research article remarks on residual stress measurement …

Research ArticleRemarks on Residual Stress Measurement by Hole-Drilling andElectronic Speckle Pattern Interferometry

Claudia Barile Caterina Casavola Giovanni Pappalettera and Carmine Pappalettere

Dipartimento di Meccanica Matematica e Management Politecnico di Bari Viale Japigia 182 70126 Bari Italy

Correspondence should be addressed to Caterina Casavola casavolapolibait

Received 16 April 2014 Revised 20 June 2014 Accepted 22 June 2014 Published 6 July 2014

Academic Editor Guangxiong Chen

Copyright copy 2014 Claudia Barile et al This is an open access article distributed under the Creative Commons Attribution Licensewhich permits unrestricted use distribution and reproduction in any medium provided the original work is properly cited

Hole drilling is the most widespread method for measuring residual stress It is based on the principle that drilling a hole in thematerial causes a local stress relaxation the initial residual stress can be calculated bymeasuring strain in correspondence with eachdrill depth Recently optical techniques were introduced to measure strain in this case the accuracy of the final results dependsamong other factors on the proper choice of the area of analysis Deformations are in fact analyzed within an annulus determinedby two parameters the internal and the external radius In this paper the influence of the choice of the area of analysis was analysedA known stress field was introduced on a Ti grade 5 sample and then the stress was measured in correspondence with differentvalues of the internal and the external radius of analysis results were finally compared with the expected theoretical value

1 Introduction

Almost every material presents as results of production andmanufacturing processes an intrinsic field of stresses knownin the literature as residual stresses In some cases they arealso intentionally introduced in order to increase fatigueresistance and fracture strength Due to the relevance of thetopic a number of techniques have been developed in orderto measure residual stresses X-ray diffraction can be used tomeasure interatomic strain caused by the presence of residualstresses (RS) introduced by different industrial processessuch as those connected to shot peening [1] or friction stirwelding [2] This measurement allows calculating residualstresses once Youngrsquos modulus and Poisson coefficient areknown This kind of measurement is not destructive but inmany cases it is limited to a few microns below the exposedsurface Penetration depth is increased up to tens of mmin systems using hard X-rays from synchrotron facilities[3] Alternatively also diffraction of neutron beams can beexploited to determine residual stress as it was demonstratedon powder metallurgy component in [4] and in weldedspecimen in [5] Both these systems are very expensive anddifficult to implement

Magnetic properties such as Barkhausen noise canalso be exploited to evaluate residual stresses but only inferromagnetic materials [6] however the Barkhausen noisedepends on the hardness of the sample and for this reason itrequires a calibration procedure in order to get quantitativeresults [7] In the ultrasonic methods the change in velocityof an ultrasonic wave propagating in the specimen can berelated to the level of internal stress in the component [8]Photoelasticity can be applied for measuring residual stressesin glasses [9] Also Raman and fluorescence spectroscopictechniques have been applied to evaluate the presence ofresidual stresses [10] Nowadays the hole-drilling method(HDM) is the most widespread one [11] and it is also ruledby a standard [12] In this technique residual stresses arecalculated starting from the on-surface measured strainsafter drilling a hole in the specimen This technique issemidestructive and nowadays there are some attempts toget stress relaxation in a nondestructive way by focusing alaser source on the sample to be analyzed to get local stressrelaxation [13] In hole-drilling method strain measurementis performed by strain gage rosette this kind of procedureis quite reliable however it involves some drawbacks Forexample the surface of the sample needs to be prepared

Hindawi Publishing Corporatione Scientific World JournalVolume 2014 Article ID 487149 7 pageshttpdxdoiorg1011552014487149

2 The Scientific World Journal

in order to attach the rosette and also the positioning ofthe drill bit must be accurately performed in order to avoideccentricity errors [14] Furthermore only a few data pointscan be obtained by strain gage and this can introducesignificant errors in the stress calculation procedure Finallyit should be considered that the strain gage rosette hasa cost which cannot be considered negligible particularlyin the case where a relevant number of measurements arerequired Optical methods based on laser interferometry arepromising in replacing strain gage in view of the fact thatthey can allow obtaining high sensitivity measurements froma huge number of data points without contact and allowingaccurately defining on the recorded image the position ofthe drilled hole [15] Moire interferometry [16] phase shiftshearography [17] and electronic laser speckle interferometry(ESPI) were successfully adopted in many different situationsto measure displacement fields ESPI was used to measuredisplacement field in anisotropic specimen made by selectivelaser melting [18] sinterization [19] or laminated wood [20]Now ESPI is under study for the specific application inmeasuring residual stresses [21] Specific studies were carriedout in order to establish the effects of different parametersinvolved in the residual stress measurement process on thefinal accuracy In [22] effects of several process parameters inRS measurement in titanium plates were investigated whilethe effects of errors in the determination of the geometricalparameters characterizing the optical system are discussed in[23] Finally the effects connected to a proper choice of thedrilling speed are reported in [24]

In this paper the attention is focused on the phase shiftingelectronic speckle pattern interferometry used in combi-nation with hole-drilling technique for measuring residualstress profileThis method being an interferometric methodcan guarantee sensitivity of the order of a fraction of theillumination wavelength it does not require any particularpreparation of the surface being applicable also to roughand curve surfaces and it does not require application of agrating as in moire interferometry [25] The main drawbackof this technique is the necessity to properly isolate the systemfrom vibrations in order to obtain fringe pattern with goodcontrast

Several works are present in the literature where ESPIhas been used to measure strain and calculate residual stressFocht and Schiffner used an ESPI set-up to measure strainin thin metal sheets [26] a double beam ESPI interferometerwas used by Baldi [27] to evaluate stresses in orthotropicmaterials Dıaz et al measured by ESPI the stress relieved byhole drilling in an aluminum plate subjected to a uniformuniaxial tensile stress [28] Viotti and Kaufmann evalu-ated the accuracy and the sensitivity of this hole-drillingand digital speckle pattern interferometry (DSPI) combinedtechnique on aluminum thin plates subjected to a uniformuniaxial tensile stress [29] and described a portable deviceto measure residual stresses [30] a mathematical methodwas proposed by Schajer and Steinzig for calculating residualstresses from hole-drilling electronic speckle pattern inter-ferometry (ESPI) data independent of rigid-body motions[31] a method to cancel rigid body displacements that can beintroduced when a hole-drilling and digital speckle pattern

interferometry (DSPI) combined system is used to measureresidual stresses was proposed by Dolinko and Kaufmannbased on a least-square approach [32]

In this paper some experimental considerations abouterror sources in the ESPI hole-drilling method will bediscussed In particular the influence of the area of analysison the final results in terms of measured residual stresses willbe discussed The inner and the outer radius of the area ofanalysis will be changed and the corresponding variations inthe measured residual stress profile will be calculated

2 Materials and Methods

21 Experimental Set-Up for HDM+ESPI Measurements TheESPI hole-drilling measurement system used in this work isschematically reported in Figure 1 The laser source used inthe proposed set-up was a diode-pumped solid-state (DPSS)laser source with a wavelength 120582 = 532 nm Radiationwas split into two beams and focused on two monomodeoptical fibersOne beam is collimated through a biconvex lensand illuminates the sample while the second beam passesthrough a phase shifting piezoelectric system and then goesto a 640 times 480 pixel charge-coupled device (CCD) camerawhere interferes with the light diffused by the optically roughsurface of the specimen Camera is equipped with an opticalimaging system allowing fine focusing of the image Theintensity recorded by the camera at each (119894 119895) pixel can bedescribed by

119868 (119894 119895) = 119868ref (119894 119895) + 119868obj (119894 119895) + 2radic119868ref119868obj cos (120593 (119894 119895)) (1)

where 119868ref(119894 119895) is the intensity of the reference beam 119868obj(119894 119895)is the intensity of the object beam deformations and 120593(119894 119895) isthe initial phase difference

If the surface of the specimen undergoes some displace-ment along direction of the sensitivity vector describing thesystem the speckle pattern experiences a phase differenceΔ120593(119894 119895) and the new intensity 119868(119894 119895) recorded by the CCDcamera at each (119894 119895) pixel can be described by

119868 (119894 119895) = 119868ref (119894 119895) + 119868obj (119894 119895)

+ 2radic119868ref119868obj cos (120593 (119894 119895) + Δ120593 (119894 119895)) (2)

where Δ120593(119894 119895) is the additional phase difference connectedwith the displacement 119889 experienced by the surface

In order to extract the value of the additional phase thefour-step phase shifting algorithm was implemented [33] Inorder to do this in correspondence with the undeformedsample and for each stage of deformation of the specimen fourimages were recorded A given image differs from the nextone in view of the fact that piezoactuator is moved so that

The Scientific World Journal 3

Sample

Laser source+

phase shifting system

Drilling system

controller

Illumination beam

CCD

Reference beam

Fiber optics Zoom lens

Drilling system

Figure 1 Experimental set-up for ESPI measurements of strainsrelaxed by HDM

a known supplementary phase value equal to 1205872 is addedEquations (1) and (2) can be rewritten as

119868119894(119894 119895) = 119868ref (119894 119895) + 119868obj (119894 119895)

+ 2radic119868ref119868obj cos(120593 (119894 119895) + 119899120587

2)

119868119894(119894 119895) = 119868ref (119894 119895) + 119868obj (119894 119895)

+ 2radic119868ref119868obj cos(120593 (119894 119895) + Δ120593 (119894 119895) + 119899120587

2)

(3)

with 119899 varying from 1 to 4The additional phase Δ120593(119894 119895) can now be obtained by

Δ120593 (119894 119895) = tanminus1((1198681minus 1198683) (1198681minus 1198683) + (1198682minus 1198684) (1198682minus 1198684)

(1198681minus 1198683) (1198682minus 1198684) minus (1198682minus 1198684) (1198681minus 1198683))

(4)

and finally displacement can be obtained by reminding thebasic equation

Δ120593 = sdot 119889 (5)

where is the sensitivity vector of the interferometric systemThe hole is drilled by means of an air-cooled high-

speed turbine Maximum rotation speed achievable withthis turbine was 50000 rpm The rotation of the turbine iselectronically controlled so that it stays constant along allthe drilling process Experimental tests shown in this paperwere all run at 50000 rpm in view of the fact that preliminaryexperiments displayed higher quality of the hole for thismaterial at higher rotation speeds In order to perform andcontrol the displacement of the turbine it was mounted overa linear actuator having a travel of 25mm and a resolutionof about 005 120583m with bidirectional repeatability less than15 120583mThe cutter is made by tungsten coated with TiCN andhas a nominal diameter 119889 = 159mm

Experimental measurements were performed on a tita-nium grade 5 specimen (2485mm times 425mm times 30mm) In

Figure 2 Screenshot of the area of analysis included between theouter circle (dashed line) and the inner circle (dotted line)The solidline identifies the drilled hole

order to generate a known stress field to be used as a referenceto test the effect of using different analysis parameter thespecimen was introduced in a four-point bending frameThe distance between the inner supports was 133mm thedistance between the external supports was 208mm Thedistance between the centers of the holes was set to 12mmto minimize the interaction between the holes Holes weredrilled at 120mm from each edge in order to minimizeedge effects Before loading the sample however preliminaryX-ray residual stress measurement was performed in orderto evaluate that initial stress field on the specimen and avery low value of about 10MPa was found Subsequently theHDM+ESPI method was utilized to confirm this ldquounloadedrdquostress field (the hole was drilled till 04mm depth in thecenter of the specimen each step was 005mm) The samedrilling parameters were also utilized for the subsequent testson the loaded specimen Concerning the calculation step119889calc = 01mm was chosen in order to minimize the effectsconnected with random errors in strain [34 35] and to betterevidence effects connected with the choice of the area ofanalysis

22 Considerations on the Area of Analysis Strain measure-ments around the drilled hole were performed within anarea delimited by two circles concentric with the drilled holeFigure 2 displays how the area of analysis is defined Thecenter of the area of analysis coincides with the center ofthe circle defining the edges of the drilled hole Around thiscircumference twomore concentric circles are defined Pixelsthat will be effectively analyzed are only those included in theannular region delimited by those two circles

The size of the analysis area can affect the results interms of residual stresses In fact if the radius of the innercircle is too small a region where plasticity effects occur isincluded in the calculation Also particles generated duringthe drilling process can lay down around the holes Theyconstitute a noise for the interferometric measurement sothat this occurrence can act as a further source of error Onthe other side if the outer radius is too large a region ofvery small deformations can be included and this can leadto erring in the residual stress results


y

x

01mmy

x

02mm

y

x

03mm y

x

04mm

Figure 3 Speckle correlation fringes recorded after four drilling increments obtained by subtraction the real time recorded pattern from theinitial reference pattern

In order to evaluate the influence of the analysis area thestress fieldwas initiallymeasured in correspondencewith twofixed values of the inner radius (119877int) and of the outer radius(119877ext) which delimitates the analysis area around the hole

In order to highlight the influence of each variable on theobtained stress values two situations were analyzed

The first analysiswas performed by changing the innerradius in the range 119877int isin [111mm 159mm] and keeping119877ext unchanged Analogously the outer radius ratio waschanged in the range 119877ext isin [207mm 262mm] whilekeeping 119877int constant The upper value 262mm of the 119877extrange was limited by the image dimension

3 Results

Recorded correlation fringe patterns obtained in correspon-dence with four different drilling steps are shown in Figure 3Fringes are obtained by subtracting from the referencespeckle pattern recorded on the sample before starting thedrilling process the speckle pattern of the surface deformedas a consequence of the relaxed stress at each drill incre-ment Main qualitative features detectable in Figure 3 are theincrement of the number of fringes in correspondence withhigher drilled depths as a consequence of incremental stressrelaxation and the presence of an axis of symmetry parallel tothe 119909-axis which indicates the presence of an uniaxial state ofstress

Also a reference system is reported the 119909-axis is orientedalong the longitudinal direction of the specimen and the119910-axis is oriented along the transverse direction of thespecimen Figure 4 shows the difference between the stress

01234

2 22 24 26

Meas 2Meas 1

Meas 3

minus1

minus2

minus3

minus4

minus5

minus6

minus7

Δ120590xx

(MPa

)

Rext (mm)

Figure 4 Plot of the average stress measured along the drilleddepth in correspondence with different values of119877ext and by keepingconstant the value of the internal radius ratio 119877int = 159mm Thethree curves refer to three different holes in the same specimenunder the same stress conditions

measured along the drilled depth and theoretical one calcu-lated by keeping constant the radius of the internal circle ofanalysis 119877int = 159mm and by changing the value of 119877ext

Table 1 reports the theoretical average stress values andthe measured average stress values obtained by keepingconstant 119877int = 159mm and by changing 119877ext Percentageerrors are also indicated


Table 1 Values of the measured average stress obtained by keeping constant 119877int = 159mm and by changing 119877ext The expected theoreticalvalue was 120590

119909119909= 138MPa Also standard deviations are reported as a result of three different drilled holes

119877ext = 207mm 119877ext = 223mm 119877ext = 238mm 119877ext = 262mm120590119909119909

[MPa] 1353 1367 1373 13731003816100381610038161003816Δ1205901199091199091003816100381610038161003816 () 19 09 05 04

Table 2 Values of the measured average stress obtained by keeping constant 119877ext = 262 mm and by changing 119877int The expected theoreticalvalue was 120590


119877int = 111mm 119877int = 127mm 119877int = 143mm 119877int = 159mm120590119909119909

[MPa] 1082 1291 1355 13861003816100381610038161003816Δ1205901199091199091003816100381610038161003816 () 216 64 18 04

A further investigation on the analysis area was per-formed as follows the radius ratio of the external circle ofanalysis 119877ext = 262mm was kept constant and the value of119877int was changed (Figure 5) Table 2 reports the theoreticalaverage stress values and the measured average stress valuesobtained by keeping constant 119877ext = 262mm and bychanging 119877int Percentage errors are also indicated

In performing evaluation of the influence of the choiceof the internal radius of analysis it was changed from 119877int =111mm up to the maximum value 119877int = 254mm whichcorresponds to a 563 of overall variation

4 Discussion

It can be observed in Figure 4 that by changing the externalradius of analysis from 119877ext = 207mm up to the maximumvalue119877ext = 262mmwhichmeans a 21of overall variationthe system shows an almost unchanged value of measuredstress In all cases the found value is compatible with theexpected theoretical value and no substantial influence of theset external radius on the final result can be evidenced

As the analysis area expands by increasing 119877ext therelaxation effect produced by the hole is less intensive at theouter however the presented results show that for the givenmaterial under analysis and for the given level of appliedstress the ESPI system used in this experiment is able toaccurately measure the value of deformation also in the outerregion As a consequence final results appear to be insensitiveto the choice of the external radius

Differently from what observed with changing the exter-nal radius Figure 5 and Table 2 clearly show the influence ofthe internal radius on the final calculated stress value Whilethe measured stress value obtained by using 119877int = 159mmis perfectly in agreement with the theoretical expected valuea significant discordance starts to be observed by using 119877int =127mm In this situation a 64 of difference is observedwith respect to the theoretical value Situation worsens evenmore if the internal radius is further reduced down to119877int = 111mm Choosing this internal area of analysisthe calculated average stress value is strongly different from

0

10

1 12 14 16

Meas 1Meas 2Meas 3

minus10

minus20

minus30

minus40

minus50

Rint (mm)

Δ120590xx

(MPa

)

Figure 5 Plot of the average stressmeasured along the drilled depthin correspondence with different values119877int and by keeping constantthe value of the external radius ratio 119877ext = 262mm The threecurves refer to three different holes in the same specimen under thesame stress conditions

the theoretical one and the discrepancy can be evaluated tobe of the order of 216

The variation of the calculated stress with varying theinternal radius of analysis can be justified in view of twoaspects First of all it should be considered that plasticizationeffects could occur around the hole as a consequence of thedrilling process By varying the positioning of the internalradius 119877int it is possible to include at a different extent theplasticized region in the analysis area and this can affectthe final result in terms of measured stress Furthermoreit should also be taken into account that small particlesare generated in proximity of the hole during the drillingprocess A part of these flecks can remain attached to thesurface especially in the surrounding of the border of the holeThe presence of these particles acts as a noise error for theelectronic speckle pattern measurement In fact they can be


considered like a local modification of the surface topologyand they can also saturate the intensity of the correspondingimage point By reducing too much the internal radiusof analysis much of these particles can be included inthe calculation and as a consequence the accuracy of themeasurement is strongly reduced

5 Conclusions

The influence of the variation of analysis parameters forresidual stress measurement by hole drilling coupled withESPI was carried out in this work This is an important taskin view of the fact that differently from strain gage rosettewhere extensimeters are properly placed at a given distancefrom the drilled hole in ESPI the choice of the pixel thatmust be included in the analysis is left to the operator Thestudy of the influence of the area of analysis has shown thata 242 variation of the external radius of analysis does notaffect the final result This means that at least for the levelof stress considered in this material the ESPI system hasenough sensitivity to properly detect strain independentlyfrom the distance where the outer area of analysis is placedOn the contrary it was found that a 428 variation of theinternal radius of analysis can introduce errors of up to about216 on the measured stress If compared with anotheranalysis performed on the accuracy of the hole drilling + ESPIapproach this appears to be a relevant aspect In fact in thestudy of the effects of the drilling speed [24] it was foundthat it does not affect essentially the final results even if itcan result in higher dispersion at low velocity (up to 20)due to the poor quality of the drilled hole in that conditionThe study of the errors introduced by a bad determinationof the geometrical parameters [23] indicates that the mostcritical parameters that is to say the angle between theoptical axis of the camera and the normal to themeasurementsurface can introduce errors up to 5 This occurrencesuggests that attention should be paid to the proper choiceof the internal radius of analysis being a critical parameterthat can severely affect the quality of the measurement Thereason of the influence of this parameter can be trackedback mainly to two effects the plasticization that occurs nearto the hole due to both the stress concentration generatedby the drilled hole and by the drilling operation itself andalso the possible presence near the edges of the hole ofsmall metallic particles that alter the surface and that canas a consequence introduce errors in the phase calculationfrom the speckle pattern This last effect could be reducedby implementing proper measurement procedures whichinclude dust removal from the surface at the end of eachdrill increment Air compress can be used for this scopeAttention should however be paid to not moving the sampleduring this operation in order to not introduce any fictitiousdisplacementAlso care should be paid to not stain the surfacewith any impurity that could be present in the compress air Itcould be interesting as a futurework to explore the extensionof the plasticization region by performing measurementsat different drilling rotation speeds in order to change thethermal input to the specimen during the drilling process

Conflict of Interests

The authors declare that there is no conflict of interestsregarding the publication of this paper

References

[1] S A Martinez S Sathish M P Blodgett S Mall and SNamjoshi ldquoEffects of fretting fatigue on the residual stressof shot peened Ti-6Al-4V samplesrdquo Materials Science andEngineering A vol 399 no 1-2 pp 58ndash63 2005

[2] M Iordachescu J Ruiz-Hervias D Iordachescu P Vilaca andJ Planas ldquoFriction stir processing of AA6061-T4 -cold rolled vsas castrdquo in Proceedings of the 8th International Conference onTrends inWelding Research pp 90ndash95 ASM International PineMountain Ga USA 2008

[3] A Pyzalla ldquoMethods and feasibility of residual stress analysis byhigh-energy synchrotron radiation in transmission geometryusing a white beamrdquo Journal of Nondestructive Evaluation vol19 no 1 pp 21ndash32 2000

[4] L C R Schneider S V Hainsworth A C F Cocks and ME Fitzpatrick ldquoNeutron diffraction measurements of residualstress in a powder metallurgy componentrdquo Scripta Materialiavol 52 no 9 pp 917ndash921 2005

[5] M J Park H N Yang D Y Jang J S Kim and T E JinldquoResidual stress measurement on welded specimen by neutrondiffractionrdquo Journal of Materials Processing Technology vol 155-156 no 1ndash3 pp 1171ndash1177 2004

[6] X Kleber and S P Barroso ldquoInvestigation of shot-peenedaustenitic stainless steel 304L bymeans ofmagnetic Barkhausennoiserdquo Materials Science and Engineering A vol 527 no 21-22pp 6046ndash6052 2010

[7] C Casavola C Pappalettere and F Tursi ldquoCalibration ofbarkhausen noise for residual stress measurementrdquo in Con-ference Proceedings of the Society for Experimental MechanicsSeries vol 4 pp 255ndash266 2013

[8] A Karabutov A Devichenskyb A Ivochkina et al ldquoLaserultrasonic diagnostic of residual stressrdquo Ultrasonics vol 48 no6-7 pp 631ndash635 2008

[9] H Aben L Ainola and J Anton ldquoIntegrated photoelasticityfor nondestructive residual stress measurement in glassrdquoOpticsand Lasers in Engineering vol 33 no 1 pp 49ndash64 2000

[10] G de Portu L Micele Y Sekiguchi and G Pezzotti ldquoMeasure-ment of residual stress distributions in Al

2O33Y-TZPmultilay-

ered composites by fluorescence and Ramanmicroprobe piezo-spectroscopyrdquoActaMaterialia vol 53 no 5 pp 1511ndash1520 2005

[11] C Casavola L SCampanelli and C Pappalettere ldquoExperimen-tal analysis of residual stresses in the selective laser meltingprocessrdquo in Proceedings of the 11th International Congress andExhibition on Experimental and Applied Mechanics pp 1479ndash1486 Society for Experimental Mechanics Orlando Fla USAJune 2008

[12] ASTM E 8372008 Determining Residual Stresses by the hole-drilling strain-gage method

[13] C Barile C Casavola G Pappalettera and C PappalettereldquoFeasibility of local stress relaxation by laser annealing and X-ray measurementrdquo Strain vol 49 no 5 pp 393ndash398 2013

[14] P Barsanescu and P Carlescu ldquoCorrection of errors introducedby hole eccentricity in residual stress measurement by the hole-drilling strain-gage methodrdquo Measurement vol 42 no 3 pp474ndash477 2009


[15] G S Schajer and T J Rickert ldquoIncremental computation tech-nique for residual stress calculations using the integralmethodrdquoExperimental Mechanics vol 51 no 7 pp 1217ndash1222 2011

[16] A Martınez R Rodrıguez-Vera J A Rayas and H J PugaldquoFracture detection by grating moire and in-plane ESPI tech-niquesrdquo Optics and Lasers in Engineering vol 39 no 5-6 pp525ndash536 2003

[17] M Y Y Hung K W Long and J Q Wang ldquoMeasurement ofresidual stress by phase shift shearographyrdquo Optics and Lasersin Engineering vol 27 no 1 pp 61ndash73 1997

[18] C Barile C Casavola G Pappalettera and C PappalettereldquoMechanical characterization of SLM specimens with speckleinterferometry and numerical optimizationrdquo in Experimentaland Applied Mechanics vol 6 of Conference Proceedings ofthe Society for Experimental Mechanics Series pp 837ndash843Springer New York NY USA

[19] C Barile C Casavola G Pappalettera and C PappalettereldquoExperimental and numerical characterization of sinterizedmaterials with speckle interferometry and optimization meth-odsrdquo in Proceedings of the 10th Youth Symposium on Experimen-tal SolidMechanics (YSESM rsquo11) pp 35ndash36 Chemnitz GermanyMay 2011

[20] C Barile C Casavola G Pappalettera and C PappalettereldquoHybrid characterization of laminated wood with ESPI andoptimization methodsrdquo in Proceedings of the Society for Experi-mental Mechanics Series Conference vol 3 pp 75ndash83 2013

[21] C Barile C Casavola G Pappalettera and C PappalettereldquoResidual stress measurement by electronic speckle patterninterferometry a study of the influence of analysis parametersrdquoStructural Integrity and Life Integritet i vek Konstrukcija vol 12no 3 pp 159ndash163 2012

[22] C Barile C Casavola G Pappalettera and C PappalettereldquoAnalysis of the effects of process parameters in residual stressmeasurements on Titanium plates by HDMESPIrdquo Measure-ment Journal of the International Measurement Confederationvol 48 no 1 pp 220ndash227 2014

[23] C Barile C Casavola G Pappalettera and C PappalettereldquoResidual stress measurement by electronic speckle patterninterferometry a study of the influence of geometrical parame-tersrdquo Structural Integrity and Life vol 11 no 3 pp 177ndash182 2010

[24] C Barile C Casavola G Pappalettera C Pappalettere andF Tursi ldquoDrilling speed effects on accuracy of HD residualstress measurementsrdquo in Residual Stress Thermomechanics ampInfrared Imaging Hybrid Techniques and Inverse Problems vol8 of Conference Proceedings of the Society for ExperimentalMechanics Series pp 119ndash125 Springer 2014

[25] C Barile C Casavola G Pappalettera and C PappalettereldquoMechanical characterization of SLM specimens with speckleinterferometry and numerical optimizationrdquo in Proceedings ofthe Annual Conference and Exposition on Experimental andApplied Mechanics vol 3 pp 2523ndash2529 Society for Experi-mental Mechanics Indianapolis Ind USA June 2010

[26] G Focht and K Schiffner ldquoDetermination of residual stressesby an optical correlative hole-drilling methodrdquo ExperimentalMechanics vol 43 no 1 pp 97ndash104 2003

[27] A Baldi ldquoFull field methods and residual stress analysis inorthotropic material I linear approachrdquo International Journalof Solids and Structures vol 44 no 25-26 pp 8229ndash8243 2007

[28] F V Dıaz G H Kaufmann and O Moller ldquoResidual stressdetermination using blind-hole drilling and digital specklepattern interferometry with automated data processingrdquo Exper-imental Mechanics vol 41 no 4 pp 319ndash323 2001

[29] M R Viotti and G H Kaufmann ldquoAccuracy and sensitivityof a hole drilling and digital speckle pattern interferometrycombined technique to measure residual stressesrdquo Optics andLasers in Engineering vol 41 no 2 pp 297ndash305 2004

[30] M R Viotti A E Dolinko G E Galizzi and G H KaufmannldquoA portable digital speckle pattern interferometry device tomeasure residual stresses using the hole drilling techniquerdquoOptics and Lasers in Engineering vol 44 no 10 pp 1052ndash10662006

[31] G S Schajer and M Steinzig ldquoFull-field calculation of holedrilling residual stresses from electronic speckle pattern inter-ferometry datardquoExperimentalMechanics vol 45 no 6 pp 526ndash532 2005

[32] A E Dolinko and G H Kaufmann ldquoA least-squares methodto cancel rigid body displacements in a hole drilling and DSPIsystem for measuring residual stressesrdquo Optics and Lasers inEngineering vol 44 no 12 pp 1336ndash1347 2006

[33] M Steinzig and E Ponslet ldquoResidual stress measurement usingthe hole drilling method and laser speckle interferometry partIrdquo Experimental Techniques vol 27 no 3 pp 43ndash46 2003

[34] C Casavola G Pappalettera C Pappalettere and F TursildquoEffects of strainrsquos error on residual stresses calculated byHDMrdquoin Proceedings of the Society for Experimental Mechanics SeriesConference vol 4 pp 395ndash402 2013

[35] C Casavola G Pappalettera C Pappalettere and F TursildquoAnalysis of the effects of strainmeasurement errors on residualstresses measured by incremental hole-drilling methodrdquo Jour-nal of Strain Analysis for Engineering Design vol 48 no 5 pp313ndash320 2013

International Journal of

AerospaceEngineeringHindawi Publishing Corporationhttpwwwhindawicom Volume 2014

RoboticsJournal of

Hindawi Publishing Corporationhttpwwwhindawicom Volume 2014


Active and Passive Electronic Components

Control Scienceand Engineering

Journal of



RotatingMachinery


Hindawi Publishing Corporation httpwwwhindawicom

Journal ofEngineeringVolume 2014

Submit your manuscripts athttpwwwhindawicom

VLSI Design



Shock and Vibration


Civil EngineeringAdvances in

Acoustics and VibrationAdvances in



Electrical and Computer Engineering

Journal of

Advances inOptoElectronics


Volume 2014

The Scientific World JournalHindawi Publishing Corporation httpwwwhindawicom Volume 2014

SensorsJournal of


Modelling amp Simulation in EngineeringHindawi Publishing Corporation httpwwwhindawicom Volume 2014


Chemical EngineeringInternational Journal of Antennas and

Propagation




Navigation and Observation



DistributedSensor Networks



in order to attach the rosette and also the positioning ofthe drill bit must be accurately performed in order to avoideccentricity errors [14] Furthermore only a few data pointscan be obtained by strain gage and this can introducesignificant errors in the stress calculation procedure Finallyit should be considered that the strain gage rosette hasa cost which cannot be considered negligible particularlyin the case where a relevant number of measurements arerequired Optical methods based on laser interferometry arepromising in replacing strain gage in view of the fact thatthey can allow obtaining high sensitivity measurements froma huge number of data points without contact and allowingaccurately defining on the recorded image the position ofthe drilled hole [15] Moire interferometry [16] phase shiftshearography [17] and electronic laser speckle interferometry(ESPI) were successfully adopted in many different situationsto measure displacement fields ESPI was used to measuredisplacement field in anisotropic specimen made by selectivelaser melting [18] sinterization [19] or laminated wood [20]Now ESPI is under study for the specific application inmeasuring residual stresses [21] Specific studies were carriedout in order to establish the effects of different parametersinvolved in the residual stress measurement process on thefinal accuracy In [22] effects of several process parameters inRS measurement in titanium plates were investigated whilethe effects of errors in the determination of the geometricalparameters characterizing the optical system are discussed in[23] Finally the effects connected to a proper choice of thedrilling speed are reported in [24]

In this paper the attention is focused on the phase shiftingelectronic speckle pattern interferometry used in combi-nation with hole-drilling technique for measuring residualstress profileThis method being an interferometric methodcan guarantee sensitivity of the order of a fraction of theillumination wavelength it does not require any particularpreparation of the surface being applicable also to roughand curve surfaces and it does not require application of agrating as in moire interferometry [25] The main drawbackof this technique is the necessity to properly isolate the systemfrom vibrations in order to obtain fringe pattern with goodcontrast

Several works are present in the literature where ESPIhas been used to measure strain and calculate residual stressFocht and Schiffner used an ESPI set-up to measure strainin thin metal sheets [26] a double beam ESPI interferometerwas used by Baldi [27] to evaluate stresses in orthotropicmaterials Dıaz et al measured by ESPI the stress relieved byhole drilling in an aluminum plate subjected to a uniformuniaxial tensile stress [28] Viotti and Kaufmann evalu-ated the accuracy and the sensitivity of this hole-drillingand digital speckle pattern interferometry (DSPI) combinedtechnique on aluminum thin plates subjected to a uniformuniaxial tensile stress [29] and described a portable deviceto measure residual stresses [30] a mathematical methodwas proposed by Schajer and Steinzig for calculating residualstresses from hole-drilling electronic speckle pattern inter-ferometry (ESPI) data independent of rigid-body motions[31] a method to cancel rigid body displacements that can beintroduced when a hole-drilling and digital speckle pattern

interferometry (DSPI) combined system is used to measureresidual stresses was proposed by Dolinko and Kaufmannbased on a least-square approach [32]

In this paper some experimental considerations abouterror sources in the ESPI hole-drilling method will bediscussed In particular the influence of the area of analysison the final results in terms of measured residual stresses willbe discussed The inner and the outer radius of the area ofanalysis will be changed and the corresponding variations inthe measured residual stress profile will be calculated

2 Materials and Methods

21 Experimental Set-Up for HDM+ESPI Measurements TheESPI hole-drilling measurement system used in this work isschematically reported in Figure 1 The laser source used inthe proposed set-up was a diode-pumped solid-state (DPSS)laser source with a wavelength 120582 = 532 nm Radiationwas split into two beams and focused on two monomodeoptical fibersOne beam is collimated through a biconvex lensand illuminates the sample while the second beam passesthrough a phase shifting piezoelectric system and then goesto a 640 times 480 pixel charge-coupled device (CCD) camerawhere interferes with the light diffused by the optically roughsurface of the specimen Camera is equipped with an opticalimaging system allowing fine focusing of the image Theintensity recorded by the camera at each (119894 119895) pixel can bedescribed by

119868 (119894 119895) = 119868ref (119894 119895) + 119868obj (119894 119895) + 2radic119868ref119868obj cos (120593 (119894 119895)) (1)

where 119868ref(119894 119895) is the intensity of the reference beam 119868obj(119894 119895)is the intensity of the object beam deformations and 120593(119894 119895) isthe initial phase difference

If the surface of the specimen undergoes some displace-ment along direction of the sensitivity vector describing thesystem the speckle pattern experiences a phase differenceΔ120593(119894 119895) and the new intensity 119868(119894 119895) recorded by the CCDcamera at each (119894 119895) pixel can be described by

119868 (119894 119895) = 119868ref (119894 119895) + 119868obj (119894 119895)

+ 2radic119868ref119868obj cos (120593 (119894 119895) + Δ120593 (119894 119895)) (2)

where Δ120593(119894 119895) is the additional phase difference connectedwith the displacement 119889 experienced by the surface

In order to extract the value of the additional phase thefour-step phase shifting algorithm was implemented [33] Inorder to do this in correspondence with the undeformedsample and for each stage of deformation of the specimen fourimages were recorded A given image differs from the nextone in view of the fact that piezoactuator is moved so that


Sample

Laser source+


Drilling system

controller

Illumination beam

CCD

Reference beam


Drilling system



119868119894(119894 119895) = 119868ref (119894 119895) + 119868obj (119894 119895)

+ 2radic119868ref119868obj cos(120593 (119894 119895) + 119899120587

2)

119868119894(119894 119895) = 119868ref (119894 119895) + 119868obj (119894 119895)


2)

(3)




(4)


Δ120593 = sdot 119889 (5)









y

x

01mmy

x

02mm

y

x

03mm y

x

04mm





3 Results



01234

2 22 24 26

Meas 2Meas 1

Meas 3

minus1

minus2

minus3

minus4

minus5

minus6

minus7

Δ120590xx

(MPa

)

Rext (mm)








[MPa] 1353 1367 1373 13731003816100381610038161003816Δ1205901199091199091003816100381610038161003816 () 19 09 05 04




[MPa] 1082 1291 1355 13861003816100381610038161003816Δ1205901199091199091003816100381610038161003816 () 216 64 18 04



4 Discussion




0

10

1 12 14 16

Meas 1Meas 2Meas 3

minus10

minus20

minus30

minus40

minus50

Rint (mm)

Δ120590xx

(MPa

)






5 Conclusions




References











2O33Y-TZPmultilay-






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










Sample

Laser source+


Drilling system

controller

Illumination beam

CCD

Reference beam


Drilling system



119868119894(119894 119895) = 119868ref (119894 119895) + 119868obj (119894 119895)

+ 2radic119868ref119868obj cos(120593 (119894 119895) + 119899120587

2)

119868119894(119894 119895) = 119868ref (119894 119895) + 119868obj (119894 119895)


2)

(3)




(4)


Δ120593 = sdot 119889 (5)









y

x

01mmy

x

02mm

y

x

03mm y

x

04mm





3 Results



01234

2 22 24 26

Meas 2Meas 1

Meas 3

minus1

minus2

minus3

minus4

minus5

minus6

minus7

Δ120590xx

(MPa

)

Rext (mm)








[MPa] 1353 1367 1373 13731003816100381610038161003816Δ1205901199091199091003816100381610038161003816 () 19 09 05 04




[MPa] 1082 1291 1355 13861003816100381610038161003816Δ1205901199091199091003816100381610038161003816 () 216 64 18 04



4 Discussion




0

10

1 12 14 16

Meas 1Meas 2Meas 3

minus10

minus20

minus30

minus40

minus50

Rint (mm)

Δ120590xx

(MPa

)






5 Conclusions




References











2O33Y-TZPmultilay-






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation










y

x

01mmy

x

02mm

y

x

03mm y

x

04mm





3 Results



01234

2 22 24 26

Meas 2Meas 1

Meas 3

minus1

minus2

minus3

minus4

minus5

minus6

minus7

Δ120590xx

(MPa

)

Rext (mm)








[MPa] 1353 1367 1373 13731003816100381610038161003816Δ1205901199091199091003816100381610038161003816 () 19 09 05 04




[MPa] 1082 1291 1355 13861003816100381610038161003816Δ1205901199091199091003816100381610038161003816 () 216 64 18 04



4 Discussion




0

10

1 12 14 16

Meas 1Meas 2Meas 3

minus10

minus20

minus30

minus40

minus50

Rint (mm)

Δ120590xx

(MPa

)






5 Conclusions




References











2O33Y-TZPmultilay-






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation













[MPa] 1353 1367 1373 13731003816100381610038161003816Δ1205901199091199091003816100381610038161003816 () 19 09 05 04




[MPa] 1082 1291 1355 13861003816100381610038161003816Δ1205901199091199091003816100381610038161003816 () 216 64 18 04



4 Discussion




0

10

1 12 14 16

Meas 1Meas 2Meas 3

minus10

minus20

minus30

minus40

minus50

Rint (mm)

Δ120590xx

(MPa

)






5 Conclusions




References











2O33Y-TZPmultilay-






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation











5 Conclusions




References











2O33Y-TZPmultilay-






























RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation

































RoboticsJournal of





Journal of



RotatingMachinery





VLSI Design



Shock and Vibration







Journal of



Volume 2014


SensorsJournal of





Propagation









research article remarks on residual stress measurement …

Documents