chiller, check for fouling. Isolate, drain, once conditions are safe, pull out thebundle and check for fouling inside and outside tubes; use optic fiber probeas needed, 2091Feed pump. Isolate, when conditions are safe, open and inspect, 2256Line from the heads tank to the chiller for “thermosyphon operation”: isolate,when safe, open and inspect for blockages. Use optic fiber probe as needed,1161

. Take “corrective” action,Replace the two valves on the vapor bypass around the chiller, 776Replace the globe valve on the thermosyphon line from the head tank with aball valve, 214

373


Case’47: The fluctuating production of acetic anhydride [8, reactor, vaporizer, conden-sers, liquid ring vacuum pump, control; petrochemical]To produce acetic anhydride from acetic acid, via the Wacker process, about 0.6 kg/sof acetic acid is vaporized and sent into a series of three cracking coils that are oper-ated under vacuum. The three cracking coils are inside a furnace as is illustrated inFigure 8-26. But producing acetic anhydride is not a simple process. The acetic acidis cracked in the furnace to produce ketene and water. The reaction is reversible. Ifthe water is not condensed and removed rapidly from the reactor effluent, thereverse reaction could occur and the product from the reactor would be acetic acid –the reactant you started with. If the removal of the water is successful, then gaseousketene leaves the reactor condenser system and reacts with fresh liquid acetic acidin an absorber to produce acetic anhydride. The condensers for the water are a seriesof Liebig condensers using first water and then brine as the coolant. A further com-plication is that ketene dimerizes and forms a gunk that plays havoc with the opera-tion of the reciprocating vacuum pumps.

AcOHLIQUID

STEAM

ACETICACID

VAPOUR

FURNACEWATER WATER

VACUUMPUMPS

AcOH

Figure 8-26 The furnace reactor and condensers to produce acetic anhydride in Case’47.

PI

201

PI

101

FI

201

LIQUID

ACETICACID

TO CONDENSATEHEADER

TO CRACKING

FURNACE

STEAMMAIN

LC

A

201

FRC

202

��

Figure 8-27 The vaporizer for the feed to the furnace in Case’47.

374


The system has experienced serious upsets. The amount of acetic anhydride prod-uct fluctuates greatly. Some suspect the vaporizer that is shown in Figure 8-27.Some think that the condensation of the water from the ketene is the source of thefluctuations. Still others point to the vacuum pump. Perhaps it is the absorber. Sortit out! and sort it out fast!

Case’47: Fluctuating production of acetic anhydride

. MSDS, 339








Cracking reactor, 46Condenser, 117Vacuum pumps, 363Orifice meter FRC/201, 474Absorber-reactor, 985Vaporizer, 622Steam trap, 541Gas-fired furnace for the reactor coils, 1433

Vendor files: Vacuum pump, 1038Commissioning data, P&ID, internal reports, 1023Handbook, Properties of acetic acid, 1267Trouble-shooting files, 1363


Pressure profile,Pressure drop through reaction coil, 1681Direction of flow if there is a leak: in the condenser, 1981

Mass balance, on acetic anhydride produced, 2104

375


Energy balance: sink= source,On condensers: heat removed in water/brine= heat to be removed from con-densing reactant, 2454

Thermodynamics,Vapor-liquid equilibrium in KO pots between condensers, 2963

Rate,Rate and catalyst addition in reaction coils, 2670Boiling characteristics in vaporizer: nucleate or film, 2553

Equipment performance,Condensers: heat-transfer coefficients, 2515


Visit control room: control-room data: values now and from past records,Steam pressure PI 101, 2973Acetic acid flowrate FRC 201, 2590

Process operators, Please describe the operation so far, 2153Contact with on-site specialists,

Steam plant: pressure and steadiness of steam supplied to battery limits,1607Steam plant: condensate return header, 1927Source of acetic acid; specs agree with those used in the design, 229Cooling water to condensers, 56

Visit site, read present values, observe and sense.Inspect the tab on the orifice plate to determine if the plate is in backwardsand to note the size of the orifice, 370Does condensate discharge into the top of the condensate header, 931Variation in the pressure and flowrate of the fuel to the cracking furnace, 768Upstream and downstream distance from the orifice plate to see if distancemeet usual design specs, 598Duration of the cycle of the variation in production, 1164


Glove test on the steam trap to estimate the temperature of the trap upstreamand downstream, 1706Stethoscope on trap to listen is discharge of condensate, 2203Tap the side of the evaporator to try to discern the level of acid in the vaporiz-er, 2057

Consistency checks,Pressure on steam header on-site and the pressure at the steam drum, 2402

Sensors: calibrate, LIC on vaporizer, 855FRC 201, 1256Pressure in acetic acid vapor line PI 201, 1872Steam pressure to coil PI 101, 2220

Control system,Controller on manual and check the FRC 201, 2604

376


Controller on manual and read PI 201, 2933Controller on manual and check for steadiness in the downstream produc-tion, 560

Call to vendors, licensee, Vacuum pump, 2905More complicated tests,

Gamma scan on vaporizer to locate level relative to the coil, 715Open and inspect,

Evaporator and look for fouling, 2493Isolate coil in vaporizer and pressure test for leaks, 2148Orifice plate FRC 101 to see if it is put in backwards, 2702Isolate, inspect cracking coils for carbon formation inside the tubes, 2548

. Take “corrective” action,Retune the control system on the vaporizer, 253Replace the bucket trap with a float trap, 889

Case’48: The column that just wouldn’t work (courtesy of T. E. Marlin, ChemicalEngineering, McMaster University) [8, sequence of distillation columns withauxiliaries; depropanizer-debutanizer]The process is the depropanizer-debutanizer described in Case’8, Chapter 2. AP&ID is given in Figure 2.4 accompanying Case’8.

This is the startup of the unit after the annual turnaround. During the turn-around several valves were replaced, pumps were dismantled and reassembled, thecolumn internals were inspected, the heat exchangers, condensers and reboilerscleaned and the sensors calibrated. This was done for both the debutanizer, C-9, andthe depropanizer, C-8.

Shortly after the unit starts up the operator is in a panic because, according to thelaboratory analysis, the depropanizer overhead product has far too much C4 and thedebutanizer overhead has far too much C3. The plant is losing $1000s per hour andthe plant manager is furious. Fix the problem.

Case’48: The column that just wouldn’t work

. MSDS, 1495




. Why? Why? Why?, Best goal? To meet specs on overhead and bottoms, 1366


377



Was the vortex breaker in the reflux drum V-30 in “good” shape, 24




Vendor files: Condenser, reboiler and preheater, 2915Reflux pump, F-27, 1755Steam traps, 1535

Commissioning data, P&ID, internal reports, 1863Handbook and Google, Cox charts, 1162Trouble-shooting files, 664


Visit control room: control-room data,Feed to the C8, depropanizer. FC/1, 1820Reflux flowrate, FIC/4, 1712Overhead product flowrate of propane, FIC/5, 1628Pressure drop Dp I/1, 2309Level bottoms LIC/2, 2026Level in reflux drum: LIC/3, 2705Steam to reboiler, FIC/3, 2760Temperature of feed into the column, TIC/2, 2974Analyzer A-1; concentration of C3 in bottoms of C8 and feed to C9, 1573Temperature bottoms TI/4, 1515Temperature mid-column TIC/5, 1077Temperature top, TI/3, 1374Pressure on overhead drum, PIC/10, 673

Process operators, This startup shift, 943Any change to the feedstock to this unit?, 974

Operating procedures, Column temperature too high; then increase the reflux,321

Call to others on-site, Utilities: is the steam-production rate and steam tempera-ture and pressure the usual values?, 123

Utilities: is the cooling-water flowrate and temperature to our unit what weexpect?, 39Operators of plants downstream receiving the butane, propane and pentane,2731Operators of upstream plant providing feed, 1732

378


Visit site, read present values, observe and sense.Column pressure, PI-4, 412Pressure relief to flare PSV-1, 2425Look at the flare, 882Temperature mid-column TI-8, 2183Valve position for column feed, FV-1, 1766Valve position for reflux, FV-4, 1913Valve position for propane overhead product to downstream processing,FIC-5, 1695Pressure on the reflux pump exit, F-27 as shown on PI-11, 2277Is there noise of cavitation around reflux pump F-27, 2127Are the leads to the motor drive on reflux pump F-27 correct so that themotor turns in the “correct” direction, 2925Check that the direction of rotation of the pump shaft is correct, 2593Valve position for steam to preheater E-24, 2138Valve-stem position on PV-10, 2366Level in feed drum, V-29; LI-1, 2106Isolation valves around the reflux control valve, FIC/4, 1655Isolation valves around the reflux pump, F-27, 1747


Open bypass around the reflux control valve, FIC/4; note any change in over-head temperature TI-3 and flow of reflux FIC/4, 1217


Drum V-29 to pumps, F-25, 26, 529Dp across pump, F-25, 26 converted to head, 96Pump F-25–26 exit to feed location, 463Drum V-30, pump F-27 and reflux into column, 164Pump F-27, 868Vapor from top of column to vapor space in V-30, 582Thermosyphon reboiler process fluid side, 1405

Mass balance, Over column, 1287Energy balance:

Heat load for condenser. Heat load condensed=heat load picked up in cool-ing water, 1015Steam load to reboiler based on 1 kg steam boils 5 kg organic, 1088Water flowrate to overhead condenser consistent with steam flowrate at bot-toms. 15 L water / kg steam, 1426With data from hand-held voltmeter, clamp-on ammeter and power-factormeters calculate the power used by pumps F-25,–26 and compare with specs,890

Sensors: check response to change,Temperature at the top of column C8: TI 3, 572Analyzer A-1, 76

379


Flowmeter for reflux, FE-4, 375Pressure at the top of column C8, PI 4, 747

Sensors: calibrate,Temperature at the top of column C8: TI 3, 1690Analyzer A-1, 2257Flowmeter for reflux, FE-4, 2270Pressure at the top of column C8, PI 4, 2312

Control systems,Output signal from controller to reflux valve FIC/ 4, 1246Put reflux control system, FIV/4, on manual; increase the flowrate, 1795

Samples and measurements,Sample feed and analyze composition and compare with usual, 2158Sample overhead concentration in vapor from column every 10 minutes for 1hour. Analyze for composition, 2196Sample bottoms concentration from the tower every 10 minutes for 1 hour.Analyze for composition, 2088

. More ambitious tests, Gamma scan on column to look for collapsed tray(s),2730

. Open and inspect,Condensers; E-25, check baffles and for fouling, 2694Access hole to column and check if trays are level and sealed; for those traysthat can be seen from access hole, 2855Reflux pump: check for damage to impeller, wear rings are not worn, F-27,2891Reflux control valve: check stem, trim, FV/4, 2803

. Take “corrective” action,Shut down the column; take out all the trays and add weirs to “seal” the self-sealed downcomers, 2284Replace the check valve on the exit line of the reflux pump, F-27, 2349Replace the control valve on the reflux, FIC/4, 2039Reduce feedrate to the column to 1/2, 1647Put on safe hold, 1604

Case’49: The case of the faulty stretcher pedal (from D.R. Winter, UniversalGravo-plast, Toronto, 2004) [Difficulty 8; involves feed bin, molding machine,mold and mold design. Context: injection molding of thermoplastics]The client manufactures hospital stretchers, sometimes called gurneys. The plasticcomponent the client wishes molded is a foot pedal that would stop the stretcherfrom rolling. The peal is 24 cm long with a central hub about which the pedal func-tions like a teeter-totter so that the pedal can be pushed from either direction. Thefoot part of the pedal is 6.5 cm across and ribbed. The pedal is 2.6 cm thick withwebs of 3 mm thickness. The central hub is 3.2 cm in diameter with wall thick-nesses of 1 cm. Ribs support the under part of the pedal because the torque created

380


is substantial. A metal core was used in the mold to create the hole for the axle, overwhich the pedal was fitted.

The resin selected was an alloy of polycarbonate and ABS that provides an impactstrength of 640 J/m. A mold was designed, tests were done and the customerapproved the prototype. Production was started.

After several months in the field, some pedals were breaking in Hong Kong, Cali-fornia, British Columbia. The breaks seemed to be independent of the hospital loca-tion. An analysis of the broken pedals suggested that the breakage occurred at thehub; inspection showed weld or knits lines. Solve the problem! Figure 8-28 is asketch of the pedal.

Figure 8-28 The pedal.

Case’49: The case of the faulty stretcher pedal

. MSDS,300



. More about the product and the mold, 2093






Injection molding machine, 2473Mold, 2704

381


Feed for this product, 1862Commissioning data, P&ID, internal reports, Prototype development, 1522Handbook, Data sheets for resin, 1691Trouble-shooting files, 2118


Visit control room: control-room data: values now and from past records,Fill cycle, 2314Cool cycle, 2892Open cycle, 2629Total cycle, 384Feed temperature into mould, 721Injection pressure, 904Extruder: rear-barrel temperature, 1076Extruder: head temperature, 1419Shot to cylinder size, 1939Concentration of foamer, 1622Backpressure, 1980Mold temperature, 2447

Process operators, 2007Operating procedures, Shutdown procedures used, 2591

Cleaning procedures used, 2673Were the feed materials from the same batches of resin, coloring agent andblowing agent as was used for the prototype?, 149


On-site simple tests:Increase injection speed by 10%, 623Decrease injection speed by 10%, 873Change injection temperature, increase to 243 �C and keep the cooling timethe same, 990Change feed temperature, increase to 243 �C and increase the cooling time,1022Clean the mold and extruder before the run by sending through a charge ofacrylic. Then use standard conditions, 1308Increase the foamer from 0.5 to 2.1% and maintain the same conditionsotherwise, 1479Increase the foamer from 0.5 to 0.8% and maintain the same conditionsotherwise, 1608Change mold to 1 feed gate near the hub with a fill cycle of 8 s; all the restthe same, 1741Increase the mold temperature to 65 �C, 1965Increase the mold temperature to 90 �C, 2267Increase the cold water flow to the mold, 2053Decrease the cold water flow to the mold, 2693

382


Increase the backpressure from 400 to 500 kPa, 2556Change the foamer from an exothermic to an endothermic type, 95Increase rpm from 160 rpm to 190 rpm, 373Use the same mold, same resin, same operator but move from a differentmolding machine, 782Use the same mold, same resin, same machine, same conditions but differ-ent operator, 965Use the same mold, same operator, same machine, same conditions butresin from a different supplier, 1159Use the same mold, same resin, same machine, same conditions, sameoperator but delete the coloring agent, 1126Separately dry the resin pellets just before using and maintain the same con-ditions, 1754

Gather data for key calculations,Calculations of the heat loss through the walls for various parts of the mold,160

Redesign of product and mold, One feed gate instead of two, 653Longer plastic neck around the axle from 2.5 to 3.5 cm, 807Realignment of axle so that plastic over the axle is centered over the pressureon the pedal, 1388

Sensors: check response to changeTemperature sensor at nozzle, 1049Temperature sensor at first stage, 1726

Sensors: use of temporary instruments,Use a pyrometer or laser sensor to measure melt temperature and comparewith temperature sensor, 1597

Call to vendors, licensee, or suppliers,Properties of feed polycarbonate ABS blend: 2249Properties of foamer, 2784Properties of coloring agent, 2172

Samples and measurements,Measure moisture in resin, 1783

More complicated tests, Change the mold from a aluminum to P-40, 1557

Case’50: The cleanup column (courtesy W. F. Taylor, B. Eng. 1966, McMaster Univer-sity) [9, vacuum distillation column plus auxiliaries; ammonia]The solvent sulfinol is stripped of unwanted degradation product heavies in a solventrecovery column. The solvent is then recycled for reuse. The column, 0.75 m diam �10.5 m high, treats a small sidestream of organic solvent to remove impurities asthe bottoms. The feed flowrate is 0.06 to 1. L/s. Feed enters at tray’5 and the recov-ered solvent is taken off at the tray above,’4, and pumped, via a positive displace-ment pump, 114-J, to the sulfinol storage tank, about 15 m away.

The top four trays act as a water wash to prevent overhead losses. The bottom tentrays do the separation. Live stripping steam is injected into the bottoms via valveHCV-13. Most of the steam eventually goes overhead. The wash water fed to tray’1

383


is sufficient to condense all the organic product and some water; the product con-tamination is about 20 to 50% water. However, most of the overhead stream goingto the vacuum equipment is water. The column is shown in Figure 8-29.

P

100

TR

100

T

101

P

201

T

201

PI

101

Toatmos.

200#

steam

200#

steam CWCW

CW

Product

FIC

FEE

D

114-J115-J

Product

to

Storage From Storage

To Storage

DrumDisposal

ofBottoms

HCV-13

50# SpargeSteam

Condensate

200# Steam

138-C

103E

14

5

4

Reflux Water

FIC-64

Figure 8-29 The vacuum distillation unit for Cases’50 and 51.

The vapor pressure of the steam is about 100 times that of the organic product atthe tower conditions. The vapor pressure of the organic product, in turn, is about 10times that of the bottoms. The tower operates under vacuum developed by a two-

384


stage steam ejector system. This particular column is operated for a day once a weekso that production downtime costs are not a significant contribution to any prob-lems on this plant.

The problem:The tower has never worked properly. Overhead losses of organic into the hot wellare probably high, but they have not been monitored. Operating conditions arerarely stable. No standard operating procedure has been developed. The bottomsproduct typically contains 10% organic solvent, whereas the column was designed toproduce a bottoms concentration of < 2% sulfinol.

But today is particularly frustrating. It was a cold and blustery night and productpump 114-J is just not delivering. The amount of “overhead product” being pumpedto storage is almost zip!

Case’50: Cleanup column

. MSDS, 719









Column, 1122Feed pumps system, 1205Product pump system, 114-J, 1305Vacuum system, 1354

Vendor files: Pump 114-J, 1486Commissioning data, P&ID, internal reports, 1964Handbook, Sulfinol composition, 1900

Expected density of liquid product, 1750Steam tables, 1650


385



Rate,Use DT to estimate whether boiling in reboiler is nucleate or film, 2908

Equipment performance, insufficient data available to do calculations


Visit control room: control-room data: values now and from past records,Pressure of sparge steam P/ 201, 2605Temperature of sparge steam, T/201, 2230Temperature of water in the barometric leg, 1850Temperature of the overhead TR/ 100, 1950Pressure at the top of the column PI/100, 1674Pressure in the overhead line to the ejectors. PI/ 101, 847Flowrate of feed to column PI/101, 946

Process operators,When the process was shut down the last time, was the pump 114-J drainedto prevent possible freezing in the idle pump, 695Please describe what has been done so far, 650

Contact with on-site specialists,Utilities: steam pressure and flow to unit stable, 132Utilities: cooling water to reflux cooler and to the barometric condenser: tem-perature and flow, 419Utilities: reflux water; temperature, flowrate and composition, 1973

Visit site, read present values, observe and sense.Is the steam tracing on, 2374Compare speed of pump 114-J with specs, 2475Compare pump stroke on 114-J with specs, 2868Pressure drop across trays 1 to 14, 797Pressure drop across trays 1 to 4, 2517Pressure drop across trays 4 to 14, 1393Flowrate of reflux water, FIC/64, 2030Temperature at the bottom of the column, 2413Temperature of the feed to the column, 1612Listen to pump 114-J for sounds of cavitation, 1854Read level glass on the sump at tray 4, 1525


Close bypass valve on 114-J, 1443Shut-off steam tracing, wait for h and note performance, 1039Block off pump 114-J and open bypass direct to storage. Observe liquid levelon tray 4 and the “bulls eye” on the line to product storage, 761Open suction line on 114-J; drain, 524Reduce the flowrate of reflux water, check bottoms temperature and sample/analyze bottoms, 211

386


Install a pressure gauge on the suction of 114-J and note readings under pres-ent, usual operation and compare with past, 393Pressure at top of the column; compare P/100 with PI/101, 282


Calculated NPSH supplied to pump 114 J, 2141Pressure profile from tray 4 to storage, 2281

Mass balance,On sulfinol over the still, 2802

Samples and measurements,Measure flowrate of water from the hot well and compare with design value,2895Sample water from the hot well and analyze concentration of organic, 2691

More complicated tests,Stop feed to column. Maintain vacuum conditions. Unhook suction linefrom pump 114-J; hook up water line to the suction line and blow water back-ward up the suction line and watch the level via the sight glass on tray 4,2785Install a pressure gauge on the suction of 114-J, disconnect the dP and mea-sure the pressure in the column at tray 4. Hook up a portable liquid pump;start pump, measure liquid flowrate and compare the Dp measured with anestimate Dp for flowrate, 2857Gamma scan around tray’4 to locate liquid level in on tray 4 and in theproduct sump, 1420

Open and inspect,Inspect foot valve on 114-J and compare valve size with the size of suctionline. Correct as needed, 2117Check for plugs, obstructions in the suction pipe from Tray 4 to pump 114-J,2386Check the strainer on the suction line to pump 114-J for plugs and obstruc-tions, 1227Open column and look for blockage in the nozzle leaving product sump andthe condition of the vortex breaker, 221

. Take “corrective” action,Replace two valves on the steam to both ejectors, 834Replace bypass valve on 114-J, 1880Replace faulty pressure relief valve on pump 114-J, 2177

Case’51: The cleanup column revisited (courtesy W. K. Taylor, B. Eng. 1966,McMaster University) [9, vacuum distillation column plus auxiliaries; ammonia]The diagram of this unit is given in Case’50.

The solvent sulfinol is stripped of unwanted degradation product heavies in a sol-vent-recovery column. The solvent is then recycled for reuse. The column, 0.75 mdiam � 10.5 m high, treats a small sidestream of organic solvent to remove impuri-

387


ties as the bottoms. The feed flowrate is 0.06 to 1.0 L/s. Feed enters at tray’5 andthe recovered solvent is taken off at the tray above,’4, and pumped, via a positivedisplacement pump, 114-J, to the sulfinol storage tank, about 15 m away.

The top four trays act as a water wash to prevent overhead losses. The bottom tentrays do the separation. Live stripping steam is injected into the bottoms via valveHCV-13. Most of the steam eventually goes overhead. The wash water fed to tray’1is sufficient to condense all the organic product and some water; the product con-tamination is about 20 to 50% water. However, most of the overhead stream goingto the vacuum equipment is water.

The vapor pressure of the steam is about 100 times that of the organic product atthe tower conditions. The vapor pressure of the organic product, in turn, is about 10times that of the bottoms. The tower operates under vacuum developed by a two-stage steam-ejector system. This particular column is operated for a day once a weekso that production downtime costs are not a significant contribution to any prob-lems on this plant.

The problem:The tower has never worked properly. Overhead losses of organic into the hot wellare probably high, but they have not been monitored. Operating conditions arerarely stable. No standard operating procedure has been developed. The bottomsproduct typically contains 10% organic solvent, whereas the column was designed toproduce a bottoms concentration of < 2% sulfinol.

Get the bottom composition within design specs. And, by the way, the productpump 114-J is sporadic; just not delivering the amount of product we expect, consis-tently.

Case’51: More about the cleanup column

. MSDS, 719









388


Column, 1122Feed pumps system, 1205Product pump system, 114-J, 1305Vacuum system, 1354

Vendor files: Pump 114-J, 1486Commissioning data, P&ID, internal reports, 1964Handbook, Sulfinol composition, 1900

Expected density of liquid product, 1750Steam tables, 1650



Rate,Use DT to estimate whether boiling in reboiler is nucleate or film, 2908

Equipment performance, insufficient data available to do calculations.


Visit control room: control-room data: values now and from past records.Pressure of sparge steam P/201, 2605Temperature of sparge steam, T/201, 2230Temperature of water in the barometric leg, 1850Temperature of the overhead TR/100, 1950Pressure at the top of the column PI/100, 1674Pressure in the overhead line to the ejectors. PI/101, 847Flowrate of feed to column PI/101, 946

Process operators, Please describe what has been done so far, 2650Contact with on-site specialists,

Utilities: steam pressure and flow to unit stable, 132Utilities: cooling water to reflux cooler and to the barometric condenser: tem-perature and flow, 2419Utilities: reflux water; temperature, flowrate and composition, 1973

Visit site, read present values, observe and sense.Is the steam tracing on?, 1376Compare speed of pump 114-J with specs, 2475Compare pump stroke on 114-J with specs, 2868Pressure drop across trays 1 to 14, 2644Pressure drop across trays 1 to 4, 2517Pressure drop across trays 4 to 14, 2133Flowrate of reflux water, FIC/64, 2030Temperature at the bottom of the column, 2413Temperature of the feed to the column, 1612Listen to pump 114-J for sounds of cavitation, 2854Read level glass on the sump at tray 4, 2225

Check diagram and P&ID versus what’s out on the plant, 1336

389


On-site simple tests:Close bypass valve on 114-J, 1443Block off pump 114-J and open bypass direct to storage. Observe liquid levelon tray 4 and the “bulls eye” on the line to product storage, 761Open suction line on 114-J; drain, 524Reduce the flowrate of reflux water, check bottoms temperature and sample/analyze bottoms, 2221Install a pressure gauge on the suction of 114-J and note readings under pres-ent, usual operation and compare with past, 393Pressure at top of the column; compare P/100 with PI/101, 282


Calculated NPSH supplied to pump 114 J, 2141Pressure profile from tray 4 to storage, 2281

Mass balance,On sulfinol over the still, 2802

Samples and measurementsMeasure flowrate of water from the hot well and compare with design value,2895Sample water from the hot well and analyze concentration of organic, 2691

More complicated tests,Stop feed and reflux to column. Maintain vacuum conditions. Unhook suc-tion line from pump 114-J; hook up water line to the suction line and blowwater backward up the suction line and watch the level via the sight glass ontray 4. When it reaches the top of the sight glass, stop the flow and observethe level over the next 10 minutes, 785Gamma scan around tray’4 and in the stripping section to locate liquidlevel in on tray 4 and possible tray collapsed, 404

Open and inspect,Inspect foot valve on 114-J and compare valve size with the size of suctionline. Correct as needed, 2117Check for plugs, obstructions in the suction pipe from Tray 4 to pump 114-J,2386Check the strainer on the suction line to pump 114-J for plugs and obstruc-tions, 1227Open column and look at the product sump and the trays in the strippingsection, 1120

. Take “corrective” action,Replace two valves on the steam to both ejectors, 834Replace bypass valve on 114-J, 1880Replace faulty pressure relief valve on pump 114-J, 2177

390


Case’52: The case of the swinging loops (courtesy of W. K. Taylor, B. Eng. McMaster,1966) [9, reactor, compressor, separator; ammonia]Ammonia is produced on two interconnected reactor loops for the production ofammonia are given in Figure 8-11, the figure in Case’29. Feed gas consists ofhydrogen and nitrogen in the proper 3:1 ratio with about 1% methane as an inert.In this ammonia synthesis reaction about 10% conversion occurs per pass throughthe reactor. Feed gas is compressed to 34.5 MPa abs and fed to a common headerthat feeds two reactor loops. Liquid product is condensed and removed from thesystem; gas is recycled back to the loop via the recycle stage compression. The reac-tor operates at 500 �C. There is an internal gas-gas heat exchanger within the reactor.The DTdue to exothermic reaction is about 50 �C.

Each compressor is a multistage reciprocating constant-speed machine rated atabout 3000 kW. Bypass valve B is operated to control the Dp across the recycle stagethat must not exceed 3.5 MPa. Opening valve B lowers the Dp and the flow of recyclegas to the loop. The recycle flow is about five times the flow of fresh feed.

Bypass valve A is operated to trim the loop: closing this valve forces more gas overto the reactor; opening the valve causes gas to bypass the loop. Valve A is used tocontrol the reaction temperature. If too much gas is fed to the reactor and the cata-lyst is inactive, the high flow might extinguish the reaction. Similarly, if the flow tothe reactor is too low, the reaction will go further because of the longer reactiontime; the reactor will overhead because there is not enough flow to carry away theheat of reaction. Normally valve A is open slightly during plant operation.

Methane is an inert coming in with the feed. The methane concentration is keptabout 15% in the loop gas to the reactors by maintaining a small purge.

The pressures, levels in the separators and the temperature profile in the reactorare shown in the control room.

The design provides operating flexibility. If one compressor breaks down theother machine can feed both loops thus keeping the reactors at operating tempera-tures while repairs are done. This avoids costly startup expense. Isolation blockvalves on the compressors are not shown on the figure. Furthermore, both loops areequalized in pressure thus evening out any slight variations introduced by the com-pressors.

The problem:The plant has been in operation with everything apparently running smoothly. Pro-duction rates, however, are only 80% of design. When the rates are increased byincreasing the front-end feedrate and closing the compressor kickbacks, then thewhole process becomes very hard to control. Operating logs report the following “...all the gas goes to the North loop for a while and then it swings and all the gas goesto the South loop ...“ “... the reactor is overheated ... and the loops flip flop againfrom South to North ... or sometimes from North to South. Usually it starts with theSouth”. You are brought in to sort out what the operators call “The Swinging Loops”.This costs us $40 000/ day.

391


Case’52: The swinging loops

. MSDS, 186








Reactor, 608Internal heat exchanger, 903Reciprocating compressor, 153Refrigeration system, 335Condensers: refrigerant, 1928Condensers: water, 2929Gas-liquid separators, 2278Valves A and B, 1756

Vendor files: Reciprocating compressors, 1907Refrigeration system, 1353

Commissioning data, P&ID, internal reports, 1158Handbook, Thermal properties of hydrogen and ammonia, 1053Trouble-shooting files, 1467


Pressure profile,Direction of leak: condensers with CW, 107Direction of leak: condensers with refrigerant, 64Around the loop and fresh feed into the loop, 260

Mass balance,Bleed to maintain the inerts in the recycle at 15%, 406

Thermodynamics, 489Equipment performance,

Controlling any hot spots, 217Maximum temperature before the catalyst is damaged, 1309

392



Visit control room: control-room data: values now and from past records.Valve A in South loop, 1024Valve A in North loop, 1074Valve B in South loop, 1113Valve B in North loop, 1208Temperatures in bed in South reactor, 1414Temperatures in bed in North reactor, 1496Production of liquid ammonia South loop, FR, 2417Production of liquid ammonia North loop, FR, 2258Dp across catalyst bed South reactor, 2113Dp across catalyst bed North reactor, 2060Hydrogen concentration in feed gas to the reactor: South loop, 2021Hydrogen concentration in feed gas to the reactor: North loop, 2156Temperature at the exit of the cooling water exchanger, South loop, 2804Temperature at the exit of the cooling water exchanger, North loop, 2607Temperature at the exit of the refrigeration exchanger, South loop, 2778Temperature at the exit of the refrigeration exchanger, North loop, 2958Calrod startup heaters to both South and North loops, 2333

Visit control room: control-room data:As soon as one of the loops starts to swing record the values and note the“usual values”. Here are the data when the South loop swings taken 4 minafter the swinging was first noted by the operators, Valve A in South loop,2354Valve A in North loop, 1852Valve B in South loop, 1982Valve B in North loop, 1845Temperatures in bed in South reactor, 1602Temperatures in bed in North reactor, 920Production of liquid ammonia South loop, FR, 683Production of liquid ammonia North loop, FR, 556Dp across catalyst bed South reactor, 131Dp across catalyst bed North reactor, 369Hydrogen concentration in feed gas to the reactor: South loop, 289Hydrogen concentration in feed gas to the reactor: North loop, 1824Calrod startup heaters to both South and North loops, 1884

Process operators,What is the frequency of the swinging loops, 2318What is the cycle time of the swinging loops, 2392What do you do when you first notice a hot spot, 2872Can you predict when one loop will start swinging, 2903What do you think is the most noteworthy observation when swinging isoccurring, 2623

393


Is it always the South loop that gets the hot spot that seems to trigger theswinging loops or does it happen with equal frequency for the North loop,1620

Operating procedures,What to do if there is a hot spot, 1657New procedure; if reactor “hot spot” is 10 to 20 �C; immediately close valve Ain that loop, 1181

Contact with on-site specialists,Supervisor for construction crew for the North loop and the tie-ins: precau-tions that were taken to ensure that lubricating oil and or water was not leftin any of the lines especially the tie-in lines, 706Any possible preservative oil or residual valve stem lubricant left on any ofthe tie-in lines/valves for the recent construction, 806

Visit site, read present values, observe and sense.Change set point on valve A, South and North loops, and observe whetherthe valve stem moves, 988Change set point on valve B, South and North loops and observe whether thevalve stem moves, 80Observe the “kickback” behavior of the compressor on the South loop, 206Observe the “kickback” behavior of the compressor on the North loop, 655Observe-listen for surge in compressors on South loop, 887Observe-listen for surge in compressors on North loop, 800

Check diagram and the P&ID versus what’s actually out on the plant, 1216On-site simple tests:

Record evidence and actions over a three hour period whenever the swingingloop behavior “just starts”, 1384

Consistency checks,For temperature sensors in catalyst bed; do they make sense relative to eachother for Loop South reactor, 1145For temperature sensors in catalyst bed; do they make sense relative to eachother for Loop North reactor, 1725

Trend checks,Cycle time: duration and frequency, 1268

Sensors: check response to changeTemperatures North and South loop, 720Ammonia production flowrate: North and South loop, 580Level on liquid receivers: North and South loop, 1059Temperatures on exit of condensers: both water and refrigerated condensers.North and South loop, 1437Hydrogen analyzer in loop circuit. North and South loop, 1714

Sensors: calibrate,Eight temperature sensors North loop reactor, 2244Eight Temperature sensors South loop reactor, 2146Ammonia production flowrate: North and South loop, 2444Level on liquid receivers: North and South loop, 1686

394


Temperatures on exit of condensers: both water and refrigerated condensers.North and South loop, 1534Hydrogen analyzer in loop circuit. North and South loop, 1239Kickback relief on North loop compressor, 1979Kickback relief on South loop compressor, 1678

Call to vendors, licensee,Supplier of the catalyst: anything to watch for that might poison the catalyst,1873Was the same batch of catalyst shipped for the new reactor in the North loopas was used in the previously commissioned South loop system, 2375

Samples and measurementsShut down the system; sample the “reduced” catalyst from both reactors andcompare the activity of each, 1295Sample the feed to South reactor when it has the hot spot; the samples are tobe every 2 minutes for the first 10 minutes of the swinging loop cycle. Ana-lyze for inerts, 1340Sample the feed to North reactor when it has the hot spot; the samples are tobe every 2 minutes for the first 10 minutes of the swinging loop cycle. Ana-lyze for inerts, 633Fresh feed to South loop; analyze for inerts. Sample once per hour for 24 hoursand once every two minutes for ten minutes during the swinging loop, 940Fresh feed to North loop; analyze for inerts. Sample once per hour for 24hours and especially during the swinging loop, 2430

More complicated tests,Isolate the North loop and run commissioning tests the same way we haddone before turnaround for the South loop alone, 2756

Open and inspect,Open the reactors and note the height of catalyst in the two reactors, 1710Valve A; South loop and look for chatter anything that might vibrate or oscil-late to cause fluctuations, 1949Valve A; North loop and look for chatter anything that might vibrate or oscil-late to cause fluctuations, 2949Valve A from both North and South loops and pressure test for leak acrossthe butterfly valves, 350Valve B from both North and South loops and pressure test for leak acrossthe butterfly valves, 100

. Take “corrective” action,Increase the bleed rate in the South loop to decrease the inert concentrationin the recycle from 15% to 14%, 867Replace valve on purge line in the South loop, 1867Replace the butterfly valves (used for valves A and B on both North andSouth loops) with full size globe valves, 2865Replace the temperature sensors in the South loop reactor, 2737Replace kickback valve system on the South loop reactor, 763

395


8.3Summary

I hope you enjoyed your journey to improve your skill and confidence as a troubleshooter. I hope you had a chance to use the triad method; this provides a rich experi-ence for you to see how others handle a situation, to spend the time to really under-stand a process so that you can respond quickly, as the expert system, to any questionsthe trouble shooter might pose.

Working the cases on your own is a great experience as well. I trust that the rangeof cases gave you enriched insights about processes and trouble shooting. I welcomeyour feedback on how to improve the cases and, of course, cases and answers ofcases you solved.

396

397

Here we summarize the highlights of this book so that we can define where we arenow. Then we emphasize why reflection and self-assessment are key – yet time-con-suming, challenging and frustrating – activities to develop confidence. Ideas aregiven in Section 9.3 on how to use goal setting as a neat way to develop skill beyondthat discussed in this book. Sections 9.4 and 9.5 focus on how to go beyond thisbook for knowledge about process equipment (as given in Chapter 3) and for inter-esting and challenging trouble-shooting cases (as given mainly in Chapter 8).

9.1Summary of Highlights

This book summarizes research about what is known about trouble shooting andwhat you can do to improve your skills. This is not a series of anecdotes of my trou-ble-shooting experience or how I personally solve problems. This book is flexiblydesigned to address your interests and needs. This book is filled with activities!

. Five key skill areas needed by trouble shooters: The five areas of skills neededin trouble shooting are 1) problem solving, 2) knowledge of process equip-ment, 3) process safety and properties of materials, 4) “systems thinking”,and 5) people skills. Although research has shown that knowledge of processequipment (common faults, typical symptoms) is of vital importance in deter-mining success in trouble shooting, equal emphasis is given in this book todevelop your skill in the other areas and to give you practice in using all theskills to trouble shoot. Section 1.3 gives an inventory for you to reflect on andrate your skills in these five areas. Five cases, Cases’3–7, follow in Section1.6 on which you can try out your skills.

Where were your five strengths?What were the areas you wanted to work on?How did you rate your skill after you had worked on some of the five cases in Sec-tion 1.6?

9

What Next?


9 What Next?

. In problem solving: the four key components of the process are: 1) the overallproblem-solving process as characterized by frequent monitoring, emphasison checking and double checking, being organized and systematic and keep-ing the problem in perspective. 2) the elements of data handling and criticalthinking: data gathering and resolution, based on fundamentals, with validreasoning and being complete. 3) how well we synthesize and put all theideas together: having five to seven working hypotheses and being flexible.4) decision-making: based on criteria, priorities and avoiding bias. This pro-vides, in Table 2-1, target skills for us to emulate. To help quantitatively seeprogress in skill development an evaluation form (Worksheet 2-2) and a Trou-ble-Shooter’s Worksheet (Worksheets 2-1 and 2-4) are given. The exampleuse of the Trouble-Shooter’s Worksheet was illustrated for Case’8: “thedepropanizer: the temperatures go crazy”.

The target skills are described in Table 2-1. Reconsider your rating in Chapter 1based on your better understanding of these problem-solving skills.Did the Trouble-Shooter’s Worksheet help you? If not, develop your own guide tohelp you to be systematic.

. Knowledge of process equipment: symptoms–probable causes. Scatteredthrough the literature and one’s experience are details of symptoms (andcause) from malfunctioning processes. Sometimes the information is con-founded because the cause is not the root cause. Chapter 3 attempts to orga-nize this information for ease in use with cross referencing for systems ofequipment. Consistent SI units of measurement, terminology and organiza-tion are used, following the book Process Design and Engineering Practice.

How does this collection of practical details match yours?How might you combine this collection with yours?How might you expand and keep it up-to-date?

. Want feedback about trouble shooting? ... then Chapter 4 gives story timedescribing the adventures of five engineers as they trouble shoot problemsthat range in complexity. But these aren’t your usual stories. The story isinterrupted. Each of the five scripts consists of about three parts with eachpart concluding with a few questions for you to consider. This question breakwas introduced to give you a chance to reflect on how you would havehandled the case, and to decide what you should do next. At the end of eachcase an assessment is given of the problem-solving processes used by each ofthe trouble shooters.

How was your approach similar to, different from Michelle, Pierre, Dave, Saadiaor Frank?How does your rating of their approach agree with my rating of their approach?How well could you describe the approach you took?How would you rate your approach?

398

9.1 Summary of Highlights

. If you want to improve your confidence and skill in problem solving ... thenChapter 5 provides activities, with feedback, to develop skills and confidencein– awareness or the ability to describe your problem-solving processes;– strategies or the ability to see patterns in the process;– exploring the context using theWhy? Why? Why? process;– being creative;– being skilled in performance assessment.

For each, target skills were listed, some examples were provided, activities todevelop the skills were described and forms of evidence were given. Cases’9and’10 (the bleaching plant and to dry and not to dry) were introduced.

What problem solving strategy do you use? How does it compare with the MPS 6-step?Is the Why? Why? Why? approach effective for you?When are the best times to apply it?For creativity, are you willing to spend the extra time to write down a wide range ofridiculous ideas?For creativity, what triggers work best for you?

. If you want to improve your confidence and skill in data collection and criticalthinking ... then Chapter 6 provides activities, with feedback, to develop skillsand confidence in– gathering information, selecting diagnostic actions, testing hypotheses/

causes. Criteria are given for selecting tests. Biases and mistakes made ingathering and interpreting data are confirmation bias, over-interpretation,under-interpretation and mis-interpretation, availability bias, prematureclosure, anchoring. The Jungian typology dimension P-J provides a usefulindicator of one characteristic of your personal style for gathering andinterpreting data.

– being consistent in our use of words, and our use of knowledge aboutprocesses and process equipment. Specifically the focus was on identify-ing fact versus opinion, gathering accurate cause–symptom data andastutely reversing the connection to link symptom–cause. Our usageshould be consistent with the rules of English, mathematics, the funda-mentals of science and engineering and practical experience.

– classification; the process of dividing large sets of information into mean-ingful parts. In making the division we should use a single basis/criteriaper level, use no single entries and avoid faulty coordination or subordi-nation. This was illustrated for the task of classifying the starting infor-mation into “symptoms” and “triggering” events.

– identifying patterns.– reasoning. A systematic 9-step process is suggested and illustrated in the

context of part of Michelle’s reasoning in Case’3. Case’11 (the lazytwin) was introduced.

399

9 What Next?

Did the list of actions given in Section 6.1 agree with yours?What types of actions do you prefer to select when you trouble shoot?What is your style in matching hypotheses with symptoms?What methods do you use to select tests that will correctly test hypotheses?Did the two inventories about bias provide useful insight? If not, what else mightyou do to identify your preference and style?How easy was it for you to separate facts from opinions?How easy was it matching hypotheses consistent with symptoms?What special techniques do you use to identify “symptoms”?What methods do you use to spot patterns in data?Did the diagramming of an argument help you?

. If you want to improve your confidence and skills in interpersonal skills ...then visit Chapter 7 that provides activities, with feedback, to develop skillsand confidence in:– communicating,– listening,– applying the basic fundamental principles of interpersonal relationships,– building trust and– understanding our own uniqueness and the uniqueness of others.– Factors that affect our performance and the performance of others

include:– unwillingness to admit having made a mistake,– stress,– level of frustration and lack of motivation.– preference to follow their own approach even though it may result in

incorrect and even unsafe operation of the process.– preference to infer and interpret what we see or hear.

A questionnaire is included to give us a chance to evaluate the environmentin which we trouble shoot. Cases’12–14 (the drop boxes, the lousy controlsystem, the condenser that was just too big!) were introduced. Cases’15–18were included as exercises.

Which activities were most useful to you?How strong is your tendency to infer?How does your environment rate?What was the best new idea you learned from this Chapter?

. If you want to improve your skill and confidence in trouble shooting ... thenwork the cases in Chapter 8, reflect on the process you used, check on thenumber of target skills you displayed and set goals for improvement. You canexperience the cases as triads or as an individual. If you get the chance selectthe triad experience! The cases have been carefully selected to provide varyinglevels of difficulty, different types of situations (from startup to usual opera-tions) and different varieties of equipment. Start looking at cases at Level 3and 4. The lower ratings start with Case’19.

400

9.2 Reflection and Self-Assessment are Vital for the Development of Confidence

Which case did you enjoy the most?From which case did you learn the most?When you experienced the triad activity, which role did you prefer?What was the biggest challenge in reflecting on and evaluating the approach youused?Did you complete a Trouble-Shooter’s Worksheet when you started? If not, whynot?

This book was designed as a challenging and enjoyable educational experience.

9.2Reflection and Self-Assessment are Vital for the Development of Confidence

Research has shown that the quality of the answer to a problem and the problem-solving skill and confidence improves if we:

. pause during the process and write down reflections of what you have doneso far and where you are going next.

. have clear goals describing the skill, have measurable criteria about how youwill know when you have the skill and have opportunities to collect evidence.

The design for self-assessment and confidence building is illustrated in Chapter 5.

. The target skills are listed. These are the proven behaviors of successful trou-ble shooters. Write these in terms that can be observed. Try to remove anyambiguity. Include some quantitative criterion to measure achievement ofthe target skill. This was done, for example in Sections 5.1.1, 5.2.1, 5.4.1. In thecontext of the overall process of trouble shooting, this was presented inTable 2.1.

. An activity is posed. The activity gives you a chance to display the behavior.In Chapter 5 the tasks were described in Sections 5.1.2, 5.2.2, with the materials tobe used given in Sections 5.1.3, 5.2.3. For the overall process of trouble shoot-ing, the overall process was illustrated first through the completed Trouble-Shooter’s Worksheets (for Case’8 in Chapter 2, and the Cases’2–4 inAppendix C) and through the examples given in Chapter 4 of Michelle,Pierre, Dave, Saadia and Frank. For your approach, two options were pre-sented in Section 8.1: the triad and the individual activity. Cases were scat-tered through the various chapters with a list of options given in Section 8.1.

. Feedback forms and forms of evidence are created. In Chapter 5 the forms ofevidence were listed in Sections 5.1.4, 5.2.4. For the overall trouble-shooting pro-cess, the forms of evidence will include: the problem statement (and themarks, underlines and notations you made directly on this); your troubleshooter’s notebook or worksheets that you used; the pauses and reflectionsyou wrote as you worked through the case, the Trouble-Shooter’s Worksheet(and especially the hypotheses–symptoms chart), the written requests forinformation (if you worked in triads) or the sequence of codes for the ques-

401

9 What Next?

tions (if you worked as an individual). The most important form of evidenceis Worksheet 2-2. This is significant because it relates directly to the targetbehaviors of successful trouble shooters, noted in Table 2.1. Use this to giveyourself-feedback about the process you used.

9.3Going Beyond this Book: Setting Goals for Improvement

Prepare yourself for success and use reflection and self-assessment effectively.

9.3.1Prepare Yourself for Success

. Continually update your knowledge of process equipment. Research hasshown that the key is a broad knowledge of process equipment. Continuallydraw on your experience and the experience of colleagues to enrich thatexperience.

. Praise yourself for where you are now. Too often we focus only on the nega-tive and things we can’t do well. Throughout this book you have noted thatthe feedback was always five strengths and two areas to work on. At thistime, write down your five strengths1. ______________________________________________________________2. ______________________________________________________________3. ______________________________________________________________4. ______________________________________________________________5. ______________________________________________________________

. Set achievable goals. These should be expressed in terms of the target skillsof successful trouble shooters, described in Table 2.1. These should be consis-tent with the two areas you want to work on. You might identify your goalsas:1. to maintain the five strengths noted above andto shift one of the following “areas to work on” to a strength within the nextyear.My two areas to work on are:1. ______________________________________________________________2. ______________________________________________________________

. Arm yourself for success.In Section 6.1.2b I listed the stuff I had available in my office related to trou-ble shooting. A notebook, paper or hand-held electronic, is essential. Gloves,tape measure and string and always good additions. The first time you appearwith a stethoscope will require courage! But after you have solved a bunch oftricky problems because of the stethoscope, others will want one too.

402

9.5 Beyond this Book: Sources of Other Cases

9.3.2Use Reflection and Self-Assessment Effectively

Follow the principles of self-assessment, given in Sections 5.5 and 9.2. Assessmentor performance review is not a dirty word! Conscientious self-assessment is theguide to growth.

9.4Going Beyond this Book: Updating your Rules of Thumb and Symptom ‹ Cause Datafor Process Equipment

Rules of thumb are generalized, usually numerical, values of “usual” practice. Al-though we all generate such “experience” numbers intuitively, it helps to organizeand write these down. You might create files for different equipment, following, forexample, the titles used in Chapter 3. As you read articles in such journals as Chem-ical Engineering Progress, Chemical Engineering and Hydrocarbon Processing record therules of thumb in your paper or electronic files. In creating the files, decide on asystem of units that you will consistently use. Take the time to rework informationfrom other systems of units. Create your own set of rules of thumb for the processesand unit operations with which you work.

Extend the rules of thumb to include cause fi effect data. Check through the ven-dor files on the web or that are in the equipment files. Some sources in the literatureinclude Bloch and Geitner (1983), and McNally Institute at http://www.mcnallyinsti-tute.com (2003).

9.5Beyond this Book: Sources of Other Cases

Some computer simulation games for trouble shooting have been developed byDoig and colleagues for the SYSCHEM process (1977, 1980). Trouble-shooting casesreported in the literature tend to describe the problem, the process used to discoverthe fault and the corrective action taken. Although this format is not easy to useto help polish your skill, these cases broaden our perspective, provide morecause fi effect data and can be converted into the format used in this book to aid inskill development.

Some sources include Liberman (1985), Saletan (1994), the Riance series (1983 ff),the articles by Henry Kister that appear in Hydrocarbon Processing or at the AIChEconferences and the Marmaduke series in Power Magazine 1950 ff with some rep-rinted by Elonka (1979).

403

405

PrefaceBranan, C (1998) “Rules of Thumb for Chemi-

cal Engineers,” 2nd edition, Gulf Publish-ing Co.

Gans, M. et al., (1983) “Plant Start-up step bystep by step,” Chem. Eng. 90, 20, Oct 3p 74.

Kister, H.Z, series of articles in HydrocarbonProcessing 1979 ff.

Lieberman, N.P. (1985) “Trouble-shooting pro-cess Operations” 2nd edition, PennWellBooks, Tulsa OK.

Saletan, D, (1994) “Creative Trouble Shootingin the Chemical Process Industries,”Chapman and Hall, London.

Chapter 1Example articles describing a personalapproach to trouble shooting in engineering:Gans, M. et al., (1983) “Plant Start-up step by

step by step,” Chem. Eng. 90, 20, Oct 3p 74.

Laird, D, B. Albert, C. Steiner and D. Little(2002) “Take a Hands-on approach torefinery troubleshooting,” Chem. Eng.Prog, June, 68–73.

Smith, K. (2002) “Refine your approach to pro-cess trouble shooting and optimization,”Hydrocarbon Processing, June, 63–66.

Taylor, W.K. (1980) “Trouble Shooting at Cana-dian Industries Limited” Chemical Engi-neering Education, Spring, p 88–89.

. Books to help improve knowledge aboutsafety and hazards.

Kletz, T.A. (1985) “What went Wrong?” GulfPublishing Co, Houston, TX.

Kletz, T.A. (1983) “Hazop and Hazan–Noteson the Identification and Assessment ofHazards,” Institution of Chemical Engi-neers, London, UK.

Woods, D.R. (1995) “Data for Process Designand Engineering Practice,” Prentice Hall.

. Example references summarizingresearch about trouble shooting:

Dubeau, C.E. et al. (1986) “Premature Conclu-sions in the Diagnosis of Iron-deficiencyAnaemia: cause and effect”, Medical Deci-sion Making, 6, 3, 169–173.

Elstein, A.S, L.S. Shulman and S.A. Sprafka(1978) “Medical Problem Solving: an anal-ysis of clinical reasoning”, Harvard Uni-versity Press, Cambridge MA.

Groen, G.J. and V.L. Patel (1985) “MedicalProblem Solving: some questionableassumptions” Medical Education 19,95–100.

Johnson-Laird, P.N. and W.C. Wasan, eds,(1977) “Thinking: Readings in CognitiveScience”, Cambridge University Press,Cambridge, MA.

Kassirer, J.P, and G.A, Gorry (1978) “ClinicalProblem Solving: A Behavioral Analysis”,Ann. Int. Medicine, 89 245–255.

Kern, L, and M. E. Doherty (1982) “Pseudo-diagnosticity” in an Idealized MedicalProblem Solving Environment”, J. MedicalEducation, 57 100–104.

McGuire (1985) “Medical Problem Solving: ACritique of the Literature”, J. Med. Educa-tion, 60, 587–595.

Nisbett, R. and L. Ross (1980) “Human Infer-ence: strategies and shortcomings ofsocial judgement”, Prentice Hall, Engle-wood Cliffs, NJ.

Literature References



Spencer, Joanne, (1988) “ Critique of Problemsolving and trouble shooting approaches”,Department of Chemical Engineering,McMaster University, NSERC summerstudent report.

Tversky, A, and D. Kahneman (1974) “Judge-ment and uncertainty: heuristics andbiases”, Science, 185, 1124–1131.

Voltovich, A.E. et al. (1985) “Premature con-clusions in diagnostic reasoning”, J. Med.Education, 60, 302–307.

Whitman, N. et al. (1986) “Problem Solving inMedical Education: Can it be Taught?”,Current Surgery, 43 453–4.

Wolf, F.M, L.D. Grappen and J.E. Billi (1985)“Differential Diagnosis and the Compet-ing-Hypothesis Heuristic”, J. AMA, 253,19, 2858–2861.

Chapter 2Covey, S.R. (1990) “Seven Habits of Highly

Effective People,” Fireside Book, SimonSchuster.

Holmes, T.H. and R.H. Rahe (1967) “TheSocial Readjustment Rating Scale,” J ofPsychosomatic Research, Aug, 213–218.

Kepner, C.H. and B.B. Tregoe (1985) “TheNew Rational Manager,” McGraw Hill,New York.

Woods, D. R. (1994) “Problem-solving skills”,Chapter 3 in “Problem Based Learning:How to Gain the Most from PBL” Woods,Waterdown, ON, Canada.

Woods, D.R. (2000) “An Evidence-based strat-egy for problem solving,” J. of Engineer-ing Education, Oct, 443–459.

Woods, D.R. et al, (1997) “Developing Prob-lem-solving skills: The McMaster ProblemSolving Program,” J of Engineering Edu-cation, 86, 2, 75–91 and http://www.che-meng.mcmaster.ca/innov1.htm and clickon MPS.

Woods, D.R. (1988) “Novice versus ExpertResearch suggests ideas for implementa-tion,” J College Science Teaching, Sept,p 77–79, 66–67; Nov, p 138–141; Dec,p 193–195.

Chapter 3Gans, M. et al., (1983) “Plant Start-up step by

step by step,” Chem. Eng. 90, 20, Oct 3p 74.

Griff, A. (1968) “Plastics Extrusion Technolo-gy”, Reinhold Book Co.

Lieberman, N.P. (1985) “Trouble-shooting Pro-cess Operations” 2nd edition, PennWellBooks, Tulsa OK.

Rauwendall, C. (1986) “Polymer Extrusion”,Hanser Publishers, Munich.

Vlachopoulos, J. et al, (2001) “The SPE Guideon extrusion technology and troubleshoot-ing,” The Society of Plastics Engineers,Brookfield, CT.

Woods, D.R. (1995) “Process Design and Engi-neering Practice, Prentice Hall.

Francis, D, and D. Young, “Improving WorkGroups: A Practical Manual for TeamBuilding,” University Associates, SanDiego, CA. (1979).

Woods, D.R, “Group skills,” Chapter 5 in“Problem-based Leaning: how to gain themost from PBL,” Woods Publisher, Water-down ON Canada distributed by McMas-ter University Bookstore, Hamilton, ON(1994).

Fisher, K, et al. (1995) “Tips for Teams,”McGraw Hill, New York.

Kirton, M.J. (1976) “Adaptors and innovators:a description and measure,” J. AppliedPsychology, 61, 622–629.

Keirsey, D. and M. Bates, “Please understandme: character and temperament types,”Gnosology Books, Del Mar, Ca (1984) andhttp://www.keirsey.com.

Schutz, W.C, “FIRO: a three-dimensional the-ory of interpersonal behavior,” Holt Rine-hart and Winston, New York, NY, 1958,with the instrument and scoring availablefrom Whetton, D.A, and K.S. Cameron,“Developing Management Skills,” ScottForseman, Glenview, IL,1984, p. 80.

Tannen, D, “You Just Don’t Understand:Women and Men in Conversation,” Bal-lantine, New York, 1990.

Johnson, D.W, “Reaching Out,” Prentice Hall,Englewood Cliffs, NJ, 1986.

Chapter 4Lieberman, N.P. (1985) “Trouble-shooting pro-

cess Operations” 2nd edition, PennWellBooks, Tulsa OK.

Kister, H.Z., Series of articles in HydrocarbonProcessing 1979 ff.

406


Gans, M. et al., (1983) “Plant Start-up step bystep by step,” Chem. Eng. 90, 20, Oct 3p 74.

Chapter 5Basadur, M. (1995) “Simplex: a flight to crea-

tivity”, Creative Education Foundation,Buffalo, N. Y.

Gadsby, R.E. and J.G. Livingstone (1977) “Cat-alysts and some incidents they have sur-vived,” CEP Ammonia Plant Safety, 20,p 70 ff.

Kimbell, R. et al. 1991 “Assessment of Perfor-mance in Design and Technology,” SEACreport, UK.

Krishnaswamy, R. and N.H. Parker (1984)Chemical Engineering, April 16, p 93–98.

Leifer, L, (1997) “Design team performance:metrics and the impact of technology,” in“Evaluating organizational training: mod-els and issues,” S.M. Brown and C. Seid-ner, eds, Kluwer Academic publishers.

Lombard, J.F. and R.A. Culberson (1972)“Defining Reformer Performance,” CEPAmmonia Plant Safety, 15, p 29–35.

Woods, D.R. (1988) “Novice versus ExpertResearch suggests ideas for implementa-tion,” J College Science Teaching, Sept,p 77–79, 66–67; Nov, p 138–141; Dec,p 193–195.

Chapter 6Branan, C. (1998) “Rules of Thumb for Chem-

ical Engineers,” 2nd edition, Gulf Publish-ing Co.

Halpern, Diane (1996) “Thought and Knowl-edge; an introduction to critical thinking”3rd edition, Lawrence Erlbaum.

Scriven, M. (1976) “Reasoning”, McGraw Hill.Walas, S.M (1988) “Chemical Process Equip-

ment,” Butterworths.Woods, D.R. (2001–3) “Rules of Thumb” John

Wiley and Sons, forthcoming.Woods, D.R. (1995) “Process Design and Engi-

neering Practice,” Prentice Hall.

Chapter 7Johnson, D.W. and F.P. Johnson (1986) “Join-

ing Together,” Prentice Hall, EnglewoodCliffs, NJ.

Jungian typology or MBTI: see Keirsey, D. andM. Bates (1984) “Please Understand Me”

Prometheus Books, Del Mar CA andhttp://www.keirsey.com.

Kirton KAI see Kirton, M. (1976) “Adaptorsand Innovators: a description and mea-sure,” J. Applied Psychology, 61, no. 5,622–629.

Kletz, T.A. (1986) “What went Wrong? Casehistories of Process Plant Disasters,” GulfPublishing Co, Houston, TX.

Powers, G.J. and S.A. Lapp (1983) “Fault TreeAnalysis,” Carnegie Mellon UniversityShort Course, Pittsburgh.

Woods, D.R. (1994) Chapter 5 in “Problembased learning: how to gain the mostfrom PBL,” Woods, Waterdown, ON,Canada.

Chapter 8Barton, J. and R. Rogers (1997) “Chemical

Reaction Hazards”, 2nd edition, Gulf Pub-lishing Co, Houston TX.

Krishnaswamy, R. and N.H. Parker (1984)“Corrective Maintenance and Perfor-mance Optimization”, Chemical Engineer-ing, April 16, p 93–98.

Yokell, S. (1983) “Trouble shooting shell andtube heat exchangers”, Chem. Eng. 90no 15, p 57–75.

Chapter 9

. Sources of symptom-cause information:

Bloch, H.P. and F.K. Geitner (1983) “PracticalMachinery Management for ProcessPlants, part 2: Machinery Failure analysisand troubleshooting,” Gulf Publishing.

. Simulation:

Doig I.D. (1977) “Training of Process PlantMalfunction Analysis,” Chemeca 77, Can-berra, 14–16 Sept, p 144–148 and (1980)“Trouble-shooting systems and experi-ences at New South Wales,” ChemicalEngineering Education, Summer 1980,p 130.

. Sources of other cases:

Drew, J.W. (1983) �Distillation ColumnStartup,” Chem. Eng. Nov 14, p 221.

Elonka, S.M. (1979) “Marmaduke Surface-blow’s Salty Technical Romances” R. Krei-

407


ger Publishing, New York. Collection ofarticles from Power Magazine.

Karassik, I.J. (1981) “Centrifugal PumpClinic,” Marcel Dekker Inc. New York.

Kister, H.Z., Series of articles in HydrocarbonProcessing 1979 ff.

Lieberman, N.P. (1985) Trouble-shooting pro-cess Operations, 2nd edition, PennWellBooks, Tulsa, OK.

Riance, X.P. (1983 ff) “Learning–the hard way”series of articles in Chemical EngineeringMagazine, in 1985: Oct 3, Oct 31, In 1985:May 13, June 10; In 1986, Feb 17. In 1987,Jan19. In 1988, Feb 15 1986.

Saletan, D, (1994) “Creative Trouble Shootingin the Chemical Process Industries,”Chapman and Hall, London.

Shah, G.C. (1979) Trouble shooting reboilersystems,” Chem. Eng. Prog., July 53–58.

Yokell, S. (1983) “Trouble shooting shell andtube heat exchangers”, Chem. Eng. 90no 15, p 57–75.

. Other articles for rules of thumb andtrouble shooting:

Eckert, J.S. (1979) “Design of Packed Distilla-tion Columns” Section 1.7 in “Handbookof Separation Techniques for ChemicalEngineers”, P.A, Schweitzer, ed, McGrawHill, New York. p 1.221–1.240.

Ellerbe, R.W. (1979) “Batch Distillation” Sec-tion 1.3 in “Handbook of Separation Tech-niques for Chemical Engineers”, P.A,Schweitzer, ed, McGraw Hill, New York.p. 1.164–1.167.

Farminer, K.W. (1988) “Defoaming” in “Ency-clopedia of Chemical Technology andDesign”, J McKetta, ed, Marcel Dekker,NY.

Gans, M. and Fitzgerald, F.A. (1966) “PlantStartup” Chapter 12 in “The ChemicalPlant” R. Landau, ed. Reinhold PublishingCo. New York.

Gardner, K.A. (1974) “Anticipation of Operat-ing Problems in the Design of Heat Trans-fer Equipment” in “Heat Exchangers:Design and Theory Sourcebook” N. Afganand E.U. Schlunder, eds, Scripta Book Co,McGraw Hill, New York.

Godard, K.E. (1973) “Gas Plant Startup Prob-lems” Hydrocarbon Process. Sept. p 151–155.

Griffith, S. and Keister, R.G. (1970) “ThisButadiene Unit Exploded” HydrocarbonProcess, 49, no. 9, p 323.

Kister, H.Z. (1979) “When Tower Startup hasProblems” Hydrocarbon Process Febp 89–94.

Kletz, T. (1979) “Learn from these HPI fires”Hydrocarbon Processing Jan p 243–250.

Kletz, T.A. (1986) “Hazop and Hazan” 2nd edi-tion, The Institution of Chemical Engi-neers, London, UK.

Kletz, T.A. (1985) “What Went Wrong? CaseHistories of Process Plant Disasters” GulfPublishing Co, Houston, TX.

Lapp, S.A. and Powers, G.J. (1977) “ComputerAided Synthesis of Fault-trees” IEEETransactions on Reliability, April p 2–13.

Lefevre, L.J. (1986) “Ion Exchange: Problemsand Troubleshooting” Chem Eng, July 7,p 73–75.

Lieberman, N.P. (1983) “Process Design forReliable Operations” Gulf Publishing Co.Houston, TX.

Lieberman, N.P. (1985) “Trouble-shooting pro-cess Operations” 2nd edition, PennWellBooks, Tulsa OK .

McLaren, D.B. and Upchurch, J.C. (1970)“Guide to Trouble-free Distillation” Chem.Eng. 77, No 12, June 1, p 139–152.

Penny, W.R. (1970) “Guide to Trouble-freeMixers”, Chem. Eng. 77, No 12, June 1,p 171–180.

Powers, G. J. and Lapp, S.A. (1983) “Fault TreeAnalysis” Carnegie Mellon UniversityShort Course, Pittsburgh.

Powers, G.J. and Lapp, S.A. (1981) “A ShortCourse on Risk and Reliability Assess-ment by Fault tree Analysis” CarnegieMellon University.

Powers, G.J. and Lapp, S.A. (1976) “Comput-er-aided Fault Tree Synthesis” Chem. Eng.Prog. April p 89–93.

Reason, J. and Mycielska, K. (1982) “AbsentMinded? The Psychology of Mental Lapsesand Everyday Errors” Prentice Hall, Engle-wood Cliffs, NJ.

Reginald, S. and Gupta, J.P. (1986) “A Note on1.2 Heat Exchanger Trouble Shooting”Int. Comm. Heat Mass Transfer, 13,p 235–243.

Riance, X.P. (1983) “Learning- the Hard Way”Chem. Eng. Oct 3, p115–116.

408


Riance, X.P. (1987) “More Learning–the HardWay: part 4, Chem. Eng. Jan 19, p 131–133.

Shah, G.C. (1978) “Trouble Shooting Distilla-tion Columns” Chem. Eng. July 31 p 70–78.

Swartz, A. (1988) “Evaporator Operation: Trou-ble shooting” in Encyclopedia of ChemicalTechnology and Design”, J McKetta, ed,Marcel Dekker, NY.

Talley, D.L. (1976) “Startup of a Sour GasPlant” Hydrocarbon Process. Aprilp 90–92.

Troyan, J.E. (1960) “Trouble Shooting NewProcesses” Chem. Eng. Nov 14, p 223–226.

Troyan, J.E. (1961) “Trouble Shooting NewEquipment” Chem. Eng. March 20,p 147–150.

Troyan, J.E. (1961) “More on Trouble ShootingNew Equipment: Pumps, compressorsand agitators” Chem. Eng. May 1, p 91–94.

Wetherhorn, D. (1970) “Guide–Trouble-freeEvaporators” Chem. Eng. 77, No. 12,June 1, p 187–192.

Appendix AChin, T.G. (1979) “Guide to distillation pres-

sure control methods”, Hydrocarbon Pro-cessing, October, p 145–153.

Goyal, O.P. (2000) “Evaluating troubleshoot-ing skills”, Hydrocarbon Processing,October, p 100C.

409

411

For each of the ten questions, select the best answer. Then score your answers andobtain a total score. Although Goyal’s published Test about Trouble Shooting has 40questions that consider human skills, problem solving or conceptual skills and tech-nical knowledge, here we include ten of the questions about technical knowledge.see O.P Goyal “Evaluating troubleshooting skills”, Hydrocarbon Processing, Oct.,2000, p 100-C to 100-N.

1. Suddenly, emergency situations have occurred simultaneously on manyfronts:a) Control valve on fuel to furnace is stuck open.b) Side-stream pump-around on the distillation column is showing dark

color; crude leak suspected.c) Alarm sounding loud – high level in the steam drum.d) Flare pilots are extinguished.e) Safety valve leaking at the top of the fractionating tower.f) Flames impinging on tubes in the crude heater.g) Chlorine leakage in the utilities building.h) Rupture of fire/water main at offsite area.i) Steam trap blowing live steam at 0.5 kg/s.j) Vacuum constantly falling in a vacuum distillation column.k) Instrument air supply acting erratically.l) Rundown temperature of gasoline running 5 �C higher than maximum

allowed.

Assuming that all these cannot be simultaneously attended to, indicate the topfive, most sensitive operating problems that require immediate attention by select-ing any one of the following combinations

(i) a, g, h, k, l (ii) a, c, f, g, h (iii) b, e, j, k, l (iv) b, d, e, i, j

2. A two-stage reciprocating compressor is operating at 70 kPa abs inlet and 620kPa abs outlet. If the pressure drop across the intercooler is 14 kPa, the pres-sure at the inlet to the second stage will be approximately:

a) 340 kPa abs. b) 270 kPa abs. c) 200 kPa abs.

Appendix AFeedback about Experience with Process Equipment(from Goyal (2000) and reproduced with permission fromHydrocarbon Processing)


Appendix A Feedback about Experience with Process Equipment

3. For an exchanger with no phase change, given that T1 = 213 �C; T2 = 57 �C;t1 = 29 �C and t2 = 41 �C. If flowrates, area and configuration are kept constant,if T1 is reduced to 201 �C and t1 is increased to 35 �C; then (a) the new temper-atures are T2 = 60 �C; and t2 = 45.8 �C; (b) insufficient data are available tomake a calculation; (c) data are sufficient but I don’t know how to calculate;(d) the new temperatures are T2 = 60 �C; and t2 will remain= 41 �C; (e) thenew temperatures are T2 will remain= 57 �C; and t2 = 45.8 �C.

4. For most distillation cases, an increase in pressure means: a) decrease or b)increase in column capacity and (i) better or (ii) worse fractionation. Select(a) or (b) and (i) or (ii).

5. If the circulation rate of a cooling tower is substantially decreased by pinch-ing water to the condensers/coolers, the tower water-inlet temperature will:(a) go up; (b) come down; (c) remain nearly the same. The tower water-outlettemperature will (i) go up; (ii) come down; (iii) remain nearly the same. Select(a), (b) or (c) and (i), (ii) or (iii).

6. A cylindrical furnace has been insulated with two types of refractory materialof equal thickness, such that the low thermal conductivity material is on thefurnace side and the higher conductivity is on the outside. (a) what has beendone is correct to minimize heat loss; (b) since these are in series, it makesno difference which one is next to the furnace, c) what has been done isincorrect; the opposite sequence should have been used.

7. A shell and tube, horizontally installed heat exchanger with two tube passeshas pressure gauges (P1 and P2) installed, respectively, on the inlet pipe atthe lower nozzle and on the outlet pipe from the upper nozzle of the channel.The pressure gauges are approximately 1.4 m apart. When water flows in thetubes the difference in readings of P1 inlet and P2 outlet is 41 kPa. Each noz-zle takes 7 kPa pressure drop. Estimate the difference in pressure gauge read-ing as P1–P2 for the following cases:A) Tube side passes changed from 2 to 4. Answer a) 135 kPa; b) 82 kPa; or

c) 40 kPa.B) Tube-side flowrate is doubled. Answer a) 124 kPa; b) 82 kPa; or c) 40 kPaC) Tube-side flow direction is reversed. Answer a) –14 kPa; b) 0 kPa; or

c) 14 kPa.8. For a centrifugal pump:

a) flow is directly proportional to the impeller diameter and the power to itssquare.

b) head is directly proportional to the square of the rpm and the flow to thecube.

c) flow is directly proportional to the rpm and the impeller diameter and thepower to the cube of each of these.

9. At the inlet of a centrifugal compressor, if the pressure is throttled maintain-ing the actual volumetric flowrate constant at the inlet,A) the net mass flowrate will a) decrease; b) increaseB) the net standard volumetric flowrate will a) decrease; b) increase

412


C) keeping other parameters the same, an increase in the specific heat ratiowill: a) decrease; b) increase the discharge temperature significantly.

10. For steam ejectors:A) The higher the design steam pressure for an ejector, the a) higher; b)

lower the steam consumption, particularly for one- or two-stage ejectors.B) Running a steam ejector at steam pressures slightly above design a) does;

or b) does not increase suction capacity.C) At steam pressure > 25% above design, the ejector capacity actually starts

to a) decline; or b) increase.D) At steam pressures even a few kPa below design pressure, significant

reductions may be expected in a) both suction capacity and compressionratio; or b) suction capacity alone; or c) compression ratio alone.

Answers and scoring:

1. (i) +1; (ii) +2; (iii) 0; (iv) –1.2. (a) –2; (b) –1; (c) +3 [answer: 200kPa abs]3. (a) +3; (b) –1; (c) 0; (d) –1; (e) –1 The heat lost in the hot stream must= heat

gained in cold stream.4. (a) –1; (b) +1; (i) –1; (ii) +1 Except in rare cases, increasing the pressure will

increase the capacity but reduce the fractionation.5. (a) +1; (b) –1; (c) 0; (i) 0; (ii) +1; (iii) +1. The water exiting the condensers will

go up so the water temperature entering the tower will go up. The exit watertemperature from the cooling tower will not go up; the temperature will like-ly be reduced or it may remain the same depending on other parameters.

6. (a) +2; (b) 0; (c) –1.7. A: a) +1; b) 0; c) –1. The static head is 13.5 kPa. Calculation: 8 (41–7–7–13.5)

+ 14 + 13.5 = 136.5 kPa.B: a) +1; b) 0; c) –1. Calculation: 4 (41–7–7–13.5) + 4 (14) + 13.5 = 123.5 kPa.C: a) +1; b) 0; c) –1. Calculation: 13.5–7–7–(41–7–7–13.5) = –14 kPa.

8. a) 0; b) –1; c) +2.9. A: a) +1; b) –1�2 When the pressure at the suction decreases, the actual volu-

metric flowrate is maintained resulting in a decrease in the mass flowrateand also a decrease in the standard volumetric flowrate.B: a) +1; b) –1�2

C: a) –1; b) +1. Temperature increases significantly.10. Steam ejectors designed for high inlet steam pressure consume less steam

compared to those designed for low inlet-pressure steam. Operating an ejec-tor at the design pressure is very important. A) operating at a steam pressurehigher than design does not increase the suction capacity. a) 0; b) + 1�2.B) operating at a steam pressure > 25% more than design makes the capacitydecline. a) –1�2 ; b) +1�2. C) operating at a pressure slightly less than designpressure causes both suction capacity and compression ratio to suffer signifi-cantly. a) + 1�2; b) –1�2. D) a) +1�2; b) 0; c) 0.Minimum score: –15Maximum score: 24

413


General guidelines:

> 22 Outstanding experience with process equipment; you will probably addyour insight to the ideas in Chapter 3.

17–22 Very good understanding of process equipment; please add your ideas toChapter 3.

10–17 Average understanding of process equipment; you might want to checkideas in Chapter 3 as you trouble shoot.

< 10 Beginning understanding of process equipment; use ideas in Chapter 3to help you trouble shoot.

414

415

In Section B.1 are summarized some symbols commonly used on Process andInstrumentation Diagrams, P&IDs. Section B.2 gives you an opportunity to practiceanalyzing a P&ID.

B.1Symbols for P&ID

Appendix BImproving “Systems Thinking”


SYMBOL KEY FOR PROCESS

SKETCHES

XXX

100

XXX

100

CONTROL ROOM BOARD

MOUNTED INSTRUMENT

LOCAL BOARD MOUNTED

INSTRUMENT

}MEASURED VARIABLE:

L: LEVEL

P: PRESSURE

A: ANALYZER

F: FLOW

T: TEMPERATURE

EXAMPLES:

PRC

100

PRESSURE RECORDER

CONTROLLER: I.D. NUMBER 100,

MOUNTED IN CENTRAL CONTROL

ROOM.

FI LOCAL FLOW INDICATOR

TAH

TEMPERATURE ALARM:

ACTIVATED AT HIGH

TEMPERATURE.

FUNCTIONS:

I: INDICATOR

C: CONTROLLER

R: RECORDER

S: SENSOR

A: ALARM

(H =HIGH, L=LOW)

Appendix B Improving “Systems Thinking”416

VALVE

DIAPHRAGM VALVE:

CONTROLLED BY AIR LINE

CHECK VALVE

S SOLENOID VALVE

CONTROLLED BY

ELECTRICAL SIGNAL

CENTRIFUGAL PUMP

AIR LINE

ELECTRICAL SIGNAL

ELECTRICAL TO PNEUMATIC

SIGNAL CONVERTER

PACKED TOWER OR

PROCESS VESSEL (SUCH

AS A REACTOR). SIZE

VARIES.

GENERIC TOWER OR

PROCESS VESSEL (SUCH

AS A REACTOR). SIZE

VARIES.

BURST DIAPHRAGM

HEAT EXCHANGER,

COUNTERCURRENT (BOX

DIRECTLY AFTER

INDICATES STEAM TRAP)

HEAT EXCHANGER,

COCURRENT (BOX

DIRECTLY AFTER

INDICATES STEAM TRAP)

FURNACEPI

AXIAL

COMPRESSOR

THREE-WAY VALVE

SPRING LOADED

SAFETY VALVE

INSTRUMENT SIGNAL

Figure B1

B.1 Symbols for P&ID 417

SYMBOL KEY FOR PROCESS

SKETCHES

HEADER (STEAM OR

CONDENSATE)

STEAM TRAP

MOTOR

FLOW METER

ORIFICE PLATE

ELECTRICAL

GENERATOR

M

T

POSITIVE

DISPLACEMENT

PUMP

CONDENSER

DRAIN TO SEWER

IMPELLER

VAPORIZER

OR

REBOILER

FILTER AND DRAIN

CYCLONE

DAMPER

OR

BUTTERFLY VALVE

STEAM

EJECTOR

FLARE STACK

SHELL AND TUBE

HEAT EXCHANGER

FAN

OR

BLOWER

FAN

TURBINE USED TO

PROVIDE WORK

TO DRIVE

COMPRESSOR

Figure B2

Appendix B Improving “Systems Thinking”

B.2Systems Thinking

For the P&ID given in Figure 2-4 of the depropanizer, Section 2.4 Case’8, system-atically work your way through the following questions. For more see http://www.chemeng.mcmaster.ca/courses/che4n4/Process operability/

1. What’s going on? For column C-8; how can you tell?

2. What’s unique? draw simplified sketch; compare C-8 with C-9.

. So what?

3. Setting temp and pressure: what is the pressure at top of C-8? How can youtell?

We try to operate columns at atmospheric pressure.

. Why use this pressure? Cox chart.

. What does this tell us about the species? Cox chart

. Pressure at the bottom of the column. Bigger than top? smaller than top?same as top? Why?

Estimate:

. Why is this a good pressure for the bottom?

. What’s the species at the bottom?

Quick check on the debutanizer.

4. Controlling pressure and temperature

Pressure: For deprop C-8; how control?What if the sequence of the control valve and sensor is reversed? Control (see in

control room?)AlarmSIS: fail open?ReliefFor debutanizer C-9, how control?

. What if it is a vacuum?

Use vacuum for those species with a high decomposition index: foodstuffs and “higherboiling pharmaceuticals”

Temperature: For Deprop C-8, how control: top?bottom?

. Why sensor location?

. What signals: electrical or pneumatic? how can you tell?

. At the bottom, what type of reboiler?

. Is it an equilibrium stage?

418

B.2 Systems Thinking

. Is it nucleate boiling?

. Is steam superheated? H-S diagram.

. Estimate the flowrate of steam

. What type of steam trap? Does it make any difference which type of trap isused?

. Compare with Debut C-9

Level:

. Is the level controlled in bottoms?

. in overhead drum V-30?

. in feed drum?

5. Checking the fluid mechanics

Why do fluids move? Consider Bernoulli’s

D<v>2

2þ Dp

�� 4f

LD

� �<v>2

2þ Dz ¼ power needed

and the microscopic equation of motion:

�DvDt

¼ �rp þ lr2v þ �g

inertia pressure friction body forces

to see the fundamentals of why fluids move. Fluids move because of a pressure differ-ence; or because they are dragged along by something moving (eg a steam ejector), becauseof momentum and because of body forces acting on the mass. Pressures occur because1. pump or blower, 2. suck or pull vacuum, 3. heat up liquid in closed vessel 4. densitydifference on two ends of a tube.

Pressure Profiles: Estimate: Tabulate Dp across the following equipment: Packedbeds:

Fittings:Trays and packing:Turbulent flow of gas flow and liquid:Major Dp across control valves:Create a summary page of rules of thumb Dp.

. Find the thermosyphon reboiler on the bottom of tower C-8. Track the pres-sure. Use 100 kPa approach by arbitrarily setting the pressure= 100 kPa atone location.

. Track the pressure as you go from the bottom of C-8 up the column andaround to the vessel V-30.

. Consider the reflux drum V-30 and the flow of reflux back into column C-8.How could we make the reflux flow by gravity? Self-standing column. skirtand configuration.

419

Appendix B Improving “Systems Thinking”

Vacuum: Steam ejectors; work on principle of constant S and thus have available theenthalpy difference per kg of steam but 15% efficiency.

Pumping liquids: express the pressure losses in terms of “head”; the head lossdiagram is independent of the liquid.

. Do amass balance around column C-8 on PID-2A.

. What is the reflux ratio?

Pump F-26

Estimate the pressure drop from pump F-26 into depropanizer C-8.

For a centrifugal pump

1. head: independent of fluid if it is liquid.2. efficiency: where it is located. Sometimes shown as “concentric rings” on the dia-

gram.3. power and how we can estimate the efficiency from the head capacity and the

power: 60%.4. can shut the exit valve off completely and the pump is still OK and gives the max.

head. This is a handy way to help you identify which pump is installed in a circuitif the ID plate is removed or if you are troubleshooting and suspect it is not deliver-ing what it should. Note that the pressure gauge must be installed upstream of thevalve.

5. pumps usually run at 1800 rpm or 3600 rpm. At the former, the heads are usuallyabout 10 m; at the latter about 40 m.

6. Key idea, the pump operates along the operating curve! You change the Dp in thesystem and the pump moves to a new operating location on the operating curve.

. Select, from a manufacturer’s catalog, pump F-26, on the feed to the depropa-nizer.

. What changes if we use this pump, F-26, from the Deprop circuit to operatewith a) benzene. b) water and c) with mercury.

Net positive suction head required: “head” needed to keep liquid from boiling in theeye of the impeller. Data must be supplied from manufacturer (for water at 20 �C). Mustsupply “more than this”. Options for control:

1. large suction line,2. negligible stuff to cause Dp on the suction line,3. have a large height for static head,4. use vortex breakers and submersion.

. Spot places where we could have trouble on this plant from NPSH.

420

B.2 Systems Thinking

Control configurations

. Note the bypass loop around the control valve so that we can operate tempo-rarily on bypass.

. Note change in diameter of the control valve compared with the diameter ofthe line in and out.

. Why?

. Note drain valve so that we can take out the valve safely.

421

423

Here are some example Trouble-Shooter’s Worksheets for the Cases posed inChapter 1.

Appendix CFeedback on the Cases in Chapters 1, 2 and 7


Trouble-Shooter’s Worksheet � copyright 2003 Donald R. Woods andThomas E. Marlin

Case’3: The case of the cycling column1. Engage: Write down what is said; what you sense, smell, hear. If someone istelling you, then use skilled reflective statements to ensure you accuratelyobtain the information.

. Emergency priority: Safety? Hazard? Equipment damage?shut down&; safe-park &. If not, &� then:

. Draw a sketch of the process and mark on values. Provide a descriptionin words of what is going on. This is a distillation column. The bottomsis heated via a thermosyphon reboiler. The liquid handled is a hydrocar-bon, iC4. Liquid from the bottom of the column fills the tubes in thevertical shell and tube reboiler. Steam is applied to the shell side. Someof the hydrocarbon in the tubes boils. The resulting density differencecauses fresh bottoms liquid to enter the tubes and the evaporation con-tinues. Usually 5 to 25% is evaporated per pass. This is rarely used invacuum or high-pressure service. The steam comes off the top of theheader, as it should. The condensate from the bucket trap goes into thetop of the condensate header, as it should. The flow of steam is con-trolled by the temperature.

. Symptoms:a) The level in the sight glass, hopefully representing the level in the

bottom of the column, rises slowly about two feet above the normaloperating level and then quickly drops to about two feet below nor-mal. The operators say “the column cycles madly”.

b) this is just after annual maintenance.

Appendix C Feedback on the Cases in Chapters 1, 2 and 7424

. Manage any panic you might feel by saying “I want to and I can. I have astrategy that works. Let’s systematically follow it.” Yes! I can do my best!

. Monitor: Have you finished this stage? Can you check? What next? I havebeen cautious. I restated the words from the statement.


IS IS NOT

WHAT Level in the bottoms is cycling with afour-foot variation in level.Performance is controlled bytemperature.

(should be happening but it’s not)Level should be “level”Performance is not controlled by level.

WHEN ?& just after turnaround maintenance ?&� didn’t cycle before turnaroundWHERE ?& bottoms of column ?&� no cycling reported in feed, feed con-

centration.WHO ?& current shift operators ?&� this is first shift to start this plant

– startup new process & suggest use Basics– startup after maintenance or change &� suggest use Change– usual operation but changes made in operation but not in equipment& suggest use Basics

– usual operation& suggest use Basics.

. Monitor: Have you finished this stage? Can you check? What next? I havethree � that need to be answered. This sounds like a change problem.

3. Explore: Gather information to be gathered ?& in Define stage &. Done andrecorded in chart.

Exercise? & or a problem?&� .Strategy: change &� or basics &.Perspectives. Why? Why? Why?

__________________________________________________________Why? ›

__________________________________________________________Why? ›

4. so product can be soldWhy? ›

3. get on-spec tops and bottomsWhy? ›

2. level out performance of columnWhy? ›

Start fi 1. stop great variation in level and in steam flow

Appendix C Feedback on the Cases in Chapters 1, 2 and 7 425

Main goal is to “get on-spec tops and bottoms”

. Prioritize: product quality &�; production rate &; profit &.

. Goal: safe-park? &�: short term now with long term later &; long termnow &

Action to be achieved: Specific terms and Measurableget on-spec tops and bottoms

Attainable? could put it on manual control or on safe-parkReliable?Timely?Safe?$

. Check consistency of data/symptoms: interdata consistency?OK& no &data consistent with fundamentals?OK& no &

. Type of problem: startup new process &� maybe mechanical/electricalfailureusual operation &: ambient temp? &maybe fluids problems;high temperature? & then maybe materials problemsSystem? failure of heat exchanger &�> rotating equipment& >vessels &> towers&


What if? _______________________ then _____________________________What if? _______________________ then _____________________________What if? _______________________ then _____________________________

. List changes made &� and/or List trouble-shooting causes based onsymptom &�. List both.

Changes made: During the shutdown

1) The condensate from this inverted bucket trap, that previously dis-charged to atmosphere, was repiped to discharge into a 200-kPa conden-sate header. The condensate discharged into the top of the header. Steamto the reboiler was “saturated” at 1.7 MPa.

2) For this unit the only other maintenance was instrument checks. Nofaults were found in the instruments and no changes were made to set-tings in controllers.

3) And visual inspection of the trays. The visual inspection involved open-ing the access holes.Trouble shooting background. From Chapter 3 and from Michelle’saccount in Chapter 4. Not reproduced here.


. Brainstorm root causes:collapsed tray, change in downstream pressure affecting bucket trap,instruments wrong; liquid level is not cycling, control system, restrictionin the vapor line, cycling change in concentration of high boilers in thefeed, temperature sensor in the feed zone.

In prioritizing these, focus on the change: change the exit pressure of thebucket trap.

Hypotheses: list in Chart; Symptoms: code and list in chart; Analyzewith S supports; D disproves and N neutral or can’t tell.

Symptom a. liquid level increases by 2 ft and then decrease by 2 ftb. occurs after shutdown maintenancec.d.e.


a b c d e A B C D

1. change in discharge pressure to bucket trap S S 4 4 4

2. instrument reading wrong S

3. cycling concentration of heavies in feed S N

4. control system S 4

5. restriction in vapor line S

6. temperature sensor in feed zone S N

7.

Diagnostic actions:

A. control system on manualB. open trap bypassC. discharge condensate to atmosphereD. check out sizing calculations for orifice in trap



Case’4: The platformer fires1. Engage: Write down what is said; what you sense, smell, hear. If someone istelling you, then use skilled reflective statements to ensure you accuratelyobtain the information.

. Emergency priority: Safety? Hazard? Equipment damage?shut down&; safe-park &. If not &�, then:

. Draw a sketch of the process and mark on values. Provide a descriptionin words of what is going on. Liquid naphtha is preheated in a shell andtube exchanger and enters the packed bed, reactor where the naphtha isconverted to platformate. The products from the reactor are high octanegasoline plus hydrogen-rich gas at 4.8 MPa g and 500 �C. These hotgases are cooled in the feed preheater. The preheater shell is 1 m diame-ter. Stainless steel.

Three weeks since startup. Four flash fires along the flanges. Maintenancehas tried unsuccessfully to tighten the flange to prevent leaks once the reactorheats up.

. Manage any panic you might feel by saying “I want to and I can. I have astrategy that works. Let’s systematically follow it.” This one sounds chal-lenging but I’ll do my best.

. Monitor:Have you finished this stage? Can you check? What next? Yes.


IS IS NOT

WHAT Flash fires along flanges.Six bolts broken trying to tighten theflange

(should be happening but it’s not)No flash fires. A flange that is easy toseal mechanically.

WHEN ?& During three weeks since startup ?& not operated before so have noinformation

WHERE ?& Along the flanges of the preheater-effluent cooler

?& in the platformer reactor orelsewhere

WHO ?& no evidence that it occurs only onone shift

?&

– startup new process &� suggest use Basics– startup after maintenance or change & suggest use Change


– usual operation but changes made in operation but not in equipment& suggest use Basics

– usual operation& suggest use Basics.

. Monitor: Have you finished this stage? Can you check? What next? ThinkI’ve finished

3. Explore: Gather information to be gathered ?& in Define stage &.Exercise? & or a problem?&� . Never seen anything like this before.Strategy: change & or basics&�Perspectives. Why? Why? Why?

6. so that I can be paidWhy? ›

5. so we can sell platformate and make profitWhy? ›

4. so the whole process can operateWhy? ›

3. make it safe; prevent flames to other parts of the processWhy? ›

2. prevent fires on the effluent exchangerWhy? ›

Start fi 1. prevent hydrogen-rich gas from leaking out the flange

. Prioritize: product quality &; production rate&; profit &; safety &�

. Goal: safe-park?&: short term now with long term later&� ;long term now&

Action to be achieved: Specific terms and Measurable:level 2: prevent fires

Attainable? Fires are hydrogen + spark + oxygenReliable?Timely?Safe?$

. Check consistency of data/symptoms: interdata consistency?OK& no &. Fires are hydrogen + spark + oxygendata consistent with fundamentals?OK &� no &. Hydrogen diffuses very easily through small crevices.These are very high temperatures and pressures. The thermal expansionduring startup will be extensive.

. Type of problem: startup new process &�; maybe mechanical electricalfailure


usual operation &: ambient temp? & maybe fluids problems; high tem-perature? & then maybe materials problemsSystem? failure of heat exchanger &� > rotating equipment &>vessels &> towers&


What if? heated the outside of the exchanger then negligible differentialthermal expansionPut exchanger in a furnace?

What if? let the hydrogen leak then focus on preventing oxygen from mix-ing with hydrogen

What if? _________________________ then ________________________

. List changes made & and/or trouble-shooting causes based onsymptom &�fire: combustible like hydrogen + oxygen/air + spark/

Identification of cause is not the issue; prevention is

. Brainstorm root causes: possible ways to keep oxygen from mixing withthe hydrogen.nitrogen, steam, not airideas to contain the hydrogen once it leaks: a blanket or outside shellthat contains the hydrogen and from which to extract it elsewhere;packed hydrogen adsorbent; spray liquid over exchanger that rapidlyadsorbs hydrogen; react hydrogen.


Symptom a. four fires along flangesb. break six bolts trying to tightenc. during first three weeksd.e.


a b c d e A B C D

1. hydrogen leaks out of the flanges S N N

2. cannot tighten the flanges any tighter toprevent the leak

S S



Case’5: Sulfuric acid pump problem1. Engage: Write down what is said; what you sense, smell, hear. If someone istelling you, then use skilled reflective statements to ensure you accuratelyobtain the information.

. Emergency priority: Safety? Hazard? Equipment damage?shut down&; safe park &. If not &, then:

. Draw a sketch of the process and mark on values. Provide a descriptionin words of what is going on. Acid comes from different parts of theplant and is collected in a common storage tank. A pump, located abovethe tank, pumps the acid to an elevated tank. Vertical dimensions aregiven.




IS IS NOT

WHAT When the level of acid in the storagetank drops to 0.7 m in the bottom ofthe storage tank, the operator says thepump makes a “crackling” noise thatsounds like cavitation.

(should be happening but it’s not)Pump cavitation should not happen

WHEN ?&When the level of acid in thestorage tank drops to 0.7 m in thebottom of the storage tank.

?&When the level of acid in the storagetank is above 0.7 m in the bottom of thestorage tank.

WHERE ?& In the acid pump ?&� In other pumpsWHO ?& ?&


– usual operation&� suggest use Basics.



3. Explore: Gather information to be gathered ? &� in Define stage &. Done.Exercise? &� or a problem? &. Sounds like a cavitation problem and NPSH sup-plied & NPSH required

Strategy: change & or basics&�.Perspectives. Why? Why? Why?

6. Personal Happiness and BlissWhy? ›

5. pay my salary plus bonusWhy? ›

4. improve profitsWhy? ›

3. keep costs downWhy? ›

2. minimize erosion and allow longer pump cycleWhy? ›

Start fi 1. stop pump cavitation

. Prioritize: product quality & ; production rate&�; profit &�

. Goal: safe-park?&: short term now with long term later&� ;long term now&

Action to be achieved: Specific terms and Measurable:stop crackling noise when level drops to 0.7 m.

Attainable? probablyReliable? temporary should workTimely?Safe?$

. Check consistency of data/symptoms: inter data consistency?OK &� no &. Onset of “crackling noise” should coincide with reducedpumping flowrate.data consistent with fundamentals?OK &� no &. “Crackling noise” consistent with cavitation consistentwith insufficient NPSH that is consistent with excessive suction lift thatis consistent with level in tank too low.

. Type of problem: startup new process & maybe mechanical electricalfailureusual operation &�: ambient temp? &� maybe fluids problems; high tem-perature? & then maybe materials problemsSystem? failure of heat exchanger &> rotating equipment &� >vessels &> towers&



What if? acid more dense then more likely to cavitate earlier or > 0.7 mWhat if? atmospheric pressure drops then more likely to cavitate earlier

or > 0.7 mWhat if? vent plugged then non flow

. List changes made & and/or trouble-shooting causes based onsymptom &�

See Chapter 3 or list Dave generated.

. Brainstorm root causes:liquid too hot, non-condensibles, air leakage into suction, vortex entrain-ing gas, density change in liquid, excessive Dp on suction side, suction ve-locity too high, increased rpm, increased liquid capacity.

The entries in italics are symptoms and not a root causes. Many of these donot relate to suction-lift situations.

. Hypotheses: list in Chart; Symptoms: code and list in chart;Analyze with S supports; D disproves and N neutral or can’t tell.

Symptom a. crackling noise related to cavitationb.c.d.e.


a b c d e A B C D

1. NPSH supplied is less because atmosphericpressure low

S 4

2. NPSH supplied is less because of entrainedgas because of vortex

S 4

3. NPSH required is higher than expected a)because higher capacity because of lesssystem resistance

S

4. NPSH supplied less because liquid densitytoo high

S 4

5.

6.

7.


Diagnostic actions:

A. increase pressure on top of storage tankB. reduce liquid pumping capacity by shutting down on discharge valve (Will

reduce the NPSH required and should reduce tendency to form vortex)C. sample acid and measure densityD.

Cases’15 to 18 are a few of the cases that are related directly to human error andissues raised in Chapter 7. There may be other problems given in Chapter 8 wherethe cause is human error.

In this Appendix we simply provide the cause of the problem for Cases’15 to 18.

Case’15: The flooded boot (from Bill Taylor)This is a case of initial construction error. Workers left a block of wood in the 15-cmdiameter (6†) line between the ion exchanger and the pump. Material left in lines iscommon. For example, a line rigger temporarily places bolts in a pipe while taking abreak. A colleague arrives back from break, continues assembling the line, althoughhe cannot find the bolts so gets new bolts. The plant starts up and the bolts arecarried into the blades of a downstream centrifugal compressor. Six months delayoccurs in getting the repaired compressor on-line! Sandwiches, rags, wrenches andstuff ... don’t be surprised by what might be left in lines.

Of course, after commissioning runs with water and air, expect pockets of theseto remain in low places and unvented locations.

Case’16: The case of the dirty vacuum gas oil (From Bob Farrell)The problem is that crude oil is leaking into the vacuum oil. In an effort in increasecapacity beyond the usual, the production manager increased the pressure in thecrude line such that it exceeded the rating of the exchangers. A leak resulted.

This is a case of operator/supervisor error

Case’17: Is it hot or is it not? (case from Jonathan Yip)This is a case of an operator ignoring the readings of the instruments. Within ashort time after the operator had manually suppressed the purge valve, thinking itwas a false temperature indication, a field operator felt something was wrong whenhe heard unusual cracking noise from the column. Looking at the DCS screenagain, he realized that a section of the column had temperatures above the range. Afire had broken out in the column. He immediately activated the nitrogen purgeand started to cool down the column with water spray.

Investigations afterwards showed that the pump isolation valve was leaking dueto a worn seal. This allowed air to migrate into the column. Since the fatty acidswere above their autoignition temperature, the acids ignited in the presence of air. A


further survey of the data showed that the column pressure had been graduallyincreasing or losing its vacuum over the same period suggesting a air leak. Theinternal damage to the column from the fire was extensive. The entire column hadto be replaced.

Case’18: The streptomycin dilemma

This is a case of an operator ignoring operational procedure and substituting hisown to make life easier. The expected procedure was:

1. brush out the tubes; 2. rinse. 3. fill the tubes with water and 4. add steamto the shell and boil the water. What the operators did was 1. brush out thetubes; 2. rinse and then 3. use a steam hose, with a long pipe attached to theend of it, to blow live steam into the tubes individually. They reasoned thatthe live steam “would do a better job of sterilizing” than the old slower meth-od. They did not understand the impact of the high temperatures on the ther-mal expansion of the tubes.

When asked if they followed the expected procedures, all operators said yes. Noneadmitted to using live steam. This was only discovered when engineers unexpect-edly visited the site when the tubes were being “sterilized”.

In this case, the environment did not allow for admission of error; the rationalefor the sterilization procedure had not been given. The consequences of using livesteam had not been spelled out.

434

435

Consult the code corresponding to the question or activity you want. Keep a recordof the sequence in which you pose each question. You can later compare thesequence of questions you asked with those of an experienced trouble shooter. Sug-gestion, work question by question instead of collecting four or five questions andobtaining answers to them all at once.

1 Not needed. $6502 This is the commissioning startup of a new unit. $503 Responds to change. $2004 No. $305 Hot 30 �C and humid: 80% relative humidity. This week started out with mod-

erate temperatures but the temperature and humidity have climbed to today’shigh.

6 Not needed. $507 Estimate seems OK. No major blockage. $1008 Not needed. $1509 Water is used in the seal pot in the summer; kerosene is to be used in the

winter. Kerosene is obtained from the crude unit. $50010 Should work well. Baffles were in vertical. Vessel has air bleeds. Before you

started up you would have opened the bleed to ensure all the air was out. $20011 Responds to change. $20012 Usual cycle is to power the heaters, add feed to hopper, start mixer, mix 35 s,

stop rpm and screw advances to push melt into the mold, screw reverses, feeddrops from hopper into mixer and so on. The resin is predried for 3 hours bydirect contact with hot air at 93 �C. $100

13 Fully open. $30014 IS: on this reactor. IS NOT: elsewhere in the plant. $50. If you didn’t put plant

on SIS or SIS + evacuation before you asked this question, the plant explodeswith loss of life. Penalty $3 000 000

15 Perhaps shut down, depending on the length of time of operation since thelast attempted flow increase.

16 Allowance of 140 kPa across flow control valve FV-1 NPSH required= 1.8 mwater. $150

Appendix DCoded Answers for the Questions Posed to Solve the Cases


Appendix D Coded Answers for the Questions Posed to Solve the Cases

17 IS: This shift. $300018 Not apparent. $200019 Not for 8 months. $5020 Should do the job. $20021 Usual design with direct water contact. Use river water. $12022 No change. Flowrate and temperature should be the usual values. The temper-

ature should be about 20 �C. $200.23 Cold –1 �C; cloudy, snow predicted. Previous week, cold with some snow. $2024 Looked fine. $20025 Equipment should do the job.26 Check on design calculations show liberal allowance for the pressure drop

across the control valve so that it operates mid-range. Pump should have notrouble supplying the design reflux. $100

27 IS: Just after change to new catalyst. IS NOT: problem with old catalyst. $15028 OK, recalibration not necessary. $70029 Clear, not clogged. All clearances are OK. $200030 Raining. Yesterday: hot, dry, 30 �C. Showers three days ago. Cleared up and

has been hot ever since until today’s rain. $1031 Clean. It just had new bags installed. Reverse jets had been serviced.32 Not needed. $30 00033 No problems. $10034 Ethylene NFPA: 1, 4, 2; n-butane: 1, 4, 0. Butane skin-contact with gas may

cause frostbite (liquefied gas). May cause slight eye irritation, inhalationcauses drowsiness, excitation or unconsciousness due to anesthetic andasphyxiation properties of this gas. $50

37 Typically rule-of-thumb value is 5 kPa; instruments read 50 kPa.38 Samples show a decrease in number of particles with time. No data available

for previous operation. $210039 Yes; everything seems to be OK on our unit. $20040 OK. Recalibration not necessary.41 3 days ago completed the annual work on the baghouse: new bags installed.

Reverse jets had been serviced.42 Not needed. $100043 Not needed44 Fluctuates. $20045 Selected for flow and pressure drop around the hydrocyclone. $18046 Well-designed for the design capacity. Since acetic acid vapor partially

dimerizes (so that the molar mass varies between 60 and 120) a molar mass of102 was used for all calculations involving acetic acid vapor. Heat transfer coef-ficient 20 to 40 W/m2 K. Triethyl phosphate added as catalyst about 0.3% w/win the acetic acid feed.

47 About 400 ppm for each 24 h composite for each of the seven days. $7048 Not needed. $3000

436


49 Responds to change. $300. If you didn’t put the plant on SIS or SIS + evacua-tion before you asked this question, the plant explodes with loss of life. Penalty$3 000 000.

50 Responds to change. $200051 Hot liquid fatty acids will oxidize and cause spontaneous combustion. $20052 Crystals and liquids do not pose a hazard. $3053 Responds to change.54 Crystals in compact layer instead of being movable crystals. Predict that this

would cause a decrease in filtration rate compared with design and higher exitmoisture content in crystals peeled from the centrifuge. $1400

55 Ethylene 99.99%. no butane. $500056 Your unit receives the first cut because this is so critical for you. Flowrate is

steady. Temperatures are steady and the usual values for this time of year.57 Everything OK. $650058 Not needed. $20 00059 Well designed; should do the job. $200060 From the package: strengths of cast samples well below specs. Particle-size dis-

tribution missing the very fine particles < 10mm.61 None available. $5062 Amps as expected except when low flowrates of acid are required. $100063 Should be OK even with increase in column pressure by 100 kPa; allowed for

in the Dp across control valve FV-1. $10064 Process stream into the refrigerant. $30065 Reads as expected for the naphtha flowrate giving the same space velocity for

the new conditions as the old conditions. $15066 Estimate seems OK. No major blockage. $120067 Not needed.69 Level above level of tubes; design level. $5070 The diagram is a reasonable schematic. Missing are the isolation block valves

for the flare-gas compressor; the isolation block valves, drain and bypass withvalve around the kickback control valve. The pressure gauges have pigtailswith a shutoff valve with a tee nipple and valve installed between the gaugeand the pigtail to allow backflushing of pigtail. The knockout pot has a demis-ter and a level indicator; the flare pot has isolation block valves. The flare is anelevated flare with a molecular seal and an elevated flare burner. There is aflow meter on the gas to the flare that sends a signal to a ratio controller con-trolling the steam flow to the flare. Liquid from the knockout pot is pumpedto oil recovery or slops. A gas purge line goes to the molecular seal.

71 Moisture content was within the design range although slightly higher thanbefore the change to the washing cycle was made. $60

73 Continuous. No break or short circuit in heater. Heater should work. $23 00074 Not needed. $50 00075 Design specs. seem OK. Centrifugal compressor; designed to operate without

surge based on typical gas composition. $50076 Responds to change.

437


77 On spec. No complaints from other customers for that batch. $35078 DPI decreasing. $5079 No change; still cycles. $120080 Valve stem responds to change. $60082 Closes within –18%. $60084 Not needed. $15085 See Chapter 3: Distillation Section 3.4.2, perhaps pertinent condensers and

reboilers, Section 3.3.3; pumps, Section 3.2.3 and controllers Section 3.1.1,$50

86 No change in feed sent to your unit $120087 Allowance for fouling on tube side, water= 0.0002 m2 K/W; shell-side

stream=0.0005. Care to prevent temperature cross-over. Simulation showsthat unit was operating very close to design values for usual range of feed-stocks and conditions. $400

88 Not needed., $200089 Vent bleed installed to break any syphon. The elevation of the header is such

that the tubes at all levels should be flooded. $15090 Instrument error in flowmeter/ motor fails to start when remote start button

is pushed/ pump A has worn wear rings causing internal flow circulation/ onpump A the motor is turning backwards so that the impeller is turning in thewrong direction/ air lock in pump/ debris or stuff plugging the suction line ofpump A/ reverse flow through check valve on pump B/ faulty block valves.$40

91 No change in operation. Pressures and temperatures still increasing. $25092 25 �C. $15093 Same as usual concentration. $250094 No improvement; indeed, when level drops below complete coverage of the

bundle, T/6 reads 15 �C and rising. $200095 Product cracked coming out of the mold. $90096 Agrees with head capacity. $20097 Sensor fault/ poorly tuned control system/ sticky control valve/ control-valve

hysteresis/ caustic waste flows backwards into the pump/ electrical interfer-ence with the control system/ cavitating pump/no vent on the storage tank;vacuum created in the storage tank/ density of acid > expected and motor over-load. $100

98 Not needed. $50 00099 Steady. $120100 Both valves leak when fully closed. % leakage varies from valve to valve and

ranges from 1 to 3%. $40 000101 Previous flows of air and fuel gave 10% excess air for stoichiometric combus-

tion. If didn’t put on safe-park as first step, then dangerous potential fire/explosion conditions are created while you experiment. $500 000

102 Flow is the control value of 14 L/s. $50103 Steady and usual value. $50104 Difficult because the steam flowrate is not recorded. $50

438


105 Responds to change. $200106 Well within design specs even though this is a hot and humid day. $3200107 Process stream into the cooling water. $300108 Yes, within –15%. $1000110 Sodium hydroxide: Health Rating: 3 – Severe (Poison) Flammability Rating: 0 –

None. Reactivity Rating: 2 – Moderate. Contact Rating: 4 – Extreme (Corro-sive); Poison! Danger! Corrosive. May be fatal if swallowed. Harmful ifinhaled. Causes burns to any area of contact. Reacts with water, acids andother materials. Lab Protective Inhalation: Severe irritant. Effects from inhala-tion of dust or mist vary from mild irritation to serious damage of the upperrespiratory tract, depending on severity of exposure. Symptoms may includesneezing, sore throat or runny nose. Severe pneumonitis may occur. Inges-tion: Corrosive! Swallowing may cause severe burns of mouth, throat, and sto-mach. Severe scarring of tissue and death may result. Symptoms may includebleeding, vomiting, diarrhea, fall in blood pressure. Damage may appear daysafter exposure. Skin Contact: Corrosive! Contact with skin can cause irritationor severe burns and scarring with greater exposures. Eye Contact: Corrosive!Causes irritation of eyes, and with greater exposures it can cause burns thatmay result in permanent impairment of vision, even blindness. Chronic Expo-sure: Prolonged contact with dilute solutions or dust has a destructive effectupon tissue. Aggravation of Pre-existing Conditions: Persons with pre-existingskin disorders or eye problems or impaired respiratory function may be moresusceptible to the effects of the substance. Airborne Exposure Limits: – OSHAPermissible Exposure Limit (PEL): 2 mg/m3 Ceiling – ACGIH ThresholdLimit Value (TLV): 2 mg/m3 Ceiling. Chlorine: NFPA: 4,0,1. Corrosive and poi-sonous gas. Contact may cause severe irritation or corrosive burns to the eyes,skin and mucous membranes. Inhalation may result in chemical pneumoni-tis, pulmonary edema and respiratory collapse. Nonflammable. Oxidizer. Mayreact violently with reducing agents. Can accelerate combustion. Store below50 �C. OSHA-PEL 1 ppm; TLV-ACGIH 0.5 ppm TWA; 1 ppm STEL Sodiumhypochlorite: Contact with acids releases poisonous gas (chlorine). Light sensi-tive. Incompatible with strong acids, amines, ammonia, ammonium salts, re-ducing agents, metals, aziridine, methanol, formic acid, phenylacetonitrile.Corrosive, causes burns to skin and eyes. Harmful by ingestion, inhalationand through skin contact. Skin irritant. Toxicity data ORL-MUS LD50 5800mg kg–1 ; ORL-WMN TDLO 1000 mg kg–1; IVN-MAN TDLO 45 mg kg–1. $50

111 Not needed. $3000112 Responds to change; PI indicates increase in pressure. $25113 Methane: NFPA: 1, 4, 0; dangerous fire and explosion hazard. Does not contain

oxygen and may cause asphyxia if released in a confined area. Can cause irrita-tion and central nervous system depression at high concentrations. Nitrogen:NFPA 0, 0, 0; Does not contain oxygen and may cause asphyxia if released in aconfined area.Hydrogen: NFPA: 0, 4, 0.

115 Hopper clean, blower works well, line appears open.116 Not needed

439


117 Carefully designed to provide the minimum surface area/ unit volume so as tominimize the potential for ketene to react reversibly with condensed film ofwater. Area should handle the condensation duty. Gas-liquid KO pots (withbarometric legs) provided between different sections of the condenser to facil-itate prompt removal of the condensed water.

118 Recalibration not needed. $3500119 Non-explosive ranging from 17 to 22% of the lower explosive limit for all tests

and at all times. $700120 Allowance for fouling on the shell side= 0.00001 m2 K/W; on the tube

side= 0.00001. Use U tube instead of fixed tube sheet, because of thermalexpansion. Vapor-liquid ports joining the shell sides were designed for maxi-mum upward butane flow and full condensate downflow based on downcomervelocity considerations to give annular flow. The design is to evaporate the eth-ylene at 3.9 MPa g (–3.8 �C) and superheat the vapor above 1.6 �C. $50

121 Not needed. $5000122 Not needed. $650123 Yes; everything seems to be OK on our unit. $200124 No change, no improvement. $400125 No improvement. $40 000126 Design checks out OK. No errors made. As expected, a trim cooler was

included on the IPA liquid line. $1000127 –99 –3 �C128 No leaks observed.129 Steam would leak into the process fluid. $50130 Awkward. Control is difficult but temperatures are within –2 �C of targets.

Steam usage on compressor increases. $2000131 85 kPa. Usually 54 kPa. $600132 Yes, steam pressure is 3.5 MPa g and brought to the BL of your plant. On site

you reduce it to 1.38 MPa.g and further reduced to 350 kPa g for the steamtracing. I keep saying you should be put in local steam turbine drives; you arejust wasting thermal energy by dropping the pressure across reducing valves!

133 Should give the correct% vaporization per pass. $700134 400 kPa g plus pump compared with 20 kPa g on ethylene side: process fluid

leaks into ethylene.135 Pressure is slightly less than usual. On the head-capacity curve suggesting a

higher than usual flowrate. However, the pressure is gradually increasing.$300

136 7 s. $150137 Nothing. We were shut down for maintenance.138 Checks suggests looks OK. Maybe check the equipment selected when on site

to ensure it matches design specs. $200139 No gauge present. $120140 Estimate of power needed for pumps A and B, 10 kW each, agree with

installed motors and usual efficiency of 70%. $50141 Responds to change. $300

440


142 Pump should supply the design flowrate with ease. A pressure drop of 70kPaacross the control valve was allowed for the design. $120

143 Instrument fault/ maldistribution/ hot-gas recirculation/ fouled tubes/ bladewrong pitch/ fan not working/ insufficient tube area/ buildup of non-conden-sibles in the bottom row of tubes/ tubes not sealed/ no vent break/ poor con-trol system/ control system not well-tuned.

144 Equipment should do the job. Indeed, should overheat because some “small”fouling resistance was used in the design and since this is startup should beclean. $800

145 Well designed; trim selected carefully for caustic; should operate mid-positionfor design flowrate. $100

148 Not needed.149 We used new batches of resin. We are still using from the same batch of color-

ing and foaming agents. $250150 Based on the performance curve from the original vendor on file and the pres-

sure calculations for the configuration used for this fired heater, the blowershould be able to supply the air needed for the new conditions. There are noapparent blockages in the line. If didn’t put on safe-park as first step, thendangerous potential fire/explosion conditions are created while you experi-ment. $500 000

152 Trace, trace, trace, 10% v/v, trace, trace, trace, 8% v/v, trace, trace. $10 000153 Four stage, reciprocating with kickback protection at the exit of the third stage.

10% overdesign. $800155 Decreasing. $50156 No observable fouling on either inside or outside although some specs of rust

are on the inside suggesting eventual buildup. When put back in operation,not change in control and temperature still cycles. $30 000

157 Six months ago, inspected internals of column. Corrosion negligible; every-thing looked OK. Pumps overhauled with routine checking; instruments cali-brated and control system checked and retuned, if needed.

159 Town-gas pressure 1.1 MPa is < steam pressure, 2.8 MPa. $400160 Heat loss through the metal hub insert is 20% lower than through other parts

of the mold simply because of the greater width. $600162 Moisture coming in because the dryers are not working/ steam leak in the

reboiler on the bottoms of the de-ethanizer/ town gas coming in too “wet”/steam leak from E310 into the town gas/ process gas entering the site is toowet/ column was not designed to handle this high a flow/ collapsed trays. $50

164 Should be OK; allowed for in the Dp across control valve FV-4. Suction sideOK, complete with vortex breaker. $300

166 Allowance for fouling on tube side= 0.0005 m2 K/W; shell-side steam= 0.0 0001. $15

168 No change. $5000169 No change in operation. Pressures and temperatures still increasing. $250170 P210 = 0.5 m; P220 = 34 m. $50

441


171 Diagram is missing most of the details. All control valves, for example, haveblock valves, drain, and bypass with valve. The process pump has isolationblock valves, pressure gauge; all pressure taps have a tee nipple and valve forflushing; pressure gauges on hot process stream have a pigtail before thegauge. The fuel line has a pressure gauge before the burner. Automatic emer-gency steam-out connections for set of tubes. Draft gauges upstream anddownstream of the damper. Observation ports in radiant and convection sec-tions. If didn’t put on safe-park as first step, then dangerous potential fire/explosion conditions are created while you experiment. $500 000

173 Steady and usual at 109 �C. $50176 No variation that we can detect. $120177 121 �C saturated corresponding to the pressure. $200178 Extensive manuals including maintenance schedule and table of trouble-

shooting diagnostics. $6000180 Not needed. $2000181 Just completed; this is first startup. $50183 On spec. No complaints from other customers for that batch. $350184 Responds to change. $1000186 Ammonia: NFPA health 3; flammability 1; reactivity, 0. Toxic, corrosive gas.

Overexposure can be fatal. Low dosage: irritation to nose and throat. > 5000ppm may result in rapid death due to suffocation or fluid in the lungs. Flam-mable in air for concentrations 15% to 28% v/v. Gas can ignite explosively ifreleased near an active fire. The explosive range broadens 1) if hydrogen ismixed with the ammonia and 2) at higher temperatures and pressures. Pres-ence of oil and combustibles increases fire hazard. Ignition energy > 0.68 J.Autoignition temperature 651 �C which is lowered from 842 to 651 �C by thepresence of iron. At atmospheric pressure, ammonia decomposes to hydrogenat temperatures > 450–500 �C. Gas has explosive sensitivity to static charge.Ammonia is highly reactive with most metals, especially mercury, gold or sil-ver compounds. Reacts violently with tellurium tetrabromide and tetrachlor-ide, chlorine, bromine, fluorine and with acid halides, ethylene oxide andhypochlorites. Hydrogen: 0, 4, 0; Dow 21; methane: 1, 4, 0; nitrogen: 0, 0, 0.$350

188 Temperatures and absolute pressures have been gradually increasing. $30190 UA=q LMTD. Actual= (F cp)E100 90/ 45 �C or 2 (F cp)E100. Design= (F cp)E100

96/ 38 �C or 2.5 (F cp)E100. Actual UA is 80% of design. $300191 IS: the operator of the machine several months ago when the product was

produced. $60192 Consistent with reading on temperature sensor. $500193 Pressure gauge reads 13 kPa abs. $1000195 IS: Overflow to DAF less than expected. $20196 Cycle 1, 3 min:’1: 5.1%; 5.6%;’3: 6.2%;’4: 6.7%;’5: 7.1%;’6: 7.6%;’7:

7.2%;’8: 7.7% ;’9: 7.9%. Cycle 2&4: 8.0%; 8.1; 8.8; 9.0; 9.2; 9.1; 9.7; 10.3;10.5. Cycle 3: 15.5; 15.5; 16.0; 17.3; 18.7; 19.9; 20.0; 20.1; 20.2. Cycle 5: 18.3;

442


19.4; 18.7; 19.8; 20.5; 22.3; 23.7; 24.2; 25.1. No data available for previousoperation. $4000

199 Flows and temperatures are constant and the usual values. $120200 Responds to change: shows condensation temperature of IPA for the pressure.

$2000201 Leveling out of the bottoms composition at 0.013% –15%. $500202 Characteristic noise of open relief valve. Valve open. Flashy flare suggests that

the pressure relief has opened somewhere on the site. $300203 Not needed204 About 200 ppm for each of the 24 h composite for each of the seven days. $70205 Could the baffles have been put in so that the windows were top and bottom

instead of vertical? That would really foul up the hydraulics, but would this beconsistent with the evidence? The fouling coefficients allowed are minimaland extra area was not allowed. When first put into service with clean tubes,the product would have been overcooled! $100

206 Behaves as expected; no surprises. $2000208 Fully closed. $300210 Responds to change. $200211 Bottoms temperature increases by 3 �C; bottoms composition is 8% organic;

flowrate from pump 114-J of overhead is still negligible.212 Reverse jet bag filter; air:cloth ratio 2:1. No bypass installed because this is

integral part of the conveying system. Bags replaced annually.213 No observable reasons. Inside looks a little worn but otherwise OK. Impeller

looks OK and the key is in the keyway. $3000214 No change215 All the hydrocarbon vapors are very flammable. $150217 The reactor can overheat if the flow of reactants is too small. This not only

does not remove the heat of reaction but it also extends the residence time.Therefore, if there is a hot spot, close valve A to increase the gas flow to thereactor. $600

218 IS: On furnace process exit stream. IS NOT: Elsewhere on this plant or others.219 Our lab analyses show variability around the specifications. $100220 No apparent recirculation. $120221 Vortex breaker is welded across the exit nozzle such that most of the cross sec-

tion is blocked. $10 000222 Trace. $10 000223 See Chapter 3: thickener, Section 3.5.9; pump Section 3.2.3; belt filter Section

Section 3.5.12. $20224 See Chapter 3: heat exchangers, Section 3.3.3. $50225 Feed forward should not cause unstable behavior. This is feedback control that

can become unstable with poor tuning. $50227 Not needed. $3000229 We make the acid on-site; there has been no change over the past ten years;

specs are the same as was used in the design.

443


231 Pressure and overhead temperature reduces slightly but not to the designvalue. Pressure control still sluggish. $8000

232 Why1? to get column operating on spec during the hot weather. Why2? to pro-duce quality product reliably and safely. Why3? to meet customer’s demands.Conclude: focus on “to get column operating on spec during the hot weather”.$20

233 No controls, other than the controls within the screen and centrifuge system.The bottoms from the crystallizer are pumped to the nozzle feed to the DSMscreen. The crystals exit from the screen and flows by gravity into the centri-fuge except when the screen is being washed (when there is no flow to thecentrifuge). The discharge from the centrifuge to the dryer is continuousexcept that there is no crystal flow during the wash cycle in the centrifuge.There is no spare equipment. The dryer runs continuously. $30

234 None, this is startup of new plant. $50236 The block valves on both inlet and outlet of the reflux pump are one size small-

er than the pipe size. Check with design plans show these should be the samesize as the pipe size. The equipment for the control system agrees with specs.A vent break is included on the exit liquid manifold of the condenser. For thereflux drum, the gas-liquid inlet nozzle is on the top and at the end of thereflux drum opposite to the top nozzle for the vapor outlet to flare and thebottom nozzle for the liquid. The diagrams show a demister pad and a vortexbreaker supposedly in the drum but we are unable to tell from the outside.$180

237 IS: Neutralizer temperature < 135 �C; product concentration 79%. IS NOT:Neutralizer temperature= 135 �C; product concentration 83%. $50

239 Flashing erratically. $500240 No change in conveying ability. Gauge reads 13 kPa abs. $55 000241 Responds to change. $300242 Most details are present on the diagram. Missing, however, are for the recipro-

cating pump, F1400: the isolation block valves and kickback pressure relieffrom discharge to suction; for the collection drum V1401, vent break anddrain valve; for connecting line between the condenser E1400 and coolerE1404, a block valve; for the collection vessel between the booster ejector andthe contact condenser, V1402, a drain valve and vent break; for the barometricleg from the condenser, E1402, a seal pot; for the wet vacuum pump, isolationblock valves. $40

243 Not needed. $2000245 P&ID supplied; no additional data available. P&ID is less than a year old. $30247 Responds. $120249 No improvement. $300250 Analysis 41% ammonia; 29% water vapor, 31% CO2 all –3%. Total 101–3%;

OK within error. $50251 Fan runs at correct speed; balanced operation; no wobble; pitch of blades

changes from negative to extreme positive. $10 000252 Butane in storage tank is “on-spec”. $5000

444


253 No change. Still upset and fluctuating254 Why1? to minimize the heat load in the boiler Why2? to supply steam for en-

ergy efficiency. Why3? so that our corporation continues profitably. Conclude:focus on Goal: “to heat the boiler water to 70 �C”. $50

255 No indication of a collapsed tray; negligible liquid in column. ($5000 for scan+ $1,600 for time for scan + $20 000 for lost time to arrange for scan) $26 600

256 Usual allowance for fouling on tube and shell side. Vent/blowdown lines onboth the tube and shell side. Vent/blowdown lines discharge to sewer. Eachvent line has a single gate valve that is normally shut. Process pressureis > steam pressure. Fuel-gas pressure, 3.1 MPa > 2.8 MPa steam. $200

257 Usual allowance for fouling on tube and shell side. Vent/blowdown lines onboth the tube and shell side. Vent/blowdown lines discharge to sewer. Eachvent line has a single gate valve that is normally shut. Town-gas pressureis < steam pressure. Fuel-gas pressure, 1.1 MPa < 2.8 MPa steam. $1000

258 Fully open. $300260 Requires three stages of compression to bring the feed gas up to pressure of

34.5 MPa but only one stage of the compressor to pump the recycle gasaround the loop. $500

262 Not needed263 No upsets here. Flow, temperature and concentration should be the usual val-

ues with temp. = 108 �C; concentration 41% ammonia; 29% water vapor, 30%CO2. $50

264 Because the overheads from the demethanizer are essentially “town gas” theoverhead is used as an on-site source of regeneration gas for the dryers. Moregeneral information is in Chapter 1 of Process Design and Engineering Practice(1995) by Woods. $15

265 No breaker. No plans for one in the design. $1000266 Not needed. $10 000267 No change; when on automatic we have an unstable system whose oscillations

are ever-increasing. $1600268 Not needed. $24 000269 None recently. $50270 When: not since the turnaround.271 Not needed. $650272 Usual hydrocarbons. Flammability 2. $50273 Fatty acids are pumped in the bottom of the column C1400, rise through the

tubes of the reboiler. The heating medium on the shell side is Dowtherm. Thevapors are condensed in E1400, with the overhead product dropping intoV1400 where it is subcooled by the coil E1404. The product is removed fromthe system by reciprocating pump F1400. The distillation occurs under a highvacuum supplied by a booster ejector and a single-stage steam ejector F1402and F1403 with an interstage barometric, direct contact condenser E1402.Condenser E1400 is a closed circuit of boiling-condensing water. The tempera-ture of the boiling is controlled by setting the pressure P2. The backup con-denser, E1401, merely condenses the more volatile contaminants that have

445


been removed from the fatty acids. These are drained off periodically fromvessel V1401. Cooling water comes from the header and some flows intoE1401 to condenses the volatiles, and some flows into the tank in which thesteam coil is submerged, E1403. The bottoms of column C1400 recycles tosecond column. Both the recycle lines and the column are not shown. Theside-draw above the reboiler on Column C1400 recycles to the feed. This alsois not shown on the diagram. $250

274 IS: thickener-pump-belt filter system. $50276 The valves look OK. Went ahead and changed them as ordered. No change in

performance. $6000277 Nothing unexpected. $200 000278 This has been a bear to operate. The overhead temperature has been increas-

ing gradually and I have been increasing the reflux accordingly but the tem-perature keeps increasing. Yet the analyzer is showing the usual overheadcomposition. $50

279 Standard ones we have always used. $150280 Alumina. cycle time 12 h; two in series on-line; one off-line being regenerated

with fuel gas. Fuel gas should have < 6 ppm moisture. Alumina loading 0.14–0.22 kg water/kg dry adsorbent. $400

281 Hollow sound that changes to a deeper sound consistent with liquid level ris-ing in the effect. This corresponds to the cyclical drop in temperatures and thegradual increase in pressure. $600

282 Both fluctuating but consistent.283 Fuel oil: NFPA 1, 2, 0. Explosive limits in air 0.5 to 7% v/v. Flash point > 53 �C;

can react vigorously with oxidizing materials. Process fluid: NFPA 1, 3, 0.284 Single-speed motor with variable pitch on the blades. Sensor measures the

temperature of the exit IPA liquid. $1000286 Allowed 130 kPa Dp for the control valve. NPSH required= 2 m water. $200287 Was OK. Recalibration not needed. If didn’t put on safe-park as first step, then

dangerous potential fire/explosion conditions are created while you experi-ment. $500 000

288 Checks suggests looks OK provided the gas-liquid separation is sufficient andthat there is no carryover to flare. Maybe check location of nozzles when on-site. $200

289 Oscillating around 65% and increasing; usually 62% and steady. $600290 15 �C cool and windy; this week started off mild, about 18 �C but cooled down

as a cold front moved in. $30.291 Red, indicating stopped. $200. If you didn’t put the plant on SIS or SIS + eva-

cuation before you asked this question, the plant explodes with loss of life.Penalty $3 000 000.

292 Clear. No obstruction. $6000293 No files for this unit. $50294 They appear to be open. $50295 No improvement. $15 000

446


296 Operators downstream report:. Temperature= 127 �C; should be 130 to 140 �C;concentration= 78% and should be 83%. Our plant: temperature= 129 �C inthe neutralizer; concentration= 79%. Reasonably consistent. $200

298 Not needed. $6000300 Polycarbonate, ABS Styrenics. NFPA: 0, 0, 0. $60301 No upsets. Feed rate and composition are within specs. $60303 Decreasing from 1.73 MPa. $500304 Responds to change. $300306 Responds to change. $300. If you didn’t put the plant on SIS or SIS + evacua-

tion before you asked this question, the plant explodes with loss of life. Penalty$3 000 000.

307 Based on the measured flame temperature in the radiation section, the tem-perature is not sufficient to heat the process liquid. If didn’t put on safe-parkas first step, then dangerous potential fire/explosion conditions are createdwhile you experiment. $500 000

308 Not needed. $5000311 Controlling flow via blade pitch is “slow and clumsy”. $1000312 The pressure on the head tank must be > 22 kPa g because if it is less than this

pressure, then the suction of the downstream compressor is subatmospheric.This would allow air to leak into the ethylene. The vessel includes a demister,a vortex breaker. The design provides enough residence time for good liquidlevel control and for good gas-liquid separation.

313 Confirms that the temperature read on TI 202 is high. $1200315 Not needed. $3000316 IS: operators on the debut; IS NOT: operators on other plants. $50317 See Chapter 3: thickeners, Section 3.5.9; sensors and control Section 3.1.3;

pumps, Section 3.2.3. $50318 Largest particle is 1.3� aperture and is log normally distributed with a geo-

metric mass weighted average of 0.6� aperture size and a geometric standarddeviation of 2.1. No data available for previous operation. $1800

319 Not needed. $10 000320 NFPA ratings are: methane, 1, 4, 0; propane, 1, 4, 0; propylene, 1, 4, 1; hydro-

gen, 0, 4, 0; ethylene, 1, 4, 2; Explosive limit for hydrogen 4.1–74.3% in air.$200

321 I did that but either the controller isn’t responding, or the meter reading iswrong. The reflux is not increasing! $50

322 Yes. $20323 105 kPa gauge. Our initial assumption of 200 kPa g was wrong. This changes

the estimates made in the office. $200324 Single-speed motor with variable pitch on the blades. Sensor measures the

temperature of the exit from the condenser. $70325 Noisy from the fans and air flow. Nothing unusual. $150326 Loud knocking noise. $60327 Accurate; no recalibration needed. $3000328 Based on the values available this should flow well. $300

447


329 49 �C. $400330 Not needed. $24 000331 Diagram reasonably consistent with what’s on the plant. No major surprises.

$200332 IS: not known when this started. $50333 Continuous. No break or short circuit in heater. Heater should work. $23 000334 I’m a new operator. I thought I was doing the right thing by increasing the

pressure. I certainly didn’t expect everything to go wild. $50335 Closed cycle, standalone unit. Should do the job. $700336 Robust unit. No packaging problems are common.337 IS: flowrate meter reads about 80% of design rate when pump A is pumping.

100% design rate when pump B pumping. IS NOT: pump A at 100%338 Noise from the fans and air flow. Nothing unusual. $2000339 Health Rating: 3 – Severe (Poison) Flammability Rating: 2 – Moderate; Reac-

tivity Rating: 2 – Moderate; Contact Rating: 4 – Extreme (Corrosive) Inhalationof concentrated vapors may cause serious damage to the lining of the nose,throat, and lungs. Breathing difficulties may occur. Neither odor nor degree ofirritation are adequate to indicate vapor concentration. Ingestion: Swallowingcan cause severe injury leading to death. Symptoms include sore throat, vomit-ing, and diarrhea. Ingestion of as little as 1.0 ml has resulted in perforation ofthe esophagus. Skin Contact: Contact with concentrated solution may causeserious damage to the skin. Effects may include redness, pain, skin burns.High vapor concentrations may cause skin sensitization. Eye Contact: Eye con-tact with concentrated solutions may cause severe eye damage followed by lossof sight. Exposure to vapor may cause intense watering and irritation to eyes.Chronic Exposure: Repeated or prolonged exposures may cause darkening ofthe skin, erosion of exposed front teeth, and chronic inflammation of thenose, throat, and bronchial tubes. Aggravation of Pre-existing Conditions: Per-sons with pre-existing skin disorders or eye problems, or impaired respiratoryfunction may be more susceptible to the effects of the substance.

340 Yes, put on safe-park. Soot buildup in the furnace could combust at a latertime when excess air is present.

341 The steam is more superheated than usual but we have no sensors on the lineso we cannot tell actual temperature.

343 about 400 kPa; ammonia < 20 kPa g to have ammonia temperature < 30 �C.Therefore process leaks into ammonia.

344 Contamination near the hub/ moist resin/ uneven temperature over themold/ injection too slow/ injection too fast/ faulty design of the product/faulty design of the mold/ too much cooling/ too little cooling/ faulty resin/incorrect foaming agent/ correct foaming agent but wrong concentration/inadequate mixing/ cooling cycle too long/ cooling cycle too short/ feed tem-perature too low/ mold too cold. $200

345 I can manage by operating on manual but I have other things I must do. $120346 Recalibration not needed. $3500

448


347 Minimum number of bends; relatively straight and any bends have long radiiof curvature. $2000

348 Extensive leaks in 10 out of 340 tubes. $20 400349 Water: NFPA: 0, 0, 0. $50350 Both valves leak when fully closed.% leakage varies from valve to valve and

ranges from 2 to 3.2%. $40 000351 Sensor OK. Calibration checked out OK. $8000352 Ammonia: NFPA health 3; flammability 1; reactivity, 0. Toxic, corrosive gas.

Overexposure can be fatal. Low dosage: irritation to nose and throat. > 5000ppm may result in rapid death due to suffocation or fluid in the lungs. Flam-mable in air for concentrations 15% to 28% v/v. Gas can ignite explosively ifreleased near an active fire. The explosive range broadens 1) if hydrogen ismixed with the ammonia and 2) at higher temperatures and pressures. Pres-ence of oil and combustibles increases fire hazard. Ignition energy > 0.68 J.Autoignition temperature 651 �C which is lowered from 842 to 651 �C by thepresence of iron. At atmospheric pressure, ammonia decomposes to hydrogenat temperatures > 450–500 �C. Gas has explosive sensitivity to static charge.Ammonia is highly reactive with most metals, especially mercury, gold or sil-ver compounds. Reacts violently with tellurium tetrabromide and tetrachlor-ide, chlorine, bromine, fluorine and with acid halides, ethylene oxide andhypochlorites. Hydrogen: NFPA: 0, 4, 0; Dow 21. Extremely flammable. Explo-sive limit for hydrogen 4.1–74.3% in air. $3000

353 Responds to change. $200354 Responds to change355 No apparent fouling. $20 000356 Reads as expected $150359 Signal is 0%. This is a fail-close valve so the valve should be wide open. $300360 Exit moisture content in crystals from dryer increases to 6%. $3000362 Level above level of tubes; design level. $50363 Dry reciprocating vacuum pump. Specs suggest that a series of pumps in par-

allel should provide the vacuum plus reserves for on-line maintenance antici-pated because of the high probability that ketene will dimerize into a gunkygoo that will have to be mechanically removed from the pumps periodically.Vacuum must be kept constant.

364 68 �C. $400365 Steady and usual design value. $50366 Other customers were shipped material produced with that number. No, we

have not received any complaints or comments about “clumpiness”. Ourrecords show that the material more than satisfied our specifications. $3000

367 Overhead product within spec most of the time. $1800368 Calibrates OK. $1200369 45 kPa. Usually 54 kPa. $600370 Tab marking unclear; cannot tell.371 Check out OK. Accurate. $20 000373 Product breaks. $800

449


375 Responds to change376 No major hazards with fluids being pumped.377 Downstream complaints of insufficient product. $1200378 No improvement. $35 000379 Typical hydrocarbon NFPA: 1, 3, 0. $50380 Allowance for fouling on tube side= 0.0 0001 m2 K/W; shell-side steam

= 0.0006. $250381 At the usual mid-range value before and after safe-park. If didn’t put on safe-

park as first step, then dangerous potential fire/explosion conditions are creat-ed while you experiment. $500 000

382 Looks OK.384 69 s. $200385 Hydrogen has relatively high heat capacity; about 10 kJ/kg K compared with

about 2 for ammonia vapor, about 1 for nitrogen; hydrogen thermal conductiv-ity is about 0.2 W/m K compared with about 0.025 for ammonia vapor andnitrogen. The Pr numbers are similar. Hence, the thermal properties of a mix-ture of hydrogen, nitrogen and ammonia depend on the gas composition.$6000

386 Higher than usual. $20387 Steady and usual. $50388 Agrees within –10%. $1200389 IS: safety relief valve popped? IS NOT: any other sensor signal. $50390 160 �C. $30391 Keep the gas pressure drop < 1.5 kPa. Replace bags annually.392 Exit-gas composition from the reformer will have less hydrogen than you had

previously. Otherwise, should work very well; negligible coking; good conver-sion and selectivity. $800

393 Reads 67 kPa abs; at other times it might read between 16 and 33 kPa abs.$300

394 Responds. $250395 Usual type. $15396 No mercury used. $300397 Agrees.398 IS: mass balance does not balance on the IPA. IS NOT: all other conditions

except change in type of condenser seem to be the same. No change in sup-plier of feed. $650

399 Not needed. $15 000400 Needed. $200401 Negligible change. $1000402 No change; still cycles. $1200403 533 –10% L/s. $90404 Tray collapsed in stripping section; vapor space above tray 5 seems “more

dense” than expected vapor space. Indeed, more dense vapor space was foundbetween other trays in the stripping section suggesting weeping. $5000

405 No. $30

450


406 Yes, We can ensure that the inerts don’t exceed 15% by opening the purge linein the recycle loop in each separate loop. $500

407 Clear. No obstruction. $1000408 Vent hole is clear; orifice and seat look OK; mechanism looks OK. Upon

restart, cycling and poor control continue. $46 000409 No change in operation. Pressures and temperatures still increasing. $250410 5 �C. $2000411 Rain: load= latent heat� all condensed + all condensed � 2 kJ/kg K � 3 �C. Hot:

load= latent heat � partial condensed. Cannot do any more calculations basedon the information in the problem statement.

412 Steady at 1.73 MPa. $500413 Calibrates OK. $300414 No files for this unit. $50416 Not needed. $500417 Flashing; not usual. $50418 Not needed.419 Temperature is 13 �C; flow is steady420 On this older part of the site, the water flows in drains that join the main

drain at “drop boxes”. The main drain goes to disposal drain. The drop boxesallow mixing and sampling of the water. Rain water also flows from variousdrain across the unit into the drain system. This is an older part of the site; pand s-traps have not been installed. The safety inspector checks on “explosiv-ity” of the vapor in all the drop boxes on site at least once per shift. $15

421 This is the commissioning startup of a new unit. $50422 About 50%. this is the design amount. We really want to find a good use for

the off-gas and minimize the use of “pure” ammonia. $600423 Ethylene: NFPA: 1, 4, 2; butylene: 1, 4, 0; methyl chloride: 2, 4, 0; ammonia: 3,

1, 0; water 0, 0, 0.424 Some improvement. Concentration of propane in flare gas decreases to 80%

above design. $5000425 I tried to keep the cycle time, the pressures and temperatures the same as we

used for the prototype. $150426 Stiction is the sticking and friction related to valve movement and measured

as the difference between the driving values needed to overcome static frictionupscale and downscale. Likely cause of small amplitude, continuous cycling.$50

427 Operates as flooded underflow. Underflow returns the particles to crystallizer;overflow returns to pregnant liquor storage. Dp and body shape seem well de-signed. $250

428 Heat loss based on given data of process gas gives calculation of the steamgenerated under actual conditions as= about 70 kg/s; consistent with gas cool-ing being only 90 �C instead of 150 �C

429 Warm and cloudy 20 �C. Previously this week it was hot and clear. $50431 Varies. Sometimes trap hot with condensate intermittent discharge; other

times, cold trap and long time between discharges. Sporadic. $500

451


432 Robust unit. If the mixing time� speed of rotation= 300 min. rpm then shouldbe OK.

433 Yes, I have noticed that AC/1 shows variation. $100435 Top temp –18 �C usual; bottom temp same as usual. $200436 Indicating slightly higher concentration than usual. $50437 P&ID for the columns given in Case’24. $1000438 Not needed. $50439 Calibrates OK; recalibration is not necessary. $3000440 Not needed. $4000441 Equipment should do the job.442 Difficult because the flowrates are continually changing during the upsets.

$50443 Lower than expected. If didn’t put on safe-park as first step, then dangerous

potential fire/explosion conditions are created while you experiment. $500 000444 Usual allowance for fouling on tube side and shell sides. Vent/blowdown lines

on both the tube and shell side. Vent/blowdown lines discharge to sewer. Eachvent line has a single gate valve that is normally shut. Process pressure is > uti-lity pressure. E107: Process feed, 3.3 MPa> propylene refrigerant pressure.E108: Process feed, 3.29 MPa > ethylene refrigerant pressure. $200

445 Probably should only be considered when overhead process-gas temperatureis < 50 �C; response is usually sluggish because condenser lag is in the controlloop. Watch for fouling on the cooling-water side. $200

446 Why1? to start the production of ammonia in the reactor. Why2? to supply thedemand for ammonia. Why3? so that our corporation continues profitably.Conclude: focus on Goal “to heat up the reactors to 500 �C”. $50

447 Hot 35 �C and humid: 80% relative humidity. Four days ago, it rained and wasmoderate, 28 �C. Since then, it has cleared and gotten hotter and hotter. $650

448 The stuff we are receiving is not acceptable. Temperature= 127 �C; should be130 to 135 �C; concentration= 78% and should be 83%. $120

449 No level. $50450 Not needed. $50451 Usual allowance for fouling on tube side and shell sides. Vent/blowdown lines

on both the tube and shell side. Vent/blowdown lines discharge to sewer. Eachvent line has a single gate valve that is normally shut. Process pressure is > uti-lity pressure. E107: Process feed, 3.3 MPa> propylene refrigerant pressure.E108: Process feed, 3.29 MPa > ethylene refrigerant pressure. $1000

452 Missing. Perhaps misfiled. $6000453 Valves OK. No evidence of corrosion, not plugging of valves. No leakage, valve

stem moves easily. $2000454 No change. $4000455 No noise sounding like cavitation. $200456 Flashing.457 Within specifications. $35000458 0.532 MPa and increasing. $500459 OK. Recalibration not necessary.

452


460 Increasing. $50461 10.1; 10.5; 9.9% w/w on three samples. $10 000462 (E114) Usual allowance for fouling on tube and shell side. Vent/blowdown

lines on both the tube and shell side. Vent/blowdown lines discharge to sewer.Each vent line has a single gate valve that is normally shut. Process pressure> ethylene refrigeration pressure. $250

463 Should be OK. allowed for in the Dp across control valve FV-1. $100464 OK, recalibration not necessary. $700465 Estimate suggests no major blockage. $1200466 Utility pressure > atmospheric. $200467 IS: cyclical and –10 �C. IS NOT: steady and –1 �C. $150468 Centrifugal pump. Designed for range of flows and to handle the vertical

height. $200469 Was OK. Recalibration not needed. If didn’t put on safe-park as first step, then


470 levels lower than expected. $20471 Control should work well or should be tested. $500472 Whistling noise; no detectable “vibration” noise. $3000473 Naphtha: NFPA ratings for Health, 1, flammability, 3, spontaneous 0; Dow

rating 16; cyclohexane: 1, 3, 0; Dow 16; benzene: 2, 3, 0; Dow 16; hydrogen: 0,4, 0; Dow 21; hexane: 1, 3, 0; 2 methyl pentane: 1, 3, 0. $150

474 Sharp-edged orifice sized so that the design flow corresponds to conditions fora constant drag coefficient through the orifice.

475 As expected. $3000476 Not yet but if trouble shooter takes a long time then this should be done.477 Some condensate but mainly wet steam. $400478 27 actual trays; anticipate lower tray efficiency in the rectification because of

inerts; also allowed two extra trays. Simulation shows that unit was operatingvery close to design values for usual range of feedstocks and conditions. $350

479 Not needed. $6000480 Slightly worse operation. $800481 Thermodynamic traps on all three locations: reboiler, turbine and preheater.

No bypass supplied; upstream strainer. $300482 Suggested that a common fault is a faulty steam trap causing condensate to

build up in dryer and reduce effective area for heat transfer. $120483 Pages 4-90 to 93 in Woods “Process Design and Engineering Practice,” Pren-

tice Hall, 1995. Gives economics, simplified P&ID and description (althoughthe reaction temperatures in the current problem are higher than thosedescribed in this references). $500

484 Extensive manuals including maintenance schedule and table of trouble-shooting diagnostics. $6000

485 Yes, all lines insulated. $1000486 The pump F1400 seems to pump some of the product out and then the flow

seems to stop. $30

453


487 Fan changes but a very slow response time. $600488 Why1? to operate all columns at 150 Mg/d safely and on spec products. Why2?

to supply the demand for ethylene elsewhere on-site. Why3? so that our cor-poration continues profitably. Conclude: focus on Why1? “to operate all col-umns at 150 Mg/d safely and on spec products”. $50

489 If the flowrate and reactant concentration to the reactor are too low, then thereaction quickly reaches equilibrium before the reactants leave the bed. $600

490 Not needed. $2000491 Cold –10 �C. Indeed this whole week has been cold, with snow flurries two

days ago. $50492 When: before the current operation that has problems with the moisture con-

tent in the dryer outlet. What was done? To improve the washing, the diameterof the wash pipe and the associated washing pump system were altered sothat the flowrates of wash water were increased by 35%. The washing timewas to be kept the same, 3 min. This was done. No changes to the dryer or tothe screen. $30

493 Greater than usual. $50494 Warm, 23 �C, dry, windy. Similar weather all week. $20495 Fan very noisy. No improvement in yield of IPA. Calculated yield about 69.8%.

$20 000496 IS: Operators on other units. IS NOT: operators on reformer unit. $150497 Overhead pressure on the column is controlled by the amount of non-conden-

sibles sent overhead from the reflux drum. The temperature of the condensedoverhead is controlled by adjusting the pitch on the blades of the fan. Thereflux rate is adjusted by the level in the reflux drum. There is no overheadliquid product. $20

498 Problem still occurs. $2000499 Since cooling water or air cannot be used as the coolant, various “refrigerants”

are used: ethylene, propane, propylene. Each refrigerant is part of a refrigera-tion unit consisting of the four parts: 1) an “evaporator-condenser”, 2) a refrig-erant compressor, 3) a refrigerant condenser and 4) a let-down valve. In theevaporator-condenser, the liquid refrigerant inside the tubes evaporates andcausing the process vapor on the shell side to condense. To allow the refriger-ant to be reused in a closed cycle, the refrigerant vapor is then compressed,condensed and the resulting liquid reduced in pressure to return around therefrigeration cycle. The pressure on the refrigerant side of the evaporator isadjusted to provide the correct temperature driving force for condensation ofthe process vapors. More general information is in Chapter 1 of Process Designand Engineering Practice (1995) by Woods. $100

500 Demister should remove mist droplets > 500 mm. $500501 Guarantee flow needed. Typical moisture content < 6 ppm. Heating value 37

MJ/m3 $1000502 18 �C. This agrees with design assumptions. $400503 We followed the startup procedure carefully since we have a new catalyst with

lower recycle rates of hydrogen and higher feed flows of naphtha. We think we

454


are doing a good job of maintaining the same space velocity in the reformer.$150

504 Operation went smoothly. No problems. No different from a month ago.505 Steady at 116 �C. $300506 Remote. $15507 Accuracy of float sensor –0.5 cm; accuracy of point gauge –0.045 cm. $120508 Nothing unexpected.509 About two to ten minutes. $200510 Should do the job. $100511 See Chapter 3: Section 3.9.6. $150513 Steady and usual. $50515 27 actual trays; anticipate lower tray efficiency in the rectification because of

inerts; also allowed two extra trays. $350516 3 days ago completed the annual work on the baghouse: new bags installed.

Reverse jets had been serviced.517 No change; still cycles. $1000518 All fully open (according to the handle and the valve stem). $20519 27 actual trays; anticipate lower tray efficiency in the rectification because of

inerts; also allowed two extra trays. $350520 Steam inside tubes at 12 MPa with gas on outside at 4.5 MPa; steam leaks into

process gas.521 Same as expected; 3/4 full and reasonably steady. $1200522 Responds to change. $20523 Not needed. $8000524 Some product drains off but quickly stops; prime is lost and pump fails to

pump.527 10 ppm. $4000529 Estimate seems OK. No major blockage. $300530 1.48 Mg/m3. $100531 1.04 Mg/m3; normal boiling temp. 134 �C. $100. If you didn’t put plant on SIS

or SIS + evacuation before you asked this question, the plant explodes withloss of life. Penalty $3 000 000

532 For the reduced flowrate specified in the startup manual and for the gas com-position of 3:1 hydrogen to nitrogen, the cal rod heaters should bring the reac-tor temperature from 300 �C up to 500 �C in four hours. 8 am to 12 noon=4 hours. $6000

533 See Chapter 3: crystallizer, Section 3.4.3; dryer, Section 3.5.5; hydrocyclone,Section 3.5.8; filtering centrifuge Section Section 3.5.11; screen, Section 3.5.6.$30

534 No idea. Good questions. That’s something we do not monitor, nor do wehave records. I just know that the control of the pressure always tended to besluggish right from the beginning! $150

535 See Chapter 3: vacuum systems, Section 3.2.2, solids conveying, Section 3.2-4;hopper design, Section 3.10. $2000

455


537 IS: Everything seems to be the usual behavior. IS NOT: different than usual.$50

538 Operation went smoothly. No problems. No different from a month ago.540 Full documentation is available. $15541 Allow 170 kPa across the steam control valve, gives pressure in coil= 1530 kPa.

Inverted bucket sized on Dp= 1530 – backpressure in condensate header of200 kPa= 1330 kPa. Condensate= 0.12 kg/s and select based on 8 times or1 kg/s. Strainer included.

542 Well designed; trim selected carefully for gaseous chlorine; should operatemid-position for design flowrate. $100

543 Yes, excessive fouling. $4000. If you didn’t put the plant on SIS or SIS + eva-cuation before you asked this question, the plant explodes with loss of life.Penalty $3 000 000.

545 No feed to the dryer during the centrifuge wash cycle either before or after thechange. $700

546 Hot summer day. Same weather all week. $50549 Should be OK; allowed for in the Dp across control valve FV-4. Suction side

OK, complete with vortex breaker. $300550 P210 = 28 m; P 220 = 22 m. $50551 Not available. $15552 Nothing has changed but the problem seems to be getting worse as the shift

progresses. $50553 Systems respond as expected. If didn’t put on safe-park as first step, then dan-

gerous potential fire/explosion conditions are created while you experiment.$500 000

554 Explosive ranging from 170 to 195% of the lower explosive limit for all testsand at all times. $700

555 See Chapter 3: reactor Section 3.6.2; exchangers/boilers Section 3.3.3556 About 78% design rate and decreasing. usually at current production rate this

would be 87% design rate and steady. $1000558 Cycle 1, 10 min:’1: 3%;’2: 2.9%;’3: 3.3%;’4: 3.8%;’5: 4.2%;’6:

4.7%;’7: 4.9%;’8: 4.9%;’9: 5.3%;’10: 5.2. Cycles 2 and 3 similar:’11:5.4%;’12: 5.7%;’13: 6.0%;’14: 6.1%;’15: 6.3%;’16: 6.4%;’17:6.6%;’18: 6.9%;’19: 7.1%; and’20: 7.2%. $3500

559 No upsets. We can supply on-spec feed for up to 160 Mg/d production with noproblem. $300

560 Downstream production steady; no fluctuations.561 No feed to the centrifuge during the screen-wash cycle either before or after

the change. $1200563 Zero. $50564 IS: other units. IS NOT: on reformer unit. $150566 IS: past few months. Gradual not sudden. IS NOT: previous year although the

slow response has been present even then. $50567 Area sufficient to do the job with the following fouling factor allowances: shell

side 0.0002; tube side 0.0005 m2 K/W. Considered propylene boiling to range

456


over nucleate and film boiling shell side 0.0002; tube side 0.0005 m2 K/W.Indeed, since this is a clean bundle, we would expect overcooling when wefirst start up because the allowed fouling does not interfere with the heattransfer. The observed overcooling is expected. $200

569 Equipment should do the job.570 No. I have operated this plant for six years and have never seen this before.

$50571 Fluctuates. $200572 Responds to change.574 IS: bottoms and pump 114-J. IS NOT: other parts of the unit.576 Commissioning of the compressor last summer went well. All systems per-

formed well regardless of the variation in atmospheric pressure and the usualrange of flare-gas compositions. The seal pot used water since it was summeroperation. Recommended change over for October to May to keroseneobtained from the crude unit. $500.

577 OK. No calibration needed. $30 000579 515 –10% L/s. $85580 Respond to change. $500582 Estimate suggests no blockage. $400584 Both start to decrease.585 As expected.586 Responds to change. $100588 Small amounts of rust flecks but not enough to block strainer. Removed what

was there. Upon restart, cycling and poor control continues. $6000589 Sulfinol removes the carbon dioxide from the reformer gas to produce a rela-

tively pure feed of nitrogen-hydrogen to the ammonia synthesis reactor. Thesulfinol needs to be regenerated occasionally. This is the cleanup column.

591 Allowance for fouling on tube side, water= 0.0002 m2 K/W; shell-sidestream=0.0005. Care to prevent temperature cross-over. $400

592 Should work fine. Head-capacity curves on file. rpm 1800. $500593 Temperature about 19 �C. $1000594 4.2 MPa. $150595 Clean, trim looks fine; snug bonnet. Size is two sizes smaller than line size.

$1000596 Why1? to prevent possible shutdown. Why2? to keep plant operating safely

and prevent upsets in operation. Why3? so that plant can handle the wastewater it receives. Conclude: focus on “to keep plant operating safely and pre-vent upsets in operation”. $50

597 Why1? to safely recycle vapors for reprocessing and prevent them from goingto the flare. Why2? to convert vapors into valuable products and keep theneighbors happy by preventing flares. Why3? to maintain good communityrelationships and keep the company economically viable. Conclude: focus on“to safely recycle vapors for reprocessing and prevent them from going to theflare”. $50

598 Upstream and downstream distances consistent with good design.

457


599 Mid-point. $300600 Was OK. $2000601 Steady and usual design value. $50602 Both motors start. Flow, from FRC/ –100, increases when each is turned on.

$30603 Within –10%. $700604 Nothing changed from usual operation except the liquid level in the floccula-

tion tank is not at the normal level. $20605 Agrees with head capacity. $200606 No change. $600607 OK, recalibration not necessary. $700608 Specialized design. Should do the job. Some designs use a standby furnace to

bring the system up to temperature. This design uses cal rod heaters. Thermo-couples in the catalyst bed and in the feed gas entering the bed. Temperaturesensors at four levels in the catalyst bed. $750

609 Allowance for fouling on the shell side= 0.0 0001 m2 K/W; on the steamside= 0.0006. Use U tube instead of fixed tube sheet, because of thermalexpansion. Vapor-liquid ports joining the shell sides were designed for maxi-mum upward butane flow and full condensate downflow based on downcomervelocity considerations to give annular flow. $50

610 OK, recalibration not necessary. $800611 Fluctuates. $200612 Wet and rusty flecks come out with steam. $1000613 For some flowrates the pressure differential is such that the pump can be

bypassed. The pump is adequately sized to handle higher flowrates.614 Almost closed. $200615 Valve stem responds as expected. $230616 Responds to change. $200617 No noticeable change in sound. $800619 Steady at 116 �C. $300621 IS: Overhead of column.622 Designed to provide good residence time; sufficient disengaging space; suffi-

cient area for vaporization, liquid level so that the coil is submerged. Coil areasufficient and designed for film boiling.

623 No change, still see weld lines. Product breaks. $900624 Purge specified in the manual should maintain the desired level of methane

in the gas. $6000625 See Chapter 3: instruments Section 3.1.3, compressor, Section 3.2.1, knockout

pots, Section 3.5.1, $150627 No leaks observed.630 Consistent with overhead temperature and the overhead specs; increased over

usual. $100631 Lumps persist. $6,500632 Equipment should do the job. $2000

458


633 Sample 1: 13%; sample 2: 11%; sample 3: 12%; sample 4: 14%; sample 5:13.5%. $20 000

635 Calibrates OK. $1500636 Usual design. $400637 Why1? to continue to operate the plant without being shut down by the inspec-

tor. Why2? to meet production deadlines. Why3? to keep the company eco-nomically viable. Conclude: focus on “to continue to operate the plant withoutbeing shut down by the inspector.” $30

640 After one hour, the temperature has risen 50 �C to 375 �C. $30 000641 Pump should do the job. Based on current PI pressure the pump should be

pumping 11.5 L/s. $50642 Steam-gauge pressure upstream of control valve= 230 kPa for saturation tem-

perature of 136.8 �C. This would give a DT of 136.8–109 = 27.8 �C. However,with an estimated Dp across the control valve of 30 kPa, then the DT= 25 �C ifthe steam was saturated at this pressure. Expect wiredrawing so the steam willbe slightly superheated until it reaches saturation. The DT seems to be in thenucleate boiling range. $200

643 Expected value. $50645 Steam leak into the acid. $90647 Should do the job.648 Working OK sometimes. Sometimes the flowrate= 14 L/s. Other times the

flowrate < 14 L/s. What’s going on? We expect reliable flow from you. $20650 There are no standard operating procedures. However, I have been operating

this unit for many years so I think I understand its peculiarities although itnever has worked correctly. However, today it’s much, much worse than I haveencountered before. Started it up as usual; carefully set values at usual values;ensured the steam tracing was on because it’s so cold out there. But it justwon’t draw off the product.

651 Unable to control temperature with cooling-water feed valve full open. $300. Ifyou didn’t put the plant on SIS or SIS + evacuation before you asked this ques-tion, the plant explodes with loss of life. Penalty $3 000 000.

652 Yes. From past records, the timing varies and the concentration values differbut surges in high lab values for the bottoms “correspond with” surges in highvalues of C3 on AC/1. $650

653 Improvement, extension of the period before break but still not satisfactory.$1000

655 Behaves as expected; no surprises. $2000656 No change. $500657 No contaminants. $3500659 The setting on PIC/10 should be such that the column pressure is 1.7 MPa

but does not exceed 1.8 MPa, the relief pressure for PSV 1. $50660 After two hours, no increase in temperature above 325 �C. $45 000661 Process fluid leaks into furnace. If didn’t put on safe-park as first step, then


459


662 Vacuum leak/ variation in steam pressure to booster or other ejectors/ liquidnot subcooled enough and it is flashing in pump F1400/ wet vacuum pumpnot pumping at capacity/ leak in exchanger E1400 causing water to flow intovacuum system/ instrument error/ more volatile material in feed/ vapor lockin suction line because of no vent break/ leg from barometric condenser notsealed/ booster not working right and air is sucked down from the boosterinto the pump F1400/ dry vacuum pump F14 put into service without 30-min-ute warmup.

663 No improvement. $6000664 See Chapter 3: distillation Section 3.4.2, perhaps pertinent condensers and

reboilers, Section 3.3.3; pumps, Section 3.2.3 and controllers 3.1.1. $50666 Already insulated. $500667 Polypropylene: melt point 162 �C; melt flow rate 35 g/10 min; Izod impact

strength, 70 J/m (1.2); HDT 105 �C; barrel temperature 180 to 220 �C; nozzletemperature, 200 to 220 �C; melt temperature 200 to 240 �C; injection pressure60 to 80 MPa; mold temperature 15 to 50; target 25 �C. $100

668 Was OK. $2000670 Equipment should do the job. $2000672 Ethylene leaks into the butane. $50673 Steady and usual. $50675 We are receiving more because of the higher through put you are using. We

haven’t checked our knockout pots on the fuel gas lines to see if there is anycarryover. $300

676 No difference. Solution meets specifications. $1000677 Not needed. $1000678 The check valve on the exit of pump B is faulty. The other looks OK. Went

ahead and changed both of them as ordered. Performance back to normal.$6000

679 IS: on top of column. IS NOT: elsewhere on the unit. $50681 OK recalibration not necessary. $800683 About 93% design rate and increasing; usually at current production rate this

would be 87% design rate and steady. $1000684 Checks out OK. Recalibration not needed. $800686 As usual. Same as before the shutdown. $300688 No change, no improvement. $400689 Everything OK. no apparent steam leaks. Buildup of cake on the tube near the

inlet. Flights in good shape. $4500690 Pressure, when converted to head based on the assumed density of the process

fluid, agrees with the no-flow condition on the vendor’s head-capacity curve. Ifdidn’t put on safe-park as first step, then dangerous potential fire/explosionconditions are created while you experiment. $500 000

691 Equal to the static head about 15 m whereas head-capacity curve is about 20m. $300

694 Negligible amount drains from KO pot. $200695 Yes.

460


697 Sharp-edged orifice plate sized to give CD= constant for range of flows; ade-quate straight length of pipe upstream and downstream. $100

698 Details of all utilities, instrument and lubrication lines are not on the diagram.There are a drain at the bottom of the neutralizer and the usual vents, sightglass and pressure-relief valves on the neutralizer. There is no measurementof flow on the cooling water. Nitric acid is steam heated in the storage tank viaa trombone heater with the condensate going to drain; demineralized watergoes to the lantern ring-packing seal in the transfer pump. All sample lineshave water flush; the pigtails on the pressure gauges have tee-valve for waterflush. $600

699 Responds to change. $200700 Loud crackling noise, like cavitation, when< 6 g/L NaOH equivalent acid addi-

tion is required. $1200701 Working well. $400702 OK, recalibration not necessary. $700703 18 –10% L/s. $80705 OK; recalibration not necessary. $10 000706 Followed standard procedure; we were all especially aware of the impact of

trace amount of oil left in the system and did our best to ensure none waspresent. $1200

707 Deep thickener with central rake, designed for influent solids 1 to 4% solids;underflow 3 to 6% solids with a mass loading of 0.4 to 0.55 g/m2 s. $30

709 TDI from the new supplier doesn’t contain benzoyl chloride. The TDI fromthe previous supplier had benzoyl chloride as a reaction modifier. $3000. Ifyou didn’t put the plant on SIS or SIS + evacuation before you asked this ques-tion, the plant explodes with loss of life. Penalty $3 000 000.

711 Filter cycle: pressure nozzle feed, Model 1600 DSM 120�, 10-min screen,wash cycle 3 min. $100

713 500 kPa g –10. $50714 Appears to be fully open. The direction of flow through valve is consistent

with expected flow.715 Level indicates the coil is covered716 No leaks apparent. $100717 No noise. Valve not open. No evidence from flare that the pressure relief has

opened. $300719 NFPA: 1, 1, 0; irritant to the eyes.720 Respond to change. $500721 227 �C. $200722 Four months ago. None on this unit. $20723 Operator had turned this wide open. $200. If you didn’t put the plant on SIS

or SIS + evacuation before you asked this question, the plant explodes withloss of life. Penalty $3 000 000.

725 Allowance for fouling on tube side, water= 0.0002 m2 K/W; shell-sidestream=0.0005. Care to prevent temperature cross-over. $400

727 205 kPa g and steady. $900

461


728 Nothing. We were shut down for maintenance.729 No ice. $10 000730 4.2 MPa. $150731 Yes. $50732 Not needed. $4000735 50 �C. $300736 Usually 90 �C and relatively steady. $300738 Why1? so that other units can function well. Why2? to keep the overall plant

operating safely and producing specification product. Why3? so that companyis economically profitable. Conclude: focus on the goal statement: “to providethe usual amount of heat to the process streams of other units”. $50

739 Design checks out OK. No errors made. No water-cooled trim cooler wasincluded. $100

740 UA=q LMTD. Actual= (F cp)E101 66/ 32 �C or 2.1 (F cp)E101. Design= (F cp)E10158/ 36.5 �C or 1.6 (F cp)E101. Actual UA is 30% higher than design. $300

742 Should work OK. If you are concerned about surge, check the amps. $500743 Water would leak into the neutralizer. $60744 Not needed. $3000746 At half-load: UA=Feed flow. cp. (33.5/ 27); At full load: UA=Feed flow.

cp. (31/56); UA full load is 0.44 of UA for half-load.747 Responds to change749 No improvement. $8000750 The temperature in the neutralizer started to decrease. We took extra samples

of the product and found this was more dilute than usual. The exit cooling-water temperature is slightly less than usual. Everything else seems normal.$120

751 See Chapter 3: reactor, Section 3.6.2; compressor Section 3.2.1; sensors, valvesand control Section 3.1.3; knockout pot Section 3.5.1; exchangers, Section3.3.3; refrigeration cycle Section 3.3.4. $6000

752 Increasing concentration of heavies. Overhead off-spec. $50753 Bitter cold, –15 �C, blustery with snow squalls. The weather has been slightly

warmer in the week but has gotten colder each day.756 Recalibration not necessary758 IS: pressure above setpoint and controller output= 0%. IS NOT: steady and

controller output mid-range. $50761 No observable change in level on tray 4; no observed flow in the bulls eye.763 No change in swinging-loop phenomena. $80 000765 Design specs. seem OK. Suction nozzle 10 cm nominal diameter; discharge

nozzle 7.5 cm diameter Sludge-handling pump. $20766 Steady and usual value. $50767 Motor running. Shaft rotating in the “correct” direction of rotation. $400768 Both are steady769 We’re not getting much and what we get is off-spec. $50770 Might be better with a float trap with a separate trap from each of the three

stages. The “design” condensate load should be more than double your design

462


load of 1.6 kg/s. However, for an inverted bucket trap you must ensure thatyou have the correct Dp across the trap.

771 Specs on size distribution of particles.772 Work OK. $120774 Some improvement. Concentration of propane in flare gas decreases to 50%

over the design value. $6000775 No change. Still cannot control. $10 000776 No change.777 Same as expected. $1200778 No gauge, can’t tell. $30779 Yes. What’s going on? $30780 Not needed. $50781 3.15 L/s with diameter 5.5 cm= 1.3 m/s. $50782 Sink marks; product breaks. $2000783 When: 8 months ago. During this turnaround, routine checks were done on

equipment. $400785 Liquid appears in the sight glass and rises. When the flow is stopped, the level

gradually falls, in the sight glass, slowly over the 10 minutes. $1000786 Negligible change. $1000789 Steady and usual 45 �C. $50790 Nothing unexpected.791 No water leaks observed. $20 000792 The pressure at the PRCV must be > 22 kPa g to prevent subatmospheric con-

ditions at the compressor suction. If subatmospheric pressure occurs, there isa chance that air will leak into the ethylene stream.

793 There is still no flow of sludge to the filter. $40795 No improvement, $6000797 Usual value.798 No improvement. $300799 Same size.800 No evidence of surge. $2000801 From the pressure gauge, an estimate of the pressure drops and from the

pump head-capacity curve from the original vendor on file the pump shoulddo the job and no apparent blockages are in the line. If didn’t put on safe-parkas first step, then dangerous potential fire/explosion conditions are createdwhile you experiment. $500 000

802 50 �C. $300803 Confirm the temps and pressures at the bottom and top of column are consis-

tent with the compositions expected. $100804 Responds to change.806 Suppliers were aware of our specs; we were all especially aware of the impact

of trace amount of oil left in the system and did our best to ensure none waspresent $2000

807 Improvement, extension of the period before break but still not satisfactory forthe customer. $15000

463


809 IS: On the ethylene exit line. IS NOT: upstream or in butane unit. $50811 78 –1%. $2300812 Robust unit. Should work well for cement.813 Alarm faulty. $300814 Fluctuates. $200816 Continuous crystal discharge by peeler. No crystal discharge during the wash

cycle (that is used to clear the fines from blinding the media.) cycle times,10 min filter; wash media cycle 3 min $120

818 Fortunately a spare was available in stores. No change. When capacity wasincreased to 150 Mg/d we encounter the same problems as before. $80 000

820 Lower than expected. Occasion puffs of “afterburn” with local temperaturesmuch higher than normal in the convection section. If didn’t put on safe-parkas first step, then dangerous potential fire/explosion conditions are createdwhile you experiment. $500 000

821 Difficult because the flowrates are continually changing during the upsets.$50

823 Should do the job.824 Inverted bucket with upstream strainer. No bypass. Sized on double the con-

densate flow. $200825 Check on design calculations show liberal allowance for the pressure drop

across the control valve so that it operates mid-range. Pump should have notrouble supplying the design reflux. $100

827 IS: Gas samples are explosive and pose immediate hazard. IS NOT: gas sam-ples have non-explosive gaseous mixtures. $30

831 The hopper should be well designed. It has rained recently and if moisture gotinto the powder you might run the risk of bridging. $3000

832 Hot 33 �C and humid. This hot spell has continued all week. $150834 No improvement. $7500838 No observable decrease in pressure. $6000842 No change, no improvement. $400843 Emergency/ action to prevent explosion. Yes, there are a variety of hypotheses

to explore later to discover why the temperature runaway: cooling-water fail-ure/ coolant-inlet temperature too high/ cooling-surface fouled/ addition toofast/ operating procedures not followed/ sensors wrong/ different supplier ofTDI/ contamination of TDI in the storage tanks/ different polyol/ contamina-tion of polyol in the storage tanks or lines. $300. If you didn’t put the plant onSIS or SIS + evacuation before you asked this question, the plant explodeswith loss of life. Penalty $3 000 000

844 Equipment should do the job. $2000846 Not needed. $15 000847 Usual value but fluctuating; past records show similar values.849 Bypass valve shut; block valves open. $80850 Calibrates OK. $8000851 Working well. $400852 Designed to shut down if the suction pressure exceeds 112 kPa. $500

464


853 Fully open. $300854 OK. Recalibration not necessary.855 OK. Recalibration not necessary.857 Should be OK. allowed for in the Dp across control valve FV-1. $100858 Sound changes about half-way up the condenser. $600860 More cycling. $600861 Rake speed is kept consistent with feed concentration of 400 ppm but because

the torque has increased I increased the sludge pump rate, even up to wide-open, as given in Standard Procedures. The supernatant was still around200 ppm and the belt filter was flooded. $120

863 Half-load: DT goes from 75 to 5 �C. Starts at film boiling, flux decreases andthen increases as shifts to nucleate near the exit. Full load: DT goes from 75 to41 �C. Starts at film boiling, and flux decreases near the exit.

864 Fluctuates. $200865 The reverse jet works automatically. When trouble shooter is out on the plant

the cycle seems to be working.866 Fluctuates. $200867 No change in swinging-loop phenomena. Loss of reactant gas to bleed. $100

000868 OK; head-capacity curve OK relative to estimated pressure requirement. $240871 Can’t tell. No level indicator nor sight glass was installed. $30872 Propane flowrate is way below normal; the quality is OK. $120873 Cycle time increases; product breaks more easily. $1000874 Startup seemed to be methodical and carefully following the guidelines sup-

plied by the vendor of the new catalyst. $150875 Appears to be fully closed. The direction of flow through valve is consistent

with expected flow.876 45 �C. $3400878 None available. $30880 Three months ago; usual maintenance on pump. $200881 Trace amounts of water; no ethylene. $5000882 Usual. $50883 U tube, horizontal. $100885 Low level that is rising. $200886 93 �C; about 3 �C higher than usual. $100887 No evidence of surge. $2000889 No change. Still upset and fluctuating890 Power agrees with expected draw for the flowrate. $200892 8, 83, 105, 43 ppm $6000894 No change. $2000896 Amps as expected excepted when low flowrates of acid are required. $1000897 P3 is < atmospheric, as it should be. Therefore air leaks into the furnace;

excess fuel would not leak out of the furnace. If didn’t put on safe-park as firststep, then dangerous potential fire/explosion conditions are created while youexperiment. $500 000

465


898 Just start the pump and let it run. $10899 No improvement. $20 000900 Heavies concentration exceed specifications. $800901 Not successful. $1000902 Boiling water on outside tubes at 12. + hydraulic head MPa with gas on inside

of tubes at 4.55 MPa; water leaks into process gas and becomes steam.903 Part of the design of the reactor. Incoming gas flows across tubes carrying hot

reactor effluent; then on the tube side with the hot gas from the catalyst bedon the shell side. The heated feed gas then goes up and then down throughthe catalyst bed. Thermocouples monitor the temperature of the exit gas as itcools down. $800

904 54 MPa. $200906 Fixed roof, usual design with breather, heater, level gauge, sensors. Tried to

design to minimize water condensation inside the vessel. Ensured roof hasgood seal to prevent rain from entering vessel. Should do the job. $300

907 Design load= Feed flow. cp. 67 �C; Actually get (at half-load) = Feed flow.cp. 33.5 �C; Actually get (at full load) = Feed flow. cp. 31 �C; Slightly better forhalf-load than full load.

909 Before the change the feedrate to the dryer was relatively constant for 10 min-utes; then feed dropped to zero for the 3 minutes of wash cycle; then cyclerepeats. After the change, the feedrate to the dryer only lasts for 3 minutes,then the wash cycle for the centrifuge, then feed to the dryer and so one. Dur-ing the 3 min feed to the dryer, the rate seems to be all over the place. First, itseems about the same as before; another time it is about 30% higher andanother time it is about double. $1200

910 19 �C. $300911 No formal records kept. The operator had worked with polypropylene and UV

stablizers often, previously. No lumps appear in prototype when viewed bytransmitted light. $100

912 Valve A fully open. No change in temperature. $6000913 Everything OK. $3000914 Propylene: NFPA: 1, 4, 1. Extremely flammable. $50915 Fluctuates. $200916 No upsets. Steam pressure 1.2 MPa g. 3 �C superheat.917 Should be OK even with increase in column pressure by 100 kPa; allowed for

in the Dp across control valve FV-4. Suction side OK, complete with vortexbreaker. $300

919 Not needed.920 495 �C gradually increasing to 510 �C. Bed is cold. Usually 510 increasing to

550 �C at exit of bed. $1000922 Nothing, except we are not getting the amount of heating we expect. $150925 Not easy without shutting down plant. $3000926 Estimates agree with Dp. $2000

466


929 Before “safe-park” = 0%; under “safe-park” conditions= 5%. If didn’t put onsafe-park as first step, then dangerous potential fire/explosion conditions arecreated while you experiment. $500 000

930 Full documentation is available. Model 2GYUO;, 7 L/s; 10 m head, 1 kW;1750 rpm, efficiency= 70%; NPSH required at 7 L/s= 1.1 m water. At zeroflowrate, head= 12 m. $15

931 Yes932 No difference. Solution meets specifications. $1000933 Responds to change. $20935 Bypass valve was closed; both block valves were open. $800936 No pressure decrease. $15 000938 Very fine particles, < 150 mm, free flowing with angle of repose 15 to 30�,

mildly abrasive with Moh’s hardness 2–3. $1000940 Sample 1: 0.95%; 2: 0.98%; 3: 1.10%; 4: 1.00%; 5: 0.95%; 6: 1.05%; 7: 1.06%; 8:

0.96%; 9: 0.88%; 10: 0.99%; 11: 1.10%; Insert swing: samples during 10-minuteswinging loop: S1: 0.98%; S2: 0.99%; S3: 0.99%; S4: 0.98%; S5: 0.99% 12:1.06%; 13: 1.02%; 14: 1.01%; 15: 1.02%; 16: 1.10%; 17: 0.94%; 18: 0.98%; 19:0.95%; 20: 0.89%; 21: 1.01%; 22: 1.02%; 23: 1.01%. The swinging occurred inthe middle of hour 11. $80 000

941 Operation went smoothly. No problems. No different from a month ago.943 We’ve tried everything. We have the reflux set at the highest value but, as you

can see, the reflux rate, FIC/4, is below normal. $50944 Level mid-drum; design level. $50945 Negligible variation detected over five minute. $1200946 Usual value and steady; past records show similar value.947 Not needed. $4000948 Agrees. $200949 Not needed. $6000950 Moderate; 21 �C; cloudy; humid. Typical weather for all week. $50951 This is the startup after a shutdown. New catalyst was packed in the reformer.

Routine checks made of fouling on tubes in exchangers; checking of the sen-sors. Trapped air was vented from all exchangers before startup. Larger-sizeimpeller put in naphtha pump to handle the increased flowrate. $400.

952 No change. $3000953 Clean, clear, works well. $500954 Standard design, 50 cm width; stilling well and float-operated recorder for

height of liquid level. $400955 FC-1 = 6.8 L/s. Before we put the system on “safe-park” it read 8.2 L/s. If didn’t

put on safe-park as first step, then dangerous potential fire/explosion condi-tions are created while you experiment. $500 000

956 50.2; 49.8; 50.4% w/w. $10 000957 0.8 Mg/m3. Color, crystal clear. $150958 Trace amounts of butane. $5000959 Estimate seems OK. No major blockage. $300960 21 �C. $300

467


961 Within normal specs. $800962 Steady at 70 �C. $200963 Responds to change. $100964 IS: after two months operation; summer. IS NOT: immediately at startup al-

though “control was not very good”. $50965 Product breaks. $1500966 Allowance for fouling on tube side= 0.0001 m2 K/W; shell-side steam

= 0.0 0001. Uses tower water. $160967 See Chapter 3: boilers and condensers, Section 3.3.3; sensors and control, Sec-

tion 3.1.3. $50969 Allowance for fouling on tube side= 0.00001m2K/W; shell-side steam

= 0.0006. Simulation shows that unit was operating very close to design valuesfor usual range of feedstocks and conditions. $250

972 Not needed. $50973 No idea. There is no pressure gauge on the tank. $15974 None that we know about. The feed has been in storage tanks during the shut-

down. $50.976 See Chapter 3: instruments and controllers, Section 3.1.3; pumps, Section

3.2.3. $50977 Calibrates OK. $1200. If you didn’t put the plant on SIS or SIS + evacuation

before you asked this question, the plant explodes with loss of life. Penalty $3000 000.

979 IS: feed flowrate is gradually being reduced. IS NOT: the normal feedrate. $50980 Steady and usual. $50981 Estimate 1.5 m/ (1.3 m/s) = 1.1 s. $50982 No additional data available. $200983 This system has worked extremely well in all our installations. Where we have

had trouble has been when the on-site utilities have not been reliable: steam,cooling water, condensate handling. Oh, we did have two cases where theproduct foamed severely. That caused cycling and reduced capacity!

984 Not needed. $2000985 Packed column providing good contact between ketene vapor and glacial liq-

uid acetic acid to produce acetic anhydride.986 IS: always since startup.987 (E113) Usual allowance for fouling on tube and shell side. Vent/blowdown

lines on both the tube and shell side. Vent/blowdown lines discharge to sewer.Each vent line has a single gate valve that is normally shut. Process pressure> propylene refrigeration pressure. $250

988 Valves stems respond to change. $600989 For ethylene, 1.6 �C corresponds to 4.4 MPa abs; 8.5 �C, to 5.1 MPa abs;

–3.8 �C, to 3.9 MPa g. For n- butane: 0.58 MPa abs corresponds with 57.8 �C;0.97 MPa abs, to 78.5 �C. For steam at 0.8 MPa g corresponds with 170 �C. $50

990 Bumps appear on the product opposite to the knock out pins. Product stillbreaks. $800

468


991 This process always worked OK. $400. If you didn’t put plant on SIS or SIS +evacuation before you asked this question, the plant explodes with loss of life.Penalty $3 000 000

992 Head-capacity curve available. $40993 Value of 3.5 m compared with 2.5 m required. $100994 Less than design. $50995 Sized on condensate discharge with a safety factor of 2 times the nominal

flowrate of condensate. $150996 IS: OK at 8 am; IS NOT: OK at noon or later. $3000997 Rectangular box 0.5 m � 1 m with coarse screen to remove metal and large

material that might damage impeller of downstream pump. $20998 Control system tuned by process-control specialists. $400999 Not needed. $5001000 503, 547, 602, 698, 587 ppm. $60001001 Equipment should do the job.1002 IS: operators on this plant and operators on downstream IPA processing

plant. IS NOT: at this time no complaints from upstream or downstream ventscrubber personnel. $650

1003 Estimate of head required agrees with that supplied by pump. $10001004 Resin polypropylene plus UV stablizer and pearlized pigment. No blowing

agent. $501005 Blower-performance curve available, specs available. No pressure gauge on the

exit line so we cannot tell. If didn’t put on safe-park as first step, then danger-ous potential fire/explosion conditions are created while you experiment.$50 000

1006 Use either of the pumps A or B. We have the extra pump available in case onepump needs maintenance. Then we can swing the other pump on-line andkeep the process going. Occasionally, if capacity demands increase beyondwhat one pump is capable of producing, start both pumps and operate in par-allel. $30

1007 No improvement. $15 0001008 Why1? to contact Fuller’s earth with oil in the deodorizer; Why2? to contact

Fuller’s earth with oil; Why3? to remove the color bodies from the oil; Why4?to provide oil suitable for downstream processing. Conclude: focus on goal “tosuck Fuller’s earth into the deodorizer.”

1009 Before: screen 10 min; wash 3 min. Now: screen 3 min; wash 3 min; $301010 Given in Chapter 2 and on page 2-68 of “Process Design and Engineering

Practice”, Woods, Prentice Hall (1995).1011 Bypass valves: butterfly valves. These do not seal tight enough and cause gas

leakage even when they are “shut”. Replaced with globe valves. $60001012 Reads 24 �C. $10001013 Responds to change. $1001014 Steam pressure the nominal 1.5 MPa g; we have a slight superheat here so

that the steam should be saturated at your unit. $500

469


1015 Yes, for this production rate. Water should be able to condense all the vaporproduct. $50

1016 No noise sounding like cavitation. $2001017 See Chapter 3: distillation, Section 3.4.2; condensers, Section 3.3.3; control

and sensors, Section 3.1.3. $7001018 IS: exit gas is 730 �C. superheat temperature is 443 �C. IS NOT: exit gas design

value of 600 �C; superheat is not 353 �C.1019 Estimate 1.25 kg/L � 1.3 m/s � 0.055 m/ 30� 10–3 = 2900. Probably turbulent. I

would be more comfortable if the Re was three times higher. $501020 Yes. $301021 None done. This is just startup. Plan to repack the circulation pumps on the

absorber-reaction system every time the furnace is decoked about every10 days.

1022 Cooling time lengthened to 51–60 s and product looks OK although the pro-duction decreases because of the increased cycle time per piece. Further testsshow the product still breaks. $800

1023 All air and water tests done prior to startup showed no leaks. Care was takento remove the air and water from all process equipment prior to startup.

1024 Set at 20% open; usual value and steady. $6001025 Fluctuating wildly. $3001026 For the air: Air flowrate is unknown; we can measure the air temperature and

humidity in and out, estimate the heat capacity but we really can’t estimatethis without plant data. The files show that the new exchanger should con-dense all the IPA with ease. $3200

1027 Sharp-edged orifice plate sized to give CD= constant for range of flows; ade-quate straight length of pipe upstream and downstream

1028 Stage 1: 0.58 kg/s; from stages 2 and 3 combined 1.0 kg/s (neglecting about10% loss as flash steam) or total condensate load of 1.58 kg/s. $700

1029 Not needed.1030 Head= 46.5 m at zero flow. $501031 Area sufficient to do the job with the following fouling factor allowances: shell

side 0.0002; tube side 0.0003 m2 K/W. Considered propylene boiling to rangeover nucleate and film boiling. $200

1032 Cycle 1, 10 min. Cycles 2 & 3 similar:’1: 5.1%;’2: 5.6%;’3: 6.2%;’4:6.7%;’5: 7.1%;’6: 7.6%;’7: 7.2%;’8: 7.7%;’9: 7.9%; $3700

1034 Reads OK. $10 0001036 Allowance for fouling on tube side= 0.0001 m2 K/W; shell-side steam

= 0.00001. Slightly overdesigned with about 10% more area than needed. $7001037 This is not measured but it comes from storage and the temperature should

be pretty steady. $8001038 Dry reciprocating; should supply needed vacuum.1039 Flowrate of product increases to 0.08 L/s and is steady but still is not up to the

expected 0.15 L/s1041 Before: screen 10 min; wash 3 min. Now: screen 10 min; wash 3 min.

unchanged. $30

470


1042 No improvement.1043 P&ID supplied; P&ID is less than a year old. Simulation report shows that

unit was operating very close to design values for usual range of feedstocksand conditions. $30

1045 Should produce target amount of dry sludge and target filtrate of 100 ppm.$400

1046 Flow is constant and at design value. temperature going to the chiller unit hasTemp= 100 �C. We are not pleased that the temperature of the stream that weare getting from your chiller unit is 10 to 11 �C. Do something on your chillerunit or we are going to have to shut down! $500

1048 325 �C. $30001049 Responds to change. $2001050 Slightly less. $2001051 Thermowell location and “cleanliness” looks OK; vortex breaker is clean and in

place. Level sensor looks OK. $25 0001052 Higher than usual 0.522 MPa and increasing. $501053 Hydrogen has relatively high heat capacity; about 10 kJ/kg K compared with

about 2 for ammonia, methane or nitrogen; hydrogen thermal conductivity isabout 0.2 W/m K compared with about 0.02 to 0.03 for ammonia, methaneand nitrogen. The Pr numbers are similar. Hence, the thermal properties of amixture of hydrogen and others vapors depends on the gas composition. $400

1055 No change. $30001056 Yes. $701058 NPSH supplied= 3.5 m; end suction centrifugal; impeller one size smaller

than housing in anticipation of future expansion. Pressure allowance for con-trol valve. Check valve but no pressure gauge on exit line. $50

1059 Respond to change. $5001060 Moderate, 20 �C; cloudy. This past week has been moderate with temperatures

18 to 22 �C; rain two days ago. $201061 Should do the job.1062 Estimate agrees with Dp. $20001063 None. $501065 Hot 25 �C; sunny. Similar all week; we have had a hot and dry spell. All in the

turnaround complained. $201067 Estimate seems OK. No major blockage. $12001068 Open. $5001069 As expected. No improvement in the mass balance. $80001070 0.75 Mg/m3. $10001071 Responds to change.1072 6 kW. $1001073 Valve A fully open. No change in temperature. $60001074 Set at 20% open; usual value and steady. $6001075 Density 1.478 –10% Mg/m3. $10 0001076 221 �C. $2001077 Steady and usual. $50

471


1078 Strengths of cast samples within specs. Particle-size distribution within specs.Concentration of components within specs.

1079 Preheater 1: 175 kW; preheater 2: 120 kW; Stage 1: 1100 kW; stage 2: 1080 kW;stage 3: 930 kW. $1000

1080 Two days ago. Before the turnaround, the South loop was constructed first;was commissioned and performed well at design rates. The North loop hasjust been completed, the reactor catalyst has been reduced and is now readyfor operation. The tie-ins have been made to connect the two loops to allow forgreat operating flexibility. The combined “two-loop” plant has operated for twodays during which time every time the flows were increased beyond 80%design capacity, the swinging-loop behavior occurs. $800

1081 Yes. $501082 Today; warm and humid; past week has been torrential rain. $501083 Equipment should do the job without rat holing or bridging.1084 0.5 MPa. $1501085 After 1 hour the temperature is 335 �C; up 10 �C. $25 0001086 For the new reactor conditions there will be less energy available for export via

exchangers E202 and E203. $10001087 Responds. $4001088 Agrees; seems consistent. $2001089 No change in operation. Pressures and temperatures still increasing. $2501091 Good flowrate; inlet temperature 15 �C. $2001092 TC-3 = 205.9 �C. If didn’t put on safe-park as first step, then dangerous poten-

tial fire/explosion conditions are created while you experiment. $500 0001093 Some improvement. Now can control –4 �C. $50 0001094 No change. Still cannot control. $10 0001095 Currently at full rate.1096 Flows required for the reaction are consistent with design values. $501097 When I opened the valve, I would have expected to hear a noise of air coming

in and, with the light on, when I looked through the viewport I could see thesurface of the oil clearly and I could see the discharge end of the Fuller’s earthconveyor. No solids or powder were coming out of the pipe. I didn’t hear anynoise. $ 4000

1099 No change. $501100 Goes into the top. $1201101 Not available. They were put in so long ago no one remembers. $6001102 dPI slightly higher than usual. $501103 Not needed. $50001104 OK. recalibration not necessary. $30001105 No improvement.1106 When: 4 months ago: Checked over the control system and fans. Little adjust-

ments or changes made; Pitch on the blades seems to change when set pointchanged. $50

1107 Slight variation: 0.035, 0.041; 0.039%. $3000

472


1108 IS: over the past few months PIC set point needs to increase. Low-temperaturetrip when full load of ethylene fed to the unit. Outlet ethylene temperature hasbeen slow to recover from change in flowrates. IS NOT: PIC set at 0.8 MPa;ethylene temperature quickly recovers from change in flow rate. $50

1109 Fully open. $1101110 Thermodynamic traps on all three locations: reboiler, turbine and preheater.

No bypass supplied; upstream strainer. $3001111 Sensors wrong, T3/ sensor fault F2/ heat-transfer area too small/ heat-transfer

area fouled/ poor tuning of controller/ air flow too small to support combus-tion/ flameout/ gas velocity on outside too small/ excess air cooling the radiantand convection sections/ damper failed closed/ liquid-fluid velocity on insidetoo small/ decrease in thermal properties of process fluids/ process fluid flowincreased.

1112 Allowed 130 kPa Dp for the control valve. NPSH required= 2 m water. Sup-plied 4.5 m organic minimum. $200

1113 Set at 10% open; usual value and steady. $6001114 No feed, crystals or water was observed to flow from the screen to the centri-

fuge during the wash cycle for the screen. $7001116 See Chapter 3: pumps, Section 3.2.3; reactor Section 3.6.3; exchangers. Section

3.3.3. $3001117 74.3%. $30001119 Basics are OK but it is missing key details. There are block valves, a drain and

bypass with control manual valve for all control valves: CW, steam, Level.There are local temperature and pressure sensors on the top of the column.The temperature is shown in the control room. Vortex breakers are specifiedfor the reflux drum. The condensed liquid flows into a nozzle at the top of thereflux drum. The vent-break line includes a small orifice plate (instead of avalve to adjust the pressure balance). The valves are CW: Fail Open, FO;steam: FC; Level, FC. The steam trap on the reboiler is a float trap. $120

1120 Product sump has holes; liquid weeps from the sump. The trays are corrodedwith the hole size increased in all of the trays. One tray collapsed. $3000

1122 Should do the job. Enough flexibility in operating conditions that the bottomsconcentration of < 2% organic should be met and should provide negligibleorganic overhead into the steam-ejector system. Adjustable parametersinclude the boilup rate, the vacuum, the live steam, the reflux water rate.Furthermore, we can recirculate the overhead product back as feed. This plantis very flexible.

1123 Heavy fouling. $10 0001124 Well designed based on pressure, residence time plus vortex breaker and dem-

ister pad. $1501125 None seem to have been prepared for this new addition. Someone assumed it

would just be like the “old” unit. But it isn’t! $2001126 Product breaks. $10001127 Checked that the height in the seal pot that was used by maintenance= that

calculated for kerosene using handbook value for density. $400

473


1128 Stage 1: 123 �C steam; 109 to 115 �C glycerine inside tube which is < 25 �C; it islikely nucleate boiling; similarly 114–(98 to 108) �C in the second stage; 103–(80–60) �C in the third stage. $140

1129 No upsets. Everything is working normally. $3001130 Last maintenance 3 months ago; no planned maintenance for two months. If

didn’t put on safe-park as first step, then dangerous potential fire/explosionconditions are created while you experiment. $500 000

1131 Two months ago in May. Startup of new plant. $751132 IS: Control engineers. IS NOT: others. $1501133 Wide open. $1001134 The diagram seems to be a good representation of what is on-site. No sur-

prises. Around all the control valves are isolation block valves, drain, bypasswith valve. The steam drum has a level sensor with low and high level alarms.

1135 Vertical depth of the tube bank= 1.2 m. Top of syphon to top of the tubebank= 0.1 m. Bottom of the tube bank to the top of the overhead receiver= 1.2 m. $3200

1136 The ethylene vapor from the ethylene head tank goes to a downstream com-pressor as part of the ethylene chiller system.

1137 1800 nominal rpm, end suction; shrouded impeller; direct drive. NPSH re-quired for design condition= 2.5 m. $60

1138 Gurgling noise consistent with condensate-steam flow. $5001139 FC-1 looks OK; F-2, tab markings obscure; F-3, looks OK. If didn’t put on safe-


1140 Temp of –5 �C and pressure 500 kPa g; consistent with sat. propylene vapor;suggests instruments OK and no contamination. $75

1142 Cavitation, vapor lock on suction, excessive pressure loss on suction side. $2001143 Rpm � minutes= 300.1145 Seem consistent with expected gradients and responses. $4001147 Butane outside the tubes should be film boiling because the temperature dif-

ference for steam–butane is > 50 �C. $501148 This is first time startup. $501149 43 �C and usually 45 �C. $1201150 Loud crackling noise, like cavitation, when low acid additions are needed.

$5001152 Gradually increasing to 52 �C from the usual 47 �C. $501153 No change and poorer operation in the long run. $30 0001154 Yes, we’ve had occasional upsets that would have released stock into the water.

“But, you folks measure that and can handle that!” These have been going onfor about 20 days. $300

1155 55 �C. $5001156 Steady over 10-minute period at 110 �C. $501157 IS: Exit reactant feed temperature. IS NOT: elsewhere. $1501158 Detailed report of the startup of the South loop only. The plant worked well;

met all specifications. Everything worked at the design rate. At full design

474


rate, the conditions for the South loop were: lowest pressure in loop: 2.75 MPag; temperature at the exit of the cooling water exchanger= 30 �C; at the exit ofthe refrigeration exchanger= 10 �C; loop-gas analysis at the inlet to the reactor:H2 = 62%; N2 = 21%; methane= 13%; ammonia= 4%; discharge temperatureof the fourth and recycle stages of the compressor= 40 �C; valve A= 20% open;valve B = 10% open; inlet temperature to catalyst bed in reactor= 505 �C; exittemperature from catalyst bed in reactor= 550 �C; Dp across catalyst bed= 76 kPa g. A second internal report recommends linking the two loops to pro-vide flexibility in operation and provides the economics benefits of such opera-tion. $2000

1159 Product breaks. $10001160 Correct. $301161 Partially plugged with construction garbage: gloves, bolts and gaskets.1162 Confirm the temps and pressures at the bottom and top of column are consis-

tent with the compositions expected. $1001164 Sporadic fluctuation; no observable cycle time.1166 Boiling temperature of IPA= 88.4 �C. $8001167 Observed high frequency (and large fluctuations) suggest a controller prob-

lem. $501169 No improvement. $1001171 Estimate suggests no blockage. $4001172 Feed concentration the same 50% w/w IPA. $80001173 0.525 MPa and increasing. $5001175 Responds to change.1177 No noise. Valve not open. No evidence from flare that the pressure relief has

opened. $3001178 For the given areas, the fouling coefficient allowances and the conditions

shown in the given information, the three stages should work fine. $6001179 Confirm unsatisfactory yields of IPA. $12001180 Valve stem responds as expected. $2401181 Case 1: 8:28 am: South loop temp 563 �C; closed valve A for South loop by 8:29

am; 8:33 am; hot-spot temperature 555 �C and swinging seems to stop. Case 2:4:32 pm: North loop; temp 562 �C; closed valve A for North loop by 4:33 pm;4:35 pm hot-spot temperature 572 �C and swinging just gets worse. $80 000

1183 221 �C. $1501185 The setting on PIC/10 should be such that the column pressure is 1.7 MPa

but does not exceed 1.8 MPa, the relief pressure for PSV 1 $501186 DHf aq, 500 = –338 kJ/mol. $3001187 Should be about half-load to be consistent with the evidence.1189 Clean, impeller as expected, bearings OK. If didn’t put on safe-park as first

step, then dangerous potential fire/explosion conditions are created while youexperiment. $500 000

1190 No leaks. $80 0001191 Reynolds number= 8 � 104; turbulent. $501193 6 kW. $100

475


1194 With the centrifuge cycle at 10 min but with the increased volume of washwater for the centrifuge, the crystals leaving the dryer have moisture contentaverages at 3.5%. $ 2000

1195 170 and 171; a few minutes later 201 and 202, respectively. $200. If you didn’tput the plant on SIS or SIS + evacuation before you asked this question, theplant explodes with loss of life. Penalty $3 000 000

1196 Not needed. $5001199 A centrifugal pump should handle this condition but perhaps a reciprocating

pump would better serve this application. Check that the pump is locatedclose to the exit of the thickener and ensure you do not have NPSH problems.$75

1200 No change. $20001201 No evidence other than the log. Seemed to follow standard procedure. The

level in the poly was correct when I came on shift. $150. If you didn’t put theplant on SIS or SIS + evacuation before you asked this question, the plantexplodes with loss of life. Penalty $3 000 000

1202 Slightly more C3 in this sample than usual. $20001203 Design should handle usual flowrate and 90% reduction of target influent sus-

pended solids. No coagulant used. $5001204 None done. This is just startup.1205 Should do the job. Feed temperature varies from room temperature to 100 �C.1206 The steam-ejector system consists of a booster ejector, and two ejectors with

barometric condensers. The steam for all ejectors comes off the top of thesteam main. The cooling water for the condensers and the deodorizer comesfrom the cooling tower. There are eight, 25-ton deodorizers inside this part ofthe plant and on the second floor. The Fuller’s earth is in a hopper in an adja-cent building. The conveying line has very few bends; all bends have a largeradius of curvature.

1207 OK; recalibration not necessary. $10 0001208 Set at 10% open; usual value and steady. $6001209 Improved control but still the same type of general oscillations. $35 0001210 No evidence of fouling. $30001211 If there was no tube leak, water would dribble out. If there is a tube leak, fuel

gas will gush out. A vapor smelling like fuel gas gushed out. $20001212 New installation. Carefully pressure tested, cleaned out, inspected before

startup. The rest of the system was checked over, adjusted as needed beforethe new system was started up. $50

1213 Responds to change.1214 Top temp –100 �C usual; bottom temp same as usual. $2001216 Diagram is schematically correct. Around all control valves are block valves,

drain and bypass with valve. The reactor design uses cal rod heaters. Thermo-couples in the catalyst bed and in the feed gas entering the bed. Temperaturesensors at four levels in the catalyst bed. The catalyst bed, the gas preheaterand the water cooling exchanger are integrated into one, high pressure vessel.

476


Level indicators on the gas–liquid separators are used to control the flow ofliquid. $10 000

1217 TI-3 decreases to 48 �C promptly; the flowrate FIC/4 increases rapidly to usualvalue. $600

1218 No improvement. $25 0001219 Clean. $20001220 1.5 MPa g steady. $7001222 IS: at the hub at the weld line. $601223 Design capacity with allowance for pressure drop through the chiller on the

tube side and the elevation and frictional losses in the fittings and piping.Should do the job. Able to bypass the pump if the pressure in the feed drum issufficient for the desired flowrate.

1224 Should do the job. $451225 FC-5 reads the usual value and it reads the same value just before the system

was put on “safe-park”. If didn’t put on safe-park as first step, then dangerouspotential fire/explosion conditions are created while you experiment. $500 000

1227 Clean; no improvement. $2001228 I brought this on-line following the procedures. However, shortly thereafter it

started cycling like mad. I was keeping the feed steady. The feed compositionis on spec. But the product isn’t! That’s when I phoned you to sort this out.$120

1230 Small amount of scale removed. When operation resumed, negligible differ-ence in system behavior. Still getting off-spec material on some hot days.$4000

1235 Should handle new conditions easily. $9001237 Total area= 100 tubes � 2.83 cm2. Assume tube velocity to minimize fouling

= 1 m/s. Estimate of total water flow= 28 L/s. $1001238 OK. recalibration not necessary. $30001239 OK. Recalibration not necessary. $30001240 CH4: 16%; C4 s: 4%; C2H6: 12%; C5 +: 3%; C3H7: 6%; C3H8: 7%; N2: 48%;

H2S: 3%. $120001244 Thermodynamic trap; upstream strainer. No bypass supplied. $3001245 P&ID supplied; the data are being generated now during the commissioning.

$301246 Output is 100%; fully opened for a fail-safe valve, FC. $1001247 Usual. $1001249 Midrange and steady. $1201250 None on the gas feed lines from the ammonia and urea plants; steam lines

hooked up to the heater in the storage tank (with condensate going to drain);demineralized water to lantern ring-packing seal in transfer pump. $300

1252 49 �C. $4001253 Pressure is slightly less than usual. On the head-capacity curve suggesting a

higher than usual flowrate. However, the pressure is gradually increasing.$300

1254 Some fouling. Water and air pressure tests show no leaks. $3000

477


1255 Nothing unexpected because we had not increased process liquid flowrate. Ifdidn’t put on safe-park as first step, then dangerous potential fire/explosionconditions are created while you experiment. $500 000.

1256 OK, recalibration not necessary.1257 Slight decrease in response, more sluggish and slight decrease in ability to

evaporate ethylene. $5001258 No interruptions. Pressure should be 550 kPa g. $501259 Just before we started up, changes only made to the condenser and associated

piping. $6501260 No diagram supplied. Blends are stored inside in vertical hoppers, star feeders

at the bottom leading to batchwise, dense-phase conveyors that bring the dif-ferent blends to blender. The lines are gently curved. The mixer agrees withthe specs. The blended product is conveyed by dilute phase, pressure convey-ing to a reverse jet bag filter located above the hopper for the baggingmachine. The conveyed material enters the bottom side of the filter, rises upthe inside of the bags and the “clean” air goes out through the bags andexhausts to atmosphere. The equipment agrees with specs. The baggingmachine agrees with specs. No general surprises. The area is surprisinglyclean and dust free.

1261 This is a fail-open valve. The signal is 0. $200. If you didn’t put the plant onSIS or SIS + evacuation before you asked this question, the plant explodeswith loss of life. Penalty $3 000 000.

1263 Accurate; no recalibration needed. $30001264 23 �C; cloudy and overcast; rain forecast. It sprinkled over the last couple of

days. $3001266 Vent-break line already in place. $501267 Boiling temperature 117.9 �C; pKa= 4.76; heat of vaporization 405 kJ/kg. Acetic

acid vapor partially dimerizes (so that the molar mass varies between 60 and120).

1268 Frequency: once every 24 hours; sporadic; major changes within 5 minutes ofwhen hot spot first appears; takes about 1 to 3 hour to settled things outassuming “all the right adjustments have been made by the operators”. $600

1270 Pumped to the exchanger. $1501271 Slightly over sized. The liquid is subcooled so that there should not be any

NPSH problems. The role is to remove liquid product so that liquid does notback up and overflow into the vacuum system. $150

1273 No noise sounding like it is passing. $2001274 This is first time startup. $501275 High inlet temperature and humidity; Exit temperature and humidity as

expected. $40001276 Was OK. $20001278 Steady pressure at normal values; amount of superheat about 3 �C. $501279 Damper appears to be 1/3 closed. If didn’t put on safe-park as first step, then


478


1281 Nothing unexpected. $200 0001282 Not needed.1284 Valve stem moves to open valve. There is 100% valve travel. $30001285 Lumps persist. $25001286 No changes. Steam on the drum is steady, and we are supplying steam with a

slight superheat, as usual. $3001287 Balances –10%. $501289 Filled with sludge. $3001290 Flowrate relatively constant; the pH is jumping all over the place! $1001291 Pulses of excessive gas loading on the vent scrubber. $12001293 Same as we supplied you with over the past ten years. $2501295 Both have equal activity. $80 0001297 Sized to do the job. 10 L/s at head of 10 m ( = 100 kPa). $301298 Not needed. $60001299 Slightly more than usual. $751300 No improvement. $30001301 Signal is 0%. This is a fail-close valve so the valve should be wide open. $3001302 –25 –2 �C1303 Same as TC/5 within –1 �C. $5001304 The diagram seems to be a good representation. In addition, there is a control

system. The steam trap is an inverted bucket with an upstream strainer; isola-tion valves, no bypass. The barometric leg is submerged in a “hot well”. Vacu-um system consists of a booster ejector with two ejectors with two, direct-con-tact interstage barometric condensers. These condensers feed the hot well viabarometric legs. All the steam lines come off the top of the steam main. $200

1305 Reciprocating pump. NPSH needed << calculated NPSH supplied; steamtraced suction line (with 350 kPa g steam) to prevent freezing in suction line;upstream vortex breaker in sump on tray 4; design pumping rate 0.07 to 0.15L/s. The diagram does not correctly show that there is a strainer in the suctionline to 114-J. An analysis of the pressure profile from the tray 4 to storageshows that the product should flow by gravity to the storage tank without theneed for pump 114-J.

1306 128 �C. The usual value is 130 to 140 �C. $1201307 Responds to change. $1001308 Product breaks. $11001309 605 �C. $12001310 The feed concentration seemed to be about 200 ppm but because the torque

has increased I increased the sludge pump rate as given in Standard Proce-dures. The supernatant was still around 120 ppm and the belt filter wasflooded. $120

1311 Not enough steam/ wash water carryover from the centrifuge/ cycle fromscreen not coordinated with cycle in centrifuge/ feed crystals too wet from thescreen/ rotational speed of the dryer has increased/ more fines in the centri-fuge causing the filter cycle to be too long; fines carryover to the centrifugecausing blinding in the centrifuge/ centrifuge rpm faster than usual/ crystal

479


size change; crystals change shape so that filtering is different/ centrifugeoperates but has run out of feed/ dryer feed too cold/ vendor supplied faultyequipment

1312 Raw material: particle-size distribution within specs.1313 Slightly more than usual. $1501314 Fcondensed. 200 kJ/kg= 200 Fcondensed.1315 Not needed. $40001316 Pressure negligible. $10001317 The pressures and temperatures are rising and we can’t seem to control these.

The pump F1400 is knocking its head off but doesn’t seem to be pumpinganything out. I wish I wasn’t on this shift!

1318 Sampling error/ analysis error/ hazard coming from upstream plants/ leaksin vent valves from any location having “fuel gas” type materials upstream ofthe valve, e.g., the shell side of most exchangers/ leak in the valve on theknockout pot/ leak in the tubes from fuel gas to water in E131.

1319 IS: much less than expected and not in production. IS NOT: 500 �C at noonand steady production. $3000

1320 IS: after 3 hours since startup. IS NOT: immediately at startup. $501321 Why1? to ensure safe operation; Why2? to provide steady, reliable operation;

Why3? to meet production requirements; Why4? to sell products and keepcompany financially viable; Conclude: focus on Why1? “To ensure safe opera-tion.” $50

1322 Good flowrate. $2501323 No change. Flowrate and temperature should be the usual values. The temper-

ature should be about 18 �C. $200.1325 See Chapter 3: reactors, Sections 3.36, 3.3.7; mixers, 3.7.1; control and valves,

Section 3.1.3 $200. If you didn’t put plant on SIS or SIS + evacuation beforeyou asked this question, the plant explodes with loss of life. Penalty $3 000 000

1326 See Chapter 3: sensors and controls, Section 3.1.3; pumps, Section 3.2.3. $2001327 Not needed.1328 Steady flows and temperatures expected for this time of year. Since this is

spring, the temperatures are lower than usual. $601329 Everything started fine. We have not changed anything. $501330 I think early on today’s shift, but now there is no flow. $201331 Not needed. $50001332 IS: in neutralizer. IS NOT: reported elsewhere as yet. $501333 Recalibration not necessary.1334 Yes, I have noticed that AC/1 shows variation suggesting variable C3 in the

debut overheads or. maybe the analyzer is just jumping around. But that sys-tem controls itself – so I don’t worry. $50

1335 Why1? to provide steady, reliable operation; Why2? to meet productionrequirements; Why3? to sell products and keep company financially viable;Conclude: focus on goal “To keep bottom level steady and stop tray-tempera-ture cycling.”

480


1336 Diagram is reasonably complete. Additions include a strainer just upstreamon the suction side of the pump 114-J. The suction line also a bypass of pump114-J. Block valves are also on the exit of pump 115-J to direct the flow todrum disposal or to product storage. A drain valve is present beside the feedcontrol valve FIC /101 (which is incorrectly shown as PI/101). The controlvalve on the reflux water has block valves, drain valve and bypass with a valve.The cooling water on the reflux water comes from a common header with theCW to the barometric condensers and the water goes to the hot well. Steam isgenerated on-site at 3.5 MPa g, brought to the battery limits and pressurereduced to 1.38 MPa g for use within the battery limits. A temperature sensoris present on the steam to the reboiler. A temperature sensor is on the vaporfrom the reboiler into the column. A pressure sensor is at the bottom of thecolumn with a separate tap for the differential pressure. There is a level indica-tor at the bottom of the column. All process steam comes off the top of thesteam headers. The pipe rack is 3.6 m above grade. Just before the productstorage tank is a “bulls eye” in the line that allows one to see the flow to stor-age. There is a sight glass at tray 4 that includes the vertical distance includingthe product sump and the seal pan for the downcomer from tray 4. The liquidfrom tray 4 overflows the weir on the downcomer seal into the product sump.The weir on the product sump overflows onto tray 5.

1337 Responds to change. $2001338 Lost. Cannot be located. $5001340 Sample 1: 13%; sample 2: 10%; sample 3: 7%; sample 4: 3%; sample 5: 8%.

$20 0001342 Diagram reasonable. It is missing the isolation block valve on the pump suc-

tion. Pressure gauge has a pigtail with a shutoff valve with a tee nipple andvalve installed between the gauge and the pigtail to allow backflushing of pig-tail. $40

1343 Moist and humid today: a continuation of the weather over the last week. $601345 Not needed. $10 0001347 Alarm sounds indicating high pressure. $30001349 Oscillating corresponding to the recorded output. $1201350 The obvious ones: the cooling water in and out; steam out the top. There are

water lines available to clear sample lines but these are not directly attached tothe neutralizer. $300

1351 During the turnaround. Previously about four months before the turnaround,some routine maintenance was done on the naphtha feed pump and gas feedline to the furnace. $300

1352 Samples showed the propane concentration in the fuel gas was about 1.5times higher than usual but this could be simply sampling and analyticalerror. $100

1353 Extensive manuals including maintenance schedule and table of trouble-shooting diagnostics. $800

1354 Should do the job. Steam pressure to ejectors is 1.38 MPa g.1355 Fcondensed. 2. (120–60) = 120 Fcondensed.

481


1356 Confirm unsatisfactory yields of IPA. $12001357 Usual concentration of butane. $501359 Five months ago; routine cleaning of exchangers, checking and calibrating

instruments; usual maintenance and adjustments of the ethylene feed pump.$50

1360 Why1? to be able to sell propane not fuel gas; Why2? to keep the companyprofitable; Why3? to pay my salary; Conclude: focus on goal “To reduce thepropane loss to fuel gas.” $50

1362 Temperature quickly drops from 325 �C. $10 0001363 See Chapter 3: sensors and control, Section 3.1.3; evaporators, Sections 3.3.3

and Section 3.4.1; reactor, homogeneous (no trouble shooting suggestionsavailable); condensers, Section 3.3.3; vacuum pumps, Section 3.2.2.

1364 Correct residence time for gas-liquid disengagement and residence time forcontrol. Demister pad and vortex breaker included. $50

1365 Tricky. The leak goes either way depending on the PIC setting. For a high PICsetting the leak would be butane into the condensate; for a low PIC setting thesteam would leak into the butane. There is no butane detector on the conden-sate. $50

1366 Why1? to provide steady, reliable operation; Why2? to meet productionrequirements; Why3? to sell products and keep company financially viable;Conclude: focus on goal “To meet specs on overhead and bottoms.” $50

1367 Lumps persist. Time for injection increases; overall cycle increases. $30001368 Yes. $20001369 I can’t seem to control anything! I’ve tried putting it on manual but it is so

sensitive and I seem to get what I get no matter what! That’s why I called you.$120

1371 Pressure and overhead temperature reduces to design. Pressure control stillsluggish but it is in the design range. $12 000

1372 Equipment should do the job. Simultaneous fill and weigh.1374 Steady and higher than usual, 50 �C. $501375 Clear. No obstruction. $70001376 No.1377 Top temp –40 �C usual; bottom temp higher than as usual. $2001378 Steady, 1.8 –0.1 kg/s. $601379 The oscillations began small and increased in amplitude until it met sensor

range. Problems related to valve stiction and hysteresis more likely to lead tocontinuous cycling at smaller amplitude. $50

1381 Head-capacity curve available. NPSH requirements available. Operating at1800 rpm nominal. No pressure gauge on exit line so cannot check. If didn’tput on safe-park as first step, then dangerous potential fire/explosion condi-tions are created while you experiment. $500 000

1382 Allowance for fouling on tube side= 0.00001m2K/W; shell-side steam= 0.0006. $250

1384 2:01 am South reactor: hot spot noted 580 �C; 2:02 Production in South loopincreasing; 2:02 North reactor “drying” with temp falling below 500 �C; 2:03

482


am. Startup heater put ON for North loop reactor; valve A on North loopOPEN to decrease gas flow and help reheat North loop; 2:04 am hydrogen ana-lyzer on South loop oscillating and showing a slow increase in hydrogen; 2:05am production in North loop decreasing; hydrogen in North loop oscillatingand showing a gradual decrease in hydrogen; 2:06 am valve A on the Southloop CLOSED. 2:07 am South reactor temperature 590 �C; hydrogen and pro-duction increasing on South loop; 2:20 am North reactor temperature 500 �C;2:22 am startup heater on North loop OFF; valve A on North loop closed to20% open setting; 2:30 am South loop reactor temperature 560 �C; hydrogenand production in South loop decreasing toward design values; 2:45 am Southtemp 555 and North temp 545 �C; 3:05 am both and North and South reactors550 �C exit temperature. No upsets or unexpected behavior over next twohours. $40 000

1385 Same before and after change. $12001387 No evidence of solidified fatty acid. $30001388 Improvement, extension of the period before break but still not satisfactory.

$12 0001389 Last turnaround three months ago. Process has been operating “fine” since

then. If didn’t put on safe-park as first step, then dangerous potential fire/explosion conditions are created while you experiment. $500 000

1390 Insufficient liquid in column. $50001391 No improvement. $25 0001393 Usual value.1395 Last summer; a compressor installed to recycle flare gas as fuel gas. New dem-

ister pad installed in the knockout pot. $3001397 –59 –2 �C1399 No change; gets worse. $80 0001400 None done recently. four years ago the rpm of the pumps was changed from

1800 rpm to 3600 rpm (and the motors changed accordingly from 1 kW to10 kW) to account for pressure for increased production. $15

1401 Files missing. Engineers say that the reactor was overdesigned and shouldhandle the increased feedrate “No problem!” If didn’t put on safe-park as firststep, then dangerous potential fire/explosion conditions are created while youexperiment. $500 000

1402 No sounds of cavitation. $2001403 1.6 m/s, reasonable. $501405 Should give the correct % vaporization per pass. $7001407 IS: On the debut. IS NOT: upstream on the deprop. No problems stated on

reactor $501408 This has been a bear to operate. The overhead temperature has been increas-

ing gradually and I have been increasing the reflux accordingly but the tem-perature keeps increasing. Yet the analyzer is showing the usual overheadcomposition. $50

1409 Steady and usual design value. $7001410 None

483


1411 We follow the new operating procedures that affect only the centrifuge. Washtime remains the same; volume of wash water increased by 35%. Filteringtime reduced by 1/3 to=wash time. $30

1412 75.3%. $30001413 Reads 438 –3 �C1414 510 increasing to 550 �C at exit of bed; usually consistent and “reliable”. $6001415 Warm 20 �C; similar temperature and slight wind all week. $151416 IS: nothing happened; IS NOT: Fuller’s earth seen flowing into deodorizer.1417 Operation went smoothly. No problems. No different from a month ago.1419 227 �C. $2001420 High liquid level on tray 4 and in the product sump. $50001421 Not for 8 months. $501422 The temperatures recorded in the control room indicate that the feed tempera-

tures are “normal”. The exit “cooling water” temperatures are extremely high.$200. If you didn’t put the plant on SIS or SIS + evacuation before you askedthis question, the plant explodes with loss of life. Penalty $3 000 000

1423 3 to 7.5 m/s as expected from the vendor data. $30001424 IS: at startup and during the first week. IS NOT: corrected after one week with

focus on control. $1501425 Based on the fluid velocity of 1.6 m/s and a distance between injection and the

sensor of 2.4 m=1.5 s. $501426 Agrees; seems consistent. $2001427 Fluctuates. $2001428 Not needed. Depends on the safety officer and magnitude of the values. $20001429 Half-load: inside h= (0.5)0.8 hfull= 0.57 hfull ; outside: heat flux (half-load) >>

Heat flux (full load) because of shift to nucleate boiling for half-load.1430 See Chapter 3: distillation, Section 3.4.2, perhaps pertinent condensers and

reboilers, Section 3.3.3; pumps, Section 3.2.3 and controllers 3.1.1. $501431 Coke on the outside of the tubes and on the inside furnace walls. If didn’t put

on safe-park as first step, then dangerous potential fire/explosion conditionsare created while you experiment. $500 000

1432 Estimate seems OK. No major blockage. $1001433 Well designed; should do the job. Use hydrogen fuel. Thermal effi-

ciency= 32%.1434 Consistent with overhead temperature and the overhead specs; relatively

steady since startup, about 230 kPa g. $2001436 Decreasing. $501437 Respond to change. $5001440 Seven months ago. Routine calibration of all instruments; changed the seal on

the transfer pump and adjusted clearances; cleaned the outside and inside ofthe cooling coils. Inspected neutralizer for corrosion. $150

1442 Line clear; no plugs. $35 0001443 No improvement in process.1444 Why1? to save lives and equipment, keep insurance down, good operations.

Why2? to stay in business. Why3? to pay my salary and give me continued

484


peace of mind. Conclude: focus on Why1?: “to save lives and equipment, keepinsurance down, good operations.” If didn’t put on safe-park as first step, thendangerous potential fire/explosion conditions are created while you experi-ment. $500 000

1445 Aluminum with two feed gates. Two cooling lines each side. $1501446 Midrange and steady. $1201447 Dark black with yellowish tinge. $10001448 Should work well. Baffles were in vertical. Vessel has air bleeds. Before you

started up you would have opened the bleed to ensure all the air was out. $2001449 This plant has been very steady. No changes.1450 Variable speed; air cooled; forced draft. No trim cooler. Good design of hydrau-

lics with vent break (not shown on diagram). 20% excess area. $501451 Calculated height of kerosene in the seal pot was checked to ensure that this

was consistent with a pressure differential of 6.8 kPa. No visual or level trans-mitter was installed that allows us to verify the height. $300

1452 Everything OK until temperature TC/3 started to droop. Looked at the flare.It’s black. I went on the plant but couldn’t see anything obvious. My first reac-tion is to reduce the process flow but called you before I reduced flowrate. Ifdidn’t put on safe-park as first step, then dangerous potential fire/explosionconditions are created while you experiment. $500 000

1453 Our lab: 0.028; Excell Lab: 0.030; Univlab: 0.026. $20 0001454 Slightly more than usual and consistent with the increased pressure. $751455 Well designed based on pressure, residence time plus vortex breaker and dem-

ister pad. Complete with syphon break, vents and drains. $1501456 Should handle range of flows for design conditions. $5001458 Feedrate increases slightly with time averaging about 65 during one cycle and

then averaging about 78 during the next cycle and then fluctuating in therange 60 to 85 depending on when the analysis is done. Before the change itwas about 68 during any cycle. $2000

1459 7 MPa; usual value for startup. $30001460 No change. $90001461 All seems to be working properly. $301462 Diagram shows general layout. Both control valves have isolation block valves,

drain and bypass with valve. There is no pressure gauge on the reflux pumpbefore the check valve on the discharge line. The condenser has four rows offinned tubes. The reflux drum has a demister and vortex breaker. There is avent break on the exit liquid header of the condenser to “seal” the tubes. Thereis no pressure gauge on the overhead receiver. $100

1464 OK; head-capacity curve OK relative to estimated pressure requirement. $2401465 Head= 45.0 m at zero flow. $501466 Not pertinent, just completed a turnaround. $501467 See Chapter 3: reactor, Section 3.6.2; compressor, Section 3.2.1; sensors, valves

and control Section 3.1.3; knockout pot, Section 3.5.1; exchangers, Section3.3.3; refrigeration cycle, Section 3.3.4. $500

485


1470 Everything works fine as long as we can depend on pump B operating well.We really are in big trouble if anything happens to B and we are expected touse A. As it now stands, we don’t have a backup! $15

1472 Clean; negligible corrosion. No apparent blockages in sparger holes. $30 0001473 Internals look normal; clearance=normal; wear rings= good shape; negligible

erosion; key in place. No pluggage. $40001474 Head-capacity curve suggests the increased naphtha flowrate can be handled

easily. $7001475 Should work well. Baffles were in vertical. Vessel has air bleeds. $2001476 Fluctuates. $2001477 IS: reported variation in concentration of propane in deprop bottoms > specs.

IS NOT: everything else is running smoothly1478 Well designed based on pressure and residence time. $1501479 Product continues to swell after it is ejected from the mold; resulting product

is mis-shapen. $9001480 Very fine particles, < 150 mm, free flowing with angle of repose 15 to 30�,

mildly abrasive with Moh’s hardness 2–3. Harmful by inhalation. Contains sil-ica, crystalline cristobalite. Irritating to eyes and respiratory system. $500

1481 IS: 7 hours after startup and after an upset on the upstream deprop that senttoo much C3 to the debut; IS NOT no information known because this isstartup. $50

1482 Flow is the control value of 14 L/s. $501483 Two months ago. $601484 When: 6 months ago; routine checks done during the turnaround. $4001485 41% ammonia; 29% water vapor, 30% CO2 all –3%. $21001486 Reciprocating pump. NPSH required is specified. Details for maintenance,

trouble shooting and precautions about operating with a closed valve on dis-charge and the importance of checking that the pressure relief from dischargeto suction is working. Nominal flow 0.15 L/s but this can be decreased by set-ting the recycle valve. Precaution: drain the pump cylinders of all liquid whenshut down to prevent liquid from freezing in the pump in cold weather.

1487 IS: T3 is lower than expected. IS NOT: changes elsewhere that have beennoted

1488 Unable to sample directly. The vent line is at the highest point, namely on theinlet cooling water. The exit discharges below ground. $200

1489 Was OK. $20001490 Dryer should do the job $1001491 Steady and slightly higher than usual. $501492 This thing cycles no matter what I try. $5001493 Hydrates are ice-like solid clathrates with the water forming the cage around

the hydrocarbon molecule. Species like methane, ethane, isobutane, propane(and butane in the presence of other hydrocarbons), plus water, can formhydrates. Hydrates form at high pressures and relatively low temperatures.For example, at 3.1 MPa hydrates form with isobutane at temperatures < 3 �C;with methane, temperatures < 3 �C; ethane < 14 �C; propane < 6 �C. Hydrate for-

486


mation can be inhibited by alcohols (such as methanol), glycols and ionic salts(including NaCl). Hydrates do not form with very soluble gases such asammonia and hydrogen chloride. $500

1494 Some improvement. Now can control –9 �C. $120 0001495 NFPA ratings: Health, Fire, Stability: Propane: 1, 4, 0; Methane: 1, 4, 0; Hydro-

gen: 0, 4, 0; Butane: 1, 4, 0; Pentane: 1, 4, 0; Extreme flammability hazard;avoid sparks, open flames. Hydrogen: Evacuate all personnel from affectedarea; do not enter areas containing flammable mixtures of hydrogen and air.Vent area to prevent form of flammable or oxygen deficient atmosphere. Elim-inate all potential sources of ignition. Butane skin-contact with gas may causefrostbite (liquified gas). May cause slight eye irritation. Inhalation causesdrowsiness, excitation or unconsciousness due to anesthetic and asphyxiationproperties of this gas. $50

1496 510 increasing to 550 �C at exit of bed; usually consistent and “reliable”. $6001497 Checked that it is not on surge or sonic conditions based on the average flare-

gas composition given in the problem statement. Confirmed that the decreasein gas temperature, because of cold temperature, is not a factor. Realize thathigh molar mass in feed decreases range of operation. Surge is related topower used. $800

1499 Installed pump and motor should handle it OK. $9001500 Not needed. $20001501 Same as expected. $12001502 Going crazy. $501503 Not needed. $20 000. If you didn’t put the plant on SIS or SIS + evacuation

before you asked this question, the plant explodes with loss of life. Penalty$3 000 000.

1504 Water has relatively high heat capacity; at the low-pressure steam has a highlatent heat. Ethyl acetate vapor heat capacity about 1.24 kJ/kg K comparedwith steam at 2 kJ/kg K. Pr= 0.8 for ethyl acetate and 1.1 for steam. If thesteam was at “atmospheric pressure,” then there is not enough pressure topush the steam through the pipe and exchanger. Usually 200 kPa g is theminimum practical pressure; sat. temperature for steam= 134 �C. Heat capaci-ty for water= 4.2 kJ/kg K. $50

1505 Cold, –28 �C, dry, windy. January. The weather has been like this all week.$150

1506 Steady and mid-vessel. $2001507 Closed cycle, standalone unit. Should do the job. $60001508 Now: IPA in= 2250 kg; IPA out bottoms= 225 kg; IPA out overhead to further

processing= 1350 kg or total of 1575 kg out. $42001509 Sounds “hollow” all the way up the condenser. $6001510 Identical to pump A. Head capacity and NPSH data as expected. Allowance of

140 kPa for the control valve. $151511 Usual amp draw under existing circumstances. $3001512 No change. $600

487


1513 Valve position is fully open; valve is put in correct direction; valve is two sizessmaller than the line. $100

1514 Feedrate constant now; feedrate constant before the change, Both feedrates thesame= about 100. $2000

1515 Steady and lower than usual at 104 �C. $501516 Feed flow rate 531 –1 L/ s; effluent flow rate 514 –1 L/s. $20001517 Area and fan should be fine for your operation. Allowances were made for the

hot day-time temperatures like today. Should be no problem. Should ensure,however, that the bottom series of tubes in the condenser are flooded to “seal”and prevent the vapor from passing through the tubes uncondensed. Toensure this, the liquid header should either be near the top of the tube bank orshould have an exit pipe that is elevated. Naturally you will provide a ventbreak at the top of this seal so that the condensate can syphon out continu-ously. If no vent break is provided, the liquid seal is periodically broken, all theliquid syphons out and uncondensed vapor escapes. $800

1518 Head-capacity curves available. Strainer needed on the suction to remove met-als and materials that might damage the impeller. Net positive suction headinformation given. $30

1519 Should work well. Baffles were in vertical. Vessel has air bleeds. Before youstarted up you would have opened the bleed to ensure all the air was out. $200

1520 Cycle 1 3 min:’1: 2.1%;’2: 2.6%;’3: 3.2%;’4: 3.7%;’5: 4.1%;’6:4.6%;’7: 5.2%;’8: 5.7%;’9: 5.9%. Cycles 2 and 3: 6.0%; 6.1%; 6.8%; 7.0%;7.2%; 7.1%; 7.7%; 8.3%; 8.5%. Cycle 4: 8.5%; 8.5%; 8.0%; 9.3%; 8.7%; 8.9%;9.0%; 9.1%; 9.2%. Cycle 5: 4.3%; 5.4%; 6.7%; 7.8%; 7.5%; 8.3%; 8.7%; 9.2%;9.1%. No data available for previous operation. $3500

1521 Same as expected; 3/4 full and reasonably steady. $301522 The major mold concern was the thick wall at the hub. Two gates were placed

in the mold at the hub to account for the thick walls. Several different moldingmachines were available on site to produce this product. A cold runner systemwas used. Eventually a set of production conditions were developed to producea reliable product. $200

1524 Allowed 130 kPa Dp for the control valve. NPSH required= 2 m water. Endsuction. $200

1525 Liquid has filled the glass; suggesting either a plugged sight glass, faulty read-ing or a very high level of liquid that would “overflow both the seal pan on thedowncomer from tray 4 and the product sump”.

1526 Dp across the new catalyst consistent with old catalyst. Max temperature fornew catalyst comparable with max temperature of old. Procedure for loadingcatalyst well documented. Same catalyst support used as previously. $650

1528 Should work well provided there is no vapor lock in the feed line. Caution:never operate against a closed discharge valve. Include a bypass. $200

1529 TI/2 = –59 –2 �C; downstream operator reports at design flows the tempera-tures are –60 to –55 �C;

1530 Hot and humid all week including today. $501531 Not needed. $4000

488


1532 Furnace: sized to take the increased flowrate. Heat-exchange tubes: sized totake the increased flowrate. With oil firing the emissivity goes through a max-imum in the early parts of the flame. If didn’t put on safe-park as first step,then dangerous potential fire/explosion conditions are created while youexperiment. $50 000

1533 Float trap provides continuous discharge, is suitable for low pressures; notsuitable for high pressure and not suitable where there is a chance for waterhammer. Range 0.06 to 5 kg/s. At 2.7 kg/s, even our largest size trap may notbe best because we recommend double the size and double your load isslightly more than the capacity for this trap. $200

1534 OK. Recalibration not necessary. $30001535 Thermodynamic traps on all three locations: reboiler, turbine and preheater.

No bypass supplied; upstream strainer. $3001538 UA design=heat load/LMTD=118/ 450 = 260; UA actual = heat load/LMTD

=81/550 = 147 or 56% of design. Either the area is dramatically reduced; heat-transfer coefficient on gas side is less (especially in second section) or highfouling coefficient.

1540 See Chapter 3: distillation, Section 3.4.2; pumps, Section 3.2.3; reboiler andexchanger, Section 3.3.3; vacuum system, Section 3.2.2; sensors and control,Section 3.1.3.

1541 No apparent leaks. $20 0001542 Warm, 22 �C; sunny. This nice weather has continued all week.1543 Yes. Valve stem closes slightly. $4501545 Moderate. $1501546 324 �C. $30001547 Unlikely that flashing would occur. $3001548 Responds to change. $2001549 No change. $3001550 Much lower total amount exchanged than expected. Heat lost in reformer exit

gas is same as that gained by stream to other units –10%. $8001551 Steady and mid-vessel. $2001552 Recalibration not needed, no adjustments made. $20001553 OK. Recalibration not necessary. $30001554 IS: this operator. $501555 Negligible fouling; no condensed liquid; baffle spacing correct and baffles

secure. $60001556 Orifice plate in correctly. No improvement.1557 Product breaks. $10 0001558 Balances –10% although too high a percentage of propane is going to fuel as

gas instead of forward as liquid. $501559 Clean; no improvement in process. $50001560 Unmixed without mechanical mixer. $30 0001561 Controller output suggests negligible stiction of < 0.1%. $16001563 Steady and usual value. $501564 150 amp. Usual value. $6000

489


1566 Should do the job. $20001569 Shell and tube. Should work well. Baffles were in vertical. Usual allowance for

fouling on tube side and shell side. Correction factor for not-true countercur-rent flow is > 0.85. Vent/blowdown lines on both the tube and shell side. $700

1571 IS: at drop boxes B and C. IS NOT reported elswhere on the site by the safetyinspector. $30

1572 Polypropylene NFPA: 0, 0, 0. Nuisance dust. Autoignition about 357 �C. Elec-trostatic charge can potentially build up when handling. $50

1573 Much higher in C3 than design. $50.1576 Almost completely plugged. $1501577 Everything OK. No evidence of leaks. $62001580 Pulp fibers had formed a slight plug at the end of the sampler tube. Cleared it

out and carefully put the sampler line in the feed. Suspended solids 482 ppm.$14 400

1582 Yes, T8 >T2. $501583 IS: sporadically; IS NOT: never. $3001584 Not needed. $80001585 No change. $6001587 Both correct. $301589 During the turnaround, routine inspection of the column internals, checking

pumps. calibrating sensors. $6501590 1.02 Mg/m3. $10001592 Both seem to function OK. Are left CLOSED.1594 Yes. The downcomer appears to be “sealed”. The water temperature seems

“normal”. $601595 Identical to pump B. Head capacity and NPSH data as expected. Allowance of

140 kPa for the control valve. $151596 Why1? to prevent possible explosion. Why2? to save lives and keep equipment

intact. Why3? employee’s lives and health are paramount, to prevent lawsuits,to protect equipment, to maintain reputation for safety. Conclude: focus onWhy1? “to prevent possible explosion”. $50. If you didn’t put plant on SIS orSIS + evacuation before you asked this question, the plant explodes with lossof life. Penalty $3 000 000

1597 Pyrometer reads 228 –3 �C; temperature sensor reads 227 �C. $3001598 No improvement. $18 0001599 Mid-range. $901600 Design should handle separation well. Level sensors. Sufficient residence time

for both gas and liquid with well-designed demister mesh pads and vortexbreakers for gas and liquid, respectively. $6000

1601 The pressure is stable and sufficient to ensure good combustion in the burner.If didn’t put on safe-park as first step, then dangerous potential fire/explosionconditions are created while you experiment. $500 000

1602 510 increasing to 580 �C at exit of bed. Usually 510 increasing to 550 �C at exitof bed. Hot spot in bed! $1000

490


1604 Column settles out; concentrations of overhead and of bottoms are consistentwith design but we cannot meet the downstream production required. $4000

1605 Column gradually settles down. All values return to normal. $6001606 With the overhead temperature increasing I have increased the reflux to the

maximum but nothing seems to happen. This is out of control! $501607 Pressure about 1.72 kPa g; 5 �C superheat; no upsets and steam supply should

be reliable and constant.1608 Evidence of splay marks on the surface. Product still breaks. $8001610 Decreasing. $501611 Not needed. $5001612 18 �C and steady; past records show this ranges all over the place from 15 to

100 �C.1613 Excessive off-specification gaseous overhead stream especially on hot dry days.

$201614 OK. $25001615 Overflow effluent= 210 ppm; belt filter still flooded and cannot handle feed.

$60001616 Sufficient area supplied for design flowrate; natural circulation; designed for

mix of nucleate and film boiling on the shell side – the mechanism and fluxshifts as the liquid goes through the tubes. Changes from high-flux film tolower-flux film, then transition and then to high-flux nucleate. Considerationgiven for the vapor space to keep vapor to exit port reasonable.

1617 No. $601618 As best I can tell. $150. If you didn’t put the plant on SIS or SIS + evacuation

before you asked this question, the plant explodes with loss of life. Penalty$3 000 000

1619 0%. $801620 We’ve had four days of operation since startup after the turnaround. It has

been 3 times the South loop and once the North loop. $5001621 Being collected now during the commissioning. No other internal reports. $501622 0.5% w/w added to the resin hopper; this is added especially because of the

thickness of the hub 1.25 mm. $2001623 T4= 207.2 �C. If didn’t put on safe-park as first step, then dangerous potential

fire/explosion conditions are created while you experiment. $500 0001624 No lumps. $30001625 –101 �C1626 See Chapter 3: solids conveying, Section 3.2.4; fans and blowers, Section 3.2.1;

bag filters and cyclones, Section 3.5.2; mixing solids, Section 3.7.3.1628 Steady but higher than design. $501630 Pressure tap line is clean and clear. $3501631 Equipment should do the job. $1001632 Usual range 0.06 to 5 kg/s. Should do the job. Water hammer is unlikely;

usually sized on double the condensate rate. Here condensate rate is about2.7 kg/s so it should be OK. $800

1633 Steady but much below design. $50

491


1634 Responds to change. $1001635 The pressure should provide a head of about 47m of sludge; the static height it

blows against is about 15 m. Therefore there should be sufficient air pressureto overcome the head and the friction to blow out the line. $50

1636 Should work well. Baffles were in vertical. Vessel has air bleeds. Sealing stripsused on baffles. $500

1638 Clear, no plugs. $60001642 Mild, 18 �C; sunny; predicted rain; Spring time. Mixture of sunny and overcast

weather over the past week. $501643 Exit overhead is 0.65 kg/s which produces, according to flash drum, is 0.43

overhead and 0.24 liquid underflow or 0.67 kg/s. This is within the accuracy ofthe measurements. $600

1644 Not out of alignment. Alignment OK. No change. Level in flocculation tankstill below normal. $2000

1647 Column settles out; concentrations of overhead and of bottoms are consistentwith design but we cannot meet the downstream production required. $4000

1648 Calculated power consistent with power draw for “no flow”. $1201649 No fouling. $50001650 350 kPa g steam; sat. temp. = 148 �C; 1.48 MPa g; sat. temp= 198 �C. 30 �C

= 4.2 kPa abs1651 Thermowell location and “cleanliness” looks OK; downcomer is sealed; tray is

level. $20 000.1652 Clean; new looking. Water and air pressure tests show no leaks. $20 0001653 See Chapter 3: distillation, Section 3.4.2, perhaps pertinent condensers and

reboilers, Section 3.3.3; pumps, Section 3.2.3 and controllers 3.1.1. $501654 IS: the acetic acid-acetic anhydride system.1655 Partly closed. $3001656 Hot, sunny. 28 �C. This week has been one deluge of water after another. I

didn’t think it could rain that much! Today, however, it has cleared up toreward us with a nice sunny day. $50

1657 Close valve A to increase the gas flow and sweep the hot spot out of the catalystbed. $300

1658 Same before and after. Bottoms concentration IPA= 10% w/w. $80001660 Allowance for fouling on tube side= 0.0001m2K/W; shell-side steam

= 0.00001. This is the main condenser. The overhead product is drawn off bythe reciprocating pump. $130

1662 Lumps persist. $30001664 Higher than usual value for startup. $30001666 Agrees with head capacity. $2001667 Why1? so that the design flowrate of feed can go to the reactor. Why2? so that

the reactor can produce the design rate of product. Why3? so that we haveproduct to sell to customers. Conclude: focus on Why1? “to supply the designflowrate of feed to the reactor.” $50

1668 This is a closed system. Unable to get a reasonable check on the amount ofcondensate. No drain valve to use to measure condensate. $300

492


1669 OK. Recalibration not necessary. $20001670 Equipment should do the job. $1001671 Estimate seems OK. No major blockage. $1001673 Allowance for fouling on tube side= 0.00001m2K/W; shell-side steam

= 0.0006. $2501674 Usual value but fluctuating; past records show similar values.1675 No change. $1101676 Should give the correct% vaporization per pass. $7001677 Responds to change.1678 OK. Recalibration not necessary. $30001680 IS: on deprop. IS NOT: other units. $501681 The Dp for clean, carbon-free reaction coils= 5 kPa; but rapidly increases to 10

to 20 kPa because of carbon formation. This seems consistent with design val-ues and with values used to size/select the vacuum pumps.

1682 Erratic.1683 IS: startup of new double-load system. $501685 Same as expected. Consistent with the overhead composition at this pressure.

$12001686 OK. Recalibration not necessary. $30001688 TI/3 = –102 –2 �C; TI/4 = –101 –2 �C.1690 Accurate; no recalibration needed. $30001691 Polycarbonate’s impact strength is particularly sensitive to moisture in the

feed resin. Blend of polycarbonte and ABS: melt volume flowrate: 12 cm3/10 min; molding shrinkage parallel, 0.5 to 0.7%; molding shrinkage normal,0.5 to 0.7%; tensile stress 60 MPa; Izod notched impact strength (at 23 �C)640 J/m; Izod notched impact strength (at –30 �C) 267 J/m; HDT (0.45 MPa)127 �C; (1.84 MPa) 110 �C; injection-molding melt temp, 260 �C; mold temp.80 �C; injection velocity 240 mm/s. Drying temperature: 105 to 110 �C; dryingtime 3 to 4 h; max. moisture content 0.04%; melt and nozzle temp. 275 to300 �C; rear barrel, 250 to 290 �C; middle barrel, 255 to 295 �C; front barrel,260 to 300 �C; mold temp. 60 to 90 �C; back pressure 0.3 to 0.7 MPa; screwspeed, 40 to 70 rpm; shot to cylinder size, 30 to 80%; vent depth 0.038 to 0.076mm; hold pressure 50 to 75% of injection pressure; injection pressure 70 to140 MPa; clamp 45 to 75 MPa. $150

1693 Temperature increases consistent with vapor–liquid data for ethylene.1695 3/4 open whereas usual is mid-range. $3001697 Responds to change. $2001699 Explosive conditions 180% above lower limit were found at drop boxes B and

C. The composition is unknown; the test instrument showed it was explosive.No explosive conditions were found upstream at box A when it was tested as20% of explosive lower limit. It must be your plant. $300

1700 Everything OK. Clean as a whistle. $20 0001701 Steady and usual value. $501702 Temperature starts to drop: 185 and 130 �C; Temperature near lowest tempera-

ture: 175 and 128 �C; temperature starts to rise; for a brief period the tempera-

493


tures are 190 and 183 �C, then quickly drop to about 185 and 130; then cyclesrepeats. However, this is not a reproducible cycle. Sometimes the tempera-tures on either side are about equal, at 128 –4 �C. It seems to depend on thetime of day. $8000

1703 No improvement. $70001704 IS: sporadically any shift. $3001705 Cycle 1, 10 min Cycle 2 & 3 similar within experimental error in sampling and

analysis:’1: 0.5%:’2: 0.6%;’3: 1.2%;’4: 1.7%;’5: 2.1%;’6: 2.6%;’7:3.2%;’8: 3.7%;’9: 4.9%;’10: 5.0%. $4200

1706 Hot upstream, cool downstream.1707 Yes, T6 >T2. $501708 IS: this plant and downstream IPA processing. $6501710 Height is the same in both reactors. $13 0001711 Alumina dryer includes several beds of alumina; with two on-line and one

regenerated via hot gas heated by steam. All control valves have block valves,bypass plus valve, drain valve. FRC/1 is feedback control. The vapor and liquidethylene are attached to a compressor, condenser and drop-down valve in aconventional refrigeration cycle.

1712 Much lower than design. $501714 Respond to change. $5001715 The piping is complex and consists of a range of 3-way and 4-way valves to

allow any of the dryers to be on-line or regenerated. On-line is straightforward.The feed gas enters the top of the first dryer, exits the bottom, enters the top ofthe second dryer in series and then out the bottom to a knockout pot. Regen-eration is more complex because two different sequential activities occur dur-ing regeneration. First, hot “fuel gas” goes through the bed and sends off theadsorbed water. When most of the water has been desorbed from the bed ofalumina, the bed must now be cooled before it can be put back on-line. Fuelgas is used for both these functions. First, the desorption is done with “hot”fuel gas; then the cooling is done with “cooled” fuel gas. The fuel gas fromboth functions is returned to the fuel gas system. Case’24 gives the details ofa similar – but slightly different – system. More general information is inChapter 1 of Process Design and Engineering Practice (1995) by Woods. $100

1717 Ethylene inside the tubes should be film boiling because the temperature dif-ference for ethylene–butane is > 50 �C. $50

1718 Lumps persist. $20001719 Usual composition caproic, capryllic and capric fatty acids with small amounts

or lauric and myristic fatty acid. This feed does not contain more volatiles thanexpected. $1500

1720 I am following the usual startup procedure. Nothing is different about thecooling-water condensers. $3000

1722 Same as expected. $12001723 This is startup of a new plant. No maintenance has been done recently.1724 dPI slightly higher than usual. $501725 Seem consistent with expected gradients and responses. $300

494


1726 Responds to change. $3001727 Not needed; old impeller is 15 cm diameter and shows little erosion or corro-

sion. This is the correct impeller based on vendor files. No change. Level inflocculation tank still below normal. $8000

1729 See Chapter 3: fired heater/furnace, Section 3.3.2; blower Section 3.2.1; sen-sors and controllers, Section 3.1.3. If didn’t put on safe-park as first step, thendangerous potential fire/explosion conditions are created while you experi-ment. $500 000.

1730 IS: The crude unit was having troubles a couple of months ago but not now. Itwas producing off-specification products. $150

1731 Usual value. $801732 Everything is normal; you should be receiving on-spec feed at the design rate.

$2001733 Shell and tube. Allowance for fouling on tube side= 0.0001 m2 K/W; shell-side

steam=0.00001. $60001734 q process fluid= q latent heat. For E100: Fp,E100 (latent heat) or FpE100

438 kJ/kg. Hence, Fp,E100 = q process / 438; Actual Fp,E100 = (F cp)E100 90/ 438or 0.205 (F cp)E100. Design Fp,E100 = (F cp)E100 96/ 438 or 0.219 (F cp)E100. ForE101: Actual Fp,E101 = (F cp)E101 66/ 438 or 0.1506 (F cp)E101. Design Fp,E101 =(F cp)E101 58/ 438 or 0.132 (F cp)E101. Total propylene flow: Actual= 0.205(F cp)E100 + 0.151 (F cp)E101. If the flow and heat capacities of the processstreams are about equal then= 0.356. Design= 0.219 (F cp)E100 + 0.132(F cp)E101. If the flow and heat capacities of the process streams are aboutequal then= 0.351. $350

1735 No. Same supplier we’ve had for years. $1201737 Not successful. $10001739 Yes. $30001740 Not needed. $30001741 Faint weld line, but product still broke. $8001742 First section: Design (F cp) (1000–750) versus (F cp) (1000–820) or 250 versus

180 or 72% actual versus design. Second section: Design (F cp) (750–600) ver-sus (F cp) (820–730) or 150 versus 90 or 60%. First section Design versus sec-ond section; 250 versus 150 = 62.5% of heat load lost in first section. First sec-tion actual versus second section actual: 180 versus 90 or 66% of heat load infirst section. This incorrectly assumes that cp is independent of temperaturebut this gives us an idea of what is going on.

1743 Yes, we received a discount price of 20% off for the TDI from another supplier.Indeed, today is the first day you are using the new shipment of TDI. Thepolyol we are purchasing from the same supplier that we have always used.$300. If you didn’t put the plant on SIS or SIS + evacuation before you askedthis question, the plant explodes with loss of life. Penalty $3 000 000

1744 Not needed. $30001745 No change. $3001747 Fully open. $2001749 Temperature= 41 �C. $1800

495


1750 1.1 kg/L1751 Hazardous because of pressure, temperature and composition. Seems to be

expected process gases. $20001752 Very much so; it just can’t handle this underflow. Water is everywhere. $7001753 No change in performance. $3001754 Negligible improvement. Product breaks. $30001755 End suction. Head-capacity curve available. Details of NPSH required are sup-

plied.1756 Bypass valves: butterfly valves. $3001757 It was idle for 6 to 12 hours. I suspected a lot of settling had occurred, so I

blew back the lines into the thickener with air. $501758 Since the new installation. $6501759 Shell side 228; tube 164; leak shell to tube side or steam into glycerine. Same

pattern occurs in other effects. $701761 Hot 32 �C; sunny. It has been this way for the whole week. $201762 Insufficient vacuum/Dp too low/ pressure sensor wrong/ plugging conveying

line/ valve not open/ no air to convey the powder/ no powder in the hopper/powder is bridging in the hopper. $500

1763 Some improvement. Now can control –6 �C. $75 0001765 Fluctuating. $501766 Steady and mid-range. $3001767 Diagram is a reasonable general picture. What’s missing are the vent/blow-

down lines on both the shell and tube side of all exchangers; the drain fromthe knock-out pot that goes to the sewer. Each line has a single gate valve thatis normally shut. All control valves have block valves, drain and bypass withvalve. All equipment has isolation block valves on the inlet and outlet nozzles.All steam comes off the top of the header; all condensate going into the con-densate main enters the top of the main. Steam, town gas, condensate andprocess lines are on elevated pipe rack. $50

1768 Usual allowance for fouling on tube side and shell side. Correction factor fornot-true countercurrent flow is > 0.85. Vent/blowdown lines on both the tubeand shell side. Vent/blowdown lines discharge to sewer. Each vent line has asingle gate valve that is normally shut. Process pressure is > utility pressure.Fuel-gas pressure, 1.1 MPa > 0.65 MPa water. $200

1769 Excessive fouled with blobs of stuff. $4000. If you didn’t put the plant on SISor SIS + evacuation before you asked this question, the plant explodes withloss of life. Penalty $3 000 000.

1770 IS: hot and sunny. IS NOT: rainy day1771 P&ID supplied; Simulation report suggests that all equipment on this unit is

working close to design capacity. P&ID is less than a year old. No details avail-able in this office for upstream unit where catalyst was changed. $30

1772 IS: On the debut. IS NOT: upstream on the deprop. those problems have nowbeen corrected. $50

1773 Usual allowance for fouling on tube side and shell side. Correction factor fornot-true countercurrent flow is > 0.85. Vent/blowdown lines on both the tube

496


and shell side. Vent/blowdown lines discharge to sewer. Each vent line has asingle gate valve that is normally shut. Process pressure is > utility pressure.Fuel-gas pressure, 1.1 MPa > 0.65 MPa water. $1000

1774 Negligible. $30001775 Reflux rate increases but not to required rate; level in drum holding constant.

$20 0001776 Insulation seems to be sealed; intact. $1201777 Well designed. Demister included. Globe valve on periodic drain to sewer.

$6001778 DT> 25 �C (1000 or 750 or 600 less 325). Therefore all film boiling.1780 –102 –2 �C1781 Designed to handle the usual 50 kg/h of non-condensibles. Should do the job

well. $1101782 Sized to take the increased flowrate. If didn’t put on safe-park as first step,

then dangerous potential fire/explosion conditions are created while youexperiment. $50 000

1783 < 0.3%. $8001785 OK, recalibration not necessary. $7001787 See Chapter 3: pump, Section 3.2.3; sensors and control system, Section 3.1.3.

$501788 Design: F 4.2 kJ/kg K (70–18) �C or 218 � F; Actual: F 4.2 (42–18) �C or 100 �

F. This is a loss of F cp (70–42) or F 4.2 � 28 �C = 118 � F. The heat ratio ofactual/design= 14/52 = 0.27. $200

1789 Valve stems suggest that all valves are closed. No apparent leaks. $12 0001791 About one month ago, changed the oil in the hydraulics. Continually add a

suitable lubricant to the tie bars. $601792 Estimate seems OK. No major blockage. $3001795 Valve is fully open; valve responds to change. $4001796 End suction, centrifugal. 1800 rpm with impeller one size smaller than casing.

NPSH data and head-capacity curve available. An allowance was made of 70kPa Dp across the control valve. Should do the job. Special attention to seal toprevent acid leakage out. $300

1799 Not needed. $30001800 IS: excessive Dp in T103, the C2 splitter. IS NOT: other major variables1801 The blower can easily supply the amount of air to provide at least 10% excess

air for the range of fuel rates expected. If didn’t put on safe-park as first step,then dangerous potential fire/explosion conditions are created while youexperiment. $500 000

1802 OK, recalibration not necessary. $7001803 IS: temperatures in reactor bed, “hot spots”; production of ammonia. IS NOT

elsewhere on plant. $3001804 No. We cannot afford to have duplicate. $301805 Inadequate water flush/ flush does not go into the correct lines/ pump

stopped/ strainer plugged/ line plugged and cannot be cleared with the flush/

497


flush not working/ block valves leaking/ valves around pump closed/ pumpclogged. $50

1807 Much lower total amount exchanged than expected. Heat lost in reformer exitgas is same as that gained by stream to other units –10%. $800

1808 Sensor OK. Calibration checked out OK. $80001809 “We have not detected any contamination. The pressure is 1230 kPa g.” You

therefore conclude that the leak may be from the process stream into the pro-pylene on the very unit that is underperforming. $250

1810 Yes, insulated. The insulation seems to be intact. $5001811 The raw material for the cat-cracking unit, from storage tank T200, is pumped

and preheated via pump A followed by a preheater. A spare pump B is hookedup in parallel so that the process can continue when pump A is being serviced.$50

1812 No apparent fouling inside or outside of tubes. $20 0001813 Work OK. $1201815 323 �C. $30001817 Both appear to be closed. Valve direction consistent with direction of flow.1818 See Chapter 3: evaporators, Section 3.4.1; heat exchangers, Section 3.3.3; vacu-

um, Section 3.2.2; steam traps, Section 3.5.1; pumps, Section 3.2.3. $501819 Overflow effluent= 330 ppm; torque increased. $80001820 Steady and usual. $501821 Yes, consistent.1822 None done recently; but pump B is scheduled for maintenance next week. $151823 None available. $501824 Oscillating around 58% and decreasing; usually 62% and steady. $6001827 IS: both chiller exchangers. $501828 See Chapter 3: distillation, Section 3.4.2; exchangers, Section 3.3.3; sensors

and control, Section 3.1.3. $501829 Not upsets here. Flow, temperature and concentration should be the usual val-

ues with temp. = 98 �C; concentration 100% ammonia. $701831 About one month ago, changed the oil in the hydraulics. Continually add a

suitable lubricant to the tie bars. $501832 Four stage, reciprocating with kickback protection at the exit of the third stage.

10% overdesign. $60001834 Monitor the level in feed drum V-29 so that the amount fed to the column

does not exceed the flow to the drum. $501835 Yes, although it is not shown on the diagram. $5001837 The polymerizer is a jacketed stirred tank. The top entry stirrer has double

mechanical seals; the amps are recorded. Besides a view port and light andaccess hole, at the top are two nozzles to vent. The one nozzle has a pressure-relief valve; the other backup emergency line has a manual block valve.Another top nozzle is hooked to a manifold with three lines with block valves:vacuum; vapor recovery and equalizer line to the vapor space in the feed ves-sels. Another top nozzle is manifolded to three lines, each with manual blockvalves: emergency air, compressed air and nitrogen. A easy-clean angle block

498


valve is attached to the nozzle for each of the last three nozzles on the poly-merizer. A pressure-recorder sensor is also connected to the top of the reactor.The pressure gauge has a pigtail with a shutoff valve with a tee nipple andvalve installed between the gauge and the pigtail to allow backflushing of pig-tail. Inside the vessel are four vertical wall baffles extending 0.1 D into thevessel, with space between the baffle and the wall for easy cleaning. At thebottom of the polymerizer is an easy-clean angle valve. Four/three lines attachto this valve: a premix feed line, a blowdown line/ feed line to downstreamprocessing and a process water line. These each have manual block valves andcheck valves. Hot or cold water is fed to the jacket. The temperature sensorsindicate the temperature in the liquid in the reactor and a second sensor isused to control the flow of heating or cooling water via a TRC. The tempera-tures of the fluid entering and leaving the jacket are measured. The TRC valvehas block valves, drain and bypass with a valve. There is a pressure-relief valveson the jacket. $200. If you didn’t put the plant on SIS or SIS + evacuationbefore you asked this question, the plant explodes with loss of life. Penalty$3 000 000.

1838 They both are decreasing. The cooling-water exit temperature is decreasingand the steam production rate is decreasing. All else seems to be the same.$50

1839 Usual. $501841 Responds to change. $3001842 None reported. $1501845 Set at 10% open; usual value 10% and steady. $8001849 Temperature= 127 �C. $18001850 Usual value.1851 Seems steady at 1.2 MPa g. $501852 Set at 40% open; usual value 20% open. $8001853 Preheater 1: tube is > 200 kPa; shell is about 120 kPa; leaks from tube into

shell; steam into glycerine; Preheater 2: tube is > 190 kPa; shell is about170 kPa; steam leaks into glycerine. $60

1854 Sounds like cavitation.1855 Amps consistent with “no flow”. $801856 Fuller’s earth dumps into the bleacher at the design rate. $100 0001857 165 �C. $1501858 Mild, 18 �C and raining. Previously this week we have had thunderstorms and

wind. $1501859 IS: at exit of pipe reactor. IS NOT: elsewhere. $501860 Should do the job. $20001861 Reads 995 –7 �C1862 Resin premixed as polycarbonate and ABS; coloring agent and exothermic

blowing/foaming agent. $801863 P&ID supplied; P&ID is less than a year old. $301864 Rising. $501865 No evidence of fouling. $3000

499


1866 Well within design specs even though this is a hot and humid day. $3001867 No change in swinging-loop phenomena. $100 0001868 OK; head-capacity curve OK relative to estimated pressure requirement. $2401869 Appears to be fully open. The direction of flow through valve is consistent

with expected flow.1870 Steady and higher than usual, 105 �C instead of 101 �C. $501871 Glycerine: NFPA: 1, 1, 0. $501872 OK, recalibration not necessary1873 Watch out for oil contamination. The lubricating oil from the compressors will

poison the catalyst if it ever reaches the catalyst. $20001874 540 ppm. $40001875 No change. $1101876 Equipment should do the job. $1001877 In= 18 �C; Out= 42 �C. $13001878 Shut. $4001879 No apparent leaks on the four accessible flanges. Six flanges on the overhead

line to the condenser are relatively unaccessible. No apparent leaks on threevalve stems. $3200

1880 Valve was OK. No improvement. $10001881 Not needed at the beginning until after we have collected data. Indeed if put

on safe-hold we won’t be able to collect data to figure out what is wrong.$3000

1882 Not needed. $30001884 South loop: OFF; North loop: ON. $6001886 No observable reasons. Inside looks a little worn but otherwise OK. Impeller

looks OK and the key is in the keyway. $30001888 No valve. Would have to stop the process and insert a smaller orifice in the

resistance disk in the line. $3001889 Temperature-sensor error/ power failure/ cal rod heater failure/ gas flow too

high/ refrigerant condenser too cold/ cooling-water condenser too cold/ hydro-gen concentration < 64%/ startup instructions not followed/ startup pressuretoo high.

1892 IS: DAF tank.1894 Careful and methodical. Should not be a problem. If didn’t put on safe-park as

first step, then dangerous potential fire/explosion conditions are created whileyou experiment. $500 000

1896 IS: first time startup; IS NOT: ever worked1900 Solvent to remove carbon dioxide; mixture of tetramethylene sulfone, dietha-

nolamine and water; boiling temperature of pure diethanolamine= 270 �C.1901 Responds to change. $2001902 Process pump will be able to handle the anticipated flowrates and maintain

liquid flowrates > 1.5 m/s. If didn’t put on safe-park as first step, then danger-ous potential fire/explosion conditions are created while you experiment.$500 000

500


1903 From the reactant flowrate, measured upstream; the calculated steam usage(and condensate flow) are reasonably consistent with the design values. $3000

1904 Fouled on the outside of the tube. Scrape the scaling and analyze. Negligiblefouling on the inside of the tubes.

1906 Motor running. Shaft rotating. Pump noisy. $4001907 Extensive manuals including maintenance schedule and table of trouble-

shooting diagnostics. $8001908 Temperature leaving the cooling tower is about 23 �C; flowrate from the towers

is about usual for daytime; the temperature is lower at night. We don’t have adirect line to your units so we cannot say exactly how much is going to yourplant. $200

1909 Plate is installed correctly and orifice diameter correct. $20001910 Confirm the temps and pressures at the bottom and top of column are consis-

tent with the compositions expected. $1001911 Largest particle is 1.2 � aperture and is log normally distributed. No data avail-

able for previous operation. $21001912 Nothing unusual. $501913 100% open. $300.1914 Operation returns to design operation. PIC= 0.8 MPa for “usual” design flow-

rate. Response to change remains sluggish. However, as time passes,increased settings on PIC are needed to maintain even the usual capacity.$10 000

1915 Density 1.53 kg/L; viscosity 85 mPa s. $501916 Slight improvement. 73% yield. $32 0001917 < 150 mm diameter, angle of repose 15 to 30� (free flowing), moderately abra-

sive, fluidizes, hygroscopic and packs under pressure.1918 Moderate, sunny; today and all week. $501920 OK. $25001921 Seven hours into first time startup after changes: new catalyst; new feed sup-

ply. $501923 Internals look normal. Clearances are normal; wear rings are in good shape;

no pluggage; negligible erosion. $3001924 Conditions get worse. Condensate seems to build up in the exchanger. $42001925 The operating procedures were followed. $30001926 dPI Steady and usual. $501927 Pressure steady about 200 kPa g; no upsets.1928 Shell and tube. Allowance for fouling on tube side= 0.0001 m2 K/W; shell-side

steam=0.00001. $8001929 Functions OK. Is left fully OPEN.1930 0.5 mm diameter perforations; usual for steam. $4001931 Agrees. $3001932 “We have not detected any contamination. The pressure in the process line is

230 kPa g.” You therefore conclude that any leak would be from the propyleneto the process fluid. $300

1934 8.2 MPa. $150

501


1935 Temperature and flowrate are steady and about the same as usual. $2001937 Feedback control; orifice plate sensor; PID.1938 Initial increase and subsequent decrease in temperature. After about one to

two minutes (allowing sufficient time to purge the fire box of the excess fuel)the temperatures level out and production returns to normal.

1939 40%. $2001940 Orifice is sized so that, for the range of anticipated flowrates, the discharge

coefficient is independent of Reynolds number. $3001942 Half the design value.1943 Calculated yield 42%, 48%, 32%. $40 0001944 Expensive and no improvement. $82001945 No. We have had to order in new batches of all feed materials. $1501946 326 �C. $30001947 Fluctuating about 45 �C. $501948 Pipe is clear; injection line meets design specifications. $60001949 Nothing that looks like it would vibrate or chatter. $40001950 30 to 35 �C; past records show similar values.1951 Fully open. $3001952 No evidence of fouling. Problem persists. $15 0001953 Intense thunderstorm. Previously this week has been rainy most of the week.

$301954 Raining. It has been cool and mild, slightly sunny all week. $30. If you didn’t

put plant on SIS or SIS + evacuation before you asked this question, the plantexplodes with loss of life. Penalty $3 000 000

1955 Responds to change.1956 Much, much less hydrogen for the new conditions compared with previous

conditions. $40001957 Steady and usual value. $501958 Design superheat= 100 kg/s � 7.5 kJ/kg K � 28 �C = 21 MJ/s. Actual= 70 kg/s �

8.5 kJ/kg K � 118 �C = 68 MJ/s or 220% more heat transferred. Question!1959 Eight months ago. Routine check of rake, skimmer, calibration of instruments

and sampler. Usual turnaround checks and adjustments on pump and beltfilter. $60

1960 Shell and tube. Allowance for fouling on tube side= 0.0001 m2 K/W; shell-sidesteam=0.00001. $6000

1961 Standard literature giving usual steam usage for different vacuum conditions.$200

1962 IS: operators on the debut; IS NOToperators on other plants. $501963 OK. recalibration not necessary. $30001964 Not available.1965 Product did not break; overall cycle time increased from 68 to 78 s. $20001966 Estimate suggests no blockage. $4001967 No change. Still cannot control. $10 0001968 Cycle 1 10 min:’1: 3%;’2: 2.9%;’3: 3.3%;’4: 3.8%;’5: 4.2%;’6:

4.7%;’7: 4.9%;’8: 4.9%;’9: 5.3%;’10: 5.2. Cycles 2 and 3 similar:’11:

502


5.4%;’12: 5.7%;’13: 6.0%;’14: 6.1%;’15: 6.3%;’16: 6.4%;’17:6.6%;’18: 6.9%;’19: 7.1%;’20: 7.2%. No data available for previous operat-ing conditions. $2500

1969 I followed the standard startup procedure. I have used the same procedure forthe past six startups. $3000

1971 Inverted bucket with bypass and upstream strainer. Blows intermittently.Glove test shows hot upstream and cold downstream. Confirm that it shouldbe working OK. $500

1973 Usual values. No changes here.1974 Tried to operate as usual but pressure gauge in flare line= 112 kPa abs; com-

pressor stopped. $200.1975 Could have increased the capacity by installing a larger-diameter impeller or

shifting to 3600 rpm instead of the original 1800 rpm. NPSH requirementsfor larger-diameter impeller similar to previous. $400

1976 IS: at production rates > 80% the whole process is difficult to control. Usuallystarting with the South loops the production, and temperatures “swing”. ISNOT: steady production and temperatures on both loops. $300

1977 Should do the job; unclear as to why the washing system doesn’t work. $20001978 IS: gas temperature entering reactor. IS NOTelsewhere on plant. $30001979 OK. Recalibration not necessary. $30001980 400 kPa. $2001981 Water would flow into the reaction product.1982 Set at 10% open; usual value 10% and steady. $8001983 Cold runner mold with reciprocating screw injection; L/D of 20:1 with com-

pression ratio of 2 to 3:1. 100 ton, Toggle unit. Includes resin hopper. $1501984 Responds to change.1985 Direction of rotation and arrow are consistent. $1001986 No power failure. $150. If you didn’t put the plant on SIS or SIS + evacuation

before you asked this question, the plant explodes with loss of life. Penalty$3 000 000

1987 Yes. $1101988 New plant startup; no data. $501989 Allowance for fouling on tube side, steam= 0.00001 m2 K/W; shell-side stream

= 0.0005. Care to prevent temperature cross-over. $2001990 Flash steam flows to a steam header that runs from the ethyl acetate plant to

this location. Steam line comes off the top of the header. Steam traps alongthe line to try to ensure steam is saturated and clean when it arrives at theexchanger. $150

1991 Motor running. shaft rotating in the “correct” direction of rotation. $4001992 Seems to be independent of operators and the shift. $1501993 IS: undercooling on E100; overcooling on E101. IS NOT: meeting target exit

temperatures on either exchanger. $501994 Hot 30 �C, humid; yesterday was the same. $301995 Not needed. $500

503


1996 Design checks out OK. No errors made. Top temperature and pressure areconsistent with the expected overhead concentration. $200

1997 OK, recalibration not necessary. $7001998 No. $501999 Sized to take in increase flowrate. If didn’t put on safe-park as first step, then


2000 Design checks out OK. No errors made. No changes to feed location, feed tem-perature, type of trays. Top temperature and pressure are consistent with theexpected overhead concentration of IPA. $1000

2001 IS: just this morning. IS NOTprevious morning. $302002 Yes. $4002003 Adiabatic flame temperature calculated for 10% excess air but the measured

temperature in the combustion chamber is much lower. If didn’t put on safe-park as first step, then dangerous potential fire/explosion conditions are creat-ed while you experiment. $500 000

2004 Although this is a new plant, no information about commissioning and testswere recorded. $30

2005 For current actual conditions= 0.71 kg/s –10%. $12002006 Allowance for fouling on tube side, water= 0.0002 m2 K/W; shell-side

stream=0.0005. Care to prevent temperature cross-over. $4002007 I tried to keep the cycle time, the pressures and temperatures the same as was

used for the prototype. $2502008 Feels warm. $2002009 Cooling time lengthened to 51–60 s and product looks OK although the pro-

duction decreases because of the increased cycle time per piece. Lumps per-sist. Brown streaking at the same location. $2000

2010 Part of the design of the reactor. Incoming gas flows across tubes carrying hotreactor effluent; then on the tube side with the hot gas from the catalyst bedon the shell side. The heated feed gas then goes up and then down throughthe catalyst bed. Thermocouples monitor the temperature of the exit gas as itcools down. $6000

2011 Higher than design. $502012 As measured on the gauges Dp=110 kPa. $10002013 IS: reflux. IS NOT: elsewhere on the unit as yet. $502014 0.205 MPa g, the condensate header: 134 �C; Steam main: 1.5 MPa g: 201 �C.

At 1.2 MPa g: 192 �C. $3002015 Warm 25 �C; clear, windy. It was raining and thunderstorm about four days

ago. Since then it has cleared and settled into nice warm weather.2016 Estimate seems OK. Flowrate from the “other units” seem same as previously.

$7002017 Internal report of commissioning showed that all units function well and

work on design specification. The steam generated from the waste-heat boilerwas 50 to 120 kg/s with 28 to 30 �C superheat.

2018 Not needed. $3000

504


2019 No leaks. $50002020 No change. $12002021 62%. $6002022 Alumina. cycle time 12 h; two in series on-line; one off-line being regenerated

with fuel gas. Fuel gas should have < 6 ppm moisture. Alumina loading 0.14–0.22 kg water/kg dry adsorbent. $100

2024 Confirm the temps and pressures at the bottom and top of column are consis-tent with the compositions expected. $100

2026 Steady and usual. $502027 Works OK with flow 14 L/s when either pump B or A is working. Issue repair

order for next maintenance.2029 Current level= 2.5 m; usual level= 2.5 m. $402030 Steady and design value.2031 Responds to change. $1002032 Acknowledge that they recently opened a new leg of the line from Western

Canada. The line was hydraulically tested before the line was put into service.2034 Yes. $30002035 Not needed. $30002036 Steady and higher than usual. $502037 Why1? to ensure safe operation, correct production rate and quality product.

Why2? to meet customer’s demand. Why3? to keep company economicallyviable. Conclude: focus on “to ensure safe operation, correct production rateand quality product”. $40

2039 No improvement. $12 0002040 Yes. $1202041 Vendor has supplied standardized triple effects to many satisfied customers.

Vendor cautions that the purchaser install the unit “appropriately” into theexisting local environment. $300

2042 Valve stem is about 1/4 open.2043 No. $502044 None recently. $502046 Usual, no sounds attributed to “cavitation”. If didn’t put on safe-park as first


2047 Responds to change2048 No change. $13 0002050 Yes, we do expect the town gas to have < 6 ppm moisture. However, some

slugs of water may have been left over from the hydraulic test of the new lineand the result might be periodically high moisture levels but this should levelout to < 6 ppm as the residuals slugs of water disappear.

2051 P&ID supplied; Simulation report suggests that all equipment on this unit isworking close to design capacity. P&ID is less than a year old. $30

2052 The operator turns on these lines, I think I hear flow. I shut off the flush afterabout five minutes. There is still no flow of sludge to the filter. $40

2053 Cool cycle decreases from 45 to 42 s. Product still breaks. $800

505


2054 High-pressure alarm ceases; overhead temperature and pressures realign todesign values. $5000

2055 No apparent fouling inside or outside of tubes. $18 0002056 Silica-based catalyst.2057 Level seems to be constant and suggests coil is covered with liquid.2059 Hazardous because of pressure, temperature and composition. Seems to be

expected process gases. $20002060 54 kPa. $6002061 IS: since startup.2062 Pressure gauges not installed. Expect gas Dp, on average, = 0.025 kPa. Should

increase when gas flow increases. $502063 Inlet temperature consistent with data from utilities; outlet temperature below

design. $12002065 Not needed. $502066 IS: startup. $302067 For all exchangers, reboilers and condensers (E130, 131, 107, 108, 113, 114)

Process pressure is > utility pressure. For example, E131: fuel-gas pressure,3.1 MPa > 0.65 MPa water. $100

2068 Not needed.2069 Process liquid would leak into the cooling water. $502070 Design checks out OK. Demister included. $502071 PC-100 = 517 kPa; PI-200 = 510. Dp= 7 –1 kPa which is reasonable. $752072 Checks OK. Recalibration not needed. $6802073 Responds to change. $202075 No change in sample properties.2076 IS: The kickback valve on the compressor seems to be working OK now. $1502077 Midrange both before and after safe-park. If didn’t put on safe-park as first


2078 Slightly higher flow than usual. $502079 Well-designed for center-line injection. $1002080 Not needed.2081 IS: independent of operators. $502083 The scaling from the inside of the tube contains silica.2084 Mid-tank; usual value.2085 When: 8 months ago. Previously the regeneration gas for the dryers was sup-

plied by an independent, off-site utility who supplied gas via pipeline. How-ever, since we have basically “town gas” as the overhead from the demethani-zer we stopped purchasing the town gas and repiped the overhead from T101through heaters to the regeneration line. In addition during this turnaround,routine checks were done on equipment. $200

2086 Valve stem is fully open. $302087 Lumps persist. $12002088 All five samples are consistent with excessive C3 in all samples –10%. $4000.2090 Temperature= 93 �C. $1900

506


2091 Negligible fouling.2092 Sufficient pressure difference. Should work. $8002093 Two gates, aluminum mold. Product to have impact strength of 640 J/m at

room temperature. $702094 No. $802095 IS: startup. $502096 IS: startup, no evidence that it ever worked.2097 End suction, centrifugal, 1800 rpm. Should have no NPSH problems. $3002099 No sounds of cavitation. $1202100 Internals look normal. Clean as a whistle. Some evidence of slight erosion just

downstream of valve. $20 0002101 Flue gas is hazy, black flue gas. If didn’t put on safe-park as first step, then


2102 Yes. From past records, the timing varies and the concentration values differbut surges in high lab values for the bottoms “correspond with” surges in highvalues of C3 on AC/1. $700

2103 Much greater than design. $6002104 Amount fluctuates erratically.2105 Usual value. $1202106 Controlled and steady. $2002107 Sample’1: 650 ppm;’2: 720 ppm;’3: 690 ppm;’4: 705 ppm;’5: 698 ppm.

$20002108 Test’1: flow= 0.5 kg/s; Test’2: flow= 0.01 kg/s; Test’3: flow= 2.1 kg/s.

$10002109 0.97 MPa. $1002111 Reads higher than expected. $1502112 52 trays= 120 kPa. $2002113 54 kPa. $6002114 Confirm the temps and pressures at the bottom and top of column are consis-

tent with the compositions expected. $1002115 Startup of new process. $3002116 48 trays= 104 kPa. $2002117 Looks OK. No improvement in operation. $10002118 See Chapter 3: injection molding, Section 3.9.6. $2002119 No blockage. $15 0002120 Temperatures in bed are consistent with the exit temperature of about 1000 �C.2122 As expected for increased flow of naphtha to provide the same space velocity.

$3002123 Steady. $3002124 No reports available. $3002125 Yes. $1002126 Diagram is reasonably complete. Missing are isolation block valves for the

strainer trap and a drain line valve within the block valves. Upstream there is a

507


flowmeter to measure the boiler feedwater. The steam comes off the top of thesteam header. $50

2127 No. $5002128 IS: cycling steam flowrates, temperatures and pressures. IS NOT: glycerine

feed. $502129 Dimensions are correct. Missing are the isolation block valves around the

pump; the pressure gauge on the exit of the pump before the block valve; theblock valves, drain and the bypass with valve on the control valve; and thedrain, feed nozzles and vent on the acid storage tank. Also missing are thesafety showers, safety equipment storage for gloves, protective clothing relatedto the hazards associated with this process. $120

2130 Usual allowance for fouling on tube side and shell sides. Vent/blowdown lineson both the tube and shell side. Vent/blowdown lines discharge to sewer. Eachvent line has a single gate valve that is normally shut. Process pressure is > uti-lity pressure. 3.3 MPa> LP steam at 0.85 MPa. $1000

2132 Difficult because the steam flowrate is not recorded. $502133 Slightly higher than usual value.2134 Self-standing column with the air-cooled condensers about 10 m above

ground. The overhead line from the top of the column goes vertically downand then across to the feed header of the condenser. The exit from the liquidheader containing the condensed IPA goes directly to the overhead receiver.From the overhead receiver, the condensed IPA is pumped to downstreamunits for further processing. Any uncondensed inerts from the overhead recei-ver go to the vent scrubber system. There are sensors on most lines. There areisolation block valves around all pieces of equipment. For all control valvesthere are isolation block valves, drain and bypass plus valve. $650

2136 No improvement. $12002138 Steady and mid-range. $3002139 No change. $30002140 Yes. If didn’t put on safe-park as first step, then dangerous potential fire/explo-

sion conditions are created while you experiment. $500 0002141 More than enough supplied: 2.5-cm diameter line, welded connections with

four elbows. Vertical height 11.1 m.2143 Typical feed is 3 to 6% solids. Polymer conditioner added to aid filtration. Typi-

cal product is 30% “heavy solids”; 20% light solids; 50% water. $502144 Pressure on the steam drum is steady at 3.4 MPa g. We’ve had no upsets. The

flow is steady. The degree of superheat is not excessive but sufficient to pre-vent you from having wet steam in your turbine. $300

2145 Not needed. $10 0002146 OK. Recalibration not necessary. $30002147 Selected based on a condensate load of 2.5 kg/s; discharge to atmosphere;

common header from all stages with single inverted bucket; sized orificebased on Dp= 125 kPa. Upstream strainer. $120

2148 No leaks.2149 No problem found in the linkage. No change in operation.

508


2150 Responds to change. $20002151 Responds to change. $1002152 Trace, trace, trace, off-spec, off-spec, off-spec, trace, off-spec, trace, trace. $20002153 It has really been trouble since startup! It seems to randomly fluctuate all over

the place. I can’t sort out any trend. The feed is steady as a rock; the pressureis steady; everything else seems to be steady except for the production rate!

2154 OK. recalibration not necessary. $30002155 Seal is partially failed. Clearance OK. $20002156 62%. $6002157 q= (F cp)E101 DT; Actual= (F cp)E101 (70–4) or (F cp)E101 66; design= (F cp)E101

(68–10) or (F cp)E101 58 or Actual is 14% higher than design. $3002158 Same as usual concentration. $25002159 Not successful. $10002160 Yes, responds. $2302161 IS: gradual decrease in flow and today there is no flow; IS NOT: past three

weeks. $502162 Why1? to close the mass balance on the IPA column. Why2? to ensure that

there is not a loss of chemicals from any unit. Why3? so that our corporationcontinues profitably. Conclude: focus on Goal: “to prevent the apparent loss ofIPA”. $650

2163 Standard literature. $2002164 Diagram is a reasonable general picture. What’s missing are the vent/blow-

down lines on both the shell and tube side of all exchangers; the drain fromthe knock-out pot that goes to the sewer. Each line has a single gate valve thatis normally shut. All control valves have block valves, drain and bypass withvalve. All equipment has isolation block valves on the inlet and outlet nozzles.All steam comes off the top of the header; all condensate going into the con-densate main enters the top of the main. Vapor samples can be taken from alldrop boxes and sewer gates. $50

2167 Not needed for safety purposes. $2002168 Need safety interlock and emergency shutdown. $2002170 IS: boiler on exit of reformer gas. IS NOT: upstream2171 Some liquid. No change in pressure drop in column T103. $3002172 On spec. No complaints from other customers for that batch. $3002173 Tapping suggests height is mid-vessel, at the design location.2174 Minor improvement. Concentration of propane in overhead to flare decreases

to 1.25 � design value. $50002176 Pressure gauge reading wrong/ cavitating pump/ wear rings worn causing

internal circulation/ impeller worn/ blockage in line/ change in density of liq-uid so that pressure gauge reading converts to a different value of “head”/faulty check valve/ faulty block valves/ pressure-sensor line plugged.

2177 Pressure-relief valve is not faulty. No improvement. $10002179 Confirm that the system is temperamental and difficult to control. Sometimes

we just don’t condense enough of the high boiler and the vapor from the over-

509


head drum has too many high boilers in it. We have to flare it. It seems towork OK in the winter and when its raining. $30

2181 On file, a thick manual but it does not include a section on trouble shooting.The specifications suggest that the new condenser should do the job very wellwithout a trim cooler. $1000

2183 Steady. $3002184 Chin lists A. three methods that vary the vapor rate; B. seven methods that

vary the area; C. three methods that vary the temperature difference; D. sixmethods that vary the heat flux and E two other methods. For this clean andmoderate pressure system, where there is “full condensation” the recom-mended options should be from options B, C, or D. Chin recommends B2.Method D1a was selected. $230

2185 No change reported upstream. $202186 Responds to change. $2002187 Bumps appear on the product opposite to the knock out pins. Lumps persist.

brown streaking at the same location. $20002188 Since the startup following the new supplier of TDI, I was meticulously care-

ful to follow the standard procedure. $200. If you didn’t put the plant on SISor SIS + evacuation before you asked this question, the plant explodes withloss of life. Penalty $3 000 000

2190 Clear. $25002191 P3= expected value, slightly below atmospheric pressure. If didn’t put on safe-


2193 Operators confirm at design flows the temperatures are –60 to –55 �C; at halfthe design flow the temperatures are –90 to –96 �C.

2196 All five samples are consistent with excessive C4 in all samples –10%. $45002198 Within specifications. $30002199 Design= 218 � F =UA LMTD or UA/F = 218/104 = 2.1; Actual= 100 � F =UA

LMTD or UA/F = 100/ 90 = 1.1. The UA actual/ UA design= 0.5. If the flowthrough the tubes, F, has not changed, then either the outside heat-transfercoefficient dropped by half or the area became flooded and half the area islost. $500

2200 IS: at high capacities of 150 Mg/d. IS NOT: low flowrates, 70 Mg/d.2201 Both valves seem to be fully open. $902202 OK, recalibration not necessary. $7002203 Sounds like the inverted trap is functioning very well and consistently.2204 OK. rechecking not needed. $8002205 Two gauges agree within –4 kPa. $3002206 Moisture < 0.03%. $35002207 No leaks, no blockages, no fouling. No improvement in process operation.

$40002208 In E100: DT ranges from 105 to 10 �C; hence goes through nucleate and film

boiling. In E101: DT ranges from 73 to 15 �C; hence goes through nucleateand film boiling. $50

510


2209 IS: Crystal moisture is 4.5% coming out of rotary dryer. IS NOT: <1.5%2211 Why1? to provide steady, reliable operation with bottoms meeting specifica-

tion; Why2? to meet production requirements; Why3? to sell products andkeep company financially viable. Conclude: focus onWhy1? “to provide steady,reliable operation with bottoms meeting specification”. $50

2212 Hazardous because of pressure, temperature and composition. Seems to beexpected process gases. $2000

2213 No sounds of cavitation. $902214 Not needed. $50002215 No data for this configuration. Internal report for initial installation of single

chiller unit shows all performance data agreed with specifications. No sur-prises. Very reliable, very stable unit. $300

2216 Test OK. $402217 Appears to be closed.2218 See Chapter 3: control, Section 3.1.3; condensers, Section 3.3.3. $502219 No upsets. Steady operation. Usual pressure delivered to site. $302220 OK, recalibration not necessary.2221 Bottoms temperature increases by 3 �C; bottoms composition is 8% organic;

flowrate from pump 114-J of overhead is still very low.2222 Set at 22 kPa g.2223 See Chapter 3: For these symptoms, it is difficult to identify specific sections.

Perhaps related to steam distribution, Section 3.3.5; and steam traps, Section3.5.1; exchangers, Section 3.3.3 or valves, Section 3.1.3. $50

2225 Level is not steady, sometimes it doesn’t even show in the sight-glass.2226 I am following the usual startup procedure. Nothing is different about the

refrigeration condensers. $30002227 30% higher than design. $502228 This diagram presents just the basic connections among the equipment. Miss-

ing are sensors, controls, isolation block valves, vents, drains, safety relief andSIS systems. As usual, there are isolation block valves, drain and bypass withvalve around all control valves. A line brings fresh hydrogen into the feed lineto the reactor. The sensors shown on the diagram are the bare minimumneeded to understand, and solve, this case. $300

2230 148 �C with some fluctuations; past records show similar values.2231 Routine check on stirrer two weeks ago. $50. If you didn’t put plant on SIS or

SIS + evacuation before you asked this question, the plant explodes with lossof life. Penalty $3 000 000

2232 Respond to change. $5002233 56 –1%. $20002234 No controls, other than the controls within the screen and centrifuge system.

The bottoms from the crystallizer are pumped to nozzle feed to the DSMscreen. The crystallizer is a continuous process. The screen is a batch processbecause it needs to be washed and cleaned periodically. The crystals exit fromthe screen flows by gravity into the centrifuge except when the screen is beingwashed (when there is no flow to the centrifuge). The centrifuge is a batch

511


process because of the washing cycle. The discharge from the centrifuge tothe dryer is continuous except that there is no crystal flow during the washcycle in the centrifuge. The dryer is a continuous process. $30

2237 IS: since we started ISO testing. IS NOT: before when we didn’t sample thisfrequently

2238 Two gates, QC-7 aluminum mold. $502240 IS: at startup after major turnaround. IS NOT: before shutdown. $502243 Yes. $202244 OK. Recalibration not necessary. $30002245 IS: oil deodorizer. IS NOT: apparently elsewhere on plant.2248 Our customer demand was for full load. I didn’t think we could handle it.

However, I set the PIC at our max. But we just couldn’t evaporate and heat theethylene enough. The unit just shut down because we couldn’t meet the mini-mum pipeline temperature! $100

2249 On spec. No complaints from other customers for that batch. $4002250 Hydrochloric acid NFPA 301; causes burns and irritation to eyes, skin and

lungs. Decomposes to HCl vapor. $502251 Slightly higher flow than usual. $502252 TDI: NFPA 211; vapor is extremely irritating, may cause respiratory reaction,

eye irritation. Contamination or excessive heat may result in dangerous pres-sure. TDI caused cancer in animals. For stability, avoid contact with acids,bases, amines and oxidizing agents. Thermal decomposition may produce car-bon monoxide and/ or oxides of nitrogen. STEL 0.02 ppm; TWA 0.005 ppm$50

2253 Special design to include internal baffle. Care needed to account for extremesin thermal expansion. Allowance for fouling on shell side, water-boil-ing= 0.0002 m2 K/W; tube-side process gas= 0.0005. An allowance of 10%more area than needed for design flowrate. Single pass on the tube side withunique baffle design for secondary part of the boiler.

2254 See Chapter 3: sensors and control, Section 3.1.3; heat exchange, Section3.3.3; steam traps, Section 3.5.1. $700

2255 P&ID shows control system for the unit.2256 Appears normal; clearances and settings appear normal.2257 Accurate; no recalibration needed. $30002258 About 87% design rate; steady. $6002260 No change. $90002261 Sensor is OK; negligible noise in the sensor. $8002263 No leaks. $50002264 Temperature= 101 �C. $18002265 Before pump= 10.1 –5% glycerine; into the first stage is 9.9% glycerine. $30002266 Typical hydrocarbon with NFPA rating 1, 3, 0.2267 Product did not break; overall cycle time increased from 68 s to 88 s. $20002268 Not needed. $20002270 Accurate; no recalibration needed. $3000

512


2272 Today is warm and humid. When the previous parts had been made, theweather was variable spring weather: windy, rain, cloudy, sun with variabletemperatures 15 to 22 �C. $50

2273 2.7 kPa absolute. $302274 No upsets. The flowrate and temperatures are consistent with what you

usually receive. $300. If you didn’t put the plant on SIS or SIS + evacuationbefore you asked this question, the plant explodes with loss of life. Penalty$3 000 000

2277 Slightly higher than usual. $4002278 Design should handle separation well. Level sensors. Sufficient residence time

for both gas and liquid with well-designed demister mesh pads and vortexbreakers for gas and liquid, respectively. $800

2279 IS: Out of rotary dryer.2280 No leaks, no blockages, no evidence of foaming, no fouling. No improvement

in process operation. $40002281 The product should flow by gravity to the storage tank without the need for

pump 114-J.2282 Usual cycle is to power the heaters, add feed to hopper, start mixer, mix 35 s,

stop rpm and screw advances to push melt into the mold, screw reverses, feeddrops from hopper into mixer, mix 35 s, and so on. The resin is predried for3 hours by direct contact with hot air at 95 �C. Resin is dried, and componentsare fed to the hopper. The feed mix enters the extruder that melts and mixesthe melt through a combination of barrel heaters and screw mixing. After themelt is mixed the melt is pushed through the nozzle, sprue and runner systemand enters the mold through the gates. Air, initially present in the mold, isdischarged from the mold through the air vents in the mold. Coolant runsthrough the upper and lower mold. Once the cooling cycle is completed, themold is open, the part is discharge; the mold is closed and the cycle repeats.$150

2284 No change. $200 0002287 Diagram is a reasonable representation. $2002289 339 kPa; usually much, much less. $200. If you didn’t put the plant on SIS or

SIS + evacuation before you asked this question, the plant explodes with lossof life. Penalty $3 000 000.

2290 Yes. $1202293 Fan runs at correct speed; balanced operation; no wobble; pitch of blades

changes from negative to extreme positive. $3502294 Based on the flowrates from the unit, the space velocity seems to be the same

as previously used. The mass flowrate through the exchangers should be thesame as previous. $900

2298 No improvement.2299 All OK. $1202300 No improvement. $300

513


2301 Flow FC/1 = 6.8 L/s. Flow F/7 = 6.5 L/s. If didn’t put on safe-park as first step,then dangerous potential fire/explosion conditions are created while youexperiment. $500 000

2303 Moderate, 21 �C; severe thunderstorms. The previous week has alternatedsunny and cloudy days. The temperature has been about the same. $20

2304 Steady and usual design value. $502306 129 �C. The usual value is 135 �C. $1202309 dPI Steady and lower than usual. $502310 231 �C. $1502312 Accurate; no recalibration needed. $30002314 5 s; slightly longer than usual because of the thickness of the product, espe-

cially the hub. $2002315 Not needed. $40002316 Estimate suggests no major blockage. Flowrate is consistent with space veloci-

ty used previously. $8002317 IS: “all the non-condensed gas goes to flare”. IS NOT: negligible going to flare.2318 About once every 24 hours; sometimes on my shift; sometimes on someone

else’s. $3002319 Responds to change. $1002320 High and rising. $802321 99.9 –1%. $20002322 90 kPa (9 m head) $152324 Well designed based on pressure, residence time plus vortex breaker and dem-

ister pad. $1502325 Signal output is 100% indicating fully open. This is a fail-close valve. $302327 Hydrogen: 70.2%; nitrogen 23.1%; methane and other 5.9%. $20 0002329 I’ve only been on this job for 3 months. I have never seen anything like this

before. $502330 No. $502331 Not needed. $10 0002332 This is startup.2333 Both OFF. $3002334 Valve-stem movement suggests valve works and is fully open. $20002335 Why1? to ensure safe and reliable operation; Why2? to meet production

requirements; Why3? to sell products and keep company financially viable;Conclude: focus on Why1? “To ensure safe and reliable operations.”

2336 IS: on deprop. IS NOT: other units.2337 IS: cool to –96 �C at half flowrate and –60 �C at full flowrate. IS NOT: cool to

–96 �C at full flowrate2338 Flow is constant and at design value; temperature going to the chiller unit has

Temp= about 70 �C. We are not pleased that the temperature of the stream thatwe are getting from your chiller unit is so cold! We need it to be 10 to 11 �Cand not the 4 to 5 �C that we are currently getting. Do something on your chil-ler unit or we are going to have to shut down! $500

2339 No change. $10 000

514


2340 Tapping suggests that the height is half-way up the tube bundle. Half thetubes are covered.

2341 Not needed. $24 0002343 No sounds heard. $202344 Units are neither over- or under-designed. $2002346 Estimates agree with Dp. $2002347 Under design conditions, the cooling coil removed the appropriate amount of

heat. $6002349 No improvement. $60002350 OK. Recalibration not necessary.2352 Responds to change2353 Sample: 0.2% suspended solids. $3002354 Set at 0% open; usual value 20% open. $8002356 Confirm that the system is temperamental and difficult to control. Sometimes

we just don’t condense enough of the high boiler and the vapor from the over-head drum has too many high boilers in it. We have to flare it. It seems towork OK in the winter, when its raining and often on the night shift. $40

2358 IS: pump and control system IS NOT: apparently upstream. $502359 That particular batch tended to blob. Several complaints from other customers

for that batch. $3502360 The increased duties in the secondary reformer cause traces of silica from the

catalyst to be vaporized in the presence of water vapor and carried along in theprocess gas stream. As little as 1 ppb of such silica can foul the surfaces of thecritical waste-heat boiler. The first effects of this fouling do not normallyappear until after three months of operation and the tendency disappears afterabout nine months. See Chem. Eng. June 5, 1967, p 125.

2361 IS: cycling of the amount of product. IS NOT: steady production2363 All valves appear to function OK. $8002364 Yes. All valves appear to be closed on tube side. $8002366 Steady and almost fully closed. $2002367 23 actual trays; allowed two extra trays. $3502368 IS: when only A is running; B is stopped. IS NOT: when B is running and A is

stopped2369 The level seems to be about 1 m above the exit nozzle and it is rising. $3002370 Takes about 2 days. Nothing to be seen. $40 0002371 Steady as usual at this end; the production might be a little lower than usual.

The feed you receive should be on spec. –1%. $502372 Diagram reasonably complete. Isolation block valves on process line entering

and leaving the shell side of the preheater. Isolation block valves, drain andbypass with valve around the control valve V100. $50

2373 Diagram is a reasonable representation. In addition, the torque of the rake ismeasured and displayed in the control room. The sludge flow is measured atthe outlet of the pump. The pump has isolation valves with a pressure gaugeon the discharge line upstream of the block valve

2374 Yes.

515


2375 Yes, both were batch from BZ684. $20002377 61 �C. $302379 Connect with flow controller FC-5. Good move! However, if didn’t put on safe-


2380 Allowance for fouling on tube side= 0.00001m2K/W; shell-side steam= 0.0006. Should have sufficient area to boil off propane to give on-spec bot-toms. $250

2381 Seven hours into first time startup; changes have been made to correct faultson the upstream deprop. Case’41 $50

2382 112 kPa. $4002383 Insulation in good condition; noise and temperatures as expected. It’s a hot

day out here and it’s even hotter by these exchangers. $3002384 No distinctive noise. Valve not open. No evidence from flare that the pressure

relief has opened. $3002385 No change in either light ends or in flowrate. $1782386 Looks OK. No improvement in operation. $10002387 502 ppm. $40002389 Some improvement; by adjusting level below complete coverage of the bundle,

T/8 reads 9 –2 �C; With care we can get it closer to the target 10. $20002390 Responds to change. $2002391 IS: problem is at boiler feedwater heater exit. IS NOT: elsewhere on the unit.

$502392 The temperatures and flows go wild in the 5 minutes after we first notice a hot

spot but it takes us about 1 to 3 hours to get things settled out after an upset.$300

2394 Feedrate the same: feed concentration the same 50% w/w IPA. The feed rateis 2.78 kg/s –10%. $2000

2395 Tray collapsed in the stripping section/ too much bottoms fed to the debutani-zer/ too much overheads in the feed/ feed valve FV-1 stuck/ pump F-26 notworking/ check valve on the idle pumps allows backflow/ no feed left in feedvessel V29.

2399 IS: C2 splitter, T103. IS NOT: T101 and T1022400 Not needed. $20 0002401 Steady and higher than usual, 108 �C instead of 101 �C. $502402 Header 1.7 MPa g; steam drum 1.72 MPa g.2403 Diagram seems to be a reasonable schematic representation. For all the con-

trol valves there are isolation block valves, a drain valve and a bypass withvalve. The chlorine storage tank has drain and fill lines attached to nozzles onthe tank. The pump has isolation block valves with a pressure gauge on thedischarge line followed by a check valve. $250

2404 Reads as expected $1502405 Steady at 1.73 MPa. $5002406 Not needed. $10 0002407 Level in indicator shows that the butane should cover the full bundle. $100

516


2408 IS: now. IS NOT: values recorded for making 100 kg/s of steam.2409 Responds to change. $2002410 Acetone, n-hexane, methanol. $2002411 No. $202412 PID and carefully tuned. $4002413 180 to 185 �C; fluctuates.2414 Flowrate and temperatures are as expected. We increased the biocide slightly

because the days are getting warmer, but nothing surprising is happeningover here. I hear you are having fun! $230

2415 Higher than usual value for startup. $30002416 Dirty machine/ dirty hopper/ moist feed/ too many volatiles in feed/ lubricant

or oil on mold/ incorrect mold lubricant/ feed contaminated during materialhandling/ faulty raw material from supplier/ poor shutdown procedures/screw rpm too high/ sensor error/ shutoff valve dirty or clogged/ injectionpressure too high/ gates too small. $50

2417 About 87% design rate; steady. $6002418 Higher than expected. Higher than before the shutdown. $3002419 Temperature is 23 �C; flow is steady.2420 No. $202421 No change; still cycles. $10002422 IS: with this shift starting up. IS NOT: with previous shifts last week. $50. If

you didn’t put plant on SIS or SIS + evacuation before you asked this question,the plant explodes with loss of life. Penalty $3 000 000

2423 See Chapter 3: distillation, Section 3.4.2; exchangers, Section 3.3.3; sensorsand control, Section 3.1.3. $100

2425 No characteristic noise. Valve not open. No evidence from flare that the pres-sure relief has opened. $300

2427 Response to change has been sluggish since startup. Over the past fewmonths PIC at 0.8 MPa for the “usual” ethylene load. Brief period about twomonths ago when demand was below usual and PIC setting was at 0.7 MPa.Since then, for the usual demand the PIC settings have been increasingslightly as time progresses until recently the PIC setting was 0.89 MPa for theusual load. $150

2428 Usual value even at the increased flowrate. If didn’t put on safe-park as firststep, then dangerous potential fire/explosion conditions are created while youexperiment. $500 000

2429 Process pressure > atmospheric. $2002430 Sample 1: 0.96%; 2: 0.99%; 3: 1.04%; 4: 1.02%; 5: 0.98; 6: 1.01%; 7: 1.04%; 8:

1.06%; 9: 0.89%; 10: 0.97%; 11: 1.08%; 12: 1.05%; 13: 1.03%; 14: 1.02%; 15:1.05%; Insert swing: samples during 10-minute swinging loop: S1: 1.02%; S2:1.03%; S3: 1.02%; S4: 1.03%; S5: 1.03% 16: 1.08%; 17: 0.99%; 18: 0.96%; 19:0.93%; 20: 1.01%; 21: 1.02%; 22: 1.04%; 23: 1.06%. The swinging occurredduring the start of hour 15. $80 000

2431 No upsets. Steam pressure is about the same at the two locations. $5002432 As measured on the gauges Dp= 140 kPa. $1000

517


2433 All valves appear to function OK on shell side. $8002434 Drop box B: 55% hydrocarbon consisting of 64% methane; 32% hydrogen and

2% ethylene. This is the same composition as the overheads from tower T 101that is used as regeneration gas for the dryer. Drop Box C: same as for B with-in experimental error. $1500

2435 12. MPa g, saturated steam temperature 325 �C. Latent heat of steam, 1188 kJ/kg;heat capacity steam 7.5 kJ/kg K.

2437 No change; original valve looked OK. $45 0002439 Not needed. $20002440 No apparent contamination. 99.9% ethylene.2441 Signal output is about 80%. This is a fail-close valve. $302442 All five samples are about 2.5 times design value –10%. $45002443 Higher than usual 0.532 MPa and increasing. $502444 OK. Recalibration not necessary. $30002446 Allowance for fouling on tube side= 0.0001m2K/W; shell-side steam

= 0.00001. As the name suggests, to condense any vapors that have not con-densed and to ensure that these do not solidify in the knockout pot V1402 orin the barometric condenser E1402. $120

2447 25 �C. $2002448 Usually very fast. If there is a mega-change in residence time, then perhaps a

consideration. $1202449 On specification. $1602450 Steam 228 kPa abs, 123 �C; 164 kPa abs, 114 �C; 115 kPa abs, 103 �C. $1002451 Without data from the instruments on the unit, office-based calculations of

the exchangers before the change suggest the exchanger should do the job. Ifthe flowrates or hydrogen or water composition of the exit gas from the refor-mer change, then we need to revisit the calculations. $200

2452 Specialized design. Should do the job. Some designs use a standby furnace tobring the system up to temperature. This design uses cal rod heaters. Thermo-couples in the catalyst bed and in the feed gas entering the bed. $6000

2453 Allowance for fouling on shell side, process gas= 0.0005 m2 K/W; tube-sidesteam superheating= 0.0005.

2454 Consistent. Heat picked up in cooling water= heat to be removed from reac-tant gas.

2456 Sounds as though air is flowing; gauge pressure gradually increases. $15 0002457 On specification. $1602458 Design checks out OK with pressure drop allowance of 140 kPa for the control

valve. Check valve on discharge. No gauge. $702460 Yes. $202461 Steady and slightly higher than usual. $502462 100 kPa. $3002463 Plugged at sampler inlet. $1202465 No improvement. $30 0002466 46 –1%. $20002467 Controller output suggests negligible stiction of < 0.1%.

518


2468 Cooling surface area is double that needed. Fully equipped with charge ves-sels, instrumentation, pressure relief and SIS. $200. If you didn’t put plant onSIS or SIS + evacuation before you asked this question, the plant explodeswith loss of life. Penalty $3 000 000

2469 Responds to change. $2002470 OK. Recalibration not necessary.2471 This plant has never worked right.2472 Utility pressure > atmospheric. $502473 Cold runner mold with reciprocating screw injection; L/D of 20:1 with com-

pression ratio of 2 to 3:1. Toggle unit. Includes resin hopper. For this product,the shot is 40% of the machine capacity. Screw diameter, 6 cm; maximumrpm, 270 rpm; maximum pressure 137 MPa; stroke length 20 cm; injectionspeed 600 cm3/s. $200

2475 Speed same as specs.2477 Yes. All valves appear to be closed on shell side. $8002481 IS: lumps are in the product when viewed by transmitted light. IS NOT: seen

when viewed by reflected light. $502482 Not needed. $30002484 This is not a double-suction pump. This is an end-suction pump! Can’t hap-

pen. $502485 U tube, horizontal. $1002486 Allowed 130 kPa Dp for the control valve. NPSH required= 2 m water. Simula-

tion shows that unit was operating very close to design values for usual rangeof feedstocks and conditions. $200

2487 Amps reading agrees with expected values. $4002488 Usual design value. $6002489 Units are neither over- or under-designed. Mechanical baffles and extra head

space allowed to account for any foaming. $2002490 Typical for the overhead material: 0.1 kPa; T = 85 �C; 0.5 kPa; T = 118 �C. Liquid

heat capacity 2 kJ/kg K; latent heat about 200 kJ/kg. $802491 Much greater concentration of C4 and C5 than before. $6002492 If less neutralization occurs, then there will be less heat to remove. Could we

be removing too much heat and overcooling? $902493 No fouling.2494 Warm and cloudy 20 �C. We had a thunderstorm about four days ago; then the

weather continued clear and mild. $102495 TIC- 3 increases; TI-5 increases.2496 No change.2497 Warm, sunny 28 �C. Some rain sprinkles at the start of the week; but it cleared

up to this nice weather. $22498 Valve at full open position. Valve is one size smaller than line size. Full bypass,

block, sample lines are in place. $1202499 The diagram is correct; there is no pressure gauge. However, the block valves

on both inlet and outlet are one size smaller than the pipe size. Checks withdesign plans show these should be the same size as the pipe size. $120

519


2500 No improvement. $21 5002501 All columns plus ancillaries should be able to handle up to 160 Mg/d ethylene

capacity. $8002502 Same as expected. $302503 Lower than expected. Lower than before the shutdown. $3002504 Steady and almost fully open. $2002505 Alarm is not on. $502506 Ampere draw lower than expected for design flow. Estimated power corre-

sponds to flowrate about 8.5 L/s –30%. $1202507 Accurate; check out OK. $20 0002508 New plant. $502509 Everything started fine. However, we really seem to have problems with acid

addition when low flows are required. The pump sounds as though it is cavi-tating under these conditions. We tried putting the controller on manual butthis didn’t seem to help. $150

2510 Not needed. $20002511 C1400 about 0.6 kPa; now 5 kPa; E1400 vapor side, about 0.55 kPa; now

3.4 kPa; E1401 vapor side about 0.45 kPa; now 2.7 kPa. $1202512 When: just completed. What: cleaned all the condensers; replaced valves A by

globe valves instead of the previous butterfly valves. The butterfly valves didnot seal. Routine maintenance on the compressors. Checked the calibrationsof all temperature sensors. $5000

2513 Pump has no pluggages; adjusted the clearances, although negligible adjust-ments needed. $5500

2514 Since cooling water or air cannot be used as the coolant, various “refrigerants”are used: ethylene, propane, propylene. Each refrigerant is part of a refrigera-tion unit consisting of the four parts: 1) an “evaporator-condenser”, 2) a refrig-erant compressor, 3) a refrigerant condenser and 4) a let-down valve. In theevaporator-condenser, the liquid refrigerant inside the tubes evaporates andcausing the process vapor on the shell side to condense. To allow the refriger-ant to be reused in a closed cycle, the refrigerant vapor is then compressed,condensed and the resulting liquid reduced in pressure to return around therefrigeration cycle. The pressure on the refrigerant side of the evaporator isadjusted to provide the correct temperature driving force for condensation ofthe process vapors. $400

2515 Heat removed consistent with theory.2516 Fully opened. $3002517 Usual value.2518 Shell and tube. Should work well. Baffles were in vertical. Usual allowance for

fouling on tube side and shell side. Correction factor for not-true countercur-rent flow is > 0.85. Vent/blowdown lines on both the tube and shell side. $700

2519 OK. Recalibration not necessary. $20002520 IS: 7 hours after startup; IS NOT: no information known because this is

startup. $502521 7 s. $150

520


2522 Yes. $3000. If you didn’t put the plant on SIS or SIS + evacuation before youasked this question, the plant explodes with loss of life. Penalty $3 000 000.

2523 IS: after short time of use. $602524 Standard design. Should do the job easily. $1002525 1800 rpm, 15 cm diameter impeller, 0.75 kW; 10 L/s at head of 10 m. NPSH

(required) = 1.2 m. Head-capacity curve available. At zero flow, h= 45 m. $452526 1000 �C.2527 IS: Amount of heating of streams to other units is less; IS NOT: the usual

amount of heating. $1502528 Slight hydrocarbon contamination from somewhere on site during the last

couple of months. $1202529 IPA: NFPA: 1, 3, 0. Dow value: 16. $6502530 No change. Slightly different cycle but still cycles. $150 0002531 Should give the correct % vaporization per pass. $7002532 No. $1202533 Inverted bucket. $502534 Overflow effluent= 350 ppm; belt filter OK. $50002535 IS: temperature gauge is higher than expected; stirrer should be operating; IS

NOT: controlled at 127 �C; stirrer is not operating (based on the lights). $50. Ifyou didn’t put plant on SIS or SIS + evacuation before you asked this question,the plant explodes with loss of life. Penalty $3 000 000

2537 IS: Everything upstream and downstream seems to be the usual behavior. ISNOT: different from usual. $50

2538 No leaks.2539 Negligible non-condensibles. Concentration same as always (within the usual

sampling and analytical error). $90002540 Respond to change. $5002542 F/2 reads ... and it reads 50% higher just before the system was put on “safe-

park”. If didn’t put on safe-park as first step, then dangerous potential fire/explosion conditions are created while you experiment. $500 000

2543 No details available. $502544 See Chapter 3: refrigeration, Section 3.3.4; sensor and control, Section 3.1.3;

turbine, Section 3.3.1, exchangers, Section 3.3.3. $1202545 540 ppm. $40002546 From a lake that is fed by several streams. $1002547 Recommends no bypass; (bypass is included in this design); use of a drain

valve inside the block valves (no drain valve included). Inverted bucket is therecommended type although float trap is another option. The other recom-mendations are consistent with the design chosen. $1000

2548 Yes, carbon formation but not more than expected for three days of operationsince startup.

2549 Reads 1.5 �C. $1002550 No change. $60002551 Hydrogen has relatively high heat capacity; about 10 kJ/kg K compared with

about 2 for naphtha or organic vapors; hydrogen thermal conductivity is about

521


0.2 W/m K compared with about 0.02 for naphtha, benzene and other organicvapors. The Pr numbers are similar. Hence, the thermal properties of a mix-ture of hydrogen and other organics depends on the gas composition. $200

2552 Clean, clear valve stem; trim in good condition. $50002553 Film boiling since DTapprox. (220–120) = 100 �C.2554 Relatively clean. $12 0002555 For pure methyl chloride: 630 kPa abs; 530 kPa g. Since the mixtures is a gas,

the concentration of butylene decreases the vp to < 450 kPa g.2556 Fewer sink marks, less air trapped in part. Pack in the mold is improved. Prod-

uct breaks. $8002557 Valve appears to function OK. $2002558 Nothing much. It should be straightforward. Exit pressure from the third

stage was 7 MPa during startup. Dp across the reactors was higher than Iusually see during startup. $3000

2559 Natural circulation; well designed. Should do the job. Heat flux 50 kW/m2;overall heat transfer 1 kW/m2 K. Hydrocarbon is relatively clean; moderatepressure at 230 kPa g. Fouling assumed 0.0003 m2 K/W for process fluid;0.0001, for steam. Float trap for the condensate. $100

2562 The design has the exit pipe inserted into the hopper. However, the sides aredesigned such that neither ratholing nor bridging should occur. $2000

2564 None on this unit.2567 IS: effluent overflow and underflow from mill clarifier. IS NOT: elsewhere.

$502568 IS: at suction to booster ejector. $302569 No change. Level in flocculation tank still below normal. $50002570 Usual concentration of butane. $502571 New plant. $502572 Very sluggish response. $8002574 Before “safe-park” the signal output was “high” ; after “safe-park” the output is

“mid-way”. If didn’t put on safe-park as first step, then dangerous potentialfire/explosion conditions are created while you experiment. $500 000

2575 Lumps persist. $30002576 Can’t see any leaks $24002577 Steady at 31 –1 �C. $40002578 Not needed. $40 0002579 Why1? to provide smooth control of pH. Why2? to maintain the pH within

target values for the feed to the effluent treatment plant. Why3? so that theeffluent plant can handle the waste water it receives. Conclude: focus onWhy2? “to maintain the pH within target values for the feed to the effluenttreatment plant”. $50

2580 Steam at 240 kPa condenses/ boils about 115 kPa so that DT for condensationis about 5 �C. Seems consistent for the design conditions. Now there is a largeDT. $40

2581 47 –1%. $20002582 Responds to change. $200

522


2583 Diagram is reasonably correct and complete. There is a block valve isolatingthe pump from the KO drum; vents on the top of the cooling-water conden-sers, the propylene drum and exchangers E100 and E101. There are blockvalves, drain and bypass-with-valve around both control valves on LC 3 and 4and on the steam to turbine. The turbine-compressor has the usual controlsand instrumentation. The steam comes off the top of the steam header andthe low-pressure steam from the exhaust goes into a low-pressure steam linefor use in heating other parts of the plant. $2000

2584 No.2585 No change. $6002586 Negligible fouling; no condensed liquid; baffle spacing correct and baffles

secure. $60002587 Trace; within spec. $10002588 IS: ever-increasing oscillation in OPRC output. IS NOT: steady. $502590 Steady at 0.6 kg/s.2591 Standard ones we have always used. $2502592 No dirt or contamination. Valve is clean. $30002593 Yes, the impeller shaft is rotating in the direction consistent with the “arrow”

on the casing. $3002594 OK; head-capacity curve OK relative to estimated pressure requirement. $2402596 Why1? to get flow to DAF to expected value Why2? to keep plant operating

safely and prevent upsets in operation. Why3? so that plant can handle thewaste water it receives. Conclude: focus on “to get flow to DAF to expectedvalue”. $50

2597 Checks OK. Recalibration not needed. $6802598 Feed drum is under pressure (because of pressure relief); probable Dp= 40 kPa

across drier followed by 5 to 10 kPa across condenser= 400 kPa g on the feeddrum.

2599 Baffle should be fully closed.2600 No improvement. $45 0002601 Zero. $50 0002602 Pump has no pluggages; adjusted the clearances, although negligible adjust-

ments needed. $55002603 Reads 0.8 MPa. $1002604 FRC reads wild fluctuations.2605 342 kPa g with some fluctuations; past records show similar values.2606 Steady and mid-range. $3002607 30 �C. $3002608 IS: sludge not flowing to filter. IS NOT: feed should be going to filter. $502609 Yes. $1202610 Yes, both read 112 kPa abs –2 kPa. $5502611 Improves. Level in the reflux drum gradually decreases to normal. Reflux flow-

rate still “flat out”. $20002612 Not needed. $502614 Midway. $200

523


2616 Tapping suggests height is the same as level gauge reading.2617 Yes. Valve appears to be closed. $2002619 Not needed. $40002621 IS: just after startup. IS NOT: no previous information. $502622 Overflow effluent= 95 ppm; belt filter operation OK for a while. Then, every-

thing changes and does not work no matter what we try. $30002623 When the production rate in the “hot loop”, say the South loop, is 20% higher

than that in the other loop, say the North loop, then there is no more reactionin the North loop. $300

2624 Nothing done recently. $102626 Moderate, windy. 20 �C; This weather has occurred all week. If didn’t put on

safe-park as first step, then dangerous potential fire/explosion conditions arecreated while you experiment. $500 000

2627 IS: in August; IS NOT: in January when simulation done2628 Should do the job. $3002629 11 s. $2002630 As viewed from the view port, the liquid level appears to be 20 cm below the

Fuller’s earth entry pipe. $20002631 Seems to be near the bottom but very difficult to tell. $3002632 Designed on condensate flowrate and for pressure differential across trap.

Upstream strainer. No bypass. Discharge to header. $6002633 Weather today and all week has been moderate and sunny with highs of 23 to

25 �C2634 OK. Recalibration not necessary.2635 Recalibration not needed. $12002636 IS: operator of this plant. IS NOT: Others.2638 Usual value. $1202640 Butene about 2.2, methyl chloride about 1.5; water about 4.3 and ammonia

about 4.5 kJ/kg K2641 IS: the seal pot did work OK last summer after the compressor had been

installed. $1502642 IS: after they modified the process for more wash water in the washing cycle

of the centrifuge.2644 Slightly higher than usual value.2645 Usual. $3002647 Bypass valve closed. Excessive scaling on the inside of the tubes.2649 IS: no specific location in the part is identified. $502650 There are no standard operating procedures. However, I have been operating

this unit for many years so I think I understand its peculiarities although itnever has worked correctly. However, today it’s much, much worse than I haveencountered before. Started it up as usual; carefully set values at usual values;ensured the steam tracing was off. But the flowrate pumped out is just not upto expectations and the bottoms concentration is still too high.

524


2651 Allowance for fouling on shell side, process gas= 0.0005 m2 K/W; tube-sidesteam superheating= 0.0005. Simulation shows that this unit should do thejob; it has done so in the past.

2652 No improvement. $40002653 Not needed.2655 Serious burns. The hot water sprayed over the operator when the top up valve

was opened. $50002656 Within specifications. $30002658 Pressure higher in the propylene so propylene would leak into the cooling

water. $502660 Flush lines and clean out strainer as needed. Done three weeks ago. Extensive

preventative maintenance done on the belt filter press. $202662 Four rows of finned tubes. Fan diameter-bundle width; gas Dp= 0.025 kPa;

Net free area for air flow =50% face area of bundle. Effective MTD= 20.5 �C;air velocity 3 m/s. $100

2663 Pyrometer reads 232 –3 �C; temperature sensor reads 231 �C. $3002664 Valve stems turn. However, valves are one size smaller than line size. $1502665 Not needed. $40 0002667 NFPA ratings are: methane, 1, 4, 0; propane, 1, 4, 0; propylene, 1, 4, 1; hydro-

gen, 0, 4, 0; ethylene, 1, 4, 2; Explosive limit for hydrogen 4.1–74.3% in air; thelower explosive limits are: for methane 5.0% v/v; propane 2.1% v/v; propylene,2.0% v/v and ethylene 2.7% v/v. $50

2668 Moderate: 19 �C; thundershowers predicted. It’s spring! This week has beenchilly temperatures and overcast. $50

2670 Correct amount of TEP added. Carbon buildup as expected.2671 Not needed. $20002672 Test OK. $402673 Run a charge of acrylic through the extruder. Yes, I do this whenever we start

up each day. $2502674 Occasionally flashes. $502676 IS: effluent 200 ppm; flooded belt filter. IS NOT: 40 ppm; not flooded belt fil-

ter. $502677 No. $1502678 No blockages. Clean. No improvement in process operation. $30002680 We are gathering such data now. No previous reports are available.2681 IS: high-pressure alarm on debut; IS NOT: any other sensor signal. $502682 OK; should work. $502683 No improvement. $10202685 Same as expected. Consistent with the overhead composition at this pressure.

$302686 Condensate buildup/ trap malfunction/ sensor error/ steam pressure too low/

inerts in the steam/ inert buildup in the exchanger/ fouling of the tubes onthe outside/ fouling of tubes on the inside/ steam superheat too high/ steamflowrate < design/ water flowrate > design. $50

2687 No change; still cycles. $2000

525


2688 Checked out TIC-3 and it should have the baffle completely shut. I really don’twant to reduce the flowrate through the reformer because we have highdemand for ammonia these days.

2690 35 m/s, about 3 kg/kg loading and pressure drop is less than that provided.The vacuum in the bleacher should be more than adequate. $2000

2691 Negligible organic. $5002692 Not needed. $40002693 Cool cycle increases from 45 to 47 s; product breaks. $8002694 Relatively clean. $12 0002695 None. We are gathering data as the plant is operating. We have already noted

that the pressure-control system is “not very good” and appears to be sluggish.$200

2696 Inverted bucket; intermittent discharge. Sized for double the design flowrateof condensate. Orifice size selected for a Dp of 1.2 MPa with an allowance of100 kPa pressure drop across the control valve.

2697 Mid-range consistent with full design flow.2698 Mass balances closes within 10%. $1002699 Design F cp DT=F (steam) � latent heat or 28 L/s � 1 kg/L � 218 = 6175 kJ/s

= F(steam) � 2263. Hence F(steam) estimated as 2.7 kg/s; for actual F(steam) = 0.75 kg/s. $500

2700 IS: startup of new plant. $502701 Steady and mid-range. $3002702 Sharp edge facing upstream. OK.2703 No change. $50 0002704 Aluminum with carbon steel for the center of the hub with two feed gates

near the hub. Two cooling lines each side. $2002705 Steady and mid-vessel. $50.2706 Should do the job. Two extra plates in column to allow for growth. Top pres-

sure= about 230 kPa g. $2002708 q= (F cp)E100 DT; Actual= (F cp)E100 (100–10) or (F cp)E100 90; design= (F cp)E100

(101–5) or (F cp)E100 96 or Actual is 94% of design. $3002711 Negligible improvement in heat transfer; increase in gas pressure drop and

resulting upstream adjustments to flows. $20 0002712 Traditional fixed-bed catalytic bed with water jacket and numerous thermocou-

ples so that we can monitor the bed temperatures from the control room. Sili-ca-based catalyst.

2714 Not needed although hazardous conditions might arise. Caution is needed.$50

2715 Not needed.2716 Ammonia: NFPA health 3; flammability 1; reactivity, 0. Toxic, corrosive gas.

Overexposure can be fatal. Low dosage: irritation to nose and throat.> 5000 ppm may result in rapid death due to suffocation or fluid in the lungs.Flammable in air for concentrations 15% to 28% v/v. Gas can ignite explo-sively if released near an active fire. The explosive range broadens 1) if hydro-gen is mixed with the ammonia and 2) at higher temperatures and pressures.

526


Presence of oil and combustibles increases fire hazard. Ignition energy> 0.68 J. Autoignition temperature 651 �C which is lowered from 842 to 651 �Cby the presence of iron. At atmospheric pressure, ammonia decomposes tohydrogen at temperatures > 450–500 �C. Gas has explosive sensitivity to staticcharge. Ammonia is highly reactive with most metals, especially mercury,gold or silver compounds. Reacts violently with tellurium tetrabromide andtetrachloride, chlorine, bromine, fluorine and with acid halides, ethylene oxideand hypochlorites. Nitric acid: 2, 0, 1. $50

2718 Sized with valve position mid-range for design flowrate. $4002720 Calibration not necessary. No change in operation. $30002722 Same as expected. $302723 Should be OK. allowed for in the Dp across control valve FV-6. Suction side

OK, complete with vortex breaker. $3002724 Lumps persist. $12002726 For all exchangers, reboilers and condensers (E 131, 107, 108, 113, 114) Pro-

cess pressure is > utility pressure. For example, E131: fuel-gas pressure,1.1 MPa > 0.65 MPa water. $400

2728 IS: pumps and flowmeter.2729 IS: product breaks. $602730 None found. $30 0002731 We are not receiving the usual amount of product and what we are getting is

off-spec! What’s going on? $3002733 No change. $50002734 Responds to change. $202736 Calibrate OK. $1200. If you didn’t put the plant on SIS or SIS + evacuation

before you asked this question, the plant explodes with loss of life. Penalty$3 000 000.

2737 No change in swinging-loop phenomena. $200 0002738 38 to 42 s; based on 1.5 s/0.1 mm wall thickness. $1502739 Functions OK. Is left fully CLOSED.2740 No improvement. If didn’t put on safe-park as first step, then dangerous

potential fire/explosion conditions are created while you experiment. $500 0002742 Clear, no obstructions. $80002744 IS: This day. IS NOT: noticed before2746 Booster ejector plus two ejectors with direct contact interstage condensers

plus well-designed hot well should provide reliable vacuum. The steam sup-plied for the ejectors is usually reliable; the steam pressure has neverbeen < 95 and >120% of specifications; the steam has never been superheatedby more than 10 �C. $100

2749 45 �C. This is within the usual range. $1202750 No Fuller’s earth is conveyed. $150 0002751 Interesting new set of data. No improvement. $50 0002752 Clean, clear valve stem; trim in good condition. $50002753 Warm, sunny all week, including today.

527


2755 IS: 35 min after polyol addition completed. IS NOT: earlier or later. $50. If youdidn’t put plant on SIS or SIS + evacuation before you asked this question, theplant explodes with loss of life. Penalty $3 000 000

2756 The plant worked well; met all specifications. Everything worked at the designrate. At full design rate, the conditions for the North loop were: lowest pres-sure in loop: 2.75 MPa g; temperature at the exit of the cooling water exchan-ger= 30 �C; at the exit of the refrigeration exchanger= 10 �C; loop-gas analysisat the inlet to the reactor: H2 = 62%; N2 = 21%; methane= 13%; ammonia= 4%;discharge temperature of the fourth and recycle stages of the compres-sor= 40 �C; valve A= 20% open; valve B = 10% open; inlet temperature to cata-lyst bed in reactor= 505 �C; exit temperature from catalyst bed in reac-tor= 550 �C; Dp across catalyst bed= 75 kPa g. $300 000

2757 We’re not getting much and what we get is off-spec. $502758 IS: when on automatic control. IS NOT: on manual. $502760 Steady and lower than usual flowrate. $50.2764 Pressure gradually rose to 45 m; (450 kPa). This corresponds to head at zero

flow as given on the vendor’s head-capacity curve. $502765 Allowance for fouling on tube side= 0.0001 m2 K/W; shell-side steam

= 0.00001. $1502767 No change in conveying. $50002769 Diagram is a good representation of the plant. No surprises. Steam lines come

off the top of all steam headers. Valves accessible. There is no measure of thesteam flow to the reboiler on the debutanizer, E-30. $200

2771 Temperature of melt decreases to 223 �C; injection time increases; cool timedecreases to have a net effect of the same cycle time. Lumps persist in product.$2000

2774 Not needed. $20 0002775 55 –1%. $20002777 Upstream 178 and downstream 128 �C. $40002778 10 �C. $3002779 Why1? to produce crystals within specifications. Why2? to provide consistent

and reliable product (for another unit) or for sale. Why3? so the company con-tinues to prosper. Conclude: focus on Why1: “to produce crystals within speci-fications”.

2780 Temperature-sensor error/ other units sending process fluids that are colderthan usual/ exit temperatures from E201 is lower than usual/ flowrate of gasfrom the reactor is slower/ catalyst degradation in the reformer with carryoverand fouling of the shell side of the exchanger/ baffles loose and gas flow acrosstubes is < design/ pressure increase downstream causing less gas flow fromthe reactor/ inert gas blanketing the tubes on the shell side/ exchangers notvented before startup/ vapors condensing in the exchangers and causing a lossof area/ changes on the tube side. $300

2781 Mass of overhead pumped out by F1400 is 88% less than expected overhead.$50

2783 Mid-range. $200

528


2784 On spec. No complaints from other customers for that batch. $3502785 Water flowrate is slower than expected; liquid level seems to increase on tray

4.2786 No upsets. We have had to supply more water since you started the new wash-

ing. $302787 Diagram does not show the water-cooled liebig condenser directly after the

furnace with the vapor–liquid separator to remove the water via a barometricleg (as shown in the diagram); then a brined cooled Liebig condenser plusseparator and barometric leg. After the dry reciprocating vacuum pumps theketene vapor rises through a packed column where the ketene reacts withacetic acid to form acetic anhydride. The cracking furnace is fired by hydro-gen.

2788 Nothing surprising. Indeed, the chiller operation is relatively self-contained soour interaction is minimal.

2789 No fouling. $50002790 Observe liquid holdup on the C2 splitter on trays 3 and above. No evidence of

collapsed trays; just the top of the column seems flooded. ($5000 for scan +$1600 for time for scan + $20 000 for lost time to arrange for scan) $26 600

2791 Not that we can see on the outside. $200. If you didn’t put the plant on SIS orSIS + evacuation before you asked this question, the plant explodes with lossof life. Penalty $3 000 000.

2793 Awkward but overhead product within spec most of the time. $5002794 Well designed; should operate mid-position for design flowrate. $1002795 Usual. $1002797 OK. Recalibration not necessary. $20002798 See Chapter 3: valves, Section 3.1.3; pump, Section 3.2.3. $302799 No. The oscillations are erratic. $2002800 Indicates when pump is operating as expected. However, no improvement in

operation at this time. $5002801 IS: for last 20 days. IS NOT: before then. $502802 Difficult because the water composition varies between 20 to 50%; no sam-

pling has been done in the hot well and the system tends to fluctuate. Flowrateof live steam is not measured.

2803 We replaced this valve during the shutdown. It sparkles. $30002804 30 �C. $3002805 Vessel is insulated. Reactant is clear. Block valves, a drain and bypass valves

are present for the steam control valve but are not shown on diagram. Nodrain valve is included with the steam trap. The feed flowrate of feed to thetank is measured upstream. $300

2807 Reads 3.9 MPa. $1002808 Responds to change. $202809 See Chapter 3: distillation, Section 3.4.2, perhaps pertinent condensers and

reboilers, Section 3.3.3; pumps, Section 3.2.3 and controllers, 3.1.1. $502810 Negligible fouling on the outside of the tubes.

529


2811 Area 20% oversized. Coil supported along only one side of the vessel so that itwon’t tear away when heated up. Thermal expansion anticipated andaccounted for. $600

2812 All valves appear to function OK. $3002813 End-suction centrifugal; 1800 rpm; sized to operate at design flowrates with

appropriate allowances for Dp across control valve. $1002814 Process pressure > atmospheric. $502815 Same as usual. $602816 Lower than expected. Lower than before the shutdown. $3002818 Clean. $20002820 IS: after 2000 parts had been shipped. $502821 Checks that the steam flow available should be able to reach specification bot-

toms composition. $502824 Wide open. $200. If you didn’t put the plant on SIS or SIS + evacuation before

you asked this question, the plant explodes with loss of life. Penalty $3 000 000.2825 Clear, no obstructions. $80002826 Should do the job. Neutralization reaction is rapid. Good mixing supplied by

the rising gas from the spargers. $3002827 See Chapter 3: adsorption-drier, Section 3.4.7; exchanger and thermosyphon

boilers, Section 3.3.3; refrigeration loop, Section 3.3.4; sensors and controls,Section 3.1.3; pump. Section 3.2.3.

2830 Polymer is added to the feed to the filter to aid the filtration process. $202831 Steady and usual. $502832 Over the past few months we kept the PIC at 0.8 MPa. We did have a time

when the load was small and we operated for a short time at 0.7 MPa. Afterthat, however, we had to gradually increase the PIC settings for even the usualloads (where in the past 0.8 MPa would have handled the load well). $100

2834 Accurate; check out OK. $20 0002836 OK. No calibration needed. $30 0002838 IS: First operators.2839 The diagram is reasonably correct in the overall layout. In addition, there are

clean outs and flush on the pump. Isolation block valves on strainer andpump. A pressure gauge is on exit from the pump. A metering pump providesthe polymer addition. Both thickeners have overflow launders, central feedand rotating central rake. There is a pressure gauge on the blowout air lineupstream of the block valve. $50

2840 As measured on the gauges Dp= 243 kPa. $10002841 Temperature= 125 �C. $6002843 Mid-range. $3002844 Equipment should do the job. $1002845 Valve stem is settled in to be about 60% open and consistent with design value

for 14 L/s. $302846 Same before and after. Bottoms flowrate= 1.38 kg/s –10%. $12002848 Steady. $1202849 No change. Lumps persist. $8500

530


2850 Equipment should do the job. Concern about NPSH and excessive pressureloss on the suction side. $800

2851 F/7 = 6.5 L/s and did read 7.8 L/s before the system was put on “safe-park”. Ifdidn’t put on safe-park as first step, then dangerous potential fire/explosionconditions are created while you experiment. $500 000

2852 No change. Level in flocculation tank still below normal. $60002853 Horizontal pressure vessel; level control ensures that water is always main-

tained in the steam drum. Natural circulation lines feed water to the shell sideof the waste-heat boiler. Traditional design.

2854 Does not sound like cavitation.2855 The trays we can see are level; check on the clearance between downcomer

bottom and tray suggests that tray should be sealed. $30 0002857 Dp measured is ten times estimate. Suspect blockage in line from product

sump to pump suction.2859 Clear and clean. $30002861 IS: exit water temperature is 42 �C; IS NOT: 70 �C. $502863 Estimate seems OK. No major blockage. $3002865 No change in swinging-loop phenomena. $100 0002866 Negligible fouling; no condensed liquid; baffle spacing correct and baffles

secure. $60002867 Designed to handle design flowrate; horizontal configuration; good baffle

design with vertical windows; sealing strips used. Slight tilt to facilitate con-densate drainage. Usual vents and drains. Fouling assumed 0.0003 m2 K/Wfor process fluid; 0.0002, for clean cooling water. $150

2868 Stroke same as specs.2869 Several tubes have slight leaks. $20 0002870 The solution leaving the reactor is 83%. This is concentrated to 99% in a

downstream falling-film evaporator. This concentrated solution is then prilledto produce pellets. $50

2871 All five samples are the usual concentration. $4000.2872 Call you, as instructed, since this is costing us megabucks! Then promptly try

to close valve A in the hot loop and open valve A in the cold loop; I may haveto put on the calrod startup heater if the cold loop gets too cold. $300

2873 Lumps persist. $40 0002874 Not needed. $40002875 No water or ammonia in the sample.2876 Yes. All valves appear to be closed on shell and tube side. $3002877 Whistling sound of fluid flowing. $1002878 Usual values; suggests that the damper is 1/3 closed. If didn’t put on safe-park

as first step, then dangerous potential fire/explosion conditions are createdwhile you experiment. $500 000

2879 No improvement. $5002880 10 ppm. $4000

531


2881 Wide open before “safe-park”. At “safe-park” mid-range. If didn’t put on safe-park as first step, then dangerous potential fire/explosion conditions are creat-ed while you experiment. $500 000

2882 No upsets. No calls from the safety inspector.2883 101.6 kPa $1502885 (E114) Usual allowance for fouling on tube and shell side. Vent/blowdown

lines on both the tube and shell side. Vent/blowdown lines discharge to sewer.Each vent line has a single gate valve that is normally shut. Process pressur-e > ethylene refrigeration pressure. $1500

2886 Higher than usual concentration. Particles are < 10 mm but cannot tell if thereare many < 1 mm because of the optical limits of our analyzer.

2888 The piping is complex and consists of a range of 3-way and 4-way valves toallow any of the dryers to be on-line or regenerated. On-line is straightforward.The feed gas enters the top of the first dryer, exits the bottom, enters the top ofthe second dryer in series and then out the bottom to a knockout pot. Regen-eration is more complex because two different sequential activities occur dur-ing regeneration. First, hot “town gas” goes through the bed and sends off theadsorbed water. When most of the water has been desorbed from the bed ofalumina, the bed must now be cooled before it can be put back on-line. Towngas is used for both these functions. First, the desorption is done with “hot”town gas; then the cooling is done with “cooled” town gas. The town gas fromboth functions is returned to the fuel gas system. Case’12 gives the details ofa similar – but slightly different – system. Also this downtime costs more thanCase’12. $400

2889 Estimate: 7.6 m/ (1.3 m/s) = 5.75 s. $502890 No improvement. $9802891 We already did this during the turnaround. Looks fine; minor or no adjust-

ments needed. When we bring this back on line there is no change. $80002892 45 s; based on 1.5 s/0.1 mm wall thickness; slightly longer than usual because

of the thickness of the product, especially the hub. $3002893 Set to blow at 340 kPa. $200. If you didn’t put the plant on SIS or SIS + eva-

cuation before you asked this question, the plant explodes with loss of life.Penalty $3 000 000.

2894 Damaged spring; check valve partially blocking the flow. $5002895 Agrees with design values –15%.2896 IS: thermosyphon chiller output. IS NOT: elsewhere.2897 Zero. $30002898 No change. $20 0002899 Not needed. $30 0002900 Not needed. $24 0002901 No significant fouling; baffles in good condition, vortex breaker in place.

$10 0002902 Gradually increasing to 57 �C from the usual 47 �C. $502903 No. I sure wish I could. $3002904 No improvement in operation. $8000

532


2905 Concerned with the heavy maintenance caused because of the ketene dimeri-zation. Would recommend the use of a liquid ring pump using acetic acid asthe sealant.

2906 Wrong height of liquid in the seal pot/ compressor under surge or sonic con-ditions/ pressure sensor wrong/ control system fault/ excessive pressure inthe flare line from upstream units/ kickback valve faulty/ error in matchingseal-pot level with expected Dp/ plugged line from seal pot to flare/ notenough steam to the flare. $150

2907 IS: lab analysis of the overhead shows too much C4; in the bottoms is toomuch C3. IS NOT: anything else different. $50

2908 DT in reboiler= 198–(180 to 185) �C = 13 to 18 �C and hence probably nucleateboiling.

2909 Has tendency to oscillate. $502910 Head-capacity curve and NPSH data available. $702911 Yes, for this production rate provided water flow is usual and inlet water tem-

perature < 20 �C and exit water temperature is < 40 �C. $502912 IS: independent of the operator.2913 Temperature –5 �C; 600 kPa abs. Latent heat= 438 kJ/kg. $752914 Sharp edged orifice facing correct direction, based on tab markings.2915 Should work well. Baffles were in vertical. Vessel has air bleeds. $2002916 No significant fouling; baffles in good condition, vortex breaker in place.

$10 0002917 Valve starts mid-range and gradually increases to almost wide open as the tem-

perature drops and until the temperature reaches close to the high; then itreturns to mid-range and the cycle repeats.

2918 IS: overhead pressure gauge P1 is too high, 2.7 kPa. IS NOT: 0.4 to 0.7 kPa.$30

2919 Reads 1.3 �C. $5002920 Hopper is 3/4 full. $10002921 No improvement. $8002922 Reaction rate is very fast. $502923 No change. $20 0002924 End suction, centrifugal pump; 1800 rpm; NPSH required at design flow-

rate= 1 m; mechanical seal.2925 Yes, the leads are correct. $4002926 Leaks in the overhead line, loss through the bottoms, incomplete condensa-

tion: condenser undersized, air flowrate too small, air too hot, air recirculation,inadequate liquid seal in condenser, dirty heat-exchanger surface, incorrectcalculation. $800

2927 1725 rpm. $502928 Just started up. None. $502929 Shell and tube. Allowance for fouling on tube side= 0.0001 m2 K/W; shell-side

steam=0.00001. $9302930 IS: pump only 2/3 design; level in reflux drum increases. IS NOT: pump

design capacity and level in reflux drum constant. $50

533


2931 No blockages; impeller key in place; correct clearances. No improvement inprocess operation. $2000

2932 Not needed. $20002933 Steady and at the design value.2934 Not needed. $40002935 All correct. $202936 62 trays= 136 kPa. $2002937 Dilution water could come in with any of the feed streams, from a leak in the

cooling coil, condensate running back down from the overhead steam line,steam leak for the heater into the nitric acid in the storage tank and whatabout the heavy rains we have been having all week? $300

2938 Fail open. Direction agrees with flow direction. $2002939 No change in operation. Pressures and temperatures still increasing. $2502940 IS: control is erratic; acid flow may stop. IS NOT: steady control; acid flow as

needed. $502941 For the naphtha pumps, the NPSH supplied in this installation >> than re-

quired. No air or water tests were done before starting up the unit. $5002942 Controlled and steady. $2002943 Diameter and length sized to give required residence time for mixing and

reaction. OPRC sensor placed well downstream of “end of reaction”. $1002945 No change. $200 0002946 71 kg/s (compared with design of 100 kg/s).2947 Level covers half of the tubes.2948 Yes, reduce flowrates to 70 Mg/d unless you need it to run at high flowrates to

get key operating data. $20 0002949 Nothing that looks like it would vibrate or chatter. $40002950 OK. Recalibration not necessary.2951 Estimate seems to agree with information from the unit. Suggests flowrates as

expected for new conditions and same as previously. $6002952 No changes over the past year. Still producing slightly superheated steam at

0.8 MPa g. $1002953 Allowance for fouling on shell side, water-boiling= 0.0002 m2 K/W; tube-side

process gas= 0.0005. An allowance of 10% more area than needed for designflowrate. Single pass on the tube side with unique baffle design for secondarypart of the boiler. Simulation shows that this unit should do the job; it hasdone so in the past.

2956 Design method D-1a in Chin’s classic article in Hydrocarbon Process, Oct 1979,p 151. Controls the flowrate of coolant. $200

2957 End suction, centrifugal. 1800 rpm with impeller one size smaller than casing.NPSH data and head-capacity curve available. To prevent acid leakage throughthe packing seals, demineralized water is fed to the packing seals via lanternrings. The water is kept under pressure so that there is a Dp= 510 kPa acrossthe seal. This means a small amount of water goes across the seal instead of asmall amount of acid leaking out through the seal. $600

2958 10 �C. $300

534


2959 Zero. $30002960 Sounds as though it is working properly. Valve direction= flow direction. $2002961 Looks fine. $20 0002962 Not needed. $80 0002963 Non-ideal but data match theory.2964 IS: especially on third effect but on all effects and after preheaters. IS NOT:

feed inlet or steam inlet. $502965 Plant operates as it should. Level in flocculation tank rises to “normal level”.

$55002966 Not needed. $20 0002967 Use the heat load gained by the steam to back calculate the process-gas tem-

perature leaving the superheater. At design conditions this gives 600 �C goingto 530 �C out. For actual conditions this gives 730 �C going to 503 �C out.These can then be used to estimate LMTD and thus UA. UA design= 21 MJ/s/226 = 93. UA actual= 68 MJ/s/ 232 = 293. For the higher temperatures on theactual conditions, the thermal properties will be higher; the steam-side coeffi-cient will be 25% lower because the steam flowrate is 70% slower based onRe0.8.

2968 IS: negligible overhead product and bottoms 10% organic. IS NOT: designrate of overhead and bottoms concentration < 2% organic.

2969 See Chapter 3: vacuum, Section 3.2.2; distillation, Section 3.4.2; condensers,Section 3.3.3; reciprocating pump, Section 3.2.3. $60

2970 Well designed based on pressure, residence time plus vortex breaker and dem-ister pad. Simulation shows that unit was operating very close to design valuesfor usual range of feedstocks and conditions. $150

2971 Four months ago; routine cleaning of all heat exchangers including the waste-heat boiler and steam superheater.

2972 Takes about 2 days. Nothing to be seen. $40 0002973 1.7 MPa g and steady.2974 Steady and usual value, 75 �C. $50.2975 Not needed. $20 0002976 No improvement in operation. Steam is still superheated. Gas temperature

still very high.2977 Warm, 20 �C; humid. Rain two days ago. Warm and humid the rest of the

time. $5002978 IS: lab analysis of the overhead shows propane concentration 21�2 times high-

er than design. IS NOT: anything else different2979 No noise until the temperature starts to rise; then hear a high-pitched whis-

tling that stops when the temperature hits the top of the cycle; followed bylower-pitched bubbling, then no noise. Cycle repeats. $6000

2980 The flooded evaporator works as follows. Sufficient area is supplied so thatwhen there is no condensate in the tubes, the area available for heat transfercan evaporate all the ethylene (for the butane temperature corresponding tothe higher pressures). Usually the condensate fills the tubes so that the area ishalf the maximum. Thus, to increase the butane evaporation, the pressure

535


PIC is increased and the valve on the condensate opens to drain the conden-sate out of the tubes. There is a knockout pot in the ethylene line to pipelineto remove any entrained liquid. The liquid is recycled to the storage tank. $50

2981 “Process Design and Engineering Practice” D.R. Woods, Prentice Hall (1995)p 4-90 to 4-93

2982 When: 6 months ago; routine checks done during the turnaround. $502983 Column pressure decreases. Flare reduces. Impurities reported in pentane

product. $50002984 Agrees with head capacity. $2002985 None, this is startup of new plant. $502986 4 months ago, cleanout buildup on the cooling coils in the reactor. $50. If you

didn’t put plant on SIS or SIS + evacuation before you asked this question, theplant explodes with loss of life. Penalty $3 000 000

2987 When: 7 months ago. Nothing done on this section of the plant. $102988 Soap tests inconclusive. $50002989 Slightly more open than usual. $2002990 Well designed. Demister included. Globe valve on periodic drain to sewer.

$1502991 18 �C; cloudy and overcast; rain forecast. Cool and cloudy all week. $30002992 See Chapter 3: distillation, Section 3.4.2; adsorption, Section 3.4.7; knockout

pots, Section 3.5.1; heat exchangers, Section 3.3.3. $2002993 None available. In a sense this startup is the commissioning. Maybe next

time! $6502994 Usual. $502995 IS: increase flowrate above level. IS NOT: when lower flowrate used.2996 Hot!!! $300. If you didn’t put the plant on SIS or SIS + evacuation before you

asked this question, the plant explodes with loss of life. Penalty $3 000 000.2997 Designed on condensate flowrate and for pressure differential across trap.

Upstream strainer. No bypass. Discharge to header. $1502998 See Chapter 3: reactor, Section 3.6.2; furnace, Section 3.3.2; exchangers, Sec-

tion 3.3.3; pump, Section 3.2.3; control and valves, Section 3.1.3. $2002999 (E113) Usual allowance for fouling on tube and shell side. Vent/blowdown

lines on both the tube and shell side. Vent/blowdown lines discharge to sewer.Each vent line has a single gate valve that is normally shut. Process pressure> propylene refrigeration pressure. $1500

3000 Oversized. This should easily do the job. $120

536

537

Cases’1 and ’2 are discussed in Chapter 1. Cases’3 to ’7 are posed in Chapter1 with some answers in Chapter 4. Chapter 4 illustrates the process used by troubleshooters with varying degrees of skill. In this Appendix, we provide feedback aboutCases’8 to ’52.

For each case is given: an estimate of the degree of difficulty, the cause, possiblecorrective action, and two sections related to the TS process: TS process hypothesesand TS process: possible diagnostic actions.

The Estimate of difficulty is very subjective. The criteria I used in assigning theratings were: 1) the ease with which Level 4 undergraduate engineering studentshad in “solving the problem’, 2) the ease industrial participants had with the casesand 3) the level of “experience with process equipment” required in solving the prob-lem. For the latter, I used the self-test data from Chapter 1. If the experience withprocess equipment rating is below five, then the cases are rated as relatively easy. Rat-ings 7 to 9 were used where the experience with process equipment ratings wereabout seven to ten. In assigning the ratings I assume that the MSDS data, “Moredetails about the process”, handbook data and guidelines for trouble shooting (fromChapter 3) are not needed by the trouble shooter. The trouble shooter knows thisalready. Nevertheless, this information is available for many of the cases, and shouldbe used when working cases where you are uncertain about such information.

Case’8: The depropanizer: the temperatures go crazy (courtesy of T.E. Marlin,McMaster University)Estimate of difficulty: 8/10. This gets a high rating for “experience with processequipment on process control and distillation”.

Cause: No level control on the feed vessel, V29. The fact that the fault occurredbecause of the change in pressure was coincidental.

Possible corrective action: Immediate: Re-establish the feed flow rate by: placingthe TC, FC and LC controllers for the tower on manual. If enough material exists,place the column on safe-hold (total reflux) until the feed is started; place the con-

Appendix EDebrief for the Trouble-Shooting Cases


Appendix E Debrief for the Trouble-Shooting Cases

troller FC-1 on manual with the valve partly opened; prime the pump; adjust FC-1until the flows in and out of the feed drum, V29, are equal. After the flow has beenre-established, the operator must closely monitor and control the level manually.

In the longer term: The level L-1 should be controlled by adjusting the FC-1 set-point. A cascade control system should be installed. Also, the feed flow rate has nolow alarm. Although a SIS exists to start the backup pump, a FAL should be placedon the FC-1 measurement. In addition, the feed drum has no alarms. LAL and LAHshould be placed on the LC-a measurement.

TS Process: illustrative hypotheses: tray collapsed in the stripping section/ toomuch bottoms fed to the debutanizer/ too much overheads in the feed/ feed valveFV-1 stuck/ pump F-26 not working/ check valve on the idle pumps allows back-flow/ no feed left in feed vessel V29.

TS Process: possible diagnostic actions: 1666, 63, 563, 155, 78, 1436, 1765, 1502,1947, 1610, 1834, 691, 1906, 2769, 1737, 1390.

Case’9: The bleaching problem

Estimate of difficulty: 7/10. This is a relatively moderate rating because of the experi-ence with process equipment required about pneumatic conveying. This was adesign problem in that this mistake should not have occurred.

Cause: No conveying air is getting to the inlet of the conveying line. Without air,there is nothing to transport the Fuller’s earth.

Possible corrective action: Immediate action: either dump Fuller’s earth into thebleacher directly or insert air spargers into the hopper so that air is supplied to con-vey the solids. Longer term: redesign with fluidizing panels on the bottom cone ofthe hopper or install a concentric annular pipe that allows air to flow down the insertpipe to the pipe inlet.

TS Process: illustrative hypotheses: insufficient vacuum/ pressure sensor wrong/plugged conveying line/ valve not open/ no air to convey the powder/ no powder inthe hopper/ powder is bridging in the hopper.

TS Process: possible diagnostic actions: 2977, 2690, 2562, 59, 1368, 1097, 2431,2920, 2630, 1206, 2767, 2334, 1856.

Case’10: To dry or not to dry (based on Krishnaswamy and Parker, 1984)Estimate of difficulty: 8/10. This requires systems thinking and experience with pro-cess equipment for drying, screens and centrifuges. Many senior students have hadlittle experience with equipment related to liquid solid separations.

Cause: The washing cycles between the upstream screen and the centrifuge werenot synchronized. As a result the centrifuge was getting no feed when the screenwas being washed. During this time the crystal layer in the centrifuge under thescraper compacted. This resulted in a reduced filtration rate and higher crystal mois-

538


ture content. This persisted until the wash cycle in the centrifuge. Figure E-1 (rep-rinted with permission from R Krishnaswamy and N.H. Parker, Chemical Engineer-ing, April 16, 1984, p 93 to 98, copyright � McGraw Hill) shows, at the top, the per-formance before the change when the washing cycles were synchronized. At the bot-tom is shown the performance after the change when the cycles were not in sync.

Figure E-1 The top shows performance conditions – before the change – for threepieces of equipment: for the screen the feedrate and the % of the feed over, as a func-tion of time; for the centrifuge, the feedrate and discharge moisture, %, as a functionof time and for the dryer, the feedrate, the % moisture in the feed and the % moisturein the discharge, as a function of time. The bottom picture shows the same data forthe three pieces of equipment after the change.(Excerpted by special permission fromChemical Engineering (April 16, 1984) Copyright � (2004) by Chemical WeekAssociates, New York, NY, 10038). This information is useful for solving Case’10.

Possible corrective action: bring the washing cycles into sync.TS Process: illustrative hypotheses: not enough steam/ wash water carryover

from the centrifuge/ cycle from screen not coordinated with cycle in centrifuge/feed crystals too wet from the screen/ rotational speed of the dryer has increased/more fines in the centrifuge causing the filter cycle to be too long; fines carryover to

539


the centrifuge causing blinding in the centrifuge/ centrifuge rpm faster than usual/crystal size change; crystals change shape so that filtering is different/ centrifugeoperates but has run out of feed/ dryer feed too cold/ vendor supplied faulty equip-ment.

TS Process: possible diagnostic actions: 492, 4, 405, 816, 711, 1411, 603, 1009,1041, 1804, 1971, 360, 1114, 1458, 909, 54, 1968, 1520, 196, 1705.

Case’11: The lazy twin (courtesy W.K. Taylor, B. Eng. McMaster University, 1966)Estimate of difficulty: 5/10

Cause: The check valve on the exit line from pump B is faulty and allows liquidfrom pump A to recycle around when pump A is on-line and B is off-line

Possible corrective action: Immediate action: close the block valve V205 on theexit line of pump B whenever pump B is idle. Longer term: replace the check valve.

TS Process: illustrative hypotheses: instrument error in flowmeter/ motor fails tostart when remote start button is pushed/ pump A has worn wear rings causinginternal flow circulation/ on pump A the motor is turning backwards so that theimpeller is turning in the wrong direction/ air lock in pump/ debris or stuff plug-ging the suction line of pump A/ reverse flow through check valve on pump B.

TS Process: possible diagnostic actions: 930, 1470, 1998, 518, 550, 170, 102, 1482,2027.

Case’12: The drop boxes (courtesy John Gates, B. Eng. McMaster University)Estimate of difficulty: 3/10. The key is to realize that the leak is from the high-pres-sure side to the low pressure side; identify the location where a hydrocarbon streamis on the high-pressure side.

Cause: Leak in E131 from the “fuel gas” process side to the water on the tubeside. The contaminated water goes to sewer.

Possible corrective action: Isolate E131. Water pressure test and locate leaks. Plugthe tubes that leak.

TS Process: illustrative hypotheses: sampling error/ analysis error/ hazard com-ing from upstream plants/ leaks in vent valves from any location having “fuel gas”-type materials upstream of the valve, e.g., the shell side of most exchangers/ leak inthe valve on the knockout pot/ leak in the tubes from fuel gas to water in E131.

TS Process: possible diagnostic actions: 2085, 2067, 1699, 2617, 1211, 2433, 119,2424.

540


Case’13: The lousy control system (courtesy ESSO Chemicals)Estimate of difficulty: 4/10. This is relatively easy because the operator had identifiedthe problem correctly.

Cause: The temperature of the condenser effluent is controlled by varying thepitch of the fan. This is, indeed, a slow and clumsy method. The response tochanges in ambient temperature meant a loss of effective control of this tempera-ture.

Possible corrective action: Immediate: run cold water onto the outside of thetubes. Longer term: this was resolved by installing a bypass control valve around thecondenser to control the effluent temperature. Another option is to install a water-cooled trim cooler.

TS Process: illustrative hypotheses: instrument fault/ maldistribution/ hot-gasrecirculation/ fouled tubes/ blade wrong pitch/ fan not working/ insufficient tubearea/ buildup of non-condensibles in the bottom row of tubes/ tubes not sealed/ novent break/ poor control system/ control system not well tuned.

TS Process: possible diagnostic actions: 2179, 1613, 220, 89, 2152, 487, 367, 2793.

Case’14: The condenser that was just too big

Estimate of difficulty: 3/10.Cause: A vapor lock occurs in the suction line to pump F1400 because there is no

vent-break line connecting the inlet to the pump to the inlet to the booster ejector.Such a vent line needs to be designed carefully to prevent liquid from going up thevent line instead of into the suction of the pump. The vertical height at the entranceto the booster ejector line must be higher than the difference in height of level ofthe entrance to condenser E1400 and the pressure difference between the exit fromexchanger E1400 and the entrance to the booster ejector.

Possible corrective action: Install vent-break line as illustrated in Figure E-2.TS Process: illustrative hypotheses: vacuum leak/ variation in steam pressure to

booster or other ejectors/ liquid not subcooled enough and it is flashing in pumpF1400/ wet vacuum pump not pumping at capacity/ leak in exchanger E1400 caus-ing water to flow into vacuum system/ instrument error/ more volatile material infeed/ vapor lock in suction line because of no vent break/ leg from barometric con-denser not sealed/ booster not working right and air is sucked down from the boos-ter into the pump F1400/ dry vacuum pump F14 put into service without 30-minutewarmup.

TS Process: possible diagnostic actions: 400, 2273, 2377, 390, 486, 188, 871, 2511,2781, 2580, 1317, 1020, 1594, 2201, 326, 242, 2369, 2939.

541


TI

TI

TI

TI

PI

PI

PI

Coolin

gW

ate

r

reci

pro

cating

pum

p

60ºC

Back

up

Condense

r

Boili

ng

Condense

r

240 k

Pa

to/f

rom

Dow

therm

syst

em

Fatt

y aci

d

Feed

240ºC

to260ºC

Coolin

g w

ate

r exi

t

Coolin

g w

ate

rC

oolin

g w

ate

rre

turn

0.4

kP

a

Boost

er

steam

well

wate

r

baro

metr

icco

ndense

r

wet

vacu

um

pum

p

120ºC

Coil

in t

ank

va

cu

um

bre

ak

lin

e

ne

ed

ed

542

Figu

reE-2

Thevent-break

system

forCase’14.


Cases’15 to 18 are discussed in Appendix C.

Case’19: The Case of the reluctant belt filter (supplied by Mike Dudzic, B. Eng. 82,McMaster University)Estimate of difficulty: 2/10.

Cause: The strainer is plugged.Possible corrective action: Clean the strainer. Long term: “ensure clean the strai-

ner” is included in operating procedures and training.TS Process: illustrative hypotheses: inadequate water flush/ flush does not go into

the correct lines/ pump stopped/ strainer plugged/ line plugged and cannot becleared with the flush/ flush not working/ block valves leaking/ valves around pumpclosed/ pump clogged.

TS Process: possible diagnostic actions: 1635, 1330, 2329, 2052, 1258, 322, 793,2672, 1855, 1576.

Case’20: The case of the fussy flocculator pump (courtesy Jonathan Yip, B. Eng. 97,McMaster University)Estimate of difficulty: 3/10

Cause: Damaged check valve that didn’t open properly combined with faulty loca-tion of the pressure gauge.

Possible corrective action: replace the check valve and relocate the pressure gaugebefore the check valve.

TS Process: illustrative hypotheses: pressure gauge reading wrong/ cavitatingpump/ wear rings worn causing internal circulation/ impeller worn/ blockage inline/ change in density of liquid so that pressure gauge reading converts to a differ-ent value of “head”. This fault happened gradually so that it does not seem to berelated to maintenance or other changes. Need to visit the site and see the pump.Locate the head capacity chart from the vendor.

TS Process: possible diagnostic actions: 2525, 2322, 2343, 2243, 1342, 2764, 1099,2506, 2894, and 2965. If a dominant J trouble shooter (as described in Section6.1.3c) decided to change things and fortuitously selected diagnostic action 2965,then the “problem” would have been solved. However, a few simple tests and gooduse of a head-capacity curve and fundamentals will remove the fortuitousness. Test2506 is very telling because, although the answer is relatively inaccurate, the power(and flow) is less than expected. Another clue is the slow response to action discov-ered by action 2764. By this time, if not earlier, the trouble shooter should realizethat the pressure gauge is downstream of the check valve and thus is not reflectingthe true response at the exit flange of the pump.

543


Case’21: The case of the flashy flare (courtesy Mark Argentino, B. Eng. 1981,McMaster University)Estimate of difficulty: 3/10. The solution does not require extensive knowledgeabout processing equipment. The key is second-year fluid statics.

Cause: Off-specification “kerosene” with density 1.02 was used in the seal potinstead of the usual specification-grade kerosene of density 0.8.

Possible corrective action: Use specification-grade kerosene.TS Process: illustrative hypotheses: wrong height of liquid in the seal pot/ com-

pressor under surge or sonic conditions/ pressure sensor wrong/ control systemfault/ excessive pressure in the flare line from upstream units/ kickback valvefaulty/ error in matching seal-pot level with expected Dp/ plugged line from seal potto flare/ not enough steam to the flare.

TS Process: possible diagnostic actions: 2641, 1505, 1395, 1451, 9, 1497, 1842,1974, 2610, 70, 2487, 1284, 1590. Another skilled trouble shooter used, in addition,471, 500, 1511, 694, 410, 1240.

Case’22: The pH pump (courtesy of Scott Lynn, Chemical Engineering, University ofCalifornia, Berkeley)Estimate of difficulty: 4/10. This is a basic design flaw in installing a pump. Thecontrol valve should never be put on the suction side, especially for a liquid with ahigh vapor pressure such as hydrochloric acid.

Cause: The control valve is on the suction side of the pump. The pump cavitates.Possible corrective action: Move the control valve to the discharge side. Ensure

that there is a vent on the storage tank and that the acid is injected into the top ofthe waste line to prevent waste from flowing back through the pump. Install a checkvalve in the pump-discharge line.

TS Process: illustrative hypotheses: sensor fault/ poorly tuned control system,/sticky control valve/ control-valve hysteresis/ caustic waste flows backwards into thepump/ electrical interference with the control system/ cavitating pump/ no vent onthe storage tank; vacuum created in the storage tank/ density of acid > expected andmotor overload.

TS Process: possible diagnostic actions: 1425, 1191, 2509, 1290, 2002, 1068, 1835,1682, 1150, 2129, 2467, 700.

Case’23: The hot TDI (based on Barton and Rogers, 1997)Estimate of difficulty: 4/10. The hazardous nature of this situation should be recog-nized quickly.

Cause: The supplier of the TDI did not have benzoyl chloride as a reaction modi-fier. The reaction ran away, the pressure built up to 3.4 MPa g and the reactor ejectedthe batch. If, for any reason, the relief system failed, the reactor would have

544


exploded. Without the modifier present, the exothermic reaction ran away, solidproduct formed causing the stirrer to stop. The high temperature caused gas forma-tion and pressure buildup.

Possible corrective action: add reaction modifier or return to the previous supplierof TDI or redesign the coolant system and the operating procedures to prevent therun away.

TS Process: illustrative hypotheses: emergency/ action to prevent explosion. Yes,there are a variety of hypotheses to explore later to discover why the temperaturerunaway: cooling-water failure/ coolant-inlet temperature too high/ cooling-surfacefouled/ addition too fast/ operating procedures not followed/ sensors wrong/ differ-ent supplier of TDI/ contamination of TDI in the storage tanks/ different polyol/contamination of polyol in the storage tanks or lines.

TS Process: possible diagnostic actions: 2034 perhaps 1739. Later, 1195, 2289,2893, 1201, 1422, 1986, 2188, 1743, 2824, 723, 1837, 709.

Case’24: Low production on the ethylene plant (courtesy John Gates, B. Eng. 1968,McMaster University)Estimate of difficulty: 4/10. The unit operations involved are familiar. Althoughthere are two faults, they should be relatively easy to spot.

Cause: Two causes: 1) the town gas is off-specification because it is picking upmoisture from the slugs of water left over from the hydraulic testing of the line and2) high-pressure steam is leaking into the town gas from heater 130. The town gassaturates the dryers during the regeneration stage and not sufficient moisture isremoved during the cooling phase. The result is that the dryers are adding water tothe process gas, instead of removing water.

The result is that gas hydrates are solidifying in the column, in particular near thetop of the demethanizer and the top of the C2 splitter where the pressures are highand temperatures are below 0 �C. The terminology used in the industry is “icingup”.

Possible corrective action: Three things to correct: 1. removal of the hydrates tokeep the column going. 2. stop the steam leak and 3. insist that the town gas sup-plier give town gas with < 6 ppm moisture. The first two are directly under your con-trol. Eventually, at the next turnaround the third problem can be solved by using ourown fuel gas (town gas) generated on site instead of relying on an outside supplier.More specifically: 1). About two m2 of methanol was pumped into the reflux lines ofeach column affected. The methanol came down the column and eventually wasrecycled with the ethane to the cracking furnace. Fortunately, a single treatmentcleared up the problem and the column could operate at 150 Mg/d with acceptableDp.

2) The leaking tubes in heater E130 were plugged. A new bundle was preparedfor the next turnaround and another spare was ordered for the next possible recur-

545


rence. 3) Samples were taken of the trends in the town gas. We met with the vendorto resolve the lack of specs on the town gas provided.

TS Process: illustrative hypotheses: moisture coming in because the dryers arenot working/ steam leak in the reboiler on the bottoms of the de-ethanizer/ towngas coming in too “wet”/ steam leak from E310 into the town gas/ process gas enter-ing the site is too wet/ column was not designed to handle this high a flow/ col-lapsed trays.

TS Process: possible diagnostic actions: 2948, 257, 2130, 2501, 280, 501, 2936,2112, 1214, 1377, 2432, 2840, 2043, 559, 1912, 1767, 892, 527, 2050, 2545, 1874, 348.

Case’25: The case of the delinquent exchangers (based on Yokel, 1983)Estimate of difficulty: 4/10. Although this is a larger system than some of the earliercases, the cause should be relatively apparent from the typical symptoms given inChapter 3.

Cause: Under the new conditions the concentration of hydrogen in the reactoreffluent is greatly reduced. Since the thermal conductivity and heat capacity ofhydrogen are five to ten times higher than most other vapors, there is a significantreduction in the heat transfer because of the reduced concentration of hydrogeneven though the flowrates are unchanged.

Possible corrective action: recalculate the ratings of the exchangers to see if any-thing can be done to alter the current exchangers; if not, look for other sources ofheat or install additional exchangers.

TS Process: illustrative hypotheses: temperature-sensor error/ other units sendingprocess fluids that are colder than usual/ exit temperatures from E201 is lower thanusual/ flowrate of gas from the reactor is slower/ catalyst degradation in the refor-mer with carryover and fouling of the shell side of the exchanger/ baffles loose andgas flow across tubes is < design/ pressure increase downstream causing less gasflow from the reactor/ inert gas blanketing the tubes on the shell side/ exchangersnot vented before startup/ vapors condensing in the exchangers and causing a lossof area/ changes on the tube side.

TS Process: possible diagnostic actions: 2111, 2404, 356, 922, 686, 2418, 2816,2503, 2059, 2212, 1751, 2294, 1807, 1956. Another trouble shooter used the follow-ing sequence: 473, 2111, 922, 874, 832, 951, 686, 2418, 2816, 2503, 1807, 1550, 2383,1555.

Case’26: The case of the drooping temperatures (Used with permission fromT. E. Marlin, McMaster University).Estimate of difficulty: 5/10.

Cause: Insufficient air is being supplied to give the correct fuel-air mixture.Hence, the flame temperature decreases; less heat transfer and a resulting lower

546


temperature in the hydrocarbon stream leaving the furnace. As the temperaturedecreases, the temperature controller increases the fuel flow rate. This makes thesituation worse and creates a hazardous situation with a lot of excess fuel at a hightemperature. The excess natural gas may accumulate in the furnace and leave thefurnace with the flue gas. It’s hazardous because there might be an explosion alongthe furnace where the fuel-oxygen mixture is high. In particular, if the furnace has anegative draft and air leaks in through the observation holes then flashes or hotspots could occur here.

Possible corrective action: Immediate: either reduce the flowrate and return tosafe-park conditions where complete combustion occurs or activate the SIS to stopthe furnace. 2) use theoretical calculations to determine the required air flow for afurnace efficiency of 85 to 90% and 10% excess air. Daily inspection is recom-mended. Longer term: install a sensor to measure the percentage oxygen in the fluegas and this could be used by a feedback controller by adjusting the setpoint on theair flow controller FC-5 to adjust the damper close or open as needed. This is a veryserious situation. For insurance purposes, an audit would identify this as an inse-cure condition, request improvement and, in the meantime, follow up closely. Theinsurance premium would also increase.

TS Process: illustrative hypotheses: sensors at the outlet from the fired heater iswrong, T3/ sensor fault F2/ heat-transfer area too small/ heat-transfer area fouled/poor tuning of controller/ air flow too small to support combustion/ flameout/ gasvelocity on outside too small/ liquid-fluid velocity on inside too small/ decrease inthermal properties of process fluids/ process fluid flow increased.

TS Process: possible diagnostic actions: 340 and then 955, 2851, 2542, 1225, 1092,1452, 2191, 2101, 2881, 381, 2878, 1279, 171, 443, 820, 2301, 2574, 2077, 929.

Case’27: The IPA column

Estimate of difficulty: 5/10.Cause: There is no vent break on the exit liquid header. The condensate periodi-

cally syphons out of the condenser and leaves the tubes without a liquid seal. IPAgoes through the condenser uncondensed until enough condensate can build up toseal the tubes again.

Possible corrective action: Put in a vent breakTS Process: illustrative hypotheses: leaks in the overhead line, loss through the

bottoms, incomplete condensation: condenser undersized, air flowrate too small, airtoo hot, air recirculation, inadequate liquid seal in condenser, dirty heat-exchangersurface, incorrect calculation.

TS Process: possible diagnostic actions: 2134, 1501, 1685, 521, 1291, 18, 338, 239,106, 1879, 1135, 152, 1259, 1517, 1916.

547


Case’28: The boiler feed heater (based on a case supplied by P. L. Silveston,Chemical Engineering Dept., University of Waterloo)Estimate of difficulty: 5/10.

Cause: Inert gas coming in with the flash steam gradually builds up in theexchanger and blankets off some of the tubes.

Possible corrective action: vent the exchanger periodically or install a deaerationunit in the steam upstream at the ethyl acetate plant.

TS Process: illustrative hypotheses: condensate buildup/ trap malfunction/ sensorerror/ steam pressure too low/ inerts in the steam/ inert buildup in the exchanger/fouling of the tubes on the outside/ fouling of tubes on the inside/ steam superheattoo high/ steam flowrate < design/ water flowrate > design.

TS Process: possible diagnostic actions: 1237, 1788, 2199, 2699, 108, 1329, 323,177, 477, 364, 876, 617, 1632, 1533, 388.

Case’29: The case of the reluctant reactor (courtesy of W.K. Taylor, B. Eng. McMaster1966)Estimate of difficulty: 5/10

Cause: Valve A was too small. Even though it was wide open during startup, stillnot enough gas was bypassing the reactors on startup. Too much gas was forcedthrough both reactors during the heating step and, under these conditions, the tem-perature will not rise above 325 �C.

Possible corrective action: In the long term the correct size of valve should beinstalled. In the short term, to reduce the flow of gas to the reactor loops, shut downone compressor and provide loop gas to both North and South loops from one com-pressor. With the gas flow to each reactor cut in half, the reactors heated up rapidlyand in control. When the reaction temperature had been achieved in both reactors,the second compressor was started up and normal operations began.

TS Process: hypotheses: temperature-sensor error/ power failure/ cal rod heaterfailure/ gas flow too high/ refrigerant condenser too cold/ cooling-water condensertoo cold/ hydrogen concentration < 64%/ startup instructions not followed/ startuppressure too high.

TS Process: possible diagnostic actions: 1011, 532, 1459, 1412, 1117, 1048, 1546,1815, 1946, 1664, 2558, 1564, 1362, 1073, 1085, 660, 640.

Case’30: The case of the reluctant reflux (courtesy of ESSO Chemicals)Estimate of difficulty: 6/10.

Cause: Undersized valves on both the liquid block around the pump and the con-trol valves. No one checked the system resistance requirement when the valves werereplaced; the appropriate pump was not installed.

548


Possible corrective action: Immediate: reduce the condenser duty to keep thereflux drum from filling up. This will change the top specs unless a new set of oper-ating conditions are created. If this is not effective, shut down the plant.

TS Process: illustrative hypotheses: pump cavitating/ instrument error/ poorlytuned controller/ plug in line/ blockages in condenser/ motor electrical leadscrossed/ valve stuck shut or only partly open.

TS Process: possible diagnostic actions: 1058, 1137, 993, 460, 752, 1864, 1606,1513, 1985, 2071, 2213, 236, 2664, 2611.

Case’31: The ethylene product vaporizer (courtesy of C. J. King, Chemical EngineeringDept., University of California, Berkeley)Estimate of difficulty: 6/10.

Cause: Steam leak; steam then condenses, rises in the boiler and blocks off theeffective area for heat transfer.

Possible corrective action: Isolate the steam heater. Pull the bundle. Water pres-sure test to locate the leaks. Plug off the leaking tubes.

TS Process: illustrative hypotheses: buildup of inerts in the butane/ butane levelwrong/ leaks in the header/ faulty temperature shut-off switch/ impurities in thebutane/ vapor lock on ethylene evaporation inside tubes/ sluggish flow up anddown for butane vapor/ unstable flow with most of the butane going up the firstopening and down the second/ ethylene coming in subcooled/ ethylene pressure> expected/ loss of butane.

TS Process: possible diagnostic actions: 1359, 2427, 2248, 2952, 2528, 2603, 2807,2549, 2109, 2407, 2498, 331, 1549, 1745, 1257, 105, 353, 2390, 2469, 2919, 2572, 881,958, 252, 1812, 2869.

Case’32: The alarming alarm (courtesy T. E. Marlin, Chemical Engineering,McMaster University)Estimate of difficulty: 6/10.

Cause: Increase in feedrate to the column.Possible corrective action: Immediate: decrease the feed flowrate. Longer term:

increase the area for heat exchange for the condenser or, perhaps, increase the oper-ating pressure in the column.

TS Process: illustrative hypotheses: false alarm/ pressure sensor error.TS Process: possible diagnostic actions: 2227, 2304, 2011, 2461, 1724, 1870, 1152,

1357, 1052, 1408, 1323, 1173, 910, 802, 258, 13, 359, 2769, 858, 2103, 2054.

549


Case’33: Chlorine feed regulation (courtesy of Scott Lynn, Chemical EngineeringDepartment, University of California, Berkeley)Estimate of difficulty: 6/10.

Cause: Excessive residence time between the redox-potential sensor and the feedpoint.

Possible corrective action: Immediate: operate on manual. Longer term: relocatethe sensor closer to the feed point.

TS Process: illustrative hypotheses: fluctuation in caustic addition/ fluctuation inwater flow/ plugged chlorine inlet pipe/ vibration in the chlorine inlet tube/ poorlytuned control system on chlorine addition/ poorly tuned controller for caustic addi-tion/ pressure variation in vapor space in the chlorine storage tank/ variation in tem-perature outside the chlorine storage tank/ welder working close to the chlorine stor-age tank/ oxidation-redox sensor error/ sensor location wrong.

TS Process: possible diagnostic actions: 2943, 2546, 1167, 1379, 1096, 1019, 781,981, 2889, 225, 99, 2848, 2609, 345, 176, 1349, 1446, 1249, 2299, 2403, 1561, 1931,267, 2261, 1209.

Case’34: The cement plant conveyor

Estimate of difficulty: 6/10.Cause: Under the new conditions in the baghouse, the fines were being removed

from the blend just before the product was packaged.Possible corrective action: Change bags or filtering conditions so that fines are

not removed.TS Process: illustrative hypotheses: humidity causing particle clumping or bridg-

ing/ bridging in the feed hopper/ conveying process grinds the particles/ mixingprocess grinds the particles/ large-size particles missing from product/ fines miss-ing from product/ new filter cloth lets fines escape/ fines hang up in packaging hop-per/ fines fluidize in the packaging hopper.

TS Process: possible diagnostic actions: 5, 516, 538, 504, 1260, 2075, 1312, 1078,60, 2886.

Case’35: The cycling triple effect evaporator

Estimate of difficulty: 6/10.Cause: single steam trap serving all three effects instead of a separate trap for

each effect. The pressure in the condensate main= highest pressure connected to it.The condensate will only drain from this first stage; it builds up in the other stagesuntil the heat transfer drops and the pressure increases to a level that will allow drai-nage. Then, the second effect drains suddenly; the pressure drops and the heattransfer increases dramatically, increasing the flowrate to the 3rd effect and dropping

550


the pressure in the first effect. This cycling is shown in levels, temperatures, pres-sures and flows.

Possible corrective action: Put a separate condensate trap on each effect.TS Process: illustrative hypotheses: fouled tubes/ blockage/ instrument error/

control system poorly tuned/ flooded steam traps/ wrong type of trap/ inerts in thesteam/ poorly tuned control system/ fluctuations in the concentration in the feed/fluctuations in the feedrate/ fluctuations in the vacuum system/ steam to the ejec-tors is below design pressure.

TS Process: possible diagnostic actions: 2746, 2489, 2344, 2147, 2450, 1028, 1228,1278, 1328, 1378, 1476, 509, 44, 139, 1304, 281, 431, 860, 2108.

Case’36: The really hot case (courtesy W. K. Taylor, B. Eng. 1966, McMasterUniversity)Estimate of difficulty: 7/10.

Cause: Silica from the catalyst in the secondary reformer is vaporized and carriedover to deposit on the tubes of the waste-heat boiler.

Possible corrective action: Immediate: clean the tubes regularly. Longer term:lower the operating temperature in the reformer or find a new catalyst that is morerobust.

TS Process: illustrative hypotheses: flow blockage/ fouling of waste-heat boilertubes/ temperature or flow instrument error/ steam generation is < design/ control-ler fault/ leak/ bypass not entirely closed/ insufficient water in the steam drum.

TS Process: possible diagnostic actions: 2971, 37, 428, 1742, 1958, 1538, 2967,2526, 2946, 2599, 1134, 2120, 2495, 584, 2360, 2083, 2647.

Case’37 The mill clarifier (courtesy Don F. Fox, B. Eng. McMaster University)Estimate of difficulty: 7/10.

Cause: upsets upstream combined with a plugged sampler on the feed so thatincorrect feed concentration is recorded (400 ppm instead the correct value of700 ppm). The feed flowrate remains constant but the suspended solids concentra-tion fluctuates in the feed. The rake turning rate is too slow and does not move theexpectedly large concentration of solids into the central takeoff for the sludge pump.Ratholing occurs.

Possible corrective action: Immediate: shut off the rake and the sludge pump.Then increase the rake speed until the speed is consistent with 0.2 kg/s solids; startthe sludge pump. When the sludge is settling and the rate of raking= rate ofremoval the rake torque should read close to zero. The operator needs, in theinterim, to watch feed conditions with low suspended solids. Then, for light loadingof solids, the sludge pump should be stopped to allow the sludge to build up for ashort time until the torque shows an increase. Then the sludge pump is turned on.

551


Longer term: better sampling technique, so that the sampler is not plugged and sothat the operators have a more accurate indication of the actual concentration of sus-pended solids. Need to identify and correct upsets that occur upstream.

TS Process: illustrative hypotheses: upsets upstream/ rake too fast/ ratholing inthickener/ faulty design of thickener/ error in Parshall flumes/ incorrect flumedesign/ sampler error/ sampling at wrong location/ sampler line plugged/ velocityin sampler tube too low so that it gets a faulty sample/ excessive water flow becauseof rain this past week/ sludge pump takeoff nozzle poorly located/ damaged sludgerake.

TS Process: possible diagnostic actions: 1656, 1959, 493, 47, 204, 403, 703, 579,861, 1310, 1154, 1133, 1752, 2373, 1580, 2107, 2353.

Case’38: More trouble on the deprop! (courtesy T. E. Marlin, Chemical Engineering,McMaster University)Estimate of difficulty: 7/10.

Cause: The cooling water entering the battery limits is too hot in the summer. Asa result, the heat transfer is reduced and the concentration of propane in the productincreases.

Possible corrective action: Immediate: direct water onto the shell of the exchan-gers using a fire hose or reduce plant production or increase the pressure in the col-umn (if this is possible) or live with the trouble. Longer term: repipe the coldestwater to this unit or increase the area in the condenser or add a heat exchangerusing refrigeration.

TS Process: illustrative hypotheses: sample error/ temperature or pressure sensorerror/ increased concentration of very light components in the feed/ reduced towerpressure/ high feed rate/ fouled heat exchanger/ reduced flow of cooling water/increase in inlet temperature of the cooling water.

TS Process: possible diagnostic actions: 2831, 1633, 789, 513, 552, 1352, 1908,2405, 2504, 2516, 2008, 2769, 1512, 1558, 2911, 1012, 2174, 424, 774.

Case’39: The case of the lumpy sunglass display (from D. R. Winter, UniversalGravo-plast, Toronto, 2004)Estimate of difficulty: 7/10

Cause: the new batch of pearlized additive sent by the supplier was faulty.Possible corrective action: The supplier sent a fresh batch of pearlized additive.TS Process: illustrative hypotheses: Dirty machine/ dirty hopper/ moist feed/ too

many volatiles in feed/ lubricant or oil on mold/ incorrect mold lubricant/ feed con-taminated during material handling/ faulty raw material from supplier/ poor shut-down procedures/ screw rpm too high/ sensor error/ shutoff valve dirty or clogged/injection pressure too high/ gates too small.

552


TS Process: possible diagnostic actions: 2272, 1983, 1445, 911, 136, 2738, 2521,2310, 1934, 1545, 1857, 1183, 1084, 730, 594, 92, 425, 1945, 2087, 2724, 2575, 2771,1662, 1718, 1624, 2359.

Case’40: The cool refrigerant (courtesy T. E. Marlin, Chemical EngineeringDepartment, McMaster University)Estimate of difficulty: 7/10.

Cause: Over-specified control system. The heat transfer in the two exchangersdepends on the areas and the refrigerant temperature. Since both of these are appar-ently fixed, only extreme good luck would result in both process temperatures attain-ing their desired values. The process operation must be changed to increase the heattransferred in E100 and to reduce the heat transferred in E101. Thus, an extradegree of freedom is required.

C.W.

PC

1

H.P. Steam

periodic flow

F

1

T

2

L

1

L

2

LC

3

T

5

T

6

chilled

process

E 100

K.O. Drum

compressorsteam

turbine

LC

4

E 101

T

7

T

8

chilled

warm

Figure E-3 Automation of the intermediate corrective action for Case’40.

553


Possible corrective action: Immediate: The set point of each level controller canbe adjusted to achieve the desired process temperature. Level L4 will have to bedecreased. Level L3 will be increased. If Level L3 is already at its maximum, therefrigerant pressure P1 would have to be reduced. The actions could be automatedwith cascade designs T6 fi L3 and T8 fi L4. This is shown in Figure E-3. Longerterm: a design change to provide independent changes for the temperature of therefrigerant for each exchanger. Control valves could be installed in the vapor linesfrom each exchanger. By adjusting the valves, the pressure of the refrigerant in eachof the exchangers can be set independently. Use cascade designs T6 fi v1 andT8fi v2 as illustrated in Figure E- 4.

C.W.

PC

1

H.P. Steam

periodic flow

F

1

T

2

L

1

L

2

LC

3

T

5

T

6

warm

process

chilled

process

E 100

K.O. Drum

compressorsteam

turbine

LC

4

E 101

T

7

T

8

chilled

warm

v2

v1

Figure E-4 Design change for the long-term correction for Case’40.

TS Process: illustrative hypotheses: exchangers fouled/ temperature sensorswrong/ exchanger design fault/ faulty control design/ not enough area/ level controlfault/ buildup of inerts in E101.

TS Process: possible diagnostic actions: 2708, 2157, 1734, 190, 740, 1140, 69, 362,449, 944, 643, 713, 1369, 1046, 1809, 1932, 2338, 2783, 2843, 2583, 688, 124, 94,2389, 130.

554


Case’41: The ever-increasing column pressure (courtesy T.E. Marlin, ChemicalEngineering Department, McMaster University)Estimate of difficulty: 7/10.

Cause: Fouling on the water side because the pressure controller has been adjust-ing the cooling-water flow rate, and, at times, the flow has been too low, the cooling-water temperature exceeded 50 �C. The initial pressure control was poor because thecontrol system was poorly selected: control of the pressure based on the dynamics ofthe condenser can be a slow process.

Possible corrective action: Immediate: try to prevent too high a pressure in thecolumn: reduce the column feed rate and column overhead rate or reduce the refluxflow rate and thus reduce the overhead vapor rate. The latter gives an increase in theimpurity of the overhead product. Longer term: clean the exchanger and redesign acontrol system that prevents overheating of the cooling water. An example is shownin Figure E-5 where the valve on the exit of the exchanger is manipulated to changethe level of condensed liquid in the condenser. This design gives relatively fastresponse and should result in better pressure control.

M.P.

steam

Figure E-5 Improved control for Case’41.

555


TS Process: illustrative hypotheses: sensors wrong/ cooling-water control valvemalfunction/ poorly tuned pressure controller/ condenser underdesigned/ cooling-water feed too hot/ fouled condenser tubes/ inerts on shell side/ condenser notvented when started up/ restriction in condensate line to reflux drum and conden-sate building up in condenser/ cooling-water flow too low/ feed composition in lightcomponents increased/ overhead temperature changed.

TS Process: possible diagnostic actions: 546, 2205, 1109, 1313, 1299, 1619, 1454,1119, 1753, 2631, 1123, 1371, 231.

Case’42: The case of the weak AN (courtesy W. K. Taylor, B. Eng. 1966, McMasterUniversity)Estimate of difficulty: 7/10.

Cause: the packing seal in the acid feed pump is worn and large amounts ofdemineralized water being fed to the seals enter the acid. Normally the flow of watergoes to the outside of the packing seal to prevent acid from leaking out into thepump. The pressure difference between the water side of the seal and the acid sideis 500 kPa. Normally the amount of water going through the seal into the acid feedis so small that there is no detectable change in the acid concentration. However,the seal is so worn that the water flow is large; the acid is diluted; the reactions aredecreased.

Possible corrective action: replace the packing seal and adjust the clearances.TS Process: illustrative hypotheses: temperature-sensor error/ sampling error/

analysis error for AN/ cooling water leak into the neutralizer-reactor/ ammoniavapor feed is moist/ urea plant off-gas has too much water/ rain leaked into thenitric acid storage tank and the acid in the storage tank is dilute/ steam generated inthe reactor-neutralizer condenses and runs back into the reacting liquids/ poor mix-ing in the reactor/ increased flow of cooling water/ cooling water inlet temperaturecolder than usual.

TS Process: possible diagnostic actions: 1082, 1440, 880, 2957, 2749, 2105, 2306,1306, 1149, 1050, 750, 570, 448, 263, 1829, 1735, 698, 2233, 2466, 2775, 2581, 2321,1485, 2155.

Case’43: High pressure in the debut! (courtesy T. E. Marlin, Chemical EngineeringDepartment, McMaster University)Estimate of difficulty: 7/10

Cause: the buildup of non-condensable propane in the debutanizer overhead con-denser. The propane had come through from the upstream depropanizer during itsupset. For the debutanizer overhead condenser, the area was insufficient or the pres-sure not high enough or the cooling-water temperatures or flowrates were insuffi-cient.

556


Possible corrective action: lower the column pressure by reducing the reboilerduty and/ or ask the operator to open the “vent to fuel” manual valve on the conden-ser for a few minutes to purge the condenser. Long term: no action is needed.

TS Process: illustrative hypotheses: faulty pressure-relief valve/ wrong setting onthe relief valve/ pressure sensor wrong/ flowrate of cooling water too low/ plugs inthe cooling water line/ fouled tubes/ area in the condensers too small/ control valveon condensate stuck shut and condensate floods condenser/ inert vapor in the con-denser/ air not purged from the condenser during startup/ excessive boilup.

TS Process: possible diagnostic actions: 365, 994, 1491, 1102, 2401, 2902, 2570,2505, 2443, 278, 22, 458, 202, 960, 735, 853, 1301, 2769, 1605, 813, 2983.

Case’44: Reactant storage

Estimate of difficulty: 8/10.Cause: Five faults: 1) no mixer installed to provide convection for good heat trans-

fer; 2) contaminated steam because the steam line comes off the bottoms of theheader; 3) waterlogged coil because the steam comes in the bottom of the coil,4) flooded steam trap because the condensate line “into the header” comes into thebottom of the header and 5) cyclical condensate buildup in the coil because of poordesign for the removal of condensate from the “bottom of a coil”.

Possible corrective action: Immediate: the corrective changes require reconstruc-tion. Add a mixer; relocate the steam line to the top of the steam main; relocate thecondensate line into the top of the condensate header; install a condensate lift mech-anism to allow steady withdrawal of the condensate from the bottom of the coil. Seesketch in Figure E-6.

FALL

FROM

COIL

TO

TRAP

Figure E-6 Condensate lift mechanism showing closeup offitting for Case’44.

557


TS Process: illustrative hypotheses: no mixing/ wrong steam trap/ flooded steamtrap/ condensate buildup and then cycling out of coils/ sensor fault/ sensor locationwrong/ steam valve too big/ dirty steam/ corrosion products in steam/ steam trapbypass open/ cycling condensate from other units flowing back into this unit viacondensate header/ fouled heat coil.

TS Process: possible diagnostic actions: 2805, 2696, 2014, 1220, 1409, 1492, 1037,1014, 727, 935, 2287, 2421, 612, 402, 2917, 2092, 2577, 2979, 1702, 1093, 1494.

Case’45: The deprop bottoms and the ISO dilemma (courtesy T. E. Marlin, ChemicalEngineering Department, McMaster University)Estimate of difficulty: 8/10.

Cause: The bottoms composition is not controlled in real time. The C-8 tempera-ture, as measured in tray’9 downcomer, is measured and controlled (by feedback)by adjusting the steam flow to the reboiler. This tray is not a perfect indicator of thebottoms composition. Thus, changes in the C-8 feed composition result in changesin the bottom composition.

Possible corrective action: Immediate: operator adjust the set point on TC-5 basedon the real-time sensor analysis A-1 on the overheads of the downstream debutani-zer: if C3 is too low, decrease the TC-5 set point and vice versa. Longer term: AC-1cascade to TC-5. The AC-1 primary will correct the deficiencies in the relationshipbetween TC-5 and the bottoms composition.

TS Process: illustrative hypotheses: sampling error/ analysis error/ variablechanges in feed concentration/ fouled reboiler/ faulty reboiler design/ bottoms con-troller poorly tuned / controller faulty/ temperature sensor faulty/ electrical storminterfering with signals/ poor design of control system.

TS Process: possible diagnostic actions: 2380, 803, 1563, 1156, 2078, 436, 2251,1334, 219, 1303, 2989, 2769, 652, 201, 2102.

Case’46: The not so cool chiller (courtesy Scott Lynn, Chemical Engineering,University of California, Berkeley)Estimate of difficulty: 8/10.

Cause: construction garbage left in thermosyphon line causing limited ethylenecirculation such that the bundle in the chiller was only half covered with liquid.

Possible corrective action: clean out the line.TS Process: illustrative hypotheses: temperature-sensor error/ level sensor error/

flowrate too low/ bypass open on pump/ leak from process fluid into the refrigerantloop/ boiling shifts from nucleate to film boiling along the length causing vaporinstabilities/ ammonia condenser not condensing / ammonia leaking into the pro-

558


cess fluid/ design-problem area too small/ plug in vertical thermosyphon line/ valvepartially closed in thermosyphon line.

TS Process: possible diagnostic actions: 2598, 907, 863, 746, 1429, 1302, 1397,1780, 1942, 2222, 2471, 2193, 2584, 2697, 2217, 2042, 1625, 1817, 714, 875, 1869,1711, 1693, 2340, 1821, 2875, 1161.

Case’47: The fluctuating production of acetic anhydride

Estimate of difficulty: 8/10.Cause: The control system was poorly designed because the sensor for the steam

flowrate is a flowmeter on the vapor. This might work for well-behaved vapors; butacetic acid is not well behaved. Some acetic acid vapor dimerizes so that the molarmass of the vapor is somewhere between 60 (for the monomer) and 120 (for thedimer). For design purposes we usually used 100. However, the control is based on asignal from an orifice meter measuring the flow of the vapor. But the apparent massflowrate (estimated from the signal from the orifice meter) is a function of the gasdensity and the measured Dp. Therefore, even if the mass flowrate of the vapor isconstant, the apparent flowrate read by the orifice meter is erratic because of thevariation in the molar mass of the vapor.

Possible corrective action: Shift the controller to control the steam flow by the liq-uid flowrate and adjust the liquid flowrate via level control. Or better still, add a gasanalyzer to determine the density of the gas. This datum is multiplied by the pres-sure drop given by the orifice meter, a square root function is taken of the resultingdata and sent to the steam flow controller. This is illustrated in Figure E-7.

PI

201

PI

101

FI

201

LIQUID

ACETICACID

TO CONDENSATEHEADER

TO CRACKINGFURNACE

STEAMMAIN

LC

A

201

FRC

202

��

Figure E-7 Modified control system for the vaporizer for Case’47.

559


TS Process: illustrative hypotheses: poorly tuned controller/ location of orifice me-ter too close to bends/ wrong orifice diameter/ condensate cycles because of poortrap or flooding of the exit of the trap/ cycling steam/ level cycles/ cycling in thedownstream vacuum pumps/ uneven temperatures in the cracking furnace causingcycling reaction rate/ cycling brine and water to the condensers causing upstreamcycling/ cycling in “barometric legs” for the condensed water from the system.

TS Process: possible diagnostic actions: 2973, 2590, 2153, 1607, 1927, 1164, 2787,1706, 2057, 2604, 2933, 560.

Case’48: The column that just wouldn’t work (courtesy T. E. Marlin, ChemicalEngineering Department, McMaster University)Estimate of difficulty: 8/10.

Cause: One or both of the isolation valves around the depropanizer reflux valve werenot completely opened. The resistance to flow is too high. The FC-4 controller hasopened the control valve 100%but the reflux flow is much below the desired value.

Possible corrective action: Immediate: open the isolation valves. Longer term:check before the process is started up.

TS Process: illustrative hypotheses: dry trays/ flooding/ insufficient reflux/ lowfeedrate/ high boilup/ feed temperature too high/ reflux pump cavitating/ sensorfault/ analysis fault/ poorly tuned controllers/ transient vapor puff from horizontalthermosyphon reboiler/ bypass open on reflux control valve.

TS Process: possible diagnostic actions: 1820, 1712, 1628, 2309, 2760, 2974, 1573,1515, 1374, 943, 974, 1766, 1913, 1695, 2277, 2127, 2925, 2593, 2366, 1655, 1747,2769, 1217, 1015, 1246, 1647, 1604.

Case’49: The case of the faulty stretcher pedal (from D. R. Winter, UniversalGravo-plast, Toronto, 2004)Estimate of difficulty: 8/10

Cause: Wrong location of gate and mold temperature was too low.Possible corrective action: Use only one gate with a location at the hub; control

the mold temperature at 65 �C.TS Process: illustrative hypotheses: Contamination near the hub/ moist resin/

uneven temperature over the mold/ injection too slow/ injection too fast/ faultydesign of the product/ faulty design of the mold/ too much cooling/ too little cool-ing/ faulty resin/ incorrect foaming agent/ correct foaming agent but wrong concen-tration/ inadequate mixing/ cooling cycle too long/ cooling cycle too short/ feedtemperature too low/ mold too cold.

TS Process: possible diagnostic actions: 2473, 1522, 2314, 2892, 2629, 384, 721,904, 1076, 1419, 1939, 1980, 2447, 2007, 149, 623, 1308, 1741, 1965, 2267, 782, 965,1159, 1126, 1754, 1597, 1783.

560


Case’50: The cleanup column (courtesy W. K. Taylor, B. Eng. 66, McMasterUniversity)Estimate of difficulty: 9/10.

Cause: Two causes: 1) the steam tracing was poorly designed and overheated thesuction line to the pump and 2) the vortex breaker was poorly designed and coveredmost of the nozzle exit cross section leaving the column at tray’4.

Possible corrective action: Immediate: shut off the steam tracing. This helps, butthe pumping is still below expectation. Long term: correct the vortex breaker. Italmost plugged the exit from the column! Figure E-8 shows a photo of the vortexbreaker.

Figure E-8 Vortex breaker blocks the outlet in Case’50.

TS Process: illustrative hypotheses: product not in the well of downcomer becauseof: downcomer blocked at top/ overhead leaks out of well and goes to bottoms/ traycollapsed and liquid bypasses the well/ no feed to the column/ feed bypassing todrum disposal/ insufficient steam stripping/ insufficient boilup/ reflux water offand organic goes out the top/ feed composition is primarily heavies. Product in thewell of downcomer but cannot be pumped out because of: pump off/ pump notprimed/ pressure relief causing liquid to recycle around the pump/ pump cavitat-ing/ steam tracing too hot/ vacuum leak of air into system/ wrong pump selectedfor the job/ insufficient head/ line blocked/ product solidified in line/ junk left inline after maintenance/ physical blockage from vortex breaker/ no vortex breaker.

TS Process: possible diagnostic actions: 753, 157, 2605, 2230, 1850, 1950, 1674,847, 946, 695, 650, 2374, 2475, 2868, 797, 2030, 2413, 1612, 1854, 1525, 1336, 1039,761, 524, 211, 393, 282, 2141, 2281, 2785, 2857, 1420.

561


Case’51: Revisiting the cleanup column (courtesy W. K. Taylor, B. Eng. 66, McMasterUniversity)Estimate of difficulty: 9/10.

Cause: weeping sump at Tray’4; collapsed trays in the stripping section; weep-ing/ leaking trays because the sieve holes have corroded.

Possible corrective action: Repair the sump; replace the trays. Long term: identifythe cause of corrosion and remedy by removing the cause or selecting differentmaterials of construction. Figure E-9 is a sketch of the sump.

25" 12½"

1"

LEVEL GLASS

TOWER CONNECTION

TOWER CONNECTION

To 114-J

2'8" ID

Product Sump

13"

6"

seal pan

4th Tray

TOP VIEW

not to scale

seal

sump

LEVEL GLASS

CONNECTIONS

4th TRAY

NOZZLE

Figure E-9 Sketch of the sump in Case’51.

TS Process: illustrative hypotheses: sampling error/ analytical error/ insufficientboilup/ stripping steam flowrate too low/ vacuum fault/ trays collapsed/ downco-mers not sealed/ sieve holes corroded and trays weeping/ product sump leaking.

TS Process: possible diagnostic actions: 2605, 2230, 1950, 1674, 847, 946, 2650,1376, 2644, 2517, 2133, 2030, 2413, 2854, 2225, 1336, 761, 524, 2221, 393, 785, 404,1120.

562


Case’52: The case of the swinging loops (courtesy W. K. Taylor, B. Eng. 66, McMasterUniversity)Estimate of difficulty: 9/10.

Cause: Trying to balance the flow to more than one downstream unit from a com-mon header is always tricky. The pressure drop in both downstream branches mustbe exactly equal; otherwise the flow will go preferentially to the branch with the leastresistance. A second, complicating element, is that the branches include a reactorthat, if one reactor develops a hot spot, the reaction rate increases, and the flow tothat reactor increases. The third complication is that the configuration of the feedgas into the manifold is not a mirror image (and fresh feed contains 1% inertwhereas recycle feed contains 15% inert). Thus, if more total feed is called for by theSouth loop (which is usually fed by fresh feed and recycle) then the source of the“extra” feed for the South loop is “fresh feed” from the North compressor. In thisway the “extra” feed to the South loop is richer in fresh reactants and has less inerts.Contrast this with what happens if more total feed is called for by the North loop.Then, the source of “extra” feed for the North loop is the “recycle feed” from theSouth compressor. This recycle feed contains 15% inerts so the net result is a feed tothe North reactor that is slightly higher in inerts. When production in the Southloop is 20% higher than the North loop, then the South loop gets all of North loop’sfresh feed (because of the 5:1 recycle to feed ratio). The North reactor gets no freshfeed, the North reactor quickly reaches equilibrium and stops reacting for lack ofreactants. It is little wonder that it is usually the South loop that overheats, althoughoccasionally the North loop does when the operators are trying to smooth out theloops. The situation is not dangerous because the temperature never exceeds 590 �C(which is well within the catalyst and vessel limitations).

Possible corrective action: Immediate: try to keep the South reactor from gettingtoo hot by operating it 5 to 10 �C cooler than the North one. Monitor it closely.Interim: add a high-temperature alarm. Longer term: install isolation valves so thateach compressor system feeds one system.

TS Process: illustrative hypotheses: fluctuating valve stem in control valve A/ poormanual control of valve B/ fluctuating kickback on one of the compressors/ fluctuat-ing valve B on the exit of the recycle compressors/ during startup both reactors arenot heated up to identical inlet temperatures/ depth of catalyst differs between thetwo reactors/ analysis of pressure drops in the two loops shows that Dps are notidentical for the same flow/ recycle gas in one loop is cooled more than that in theother loop/ hydrogen content in one loop leaving the reactor is different so that thecooling is more in one loop than the other.

TS Process: possible diagnostic actions: 1080, 1158, 2354, 1852, 1982, 1845, 1602,920, 683, 556, 131, 369, 289, 1824, 1884, 2318, 2392, 2872, 2903, 2623, 1620, 1657,1181, 1216, 1384, 1268, 1340, 633, 2756.

563

565

F.1Enrichment Tasks for Awareness Development in Section 5.1

Task 5.1-BTasks related to science and engineering

B.2 Bernoulli’s equation, expressed in the old units of ft lbf / lbm, is:

D<v>2

2gþ Dp

�þ Dz ¼½ � ft:lbf

lbm

wherev = velocity, ft/sp = pressure, lbf/ft

2

� = density, lbm/ft3

z = elevation or height, ftg = local gravitational acceleration, ft/s2

Consider the term Dz.

a) since the units of z are not the same as the units in the equation, it shouldnot be included.

b) the term should be included, the units of ft are acceptable because the correctunits are understood.

c) consistency of units does not apply to this equation; go ahead and use theterm as it is.

d) consistency of units does not apply; the equation was derived in 1738 whenconsistency was not an issue. Use it the way Bernoulli derived it.

e) the term must be multiplied by g / gc.f) other.

Appendix FOther Tasks for the Skill-Development Activities in Chapter 5


Appendix F Other Tasks for the Skill-Development Activities in Chapter 5

Task 5.1-CGeneral tasks more related to trouble shooting

C.2 I am to write a report to my supervisor telling her about what I have done thispast month on my research project. Which of the following topics are most impor-tant to be included?

1) the pump for the equipment broke and only half of the experiments could becompleted,

2) the mechanics could not fix the pump; the new parts will arrive in 5 weeks,3) the results for the experiments that were completed seem to show that the

most important effect is the temperature – not the concentration of the reac-tants. But, we cannot do a statistical analysis of the data until all of the experi-ments are completed.

4) I went to the library and studied the literature for another project because Icould not do the experiments.

5) the technician will not be able to do the experiments in 6 weeks because he willbe in Kyoto.

Task 5.1-DEngineering related

D.2 A researcher is studying the catalytic combination of two reactants A and B toform a single compound C. The reaction is first order with respect to the reactants.Ten minutes after the reaction has started, the researcher accidently adds chemicalX that combines rapidly with A. Enough X is added to react with about 1/2 of A.Which of the following is most likely to occur when the clock reads 12 minutes afterthe reaction has started?

1) the reaction of A and B to produce C would proceed more rapidly.2) the reaction of A and B to produce C would proceed at the same rate.3) the reaction would stop; that is, no more C would be produced.4) the reaction of A and B to produce C would proceed more slowly.5) other.

Task 5.2-CTerry Sleuth and the Case of the Stinky Margarine

The spring rains turned the landscape into a muddy mire. What miserable weather,Pete thought. Why even the creeks by his house were overflowing the banks. Hecould imagine the quagmire down by the Burlington treatment plant where some ofhis colleagues were checking out the operation. He was glad he was indoors.

Rinnnnng ... “Water Consultants” answered Pete. “Hmmm. Please slow down abit so that I can understand what the problem is.”

566

F.1 Enrichment Tasks for Awareness Development in Section 5.1

Pete sipped on his coffee ... He saw Terry bustle in from the computer and wavedTerry over. Motioning, he asked Terry to lift up the extension and hear the latest chal-lenge that was besetting the engineers and scientists at Water Consultants.

“So you make margarine ... ahum.. and now the margarine tastes like ammonia.But why do you think that we might help? We specialize in water and water quality!We specialize in water treatment, adsorption, ultrafiltration, and trace organic analy-sis but not margarine production” ...

“Oh, you’ve tried everyone else and no one seems to have any answers so youthought that maybe we can help.” said Pete as he cast a puzzled look at Terry andthen suggested “Maybe my colleague has some questions or suggestions for you;I’ve asked Terry Sleuth to listen in on our conversation” as Pete dodged the issueand passed the situation neatly over to Terry.

“Perhaps you will refresh my memory on how you make margarine; you purchasevegetable oils, clean them up and then blend them with milk.” offered Terry Sleuth,“What do your lab analyses of the incoming oil and milk show?”

“OK, so they seem to be the same as usual. How do you clean up the oils?” Terryjotted down “ deodorize under a vacuum and with live steam, adsorb undesirableson Fuller’s earth, filter out the Fuller’s earth and cool down the oil”. Pete’s earsperked up when he heard the word adsorb with Fuller’s earth because that was hisspecialty. Maybe they could help. Terry seemed lost in thought as Terry looked outthe window at the quagmire and asked, seemingly to give more time for thought,“Do you produce your own steam from the plant with the intake from Butcher’screek?”. “OK you do;” “And you pump the cooling water from the same source?Hmmm.” “Do you filter the cooling water before you send it to the margarine plant?It’s rather dirty these days with the terrible weather outside”. “OK so you filter it andsend it to the cooling-tower circuit.” “All your coolers have no direct contact witheither the milk or the oil, Right?” “Right.”

Pete asked “Do you regenerate the Fuller’s earth for reuse?” Hmmm.Terry said “I think we can solve your problem; we’ll be over in an hour to take a

few samples. We should have an answer for you in about 3 hours”.What samples did Terry take? what analyses were done and what advice did Water

Consultants give to solve the case of the stinky margarine?

Task 5.2-DTerry Sleuth and the Case of Boorish Bob

Bob jogged by Terry’s office. The odor of locker-room socks and just plain BO waftedthrough the door. Terry opened the window.

As Terry turned back to the desk, the perspiring face of Bob appeared in the opendoorway. “Joggin’ really makes you feel good,” announced Bob proudly.

“And talkin’ of feeling good, I have a meeting in half an hour with our client Mar-lene from Transpix. She asked us to calculate the weight per cent of zinc chloridethat they should be using for the additive solution they are using. I’ve got those cal-culations all done. They were really simple.” With that, he threw his calculationsand results down on Terry’s desk.

567

Appendix F Other Tasks for the Skill-Development Activities in Chapter 5

“Look em’ over. You’ll see they are great! Many of the others here in this organiza-tion don’t take the care I do to check and double check. Just as one last check, youcheck me out! After all, Marlene is an important client and we don’t want to makeher upset, do we?” gloated Bob.

Terry breathed deeply the fresh air from the window and then looked at the calcu-lations. “What was Marlene’s question?” asked Terry.

“She has a solution at 68 �F (20 �C) that has a density of 1.4890 g/mL that contains0.0411 lb-mole of zinc chloride per US gallon of solution. She wants that expressedas weight%.” smugly beamed Bob. His bad breath almost made the paper curl.

“I looked up the density of water at 20 �C, (0.9982 g/mL) and the density of thesolution to be 1.4890 Mg/m3” boasted Bob. “A US gallon is 8.337 lbm but the realthing to remember is that a gallon is a volumetric unit of measure.”

Tired of Bob’s tirade, Terry looked at Bob’s calculations:

0:0411· 136:28

8:337 · 1:48900:998

¼ 45:2%

Terry looked at Bob, frowned, and said ... What did Terry say?

568

569

. From Chapter 6

6.1 a. stop the operation, open the pipe at the various fittings and look: veryexpensive.

b. increase the flowrate and note the difference in pressure drop: ineffectiveactivity unless one hopes to dislodge the obstruction.

c. decrease the flowrate and note the difference in pressure drop: ineffective.d. estimate the pressure drop and compare with measured values: recom-

mended early test.e. stop the operation, open one end and send a plumber’s worm through the

line: expensive.f. and maybe there might be some others.

6.2 1) check the pressure difference between the refrigerant side and the processside. Easy to do. If the refrigerant pressure is greater than the processside, then the leak will be into the process and not into the refrigerant.

2) sample the refrigerant and analyze for impurities using LGC. Expensive,time consuming.

3) read the temperatures and pressures and compare these with the data onpressure–enthalpy charts for the refrigerant. Easy, pressure should behigher and temperature lower.

4) read the pressure gauge on the compressor suction. Easy, pressure higher.Select 1, 4, 3, in that order.

6.4 Typically “only card U” is chosen whereas the correct test is “cards U and 9”.If card 9 is not chosen, you might tend to have a confirmation bias in thatyou look only for a test that proves the case and not an additional test tocheck the negative, namely that card 9 does not have a vowel.

6.5 The Tom Dayton murderThe answer is’3, conclude that Tom died from self-inflicted shot based onm, p, q, hh, r, s, v, w, kk bb.If you chose’1, perhaps you might have a bias in either selecting theevidence or reaching conclusions. Look over the characteristics given in Sec-tions c-i and c-ii.

Appendix GSelected Responses to the Activities in Chapters 6 and 7


Appendix G Selected Responses to the Activities in Chapters 6 and 7

6.15Adiabaticity: Two simultaneously contradictory beliefs: heat is what is “in” hotthings and heat is energy flowing from hot to cold (from HELP P, “bugs” session20).

. From Chapter 7.

Activity 7.1a

Audience: Engineer to operator. The operator should be told the purpose of the test,how and why the results will be used. Is the operator permitted by the union toobtain samples, or is it the lab technician or instrument shop?

Content: The instructions should have been clearly stated. Location of sampleport; procedure for gathering the sample safely, how to clear the sample line and geta representative sample. What volume of sample to take? Can the samples bestored? What sample bottles to use? How to label them? Time period between sam-ples? What analyses are to be done? When you would like the results?

Organization: difficult to assess because the instructions are so short.Style: rather condescending attitude communicated.Format: should have been in writing.The Engineer should have phoned the lab, checked on the types of analyses that

could be done and the time frame, the volume of samples required and obtainedsample bottles from the lab. Then Jose should have verbally explained what samples,been present when the samples were taken and labeled the samples. The writteninstructions should have been part of the approach taken.

Activity 7.1b

The Pulp Mill engineer verbally told the instrument shop to “check the flowmeters.”The results reported to Andre verbally were “Some of the instruments were awayoff; they forget which ones but that’s OK because they are all OK now.” This meantthat all the test data taken when the instruments were “way off” we useless becausea) we don’t know which values were correct and which were wrong and b) for thosethat were wrong, we don’t know what the correction factor should be. Thanks to IanShaw, B Eng. PhD McMaster for this example.

Activity 7.2

Ahmed shows no listening in this situation. It is hard to identify five strengths. Theymight be: 1. He acknowledges the operator. 2. He goes to the control room firstinstead of going directly out on the plant to see the VC. 3. He calls the operator byname. 4. He acknowledges that he is new and 5. He talks aloud about his observa-tions. Areas to work on: many! but start simply with the two most important ones:1. Take time to introduce himself. 2 Learn to listen and seek input from the opera-tors.

570

Appendix G Selected Responses to the Activities in Chapters 6 and 7

Activity 7-4

Tanya: RIGHTS claimed right to have an opinion,Honored: initially tried to honor Marcos’s right to express an opinion (iii)but failed to honor Marcos’s right to have an opinion; (vi) (vii) lack ofrespectDestroyers: criticism (vi) (vii)

Marcos: RIGHTS claimed right to have an opinion.Honored: lack of respectDestroyers: defensive (ix) (x); criticism/sarcasm (xi) (xii)

Let’s try this again:Tonya: “OK, my six hypotheses are (i) 1. tray collapsed stripping section, 2. too

much bottoms fed to debutanizer, 3. too much overheads in feed, 4. feedvalve FV1 stuck, 5. pump F-26 not working, 6. not enough feed to thecolumn (ii). What do you think? (iii)

Marcos: “Your six hypotheses are a good start. Actually, I really support your firsthypothesis I encountered something like that on the S256 plant last year.Same evidence and it was a collapsed tray.

Tanya: “Maybe. But checking out a tray is more complicated than some simpletests that might rule out some of the other hypotheses. Why don’t wecheck out the temperatures, pressures and flows that we see on the plant.That shouldn’t take long.”

571

573

Here are some example “causes” or faults that might occur for centrifugal pumps,valves, controllers and instruments, pipes and fittings, steam traps, shell and tubeheat exchangers, plate heat exchangers, distillation and tray absorbers and sedimen-tation centrifuges.

. Centrifugal pumps

1. instrument wrong.Design. 2. discharge pressure required is greater than that provided by pump.

3. total system head required is much smaller than head provided by pump. 4. suc-tion pressure is too low/too high. 5. available NPSH is too low or suction lift is toohigh. 6. undersized suction pipe. 7. excessive vapor entrainment. 8. incorrect sumpdesign. 9. viscosity higher than expected. 10. density different from expected.11. incorrect piping layout. 12. sump level below intake. 13. loose valve or disk insystem. 14. resonance between rotating speed and foundation. 15. incorrectly de-signed baseplate. 16. system requirements too far out on the head-capacity curve.17. operates at very low capacity. 18. one pump in the system affects another.

Installation: 19. misalignment. 20. wrong. damaged or improperly installed bear-ings. 21. wrong. improperly installed packing. 22. wrong type or amount of lubri-cant. 23. mechanical seals exert excessive pressure. 24. glands too tight. 25. rotor notbalanced. 26. shaft bent.

Operation: 27. air entering or not dearated initially. 28. impeller backwards indouble-suction pump. 29. impeller clogged with solids. 30. impeller damaged, vanesworn or missing. 31. impeller diameter different from expected. 32. no key betweenshaft and impeller. 33. speed too high or too low. 34. running at critical speed.35. rotation in wrong direction; leads reversed on power to drive. 36. faulty casing,cracks, leaks. 37. insufficient clearance between impeller and volute tongue.38. wrong shape for volute tongue. 39. leakage through the wear surfaces (wearrings) process fluid out/ lube oil into process/ air leak in. 40. internal waterwaysrough and burred. 41. plugged inlet or exit. 42. suction strainer plugged. 43. motorfaulty, undersized. 44. drive connection fails, slips. 45. casing and wear rings worn.46. defective casing gasket. 47. no cooling water to water-cooled stuffing boxes.48. dirt and grit in sealing liquid. 49. motor windings fail. 50. motor bearings fail.51. change in or wrong frequency of voltage to the motor. 52. wrong phases hooked

Appendix HData about “Causes” for Selected Process Equipment


Appendix H Data about “Causes” for Selected Process Equipment

up to the motor. 53. short or fuse in power to motor. 54. motor switch not turned on.55. power failure on site. 56. power failure at the utility. 57. faulty signal to relay tostart motor. 58. faulty relay for signal from process to the pump motor. 59. notprimed.

. Valves, controllers and instruments

Measurement instruments: 1. wrong sample location. 2. sample withdrawn is notrepresentative. 3. sample mixed up with other samples; incorrectly identified.4. error/ contamination in the sampling. 5. insufficient or no purging done. 6. nosample withdrawn because line plugged. 7. faulty measurement made on sample;instrument wrong. 8. instrument installed incorrectly. 9. instrument corrections notmade to the answer. 10. calibration/ standard conversion wrong. 11. result reportedincorrectly or not reported at all; or confused with other sample or signals reversed.12. periodicity to behavior and sampling done at the wrong time. 13. instrumentsdamaged during startup trials or between startup and current situation. 14. flashingoccurs in instrument.

Valve: Design: 15. valve too small. 16. valve too large. 17. valve trim wrong mate-rial. 18. valve wrong type. 19. air to open when need air to close. 20. valve positionerneeded but not installed. Installation: 21. installed backwards for non-vacuum con-ditions. Mechanical: 22. stuffing box too tight. 23. stuffing box too loose. 24. wrongmaterials in stuffing box. 25. valve stem bent. Operation: 26. valve bypassed.27. block valves closed. 28. valve seat eroded away. 29. valve stem eroded and vibrat-ing from flow. 30. valve stuck open, closed or midway. 31. receives the reverse signal.32. accumulation of dirt in the valve. 33. valve stem falls off. 34. faulty valve posi-tioner.

Controller: 35. wrong set point. 36. no integral action and hence no return afteroffset. 37. loss of instrument air or electric power. 38. signal reversed. 39. controllerstuck. 40. sends out low signal; ie fails low. 41. sends out high signals; i.e. fails high.42. set on manual.

Recorder/indicator: 43. displays wrong values. 44. pens mechanically stuck andcannot cross. 45. chart motor not functioning. 46. chart pens out of ink. 47. chartstarted and date one-day off.

. Pipes and fittings

Design: 1. wrong diameter. 2. wrong wall thickness. 3. insufficient support. 4. noallowances for thermal expansion. 5. not insulated. 6. faulty location of lines: steamoff the bottom of the header; condensate into the bottom of a header. 7. insufficientdrain lines. 8. low points and undrainable sections in lines. 9. valves and instru-ments at wrong elevations. 10. instruments installed at wrong locations. 11. noallowance for purge. 12. item on diagram but not installed. 13. inadequate statementof specifications on the contract.

Installation: 14. left garbage in the pipe. 15. installed filter, strainer, screen back-wards. 16. not pressure or vacuum tight. 17. installed wrong pipe. 18. made on-sitechanges without recording them. 19. substitution of inferior materials without/withnotification. Operation: 20. line fails/cracks. 21. corrosion products build up or con-

574


taminates. 22. purge operated for too short a time or not at all. 23. line sags. 24. linereceives abnormal/unexpected stress and ruptures. 25. thermal expansion greaterthan expected; pipe tears loose from the support. 26. excessive steam tracing.27. insufficient steam tracing.

. Steam traps

Design: 1. wrong type of trap installed. 2. wrong size of trap. 3. wrong orifice sizefor the Dp across the trap. 4. traps in the system interact and interfere with theoperation of each other. 5. no strainer included. 6. condensate cannot get to the trap.

Installation: 7. installed backwards. Operation: 8. filter plugged. 9. trap inopera-tive because it is frozen. 10. bypass left open. 11. mechanically stuck internals.12. mechanically loose and unattached internals. 13. disk installed upside down.14. exit flooded. 15. exit plugged. 16. trap internally plugged with dirt.

. Shell and tube heat exchangers, condenser, reboilers

1. Instrument fault.Design: 2. overdesign by adding more area for future. 3. overdesign by use of

wrong fouling allowance. 4. overdesign “to be on the safe side”. 5. overdesign byadding large fouling factors to extend runs between cleaning. 6. underdesign.7. inadequate allowance for thermal expansion. 8. no inlet protection baffle. 9. clear-ance between the shell and cross-flow baffles too large. 10. no air vents. 11. no safetyrelief inside blocked pipes.12. baffle spacing and baffle window area inconsistent togive turbulent flow in both. 13. liquid velocity < 1 m/s. 14. baffle loose. 15. didn’taccount for hydrogen concentration in gas. 16. decrease in pH causing corrosion.17. increase in pH causing fouling. 18. DT high; designed for nucleate but film boil-ing occurs. 19. water temperature exceeds 50 �C. 20. temperature cross-over.21. increase in clearance between baffles and shell. 22. no sealing strips betweenbaffle and shell. 23. maldistribution.

Installation: 24. improper fitting of gaskets in headers. 25. headers damaged dur-ing installation. 26. overpressure during trial tests. 27. garbage left in lines, tubes,shell. 28. lines installed backwards with wrong fluid on the shell side. 29. installedwith cocurrent flow when design was countercurrent. 30. condenser installed verti-cally when design said horizontal.

Operation: 31. tubes fouled inside. 32. outside tubes fouled. 33. air in tubes.34. air in steam. 35. air not bled. 36. superheated steam. 37. change in hydrogenconcentration in gas. 38. shift in DT in boiler causing film boiling. 39. wet and dirtysteam. 40. steam trap fault. 41. insulation damaged or wet. 42. decrease in flowrates.43. increase in flowrates.

. Plate exchangers

Instruments: 1. faulty readings.Design: 2. under vacuum. 3. control valves on exit lines. 4. temperature > 120 �C;

5. pressure > 2.5 MPa.Operation: 6. temperature too high. 7. temperature spike. 8. cold fluid stopped

but hot fluid continues. 9. superheated steam. 10. under vacuum.

575


. Distillation and absorption columns

Instruments: 1. faulty readings. 2. faulty samples.Control: 3. improper choice (two columns in series on level control; controlling

both top and bottom on temperature). 4. pressure throttling the wrong side of theinstrument. 5. faulty placement of sensor.

Design: 6. trays not held down properly. 7. inadequate support-bearing area fortrays. 8. trays not level. 9. insufficient provision for flashing feed. 10. pump prob-lems (see Pumps). 11. inadequate venting of condenser/ reboiler. 12. gas pockets inliquid lines. 13. insufficient pressure equalization to pump out/ drain system.14. reflux subcooled. 15. feed and reflux liquid incorrectly introduced to column (atwrong location or not into a downcomer). 16. access holes aligned with down-comers. 17. didn’t account for polymerization or crystallization within column.18. series of columns with conditions shifting from oxidation to reduction. 19. wrongmaterials of construction (because, for example, prelim bench tests didn’t includetrace contaminants). 20. incorrect allowance for parasitic reflux in uninsulated col-umns. 21. didn’t seal downcomers during startup. 22. didn’t allow for drainage fromthe seal well. 23. inadequate allowance for thermal expansion. 24. trays not designedfor liquid loading, only vapor loading. 25. didn’t account for surface phenomenacontributing to foaming, wetting and Marangoni instabilities. 26. no splash baffle.27. didn’t use split flow when needed. 28. vortex breakers missing.

Installation: 29. Poor-quality construction (edges rolled when too cold and cracksconcealed by filling with lead, which then contaminated the system). 30. sieve holesbigger than design. 31. trays not installed correctly (not level, bent, poor seals fordowncomers, not held down). 32. junk left behind in column. 33. bubble caps loose.

Operation: 34. high superheat in live steam. 35. high superheat in steam to reboi-ler. 36. pump problems. 37. reboiler problems. 38. condenser problems. 39. pluggedvapor lines, downcomers and/or lines. 40. cannot get downcomers to seal. 41. can-not get trays to drain at shutdown. 42. variable feedrate or composition. 43. foulingof trays. 44. polymerization products inside column. 45. wrong feed tray. 46. insuffi-cient reflux. 47. cycling. 48. boilup rate below tray stability limit. 49. bumping oftrays because of flashing liquid after upset. 50. flooding (plate collapse, excessivevapor rate, excessive liquid rate, low surface tension, foam stabilizing agents, parti-cles, pH). 51. changes in utilities. 52. shift in equilibrium. 53. components in feed.54. vapor locks in fluid lines (caused, for example, by steam tracing). 55. excessiveventing of process materials. 56. excessive liquid entrainment from the top tray.57. maldistribution on trays. 58. vortex breakers not working. 59. syphoning indown pipes.

. Sedimentation centrifuge

Installation: 1. flexible piping not used. 2. misalignment. 3. conveyor bowl notbalanced. 4. not level. 5. liquid dams not set alike or set incorrectly.

Operation: 6. blown fuse. 7. overload relays tripped. 8. motor overheated. 9. bro-ken shear pin. 10. lube oil flowswitch tripped. 11. broken isolators. 12. motor onflexible mounts. 13. motor bolts loose. 14. bearing failure. 15. damaged plows.

576


16. damaged conveyor hub.17. solid product buildup in conveyor hub. 18. conveyorflights worn or portion of blade missing. 19. trunions cracked or broken. 20. con-veyor bowl cracked or broken. 21. leaking effluent weirs. 22. plugged solids in theeffluent hopper. 23. feed temperature too low. 24. feed rate too high. 25. effluenthopper not vented. 26. solids concentration too high. 27. foreign material stuck inbowl. 28. worn bowl strips. 29. loose or broken trunion bolts. 30. bowl inadequatelywashed. 31. clearance too large for blade tip to bowl wall. 32. bowl inside rough.33. wrong size shear pin. 34. no strip installed. 35. interaction with other batchwiseprocess in the system.

577

579

Relating cause fi symptom.Some example causes are listed, for selected pieces of equipment, in Appendix H.

In this section, we list a range of symptoms (readings, sounds, visuals, sampleresults, behaviors) and then, for selected causes, we give the symptoms.

. Centrifugal pump

Symptoms: A: no liquid delivered. B: insufficient capacity; flow lower thanexpected. C: intermittent operation; loses prime shortly after start. D: insufficientdischarge pressure; pressure less than expected. E: short bearing life. F: short me-chanical-seal life. G: vibration and noise. H. power demand excessive; higher thanexpected. I: pump overheats and or seizes. J: stuffing box leaks excessively, K: “crack-ling” noise.

From Appendix H, for centrifugal pumps, the cause is listed in bold face followedby a letter code for the likely symptoms.

Cause ’5 (available NPSH is too low or suction lift is too high): A, B, C, G (I) K;cause ’6: A, B, C, G, (I) K. 7: A, B, C, G, (I) K. 8: A, B, C, G, ?K. 9. B, D, H. 10: H.12: A. 14: destruction and rupture. 15: G. 16: H, I. 17: G, I. 18: A, B, D, I. 19: E, H,F, G, I. 20: E, G, I. 21: H, F, J, cross-contamination. 22: bearings. 23: H, F. 24: H, F.25: E, G, F, I, J. 26: E, G, F, H, J. 27: (A) B, C, D (G) K. 28: B, H. 29: B, D, G. 33: toofast H; too slow (A) B, D. 34: G. 35: (A) B, D, H. 44: A. 45: B, D, H. 46: B, D. 47: F, I.48: F, I. 59: A, I.

The above form of information is extremely useful for the creation of the hypoth-esis, evidence, action table used in the TS Worksheet.

The information can be presented in another format; namely a symptom ‹causes list.

Thus, for centrifugal pumps,Symptom A could be caused by 5 6, 7, 8, 12 16, (27), (33), (35), 44 59. This list can

be prioritized by listing the causes in the most likely sequence. This is the form ofthe information that is summarized in Chapter 3.

Appendix IFeedback about Symptoms for Selected Causes


581

Skill in trouble shooting is needed throughout your career. Trouble shooting is theapplication of your problem-solving skills in a particular context where: a) safety andeconomics are the major drivers, b) knowledge about how equipment and systemswork is crucial and c) you work in a context of someone else’s turf, namely the pro-cess operator. I hope you have a chance to develop confidence in your skills as anundergraduate student. In this section are given some guidelines, especially for stu-dents, on how to use the material in this book to develop your skill. Section J-1 sug-gests that you start with an overview, before you start your journey.

Then, since defining the problem is often a challenging task, start by working aproblem where the TS worksheet gives a reasonable definition of the problem, Sec-tion J-2. Section J-2 elaborates on the activities and questions and clarifies the style Iused in posing the activities. This information is important background. Details(beyond those given in Chapter 8) that are of particular interest to you are givenhere.

Next, I would become familiar with a variety of problems and approaches byworking through, and reflecting on, the trouble-shooting processes used byMichelle, Pierre, Dave, Saadia and Frank in Chapter 4, as recommended in SectionJ-3.

Sections J-4 and J-5 describe activities of developing your prerequisite skills andthen how to use the Cases of Chapter 8. Finally, Section J-6 describes the context inwhich you will be applying your trouble-shooting skills in the process industry.

J-1Getting Started: Get the Big Picture

The Preface gives you an overview of the style and goals of this book. Proceed toChapter 1 and the first, recommended activity that is the pretest in Chapter 1, Sec-tion 1.3. This helps you identify your starting skills in the six areas. You can use thisto select activities to bolster your skills in areas needing attention. You can also retestyourself after you have worked about ten cases, and then after twenty and so on.

Appendix JGuide for Students: How You Can Get the Most from this Book


Appendix J Guide for Students: How You Can Get the Most from this Book

J-2Try a Trouble-Shooting Case where the Problem is Reasonably Well Defined

Next, read the Trouble-Shooting approaches suggested in Chapter 2 and scanthrough the worksheet for Case’8, given in Chapter 2. This is a relatively challeng-ing problem, rated 6 out of 10. Once you have read over the worksheet, I would goto Chapter 8 and try to solve the case by selecting the actions for Case’8. For eachcase the actions are organized approximately in the sequence you might select theaction, although you can jump around as you see fit.

Let’s take some time to briefly look at the possible actions listed with each case.For all actions, the two most important criteria are safety and economics.The safety is considered first for each case.To illustrate the economic impact of each activity, for many of the cases I have giv-

en a “cost” for the activity. Unless otherwise specified, I have used $600/h as a costfor loss in production. To this needs to be added the cost of the people and equip-ment needed.

For example, to obtain an answer to “what’s the weather today and in the past?”costs $50, representing about a 10-minute activity. One could argue that it only takesless than a minute to look outside and get the weather, but the I use 10 minute torepresent actually locating a measurement of the temperature and some reflectionabout the weather over the week. This might require a phone call, information fromthe newspaper or the web.

Another example might be the sampling and analysis. The elapsed time I esti-mate to be about 6 hours. Arrangements have to be made to have someone (trainedin gathering samples) coming to the site, locating the sampling line, flushing theline, taking the sample and returning the sample to the lab. The lab has its ownroutine for completing the regular analyses. This “new” sample is non-routine andwill have to be fit in when possible. In addition, the laboratory will charge the unitfor the “additional” analysis. Hence 6 h � $600/h plus $100 per analysis= $3700.The more complex the analysis, the higher the analytical charges. Some analyseswill have to be done off-site because the lab does not have the facilities. For the morecomplex tests and the use of off-site labs, the cost per sample could escalate to oneday delay and a sample cost of $300 for a total of about $15 000/sample.

For each case I tried to assign reasonable costs.Now consider the actions themselves. The actions are in four sections. The first

actions relate to safety! The second set of actions help you understand the back-ground. The third set are factual gathering of information about the process as itoperates now.

By this time you have information to complete the TS Worksheet. This will thenguide the fourth set that are various tests and actions to test hypotheses, identify thepossible fault, correct and prevent the fault from reoccurring. Throughout all actionssafety and economic are the driving forces.

Here are more details of the four sets of actions that you may select.

582

J-2 Try a Trouble-Shooting Case where the Problem is Reasonably Well Defined

The first set of actions

The first priority is safety: recognize hazards (via MSDS) and take action.

. MSDS

Usually all engineers working on any site should know this information already.That’s one of the first things you learn for the processes with which you are working.If you don’t know the hazardous properties of the materials on the plant, then checkGoogle, the MSDS sheets for the plant or safety manuals. Here, I usually giveMSDS material that I have downloaded from manufacturer’s sites. Sometimes thekey information is the NFPA ratings. The National Fire Protection Association,NFPA, has assigned ratings from 0 to 4 for health, fire and spontaneous reaction/explosion for individual chemicals. Thus an NFPA rating of 0, 4, 0 would mean thatthe species is not an issue for health, it is extremely flammable and it is stable. Forexample, hydrogen cyanide is 4, 4, 0; water is 0, 0, 0. Nitroglycerine is 2, 1, 4. TheNFPA ratings apply to individual species. I tried to include information about howone chemical might react with another when this is important.

. Immediate action for safety and hazard elimination. The first decision relatesto safety. The initial evidence might suggest a health, a fire or an explosivehazard. Act now! There are four options:– Continue with the trouble shooting without implementing actions related to

safety: there is no hazard.– Put on safe-park: this keeps the process going but under conditions that

are safe. This could mean isolating a distillation column and keeping iton total reflux; or reducing the throughput to conditions that previouslydid not pose a hazard.

– Safety interlock shut down, SIS: this should happen automatically if thecontrol system has been designed correctly with the four levels ofresponse expected. However, sometimes this has to be actively initiated.This gives you a chance to reflect on the situation and decide if thisshould have happened.

– SIS plus evacuation: the SIS should happen automatically if the controlsystem has been designed with the four levels expected. Now, because ofthe hazard posed we should add evacuation.

The second set of actions: background

These are mental or office actions to try to put the problem into perspective andstart the process of defining the trouble.

. More about the process. For some problems, fairly detailed knowledge aboutthe process (beyond that given on the diagram or in the problem description)might help you feel more comfortable working on the case. I assume thatyou know a reasonable amount about the process described in the case. Ifmore knowledge might be helpful, then I have included this activity to helpbring everyone to a comfortable level of basic understanding of the process.For some cases, this question is not included. For those where it is included,

583


you probably don’t need this additional information; but if you need somereassurance about what the process is about, this additional informationmight help. I have not included key information that is needed to solve thecase in this activity.

. IS and IS NOT: (based on given problem statement). This is an excellent wayto systematically consider the starting information that is given in the prob-lem statement. You can complete this on your own, without incurring a cost.These are included here a) to remind you to do this; b) give you feedbackabout how you did the task.– What?– When?– Who?– Where?– Sometimes the Who? is not that helpful; it depends on the case.

. Why? Why? Why?

The Why? Why? Why? may be helpful to put some of the cases in the larger con-text. You can do this activity on your own and use this question to give you feedback.

. Weather Today and past. All cases refer to weather conditions in Ontariowhere there are four seasons with snow and freezing weather Decemberthrough March; hot humid summers June to August. Why is the weatherimportant to know? Cold weather gives the potential for freezing of steamtraps (float type); the need for steam tracing of lines to be turned on; colderwater from the cooling tower, river or wells. Storms mean the local atmo-spheric pressure is low – and implies that pumps from sumps (open to theatmosphere) might experience NPSH problems. Damp weather can meanthat particles in hoppers might clump together. Lightning might be givingelectrical interference with the instrumentation.

The weather may have a dramatic effect on the quality of the water. As Doug Pear-son says, “A major storm (local or distant) could, through excessive run-off, affectthe quality of water: excess turbidity, debris, sediments and changes in temperatureof pH.” In spring the water usually is turbid and may have excessive amounts ofhumic acid. If this water, containing high amounts of humic acid, is used to gener-ate steam, the steam may contain ammonia.

Equipment that works in January may not work in August simply because the at-mospheric temperatures are so different.

. Maintenance: turnaround. Three conditions might apply; new plant startup,startup of an existing plant that has just been through its annual turnaroundand operation after some maintenance has been done.

. That the problem is with a first time or startup of a new process is usually givenin the statement of the problem. Therefore, no separate question is posedrelated to this. Startup of a new process triggers faults in design, constructionwaste left in the plant, pretest air and water left in the lines, faulty operating

584


instructions and operators inexperienced with the subtleties of this particularprocess.

. The Maintenance: turnaround activity relates to startups after the annual turn-around. During the turnaround the minimum is usually– inspection of most pieces of equipment (such as sending someone down

through the central part of a distillation column to check on the level andcondition of the trays),

– replacing worn parts,– the installation of changes to the process, repiping, and changing the

operation to implement ideas to optimize or improve operation,– checking the calibration of sensors,– cleaning exchangers,– changing the catalyst.It is important to know “When and what done?”

. Maintenance: routine. During routine maintenance something will havechanged. It is important to know what and the extent. It could be that all theisolation valves were not correctly opened after the maintenance was com-pleted; that the key was not correctly fitted into the drive shaft; that the pumpwas not primed.When and what done? The answers to this question will not admit to suchmistakes. However, this will open your mind to possible things to check andlook at.


The experienced engineer will have already internalized knowledge about whatshould be happening on the process before he/she encounters a case. However, as astudent it is wise to gather this background information before jumping headlonginto the case. The general sources include the simulation/design computer back-ground; records of the design; the collection of information from the equipmentvendors and any internal reports on past tests, or trials done on the plant.

Some values of key properties might be useful to know. These would be availablefrom a Handbook, from Google or from internal files. You might peruse your trou-ble-shooting files that give symptom ‹ cause information.

Based on the information you have already, you often can do some simple checksand calculations.

Here are the details of each of these actions.

Design and simulation files (allowances made for fouling, overdesign and uncer-tainties). Here I tried to list all the possible pieces of equipment involved in the case.All equipment are designed according to the Codes and Standards; the informationgiven here are decisions within a designer’s judgement.

Vendor files. This is the practical information and some specifications from sup-pliers of heat exchangers, steam ejectors, pumps and equipment that would be pur-

585


chased from a vendor. The information may be slightly different from the informa-tion in the design files.

Commissioning data, P&ID, internal reports. Often new plants are constructedunder contract. The contract may include penalty clauses stating a financial penaltythat is paid if the process does not perform up to specification within a stated periodof time. For insurance and governmental regulations, certain performance stan-dards have to be met. Usually records are kept of the performance trials. For thestartup of a new plant, such records are being developed. For processes that havebeen operating for a while, the startup and internal reports and the P&ID diagrammay still be pertinent. They may or may not be available. In some cases, the opera-tion has changed so much that the information is not helpful.

Handbook. The data given here are usually physical and thermal properties of thechemicals in the case: vapor pressure–temperatures, steam tables, thermal proper-ties. Such information may be needed in calculations and estimates you do.

Trouble-shooting files. Here I refer to specific sections of Chapter 3 that pertain tothe equipment in the case.

. Calculations and estimations (that can be done in the office before specialtests are done) The calculations reported here are simple and not ones thatdepend on numbers other than those given in the case statement or that yourecall from rules of thumb. (Later you might do more improved calculationsfrom the data you gather.) The given information varies from Case to Case.Nevertheless, some simple checks and calculations can provide neat insight.These include:– Pressure profile: in most cases the pressures on either side of a barrier

are given and you can state the direction of flow that would incur if thereis a leak in the barrier. For example, 1 MPa steam boiling hydrocarbon ina column at pressure of 0.6 MPa; then the steam would leak into thehydrocarbon.

– Mass balance: this is an important fundamental to help us understandwhat is going on. However, usually at the start of a Case we do not haveenough information to do the calculations. We will be able to do a massbalance later after we have gathered data.

– Energy balance: sink= source; the heat lost by one fluid= heat gained bythe other. Often we have enough information to allow us to do this check.

– Thermodynamics: we can often use the principles of thermodynamics topredict trends, in general.

– Rate: we can use DTs to estimate whether nucleate or film boiling mightpredominate. Similarly, we might be able to estimate rates of mass andheat transfer.

– Equipment performance: we might be able to estimate the performanceof equipment; usually, however, we need more information than thatgiven in the Case statement.

586


The third set are factual gathering of information about the process as it operates now.


Whenever you go out to the plant, you must realize that you are going onto theoperator’s territory. You need to check in with the operator first. Unless youarranged over the phone to meet the operator at the piece of equipment, you will goto the control room first and check in. You might scan the data reported in the con-trol room, discuss the operating actions with the operator and explore the operatingprocedures.

Visit control room: control-room data: values now and from past records. Here wecan obtain values that the instruments read and look at past records.

Process operators; You can ask for information about what happened (from theirperspective) and they might offer ideas as to what is the fault. The operators areusually a very valuable source of information.

. Operating procedures Knowledge of the usual operating procedure to beused for this condition might help.

By this time you have information to complete the TS Worksheet. If you are work-ing on Case 3 4 5 or 8 where the TS Worksheet is published in the book, then youmight start selecting actions from this point onwards.

. Check with colleagues about hypotheses

In practice, you may not have a chance to check with someone else about yourhypotheses. However, in this book and series of actions, you might elect to get somefeedback about your list of hypotheses that you wrote on your TS worksheet. As youdevelop your confidence you will no longer want to get the feedback. Indeed, I haveelected not to give this feedback for cases after Case’29.

You may wish to create a symptom ‹ cause diagram similar to the one illustratedin Figure 6-2, in section 6.5.5c.

The fourth set are various tests and actions selected to test hypotheses/or correctBy this time, you have identified more than seven hypotheses as to the fault and

probably created your symptom ‹ cause diagram.Apply a strategy of simple tests.Inexperienced trouble shooters tend to:

1. Become fixed on “it’s got to be this” and want to take a corrective action imme-diately. Although these tend to be dominant J stylists (described in Section6.1.3.3, section a) inexperienced trouble shooters also often demonstrate thispreference. For these people, I have included Take “corrective” action listed atthe end. Sometimes this does correct the fault. Sometimes it doesn’t. Iemphasize that the section called Take “corrective” action is not often the so-lution to the problem.

2. Immediately open and inspect a piece of equipment. This is usually not anearly step. Opening and inspecting equipment costs usually more than$4000. There are many, many very simple tests you can do to help you iden-

587


tify whether or not open and inspect is needed; and to identify which equip-ment to open and inspect.

Work systematically. Start simply.

Check for consistency as a check of the sensorsSince sensors may be incorrect, one of the first checks might be to look for consis-tency between sensors reading the same variable on the same stream. In the formatof this text, I rarely include a section called “consistency tests”. I expect you to iden-tify neat ways to check for consistency. Methods you can use include:

1) to compare two sensors at the same location,2) agreement between composition, temperature and pressure (say at the top or

bottom of a distillation column),3) agreement between temperature and pressure on a pressure–enthalpy dia-

gram for pure refrigerant,4) to check that the conditions on a stream are the same at two locations. This

latter type of information we often obtain by contacting the operators of theutilities or of the plants upstream or downstream from our process.

Other actions you can select are:

. visit the site,

. check that your P&ID agrees with reality,

. do on-site simple tests,

. gather data and do calculations,

. checking the sensors and the control system,

. get information from vendors,

. sample and analyze,

. do more complicated tests,

. open and inspect, and

. take corrective action.

Here are the details.Use fundamentals to guide the selection of the information you gather. I do not

include an activity called “Calculate fundamentals”. You are expected to gather theinformation needed and perform these. However, to make life easier, I do identifythe information gathered and include the results of the calculations.

Once you have zeroed in on the fault you might need to open and inspect. Or, youmight want to take corrective action. Here are more details of each.

Check for consistency through Contact with on-site specialists. These mightinclude operators running the various utilities (steam, cooling water, refrigeration,power, waste water, flare, storage) and those running various processes that interactwith your unit. The latter are particularly useful. They help provide “consistency”tests for data. In other words, the reading on the temperature sensor of a streamleaving your unit should agree closely with the temperature recorded for the same

588


stream when it reaches another unit (unless additional processing occurred on thatstream between units).

The process-control specialists are contacted separately.Visit site, read present values, observe and sense. Walking out of the control room

and visiting the actual equipment can provide great insight simply from our senses:the noises, smells and sensors. The focus here is on the senses and reading instru-ments whose values are not recorded in the control room. (The actions do notinclude actual adjusting valves, measuring temperatures, and simple tests.)

The sounds to listen for include pump cavitation, pressure-relief valves “blowing”.The level of liquid in a vessel can sometimes be sensed by tapping on the side of thevessel and listening for the change in sound (consistent with the gas–liquid inter-face).

The sights to look for include the values of the instruments, rust stains on the out-side of overhead condensers (indicating that in the past, water flowed over the out-side to try to improve condensation), the condition of the flare, the intermittency ofthe steam discharge from traps, the condition of the insulation, whether steam linescome off the top of the main (the steam tends to be wet and contain rust if it comesoff the bottom of the main), whether the condensate lines entering the condensateheader go in the top (bottom entry means that the traps are flooded with condensatefrom other locations on the site), the location of the valve stem, the signal to thevalve. The sights also include checking that the motor shaft is rotating in the direc-tion indicated on the pump casing; that the direction of flow through the valve is thesame as marked on the valve, that the tab on the orifice plate suggests that the sharpedge of the orifice is on the upstream direction and the orifice is the correct diame-ter.

The smells include anything unexpected; hot smell (suggesting an overheatedmotor); hydrocarbon smell suggesting a leak.

Check diagram and P&ID versus what’s out on the plant. You should have donethis before trouble occurred. As soon as you are assigned to a unit, spend the timecreating a record of all the lines and layout on the plant. You should be able to iden-tify what is in every line and state the direction of flow. Many P&IDs in the office areout-of-date. You need an accurate idea of the actual layout and interconnects. For thepurposes of the cases, I provide that information, as it pertains to most of yourpotential hypotheses, in this activity. This is included here, instead of earlier under“office” actions because regardless of how up-to-date you think your information is,it is always wise to check on the plant at the time the trouble-shooting case arises.

On-site simple tests. This is usually the bread and butter activity. You might use abucket of cold water and measure the amount of condensate. You might ask for“turn and seal” tests on valves to check that they seem to be working correctly. Soaptests on flanges can be used to detect leaks from pressurized systems. Leaks in avacuum systems may be identified by isolating the system and noting the change invacuum over time. Couplings can be checked for wear and misalignment. For eachcase, I have tried to include options that might be useful to check a wide variety ofhypothetical causes.

589


Trend checks. Look for variables that are linked by fundamentals and check thatthey behave accordingly. For example, for distillation an increase in pressure usuallycorresponds to an increase in temperature for constant composition. If the processis cycling, ask questions about the frequency and amplitude. Look for patterns. Foreach case, I have tried to include options that might be helpful.

Gathering data for equipment performance calculations. The focus here is ongathering the information needed to apply the fundamentals: to complete a massbalance, to rate a heat exchanger, and to check on the pressure profile. For eachcase, I have tried to include options that might be useful to check a wide variety ofhypothetical causes.

Sensors: check response to change. Before an instrument is replaced, two simpletests can be done: 1) check the response to change and 2) check on calibration.When conditions are changed slightly, the sensors should respond accordingly. Dothey? If they do not, they might be broken. Secondly, the sensor might respond tochange but the numerical reading might not be accurate. Accuracy is checked viacomparing with measurements from temporary instruments or via calibration(mentioned later).

Sensors: use of temporary instruments. Laser or contact temperature sensors canoften add confirmation to the temperatures at key locations in the plant. Since steamtraps often cause problems, the distinctive noise of traps in operation (heardthrough a stethoscope) can often help identify the problem. Clamp-on ammeters,and power-factor meters are useful.

Sensors: calibrate. If the sensor is suspect, often it can still be used if it is recali-brated. This three to four hour task might be as simple as submerging the thermo-couple is ice water. Specialists are usually required.

Control system. If a control problem is suspected, putting the system on manualis a good tactic to use. Sometimes retuning the control system corrects the problem.

Call to vendors, licensee, or suppliers. A phone call may often provide valuableinsight about a catalyst, a new source of raw material, or trouble-shooting experiencefrom a vendor. The difficulty often is that the key person who may have the informa-tion is not available when you need them. The cost allocated to this activity variesarbitrarily from case to case to account for the uncertainty of getting answers whenyou need them.

Samples and measurements. This sounds easy. However, for cycling or oscillatingsystem the challenge is to obtain samples that are going to help identify the cause ofthe cycling.

More complicated tests. More complicated tests include gamma scans; pressuretesting exchanger tubes for leaks.

Open and inspect. Isolate the equipment, make it safe to open. Open and inspect.This is expensive and usually a last resort.

. Take “corrective” action. Sometimes this corrects the fault. Sometimes itdoesn’t. I emphasize that the section called Take “corrective” action is notoften the solution to the problem.

This ends the list of possible actions. Now, what about the actual fault?

590

J-6 Trouble Shooting on the Job

The fault and closure for the case. Identifying the fault, correcting the fault andpreventing the fault from reoccurring are left to you. There usually are no diagnostictests listed for the case that tells you the fault. There is sufficient evidence from theactions to help you identify the fault.

Write out the fault, think about how you would correct and prevent the fault andthen check with the feedback in Appendix E.

Next, complete the form about the TS process, given in Chapter 2, Table 2-3.Reflect on the thinking process you used, set goals for improvement for the nextcase.

J-3See How Others Handle a Case

Then I would read the five cases at the end of Chapter 1, Cases’3 to 7, and workthrough the stories of the five trouble shooters given in Chapter 4. Spend time withthe reflective pauses given in each story. Compare how you approached the situationwith the approach taken by Michelle, Pierre, Dave, Saadia and Frank.

J-4Pause, Reflect on the Pretest, and Invest Time Polishing Specific Skills

Now that you are familiar with the process and have tried six, guided actions, reflecton the results of the pretest, Section J-1, identify any specific prerequisite skills youwant to polish and invest time developing your confidence with that skill beforetackling the structured set of Cases in Chapter 8.

J-5Work your First Cases Starting with Case’19

The simplest cases start with Case’19. Use the coding related to difficulty to guideyou. Select processes with which you are more familiar. Start simply.

J-6Trouble Shooting on the Job

In this book the actions are selected and answers are given. However, it is neverstated who actually does the tests. On the job the engineer rarely does any of thetests. The engineer asks questions, requests that tests be done, interprets the resultsand proposes actions, but rarely does the engineer do such things as turn valves orchange set points. The cardinal rule is that the process site is the responsibility ofthe process operator. This is their turf. Engineers are expected to check in with the

591


process operators before they go out and look at equipment. Engineers do not altervalves; the operators do.

Building and developing trust with the process operators is crucial for successfultrouble shooting.

J-7Summary

Here is provided a simplified guide, for students, on how to use this book. Start bygetting the big picture and self-assessing your prerequisite skills. Then start simply;try trouble shooting Case’8 for which a TS Worksheet has been completed. Detailsare given of the rationale and structure used in developing the cases and for present-ing the optional actions available for each case. Then, reflect on how your trouble-shooting processes compare with approaches of others by working through the fivecase stories in Chapter 4.

Polish the prerequisite skills and then start simply and work through the cases inChapter 8.

Finally, in working the cases in this book details were not given as to who per-formed the tests. Indeed, it is incorrectly inferred that the engineering trouble shoo-ter did the tests. Rarely do process engineers actually do the tests; tests are done bythe process operators or by specialists.

592

Trouble-Shooter’s Worksheet 2-1: Succinct summary(D. R. Woods and T. E. Marlin)

1. Engage: Gather initial information.

. Establish if emergency priority: safety? damage? shut down or safe-parkor continue?

. Describe what’s going on.

. Manage panic: “I want to and I can.”


2. Define the stated problem: based on given information. If the information isnot known at the stage, gather it later.

IS IS NOT

WHAT (should be happening but it is not)

WHEN ?&?&

WHERE ?&?&

WHO ?&?&

Identify situation as 1) startup new process; 2) startup after maintenance orchange, 3) usual operation. Monitor: Have you finished this stage? Can youcheck? What next?

3. Explore: Exercise? or problem? Strategy for change or basics? Useful tobroaden withWhy? Why? Why?

Gather information. Perspectives: customers? suppliers? weather? changedeconomics? politics? environment?

. Prioritize: product quality, production rate or profit?

. Goal: safe-park? short term? long term? SMARTS$

. Data consistency? Pertinent fundamentals? Likelihood of problem type.

. Explore with What if?

. List changes made and/or list trouble-shooting experience: root causesbased on symptom. (Chapter 3)


. Brainstorm hypotheses

. Hypotheses and evidence of symptoms:

Evidence of symptoms:a.

b.

c.

d.

e.


a b c d e A B C D

1

2

3

4

5

S = supports; D = disproves, N = neutral

Diagnostic actions:A.

B.

C.

D.

4. Plan5. Do it6. Look back


Worksheet 2-2: Summary observation form for feedback about Trouble Shooting.

TS name _____________________ Case _____ Initials ES _____ Obs ___Rough work area:Process: how Data/analysis: whatMonitoring _____________________ Data resolution _______________Checking _______________________ Fundamentals? _______________Systematic ______________________ Reasoning ___________________Subs and perspective _____________ Completeness ________________

Decision making: how Synthesis: whatPriorities ______________________ Hypotheses __________________Bias ___________________________ Flexibility ____________________

Rating and Feedback

Clarity of Communication

MostSomeNone All

Process used:

MostSomeNone All

Data collections and analysis:

MostSomeNone All

Synthesis:

MostSomeNone All

Decision-making:

MostSomeNone All

Five Strengths:________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Two areas for improvement________________________________________________________________________________________________________________________________


Trouble-Shooter’s Worksheet 2-3: Case’8: The depropanizer: the temperaturesgo crazy (Donald R. Woods and Thomas E. Marlin)


. Emergency priority: Safety? Hazard? Equipment damage?shut down &; safe park &. If not &, then:

. Draw a sketch of the process and mark on values. Provide a descriptionin words of what is going on. This is a simple distillation column but allthe piping and instrumentation details make it look complex. On theLHS, feed from upstream processing enters drum V29. This feed ispumped (via either a steam driven or motor driven centrifugal pump,F25, 26) through a preheater, E24, and into the depropanizer at tray 18.As the name suggests, the purpose of this column is to take overhead“propane and all lighter species”. Let’s follow the overhead. The overheadis condensed in two condensers in series, E25, collected in overheaddrum V-30 with the non-condensibles (such as methane and hydrogen)removed from the drum and vented to the fuel-gas system. The pressureon the column, C8, is controlled by the valve on the vent system, PV 10.Condensed propane is pumped, F-27, from the drum V-30 forward asproduct, through product cooler E-26, and returned to the column asreflux. The reflux is flow controlled. Following the bottoms: a thermosy-phon reboiler is steam heated. The bottoms flows forward to the nextcolumn, the debutanizer. No pump is needed because of the pressuredifference between the depropanizer, 1.7 MPa, and the debutanizer, 0.48MPa. I’m not sure at this stage if this is a control “problem” so I won’telaborate further on the system at this time. I also will focus on thedepropanizer, and not explore the debutanizer at this time.


. Monitor: Have you finished this stage? Can you check? What next? I’vesystematically followed my way around the flow diagram. I think Iunderstand enough for now.

2. Define the stated problem: Systematically classify the given information usingIS and IS NOT. If the information is not known at the stage check? & toremind you to gather this information


IS IS NOT

WHAT tray temperatures “go crazy”;bottoms level decreases.

(should be happening but it’s not)tray temperatures steady; bottoms level steady.

WHEN ?& 10 minutes after thepressure in column C8increased by 0.1 MPa.

? & before the pressure increase; running wellfor several shifts. First plant startup.

WHERE ?& depropanizer, C8. ? &� maybe upstream; no information aboutdownstream debutanizer, yet!

WHO ?& new operator. ? & not with previous operators.



. Monitor: Have you finished this stage? Can you check? What next? Yes. Ithink I’ve finished.

3. Explore: Gather information to be gathered ? & in Define stage &. Perhapsquestions about downstream effects. The decrease in liquid level could bebecause the flow has increased to the debutanizer (because the Dp) hasincreased.

Exercise? & or a problem? &�. I haven’t seen anything like this beforeStrategy: change & or basics &�.Perspectives. Why? Why? Why? no information at this time that suggests

this might be useful.____________________________________________________________

Why? ›____________________________________________________________

Why? ›____________________________________________________________

Why? ›____________________________________________________________

Why? ›____________________________________________________________

Why? ›

Startfi _________________________________________________________


. Prioritize: product quality &�; production rate &; profit &

. Goal: safe-park? &�: short term now with long term later &�;long term now &

Action to be achieved: Specific terms and Measurable:level out the temperatures and the bottoms level

Attainable? total reflux is a start but I hope it’s attainableReliable? depends on my short-term solutionTimely? will work on solving it quicklySafe? cannot think of major hazard now$


. Type of problem: startup new process &� maybe mechanical electricalfailureusual operation &�: ambient temp? &maybe fluids problems;high temperature? & then maybe materials problemsSystem? failure of heat exchanger &> rotating equipment &>vessels &> towers &

. Identify key and What if?

What if? temperatures “going crazy” = temperature cyclingthen focus on cycling symptoms / causes

What if? only temperature “cycling” and no decrease in levelthen bottoms and tops temperature and pressures should be

cycling tooWhat if? only bottoms level drop and no temperature “going crazy”

then root cause related to bottoms level dropWhat if? column pressure increases

then condensation temperature at top increases; DT condenserincreases and condensation should be easier; boiling tem-perature at bottoms increases; DT reboiler decreases somight shift from film to nucleate boiling giving higher heatflux, causing increased boilup or if nucleate to start withthen insufficient area and boilup decreases.

. List changes made & and/or trouble-shooting causes based on symp-tom &�.

“Crazy temperatures” and decreasing bottoms level sounds like a control prob-lem.

From Chapter 3, no symptoms listed for bottoms level dropping, but symp-toms related to “oversized condenser” and “undersized reboiler” are:


“Insufficient boilup”: [ fouling on process side]*/ condensate flooding, seesteam trap malfunction, Section 3.5 including higher pressure in the conden-sate header/ inadequate heat supply, steam valve closed, superheated steam/boiling point elevation of the bottoms/ inert blanketing/ film boiling/ increasein pressure for the process side/ feed richer in the higher boiling components/undersized reboiler/ control system fault/ for distillation, overdesigned conden-ser.

But if there was “insufficient boilup”, then the bottoms level should beincreasing and not decreasing. This doesn’t make sense??

For:“Cycling of column temperatures:” controller fault/ [ jet flooding]*/ [downcomer

flooding]*/ [ foaming]*/ [dry trays]* with each of the []* items listed as separatesymptoms with their own root causes.

[ jet flooding]*: excess loading/ fouled trays/ plugged holes in tray/ restrictedtransfer area/ poor vapor distribution/ wrong introduction of feed fluid/ [ foam-ing]*/ feed temperature too low/ high boilup/ entrainment of liquid because ofexcessive vapor velocity through the trays/water in a hydrocarbon column.

[downcomer flooding]*: excessive liquid load/ restrictions/ inward leaking ofvapor into downcomer/ wrong feed introduction/ poor design of downcomerson bottom trays/ unsealed downcomers/ [ foaming]*

[ foaming]*: surfactants present/ surface tension positive system/ operatingtoo close to the critical temperature and pressure of the species/ dirt and corro-sion solids.

[Dry trays]*: flooded above/ insufficient reflux/ low feedrate/ high boilup /feed temperature too high.

. Brainstorm root causes: summary of major ideas generated:

change in feed, too much overheads in feed, not enough feed, tray collapsedin stripping section, too much vaporized feed, pump F26 failure, pump F26cavitates, increased pressure and DT increases, dry tray, flooded in rectification,insufficient reflux, low feedrate, feed temperature too high, boilup too high, toomuch feed to debutanizer, leak in bottoms, vaporizer flashes 90% (instead of67%), failure of check-valve on idle pump outlet, boilup controller fault.

Some of these are symptoms and not root cause, e.g. “not enough feed” “dry trays”


Symptom a. 10 min after column pressure increased, column temperatures gocrazy

b. 10 min after column pressure increased, bottoms level decreasesc.d.e.



a b c d e A B C D

1. tray collapsed stripping section S S 4

2. too much bottoms fed to debutanizer N S 4

3. too much overheads in feed N S 4

4. feed valve FV1 stuck S S 4

5. pump F-26 not working S S 4

6. check valve on idle pump allows backflow S S 4

7.

Diagnostic actions:

A. readings of instruments on columnB. visit site and listen to pump for cavitationC. visit site and see location of valve stem on FV-1D. shut isolation valves on idle pump

4. PlanSelect “read instruments” as the first task because it is inexpensive and shouldhelp test many of the hypotheses. Many of the key variables are displayed in thecontrol room.

5. Do itGo to the control room, notebook in hand.

6. Look back


Trouble-Shooter’s Worksheet 2-4:(Donald R. Woods and Thomas E. Marlin)


. Emergency priority: Safety? Hazard? Equipment damage?shut down &; safe park &. If not, & then:

. Draw a sketch of the process and mark on values. Provide a descriptionin words of what is going on.



2. Define the stated problem: Systematically classify the given information usingIS and IS NOT. If the information is not known at the stage check ? & toremind you to gather this information

IS IS NOT

WHAT (should be happening but it is not)

WHEN ?&?&

WHERE ?&?&

WHO ?&?&





3. Explore: Gather information to be gathered ? & in Define stage &Exercise? & or a problem? &.Strategy: change & or basics &.Perspectives. Why? Why? Why?

____________________________________________________________Why? ›

____________________________________________________________Why? ›

____________________________________________________________Why? ›

____________________________________________________________Why? ›

____________________________________________________________Why? ›

Startfi _________________________________________________________

. Prioritize: product quality & ; production rate &; profit &

. Goal: safe-park? &: short term now with long term later &;long term now &

Action to be achieved: Specific terms and Measurable: __________________Attainable? ______________________________________________________Reliable? ________________________________________________________Timely? _________________________________________________________Safe? ___________________________________________________________$ _______________________________________________________________


. Likelihood of problem: startup new process &maybe mechanical electrical failureusual operation &: ambient temp? &maybe fluids problems;high temperature? & then maybe materials problemsSystem? failure of heat exchanger &> rotating equipment & >vessels &> towers &

. Identify key and What if?

What if?thenWhat if?then


What if?then

. List changes made & and/or trouble-shooting causes based on symp-tom &.

. Brainstorm root causes:


Symptom a.b.c.d.e.


a b c d e A B C D

1

2

3

4

5

6

7

Diagnostic actions:

A.B.C.D.


Worksheet 3-1: Rating form for teams

Assessment of your team and team meeting Name: _________________Date: __________________

Purpose of team: ______________________________________ unclear &Purpose of this meeting ________________________________ unclear &

Agenda for this meeting: detailed, clear and circulated ahead of time &bare minimum circulated ahead of time &none &

Three-minute team task to seek consensus about the rating of the Task andMorale:

. Teamwork: Task all members clear about and committed to goals; allassume roles willingly; all influence the decisions; know when to dis-band for individual activity; all provide their unique skills; share informa-tion openly; the team is open in seeking input; frank; reflection andbuilding on each other’s information; team believe they can do theimpossible; all are seen as pulling their fair share of the load.



& & & & & & &1 2 3 4 5 6 7

. Teamwork: Morale: Trust high, written communication about any indi-vidual difficulties in meeting commitments; cohesive group; pride inmembership; high esprit de corps; team welcomes conflict and uses meth-odology to resolve conflicts and disagreements; able to flexibly relievetension; sense of pride; we attitude; mutual respect for the seven funda-mental rights of all team members; Absence of contempt, criticism,defensiveness and withdrawal.




& & & & & & &1 2 3 4 5 6 7

Each, in turn, gives a 30-second summary of his/her perception of his/hercontribution. This is presented without discussion.

Individual, 30 second reporting of his/her contribution to this meeting:____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Four-minute team task to reach consensus about the five strengths and the twoareas for growth.

Strengths of your team

Areas to work on for growth

D.R. Woods (2005)


Worksheet 5-1: Reflections: A place for you to record your ideas about the TAPPSMethod:

Being the talker: What did you enjoy most? What was most difficult about thetask? What did you discover by being the talker? What did you discover frominteracting with a listener? What were your strengths? Focus on accuracy? Veryfew silent periods? Being active? Good communication?

________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Being the listener: What did you enjoy most? What was most difficult aboutthe task? What did you discover by being the listener? What did you discoverabout problem solving by comparing the talker’s approach to yours? What wereyour strengths as listener? Quality of your prompts and degree of interaction?Tone of interaction? Good communication? Non-intrusiveness?

________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________


Worksheet 5-5: Evidence for listening: Feedback to listener & the listener willencourage verbalization, an emphasis on accuracy, active thinking and encouragethe problem solver to move the marker correctly on the strategy board. Your inter-ventions will be judged by the problem solver to be helpful, and not judged to bedisruptive.

Activity 1: Talker ____________ Case ___________ Listener _____________

encourage verbalization: not needed interruptive OK really helpedencourage emphasis on accuracy: not needed interruptive OK really helpedencourage active thinking not needed interruptive OK really helpedinterventions: not needed interruptive OK really helped

Comments:______________________________________________________________________________________________________________________________________


Worksheet 5-4: Record of the talker’s strategy with . for monitoring statements.

Talker __________________ Case _________ Listener ____________________

Stage

Engage: “I want to and I can!”

Define-the-stated problem: Sort the givenproblem statement

Explore the problem to discover what theproblem really is

Plan

Do it

Look back: elaborate, check

0 2 4 6 8 10 12

Time, minutes


Worksheet 5-3: Feedback to the listener:

problem _______________________ listener __________________________

I found the listener:

. The quality of the comments:

–10 –8 –6 –4 –2 J –2 –4 –6 –8 –10


. The attitude displayed:

–10 –8 –6 –4 –2 J –2 –4 –6 –8 –10


. The listener’s emphasis was listening to me &; helping me verbalize &;helping me solve the problem&; solving the problem for me&.

validated by talker __________________________________________________


Worksheet 5-2: Feedback from the listener to the talker about the process used.

Awareness ____________________________ ___________________________problem listener

Number of silent periods 0 1 2 3 4 5 >5Number of checks, double checks >5 5 4 3 2 1 0Amount of writing/ charting >5 5 4 3 2 1 0Comments:______________________________________________________________________________________________________________________________________

Validated by: __________________________________


Worksheet 6-1: The Tom Dayton murder (adapted from Sherlock Holmes).

Tom Dayton had many enemies. (a) He was a scalawag and a prankster whonever passed up an opportunity to embarrass someone through a practical joke(b). It was Tom who invented the joy buzzer and the whoopee cushion, andsome even credit him with having originated fake vomit.

It is well known that Tom’s favorite target was his old headmaster, StanleyBosworth, (c) at Bromley School. Stanley Bosworth was the victim of some ofTom’s most elaborate pranks. Tom’s eventual marriage to Bosworth’s daughter,Melissa, was considered by many to be Tom’s ultimate joke (d) on the respectedheadmaster.

Among the more prominent victims of Tom Dayton’s past pranks were JudgeWalter Brighton (e), Lord and Lady Morton (f) of Westchester, banker MortimerFawcett (g), Doctor Fabian Peerpoint (h), and tobacco merchant Dawes Flescher(i).

All of the above were present at the dinner party held on the Bosworth yachtin honor of Stanley Bosworth’s 60th birthday (j).

Following an uneventful dinner, most of the guests retired to their state-rooms to freshen up. The clock in the dining room struck 10 pm (k) when ashot rang out (l). Most later claimed they heard a second shot (m). All aboardthe yacht, including the yacht’s captain, Jonas Fenton, (n) and cook, the curvac-eous Mildred Weekson (o), arrived at Tom Dayton’s stateroom to find him dead– shot in the forehead (p).

A smoking revolver lay near the doorway (q); Tom’s body lay on the flooracross the room (r), just below an open porthole (s). Scrawled in the dust nearDayton’s body were the initials SB (t). In the corner of the room was a suitcasefilled with $500,000 (u). No bullets were found in the walls, ceiling or floor ofTom’s room (v). Who killed Dayton?

During the investigation the following factual evidence was produced:

w. Although just about everyone claim they heard two shots, Tom Daytonhad one bullet in his head.

x. Lady Morton was being blackmailed.y. The clock in the dining room was 15 minutes slow.z. The $500 000 in the suitcase was counterfeit and was accompanied by a

withdrawal slip for $1 000 000 from Mortimer Fawcett’s bank.aa. Bosworth angrily stated at dinner that he felt Tom was mistreating his

daughter.bb. The actual murder weapon was found under water below the porthole.cc. Dr Fabian Peerpoint advocates mercy killing.dd. The smoking revolver in the room belonged to Stanley Bosworth.ee. Tobacco merchant Dawes Flescher is a talented mountain climber.ff. Jonas Fenton, the yacht’s captain, went to school with Tom Dayton.


gg. Dr Fabian Peerpoint revealed that Tom Dayton was terminally ill withonly a few months to live.

hh. The bullet in Tom’s head did not come from the smoking revolver on thefloor in Tom’s room.

ii. Melissa says that she was visiting her father in his stateroom from 9:45pm until she heard the shot.

jj. Mildred Weekson was having a secret affair with Tom Dayton for thepast two years; and with Lord Morton.

kk. The shot that killed Tom Dayton was fired from outside the porthole.Directly below the porthole is water.

Based on the evidence so far would you:

1. accuse ____________ of murder based on evidence (list the letters of theevidence supporting your conclusion) ____________________________.

2. accuse Dr. Peerpoint of mercy killing based on evidence (list the letters)_____________________________________________________________

3. conclude that Tom died from __________ based on evidence.(list the letters) ________________________________________________

or... .4. require that the following information is needed before any conclusion

can be drawn:ll) what pranks had Tom played on _________________________mm) check for poison in Tom’s bodynn) other _______________________________________________


Worksheet 7-1: Trust.

Trust is having the confidence that you can mutually reveal aspects of yourself-and your work without fear of reprisals, embarrassment or publicity.

Trust works both ways: you trust them and they trust you. Trust is not devel-oped overnight, trust takes time to develop. Trust can be destroyed by one incor-rect act.

Check your current status

Building your trustworthiness getting them totrust you

alreadydo this

needssome work

need lotsof work

unsureif this isfor me

1. Do what you say you will do. & & & &2. Be willing to self-disclose: don’t hide

your shortcomings; share yourself-honestly.& & & &

3. Listen carefully to others and reflectto validate your interpretation.

& & & &

4. Understand what really matters toothers; do your best to look out for theirbest interests.

& & & &

5. Ask for feedback. & & & &6. Don’t push others to trust you more

than you trust them.& & & &

7. Don’t confuse “Being a buddy” withtrustworthiness.

& & & &

8. Tell the truth. & & & &9. Keep confidences. & & & &10. Honor and claim the 7 RIGHTS. & & & &11. Don’t embarrass them. & & & &

Checking your trustworthiness do they trust you? always mosttimes

some-times

don’t thinkapplies

1. Do they disclose confidential informationtrusting that you will keep it confidential?

& & & &

2. Do they assign you challenging tasks to dowithout frequently checking up on you?

& & & &

3. Do they honor your RIGHTS? & & & &4. Do they seem to look out for your best

interests?& & & &

5. Honest and forthright. & & & &6. Do not leave you feeling that they haven’t

told you everything about the situation;they seem to be holding back.

& & & &

copyright 1999 �, Donald R. Woods


Worksheet 7-3: Feedback about your environment.

To what extent do you agree with the following descriptors of your environmentwhere you usually “trouble shoot”.

People are willing to admit error: “The people that I work with are veryunwilling to admit errors; they blame others, they pass the buck and, if neces-sary, would purposely mislead me rather than to admit error.”


Encourage risk taking: “Risking is rewarded. We are expected to take risksabout 10 times a day. Risks should be wisely, not indiscriminately selected. But,nevertheless, we are not only encouraged but we are rewarded for risk taking.”


General stress at work they are under: “Their environment is very stressful.People have many deadlines and interruptions. The consequences of makingmistakes is very high. The issues are complex. The environment changes oftenand includes a lot of uncertainty.”


General stress at work you are under: “My environment is not stressful. I donot have many deadlines or interruptions. The consequences of making mis-takes are low. The issues are straightforward. The environment is safe, stableand secure.”



People’s listening and responding: “The people are open, communicate well,can clearly identify facts, will offer opinion when asked for, and are very compe-tent but are not aggressive “know-it-alls””.



Worksheet 7-2: The problem given in Case’12 arises at 8:30 am. The followingbackground and resources are available. Identify which ones you think are directlyunder your control. If you think you have control over the item, circle Y;if you do not have control over it, circle N.

1. The safety officer, Hack, is overbearing, not liked andgets carried away about simple things. Y N

2. The laboratory can analyze liquid samples with their equipmentbut gaseous samples cannot be analyzed because theirinstrument is broken. Y N

3. The lab schedule is busy with top priority analyses.Samples could not be analyzed until after 3:00 pm. Y N

4. The union prevents you from analyzing any samples;if you do, there will be a strike. Y N

5. The upstream styrene plant is operating at 80% capacity. Y N6. The upstream ethylene cracking furnace is operating

at 95% capacity. Y N7. The upstream propylene plant is shut down. Y N8. The operator on the ethylene plant is cooperative. Y N9. Samples can be taken at any of the sewer gates within

the Battery Limits and at A, B, and C. Y N10. No blueprints are available for the sewer system. Y N11. Your performance review to establish your salary is being

done next Thursday at 4:30 pm. Y N12. You have to prepare your “record of progress” record

for the performance review on Thursday. Y N


12/1 rule 4820 min rule 4880/20 rule 2285/15 rule 22

aAbsorption, gas– cases involving 383, 387– causes 576– fundamentals 67– symptoms-cause 73–78Accuracy, need to focus on 20– example check and double check

– in Case’5 141, 142, 143– in Case’6 147, 151

Activated sludge, symptoms-cause 106Active, importance of being mentally active

20– Case’7 158–160– example Case’6 145– to aid in TS 25AAI, Adhesion angle index, defined 108Adsorption, gas, fundamentals 67– Case’6 12–13, 144ff

– cases involving 12, 247, 299, 370– symptoms-cause 77Adsorption, liquid, fundamentals 67– symptoms-cause 77

– cases involving 181, 247–248Aerobic bioreactors, symptoms-cause 96– foaming 129AI, Arching Index– cited 56, 57– defined 108Alkylation, decanters, symptoms-cause 83– reactors, symptoms-cause 99Ambiguity,– in terminology 226– in words or instructions 218

Amine absorbers, causes of corrosion in 46– symptoms-cause 73–75, 129Amine flash drum, symptoms-cause 85Ammonia process, cases involving 1, 171,

317, 339, 357, 383, 387, 391Anaerobic digesters, symptoms-cause 97Analysis (as part of the Define stage) see also

Classification– example use in TS Worksheet for Case’8

36– second stage in a generic problem solving

process 20, 21Anderson, R. B., cited 9Anxiety, and the Engage stage 21Argentino, Mark, cited 291Assertiveness, how to 48Assessment, self, see Self assessment and

FeedbackAssumptions,– and reasoning 224– definition 224, 230– diagram 229– formulate 230Attitude toward problem solving 22Awareness of problem solving process,

importance of, 20– activity to improve your skill, 166–172– feedback form, 172–173– target skills, 166

bBag filters, symptoms-cause 81– cases involving 222, 333Bagging machine, cases involving 222, 333Barton, J., cited 297, 407Basadur, M., cited 180, 407Basics strategy for TS:– contrast with change strategy 23– description 23

Index I 1

Index


– diagnostic actions to help select 197– example questions to ask 24– example use, Case’4 139

– Case’5 141– Case’6 145

– how to apply 24– use of fundamentals in 23– when to use 4, 23Bayes theorem, use 25BDI, Bin density index– cited 56, 58– defined 108Bernoulli’s equation, use 51, 58, 419, 565Bias, personal 25, 205– in collecting evidence 205– in reaching conclusions 205– self test 206– types:

– anchoring 25, 205– availability 205– confirmation 25– inadequate synthesis 206– misinterpretation 206– need to avoid 195, 205– overinterpretation 25– premature closure 25, 205– pseudodiagnosticity 25, 205– representative 25, 205– test 206– underinterpretation 206

Bins, symptoms-cause 126– cases involving 182, 222, 347Blenders, solid, symptoms-cause 108–109– cases involving 333Bloch, H.P. 44, 403, 408Blowers, symptoms-cause 52Boiling film 63– nucleate 63 198 267Brainstorm, see CreativityBubble formation, symptoms-cause 110Bubble reactors symptoms-cause 98Bucket elevators, symptoms-cause 56

cCarry out the plan see Do itCases, overall list of, with ratings 268–271– Case’1 1

– answer 2– Case’2 1


– answer 133–138– critique 212

– reasoning 224–232– evidence 227, 228– meaning 226– diagram of cause-effect 228, 229– classification 224– conclusion 225– context 225– worksheet 423

– Case’4 10– answer 138–140– cited 180– worksheet 427

– Case’5 11– answer 141–144– patterns 223– worksheet 430

– Case’6 12– answer 144–156– worksheet 202, 217, 217–218

– Case’7 13– answer 157–162– cyclical 223– Why? 180

– Case’8 35– actions 271– answer 36–39, 537– destroyers activity 240– listening activity 239

– Case’9 181– actions 274– answer 538– brainstorm 190– identify symptoms 220

– Case’10 186–87– actions 278– answer 538– cyclical 223– example approaches 204– patterns 223

– Case’11 214– actions 282– answer 540– hypotheses 216– root cause 217

– Case’12 247– actions 280– answer 540



IndexI 2

– Case’15 255– answer 33


























– Case’41 354












– answer 563– Other cases 403– Feedback about hypotheses in cases 267

– Rating of cases 262, 263, 537– Selecting cases 263

Cause (of trouble) see alsoHypothesis andrelated terms Cues and Symptoms

– actual cause in case 264, 591– relating cause to magnitude of the extent of

symptoms 43– relating to when: startup of new process,

startup after shutdown, usual operations44

Cause-effect information, see also specificequipment for details

– activity to check consistency 213– activity to improve skill 213– consistency between 213– use of 195Cellular PE, extrusion 120Centrifuges, filtering, symptoms-cause 87– brainstorming 187– cases involving 186Centrifuges, sedimentation, symptoms-

cause 87– causes 576Change strategy for TS:– contrast with basics strategy 23– description 23– example questions to ask 23– when to use 4, 23, 197

Index I 3

Check, see Look back stage in generic strategyCheck and double check, see AccuracyChin, T.G. 409, 534CI, Chute index– cited 56, 57– defined 108Classifying information 219 related term

Analysis– activities to improve skill 219–220– as first step in reasoning 224– example 224– definition 219– how to 219, 399– need for 20, 21, 195, 219, 399– of ideas from brainstorming 220– of starting information 219– of triggering events 220, 399Coagulation, symptoms-cause 111Coalescers, fundamentals 110– symptoms-cause 110–111Coating, symptoms-cause 126Commissioning 266– diagnostic actions 586Communication:– activities to improve skill 237–238– criteria for effective 237– definition 237– preference to interpret 249– self test of skill 8Compressors:– cases involving 1, 291, 304, 317, 350, 391– centrifugal, symptoms-cause 52– frequency of faults 44– MTBF 44– reciprocating, symptoms-cause 52Conclusion (inference, proof or disproof of

hypothesis):– example 228– reasoning process to test validity 223–232– write down 224Condensers, shell and tube, symptoms-cause

62– cases involving 251, 252, 255–256, 312,

317, 321, 323, 370, 374– causes 575Consistency (among evidence):– activities to improve skill 209–228, 227,

399– diagnostic actions to check 267–268,

588–589– examples 199, 227– how to check with evidence 227– importance of, for reasoning 209

– need for 195, 209– to assess reasoning 230– types of

– cause-effects 212, 399– evidence 227, 399– experience with equipment 228, 399– rules of Mathematics and English 217,

399– words/definitions 210–212, 399

Contamination (impurities) andcrystallization 72

Context– definition 224– example 224– how to identify 225– of reasoning 224Control, process:– activities to improve skill 418, 421– four elements of 3– self test about knowledge of 8– symptoms-cause 47Control room: need to visit 198, 267, 589Controller, process, causes 574Conveying, solids, see also Pneumatic

conveying, and feeders– cases involving 182, 186, 222, 333– symptoms-cause 34Corrosion:– general reasons why corrosion becomes a

cause 45– symptom-case data 45– types of corrosion and their frequency 45Costs, see also Financial penalty of diagnostic

action 264 and 582Covey, S., cited 22, 406Cox chart, use 418Cracking, catalytic, reactor, symptoms-cause

101Creativity, and Brainstorming and Hypothesis

generation:– activity to improve your skill 190– and skill in classification of ideas 219– case’7 171– checklist of triggers 183–186– example Case’10 187–190– feedback form 191– need for 20,183– target skills 183– to identify root cause, example on TS

Worksheet for Case’8 38Criteria:– and goals 401, 193– need for must and want 22

IndexI 4

Critical thinking, see also Reasoning, andDiagnostic actions:

– activities to improve skill in consistency209, 399

– list of subskills 195– need for 20– reflections about 400Crystallization, solution: fundamentals 67– cases involving 13–15, 157ff, 186– symptoms-cause 72– vacuum and circulating system, symptoms-

cause 72Cues (for storing experience in Long Term

memory) 20Cues/evidence (for trouble) see also Diagnostic

actions– how to critique of evidence 227, 231:

– check for consistency 227– diagram 228–229– test pertinence of 228

– number of cues used per case 34– suggestions for weighting 25– suggestions to prevent overlooking 25Cycling, related term Surging:– batch cycles 221, 222– symptoms-cause 130, 63–65, 71, 101Cyclone, GS, symptoms-cause 81–82– cases involving 333

dData handling, see also Diagnostic actions:– activities to improve skill 399– and cycling 222– elements of the TS process 33, 398– errors in data 208– interpreting data 208, 249–250– list of detracting and enriching behaviors

30– options for gathering data 196ff– personal bias in 204–208– reflections about your approach 400– self test of skill 5Decanters– cases involving 179– symptoms-cause 82–84Decision making:– elements of the TS process 22, 33– list of detracting and enriching behaviors

32– self test of skill 6Define: second stage in a generic problem

solving process:– and skill in classification 219

– context, how to identify 225– critique use of 163– description 20, 21, 28– example use in TS Worksheet for Case’8

36– for Case’5 141– for Case’6 145– prompts for use in the TS Worksheet 26,

40Dehydration, glycol system, symptoms-cause

73, 74, 75, 76, 129Dehydrogenation, safety 96, 98, 99Demisters– fundamentals 110– symptoms-cause 110Depropanizer:– Case’8 description 35– cases involving 35, 213, 327, 345, 362, 367,

377– P&ID, 34– TS Worksheet for 35Design fault, occurrence 44Design data fault, occurrence 44Desorption, gas, fundamentals 67– cases involving 181, 383, 387– symptoms-cause 75Diagnostic actions– activity to improve skill 201, 202– and cycling 222– chart relating to hypotheses and evidence

on TS Worksheet, seeHypothesis, chart– consistency tests 267–268, 588– costs incurred 264, 582– criteria for selecting 196, 264– equipment to help 203–204– errors in evidence 208– examples for TS Case’8 39, 201– how to select 24, 196, 264, 587– interpreting data 208– options 196:

– recent events 197, 265, 583– safety 196, 264, 583– to correct 200, 264, 268– to help select TS strategy 197– to understand what’s going on 198, 267,

585, 587– to test hypotheses 198, 267, 587

– personal style in selecting 204– reflection about 400– relating to hypotheses 24– site visit 589– time required for selected 204– trends and patterns 590

Index I 5

– using senses 589Differences inconcentration for mass transfer 67– exchange equilibrium for separations 67– molecular geometry for separations 67– partition coefficient for separations 67– pressure for fluid flow 51– solubility for separations 67– temperature for heat transfer 58– vapor pressure for separations 67Distillation column– activities to improve skill 418– cases involving 170, 10, 35, 211, 222, 247,

251, 252, 256, 299, 321, 327, 345, 354, 362,367, 372, 377, 383, 387

– causes 576– frequency of failures of 44, 201– good practice 69– self test of knowledge about 6– symptoms-cause 69–72Do it, fifth stage in a generic problem solving

process,– description 20, 21, 28– example on TS Worksheet for Case’8 39Doig, L 403, 408Drives, see Engines, Motors and Turbines– for extruders 120, 121Dryers, symptoms-cause 85– cases involving 186, 347Dudzic, Mike, cited 286, 543Dutta, S., cited 101

eEdgar, M.D. 9Ejectors, steam:– cases involving 13, 181, 335– symptoms-cause 53–54Elaboration, as part of the Look back stage 20– importance of in problem solving 17Electrical failure, occurrence 44Elonka, S.M. 403, 408Elstein A.S. 24, 405Emulsion, formation of stable, cause 129Energy, conservation of 58Energy, mechanical, fundamentals 58Energy, thermal, seeHeat exchangers, and

furnacesEnergy exchange, fundamentals 58Engage, first stage in a generic problem

solving process– critique use of 163– description 20, 21, 28– example use in the TS Worksheet for

– Case’5 141– Case’6 145– Case’8 35

– prompts for use in the TS Worksheet 26,40

Engines, symptoms-cause 58Entrainment (GL), symptoms-cause 74, 84Entrainment (LL), symptoms-cause 74, 85Environment in which you TS:– impact on your approach 253– self test 254–255Environmental impact of process: self test

about knowledge of 7Equilibrium, phase, use of 67Esso Chemicals, cited 251, 321, 541, 548Esterification, safety 96, 98, 99Ethylene 247, 299Evacuation 196, 265– as diagnostic action 265Evaporators, general, fundamentals 67– cases involving 257, 335– forced circulation, symptoms-cause 68– multiple effect, symptoms-cause 68–69– symptoms-cause 67– vapor recompression, symptoms-cause

68– vertical falling film, symptoms-cause 65,

68Evidence, see CuesExercise solving:– contrast with problem solving 17– definition 17– diagram of 19– example use, Case’7 157– frequency of, for experienced trouble

shooters 17Experience, past, and use in problem solving,

trouble shooting 17Experience with process equipment, see

specific equipment for details– and consistency 218– keeping up-to-date 402–403– need for and importance of 4, 397, 402– reflections about your data base 398– self test of 6, 411– stuff left in lines 433, 584Explore, the third stage in a generic problem

solving process,– and skill in classification of information

219– critique use of 163– description 20, 21, 28– example for Case’4 139

IndexI 6

– example use in TS Worksheet for Case’836

– example use of Why? Why? Why? 180– prompts for use in the TS Worksheet 26,

41Extruder, symptoms-caus see also Reactive

extrusion 120–125– blown film, symptoms-cause 121, 122, 124– cast, symptoms-cause 122– coating, symptoms-cause 122– coating wire and cable, good practice 120– filament, symptoms-cause 122, 124– pipe, symptoms-cause 122– sheet, symptoms-cause 122– single screw, good practice 120– symptoms-cause 123– twin screw, good practice 120– vented twin screw, good practice 120

fFabrication fault, occurrence 44Facts, related to Opinions and Opinionated

facts:– activities to improve skill 211, 250–253– definition 210– examples 211– facts versus opinions 208– opinionated facts

– definition 210– examples 211

– opinions– definition 210– examples 211

– preference to interpret 249Fans, symptoms-cause 52– cases involving 308, 312, 321Farrell, R. J., cited 256Faults, see CausesFCCU, fluid catalytic cracking unit, see

Cracking, catalyticFDI, Feed density index– cited 56, 57, 58– defined 108Feedback:– self tests:

– experience with process equipment411

– of critical thinking skills 233– of PS skill 194– of trust 242– personal style, confirmation bias 206– personal style in TS 206– skill in five elements of TS 5–8

– forms:– brainstorming 191– criteria identification 193– example data 177– goal setting 193– listener 180– PS strategy 178– reflections about critical thinking skills

233– reflections about PS skills 194– reflections for TAPPS 172– self assessment 193– talker-listener in TAPPS 173– teamwork 50–51– trust 242– TS environment 254–255

Feeder, belt, symptoms-cause 57– cases involving 186, 347– from bottom of hopper, symptoms-cause

57– screw conveyor, symptoms-cause 57– solids volumetric, symptoms-cause 57Filter (LS), symptoms-cause 88– cases involving 286, 342Financial penalties 1, 4, 266– cases involving 321Fisher, K., cited 48, 406Flash drum, three phase separator (GLL),

symptoms-cause 84–85Flexibility, need for 22– as part of the synthesis elements of TS 21,

22Flocculation, symptoms-cause 111– case involving 289Fluid dynamics:– faults, occurrence 44– fundamentals 51– types of faults 51–52Fluoroplastics, extrusion 120Foaming, symptoms-cause 62, 67, 69, 71, 74,

80, 93–98, 110, 128– reboiler selection 63Fouling, symptoms-causes 63, 64, 67, 68, 69,

128Fox, Don F., cited 342,-551Francis, D., cited 48, 406Freon, foaming 129FRI, Flow ratio index– cited 56– defined 108Fundamentals:– and consistency 218– energy exchange 58

Index I 7

– example in Case’4 139– fluid flow 51– separations of homogeneous phases 67– use of, for the basics strategy for TS 23Furnaces:– cases involving 171, 256–257, 304, 308,

374– fundamentals 58– symptoms-cause 60

gGans, M., cited 43, 133, 405–408Gates, John, cited 247, 299, 540, 545Garbage left in process, see Experience with

process equipmentGas breakup, symptoms-cause 110Gauly, R., cited 91, 101Geitner, F. K., cited 403, 408Goals, express as results 22– need for 401– setting goals for improvement 42, 402Goyal, O.P., cited 411Grit chamber, symptoms-cause 86

hHalpern, D., cited 229, 407Handbook data 266Hazard seeHealth and safetyHAZOP, use of 3HDPE, high density polyethylene, extrusion

120Health and safety:– and Case’4 140– combinations of chemicals 131–132– dust explosions, example data 131– flammability, example data 131– health, example data 130– impact on diagnostic action 3, 196, 264– importance in TS 2– need for knowledge about 5– reactions 131– self test about knowledge of 6–7– shutdown to avoid hazard 3, 264– stability, example data 131– symptom-cause 130–132– symptoms of hazard in STR reactors 96Heat exchangers:– cases involving 10, 12, 171, 247, 299, 304,

308, 314, 335, 339, 350, 357, 364, 391– frequency of faults 44, 201– fundamentals 58– types:

– air cooled, symptoms-cause 55

– trim coolers 65– causes 575– plate, symptoms-cause 65– shell and tube:

– causes 575– fundamentals 58– good practice 61– self test experience 6– symptoms-cause 61

– spiral plate, symptoms-cause 65Heat of reaction, and hazards 131Heat transfer fluids, high temperature,

symptoms-cause 66–67– cases involving 252HI, Hopper index– cited 56, 57– defined 108Holmes-Rahe scale for stress 22, 406Hoppers, see BinsHydrocyclone, LL, symptoms-cause 84Hydrocyclone, LS, symptoms-cause 86– cases involving 186Hydrolysis, safety 96, 98Hydrotreating, reactors, symptoms-cause 95Hypothesis,– and cues 25– and reasoning process to validate 223– and skill in classification 219– chart relating to evidence and diagnostic

actions 24– Case’3 426– Case’4 429– Case’5 430– Case’6 148, 202–203– Case’7 158, 159, 171– Case’10 171– Case’11 216– example for TS Case’8 39– example on TS Worksheet 42

– check with colleagues 587– feedback for all cases 537ff– feedback for some cases 267– generate early 24– generating multiple, as part of the synthesis

element of the TS process 21, 22– need to generate 195– number of active 24– self rate of skill 5

i“I want to and I can”, use of 20– Case’5 141

IndexI 8

– example use in TS worksheet for Case’836

Ice formation: in steam ejectors 53“If...then...” logical statements, characteristics

of 214Impact sensitivity 131Indicator, see InstrumentsInference, see ConclusionInjection molding, symptoms-cause 112–119– cases involving 347, 380Instruments, sensors, indicators, recorders– causes 574– frequency of failure 201– simple consistency tests 267–268, 588– symptom-cause data 46Insurance, and data 266, 586– case involving 321Interpersonal skills, see People skillsIon exchange IX:– cases involving 256– fundamentals 67– symptoms-cause 77–78IS and IS NOT:– as diagnostic action 265, 584– example use in TS worksheet

– Case’3 135, 136– Case’4 139– Case’5 141– Case’6 145– Case’7 158– Case’8 36

– use in strategy 23, 197

jJohanson, J. R.– cited 56, 108– description 20, 21, 29Johnson, D. A., cited 135Johnson, D. W., cited 210Johnson, D. W., and Johnson, F. P., cited 243Johnson-Laird, P. N., cited 206, 405Jungian typology, (MBTI) and style 204, 234,

243, 399

kKepner-Tregoe 23, 406King, C. J., cited 12,323Kirton, M. J., cited 243, 406Kister, H., cited 135, 403, 405, 406, 408Kletz, T., cited 245, 253, 407–409Knock out pots, symptoms-cause 80– cases involving 247, 291, 299, 350

Knowledge about process equipment, seeExperience with process equipment

Koros, R. M., cited 95Krishnaswamy, R., cited 173, 407, 538

lLapp, S. A., cited 245, 407, 409Latex crumb, flocculation, symptoms-cause

111Lieberman, N. P., cited 101, 133, 403,

405–409Limitations, personal, need to learn and

identify 22Listening skills:activities to improve skill 239, 250–252– characteristics of 238– feedback via TAPPS 173– need for skill 249– self rate 8– self rate the environment 255– the SIER model 238– three skills 238

– activities to improve skill 239– attending, characteristics of 238– how to 239– reflecting, characteristics of 239– tracking, characteristics of 238

LLDPE, extrusion 120LM ID, log mean temperature difference 58,

61Look back description 20, 21, 29Luckenbach, E. C., cited 101Lynn, Scott, cited 294, 330, 370, 544, 559,

550

mMaintenance, as diagnostic action 265, 584,

585Marangoni effects, causes 76, 83Marlin, T. E., cited 35, 327, 309, 345, 350,

354, 357, 362, 367, 377, 537, 546, 549,552,553, 555, 556, 558, 560

Mass balance 266Material failure fault, occurrence 44MBTI, see Jungian typologyMcNally Institute, cited 403Mean time between failure, MTBF, use of

data 44Mechanical failure, occurrence 44Membranes– fundamentals 67– symptoms-cause 79

Index I 9

Mental representation of the problem:– as part of the Explore stage 21– importance in problem solving 17Methyl ethyl ketone, and foaming 129Microfiltration, seeMembranesMixers: mechanical agitators, L. Mixing,

solids, see Blending– cases involving 181, 297– frequency of faults 44– symptoms-cause 107–108MM, molar mass 103, 124Monitoring:– example application in TS Worksheet for

– Case’5 141, 142– Case’6 145– Case’8 36

– example data of 177– importance of in problem solving 20, 260Motors, electric:– MTBF 44– symptoms-cause 59MPI, main plant items 7MSDS, rating 130, 263, 264, 583MTBF seeMean Time between failures

nNano filtration, see MembranesNeutralization, safety 96, 98, 99NFPA, National Fire Protection Agency,

rating 130, 264, 583NIPR, net inlet pressure required 54Nitration, safety 96, 98, 99Nominal group 49NPSH, net positive suction head 54, 64,

201– activities to improve skill 420– and Case’5 141– definition 420

oOn-going process, typical causes for 44Operators see also People– example interaction with,

– by Ahmed 239– by David 143– by Frank 158– by Jose 238

– importance of talking to 198, 591, 587– on other plants 268, 588Opinionated facts, see FactsOpinions, see FactsOxidation, safety 96, 98, 99

pP&ID (process and instrumentation diagram):– as diagnostic action 268, 589– for depropanizer 34– for ethylene 300– information from 266– symbols on 415Pareto’s principle, use of 22Parker, N. H., cited 186, 407, 538Pattern recognition:– look for as diagnostic action 590– need for skill in 195– types of patterns:

– in cues/evidence 223– in symptoms 221

Pearson, Doug, cited 584Pelleting, strand type, symptoms-cause

111–112– water ring, symptoms-cause 112Penalties 266, 586– cases involving 321People:– activities to improve skill 240– destroyers (four) of relationships 240– fundamental interpersonal RIGHTS 239

– definition 239–240– list of nine skills needed 47– performance:

– factors affecting 244–252– impact of alienation 249– impact of “I know best” 249– impact of listening skills, see Listening– impact of motivation 249– impact of personal style, see Personal style– impact of pride 244–245– impact of stress on 22, 221, 245–246

– activities to improve skill 246–247– self rate environment 254

– impact of the environment on 253–255– self rate 254–255

– impact of unwillingness to admit error244–245– self rate environment 254

– impact of willingness to risk, self rateenvironment 254

– types of errors made 244People problems, see Problems involving peopleperformance review/assessment 249, see

also Self assessmentproblem involving 2– five elements of 8– reflections about 400– self rate 8

IndexI 10

– skill with, need for 2– trust, see TrustPersonal style:– bias 25, 205–206– effect on selecting actions 200, 204–208– identify via Jungian typology (MBT1)

204–205, 243– Saadia’s dominant P 156– Frank’s dominant J 161

– example data 243– impact on selection of diagnostic

actions 204–208– via Johnson and Johnson inventory 243– via Kirton inventory 243

– need to identify 22– preference to infer 249ff– self rate 8– self tests about bias 206– skill in listening 238– tendency to interpret 249–250

– activities to improve skill 250–252– feedback 253– self rate environment 255

PET, extrusion 120pH, acidity, as cause of corrosion 45, 46– as cause of stability 69, 71, 72, 74, 76, 78,

79, 82, 83, 87, 88p-H, pressure enthalpy diagram, use for TS

refrigeration 65Pinch analysis, use of 61Pipes, causes 574Pistorius, J. T., cited 95Plan, fourth stage in a generic problem

solving process– description 20, 21, 28– example on the TS Worksheet for Case’8

39Platforming, reactors, symptoms-cause 91,

93– cases involving 10Pneumatic conveying, symptoms-cause

56–57– cases involving 182Polymerization:– cases involving 178, 297, 370– reactors, STR, symptoms-cause 98– safety 96, 98, 99Powers, G., cited 245, 409Pressure profile:– activities to improve skill 419– self rate skill 7– use of 266

Problems involving people, impact of rulesand regulations 22

Problems, example “problem” in TSWorksheet

– Case’3 137– Case’4 139– Case’5 141– Case’6 145– Case’8 36– need to define the real problem 20, 21,

181– versus exercises 17Problems, TS see also Cases– as related to different types of equipment

44– four common characteristics of 2– high temperature 44– types: change versus on-going process 23

– for on-going process 3, 4, 44– posing safety and health hazards 3, 264– when startup after maintenance

shutdown 4, 44, 265– usual causes 4, 265

– when startup new process 3, 44, 265– usual causes 3

Problem solving, generic See also TroubleShooting, mental process; See also subsetskills of Analysis, Awareness, Creativity,Critical thinking, Strategies for problem solving

– activities to improve skill 399– and type of TS problem:

– as related to different types ofequipment 44

– high temperature 44– on-going process 44– startup after shutdown 44– startup of new process 44

– contrast with exercise solving 17– definition and description of mental

process 17– diagram of process 18– list of characteristics 19, 20– list of detracting and enriching behaviors

29– reflections about your skill 399– self test 194– stages in a strategy 20, 21– strategy for general problem solving 20,

21Process and Instrumentation diagram, see

P&IDProcess control, see Control, process

Index I 11

Process equipment, knowledge about, seeExperience with process equipment

Pumps, centrifugal:– activities to improve skill 420– cases involving 11, 35, 186, 214, 251,

255–256, 286, 289, 294, 304, 308, 321, 330,336, 342, 350, 357, 370, 383, 387

– causes 573–574– fundamentals 51, 420– frequency of faults 44, 201– MTBF 44– self rate experience 6– symptoms-cause 54, 579Pumps, dry vacuum, symptoms-cause 53– gear, symptoms-cause 55– liquid piston vacuum, symptoms-cause

53– mono, symptoms-cause 56– reciprocating, symptoms-cause 54– rotary, symptoms-cause 54– rotary screw, symptoms-cause 56PVC, polyvinyl chloride, extrusion 120

qQuestions to ask, see Diagnostic actions

rRag at interface for SX or decanters– causes of 76RAS, Rough wall angle slide– cited 56, 57– defined 108Reactors– cases involving 1, 10, 171, 178, 182, 297,

304, 308, 317, 330, 339, 357, 374, 391– frequency of faults, 44, 201– introduction 88–89– safety, 96–99– types

– CSTR, mechanical mixer, symptoms-cause 99–100

– CSTR-PFTR with recycle, symptoms-cause 106

– PFTR, bubble reactors, tray columns,symptoms-cause 94

– PFTR, fixed bed, adiabatic, symptoms-cause 91–93

– PFTR, thin film, symptoms-cause 96– PFTR, trickle bed, symptoms-cause

94–96– PFTR multitube fixed bed, non adiabatic,

symptoms-cause 89–91

– Reactive extrusion, symptoms-cause106–107, 123, 125

– STR, batch, symptoms-cause 96–98– STR, fluidized bed, symptoms-cause

101–106– STR, semibatch, agitated bubble,

symptoms-cause 98– STR, semibatch, bioreactor, symptoms-

cause 98– STR, semibatch, symptoms-cause 98

– processes:– alkylation 83, 99– platforming 91, 93– polymerization 178– symptom cause 98

Reasoning related term Critical thinking– activities to improve skill 223–232– common biases in 205– nine-step process 223–232, 299Reboilers, general, fundamentals 63– cases involving 10, 35, 323, 370– causes 575– symptoms-cause 63– types: forced circulation, symptoms-cause

64– kettle:

– fundamentals 63– symptoms-cause 63–64

– thermosyphon:– fundamentals 63– symptoms-cause 64– when used 64

Reflections– activity 5, 172, 194, 233, 251, 262– need for 401, 403Reforming, symptoms-cause 89, 91– cases involving 171, 339Refrigeration– symptoms-cause 65–66– cases involving 202, 247, 299, 323, 350Reverse osmosis, RO, seeMembranesRI, Ratholing index– cited 56, 57– defined 108Riance, X. P., cited 403, 409Rights, personal:– list of 47, 239–240– self rate 8Risk, willingness to 21, 22Rogers, R., cited 297, 407, 544Root cause, see SymptomRotating equipment, frequency of faults 44– MTBF 44

IndexI 12

RTD, residence time distribution 107, 124,125

Rules of thumb:– importance of 218– listing of by process equipment name,

Chapter 3, 43ff

sSafe park, use of 1, 3, 134, 141, 195, 197, 265Safety, seeHealth and safetySafety interlock shutdown, SIS 265Saletan, D., cited 403, 405, 408Samples, collecting 200SBI, Spring back index– cited 56, 57– defined 108SCAMPER, trigger for creativity 184Screens, dewatering– cases involving 186, 286– for Liquid Solid separation, symptoms-

cause 85–86– for Solid Solid separation, symptoms-

cause 88Scriven, M., cited 209, 229, 407Self assessment, related topics Feedback– activity to develop skill 192– feedback form 193– how to develop skill in 192, 401– need for 403– of bias 206– of creativity 191– of critical thinking skills 233– of experience with process equipment 411– of five component skills in TS 5–8– of personal style 206– of personal style, confirmation bias 206– of PS skills 194– of TAPPS 172– of trust 242– target skills 192– using to improve 401Sensors, see InstrumentsSeparation of species in heterogeneous

phases, introduction to 79– process equipment, symptoms-cause

80ffSeparation of species in homogeneous phase,

fundamentals 67– process equipment, symptoms-cause 67ffSeparator, gas-liquid, cases involving 304,

317, 321, 326, 350, 370, 391Shaw, I. D., cited 570Short term Memory, STM 25

Shut down (the process), and safety, see alsoHealth and safety 1

– troubles after, for maintenance 4, 44, 265Silveston, P. L., cited 314SIS, safety interlock shutdown 3, 265Size enlargement, fundamentals 110– process equipment for 110ffSMARTS: example use in TS Worksheet for– Case’3 136, 423– Case’4 428– Case’5 141, 431– Case’8 37Solids conveying, see Conveying, solidsSolvent extraction, SX, fundamentals 67– and coalescers, symptoms-cause 110– SX, column extractors, symptoms-cause

76– SX, centrifugal, symptoms-cause 77– symptoms-cause 76Sonic velocity, and compressors 52Sour water scrubbers, SWS, causes of

corrosion of 46– symptoms-cause 75, 76So What?, use 232Startup, of new process:– as diagnostic actions 265– typical causes of trouble 44, 265– use of basic strategy for 23Startup after shutdown:– as diagnostic actions 265– typical causes of trouble 44, 265– use of change strategy for 23Steam, usage 198Steam systems, causes of corrosion in 36– cases involving steam generation 339– good practice 58– symptoms-cause 66Steam traps, symptoms-cause 80– causes 575Stiction 47STM, see Short Term MemoryStrategies for problem solving,– activities to improve skill 173– description 20, 21Strategies for TS, related entries Basics and

Change– selecting basics versus change 197Stress:– activities to improve skill 246–247– effect of high distress on propensity to

make mistakes 22, 221, 245–246– need for some stress 22– need to manage 20, 22

Index I 13

– self rate environment 254– suggestions on how to manage 22Stripping, see Desorption, gasStyle, preferred, see Personal styleSubproblems, identify 20Sulfinol 383, 387Sulfolane, and foaming 129Sulfonation, safety 96, 98, 99Sulfuric acid systems, causes of corrosion in

46Surfactants (causing foaming), examples 74– causing emulsion stability, examples 84Surface phenomena, importance 67Surface tension:– and absorption 73– and coalescers 111– and wetting 94– critical, use in absorption 73– impact 69, 71, 73, 83, 110, 129– in packed columns 109– in reactors 94– negative, definition 69– positive, definition 69Symptoms, (one type of cue):– characteristics of 2– definition 212– how to identify symptom 220– root cause versus symptom 24, 43, 195

– activities to improve skill in identifyingroot cause 217, 225

– examples, Case’8 38, 216Symptom-cause data, see specific pieces of

equipment– diagraming 228, 229– symptom-cause data for

– Case’3 228–229– Case’8 38

– symptom-hypotheses-diagnostic actionchart, seeHypothesis, chart

symptoms, list of for– Case’6 147– Case’8 39Synthesis– element in the TS process 21, 22, 33– list of detracting and enriching behaviors

31– self rate skill 5Systems thinking:– activities to improve skill 418– definition 5, 7, 127– elements in 7– need for 2– self rate skill 7

tTabletting, symptoms-cause 111TAPPS (talk aloud pairs problem solving)– use of, to improve awareness 165–168– use of, to improve the application of

strategies 174–177Tanks, frequency of faults 44Taylor, W., cited 13, 214, 255, 317, 339, 357,

383, 387, 391, 540Teamwork:– feedback form for meetings 50–51– symptoms-cause 48–49Temperature, high, fiult in systems operating

at 44Tests, selecting and designing, see Diagnostic

actionsThermodynamics 267Thickener, symptoms-cause 86–87– cases involving 286, 342Three phase separators, GLL, symptoms-

cause 84Time management:– how to 22– need for good 22Transmitters, symptom-cause data 47Trends in evidence, see PatternsTrim coolers, use of 61Trouble shooting, mental process:– activities to improve skill, triad 259–262– as a problem solving process 17ff– assessment of skill 33, 262– classification of components:

– data and analysis elements 19, 398– Case’4 140– Case’5 144– Case’6 156– Case’7 161–162– example rating form 33– example critique for Case’3 138– list of detracting and enriching

behaviors 30– decision making elements 19, 22, 398

– Case’4 140– Case’5 144– Case’6 156– Case’7 161–162– example rating form 33– example critique for Case’3 138– list of detracting and enriching

behaviors 32– problem solving elements 19, 20, 21,

398– Case’4 140

IndexI 14

– Case’5 144– Case’6 156– Case’7 161–162– example rating form 33– example critique for Case’3 137– list of detracting and enriching

behaviors 29– synthesis elements 19, 22, 398

– Case’4 140– Case’5 144– Case’6 156– Case’7 161–162– example rating form 33– example critique for Case’3 138– list of detracting and enriching

behaviors 31– definition 1, 2– five key skills:

– experience with process equipment 4,398– self test 5

– people skills 5, 397, 400– self test 8

– problem solving 4, 397, 398– self test 5

– process safety and properties ofmaterials 5, 397– self test 6

– systems thinking 5, 397– self test 7

– individual 262– possible immediate corrective options 1– reflection about activities 401– reflection about skill 398– self rate five strengths 402– strategies, see Basics strategy for TS and

Change strategy for TS– worksheet, description 27–29, 587

– succinct version 26–27– detailed version 40–42– examples

– Case’3 423– Case’4 427– Case’5 141, 430– Case’6 145– Case’8 35–39

Trouble shooting problems, see Problems, TSTrust, building and developing:– activities to improve skill 241, 242– components of 240– definition 242– need for 47, 240– self rate skill 9, 242

Turbine, MTBF 44, 201– cases involving 35, 171, 255, 339, 350– steam, symptoms-cause 59Turnaround, as diagnostic action 265, 584,

585– trouble after 4, 44Turner, J., cited 80Tyler, Ted, cited 222Types of trouble shooting problems, see

Problems, TS

uUltrafiltration, seeMembranesUniqueness, see Personal style

vVacuum, equipment to produce, symptoms-

cause 53–54– and crystallizers, symptoms-cause 72– and evaporators, symptoms-cause 68– cases involving 13, 181, 252, 255–256, 335,

374, 383, 387Valid conclusions, how to reach, see ReasoningValves, causes 574– block, symptom-cause data 47– check, symptom-cause data 47– control, symptom-cause data 47– rotary star, symptoms-cause 56Varadi, T., cited 94Vendors, vendor files, use of 197, 266– call 268Vessels, pressure, frequency of faults 44

wWasan, W.C., cited 405Water, makeup contains corrosion, products,

causes 46Water treatment, flocculation, symptoms-

cause 111Weather, importance of 197, 584– as diagnostic action 265, 584– example impact 221Wetting– and absorbers, symptoms-cause 73– and decanters, symptoms-cause 83– and demisters 110– and packed columns 109– and reactors 94– and SX, symptoms-cause 76What if?– example use in TS Worksheet for

– Case’4 429

Index I 15

– Case’5 142, 432– Case’8 37

– to create counterarguments 232– trigger for brainstorming 185– when to use 198Why? Why? Why?– activity to improve skill 181– as diagnostic action 265– example use on TS Worksheet

– Case’4 139– Case’7 180– Case’8 36

– prompt to use on TS Worksheet 26, 41– use of during the explore stage of problem

solving 20

Winter, D. R., cited 347, 380, 552, 561Woods, D. R., cited 130Worksheet, see Trouble shooting, worksheet

yYip, Jonathan, cited 256, 289, 543Yokell, S., cited 546, 409Young, D., cited 48, 406

zzpc, zero point of charge (isoelectric point),

and stability 69, 71, 74, 76, 78–80, 83, 93,111

IndexI 16

donald r. woods - skills4success.cosaharapcc.skills4success.co/spcc/er/successful trouble shooting...

Documents