materials

Article
Modeling and Experimental Research on Resistance
Spot Welded Joints for Dual-Phase Steel
Dawei Zhao 1, *, Yuanxun Wang 2 , Peng Zhang 2 and Dongjie Liang 3
1 School of Mechanical and Vehicle Engineering, Linyi University, Linyi 276005, China
2 Department of Mechanics, Huazhong University of Science and Technology, Wuhan 430074, China;
wooshin@163.com (Y.W.); ldj_hust@126.com (P.Z.)
3 Guangxi Zhuang Autonomous Region Institute of Metrology & Test, Nanning 530007, China;
dj75037882@163.com
* Correspondence: zhaodawei0322@163.com; Tel.: +86-183-0659-3712




Received: 13 March 2019; Accepted: 1 April 2019; Published: 3 April 2019


Abstract: Dual-phase steel has been employed in the automotive industry as it has the advantages of
high strength, satisfying ductility, low yield ratio, and so on. A novel framework for the weld nugget
size prediction and control using finite element modeling and experimental research was proposed
in this paper. The two-dimensional axisymmetric numerical analysis model was established and the
phase transition on thermal expansion coefficient was taken into account. The whole welding process
was simulated and discussed using thermal elastic-plastic theory. To validate the predictive methods
of developed weld nugget size and confirmation experiments were implemented with the same
input parameters in the ranges of process parameters. The simulated weld nugget sizes were in good
agreement with the experimental results except for extreme welding conditions. The microstructure
of the welding zone was also investigated based on metallographic experiments and temperature
field analysis. The welding parameters were adjusted using the model proposed in this paper so as
to obtain the nugget size with pull out failure mode.

Keywords: advanced high-strength steel; resistance spot welding; simulation analysis; nugget diameter

1. Introduction
Currently, a great deal of attention has been focused on the automotive industry to reduce fuel
consumption and gas emission by means of reducing the weight of vehicles. One such initiative
is the development of welding joints with high-strength steels. As one kind of high-strength steel,
dual-phase steels (DP500-1000) possess the advantages of high strength, good corrosion resistance,
satisfying press performance, and so on, and have been gradually employed in modern car body
manufacturing [1,2]. Resistance spot welding (RSW) is the main joining technique widely employed
in car body manufacturing. It is a multifaceted process involving the metallurgy, mechanics,
thermoelectricity, fluid-structure interaction, and many other coupling factors [3,4]. In recent years,
a great amount of research work on the resistance spot welded joints of DP steel was available.
Hayat et al. [5] evaluated the fracture toughness of galvanized DP600 steel sheets under RSW against
the welding current and welding time with the thickness of 1.20 mm. It was observed from the
results that the relationship between fracture toughness and welding current, welding time was
a second-degree polynomial; fracture toughness of spot welds was not only dependent on the
nugget diameter but also relied on sheet thickness, tensile rupture force, welding time, and current.
Kaščák et al. [6] investigated the influence of the primary welding parameters on the weld quality by
testing microhardness and tensile shear load bearing capacity of resistance spot welds of DP600 steel
sheets. Huin et al. [7] characterized the fracture mechanisms for the RSW joints of two common steel

Materials 2019, 12, 1108; doi:10.3390/ma12071108 www.mdpi.com/journal/materials

Materials 2019, 12, 1108 2 of 16

grades DP600 and martensitic steel Usibor 1500. Cross tension and tensile shear tests were carried out,
while the resulting strengths, microstructures, failure modes, and fracture mechanisms were compared.
It was found that the failure load of heterogeneous spot welds increased with the nugget size and the
sheet thickness regardless of the failure mode, as long as no splash defects occurred. Aktas et al. [8]
studied the effect of welding current and welding time on mechanical properties of DP600 steel sheet
joints to detect the optimum welding parameters for maximum joint strength. Kim et al. [9] used a
conically shaped hollow electrode for the lower electrode to improve resistance spot weldability of
three-sheet welding, including high strength steel. Shear strength test, analysis of welding signal,
and weld cross-section were conducted. It was found that the proposed lower electrode had some
advantages in welding the three-sheet welding compared with conventional electrodes or hollow
electrodes. Hernández et al. [10] assessed the influence of polarity during resistance spot weld of
dissimilar lap joints between DP 600 dual-phase steel and AISI 304 stainless steel on mechanical
properties and failure modes. Jaber et al. [11] investigated the microstructure, failure mode transition,
peak load, and energy absorption of spot welded DP600 dual-phase steel during the tensile-shear test.
It was found that the welding current has profound effect on the load-displacement characteristics.
In fact, the weld joints are accomplished by the effects of a series of welding process parameters,
which mainly includes the welding current, electrode force, welding time, and so on [12–14].
The mechanical property of the weld assembly is basically dependent on the integrality of each
welding joints. During the welding process, deformation of the workpieces, the stress and strain in
the welding zone will be produced and changed. These mechanical properties have great influences
on the performances of welding joints, including nugget formation and its failure strength [15].
Consequently, to understand the temperature distribution in the welding process, the deformation
and phase transformation of the microstructure is extremely important. Nevertheless, the process
of spot welding is such a highly non-linear, multi-parametered, and complicated process containing
a large number of random factors. Only through experimental study can we obtain insightful and
knowledgeable information about welding process [16]. Combining the finite element numerical
simulation with experimental research provides a very valuable method for welding process
parameters recommendation to get sound welding joints in actual production.
The main objective of this study is to develop a two-dimensional axisymmetrical thermal-electrical
finite element model to simulate the welding process for DP600 steel sheets using the finite element
software Ansys (Canonsburg, PA, USA). The simulation results were compared with the test results to
verify the validity of the model. The microstructure of the welding zone was also investigated based
on metallographic experiments and temperature field analysis of the thermal-electrical-mechanical
finite element model. The appropriate welding parameter combinations guaranteeing the welding
quality were finally proposed. The numerical simulation can predict the nugget diameter with the
changes of process parameters in the welding process rather than based on experiences without theory,
which also reduces the expensive experimental cost and provides guidance for further research in the
actual production.

2. Materials and Methods

The material investigated was 1.7 mm thick ferrite-martensitic dual phase steel DP600. The typical
chemical composition of DP600 steel (wt%) is C 0.11, Mn 1.6, Si 0.182, Ni 0.027, Nb 0.0037, Ti 0.002, Cr
0.34, Mo 0.098, and V 0.0035. The basic mechanical properties of DP600 steel [17] are shown in Table 1.
The welding samples for the experiments were made by shearing form the steel sheet plates against
the direction of rolling; the dimensions were 40 × 120 mm and with overlapping 40 mm. Then the
samples were cleaned by acetone and the welding tests were performed using pneumatic spot welding
machine (Panasonic Welding Systems Company, Tangshan, China). Two-pulse alternating welding
current was employed to produce the welded joints; the frequency of alternating current was 50 Hz.
The copper alloyed truncated electrode was used and its diameter was 6 mm. The sizes of steel sheets
for lap welding were 120 mm × 40 mm and the overlapped length was 40 mm. The welding process
Materials 2019, 12, 1108 3 of 16

parameters were determined on the basis of literature review and preliminary welding tests [18].
The holding time and squeeze time were kept at 10 cycles (1 cycle = 0.02 s), the welding process
parameters employed in the welding experiments were listed in Table 2. The electronic universal
testing machine was used to conduct quasi-static tensile shear tests for the welding joints. The loading
speed was 10 mm/s. The mechanical properties of the specimens were obtained according to the
load-displacement curves. Peak load is the maximum value of load-displacement curve, while the
maximum displacement is its corresponding displacement value. The total failure energy is the
area under load-displacement curve [19], while the penetration rate can be obtained according to
Zhao et al. [20]. The failure modes of samples were determined from the rupture interfaces of the
samples after the tensile shear tests. Metallographic cross section for each weld was polished by
automatic grinding and polishing machine. Abrasive cloths with different size of particles were used
for the polishing. After polishing, they were etched using natal for nugget size measurement and
microstructure observation to judge the welding quality. The Axiovert 200 Microscope produced by
Carl Zeiss (Heidenheim, German) was used for metallographic observations.

Table 1. Mechanical properties of DP600 steel.

Specified Percentage
Plastic Strain Strain Hardening
Tensile Strength Non-Proportional Elongation after
Ratio Exponent
Elongation Stress Fracture
Rm Rp0.2 A80 r n
MPa MPa % % %
633 407 23 67 14

Table 2. Welding parameters employed during the RSW process.

Welding Parameters Symbol Unit Value

Electrode force F kN 2.0, 2.5, 3.5, 4.0
Welding time T cycles 8, 11, 14, 17
Welding current I kA 6, 8, 10, 12

During the finite element analysis process, the electrode force was first applied to the steel
sheets and it resulted in initial deformations and contact area. Then the temperature distribution was
considered for an increment from fully coupled electrical thermal finite element analysis. The electrical
and thermal boundary conditions were also considered in the model. The Joule heating resulted
from the welding current at the faying surfaces of DP600 steel sheets and electrode were calculated.
Consequently, the temperature distributions in the welding zone would be achieved and the final
results were obtained using the mechanical analysis after several loops. Therefore, the heat flow
problem of resistance spot welding is a heat conduction related to melting and subsequent solidification.
Although the nature of the heat flow is three-dimensional, it can also be simulated as the form of the
two-dimensional axisymmetric model, because the circular section of the electrode can be applied on
both current and compressive force at the same time. The advantage of the axisymmetric formula
is that its symmetry θ, the azimuth angle, allows for heat transfer and flow changes in radial (r)
and vertical (z) directions [21]. Therefore, in current work, thermal analysis is considered to be an
axisymmetric transient heat conduction.
Supposing the temperature is constant, the control equation of transient heat conduction in
cylindrical coordinate system can be expressed as [22]:
   
∂ ∂T γ ∂T ∂ ∂T ∂T
γ· + · + γ· + qv = C p ·ρ (1)
∂r ∂r r ∂r ∂z ∂z ∂t
 Materials 2019, 12, 1108 4 of 16

Materials 2019, 12, x FOR PEER REVIEW 4 of 16

Among them, z is the axial coordinate of the cylindrical coordinate system, and r is the radial
coordinate. qv , t, T, γ, ρ and Cp respectively indicate heat yield, time, temperature, thermal conductivity,
The steady-state current conduction equation in the sheet electrode system can be obtained by
density, and specific heat, whose units are all international units, and all of them are functions
in theofaspect of electrical voltage [21]:
temperature.
The steady-state current 𝜕 1 𝜕𝑈equation1 in the
conduction 𝜕𝑈sheet𝜕electrode
1 𝜕𝑈 system can be obtained by in
the aspect of electrical voltage [21]: ∙ + ∙ + ∙ =0 (2)
𝜕𝑟 𝜌 𝜕𝑟 𝜌 ∙ 𝑟 𝜕𝑟 𝜕𝑟 𝜌 𝜕𝑧
   
where 𝜌 is the resistivity of∂the1material, ·
∂U 1 ∂U
+ U is · the
∂ 1 ∂U
+ electrical
· potential.
=0 (2)
∂r ρ E ∂r ρ E ·r ∂r ∂r ρ E ∂z
A two-dimensional axisymmetric model was established as the thermoelectrical analysis
wheremodel
geometry ρ E is thetoresistivity
analyzeofthe material, U
thewelding is the electrical
process potential.
using the software Ansys, as shown in Figure 1. The
A two-dimensional axisymmetric model was established as the thermoelectrical analysis geometry
radius of the model was 20 mm and the thickness of model was 3.4 mm, which was two times of the
model to analyze the welding process using the software Ansys, as shown in Figure 1. The radius of
DP600 sheets thickness. Thermoelectric coupling elements were employed to mesh the model and
the model was 20 mm and the thickness of model was 3.4 mm, which was two times of the DP600
solid sheets
elements wereThermoelectric
thickness. utilized to simulate the thermo-elastic-plastic
coupling elements were employed to mesh behavior of and
the model the solid
faying surfaces
between the sheets
elements and electrodes.
were utilized to simulate theThe thickness of the solid
thermo-elastic-plastic element
behavior at thesurfaces
of the faying contactbetween
of the sheet-sheet
thewas
surface sheets and electrodes.
0.035 mm and the The one
thickness of sheet-electrode
at the the solid element atsurface
the contact
wasof0.07
the sheet-sheet
mm. surface
was 0.035 mm and the one at the sheet-electrode surface was 0.07 mm.

Figure 1. Finite element model employed to simulate RSW process: (a) overall view; and (b) enlarged view.
Figure 1. Finite element model employed to simulate RSW process: (a) overall view; and (b) enlarged
view.The solid element Plane42 was selected in the squeeze stage. The contact surfaces between
the electrode/workpiece and workpiece/workpiece with the action of the electrode force were
flexible
The contacts,
solid and Plane42
element the deformation and movement
was selected in theofsqueeze
the material were
stage. tracked
The by the
contact elementsbetween the
surfaces
TARGE169/CONTA172. The thermoelectric element Plane67 was selected for the electrical-thermal
electrode/workpiece and workpiece/workpiece with the action of the electrode force were flexible
coupled analysis. The contact resistance is represented by a physical unit attached onto the surfaces
contacts, and theand
of workpieces deformation
electrodes byand movement
artificially adding of the ofmaterial
a layer were
1/20 of the trackedthickness.
workpiece by the elements
TARGE169/CONTA172.
The structural element The thermoelectric
Plane42 was selected element Plane67
in the thermal wasstage.
analysis selected for same
As the the electrical-thermal
in the
coupled analysis.
squeeze phase,The contact resistance
the elements is represented
TARGE169/CONTA172 wereby a physical
employed unit attached
to monitor onto the surfaces
the deformation
and movement
of workpieces andofelectrodes
the contact surface.
by artificially adding a layer of 1/20 of the workpiece thickness. The
The cooling water temperature inside the electrodes was assumed to be 10 ◦ C and the ambient
structural element Plane42 was selected in the thermal analysis stage. As the same in the squeeze
temperature was 20 ◦ C. As the temperature was above 1500 ◦ C, DP600 sheet began to melt and this
phase, the value
critical elements TARGE169/CONTA172
was thought were employed
to be the threshold temperature to monitor
for nugget formation. the deformation
The surfaces of the and
movement
electrodeofand
thethin
contact
sheetssurface.
pass heat to the air by convection. The convective heat transfer coefficient for
The cooling
the cool water water temperature
in the electrodes × 104 the
inside
is 4.187 W/m 2 ·K. The surface heat dissipation coefficients for
electrodes was assumed to be 10 °C and the ambient
electrode was
temperature and metal
20 °C.sheets are presented
As the temperature in Figure
was2. above 1500 °C, DP600 sheet began to melt and this
critical value was thought to be the threshold temperature for nugget formation. The surfaces of the
electrode and thin sheets pass heat to the air by convection. The convective heat transfer coefficient
for the cool water in the electrodes is 4.187 × 104 W/m2·K. The surface heat dissipation coefficients for
electrode and metal sheets are presented in Figure 2.
 2

where n is the number of iterations, Tn is the final temperature, H(Tn) is the enthalpy as the
corresponding temperature is Tn, Ci is the initial specific heat at the ith interval, Ci+1 is the final specific
heat at the ith interval, ΔTi is the interval of temperature. In this case, the enthalpy of steel sheets with
different
Materials temperatures
2019, 12, 1108 can be solved and it is presented in Figure 3. 5 of 16

Materials 2019, 12, x FOR PEER REVIEW 5 of 16

where ρ is the density of welding joints, C(T) is the specific heat of weldment. According to the
definition of definite integral, Equation (4) can be written as follows:

𝐶 +𝐶
Figure2.2. Surface 𝐻 𝑇 =of
Surface coefficients of𝜌heat
heattransfer ∆𝑇 (4)
Figure coefficients 2 of
transfer ofelectrodes
electrodesand
andsteel.
steel.

Additionally,
where one of the
n is the number most remarkable
of iterations, Tn is thecharacteristic of the welding
final temperature, process
H(Tn) is the is thatasthe
enthalpy the
coexistence of solid and liquid phase, solid state phase transition and solid-liquid phase transition in
corresponding temperature is Tn, Ci is the initial specific heat at the ith interval, Ci+1 is the final specific
the welding zone, which is supposed to produce an effect on the temperature field. Consequently,
heat at the ith interval, ΔTi is the interval of temperature. In this case, the enthalpy of steel sheets with
enthalpy
differentwas employedcan
temperatures to solve the problem
be solved and it is of latent heat
presented during3.the thermoelectric analysis and
in Figure
modeling the welding process [21]. Enthalpy is defined as:
Z
H= ρC ( T )dT (3)

where ρ is the density of welding joints, C(T) is the specific heat of weldment. According to the
definition of definiteFigure
integral, Equation
3. Enthalpy (4) can
values be written
of steel as follows:
sheets with different temperatures.

n
The thermal, mechanical, and electrical C + Cofi+the
H ( Tn ) = properties
ρ∑ i 1
∆Tbase
i
material and electrode had to(4) be
available so as to obtain accurate results [23]. As i =the
1
2
electrical and physical values are affected by the
temperature, they are commonly acquired from literature review and supposed to be homogeneous.
where n is the
Mechanical numberofof
properties electrode Tand
theiterations, n issteel
the sheets
final temperature, H(Tn ) is were
at room temperature the enthalpy as the3
listed in Table
corresponding temperature is T n , C i is the initial specific heat at the i th interval,
[24,25]. The thermo-electrical properties parameters of the steels and electrode are presented in Figures C i+1 is the final
specific heat at the i th
4–7 [26–28]. The propertiesinterval, ∆T
of DP600 is the interval of temperature. In this case, the
i with very high temperature were employed according to Bézi et enthalpy of steel
sheets with different Figure 2. Surface
temperatures coefficients
can be solved ofand
heatittransfer of electrodes
is presented and steel.
in Figure 3.
al. [29].

Table 3. Mechanical properties of electrode and DP600 at room temperature.

Yield Elasticity Strain Hardening Poisson’s

Materials Density
Strength Modulus Modulus Ratio
ρ
Symbol σs (MPa) E (GPa) Et (MPa) ν
(kg/m3)
Electrode 230 115 566 0.35 8900
DP600 360 213 611 0.3 7800

Figure3.3.Enthalpy
Figure Enthalpyvalues
valuesofofsteel
steelsheets
sheetswith
withdifferent
differenttemperatures.
temperatures.

The
Thethermal,
thermal, mechanical, andelectrical
mechanical, and electricalproperties
propertiesof ofthethe
basebase material
material andand electrode
electrode had had
to be
toavailable
be available soobtain
so as to as to accurate
obtain accurate results
results [23]. [23].
As the As the
electrical andelectrical
physicaland physical
values valuesbyare
are affected the
affected by thethey
temperature, temperature, they are
are commonly commonly
acquired fromacquired
literaturefrom
reviewliterature review and
and supposed supposed
to be to be
homogeneous.
homogeneous. Mechanical
Mechanical properties of properties of the
the electrode andelectrode and steel
steel sheets sheets
at room at room temperature
temperature were listedwere listed3
in Table
in[24,25].
Table 3The[24,25]. The thermo-electrical
thermo-electrical propertiesproperties
parameters parameters of the
of the steels andsteels and electrode
electrode are presented
are presented in Figures
in4–7
Figures
[26–28].4–7The
[26–28]. The properties
properties of DP600 of DP600
with verywith
highvery high temperature
temperature were employed
were employed accordingaccording
to Bézi et
toal.Bézi
[29].et al. [29].

Table 3. Mechanical properties of electrode and DP600 at room temperature.

Yield Elasticity Strain Hardening Poisson’s

Materials Density
Strength Modulus Modulus Ratio
ρ
Symbol σs (MPa) E (GPa) Et (MPa) ν
(kg/m3)
Electrode 230 115 566 0.35 8900
Materials 2019, 12, 1108 6 of 16

Table 3. Mechanical properties of electrode and DP600 at room temperature.

Yield Elasticity Strain Hardening Poisson’s

Materials Density
Strength Modulus Modulus Ratio
Symbol
Materials (MPa)REVIEW E (GPa)
σs PEER
2019, 12, x FOR Et (MPa) ν ρ (kg/m3 ) 6 of 16
Electrode
Materials 230 REVIEW
2019, 12, x FOR PEER 115 566 0.35 8900 6 of 16
DP600
Materials 360 REVIEW
2019, 12, x FOR PEER 213 611 0.3 7800 6 of 16

Figure 4. Thermal conductivity of the electrode (a) and DP600 (b).

Figure 4. Thermal conductivity of the electrode (a) and DP600 (b).
Figure 4. Thermal
Figure conductivity
4. Thermal of the
conductivity electrode
of the (a) (a)
electrode andand
DP600 (b).(b).
DP600

Figure 5. Resistivity of the electrode (a) and DP600 (b).

Figure 5. Resistivity of the electrode (a) and DP600 (b).
Figure 5. Resistivity of the electrode (a) and DP600 (b).
Figure 5. Resistivity of the electrode (a) and DP600 (b).

Figure 6. Specific heat of the electrode (a) and DP600 (b).

Figure 6. Specific heat of the electrode (a) and DP600 (b).
Figure 6. Specific heat of the electrode (a) and DP600 (b).
Figure 6. Specific heat of the electrode (a) and DP600 (b).
 Materials 2019, 12, x FOR PEER REVIEW 7 of 16

Materials 2019, 12, 1108 7 of 16

Materials 2019, 12, x FOR PEER REVIEW 7 of 16

Figure 7. Thermal expansion coefficient of the electrode (a) and DP600 (b).

The contactFigure resistance

Figure is another
7. Thermal
7. Thermal expansion factor
expansion affecting
coefficient
coefficientofthe
the Joule heat
electrode
of the electrode(a)in
and
(a)the welding
DP600
and DP600 process and it is
(b).(b).
normally obtained through measuring in experiment. The initial value of contact resistance at room
The contact resistance is another factor affecting the Joule heat in the welding process and it is
temperature can beresistance
The contact worked is out via Equation
another (5) [30].the
factor affecting InJoule
mostheatcases,
in theas welding
for a specificprocess welding
and it is
normally obtained through measuring in experiment. The initial value of contact resistance at room
condition,
normallythe electrode
obtained force ismeasuring
through set as a constant value, the
in experiment. Thecontact
initialresistance can be simplified
value of contact resistance to the
at room
temperature
function of can be workedas
temperature, outshown
via Equationin (5) [30].(6)
Equation In [31,32].
most cases,It as for
can be atranslated
specific welding
into condition,
the contact
temperature can be worked out via Equation (5) [30]. In most cases, as for a specific welding
the electrode
resistivity byforce
means is set as a constant
of Equation value, the contact resistance can be simplified to the function of
condition, the electrode force is(7):set as a constant value, the contact resistance can be simplified to the
temperature, as shown in Equation (6) [31,32]. It can be translated into the contact resistivity by means
function of temperature, as shown in Equation (6) [31,32]. It can be translated into the contact
of Equation (7): 𝑅 = 𝑟𝐹 (5)
resistivity by means of Equation (7): −m
Rc = rF (5)
𝑇 −𝑇
𝑅 =𝑅 Tm = − 𝑟𝐹 T ∙𝑅 (6)(5)
R T = 𝑇 − 𝑇 · Rc (6)
Tm 𝑇− T−0 𝑇
𝑅 = R𝑅 ·∙S𝑆 ∙ 𝑅 (6)
ω𝜔== 𝑇 T𝐿− 𝑇 (7)
(7)
L
wherer risis
where thethe coefficient
coefficient indicating
indicating the nature
the nature and𝜔theand 𝑅 the∙ 𝑆 surface state of DP600 steel sheets, it is
= surface state of DP600 steel sheets, it is generally(7)
generally acknowledged
acknowledged the value is 1.375 × 10−is2 between
the value 1.375 × 10 copper 𝐿and steel
−2 between copper and steel
(electrode (electrode
and steel contactand steel
surface),
contact
where ×r 10 −
2.75 surface), 2
is the and 2.75steels
coefficient× 10indicating
−2 between the steels (workpiece
nature and theinterfaces) at room
surface state temperature
of DP600 [30]. Rc it
is is
and between (workpiece interfaces) at room temperature [30]. R c is thesteel sheets,
initial contact
thegenerally
initial at
resistance contact
room resistance
temperature, at room
m is temperature,
an index, the m is
value
−2 an index,
acknowledged the value is 1.375 × 10 between copper and steel (electrode and steel
of which theis value of
frequently whichin is
the frequently
range of in
0.5
the1,
to range
contact of 0.5 ofto
surface),
the value 1,
is the
mand value
2.75
set 10of
as ×0.7 m is paper.
−2 between
in this set as 0.7
F isin
steels thiselectrode
(workpiece
the paper. Fforce,
is theTelectrode
interfaces) atisroom
m melting force, Tm is melting
temperature
temperature [30].ofRc is
temperature
the initial
DP600 steel and of DP600
contact steel
resistance
its value canand atits
be setvalue
room can ◦be
C, set
as temperature,
1500 T0 as 1500
is an°C,
ismroom T0 isthe
index,
temperatureroomvalue temperature
whichof which
set as is20 ◦ C, R
which set
frequently T as
is in
20 °C,
contact R
the range is contact
of 0.5as
resistance
T resistance
tothe
1, the as the temperature
value of misisT,set
temperature is T,
ascross-sectional
S is S is
0.7 in this paper. cross-sectional
areaF of
is the area
electrode
welding of
current welding
force, Tm L
path, current
is is
melting
the
path,
length Lof
is contact
the length
temperature ofarea,of ω
DP600 contact
steel
is area,
theand ω value
its
resistivity. is theHence,
resistivity.
can bethe Hence,
1500 the
set reference
as °C, reference
T0 is room
contact contact resistivity
temperature
resistivity values which
at values
various set as
at 20
various
°C, RTtemperatures
temperatures iscancontact can be as
resistance
be determined determined
by the
linear by linear
temperature
interpolation is T,interpolation
S isand
[33] [33] and
cross-sectional
the specific theofspecific
area
contact welding contact
resistances current
of
resistances
path, L is
interfaces of
of the interfaces
length of and
sheet/sheet of sheet/sheet
contact and
area, ω is theare
electrode/sheet electrode/sheet
resistivity.
shown inHence, are shown
Figurethe in Figure 8.
8. reference contact resistivity values
at various temperatures can be determined by linear interpolation [33] and the specific contact
resistances of interfaces of sheet/sheet and electrode/sheet are shown in Figure 8.

Figure8.8.Contact
Figure Contactresistivity
resistivity(F
(F==3.5
3.5kN).
kN).

The
The main
main process
process parameters determining
parametersFigure
determining the quality of
of=spot welding
welding are
are electrode
electrode pressure,
8. Contactthe quality(F
resistivity spot
3.5 kN). pressure,
welding
welding current and welding time. As the simple sinusoidal alternating current was employedin
current and welding time. As the simple sinusoidal alternating current was employed inthe
the
welding experiments and the frequency is 50 Hz. During the cooling stage in the welding process, the
The main process parameters determining the quality of spot welding are electrode pressure,
welding current and welding time. As the simple sinusoidal alternating current was employed in the
 Materials 2019, 12, x FOR PEER REVIEW 8 of 16

Materials 2019, 12, x FOR PEER REVIEW 8 of 16

welding
Materials experiments
2019, 12, 1108 and the frequency is 50 Hz. During the cooling stage in the welding process, 8 of 16
the welding
welding current and
experiments was the
cut frequency
off. Thus, isthe
50 welding
Hz. During current in the finite
the cooling stage element analysis
in the welding can be
process,
expressed
the welding as follows
current (shown
wasoff. in
cutThus,Figure
off. the
Thus,9):
the welding
welding current was cut welding current current in the
in the finite finite analysis
element elementcan
analysis can be
be expressed
expressed as follows (shown
as follows (shown in Figure 9): in Figure 9): 𝐼 𝑠𝑖𝑛 100𝜋𝑡 , 0 ≤ 𝑡 ≤ 𝑡
𝐼= (8)
𝐼 𝑠𝑖𝑛 100𝜋𝑡0, 𝑡 >
, 0𝑡≤ 𝑡 ≤ 𝑡
𝐼=( (8)
0, 𝑡 >0𝑡≤ t ≤ tm
Im sin100πt,
I= (8)
0, t > tm

Figure 9. The waveform of welding current.

Figure 9. The waveform of welding current.
Figure 9. The waveform of welding current.
3. Results
3. Results
3. Results
Figure 10 shows the contact pressure of the electrode/sheet and sheet/sheet contact surfaces as
Figure 10 shows the contact pressure of the electrode/sheet and sheet/sheet contact surfaces as
the electrode force
Figure 10forceshows is the
3.5 contact
kN in the squeeze stage. The holding time andisis set as 10 cycles. The pressure
the electrode is 3.5 kN in thepressure
squeezeof the electrode/sheet
stage. The holding time sheet/sheet contact
set as 10 cycles. Thesurfaces
pressureas
stress
the is unevenly
electrode force distributed
is 3.5 kN in along
the the
squeeze contact
stage. radius,
The which
holding timeis caused
is set asby10its uneven
cycles. The axial stress
pressure
stress is unevenly distributed along the contact radius, which is caused by its uneven axial stress
distribution.
stress As for the electrode
is unevenly force and DP600 steel sheet, the contactby pressure distributes more
distribution. As fordistributed
the electrode along
forcethe
and contact
DP600radius, whichthe
steel sheet, is contact
caused its uneven
pressure axial stress
distributes more
smoothly
distribution. near the
Asthe center
for center of
the electrode electrodes,
force and and then
DP600 it increases rapidly until stress concentration is
smoothly near of electrodes, and then steel sheet, the
it increases contact
rapidly pressure
until stress distributes
concentration more is
observed near
smoothly on edgetheofof the electrodes.
center of electrodes,Theand contact
then pressure rapidly reduces
it increases tostress
zero once it is beyond
observed on edge the electrodes. The contact pressure rapidlyrapidly
reducesuntilto zero onceconcentration
it is beyond the is
the electrode
observed edge. edge.
on edge The conductive area of the contact point between the electrode and steel sheet is
electrode Theofconductive
the electrodes. The
area of the contact
contact pressure rapidlythe
point between reduces to zero
electrode andonce
steelitsheet
is beyond
is the
the same
the electrodeas the electrode
edge. The face area.
conductive The pressure stress of the contact surface between the upper steel
same as the electrode face area. The area of the
pressure contact
stress point
of the between
contact surface thebetween
electrode theand steelsteel
upper sheet
andis
andsame
the lower as steelelectrode
the slowly declines
face area.until
The itpressure
reachesstress
zero, ofthethevirtual
contact contact
surface radius,
between which
the is about
upper 4.6
steel
lower steel slowly declines until it reaches zero, the virtual contact radius, which is about 4.6 mm. It is
mm.lower
and It is much larger than the electrode radius. zero, the virtual contact radius, which is about 4.6
much largersteel
thanslowly declines
the electrode until it reaches
radius.
mm. It is much larger than the electrode radius.

Figure 10.Contact
Figure10. Contactpressure
pressuredistribution
distributionofofthe
thewelding
weldingjoints
jointsininthe
thesqueeze
squeezestage.
stage.
Figure 10. Contact pressure distribution of the welding joints in the squeeze stage.
Figure
Figure1111
shows
showsthethe
temperature
temperature distributions in different
distributions welding
in different time calculated
welding in the welding
time calculated in the
process.
welding The corresponding
process. The welding
corresponding current
welding is 10 kA,
current the
is electrode
10 kA, the force is
electrode3.5 kN,
force andthe
is 3.5 kN,welding
andthe
Figure 11 shows the temperature distributions in different welding time calculated in the
time is 14time
welding cycles (0.28
is 14 s). (0.28
cycles Froms).Figure
From 11a, at11a,
Figure the at
initial
the stage,stage,
initial the temperature
the temperatureat the surfaces
at kN, of
the surfaces
welding process. The corresponding welding current is 10 kA, the electrode force is 3.5 andthe
sheet/sheet and
of sheet/sheet electrode/sheet is much higher than other places, the contact resistance is the main
welding time isand electrode/sheet
14 cycles is much
(0.28 s). From higher
Figure 11a, than
at theother
initialplaces,
stage,the
thecontact resistance
temperature is surfaces
at the the main
factor
factor generated
generated welding
welding heat
heat and
and its
itsvalue
value isisthe
the largest
largest atatthe
theinterfaces
interfacesofofsheet/sheet
sheet/sheet (Figure
(Figure 8).
8).
of sheet/sheet and electrode/sheet is much higher than ◦ other places, the contact resistance is the main
The highest temperature at this stage is about 450 C at this time. As the carbon content of DP600
factor generated welding heat and its value is the largest at the interfaces of sheet/sheet (Figure 8).
Materials 2019, 12, x FOR PEER REVIEW 9 of 16

The highest temperature at this stage is about 450 °C at this time. As the carbon content of DP600
steel is about 0.11%, the temperature of austenitic initialization Ac1 is 717 °C, and the temperature of
Materials 2019, 12, 1108 9 of 16
complete transition into austenitic Ac3 is 854 °C [33,34]. The microstructure of the welding joint at this
stage is the same as DP600 steel since its temperate was below Ac1. When the welding time is 0.05 s,
steel is about 0.11%, the temperature of austenitic initialization Ac1 is 717 ◦ C, and the temperature of
the temperature at the interfaces of electrode/sheet is much smaller than that of sheet/sheet. From
complete transition into austenitic Ac3 is 854 ◦ C [33,34]. The microstructure of the welding joint at this
now on,stage
the dominant factor of the influence on welding heat is the bulk resistance of the upper and
is the same as DP600 steel since its temperate was below Ac1 . When the welding time is 0.05 s,
lower sheets, and the at
the temperature electrical resistivity
the interfaces of DP600 issteel
of electrode/sheet much issmaller
much thanlarger
thatthan copper alloy
of sheet/sheet. Fromelectrode.
At this stage, the largest temperature at the faying surfaces is about 1220 °C and it is lower
now on, the dominant factor of the influence on welding heat is the bulk resistance of the upper and than the
meltinglower
pointsheets,
1500and °C,the electrical
while theresistivity
two sheets of DP600
are steel is muchadhesion
in plastic larger thanstate
copperwith
alloythe
electrode.
action of the
At this stage, the largest temperature at the faying surfaces is about 1220 ◦ C and it is lower than the
electrode pressure imposing on them. Then, as time goes on and the application of welding current
melting point 1500 ◦ C, while the two sheets are in plastic adhesion state with the action of the electrode
continues, the temperature at the faying surfaces of the welding zone is above 1500 °C, molten metal
pressure imposing on them. Then, as time goes on and the application of welding current continues,
appearstheand graduallyatenlarges
temperature the fayinguntil
surfacestheofwelding
the welding current
zone isisabove
cut off.
1500The temperature
◦ C, molten of most of the
metal appears
weldingand zone is above
gradually Ac3until
enlarges andthe the austenitic
welding currentisis filled
cut off. with this region.
The temperature Theoftemperature
of most the welding of the
zone is above A
electrode surface is smaller than 500 °C and, meanwhile, the shape of molten metal issurface
c3 and the austenitic is filled with this region. The temperature of the electrode oval, which is
is smaller than 500 ◦ C and, meanwhile, the shape of molten metal is oval, which is caused by the
caused by the cooling water in the electrodes. The heat affected zone is usually identified as the zone
cooling water in the electrodes. The heat affected zone is usually identified as the zone of austenitic
of austenitic initialization contained. The region covered by the contact radius of the electrode and
initialization contained. The region covered by the contact radius of the electrode and DP600 sheet is
DP600 sheet
the heatis affected
the heatzone. affected zone.

Figure 11. Temperature profile of the welding zone in the welding process (a) 0.007 s, (b) 0.05 s,
Figure 11. Temperature profile of the welding zone in the welding process (a) 0.007 s, (b) 0.05 s, (c)
(c) 0.13 s, and (d) 0.28 s.
0.13 s, and (d) 0.28 s.
Figure 12 shows the key point selected in the welding zone in order to investigate the cooling rate
Figure
of the12 showsjoints
welding the in
key
thepoint
cooling selected in the
stage after thewelding
welding zoneis in
current cutorder to investigate
off. Figure the cooling
13 presents the
changes
rate of the of the joints
welding temperature
in theoncooling
the four stage
points.after
As thethe
welding
weldingcurrent is cut off,
current the electrode
is cut force13
off. Figure is presents
still imposed on the welding specimen with the purpose of relaxation of residual stress; meanwhile the
the changes of the temperature on the four points. As the welding current is cut off, the electrode
cooling water is employed in order to cool the welding zone. This stage has an unignorable effect on
force is the
still imposed on
microstructure the welding
transformation specimen
of the with
fusion zone. theinitial
At the purpose of relaxation
stage, the cooling rate isofvery
residual
high stress;
meanwhileand thethemolten
cooling area water is employed
progressively in order
grows smaller. to cool
At the final stage,the
it iswelding
much smaller zone.
thanThis
beforestage
but has an
unignorable effect on
the maximum the microstructure
temperature transformation
is still very high. of the
At the beginning, theminimum
fusion zone. At theofinitial
temperature the fourstage, the
◦ Ac1 , and
cooling points
rate isis about
very 740 highC and
and itthe
is above
molten areatheprogressively
temperatures ofgrows
points B,smaller.
C, and D Atare all finalAc3stage,
thebelow , it is
which means these areas are not completely austenitized while the area near point A is completely
much smaller than before but the maximum temperature is still very high. At the beginning, the
minimum temperature of the four points is about 740 °C and it is above Ac1, and the temperatures of
points B, C, and D are all below Ac3, which means these areas are not completely austenitized while
the area near point A is completely austenitized. In the first 0.2 s, the cooling rates of the four points
are all higher than the critical cooling rate of austenitic complete transformation into martensite [35].
 Materials 2019, 12, 1108 10 of 16

austenitized. In the first 0.2 s, the cooling rates of the four points are all higher than the critical cooling
rate of austenitic complete transformation into martensite [35]. As the welding joints were removed,
, 12, x FOR the temperature
PEER REVIEW of point B increases rather than decreases and the paces of temperature changes are
synchronized with point A without the effect of cooling water.

Materials 2019, 12, x FOR PEER REVIEW 10 of 16

Figure 12. Four key points selected.

Figure 12. Four key points selected.
Figure 12. Four key points selected.

Figure 13.Temperature
Figure13. Temperatureprofile
profileofof
the four
the points
four after
points the
after welding
the current
welding was
current cut
was off.
cut off.

Table
Table4 lists the the
4 lists nugget size predicted
nugget by finite
size predicted by element modeling
finite element and measured.
modeling As the welding
and measured. As the
peak current is respectively 6 kA, 8 kA, and 10 kA, the results of simulation analysis
welding peak current is respectively 6 kA, 8 kA, and 10 kA, the results of simulation analysis can can reflect
the heat the
reflect affected
heat zone (HAZ)
affected zoneof(HAZ)
the welding
of thejoints,
weldingthejoints,
simulated nugget size
the simulated is consistent
nugget with the
size is consistent
actual results, with the relative error about 2%. However, as the welding current
with the actual results, with the relative error about 2%. However, as the welding current is extremely highis
(12 kA), the high
extremely simulated
(12 kA),nugget size is larger
the simulated than
nugget theisobserved
size value
larger than the and the relative
observed errorthe
value and is relative
about
10%. As extremely large welding current is employed in welding process, the molten
error is about 10%. As extremely large welding current is employed in welding process, the molten metal is prone to
splash
metalfrom the welding
is prone to splash zone
fromandthethe nugget zone
welding size should
and thebenugget
diminished due to expulsion.
size should be diminishedHowever,
due to
this factor can’t be considered in the thermoelectricity transient analysis of the welding
expulsion. However, this factor can’t be considered in the thermoelectricity transient analysis of the process.
welding process.

Figure 13. Temperature profile of the offour

Table 4. Comparison points
simulated after
results and the welding
experimental results. current was cut off.
Maximum
Simulated
Welding Welding Electrode Measured Relative Temperature
4 lists the nugget size predicted by finite element modeling and measur
Time Current Force Nugget Size
Nugget
Error at the Nugget
Size
Center
eak current is Trespectively
I
6 kA,
F
8 kA,D and 10 DkA, theΔresults of simulation a 1 Tmax
cycles kA kN mm mm % °C
Materials 2019, 12, 1108 11 of 16

Table 4. Comparison of simulated results and experimental results.

Maximum
Welding Welding Electrode Measured Simulated Relative
Temperature at the
Time Current Force Nugget Size Nugget Size Error
Nugget Center
T I F D D1 ∆ Tmax
cycles kA kN mm mm % ◦C
2.0 4.71 5.10 8.28 1765
2.5 5.69 5.80 1.93 1867
11 10
3.5 6.42 6.20 3.43 1952
4.0 6.21 6.40 3.06 2000
6 - - - 1302
8 4.25 4.40 3.53 1770
14
10 6.88 6.94 0.87 2076
3.5
12 6.90 7.60 10.14 2286
8 6.05 5.50 9.09 1800
10
17 6.57 6.80 3.50 2185

Figure 14 presents the microstructure of the welding joints. The fusion zone and heat affected zone
are heated above Ac1 under the influence of heat generated by resistance and welding current, while
ferrite begins to change into austenite. In the cooling stage, the austenite transforms into martensite,
which has been discussed above. The microstructures of fusion zone are mainly composed of martensite
structure, as shown in Figure 14e. The cooling rates of the whole welding joints are all larger than
the critical cooling rate of hypoeutectoid steel martensitic phase transformation. When it cools to a
temperature lower than Ms , non-invasive martensitic phase transformation will be developed in the
region. The carbon content of DP600 steel is less than 0.2%, and the result of martensitic transformation
is that the fusion zone is filled with big lath martensite as it cools to room temperature. The heat
affected zone can be divided into the overheated zone (Figure 14d), fine grained region (Figure 14b),
and dual-phase region (Figure 14a) outward along the fusion zone. The overheated zone is uniformly
distributed with lath martensite, which is completely in agreement with the fusion zone. Fine grained
region is generally composed of fine and uniform martensite, and this region is farther from the center
of nugget and the microstructure is almost filled with austenitic with the action of welding current
heating. But the dwell time above the austenitizing temperature time is shorter, the grain size is
not fully grown, the region is mainly composed of slight lath martensite, a small amount of ferrite
and retained austenite after cooling. The microstructure of base metal does not change in the whole
welding process; it mainly consists of fine ferrite and martensite. Ferrite is soft phase, controlling the
material formability, while martensite is hard phase, in charge of the hardness and strength of the
material. Figure 15 exhibits the microstructures of four fusion zones with different welding parameters.
As the dwell time above austenitic initialization Ac1 is respectively 156 ms, 234 ms, 256 ms, and 267 ms
according to finite element modeling. The size of the martensite grain increases successively.
The main quality characteristics of welding joints consist of nugget size, heat affected zone width
and height, peak load, indentation depth, and failure modes in the tensile-shear test. On one hand, it
can be seen from Figures 16 and 17 that the maximum displacement, peak load, penetration rate and
failure energy are monotonically increasing with the fusion size. On the other hand, the influences of
welding parameters on the nugget size are consistent with the influences of welding parameters on
mechanical properties of joints. Then it logically follows that the welding parameters directly affect
the nugget size, while the mechanical properties of the sample are directly controlled by the fusion
size. Table 5 lists the Pearson correlation coefficients between the nugget size and penetration rate,
peak load, failure energy and failure displacement, which were calculated using the software Matlab
2013a. Thus, in a comprehensive view, the most direct control factor of the mechanical properties of
the resistance spot welding joints is the nugget size.
 The microstructure of base metal does not change in the whole welding process; it mainly consists of
fine ferrite and martensite. Ferrite is soft phase, controlling the material formability, while
martensite is hard phase, in charge of the hardness and strength of the material. Figure 15 exhibits
the microstructures of four fusion zones with different welding parameters. As the dwell time above
austenitic initialization
Materials 2019, 12, 1108 Ac1 is respectively 156 ms, 234 ms, 256 ms, and 267 ms according to12finite
of 16
element modeling. The size of the martensite grain increases successively.

Materials 2019, 12, x FOR PEER REVIEW 12 of 16

Figure 14. The The microstructure

microstructure of
of the welding
welding joints:
joints: (a)
(a) dual-phase
dual-phase region
region (I(I in
in Figure
Figure 14c);
14c); (b)
(b) fine
grained region (IIin Figure 14c); (c) metallographic structure of the weld nugget;
nugget; (d) (d) overheated
overheated zone
zone
Materials 2019, 12, x FOR PEER REVIEW 12 of 16
(III in Figure 14c); and (e) fusion zone (IV in Figure 14c).

Figure 15. The microstructures of welding joints with different dwell periods above Ac1: (a) 156 ms;
(b) 234 ms; (c) 256 ms; and (d) 267 ms.

The main quality characteristics of welding joints consist of nugget size, heat affected zone
width and height, peak load, indentation depth, and failure modes in the tensile-shear test. On one
hand, it can be seen from Figures 16 and 17 that the maximum displacement, peak load, penetration
rate and failure energy are monotonically increasing with the fusion size. On the other hand, the
influences of welding parameters on the nugget size are consistent with the influences of welding
parameters on mechanical properties of joints. Then it logically follows that the welding parameters
directly affect the nugget size, while the mechanical properties of the sample are directly controlled
by the fusion size. Table 5 lists the Pearson correlation coefficients between the nugget size and
penetration rate, peak load, failure energy and failure displacement, which were calculated using
the software Matlab 2013a. Thus, in a comprehensive view, the most direct control factor of the
Figure 15. The
15. The microstructures
Figureproperties
microstructures of
of welding
welding joints with different
joints with periods above Ac1: (a) 156 ms;
different dwell periods
mechanical of the resistance spot welding joints is thedwell
nugget size.above Ac1: (a) 156 ms;
(b)
(b) 234
234 ms;
ms; (c)
(c) 256
256 ms;
ms; and
and (d)
(d) 267
267 ms.
ms.

The main quality characteristics of welding joints consist of nugget size, heat affected zone
width and height, peak load, indentation depth, and failure modes in the tensile-shear test. On one
hand, it can be seen from Figures 16 and 17 that the maximum displacement, peak load, penetration
rate and failure energy are monotonically increasing with the fusion size. On the other hand, the
influences of welding parameters on the nugget size are consistent with the influences of welding
parameters on mechanical properties of joints. Then it logically follows that the welding parameters
directly affect the nugget size, while the mechanical properties of the sample are directly controlled
by the fusion size. Table 5 lists the Pearson correlation coefficients between the nugget size and
penetration rate, peak load, failure energy and failure displacement, which were calculated using
the software Matlab 2013a. Thus, in a comprehensive view, the most direct control factor of the
mechanical properties of the resistance spot welding joints is the nugget size.
Figure 16. Relationships between the nugget diameter, penetration rate, and failure energy.
Figure 16. Relationships between the nugget diameter, penetration rate, and failure energy.
Materials 2019, 12, 1108 13 of 16
Figure 16. Relationships between the nugget diameter, penetration rate, and failure energy.

Figure 17. The relationships between the nugget diameter, peak load, and failure displacement.

Table 5. Correlation coefficients of the five welding quality indicators.

Correlation Nugget Penetration Failure

Peak Load Failure Energy
Coefficient Diameter Rate Displacement
Nugget
1.00 0.97 0.90 0.88 0.92
diameter

Based on the experimental results, the effects of nugget size on the mechanical properties of
resistance spot welding are emphatically analyzed. As Figures 16 and 17 present the relationships
between the nugget size and the peak load, maximum displacement, and failure energy, the welding
joints can be divided into three categories according to the nugget diameter: cold weld, undersized
weld and good weld. A cold weld is usually generated as the welding heat in the welding zone is
so little that the steel just starts to melt, even though the temperature has not reached the melting
point. In this case, the mechanical performances of this kind weld are very poor. As for the undersized
weld, the steel plates thermally fused with each other; however, as it is still unsatisfying, as it cannot
bear relatively large forces. As for undersized welds, a brittle fracture is the main form of efficiency
loss. The sufficient heat input leads the adjacent surfaces to melt, resulting in an increasing nugget
volume. The larger the nugget size is, the greater the maximum displacement, peak load, and the
failure energy will be. The maximum displacement, peak load, and failure energy increase significantly
with the nugget diameter as the welding joints belong to undersized weld or good weld; while in
the undersized weld region, the mechanical properties increased slowly. This shows that changing
the nugget size has a significant effect on the mechanical properties of the welding joints. During the
loading failure process, as the nugget size gradually increases, the peak load is obviously increasing,
and shear deformation also increases significantly.
It is well documented that the size of the fusion zone is the key physical weld attribute controlling
the mechanical properties of spot welds among all the welding quality indices [36]. The failure modes
of the welding joints include: the pullout failure (PF) mode and interfacial failure (IF) mode. The PF
mode is preferred as it indicates a higher-fracture energy and better mechanical property. The larger
the nugget size is; the better mechanical properties of the welding joints will be improved and the
PF mode will occur. As the nugget belongs to a cold weld, its failure mode is completely IF. As the
nugget gets larger, its failure mode transfers from the IF mode into the PF mode. However, the costs
of the weld assembly, for example, the welding heat input demands, will be also increased in the
meantime. The critical nugget size D induced the IF transforming into the PF mode is considered to be
just satisfying without increasing the productive cost (t is the thickness of the steel sheets) [18]:

D = 3.51t (9)

To accomplish the goal of ensuring the welding quality, while reducing welding costs as much
as possible, it is necessary to optimize the welding process parameters through controlling the weld
Materials 2019, 12, 1108 14 of 16

nugget size. At present, setting and determining the welding process parameters are commonly based
on lots of experiments and work experience.
If the welding parameters are given, the nugget size of the welding joints can be predicted by
finite element analysis. For example, the thickness of DP600 steel sheets is 1.7 mm, the welding time
is 14 cycles, the welding current is 8 kA, and the electrode force is 3.5 kN. The nugget size is 4.4 mm,
which is much smaller than the critical nugget size 5.97 mm. It indicates that, this welding parameters
combination cannot guarantee the welding quality. It is necessary to adjust the weld nugget size
through changing the welding process parameters. Based on the finite element analysis, there are
several methods controlling the nugget diameter as following. The welding current can be adjusted
with the purpose of increasing the nugget size close to 6.0 mm in order to guarantee the welding
Materials
quality.2019,
The12,welding
x FOR PEER REVIEW
current can be changed into 9 kA, and the final nugget size is about 6.0 14 of
mm 16

(Figure 18 curve II). Improving the welding current peak to 10 kA in later period (nine cycles) to
III in Figure
change 18). CurveofI weld
the rangeability is thenugget
nuggetindiameter growth stage
the slow-growth in theiswelding process (shown
another method as the welding
as curve
current is 10 kA, the final nugget size is 6.94 mm, and it is satisfactory. The peak
III in Figure 18). Curve I is the nugget diameter growth in the welding process as the welding current load of the welding
joints whose
is 10 kA, the nugget size issize
final nugget slightly smaller
is 6.94 mm, and the it
critical nugget is The
is satisfactory. far less
peak than that
load of critical
of the welding nugget,
joints
even if there is a little difference between the their nugget sizes. However,
whose nugget size is slightly smaller the critical nugget is far less than that of critical nugget, even when the nuggets
diameters
if there is are above
a little the critical
difference value,the
between thetheir
mechanical
nugget performances
sizes. However, of when
the welding joints with
the nuggets very
diameters
different nuggets are close to each other [37]. From this point of view, nugget
are above the critical value, the mechanical performances of the welding joints with very different size (curve II and
curve
nuggets III are
in Figure
close to18) eachslightly largerFrom
other [37]. thanthis
the point
critical
of nugget value size
view, nugget not (curve
only indicates relatively
II and curve III in
sound mechanical performances, but also saves energy and boosts efficiency. The
Figure 18) slightly larger than the critical nugget value not only indicates relatively sound mechanical costs of Curve I
seem to be a bitbut
performances, high.
also saves energy and boosts efficiency. The costs of Curve I seem to be a bit high.

Figure 18. Controlling the nugget size.

Figure 18. Controlling the nugget size.
4. Conclusions
4. Conclusions
(1) The contact radius of the electrode/sheet contact surfaces is basically equal to the electrode
radius inThe
(1) thecontact radius
preloading of the
stages, electrode/sheet
while the value of contact surfaces
sheet/sheet is basically
contact surface isequal
muchtolarger
the electrode
than that.
radius(2)inThe theformation
preloading stages, while the value of sheet/sheet contact surface
of the nugget in theoretical analysis undergoes three stages: plasticity is much larger than
adhesion,
that.
rapid growth, and slow growth.
(2) The formation
(3) Close correlationof the nugget
is observed in theoretical
between simulated analysis undergoes
thermal fields and thethree stages: plasticity
microstructure changes
adhesion, rapid growth, and slow growth.
of joints. The longer the dwell time temperature is larger than the austenitic initialization Ac1 of the
(3)the
joints, Close correlation
coarser is observed
the lath martensite between
that simulated thermal fields and the microstructure
will be generated.
changes of joints. The longer the dwell time temperature
(4) It should be pointed out that the error is about 10% is as
larger than thecurrent
the welding austenitic initialization
is very large and A c1
the
of the joints,
expulsion the coarser
occurs. Whenthe thelath martensite
expulsion thatmolten
occurs, will bemetal
generated.
splashes from the welding zone and the
(4) It should be pointed out that the error is about
nugget size is smaller than expected, however, this can’t be considered 10% as the welding
in thecurrent is very large
finite element and
modeling.
the expulsion occurs. When the expulsion occurs, molten metal splashes from the
(5) The accuracy of the physical experiments and numerical simulations is within the acceptable welding zone and
the nugget
range. It is asize
realispotential
smaller forthantheexpected,
simulatedhowever,
results to this can’t bebyconsidered
be adopted a wide range in of
theusers
finite
forelement
process
modeling.
planning in modern manufacturing.
(5) The accuracy of the physical experiments and numerical simulations is within the acceptable
range. ItContributions:
Author Conceptualization:
is a real potential D.Z. andresults
for the simulated Y.W.; methodology:
to be adopted D.L.;
bysoftware: P.Z.; validation:
a wide range of users D.Z.
for
and D.L.; formal analysis: P.Z.; investigation: Y.W.; resources: Y.W.; data curation: P.Z.; writing—original draft
process planning in modern manufacturing.
Author Contributions: Conceptualization: D.Z. and Y.W.; methodology: D.L.; software: P.Z.; validation: D.Z.
and D.L.; formal analysis: P.Z.; investigation: Y.W.; resources: Y.W.; data curation: P.Z.; writing—original draft
preparation: D.Z.; writing—review and editing: D.L.; visualization: D.Z.; supervision: Y.W.; project
administration: Y.W.; funding acquisition: D.Z.
Materials 2019, 12, 1108 15 of 16

preparation: D.Z.; writing—review and editing: D.L.; visualization: D.Z.; supervision: Y.W.; project administration:
Y.W.; funding acquisition: D.Z.
Funding: This research was funded by Natural Science Foundation of Shandong Province, grant number
ZR2016EEM47/ZR2018PEE004and the Scientific Research Staring Foundation of Linyi University, grant
number LYDX2016BS41.
Acknowledgments: The authors are grateful for the experiment supported by the analysis and test center of
Huazhong University of Science and Technology and Dongfeng Peugeot Citroen Automobile Company Limited.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Wang, X.; Liu, B.; Liu, W.; Zhong, X.; Jiang, Y.; Liu, H. Investigation on the mechanism and failure mode of
laser transmission spot welding using PMMA material for the automotive industry. Materials 2017, 10, 22.
[CrossRef] [PubMed]
2. Fazaeli, A.; Ekrami, A.; Kokabi, A.H. Microstructure and mechanical properties of dual phase steels, with
different martensite morphology, produced during TLP bonding of a low C-Mn steel. Met. Mater. Int. 2016,
22, 856–862. [CrossRef]
3. Thakur, A.G.; Nandedkar, V.M. Optimization of the resistance spot welding process of galvanized steel sheet
using the Taguchi method. Arab. J. Sci. Eng. 2014, 39, 1171–1176. [CrossRef]
4. Amirthalingam, M.; van der Aa, E.M.; Kwakernaak, C.; Hermans, M.J.M.; Richardson, I.M. Elemental
segregation during resistance spot welding of boron containing advanced high strength steels. Weld. World
2015, 59, 743–755. [CrossRef]
5. Hayat, F.; Sevim, I. The effect of welding parameters on fracture toughness of resistance spot-welded
galvanized DP600 automotive steel sheets. Int. J. Adv. Manuf. Technol. 2012, 9–12, 1043–1050. [CrossRef]
6. Kaščák, L.; Spišák, E.; Gajdoš, I. Influence of welding parameters on the quality of resistance spot welded
joints of DP600 steels. Key Eng. Mater. 2015, 635, 143–146. [CrossRef]
7. Huin, T.; Dancette, S.; Fabrègue, D.; Dupuy, T. Investigation of the failure of advanced high strength steels
heterogeneous spot welds. Metals 2016, 6, 111. [CrossRef]
8. Aktas, S.; Ozsarac, U.; Aslanlar, S. Effect of spot welding parameters on tensile properties of DP600 steel
sheet joints. Mater. Manuf. Process. 2012, 27, 756–764. [CrossRef]
9. Kim, D.; Yu, J.; Rhee, S. Effect of a conically shaped hollow electrode on advanced high strength steel in
three-sheet resistance spot welding. Int. J. Precis. Eng. Manuf. 2016, 17, 331–336. [CrossRef]
10. Espinel Hernández, A.; Sánchez Roca, A.; CarvajalFals, H.; AntoniFerraresi, V.; Oliveira Vilarinho, L.
Influence of polarity on mechanical properties of dissimilar resistance spot welds of DP 600/AISI 304 steels.
Sci. Technol. Weld. Join. 2016, 21, 607–613. [CrossRef]
11. Jaber, H.L.; Pouranvari, M.; Salim, R.K.; Hashim, F.A.; Marashi, S.P.H. Peak load and energy absorption of
DP600 advanced steel resistance spot welds. Ironmak. Steelmak. 2017, 44, 699–706. [CrossRef]
12. Podržaj, P.; Simončič, S. A machine vision-based electrode displacement measurement. Weld. World 2014, 58,
93–99. [CrossRef]
13. Podržaj, P.; Jerman, B.; Simončič, S. Poor fit-up condition in resistance spot welding. J. Mater. Process. Technol.
2016, 230, 21–25. [CrossRef]
14. Mvola, B.; Kah, P.; Martikainen, J.; Suoranta, R. Dissimilar high-strength steels: Fusion welded joints,
mismatches, and challenges. Rev. Adv. Mater. Sci. 2016, 44, 146–159.
15. Zhang, X.; Yao, F.; Ren, Z.; Yu, H. Effect of welding current on weld formation, microstructure, and mechanical
properties in resistance spot welding of CR590T/340Y galvanized dual phase steel. Materials 2018, 11, 2310.
[CrossRef] [PubMed]
16. Moghanizadeh, A. Evaluation of the physical properties of spot welding using ultrasonic testing. Int. J. Adv.
Manuf. Technol. 2016, 85, 535–545. [CrossRef]
17. Kaščák, L’.; Kubík, R. Influence of welding parameters on the properties of spot welded joints of dual-phase
steels. Transf. Inovácií 2014, 29, 286–289.
18. Zhao, D.; Wang, Y.; Liang, D.; Zhang, P. Modeling and process analysis of resistance spot welded DP600
joints based on regression analysis. Mater. Des. 2016, 110, 676–684. [CrossRef]
Materials 2019, 12, 1108 16 of 16

19. Rao, S.S.; Chhibber, R.; Arora, K.S.; Shome, M. Resistance spot welding of galvannealed high strength
interstitial free steel. J. Mater. Process. Technol. 2017, 246, 252–261. [CrossRef]
20. Zhao, D.; Wang, Y.; Sheng, S.; Lin, Z. Multi-objective optimal design of small scale resistance spot welding
process with principal component analysis and response surface methodology. J. Intell. Manuf. 2014, 25,
1335–1348. [CrossRef]
21. Jagadeesha, T.; Jothi, T.S. Studies on the influence of process parameters on the AISI 316L resistance
spot-welded specimens. Int. J. Adv. Manuf. Technol. 2017, 93, 73–88. [CrossRef]
22. Mirzaei, F.; Ghorbani, H.; Kolahan, F. Numerical modeling and optimization of joint strength in resistance
spot welding of galvanized steel sheets. Int. J. Adv. Manuf. Technol. 2017, 92, 3489–3501. [CrossRef]
23. Mikno, Z.; Bartnik, Z. Heating of electrodes during spot resistance welding in FEM calculations. Arch. Civ.
Mech. Eng. 2016, 16, 86–100. [CrossRef]
24. Chrysochoos, A.; Berthel, B.; Latourte, F.; Pagano, S.; Wattrisse, B.; Weber, B. Local energy approach to steel
fatigue. Strain 2008, 44, 327–334. [CrossRef]
25. Zhao, Z.; Yao, L.; Guo, H. Copper and Copper Alloys Handbook; Science Press: Beijing, China, 1993; pp. 133–134.
26. Tsai, C.; Dai, W.; Dickinson, D. Analysis and development of a real-time control methodology in resistance
spot welding. SAE Trans. 1991, 100, 158–176.
27. Baumer, R.; Adonyi, Y. Transient high-frequency welding simulations of dual-phase steels. Weld. J. 2009, 88,
193–201.
28. Schenk, T.; Richardson, I.M.; Kraska, M.; Ohnimus, S. Non-isothermal thermomechanical metallurgical
model and its application to welding simulations. Sci. Technol. Weld. Join. 2009, 14, 152–160. [CrossRef]
29. Bézi, Z.; Baptiszta, B.; Szávai, S. Experimental and numerical analysis of resistance spot welded joints on
DP600 sheets. Weld. Mater. Test. 2014, 4, 7–12.
30. Gould, J.E. Modeling primary dendrite arm spacings in resistance spot welds, part I-modeling studies.
Weld. J. 1994, 73, 67s–72s.
31. Gould, J.E. Modeling primary dendrite arm spacings in resistance spot welds, part II-experimental studies.
Weld. J. 1994, 73, 91s–100s.
32. Pakkanen, J.; Vallant, R.; Kičin, M. Experimental investigation and numerical simulation of resistance spot
welding for residual stress evaluation of DP1000 steel. Weld. World 2016, 60, 393–402. [CrossRef]
33. Ramazani, A.; Bruehl, S.; Abbasi, M.; Bleck, W.; Prahl, U. The effect of bake-hardening parameters on the
mechanical properties of dual-phase steels. Steel Res. Int. 2016, 87, 1–7. [CrossRef]
34. Hayat, F. Comparing properties of adhesive bonding, resistance spot welding, and adhesive weld bonding
of coated and uncoated DP 600 steel. J. Iron Steel Res. Int. 2011, 18, 70–78. [CrossRef]
35. Ramazani, A.; Mukherjee, K.; Abdurakhmanov, A.; Abbasi, M.; Prahl, U. Characterization of microstructure
and mechanical properties of resistance spot welded DP600 steel. Metals 2015, 5, 1704–1716. [CrossRef]
36. Wang, X.J.; Zhou, J.H.; Yan, H.C.; Pang, C.K. Quality monitoring of spot welding with advanced signal
processing and data-driven techniques. Trans. Inst. Meas. Control 2018, 40, 2291–2302. [CrossRef]
37. Tamizi, M.; Pouranvari, M.; Movahedi, M. Welding metallurgy of martensitic advanced high strength steels
during resistance spot welding. Sci. Technol. Weld. Join. 2017, 22, 327–335. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).