Materials Science and Engineering A 527 (2010) 7099–7108
Contents lists available at ScienceDirect
Materials Science and Engineering A
journa l homepage: www.e lsev ier .com/ locate /msea
eformation behaviour of spot-welded high strength steelsor automotive applications
. Brauser ∗, L.A. Pepke, G. Weber, M. RethmeierAM Federal Institute for Materials Research and Testing, Unter den Eichen 87, 12205 Berlin, Germany
r t i c l e i n f o
rticle history:eceived 25 March 2010eceived in revised form 21 July 2010ccepted 28 July 2010
eywords:
a b s t r a c t
Numerical simulation of component and assembly behaviour under different loading conditions is a maintool for safety design in automobile body shell mass production. Knowledge of local material behaviouris fundamental to such simulation tests. As a contribution to the verification of simulation results, thelocal deformation properties of spot-welded similar and dissimilar material joints in shear tension testswere investigated in this study for a TRIP steel (HCT690T) and a micro-alloyed steel (HX340LAD). Forthis reason, the local strain distribution was calculated by the digital image correlation technique (DIC).
esistance spot weldingeformation behaviourdvanced high strength steelRIP steelimilar and dissimilar material spot weldBSDEM
On the basis of the hardness values and microstructure of the spot welds, the differences in local strainbetween the selected material combinations are discussed. Additionally, the retained austenite contentin the TRIP steel was analysed to explain the local strain values. Results obtained in this study regardingsimilar material welds suggest significant lower local strain values of the TRIP steel HCT690T comparedto HX340LAD. One reason could be the decrease of retained austenite in the welded area. Furthermore,it has been ascertained that the local strain in dissimilar material welds decreases for each component
spond
train field
compared with the corre
. Introduction
In the last decade a change in body shell mass production hasccurred in the automotive industry. In answer to the intensi-ying energy crisis and in order to meet customer requirementsor automobiles such as weight reduction for energy saving andnhancement of passenger safety, new materials, e.g. advancedigh strength steels (AHSS) have to be applied. These materialsre gaining in popularity due to their high strength in combina-ion with good ductility characteristics compared to traditionaligh strength steels, for example micro-alloyed steels [1–3]. An
mportant AHSS representative is the so-called TRIP (TRansforma-ion Induced Plasticity) steel dominated by a ferrite matrix withetained austenite, bainite and martensite as dispersed phases,ffering excellent mechanical properties due to the transformationf retained austenite into martensite during plastic straining [2,4].s a result, both strength and uniform strain increase owing to theppearance of a harder phase and to the additional local plastic
ielding of the surrounding grains related to the transformationtrain [5,6].
In the lightweight body shell mass production of automobiles,esistance spot welding is the most important joining technique.
Typical vehicles contain about 3000–5000 spot welds. Therefore,good resistance spot welding behaviour must be one of the keycharacteristics of any steel grade to be used in automobile produc-tion [6].
For safe design of spot-welded body shell components, knowl-edge of the failure mechanism under static and fatigue loading isof main interest [6–8]. Typically, three different failure modes canoccur in spot-welded structures, i.e. interfacial failure, plug fail-ure and partial plug failure [6,9]. Of these, the plug failure is thedesired failure mode in automobile industry. For example, Zunigaand Sheppard [10] as well as Lin et al. [11] studied the failuremodes of lap-shear specimens using optical micrographs. Lin etal. [11] and Wung et al. [12,13] proposed failure criterions undercombined three resultant forces and three resultant moments andunder combined loads, respectively, based on experimental results.The verification of such models is difficult because of the limitedavailable experimental test data. Further experimental work hasshown that in the plug failure mode, stress is concentrated at thenugget circumference and leads to necking in the HAZ and basemetal, respectively [14–16]. But, due to the localised joining zoneand, hence, localised stress/strain in the spot welds, characteris-
tic material values for the welded area are not available till today,especially for AHSS.
In order to investigate the failure mechanism in more detail,the plastic strain and the stress distribution near the nugget mustbe known [17]. However, for lap-shear specimens which are typi-
7100 S. Brauser et al. / Materials Science and Engineering A 527 (2010) 7099–7108
ur of spot-welded shear tension samples via [6].
cfoNfisTltfptttatbdiabb
drdcmtttol
2
wHordHatld
TM
Fig. 1. FE-simulation of local strain behavio
al of spot-welded structures, analytical elastic–plastic solutionsor stresses and plastic strains near the nugget are difficult tobtain. Therefore, researchers usually apply numerical methods.umerous studies have been dedicated to elastic and elastic–plasticnite element analyses in order to characterise the fracture undertatic and cyclic loading, i.e. the fatigue behaviour. Radakovic andumuluru [6] proposed a simplified three-dimensional model ofap-shear specimens and determined the local strain behaviour forhe interfacial failure and the plug failure, see Fig. 1. In the interfacialailure mode, the maximum local strain occurs in the nugget. In thelug failure mode, by contrast, the maximum local strain is foundo be at the inner surface of the sheet and decreases in direction tohe outer surface. These simulation results are in agreement withhe experimental work discussed above. Lin et al. [17], Kan [18], Pannd Shepherd [19] as well as Satoh et al. [20] conducted two- andhree-dimensional finite element analyses to examine the fatigueehaviour of spot-welded structures on the basis of plastic strainistribution near the nugget. Particularly with a view to validat-
ng and optimising such numerical simulations, local strain valuesre needed to calculate the real deformation behaviour and finallyuild up a realistic finite element model to analyse the fractureehaviour of spot-welded joints in automotive structures.
The objective of this study was to investigate the local surfaceeformation behaviour of spot-welded similar and dissimilar mate-ial welds in a shear tension test and to calculate local materialata that can be used to validate numerical simulation of static andyclic loading of spot welds. For this purpose, an optical strain fieldeasurement system with high resolution was used. To investigate
he effect of different stain values the fracture surface was charac-erised using scanning electron microscopy (SEM). Furthermore, onhe basis of EBSD (Electron Backscatter Diffraction) measurementsf the retained austenite content, explanation will be given for theocal strain values of TRIP steel HCT690T.
. Experimental
In this study two different types of high strength steelsere selected, including micro-alloyed steel HX340LAD and AHSSCT690T. The micro-alloyed steel HX340LAD was chosen becausef its extensive use in the automotive industry for similar mate-ial welds (SMW) and above all with regard to its application forissimilar material welds (DMW), especially in conjunction with
CT690T. Table 1 shows an extraction of the chemical compositionnd the mechanical properties of the tested steels. Furthermore,he carbon equivalent (CE) characterised based on Eq. (1) [21] isisted, too. All steel grades offer a thickness of 1 mm and were hotip zinc coated with an average weight of 140 g m−2.
able 1echanical properties and an extract of the chemical composition of the base materials,
Steel grade Yield strength (MPa) Tensile strength (MPa) A (%)
HX340LAD 370 450 32HCT690T 420 750 30
Fig. 2. Shear tension specimen dimension.
DMW dominate in body shell mass production. Therefore, inaddition to the base metal combinations HX340LAD/HX340LADand HCT690T/HCT690T, used as a reference, the combinationHX340LAD/HCT690T will also be investigated.
CE = %C + %Mn/6 + (%Cr + %V)/5 + %Si/15 (1)
In this investigation, two sheet samples, 105 mm long and45 mm wide, were overlapped by 35 mm and single spot-welded
in the centre of the overlapped region, Fig. 2. These shear ten-sion samples were used to calculate the local strain of conventionalspot-welded structures.
Following EN ISO 14329 [9], spot-weld failure may occur inthree modes: interfacial failure, plug failure and partial plug fail-
measured via tensile test and Emission Spectrometry.
a stationary electron beam strikes a tilted crystalline sample and
TW
Fig. 3. Examples for failure types in shear tension test.
re. In the interfacial failure mode, the failure occurs through theugget, while in the plug and the partial plug failure mode, it occursy complete or partial withdrawal of nugget from one sheet, aschematically shown in Fig. 3.
It is well established that the nugget diameter has an influencen the fracture behaviour of spot-welded joints. In simple terms, amall nugget diameter often results in an interface failure while aigger weld nugget normally leads to a plug failure or to a partiallug failure [15,16]. It is conjecturable that different failure types
ead to modified deformation characteristics. The interface failureesults in negligible deformation of the specimen surface while thelug failure is characterised by a significant deformation on thepecimen surface, see Fig. 2. Accordingly, to avoid an influence ofeld size and fracture type, respectively, on the resulting deforma-
ion, a constant nugget diameter of 4.5√
t (t = sheet thickness, 1 mm)as used which induces plug failures in all cases. The samples were
roduced using an electromotive controlled welding gun with con-tant current control and direct-current operation. All specimensere welded using medium frequency current and electrode caps
f type F16 flattened to a face diameter of 5.5 mm. The welding
Fig. 4. Schematic overview of strain field measurement w
gineering A 527 (2010) 7099–7108 7101
parameters resulting always in a nugget diameter of 4.5 mm aresummarised in Table 2.
Due to the local deformation in spot-welded shear tension spec-imens only a qualitative description of the deformability of thetested materials could be given for the shear tension test. Therefore,the deformation behaviour of spot-welded specimens was analysedusing a digital image correlation system. This non-destructive opti-cal method calculates the local strain by analysing the change of astochastically patterned surface coating. In this investigation thespecific strain εx (strain in tensile direction, Fig. 4) is measured;the local strain in the y- and z-directions are relatively low. Notethat the strain field shows the strain distribution on the specimensurface.
The measurement system consists of two 2-megapixle camerasadjusted and linked to an image and data processing system whichcalculates strain fields of specimens with stochastically patternedsurface coating, as schematically shown in Fig. 4. The frame rate ofthe taken pictures was 1 Hz. Additionally, two stroboscopes wereused and synchronised with the camera system to ensure uniformexposure of the patterned specimen surfaces. Normally, the defor-mation of one specimen surface can be measured. Following Lorenzand Kannengiesser [22] where an optical system consisting of tworeflectors was used, it was possible to observe the specimen fromthe back side, front side and side, see Fig. 4. Hence, independentlyof the specimen view, the maximum local strain of the specimencan be appointed. The shear tension test was used to characterisethe mechanical properties and the deformation behaviour of thewelds. All tests were performed at room temperature with a strainrate of 0.01 mm s−1. Metallographic tests were used to measure thenugget diameter.
Additionally, to identify the deformation behaviour of spot-welded TRIP steel HCT690T, the content of retained austenite in thearea of crack propagation was analysed using EBSD and the SorpasFE-simulation software. The purpose of using the commercial FE-simulation software Sorpas is to determine the cooling times t8/5in the HAZ fracture area and to correlate this information with thechange in austenite content which is calculated by EBSD. In EBSD,
the diffracted electrons form a pattern on a fluorescent screen. Thispattern is characteristic of the crystal structure and enables, amongother factors, the separation of face cubic centered (fcc) and bodycubic centered (bcc) lattice structures.
) Welding time in cycles Pre/post holding time in cycles
7102 S. Brauser et al. / Materials Science and Engineering A 527 (2010) 7099–7108
Fig. 5. Similar and dissimilar material spot-welded cross-sections, (a) HX340LAD, (b) HX340LAD/HCT690T, (c) HCT690T.
Fig. 6. Hardness profiles for similar and dissimilar material spot welds with schematic location of the indentations, (a) HX340LAD, (b) HX340LAD/HCT690T and (c) HCT690T.
Fig. 7. Shear tension test results, (a) load–displacement curves, (b) Failure load depending on base metal combination.
S. Brauser et al. / Materials Science and Engineering A 527 (2010) 7099–7108 7103
ear te
3
3
ia
tNwhF3pHmafc(
Hswci
Fig. 8. Local strain εx on front and back side of the specimens during sh
. Results and discussion
.1. Microstructure and hardness distribution
Weld cross-sections of the tested steel combinations are shownn Fig. 5. The three different regions, i.e. base metal (BM), heatffected zone (HAZ) and fusion zone (FZ) are labelled.
The hardness distribution provides indirect information abouthe strength and the deformation behaviour of spot-welded joints.ormally, increasing hardness results in decreasing formabilityith simultaneously increasing strength [23]. Typical weld nuggetardness profiles of the tested material combinations are given inig. 6. Due to the high cooling rates in resistance spot welding, i.e.000–10,000 K/s [24], an increase in weld nugget hardness, in com-arison to the base metal, is observed. It can also be seen that theAZ hardness decreases from the nugget edge towards the baseetal, which indicates the decrease of material strength in that
rea. In agreement with the literature [15,16], the hardness of theusion zone (nugget) and of the HAZ of TRIP steel HCT690T is appre-iably higher as a consequence of the higher carbon equivalentTable 1), in comparison to the micro-alloyed steel HX340LAD.
The hardness profile for the dissimilar material welds
X340LAD/HCT690T shown in Fig. 6b takes a discontinuous
hape which is attributable to the different properties of theelded materials. Examination of the HAZ hardness values of each
omponent (HX340LAD; HCT690T) does not reveal any significantncrease in hardness compared to the corresponding SMW. Thus,
Fig. 9. Results of front and back side strain filed measurement at the peak load (a) HX
nsion test for (a) HX340LAD, (b) HCT690T and (c) HX340LAD/HCT690T.
in the case of plug failure where the HAZ strength is of mainimportance to the shear tension strength of the weld, the DMWperformance is expected to be similar to that of the SMW ofHX340LAD, based on the assumption that the fracture happens inthe weakest part of the weld.
However, concerning the weld nugget, a deviation in hardnesscan clearly be identified which lies between the values obtainedfor the similar material weld, with a shift to the harder material(HCT690T). Consequently, in DMW the weld nugget hardness isinfluenced to a greater extent by the harder material.
3.2. Static shear tension test
The stability of spot welds is often characterised with the helpof shear tension tests. The load–displacement curves obtainedfrom shear tension tests for a nugget diameter dn of nearly 4.5 mmare plotted in Fig. 7a. For SMW, a considerable increase in failureload with ascending base metal strength is discovered. This can berelated to the fact that the strength of spot welds is determined,among other parameters like sheet thickness and nugget size(constant parameter in this study), by the nugget and especiallyHAZ hardness.
Regarding the failure loads of DMW, comparable values as inthe case of SMW of HX340LAD can be achieved; however, there isa significant drop in displacement. This confirms the assumptionsregarding the strength of DMW based on the results of hardnesstests are seen below. Fig. 7b shows this trend of increasing peak
340LAD (b)/HCT690T/HX340LAD with schematically indicated nugget location.
7104 S. Brauser et al. / Materials Science and Engineering A 527 (2010) 7099–7108
F of loH
scao
pdH
ig. 10. Maximum local strain εl ,max versus shear tension load with visualisationX340LAD.
hear tension loads with rising base metal strength. Previous workoncerning the influence of the softer material part in DMW haslso shown that the softer material component leads to a decreasef the failure load [25].
Furthermore, the decrease in displacement, shown in Fig. 7a,oints out that DMW of HX340LAD/HCT690T offer lowereformability compared to the base metal combination ofX340LAD/HX340LAD.
Fig. 11. Cross-section with hardness values (a) HX34
cal strain (a) HX340LAD/HCT690T (b) HCT690T, (c) HX340LAD/HCT690T and (d)
3.3. Strain measurement
In order to measure the surface strain of spot welds in the spotweld area, the systematic error of the measuring system was first
determined by analysing an unstressed specimen. As a result, thelocal strain varies within a range of 0.25% and −0.3%.
The results of local strain measurements in the area of maxi-mal surface strain are plotted in the diagrams depicted in Fig. 8.
0LAD, (b) HCT690T, (c) HX340LAD/HX340LAD.
S. Brauser et al. / Materials Science and Engineering A 527 (2010) 7099–7108 7105
reas, (
Dtebbist
melbcs
ttFtvlbo
aacasammtoHnIoo
r
Fig. 12. SEM-fractography of spot weld fracture a
ue to the identical microstructures as well as hardness values inhe HAZ and the fracture area, respectively, no significant differ-nces in strain values between front and back side of SMW coulde observed. Fig. 9a exemplifies the strain fields of HX340LAD onoth sides to demonstrate that not only the strain values are nearly
dentical but also the measured strain fields. In the following, thepecimen side where the maximum strain occurs will be consideredo discuss the strain behaviour.
As expected, the results of DMW (HCT690T/HX340LAD) revealaterial-dependent strain behaviour, Figs. 8c and Fig. 9b. After
xceeding 1% local strain the HX340LAD exhibits a faster increase ofocal strain than the TRIP component. The comparison of the strainehaviour of SMW and DMW shows only small differences con-erning the maximum strain values (Fig. 8) as well as the measuredtrain fields (Fig. 9).
By analysing the measured strain field during the shear tensionest as well as the cross-section of the tested sheet metal combina-ion, statements about the deformation behaviour can be made. Inig. 10, the maximum local strain εx,max is plotted versus the shearension load. Moreover, the results of strain field measurements areisualised. For visualisation of the deformation behaviour, the peakoad Fs,max (shear tension strength) was used as reference point,ecause the crack initiation and propagation was not in the focusf this study.
The strain curve of spot-welded HX340LAD (Fig. 10b) showslarge range of plastic deformation and necking which is char-
cteristic of ductile material behaviour. The maximum strain isoncentrated in the HAZ/base metal transition zone (see Fig. 10d)nd reaches values up to 15% at the peak load. In contrast, the TRIPteel HCT690T exhibits only a small range of plastic deformationnd fracture happened after achieving the peak load (11.4 kN). Thisaterial behaviour is representative of a less ductile fracture. Theaximum deformation achieved at the surface happened also in
he HAZ/base metal transition zone (Fig. 10b) and reaches valuesf nearly 5%. A reason for the lower local strain of the TRIP steelCT690T is the lower rotation of the nugget due to the higherugget hardness that hinders the deformation of the surface [14].
t should be noted that the local strain behaviour and the positionf the maximum local strain is in good agreement with the resultsf Radakovic and Tumuluru [6] shown in Fig. 1b.
The respective load–strain curves of DMW show runs compa-able to those of the corresponding curves of spot-welded SMW.
a) HX340LAD, (b) HCT690T, HX340LAD/HCT690T.
However, since the softer material significantly determines the fail-ure load in DMW, the peak load of DMW is lower than in SMWof TRIP steel HCT690T.At the maximum local strain (peak load),the HX340LAD component offers values of up to 12% while theHCT690T component exhibits a value of 3.5%, Fig. 10a and c. There-fore, a nearly similar decrease of the DMW local strain values of upto 20% is found for both steel parts in comparison to the correspond-ing SMW. Consequently, the combination of different materials inRSW results in a major decrease of the peak load associated witha drop in maximum local strain relating to the stronger materialpart. Concerning the softer material component, only the maxi-mum local strain is reduced, however without any influence on thefracture behaviour.
The differences in local strain values between the tested sheetmetal combinations are reflected in the fracture behaviour, Fig. 11.The micrographs of the cross-section for the micro-alloyed steelHX340LAD and the DMW (HX340LAD/HCT690T) show that fracturehappened in the HAZ/base metal transition zone approximately1 mm away from the nugget circumference after significant neck-ing, Fig. 11a and c. In the DMW, fracture happened in the softermaterial part (HX340LAD). Lin et al. [17] have performed finite ele-ment analyses of the failure modes of spot welds and have shownthat when necking failure occurs at the distance in the order ofthe thickness away from the notch tip, the ductility of the mate-rial near the notch or crack along the nugget circumference is high.This result corresponds well with the local strain values discussedabove.
Unlike the micro-alloyed steel HX340LAD where fracture hap-pened only in the HAZ/base metal transition zone, the TRIP steelHCT690T fracture started directly in the HAZ region with the max-imum hardness gradient (nearly 526 HV → 378 HV, Fig. 6b) andpropagated into the region with reduced hardness and strength,Fig. 10b. Based on the outcomes of [17] this behaviour could beattributed to lower ductility of the material near the notch whichleads to initiation of kinked cracks at the critical locations of thenotch seen in Fig. 10b.
To examine the deformation behaviour in more detail, SEM anal-
yses of the fracture region were performed, Fig. 12. Due to thefact that the shape of the dimples depends on the loading condi-tions, the plug failure under tensile loading predominantly resultsin equiaxed dimples while shear loading will create elongated dim-ples [15]. Fig. 12a shows the results of SEM for the micro-alloyed
7106 S. Brauser et al. / Materials Science and Engineering A 527 (2010) 7099–7108
Ft
sclftin((
TeuniOsff
F
F
atm(sr
atdbo
Fig. 14. Retained austenite (blue) in TRIP steel HCT690T base metal. (For interpre-
Fd
ig. 13. Force distribution at nugget centreline and circumference during shearensile test [15].
teel HX340LAD (SMW). It can be seen that the dimples signifi-antly elongated indicate that the fracture happened under shearoad. This result is opposite to the work of Chao [7] who studiedailure mechanisms of pullout occurring in RSW during the shearest. In contrast, the TRIP steel HCT690T exhibits equiaxed dimplesn the fracture zone which are typical of a tensile fracture mecha-ism Fig. 12b. In the case of DMW, elongated dimples are observedFig. 12c) that implies similar behaviour to SMW of HX340LADfracture under shear load).
The differences in fracture mechanisms (shear, tensile) betweenRIP steel HCT690T and micro-alloyed steel HX340LAD could bexplained by the simple model for stress distribution in spot weldsnder shear tensile load seen in Fig. 13. Shear stresses are domi-ant at the interface. At the nugget circumference, the stress nature
s tensile shear at position A and compressive at position B [15].wing to the macroscopic rotation of the weld, not only tensile
hear force F| is produced at the spot weld but also cross-tensionorce F⊥ , whereas with resign rotation angle ϕ the cross-tensionorce F⊥ increases:
⊥ = F · sin ϕ (2)
‖ = F · sin ϕ (3)
As a result, the softer material HX340LAD with higher rotationngle ϕ (∼10◦ at the peak load) is subject to a higher cross-ension force than the TRIP steel HCT690T (∼4◦). Therefore, the
icro-alloyed steel HX340LAD and the DMW HX340LAD/HCT690Trotation angle ϕ ∼ 9◦) show in contrast to the SMW of HCT690T ahear fracture behaviour with elongated dimples. The difference inotation corresponds well with that described in the work of [14].
Concerning the deformability, ductile fracture characterised by
dimple structure is observed in all cases (SMW, DMW). Even
hough the deepness of the dimples is an indication of the materialuctility [26], the differences in local strain seen in Fig. 10 cannote detected by SEM analyses for comparison due to the elongationf the dimples.
ig. 15. Temperature field at the peak temperature of spot-welded TRIP steel HCT690T wown-curves for three measurement points in the HAZ (b).
tation of the references to colour in this figure legend, the reader is referred to theweb version of the article.)
3.4. Decrease in ductility of TRIP steel HCT690T
A main reason for the high ductility of TRIP steel HCT690T is theretained austenite content which can transform into martensiteunder stress [1,2]. Fig. 14 shows the austenite distribution (bluemarked area) in the base metal measured with the help of EBSD.The fine dispersed austenite is arranged at the grain boundaries ofthe fcc lattice structures (grey area; primarily ferrite) and reacheda content of ∼17%.
The following discussion regarding the low local ductility inthe spot-welded area of TRIP steel HCT690T focuses on the HAZ,because in this investigation fracture starts in and propagatesthrough the HAZ. During resistance spot welding, the HAZ under-goes a temperature cycle which is characterised by a high coolingrate compared to other welding procedures [24]. Fig. 15a showsthe temperature field at the peak temperature of the tested TRIPsteel determined by FE software Sorpas. Furthermore, the crackpath seen in Fig. 11b is labelled. On the basis of the calculatedtemperature field, cooling-down-curves for three points near thecrack path and the corresponding cooling times t8/5 were investi-gated, Fig. 15b. The short process time of resistance spot weldingresults in times t of 80 ms (point 1) and 100 ms (point 2), respec-
8/5tively. Based on continuous cooling transformation diagrams forlow alloyed TRIP steel [27,28], only martensite can develop in theHAZ directly adjacent to the fusion zone.
ith spot weld diameter of 4.5 mm determined by FE software Sorpas (a), cooling-
S. Brauser et al. / Materials Science and Engineering A 527 (2010) 7099–7108 7107
F ed ar
rofctdtntc
rfs
4
tawct
•
ig. 16. Results of EBSD measurement near the crack path area; black and red mark
The EBSD analyses confirm the assumptions regarding theeduction of austenite during spot welding. Fig. 16 shows the resultsf EBSD measurement near the crack path area (HAZ) transcribedrom Fig. 11b. It can be seen on the one hand that the austeniteontent in the region of crack propagation is negligibly small. Onhe other hand, the austenite content increases with increasingistance from the fusion zone. That is a result of the lower peakemperature during welding (see Point 3 in Fig. 15b) which doesot lead to a complete austenite transformation into bcc struc-ures (martensite, bainite). The black marked area (Fig. 16b and) corresponds to very fine fcc structures.
Summarising, resistance spot welding leads to a nearly completeemoval of retained austenite as well as a hardness increase in theracture region (HAZ) resulting in a considerable decrease of localtrain measured with digital image correlation technique.
. Conclusions
In this study, deformation behaviour of SMW and DMW in shearension test was investigated using a system for optical strain filednalysis. This was accomplished applying hardness values of theeld and HAZ, load–displacement curves as well as load–strain
urves. SEM and EBSD analyses were carried out to characterisehe deformation behaviour in the fracture area.
The following essential conclusions can be drawn:
The failure load does not show a linear increase with the basemetal strength. Failure loads of dissimilar material welds arelocated between the analysed similar material welds with a sig-
eas correspond to fcc structure, blue marked areas correspond to bcc structure.
nificant shift to the softer material (HX340LAD) and a drop indisplacement.
• Up to the point of uniform elongation, the local strain measure-ment of SMW shows no significant differences between front andback side.
• The local strain observed for the micro-alloyed steel HX340LADreaches values of ∼15% while the TRIP steel offers values of only∼5%.
• In DMW, a nearly similar decrease of the local strain valuesaccounting for up to 20% for both steel parts is found comparedwith the results of SMW.
• With regard to the ductility behaviour, spot-welded joints arecharacterised by a significant loss of local strain accounting forup to 25% for HX340LAD and 85% for HCT690T compared to thebase metal.
• SEM results show no indication of reduced deformability of TRIPsteel compared to the micro-alloyed steel HX340LAD. But evi-dence of different loading conditions between HX340LAD andHCT690T was found.
• EBSD analysis of the retained austenite content reveals anelimination of austenite in the fracture region of spot-weldedHCT690T. The austenite reduction and the hardness increase inthe fracture region (HAZ) results in low strain values comparedto the base metal.
The strain values obtained in this investigation can be integratedinto numerical simulation studies to help analyse the deformationas well as the fracture behaviour of spot welds and are expected tocontribute to increased passenger safety in the long run.
7 and En
R
[
[
[
[
[
[
[[[[[[
[
[
[[
108 S. Brauser et al. / Materials Science
eferences
[1] S. Maggi, M. Murgia, Weld. Int. 22 (2008) 610–618.[2] Advanced High Strength Steel (AHSS) Application Guidelines Version 4.1
http://www.worldautosteel.org/Projects/AHSS-Guidelines.aspx, 13.04.2010.[3] M.D. Tumuluru, Weld. J. (2006) 31–37.[4] G. Lacroix, T. Pardoen, P.J. Jacques, Acta Mater. 56 (2008) 3900–3913.[5] M.H. Saleh, R. Priestner, J. Mater. Proc. Technol. 113 (2001) 587–593.[6] J.D. Radakovic, M.D. Tumuluru, Proceeding of the International Sheet Metal
Welding Conference XIII, 2008.[7] Y.J. Chao, J. Eng. Mater. Technol. 125 (2003) 125–132.[8] J.A. Davidson, SAE Technical Paper 830033, Society of Automotive Engineers,
Warrendale, PA.[9] ISO 14329:2000: Resistance welding – destructive tests of welds – failure types
and geometric measurements for resistance spot, seam and projection welds,2003.
10] S. Zuniga, S.D. Sheppard, in: R.S. Piascik (Ed.), Fatigue and Fracture Mechanics,
vol. 27, ASTM STP 1296, 1997, pp. 469–489.
11] S.-H. Lin, J. Pan, S. Wu, T. Tyan, P. Wung, SAE Technical Paper 2001-01-0428,Society of Automotive Engineering, Warrendale, 2001.
12] P. Wung, T. Walsh, A. Ourchane, W. Stewart, M. Jie, Exp. Mech. 41 (2001)100–106.
13] P. Wung, Exp. Mech. 41 (2001) 107–113.
[
[
[
gineering A 527 (2010) 7099–7108
14] H. Zhang, J. Senkara, Resistance Welding: Fundamentals and Applications, 1sted., CRC Press, 2005, pp. 107–146.
15] M. Pouranvari, H.R. Asgari, S.M. Mosavizadch, P.H. Marashi, M. Goodarzi, Sci.Technol. Weld. J. 12 (2007) 217–225.
16] M.I. Khan, L.M. Kuntz, Y. Zhou, Sci. Technol. Weld. J. Vol.13 (2008) 294–304.17] P.-C. Lin, S.-H. Lin, J. Pan, Eng. Frac. Mech. 73 (2006) 2229–2249.18] Y.R. Kan, Met. Eng. Quart. 16 (1976) 26–36.19] N. Pan, S. Sheppard, Int. J. Fatigue 24 (2002) 519–528.20] T. Satoh, H. Abe, K. Nishikawa, M. Morita, Trans. Jpn. Weld. Soc. 22 (1991) 46–51.21] M.J. Cieslak, ASM Handbook, vol. 6, ASM International, Materials Park, Ohio,
1990, pp. 94–95.22] S. Lorenz, T. Kannengiesser, Proceeding of the Fifteenth International Confer-
ence on the Joining of Materials (JOM 15), Helsingr-Denmark, 2009.23] DIN EN ISO 18265: Metallic materials–Conversion of hardness values (ISO
18265:2003); German version EN ISO 18265:2003.24] J.E. Gould, S.P. Khurana, T. Li, Weld. J. 85 (2006) 111–116.25] B. Hernandez, M.L. Kuntz, M.I. Khan, Y. Zhou, Sci. Technol. Weld. J. 13 (2008)
769–776.26] G. Lange, (Hg.): Systematische Beurteilung Technischer Schadensfälle, 5th ed.,
Wiley-VCH, Weinheim, 2001, ISBN 3-527-30417-7 (in German).27] M. Zhang, L. Lia, R.Y. Fu, D. Krizan, B.C. De Coomanc, Mater. Sci. Eng. A 438
(2006) 296–299.28] M. Grajcar, Opiela, J. Ach. Mater. Manu. Eng. 29 (2008) 71–78.
