looking ahead - jflfthe redesigned masthead emphasizes our commitment to new technology and new...

It is with great enthusiasm that I announce severalchanges related to the James F. Lincoln Arc WeldingFoundation and this publication. As you havealready discovered from the new cover format andtitle, this publication is now known as WeldingInnovation . The redesigned masthead emphasizesour commitment to new technology and new ideas,while the arc-like icon takes us right back to thebasic technology that makes welding viable — theelectric arc associated with the most popular weldingprocesses used today. As one might deduce fromour title change, we will no longer be targeting fourissues per year. We are still committed to makingthis informative journal available at no cost to ourreadership, but now plan to publish twice a year.In addition, the entire contents of the publication,beginning with this issue, will be available online at http://www.lincolnelectric.com.

It is also my pleasure to announce the appointmentof our Editor, Duane K. Miller, to the position ofSecretary of the Foundation. While I will be continu-ing in my role as Executive Director, Duane willassume the day-to-day responsibilities of running theFoundation, in addition to his current responsibilitiesas Editor and as Senior Project Engineer with TheLincoln Electric Company. Duane’s international rep-utation continues to grow, and he is in constantdemand as a speaker at seminars and conferencesall over the world. His new position as FoundationSecretary will give him the opportunity to interactwith our International Assistant Secretaries, coordi-nating worldwide programs aimed at increasing thequality and reducing the cost of welded components.Even as Lincoln Electric continues to expand intonew markets, it is the goal of the James F. LincolnArc Welding Foundation to help develop the weldingindustry in these emerging markets, just as it hasdone in the past and continues to do currently in theUnited States and Canada.

To supply some assistance to Duane Miller, R. ScottFunderburk has been named Assistant Editor of WeldingInnovation. He will be assuming day-to-day responsibilityfor the publication, and new ideas and comments maybe directly forwarded to Scott. Being part of the newer

Looking Ahead

Standing: Dick Sabo, Scott Funderburk.Sitting: Omer Blodgett, Duane Miller.

generation of computer-literate engineers, Scott hasrequested our readers to contact him at his emailaddress: [email protected]. I ampleased with Scott’s contributions to date — in fact, he islargely responsible for most of this edition of the maga-zine. However, I am even more excited to note that the“mentoring relationship” which is the subject of a continu-ing series of articles in this publication has moved on toyet another generation. Duane Miller has become ScottFunderburk’s mentor, while Omer Blodgett remains ourDesign Consultant, so that, in effect, a three-generationmentor/protégé relationship is now happily in place.

Amidst all these changes, certain things will continueas before. The James F. Lincoln Arc WeldingFoundation is firmly committed to its goal of advanc-ing the use of arc welding through a commitment toever-improved quality and reducing the cost of weld-ed fabrication. Through our books, seminars, awardsprograms and such publications as WeldingInnovation, we look forward to continuing to workwith people like you to build a better. . . stronger. . .welded future.

Richard S. Sabo

THE JAMES F. LINCOLN ARC WELDING FOUNDATION TRUSTEES & OFFICERS

Dr. Donald N. Zwiep, Chairman John T. Frieg, Trustee Leslie L. Knowlton, Trustee Richard S. Sabo Duane K. Miller, P.E.Worcester, Massachusetts Cleveland, Ohio Cleveland, Ohio Executive Director Secretary

Welding Innovation Vol. XIV, No. 1, 1997 1

ArgentinaRaul TimermanPhone: 54-1-753-6313

Australia and New ZealandRaymond K. RyanPhone: 61-29-772-7222Fax: 61-29-792-1387

CroatiaProf. Dr. Slobodan KraljPhone: 38-41-512-014Fax: 38-41-514-535

Hungar yDr. Géza Gremsperger Phone: 361-156-3306

IndiaDr. V.R. KrishnanPhone: 91-11-247-5139Fax: 91-124-321985

JapanDr. Motoomi OgataPhone: 81-565-48-8121Fax: 81-565-48-0030

People’ s Repub lic ofChinaDai Shu HuaPhone: 86-22-307-407Fax: 86-22-307417

RussiaDr.Vladimir P.YatsenkoPhone: 07-095-238-5543Fax: 07-095-238-6934

United KingdomDr. Ralph B.G. YeoPhone & Fax:441-709-379905

Cover: The Lágymányos bridge on the Danubeat Budapest features a unique lighting design.See story on page 20.

2 First-Ever Gas-Fired Annealing of a Nuclear ReactorExtending the life of nuclear reactors with the help of computer technology.

4 Mentoring in the Engineering ProfessionThird in a series of four: “Big Shoes to Fill.”

8 Ensuring Weld Quality in Structural ApplicationsPart III of III: A discussion of alternate acceptance criteria.

20 Sixty Years of Welded Bridges in HungaryThe impact of welding technology on bridge construction in Hungary.

The serviceability of a prod-uct or structure utilizing thetype of information present-ed herein is, and must be,the sole responsibility of thebuilder/user. Many vari-ables beyond the control ofThe James F. Lincoln ArcWelding Foundation or TheLincoln Electric Companyaffect the results obtainedin applying this type of infor-mation. These variablesinclude, but are not limitedto, welding procedure, platechemistry and temperature,weldment design, fabrica-tion methods, and servicerequirements.

INTERNATIONAL ASSISTANT SECRETARIES

Volume XIVNumber 1, 1997

EditorDuane K. Miller, P.E.

Assistant EditorR. Scott Funderburk

The James F. Lincoln Arc WeldingFoundation

Omer W. Blodgett, P.E.Design Consultant

Features

Departments

7 Opportunities:1997 Design & Welding Seminars

12 Award Programs:1996 Awards for College Engineeringand Technology Students

17 Design File: Use Undermatching Weld Metal Where Advantageous

Visit Welding Inno vation online at http://www .lincolnelectric.com/inno vate.htm

2 Welding Innovation Vol. XIV, No. 1, 1997

IntroductionThe first-ever indirect, gas-fired, radi-antly heated annealing of a nuclearreactor was a major success, thankspartly to computer simulation that vali-dated the process in advance. Afteryears of neutron bombardment,nuclear reactor pressure vesselsbecome brittle. The embrittlementproblem forced Yankee Atomic ElectricCo. to decommission its Rowe powerplant outside Boston in 1992—eightyears before its license was due toexpire. Annealing restores the ductilityand fracture toughness of the weldmetal, adding many years of operatinglife to the reactor vessel. However, theproblem is that annealing requiresheating the entire vessel to a tempera-ture of approximately 850°F for aweek. The annealing approach pio-neered by Westinghouse ElectricCorp. and Cooperheat, Inc. involvesblowing superheated air from gas-fired burners into a heat exchangerthat, in turn, heats the reactor vessel.This approach was validated, prior to asuccessful $6 million demonstrationproject, by building a functional scalemockup, and by simulating heat trans-fer and air flow using computationalfluid dynamics (CFD) software.

BackgroundThe pressure vessel contains the fueland control mechanism for the nuclearreactor and is located within the con-tainment building. Metallurgical exper-iments have shown that annealing thevessel will nearly restore its originalmetallurgical properties. This requiresdisassembling the internal compo-nents of the reactor (a standard opera-

First-Ever Gas-Fired Annealingof a Nuclear Reactor

tion during refueling) and providing ameans for in-situ heating of the 35 foottall, 15 foot diameter vessel.

The size of the reactor vessel alone isnot a major obstacle. The difficulty ofthe task relates more to the fact thatthe inside of the reactor is contaminat-ed with radiological materials that can-not be released to the environment.This eliminates the most commonmethod used to anneal large pressurevessels: circulating superheated airfrom gas burners inside the vessel andinsulating the outside of the vessel tominimize heat loss. The problem isthat the hot combustion gases wouldpick up radioactive dust and other con-taminants inside the vessel andspread them into the atmosphere.

Several nuclear reactors in Russiahave been annealed using radiantelectric heat; however, this approachwas not used for various technical andeconomic reasons. Ironically, despitethe fact that the pressure vessel waslocated within an electric generatingfacility, obtaining power for the heaterswould have been very costly. Anotherproblem with this approach is that afailed electric heater element wouldhave been virtually impossible torepair or replace inside the reactor.The use of redundant electric heaterswould have nearly doubled the costand the weight of the furnace.

Pioneering anAlternate ApproachFor these reasons, Cooperheat engi-neers decided to pioneer an alternateapproach for annealing reactors. Thismethod uses gas-fired burners to

superheat air and blow it throughsealed ducts in existing openings inthe containment building, such as theequipment hatch, into the heatexchanger inside the reactor vessel.The superheated air is then dis-charged outside of containmentthrough another duct to the atmos-phere. Since the air never comes intocontact with any contaminated sur-faces, it does not become contaminat-ed. The gas-fired heaters are locatedoutside the containment building sothey can be easily replaced in case offailure.

This approach clearly had the poten-tial to eliminate many of the problemswith electric heating. But it also raisedseveral potential difficulties of its own.The main one was insuring that theheat exchanger would be able tomaintain temperature uniformitythroughout the entire reactor vessel

annealing zone. The problem with notachieving adequate temperature uni-formity is that it can create excessivethermal stresses. Much analysis wasperformed to verify that the reactorvessel would not be over-stressed.

A scale model of the heat exchangerand reactor vessel was built and test-ed very early in the project to experi-mentally estimate heat transfercoefficients. A computational fluiddynamics (CFD) method to simulate

By Mike Sciascia, Project EngineerPaul Moodey, Design EngineerCooperheat, Inc.Piscataway, New Jersey

The Marble Hill testproved that aging

reactor vessels can berejuvenated and their

operating life substantially extended,permitting old reactorsto continue operations

for many years.


flow distribution and heat transfer with-in the heat exchanger was utilized toconfirm the experimental results. ACFD analysis provides fluid velocity,pressure and temperature valuesthroughout the solution domain forproblems with complex geometriesand boundary conditions. As part ofthe analysis, a researcher may changethe geometry of the system or theboundary conditions such as inletvelocity, flow rate, etc. and view theeffect on fluid flow patterns or concen-tration distributions. CFD also canprovide detailed parametric studiesthat can significantly reduce theamount of experimentation necessaryto develop prototype equipment andthus reduce design cycle times andcosts.

FIDAP CFD software from Fluent Inc.(Lebanon, New Hampshire) was select-ed because FIDAP uses the finite ele-ment method, which is ideal forgenerating the complex and irregulargeometries which were involved in theproposed heat exchanger design. Theflexibility of the mesh generation tool pro-vided with this software package makesit possible to handle very odd shapes.Another advantage of this program isthat a version compatible with the com-pany’s existing computer system wasavailable at a very reasonable price.

Designing the Heat ExchangerThe authors first developed an initialdesign for the heat exchanger.Consultants from Fluent modeled theflow within the heat exchanger usingthe assumption that it would be uni-form. The initial design verified theexperimental heat transfer coefficients,based on the assumptions that weremade. Next, the consultants modeledthe distribution system which providedhot gases to the heat exchanger tomake sure that it actually met the uni-form assumptions of the first analysis.The initial design had assumed that asingle injection point would provideuniform circulation within the heatexchanger. The analysis showed that,with only a single injection point, thehot gases impinged upon the wall of

the heat exchanger and created arecirculation zone. This could havecaused a cold spot that would haveprevented the heat exchanger fromuniformly heating the reactor vessel.Consultants changed the model sever-al times and re-ran the analysis toevaluate the results. The engineeringteam finally settled on the use of fourinjection points, which provided theuniform flow required to achieve uni-form heat transfer.

A SuccessfulDemonstrationWith the experimental results con-firmed, the design team was able toproceed with confidence that the newprocess would work as expected. Thefirst opportunity to use it came at ademonstration at Public Service ofIndiana’s never-completed Marble Hillplant near Paynesville, Indiana. Theannealing demonstration at Marble Hillwas carried out with the combinedresources of the Department ofEnergy’s Sandia Laboratories and anindustry consortium including theAmerican Society of MechanicalEngineers, the Electric PowerResearch Institute, Consumers Power

Company, Japan’s Central ResearchInstitute of the Electric Power Industry,Westinghouse and Cooperheat.

The reactor vessel was heated forseven days and 10 hours, as the heatexchanger reached a temperature ofabout 1,100°F. The reactor pressurevessel was brought to its peak temper-ature at a rate of about 20°F per hourand cooled down at the same rate.Each of the five gas burners producedtwo to three million BTUs of heat perhour. During the annealing process,the pressure vessel and surroundingcomponents were monitored by over500 thermocouples, strain gauges anddisplacement gauges.

An important concern during the test-ing was showing that the vessel main-tains a fairly even temperaturedistribution during annealing in orderto avoid the stresses associated withthermal variations. Another concernwas avoiding damage due to the factthat the vessel expands as it is heat-ed, but the piping and other connec-tions surrounding the vessel do notexperience the same expansion.The test was a complete success. Allmeasurements showed that the vesselmaintained an even temperature dur-ing the annealing process. Preliminaryanalytical results verify that the metalthroughout the vessel walls, welds andattached piping expanded and con-tracted without damage exactly as pre-dicted. The Marble Hill test provedthat aging reactor vessels can be reju-venated and their operating life sub-stantially extended, permitting oldreactors to continue operations formany years.

Steve Trich, General Manager ofWestinghouse’s Nuclear ServicesDivision, stated: “The technology andoverall annealing process demonstrat-ed at Marble Hill went well beyond ourexpectations. This successful demon-stration overturns a major hurdle facedby utilities with aging reactor vessels.Utility executives can confidently planto extend the life of their reactorsknowing that vessel embrittlementneed not be an impediment.”

The heat exchanger prior to its instal-lation inside the reactor vessel.


Remembering Johnny GriffithsIn October 1995, the engineering com-munity lost a unique individual whenJohn D. Griffiths, P.E., passed away.His contributions to the professioncannot be measured by the papers hewrote, nor by the number of times hespoke, nor even by his role in helping

to develop the modern format of thelatest A.I.S.C. Specifications andManuals. John’s greatest contributionwas his passion for his life’s work dur-ing a career that spanned more thanfive decades.

He came from an age when an engineerwould sign in at a hotel and proudly write“P.E.” after his name. It always saddenedhim to see engineers use disclaimers

�Big Shoes to Fill�By Mark V. Holland, P.E.

Chief EngineerPaxton & Vierling Steel Co.Omaha, Nebraska

and approver notes in an attempt todilute their responsibility. John was oftenannoyed by engineers who tried to over-complicate the behavior of steel. One ofhis favorite gambits was to ask them toexplain, through examples, how addingstiffness to a frame could cause it to col-lapse. His insight into the behavior ofsteel would have been invaluable to theengineering community in this post-Northridge environment.

Our Initial MeetingWhen I first met Johnny Griffiths, I wasa graduate student at the University ofOklahoma working on a research pro-ject sponsored by the AmericanInstitute of Steel Construction.Certainly, I had no idea that this wasthe man who, more than any other,would influence the course of my pro-fessional life. At the time, John waschairman of the A.I.S.C. research pro-ject, which focused on bolted momentend plates, a particular interest of his.He was also vice president of engi-neering at the Paxton & Vierling SteelCompany of Omaha, Nebraska.

The next time I saw John was whenmy thesis advisor, Tom Murray, and Iwere traveling through Omaha on aresearch-related business trip. Westopped in to see John at Paxton &Vierling Steel, and Tom mentioned thatI was looking for a job. John casuallysuggested that I submit a resume,which I subsequently did. It did not

really occur to me then that JohnnyGriffiths and Paxton & Vierling Steelhad very specific plans for me.

Big ShoesI was hired by Bob Owen (who wasJohn’s boss and is still my boss) andcharged with working to the pointwhere I would someday fill John

Griffiths’ shoes. At first, I did not reallyunderstand the scope of that expecta-tion. However, as John took mearound Omaha and Sioux City, intro-ducing me to his colleagues in theengineering profession as “the youngman I am training to fill my shoes,” itbegan to dawn on me. Time and timeagain, I was told “You’ve got big shoesto fill!”

This article is the third in a four-part series in which four different engineers recount their unique experiences as protégésof recognized experts in the steel fabrication industry.

Mentoring in the Engineering Profession

John D. Griffiths, mentor.

Mark Holland, protégé.

The family, friends and colleaguesof John Griffiths have establishedthe John D. Griffiths MemorialScholarship Fund at the Universityof Nebraska. The scholarshipfund is intended to preserve andhonor the memory of a great engi-neer by helping to support theeducation of qualified studentswith an interest in structural engi-neering.

Born July 8, 1909, in TakomaPark, Maryland, Mr. Griffiths grad-uated from the University ofCincinnati with a degree in civilengineering in 1934. After col-lege, he worked in a variety ofengineering capacities beforeserving in the Navy during WorldWar II from 1942 to 1946, endinghis naval service as commanderof the Civil Engineering Corps.

In a career that spanned fivedecades, Mr. Griffiths progressedfrom being one of the originalAmerican Institute of SteelConstruction regional engineers toeventually become vice president

of engineering at Paxton & VierlingSteel Company. He publishedmany papers and two books:Single Span Rigid Frames in Steel(A.I.S.C.) and Multiple Span GableFrames (A.S.C.E. Transactions).Throughout his working life, heendeavored to raise the standardof professionalism within hisindustry, while constantly encour-aging young engineers to maxi-mize their potential.

Graduate student Lamont Epp ofLincoln, Nebraska, was named thefirst recipient of the John D.Griffiths Memorial Scholarship forthe academic year 1996-97, andMr. Epp’s scholarship award hasbeen renewed for 1997-98.

Those wishing to contribute to theJohn D. Griffiths MemorialScholarship Fund at the Universityof Nebraska are asked to contactJohn Erickson, University ofNebraska Foundation, 8712 WestDodge, Suite 402, Omaha,Nebraska 68114-3434.


John’s position at the time was in theengineering-marketing area of Paxton& Vierling. He was expected to besomething of a tutor, or engineeringguru, for everybody in the area—theguy that everyone could go to with thetough questions. Yet he also had totalk to the general public and educatepeople about engineering and connec-tion design.

When I read the first article in thisseries, authored by Duane Miller, Iwas struck by the parallels betweenour relationships with our respectivementors, John in my case, and OmerBlodgett in Duane’s. Something thatDuane and I have always shared isthat we stand in the shadows of these

great men. When Omer and Johnsearched for their protégés, both ofthem were looking for the kind of per-son who could be very theoretical, andat the same time be capable of takinga complicated problem and explainingit to people from any walk of life in aclear way. John was a master at this,and Omer is, as well.

Another thing that John and Omer hadin common was the fact that both ofthem worked long past the usual retire-ment age. In John’s case, it wasbecause he could not find the right per-son to fill his role. At the time of myarrival at Paxton & Vierling in 1983, Johnwas 74, and he had already retiredtwice, but the people hired to replacehim had not worked out, so each time,he had returned to his position.

Priceless LessonsOnce I started working at Paxton &Vierling, Bob Owen made it clear thatmy time with John was to be my first

The John D. GriffithsMemorial Scholarship Fund

priority. I am still grateful to Bob forthat, because the insights and knowl-edge I gained were priceless.

Perhaps few us of know what we’regetting into when we accept a new job.I had no idea! As the weeks andmonths went by, I began to get fright-ened. People expected me to knowthe kinds of things Johnny Griffithsknew, and of course he had more thanforty years of experience in the indus-try. Furthermore, John’s thirst forknowledge never ended. So I realizedI would have to develop that same

thirst just to keep up with his constantacquisition of new knowledge.

John’s and my routine echoed themeetings over morning coffee thatDuane described in his article aboutOmer Blodgett. As in their situation,what looked like a coffee break wasreally more of a classroom...anadvanced course in not only the tech-nical fine points of connection design(though plenty of those were dis-pensed), but also in how proceduresand specifications had been devel-oped: the process, the politics and the

I had no idea that thiswas the man who,

more than any other,would influence the

course of my professional life.


philosophy behind their development...the actual business of how work getsdone. John was personal friends withpeople I had only read about in text-books.

He was a very, very proper man. If Ispoke in an unprofessional way, hewould correct me, and correct me, andcorrect me again. I wasn’t just gettingtechnical tutoring, I was being trainedin how to sit across a table from some-

one and communicate clearly and pro-fessionally. It turned out John’smother had been an English professor,and his wife was an English major. Iremember the first time I worked forhim on putting together a presentation.He gave me a list of materials to gath-er, including everything from extra lightbulbs for the slide projector, to tape tocover the electrical cords. John leftnothing to chance when it came toprofessionalism.

As the best teachers do, John taughtby example. I noticed that he neverdominated at the beginning of a con-versation. Like a chess player, hewould quietly listen to the conversa-tion, letting others talk and define theirpositions. Then, at the appropriatetime, he would use the information hehad gained during the conversation toexplain his position and persuade theothers to his way of thinking.

“Always define the problem before youtry to solve it,” John would challengeme before I tackled any issue. Hestressed that as engineers, we shouldanswer only the questions we knowthe answers to, and research the rest.He emphasized that engineers havean obligation to be honest about what

Visit Lincoln Electric�sWeld Technology Center

Online www.lincolnelectric.com/weldtech.htm

The Application EngineeringGroup of the Weld TechnologyCenter works with customers towardsolving welding related problems,and helps lower total welding costsby facilitating creative solutions.

The Engineering Ser vices Gr oupof the Weld Technology Centerassists engineers and fabricatorswith the cost-effective design andfabrication of welded connections.Weld Technology Center engineersconduct educational seminars,author technical publications, andprovide telephone consulting assis-tance.

In addition to the complete contentof Welding Inno vation , beginningwith this issue, current online offer-ings available free of chargeinclude:

Beam Deflection Welding DesignSoftwareStructural Beam to Column DesignAid

Fabricator s’ and Erector s’ Guideto Welded Steel ConstructionFree 60 page Adobe® Acrobat®Reader File.

we know, and to find out about whatwe don’t know. Even in the twilight ofhis career, John was always eager tolearn, and he inspired in me that samehunger for learning.

Even with all of his knowledge, Johnhad great respect for those working inother aspects of steel construction.He told me that I could learn more inone week working with a detailer thanin a whole semester of an engineeringcourse. When I first started withPaxton & Vierling Steel, he put me intothe shop and made sure that I workedwith the saltiest old fellow there, whoalso happened to be the most experi-enced fitter. This exposed me to agreat knowledge base of just howthings go together.

John was really good at giving me prob-lems to solve. I would spend daysworking on one of these problems,come up with an answer that I thoughtwas correct, and finally present it to him.He would look at it, perform some reallyquick, dirty calculations, and arrive atalmost the same answer. And then hewould say “Here are the rules of thumbthat will get you to the quick answer.”As a fabricator’s engineer, one has tobe able to get to the answers quickly.

But John always made sure that I couldarrive at the answer by doing the prob-lem correctly, and then he would showme the easy way.

I remember once having to computethe shear center of an odd shape.After several hours and many pages oftriple integrals, I proudly showed Johnmy work. True to form, with only a fewlines of simple calculations, he approx-imated my answer more closely thanany shop tolerance would require.What the steel industry will miss, withthe passing of John and of engineerslike him, is this practical, accurate andsimple understanding of the behaviorof steel.

John always madesure that I could

arrive at the answerby doing the problemcorrectly, and then he

would show me theeasy way.

As the best teachers do, Johntaught by example.


Production Welding — September 29-October 2Blodgett’s Design of Steel Structures — October 6-10Blodgett’s Design of Steel Weldments — November 3-7

Lincoln Electric’s seminar series continues toattract top design engineers and production weld-ing personnel from across the country and aroundthe world. All of the seminars are conducted in thestate-of-the-art Weldtech Center at Lincoln’s worldheadquarters in Cleveland, Ohio.

Omer W. Blodgett, P.E., and Duane K. Miller, P.E.,conduct the Design Seminars. Blodgett’s Designof Steel Weldments for machine tools, construc-tion, transportation, material handling, agriculturalequipment, and manufactured metal products ofall types is aimed at reducing manufacturing costsand improving performance through the efficient

use of welded steel. Blodgett’s Design of Steel Structures addresses methodsof reducing costs, improving appearance and function, and conserving materialthrough the efficient use of welded steel in a broad range of structural applications.Each 5-day Design Seminar earns 3.3 CEU credits and costs $395.00.

Production Welding is a 4-day seminar conducted by Lincoln’s staff of expertwelding engineers. It covers welding process selection, welding variables and pro-cedures, weld design, welding metallurgy, and nondestructive testing. The programis designed for welding foremen, superintendents, industrial engineers, and timestudy, quality control and inspection personnel. The seminar fee is $295.00.

Group size for each seminar is limited. Therefore, it is wise to register early.Rooms will be reserved so that the seminar group can stay together in the samehotel. Other living arrangements can be made if desired.

For further information or a registration form, write or call:

The Lincoln Electric Company22801 St. Clair AvenueCleveland, Ohio 44117-1199Attention: Marion ZagorcPhone: (216)383-2240

Opportunities

1997 Design & Welding Seminars


Ensuring Weld Quality in Structural Applications, Part III of III

By Duane K. Miller, P.E.Senior Project EngineerThe Lincoln Electric CompanyCleveland, Ohio

This three-part series on ensuring weld quality in structural applications covers the following:

Part I reexamined the roles of the Engineer, the Fabricator, and theInspector, as they relate to welded construction. The proper roles weredefined, and misunderstandings corrected.

Part II emphasized the importance of effective visual inspection and itsvital role in achieving weld quality.

Part III discusses alternate acceptance criteria and explains the Engineer’sresponsibility for invoking such criteria.

Throughout the series, reference is made to specific sections of the AWS D1.1-96 Structural Welding Code - Steel.

Defining �Quality�One of the currently popular definitionsof a quality product is as follows: “Aquality product is one that meets spec-ification requirements.” By this defini-tion, quality is integrally linked to theapplicable specification. As long asthe product meets those requirements,it is deemed “quality.” Unfortunately, ifthe specification is incorrect or inap-propriate, conformance to thoserequirements may satisfy this definitionbut would not satisfy the wants anddesires of the ultimate customer. Onlywhen a proper specification is utilized,and when the product integrity meetsor exceeds those specification require-ments, will a true quality product havebeen produced that meets customerrequirements have been produced.Therefore, a more appropriate defini-tion of “quality” would include the

concept of meeting customer expecta-tions in addition to the standard speci-fication requirements. This philosophyis summarized by A.M. Gresnight ofDelft University, the Netherlands, asfollows:

“A good weld is any weld whichdoes the job it is intended for dur-ing the service life of the structure.”

In the structural field, the customer(Owner) has a representative (theEngineer) who develops the necessaryspecifications (contract documentsand cited codes and standards) thatenable the manufacturer (Fabricator)to deliver a quality product. In the caseof fabricated steel, the commonly citedstandard for quality is the AWS D1.1-96 Structural Welding Code - Steel.

D1.1 Quality ProvisionsAn understanding of the philosophybehind the D1.1 code will help theEngineer to determine whether it willadequately address the needs of theOwner. Section 6.8, “Engineer’sApproval for Alternate AcceptanceCriteria,” states: “The fundamentalpremise of the code is to provide gen-eral stipulations applicable to most sit-uations.” The emphasis is significant.It is important to consider the scope ofthe D1.1 code, which covers struc-tures that are static and dynamic: on-and off-shore applications that utilize

IntroductionQuality is tied to a given specification.The specification must be suitable tomeet the ultimate owner’s needs. Formost welded construction, the AWSD1.1-96 Structural Welding Code -Steel provides adequate acceptancecriteria for welded construction.Unusual structures, however, maydemand additional requirements. D1.1criteria may be overly restrictive forsome lesser structures. For noncon-formances to a standard specification,alternate acceptance criteria may beutilized in order to avoid unnecessaryweld repairs. It is the Engineer’sresponsibility to evaluate the appropri-ateness of an alternate acceptancecriterion before invoking it for a specif-ic project.

Alternate Acceptance Criteria


plate, rolled shapes, and tubular mem-bers. The products covered range fromsimple single story metal-framed build-ings to 100-story-plus skyscrapers andoffshore drilling platforms. In somecircumstances, these general stipula-tions may be overly restrictive, and inother situations, they may not ade-quately address the demands of thestructure. Evaluating the suitability ofthe specification for the application iscertainly the responsibility of theEngineer.

In the commentary to Section 6.8 ofthe code, the following can be found:“The criteria provided in section 5,Fabrication, are based upon knowl-edgeable judgment of what is achiev-able by a qualified welder. The criteriain section 5 should not be considered

as a boundary of suitability for service.Suitability for service analysis wouldlead to widely varying workmanshipcriteria unsuitable for a standard code.Furthermore, in some cases, the crite-ria would be more liberal than what isdesirable and producible by a qualifiedwelder. In general, the appropriatequality acceptance criteria andwhether a deviation produces a harm-ful product should be the Engineer’sdecision. When modifications areapproved, evaluation of suitability forservice using modern fracturemechanics techniques, a history ofsatisfactory service in similar struc-tures, or experimental evidence is rec-ognized as a suitable basis foralternate acceptance criteria forwelds.” This commentary makes itclear that the code has utilized what isachievable as the acceptance criteri-on, not what is necessary for the par-ticular application. This is areasonable approach for a standard

specification, and as is indicated in thecommentary, precludes the need forwidely varying fabrication standardswhich would be difficult to monitor in atypical fabrication facility. When theweld quality does not meet these stan-dards, however, it is inappropriate toautomatically assume that the weldwill be unacceptable for service. Thisshould, however, drive the Engineer tolook to fitness-for-service type criteriafor further evaluation.

Few Engineers recognize that theD1.1 code permits the use of alternateacceptance criteria for welds.According to Section 6.8: “Acceptancecriteria for production welds differentfrom those specified in the code maybe used for a particular application,provided they are suitably documentedby the proposer and approved by theEngineer. These alternate acceptancecriteria can be based upon evaluation

of suitability for service using pastexperience, experimental evidence, orengineering analysis concerning mate-rial type, service load effects, andenvironmental factors.”

These provisions permit the Engineerto utilize alternate acceptance criteria.Since quality is integrally linked to theapplicable specification, the accep-tance criteria will have a major impacton the final product. The Engineer’s

responsibility is to assess the suitabili-ty of a standard specification to a par-ticular project, as well as to approvean alternate should the need arise.

Considering AlternateAcceptance CriteriaThere are three areas in which alter-nate acceptance criteria should beconsidered: First, there are the situa-tions where standard acceptance crite-ria are inadequate to the demands ofthe structure. Secondly, standardacceptance criteria may be overlyrestrictive for a particular application.Finally, there are cases in which fabri-cation is routinely performed to a stan-dard specification, with minornoncompliances that can be acceptedthrough the use of an alternate accep-tance criterion. All three are signifi-cant issues and will be addressedhere.

Certain structures make unusualdemands upon welds and weld quality.When new materials are employed,significant deviations from standardmaterial thicknesses are utilized, newwelding processes are employed,and/or when the design of the struc-ture involves a significant departurefrom established practices, it is pru-dent for the Engineer to critically eval-uate the suitability of standardspecifications. For example, the steel

�A good weld is anyweld which does thejob it is intended for

during the service life of the

structure.�

Figure 1. Heavy sections may require especially rigorous alternate acceptancecriteria.


fabrication industry learned manylessons when “jumbo sections” wereinitially applied to tension applicationsin trusses. Standard materials (hotrolled, carbon and/or low alloy steelshapes) in unusual thicknesses(flanges exceeding 5” in thickness)were being used in new applications(direct tension connections). The com-mon workmanship criteria set forth inthe various codes and specifications,as well as normally acceptable work-manship criteria, proved to be inade-quate in a number of structures. Inhindsight, it would have been prudentto employ more rigorous alternate

acceptance criteria for these types ofstructures. Since that time, provisionshave been written to address these sit-uations and have been presented ina variety of technical journals. Indeed,the standard specifications nowinclude more rigorous requirements.

The second situation occurs when thestandard acceptance criteria are moredemanding than is justified for the par-ticular application. An example in thestructural field would be in the fabrica-tion of steel joists. These componentsin steel buildings are usually coveredby another specification that is moreapplicable to the particular productinvolved. Application of the sameacceptance criteria as are applied toother fabricated steel structures and

mandated by D1.1 would be overlyrestrictive, justifying alternate accep-tance criteria. The Engineer should becareful when routinely suggesting thatalternate acceptance criteria beemployed which deviate from, or areless rigorous than, a national consen-sus standard such as D1.1. This prac-tice would be recommended only forspecific applications for componentswhere well established, time-provenpractices that have a history of ade-quacy have been used, and wheredeviation from these practices wouldconstitute a major hardship. The D1.1code, however, clearly gives the

Engineer the authority to accept analternate standard in these situations.

The most important use of alternateacceptance criteria, however, appliesto the third situation, where standardacceptance criteria have been utilizedfor the fabrication practice and minornonconformances have been uncov-ered. Alternate acceptance criteriacan be utilized to accept these non-conformances and eliminate the needfor unnecessary repairs. Obviously,the alternate acceptance criteria cho-sen must, as is true in the case of allengineering decisions, be applicableand appropriate for the application.Neither poor workmanship nor poorquality can be accepted. However,when the weld that does not conform

to the standard specification is suitablefor the specific situation, alternateacceptance criteria may be employedto eliminate the need for a repair.There are many reasons why this maybe desirable for all (that is, the Owner,the Engineer, and the Fabricator).Unnecessary delays may be avoided.Costly repairs are avoided. Finally,and perhaps most importantly, an“acceptable” repaired weld may actual-ly be inferior to the weld initially reject-ed as “unacceptable.” Again,according to A.M. Gresnight:

“Standards for weld discontinuitiestraditionally are based on goodworkmanship criteria. By extendingthe traditional standards with thesecond quality level, based on fit-ness for purpose, unnecessary andpotentially harmful repairs can beavoided.”

Weld DiscontinuitiesWeld discontinuities fit into two broadcategories: planar and volumetric.Planar discontinuities include cracksand lack of fusion. These are seriousdiscontinuities that are unacceptable,and particularly critical in structuressubject to fatigue. Volumetric disconti-nuities include items such as porosity,slag inclusion, and undercut. Theseare less significant, and when heldwithin certain limits, are acceptable bymost codes even under dynamic load-ing situations.

Volumetric discontinuities are readilydiscernible by nondestructive testingmethods and, in many cases, by visualinspection. Planar discontinuities areharder to detect, and may even beoverlooked by radiographic nonde-structive testing. It has been shownthat, during initial fabrication, most dis-continuities are volumetric in nature.Under repair welding conditions, whichare more demanding than original fab-rication circumstances, planar disconti-nuities are more likely to develop.Notice the progression: readilydetectable, less significant volumetricdiscontinuities observed in the originalfabrication may be removed andreplaced with welds that contain lessdetectable, but more significant, planar

Figure 2. The 7” flanges of these transfer girders require careful considerationof the NDT acceptance criteria.


discontinuities. This is not to say orimply that welds cannot be effectivelyrepaired. It does mean, however, thathaphazard demands for weld repairmay actually result in a product ofdecreased value to the Owner. Itshould also be noted that the deposi-tion of additional weld metal is likely toincrease distortion and residual stress-es in the structure. When the noncon-forming weld is adequate for theparticular application, the responsibleapproach is to utilize an alternateacceptance criterion to eliminate theunnecessary repair.

Most Engineers are unsure of the suit-ability of alternate acceptance criteria.The search for appropriate documentsthat employ the “fitness for purpose”approach is generally a frustratingexperience. Apart from finding infor-mation regarding the methods thatmay be employed for analysis, practi-cal ideas as they relate to welds areall too scarce. Mr. Robert E. Shaw, Jr.,P.E., of Steel Structures TechnologyCenter, Inc., (40612 Village OaksDrive, Novi, Michigan 48375-4462)has provided Engineers with a usefulsource of information regarding alter-nate acceptance criteria. Shaw hassystematically evaluated specific dis-continuities, the D1.1 code require-ments, and other standards that couldbe used as alternates. His summaryis excellent and provides practicaloptions to Engineers.

Undersized welds are a common prob-lem. The situation is simple: thedrawings call for a 5/16” fillet weld andthe welder deposits a 1/4” fillet. Atleast two options are available: first,additional weld metal can be deposit-ed over the surface to build it up to therequired size; secondly, an alternateacceptance criterion could beemployed that would allow thesewelds to be acceptable as deposited.It should be noted that the D1.1 codewould allow the welds to underrun thenominal fillet weld size by 1/16” with-out correction provided that the under-sized portion of the weld does notexceed 10% of the weld (D1.1-96,Table 6.1, see Part II of this series). If

the entire weld, however, is under-sized, this provision would not beapplicable. In many cases, the depo-sition of additional weld metal wouldbe routine and would not constitute amajor problem. However, the initialweld may have been produced with anautomatic, submerged arc weldingmachine when the travel speed hap-pened to be slightly too high, resultingin a slightly undersized weld. Theweld may be beautiful and meet all cri-teria except for the size. To make theweld repair, a gang of manual or semi-

automatic welders may be assigned todeposit the additional weld metal. Thefinished product may be visually inferi-or, and subject to all of the potentialdiscontinuities of the starts and stopsassociated with manual and semiauto-matic welding.

Has the product quality beenenhanced by the repair? First, it mustbe determined if the undersized weldwould have been acceptable. As isthe case in many situations involvingfabricated plate girders, the weld sizemay have been based upon the mini-mum prequalified fillet weld size pre-scribed in the D1.1 code. The designbasis was not strength, but this mini-mum size. A quality, 1/4” fillet weldwould have provided all the necessarystrength in this particular situation.The reasoning behind the minimum fil-let weld size in the code is based upon

good workmanship practices and con-trolling the heat input to preclude weldcracking. However, in this example, ithas been assumed that the initial weldis a quality weld, free of cracks, withacceptable weld contours, etc. If thisis the situation, leaving the undersizedweld in place, unrepaired, is a moreresponsible approach than demandingthe weld repair. The initial weld isprobably of higher quality than therepaired weld would be, will have lessdistortion, is less costly, and will elimi-nate unnecessary delays.

The decision to invoke alternateacceptance criteria must be made bythe Engineer. In a separate article inthis series, the roles of the Engineer,the Inspector, and the Fabricator aredefined. The Inspector cannot makethis decision and neither can theFabricator. Only the Engineer with anunderstanding of the loading, designassumptions, and overall structuralsignificance can make these types ofdecisions.

ConclusionFor most applications, the AWS D1.1-96 Structural Welding Code - Steelprovides adequate acceptance criteriafor welded construction. For weldsthat deviate from standard acceptancecriteria, engineering judgment shouldbe applied before repairs are mandat-ed. If the weld will meet the structuralrequirements for the project withoutmodification, the responsible approachof the Engineer is to utilize alternateacceptance criteria and accept thesewelds. Ultimately, a product ofimproved quality at reasonable costwill be the result of this approach.

Readily detectable,less significant

volumetric discontinuitiesobserved in the

original fabricationmay be removed andreplaced with welds

that contain lessdetectable, but moresignificant, planar discontinuities.


Jury of Awards:

Brent HallProfessor, University of Illinois, Urbana

Gary KrutzProfessor, Purdue University

Walter MassieProfessor, Technical University at DelftHolland, the Netherlands

Donald N. ZwiepChairman of the Jury,Chairman, the James F. Lincoln Arc Welding Foundation

1996 Awardsfor College Engineering and Technology Students

Cornell UniversitySan Jose State UniversitySanta Clara UniversityStanford UniversityUniversity of California, BerkeleyUniversity of Illinois, UrbanaUniversity of Maryland at College ParkUniversity of MinnesotaUniversity of WyomingVirginia Polytechnic & State UniversityWorcester Polytechnic Institute

The James F. Lincoln Arc Welding Foundation granted the following awards totaling$15,750 to undergraduate and graduate students in the 1996 Pre-Professional AwardsProgram. Grants also were made to the following schools:

(Left to right) Brent Hall, Gary Krutz, Walter Massie

BEST OF PROGRAM�$2,000 EACH

UNDERGRADUATE GRADUATE

Bradley O. CarlsonLaramie, WY

University of WyomingMechanical Engineering

Faculty: Donald A. Smith

University of California, BerkeleyCivil Engineering

Faculty: Egor P. Popov

Tzong-Shuoh YangAlbany, CA

Design of a Remotely Operated Safety Release for a Bareback Bronc Rigging

Experimental and Analytical Studies of SteelConnections and Energy Dissipators


GOLD AWARDS�$1,000 EACH

Cornell University Agricultural & Biological Engineering

Faculty: Wesley W. Gunkel

SILVER AWARDS�$750 EACH

Wayne LeeMarlboro, NJ

Diane SteinkampCentralia, IL

Michael Fasolo*Palatine, IL

*Not pictured

Min-Fan LeeBurnaby, BC, Canada

Clay Douglas Price*Laramie, WY

*Not pictured

John L. SullivanFremont, CA

Mutsuko YamadaMenlo Park, CA



University of Illinois, UrbanaGeneral Engineering

Faculty: Manssour Moeinzadeh

University of WyomingMechanical Engineering

Faculty: Donald A. Smith

Stanford University Mechanical Engineering

Faculty: Drew V. Nelson

Santa Clara University Mechanical Engineering

Faculty: Tim Hight

Mark Alan PriorStanford, CA

Christian F. JohnsonSanta Clara, CA

Lisa AbruzziniMountain View, CA

Alberto SalazarMountain View, CA

Rostam PouroushasbCollege Park, MD

University of Maryland at College ParkCivil Engineering

Faculty: Pedro Albrecht

Anheuser: Glue Optimization for Package Strength Intelligent Computer Vision Control & Target TrackingSystem Design for an Agricultural Grapevine PruningRobot

Final Report for a Pneumatic Post Driver

Silicon Wafter Sensing on Robot “End Effector” Retrofitting Fatigue-Cracked Composite Beams withExternal Prestressing, Case Study: I-79 Bridge

Integrated CD Charger & Battery Pack


Worcester Polytechnic InstituteMechanical Engineering

Faculty: Allen H. Hoffman

Virginia Polytechnic & State UniversityMechanical Engineering

Faculty: Charles Reinholtz

Michael Oliva*Quincy, MA

Susan MacPherson*Hudson, MA

*Not pictured

BRONZE AWARDS�$500 EACH

Christopher MichalakPenfield, NY

James O’SullivanGreenfield, MA


Design & Fabrication of a Backpack Access Device

San Jose State UniversityMechanical Engineering

Faculty: Buff Furman

Kristi WilliamsSan Jose, CA

John GarciaSan Jose, CA

James RitsonSan Jose, CA

Mechanical Hand for Below-Elbow Body-PoweredProsthetic Arm

University of Illinois, UrbanaGeneral Engineering

Faculty: Harry S. Wildblood

Joseph B. KirkeyChicago, IL

James T. Fisher*Orland Park, IL

*Not pictured

Rodney B. PhillipsRaleigh, NC

ADM: Corn Steep Water Evaporator Fouling

Stephen L. CanfieldNewport, VA

The Carpal Wrist, a Parallel-Architecture Robotic Wrist

University of MinnesotaCivil Engineering

Faculty: Theodore V. Galambos

Matthew W. BeckmanWarrens, WI

University of California, BerkeleyCivil & Environmental Engineering

Faculty: Egor P. Popov

Brent BlackmanHermosa Beach, CA

Seismic Analysis and Design of Multi-Bay RigidTrussed Frames

Studies in Steel Moment Resisting Beam-to-ColumnConnections for Seismic-Resistant Design


GE: Lean Engineering Appr oach to Packaging of Major AppliancesUniversity of Illinois, UrbanaFaculty: Manssour MoeinzadehJames G. Beier , Mattoon, ILWaymond W. Eng , Chicago, ILMichael S. Pape, Frankfort, IL

Minuteman: Fan Redesign f or aGasoline P owered VacuumUniversity of Illinois, UrbanaFaculty: Daniel L. MetzJoshua Fredric kson , Glenview, ILChad Peterson , Marseilles, ILMatthe w Woessner , Sterling, IL

Gearbox Coupling Design f or a Vertical Tank AgitatorPurdue UniversityFaculty: Mark T. MorganTodd L. Redlin , West Lafayette, IN

Wellhead Bolt Cleaning Mac hineUniversity of WyomingFaculty: Paul A. DellenbackErik Lundber g, Laramie, WY

A More Functional and Cosmetic Hand Pr osthesis f or AmputeesStanford UniversityFaculty: Drew V. NelsonRajiv Doshi , Stanford, CAAjit Chaudhari , Cupertino, CAClement Yeh, Honolulu, HI

Elco: Monitoring Tool WearUniversity of Illinois, UrbanaFaculty: Henrique ReisBill Huffman , Charlotte, NCMatthe w J. Jackowski , Naperville, ILAaron C. Voegele, Sheldon, IL

Design of a Hand Extension Ex erciserSanta Clara UniversityFaculty: Mark D. ArdemaDavid T. Eveland , Woodland, CADelia Sauceda-Lopez , San Jose, CAWilliam M. Richter , Oroville, CA

Eaton: Spurious Response in Engine Knoc k Sensor sUniversity of Illinois, UrbanaFaculty: Henrique ReisIan Bruce , Springfield, ILMarc Koeppel , Carol Stream, ILAdam Lac k, Orland Park, IL

Fire Saf e ProjectUniversity of Illinois, ChicagoFaculty: Mun Young ChoiBesher Ra yyahin , Chicago, ILJoseph Cold wate , Chicago, ILRemus Hotca , Prospect Heights, IL

Testor: Needle Assemb ly RedesignUniversity of Illinois, UrbanaFaculty: Wayne J. DavisJennif er Bounds , Frankfort, ILPaul Klaus , Freeport, ILAnselmo Rosa , Chicago, ILPeter Wong , Chicago, IL

Design of an Ocean Wave Ener gyExtraction De viceWorcester Polytechnic InstituteFaculty: Leonard D. AlbanoEric P. Truebe , Mirror Lake, NH

Forensic In vestigation of a Fishing Vessel SinkingU.S. Coast Guard AcademyFaculty: Vincent WilczynskiDerek Sc hade , Chesapeake, Virginia

Watlo w Gor don: Fluid Bath for Testing Thermal Response Time of Thermometer sUniversity of Illinois, UrbanaFaculty: Roland L. RuhlPeter J . Ditmar s, Naperville, ILJanice M. Holba , Frankfort, ILMelisa K. Olson , West Chester, PADan Pawlak , Champaign, IL

Design of a Retail Store P aint ShakerWorcester Polytechnic InstituteFaculty: Robert L. NortonMichael I. Walker , Auburn, MAJoshua R. Binder , Medway, MA

Sycamore: Design of Modular Drawer Suspension SystemUniversity of Illinois, UrbanaFaculty: Roland L. RuhlJennif er Antanaitis , Orland Park, ILMichael Brennan , Orland Park, ILDarren K ennell , Shelbyville, IL

Nor th American Glass: Glass QualityUniversity of Illinois, UrbanaFaculty: Harry S. WildbloodGregor y Faber , Princeton, ILEric ka Olson , Litchfield, ILErik Parks , Decatur, IL

MERIT AWARDS � $250 EACH

Effects of Dent-Dama ge on theResidual Strength & Repair of WeldedSteel Tubular BracingLehigh UniversityFaculty: James M. RiclesWilliam M. Bruin , San Francisco, CA

Temperature Sensitive Milling Burr s for Neur osur gical ApplicationsStanford UniversityFaculty: Larry LeiferDale B. Perrigo , Anchorage, AKS. Matthe w Desmond , Walla Walla, WA

Measured Stresses in Steel Cur vedGirder Bridg esUniversity of MinnesotaFaculty: Theodore V. GalambosBrian E. Pulver , Chicago, IL

Refrig eration ApplicationStanford UniversityFaculty: Larry LeiferHacene Bouadi , Palo Alto, CARober t Ware, Fairfield, CTAlfred Hernandez , El Paso, TX

Epidural Anesthesia Sim ulatorStanford UniversityFaculty: Larry LeiferJoeben Be vir t, Santa Cruz, CADavid Moore , Redwood City, CAJohn Q . Norw ood , San Francisco, CA

UNDERGRADUATE

GRADUATE


COLLEGE ENGINEERING AND TECHNOLOGY ENTRY FORMMAIL ENTRY ON OR BEFORE MIDNIGHT, JUNE 15, 1998 TO SECRETARY:

The James F . Lincoln Ar c Welding Foundation, P.O. Box 17035, Cleveland, Ohio 44117-0035Please submit a current photograph of each author, with the name written on the back, with the Entry.

1998 AWARDS

Fill in information requested and attach to title page of paper being enteredin the Program. ENTRY MUST BE TYPED.

DO NOT PUT YOUR NAME OR NAME OF SCHOOL ANYWHERE INYOUR ENTRY OTHER THAN ON THE ENTRY FORM.

Subject or title of paper(put this title on Page 1 of paper also). _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

To be entered in the following division (check only one):

o Undergraduate Division o Graduate Division

Department _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Professor or Faculty Advisor _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

School Name _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

School Address _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

This paper completed by the undersigned author(s) within the period of June 15, 1997

to June 15, 1998.

Entrant(s) signature(s) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

NOTE: In case of joint authorship of an award winning Entry, each author will receivea check dividing the total award amount equally unless otherwise indicated. Pleasecheck one of the boxes below.o Divide Equally o Divide as follows:

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

STATEMENT OF CERTIFICATION(Must be signed by faculty member)

Following statement must be signed by a professor, instructor or the faculty advisor.

The author(s) of this paper _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

(is/are) (was/were) registered at (school) _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

and to the best of my knowledge this paper was originated and prepared by thestudent(s) named on this Entry Form.Signed _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Phone _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Name(s)

Professor or Faculty Advisor (Print Name Below)

AUTHOR(S) Please Type — No Abbreviations

CHECK WILL BE SENT TO HOME ADDRESS. NOTIFICATION 10/98.Home address must be address at which you will receive mail October 1998.

Full Name

Soc. Sec. No. Home Phone

Home Address

City, State, Zip

School

School Address

City, State, Zip

Full Name


Home Address

City, State, Zip

School

School Address

City, State, Zip

Full Name


Home Address

City, State, Zip

School

School Address

City, State, Zip

Full Name


Home Address

City, State, Zip

School

School Address

City, State, Zip

Full Name


Home Address

City, State, Zip

School

School Address

City, State, Zip

Entries may not be submitted by more than five students.

Prof. Donald N. Zwiep, ChairmanProf. & Head Emeritus, Mech. Engin. Dept.Worcester Polytechnic InstituteWorcester, Massachusetts

Leslie L. Knowlton, TrusteeCleveland, Ohio

Richard S. SaboExecutive DirectorFAX: 216/383-8220

Duane K. MillerSecretary

John T. FriegTrusteeCleveland, Ohio


How Strong Does a Weld Have To Be?

The answer is fairly simple: strong enough to transfer theloads that are passed between the two interconnectedmaterials. How strong does the weld metal have to be?The answer to that question is far more complex.

In order to make a weld of sufficient size, the designer hasthree variables that can be changed to affect the weldstrength:

• weld length, • weld throat, and • weld metal strength.

Since three variables are involved, there are many combi-nations that are suitable for obtaining the correct weldstrength. It may be necessary to also consider the loadsimposed on the base metal to ensure that the completewelded joint has sufficient strength. This edition of“Design File” will focus on the variable of weld metalstrength.

For purposes of this discussion, “weld metal strength” isdefined as the yield and tensile strength of the depositedweld metal, as measured by an all-weld metal tensilecoupon extracted from a welded joint made in conformancewith the applicable AWS filler metal specification.“Matching” weld metal has minimum specified yield andtensile strengths equal to or higher than the minimumspecified strength properties of the base metal. Notice thatthe emphasis is placed on minimum specified propertiesbecause, in the case of both the filler metal and the basemetal, the actual properties are routinely higher. An exam-ple of matching weld metal would be the use of E70XXfiller metal on A572 grade 50 steel. The weld metal/basemetal properties for this combination would be 60/50 ksi

Use Undermatching Weld Metal Where Advantageous

(414/345 MPa) yield strength and 70/65 ksi (483/448 MPa)tensile strength. Even though the weld metal has slightlyhigher properties than the base metal, this is considered tobe a matching combination.

All too often, engineers see filler metal recommendationsprovided in codes that reference “matching” combinationsfor various grades of steel and assume that this is the onlyoption available. While this will never generate a noncon-servative answer, it may eliminate better choices. Matchingfiller metal tables were designed to give the recommenda-tions for one unique situation where matching weld metal isrequired, that is, Complete Joint Penetration (CJP) groovewelds in tension applications. All other applications permitsome degree of undermatching, and undermatching maybe a very desirable, cost-effective alternative for applica-tions such as Partial Joint Penetration (PJP) groove weldsand fillet welds.

The significance of weld metal strength, as compared tobase metal strength, has increased in recent years, as thenumber of higher strength steels continues to grow. WhenA36 was the predominant steel, commercially availablefiller metals would routinely overmatch the weld deposit.As steels with a 50 ksi (345 MPa) minimum specified yieldstrength became more popular (e.g., A572 grade 50 andA588), the use of the E70XX grades of filler metals provid-ed for a matching relationship. Steels with minimum speci-fied yield strengths of 70 ksi (483 MPa) through 100 ksi(690 MPa) have become more and more popular.Although matching strength filler metals are available, theoption of using undermatched weld metal, where applica-ble, is increasingly attractive.

When undermatching weld metal is utilized, the designermust ensure that weld strength is achieved, but this iseasily done with the standard equations used to determinethe allowable stress on the weld.

Practical Ideas for the Design P rofessional by Duane K . Mille r, P.E.

Design File


For fillet welds in 90° T-joints, the maximum allowable loadon the weld can be determined from the following equation:

F = (0.3) (0.707) ω (EXX) L

where,

ω = weld leg size EXX = minimum specified tensile strength

of the filler metalL = length of the weld

By substituting in the strength level of the undermatchedfiller metal, the weld strength can be determined.Undermatching may be used to reduce the concentrationof stresses in the base metal. Lower strength weld metalwill generally be more ductile than higher strength weldmetal. In Figure 1, the first weld was made with matchingfiller metal. The second weld utilizes undermatching weldmetal. To obtain the same capacity for the second joint, alarge fillet weld has been specified. Since the residual

stresses are assumed to be of the order of the yield pointof the weaker material in the joint, the first example wouldhave residual stresses in the weld metal and in the basemetal of approximately 100 ksi (690 MPa) level. In the sec-ond example, the residual stresses in the base metal wouldbe approximately 60 ksi (20 MPa), since the filler metal hasa lower yield point. These lower residual stresses willreduce cracking tendencies, whether they might occur inthe weld metal, in the heat affected zone, or as lamellartearing in the base metal.

Overmatching is undesirable and should be discouraged.Caution must be exercised when overmatching weld metalis deliberately used. The strength of a fillet weld or PJPgroove weld is controlled by the throat dimension, weldlength, and strength of the weld metal. In theory, over-matching filler metal would enable smaller weld sizes to beemployed and yet create a weld of equal strength.However, the strength of a connection is dependent notonly on the weld strength, but also on the strength of thefusion zone. As the weld size decreases, the fusion zoneis similarly reduced in size. The capacity of the basemetal is not affected by the selection of the filler metal, so

it remains unchanged. The reduction in weld size mayresult in overstressing the base metal.

Consider the three PJP groove welds shown in Figure 2.A load is applied parallel to the weld, that is, the weld issubject to shear. The allowable stress on the groove weld is0.30 times the minimum specified tensile strength of the

electrode (i.e., the “E” number). The allowable stress on thebase metal is required not to exceed 0.40 times the yieldstrength of the base metal. The first weld employs a match-ing combination, namely A572 grade 50 welded with E70electrode. The second example examines the same steelwelded with undermatching E60 electrodes, and the finalexample illustrates overmatching with an E80 electrode.

As shown in Figure 2, the allowable stress on the weld andthe allowable stress on the base metal have both been cal-culated. In the case of undermatching weld metal, theweld metal controls the strength of the joint. For matchingweld metal, the allowable load on both the weld and thebase metal is approximately the same. In the case of theovermatching weld metal, however, the base metal is thecontrolling variable. For this situation, it is important tocheck the capacity of the base metal to ensure that theconnection has the required strength.

Practical ApplicationsAlthough it is relatively easy to determine which situationsare suitable for the use of undermatching weld metal,some designers may simply choose to use the “conserva-tive” approach and specify matching weld metal for all situ-ations. However, matching weld metal may actually reduceoverall weld quality, increasing distortion, residual stresses,

Figure 1. Matching and undermatching filler metal.

Figure 2. Effect of filler metal strength level.


and cracking tendencies, including lamellar tearing. Theuse of undermatched weld metal is an important option forsuccessfully joining higher strength steels.

When welding on higher strength steels with undermatch-ing weld metal, it is important that the level of diffusiblehydrogen in the deposited weld metal be appropriate forthe higher strength steel that is being welded. For exam-ple, whereas an E6010 electrode is suitable for welding onlower strength steels that are not subject to hydrogenassisted cracking, it would be inappropriate to utilize this asan undermatching weld metal on 100 ksi (690 MPa) yieldstrength A514 or A517 since these are highly sensitive tohydrogen cracking. When undermatched weld metal isused, it must not exceed the maximum levels of diffusiblehydrogen appropriate for matching strength weld deposits.Also, any preheat requirements for matching strength rela-tionships must be maintained even when undermatchingweld metal is utilized.

There are economic considerations as well. While manyfiller metals from various welding processes are capable ofdelivering 70 ksi (490 MPa) weld deposits, the number ofoptions available to the fabricator is greatly reduced when100 ksi (690 MPa) yield strength weld metal is required(i.e., E110 class filler metals). The metallurgical character-istics necessary for the deposition of weld metal at thisstrength level may impose restrictions on the electrodedesigner which limit the attainable welding speeds and/oroperational characteristics. In contrast, the requirementsfor lower strength filler metals give the electrode designersmore latitude and may result in improved operational char-acteristics.

For many beam-like sections that are subject to bending,the resultant longitudinal shear that must be transmittedbetween the web and the flange of a built-up section is rel-atively small. Weld sizes that are based upon the transferof the stress alone may result in surprisingly small welds,welds so small that they may be difficult to make on a pro-duction basis. The heat input associated with these smallwelds may be so small that weld cracking results. Forthese reasons, the D1.1 code has established minimum fil-let weld sizes that, regardless of the level of stressimposed on the weld, should be maintained in order toobtain sound welds. For example, the minimum fillet weldsize for 3/4 in. (18mm) steel is 1/4 in. (6mm). For materialsgreater than 3/4 in. (18mm), a 5/16 in. (8mm) fillet weld isthe minimum acceptable size. The implications of this arethat, for most structural fabrications involving built-up sec-tions, the minimum fillet weld size is 5/16 in. (8mm). This isoften greater than the required fillet weld size necessary totransfer the longitudinal shear. Particularly in these appli-cations, undermatched weld metal offers a significantadvantage because the minimum required weld size canbe achieved utilizing the undermatched weld metal, andthis may also satisfy design requirements.

The following welding procedures have assumed that thegoverning factor for the design of the members subject tobending is the minimum weld size, not the allowable stresson the connection. This will not always be the case.However, comparing the welding procedures associatedwith the two connections shows the benefit of utilizing theundermatched filler metal.

Example � Matching VersusUndermatching Filler Metal

Given: T-joint formed by 3/8” and 1” plates, 5/16”fillet weld, horizontal position, automatic SAW

Find: Cost savings for using undermatching versus matching filler metal

Assumptions: SAW Flux/electrode combinations – 960/L-61 (undermatching) and 880M/LAC-M2 (matching)

Labor & Overhead = $40.00/hr.

Equations: Cost = Labor + Materials

*Prices based on current Industrial Prices in Lincoln Electric price book (April, 1997)

Solution:In this particular situation, savings of 40% were achievedby utilizing the undermatching option. Design requirementswere satisfied, and the likelihood of welding-related prob-lems such as weld cracking or lamellar tearing has beenreduced.

Cost ($/ft) = Labor & Overhead ($/hr)Travel Speed (in/min)

+ [electrode used (lb/ft) x electrode cost ($/lb)]+ [flux used (lb/ft) x flux cost ($/lb)]

12 in/ft60 min/hr

Labor & Overhead ($/hr.) $ 40.00 $ 40.00Travel Speed (in./min.) 20.0 16.0Electrode Used (lb./ft. of joint) 0.19 0.28Electrode Cost* ($/lb.) $ 0.87 $ 1.84Flux Used (lb./ft. of joint) 0.24 0.15Flux Cost* ($/lb.) $ 0.53 $ 0.93

COST ($/ft.) $ 0.69 $ 1.16

Matching880M/LAC-M2

Undermatc hing960/L-61


The Welding Industryin Hungary: A BriefHistoryProceduresThe industrial use of arc welding inHungary began in 1930 with shieldedmetal arc welding using a coveredelectrode (SMAW). Starting in the1950s, Hungarian industry began touse submerged arc welding (SAW)and gas metal arc welding (GMAW),the latter at first with CO2 and laterwith mixed gas (mostly 82/18 Ar/CO2).In the 1970s, gas-shielded flux coredmetal arc welding (FCAW-g) becamecommonplace, and by the 1990s, self-shielded flux cored welding (FCAW-s)was added to the repertoire. Mostrecently, gas tungsten arc welding(GTAW) has been preferred inHungary for welding the root pass.

EquipmentIn the early years, Hungarians usedgenerators and transformers, at firstimported from Germany and laterdomestically produced. In the 1950s,SAW equipment was imported fromthe Soviet Union and Western Europe;in time, Hungarian manufacturers alsoentered this market. Today, bothEuropean and U.S. equipment manu-facturers have distributors in Hungary.

Sixty Years of Welded Bridges in Hungary

The material in this article has been extracted, with permission, fromthe author’s paper entitled “60 Years of Welded Structures, inParticular Welded Bridges, in Hungary,” which was presented at the49th Annual Assembly of the International Institute of Welding inBudapest, Hungary, September 2-3, 1996, and is contained in the“Proceedings of the International Conference on Welded Structures,in Particular Welded Bridges.”

Consumab lesEven in the infancy of the Hungarianwelding industry, covered electrodeswere manufactured domestically. In1980, a modern factory with a capacityof 30,000 tonnes per year of coveredelectrodes and fluxes was built. Solidwires for semi-automatic (GMAW) andautomatic (SAW) welding also are produced in Hungary, but importedconsumables are used in great quanti-ty, as well.

Base MaterialsThe Hungarian steel industry wasestablished in the nineteenth century.However, the manufacture of goodweldable steel (mostly grades 37 and52, nominal tensile strength) thatwould resist brittle fracture was notstandardized and introduced until1965. One catastrophe occurred in1969, when two 30 m³, 15 Bar (217.5psi) service pressure sphericalreceivers that had been manufacturedin 1960 exploded, killing thirteen peo-ple. They had been fabricated of amaterial totally unfit for the purpose:22 mm thick plates, grade A42.21 (astructural steel used for riveted con-struction according to the Hungarianstandard MSZ 21:1950).

Education, Certification &Professional Or ganizationsPrograms of welding engineering wereintroduced in 1961/62 in twoHungarian technical universities:Miskolc and Budapest. Since then,about 700 engineers have receivedcertificates. Since 1993, Hungary’sstandards for education and traininghave met the Guidelines of theEuropean Welding Federation (EWF)and now, in cooperation with theAustrian Welding Institute, it is possi-ble for Hungarian engineers to qualifyfor and receive the EWE (EuropeanWelding Engineer) certificate.

Education of welding experts at thelevel of “European WeldingTechnologist” (EWT) takes place atboth of the aforementioned technicaluniversities, and also at two technicalcolleges, the ME DunaujvarosiFoiskolai Kar 300, and the BankiDonat Gepipari Muszaki Foiskola.

The Central Welding Section of theScientific Society of MechanicalEngineers, formed in 1949, representsHungary and its welding professionalswithin the International Institute ofWelding (IIW). In 1990, the HungarianAssociation of Welding Technologywas formed by 32 member companiesand institutes; today, there are 65member organizations. About 3,000welders per year are qualified by 30certified training schools.

Quality Assurance & Contr olThe quality and safety of welded struc-tures requires, first of all, weldingexperts who are highly trained andeducated in the fields of design, man-ufacturing and fabrication. Hungaryhas taken all the necessary steps toensure reliable quality assurance and

By Dr. Sandor DomanovszkyQuality Assurance & Welding ManagerGanz Steel Structure Co. Ltd.Budapest, Hungary


control: issuing manufacturing andwelding instructions, qualifyingwelders, destructive and nondestruc-tive testing of the welded joints, andbuilding the necessary jigs to ensurethe best position and the correctshape of structures. Many Hungariancompanies have instituted TotalQuality Management systems.

A Transitional P eriodAlong with every other sector of theHungarian economy, the structuralsteel industry is in an important periodof transition. In the post-World War IIperiod, Hungary’s production of steelstructures totalled about 150,000tonnes (1 tonne = 1000 kg) per year,with more than 60 percent accountedfor by buildings. The collapse ofsocialism has brought about privatiza-tion of Hungarian industry since 1990.However, economic times are veryhard, and many of the newly privatizedcompanies have gone bankrupt. Thecontractors who are operating in thismarket come mostly from WesternEurope. The transition is extremelydifficult, yet the Hungarian people stillendeavor to work together to solve theproblems facing our country.

Welded Bridges Special Challeng es of Orthotr opic Dec k DesignThe majority of Hungarian weldedbridges are of orthotropic deck design.The definitive characteristic oforthotropic deck design is its uniqueability to perform all of the functionspreviously derived from several sepa-rate girders forming a floor of longitudi-nal and transversely integrated girderscoordinated with the main girders.From a manufacturing point of view,the components of this complex girdersystem are more difficult to fabricateand fit than simple plane units for con-ventional structures.

For example, a traditional I-beam gird-er consists of only three main parts,and in the course of adjustment onlysix locations have to be fitted to twoadjacent girders. A typical orthotropicunit, by comparison, forms a complex

of longitudinal and transverse girdersconnected with the deck plate. Thisrequires about fifty units being fitted tofour adjoining units. Also, while theconventional girder expands in onlytwo directions, the orthotropic struc-ture expands in three, so moving andturning of the units requires more skill,making it difficult to secure dimension-al accuracy.

With conventional girder construction,assembling components such as webplates, flanges and stiffeners, andstraightening the finished girder aresimpler than the same operations onan orthotropic structure. Therefore,

careful attention must be paid to thestability of such a large structure,which will be exposed to permanentspatial stress and dynamic loading.The risk of brittle fracture is greaterthan with the conventional and oftensmaller girder, which is usually boltedand riveted and has to withstand asmaller proportion of the dynamicload. Joints that are bolted and rivet-ed on traditional girders allow for moretolerance and are easier to fit thanwelded joints on orthotropic structures.Meeting the challenges of orthotropicdeck construction in an economical,cost-effective fashion demands metic-ulous care and great skill.

Orthotr opic Dec k Assemb lyThe most complicated aspect oforthotropic deck assembly is securingthe optimum coordinate position forthe numerous fitting edges of the unitsinvolved. The manufacturing technolo-gy should focus on this critical require-ment.

There are two aspects to the require-ment: ensuring that the dimensions ofthe components are precise, andensuring the correct positioning of thecomponents.

The dimensional accuracy of the com-ponents can be achieved in severalways. By accounting in advance forthe shrinkage that will be caused bywelding and straightening (in strictcompliance with specifications andinstructions), the components can becut to size before assembly. A moredependable, but also more expensive,solution is to cut one of the edges ofthe components with an allowance ineither one or two directions (for exam-ple, the bridge axis). After weldinghas been completed, the memberscan be cut to size, and the entireorthotropic unit straightened.

There are two ways to ensure theaccurate positioning of units. With thefirst method, orthotropic componentscan be assembled in a jig; this methodis preferred if the bridge consists of anumber of identical assembly compo-nents. An alternate method is one inwhich adjacent and consecutive unitsare assembled on a suitable bench, inreverse position, carefully joining eachpiece to be fitted. This requires moreroom and more labor than the previ-ous approach, but it is secure andsimple, and can be applied when onlya few structural components areinvolved. It is also possible to com-bine the two methods when compo-nents are assembled alongside eachother. Some simple fixturing will facili-tate the correct positioning of the com-ponents.

WeldingIn an orthotropic bridge, the deckstructure contains about 60 percent ofthe welded joints. Expert welding of

The definitive charac-teristic of orthotropic

deck design is itsunique ability to

perform all of thefunctions previouslyderived from several

separate girders forming a floor of

longitudinal and trans-versely integratedgirders coordinated

with the main girders.


the orthotropic unit requires selectionof the most suitable welding process,and a decision about the order inwhich components will be welded.The process selected will usually influ-

ence the order of welding. Dependingon the specific circumstances, any ofthe following processes may be used:

• Manual arc welding• Gas metal arc welding• Submerged arc welding• The Elin-Hafergut procedure (a patented procedure developed inthe 1930s and primarily used forconnecting longitudinal stiffeners)• A combination of the above methods.

Either one of two fundamentally differ-ent approaches may be taken withregard to the order of welding. Theentire deck structure can be assem-bled before the joints are welded, orthe longitudinal girders can be joinedinitially to the deck plate, and its cor-ners welded so that cross girders canbe placed and welded subsequently.

The first approach is more common. Ithas a significant advantage over thesecond method in that during the

course of assembly, the joining of indi-vidual units (longitudinal stiffeners andcross girders) can be secured withoutdifficulty. Since, in the course of weld-ing, only lengthwise distortion occurs,

it can be minimized by prestressing inthe jig, thus leaving a negligibleamount of straightening work. Themajor disadvantage of this approach,however, is that due to frequent inter-ruptions in welding the joints, weldingcannot be automated.

The second approach was developedto eliminate the disadvantage of thefirst, making it possible for the manylong joints of the longitudinal stiffeners

to be welded automatically. This maybe advantageous, but it must be notedthat automatic welding increases pro-ductivity only if jigs are used, and thatis expensive. The major disadvantageof this method is that erecting crossgirders between the already fixed lon-gitudinal stiffeners complicates accu-rate assembly. The deck unit maytend to bend when the cross girdersare not in place during the welding ofthe longitudinal stiffeners. Althoughthe tendency to bend in one direction(preferably the transverse) can beeliminated by prestressing, the otherbending moment inevitably requiresadditional straightening work.

Naturally, both approaches require thewelding of structural units to be in jigs.This will ensure the welding of joints incorrect positions, either flat or horizon-tally, and by correct straightening, thenumber of deformations can bereduced.

Straightening and Assemb ly The straightening of orthotropic unitsis a very costly operation, due to theirlarge dimensions and complicateddesign. To reduce the amount ofstraightening required, it is importantto choose the welding technology, theprocedure, the order of welding, andthe jigging with great care.Straightening can be done with a gasflame, local hammering, or preferably,with a hydraulic press.

To facilitate the process of on-siteassembly, finished orthotropic units(separately or with the main girders inthe full cross section of the bridge) are

Figure 1. The first welded bridge in Hungary, at Gyor, opened in 1935 (span: 53.1 m).

Figure 2. The first bridge in Hungary with a fully welded orthotropic deck spans theTisza River at Szolnok, and was opened in 1962 (spans: 54.8 + 79.3 + 54.8 m).

Figure 3. The welded lattice girder composite bridge over the Tisza atTiszafüred is composed of spans measuring 30.0 +3x70.0 + 30 m.


prefabricated to the greatest degreepossible. The facilities at the fabrica-tion plant are a decisive factor.Prefabrication provides an opportunityfor checking and correction of localroot openings, and in the case of rivet-ed or bolted joints, for reaming up jointplate holes to final size. Effects on thedesigned shape of the bridge can alsobe checked at this time.

Examples of WeldedBridges in HungaryThe very first bridge structure inHungary to be created by welding wasbuilt sixty years ago and is still in ser-vice today (Figure 1). The 53.1 mspan is of lattice girder design.

Hungary’s first up-to-date weldedbridge with a fully welded orthotropicdeck structure was built over the TiszaRiver at Szolnok and opened in 1962(Figure 2). It was fabricated usingmanual metal arc welding, with all ofthe deck units assembled and weldedin special jigs. Each cross section (6m long by 8 m wide), consisting of fourmanufacturing units, was assembledon site, using a suitable bench. Toensure the flat position, the whole

structure was placed in a prestressingrotary jig. The completed welded unitswere cut to size after being set on thetwo main girder webs, and the joints ofthese were then riveted.

On the lattice girder bridge atTiszafured, dating from 1966 (Figure3), the joints of the (box structure)upper and bottom flanges were field-welded, but the diagonals were high-

tensile bolted. The two 30 m long sidestructures were fully welded on siteusing SAW.

In 1964, the 10 m long, 4 m wideorthotropic deck units of the Erzsébetcable bridge, shown during construc-tion in Figure 4, were welded different-ly from those of the Szolnok bridge. Atfirst, only the longitudinal stiffenerswere assembled onto the deck plateand welded in a prestressing-tilting jigusing SAW. After that, the cross gird-ers were built and welded using manu-al metal arc with a covered electrode.The 27.5 m wide by 380 m longbridge, shown as it appears today inFigure 5, contains 100,000 m of weld-ed joints, 70 percent of them madewith SAW, and 30 percent with manualmetal arc.

In the composite bridge at Algyo,opened in 1974, only the main girderswere laid out in the course of preassem-bly. Because the bridge (Figure 6) waserected with a floating crane by the can-tilever method, it was very important toensure the exact positioning of all of thecross girder and wind bracing joint holeson both main girders.

The orthotropic deck plate of thebridge over the Tisza at Szeged(1979) has the longest spans of anygirder type bridge in Hungary, at 52.0,97.5, 144.0 and 78.0 m (Figure 7).

Figure 4. This photo shows the last 100 tonne deck unit of the Erzsébet cablebridge being set into place in 1964.

Figure 5. The Erzsébet cable bridge at Budapest, shown today (spans: 44.3 +290.0 + 44.3 m).


This was the first instance in which thewelding of the site joints was donewith SAW, using backing plates.

In 1985, the 120 m main span latticegirder railway bridge at Csongrad

opened (Figure 8). With 3,000 tonnesof dead load, it is the heaviest bridgeon the river Tisza.

The Ganz Company began exportingbridges in the 1930s, and since thenhas delivered almost 100 bridges toforeign countries. In the 1970s, Ganzdesigned, manufactured and erectednine bridges for Yugoslavia. The mostimpressive of them is the bridge overthe Danube at Novi Sad, shown inFigures 9 and 10. Professor NikolaHajdin designed the structure with amain span of 351 m, which at the startof manufacturing was the longest inthe world and today remains thelongest on the river Danube. The totallength of the main structure with theorthotropic deck is 590 m., with com-posite side bridges of 240 + 180 m.The dead load of the steel structure isnearly 10,000 tonnes. The bridgeopened in 1981.

The Årpád bridge at Budapest wasopened originally with two tram andtwo highway lanes. In the early 1980s,the old structure was widened on bothsides with the addition of two indepen-dent bridges, each 14 m wide. Theorthotropic deck structure is thelongest bridge in Hungary, and has adead load of 8,500 tonnes. It is shownin Figure 11 as it appears today.

The highway bridge at Haros, openedin 1990, is a box girder with reinforcedconcrete deck slab (Figure 12). Toconnect the deck with the steel gird-ers, stud welding, involving almost100,000 studs, was used in Hungary

for the first time. The bridge is 22 mwide, its total length spans 805 m, andthe dead load of the steel structure is4,500 tonnes.

The LágymányosBridgeThe Lágymányos bridge on theDanube at Budapest was completed in1995 (see cover). It is 500 m long, 30m wide, and the dead load of the steelstructure is 8,000 tonnes. The boxgirder cross section was built up from15 manufactured units. Lengthwise,there are 41 sections, so the bridgeconsists of 615 manufactured units ofdifferent types. To ensure correctdimensions and shape, 300 tonnes ofvarious jig structures were built forprefabrication and 150 tonnes of vari-ous jig structures for pre-erection andsite erection work.

Construction was carried out in the fol-lowing five major stages:

• Prefabrication of the units• Preassembly of the box girder,

without the cantilevers • Application of corrosion protection• Assembly of the cross sections • Site erection (one cross section in

two units) by floating crane

The bridge is lit according to a uniquedesign: in the center of each pylon,there is a lamp which lights mirrors atthe top of the beam. These mirrorsthen reflect light onto the bridge with-out blinding the drivers below (seeback cover).

The units of the structure and the sitejoints on the orthotropic deck werefully welded using SAW and GMAWprocedures. The other site joints werehigh-tensile bolted, but all joints of thefive pylons and the slant staying gird-ers were welded.

The Lágymányos bridge designed byDr. Tibor Sigrai contains approximately150,000 m of welded joints, more than95 percent of them produced usingautomatic or semi-automatic proce-dures. Site erection was also accom-plished using self-shielded flux coredarc welding for the root pass. It wasthe first time FCAW-s had been usedin Hungary, and we had a very goodexperience with the process. Fivepipelines, from 300 to 800 mm indiameter, were incorporated into thebridge. These were welded using theGTAW process. In addition, 22,000welded studs were used to secure thetram rails.

ConclusionHungary is a small country in centralEurope, with a population just overone percent of the total population ofEurope. Due first of all to the size ofthe country, Hungary is not destined tobe a world leader in building weldedbridges. However, the technology ofwelding has made a significant impacton our infrastructure, particularly asdemonstrated by the welded steelbridges that contribute so much to ourlandscape, and to our way of life.

Figure 6. The composite bridge over the Tisza at Algyo has spans of 57.6 +102.4 + 57.6 m.


Figure 7. The orthotropic bridge over the Tisza at Szeged(spans: 52.0 + 97.5 + 144.0 + 78.0 m).

Figure 9. A section of the bridge destined for Novi Sad,Yugoslavia, being preassembled in Budapest in 1980. Figure 12. The highway bridge over the Danube close to

Budapest is a composite structure. Opened in 1990, it hasspans of 3x73.5 + 3x108.5 + 3x73.5 m.

Figure 10. The cable stayed bridge over the Danube atNovi Sad, Yugoslavia, 1981 (spans: 4x60.0 + 2x60.0 +351.0 + 2x60.0 + 3x60.0 m).

Figure 11. The Årpád bridge over the Danube at Budapest(1981), which boasts a total length of 928 m.

Figure 8. The railway bridge at Csongrád over the Tisza(spans: 107.7 + 120.0 + 107.7 +4x42.0 m).

10

7

8

9

11

12

P.O. Box 17035Cleveland, Ohio 44117-0035

TheJames F. LincolnArc WeldingFoundation

NON-PROFIT ORG.U.S. POSTAGE

PAIDJAMES F. LINCOLN

ARC WELDING FND.

This striking bridge over the Danube at Budapest features a unique lighting design. See story on page 20.

looking ahead - jflfthe redesigned masthead emphasizes our commitment to new technology and new...

Documents