volume 10 issue 2 september 2011

72
Volume 10 issue 2 September 2011

Upload: ram-jnde

Post on 12-Mar-2016

230 views

Category:

Documents


0 download

DESCRIPTION

Volume 10 Issue 2 September 2011

TRANSCRIPT

Page 1: Volume 10 Issue 2 September 2011

Volume 10

issue 2

September 2011

Page 2: Volume 10 Issue 2 September 2011
Page 3: Volume 10 Issue 2 September 2011

1

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

from the Chief Editor

DDDDDrrrrr. Krishnan B. Krishnan B. Krishnan B. Krishnan B. Krishnan BalasubramaniamalasubramaniamalasubramaniamalasubramaniamalasubramaniamProfessor

Centre for Non Destructive EvaluationIITMadras, Chennai

[email protected]@gmail.com

URL: http://www.cnde-iitm.net/balas

This issue of the Journal of Nondestructive Testing and Evaluation has technical papers on 4separate topics of interest to the NDT community. The paper on detection of disbonds inadhesively bonded structures using thermal imaging techniques brings forth a non-contacttechnique for measurement of defects in components that are now using adhesive for joiningmetals. This technique has excellent applications for thin plate like structures. The authors use2 thermography methods and confirm their results using ultrasonic pulse echo measurements.The paper on measuring vibration inside the FBR core is an interesting and novel approachsince the conventional pulse echo method is employed in a position sensing mode for themeasurement of vibration. This method certainly finds application in solving problems whereother forms of vibration measurement are not feasible. The demonstration of this techniqueunder sodium pool is a commendable achievement. The use of frequency spectrum for extractingthe vibration values is also interesting. The paper on the inversion of conductivity vs. frequencymeasurement to predict the depth profiling of conductivity values (as a function of depth)provides a powerful tool for processes such as peening and machining for measuring the materialstate such as residual stress, hardness, etc., with depth. The fourth paper provides an industryperspective on the use of ultrasonic methods for detecting entrapped foreign objects inside thepressurized tanks with access only from the outside. The authors extend the time of flightdiffraction technique (TOFD) commonly used in the detection of cracks in welded plates to anew application for pressure vessels.

The BASICS article introduces the Impact Echo technique that is commonly used in the NDTof civil structures including piers, columns, beams, etc. The Compton Backscattering techniqueintroduced in the HORIZONS holds great promise to solve the limitation of the currentindustrial radiography procedures, by providing a single side access method for the NDT ofobjects. This technology is already becoming popular in the security screening devices, as maybe found in airports, and the feasibility of applying this technique for NDT is showing excellentresults. The NDT PUZZLE challenges the reader once again. The IQFORUM has been leftout of this issue since the previous articles has failed to invoke any response from the readers.The NEWS and the PATENTS sections will keep the audience up-to-date on the recent activitiesin the field of NDT.

The NDE2011 is scheduled for 6th -10th of December 2011 at the CHENNAI TRADECENTRE. This promises to be a MEGA event with expected registrations exceeding 1500 withextensive participation of international delegates and speakers. The industrial exhibition withabout 80+ booths will have the latest NDT tools and instruments on display. There are 4 pre-conference WORKSHOPS on Aerospace, Automobile, Concrete and Boilers. I encourage allof you to register at the earliest and plan to be a part of this EVENT. For more information,please visit www.nde2011.com.

Page 4: Volume 10 Issue 2 September 2011

2

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

PresidentShri K. Thambithurai

President-ElectShri P. Kalyanasundaram

Vice-PresidentsShri V. Pari

Swapan ChakrabortyShri D.J.Varde

Hon.General SecretaryShri R.J.Pardikar

Hon. TreasurerShri T.V.K.Kidao

Hon. Joint SecretariesShri Rajul R. Parikh

Immediate Past PresidentShri Dilip P. Takbhate

Past PresidentShri S.I.Sanklecha

MembersShri Anil V. Jain

Shri Dara E. RupaShri D.K.Gautam

Shri Diwakar D. JoshiDr. Krishnan Balasubramaniam

Shri Mandar A. VinzeShri B.B.Mate

Shri G.V. PrabhugaunkarShri B.K.Pangare

Shri M.V. RajamaniShri P.V. Sai Suryanarayana

Shri Samir K. ChoksiShri B.K. Shah

Shri S.V. Subba RaoShri Sudipta Dasgupta

Shri N.V. WagleShri R.K. Singh

Shri A.K.Singh (Kota)Shri S. Subramanian

Shri C. AwasthiBrig. P. GaneshamShri Prabhat Kumar

Shri P. MohanShri R. SampathShri V. SathyanShri B. Prahlad

Ex-officio MembersManaging Editor, JNDT&E

Shri V. Pari

Chairman, NCB &Secretary, QUNEST

Dr. Baldev Raj

Controller of Examination, NCBDr. B. Venkatraman

President, QUNESTProf. Arcot Ramachandran

All Chapter Chairmen/Secretaries

Permanent InviteesShri V.A.Chandramouli

Prof. S. RajagopalShri G. Ramachandran

& All Past Presidents of ISNT

I S N T - National Governing Council

Chapter - Chairman & SecretaryAhmedabadShri D.S. Kushwah, Chairman,NDT Services, 1st Floor, Motilal Estate,Bhairavnath Road, Maninagar,Ahmedabad 380 028. [email protected] Rajeev Vaghmare, Hon. SecretaryC/o Modsonic Instruments Mfg. Co. Pvt. Ltd.Plot No.33, Phase-III, GIDC Industrial EstateNaroda, Ahmedabad-382 330 [email protected]

BangaloreProf.C.R.L.Murthy, ChairmanDept. of Aerospace Engg, Indian Institute of Science,Bangalore 560012Email : [email protected]@aero.iisc.ernet.in

ChennaiShri R. Sundar, ChairmanDirector of Boilers,Tamil NaduShri R. Balakrishnan, Hon. Secretary,No.13, 4th Cross Street, Indira Nagar,Adyar, Chennai 600 020. [email protected]

DelhiShri Ashok Singhi, Chairman,MD, IRC Engg Services India Pvt. Ltd612, Chiranjiv Tower 43, New [email protected] Dinesh Gupta, Hon.Secretary,Director, Satya Kiran Engg. Pvt. LtdBU 3 SFS Pitampura, New Delhi [email protected]

HyderabadShri M. Narayan Rao, Chairman,Chairman & Managing Director, MIDHANI,Kanchanbagh, Hyderabad 500 [email protected] Jaiteerth R. Doshi, Hon.Secretary,Scientist, Project LRSAMDRDL, Hyderabad 500 [email protected]

JamshedpurDr N Parida, Chairman,Senior Deputy DirectorHead, MSTD, NML, Jamshedpur - 831 [email protected]. GVS Murthy, Hon. Secretary,MSTD, NML, [email protected] / [email protected]

KalpakkamShri YC Manjunatha, ChairmanDirector ESG, IGCAR, Kalpakkam – 603 [email protected] BK Nashine, Hon.SecretaryHead, ED &SS, C&IDD, FRTGIGCAR, Kalpakkam – 603 102 [email protected]

KochiShri CK Soman, Chairman,Dy. General Manager (P & U),Bharat Petroleum Corporation Ltd. (Kochi Refinery),PO Ambalamugal 682 302. [email protected] V. Sathyan, Hon. Secretary,SM (Project),Bharat Petroleum Corporation Ltd. (Kochi Refinery),PO Ambalamugal-682 302. [email protected]

KolkataShri Swapan Chakraborty, ChairmanPerfect Metal Testing & Inspection Agency,46, Incinerator Road, Dum Dum Cantonment,Kolkata 700 028. [email protected] Dipankar Gautam, Hon. Secretary,4D, Eddis Place, Kolkata-700 [email protected]

KotaShri R.C. Sharma, ChairmanAssociate Director (QA),Rawatbhata 323 307 [email protected] S.V. Lele, Hon. Secretary,T/IV – 5/F, Anu Kiran Colony, PO Bhabha Nagar,Rawatbhata 323 307. [email protected]

MumbaiShri R.S. Vaghasiya, Chairman,B 4/7, Sri Punit Nagar, Plot 3, SV Road, Borivile West,Mumbai 400 092. [email protected] Samir K. Choksi, Hon. Secretary,Director, Choksi Brothers Pvt. Ltd.,4 & 5, Western India House, Sir P.M.Road,Fort, Mumbai 400 001. [email protected]

NagpurShri Pradeep Choudhari, ChairmanParikshak & Nirikshak, Plot M-9, LaxminagarNagpur - 440 022Mr. Jeevan Ghime, Hon. Secretary,Applies NDT & Tech Services,33, Ingole Nagar, B/s Hotel Pride, Wardha Road,Nagpur 440 005. [email protected]

PuneShri PV Dhole, ChairmanNDT House, 45 Dr Ambedkar Road,Sangam Bridge, Pune- 411 [email protected] VB Kavishwar, Hon Secretary,NDT House, 45 Dr Ambedkar Road,Sangam Bridge, Pune- 411 [email protected]

SriharikotaShri S.V. Subba Rao, Chairman,General Manager, Range OperationsSDSL, SHAR CentreSriharikota 524124. [email protected] G. Suryanarayana, Hon. Secretary,Dy. Manager, VAB, VAST, Satish Dhawan SpaceCentre, Sriharikota-524 124. [email protected]

TarapurShri PG Behere, Vice Chairman,AFFF, BARC, Tarapur-401 [email protected] Jamal Akftar, Hon.Secretary,TAPS 1 & 2, NPCIL, Tarapur. [email protected]

TiruchirapalliR.J. PardikarAGM, (NDTL)BHEL Tiruchirapalli 620 014. [email protected] A.K.Janardhanan, Hon. Secretary,C/o NDTL Building 1, H.P.B.P., BHEL,Tiruchirapalli 620 014. [email protected]

VadodaraShri P M Shah, Chairman,Head-(QA) Nuclear Power Corporation [email protected] S Hemal Thacker, Hon.Secretary,NBCC Plaza, Opp.Utkarsh petrol pump, Kareli Baug,Vadodara-390018. [email protected]

ThiruvananthapuramDr. S. Annamala Pillai, ChairmanGroup Director, Structural Design & Engg Group,VSSC, Thiruvananthapuram [email protected]. Imtiaz Ali KhanHon.Secretary, Engineer, Rocket Propellant Plant,VSSC, Thiruvananthapuram 695 [email protected]

VisakhapatnamShri Om Prakash, Chairman,MD, Bharat Heavy Plate & Vessels Ltd.Visakhapatnam 530 012.Shri Appa Rao, Hon. Secretary,DGM (Quality), BHPV Ltd., Visakhapatnam 530 012

Page 5: Volume 10 Issue 2 September 2011

Contents

Chief EditorProf. Krishnan Balasubramaniame-mail: [email protected]

Co-EditorDr. BPC [email protected]

Managing EditorSri V Parie-mail: [email protected]

Topical EditorsDr D K BhattacharyaElectromagnetic MethodsDr T Jayakumar,Ultrasonic & Acoustic EmissionMethodsSri P KalyanasundaramAdvanced NDE MethodsSri K ViswanathanRadiation Methods

18

23

25

27

5

About the cover page:

Editorial BoardDr N N Kishore, Sri Ramesh B Parikh,Dr M V M S Rao, Dr J Lahri,Dr K R Y Simha, Sri K Sreenivasa Rao,Sri S Vaidyanathan, Dr K Rajagopal,Sri G Ramachandran, Sri B Ram Prakash

Advisory PanelProf P Rama Rao, Dr Baldev Raj,Dr K N Raju, Sri K Balaramamoorthy,Sri V R Deenadayalu, Prof S Ramaseshan,Sri A Sreenivasulu, Lt Gen Dr V J Sundaram,Prof N Venkatraman

ObjectivesThe Journal of Non-Destructive Testing &Evaluation is published quarterly by the IndianSociety for Non-Destructive Testing for promotingNDT Science and Technology. The objective ofthe Journal is to provide a forum fordissemination of knowledge in NDE and relatedfields. Papers will be accepted on the basis oftheir contribution to the growth of NDE Scienceand Technology.

Journal of Non DestructiveTesting & Evaluation

Published byShri RJ Pardikar,General Secretary on behalf ofIndian Society for Non Destructive Testing (ISNT)

The Journal is for private circulation to membersonly. All rights reserved throughout the world.Reproduction in any manner is prohibited. Viewsexpressed in the Journal are those of the authors'alone.

Modules 60 & 61, Readymade GarmentComplex, Guindy, Chennai 600032Phone: (044) 2250 0412Email: [email protected] at VRK Printing House3, Potters Street, Saidapet,Chennai 600 015 [email protected]: 09381004771

Volume 10 issue 2September 2011

Micro IR Imaging onMEMS Devices

The cover page for this issue shows athermography image of a Micro-Electronic-

Mechanical-Sensor (MEMS) as-fabricated wafer.The MEMS technology is now becoming very

common method for fabrication of mechanicaldevices such as micro-motors, micro-pumps,

etc. that are extremely small in size.MEMS based sensors are also found to be

useful in measuring pressure, acceleration,temperature, etc. These devices are so small

that they can be embedded inside humanbodies, inside aircraft structures, and provide

real-time data logging. They are fabricatedusing micro-lithogragphy methods that are

similar to the way IC circuits in computers arefabricated, only not quite so small.

The image is obtained using a micro-lensattached to a FLIR IR Camera at 6 micron

resolution. This was done in PASSIVE mode.The fractures in the cantilever sensors can be

observed using this imaging technique.(Courtesy: Centre for NDE at IIT Madras)

47

53

57

65

68

Basics - Impact Echo Technique

Horizon - NDE by Backscatter Imaging

Chapter News

NDE Events

NDE Patents

NDT Puzzle

Technical PapersNon destructive detection of debonding in adhesivelybonded metal/ceramic composite platesSony Punnose, Amretendu Mukhopadhyay,B. Nagaraja Kowmudi, P. Rama Subba Reddy,V. Madhu and Vikas Kumar

NDE Technique for Reactor Core VibrationMeasurement in FBRsR. Ramakrishna, P. Anup Kumar, M. Anandaraj,M. Thirumalai, V. Prakash, C. Anandbabu andP. Kalyanasundaram

On the conversion of multi-frequency “apparent”conductivity data to actual conductivity gradients onpeened samplesVeeraraghavan Sundararaghavan, KrishnanBalasubramaniam

Ultrasonic Non-Destructive Evaluation (NDE) basedinternal inspection of pressure vessels for bettermaintenance practiceS.K.Nath and B.H.Narayana

Probe

33

Page 6: Volume 10 Issue 2 September 2011

4

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Madras Metallurgical Services (P) LtdMetallurgists & Engineers

Serving Industries &Educational

Institutes for thepast 35 years

24, Lalithapuram street, Royapettah, Chennai 600014Ph: 044-28133093 / 28133903 Email: [email protected]

A-3, Mogappair Indl. Area (East) JJ Nagar,Chennai 600037 Phone 044-26564255, 26563370

Email: [email protected];[email protected] www.kidaolabs.com

KIDAO Laboratories

Scaanray Metallurgical Services(An ISO 9001-2000 Certified Company)

NDE Service ProviderProcess and Power Industry, Engineering andFabrication Industries, Concrete Structures,

Nuclear Industries, Stress Relieving

Transatlantic Systems

Electro-Magfield Controls & Services &LG Inspection Services

Plot 165, SIDCO Industrial Estate, (Kattur)Thirumullaivoil, Vellanur Village, Ambattur Taluk

Chennai 600062 Phone 044-6515 4664 Email: [email protected]

We manafucture : Magnetic Crack Detectors, Demagnetizers, MagneticParticles & Accessories, Dye Penetrant Systems etc

Super Stockist & Distributors: M/s Spectonics Corporation, USA fortheir complete NDT range of productrs, Black Lights, Intensity Meters,

etc.

Betz Engineering &Technology Zone

An ISO 9001 : 2008 Company

Call M. Nakkeeran, Chief Operations,Lab: C-12, Industrial Estate, Mogappair (West), Chennai 600037

Phone 044-2625 0651 Email: [email protected] ;www.scaanray.com

Support for NDT ServicesNDT Equipments, Chemicals and Accessories

Call DN Shankar, Manager14, Kanniah Street, Anna Colony, Saligramam,

Chennai 600093Phone 044-26250651 Email: [email protected]

49, Vellalar Street, near Mount Rail Station, Chennai 600088Mobile 98401 75179, Phone 044 65364123Email: [email protected] / [email protected]

International Training Division21, Dharakeswari Nagar, Tambaram to Velachery Main Road,Sembakkam, Chennai 600073www.betzinternational.com / www.welding-certification.com

NABL Accredited Laboratory carrying out Ultrasonic test,MPL and DP tests, Coating Thickness and Roughness test.

We also do Chemical and Mechnical testsMetallographyStrength of MaterialsNon Destructive TestingFoundry Lab

Shri. K. Ravindran, Level IIIRT, VT, MT, PT, NR, LT, UT, ET, IR, AE

Southern Inspection ServicesNDT Training & Level III Services in all the

following ten NDT Methods

No.2, 2nd Floor, Govindappa Naicker Complex,Janaki Nagar, Arcot Road,

Valasaravakkam, Chennai-600 087Tamil Nadu, India

Phone : 044-2486 8785, 2486 4481E-mail: [email protected] and

[email protected] Website: www.sisndt.com

OP TECH01J, First Floor, IITM Research Park, Kanagam Road, Taramani,

Chennai 600113 India Phone : +91 44 6646 9880

Dhvani R&D Solutions Pvt. Ltd

Educational CDs -PT, UT, RT, MT, ET, BasicMetallurgy and Mechanical Testing

ASNT Level III Intensive Taining

Call93828 12624

Land044 - 2446 1159

B Ram PrakashA 114, Deccan Enclave,72, T M Maistry Street,Thiruvanmiyur, Chennai 600 041

• Inspection Solutions - CUPS, TAPS, CRISP, TASS• Software Products - SIMUT, SIMDR• Training - Guided Waves, PAUT, TOFD• Services & Consultancy - Advanced NDE, Signal Processing

- C-scans, On-line Monitoring

E-mail: [email protected] www.dhvani-research.com

Classifieds

Page 7: Volume 10 Issue 2 September 2011

5

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

displacement transducer locatedclose to the impact point is used tomonitor the surface displacementscaused by the arrival of thesereflected waves. P-waves are ofprimary importance in the impactecho testing of concrete structures,because the displacements caused byP-waves are much larger than thosecaused by S-waves at points locatedclose to the impact point [3, 5, 6].The amplitude of the reflected P-wave, Areflected is given by

(1)

Where Z1 and Z2 are the acousticimpedances of the regions in whichthe wave is approaching the interfaceand of the region beyond theinterface respectively, and A1 is theamplitude of the particle motion inincident wave. The phase of thereflected wave depends upon theratio of Z2/Z1. If Z2/Z1 is less thanone (at concrete/air interface; Z2/Z1= 5 x 10-5), then the phase reversiontakes place at the interface. Becauseof the phase reversion at both theinterfaces, waves reflected betweentwo concrete/air interfaces producesuccessive arrival of tension wave(inward displacement) at the impactsurface.

The frequency of P-wave arrivals atthe transducer is determined bytransforming the time domain signalinto the frequency domain using thefast Fourier transform technique. Thefrequencies associated with the peaksin the resulting amplitude spectrumrepresent the dominant frequenciesin the waveform.

2. EQUIPMENT

The impact echo equipment consistsof (a) a hand held unit containing animpacting device (steel ball) forproducing low frequency stresswaves (sound waves) and a pair ofpiezoelectric transducers that detectsurface displacements caused byreflected waves, (b) a high speed,

integrity of the member can beassessed. The method is subjective,as it depends on the experience ofthe operator, and it is limited todetecting near surface defects.Despite these inherent limitations,sounding is a useful method fordetecting near-surface delaminations,and it has been standardized byASTM [4]. Impact echo techniquemay be considered as a sophisticatedand instrumented advancement of thesounding technique. It involvesintroducing a transient stress pulseof predefined frequency band-widthinto a test object by selecting thesize of the spring loaded steel ballimpactor and monitoring the surfacedisplacements using a dry-coupledpiezoelectric receiver with anintegrated amplifier. The obtainedimpact pulse is analyzed online withthe help of a data acquisition systemand a computer with necessary signalanalysis software.

The impact pulse consists ofcompression (P) and shear (S) waveswhich propagate into the objectalong spherical wave fronts, and aRayleigh (R) wave that propagatesalong the surface. These waves arereflected by internal defects and theboundaries and the reflected wavespropagate back to the surface. At thetop surface, the waves are reflectedagain and they once again propagateinto the test object. Thus, a transientresonance condition is setup bymultiple reflections of wavesbetween the top surface and internalflaws or external boundaries. A

1. INTRODUCTION

Concrete is a composite materialconsisting of a binding medium withaggregates like gravel, sand etc.embedded in the medium. Ultrasonicand impact-echo test methods aretwo important methods that arewidely used for the nondestructiveexamination of concrete structures[1]. Testing of thick concretestructures using ultrasonic techniqueis often difficult due to largescattering and attenuation of thesound energy in the medium thatresults in poor signal-to-noise ratioof the reflected signal amplitudes.Also, in thick structures, ultrasonicthrough transmission technique issuggested. This needs accessibilityof both surfaces as well as properalignment of both the transducers,which is quite difficult if notimpossible. To overcome theselimitations and to reliably examinethe concrete structures, impact-echotest method was developed in themid 1980’s as a nondestructive testmethod [2, 3]. Since its development,impact echo technique has found itswide applicability in NDE ofconcrete structures [2, 3].1.1 PRINCIPLE OF IMPACT

ECHO TECHNIQUE

Tapping an object with a hammer isone of the oldest forms ofnondestructive testing techniquesbased on stress wave propagation.Depending on whether the result isa high-pitched “ringing” sound or alow frequency “rattling” sound, the

Dr. T. JayakumarMetallurgy and Materials GroupIndira Gandhi Centre for Atomic Research, Kalpakkam-603102, India

Co-ordinated by Prof. O. Prabhakar

Basics

Impact echo technique

Page 8: Volume 10 Issue 2 September 2011

6

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

analog to digital data acquisitionsystem that receives and digitizes theanalog voltage signal from thetransducers and transfers it to thecomputer, (c) a portable computer,and (d) a software program thatmonitors each test, and guides thedata processing to produce outputdisplays viz., time signal, frequencyspectrum, and frequency amplitudeas a function of % depth, thatprovides information about thestructure being tested. Figure 1shows various components of atypical impact echo system.2.1 IMPACTOR

The impactor is usually spherical orspherically tipped. The spring loadedhardened steel balls of variousdiameters in the range of 1 mm to16 mm can be used for generatingthe stress waves of differentfrequency ranges in the specimen.The highest frequency of thegenerated stress waves (fmax) can beestimated by the following equation[7]:

(2)

Where, fmax is in kHZ and D is thediameter of the hardened steel ballin meters.

The impact delivers sufficient energyinto the concrete structure so that awell defined amplitude spectrum isobtained with predominant peaks.The impact duration, tc must be lessthan the round trip travel time for aP-wave during the impact echotesting, i.e.

Where, T is the thickness of thespecimen and Cp is the P-wavevelocity in the concrete specimen.

The smaller diameter ball generateshigher frequency wave with lowenergy and is more useful for thenear surface defect detection,whereas, the larger diameter ball

generates lower frequency wave withmore energy and hence is moreuseful for inspection up to largerdepths. A 3 mm diameter impactor(the smallest practical size is about1.5 mm in diameter) produces stresswaves with useful frequencies up toalmost 90 kHz, and wavelengths assmall as 0.04 m. At the high end ofthis frequency range, the stresswaves begin to get scattered andreflected by the naturalinhomogeneous regions in concrete,such as small air inclusions andmortar/aggregate interfaces, with theresult that there is more “noise” inthe waveform and spectrum. Anotherconsideration in selecting animpactor is the relationship betweenthe wavelength and size of a flaw ordiscontinuity that can be detected. Aflaw of lateral dimension l is“invisible” to stress waves ofwavelength greater than l.Combining the fundamentalrelationship between wave speed,frequency and wavelength, Cp = fl(where l is the wavelength), with Eq.2 and by using a wave speed of 4000m/s in concrete, the minimumdetectable lateral dimension of theflaw (Lmin) is given by

Lmin = 14D (3)

Thus, the minimum lateral size of aflaw that can be detected, is about14 times the diameter of theimpactor.2.2 RECEIVER TRANSDUCER

Two broadband, piezoelectrictransducers which respond to normalsurface displacement are used. Thesetransducers are capable of detectingthe small displacements thatcorrespond to the impact generatedP-wave traveling along the surfaceduring the velocity measurement andthe reflected P-wave from theinterface during the impact echotesting. The transducer with apiezoelectric element of a smallcontact area (a tip diameter of 1.5mm) and the larger end attached to

a brass backing block has been foundsuitable. A suitable material isrequired to be used to couple thetransducer to the concrete. A leadsheet approximately 0.25 mm thickis a suitable coupling material forsuch a transducer.2.3 SPACER DEVICE

A spacer device is used to hold thetransducer at a fixed distance apartduring measurement of the wavevelocity, as shown in Fig. 1. Thetransducer tips shall be placed about300 mm apart and the actual distancebetween the tips of the transducersbe measured and recorded to thenearest 1 mm.2.4 DATA ACQUISITION SYSTEM

AND ANALYSIS SOFTWARE

This consists of hardware andsoftware for acquisition, recordingand processing of the signal outputof the two transducers. This systemcan be a portable computer with atwo channel data acquisition card ora portable two channel waveformanalyzer. The sampling rate for eachchannel is at least 500 kHz. Thesystem is capable of triggering onthe signal from one of the recordingchannels and acquisition of upto2048 data points. Higher samplingrate increases the accuracy in thetime of flight measurement andlarger data length increases thefrequency resolution.

Figure 2 shows the front panel ofthe software developed at theauthors’ laboratory for impact echotesting. It consists of acquisition ofimpact pulse for a duration of about2 – 5 μs depending upon thethickness of the structure,transforming the signal to frequencydomain and then converting thefrequency spectrum to a depth vs.amplitude graph. The software alsocalculates the expected peaks in thefrequency spectrum based on thethickness and the shape of thestructure. The software consists ofvarious additional features such as

Basics

Page 9: Volume 10 Issue 2 September 2011

7

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

Basics

Fig. 1 : Components of a typical impact echo system.

Fig. 2 : Front panel of the software for impact-echo testing showing the thickness response of a concrete block of 400 mm thickness.

selecting suitable time and frequencywindows for analysis. One of theimportant additional features in thedeveloped software as compared tothe commercial software is thedisplay of depth vs. amplitudespectrum in a linear depth scale. Thisallows clear visualization of thelocation of a defect in the structure,if any found.

3. MEASUREMENT OFP-WAVE VELOCITY

The P-wave velocity needs to bedetermined in the structure before Fig. 3 : Experimental setup for P-wave velocity measurement using procedure A.

Page 10: Volume 10 Issue 2 September 2011

8

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

the start of the impact echo testing.It can be measured in two differentways: by using a two reeiver systemfor the measurement of velocity onthe surface (procedure A) and, byusing a single transducer to measurethe thickness averaged velocity(procedure B). The details of thesemeasurements are given below:3.1 PROCEDURE A

This procedure measures the timetaken by a wave to travel betweentwo transducers positioned at aknown distance apart (300 mm)along the surface of a structure. Theschematic of the experimental set upfor this procedure is shown in Fig.3. The wave velocity is calculatedby dividing the distance between thetwo transducers by the travel time.The P-wave velocity can either becalculated directly by measuring thetravel time of P-wave or bymeasuring the travel time of R-waveand then using the relation betweenR-wave and P-wave velocities,knowing the Poisson’s ratio.

3.1.1 MEASUREMENT OFTRAVEL TIME OF P-WAVE

Figure 4a shows the signal acquiredfor measurement of P-wave velocitydirectly. The procedure fordetermination of P-wave velocity bydirectly measuring the travel time ofP-wave is as follows: As the P-wavegenerated upon the impact is

compression in nature, its arrival canbe identified by the first rise involtage (T1 in Fig, 4a). Thedifference in the arrival time of theP-waves at the two transducers (T2-T1) can be measured by moving thecursors (Fig. 4a) to locate the risingpoints in the two voltage plots andthis can be used to calculate the P-wave velocity (Cp) as follows:

Cp= d/(T2-T1) (4)

Where d is the distance between thetwo receivers.

3.1.2 MEASUREMENT OFTRAVEL TIME OF R- WAVE

Figure 4b shows the signal acquiredfor measurement of R-wave velocity.The procedure for determination ofP-wave velocity by measuring thetravel time of R-wave is as follows:The arrival of the R-wave isidentified by a sharp drop in thevoltage due to the tensile nature andhigh amplitude of the R-wave (T3and T4 in Fig. 4b). The difference inthe arrival time of R-waves at thetwo transducers (T4-T3) can bemeasured by bringing the cursors tothe first sharp dropping points ineach voltage plot (Fig. 4b). The R-wave velocity can be calculated fromthis as follows:

CR= d/(T4-T3) (5)

and P-wave velocity can be

determined using the followingequation [8]:

(6)

where, n is the Poisson’s ratio. Theaverage value of Poisson’s ratio forhardened concretes is 0.2.

3.2 PROCEDURE B

The principle of this procedure issimilar to that of impact-echo testingon a concrete structure of knownthickness. Impact on the surface ofthe concrete generates stress wavesof which P-wave is of primaryimportance. The P-wave propagatesinto the concrete and is reflectedfrom the opposite surface. Multiplereflections of the P-wave betweenthe two surfaces give rise to atransient thickness resonance with afrequency related to the thickness ofthe concrete and wave velocity inthe concrete. A receiving transducer,located adjacent to the impactposition, receives the reflected waveand the output of the transducer iscaptured as a time domain waveform.The frequency of the P-wave arrivalsat the transducer is determined bytransforming the time domain signalinto the frequency domain using thefast Fourier transform technique. Thefrequencies associated with the peaksin the resulting amplitude spectrum

Basics

(a) (b)

Fig. 4 : Signal acquired and procedure for determination of P-wave velocity by a) directly measuring the travel time of P-wave andb) by measuring the travel time of R-wave.

Page 11: Volume 10 Issue 2 September 2011

9

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

represent the dominant frequenciesin the waveform. P-wave velocitycan be determined by using thefollowing equation [2]:

(7)

Where, f is the thickness frequency(the frequency of the thickness modeof vibration), T is the thickness andâ is the shape factor for the thicknessmode of vibration. The values of âfor various structures are given inTable 1.

4. MEASUREMENT OFTHICKNESS OFCONCRETE STRUCTURE

The principle of impact echo testingfor measurement of thickness ofconcrete structure is similar to thatexplained in section 3.2 (ProcedureB for the measurement of the P-wavevelocity). Before carrying out thethickness measurement, the wavevelocity in the specimen isdetermined using either theprocedure A or B, as described inthe previous section. Even though theprocedure B is more accurate, it canbe used only if the thickness isknown in the same structure at anyother given location. Then the impactecho test will be carried out asexplained in section 3.2. Thethickness of the structure isdetermined by using the following

similar to that discussed in sections3.2 (procedure B of velocitymeasurement) and 4 (thicknessmeasurement). The presence of aflaw changes the patterns of stresswave propagation and reflection.These changes shall get reflectedboth in the waveforms and spectraobtained from impact-echo tests, andthey provide both qualitative andquantitative information about theflaws. This section focuses on signalbehavior associated with cracks and

Basics

Table 1 Various shapes as defined during impact echo testing ofconcrete structures and corresponding shape factor (ß).

Structure shape Definition Shape Factor (â)

Plate A structure with two parallel faces, for which the ratio of lateraldimension to thickness is sufficiently large, so that multiple wavereflections from the side boundaries do not reach the transducer withinthe few milli-seconds in which multiple reflections between the twofaces are being recorded. When 1024 data points are recorded at aninterval of 2 μs, this ratio is five to six. 0.96

Bar of circular cross section A structure in which the length is at least three times the diameter 0.92

Bar of square cross section A structure in which the length is at least three times the thickness 0.87

Bar of rectangular cross section A structure in which the length is at least three times the largest Function ofdimension in the cross section depth/

breadth (D/B)

Fig. 5 : A crack at a depth d gives the same response as a void at that depth

equation obtained by re-arrangingEq. 7.

(8)

5. TESTING OF CONCRETESTRUCTURES FORDETECTION OF FLAWS/DEFECTS

The experimental setup for detectionof flaws in concrete structures is

Page 12: Volume 10 Issue 2 September 2011

10

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

and voids are similar, because stresswaves are reflected from the firstconcrete/air interface encountered.Thus, a crack at a depth (d) will givethe same response as a voidwhose upper surface (nearest to theimpact surface) is at the same depth(Fig. 5).

Fig. 6 : Comparison of the solid response (left) with the response from a region with a crack of wide lateral extent.

spectrum can be generalized to anygeometry.

A crack or void within a concretestructure forms a concrete/airinterface. Cracks with a minimumwidth (crack opening) of about 0.08mm cause almost total reflection ofa P-wave. The responses from cracks

voids in plates, including the specialcase of a shallow crack ordelamination, and the response ofplates containing unconsolidatedconcrete (honeycombing). Althoughthe discussion is focused on platestructures, the interaction of stresswaves with flaws and the resultingchanges in the waveform and

Fig. 7 : Comparison of solid response (left) with response in the vicinity of a flaw of limited lateral extent.

Basics

Page 13: Volume 10 Issue 2 September 2011

11

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

Fig. 8 : The principle components of the response produced by impact on the surface of a concrete slab containing a shallowdelamination: a) flexural mode, and b) thickness mode. The contributions to the spectrum are shown in c) and d).

surface displacements are far largerthan those caused by P-wave arrivalsthey dominate the signal. The higher-frequency component due to multipleP-wave reflections across the thinlayer is weak by comparison, andthus making it sometimes difficultto detect.

A schematic representation of theeffects of flexural vibration in thinlayer is shown in Fig. 8. Flexuralvibration, shown in (a), has lowfrequency (typically 2- 6 kHz) andvery large amplitude as compared tosurface displacements caused by thearrival of reflected P-waves, shownin Fig. 8(b). Figures 8 (c) and (d)show the correspondingcontributions to the spectrum:flexural vibrations produce a high-amplitude, low frequency signals thatdominate the waveform andspectrum, while the peak resultingfrom P-wave reflections has a higherfrequency and lower amplitude, andis sometimes too small to be seen.There are two methods foramplifying this high frequency peak:(1) using a smaller impactor and (2)digital high pass filtering. In Fig. 8d,

the full thickness. However, the fullthickness frequency is lower thanthat of the solid plate because of thereduced stiffness in the vicinity ofthe crack and because the P-wavesmust travel a longer path around thecrack to reach the bottom surface.The difference between the solidresponse and the response when acrack of small lateral extent ispresent, is illustrated in Fig. 7. Whenthe depth of the flaw is greater thanabout 10 cm, the response frommultiple P-wave reflections withinthe layer above the flaw is relativelystrong. If the depth of the flaw isless than 10 cm, flexural vibrationsin the thin layer are often excited,and the response is dramaticallydifferent. This situation, for example,is frequently encountered onconcrete bridge decks, where widespread cracking –calleddelamination- occurs at shallowdepths due to corrosion in thereinforcing steel. The resulting signalincludes a large-amplitude, lowfrequency component due to theflexural vibration. The flexuralvibrations are similar to the vibrationin a drum, and because the resulting

When tests are carried out to locateflaws in a plate structure, the firststep is to determine the response ofthe solid structure. This isaccomplished by performing tests ina region where the structure is knownto be solid. If the thickness and wavespeed are known, the thicknessfrequency can be calculated, andtests can be performed until a solidresponse is obtained. The differencebetween the solid response and theresponse when a crack of widelateral extent (large surface areaparallel to the top surface) is presentis illustrated in Fig. 6. The depth ofthe flaw can be determined using Eq.9, as given below:

(9)

The impact echo testing is alsoinfluenced by the depth at which aflaw is located. When the lateraldimensions of a crack arecomparable to its depth, stress wavesare both reflected from the crack anddiffracted around it. As a result, P-wave reflections occur both withinthe layer above the crack and across

Basics

Page 14: Volume 10 Issue 2 September 2011

12

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

successive arrival of tension wave(inward displacement) at the impactsurface due to the phase reversion atboth the interfaces. Whereas, wavesreflected between a concrete/airinterface and a concrete/steelinterface produce arrival of alternatecompression wave (outwarddisplacement) and tension wave(inward displacement) at the impactsurface, which makes the frequencyof the arrival of P-wave of similarnature (compression/tension) half tothat when reflected between twoconcrete/air interfaces. It is thisdifference in the frequency ofarrivals of similar nature of P-wavethat is used for the identification ofdelamination of reinforced bars, asthe delamination changes the typeof interface from concrete/steel toconcrete/air.

Figure 10 (a) shows the typicalimpact echo signal of a reinforcedrod intact with the concrete, at adepth of 50 mm. The frequencyspectrum consists of a cluster ofpeaks centered at 17.1 kHz (116mm). These clustered peaks are thecharacteristic response of steelreinforcement rods withoutdelamination [9]. As discussed

6. TESTING OF CONCRETESTRUCTURES FORDELAMINATION OFREINFORCED BARS

The experimental setup for thetesting of concrete structures fordetection of delamination ofreinforced rods is similar to that ofimpact echo testing as describedearlier. The waves reflected betweentwo concrete/air interfaces produce

the high frequency peak is amplifiedby applying a digital high pass filterto the same signal for whichamplitude spectrum is shown in Fig.8c.

A region of unconsolidated concretetypically consists of large number ofsmall, interconnected voids,commonly referred to as“honeycombing”. Such areas includemany small concrete/air interfacesover a range of depths, and oftenthey do not have a well-definedexternal boundary. The usualresponse of a honeycombed regionto an impact-echo test includes a“displaced thickness frequency” –that is, a strong peak at a frequencysmaller than that of the solid plate –and one or more additional peakscorresponding to P-wave reflectionsfrom a range of depths within theunconsolidated region. The responseof honeycombing at about 130 mmin a 250 mm thick concrete plate isshown in Fig. 9. Multiple peakscentered at about 15 kHz (130 mmthickness) and a displaced thicknessfrequency peak at 5 kHz can be seenin Fig. 9. The expected location ofthe solid thickness frequency isshown as a vertical cursor at 8 kHz.

Fig. 10 : Response of reinforced rod (a) without and (b) with delamination

Basics

Fig. 9 : Response of honetcombing at about 130 mm (15 kHz) in a concrete block of250 mm thickness (solid thicknes frequency = 8 kHz).

Page 15: Volume 10 Issue 2 September 2011

13

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

without delamination) that, if thelocation of the reinforcement rod isknown apriori, then it can be easilychecked by impact echo testing thatwhether any delamination hasoccurred.

7. DETERMINING THEDEPTH OF SURFACEOPENING CRACKS

The experimental setup for the depthmeasurement of surface openingcrack, as shown in Fig. 11a, is similarto that for the velocity measurement.The only difference is that the crackis kept in between the twotransducers and the impact is givenin between the crack and thetransducer. When an impact isperformed, the waves generatedtravel on the surface and picked upby the transducer 1 on the same sideof the crack. Figure 11b shows thesignals received by the twotransducers for the crack depth of25 mm. The waves cannot reach thetransducer 2 directly on the surfaceand the first wave to reach transducer2 is the P-wave diffracted at thecrack tip (Fig. 11a). The time lagbetween the P-wave to reach directlyto transducer 1 and after diffractionto transducer 2 can be calculatedfrom Fig. 11 directly. Using the P-wave velocity and the time lag, thedepth of the surface opening crackcan be determined by the followingequation [10]:

Depth = ((CP x Dt)2/4 – H2)1/2 (11)

Where CP, Dt and H are the P-wavevelocity, time lag and the constantdistance between the crack and thetransducer 2 (= the distance betweenthe crack and the impact point = thedistance between the impact pointand the transducer 1), respectively.Using this equation, the depth of thecrack is found to be 28 mm, whichis very close to the actual depth ofthe crack [11]. Similarly the depthof the surface opening cracks ofdepth 50 mm and 75 mm could also

going around the rod. This has moreimportance when the D/t ratio ismuch less than 1.

Figure 10(b) shows the typicalresponse of a delaminated rod at adepth of 35 mm. The frequencyspectrum shows the dominant peaksat 6.3 kHz and 59.6 kHz (33 mm).The peak at 59.6 kHz is due to thereflection from concrete/air interfaceat the depth of 35 mm (delamination)and the peak of 6.3 kHz correspondsto the flexural vibration of theconcrete cover of 35 mm thickness.This is a typical response, if thesection above the delamination isthin (less than 100 mm), as discussedearlier [9].

Hence, it is clear from the responseof the reinforcement rod (with and

earlier; the frequency of the arrivalof the wave of similar nature is half,and hence Eq. 9 leads to the depthof the reinforced rod as double ofthat of the actual depth. Further,the exact depth of the reinforcementrod can be determined by applyingthe correction factor as per theequation given below [9]:

Depth calculated fromfrequency spectrum = t (-0.6 D/t + 1.5) (10)

Where D and t are the diameter andthe actual depth of the reinforcementrod, respectively. The abovecorrection factor is needed totake into account the interference ofthe wave reflecting from theconcrete-steel interface, steel-concrete interface and the waves

Fig. 11 : (a) Experimental setup and (b) waveform for depth measurement of the surfaceopening crack

Basics

Page 16: Volume 10 Issue 2 September 2011

14

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Fig. 12 : Photograph of (a) Mock up block-1, (b) voids and (c) reinforcement rods in it.

Fig. 13 : mpact echo response of (a) reinforcement rod and (b) duct sheath in ring beam of inner containment wall.

be determined with an accuracy ofabout ±10 % [11].

8. CASE STUDIES8.1 STRUCTURAL INTEGRITY

ASSESSMENT OF RINGBEAM OF A NUCLEARREACTOR CONTAINMENTSTRUCTURE USING IMPACTECHO TECHNIQUE

Containment structures of some ofthe pressurised heavy water reactors(PHWRs) are made of prestressedconcrete with pretensioned cables.One of the important components ofthis containment structure is the ringbeam. The inner containment domeof one of the PHWRs gotdelaminated during construction.Due to sudden release of force inprestressing cables during collapseof the dome and during subsequentdetensioning/slackening of the

cables, delaminations in ring beamwere suspected. Impact echo testinghas been carried out for assessmentof the structural integrity of the ringbeam [11].

In order to develop the test procedurefor carrying out the impact echotesting, two mock up calibrationblocks were made. The block-I (Fig.12), with a size of 4000 x 4335 mmand representing a circumferentiallength equivalent to about 4 degreessector of the ring beam containedsimulated flaws, viz. voids of sizes50, 100 and 200 mm at a depth of500 mm, surface opening cracks of25, 50 and 75 mm depth, reinforcedbars of diameter 20, 32 and 45 mmat a depth of 50 mm etc. In order tostudy the response of thedelaminations of reinforced rods,various rods of diameter 20, 32 and

45 mm each at a depth of 50 mmwere kept and shaken before thesettling of the concrete to producedisbond at the steel-concreteinterface.

The impact echo testing of this blockindicated that the 100 mm and 200mm voids at 500 mm depth could bedetected, however, the 50 mm voidat 500 mm depth could not bedetected. Further, It was alsodemonstrated that if the location ofthe reinforcement rod is knownapriori, then it can be easily checkedby impact echo testing that whetherany delamination has occurred (Fig.9). In order to establish thesensitivity of the technique i.e. thedepth at which the 50 mm diametervoid could be detected, another testblock (block-II) was made. In thisblock, 50 mm diameter voids were

Basics

Page 17: Volume 10 Issue 2 September 2011

15

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

1600 years ago. Because of thehigher attenuation of ultrasonicwaves in the pillar, it was notpossible to inspect the pillar withconventional ultrasonic technique,even at lower wave frequencies.Hence, impact echo technique hasbeen applied on the iron pillar. Asthis is the first time to the best ofour knowledge, that impact echotechnique was being applied fortesting of metallic materials, thetesting was carried out on mock upcylindrical steel block of similardiameter as that of the Iron Pillar,before testing the iron pillar. Theproportionality constant for therelation between the first peakfrequency, the p-wave velocity andthe dimensions of the structures aredetermined for the steel structures.A sketch of the Delhi iron pillar isshown in Fig. 14. Impact echo testingwas carried out on the cylindricalportion of the Delhi Iron Pillar,

was reconstructed using the samering beam and the reactor is underoperation.8.2 IMPACT ECHO TESTING ONDELHI IRON PILLAR

As discussed earlier, impact echotechnique has been developed andsuccessfully utilized for testing ofconcrete structures. For the first time,the authors have used impact echotechnique for testing of a metallicstructure [12]. Impact echo techniquehas been used for the nondestructiveevaluation of the Delhi Iron Pillartowards the characterization of itsinternal structure and to understandthe methodology adapted for makingthe Iron Pillar. The Delhi Iron Pillarhas long evoked the admiration ofantiquarians and the curiosity ofmetallurgists, principally because ofits excellent state of corrosionresistance and the method offabrication of such a huge iron object

kept at various depths such as 100mm, 200 mm and 300 mm. The voidsat 100 mm and 200 mm depths couldbe detected. However, the 50 mmdiameter void at 300 mm depth couldnot be detected. Based on theseresults, the delectability of theImpact-echo system in terms of thedepth and the lateral dimension ofthe defect was also established. Thestudy revealed that a void can bedetected, if the ratio of the depth oflocation to the lateral dimension ofthe void is less than 5. Further, theimpact echo testing on reinforcedrods of various diameter at differentdepths indicated that a reinforced rodcan be detected if the ratio of thedepth of its location to the diameterof the reinforced rod is less than 3[11].

Based on the optimized testparameters identified with the helpof studies carried out on the mockup blocks, impact echo testing wassuccessfully carried out on the ringbeam of the reactor containmentstructure for assessing its structuralintegrity. To the best of ourknowledge, this was internationallythe first time that impact echotechnique has been employed forintegrity evaluation of a criticalcomponent of the containmentstructure of a nuclear power plant.

Figure 13a shows the response of areinforced rod in the ring beam. Theimpact echo response of the rod issimilar to that of an intact reinforcedrod (Fig. 9). Further, a duct sheathat a depth of about 400 mm couldalso be detected (Fig. 13b) and noother defect indication wereobserved in the containmentstructure. The results of velocitymeasurement and impact echo testingrevealed that the ring beam was freefrom any defects/ anomalies and thereinforced rods near the surface werealso found to be properly intact withthe concrete. Based on the study, thering beam was affirmed to be freefrom any damage. Hence, the dome

Fig. 14 : Sketch of the Delhi iron pillar showing various portions of the pillar. Theschematic of a defect is also shown for which impact echo response is shownin Fig. 14.

Basics

Page 18: Volume 10 Issue 2 September 2011

16

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

interval in time domain (equal toinverse of the digitization frequency)and Δt is the travel time of the wavebetween two transducers. ForCp= 4000m/s, δt= 2 μs (digitizationfrequency =500 kHz) and with a distancebetween two transducers as 300 mm;et = ±(2x10-6/75x10-6) = ± 0.0267, i.e± 2.67% or ± 107 m/s. Similarly, themaximum error involved in thethickness (/depth of a defect)measurement due to frequency

interval is given by ; where

Δf is the frequency interval (= ,

where n is the number of samplesrecorded) and Δf is the frequencyobserved corresponding to thethickness (/depth of a defect). Forδt= 2 μs, n=2048, CP=4000 m/s, andthickness of concrete = 300 mm;ef = (± 244.14/13333.33) = ± 0.0183i.e ± 1.83 % or ± 5.5 mm. The %error associated in the measurementof thickness due to the error involvedwith frequency measurement is thefunction of the frequency and in turnthe thickness also. The % errordecreases with decrease in thethickness.

The upper limit of concrete thicknessthat can be measured by the impact-

surface is the main concern, higherdigitizing frequency, shorter datalength (to avoid low frequencyresponse) and smaller diameter ballare the best suited. In routine testing,it is usually recommended to startwith a large impactor (10 mmdiameter or larger) and proceed touse a smaller impactor, if it isnecessary to amplify or “bring up”features that are associated withfrequencies of about 20 kHz andhigher.

10. ERRORS ASSOCIATEDWITH VARIOUSMEASUREMENTS

Impact echo technique is based onthe use of digital signal analysismethods. As a result, the timedomain waveforms and frequencyspectra are composed of discretepoints with fixed intervals thatdepend upon the data acquisitionparameters such as digitization rateand total length of the time domainwaveform. This results in systematicerrors in the velocity and thicknessmeasurements. The maximum errorsinvolved in the velocitymeasurements using the procedureas discussed in section 3.1 is given

as ; where ät is the sampling

except the bottom rough portion. Thetesting was also carried out at thedecorated portion on top of the pillar.The decorated portion exhibitedhigher velocity as compared to thatin the cylindrical portion. Impactecho signals acquired at a fewlocations consisted of extra peakscorresponding to different depths,which were attributed to the presenceof random voids. A typical impactecho signal indicating the presenceof a defect in the iron pillar is shownin Fig. 15. Further, the aspect ratioof the voids (larger dimensions inaxial and circumferential directionsas compared to the radial direction)indicated that the pillar would havebeen forged in the radial directionrather than in axial direction [12].

9. ADDITIONAL GUIDELINESFOR IMPACT ECHOTESTING

The proper selection of digitizingfrequency, data length and diameterof the steel balls for giving impact,are of significant importance. Forexample, when the main concern isto get the full thickness of thestructure, lower digitizing frequency,longer data length and largerdiameter ball are suitable. Whereas,when detection of flaws near the

Fig. 15 : Impact echo signal obtained at 70 mm below the top of the cylindrical portion of Delhi Iron Pillar showing a defect (shownschematically in Fig, 13) at about 60 mm depth (tested from the surface near the defect, where the waves get reflected fromboth the defect and the diametrically opposite surface).

Basics

Page 19: Volume 10 Issue 2 September 2011

17

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

6. M. Sansalone and N.J. Carino,“Detecting Delaminations inConcrete Slabs with and withoutOverlays using the Impact-EchoMethod,” Journal of theAmerican Concrete Institute,1989; 86(2): 175-184.

7. C. Colla and R. Lausch,“Influence of source frequencyon impact-echo data quality fortesting concrete structures”,NDT&E International, 2003: 36;203–213.

8. J.M. Lin and M. Sansalone, “Aprocedure for determining P-wave speed in concrete for usein impact-echo testing using aRayleigh speed wave speedmeasurement technique”,Innovations in NondetrctiveTesting, SP-168, AmericanConcrete Institute, 1997.

9. C. Cheng and M. Sansalone,“Effects on Impact-Echo SignalsCaused by Steel ReinforcingBars and Voids Around Bars,”ACI Materials Journal, 1993; 90(5): 421-434.

10. Y. Lin and W.C. Su, The Use ofStress Waves for Determining theDepth of Surface-OpeningCracks in Concrete Structures,Materials Journal of theAmerican Concrete Institute,1996; 93(5): 494-505.

11. Anish Kumar, Baldev Raj, P.Kalyanasundaram, T. Jayakumar,and M. Thavasimuthu NDT&EInternational, 35 (2002), 213-220.

12. Baldev Raj, P.Kalyanasundaram, T. Jayakumaret al., Current Science, 88 (12)(2005) 1948-1956.

ACKNOWLEDGEMENTSThe authors are thankful to Shri S.C.Chetal, Director, Indira GandhiCentre for Atomic Research(IGCAR), Kalpakkam and Dr.Baldev Raj, the previous director ofIGCAR, Kalpakkam for theirconstant encouragement and support.

REFERENCES1. Anish Kumar, T. Jayakumar, C.

V. Subramanian and M.Thavasimuthu, “Testing ofConcrete Structures forDetermination of Strength andDetection of Flaws using LowFrequency Ultrasonic andImpact-echo techniques”, J. Non-destructive Testing andEvaluation, 1999; 19(2): 43-46.

2. Mary Sansalone, “Impact-Echo:The Complete Story”, ACIStructural Journal, 1997; 94 (6):777-786.

3. M. Sansalone and W.B. Streett,Impact-Echo: NondestructiveTesting of Concrete andMasonry, 1997; Bullbrier Press,Jersey Shore, PA.

4. ASTM C 1383-04, “TestMethod for Measuring the P-Wave Speed and the Thicknessof Concrete Plates using theImpact-Echo Method,” 2010Annual Book of ASTM StandardsVol. 04.02, ASTM,WestConshohocken, PA.

5. C. Cheng and M. Sansalone,“The Impact-Echo Response ofConcrete Plates ContainingDelaminations: Numerical,Experimental, and FieldStudies,” Materials andStructures, 1993; 26(159): 274-285.

echo technique is usually limited bythe low frequency resonant peaks ofthe receiver transducer appearing at~1.1-1.4 kHz which correspond to~1.4 – 1.8 m thickness. Impact-echotechnique has been widely used fortesting of concrete structures upto~800 mm thickness and thickness ashigh as 1.2 m concrete structure isalso reported, when a steel ball of12.5 mm diameter was used as animpactor. For thick concretestructures (thickness = 1.2 m – 2 m),thickness peak obtained in an impactecho test may be indecisive due tothe possible presence of thetransducer resonant peak. However,it can still be tested for defectdetection by performing tests fromboth the sides of the structurecovering half of the thickness fromeach side.

11. CONCLUSIONSThe paper presents the basics of theimpact echo technique including itsphysical principle, details of thehardware and software, response ofvarious types of defects, guidelinesand associated errors in the impactecho testing. With the help of typicalexamples and schematics, it has beendemonstrated that impact echotesting can be used for themeasurement of ultrasonic velocityand thickness of concrete structureas well as for the detection of voids/delaminations/defects. The depth ofthe surface opening cracks could bemeasured with an accuracy of about10%. Further, if the location of areinforcement rod is known apriori,then it can be easily checked byimpact echo testing that whether anydelamination has occurred.

Basics

Page 20: Volume 10 Issue 2 September 2011

18

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Horizon

NDE by backscatter imaging

A visual image is formed by thereflection or more generally by thescattering of light from the viewedobject to the eye, i.e., the radiationsource (light) and the detector (eyes)are on the same side of the object.Given that x-rays or gamma rays justpass through any material without anydeflection, can imaging be possiblewith scattered radiation at such highenergy X-rays or gamma rays?

Compton Scattering – Basics

Compton scattering, so named afterthe American physicist, Arthur H.Compton, who was awarded the NobelPrize in 1927 for his interpretation ofthe X-ray scatter effect that bears hisname is significant as it demonstratesthat light cannot be explained purelyas a wave phenomenon. Compton’swork convinced the scientif iccommunity that light can behave as astream of particles (photons) whoseenergy is proportional to thefrequency. Compton scatteringrequires that light is viewed as aparticle and not just a wave, becauseit is the “collision” of the photon withthe electron, and the exchange ofenergy, which accounts for the shiftin energy.

Compton scattering is the interactionof a high energy photon, as from agamma ray or high energy X-ray, withan electron, and the resulting“scattered” photon which has areduced frequency, and thereforereduced energy and is thus termedincoherent scattering. The comptonscattering formula is

(1a)

where, Ei is the energy of the incidentphoton, Ef is the energy of the

scattered photon, and mec2 is the rest

mass energy of the electron (0.511MeV). The energy of the scatteredphoton can be calculated from thisequation using the angle of deflectionand the energy of the incident photon.Figure 1 depicts the process andindicates the angular distribution of

scattered energy.

Compton scattering can also beexpressed in terms of the incident andscattering waves as

Figure 1: (top) Schematic of the Compton scattering from a free electron (from http://missionscience.nasa.gov/ems/12_gammarays.html) (bottom) Polar diagram of the scattercross-section (from http://whs.wsd.wednet.edu/)

Dr. CV KrishnamurthyCentre for NDE and Department of Physics, IIT Madras

NDE by backscatterimaging

Page 21: Volume 10 Issue 2 September 2011

19

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

HORIZON

Figure 2 : (top) Compton scattering from an atom indicatingionization events (from http://whs.wsd.wednet.edu/).(bottom) The relative intensity of photoelectric effect,Compton scattering, and pair production.

illustrated in Figure 2. In this figure,we can find that under the conditionsin which we are interested (50KeV <E < 10MeV, Z<30), Comptonscattering is the dominant mode ofinteraction of photons.

The probability of the Comptonscattering per atom of the absorberdepends on the number of electronsavailable as scattering targets, andtherefore increases linearly with theatomic number, Z. The number ofphotons scattered is then a measureof the electron density of the materialwhich deflected it. This can in turn helpidentify the presence of an inclusionor a defect as highlighted in Table 1and Table 2.

Compton Scatter Tomography

Over the last two decades, Comptonscattering based studies have beencarried out for density measurementsand thickness measurements and on

ionizing the atom, but also kineticenergy of the now free electron.Therefore the kinetic energy will beless by about 10 eV, or about 1 part in1000 since the final kinetic energy ofthe electron is 11.27 keV if θ = 180°corresponding to backscatter. The shiftin wavelength is inversely proportionalto the mass of the particle it isscattering off. So, photons whichscatter off of the nucleus would appearunshifted because their shift would beabout 1000 times smaller than theshift due to an electron (the maximumshift when the photon is backscatteredby nuclear scattering is about 0.067keV, while for electron scattering it is11.27 keV).

In the energy range from several KeVsto several MeVs, the incident x-raybeam becomes attenuated in threeprinciple ways. They are photoelectriceffect, Compton scattering, and pairproduction. The relative coefficient ofthree attenuating processes is

(1b)

where λi is the initial wavelength of

the photon, λf is the final wavelengthafter scattering, θ is the scattering

angle, and λc, the comptonwavelength, is the wavelength of theresting electron which is 2.426 10-12m.The formula is derived by consideringthe interaction of a resting electronwith an incoming photon under theconstraints of conservation ofmomentum and energy.

The formula assumes that the photonis scattering off a free electron.However within the detector theelectrons may be bound valenceelectrons. So the energy received bythe electron will not only go into

Table 2: Comparison between Compton scatter andtransmission densitometric and imaging techniques

Table 1: Features of Compton Scattering

Most likely to occur

As x-ray energy increases

As atomic number ofabsorber increasesAs mass density of absorberincreases

a) with outer-shell electronsb) with loosely bound electronsa) Increased penetration through specimen

without interactionb) Increased Compton scattering relative to

photoelectric effectNo effect on Compton scattering

Proportional increase in Compton scattering

Scattering3-D informationImaging achieved directly by scanning

Densitometry achieved on anabsolutebasis

Fractional contrast of defect structureis high especially at high photonenergies or low target densities

Energy analysis is possible - canmonitor both the Compton and elasticscattering intensitiesVariable scattering geometryBackscattering geometry available -requires access from one side only

Aspect ratio unimportant

Scanning rates are generally slow forhigh Z materials unless multibeam/multidetector arrays can be used

Transmission2-D informationImaging achieved – Indirectlyreconstruction from projectionsDensitometry requires a ‘caliper”measurement to convertattenuationinto a density valueFractional contrast of defectstructure is low due to superpositioneffects

Energy analysis of limited value

No degrees of freedomRequires access from both sides

Aspect ratio important for adequatecontrast and avoidance of shapeartefacts in CT- imagesScanning times are fast

Page 22: Volume 10 Issue 2 September 2011

20

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Figure 3: (left) geometry exploiting the scattering angle for voxel-based inspection, (right)a wide-angle Compton-scatter inspection system (from Ref [5]).

HORIZON

associated NDE of voids,delaminations and material losses.Metals, weldments, concretestructures, and composites have beeninvestigated with increasing successdue to improvements in detectors andreconstruction algorithms.

Figure 3 shows the schematics of twoconfigurations that are routinelyemployed to carry out comptonscattering based NDE to extract theinformation about density andthickness variation.

The incoherent scattering leads to a“soft-collimation” process, by virtue ofthe fact that each detected energywindow corresponds to a particulardirection of scattering. Thus, many“points” along the incident beam canbe simultaneously inspected asindicated on the right in Fig. 3. In orderto avoid ambiguity when relating theenergy to angle of scattering using Eq.(1), the detector is confined to almosta point, using a wedge- shapeddetector collimator.

Figure 4 shows an example of theresults of experiments on a pipeillustrating the potential of Comptonscattering for NDE applications.

According to ASTM E1931 - 09Standard Guide for X-Ray ComptonScatter Tomography, it is best appliedto thinner sections of lower Zmaterials. Table No: 3 provides ageneral idea of the range ofapplicability when using a 160 keVconstant potential X-ray source:

Compton Back-Scatter Imaging

Compton imaging is a visualizationtechnique that uses the kinematics ofCompton scattering for thereconstruction of a gamma radiationsource image. Compton imagingsystems, also known as Comptoncameras, are used in thenuclear power industry for siteand environmental surveys, ingamma-ray astronomy and in severalprototypes of nuclear medical imagingsystems.

Basic design of Compton imagingsystems consists of two planes ofposition sensitive and energydispersive gamma-ray detectors asshown in Figure 5.

Compton imaging requires that agamma ray must interact withelectrons at least twice—once toinduce Compton scattering and onceto allow photoelectric absorption—although more than one scatter canoccur. When an x- or gamma-rayphoton is scattered or absorbed, high-energy electrons are ejected. Thesubsequent deposition of electronenergy produces a large number ofionized atoms. The ionization fromCompton scattering or photoelectricabsorption is then recorded by thedetectors. The gamma-rayinteractions must be separated inspace sufficiently so that they can beeasily distinguished from each otherand their positions can be accuratelymeasured to obtain high angularresolution.

Electronic collimation gives thepossibility to perform gamma-rayimaging without use of mechanicalcollimators. The potential advantagesof the Compton cameras overconventional imaging techniquesinclude a large field of view, increasedefficiency, good backgroundsuppression and a more compact andlightweight imaging system. Comptoncameras are effective over a largeenergy range (140 keV to 10 MeV) andcan be used in an energy selectivemode for the separate imaging of themixed gamma-ray sources. At closeranges (less than few meters) theygive possibility for three-dimensional(3D) imaging of objects from a fixedposition without detector motion.

One of the most promising applicationshas been that XBT is able to detectmetal-free landmines buried in avariety of soil conditions includingvarious types of vegetation asindicated by Figure 6 and Figure 7.

Strikingly similar to light reflection,backscatter signals are particularlystrong whenever the incident X-raysinteract with explosives, plastics, andother biological items, which typicallycontain low Z materials. An exampleof the photo-like images producedfrom the backscattered signals bycommercial systems is shown inFigure 8.

Deciphering information by visuallyobserving raw scatter radiographs, or

Figure 4: Scattered intensity originating from interactions of 662 keV photons with(a) an iron pipe of wall thickness 2.5 mm filled with different density liquids (petrol,diesel, multipurpose engine oil API CF, water and glycerine) and (b) an iron pipe (length140 mm and opening 22 mm) having a cut of 1 mm width up to the middle of pipe,covered with insulating material in the form of powder. Detection is by a NaI(Tl) detectorwith collimators of hole-size of 4 mm in radius for source and slit size of 4 mm in widthfor detector to define the volume of interest. Analysis was carried out using an inversematrix approach rather than the usual Monte Carlo method (from Ref [11]).

Page 23: Volume 10 Issue 2 September 2011

21

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

HORIZON

a set of measurements, is not directlypossible, due the convoluted natureof scatter indications. The problem iscaused by the non-localized diffusednature of scattering; a coded aperturetechnique (see Figure 9) to detectlocalized anomalies with backscatterimaging is often employed.

The advantage of merging visualimages with the images from aCompton imager can be striking as isshown in Figure 11. In oneexperiment, the team at Livermorecombined a camera and a Comptonimager to demonstrate that thesystem can visually identify a gamma-ray source containing an isotope ofsodium.

Compton imaging systems have anumber of advantages for nuclearwaste characterization, such asidentifying hot spots in mixed wastein order to reduce the volume of high-level waste requiring extensivetreatment or long-term storage,imaging large contaminated areas andobjects etc. Compton imaging also haspotential applications for monitoringof production, transport and storageof nuclear materials and components.Mixed radionuclide sources can besuccessfully separated by selectivelyimaging of gamma rays of interest.Other advantages of scatter imaging,compared to transmission imaging,are the flexibility in locating the sourceand the detector, which do not haveto be on opposite sides, the ability touse more than one detector for thesame radiation source beam, and the3-D nature of the scattering processwhich makes it amenable to multi-planar imaging.

Compton imaging is completelypassive and gives a 3D image of an

Figure 5: (a) The energies and positions of the first two interactions define a cone of incident angles. (b) The cones can be projectedonto a plane or sphere (one circle per gamma-ray event) to produce a 2D image of the source. The source image is reconstructed by thebackprojection and intersection of number of conic surfaces in the image space. (c) Image of point gamma-ray source reconstructed bybackprojection of cones. This type of information about the possible source distribution requires specialized reconstructions techniques(from Ref [8]).

Figure 6: XBT images at two depths on the left of an anti-personnel mine (typePPM-2) shown on the right (from Ref [6]).

Figure 7: XBT images at two depths on the left of an anti-tank mine (type TM-62)shown on the right (from Ref [6]).

Table 3: ASTM E1931 – 09 Guidelines(http://www.astm.org/Standards/E1931.htm)

Material Practical Thickness Range

Steel Up to about 3 mm (1/8 in.)

Aluminum Up to about 25 mm (1 in.)

Aerospace composites Up to about 50 mm (2 in.)

Polyurethane Foam Up to about 300 mm (12 in.)

Figure 8: Image of a cargo truck carrying drugs (from Ref [9]).

Page 24: Volume 10 Issue 2 September 2011

22

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Figure 11: A Compton imager combined with a camera produced this image, whichpinpoints the location of the radioactive source - an isotope of sodium (22Na) (from Ref [8]).

For further reading:

1. R.S. Holt and M.J. Cooper,Gamma-ray scattering NDE, NDTInternational 20 (1987) 161-165

2. D. Babot, G. Berodias and G. Peix,Detection and sizing by X-rayCompton scattering of near-surface cracks under welddeposited cladding, NDT& EInternational 24 (1991) 247-251

3. Stephen J. Norton, Comptonscattering tomography, J. Appl.Phys. 76 (1994) 2007-2015

4. W.Niemann and S.Zahorodny,Status and future aspects of X-raybackscatter imaging, Review ofprogress in QuantitativeNondestructive Evaluation 17A(1997) 379-385

5. A.C. Ho, E.M.A. Hussein,Quantification of gamma-rayCompton-scatter nondestructivetesting, Applied Radiation andIsotopes 53 (2000) 541-546

6. W. Niemann, S. Olesinski, T.Thiele, Detection of buriedlandmines with x-ray backscattertechnology, NDT.net – 7 (2002)No.10

7. Joseph Callerame, X-RayBackscatter Imaging:PhotographyThrough Barriers, JCPDS-International Centre for DiffractionData (2006) ISSN 1097-0002

8. Gabriele Rennie, Imagers toprovide eyes to see Gamma Rays,S & TR on Gamma-Ray Imaging-Lawrence Livermore NationalLaboratory (May 2006)

9. h t t p : / / w w w . a s - e . c o m /p r o d u c t s _ s o l u t i o n s /z_backscatter.asp

10. G. Harding, E.Harding, Comptonscatter imaging: A tool forhistorical exploration, AppliedRadiation and Isotopes 68 (2010)993–1005

11. A. Sharma, B.S.Sandhu, BhajanSingh, Incoherent scattering ofgamma photons for non-destructive tomographicinspection of pipeline, AppliedRadiation and Isotopes 68 (2010)2181–2188

12. h t t p : / / w w w . n d t - e d . o r g /EducationResources

13. http://en.wikipedia.org/wiki/Backscatter_X-ray

object from a single viewpoint withoutneeds for scanning or access to bothsides of the object. It covers a largeenergy range, has a wide field of viewand does not require mechanicalcollimation. It also provides goodbackground suppression through thekinematics of the scattering process

and through energy filtering. The X-ray backscatter technology has thepotential for low false alarm rates anda high detection probability. Lookingat the high resolution images, atrained operator is able to identify theburied object immediately.

Figure 9: Combining a mask with an antimask whose hole pattern is the inverse of themask’s pattern effectively removes background signals outside the surveyed area (fromRef [8]).

Figure 10: The Livermore-designed gamma-ray imager is portable and can be placedinside a small truck. Two coded-aperture masks (gray bars shown in the top photo) allowthe instrument to image both sides of a road. In the bottom photo, one can see the assemblyof the detectors (cylinders shown in the middle). Compton imagers for 3D information areshown in the disk configuration (right top) with orthogonal strips on each side connectedto a preamplifier and a digital data-acquisition system to determine the three-dimensional(3D) position for each gamma-ray interaction, and in the cylindrical configuration (rightbottom) with the outside contact divided into pixels that, when analyzed with a digitalsignal processor, provide the necessary 3D information (from Ref [8]).

HORIZON

Page 25: Volume 10 Issue 2 September 2011

23

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

BANGALOREExecutive Committee Meeting & theNew EC was elected for 2011-2012CHENNAIThe following courses were conducted*UT Level-II (ASNT) course from22.04.11to 30.04.11.*MT – Level - II(ASNT),forM.M.Forgings Ltd from 24th Apr, 08th,15th, 22nd May 2011*Surface NDT ( MT & PT) Level - II(ASNT) from 20.05.2011 to 26.05.2011.No. of participants for course 20. No. ofparticipants for examination 18.*UT L - II (ASNT) course from30.05.2011 to 05.06.2011. No. ofparticipants for course 14. No. ofparticipants for examination-18*RT L - II (ASNT) course from10.06.2011 to 16.06.2011. No. ofparticipants for course 17. No. ofparticipants for examination-18Other Activities: ISNT DAY celebratedon 21.04.2011. No of participants : 225inclusive of members and their family.Chief guest Mr RG Ganesan,TechnicalDirector*The Thambithrai Award for BestTechnical Paper presented to Dr. VaidehiGanesan Scientist, IGCAR.*Best member of ISNT-CHENNAICHAPTER and PARI award, Receivedby Mr RG Ganesan, Joint secretary –ISNT Chennai Chapter. EC Meetings:*01.05.2011*11.06.2011JAMSHEDPURAnnual General Body Meetingconducted on Date: 29.04.2011 and thenew executive committee was electedunder the chairmanship of Dr. N. Parida,Scientist from NML Awards : Dr.Amitava Mitra received the prestigiousMaterials Research Society of IndiaMedal (MRSI Medal)KALPAKKAMConducted a Technical Talk on X-rayendoscopic inspection of T/TS welds inheat exchangers by eminent Prof. DrUwe Ewert on Friday, 22nd July,2011KOLKATAThe following courses were conducredRadiography Testing (RT- II) Training& Certification Course from 23rd May to29th May, 2011.No. of participants for course- 16No. of participants for examination-12-1st Magnetic Particle Testing (MPT- II)Training & Certification Course from15.07.11 to 17.07.11. Four (4) candidatesparticipated & results are awaited.- 1st

Penetrant Testing (PT- II) Training &Certification Course from 22.07.11 to24.07.11. Three (3) candidates

CHAPTER NEWSparticipated & results are awaited.-8 t h

Radiography Testing (RT- II) Training& Certification Course from 01.08.11 to08.08.11. Six (6) candidates are likelyto participate.-5th Ultrasonic Testing(UT- II) Training & Certification Coursefrom 05.09.11 to 11.09.11. ECMeetings:1. April’112.June,’11.Interaction with ISNT HO-The Chapter had interacted with theISNT HO, NCB and other Chapters onvarious occasions In connection with itsactivities, National Seminars etc KolkataChapter paid back Rs. 1.0 Lakh to ISNT-HQ in June, 2011 towards adhocpayment against excess income overexpenditure during National seminarNDE-2010 held at Kolkata in December,2010. The A/cs. Statement is in advancestage of preparation & shall be sent toHQ by end August, 2011. The duesagainst loan taken from HQ for Purchaseof Office Premises in 2005 stand at Rs.1.86 Lakh which we hope to clear bySept end, 2011 after preparartion ofBalance Sheet and income/expenditurestatement of Chapter.KOTAOrganized: Leak Testing Level -II Courseduring June, 2011.MUMBAI-APCNDT 2013 committee Meeting washeld on 8th April 2011, 19th April 2011,25th April 2011.

-conducted Welding Inspectorexamination at ITT, on 1st May 2011,

-APCNDT 2013 committee Meeting washeld on 13th May 2011 and 7th June2011.

-EC Meeting was held on 22rd June 2011APCNDT 2013 committee Meeting washeld on 29th July 2011.

-Conducted LT Level II Examination forKota Chapter on 26- 06- 2011 at Kota..PUNETechnical Talk: TOFD Techniqueintroduction & Practical Demonstrationon Weld”on 14.05.2011 by Shri. AshokTrivediEC Meeting held on 29.04.2011TRICHYNewly added Members during this period

a. Associate member : 16b. Life Members : 01c. Members : 01d. Student Members: 82

EC meeting was conducted on june 2011Following Courses were conducted:a) Radiographers level I-(In association

with BARC-Mumbai) from11.04.2011 to 29.04.2011

b) MPT-Level–II- From 16.06.2011 to19.06.2011

c) LPT – Level – II – from 13.06.2011to 15.06.2011

d) Radiography Level – II from11.07.2011 to 21.07.2011

Invited Lecture : “Introduction toSafeRad Radiography System - anunique & innovative NDTtechnique” BY Mr. Malcolm WassU.K. on 11th July 2011.

TRIVANDRUM1. Annual General Body Meeting

2010-11: was held on 28th May2011.

2. Election of office bearers and ECmembers for the period 2011-2013was conducted

3. MR Kurup Memorial Lecture 2011:for the year 2010-11 was deliveredby Shri V Srinivasan, DeputyDirector, PRSO Entity, VSSC, atHotel Maurya Rajadhani,Trivandrum on 28th May 2011.

VADODARA1) Executive Committee Meeting held

on 25th March, 20112) Annual contract for website

designing and maintenancefor one year at the cost of Rs.5,000/- (Rupees Five Thousand only) wasfinalized.

23

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

MMMMMalaysian alaysian alaysian alaysian alaysian IIIIInternational nternational nternational nternational nternational NDTNDTNDTNDTNDTConferConferConferConferConferenceenceenceenceence and E and E and E and E and Exhibitionxhibitionxhibitionxhibitionxhibition

2011 (MINDT2011 (MINDT2011 (MINDT2011 (MINDT2011 (MINDTCE 11)CE 11)CE 11)CE 11)CE 11)November 21-22, 2011, Thistle Port

Dickson Resort

www.msnt.org.my

The Malaysian Society for NDT(MSNT) cordially invites you and yourstaffs to participate and present yourpaper in the 2011 MalaysianInternational NDT Conference andExhibition 2011 (MINDTCE 11). Weexpect there will be a lot ofparticipation from NDT professionalsand with your participation we lookforward to sharing knowledge andexperience in the field of NDT.MINDTCE 11 is jointly organizedwith the Malaysian Welding andJoining Society (MWJS) and stronglysupported by PETRONAS,International Committee for NDT(ICNDT), Malaysian Nuclera Agencyand SIRIM.

The organizers invite scientists,engineers, educators, researchers andmanagers to submit paper to thiswonderful conference.

Page 26: Volume 10 Issue 2 September 2011

24

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Revised Final date for full paper submission 5th September 2011Notification of acceptance : 15th July 2011Final date for full paper submission : 5th September 2011

National NDT Awards No. Award Name Sponsored by 1. ISNT - EEC M/s. Electronic & Engineering Co., Mumbai

National NDT Award (R&D)

2. ISNT - Modsonic M/s. Modsonic Instruments Mfg. Co. (P) Ltd.,National NDT Award (Industry) Ahmedabad

3. ISNT - Sievert M/s. Sievert India Pvt. Ltd., Navi MumbaiNational NDT Award (NDT Systems)

4. ISNT - IXAR M/s. Industrial X-Ray & Allied RadiographersBest Paper Award in JNDE (R & D) Mumbai

5. ISNT - Eastwest M/s. Eastwest Engineering & Electronics Co.,Best Paper Award in JNDE (Industry) Mumbai

6. ISNT - Pulsecho M/s. Pulsecho Systems (Bombay) Pvt. Ltd.Best Chapter Award for Mumbaithe Best Chapter of ISNT

7. ISNT - Ferroflux M/s. Ferroflux ProductsNational NDT Award (International recognition) Pune

8. ISNT - TECHNOFOUR M/s. TechnofourNational NDT Award for PuneYoung NDT Scientist / Engineer

9. ISNT - Lifetime Achievement Award

Note-1: The above National awards by ISNT are as a part of its efforts to recognise and motivate excellence in NDT professionalenterpreneurs. Nomination form for the above awards can be obtained from ISNT head office at Chennai, or from the chapters. Thefilled application are to be sent to Chairman, Awards Committee, Indian Society for Non-destructive Testing, Module No. 60 & 61,Readymade Garment Complex, SIDCO Ind. Estate, Guindy, Chennai-600 032. Telefax : 044-2250 0412 Email: [email protected]

National Certification Board - Indian Society for Non Destructive TestingAnnouncement

ASNT NDT LevASNT NDT LevASNT NDT LevASNT NDT LevASNT NDT Level III Eel III Eel III Eel III Eel III ExaminationxaminationxaminationxaminationxaminationMumbai 28, 29 & 30 November 2011

ASNT NDT Level III Examination will be conducted in the following methods:

1. Basic 2. Radiographic Testing 3. Magnetic Particle Testing 4. Ultrasonic Testing5. Liquid Penetrant Testing6. Eddy Current Testing 7. Neutron Radiographic Testing 8. Leak Testing 9. Visual Testing

10. Acoustic Emission Testing 11. Thermal / Infrared Testing

It may please be noted that the basic examination by itself is not considered as a method. Basic and methodexamination(s) must be taken to become eligible to receive a certificate for that method(s). The maximum number

of examinations that can be taken is six during the three days of the Examination.

DDDDDrrrrr. B. . B. . B. . B. . B. VVVVVenkatramanenkatramanenkatramanenkatramanenkatramanASNT Level III Examination Coordinator, Modules 60 & 61, Readymade Garment Complex,

SIDCO Industrial Estate, Guindy, Chennai 600 032, India

Ph: 91 44 22500412 & 91 44 42038175 / 91 44 27480500 Ext.22306E Mail: [email protected] Alternate E Mail: [email protected]

Page 27: Volume 10 Issue 2 September 2011

25

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

We hope that this new feature added to the journal since the last twoissues has been useful for the readers in planning their activities in termsof paper submissions, registering for seminars, etc. Please send yourfeedback, comments and suggestions on this section [email protected]

November 2011International Workshop on Smart Materials & Structuresand NDT in AerospaceNovember 2 to 4, 2011 ; Montreal, Quebec, Canadahttp://www.cansmart.com/MATEST 2011 International NDT ConferenceNovember 2 to 5, 2011; Split, Croatiahttp://www.hdkbr.hrSingapore International NDT Conference & Exhibition(SINCE 2011)November 3 to 4, 2011 ; Singaporehttp://www.ndtss.org.sg/41st International Conference and NDT Exhibition; NDEfor Safety 2011 / Defektoskopie 2011November 9 to 11, 2011, Ostrava, Czech Republichttp://cndt.cz/nde_for_safety2011/NDE-TokyoNovember 16 to 18, 2011; Tokyo, Japanhttp://www.jma.or.jp/next/en/nde/outline/index.htmlMalaysia International NDT Conference & Exhibition2011 (MINDTCE ‘11)November 21 to 22, 2011 ; Malaysiahttp://www.aindt.com.au/images/stories/page_images/conferences/international/mindtce_11_brochure_revision_1.pdf2011 Aircraft Structural Integrity program ConferenceNovember 29 to December 1, 2011 ; San Antonio,Texashttp://www.asipcon.com/5th IET international conference on railway conditionmonitoring and non-destructive testingNovember 29 to 30, 2011; Derby, UKhttp://www.theiet.org/events/2011/rcm.cfm

December 2011International Conference on NDE in the Steel and Alliedindustries (NDESAI2011)December 2 to 3, 2011 ; Jamshedpur, Indiahttp://www.ndesai2011.com/National Seminar on NDE (NDE-2011)December 6 to 10, 2011 ; Chennai, Indiahttp://www.nde2011.com/59th Defense Working Group on Nondestructive TestingDecember 6 to 8, 2011; Williamsburg, VA, USAhttp://www.dwgndt.org/

September 2011International Congress on Ultrasonics (ICU 2011)September 5 to 8, 2011 ; Gdansk, Polandhttp://icu2011.ug.edu.pl/ocs233-1/index.php/icu/icu20116th International Conference on Mechanical StressEvaluation by Neutrons and Synchrotron Radiation(MECA SENS VI)September 7 to 9, 2011 ; Hamburg, Germanyhttp://www.mecasens2011.de/8th International Workshop on Structural HealthMonitoring (IWSHM 2011)September 13 to 15, 2011; Stanford, CA, USAhttp://structure.stanford.edu/workshop/Materials Testing 2011September 13 to 15, 2011 ; Telford, UKhttp://www.bindt.org/Events/Exhibitions/MT_20115th Conference in Emerging Technologies in NDT(ETNDT)September 19 to 21, 2011; Ioannina, Greecehttp://www.etech-ndt5.uoi.gr/2011 ATA NDT ForumSeptember 26 to 29, 2011; Charlotte, NC, USAhttp://www.airlines.org/SafetyOps/EM/Pages/2011NDTForum.aspx

October 2011V Pan American Conference on NDTOctober 2 to 6, 2011 ; Cancun, Mexicohttp://www.copaend5.com/en/index.phpVIth International Workshop NDT in ProgressOctober 10 to 12, 2011 : Prague, Czech Republichttp://cndt.cz/ndt_in_progress2011/2011 IEEE International Ultrasonics Symposium (IUS)October 18 – October 21 ; Orlando, FL, USAhttp://ewh.ieee.org/conf/ius_2011/International Conference & Expo on NDT 2011 (ICENDT2011)October 18 to 21, 2011; Jakarta, Indonesiahttp://www.autri.org/2011 ASNT Fall Conference & Quality Testing SHowOctober 24 to 28, 2011 ; Palm Springs, CA, USAhttp://www.asnt.org/events/conferences/fc11/fc11.htmNational Conference of the Italian Society for Non-Destructive TestingOctober 26 to 28, 2011; Florence, Italyhttp://www.aipnd.it

NDE events

25

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Page 28: Volume 10 Issue 2 September 2011

26

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

AnswAnswAnswAnswAnswers for Pers for Pers for Pers for Pers for Prrrrrevious issue - NDT evious issue - NDT evious issue - NDT evious issue - NDT evious issue - NDT WWWWWororororord Sd Sd Sd Sd Searearearearearch 2ch 2ch 2ch 2ch 2

PenetrantDetectionCharacterizationDeveloperIndicationsFluorescentDyeLiquid

CapillaryVisibleUltravioletDippingSprayingBrushingSensitivityHydrophilic

LipophilicEmulsifiersContrastDwellToxicityContaminantsSolventBleedout

BrighterTemperatureDryingCarrierCleanerDegressingRelevantWashable

DispersionRinseSoakWettingBlottingNonaqueousInterpretationEvaluation

NDT WORDSEARCH - 2

Page 29: Volume 10 Issue 2 September 2011

27

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

Here are some interesting “Firsts”from the world of patents !

FIRST PATENTThe first U.S. patent was granted in1790 to Samuel Hopkins ofPhiladelphia for “making pot and pearlashes”-a cleaning formula used insoapmaking. This patent is referredto as Patent X1. Patent numbers werenot assigned to patents until 1836.

FIRST FEMALE PATENT HOLDERIn 1809, Mary Dixon Kies, a native ofKillingly, Conn., is reported to havereceived the first U.S. patent awardedto a woman for a process of weavingstraw with si lk or thread.Unfortunately, all records of thispatent were destroyed in the PatentOffice fire of 1836. On the morningof December 15, 1836, the PatentOffice, then located at the Blodgett’sHotel in Washington, D.C., wasconsumed by fire. Among the lostpatent-related materials were anestimated 7,000 models and 9,000drawings of pending and patentedinventions.

FIRST PATENT GRANTED AFTERNUMBERING STARTSPatent numbering started on July 13,1836. Patent No. 1 was issued toSenator John Ruggles of Thompson,Maine, for a locomotive steam enginefor rail and other roads.

“A new and useful improvement orimprovements on locomotive-enginesused on railroads and common roadsby which inclined planes and hills maybe ascended and heavy loads drawnup the same with more facility andeconomy than heretofore…”

FIRST DESIGN PATENTThe first design patent was grantedto George Bruce of New York City fora typeface.

FIRST PLANT PATENTPlant Patent #1 was issued in 1931to Henry Bosenberg of NewBrunswick, NJ for a climbing or trailingrose. Said Mr. Bosenberg of hisinvention, “My invention now givesthe true everblooming character toclimbing roses.”

YOUNGEST PATENT HOLDERThe youngest person to be granted apatent is a four-year-old girl fromHouston, Texas, for an aid forgrasping round knobs.

“This invention relates to a device forgrasping drawer or cabinet knobswhich is particularly useful tophysically impaired persons whowould otherwise, due to suchimpairment, have great difficulty inopening and closing drawers or

cabinets utilizing such knobs.”

(SourceCourtesy:http://www.lib.utexas.edu/engin/patentlite/firsts/ )Given below are some interestingstatistics of patents in India (Source: IP India Annual reports ;http://www.ipindia.nic.in/) :

Continuing our endeavor to provideyou updates on NDE and Inspectionrelate patents, listed below are a fewpatents from areas related toRadiographic inspection which wereissued by USPTO in the last few years.If any of the patents are of interestto you, a complete copy of the patentincluding claims and drawings may beaccessed athttp://ep.espacenet.com/

United States Patent 7,885,381

Method for inspecting pipes, andradiographic non-destructiveinspection apparatus

NDE patentsWe hope that the section on NDE Patents, which featured in the March 2011 issue of thisjournal, has continued to trigger your curiosity on this very important topic of Intellectualproperty. We continue this section with a few more facts on patents and a listing of a fewselected NDE patents. Please send your feedback, comments and suggestions on thissection to [email protected]

Compiled by Dr. M.T.Shyamsunder, GE Global Research, Bangalore, India

Page 30: Volume 10 Issue 2 September 2011

28

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Inventors: Nagumo; Yasushi,Nukaga; Jun, Kamimura; Hiroshi,Sadaoka ; Noriyuki, Takemori;Satoshi, Kodaira; Kojirou

Assignee: Hitachi-GE NuclearEnergy, Ltd. (Ibaraki, JP)

The pipe inspection method andapparatus can be used to implementrapid, tomographic inspection of apipe set up at a narrow location. Thepipe inspection method includes: afirst step for scanning the pipe bytranslating a radiation source andradiation detector arranged opposedlyto the pipe; a second step for theradiation detector to detect radiationthat the radiation source has emitted,at given scanning distance intervals;a third step for creating a transmissionimage of the pipe, based on aradiation dose that the radiationdetector has detected; and a fourthstep for constructing a tomogram orstereoscopic image of the pipe, basedon the transmission image. Thus, itis possible to provide the pipe

inspection method and apparatus thatcan be used to implement rapid,tomographic inspection of the pipe setup at a narrow location.United States Patent 7,319,738

Delivering X-ray systems to pipeinstallations

Inventors: Lasiuk; Brian W, Griffin;Weston B, Allison; Peter S

Assignee: General ElectricCompany (Schenectady, NY)

A mobile radiographic device for usein inspecting pipelines and the like,comprising an articulating aerial boomcoupled to a mobile carriage vehicle.A pivot mount is rotatably coupled tothe distal end of the aerial boom. Aplatform having a sliding rail isoperatively coupled to the pivotmount. A mounting fixture is rotatablymounted to a cradle, which in turn iscoupled to the sliding rail of theplatform. A radiation source and aradiation detector are mounted ondiametrically opposing sides of the

fixture in order to illuminate the outersurface of a pipeline or other objectwith radiation. A first positioningmeans is provided for coarselypositioning the scanning apparatusrelative to the pipeline. A secondpositioning means is provided forfinely positioning the scanningapparatus relative to the pipeline. Thesecond positioning means is operablefrom a remote location when theradiation source is illuminating thepipeline with radiation. The first andsecond positioning means provide aplurality of degrees of freedom forpositioning the scanning apparatus.

United States Patent 7,218,706

Energy discriminationradiography systems andmethods for inspecting industrialcomponents

Inventors: Hopkins; ForrestFrank, Dixon; Walter Vincent,Bueno; Clifford, Du; Yanfeng,Mohr; Gregory Alan, Fitzgerald;Paul Francis, Birdwell; ThomasWilliamAssignee: General ElectricCompany (Niskayuna, NY)

An energy discrimination radiographysystem includes at least one radiationsource configured to alternatelyirradiate a component with radiationcharacterized by at least two energyspectra, where the component has aradiation detector is configured toreceive radiation passing through thecomponent and a computer isoperationally coupled to the detector.The computer is configured to receivedata corresponding to each of theenergy spectra for a scan of thecomponent, process the data togenerate a multi-energy data set, anddecompose the multi-energy data setto generate material characterizationimages in substantially real time. Amethod for inspecting the componentincludes irradiating the component,receiving a data stream of energydiscriminated data, processing theenergy discriminated data, togenerate a multi-energy data set, anddecomposing the multi-energy dataset, to generate materialcharacterization images insubstantially real time.

Page 31: Volume 10 Issue 2 September 2011

29

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

United States Patent 7,236,564

Linear array detector system andinspection method

Inventors: Hopkins; ForrestFrank, Galish; Andrew Joseph,Ross; William RobertAssignee: General ElectricCompany (Niskayuna, NY)

A linear array detector (LAD) forscanning an object is provided. Thedetector includes a scintillator layerconfigured for generating a numberof optical signals representative of afraction of an incident X-ray beampassing through the object. The planeof the scintillator is parallel to the X-ray beam. The LAD further includes atwo dimensional array of photo-conversion elements configured toreceive several X-rays of the X-raybeams and configured to generatecorresponding electrical signals. Anarrangement of the photo-conversionelements is independent of the X-raypaths.

United States Patent 7,689,003

Combined 2D and 3Dnondestructive examination

Inventors: Shannon; Robert E,Hatcher; Clifford, Laloni; Claudio,Forster; Frank, Davis; Fredrick MAssignee: Siemens Energy, Inc.(Orlando, FL)

An inspection apparatus applying twodimensional nondestructiveexamination images onto a threedimensional solid model of acomponent to display a virtualcomponent that may be manipulatedto perform a nondestructiveinspection. The two dimensionalnondestructive examination imagesmay be acquired from a plurality ofviews of the component in order toprovide full coverage of the surfaceto be inspected, with appropriatestitching of images in regions ofoverlap between adjacent views. Thetwo dimensional images may be coloror black and white photographs orultraviolet or infrared images, forexample. Multiple types ofnondestructive examination images,results of inspection data evaluations,and design, operational and/or

maintenance information may bedisplayed separately or jointly on thethree dimensional solid model.Surface features of interest that aremapped as defined areas on the threedimensional solid model may bedisplayed simultaneously in differentviews on 2D and 3D images of thevirtual component.

United States Patent 7,099,432

X-ray inspection apparatus andX-ray inspection method

Inventors: Ichihara; Masaru,Yoshino; Shinji, Inoue; Hiroyuki,Kinoshita; Toshio, Ohuchi; Kazuo

Assignee: Matsushita ElectricIndustrial Co., Ltd. (Osaka, JP)

The X-ray inspection device and theX-ray inspection method according tothe present invention are configuredto hold an object to be inspectedirradiated with an X-ray from an X-ray irradiation device, uses a swingingdevice for performing swingingmotion of tilting the object to beinspected at an arbitrary angle andin an arbitrary direction, images theX-ray that passes through the objectto be inspected in an X-ray detectiondevice and extracts data of a desiredcross section from the X-ray image ofthe X-ray detection device in a controldevice.

United States Patent 6,873,680

Method and apparatus fordetecting defects using digitalradiography

Inventors: Jones; James Wayne

Assignee: Siemens WestinghousePower Corporation (Orlando, FL)

A digital radiography apparatus (10)and process for providing images ofan object, for example, an exhausttransition duct (12) comprising a corematerial and an overlying thermalbarrier layer, to detect surface andinterior defects within the duct (12).Incident energy is provided by anenergy source (30), transmittedthrough the object (12), and sensedby a sensor (32). An image of theobject (12) is formed by processing

the signal from the sensor (32) in asignal processor (34) and displayingthe image on a display (36) fordetermining defects in the object(12).

United States Patent 6,637,266

Non-destructive inspection,testing and evaluation systemsfor intact aircraft andcomponents and methodtherefore

Inventors: Froom; Douglas Allen

A non-destructive inspection, testingand evaluation system and processis provided for the review of aircraftcomponents. The system provides fora structure configured to contain aninspection and testing apparatus andthe aircraft components underinspection. The structure is lined withshielding to attenuate the emissionof radiation to the outside of thestructure and has corbels therein tosupport the components thatconstitute the inspection and testingapparatus. The inspection and testingapparatus is coupled to the structure,resulting in the formation of a gantryfor supporting a carriage and a mastis mounted on the carriage. Theinspection and testing equipment ismounted on the mast which forms,in part, at least one radiographicinspection robot capable of precisepositioning over large ranges ofmotion. The carriage is coupled to themast for supporting and allowingtranslation of the equipment mountedon the mast. The mast is configuredto provide yaw movement to theequipment.

United States Patent 6,466,643

High speed digital radiographicinspection of aircraft fuselages

Inventors: Bueno; Clifford, Herd;Kenneth Gordon, Mohr; GregoryAlan, Batzinger; Thomas James,Walsh; Dennis Michael

Assignee: General ElectricCompany (Schenectady, NY)

A system and method for radiographicinspection of aircraft fuselages

Page 32: Volume 10 Issue 2 September 2011

30

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

includes a radiation source preferablylocated inside of the fuselage and aradiation detector preferably locatedoutside of the fuselage. A sourcepositioning system is provided formoving the radiation sourcelongitudinally with respect to thefuselage, and a detector positioningsystem is provided for positioning theradiation detector in longitudinalalignment with the radiation source.The detector positioning system alsomoves the radiation detectorcircumferentially with respect to thefuselage. In operation, the radiationdetector is moved over the fuselagein a circumferential direction while theradiation source il luminates anadjacent region of the fuselage withradiation.

United States Patent 7,244,955

Computed radiography systemsand methods of use

Inventors: Bueno; Clifford, CorbyJr.; Nelson Raymond, Herd;Kenneth Gordon

Assignee: General ElectricCompany (Niskayuna, NY)

A computed radiography (CR) systemfor imaging an object is provided. Thesystem includes a radiation source, astorage phosphor screen, anil lumination source and a twodimensional imager. The radiationsource is configured to irradiate thestorage phosphor screen, and thestorage phosphor screen is configuredto store the radiation energy. Theillumination source is configured toilluminate at least a sub-area of thestorage phosphor screen to stimulateemission of photons from the storagephosphor screen. The twodimensional (2D) imager is configuredto capture a two dimensional imagefrom the storage phosphor screenusing the stimulated emissionphotons. A method of reading astorage phosphor screen is alsoprovided. The method includesilluminating at least a sub-area of thestorage phosphor screen using anillumination source to stimulate

emission of photons from the storagephosphor screen. The method furtherincludes capturing at least one 2Dimage using a 2D imager, from at leasta sub-array of the storage phosphorscreen using the stimulated photons.

United States Patent 7,239,435

High-speed, high resolution,wide-format Cartesian scanningsystems

Inventors: Shahar; Arie

A scanning system for writing,printing, direct imaging, plotting,computed radiography, and scanningincludes an optical system containsat lest one modulatable radiationsource for emitting radiation, amovable collimating lens, a reflector,and a focusing lens and a mechanicalsystem containing a first mechanicalcarrier spinning about a first axis. Asecond mechanical carrier spins abouta second axis. The second axis ismounted on the first mechanicalcarrier and is arranged to rotate aboutthe first axis. A third mechanicalcarrier spins about a third axis. Thethird axis is mounted on the secondmechanical carrier and is arranged torotate about the second axis. Thesystem also has a movable surface.The collimating lens is arranged toreceive the radiation from the oneradiation source and to convert it intoat least one collimated beam whichpropagates along an optical path fromthe collimating lens to the surface viathe reflector and the focusing lens toform at least one focused radiationspot on the surface. The mechanicalsystem is arranged to cause the thirdmechanical carrier of the mechanicalsystem to carry the reflector and thefocusing lens of the optical system tomove the one focused radiation spoton the surface along a straight line.

United States Patent 6,409,383

Automated and quantitativemethod for quality assurance ofdigital radiography imagingsystems

Inventors: Wang; Xiaohui,Vanmetter; Richard L, Foos;David L, Steklenski; David J

Assignee: Eastman KodakCompany (Rochester, NY)A phantom for use in measuringparameters in a digital radiographyimage system comprising asubstantially rectangular member ofan x-ray attenuating material; a firstrectangular array of landmarksassociated with the member for usein geometry measurements; a set ofregions associated with the centralposition of the member for exposurelinearity and accuracy measurementand a set of sharp angular edges partof one of the regions for modulationtransfer function measurements.

United States Patent 6,466,689

Method and system for digitalradiography

Inventors: MacMahon; Heber

Assignee: Arch Development Corp.(Chicago, IL)

A system and method for digitalimaging. A digital radiological imageof a subject is obtained having at leastone low density region. The image isprocessed using first weighting factorsin the at least one low density regionof the digital image and secondweighting factors smaller than thefirst weighting factors in regions ofthe digital image other than the atleast one low density region. Aprocessed digital image is obtainedand a representation of the processeddigital image is produced. In theprocessing of the image, unsharpmask filtering is employed using aprocessing curve having maximumunsharp mask filtering in the at leastone low density region of the digitalimage and a constant amount ofunsharp mask filtering less than themaximum unsharp mask filtering inthe regions of the image other thanthe low density regions.

Page 33: Volume 10 Issue 2 September 2011

31

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

Page 34: Volume 10 Issue 2 September 2011

32

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Page 35: Volume 10 Issue 2 September 2011

33

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

We hope you enjoyed solving the “NDTWord Search Puzzle” which waspublished in the last issue. We receivedmany entries from the readers andbased on the maximum number ofcorrect words identified, the followingare the WINNERS

- P Sreedhar, CNDE, IIT, Madras- P Selvaraj, ISRO-SHAR, Sriharikota- Prasanna Kumar NK, ISRO-SHAR, Sriharikota- TK Manoj Kumar, Brahmos Aerospace ltd, Thiruvananthapuram

Congratulations to all the Winners.They will receive their prizes from theChief Editor of the journal shortly. Thecorrect answers to the Puzzle arepublished below.

In this issue, we have another puzzle tocontinue stimulating your brain cells! Wehope you will find this section interesting,educative and fun filled. In the nextissue, we will be featuring a very largetheme crossword puzzle. Pleasesend your feedback, comments andsuggestions on this section [email protected]

Introduction & InstructionsThe “NDT Crossword”, contains clues tothe words related to NDT. The clues aregiven as “Across and Down”. All you haveto do is identify the right words from thegiven clues and fill them appropriately onthe puzzle with a black pen. Preferablyyou may take a photocopy of the Puzzlesheet and mark your answers on that.Once completed please scan youranswered puzzle sheet as a PDF file andemail the scanned sheet [email protected] along with yourname, organization, contact number andemail address

Rules & Regulations

- Only one submission per person isallowed

- The filled in answers should be legibleand clear without any scratching oroverwriting

- The decision of the Editor-in-Chief,Journal of NDT &E is final and bindingin all matters

- The correct answers and the namesof the prize winners will be publishedin the next issue of the journal.

NAME : ___________________________

ORGANIZATION : ___________________________

PHONE : ___________________________

EMAIL ID : ___________________________

ndt puzzlendt puzzleConceptualized & Created by

Dr. M.T. Shyamsunder,GE Global Research, Bangalore

CROSS WORD PUZZLE

Page 36: Volume 10 Issue 2 September 2011

34

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Kalva Engineers Launches

DIO 1000 Precision Flaw DetectorLaunchIn India @ unbelievable price

Are you facing inspec�on challenges like

Launch Offer

Are you facing inspec�on challenges like

Iden�fy finest crack in the thickest material

Precision measurement of 0.01 mm.

A-scan together with B-scan for verifica�ong

Looking for IMPORTED Ultrasonic Flaw Detector with Latest features like

Flaw detec�on

Corrosion MappingCorrosion Mapping

DGS, AWS, DAC, TCG

Flash memory,

B Scan,

EMAT

Starmans DIO1000 from Czech Republic

Limited stocks

Hurry up !!!!! as offer valid �ll stocks lasts.

C ec epub c

For more info and for a demo please call/ email us Now ……..

“KALVA HOUSE”, Plot no 17,Road No. 8B, Mamatha Nagar, Nagole, Hyderabad – 500068, India.Ph No: 040-65216871. Tele Fax No: 040 24222476,

Official Email: [email protected] Visit us at www.kalva.net

Page 37: Volume 10 Issue 2 September 2011

35

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

A Solution for

• Complete Automation of Inspection Hi h S d Hi h R li bilit d t ti

ONLINE UT SYSTEMS FROM CZECH REPUBLIC

• High Speed, High Reliability detection • A “UNIQUE” UT / IR combination to ensure 100% surface defect

detections from 0.3 mm to 3.5 mm together with Internal defect detection.

Applications:-Ultrasonic Inspection of Rods, Tubes and Billets

Ultrasonic C scan inspection of Rods, Tubes, Billets, composite products

Infra Red Inspection of Rods Tubes and BilletsInfra Red Inspection of Rods, Tubes and BilletsUltrasonic Inspection of Weld Seams in Spirally Welded Pipes

We Sell

One stop shop for “Global” NDT needs

EquipmentsWe Sell

We Train

We Serve

NDTEquipments

Courses

Inspec�on needs

Our ClienteleBHEL, ISRO, NTPC, RIL, L&T, TPL, IGCAR, DRDO, SCR, NFC, Midhani, VSSC, KRL andlist goes on..

“KALVA HOUSE”, Plot no 17,Road No. 8B, Mamatha Nagar, Nagole, Hyderabad – 500068, India.Ph No: 040-65216871. Tele Fax No: 040 24222476,

Official Email: [email protected] Visit us at www.kalva.net

Serving NDT for 15Serving NDT for 15+ years... years...

Page 38: Volume 10 Issue 2 September 2011

36

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Page 39: Volume 10 Issue 2 September 2011

37

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

Page 40: Volume 10 Issue 2 September 2011

38

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

WWWWWant to Aant to Aant to Aant to Aant to Advt in our Jdvt in our Jdvt in our Jdvt in our Jdvt in our Journal ?ournal ?ournal ?ournal ?ournal ?

WWWWWant to rant to rant to rant to rant to reach oeach oeach oeach oeach ovvvvver 6000er 6000er 6000er 6000er 6000members acrmembers acrmembers acrmembers acrmembers across Ioss Ioss Ioss Ioss India?ndia?ndia?ndia?ndia?

SSSSSubscribe to JNDTEubscribe to JNDTEubscribe to JNDTEubscribe to JNDTEubscribe to JNDTE

AAAAAdvdvdvdvdvererererertisement tarifftisement tarifftisement tarifftisement tarifftisement tariff

Classified advt (Business card) : Rs. 600/-

BackCover (Multi Colour) : Rs. 22,000/-

Inside Cover - Front (Multi colour) : Rs. 11,000/-

Inside Cover - Back (Multi Colour) : Rs. 11,000/-

Inner Pages Full (Multi Colour) : Rs. 8,000/- US $ 300

Inner Pages Full (Black & White) : Rs. 6,000/- US $ 200

Inner Pages Half (Black & White) : Rs. 3,000/- US $ 125

Panel Display : Rs. 1,200/- US $ 50

Journal subscription(India & SAARC countries)

includes ISNT membership & subscription toJournal of Non Destructive Testing & Evaluation

Journal Subscription forR & D establishments and educational institutions :

Rs. 700/- per annum and US $ 60

Payments should be made byDemand Draft, drawn favoring

“JNDE – ISNT”, payable at Chennai.

Mr. V. PariManaging Editor

Journal of Non Destructive Testing & EvaluationModules 60 & 61, Readymade Garment Complex,SIDCO Industrial Estate, Guindy, Chennai 600032

Phone: (044) 2250 0412Email: [email protected] ;

[email protected]

Local Conference SecretariatThe Conference Company

South Africa

Tel: 27 31 303 9852Fax: 27 31 303 9529

Nina Freysen-Pretorius - [email protected] Hancke-Haysom - [email protected]

www.wcndt2012.org.za

Page 41: Volume 10 Issue 2 September 2011

39

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

Page 42: Volume 10 Issue 2 September 2011

40

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Page 43: Volume 10 Issue 2 September 2011

41

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

Page 44: Volume 10 Issue 2 September 2011

42

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Page 45: Volume 10 Issue 2 September 2011

43

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

Page 46: Volume 10 Issue 2 September 2011

44

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

This second generation OmniScan flaw detector increases testing efficiencies, ensuring

superior manual and advanced AUT application performance with faster setups, test cycles,

and reporting, in addition to universal compatibility with all phased array modules.

The OmniScan MX2 offers a high acquisition rate and new powerful software features—in a

portable, modular instrument—to efficiently perform manual and automated inspections.

THE STANDARD IN PHASED ARRAY, REDEFINED

OMNISCAN MX2Be truly in touch with phased array• Bright, large size screen• Fast, intuitive touch screen interface• Full screen mode• Advanced Weld Overlay • High-capacity data storage• Fast transfer• Designed for IP66

Exclusive Distributor in India: Blue Star Limited Tel: +91 444-244-4000 • [email protected] Visit www.olympus-ims.com/OmniScan-MX2 • [email protected]

Page 47: Volume 10 Issue 2 September 2011

45

vol 10 issue 2 September 2011Journal of Non Destructive Testing & Evaluation

Induction Heat Treatment Equipment on Sales / Rental Basis

� Thermocouple Wire (K Type) � Nichrome Heating Wire � Nickel 212 Wire � Ceramic Beads for Heating Pads

Induction Heating Systems for Pre -Heating & Stress Relieving (PWHT)

We also provide Consumables for Heat Treatment

We also provide:

Phone: +91 22 4014 2318 +91 99209 56756 [Rohit] +91 97693 20252 [Jemmy] E-mail: sales@cerberus -india.com Website: www.cerberus -india.com

Gala No. 1, Jalaram Ind Estate Ketkipada Road, Dahisar (E), Mumbai - 400068, India

CERBERUS

Pen Style Hardness Tester

Ultrasonic Thickness Gauge

Coating Thickness Gauge

Portable Hardness Tester

Page 48: Volume 10 Issue 2 September 2011

46

vol 10 issue 2 September 2011 Journal of Non Destructive Testing & Evaluation

Page 49: Volume 10 Issue 2 September 2011

47Technical Paper

Vol. 10, Issue 2 September 2011 Journal of Non destructive Testing & Evaluation

Non destructive detection of debonding in adhesivelybonded metal/ceramic composite plates

Sony Punnose, Amretendu Mukhopadhyay, B. Nagaraja Kowmudi,P. Rama Subba Reddy, V. Madhu and Vikas Kumar

Defence Metallurgical Research Laboratory, Kanchanbagh, Hyderabad – 500 058Email: [email protected], [email protected]

ABSTRACTStudies have been carried out for detecting the debonding in ceramic-metal composite laminates of Zirconia Toughened Aluminabonded to Titanium alloy (ZTA/Ti) plate. Composite panels of ZTA/Ti with simulated debonds have been tested to assess thefeasibility of thermography as a non destructive testing method for detection of debonds. The method has been studied vis-à-vis ultrasonic technique. Pulse phase and lockin thermography techniques have been used for detection of debonds. Thethermogram clearly reveals the debonding as well as the non-uniformity in bonding. Ultrasonic attenuation drop has been measuredto detect the debonding in the specimens. The study shows that both thermography and ultrasonic techniques can be adoptedfor detecting the debonding in the composite plates.

Key Words: Ultrasonic, lock-in thermography, pulse phase thermography

1. INTRODUCTION

Adhesively bonded structures and adhesive joiningtechnology are increasingly being used as alternatives totraditional methods of fastening materials. This has led toan increasing demand on nondestructively evaluating andcharacterizing this material for quality control. The mostcommon kind of damage in the adhesively bonded structureincludes localized lack of or excess of resin and debonding,inhomogenities due to the presence of spurious materialsand porosity. The quality (strength and durability) of abond depends on the interaction of the adhesive with theadherent (surface to bond). There are several factors, e.g.ambient temperature and humidity, pressure applied, thatcan affect the quality of bonding. Owing to improperfabrication parameters, different types of defects are likelyto occur, such as lack of adhesive (bubbles, air layers,foreign materials), cohesion defects (breaking within theadhesive) and bonding defects (breaking at the surface-bond interface). Various Non Destructive Examination(NDE) methods including ultrasound, radiography, acousticemission, infrared thermography (IRT) have been usedfor the detection and characterization of defects incomposites. Ultrasonic method, being one of the oldestmethods, is widely used in the industry for defectcharacterization. Among other techniques, infraredthermography is a fast growing technique for the inspectionof composite sandwich structures. As the defects in bondedstructures act as barrier to thermal diffusion, these defectscan in-principle be detected by IRT technique. Thermalimaging techniques are being widely used for detectionand evaluation of defects and delaminations in compositematerials [1-8]. Lock-in thermography is proved to be aneffective technique for quantifying the depth of the defectsin steel plates as well as detecting subsurface defects incomposite plates [9, 10]. In addition to IRT, ultrasonic

measurements are being used for the assessment ofbonding defects in different composite materials [11-17].Both these techniques have advantages as well as somelimitations in terms of its applicability. Thermography hasadvantages like its non contact nature, fast inspection rateand less health hazards [18]. On the other hand, there arefew disadvantages like difficulty to deposit uniformly alarge amount of energy in short period of time over alarge surface, effect of thermal losses, emissivity problemsand capability to detect only subsurface defects, resultingin a measurable change of thermal properties. Ultrasonictechniques are better suited for quantification of defectsthough they are slower than to thermal imaging techniquesand are contact type in nature. In this paper, we explorethe possibility of using infrared thermography (lock-inthermography and pulse-phase thermography) and studiedthese technique vis-à-vis ultrasonic techniques to detectthe debonding in ZTA/Ti bonded composites. Compositepanels of ZTA/Ti with simulated debond have been testedto assess the feasibility of thermography technique vis-à-vis ultrasonic technique as a NDE method to detect debond.

2. EXPERIMENTAL DETAILS

Two specimens of simulated debonds have been tested tostudy the feasibility of ultrasonic vis-à-vis thermographyas a NDE method for characterization of debond. Thedetails of the specimens are given in Table 1 and thephotographic images are shown in Figure 1. Thermalimaging of the two specimens was carried out using amedium wave infrared camera (M/s. Cedip, France) withInSb focal plane array detector. The camera has spectralsensitivity in the range of 3-5 μm with a spatial resolutionof 5.4 μm. The temperature sensitivity of the camera is20 mK at 25oC. Image acquisition was done in thefrequency range of 50 - 380 Hz at a full frame view of

Page 50: Volume 10 Issue 2 September 2011

48 Technical Paper

Journal of Non destructive Testing & Evaluation Vol. 10, Issue 2 September 2011

320 x 256 pixels. A fully automated software system wasemployed for capturing and processing of images. Bothlock-in as well as the pulsed phase techniques was usedand the schematic of the experimental setup are shown inFigure 2 (a) & (b) respectively.

Table 1: Details of specimens

Nomenclature Type of materials

Specimen – 1 50 mm x 50 mm x 5 mm ceramic tilebonded to metal plate of 84 mm x 84 mmx 5 mm

Specimen – 2 30 mm x 30 mm x 5 mm ceramic tile witha centre hole of 7 mm dia bonded to metalplate of 70 mm x 95 mm x 5 mm

The experiment was performed at room temperature inwhich the detector and heating sources are kept on theopposite sides. The images were acquired from the metalside. Both the specimens were heated using two Halogenlamps of 1 kW power. The lamps were kept at a distanceof ~ 0.4 m and the IR camera was kept at a distance of~ 0.4 m from the specimens. The two heating sources

were kept at an angle of 75o in order to ensure uniformheating of the specimens. The distances among thespecimen, IR camera and heating sources were keptconstant throughout the experiment. In pulsed phasetechnique (PPT) specimens were heated with a squarepulse of ~ 2 sec duration and images were captured afterthe heat pulse. Images thus captured were processed toobtain phase images within a range of chosen frequencydepending on the type and depth of debond. In lockintechnique, the procedure consisted of acquiring phaseimage at lock-in frequency while the specimen surfacewas thermally stimulated with a sinusoidal lockin heatpulse. Specimens were heated with sinusoidal modulationfor 4-5 cycles and the resulting oscillating temperaturefield in the stationary regime (that is after the transientregime) was remotely recorded. A Hameg functiongenerator was used to generate the thermal modulation.The frequency of modulation coupled with the lockin optionwas chosen from the analysis of phase images from pulsedphase experiments.

Ultrasonic measurements have been performed using contacttransducer of 5 MHz frequency in pulse echo mode, at again of 50 dB. Water was used as a couplant between thetransducer and the specimen. During the measurements,a constant pressure was maintained between the transducerand the specimen. These measurements of back wall echoamplitudes were made in the far field region. Multiplerectified back wall echo pattern has been studied bothfrom the debond and good regions in the specimens. Theseecho patterns were compared with the reference echotaken only from metal, i.e. metal air interface.

3. RESULTS

3.1 THERMOGRAPHY RESULTS

The temperature-time evolution pattern for specimen -1 atfour chosen spot on the surface has been shown inFigures 3(a) and 3(b). Though the temperature evolutionpattern shows some indication but it could not conclusively

Fig. 1 : Photographic image of Specimens

Fig. 2 : Experimental set up

(a) Lockin thermography (b) Pulsed Phase thermography are missing.

Page 51: Volume 10 Issue 2 September 2011

49Technical Paper

Vol. 10, Issue 2 September 2011 Journal of Non destructive Testing & Evaluation

reveal the presence of debond. The problem of non-uniform heat deposition masks the temperature evolutionpattern. To overcome this problem the acquired imageswere processed in the frequency range of 0.01 Hz to 0.1Hz. The frequency-phase images were analyzed fordetection of debond/hole in the specimens. Figure 4ashows pulse phase image at 0.03 Hz for specimen-1 whichclearly reveals debond and Figure 4b shows pulse phaseimage at 0.055 Hz for specimen-2 which shows thepresence of hole in the ZTA plate seen from the metalside. It can be seen from the image that the hole is seenat the centre of the plate but the size of the hole is not tothe exact scale.

For better detectability, lockin thermography has beenperformed at the frequency of pulse phase for thespecimens. But the frequency of lockin process does notmatch exactly with that of the frequency of pulse phase.Lockin images are obtained at slightly different frequencies.Figure 5a shows lockin image at 0.385 Hz forspecimen-1. The image not only reveals the debond regionbut also reveals the quality/uniformity of bonding in otherregions. The four corners are bonded more as comparedto the sides. Figure 5b shows lockin image at 0.037 Hzfrequency for specimen-2. All these images show animprovement in terms of detectability over the pulsed phaseimages.

3.2 Ultrasonic Results

Specimens were also tested using ultrasonic technique.Figures 6(a-c) and 7(a-c) shows the multiple echo pattern(Amplitude Vs time) taken from the metal surface fromthe good region and debond region for specimen-1 andspecimen-2 respectively. The scatter in the attenuationmeasurements is ±10%.

4. DISCUSSION

4.1 Thermography

Theory: When a uniform heat is deposited periodicallywith a modulation frequency of ‘ω’ on a semi infiniteplanar surface, notwithstanding the three dimensional heatflow, temperature evolution on the surface as function ofdepth (z) and time (t) for a fixed location (x, y) can beexpressed as [19]:

(1)

where is the thermal diffusion length; k is thermal

conductivity; ρ is density; C is specific heat.

From the above equation, it can be seen that the temperaturefor a particular point on the surface (z = 0) of the specimen

Fig. 3 : (a) Thermogram showing four spots (specimen-1) (b) Time-temperature graph

Fig. 4 : (a) Pulse Phase image at f = 0.03 Hz (specimen-1) (b) Pulse Phase image at f = 0.055 Hz (specimen-2)

Page 52: Volume 10 Issue 2 September 2011

50 Technical Paper

Journal of Non destructive Testing & Evaluation Vol. 10, Issue 2 September 2011

changes periodically in time and corresponding phase

of that depends on the thermal properties

of the material along which heat diffuses beneath the point.Variation in the thermal properties, along the path withrespect to the path beneath the surrounding points, canlead to a differential phase contrast on the surface of thespecimen. This is due to interference in time of thedeposited heat with the diffusing thermal front.

Now in lockin thermography the exact time dependencebetween the output signal and the reference input signal(i.e. the oscillating - also called modulated heating) ismonitored. The resulting oscillating temperature field(following the oscillating thermal stimulation) in thestationary regime (that is after the transient regime) isremotely recorded through its thermal infrared emission.For a particular (lock in) frequency, phase image is mapof f as a function of (x, y). Images are obtained instationary mode and total number of images is a function

of total periods of heating and frame rate. Images capturedduring a complete period are sampled, for the lock infrequency, using Fourier analysis and a single phase-frequency image corresponds to the summation over thetotal number of periods.

Pulsed Phase Thermography (PPT) is a processingtechnique which combines advantages of both PT(operating in the transient regime) and LT (operating inthe stationary regime). In PPT deployment [20, 21], thespecimen is pulse-heated as in PT and the mix offrequencies of the thermal waves launched into thespecimen is unscrambled by performing the Fouriertransform (FT) of the temperature decay on a pixel bypixel basis thus enabling computation of phase images asin LT. In the analysis, for each pixel (i, j), the temporaldecay f (x) is extracted from the image sequence (wherex is the index in the image sequence). Next, from f (x),the discrete Fourier transform F (u) is computed (u beingthe frequency variable). Finally, from the real R (u) andimaginary I (u) components of F (u), the phase is computed

Fig. 5 : (a) Lockin image at f = 0.385 Hz (specimen-1) (b) Lockin image at f = 0.037 Hz (specimen-2)

Fig. 7 : Signal amplitude for specimen-2

Fig. 6 : Signal amplitude for specimen-1

Page 53: Volume 10 Issue 2 September 2011

51Technical Paper

Vol. 10, Issue 2 September 2011 Journal of Non destructive Testing & Evaluation

[22]. In PPT as in LT, it is possible to explore the variousfrequencies u. However, differences exist since analysis inPPT is performed in the transient mode while in LT, thesignal is recorded in the stationary mode. This means, forinstance, the quality of images will be better in LT due tothe summation process involved in the computations. Onthe other hand, a single pulsed experiment is needed inPPT while LT sometimes requires more. The FT is usedto extract various frequencies and is expressed as [23]

(2)

where j is an imaginary number, n designates the frequencyincrement, N is the total number of thermograms and Reand Im are respectively the real and the imaginary partsof the transform. Amplitude A and phase φ data are availableas follow:

(3)

It can be seen that though the presence of debond andhole have been detected for the specimens using both thepulsed phase and lockin techniques, the lockin imagesobtained are at slightly different frequencies. In the caseof specimen-1, i.e. specimen with debond at the centre,pulsed phase and lockin images are almost at samefrequency. In specimen-2, there is a finite frequencydifference between pulsed phase and lockin images. Pulsedphase images are at higher frequencies. From therelationship of thermal diffusion length, as discussed above,it can be seen that μ is inversely proportional to ω. Thissuggests that with decrease in the modulation frequency,greater depth can be probed inside a material. But, this hasa limit in terms of detectable contrast (depends on thetemperature sensitivity of the camera) arising out of thetemperature difference between the neighboring points(depends on the spatial resolution of the camera) on thesurface of the specimen. In the present study, phasecontrast arising out of the difference in the thermalproperties of the glue and air (at a depth of 5 mm) isgetting reduced due to the high rate of three dimensionalheat diffusion in metal. In case of pulsed phase, thisbecomes an impediment towards the detection of defecteither too small. In the case of smaller defects, temperaturesensitivity becomes a matter of concern. To overcomethese problems and as first step towards the detection ofdebond, specimens were heated for higher time (~ 2 sec)and images were analyzed at higher frequencies. Higherfrequency image indicates the presence of debond at lowerdepth. In this case, it is seen as an impression of debond,as higher heating time gives rise to partition of more energytowards lower frequency components. These lowerfrequency components interfere incoherently with theoriginal thermal wave front from the defect and getsmanifested in the form of noises in the pulsed phase imageswherein the images do not reveal the proper size of thedebond. Nevertheless, pulsed phase analysis has given a

firsthand indication of the presence of debond in thespecimens with a rough estimate of the modulationfrequencies for further analysis with lockin signal.

In lockin process, the problem of decreasing phase contrastdue to faster heat transfer in metals is somewhat nullifiedand defects at greater depth can be detected. Periodicheating with lock in modulation frequency, is chosenaccording to the thermal perturbation caused by thepresence of defects, leading to the physical interference intime. Interference in time of thermal front along with theoverall summation principle discussed above leads to abetter phase contrast in case of lockin images for thespecimens.

One of the aspects of the phase images over amplitudeimages is that phase images are less susceptible to theinfrared surface features such as emissivity variation. Thisleads to better depth probing as compared to theconventional pulsed thermography with IR camera ofsimilar characteristics.

Using the phase information obtained from

the thermograms and the thermal diffusion length

calculated from the modulationfrequency, depth of the defects can be roughly estimated.But lockin and pulsed phase techniques still have somelimitations in terms of exact defect quantification, as inthis process the time information is lost. Further,experiments along with image processing with wavelettransform for quantification of debond is currently beingstudied.

4.2 Ultrasonics

From the echo pattern it can be seen that exponentialdecay from the good bonding region is markedly differentfrom the debond region. By comparing these echo patternswith the echo pattern from the metal reference, it can beinferred that echo from the debond region resembles theecho pattern of the metal reference. This is due to the factthat in case of debond; the back wall echo originates fromthe metal air interface instead of metal glue interface.Change in the nature of interface causes change in thereflection coefficient at the interface. As the ultrasonicparameters were same for all the measurements, drop inthe amplitude of the back wall echoes from the metal glueinterface was more owing to the low reflection coefficientat the interface. This method shows the promise ofdetecting debond very effectively. However, C-scan imagingcan be used for automatic and faster inspection of largeplates and hence the study can be extended further fordetection/sizing of debond in case of large components.

5. CONCLUSIONS

It could be concluded from the above experiments thatboth the techniques: Thermography and Ultrasonics can

Page 54: Volume 10 Issue 2 September 2011

52 Technical Paper

Journal of Non destructive Testing & Evaluation Vol. 10, Issue 2 September 2011

be adopted for detecting debond in the ZTA/Ti alloy bondedcomposite plates. It has been shown that thermographytechnique can be used for detection of debonding as wellas to check the quality/uniformity of bonding. Besidesthis, thermography can serve as a fast non-contact methodfor detecting debond and has the potential of automationas well as it can serve the purpose of quality checking.

ACKNOWLEDGEMENTS

The authors would also like to thank DRDO, India forfunding this activity. The constant support andencouragement from Director, Defence MetallurgicalResearch Laboratory (DMRL), Hyderabad is gratefullyacknowledged. We also thank all the technical staff ofDMRL for their valuable contributions.

REFERENCES

1. N.P. Avdelidis, A. Moropoulou and Z.P. Marioli Riga,“The technology of composite patches and theirstructural reliability inspection using infrared imaging”,Progress in Aerospace Sciences, 39 (2003) 317-328.

2. Maurice Cummings, Paul Biagioni, Philip John-Lameyand Donald J. Burden, “Thermal image analysis ofelectrothermal debonding of ceramic brackets; an invitro study”, European Journal of Orthodontics, 21(1999) 111-118.

3. S. Bagavathiappan, Y. Siva Sankar, M.C.S. Kumar,John Phillip, T. Jayakuamr and Baldve Raj, “Activeinfrared thermal imaging for quantitative analysis ofdefects and delaminations in composite materials”,Journal of Non destructive Testing & Evaluation, 8(2009) 28-36.

4. V. Dattoma, R. Marcuccio, C. Pappallettere and G.M.Smithm, “Thermographic investigation of sandwichstructures made of composite materials”, NDT & EInternational, 34 (2001) 515-520.

5. M. Menaka, S. Bagavathiappan, B. Venkataraman, T.Jayakumar and Baldev Raj, “Characterisation ofadhesively bonded laminates using radiography andinfrared thermal imaging techniques”, Insight 48 (2002)606-612.

6. K. Srinivas, A.O. Siddiqui and J. Lahiri,“Thermographic inspection of composite materials”,Proceedings of National Seminar on Non destructiveEvaluation (2006) 131-142.

7. W. Bai and B.S. Wong, “Evaluation of defects incomposite plates under consecutive environments usinglock-in thermography”, Meas. Sc. Technology, 12(2001) 142-150.

8. N. P. Acdelidis, B.C. Hawtin and D.P. Almond,‘Transient thermography in the assessment of defectsof aircraft composites”, NDT & E International, 36(2003) 433- 439.

9. C. Wallbrink, S.A. Wade and R. Jones, “The effect ofsize on the quantitative estimation of defect depth in

steel structures using lock-in thermography”, Journalof Applied Physics, 10 (2007) 101-108.

10. Sung Quek, Darryl Amond, Luke Leson and TimBarden, “A novel and robust thermal wave signalreconstruction technique for defect detection in lock-in thermography”, Meas. Sci. technology, 16 (2005)1223-1233.

11. C. Scarponi and G. Briotti, “Ultrasonic technique forthe evaluation of delaminations on CFRP, GFRP, KFRPcomposite materials”, Composited, Part B 31 (2000)237-243.

12. T. Meitzler, G. Smith, I. Wong, M. Charbeneau, E.Sohn, M. Bienkowski and A. Meitzler, “Crack Detectionin Armor Plates Using Ultrasonic Techniques,”American Society for Nondestructive Testing, MaterialsEvaluation, 66 (2008) 555-559.

13. F. Bastianini, A. Di Tommas and G. Pascale, “Ultrasonicnon-destructive assessment of bonding defects incomposite structural strengthenings”, CompositeStructures, 53 (2001) 463-467.

14. W. Hillger, “Ultrasonic imaging of internal defects incomposites”, NDTnet, 2 (1997).

15. C.V. Subramanian, M. Thavasimuthu, P. Palanichamy.D.K. Bhattacharya and Baldev Raj, “Evaluation of bondintegrity in sandwiched structures by dry couplantultrasonic techniques”, NDT & E International, 24(1991) 29-31.

16. L.S. Chang, T.H. Chuang and W.J. Wei,“Characterisation of alumina ceramics by ultrasonictesting”, Materials Characterisation, 45 (2000) 221-226.

17. M.Lethiecq and M. Perdrix, “Automatic discriminationtechniques for NDT of metal-ceramic bonds”, NDT& E International, 24(1991) 307-311.

18. Xavier Maldague, “Applications of infraredthermography in nondestructive evaluation”, New York:Willey.

19. X. Maldague, F. Galmiche, A. Ziadi, “Advances inpulse phase thermography”, Infrared physics andtechnology, 43 (2002)175-181.

20. X. Maldague, “Introduction to NDT by Active InfraredThermography”, Materials Evaluation, 6 (2002) 1063-1073.

21. H. Kaplan, “Practical Applications of Infrared ThermalSensing and Imaging Equipment”, Second Edition,Proc. Soc. of Photo-Opt. Instrumentation Eng. (SPIE),TT34 (1999) 160.

22. X. Maldague and S. Marinetti, “Pulse Phase InfraredThermography,” Journal of Applied Physics, 79 (1996)694-2698.

23. C. Ibarra-Castanedo, F. Galmiche, A. Darabi, M.Pilla,M. Klein, A. Ziadi, S. Vallerand, J.F. Pelletier and X.Maldague, “Thermographic nondestructive evaluation:Overview of recent progress’, Thermosense XXV(SPIE), 5073.

Page 55: Volume 10 Issue 2 September 2011

53Technical Paper

Vol. 10, Issue 2 September 2011 Journal of Non destructive Testing & Evaluation

NDE Technique for Reactor Core VibrationMeasurement in FBRs

R. Ramakrishna, P. Anup Kumar, M. Anandaraj, M. Thirumalai, V. Prakash,C. Anandbabu and P. Kalyanasundaram

Fast Reactor Technology Group, Indira Gandhi Centre for Atomic Research, Kalpakkam, IndiaEmail: [email protected], [email protected]

ABSTRACTThe Prototype Fast Breeder Reactor (PFBR) which is under construction at Kalpakkam, India, is a 500 MWe sodium cooledpool type reactor. The core of the PFBR consists of 1758 free standing subassemblies, out of which 181 are fuel subassemblies,supported on the grid plate. Coolant sodium flows axially from the bottom of the subassembly to top and it is in turbulent regime,which can excite flow induced vibration (FIV) of fuel subassemblies. Flow induced vibration is not desirable as it can cause failureof the fuel element clad tubes from fatigue, wear and vibration induced fretting. Excessive vibration can also cause reactivityfluctuations, rattling and power control problems. During commissioning of PFBR, it is planned to measure the subassemblyvibration in sodium at isothermal condition at 2000C.

Conventional contact type vibration sensors such as accelerometers, strain gages and displacement probes are not suited for hightemperature sodium environment and due to difficulty in sensor mounting. Ultrasonic technique is the only possible solution fornon-contact type vibration measurement in the primary sodium pool of fast reactors. Feasibility experiments were carried outin water to measure the subassembly vibration using ultrasonic technique. A full scale dummy fuel subassembly is erected in awater test loop and an electro-dynamic exciter is used to excite the subassembly with known frequency and amplitude. Ultrasonicsensor is mounted near the target (subassembly surface) and is operated in pulse-echo mode. An LVDT mounted directly on thesubassembly is used as the reference sensor to validate the results of ultrasonic technique. Ultrasonic A-scan signals are acquiredduring subassembly vibration. The vibration of the subassembly will cause a variation in the time delay between the transmittedpulse and received echo which is extracted from the ultrasonic A-scan signals. The measured vibration amplitude and frequencyfrom ultrasonic technique is compared with reference LVDT signal and the error was found to be less than 3%. The developedultrasonic technique proved to be a potential NDE method for subassembly vibration measurement during the commissioningof PFBR. In this paper the subassembly vibration measurement using ultrasonic technique is discussed with experiment detailsand results.

1. INTRODUCTION

Prototype Fast Breeder Reactor (PFBR), which is underconstruction at Kalpakkam, is a sodium cooled pool typereactor. The PFBR core consists of 1758 coresubassemblies which are supported in the grid plate. Thereare 181 fuel subassemblies in PFBR core with 217 fuelpins in each subassembly, vertically held in the form ofbundle within a hexagonal wrapper tube (hexcan). Thepins are separated by spacer wires wound around the pinshelically. The total height of the subassembly is 4.5 m outof which 3.9 m is above the grid plate. The nominal flowthrough the maximum rated subassembly is 36 kg/s.Coolant sodium flows axially from the bottom of thesubassembly to top and it is in turbulent regime, whichcan excite flow induced vibration (FIV) of fuelsubassemblies [1].

Flow induced vibration is not desirable as it can causefailure of the fuel element clad tubes from fatigue, wearand vibration induced fretting. Excessive vibration canalso cause reactivity fluctuations, rattling and power controlproblems. During commissioning of PFBR, it is plannedto measure the vibration of subassembly in sodium atisothermal condition at 200oC. Vibration measurement ofcomponents in general is carried out using accelerometers,strain gages etc. Accelerometers and strain gages must be

mounted directly on the component surface for vibrationmeasurement, which is not possible in case of sodiumimmersed components in a reactor. Hence ultrasonictechnique is the only viable solution for under sodiumvibration measurements.

Feasibility of using ultrasonic technique for the vibrationmeasurement was studied earlier in water. Ultrasonic B-scan imaging has been employed to determine the vibrationamplitude and frequency. In this measurement the ultrasonicsensor is kept stationary and the target (subassembly) isvibrated using an exciter. Fig.1 shows the B-scan imagerecorded for a subassembly excitation frequency of 2 Hzand 0.6 mm amplitude. The measured frequency from therecorded ultrasonic signal is 2.08 Hz and the amplitude isfound to be 0.59 mm [2].

Fig. 1 : B-Scan Image (Excitation 2 Hz; 0.6 mm)

Page 56: Volume 10 Issue 2 September 2011

54 Technical Paper

Journal of Non destructive Testing & Evaluation Vol. 10, Issue 2 September 2011

Based on this encouraging result obtained fromfeasibility studies, extensive experiments were carried outand signal processing techniques were developed toextract vibration signal from the ultrasonic A-scan signals.This paper discusses the details of the test section,instrumentation, and ultrasonic methodology,signal processing technique and its test results anddiscussion.

2. PRINCIPLE OF OPERATION ANDINSTRUMENTATION

Ultrasonic sensor operating in pulse-echo mode is used todetect the movement of subassembly. Ultrasonic sensor ismounted near the target (subassembly surface) and thetime delay between the transmitted pulse and reflectedecho is used to calculate the distance between the sensorand the target. The movement of the subassembly willcause a variation in the time delay, which will be arepresentation of the subassembly movement.

Ultrasonic sensor (Lithium-Niobate) operating at 4 MHz isused for the measurement. An electro-dynamic exciterattached to the subassembly is used to excite thesubassembly with known frequency and amplitude.Ultrasonic sensor is mounted near the target (facingsubassembly surface) and the time delay between thetransmitted pulse and reflected echo is used to calculatethe distance between the sensor and the target. Forvalidating the results obtained from ultrasonic technique,a Linear Variable Differential Transformer (LVDT) and anaccelerometer were attached to the subassembly. Fig.2shows instrumentation schematic employed.

3. EXPERIMENTAL PROCEDURE

A test loop was made of stainless steel, simulating thegeometrical arrangement as in PFBR. Full scale dummyfuel subassembly was fabricated and erected in the gridplate in the test section. Perspex embedded SS windowwas assembled in the test loop to view the movement oftop end of subassembly during measurements (Fig.3).

Fig. 2 : Instrumentation Schematic Setup

Fig. 3 : Test loop Perspex window Fig. 4 : Displacement Spectra

Page 57: Volume 10 Issue 2 September 2011

55Technical Paper

Vol. 10, Issue 2 September 2011 Journal of Non destructive Testing & Evaluation

Flow induced vibration (FIV) measurements were done inwater earlier to determine the vibration characteristics ofthe subassembly [3]. The first natural frequency of thesubassembly was identified as 3.6 Hz. Fig.4 shows thevibration spectra (displacement) recorded during the FIVstudies. These results gave an insight to the expectedmaximum amplitude of vibration of the subassembly andits natural frequencies.

In the current experiment, ultrasonic sensor is keptimmersed in water near top of the subassembly and it isfired continuously at a pulse repetition rate of 31 pulses/sec. Using the signal generator, power amplifier and electro-dynamic exciter, subassembly is excited and experimentswere conducted with different excitation frequenciesranging from 2 Hz to 6 Hz. The amplitude of subassemblyvibration was also varied from 0.1 mm to 1.2 mm.Experiments were repeated with different pulse repetitionfrequencies to study the effect of signal distortion.The output signals from the ultrasonic sensor and thereference LVDT were recorded using digital storageoscilloscope.

4. DEVELOPMENT OF SIGNALPROCESSING TECHNIQUE AND TESTRESULTS

As the ultrasonic sensor is fixed as the reference point,the movement of the subassembly with respect to the

reference will cause variation in the transit time betweenthe transmitted pulse and received echo. This variation intransit time represents the movement of the target(subassembly). Fig.5 shows typical time signal plot (seriesof A-scans) recorded from the ultrasonic sensor, depictingthe variation in time delay between the transmitted pulseand reflected echo.

Signal processing technique to extract the vibration signalfrom the ultrasonic pulse-echoes has been developed. Theultrasonic A-scan signal generated during the vibrationmeasurement consists of transmitted pulse and receivedecho. The time difference between the peak value oftransmitted pulse and received echo is a measure ofvibration (displacement) of subassembly. The received echois a signal with sinusoids of different amplitudes. It isvery difficult to read the peak values of transmitted pulseand received echo, directly from the raw time signal.Signal processing technique based on Hilbert transform isemployed here to extract the envelope of the ultrasonic A-scan signals. Consider ( )x t is the ultrasonic A-scan timesignal, then the Hilbert transform, y(t)=H{x(t)}, is definedas

(1)

Using the signal, and its Hilbert transform, an artificialcomplex signal, called analytic signal is generated (givenin eq.(2)) whose real part is the original time signal and

Fig. 5 : Ultrasonic time signal with pulse and echo trains

Fig. 6 : Envelope extraction using Hilbert Transform; a) Ultrasonic A-scan signal, b) Envelope of A-scan signal

Page 58: Volume 10 Issue 2 September 2011

56 Technical Paper

Journal of Non destructive Testing & Evaluation Vol. 10, Issue 2 September 2011

5. CONCLUSION

Ultrasonic technique and its signal processing algorithmhave been developed for the subassembly vibrationmeasurements in PFBR. Experiments were carried out ina water test loop using a PFBR full scale dummy fuelsubassembly. Amplitude and frequency of vibration of fuelsubassembly measured using ultrasonic technique iscompared with a reference LVDT sensor and the errorwas found to be less than 3%. The developed techniqueproved to be a potential and promising NDE method forvibration measurement, where the conventional contacttype sensors cannot be used.

REFERENCES

1. V. Prakash et.al, Experimental qualification ofsubassembly design for Prototype Fast BreederReactor, Nuclear Engineering and Design, Vol. 241,No.8, pp.3325– 3332, August 2011.

2. R.Ramakrishna et.al, Flow Induced Vibrationmeasurement using Ultrasonic Technique, InternationalConference on Sensors & Related Networks (SENNET2009), VIT, Vellore, India, Dec 07-10, 2009.

3. M. Anandaraj et.al, Flow induced vibration studies onPFBR fuel subassembly, 2nd International Conferenceon Asian Nuclear Prospects, ANUP 2010, Chennai,India, Oct 10-13, 2010.

Fig. 7 : LVDT reference signal

Fig. 8 : Vibration signal extracted from the ultrasonic A-scan signals

whose imaginary part is the Hilbert Transform of originaltime signal.

z(t) = x(t) + jy(t) = E(t)ejΨ(t) (2)

Where, E(t) is the instantaneous envelope which is themagnitude of the complex analytic signal, and Ψ(t) is thephase angle of analytic signal. Fig.6 shows the envelopeextraction of a typical ultrasonic A-scan signal acquiredduring subassembly vibration at frequency of 2 Hz andamplitude of 0.5 mm (pk-pk). After envelope extraction,the time difference (transit time) between the peak valueof transmitted pulse and received echo is found. As thevelocity of ultrasound in water is known, the displacementof target (subassembly) is calculated from the transit time.When subassembly vibrates, displacement of target changesand hence transit time changes. The transit time iscalculated for all the ultrasonic A-scan signals and thedisplacement signal is generated. Fig.7 shows the vibration(displacement) signal extracted from the ultrasonic signalsand Fig.8 shows the reference signal recorded from theLVDT for an excitation frequency of 2 Hz and amplitude0.5 mm (pk-pk). The amplitude and frequency of vibrationof subassembly measured using ultrasonic technique were0.51 mm (pk-pk) and 2.04 Hz respectively. This vibrationsignal is matching well with the reference signal fromLVDT. The signals from the ultrasonic sensor and thereference LVDT sensor were analyzed for varioussubassembly excitation frequencies and amplitudes, andthe overall error in this measurement technique is foundto be less than 3 %.

Page 59: Volume 10 Issue 2 September 2011

57Technical Paper

Vol. 10, Issue 2 September 2011 Journal of Non destructive Testing & Evaluation

On the conversion of multi-frequency “apparent”conductivity data to actual conductivity

gradients on peened samples

Veeraraghavan Sundararaghavan1 and Krishnan Balasubramaniam2

Center for Non-Destructive Evaluation, Machine Design Section, Department of Mechanical Engineering,Indian Institute of Technology-Madras, Chennai 600036, India.

2 E-mail: [email protected] Currently with Department of Aerospace Engineering, University of Michigan, Ann Arbor, MI, USA

ABSTRACTThis paper addresses the interpretation of “apparent” conductivity measurements as a function of frequency in order to determinethe actual conductivity profile as a function of depth in a conducting material. Simulations show that “apparent” conductivitiesare not indicative of actual conductivity gradients because of inherent constant conductivity approximation that is assumed atevery frequency. This paper focuses on facilitating the conversion of the multi-frequency “apparent” conductivity dataconductivity depth profiles through a Model based inversion scheme. The inversion uses a multi-layer axi-symmetric finiteelement model as the forward model and uses an optimal skin depth approximation for isolating the integral effects of theconductivity gradients on the multi-frequency “apparent” conductivity measurements. Unlike the inductance inversion methodthat has been reported elsewhere, this method does not depend on the sensor coil parameters and is robust enough to accommodatefor some common measurement uncertainties. Also, commercial multi-frequency conductivity measurement instruments can beused to obtain input data for the inverse model. Possible application of the model towards characterization of residual stressesin peened specimens is also addressed.

Keywords: Multi-Frequency Eddy Current Testing; Inverse Model; Conductivity Gradient Measurements

1. INTRODUCTION

Eddy current NDT techniques are well developed andhave been primarily applied as a means of detecting nearsurface discontinuities. The changing voltage in the sensorcoil induces eddy currents in a nearby conductor that inturn loads the coil and changes its impedance and phase.The depth of penetration of eddy currents can be controlledby the frequency of testing, due to the skin-depth effect,and hence, can be used to test components over differentdepths. Eddy current testing methodology has beensuccessfully applied over the years to address problemslike crack detection, material thickness measurements, heatdamage detection, “apparent” conductivity measurementsand for monitoring a variety of processes (ASNT (2004)).A method for assessing “apparent” conductivity involvesmeasurement of the impedance of coils, driven by a constantamplitude alternating current, placed above the samplesurface. Material is assumed to have a constant conductivityand the net conductivity is measured at various frequenciesusing eddy current absolute-coil configuration (ASTM E1004-99).

A conductivity measurement at a particular frequency hasinformation from the surface to a particular depth ofpenetration that is related to the skin-depth (d) in thatmaterial. Also, it represents an integrated value ofconductivity over this depth. Hence, when makingmeasurements of “apparent” conductivity at differentfrequencies, these measurements are not independent of

each other, since any measurement at a lower frequencyhas the information already existing in all of themeasurements made using higher frequencies.

Any NDE process may be considered to involve threesystems, each having a unique set of parameters thatdefine its characteristics viz. (a) The Input to the material,(b) The material itself, and (c) The output responsemeasured by the NDE system. Traditionally, the input andthe material parameters are assumed known and numerousForward Models have been developed that predict orestimate the output response function. Over the years,forward models are very well established and serve thekey purpose, for improved interpretation as well as tooptimize the input parameters to obtain the desired outputresponse. The other two scenarios i.e. if the output responsefunction in the form of measured data is available, toobtain one of system parameters, i.e. either the inputfunction or the material properties while the other one isassumed to be known are classified as Inverse Problems.Due to the availability of computational resources, theinverse problem solutions are becoming increasinglyfeasible. The traditional difficulties with the ill-posednessof the inverse solution (which includes lack of uniquenessor stability of the solution process) are increasinglybecoming solvable. Typical applications includemeasurement of material properties such as modulus,viscosity, temperature, hardness and stress profiles, etc.The formulation includes both numerical and analyticalsolutions in ultrasonics, eddy current and thermal imaging

Page 60: Volume 10 Issue 2 September 2011

58 Technical Paper

Journal of Non destructive Testing & Evaluation Vol. 10, Issue 2 September 2011

and the inverse solution process utilizes a variety oftechniques such as Neural Networks, Genetic Algorithms,Maximum Entropy Methods, etc (Liu, 2003).

Popular methods to invert eddy current impedance datainclude the use of regressive tools like neural networks ornumerical inversion schemes based on analytical or finiteelement based forward models. The use of neural networksfor conductivity inversion has been reported by severalauthors (Rekanos et al., 1997; Katragadda et al., 1997;Glorieux et al., 1999) with inverse models being trainedusing either experimental data or data from numericalmodels. The proposed “apparent” conductivity inversionscheme is primarily a numerical inversion scheme basedon a Finite element based forward model. Seminal workon numerical inversion of multi-frequency eddy currentimpedance measurements for characterizing coatingthicknesses and conductivities in layered materials can befound in Moulder et al. 1992. More advanced numericalinversion models (Bowler and Norton, 1993; Liu et al.,2000; Sun et al., 2002 etc.) primarily work by iterativelyadjusting relevant parameters in a forward model untilmeasured signal value is reached. Such techniques requireaccurate calibration of coil design parameters for use inthe forward model. The proposed “apparent” conductivityinversion model is different from other existing numericalmodels in that it does not employ any form of iterativerefinement and further, does not depend on coil designparameters, hence, can be used with commercialconductivity meters.

“Apparent” conductivity measurements using the eddycurrent sensor have been widely used as a basis forcharacterizing surface-treated specimens. This work wasmotivated by a study of finite element simulations thatshowed that the “apparent” conductivity does not followthe trends followed by the actual conductivity profiles.Specifically, an attempt is made here to study the effectof peening on true conductivity profiles of a specimenusing the proposed inversion scheme. Studies have beenperformed (Blodgett et al., 2003; Lavrentyev et al., 2000)to analyze the effect of peening on a material using changeof measured “apparent” conductivity with frequency. Theconductivity of the peened specimen continuously changesas a function of depth due to interacting effects of severalfactors including cold work gradients, surface roughness,and stress gradients. However, “apparent” conductivityprofiles are only able to provide a depth-averaged response

of a peened specimen. Hence, there is a need for anappropriate inversion scheme that can recover theinformation-rich actual conductivity profiles from suchmeasurements. This paper extends the inductance inversiontechnique already proposed (Sundararaghavan andBalasubramaniam, 2004; Sundararaghavan et al. 2005) toinvert multi-frequency “apparent” conductivity datameasured over nonmagnetic metals. The true conductivityprofiles show several promising trends that would possiblyallow characterization of residual stresses in peenedspecimens.

2. METHODOLOGY OF INVERSION

The critical input for the inversion model comes in theform of “apparent” conductivity measurements at variousfrequencies. The “apparent” conductivity measurement isbased on the notion that the material has a singleconductivity that contributes to the measured inductanceof the coil placed over the material. Given the “apparent”conductivity, the expected inductance change of any coilplaced over the material can be found using existing eddycurrent forward models which are either analytical (Doddet al., 1970) or based on the Finite element method. Inthis paper, finite element forward model (FEFM)(Palanisamy, 1980) is used to estimate the inductance valuesof a simulated coil placed over the material with the given“apparent” conductivity at a particular excitation frequency.For applications involving conductivity variation with depthin axi-symmetric testing situations, apart from Finiteelement based models, numerical models such as thosereported by Uzal et al. (1993) can also be used as theforward model. The proposed “apparent” conductivityprocessing procedure is depicted in Figure 1. Finite elementtechnique using the energy functional approach(Palanisamy, 1980) is used to solve the axi-symmetriceddy current governing equation for the magnetic vectorpotentials (A) in the discretized domain consisting of coil,material and air (Figure 2). The governing equation foraxisymmetric geometries is given by,

(1)

Given the “apparent” conductivity (σa) of the material atfrequency, f, the current density (Js) in the coil and knowngeometrical parameters (ro,ri,h,lo,d) as shown in Figure 2,the magnetic vector potentials (A) can be calculated overall nodes in the discretized domain. Only nonmagneticmaterials are considered, hence we use the permeability offree space, μ0. The impedance (Zcoil) of the coil whosecross section is discretized into N triangular finite elementsis then calculated as,

(2)

where j is the complex operator, Ns is the turn densityof the coil (turns/m2), Is is the current in the coil, ω is theFig. 1 : Proposed “apparent” conductivity inversion procedure

Page 61: Volume 10 Issue 2 September 2011

59Technical Paper

Vol. 10, Issue 2 September 2011 Journal of Non destructive Testing & Evaluation

angular frequency of the excitation current, and rcj, Acj,Δcj are the centroidal radius, centroidal magnetic vectorpotential and the area of the jth triangular element in thefinite element mesh respectively. The FEFM was verifiedby comparing the results with the analytical model reportedby Dodd and Deeds (1968). The self-inductance (L0) ofa coil, whose height, outer radius, inner radius and numberof turns are 6.35 mm, 9.525 mm, 3.175 mm, and 200respectively, was calculated using both techniques.Analytical solution yields a self-inductance of 3.217x10-4

Henries while the FEFM solution gives a self-inductanceof 3.216 x 10-4 Henries, a net error less than 0.1 %. Basedon these results, the confidence in the FEFM for axi-symmetric cases was established.

Once the measured “apparent” conductivities are convertedto inductance values of a simulated coil, a multi-frequencyinductance inversion model (Sundararaghavan andBalasubramaniam, 2004; Sundararaghavan, et al., 2005) isused to obtain the conductivity profiles. The conductivityprofile is assumed to be discontinuous, piecewise constantand each constant conductivity layer is modeled by severalrows of triangular elements. The multi-frequencyinductance data inversion scheme is depicted in Figure 3.During the inductance inversion process, frequencies arefirst sorted in the descending order. The highest frequency,corresponding to the least depth of penetration accordingto the optimal skin depth approximation, is used in thefirst solution step of the inverse model. Since the substrateconductivity is known, a two-layer model (optimal skindepth at the highest frequency and the substrate) can beused to separate the conductivity of the topmost layer.During this step, the finite element forward model assignsa range of conductivity values to the topmost layer andcalculates the inductance of the coil. The actual inductancevalue is then matched to a particular value of conductivityby rational interpolation of the conductivity-inductance data.

In the subsequent step of the inversion scheme, a lowercoil excitation frequency is used as the input to the inversemodel. In this case, the depth of penetration is higher thanthat of the first frequency input, and the eddy currentpenetrates the top layer whose conductivity was alreadyfound during the first step. A three-layer model with thistop layer, a second layer of unknown conductivity and the

substrate can be used to find the unknown conductivityby following a procedure similar to the first solution stepof the inversion process. In the nth step, an n+1-layeredmodel is used. The inversion method generates the depth-conductivity profile within ‘n’ steps. This technique canbe applied for measurements peened samples on whichthe stresses are inherently axi-symmetric (Sundararaghavanet al. (2005)).

Since the inverse model uses the apparent conductivitydata directly, only test specimen related parameters likebase conductivity and material thickness need to beprovided to the inverse model. The measurement isperformed at each frequency at two levels as discussedbefore, (1) Experimental measurement of “apparent”conductivity using a conductivity meter and (2) a computersimulated measurement of inductance of a simulated coilover the material using the set up shown in Figure. 2. Inthe simulated measurement, the virtual test specimen isgiven a uniform conductivity corresponding to the effectiveconductivity of the specimen as measured by theconductivity meter in step (1). The second step employsa simulated air-core coil whose parameters are given inTable. 1. This is an advantage over the multi frequencyinductance inversion scheme where customized coils werefabricated and the coil parameters and lift off have to bemeasured accurately and fed into the inverse model(Sundararaghavan et al., 2005). Since the proposed modelworks with apparent conductivities as an input, theinductance measurement is performed on computer usingthe Finite element model.

3. MODEL VALIDATION

The forward model is used to generate the “apparent”conductivity inputs for the inversion model given anyconductivity profile. The methodology for the forwardproblem is shown in Figure 4. Piecewise constantconductivity profiles were simulated on the material witha conductivity of 28 MS/m and a relative permeability of1 assigned to the substrate for all simulations. The FEFMis used to measure the inductance of a coil placed overthe material with the known conductivity profile. Thegeometrical parameters of the simulated coil used for thesimulations are specified in Table-1. A conductivity-

Fig. 2 : Finite element model configuration Fig. 3 : Multi-frequency inductance inversion scheme

Page 62: Volume 10 Issue 2 September 2011

60 Technical Paper

Journal of Non destructive Testing & Evaluation Vol. 10, Issue 2 September 2011

inductance calibration curve is used to calculate the“apparent” conductivities corresponding to the measuredinductances at various frequencies. The calibration curvefor any particular measurement configuration can begenerated using the finite element forward model.

Two different piecewise continuous conductivity profileswere simulated on the material and are plotted in Figure5. The “apparent” conductivities at 6 different frequenciesfor each of these profiles are shown alongside the simulatedprofiles. The “apparent” conductivity-frequency data wasinverted using the proposed algorithm using an optimumskin depth of 2d.

Table – 1 : Properties of the simulated pancake coil

Property Value

Number of Turns 240

Outer radius (ro mm) 4.3

Inner radius (ri mm) 2.5

Lift Off (lo mm) 4

Thickness (h mm) 15

A comparison of the actual conductivity profile and the“apparent” conductivity measurements in Figure 5 revealsthat the trends in “apparent” conductivity are not indicativeof the actual conductivity gradients in the material. Forexample in Figure 5(a), the actual conductivity gradient ismonotonically increasing whereas the “apparent”conductivity plot in Figure 5(b) does not show any suchtrend. It must also be noted here that for a monotonicchange in actual conductivity of about 25%, thecorresponding change in the “apparent” conductivity is ofthe order of only about 1%. Also, the “apparent”conductivity measurements are significantly influenced bythe top most layer conductivity. Similarly, for case inFigure 5b, a 10% change in actual conductivity bringsabout changes in the “apparent” conductivity of the orderof only 1%. This FEFM result clearly shows the reasonfor the lack of sensitivity of the “apparent” conductivitymeasurement to conductivity depth profiling during

Fig. 4 : Forward Problem: Obtaining “apparent” conductivity fromthe conductivity profile

Fig. 5 : Simulated conductivity profiles and “apparent” conductivity measurements at 6 different frequencies

Page 63: Volume 10 Issue 2 September 2011

61Technical Paper

Vol. 10, Issue 2 September 2011 Journal of Non destructive Testing & Evaluation

measurements of material properties such as stress asreported elsewhere (Lavrentyev et al., 2000; Blodgett et.al. (2003)).

The “apparent” conductivity data was then used an inputinto the inversion model and the reconstructed actualconductivity profiles are compared with the input actualconductivity profiles in Figure 6. The “apparent”conductivity inversion methodology was implemented usingC++ solvers for the eddy current forward FEM problemand a LabVIEW graphical user interface. The method iscomputationally efficient and consumes 0.1 seconds foreach “apparent” conductivity measurement on a 2.2 GHzPentium IV PC. Hence, the methodology has scope forapplicability in online monitoring where the profiles areexpected to be axisymmetric, in processes such as peeningand heat treatment.

An analysis of Figure 6 indicates that the correlationbetween the actual conductivity and the reconstructedconductivity is excellent near the surface, but thereconstruction error progressively increases with depth.The increase in reconstruction error with depth can beattributed to the skin depth approximation used in theproposed incremental layer approach. The skin depth (d)

in the proposed model was calculated based on thesubstrate conductivity since the actual conductivities arenot known prior to the reconstruction. Future work mightinvolve optimization of the skin depths based on an iterativeframework using an initial guess based on the substrateconductivity for the reconstruction. It must also be notedthat substrate conductivity is not easily obtained. Thesemay be coarsely approximated by the “apparent”conductivity measured over surface unaltered specimensat low frequencies or by using the intrinsic volumetricconductivity of the material.

4. SENSITIVITY OF “APPARENT”CONDUCTIVITY BASED INVERSION

It is difficult to quantify the error in the multi-frequencyinductance inversion scheme since the procedure isdependant on accurate measurement of multiple coilgeometry parameters. There is a need for an inversionprocedure that works regardless of the measurement sensordesign. In the “apparent” conductivity inversion technique,the only eddy current measurement input is the “apparent”conductivity. These values can be measured using severalavailable commercial conductivity meters within anaccuracy of +0.5% IACS by calibrating against reference

Fig. 7 : (a) RMS error in the reconstructed profile due to error in apparent conductivity input (b) Change in correlation of the reconstructedprofile with the actual profile due to error in apparent conductivity input

Fig. 6 : Simulated conductivity profile and the profile inverted from the “apparent” conductivity data

Page 64: Volume 10 Issue 2 September 2011

62 Technical Paper

Journal of Non destructive Testing & Evaluation Vol. 10, Issue 2 September 2011

standards at working frequencies. The success of anyinversion scheme depends on its stability within this rangeof uncertainty.

In order to test the robustness of the proposed inversionmethod, linear conductivity profiles were reconstructedusing “apparent” conductivities with simulated measurementerrors (“noise”) as input. Figure 7(a) shows the typicalchange in RMS error in the conductivity profilereconstructed using the proposed inversion scheme whendifferent amounts of “noise” are added to the “apparent”conductivity inputs. Figure 7(b) shows the change incorrelation of the reconstructed profile with the actualconductivity profile due to error in the “apparent”conductivity input. The “apparent” conductivity inversionscheme gives a RMS error of +0.09 MS/m or +0.3% ofthe substrate conductivity (s0) for an error in “apparent”conductivity measurement of +0.5% IACS. The quality ofcorrelation of the reconstructed profile with the actualprofile is excellent even for errors as large as 10% IACS.Hence, the proposed “apparent” conductivity inversionscheme is robust enough to accommodate for reasonablemeasurement uncertainties.

5. INVERSION FOR SHOT PEENEDSAMPLES

Fatigue cracks typically initiate from the surface since theoperating stresses are often maximum at the surface. Oneof the most popular methods to prevent crack initiation isto induce compressive residual stresses at the surface bymeans of peening. In the process of shot peening, a high-velocity stream of beads are used to plastically deform thesurface of a specimen. Within the plastically deformedlayer, compressive residual stresses are locked in and arebalanced by tensile residual stresses in the unaffected basemetal. Over a uniformly peened base metal, these residualstresses are known to be axisymmetric in nature allowinganalysis of “apparent” conductivity data using the proposedmodel.

The electrical conductivity of the cold worked layer islower than the conductivity of the underlying base metal.Eddy current apparent conductivity measurements over arange of frequencies in most peened materials usually displaya trend of decreasing “apparent” conductivity with increasein frequency (with the exception of certain alloys withpiezoresistive properties). At high frequencies the depth ofpenetration is low and the measurement is dominated bythe effect of the cold worked layer that effectivelydecreases the apparent conductivity. Due to large depth ofpenetration at lower frequencies, the effect of the coldworked layer near the surface is diminished and measuredapparent conductivity approaches the substrateconductivity. On the other hand, it is well known thatcompressive stresses result in an increase in conductivity.The effect of cold work is predominant close to the surfaceand the maximum compressive stresses (thus, maximumincrease in conductivities) are obtained at a certain depth

below the surface. The effect of compressive stressescan be, in principle, observed by taking “apparent”conductivity measurements at multiple frequencies thatwould selectively reach required depths in the sample.However, the intensity of eddy currents is largely affecteddue to the layers of decreased conductivity close to thesurface than the sub surface compressive stresses due tothe skin effect. “Apparent” conductivity measurementaverages out the conductivities over the depth of penetrationand hence, does not reflect the true underlying conductivityprofile. Further, effect of shot peening on measured“apparent” conductivity is quite modest at about 1 to 2%,making it difficult to quantify the change due tocompressive stresses. However, it would be interesting toobserve the trends in the actual conductivity profiles sincethe effect of cold work and compressive stresses on theconductivity at every depth inside the specimen can bequantitatively ascertained. In our previous study on waterjet peened samples, the effect of cold work was found tobe low and true conductivities provided trends consistentwith the effect of residual stresses (Sundararaghavan andBalasubramaniam, 2004; Sundararaghavan et al., 2005). Itis possible that similar trends might be observed in theprofiles obtained from shot peening although we expectcold work to be far larger under solid impact. As discussedbefore (see Fig. 5), small changes in measured “apparent”conductivity can result from large changes in trueconductivity over small depths, which can possibly providesufficient resolution to capture the effect of compressivestresses. In addition to cold work and residual stresses,surface roughness effect (Blodgett et al., (2003)) andtexture also causes a distortion to “apparent” conductivitymeasurements. The effect of crystallographic anisotropy(texture) on conductivity is assumed to be negligible andis not considered in the present study. Grid measurementmethods for independent conductivity and lift-offmeasurements have been reported (Washabaugh et al.,2000) that makes the measurements insensitive to surfaceroughness effect. Figure 8 shows one such result ofmultiple frequency apparent conductivity measurement(Washabaugh et al., 2000) with grid measurement methodsfor Al 2024 samples shot peened to Almen intensities of0.005, 0.012, and 0.017, Scale A.

In these “apparent” conductivity measurements, theunpeened sample conductivity was essentially constant withfrequency which validates the quality of the referencespecimen and provides the value of conductivity of theunaffected substrate (17.407 MS/m). The data shows aclear trend of decreasing conductivity with frequency,displaying the predominant effect of cold working. Trueconductivity profiles up to a depth of 0.384 mm wereobtained from this data set by inverting the “apparent”conductivity measurements at four frequencies of 2 MHz,1 MHz, 0.6 MHz and 0.4 MHz. These frequenciescorrespond to optimal depths of 0.168 mm, 0.24 mm,0.312 mm and 0.384 mm. The results are shown in Fig.9. Since the difference in depths between successive layersis less than 0.1 mm, the assumption of constant

Page 65: Volume 10 Issue 2 September 2011

63Technical Paper

Vol. 10, Issue 2 September 2011 Journal of Non destructive Testing & Evaluation

conductivity layers is expected to provide a goodapproximation. However, since “apparent” conductivity datawas not provided above 2 MHz, the true profile up to adepth of 0.168 mm was not known and was approximatedby a single layer of constant conductivity.

To validate the inverted conductivity profiles, apparentconductivities at 2-0.4 MHz were obtained from profilesin Fig. 9 using FEFM and are shown as dotted lines overthe experimental data in Fig. 8. The results show that thereconstruction accurately represents the trend in apparentconductivity up to a frequency of 0.6 MHz (0.312 mmoptimal depth). The reconstruction error becomessignificant over larger depths (lower frequencies) due tothe optimal skin depth approximation using the assumedsubstrate conductivity. For example, in the case of thehigh peening intensity sample (0.017 A), dotted lines inFig. 8 show a large deviation at 0.5 MHz since the optimaldepths are expected to be higher than those used in theinversion. This is because the effective conductivity ofthe substrate is lower than the assumed substrateconductivity due to a larger cold worked layer. On theother hand, the assumed substrate conductivity providesa good approximation for the 0.005A sample that has asmaller cold worked layer.

In all the peened samples, a large drop in conductivity isobserved close to the surface up to about 0.24 mm andthe drop in conductivity closely relates to the peeningintensity, the largest drop occurring for the sample withhighest peening intensity. This effect can be attributed tothe dominant effect of cold working over the near surfacelayers. The trend reverses for the 0.012A and 0.017Asamples, wherein a definite increase in conductivity isobserved from a depth of 0.24 mm, however theconductivity is still below substrate conductivity. The effectof cold work that decreases the conductivity is possiblyoffset due to the opposing effect of compressive stressescausing this increase. This effect increases with increasingpeening intensities as seen from the inverted profiles.Further experimental study of this effect (includingmeasurements at significantly higher frequencies) would

allow accurate characterization of conductivity profilesand possible calibration of residual stresses or the depthof maximum compressive stresses in peened specimens.Although the measurement beyond 0.312 mm carriessignificant reconstruction error for specimens with higherpeening intensities, the trend of a drop in conductivity atdepths between 0.312 and 0.384 mm is consistent for allthe peened samples. This might be due to the effects ofboth tensile stresses in the substrate as well as the plasticstrains that might exist at these depths. It is interesting tonote that significant changes in the true conductivity profilesare obtained even though percentage change in the apparentconductivities at these frequencies are small, which mightprovide the resolution required to capture the effect ofresidual stresses on conductivity.

6. CONCLUSION

Motivation of this work was a study of the use of“apparent” conductivity as a basis for characterizingsurface-treated specimens from processes like peening,heat treatment, cladding, coating etc. Simulations showthat the “apparent” conductivity does not follow the trendsfollowed by the actual conductivity profiles. Also, forlarge changes in the actual conductivity with depth, the“apparent” conductivity measurements show relatively poorsensitivities. Hence, there is a need to further process themeasured “apparent” conductivity in order to obtain theinformation-rich conductivity gradients. A new “apparent”conductivity inversion methodology has been presented inthis paper as an extension to the multi-frequency inductanceinversion methodology proposed by the same authors(Sundararaghavan and Balasubramaniam, 2004). Themethod is specific to planar layered geometries, iscomputationally efficient and is applicable for online testing.A study of sensitivity of the technique indicates that thescheme is robust over a range of measurementuncertainties. An advantage that this methodology enjoysis that it does not require extensive probe calibration ordesign and can be used along with any commercial sensoror conductivity meters that can work over a range of

Fig. 8 : Multiple frequency measurement results for Al 2024 alloyshot peened to Almen intensities of 0.005, 0.012, and0.017, Scale A as reported in Washabaugh et al. (2000).Dotted lines represent the apparent conductivity profilesreconstructed using specimens with conductivity profilesshown in Fig. 9.

Fig. 9 : Depth profiles of conductivity reconstructed from apparentconductivity data in Fig. 8. using the proposed inversionalgorithm

Page 66: Volume 10 Issue 2 September 2011

64 Technical Paper

Journal of Non destructive Testing & Evaluation Vol. 10, Issue 2 September 2011

frequencies. Inversion of “apparent” conductivitymeasurements of shot peened samples show large changesin conductivities near the surface even though the changesin “apparent” conductivities are quite modest. A decreasein conductivity due to cold working is possibly offset dueto the effect of compressive stresses at particular depthsand the trend is stronger with increasing peening intensities.This interesting observation might allow possiblecharacterization of residual stresses from peening.

Dependence on substrate conductivity for calculating theoptimal depth of penetration of eddy currents leads toerroneous calculation of optimal depths in applicationswhere large changes in conductivities are observed overthe depth of the specimen. Hence, the reconstructionerror progressively increases with decreasing frequenciesused in reconstruction. Future work in this area involvesimprovement of the inverse model by using an iterativeframework to eliminate the need for using optimal skindepth approximation based on substrate conductivity. Thiswould allow accurate inversion over a larger range offrequencies. Further experimental work needs to be carriedout to test the applicability of the technique for monitoringresidual stresses over peened specimens.

REFERENCES1. American Society for Nondestructive Testing,

“Electromagnetic Testing” Nondestructive TestingHandbook, Vol. 5, Udpa, S.S. and P.O. Moore, eds.,Columbus, Ohio, ASNT, 2004.

2. ASTM International, ASTM E 1004-99, StandardPractice for Determining Electrical Conductivity Usingthe Electromagnetic (Eddy-Current) Method, WestConshohocken, Pennsylvania, ASTM International,1999.

3. Blodgett, M.P., C.V. Ukpabi, and P.B. Nagy, SurfaceRoughness Influence on Eddy Current ElectricalConductivity Measurements, Materials Evaluation, Vol.61, 2003, pp.766-772.

4. Bowler, J.R. and S.J. Norton, “Theory of eddy currentinversion,” Journal of Applied Physics, Vol. 73, No. 2,1993, pp. 501-512.

5. Dodd, C.V. and W.E. Deeds, “Analytical solutions toeddy-current probe-coil problems” Journal of AppliedPhysics, Vol. 39, No. 6, 1968, pp. 2829-2838.

6. Glorieux, C., J. Moulder, J. Basart and J. Thoen, “Thedetermination of electrical conductivity profiles usingneural network inversion of multi-frequency eddy-current data,” Journal of Physics D: Applied Physics,Vol.. 32, 1999, pp. 616-622.

7. Katragadda, G., J. Wallace, J. Lee, and S. Nair, “Neuralnetwork inversion for thickness measurements andconductivity profiling,” Review of Progress inQuantitative Nondestructive Evaluation, Vol. 16A, D.O.Thompson and D.E. Chimenti, eds., Melville, AIP, July1997, pp. 781-788.

8. Lavrentyev, A.I., P.A. Stuky and W.A. Veronesi,“Feasibility of Ultrasonic and Eddy Current Methodsfor Measurement of Residual Stress in Shot PeenedMetals,” Review of Progress in QuantitativeNondestructive Evaluation, Vol. 19B, D.O. Thompsonand D.E. Chimenti, eds., Melville, AIP, July 2000, pp.1621-1628.

9. Liu, G., Y. Li, Y. Sun, P. Sacks, and S. Udpa, “Aniterative algorithm for eddy current inversion,” Reviewof Progress in Quantitative Nondestructive Evaluation,Vol. 19A, D.O. Thompson and D.E. Chimenti, eds.,Melville, AIP, July 2000, pp. 497-504.

10. Liu G.R. and X. Han, Computational Inverse Techniquesin Nondestructive Evaluation, Boca Raton, CRC Press2003.

11. Moulder, J.C., E. Uzal and J. H. Rose, “Thickness andconductivity of metallic layers from eddy currentmeasurements,” Review of Scientific Instrument., Vol.63, No. 6, 1992, pp. 3455-3465.

12. Palanisamy, R. “Finite element modeling of eddycurrent non-destructive testing phenomena,” PhD.Thesis, 1980, Colorado State University, Fort Collins,U.S.A.

13. Rekanos, I.T., T.P. Theodoulidis, S.M. Panas, T.D.Tsiboukis, “Impedance inversion in eddy current testingof layered planar structures via neural networks,”NDT&E International, Vol. 30, No. 2, 1997, pp.69-74.

14. Sun, H., J.R. Bowler, N. Bowler and M.J. Johnson,“Eddy current measurements on case hardened steel,”Review of Progress in Quantitative NondestructiveEvaluation, Vol. 21B, D.O. Thompson and D.E.Chimenti, eds., Melville, AIP, 2002, pp. 1561-1568.

15. Sundararaghavan, V., K. Balasubramaniam and N.R.Babu, “A multi-frequency eddy current inversionmethod for characterizing water jet peened aluminumalloys,” Review of Progress in QuantitativeNondestructive Evaluation, Vol. 23A, D.O. Thompsonand D.E. Chimenti, eds., Melville, AIP, 2004, pp. 651-658.

16. Sundararaghavan, V., K. Balasubramaniam, N.R. Babuand N. Rajesh, “A Multi-Frequency Eddy CurrentInversion Method for Characterizing ConductivityGradients on Water Jet Peened Components,” to appearin Int. J. of NDT&E, 2005.

17. Uzal, E., J.C. Moulder, S. Mitra and J.H. Rose,“Impedance of coils over layered metals withcontinuously variable conductivity and permeability:Theory and experiment”, Journal of Applied Physics,Vol. 74, No. 3, 1993, pp. 2076-2089.

18. Washabaugh, A., V. Zilberstein, D. Schlicker and N.Goldfine, “Absolute Electrical Property MeasurementsUsing Conformable MWM Eddy-Current Sensors forQuantitative Materials Characterization,” In:Proceedings of ROMA 2000 - 15th World Conferenceon Non Destructive Testing (WCNDT), Roma, Italy.AIPnD, 2000 (http://www.ndt.net/article/wcndt00/).

Page 67: Volume 10 Issue 2 September 2011

Ultrasonic Non-Destructive Evaluation (NDE) basedinternal inspection of pressure vessels for better

maintenance practice

S.K.Nath1 and B.H.Narayana1

1 Central Power Research Institute, Thermal Research Centre, Nagpur-441 111, Maharastra, IndiaE-mail address: [email protected]

ABSTRACTThis paper discusses about the possibility of detection of entrapped foreign object in a pressure vessel by ultrasonic inspectiontechnique. The inspection plan is designed and illustrated here. Successful in-situ implementation of this technique will help inachieving better maintenance practice for the plant components.

INTRODUCTION

Availability of any subsystem in a thermal power plantplays a very important role in uninterrupted generation ofelectricity. The plant consists of number of units based onthe total generating capacity. The typical capacity of onesuch unit may be from 110 MW to 660 MW. The steamgenerator of a unit of a power plant namely the boiler isone such important subsystem. A typical boiler consists ofvarious pressure vessels e.g. headers, drums meant forcontaining the working fluids namely water and steam ofvarying temperature and pressure in the circuit. Frequentforced outages reduce its availability. One of the majorcauses of forced outages of the boiler is the entrapmentof foreign objects inside such closed pressure vessels i.e.headers. The objects could be anything like welding rods,iron files, insulation materials, wooden pieces etc. Theseunwanted foreign objects restrict the flow of the workingfluid causing starvation in the tubes which eventually leadsto leakage. Detail failure investigation of such tubes confirmsthe presence of foreign objects inside the headers. Thusdetection and retrieval of the foreign objects in the vesselis of prime importance to reduce the forced outages ofthe boiler causing huge generation loss.

Generally fibroscopic inspection is carried out for detectingthe entrapped foreign material inside the headers. Inspectionnipples or stub joints are cut for creating the opening forinserting the fibre-optic probe inside the otherwiseinaccessible location of the headers. The inspection willdetect the foreign material if it is physically present inside;otherwise no such detection will be made. However,irrespective of the detection or not, the openings createdfor this inspection need to be closed before restart of theplant. This requires re-welding of the cut portion, stressrelieving based on thickness and radiography to check theweld quality. In case of absence of any foreign material,these are additional work serving no meaningful purposesand the same can be avoided if the information that “noforeign material is inside the header” is known by some

other means. Thus a priori information regarding thepresence or absence of the foreign material will help inoptimizing the cutting, welding, radiography work.

An ultrasonic inspection plan is developed in the presentinvestigation which will provide a priori informationregarding the presence or absence of the foreign materialinside the headers. The inspection will help in bettermaintenance planning with respect to optimum utilizationof time and resource of plants and industries.

Inspection plan

A schematic diagram of the inspection plan is illustrated inFigure 1 below.

The pressure vessel containing an unwanted foreign objectis partially filled with stagnant water. The object couldeither be in suspension in water or resting in the bottomportion of the vessel depending on its specific gravity.Two ultrasonic probes; one transmitter and another receiverare placed on the surface of the vessel in a pitch-catcharrangement and scanning is performed along thelongitudinal direction as in ultrasonic Time of FlightDiffraction (TOFD) inspection. However, in TOFD, thediffracted beam from the flaw is considered for itsdetection and sizing. Here in this case we are using the

Fig. 1 : Schematic diagram of longitudinal inspection plan

Page 68: Volume 10 Issue 2 September 2011

66 Technical Paper

Journal of Non destructive Testing & Evaluation Vol. 10, Issue 2 September 2011

reflected beam from the reflector (i.e. the foreign object).The space (S) between the two probes is optimized basedon the metal thickness of the vessel and the water levelinside it. Moreover, the space is determined in such a waythat the central axes of both transmitter and receiver probesmeet at the water-air interface. Central axes of the beamsare of maximum energy levels [1]. Thus the same areconsidered with an objective of maximizing transmittedand received energy. Longitudinal wave probes are usedbecause of the obvious reason of having liquid mediumwherein shear mode of ultrasound cannot propagate.Additional advantage of longitudinal wave is its more energycontent as compared to shear wave [2]. Details of theultrasonic wave propagation, its interaction with differentinterfaces having varying media are illustrated in Figure 2(a and b).

Probe centre spacing (S), transmitted beam path (B) basedon the simple trigonometric relationship are determined bysolving Eq. (1)-(5). The metal thickness‘t’ and the waterlevel ‘h’ used in the equations are determined separatelyby conventional normal beam pulse echo method.

(1)

S1 = t tanθ; S2 = h tanϕ (2)

S = 2 * (S1 + S2) (3)

B1 = t Secθ; B2 = hSecϕ (4)

B = 2 * (B1 + B2) (5)

θ and ϕ = Angle of incidence and transmissionrespectively

V1 and V2 = Ultrasonic velocity in metal and waterrespectively

t and h = Metal thickness and water levelrespectively

S = Spacing between transmitter and receiverprobes

B = Total beam path between transmitter andreceiver probes

The probe pair is scanned on the outer bottom surfacealong a line marked by a vertical plane through the centreline of the vessel. During scanning part of the incidentultrasound beam will travel from the transmitter to receiverprobe as creeping/surface wave [3]. As long as there is noobstruction in the path as illustrated by position ‘A’ and‘B’ in Figure 1, rest of the incident beam will propagatefrom metal to water, reflect from the water-air interface,the reflected beam will subsequently get transmitted backinto the metal and finally received by the receiver probe.The various signals generated during scanning areschematically illustrated in Figure 2 (b).

In the above Figure 2 (b) there are two clearly visiblesignals, one is the lateral wave (LWE) or creeping waveand the other is the reflected (from water-air interface)and subsequently transmitted (at water-metal interface)signal (BWE). As long as there is no obstruction in the

Fig. 2 : Schematic illustration of (a) ultrasonic wave propagation in varying media (b) signals in case of absence of foreign object (c) signalin case of presence of foreign object

Page 69: Volume 10 Issue 2 September 2011

67Technical Paper

Vol. 10, Issue 2 September 2011 Journal of Non destructive Testing & Evaluation

path of the interrogating beam, both the signals will appear.The moment the beam encounters an obstruction e.g.foreign object as illustrated by position ‘C’ in Figure 1;there will be unpredictable and irregular reflection,dispersion of the beam and in all probability no beam willreach the receiver. Thus in such a case, though there willbe the 1st signal i.e. lateral wave (LWE) as always, no 2nd

signal i.e. the reflected wave (BWE) will appear. Figure 2(c) illustrates this case.

During inspection the operator should look for the presenceof both the signals (LWE and BWE) which confirms theabsence of any foreign object. However, absence of the2nd signal i.e. BWE prima facie confirms the presence ofthe foreign object.

DISCUSSION

The present study shows the possibility of ultrasonicallydetect the entrapped foreign objects inside a pressure vessel.The inspection plan designed here is for partially filledvessel. For completely filled vessel, the inspection planwill be more or less same with minor variation in probecentre spacing (S) and beam path (B) calculation. Themetal wall of the vessel is assumed to be defect free;otherwise any such defect here also will cause the absenceof the 2nd signal leading to the misinterpretation of presenceof foreign object inside the vessel. Thus before actualinspection the metal wall should be separately examined topre-empt any such possibilities.

There could be various other signals due to multiplereflections within the metal wall or the water. In actuallaboratory experiment or in-situ application these signalsshould be clearly identified and filtered out to avoid anyfalse interpretation.

The inspection plan under discussion is applicable for apartially water filled pressure vessel. Requirement ofinspecting dry vessels also may arise. In such case,different inspection plan probably deploying the techniqueof air-borne ultrasound can be developed. However, thisis not in the purview of the present investigation.

As per the scanning plan so far discussed, the probe pairwith a fixed spacing between them is moved longitudinallyon the outer bottom surface along a line marked by avertical plane through the centre line of the vessel. Thusthe foreign object, if any lying in the inspection regionalong the length of the vessel will be detected. However,the possibility of the object lying on either side of thisinspection region is not ruled out and in such cases, thesame is likely to be missed. A circumferential scan asillustrated in Figure 3 can overcome this situation.Inspection with different spacing between the two probescan be performed to scan the entire volume of waterbody.

The inspection plan developed here may be validated by alaboratory experiment. Once validated experimentally the

same can be successfully implemented during siteinspection. The main perceivable advantage is that thisinspection will confirm the presence or absence of anyforeign object inside a vessel. In case of confirmationregarding the presence of an object by this inspection, theinspection nipple or the tube stub joint can be cut open forinserting the fibreoptic probe for further verification andretrievable purpose and subsequently the cut portion re-welded, stress relieved and radio-graphed. It is reiteratedhere that in case of confirming the absence of a foreignobject by the present ultrasonic inspection, the entirecutting/re-welding/stress-relieving/radiography works canbe judiciously avoided evolving a better maintenancepractice.

ACKNOWLEDGEMENT

The authors are thankful to the management of CPRI forconstant encouragement, motivation and support forresearch and developmental activities.

REFERENCES

1. Charlesworth JP, Temple JAG. Engineering Applicationsof Ultrasonic Time of Flight Diffraction. 2nd. Ed., 2001,Research Studies Press Ltd., England.

2. Ogilvy JA, Temple JAG. Diffraction of Elastic Wavesby cracks: Application to Time of Flight Inspection.Ultrasonics 1983; 7: 259-269.

3. Nath S.K., Balasubramaniam K., Krishnamurthy C.V.and Narayana B.H., ‘Sizing of surface-breaking cracksin complex geometry components by ultrasonic Time-of-flight Diffraction (TOFD) technique’, Insight, 49(4),200-206, (2007).

Fig. 3 : Schematic illustration of circumferential inspection plan

Page 70: Volume 10 Issue 2 September 2011

68 Technical Paper

Journal of Non destructive Testing & Evaluation Vol. 10, Issue 2 September 2011

PROBE

This article commences with a slight change in the list of SEVEN enumerated in the earlierissue. The new list is 1. Prioritization, 2. Perception, 3. Possibility Thinking, 4. Process, 5.Procrastination – A strict NO NO, 6. Pondering over the success, 7. Patience and Perseverance.This article will deal with Prioritize.

Prioritization sets the direction of life. For many of us success in life is equated to possessionof material things. The more and expensive the gadget the better it is. It is construed that Themore the possession the more is happiness and hence success. We assume that Success &Happiness are directly proportional to the amount of material possession. But in reality it isnot so. The more the possession the more is the fear and worry. Since we get used to aparticular kind of possession, the absence causes more pain (Air conditioning). Acquiringmaterial possession provides a momentary “KICK” and is accompanied by pain. Remember “NO pain No gain”.

Once you have acquired the material object, the mind is un satiated. It craves for more andmore. There is an un satiated longing or in other words “GREED” sets in. We must be able todifferentiate between “Need and Greed”. Life shall be need based rather than greed basedlike inspection. For example take an obese person or an alcoholic. It all starts with an extramorsel of food or a peg of alcohol. A smoker is converted into a chain smoker because themind that is possessed by the greed converts the smoking into a need for the body. Let usrealize that mind controls the body and not the other way. Mind is superior to body. Materialspossession is an acceptance criteria of success set by the mind.

Be it known that we are neither the mind nor the body but the inner consciousness theprimordial energy (the life) is supreme. This consciousness may be called by any name. Afterall what is there in a name. A rose by any name will always be a rose. This energy is Godparticle (Boson). This God particle is all pervading and we should be guided by it as it is thecore of a being and the being is built around it. Now the question is how do we do it? Bysitting still, keeping both the body and mind still this can be achieved. That is action ininaction. This may sound contradictory and prove to be difficult in the beginning but practicemaketh a man perfect. By being still you will dig deep into yourself and start listening to yourinner self. Once you start listening to your inner voice the inner voice will guide you.

Basically it is based on the principle of “What feeling you get when you rewind your inningsof life at the autumn of your life? Is it of satisfaction or is it of regret! As Stephen Covey theauthor 7 Principles of great men says 4 Ls - Live, Love, Learn and Leave a Legacy shall governyour life. If you consider each L as a circle then the more the 4 circles merge with each otherthen more will be the satisfaction in life. Live- The physical Needs, Love- The emotional Need(acceptability by the society), Learn – The need of the intelligence and Leave a legacy- theneed of the self.

Try this, you have nothing to loose , except probably anger, jealousy, lust, sorrow, worries andall other negativities including ego (the ones that causes secretion of unwanted chemicals inyour body) but a lot to gain including tranquility, peace, happiness, satisfaction and may beself actualization. Recently we witnessed the uprising of masses against corruption lead by asimple person which probably can be equaled to the one lead by Mahatma Gandhi.. The mottoshall be to learn to prioritize.

68 Technical Paper

Journal of Non destructive Testing & Evaluation Vol. 10, Issue 2 September 2011

Page 71: Volume 10 Issue 2 September 2011
Page 72: Volume 10 Issue 2 September 2011

Registered with the Registrar of Newspaper of India. Registration No. TNENG/2002/7666Registered with the Registrar of Newspaper of India. Registration No. TNENG/2002/7666

R. BalakrishnanConvener NDE 2011Phone : 044-6538 6075 / 4553 2115Mobile: 98406 [email protected]@gmail.com

www.nde2011.comOrganised byIndian Society for Non Destructive Testing, Chennai Chapter

nde2

NDE 2011

National Seminar & Exhibition onNon - Destructive Evaluation

December 8-10, 2011 - Chennai, India

All Payment in the form of Demand Draft in favourof NDE-2011- payable at SBI, ChennaiISNT

Name: NDE-2011-ISNTBank: State Bank of India, Guindy, Chennai 32NEFT / IFSC Code: SBIN0000956

Account No. 31652401557

Prof. Krishnan BalasubramaniamChairman, Technical [email protected]

Enrich & Excel in Engineering thro’ NDE

Silver Sponsors

Blue Star, Chennai

SupportersGold Sponsor

Pre Conference Workshop6 - 7 December 2011

RLA of Boilers and Turbines

NDE in Civil Concrete

NDE in Automobile

NDE in Aerospace