Open Research OnlineThe Open University’s repository of research publicationsand other research outputs

Measurement of residual stresses in dissimilar frictionstir-welded aluminium and copper plates using thecontour methodJournal ItemHow to cite:

Zhang, Chengcong and Shirzadi, Amir A. (2018). Measurement of residual stresses in dissimilar friction stir-welded aluminium and copper plates using the contour method. Science and Technology of Welding and Joining,23(5) pp. 394–399.

For guidance on citations see FAQs.

c© [not recorded]

Version: Accepted Manuscript

Link(s) to article on publisher’s website:http://dx.doi.org/doi:10.1080/13621718.2017.1402846

Copyright and Moral Rights for the articles on this site are retained by the individual authors and/or other copyrightowners. For more information on Open Research Online’s data policy on reuse of materials please consult the policiespage.

oro.open.ac.uk

Measurementofresidualstressesindissimilarfrictionstirweldedaluminiumandcopperplatesusingthecontourmethod

ChengcongZhang1,AmirA.Shirzadi2

1.TechnicalCentre,ShanghaiAerospaceEquipmentManufacturer,Shanghai,200245,China

2.SchoolofEngineering&Innovation,TheOpenUniversity,MiltonKeynes,MK76AA,UK

AbstractThe longitudinal residual stresses in the friction stir welded plates of 5A06 aluminium and purecopper were determined using the contour method. The results revealed the presence of hightensile and compressive residual stresses on the aluminium and copper sides, respectively. Theresidualstressesweredetectedontheweldzoneaswellasthethermo-mechanicallyaffectedzone(TMAZ)of the aluminiumplate. In contrast, the compressive residual stresses in the copper platehadamuchnarrowerwidthalongtheweldline.Peaktensilestressesupto240MPawerefoundintheTMAZofthealuminiumplate.

Keywords:FrictionStirWelding,ContourMethod,ResidualStress,AluminiumCopperJoints.

Introduction

Combinationoftheexcellentthermalandelectricalpropertiesofpurecopperwiththelightweightandlowcostofaluminiumalloyshasalwaysbeenofgreatinterestinpowergeneration,aerospace,automotive and electronics industries [1]. In order to exploit the combined potential of thesematerialsareliableandviableweldingmethod isneeded. Fusionweldingofaluminiumtocopperhasprovedunsuccessfulmainlyduetothedifferencesintheirphysicalandchemicalproperties.Forinstance, unavoidable formation of hard and brittle intermetallic compounds (IMCs) at the weldzoneresultsinlowstrengthandverylowimpactresistantjoints[2,3,4].

Frictionstirwelding(FSW),generallyregardedasasolid-stateprocess,isoneofthemostpromisingsolutions for joining dissimilar alloys e.g. aluminium to copper. Most of previous academic andindustrial investigations were aimed at optimising the FSW process parameters, improvingmicrostructures,reducingdefectsandenhancingmechanicalpropertiesoffrictionstirweldedAlandCuplates.Previousworkclearlyprovedthattheprocessparameterssuchasrelativepositionofbasematerials,tooloffset[5,6],rotational,traversespeeds[7,8,9]andtiltangleoftherotatingtool[10]as well as the material properties [11] all can affect the macrostructure, microstructure andmechanicalpropertiesoftheAl-Cujoints.ThemechanicalmixingofAlandCuduringFSWresultsincomposite-likestructureswhichconsistofcopper-richparticlesandintermetallicsdistributedwithinthe aluminium-based matrix [4, 12, 13, 14,15]. The lamellar structures seen in the friction stirwelded Al and Cu consist of different phases with a swirl pattern [1,5,6,13,16]. Meanwhile, theunmixedzoneofthejointconsistsofcontinuousIMCswithathicknessof0.2to3μm.TheIMCsattheAl-CuinterfacehavebeenidentifiedasCuAl2,CuAlandCu9Al4[7,8,17,18].

Itisreportedthatthermo-mechanicallyaffectedzones(TMAZ)aretheweakestpartsofmostfrictionstir welded Al-Cu components, where most of the failures occur due to the formation of largeamounts of IMCS [8,10,17]. Besides, inadequate interaction between the rotating tool and basematerialcanresultintheformationofveryundesirabledefectssuchasdiscontinuitiesandpores.

ThehardnessatthenuggetzoneofadissimilarAl-CuFSWisnormallyhigherthanthoseofthebasematerialsagainduetotheformationofIMCsandseverestrain-hardening.ItisalsoreportedthatthehighertheheatinputthehighertheIMCcontent,hencethehigherthehardness[11,13&19].

Residual stresses can drastically compromise the structural integrity, performance and lifetime ofwelded components. Friction stir welding is associated with a substantial amount of plasticdeformation in thematerials being joined; hence very high residual stresses are expected in the

weldzonewhichcontainsseverelydeformeddissimilarmaterialswithverydifferentYoung’smoduli.ReportsontheresidualstressesdevelopedinsimilaranddissimilarFSWjointsarescantandmostofpreviousworkconcernsAl-to-Aljoints.Regardlessofthetypeofalloy,thelongitudinalstresseswereofmaininterestinthisandpreviousworksincetheyaresubstantiallyhigherthanthosefoundinthetraversedirection[20&21].

The longitudinal residual stresses normally have an M-shape profile with peak values in theHAZ/TMAZ[20,21,22,23].Howevertherearesomecontradictoryreports;e.g.Liuetal.showedthattheresidualstressesprofileinthe8mmand4mmthickFSWAA6061-T6platesdidnothaveanM-shape profile [24]. They alsomeasured up to 168MPa residual stresses close to the edge of therotatingtoolshoulderontheadvancingside.Suchahighlevelofresidualstressisabout60%oftheyield strength of parent AA6061-T6 alloy at the room temperature, hence can substantiallycompromisethejointperformanceundertheservicecondition.Carloneetal.measuredthetensileresidual stresses in theweld line of a friction stirweldedAA2024-T3 butt joint using the contourmethod[25].Thetensilestressesontheweldlinewerebalancedbysomecompressivestressesonthebothsidesoftheweld.Twopeaktensilestressesweredetectedintheadvancingandretreatingsidesatadistanceequaltotheradiusoftheshoulderfromthecentreline.Themaximumresidualstresson theadvancingsidewas slightlyhigher thanon the retreating side, i.e.145and125MParespectively. In contrast, Primeet al. reported very low residual stresses indissimilar friction stirwelds between 25.4mm thick aluminium plates of 7050-T7451 and 2024-T351 [26]. Their resultsconfirmed the presence of residual stress with anM-shape profile and peak stresses of 43MPawhicharelessthan20%ofthebasematerialyieldstrength.

Todate,noreportwasfoundonmeasuringoranalysingtheresidualstressesinfrictionsstirweldedaluminiumandcopperplates.Inthiswork,theresidualstressesgeneratedinfrictionstirweldedAlandCuplatesweremeasuredbythecontourmethodandtheoutcomewasusedtomapthestressdistributiononaplanenormaltotheweldline,i.e.crosswelddirection.

Experimental

All experiments were performed on the thin sheets of 5A06 aluminium and T2M pure copper(99.9%) both with approximate dimensions of 300 mm×100 mm×2 mm using a standard FSWequipment at ShanghaiAerospace EquipmentManufacturer (SAEM). The chemical compositionof5A06aluminiumisgiveninTable1.

Table1.Chemicalcompositionof5A06aluminiumusedinthiswork(wt.%)

Mg Mn Ti Al5.8-6.8 0.5-0.8 0.02-0.10 Balance

TherotatingtoolwasmadeofW360steelwitha15mmdiameterconcaveshoulderanda1.9mmlong / 5mm diameter cylindrical probe. The probe was inserted in the aluminium plate with anoffset of 0.5mm. The aluminium alloywas positioned on the advancing side of the rotating toolleaving thecopperon the retreatingside.Rotational speedof1000 rpmand traversespeedof40mm/minwereused inallweldingtrials.Theplungeddepthofthetoolshoulderwas0.15mmandthe pinwas tilted 2.5° throughout thewelding cycle. Fig. 1 shows thewelding setup used in thiswork.

Fig.1:Set-upusedforfrictionstirweldingofaluminiumandcopperplates.Bondedsamplewascutacrossthebond-lineusingwireEDMtoproducethemeasurementplanesneededforresidualstressmeasurementbythe

contourmethod.

Contourmethod,whichwas invented in 2001, is an attractive technique for determining residualstresses intheengineeringcomponents[27].Themethodreliesonacquisitionandprocessingofaverylargenumberofdata.Thecontoursampleswerepreparedbycuttingeachpairoftheweldedplatesinthemiddleandnormaltotheweldingdirection(seeFig.1).Thecuttingofcontoursamplesmust be carried out using Wire Electrical Discharge Machining (EDM) to prevent any plasticdeformation [28]. The specimens in this work were cut with an EDM machine using a 0.3 mmdiameterbrass-basedwireataconstantspeedof1.6mm/min.Thespecimensweresubmergedinthe temperature-controlled deionisedwater during the cutting process. All cuts started on the Alside of the specimen and it took about 120minutes to finish each cut. Fig. 2 shows one of thesamplesintheclampedpositionandafterEDMcutacrossitsbond-line.

Fig.2:FrictionsstirweldedAlandCuplateswerecutacrossthebond-linebywireEDMmethod.Arrowshows

thecuttingdirection.

The twohalvesofeach specimenwerekeptata temperature-controlled room forat least1hourbefore measuring their surface profile. The surface profiles were measured using a coordinatemeasuring machine (CMM) equipped with a laser scanner. It took 70 minutes to collect about625,000datapointsfromeachscannedsurfacewithaspacingof0.025mminbothX&Ydirections.TheactualscanningsetupisshowninFig.3.

Fig.3:LaserscannerusedforsurfaceprofilingoffrictionstirweldedAlandCusamples.

Thedisplacementdata(surfaceprofile),collectedfromtwohalvesofeachsample,werealignedandaveragedpointbypointtoremovetheeffectsofshearstressesandcuttingartefact.Thenthedatawere“cleaned”byremovingthenoisesandoutliers,whichotherwisecouldresultinerrors.Thedatasmoothingwascarriedoutusingthecubicsplinefittingmethod,whichisacommonpracticeinthecontourmethod.Thesurfaceprofilesofasamplebeforeandaftersmoothingareshown inFig.4.ThesmootheneddatawereusedastheboundaryconditionsinthesubsequentFiniteElement(FE)analysis.

Fig.4:SurfaceprofileofanAl-Cusamplebefore(top)andaftersmoothing(bottom).

A3-DlinearelasticFEmodelofhalfspecimenwascreated.TheFEmodelofatestspecimenisshowninFig.5inwhichthemeasureddisplacementdatawereusedastheboundaryconditionsinordertocalculatetheresidualstressesnormaltothecutsurface[28,29].

Fig.5:FiniteelementmodelofthesurfacecontouroffrictionstirweldedAl-Cuplates(magnifiedbyfactorof

300).

Young’smodulusoftheweldregionneedstobedeterminedaspreciselyaspossible,sinceithasadirectinfluenceontheaccuracyofcalculatedresidualstresses.Fig.6showsthemicrostructureofanAl-Cujointwitharelativelyclearborderbetweenthetwomaterials.Theclearborderallowedusingthe Young’s modulus of parent aluminium and copper on the corresponding areas. The contourmethodislimitedto2Dmeasurementofresidualstresses,i.e.thestressesnormaltothecutsurface.However, 3Dmeasurements of residual stresses using the contourmethod should be possible bysuccessivelyremovingthinlayersandrepeatingtheentireprocess i.e.depthprofiling. Admittedly,this assumption that theweld zone consists of two distinctive areas is far from reality andmoreappropriateapproachesneedtobedevelopedinthefuturework.

Fig.6:MicrostructureoffrictionstirweldedAl-Cuplatesinlow(top)andhigh(bottom)magnifications.

FortheFEanalysis,theYoung’smodulusof67&117GPaandPoisson’sratiosof0.33&0.36wereused for 5A06 aluminium and T2M pure copper, respectively. The elastic properties of bothmaterialswereassumedtobeisotropic.

Resultsanddiscussion

The residual stressesdistributionmap,generatedby thecontourmethod, is shown inFig.7. It isevident that themagnitude and distribution of residual stresses are significantly different on thealuminium and copper sides. The longitudinal stresses (acting normal to the cut surface) on thealuminium and copper sides are tensile and compressive, respectively. Also, the width of thestressed zone in Al is farwider than that in the copper. This is due to aluminiumhaving a lowerYoung’smodulusthancopper.Thetensilestressesonthealuminiumsideareextendedacrosstheentireweld regionandTMAZ,while thecompressive stresses in copperare locatednear the jointinterface.Itwasinterestingtoseethattheregionwithresidualtensilestress(Alside)iswiderthanthetoolshoulder.ThepeaktensilestresswasfoundneartheTMAZonthealuminiumside,whichwastheadvancingsideoftherotatingtool.Thelocationofthepeaktensilestressobservedinthiswork is consistent with those of the previous work conducted on similar/dissimilar aluminiumfriction stir welds [26, 30, 31,32]. Themaximum compressive stress on the copper sidewas veryclosetotheweldcentreline.

Very high amounts of tensile stresseswere observed in the aluminium TMAZ aswell as the jointinterface.Theresidualstressesindissimilarjointsareduetohavingdifferentcoefficientofthermalexpansions. When a dissimilar weldment cools down, the material with a higher coefficient ofthermalexpansionshrinksfasterresultinginthebuildupoftensileresidualstressesandviceversafortheothermaterial[33].Thelinearcoefficientofthermalexpansion(CTE)ofaluminiumisabout30%largerthanthatofcopper,i.e.23.2×10-6K-1and17.5×10-6K-1,respectively.Thisisconsistentwiththeresultsofcontourmeasurement,whichrevealedthepresenceoftensilestressesonaluminiumsideandcompressivestressesoncopperside,asshowninFig.7.

Fromadifferentperspective, thethermalconductivityofcopper ishigher thanthatofaluminium,i.e. 401W·m−1·K−1 and237W·m−1·K−1, respectively. Therefore, theheat generatedduringweldingprocess dissipates faster in copper than in aluminium. In short, copper is a stiffer metal thanaluminium and tends to shrink less than aluminium due to having lower CTE and higher thermalconductivity.Thesedifferencesinthephysicalandmechanicalpropertiesofcopperandaluminiumledtohavingamuchnarrowerstressedzoneonthecoppersidenearthejointinterface-seeFig.7.

Fig.7:MapoflongitudinalresidualstressesonfrictionstirweldedAl-Cuplatesgeneratedbythecontour

method.

Fig.8showstheoutcomeoftwoline-scans,whichwerecarriedoutonthecontourmapatabout0.5mmfromthetopandbottomsurfaces.ThepeaktensilestressesinbothtopandbottomsurfacesoftheAlareabout240MPa.Incontrast,themaximumcompressivestressesinthecopperare350and270 MPa close to the top and bottom surfaces, respectively. Although the exact source of suchvariation remains to be investigated, it is reasonable to “speculate” that the soft Al has alreadyreached itsmaximumcontainable levelof residual stressonboth surfaces.Theyieldandultimatestrengthsof5A06aluminiumare171and345MPa,respectively;whilethoseofT2Mcopperare62and227MPa.Therefore,designengineerswouldfindhavingsuchhightensileresidualstressesquitealarming. Optimization of FSW process parameters can be considered to mitigate the residualstresses.

Extractionofnarrowcouponsmayaffecttheresidualstressinthecentreofaweldment.However,previousworkshowedthatthemaximumremainingstressatthecentreofanarrowcouponisabout95%of thatofan infinitelywidesample if thecoupon is twicewiderthanthewidthof thestress-affected zone [34]. In thiswork, the sample couponswere5 timeswider than the stress affectedzone,i.e.20mmwavelengthofresidualstressshowninFig.8comparedto100mmwidecoupons.Therefore, any relaxation in the residual stress, due to using a 100 mmwide coupon, would benegligiblecomparedtootherexperimentalerrors.

Fig.8:Residualstressesmeasuredbylinescanning0.5mmbeneaththetopandbottomsurfaces.

Conclusions

Longitudinalresidualstressesinthefrictionstirweldedaluminiumandcopperplateswereanalysedbythecontourmethod.Theoutcomesofthisworkareasfollows.

(1) The longitudinal stresses in the aluminium and copper plates proved to be tensile andcompressive,respectively.

(2) The tensile stresseson thealuminiumside is extended throughout theentireweld regionandTMAZ,whereasthecompressivestressesinthecopperplatehadmuchnarrowerwidthalongthejointinterface.

(3) Themaximumtensilestressof240MPawasdetectedneartheTMAZinthealuminiumplate.(4) TheresidualstressesinfrictionstirweldedAl-Cuplatesweretoohightoachievereliableand

structurallysoundbonds.

Acknowledgments

Oneof the authors (C.Z.) is grateful for the financial support received from theChina ScholarshipCouncil and The Open University, UK, for hosting and providing residual stress measurementfacilities. Authors would like to thank Peter Ledgard for conducting the EDM cuts; Dr JefersonOliveiraforconductingtheprofilemeasurementandDrSanjooramPaddeaforhishelpwiththeFEanalysis.

References

[1] I.Galvão,A.Loureiro,andD.M.Rodrigues,“Criticalreviewonfrictionstirweldingofaluminiumtocopper,”ScienceandTechnologyofWeldingandJoining,vol.1718,no.April.pp.1–24,2016.

[2] K.P.MehtaandV.J.Badheka,“Areviewondissimilarfrictionstirweldingofcoppertoaluminum:Process,properties,andvariants,”MaterialsandManufacturingProcesses,vol.31,no.3.pp.233–254,2016.

[3] H.J.Liuetal.,“Weldappearanceandmicrostructuralcharacteristicsoffrictionstirbuttbarrierweldedjointsofaluminiumalloytocopper,”ScienceandTechnologyofWeldingandJoining,vol.17,no.2.pp.104–110,2012.

[4] P.Xue,B.L.Xiao,D.Wang,andZ.Y.Ma,“Achievinghighpropertyfrictionstirweldedaluminium/copperlapjointatlowheatinput,”ScienceandTechnologyofWeldingandJoining,vol.16,no.8.pp.657–661,2011.

[5] I.Galvão,A.Loureiro,D.Verdera,D.Gesto,andD.M.Rodrigues,“Influenceoftooloffsettingonthestructureandmorphologyofdissimilaraluminumtocopperfriction-stirwelds,”MetallurgicalandMaterialsTransactionsA:PhysicalMetallurgyandMaterialsScience,vol.43,no.13.pp.5096–5105,2012.

[6] P.Xue,D.R.R.Ni,D.Wang,B.L.L.Xiao,andZ.Y.Y.Ma,“EffectoffrictionstirweldingparametersonthemicrostructureandmechanicalpropertiesofthedissimilarAl–Cujoints,”MaterialsScienceandEngineering:A,vol.528,no.13–14.pp.4683–4689,2011.

[7] M.F.X.MuthuandV.Jayabalan,“Tooltravelspeedeffectsonthemicrostructureoffrictionstirweldedaluminum-copperjoints,”JournalofMaterialsProcessingTechnology,vol.217.pp.105–113,2015.

[8] H.J.Liu,J.J.Shen,L.Zhou,Y.Q.Zhao,C.Liu,andL.Y.Kuang,“Microstructuralcharacterisationandmechanicalpropertiesoffrictionstirweldedjointsofaluminiumalloytocopper,”ScienceandTechnologyofWeldingandJoining,vol.16,no.1.pp.92–98,2011.

[9] H.Bisadi,A.Tavakoli,M.TourSangsaraki,andK.TourSangsaraki,“TheinfluencesofrotationalandweldingspeedsonmicrostructuresandmechanicalpropertiesoffrictionstirweldedAl5083andcommerciallypurecoppersheetslapjoints,”MaterialsandDesign,vol.43.pp.80–88,2013.

[10] K.P.MehtaandV.J.Badheka,“Effectsoftiltangleonthepropertiesofdissimilarfrictionstirweldingcoppertoaluminum,”MaterialsandManufacturingProcesses,vol.31,no.3.pp.255–263,2016.

[11] I.Galvão,D.Verdera,D.Gesto,A.Loureiro,andD.M.Rodrigues,“Influenceofaluminiumalloytypeondissimilarfrictionstirlapweldingofaluminiumtocopper,”JournalofMaterialsProcessingTechnology,vol.213,no.11.pp.1920–1928,2013.

[12] P.Xue,D.R.Ni,D.Wang,B.L.Xiao,andZ.Y.Ma,“EffectoffrictionstirweldingparametersonthemicrostructureandmechanicalpropertiesofthedissimilarAl-Cujoints,”MaterialsScienceandEngineeringA,vol.528,no.13–14.pp.4683–4689,2011.

[13] C.W.Tan,Z.G.Jiang,L.Q.Li,Y.B.Chen,andX.Y.Chen,“MicrostructuralevolutionandmechanicalpropertiesofdissimilarAl-Cujointsproducedbyfrictionstirwelding,”MaterialsandDesign,vol.51.pp.466–473,2013.

[14] H.R.ZareieRajani,A.Esmaeili,M.Mohammadi,M.Sharbati,andM.K.B.Givi,“Theroleofmetal-matrixcompositedevelopmentduringfrictionstirweldingofaluminumtobrassinweldcharacteristics,”JournalofMaterialsEngineeringandPerformance,vol.21,no.11.pp.2429–2437,2012.

[15] A.Esmaeili,H.R.ZareieRajani,M.Sharbati,M.K.B.Givi,andM.Shamanian,“Theroleofrotationspeedonintermetalliccompoundsformationandmechanicalbehavioroffrictionstirweldedbrass/aluminum1050couple,”Intermetallics,vol.19,no.11.pp.1711–1719,2011.

[16] J.Ouyang,E.Yarrapareddy,andR.Kovacevic,“Microstructuralevolutioninthefrictionstirwelded6061aluminumalloy(T6-tempercondition)tocopper,”JournalofMaterialsProcessingTechnology,vol.172,no.1.pp.110–122,2006.

[17] I.Galvão,J.C.Oliveira,A.Loureiro,andD.M.Rodrigues,“Formationanddistributionofbrittlestructuresinfrictionstirweldingofaluminiumandcopper:Influenceofshouldergeometry,”Intermetallics,vol.22.pp.122–128,2012.

[18] M.N.Avettand-Fenoël,R.Taillard,G.Ji,andD.Goran,“Multiscalestudyofinterfacial

intermetalliccompoundsinadissimilarAl6082-T6/Cufriction-stirweld,”MetallurgicalandMaterialsTransactionsA:PhysicalMetallurgyandMaterialsScience,vol.43,no.12.pp.4655–4666,2012.

[19] R.Beygi,M.Kazeminezhad,andA.H.Kokabi,“ButtjoiningofAl-Cubilayersheetthroughfrictionstirwelding,”TransactionsofNonferrousMetalsSocietyofChina(EnglishEdition),vol.22,no.12.pp.2925–2929,2012.

[20] A.Steuwer,M.J.Peel,andP.J.Withers,“DissimilarfrictionstirweldsinAA5083–AA6082:Theeffectofprocessparametersonresidualstress,”MaterialsScienceandEngineering:A,vol.441,no.1–2.pp.187–196,2006.

[21] H.JamshidiAval,S.Serajzadeh,andA.H.Kokabi,“ExperimentalandtheoreticalevaluationsofthermalhistoriesandresidualstressesindissimilarfrictionstirweldingofAA5086-AA6061,”InternationalJournalofAdvancedManufacturingTechnology,vol.61,no.1–4.pp.149–160,2012.

[22] H.Lombard,D.G.Hattingh,A.Steuwer,andM.N.James,“EffectofprocessparametersontheresidualstressesinAA5083-H321frictionstirwelds,”MaterialsScienceandEngineering:A,vol.501,no.1.pp.119–124,2009.

[23] K.Deplus,A.Simar,W.VanHaver,andB.DeMeester,“Residualstressesinaluminiumalloyfrictionstirwelds,”TheInternationalJournalofAdvancedManufacturingTechnology,vol.56,no.5–8.pp.493–504,2011.

[24] C.LiuandX.Yi,“ResidualstressmeasurementonAA6061-T6aluminumalloyfrictionstirbuttweldsusingcontourmethod,”MaterialsandDesign,vol.46.pp.366–371,2013.

[25] P.CarloneandG.S.Palazzo,“LongitudinalResidualStressAnalysisinAA2024-T3FrictionStirWelding.”TheOpenMechanicalEngineeringJournal,vol.7,no.19.pp.18–26,2013.

[26] M.B.Prime,T.Gnäupel-Herold,J.A.Baumann,R.J.Lederich,D.M.Bowden,andR.J.Sebring,“Residualstressmeasurementsinathick,dissimilaraluminumalloyfrictionstirweld,”ActaMaterialia,vol.54,no.15.pp.4013–4021,2006.

[27] M.B.Prime,“Cross-SectionalMappingofResidualStressesbyMeasuringtheSurfaceContourAfteraCut,”JournalofEngineeringMaterialsandTechnology,vol.123,no.2.pp.162–168,2000.

[28] F.Hosseinzadeh,J.Kowal,andP.J.Bouchard,“Towardsgoodpracticeguidelinesforthecontourmethodofresidualstressmeasurement,”TheJournalofEngineering.p.online-only,2014.

[29] M.B.Prime,M.R.Hill,A.DeWald,R.J.Sebring,V.R.Dave,andM.J.Cola,“Residualstressmappinginweldsusingthecontourmethod,”TrendsinWeldingResearch,Proceedings,vol.836.pp.891–896,2002.

[30] M.Peel,A.Steuwer,M.Preuss,andP.J.Withers,“Microstructure,mechanicalpropertiesandresidualstressesasafunctionofweldingspeedinaluminiumAA5083frictionstirwelds,”ActaMaterialia,vol.51,no.16.pp.4791–4801,2003.

[31] V.Richter-Trummer,E.Suzano,M.Beltrão,A.Roos,J.F.dosSantos,andP.M.S.T.deCastro,“InfluenceoftheFSWclampingforceonthefinaldistortionandresidualstressfield,”MaterialsScienceandEngineeringA,vol.538.pp.81–88,2012.

[32] V.Richter-Trummer,P.M.G.P.Moreira,J.Ribeiro,andP.M.S.T.deCastro,“ThecontourmethodforresidualstressdeterminationappliedtoanAA6082-T6frictionstirbuttweld,”MaterialsScienceForum,vol.681.pp.177–181,2011.

[33] P.J.WithersandH.Bhadeshia,“Residualstress.Part2–Natureandorigins,”Materialsscienceandtechnology,vol.17,no.4.pp.366–375,2001.

[34] Y.Traore,S.Paddea,P.J.BouchardandM.A.Gharghouri,"Measurementoftheresidualstress

tensorinacompacttensionweldsample",ExperimentalMechnics,vol.53,Issue4,pp605-618,2013.