See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/248465361

New parametric study on nugget size in

resistance spot welding using finite element
method

Article in Materials and Design January 2010

DOI: 10.1016/j.matdes.2009.06.042

CITATIONS READS

57 580

3 authors, including:

Hamid Eisazadeh Mohsen Hamedi

Clarkson University University of Tehran
13 PUBLICATIONS 79 CITATIONS 74 PUBLICATIONS 327 CITATIONS

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Piezoelectric Energy Harvesting View project

Force-Frequency Effect, Quartz View project

All content following this page was uploaded by Hamid Eisazadeh on 26 October 2015.

The user has requested enhancement of the downloaded file.

 Materials and Design 31 (2010) 149157

Contents lists available at ScienceDirect

Materials and Design

journal homepage: www.elsevier.com/locate/matdes

New parametric study of nugget size in resistance spot welding process using
nite element method
Hamid Eisazadeh a,*, Mohsen Hamedi b, Ayob Halvaee c
a
Department of Mechanical Engineering, Chabahar Maritime University, Chabahar, Sistan va Balochestan 9971756499, Iran
b
Department of Mechanical Engineering, University of Theran, North Kargar at Jalal-Exp Way, Tehran 1439957131, Iran
c
Department of Metallurgy Engineering, University of Theran, North Kargar at Jalal-Exp Way, Tehran 1439957131, Iran

a r t i c l e i n f o a b s t r a c t

Article history: Resistance spot welding process (RSW) is one of important manufacturing processes in automotive
Received 18 March 2009 industry for assembling bodies. Quality and strength of the welds and therefore body mainly are dened
Accepted 26 June 2009 by quality of the weld nuggets. The most effective parameters in this process are: current intensity, weld-
Available online 1 July 2009
ing time, sheet thickness and material, geometry of electrodes, electrode force, and current shunting. In
present research, a mechanicalelectricalthermal coupled model in a nite element analysis environ-
Keywords: ment is made using. Via simulating this process, the phenomenon of nugget formation and the effects
Resistance spot welding
of process parameters on this phenomenon are studied. Moreover, the effects of welding parameters
Nugget size
Thermalelectro-mechanical coupled model
on temperature of faying surface are studied. Using this analysis, shape and size of weld nuggets are com-
analysis puted and validated by comparing them with experimental results from published articles. The method-
Automotive body strength ology developed in this paper provides prediction of quality and shape of the weld nuggets with variation
of each process parameter. Utilizing this methodology assists in adjusting welding parameters so that
costly experimental works can be avoided. In addition, the process can be economically optimized to
manufacture quality automotive bodies.
2009 Elsevier Ltd. All rights reserved.

1. Introduction quality as specied in the handbook provided by the Resistance

Welder Manufacturers Association. Fig. 1 shows a cause-and-ef-
Resistance spot welding is commonly used in the automotive fect diagram that nugget size is a dominant factor. In general, there
industry for joining thin sheet metals. Compared with other welding is a direct correlation between heat generation and nugget size in
processes such as arc welding processes, resistance spot welding is RSW process. The contact resistances at the faying surface between
fast, easily automated and easily maintained. This welding is a com- workpieces, current density, welding time and sheet thickness
plicated process which involves interaction of electrical, thermal, mainly determine the heat generation during welding and the sub-
mechanical and metallurgical phenomena. In this process, the mate- sequent nugget size. In present research, the phenomenon of nug-
rials to be joined are brought together under pressure by a pair of get formation and the effect of process parameters on shape and
electrodes and then a high electric current is passed through the size of weld nuggets are studied. In order to gain greater nugget
workpieces between the electrodes. Due to contact resistance and diameter, the effects of RSW parameters are studied by simulation
Joule heating, a molten weld nugget is formed in the workpieces. of this process in a nite element analysis. To simulation of this
The workpieces are joined as solidication of the weld pool occurs. process, a mechanicalelectricalthermal coupled model has been
Moreover, force is applied before, during and after the application made in a FEM environment using commercially available soft-
of electric current, to maintain the electric current continuity and ware. The effects of welding time and current, electrode force, con-
to provide the pressure necessary to form the weld nugget. The total tact resistivity and sheet thickness on the temperature of faying
heat generation between two sheets per unit time is dened as the surface is investigated. Using this analysis, shape and size of weld
product of the current intensity squared, multiplied by the total nuggets are computed and validated by comparing them with
resistance and the welding efciency. experimental results from published articles. These results can be
Quality and strength of the welds are dened by shape and size used to optimize the RSW parameters.
of weld nuggets. The nugget size has referenced to the welding
1.1. Literature review
* Corresponding author. Tel.: +98 21 88003318/9354; fax: +98 21 88003029.
E-mail addresses: [email protected], [email protected] (H. Eisazadeh), In order to increase the strength of products, many researchers
[email protected] (M. Hamedi). have studied RSW process by experimental and numerical tech-

0261-3069/$ - see front matter 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.matdes.2009.06.042
150 H. Eisazadeh et al. / Materials and Design 31 (2010) 149157

Fig. 1. Cause-and-effect diagram for nugget size in resistance spot welding.

niques [518]. Nugget formation experimental studies [58] are cal methodology for estimation of interface transient heat transfer
able to provide valuable temperature data, however, the test tech- coefcient during resistance welding. Aslanlar et al. [17] have
niques have their limitations and require expensive equipment. investigated the effects of welding time on the tensile-peel
Furthermore industries are trying to reduce the expenses used strength and tensile-shear strength of welding joints in electrical
for the testing process. These shortcomings make numerical simu- resistance spot welding. More recently, Rogeon et al. [18] has mea-
lation an attractive tool for complimenting experimental tempera- sured electrical contact resistances on a specic device, allowing to
ture. For this reason, researchers have studied RSW process by rise high pressure and elevated temperature. Performed measure-
developing one-dimensional, two-dimensional and rarely three- ments concern electrodesheet and sheetsheet interfaces.
dimensional models. Gould [5] measured the nugget growth by
using metallographic techniques and proposed a nite difference 2. Theoretical analyses
based one-dimensional heat transfer model. This model demon-
strated the importance of radial heat transfer, which inhibits the 2.1. Governing equation
calculation of nugget expansion. This model did not account for
non-uniform current density distribution. Tsai and Jammal [6,7] All the equations in this study are based on the two-dimen-
have created a two-dimensional symmetric model using ANSYS sional cylindrical coordinate system. The governing equation for
to perform some parametric studies on the spot welding process. calculation of the heat generation per unit volume may be shown
Khan et al. [8] have developed a model to predict the nugget devel- as [1]:
opment during RSW of Al alloys. The model calculates time varying
1
interface pressure and is used to determine the effect of the elec- q rU2 1
trode shape and applied pressure on the nugget growth only. Khan R
were employed the iterative method to simulate the interaction where q is the heat generation per unit, R is the electrical resistivity
between coupled electrical, thermal and structural elds. Feulv- and U is the electrical potential. The governing equation for tran-
arch and co-workers [10,11] have presented a nite element for- sient temperature eld distribution, which involves electrical resis-
mulation to measure the interface contact properties. It has been tance heat, may be written as [2]:
shown that the calculated nugget appears earlier. It has also been  
1 @ @T 1 @ @T @T
noticed that the nugget was growing faster across the thickness. r a ra q qc 2
r @r @r r @z @z @t
Hou and Kim [13] developed and analyzed a two-dimensional axi-
symmetric thermo-elasticplastic FEM model; it was developed where r and z are radial and axial coordinates and, q, c and K are
and analyzed in the commercial FEM program, ANSYS. The objec- density, specic heat, and thermal conductivity, respectively. The
tive of this research was to investigate the behavior of the mechan- term Q_ refers to the rate of the internal heat generation per unit vol-
ical features during the RSW process. Through the analysis, the ume within the boundary of the region of analysis. This particular
following results were obtained: the distribution and change of term accounts for the Joule heating due to bulk resistivity in the
the contact pressure at the electrodeworkpiece interface and fay- sheet-electrode system.
ing surface, the stress and strain distribution and deformation of For stress and strain analysis, since the thermalelasticplastic
the weldment, and the displacement of the electrode. Recently, behavior is a highly nonlinear phenomenon, the stressstrain rela-
Loulou and Bardon [14,15] used a method for estimation of ther- tion is described in incremental form [3]:
mal contact conductance coefcient at the electrode tipmetal
fDrg DfDeg fCgDT 3
sheet interface. His articles describe an experimental and numeri-
 H. Eisazadeh et al. / Materials and Design 31 (2010) 149157 151

where vectors Dr and De are stress and strain increment, respec- taken as copper. Temperature-dependent physical and mechanical
tively, and DT is temperature increment. Matrix D and vector C properties of materials, including thermal conductivity, coefcient
are materials related constant. of thermal expansion, electrical resistance, specic heat, density,
enthalpy, elasticity of yield stress and Poissons ratio, were used
2.2. Boundary conditions for both electro-thermal and thermalmechanical analysis [4].
Fig. 3 and Table 1 show the some mechanical properties for AISI
As with any numerical model, it is of great importance to dene 1008 steel sheet used in present study. The welding process in-
the boundary conditions and material properties correctly to ob- volves four stages cycles: squeezing, welding, holding and cooling.
tain realistic results. Fig. 2 has described the electrical, thermal The exact welding constants used in this study are given in Table 2.
and mechanical boundary conditions used in the model. A 60 Hz
sine wave electrical current ow was assumed to be uniformly dis-
tributed at the top surface of the upper electrode and was permit- 3. Numerical technique
ted to ow across the contact areas at both the electrode
workpiece and workpieceworkpiece interfaces and, eventually In this work, the resistance spot welding process is treated as an
reaching the bottom surface of the lower electrode. Since an ac electricalthermalmechanical problem solved using the nite ele-
welding machine is used in Gould research, the welding current ment method. Commercial nite element code ANSYS were used to
during RSW can be described with standard sinusoidal signal. model the coupling between electrical and thermal phenomena
Therefore, the root-mean square value of the welding current can and between the thermal and mechanical phenomena. The ow
be given by [16]: chart of the analysis procedure in Fig. 4 has illustrated an incre-
s mental method. Based on this owchart, the subroutine denes
Z
1 p Im the squeezing cycle as a single load step and divides the welding
Irms I sin 2pft2 d2pft p 4 time into 23 load steps. The nodal temperature distribution and
p 0 m 2
the updating of deformed geometrical information are realized
where Iw is the rms value of welding current, Im is the peak value, I by the APDL language. Temperature-dependent thermal, electrical
is the real current, f is the working frequency (60 Hz in this re- and mechanical properties of material were considered including
search). contact resistance. Due to the symmetric nature of the electrode
p Based on Eq. (4) the peak value of the welding current is
Im 2Irms . As a result, the applied welding current on the upper and the workpieces, a half of the physical model for two-dimen-
end surface of the upper electrode can be given by [16]: sional analysis is created and is shown in Fig. 5.
p
I Im sin2pft 2Irms sin2pft 5
4. Results and discussion
The bottom of the lower electrode was set to be zero for the ref-
erence electrical potential.
The FEM model was employed to simulate the RSW process in
To simulate the effect of cooling water in the electrode cavity,
order to quantitatively understand the effects of the process
the temperature of the electrodewater interface was maintained
parameters on temperature distribution and the nugget size at dif-
at a constant value during the welding process. A 4670 N mechan-
ferent cycles. In this section, nugget formation in RSW process is
ical load was applied as a pressure condition at the nodes on the
predicted in squeeze, welding and holding time and compared
top face of the upper electrode. It was increased linearly during
with experimental data from Gould [5].
the squeeze time and was held constant during the welding and
In squeeze stage, the electrodes and workpieces were deformed
holding times. The radial displacement was restricted along r-axis.
after a load of 4670 N which was applied to the electrodes. Fig. 6
To simplify the analysis, sliding at the electrodesheet interface
shows the Von Mises stress distribution after the squeezing stage.
was not modeled.
The maximum stress (127 Mpa) occurred at the edge of the contact
surface between the electrode and the workpiece. This phenome-
2.3. Materials properties and welding conditions non has been also observed in other researches [5,8]. From Fig. 6,
the contact pressure on the W-E interface is fairly uniformly in
The steel sheets in this study were AISI 1008 steel with chemi- the majority around the axis, and in the domain near the electrode
cal analysis of 0.08 C, 0.32 Mn and 0.018 S, and the electrode was edge, there is severe stress concentration. The simulation results
show that the degree of stress concentration depends on the geom-
etry parameters and electrode force. Figs 7 and 8 show the temper-
ature prole and nugget growth at 8th and 14th cycle of welding
process. These gures show the results on a half model obtained
during post processing. Assuming 1530 C as the melting point of
AISI 1008 steel, the spot nugget region appears as red color. The
highest temperature was always in the middle of the workpiece.
By changing the boundary conditions, the temperature prole
could be varied which in turn changed the nugget size, i.e. the
welding quality.
In this simulation, the weld nugget begins to form at the 5th
weld cycle, see Fig. 9. It quickly grows in both the lateral and the
vertical directions in the next 23 cycles. Approximately after
three cycles it slowly grows, until about 60% of the thickness of
the workpieces is reached, because of the contact resistance was
disappear for too heat. This shows that contact resistance plays a
critical role within the some rst weld cycle duration. Also the
highest temperature in the weld nugget was always in the middle
Fig. 2. Schematic describing thermal and electrical boundary conditions. of the faying surface.
152 H. Eisazadeh et al. / Materials and Design 31 (2010) 149157

Fig. 3. Variation of material property of AISI lOO8 and copper electrode with temperature: (a) variation af yield stress and elastic modulus of AISI lOO8 and copper electrode
with temperature. (b) Variation of electrical receptivity of AISI 1008 and copper electrode with temperature. (c) Variation of thermal conductivity of AISI 1008 and copper
electrode with temperature. (d) Variation of specic of AISI 1008 electrode with temperature. (e) Variation of thermal expansion of coefcient of AISI 1008 electrode with
temperature. (f) Variation of enthalpy of AISI 1008 with temperature [4].

During the holding time in which the current is set to zero, con- Table 1
vection and squeezing forces are only external loads in the thermo- Constant physical properties used in RSW process analysis [4].
mechanical model. The nal nugget size is obtained at the end of Name Value
cooling time because there will be some deformation in the weld-
Poissons ratio of steel 0.32
ing zone when the weldment cools due to the electrode pressure Poissons ratio of copper electrode 0.3
and material shrinkage. Fig. 9 shows the temperature history in Density of steel (kg/m3) 7800
different positions. This gure has demonstrated the effect of en- Density of copper electrode (kg/m3) 8900
thalpy at the melting temperature (1530 C) and also points out Yield stress (MPa) 83
Water convection coefcient at 20 temperature (W/m2 C) 300
that considering enthalpy in nite element model cause to raise
Air convection coefcient at 25 temperature (W/m2 C) 21
temperature slowly at melting point. Fig. 10 that illustrates the
 H. Eisazadeh et al. / Materials and Design 31 (2010) 149157 153

Table 2
Welding constants used in this study.

Welding Current (kA) Welding time (cycles) Electrode force (N) Electrode diameter (mm) Electrode taper () Sheet thickness (mm)
14.2 14 4670 7.6 30 1.52

temperature distribution along the faying surface after welding study is a more realistic and advanced FEM model where utilizing
time was employed to measure the nugget width where the hori- this methodology can adjust welding parameters for achieving
zontal axis represents the distance from the nugget center. optimum nugget dimension.
In this section, result of RSW process from FEM model has been In the Joule heating formula, the welding current has the great-
compared with the experimental result of Gould et al. To maintain est effect on the generation of heat at faying surface. Therefore it is
consistency, the dimensions of the workpiece, material properties, a primary control variable in RSW process. Fig. 12 shows the inu-
welding conditions and boundary conditions used were the same ence of welding current on nugget dimension in RSW process. The
as Goulds [5]. A comparison between measured shapes of weld weld nugget forms at a welding current higher than 14.5 kA, so it
nugget from Gould and the calculated results from FEM simulation appears that the welding current has much considerable inuence
at a welding current of 14.2 kA and a welding time of 14 cycles is on weld nugget for welding currents 15.516.2 kA. Experimental
shown in Fig. 11. The thickness of the experimental molten zone is data and solutions of one-dimensional model developed by Gould
about 2.12 mm, and the calculated one is 2.23 mm. It indicates a [5] are also presented for comparison. It indicates that the calcu-
good agreement between the calculated results and measured lated results from this FEM model agree well with measured data.
data. Experimental and numerical results both point out that if the elec-
Weld quality is often stated in terms of weld nugget diameter. tric current ow exceeds the ow necessary for nugget growth,
Based on this consideration, this study is performed to observe 15.5 kA, it causes a rapid growth of nugget. The nugget growth rate
the effect of various parameters on nugget dimension and geome- decreases as the current ow increases but the nugget size raises
try. These parameters are welding current, welding time, electrode until melt spattering occurs.
force, contact resistance and sheet thickness. However, some of
these parameters have been formerly studied as recently electrode
force was recently studied by khan [8], but employed model in this

Fig. 6. The distribution of Von Mises stress during the squeeze stage.

Fig. 4. FEM program ow chart.

Fig. 5. FEM model for resistance spot welding. Fig. 7. Temperature distribution at 8th cycle of welding process.
154 H. Eisazadeh et al. / Materials and Design 31 (2010) 149157

Fig. 11. Schematic comparing the predicted arid experimental weld nugget cross-
section dimensions.

Fig. 8. Temperature distribution at 14th cycle of welding process.

Fig. 12. Comparison between FEM model and experiment data from Gould
(l.52 mm sheet: 6 weld cycles: 4760 N).
Fig. 9. Temperature histories at two locations in RSW process.

test whilst space between nuggets is considered far away therefore

the total current is forced to pass through the faying surface close
to electrode tips. Consequently current density is more aggregated
in this area and current shunting is not observed. Effect of electric
current shunting is considered in the second test by assuming
welds close to each other as in Fig. 14. When the spot weld trigger
is pressed, current ows from one electrode tip. When the current
meets the nonconductive coating at the faying surfaces; it looks for
the nearest conductive path which is through the top sheet and
back through the other electrode tip. As a result, partial current
ows through previous made weld; this may cause a signicant
decrease in weld nugget size. Effect of electric current shunting
is reduced in test the third by increasing space between weld nug-
gets (Fig. 15). In this case, the heat dissipated due to Joule effect at
the interface is proportional with current density.
Fig. 16 shows the inuence of welding time on weld dimensions
in RSW process. Experimental data and solutions of the one-
Fig. 10. Temperature distributions along the faying surface.
dimensional model developed by Gould [5] are also presented for
comparison. The predicted nugget growth agrees well with the
experimental data, and also, the one-dimensional model predicts
In addition to magnitude of welding current, current shunting proper shape of curves. However, the one-dimensional model
phenomenon affects the nugget size. Generally more current den- curves are displaced from the calculated results. Gould suggested
sity increases nugget size. Previously made welds may affect the that these discrepancies could attribute to one of two inadequacies
subsequent welding due to electric current shunting when the in the model: radial conduction in the sheet and improper charac-
welds are spaced close to each other. In order to investigate the ef- terization of faying surface resistance. Calculated results shows,
fect of electric current shunting, some tests were designed and that the weld nugget is commenced at the welding time of 8 cycles
simulated. Fig. 13 shows current density distribution in the rst and increases abruptly after the welding time of 6 cycles. The rate
 H. Eisazadeh et al. / Materials and Design 31 (2010) 149157 155

Fig. l3. Schematic of electric current ow. Current shunting has not been
considered.

Fig. 16. Comparison between FEM model and experiment data from Gould
(1.52 mm sheet: 13 kA: 4760 N).

Contact resistance has drawn the attention of many researchers.

It is inuenced by many factors like base metal, pressure, temper-
ature, surface conditions, etc. The results of simulations presented
in this research illustrate the importance of the contact resistance
on the faying surface. These results as shown in Fig. 17, express
that the value of contact resistance is a critical factor in spot weld-
ing of steel. Fig. 17 points out that increasing contact resistance
due to different factors like surface base metal and surface condi-
tion lead to increase in nugget size.
Fig. 14. Schematic of electric current ow, current shunting has considered (space In addition to contact resistance, electrode force is also an
between two nugget is 12.7 mm). important factor involved in RSW. Fig. 18 presents nugget size at
three force levels and illustrates an indirect correlation between
the electrode force and nugget size which is due to faying contact
resistance. It shows that is a relationship between contact resis-
tance and electrode force. In order to show this relationship, pre-
dicted contact areas at three load levels is summarized in Table
3. It shows that under 4670 N and 6500 N loads, a correlation be-
tween electrode force and contact area is observed. The variable
of faying contact resistance in the Joule equation is inuenced by

Fig. l5. Schematic of electric current ow, current shunting has considered (space
between two nugget is 18 mm).

of increase in weld nugget thickness is reduced when the weld

time becomes larger. Increase of electric cycles severely raises
the contact surface temperature so that the contact zone melts
leading to a big nugget but no melt spattering occurs. In fact, in-
crease of welding time gives equilibrium to the melt pool. Fig. 17. Nugget formation at different contact resistance.
156 H. Eisazadeh et al. / Materials and Design 31 (2010) 149157

of 1 mm, radial heat loss in the surrounding material is consider-

ably less important than for the thickness of 1.57 mm. This causes
radial growth of nuggets. Also, these results conrm that reducing
the sheet thickness affects the total resistance. By decreasing the
sheet thickness, the total resistance along z-axis decreases; as a re-
sult nugget development along z-axis declines. It can be concluded
that to eliminate this phenomena, electrode diameter should be
used for smaller thin sheets to increase current density.

5. Conclusions

In this study, an incremental and coupled thermal-electro-

mechanical nite element model is presented for predicting tem-
perature distribution and spot nugget size in a spot welded steel
joint. By providing the necessary information and boundary condi-
tions, simulations were performed for three combined analysis
Fig. 18. Nugget formation at different welding force.
during different stages of welding cycle. Experimental data and
solutions of a one-dimensional model developed in a published
article are also utilized for comparison. The results provide infor-
Table 3
Priedicted contact area at three level electrode force.
mation on the development of the weld nugget and thus predict
the welding quality prior to the actual welding process. The input
Electrode force (N) 5500 4670 6500 parameters to the model can be adjusted to give different sizes of
Contact area (mm) 4.78 4.66 4.9
the weld nugget. As a result, optimum setting of the welding
parameters for the desired quality and different materials of the
workpiece can be obtained through simulations, without perform-
electrode force through its effect on contact area. As electrode force ing large number of physical experiments.
is increased, the contact resistance decreases. This is due to the fact The following conclusions can be drawn from this study:
that surface of metals, on a microscopic scale, are a series of peaks
and valleys. When they are subjected to light electrode force, the If the electric current ow exceeds the ow necessary for nug-
actual faying contact will be only at contacting peaks, a small per- get growth, causes a rapid growth of nugget. The nugget growth
centage of the area. As a result, contact resistance will be high. By rate decreases as the current ow increases but the nugget size
decreasing or eliminating load level to gain greater nugget size, the raises until melt spattering occurs. Therefore, space between
expulsion phenomenon occurs. As the electrode force is increased, nuggets affects current ow which means that to eliminate
the high spots are depressed and the actual faying contact area is the current shunting more space between nuggets is necessary.
increased, thus decreasing the contact resistance. Although Increasing electric cycles remarkably raises the contact surface
increasing the electrode force causes a decrease in the faying con- temperature so that the contact zone melts leading to a big nug-
tact resistance, higher electrode force is needed to form a good get but no melt spattering occurs. In fact, increase of welding
nugget. This is because the electrode force keeps the weld intact time gives equilibrium to the melt pool.
until it solidies, cools and the weld nugget reaches its maximum Increase of load on the electrodes decreases the nugget size as it
strength where the most important advantage is preventing occu- raises the contact surface area.
rence of expulsion phenomenon. If the plate thickness is increased, the current ow needed for
The sizes of weld nugget at different sheet thicknesses are plot- the formation of appropriate weld nugget increases. Decreasing
ted in Fig. 19. Three typical sheet thicknesses are considered; the plate thickness lowers the electrode diameter.
1 mm, 1.25 mm, 1.57 mm. Fig. 19 shows that by increasing the
plate thickness weld nugget increases axially and decreases radi-
ally which is due to radial heat loss. Apparently, for the thickness
References

[1] Kraus AD, Bar-Cohen A. Thermal analysis and control of electronic

equipment. New York (NY): McGraw-Hill; 1983.
[2] Incropera FP, DeWitt DP. Introduction to heat transfer. 2nd ed. New York: John
Wiley & Sons; 1990.
[3] Timoshenko SP, Goodier JN. Theory of elasticity. 3rd ed. McGraw-Hill; 1970.
[4] ASM handbook (formerly 10th ed. and metals handbook). Properties and
selection: nonferrous alloys and special-purpose materials, vol. 2. ASM
International; 1990. p. 1920.
[5] Gould JE. An examination of nugget development during spot welding using
both experimental and analytical techniques. Weld J 1994;66(1):1s10s.
[6] Tsai CL, Jammal OA, Dickinson DW. Modeling of resistance spot weld nugget
growth. Weld J 1992;71(2):47s54s.
[7] Tsai CL, Jammal OA, Dickinson DW. Study of nugget formation in resistance
spot welding using nite element method. In: Paper presented at the trends in
welding research, 2nd international conference, Materials Park, OH, USA; 1989.
[8] Khan JA, Xu L, Chao Y, Broach K. Numerical simulation of resistance spot
welding process. Numer Heat Transfer, Part A 2000;37:42546.
[9] Richard D, Fafard M, Lacroix R, Clery P, Maltais Y. Carbon to cast iron electrical
contact resistance constitutive model for nite element analysis. J. Mater
Process Technol 2003;132:11931.
[10] Chang BH, Zhou Y. Numerical study on the effect of electrode force in small-
Fig. 19. Nugget formation at different sheet thickness. scale resistance spot welding. J Mater Process Technol 2003;139(13):63541.
 H. Eisazadeh et al. / Materials and Design 31 (2010) 149157 157

[11] Feulvarch E, Robin V, Bergheau JM. Resistance spot welding simulation: a [15] Loulou T, Bardon JP. Thermal characterization of resistance spot welding
general nite element formulation of electrothermal contact conditions. J resistance spot welding. Numer Heat Transfer, Part B 2006;49:55984.
Mater Process Technol 2004;153154:43641. [16] Li BY, Lin ZQ. Numerical analysis of magnetic uid dynamics behaviors during
[12] Feulvarch E, Robin V, Bergheau JM. Resistance spot welding process: resistance spot welding. J Appl Phys 2007;101:053506.
experimental and numerical modeling of weld growth mechanisms with [17] Aslanlar S, Ogur A, Ozsarac U, Ilhan E. Welding time effect on mechanical
consideration of contact conditions. Numer Heat Transfer, Part A properties of automotive sheets in electrical resistance spot welding. Mater
2006;49:34567. Des 2008;29:142731.
[13] Hou Z, Kim I. Finite element analysis for the mechanical features of resistance [18] Rogeon P, Carre P, Costa J, Sibilia G, Saindrenan G. Characterization of electrical
spot welding process. J Mater Process Technol 2007;180:1605. contact conditions in spot welding assemblies. J Mater Process Technol
[14] Loulou T, Bardon JP. Estimation of thermal contact conductance during 2008;195:11724.
resistance spot welding. Exp Heat Transfer 2001;14:25164.

View publication stats