This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Simulation of Friction Stir Spot Welding (FSSW) Process: Study of Friction Phenomena
Mokhtar Awang
Dissertation
Submitted to the College of Engineering and Mineral Resources at West Virginia University
in partial fulfillment of the requirements for the degree of
Doctor of Philosophy in
Mechanical Engineering
Victor H. Mucino, Ph.D. (Chairman) Bruce Kang, Ph.D. Jacky Prucz, Ph.D. Ken Means, Ph.D.
Powsiri Klinkhachorn Ph.D.
Department of Mechanical and Aerospace Engineering
Table 5. 2: Constants for Johnson-Cook material model [57].
Material Tmelt(oC) A (MPa) B (MPa) C n m
Al 6061-T6 582 293.4 121.26 0.002 0.23 1.34
59
Stress Strain Curves for Johnson-Cook Work Hardening
0
50
100
150
200
250
300
350
400
0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16
Strain
Stre
ss (M
Pa) T=100 C
T=200 CT=300 CT=400 CT=500 C
Fig. 5. 8: Plot of stress strain curve for Johnson-Cook work hardening.
5.8 Adaptive Mesh
The simulation of FSSW process uses the advantage of adaptive meshing offered in
Abaqus/Explicit. Adaptive meshing is performed in Lagrangian domains, which is defined by
element sets. In this case, the whole workpieces mesh is considered the domain of the adaptive
mesh. The adaptive mesh domain has sliding boundary regions at the top and the bottom of both
workpieces. In sliding boundary regions, the mesh is constrained to move with the material in
60
the direction normal to the boundary region, but it is fully unconstrained in the directions
tangential to the boundary region.
For the model presented in this work, the remeshing is made after 10 increments and each
remeshing algorithm includes 3 mesh sweeps for optimizing the node positions. The calculation
of the new mesh is based on volume smoothing method.
61
CHAPTER 6: DISCUSSION OF RESULTS
FE simulation results discussed in this chapter consists of frictional dissipation history,
plastic dissipation history, thermal response, plastic strain and material flow. The geometry and
the mesh used in the FE models were kept the same throughout this work.
6.1 Overview of Frictional Dissipation Energy during FSSW Process
Friction between the tool and the workpiece occurs at three tool surfaces, i.e., tip of the
pin, vertical side of the pin and tool shoulder. Fig. 6. 1 (a-c) shows three phases of friction
between the tool and the workpiece. The figure on the left illustrates the temperature profile, and
the plot on the right shows the frictional dissipation energy history.
At the initial stage, when the tool slightly comes in contact with the upper plate while
spinning, it is seen that the high temperature concentrates around the tip of the tool and heat
generation starts to develop (Fig. 6. 1 (a)). The temperature profile is also distributed uniformly
due to the heat conduction from the first plate to the second plate.
While spinning, the tool also moves downward and the pressure exerted from the tool
causes the first plate to deform slightly, as shown in Fig. 6. 1(b). This leads to less heat
conduction from the top plate to the bottom plate, which results in higher temperature profile in
the top plate. At this stage, the heat generation increases as the welding process advances.
62
As the tool keeps pressing downward, its shoulder contacts the deformed top plate and
more heat is generated through the friction between the shoulder and the workpiece (Fig. 6.
1(c)). At this phase, more heat conduction occurs between the plates since the two plates come
in contact again and more uniform temperature profile can be seen. Since the tool is defined as
an isothermal surface, the temperature profile on the tool is constant through out the analysis.
a) Phase 1, t=0.245 s.
b) Phase 2, t=0.945 s.
63
c) Phase 3, t=1.365 s.
Fig. 6. 1: Three stages of friction between the tool and the workpiece. Left: Temperature profile, Right: Friction dissipation energy history.
6.2 Energy Dissipation during FSSW Process
The FSSW technique relies on the heat generated during the process to join the
workpieces together. There are three possible heat sources generated during FSSW process, i.e.,
friction work at the tool and top workpiece interface, friction work at the interface of top and
bottom workpiece, and plastic deformation of the workpiece material. In this work, the friction
and plastic works have been investigated using frictional energy dissipation and plastic energy
dissipation histories, respectively. The angular velocity and plunge rate were 3000 rpm and 1
mm/s, respectively. Table 6. 1 shows the summary of energy dissipation during FSSW process.
Su et al. [44] have conducted an experiment on energy generation of 6.3 mm Al 6061-T6.
In the experiment, rotational speed and plunge rate were set at 3000 rpm and 2.5 mm/s,
respectively. The average energy generated by 10 mm diameter tool with shoulder at five
different plunge depths was 1.964 kJ. Based on the FE results shown in Table 6. 1, the total
64
energy dissipation at plunge depth of 1.505 mm is 1.519 kJ. The discrepancy between the two
results is due to the different thickness of the workpiece and plunge rate.
Table 6. 1: Summary of energy dissipation during FSSW process.
Energy Dissipation (kJ)
Percentage (%)
Friction at tool and workpiece interface 1.471 96.84 %
Frictional at top and bottom workpieces interface
0.000375 0.02 %
Plastic deformation 0.0478 3.14 %
Total Energy Dissipation 1.519 100 %
6.2.1 Heat Generation due to Friction Work at the Tool and Workpiece Interface
The simulation results show that the friction work at the tool and top workpiece interface
contributes the most heat to the welding process. The frictional energy contributes 96.84 % of
the total energy as shown in Fig. 6. 2. This high percentage of energy generation is expected
due to the presence of high differential velocities (slip rate) on the workpiece surface that caused
by high rotational speed and pressure from the tool have created. The peak slip rate as shown in
Fig. 6. 3 is 2284 mm/s.
Fig. 6. 2 also shows that the frictional dissipation energy increases drastically after the
time reaches 0.9450 seconds. The increase of frictional energy is due to the additional friction
from the interface of the shoulder of the tool and the surface of the top workpiece.
65
Fig. 6. 2: Frictional dissipation energy history at the tool and workpiece interface.
Fig. 6. 3: Slip rate distribution on upper surface of the top workpiece.
66
6.2.2 Heat Generation due to Friction Work at the Top and Bottom Workpieces Interface
Based on the results in Table 6. 1, about 0.02 % of the total energy is due to frictional
force at the interface between the top and bottom workpieces. Even though the amount of
energy generated at this interface is almost negligible, the interface is important in FSSW
because the actual welding occurs here. Consequently, it is relevant to trace how much energy is
used at the interface and how this is affected by the process parameters.
Fig. 6. 4 shows frictional dissipation energy history at the top and bottom workpieces
interface is about 0.000375 kJ (375 mJ) at 1.505 seconds. The friction work at the interface
between the top and bottom workpieces gives the least contribution to the total heat generation
because the interface has very small relative motion due to friction as shown in Fig. 6. 5 and Fig.
6. 6. The maximum slip rates are 3.375 mm/s and 0.3985 mm/s on lower surface of the top
workpiece (Fig. 6. 5) and upper surface of the bottom workpiece (Fig. 6. 6), respectively.
67
Fig. 6. 4: Frictional dissipation energy at the top and bottom workpiece interface.
Fig. 6. 5: Slip rate distribution on lower surface of the top workpiece.
68
Fig. 6. 6: Slip rate distribution on upper surface of the bottom workpiece.
6.2.3 Heat Generation due to Internal Friction/Plastic Work
As shown in Fig. 6. 7, plastic dissipation energy in the material is 47.8 kJoule, which is
about 3.14 % of the total energy dissipated. This energy is due to the presence of internal
friction forces, which tends to resist the motion of the material. As can be seen from the figure,
the plastic dissipation energy increases drastically after 1.3 seconds. This is because, in addition
to the plastic straining process due to the pin penetration, a large plastic deformation occurs
when the shoulder of the tool touches the surface of the workpiece at the end of the cycle time.
69
Fig. 6. 7: Plastic dissipation energy history.
6.3 Study of Friction Coefficient Dependence
The dependence of friction coefficient on surface temperature, slip rate, and contact
pressure are investigated. In this work, the same FE models were run with three different
friction coefficient dependencies by trial and error until the peak temperatures were obtained.
Rotational speed and plunge rate of the welding tool are set to be 3000 rpm and 1 mm/second,
respectively.
The frictional and plastic dissipation energy history curves are compared after 1.435
second as shown in Fig. 6. 8. The maximum frictional dissipation energy are about 1.27 kJ, 1.20
kJ, and 1.15 kJ for friction coefficient dependence of contact pressure, slip rate and temperature,
70
respectively, which are within 8.4 % difference. Appendix B-2 shows the temperature profiles
for different friction coefficient dependencies at various times.
71
a) Contact pressure dependence b) Slip rate dependence c) Surface temperature dependence
Fig. 6. 8: Frictional (dotted line) and plastic (solid line) dissipation energies history for various friction coefficient dependences.
72
6.4 Effect of Welding Tool Plunge Rate
Parametric studies have been conducted to determine the effect of tool penetrating speed
on frictional and plastic dissipation energies. Three different tool speeds; 1 mm/s, 5 mm/s, and
10 mm/s, are modeled with contact pressure friction coefficient dependent. The welding tool
rotational speed was set at 3000 rpm. The targeted welding tool displacement was 1.505 mm.
Fig. 6. 9 shows frictional and plastic dissipation energy history for three different
welding speeds. Based on the curves, frictional and plastic dissipation energy increases as the
tool velocity decreases. The frictional dissipation energy of 5 mm/s plunge rate reduces to about
29 % of the 1 mm/second frictional dissipation energy. The same decreasing trend is observed
with 10 mm/s plunge rate (14 % decrement). This is because the slower the tool velocity, the
more time it spends to spin on the workpiece thus more energy is produced. The plastic
dissipation energy also follows the same trend of frictional dissipation energy curves. In short,
the slower the penetrate speed, the higher the plastic dissipation energy. Table 6. 2 summarizes
the frictional and plastic dissipation energy for different tool plunge rates.
73
Fig. 6. 9: Energy dissipation for different tool rotational speed.
Tool Velocity (mm/s)
Frictional Dissipation Energy
Plastic Dissipation Energy
1.0
5.0
10.0
74
Table 6. 2: Summary of friction and plastic dissipation energies for different tool’s plunge rate.
Plunge Rate
(mm/sec.)
Frictional Dissipation
Energy (kJ)
Percentage
Reduction (%)
Plastic Dissipation
Energy (kJ)
Percentage
Reduction (%)
1.0 1.471 100 0.0478 100
5.0 0.425 28.9 0.0237 49.6
10.0 0.207 14.1 0.0193 40.4
6.5 Effect of Welding Tool’s Rotational Speed
A parametric study has also been conducted for various rotational speeds. Three
different welding tool rotational speeds, which are 3000 rpm, 2500 rpm, and 2000 rpm, have
been run on the FE models in order to study the effect of welding tool’s rotational speed on heat
generation. Plunge rate was set to be 1 mm/s. The model used friction coefficient dependence
of contact pressure in this case.
Fig. 6. 10(a-c) show frictional dissipation energy histories for 3000 rpm, 2500 rpm, and
2000 rpm welding tool rotational speed, respectively. Based on the results, the higher the
rotational speed is, the higher the dissipation energy. The frictional dissipation energy is
reduced by about 8.4 % when the rotational speed is reduced from 3000 rpm to 2500 rpm and
reduced by about 19.2 % when it is reduced from 3000 rpm to 2000 rpm. This is because higher
rotational speed will result in higher relative velocity of the material, consequently higher energy
will be produced. Table 6. 3 summarizes frictional dissipation energy for different tool’s
rotational speeds.
75
a) 3000 rpm b) 2500 rpm c) 2000 rpm
Fig. 6. 10: Frictional dissipation energy for various rotational speeds.
76
Table 6. 3: Summary of frictional dissipation energy for different tool’s rotational speeds.
Tool’s Rotational Speed
(rpm)
Frictional Dissipation Energy
(kJ)
Percentage Reduction
%
3000 1.471 100
2500 1.347 91.6
2000 1.188 80.8
6.6 Thermal Response
Thermal response was studied using friction coefficient contact pressure dependent. Fig.
6. 11(a-g) shows the temperature fields at various times. Based on the results, the maximum
nodal temperature is 546.40C after 1.505 seconds when the welding tool penetrates 1.505 mm
into the workpieces. The peak temperature is consistent with the theory reported in Su, et al
[44], which suggested that the peak temperature is between 0.93Ts and 0.97Ts (Ts is the solidus
temperature of Al 6061-T6).
77
a) Time = 0.0 sec., nodal temperature = 22.000C.
b) Time = 0.175 sec., nodal temperature = 55.520C.
c) Time = 0.350 sec., nodal temperature = 99.540C.
78
d) Time = 0.525 sec., nodal temperature = 139.90C.
e) Time = 0.700 sec., nodal temperature = 183.10C.
f) Time = 0.875 sec., nodal temperature = 195.00C.
79
g) Time = 1.050 sec., nodal temperature = 264.90C.
h) Time = 1.225 sec., nodal temperature = 323.10C.
i) Time = 1.400 sec., nodal temperature = 512.80C
80
j) Time = 1.505 sec., nodal temperature = 546.40C
Fig. 6. 11: Temperature contours at various times.
Fig. 6. 12 shows the temperature distribution at the surfaces of both workpieces,
measured perpendicular from the center of the workpiece to the edge. At the tip of the tool, the
temperatures of both plates are the same, which is about 485 0C. The peak temperature of the
top surface of the upper plate is almost constant at the interface of tool’s shoulder and
workpiece, which is within the radius between 1.5 mm and 5 mm away from the center of the
workpiece. Then, the temperature starts to decease to about 150 0C at the edge of the workpiece.
At the top surface of the bottom plate, lower peak temperature is observed due to
conduction heat transfer from the upper plate. Beginning from 6 mm away from the center of
the workpiece towards to the edge, the temperature of the top surface of the upper plate and the
temperature of the top surface of the bottom plate become uniform.
81
Fig. 6. 12: Plot of temperatures versus distance away from the center of the tool.
Fig. 6. 13 shows the comparison of temperature history between the FE simulation results and
the experimental results, which reported by Su et al. [44]. Both analyses were performed on Al
6061-T6 material using 3000 rpm and 2.5 mm/s of rotational speed and plunge rate, respectively.
The tool geometry and the thickness of the workpiece, however, were different. Based on the
results, the FE simulation shows the maximum tool temperature is 514.3 0C and the experimental
shows peak temperature at the tip of the tool is about 542 0C and. From this, it can be concluded
that both results are fairly in agreement with each other, which is about 5.1 % different.
82
a) FE simulation result b) Experimental result [44]
Fig. 6. 13: Experimental versus FE simulation results of temperature history at the tip of welding tool.
83
6.7 Plastic Strain
Fig. 6. 14 shows an equivalent plastic strain as a function of distance from the center of
the tool. The equivalent plastic strain at the top surface of the upper plate shows an increasing
trend within 1.5 mm distance. This peak equivalent plastic strain region is the area under the
pin, where high pressure and torque generate tremendous heat and plasticize the materials.
Between the distance of 1.5 mm and 5 mm from the center of the workpiece, the equivalent
plastic strain drops drastically. This region is the area under the tool’s shoulder. The equivalent
plastic strain becomes zero at the distance of 5 mm from the center of the workpiece towards the
edge. The equivalent plastic strain at the top surface of the bottom plate is relatively small in the
region under the pin and becomes zero towards the edge.
Equivalent Plastic Strain versus Distance from the Tool (mm)
-25
0
25
50
75
100
125
150
175
200
225
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Distance from the Tool (mm)
Equ
ival
ent P
last
ic S
trai
n PEEQ-TOPPEEQ-BOTT
Fig. 6. 14: Plot of equivalent plastic strains versus distance away from the center of the tool.
84
6.8 Material Flow
Material flow during FSSW process has been investigated. Due to disk space constraint
that requires a lot of spaces to store large output database (.odb) file, the simulation time has
been cut short to 0.8 seconds. The rotational speed and plunge rate for this study are 3000 rpm
and 1 mm/second, respectively.
Three tracer particles, which are set of nodes in the adaptive mesh domain, are defined in
the FE model as shown in Fig. 6. 15. Tracer particle “A” is set to be underneath the pin, tracer
particle “B” is located right below the shoulder and tracer particle “C" is located slightly away
from the welding tool.
Based on Fig. 6. 16, particle “A” moves in all directions. Particles “B” and “C”,
however, move quite significantly in Z direction, and very small displacement in X and Y
direction. Since a sliding boundary region has been defined on the top surface of the workpiece,
the material is constrained to move with the mesh in the direction normal to the boundary region.
The material, however, is completely unconstrained in the directions tangential to the boundary
region.
Fig. 6. 15: Locations of particle tracking.
85
a) Tracer particle “A” b) Tracer particle “B” c) Tracer particle “C”
Fig. 6. 16: Plots of particle displacements history.
86
CHAPTER 7: DISCUSSION OF FRICTION PHENOMENA AND HEAT
GENERATION DURING FSSW PROCESS
A description of friction phenomena in FSSW is addressed in this chapter, which
includes several aspects such as the friction mechanism, friction coefficient and friction heating.
7.1 Introduction
In general, two types of friction occur during FSSW process, namely contact friction and
internal friction. The former is considered more dominant in terms of energy consumption, since
the interface experiences large pressures of contact and sliding velocity from the tool motion.
Internal friction is less dominant in terms of energy consumption as shown in the simulation
results. Contact friction in FSSW takes place at the tool and the workpiece interface and the
interface of between the two plates.
7.2 Friction at Welding the Tool and Workpiece Interface
The friction between the welding tool and the workpiece is very important during FSSW
process as it contributes most of the heat during the process. In FSSW, the friction between
rotating tool and deformable workpiece can be divided into three stages as depicted in Fig. 7. 1,
i.e., first stage, when the tip of the pin and workpiece come in contact; second phase, when the
87
pin penetrates into the workpiece; third phase, when the shoulder touches and spins on the
workpiece.
a) Phase 1 b) Phase 2 c) Phase 3
Fig. 7. 1: Schematic representation of friction between the tool and the workpiece.
Friction in FSSW process begins when the tool with high rotational speed plunges into
the workpiece. The load from the tool produces heat and contact pressure at the interface of the
tool and the workpiece, which in turn produces adhesive bonding between the contacting
surfaces. These adhesive bonds drag the surface of the workpiece, but relative sliding occurs at
the interface due to velocity differential. Fig. 7. 2 shows the direction of the friction force (Ff),
which is opposite to the direction of relative velocity (∆V).
88
Fig. 7. 2: Schematic representation relative velocity of the material and the tool (FN is the normal force, Vmat is the material velocity, Vtool is the tool velocity).
As the tool moves further down, the contact friction between the vertical pin surface and
the workpiece takes place. At this stage, the material around the pin becomes soft and the
bonding between material grains becomes loose due to heat. Consequently, the material moves
in a circular direction around the tool. As the material underneath the pin presses down, the
material around the pin moves up to the surface. This phase generates most of the energy
because the contact and the internal friction occur during the full cycle time.
Sliding friction takes place when the tool’s shoulder contacts the workpiece. Similar to
the friction between the tool tip and the workpiece, asperities at this interface deform because of
the pressure distribution from the shoulder. Although the shoulder has greatest radius, and thus
greatest velocity, the tendency to generate more heat is constrained by the short period of sliding
time. Therefore, the heat generated due to friction is relatively small since the shoulder only
comes in contact with the workpiece at the end of the welding cycle.
89
The temperature that develops due to friction works softens the material in proximity and
reduces the strength of the material. Consequently, the friction force between the tool and the
workpiece decreases over the time.
7.3 Friction at Top Workpiece and Bottom Workpiece Interface
The friction at the interface of the top and the bottom workpieces is insignificant in terms
of heat generation, as compared to the friction at the tool and workpiece interface. This was
shown by the FE simulation results. Since both workpieces are clamped together, the presence
of tangential reaction force at this interface is relatively small. Sliding friction may occur at this
interface due to relative motion between the top and bottom surfaces of the joint. The interface,
however, is very important since the bonding of the materials occurs at this interface. When the
tool shoulder contacts the workpiece and penetrates slightly into the workpiece, a strong
compressive forging pressure is generated. As a result, a solid state bond is formed at the
interface between the two workpieces.
7.4 Internal Friction
Physically, when a solid material is strained, energy, associated with the work done by
the straining process is produced. This energy, in the form of heat is due to the presence of
internal friction forces, which tend to resist the motion of the material.
90
When the tool rotates under pressure on the workpiece, the bonding between the
workpiece material breaks and deforms plastically. Microscopically, this plastic deformation is a
consequence of dislocations. According to the dislocation theory, atoms in a grain boundary slip
when there is crystallographic defect that produces weak points in the bonds between atoms
within crystal structure. The forces that oppose the grains to slip are called internal friction.
Therefore, the work done by the internal friction dissipates energy in the form of heat. In
general, the heat generation due to plastic deformation in FSSW is relatively small as compared
to heat generation from the friction work between the tool and the workpiece. Fig. 7. 3 depicts
the concept of dislocation.
Fig. 7. 3: Schematic showing dislocation of the grains.
The internal friction in a solid can be measured using two methods, i.e., strain-stress
curve and decay of vibration wave. A quantity, known as damping capacity, which is a ratio of
91
the energy lost per cycle due to friction to the total strain energy of the material can be computed
from the strain-stress curve. In the second method, the energy loss can be measured by
computing a logarithmic decrement, δ of vibration decay curve.
7.5 Friction Coefficient between the Tool and the Workpiece
Because of its simplicity, the Coulomb’s friction model is used to described sliding
friction between asperities surfaces. As shown in the FE simulation analysis, friction during
FSSW varies with contact pressure, slip rate and surface temperature. Contact pressure can
influence the near surface crystallographic orientation, which affects the mechanical properties
(like shear strength). This in turn promotes easier shear. When the material of the workpiece is
softened by the heat, the relative velocity (slip rate) increases and consequently, alters the shear
strength of the material. The shear strength of the workpiece depends on the temperature.
Therefore, the friction coefficient is expected to decrease as a function of contact pressure, slip
rate, and surface temperature.
7.6 Heat Generation in FSSW
Heat generation and rising surface temperatures are generally associated with friction that
converts kinetic energy into thermal energy. The energy generated is the result of the tangential
reaction force acting over a distance. In general, the energy from friction work is expressed as
92
∫=L
0Nk dx)x(FE µ 7.1
where, E is the energy, µk is the kinetic friction, FN is the friction force, and x is the distance of
an object moved. According to Blau [40], about 90-95% of the energy due to friction is
transformed into heat and about 5-10 % of the remaining energy is used to deform the material
and some is stored as defects in the contacting material.
In FSSW process, the mechanical interaction, which is due to velocity difference between
the rotating tool and the stationary workpiece, produces heat by friction work. Friction work at
the tool and the workpiece interface can be grouped into three categories based on tool geometry
as shown in Fig. 7. 4. Each contact surface experiences sliding friction during FSSW process.
The total heat produced due to friction work at the tool and the workpiece interface is the
summation of the heat generated at the contact interface of the shoulder, the tip of the pin as well
as the side of the pin. For simplicity in the formulations, the shoulder and the tip of the pin are
assumed flat.
First, let us consider heat generation at the shoulder. The heat generation can be
computed from torque formula. The total torque at the shoulder interface can be expressed as
∫=∫=oR
iRcontact
oR
iRs dr)r2)(r(dMM πτ
7.2
where, Ms is the torque at the shoulder interface, contactτ is the contact shear stress, r is the
distance from the tool axis, and Ro and Ri is the radii of the shoulder and the pin, respectively.
Solving equation 7.2, yields,
93
)RR(32M 3
i3ocontacts −= πτ 7.3
The power exerted to the workpiece is,
ωMP = 7.4
where, n2π=ω , n is the angular velocity of the tool (radians per second).
Substituting equation 7.3 and n2π=ω into equation 7.4, the heat rate at the shoulder, 1P can be
written as
)RR(n34P 3
i3ocontact
2s −= τπ 7.5
The heat generated by the friction work at the tip of the pin and the workpiece interface, pP can
be obtained by similar approach.
3icontact
2p nR
34P τπ= 7.6
The total torque at the vertical pin surface, Mv can be expressed as
∫=L
0iicontactv dy)R2)(R(M πτ
7.7
where L is the height of the pin. Solving equation 7.7, yields
2icontactv LR2M πτ=
7.8
94
Substituting equation 7.8 into equation 7.4 and n2π=ω , the heat rate at vertical pin surface, vP
is obtained as
2icontact
2v nLR4P τπ= 7.9
Therefore, total heat generation from the tool is vpsT PPPP ++= .
2icontact
23icontact
23i
3ocontact
2 nLR4nR34)RR(n
34P τπτπτπ ++−= 7.10
Simplifying equation 7.10, we obtain
)LR3R(n34P 2
i3ocontact
2 += τπ 7.11
Since, µστ =contact , equation 7.11 can be expressed as,
)LR3R(n34P 2
i3o
2 += µσπ 7.12
where, σ is the contact pressure and µ is the friction coefficient.
In FSSW process, the heat rate equation due to plastic deformation can be expresses as
plplP ετ &= 7.13
where, plP is the heat rate due to plastic deformation, τ is the shear stress, and plε& is the plastic
straining rate.
95
Fig. 7. 4: Schematic representation of tool interfaces that generate heat.
96
CHAPTER 8: CONCLUSIONS AND FUTURE WORK
8.1 Conclusions
A fully coupled thermomechanical 3-D FE modeling of FSSW process has been
developed using an Abaqus/Explicit code. The following conclusions can be drawn from this
work.
a) The trend of friction coefficients as a function of contact pressure, slip rate, and surface
temperature has been predicted. The simulation results show that the frictional
dissipation energies for the three friction coefficient dependencies are very close, which
falls within 8.4 % different from each other.
b) The peak temperature obtained from the simulation, which is equivalent to 0.95Ts (Ts is
solidous temperature of Al 6061-T6) is consistent with the theory as reported by Su, et al.
[1].
c) The affect of plunge rate on frictional dissipation energy is also noted. The results
indicate that the lower the plunge rate, the higher the energy. The frictional dissipation
energy for a plunge rate of 5 mm/s reduces to about 29 % of the 1 mm/s frictional
dissipation energy. The same decreasing trend is observed with 10 mm/s plunge rate (14
% decrement).
d) The simulation results also show a significant affect of welding tool’s rotational speed on
frictional dissipation energy. In general, lower rotational speed yields lower frictional
97
dissipation energy. The frictional dissipation energy is reduced by about 8.4 % when the
rotational speed is reduced from 3000 rpm to 2500 rpm and reduced by about 19.2 %
when it is reduced from 3000 rpm to 2000 rpm.
e) Friction work at the interface of the tool and the workpiece generates the most energy,
which is about 96.84 %, for the FSSW process. The rest of the energies come from the
friction work between the plates (0.02 %) and the plastic deformation (3.14 %).
f) The temperature underneath the tool is almost constant at 514.3 0C and gradually
decreases toward the edge of the workpiece.
g) The plastic strain is at the highest point underneath the tool surface and zero at the
interface with no contact with the tool.
h) The material particles can be traced in the adaptive mesh domain. In general, the particle
underneath the pin move more vigorously as compared the particle at the interface with
no contact with the tool.
8.2 Contributions of the Dissertation to the FSSW Technology
A description of friction phenomena have been addressed in this dissertation. The
friction in FSSW can be divided into two categories, i.e., contact pressure and material internal
friction. The contact friction occurs at two interfaces, which are the interface of the tool and the
workpiece and the top and the bottom workpieces interface. The former experiences the most
friction, which is indicated by the most heat generation during the welding process.
98
According to the FE simulation results, the heat generation due to frictional energy is
about 96.84 % of the total energy produced. Based on the parametric studies on rotational
speeds and plunge rates, the results also show that higher rotational speed and slower plunge rate
yield higher generated heat. Therefore, by choosing the right rotational speed and plunge rate,
we can obtain a better weld quality.
A literature survey has been conducted on friction coefficient used to assess friction
between the tool and the workpiece. The friction coefficients, which depend on contact pressure,
temperature and slip rate have been assumed and applied to the same FE model. The FE results
show that the temperature history for friction coefficient temperature dependent is about 5.1 %
different compared to the experimental study done by Gerlich et al. [1]. The dissipation energy
history curves for different friction coefficient (slip rate, contact pressure and temperature
dependant) produce almost the same results. Therefore, the assumed friction coefficient curves
can be used in the FE simulation of FSSW.
8.3 Future Work
Although the contact friction between the workpieces interface is insignificant in terms of
heat generation, the interface is very important because the physical bonding between the two
workpieces occurs at this interface. It is believed that there is an inter-diffusion of material
across the interface at atomic level, which is due to heat and pressure distributions. Therefore,
the research area in FE simulation of FSSW can be extended in a subject of material flow at the
bonding interface.
99
A FE code, namely LS DYNA is capable of simulating the material flow. The
methodology, which uses LS-DYNA has been developed by Zhao [25] in her Ph.D dissertation.
In her work, a “moving mesh” scheme has been deployed with ALE formulations in order to
simulate material flow during FSW process.
100
NOMENCLATURES
TA Total contact area
A, B, C, n Johnson Cook material constants
jiD Rate of deformation
E Energy
F Applied force
fF Friction force
NF Normal force
H Material hardness
L Height of the pin
M Moment
sM Moment at the tool’s shoulder
vM Moment at the vertical surface of the pin
pM Moment at the pin
P Power
101
pP Power generated at the pin
plP Heat generation (power) due to plastic work
sP Power generated at the shoulder
vP Power generated at the vertical surface of the pin
Q Heat generation
oR Radius of the shoulder
iR Radius of the pin
T Temperature
meltT Material melting temperature
refT Reference temperature
Vmat Velocity of material
Vtool Velocity of tool
ia Asperity contact area
ib body force per unit volume
c damping coefficient
102
cp Material specific heat
if Normal force exerted on asperity
h Convection heat transfer
k stiffness coefficient
kx, ky, kz heat conductivity in x, y, and z directions
n Angular velocity, rad./s
r Distance from the tool axis
iq Heat flux per unit volume
u&& Nodal acceleration
u& Nodal velocity
u displacement
v Velocity
intw Internal energy per unit volume
C Viscous damping matrix
C(t) Time dependent capacitance matrix
F External force vector
103
fF Friction force vector
K Stiffness matrix
K(t) Time dependent conductivity matrix
M Mass matrix
M1 Second order friction coefficient tensor
Q(t) Time dependent heat vector
T& Derivative of temperature
T Nodal temperature vector
ej Arbitrary unit vector basis
ki Orthogonal unit vector basis
u&& Nodal acceleration vector
u& Nodal velocity vector
u displacement vector
v Slip velocity unit vector
σ Contact pressure
ijσ Cauchy stress
104
yσ Yield stress
plε Effective plastic strain
plε& Plastic strain rate
0ε& Normalizing strain rate
µ Friction coefficient
sµ Static friction coefficient
kµ Kinetic friction coefficient
ρ Mass density
τ Shear stress
yτ Shear strength
contactτ Contact shear stress
ω Angular velocity, rpm
∆t Time increment
∆V Relative velocity
105
ABBREVIATIONS
ALE Arbitrary Lagrangian Eulerian
CFD Computational Fluid Dynamics
DOF Degree of freedom
FE Finite element
FSW Friction stir welding
FSSW Friction stir spot welding
HAZ Heat affected zone
ORNL Oak Ridge National Laboratory
SPH Smoothed Particle Hydrodynamics
TMAZ Thermo-mechanically affected zone
2-D Two dimensional
3-D Three dimensional
106
REFERENCES
[1 ] Gerlich, A., Su, P., and North, T., 2005, “Peak Temperatures and Microstructures in
Aluminum and Magnesium Alloy Friction Stir Spot Welds”, Science and Technology of
Welding and Joining, vol. 10, p.647-652.
[ 2 ] Thomas, M., Nicholas, J., Needham, J., Murch, M., Templesmith, P., and Dawes, C.,