Journal of Materials Processing Technology 169 (2005) 320–327

3-Dimensional CFD modelling of flow round

a threaded friction stir welding tool profile
Paul A. Colegrove a,b,∗ , Hugh R. Shercliff a
a Cambridge University, Department of Engineering, Trumpington Street, Cambridge CB2 1PZ, UK
b TWI Ltd., Granta Park, Great Abington, Cambridge CB1 6AL, UK

Received 29 August 2003; accepted 22 March 2005

Abstract

This paper describes the application of the computational fluid dynamics (CFD) code, FLUENT, to modelling the 3-dimensional metal
flow in friction stir welding (FSW). A standard threaded tool profile is used for the analysis and features such as the tool rake angle, heat
generation and heat flow are included. The primary goal is to gain a better understanding of the material flow around a complex FSW tool and
to demonstrate the effect of the tool rake angle, and weld and rotation speed. The model captured many of the real process characteristics, but
gave poor predictions of the welding forces. The model also generated an excessive amount of heat, which led to a large over-prediction in
the weld temperature. These shortcomings can be overcome by using a viscosity relationship that includes material softening near the solidus
or material slip at the tool interface.
© 2005 Elsevier B.V. All rights reserved.

Keywords: Friction; Stir; Weld; Flow; Modelling

1. Introduction Shercliff and Colegrove [3] have reviewed modelling

Friction stir welding (FSW) is a relatively new welding
process, patented in 1991 by Thomas et al. [1,2]. This pro-
cess has great advantages in welding aluminium alloys that
are difficult to weld. The process gives low distortion, can
weld thick sections in a single pass and produces welds with
excellent mechanical properties.
A schematic diagram illustrating the process of FSW is
shown in Fig. 1. The key components of the FSW tool are:
• The shoulder: this is the primary means of generating heat
during the process, prevents material expulsion and assists
material movement around the tool.
• The pin: the pin’s primary function is to deform the mate-
rial around the tool and its secondary function is to generate
heat.
∗ Corresponding author. Tel.: +44 1223 332791.
E-mail addresses: pac44@cam.ac.uk (P.A. Colegrove),
hrs@eng.cam.ac.uk (H.R. Shercliff).
activities in FSW. One of the great challenges is the devel-
opment and validation of a 3-dimensional flow model. This
paper describes the development of such a model for the 5651
tool profile developed by Dawes et al. [4]. Bendzsak et al. [5]
and Askari et al. [6] have developed flow models for this
tool.
The model published by Bendzsak et al. [5] used the finite
volume method to calculate the flow around the tool. Arrow
plots and particle tracks were used to visualise the flow field,
which was greater in size than that observed experimen-
tally. The flow indicated by the model was divided into two
regions. Near the shoulder, it was largely rotational while
further down it appeared to extrude around both sides of
the tool. Spirals were observed between these two regions
and it was postulated that these could be the cause of weld
defects.
The work presented by Askari et al. [6] used the CTH
hydro-code and is the most advanced flow modelling pub-
lished to date. The model assumed that the material stuck to
the tool surface and used the Johnson and Cook [7] relation-
ship to represent the material flow stress. The model used

0924-0136/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jmatprotec.2005.03.015
 P.A. Colegrove, H.R. Shercliff / Journal of Materials Processing Technology 169 (2005) 320–327
Fig. 1. Schematic diagram of friction stir welding.
elasto-viscoplastic material properties and very good agree-
ment was obtained between the model output and marker
experiments with the flow being largely confined to the region
adjacent to the pin. The temperatures from the model com-
pared favourably with those obtained experimentally. Com-
parisons between the weld power input and forces were not
reported.
Other authors have analysed the 3-dimensional flow
around unthreaded pins using analytical [8], finite element
[9–11], finite difference [12] and Arbitrary Lagrangian Eule-
rian (ALE) [13–15] techniques. The ALE approach used by
Schmidt and Hattel [14,15] shows particular promise and can
predict void formation in the weld which is impossible with
the CFD approach used in this work.
The aim of the present work is to demonstrate the benefits
and pitfalls of using CFD to model FSW. To do this a flow
model was developed that included:
(a) coupled thermal/flow analysis;
(b) heat generation by viscous dissipation;
(c) full thermal boundary conditions, such as heat loss to the
backing plate and tool, and the convective heat loss on
the top surface of the plate;
(d) the tool rake angle.
The CFD package FLUENT [16] was selected for the
modelling work because of its ability to handle the high strain-
rates that occur near the tool surface.
2. Description of the model
2.1. Model regions
The models were 3-dimensional and assumed that the
material stuck to the tool surface. Two different material defi-
nitions are available within the FLUENT solver. Firstly, there
is the fluid region where both the momentum and heat equa-
tions are solved. Secondly, there is the solid region where only
the heat equation is solved. This is used where the material
being modelled is either a solid, or where material movement
is sufficiently small that solution of the momentum equa-
321
Fig. 2. Description of the mesh used for the models.
tion can be neglected. Neither region includes the combined
elasto-viscoplastic response of the material.
The FSW flow model considers a rotating tool with the
material translating past it at the welding speed. The work-
piece material is subdivided into the five regions shown in
Fig. 2. Firstly, a rotating mesh that moves at the rotation speed
of the tool represents a fluid region adjacent to the tool. This is
embedded in a stationary mesh slightly wider than the rotating
region, extending to the length of the plate, which models the
flow and thermal response of the material that passes through
the deformation zone. To either side a solid aluminium region
models the thermal response in the non-deforming region,
which is known to be travelling at the welding speed. Finally,
the tool and backing plate are solid regions that are used to
calculate the heat loss to these components.
2.2. Solver type and boundary conditions
The model was solved as a steady-state problem. While
this solution technique is valid for an axi-symmetric or cylin-
drical tool, the solution will only be an approximation for
a threaded tool. In this case, the solution is a ‘snap-shot’
of the flow at a particular instant in time. The approxima-
tion is reasonable provided the time-dependent terms in the
Navier–Stokes equation remain small. The benefit of this
approach over a full transient analysis is much faster solu-
tion times.
The model assumed that the material sticks to the tool
surface, so the material velocity equalled the rotation speed
multiplied by the radius. The translation is prescribed by
allowing the material to flow past the tool. Therefore, the
material enters the model at a velocity equal to the welding
speed, and at ambient temperature.
Both isothermal and non-isothermal models were solved.
The isothermal models used a representative temperature of
800 K, while the non-isothermal models included the heat
equation and calculated the heat generation from viscous
dissipation. In these models, the heat loss through the tool
was calculated by defining a 1-dimensional conduction zone
322 P.A. Colegrove, H.R. Shercliff / Journal of Materials Processing Technology 169 (2005) 320–327

at the shoulder. This calculation assumed that the far surface

was at ambient, and the material was 100 mm thick tool steel.
To model the imperfect contact between the aluminium and
backing plate a contact resistance of 1000 mm2 K/W from
Colegrove et al. [17] was used. The temperature on the bot-
tom of the backing plate far wall was set to ambient. Finally,
a convective heat transfer coefficient of 10 W/m2 K [17] was
used for the convective heat loss from the top surface of the
plate. Some of the models use a raked region around the tool
shoulder so that the tool could be inclined away from the
direction of travel. This is described in greater detail in Sec-
tion 2.4.

2.3. Material properties

The aluminium alloys 5083 and 7075-T6 aluminium

alloys were used for the analysis. The material property of
most interest was the viscosity, which was found from the
normal stress and strain-rate via the relationship described
in Appendix A. The flow stress of 5083 was found from the
Sellars and Tegart [18] constitutive law:


Q
Z = ε̇ exp = A(sinh ασ)n (1)
RT
where ε̇ is the strain-rate (/s), Q the activation energy (J/mol),
R the universal gas constant (J/kg K), T the temperature (K)
and A, α, n are material constants.
Values of Q = 161 kJ/mol, A = 1.62/s, α = 0.019 /MPa and
n = 3.2 from Sheppard and Tutcher [19] were used for the
analysis.
The flow stress for 7075-T6 was found by interpolating
experimental stress versus strain-rate data from Jin et al.
[20]. The values were interpolated on a logarithmic scale
with strain-rate and on a linear scale with temperature. Both
of these relationships were implemented in FLUENT with
user-defined code. Importantly, unlike Seidel and Reynolds
[21], neither viscosity relationship included material soften-
ing near the solidus, which led to the over-prediction of weld
power and temperature reported in subsequent sections.
Including material softening in FLUENT is difficult because
the large change in viscosity near the solidus can lead to
an unstable solution. Further development in FLUENT
used a slip boundary condition to capture the low viscosity
behaviour.
Other authors [6,14,15] have used the Johnson and Cook
[7] constitutive model, which is more numerically robust
Fig. 3. Dimensions of the 5651 tool from Dawes et al. [4] used for the
models.
because it avoids large changes in viscosity. However, the
strain-rate sensitivity at high temperature diminishes to zero
with this relationship. Experiments on 7075 aluminium alloy
have shown that this is not the case. Nevertheless, good pre-
dictions can be made with models that use this constitutive
behaviour.
The thermal properties used for the models are shown in
Table 1. The values are independent of temperature in all
cases except for the specific heat and thermal conductivity
of the aluminium alloys where temperature-dependent values
from Mills [22] were used. Note that the values from Mills for
5083 were actually for 5182, which has a similar composition.
2.4. Model types
The models generated were based on the design of the
5651 tool developed by Dawes et al. [4]. Fig. 3 shows the
dimensions of this tool. Several steps were required to gen-
erate a mesh of the material around the tool. Firstly, a solid
model of the tool was created in Mechanical Desktop [25].
This was exported into the FLUENT mesh generation pro-
gram GAMBIT [26]. In GAMBIT, the geometry was mod-
ified around the bottom and top of the tool threads to avoid
the generation of highly distorted elements. Finally, regions

Table 1
Thermal properties of the materials used in the thermal model
Material name Density (kg/m3 ) Specific heat (J/kg K) Thermal conductivity (W/m K)
5083/7075-T6 Aluminium 2660/2800b Millsa Millsa
Steel (0.5% C steel) at 150 ◦ C 7833c 465c 50c
H13 tool steel (5% chrome) at 300 ◦ C 7833c 460c 36c
a Ref. [22].
b Ref. [23].
c Ref. [24].
 P.A. Colegrove, H.R. Shercliff / Journal of Materials Processing Technology 169 (2005) 320–327
Table 2
Summary of the welding conditions used for the models
Rotation speed Welding speed (mm/min)
(rpm)
60 90 120
750 5651-raked
500 5651-raked 5651-raked 5651-raked
5651-normal (+ a model
that uses 7075 aluminium
with isothermal
properties)
Unthreaded tool
250 5651-raked
Both isothermal (800 K) and thermal models were solved using the properties
of 5083 aluminium alloy.
were created around the tool in accordance with Section 2.1,
which were then meshed.
Two models of the 5651 tool were generated: in one the
tool was normal to the workpiece (5651-normal) and in the
other it was raked at 2.5◦ away from the direction of travel
(5651-raked). Several alternative methods for raking the
tool were attempted, however, the mesh shown in Fig. 2
which used a ‘prow’ around the shoulder was the only one
that was successful. A final model (unthreaded tool) was an
unthreaded version of the ‘5651-normal’ model described
above.
As seen in Fig. 2, the inner part of the tool protruded above
the shoulder. This avoided the creation of highly distorted
elements around the fillet where the pin meets the shoulder.
Tetrahedral elements were used and a spacing of 0.3 mm was
used on the surface of the tool.
In the ‘5651-raked’ model the plate thickness was
6.35 mm, while the other two models where the tool was
normal to the surface used a plate thickness of 6.95 mm.
This is because the tool protrudes further into the plate when
it is normal to the surface. All the models used a back-
ing plate with a thickness of 40 mm and width of 140 mm.
Finally, all the models extended 70 mm ahead of the weld and
100 mm behind. In this preliminary study, tests on the influ-
ence of mesh density and the effect of plate size were not
undertaken.
Table 2 describes the models and welding parameters used.
Note that unless otherwise stated, two versions of each were
solved: one was an isothermal (800 K) model and the other
was a thermal model. Both used the material properties of
5083 aluminium alloy. The 7075 properties were only used
for one isothermal model, since the full thermal model could
not be solved due to the high heat generation and excessive
temperatures produced with these properties.
3. Results
The results primarily focus on the thermal model of the
‘5651-raked’ model that used a welding speed of 90 mm/min
and a rotation speed of 500 rpm. This is the so-called ’standard
323
model’. The other models are described where their results
are significantly different.
3.1. Arrow plots and deformation region
Fig. 4 shows vector plots of in-plane velocity on horizontal
planes at positions z = 0.1, 3.175 and 6 mm from the bottom
face, for the standard model. Referring to Fig. 4a, the material
under the dome of the tool (root) is influenced by the rotat-
ing tool above. At mid-thickness (Fig. 4b) there is a large
rotating region around the tool. Overlayed on this image is a
line where the strain-rate equals 2/s, which approximates the
boundary between the rotating and stationary flow regions.
Finally, Fig. 4c shows the rotating material under the shoul-
der. Note that in all cases, there is a stagnation point on the
advancing side, and all the material up to this point flows
round the retreating side of the tool.
Fig. 5 shows a longitudinal vertical section that compares
the size of the deformation regions between an isothermal
and a thermal model. Interestingly very little difference is
observed between the two, which contrasts with the dramatic
reduction observed in a model for 7075-T6 that included a
thermal profile in Colegrove and Shercliff [27]. In the latter
work, a large increase in material strength occurs outside
the deformation region, but this is far more gradual with the
Sellars and Tegart material properties for 5083 used in the
current work.
Streamlines of the flow around the tool are shown in Fig. 6.
Note that streamlines are instantaneously tangential to the
flow and are shown for a particular tool orientation. Parti-
cle tracks are much more difficult to calculate and require
a full transient analysis. The planar view (Fig. 6a) demon-
strates how the material is swept around the retreating side
of the tool. The deformation zone is large compared with 2-
dimensional models [28], experimental results [29,30] and
later modelling work [27]. The deformation zone is larger
than the 2-dimensional models because of the additional
stirring induced by the shoulder. The later modelling work
uses a different constitutive model and includes slip at the
tool/material interface. The large tetrahedral mesh used for
the model may also expand the flow region.
The longitudinal streamlines in Fig. 6b show very little
deviation in height as the material passes the tool. This is
in contrast to the quite large vertical movement of mate-
rial observed by Colligan [31]. The problem is caused
by the model’s over-prediction of the deformation region
size.
3.2. Pressure plot
The pressure distribution showed how raking the tool
increased the pressure behind the tool, but gave little insight
into the effect of tool shape on the process. Further details
are provided in Colegrove [32]. Force comparisons, which
are discussed below are more easily quantified and provide
more practical guidance.
324 P.A. Colegrove, H.R. Shercliff / Journal of Materials Processing Technology 169 (2005) 320–327

Fig. 5. Comparison of deformation zones between isothermal and thermal

versions of the 5651-raked model.

3.3. Temperature, force and power comparison

The temperature contours from the raked model in Fig. 7

demonstrate how the contact resistance inhibits the heat flow
to the backing plate. The peak temperature of all the thermal
models that used a welding speed of 90 mm/min and a rotation
speed of 500 rpm was approximately 860 ◦ C, which is well in
excess of the welding temperature of 550 ◦ C expected for this
material [17]. The model type had little effect on this value.
As discussed in Colegrove and Shercliff [33], one possible
reason for this discrepancy is that in a real weld slip may occur
between the tool and welded material, which reduces the heat
input and avoids material melting. An alternative explanation
suggested by Seidel and Reynolds [21] is that the material
significantly softens at temperatures approaching the solidus
which reduces the heat generation and the weld temperature
predicted by the model. Either effect will significantly reduce
the heat generation.

Fig. 4. Arrow plots across the (a) z = 0.1, (b) z = 3.18 and (c) z = 6 planes for
the 5651-raked thermal model. Note: (b) includes an overlay of the strain-
rate = 2/s boundary; although advancing and retreating side labels are shown
only for (a) they apply equally to (b) and (c). Fig. 6. Streamlines for 5651-raked thermal model (a) planar from a start
height of 3 mm; (b) transverse longitudinal section.
 P.A. Colegrove, H.R. Shercliff / Journal of Materials Processing Technology 169 (2005) 320–327 325

Fig. 7. Temperature contours for 5651-raked model. (Units of temperature

are ◦ C.)

Fig. 8a shows the effect of the welding parameters on the

peak temperature. The welding speed had little effect, while
increasing the rotation speed caused a significant increase in
the peak temperature. Obviously, this is caused by the greater
heat production from the higher rotational speed. Experi-
mentally, it has been shown that when welding 7075-T6
[33] and 6061 [34] aluminium alloys, the effect of chang-
ing the welding or rotation speeds on the peak temperature is
small, for conditions that produce sound welds. Note that the
effect of changing the rotation speed is greater at low rotation
speeds.
The weld power was found by multiplying the torque
by the rotation speed. The power input from translation is
approximately 1% of this value and can be ignored. The
results showed that the weld power was independent of model
type with 3.3 kW required for the thermal models. Fig. 8b
shows the effect of the welding parameters on the weld power.
These results reflect the peak temperature results in Fig. 8a in
that increasing the rotation speed increases the power input
to the weld, while the welding speed had little impact. The
Fig. 8. Comparisons between (a) peak temperature, (b) weld power and (c),
latter result suggests that, because of the relatively low weld- traversing force for the 5651-raked tool.
ing speed being used, the process is largely axi-symmetric
with the heat being primarily lost through the tool, backing
plate and by convection. The heat required to raise the mate-
rial to weld temperature is a relatively small component. This the isothermal model. This was a consequence of the lower
model effect is not reflected in practice [32,35]. The rotation flow stress at the higher temperature. Nevertheless, both are
speed had a much stronger influence on the heat input for a factor of 10 lower than the experimental values from John-
the isothermal models because there was no softening effect, son [37] for this alloy and tool type. It is believed that this
i.e., in the thermal models the flow strength reduces as the result could be improved by using a model that either included
temperature increases which offsets the increase in heat from viscosity softening near the solidus or material slip. Both
the higher rotation speed. would reduce the bulk movement of material around the tool,
Fig. 8c shows the effect of the welding parameters on increasing the traversing force.
the traversing force. It increased with welding speed because Finally, some interesting results were obtained when com-
of the greater amount of material needing to pass the tool. paring the forces between the three models. Fig. 9 shows how
The traversing force reduced with increasing rotation speed, the traversing force with the ‘5651-raked’ model was consid-
which is not observed experimentally [33], with the opposite erably lower than the models where the tool was normal to the
often being observed [36]. This demonstrates the inadequa- surface. The main cause of this effect is the pressure redistri-
cies of the current modelling approach. The traversing forces bution behind the tool, i.e., the higher pressure caused by the
with the thermal models were slightly lower than those with forging of material lowers the force needed to push the tool
326 P.A. Colegrove, H.R. Shercliff / Journal of Materials Processing Technology 169 (2005) 320–327

of these problems can be addressed in FLUENT by either

using a slip model or changing the viscosity so that its value
reduces dramatically near the solidus temperature.

Acknowledgements

The authors wish to express their thanks to the Uni-

versity of Adelaide, The Welding Institute (TWI), the UK
Department of Trade and Industry (DTI), the University of
Cambridge, the Post-Graduate Training Partnership (PTP),
and the Cambridge Commonwealth Trust for their financial
support. The advice given by Philip Threadgill was greatly
Fig. 9. Force comparison between the different tool types. All models used
appreciated.
a welding speed of 90 mm/min and a rotation speed of 500 rpm.

through the material. It is unclear whether the lower force is Appendix A. Determining the viscosity from the
real or an artefact of the model. normal stress and strain-rate
The unthreaded tool had a lower traversing force than
the threaded version. This result is consistent with the 2- The equation for the effective stress is given by Bathe [38]:
dimensional stick model analysis in Colegrove and Shercliff 
[28], which showed that tools with features had a higher 3
traversing force than their cylindrical counterparts. Slip mod- σ̄ = S·S (A.1)
2
els of the same profiles indicated that the tools with features
assisted the movement of material round the tool, which low- where S is the deviatoric stress tensor. The equation for the
ered the traversing force. The latter result is more consistent effective strain-rate is given by Bathe [38]:
with experimental findings. Finally, the model that used 7075 
2
properties had an even higher traversing force due to the ¯ε̇ = ε̇ · ε̇ (A.2)
3
higher flow stress of this material.
where ε̇¯ is the strain-rate tensor given by:
1 
4. Conclusions ε̇ = ∇v + (∇v)T (A.3)
2
This work has demonstrated how the CFD package, FLU- where v is the velocity vector. The equation relating the devi-
ENT can be used to analyse material flow in FSW. The atoric stress tensor to the strain-rate is given in FLUENT [16].
models primarily used a threaded tool, which was raked 2.5◦ (Note that the bulk viscosity term is ignored.)
away from the direction of travel. This model was compared
against ones that used a normal threaded tool and a normal S = 2µε̇ (A.4)
unthreaded tool. Two model types were solved: an isothermal where µ is the viscosity. The deviatoric stress is defined as:
model and one that included the viscous heat generation and
solved the resulting thermal profile. S = ␴ + pI (A.5)
All the models predicted that the material in line with the
deformation zone was swept around the retreating side of the where ␴ is the stress and p is the pressure (positive in com-
pin. The amount of material swept around the pin increased pression). Note that p is defined as:
at locations closer to the shoulder. It was noted, however, that 1
the size of the deformation zone was much larger than that σm = σii = −p (A.6)
3
observed experimentally.
In addition to the excessive deformation zone size, the Substitute equation (A.4) into (A.1):
models experienced several additional problems. Firstly, 
σ̄ = 6µ2 ε̇ · ε̇
using representative material properties for 5083 and 7075
(without viscosity softening near the solidus), there is a large 
over-prediction of the weld temperature and the weld power. 2
σ̄ = µ 9 × ε̇ · ε̇
Secondly, the rotation speed had a large impact on the peak 3
welding temperature, which is relatively small in practice. Substituting from equation (A.2) gives:
Thirdly, the traversing force was found to be an order of
magnitude lower than that measured experimentally. Many σ̄ = 3µε̇¯
 P.A. Colegrove, H.R. Shercliff / Journal of Materials Processing Technology 169 (2005) 320–327
Hence, the viscosity is given by:
σ̄
µ= . (A.7)
3ε̇¯
References
[1] W.M. Thomas, E.D. Nicholas, J.C. Needham, M.G. Murch, P.
Temple-Smith, C.J. Dawes, Friction stir butt welding, Int. Patent
Application no. PCT/GB92/02203; GB Patent Application no.
9,125,978.8, 1991; US Patent no. 5,460,317, 1995.
[2] W.M. Thomas, E.D. Nicholas, J.C. Needham, M.G. Murch, P.
Temple-Smith, C.J. Dawes, Improvements relating to friction weld-
ing, International Patent Classifications B23K 20/12, B29C 65/06,
1993.
[3] H.R. Shercliff, P.A. Colegrove, Modelling of friction stir welding, in:
H. Cerjak, H.K.H.D. Bhadeshia (Eds.), Mathematical Modelling of
Weld Phenomena 6, Maney Publishing, London, 2002, pp. 927–974.
[4] C.J. Dawes, P.L. Threadgill, E.J.R. Spurgin, D.G. Staines, Develop-
ment of the new friction stir technique for welding aluminium, TWI
Group Sponsored Report no.: 5651/35/95, Cambridge, UK, 1995.
[5] G.J. Bendzsak, T.H. North, C.B. Smith, An experimentally vali-
dated 3D model for friction stir welding, in: Proceedings of the
Second International Symposium on ‘Friction Stir Welding’, TWI
Ltd., Gothenburg, Sweden, 2000.
[6] A. Askari, S. Silling, B. London, M. Mahoney, Modeling and anal-
ysis of friction stir welding processing, in: K.V. Jata, et al. (Eds.),
Friction Stir Welding and Processing, TMS, Warrendale, PA, 2001,
pp. 43–54.
[7] G.R. Johnson, W.H. Cook, A constitutive model and data for metals
subjected to large strains, high strain rates and high temperatures, in:
Proceedings of the Seventh International Symposium on ‘Ballistics’,
The Hague, Netherlands, 1983, pp. 541–548.
[8] P. Heurtier, C. Desrayaud, F. Montheillet, A thermomechanical anal-
ysis of the friction stir welding process Materials Science Forum,
vol. 396–402, Trans Tech Publications, Switzerland, 2002, pp.
1537–1542.
[9] S. Xu, X. Deng, Two and three-dimensional finite element models
for the friction stir welding process, in: Proceedings of the Fourth
International Symposium on ‘Friction Stir Welding’, TWI Ltd., Park
City, Utah, May 2003.
[10] P. Dong, F. Lu, J.K. Hong, Z. Cao, Coupled thermomechanical anal-
ysis of friction stir welding process using simplified models, Sci.
Technol. Weld. Joining. 6 (5) (2001) 281–287.
[11] P. Ulysse, Three-dimensional modeling of the friction stir-welding
process, Int. J. Mach. Tools Manufacture 42 (2002) 1549–1557.
[12] F. Palm, U. Henneböhle, V. Erofeev, E. Earpuchin, O. Zaitzev,
Improved verification of FSW-process modelling relating to the
origin of material plasticity, in: Proceedings Fifth International Sym-
posium on ‘Friction Stir Welding’, TWI Ltd., Metz, France, 2004.
[13] L. Fourment, S. Geurdoux, M. Miles, T. Nelson, Numerical sim-
ulation of the friction stir welding process using both Lagrangian
and Arbitrary Lagragian Eulerian formulations, in: Proceedings of
the Fifth International Symposium on ‘Friction Stir Welding’, TWI
Ltd., Metz, France, 2004.
[14] H. Schmidt, J. Hattel, Modelling thermomechanical conditions at the
tool/matrix interface in friction stir welding, in: Proceedings of the
Fifth International Symposium on ‘Friction Stir Welding’, TWI Ltd.,
Metz, France, 2004.
[15] H. Schmidt, J. Hattel, A local model for the thermomechanical con-
ditions in friction stir welding, Modell. Simul. Mater. Sci. Eng. 13
(2005) 77–93.
327
[16] FLUENT, Release 6.0.20, Fluent Incorporated, Lebanon, NH,
2002.
[17] P. Colegrove, M. Painter, D. Graham, T. Miller, 3-Dimensional flow
and thermal modeling of the friction stir welding process, in: Pro-
ceedings of the Second International Symposium on ‘Friction Stir
Welding’, TWI Ltd., Gothenburg, Sweden, 2000.
[18] C.M. Sellars, W.J.Mc.G. Tegart, Int. Met. Rev. 17 (1972) 1–24.
[19] T. Sheppard, M.G. Tutcher, Effect of process parameters on structure
and properties of Al–5Mg–0.8Mn alloy (AA5456), Met. Tech. 8
(1981) 319–327.
[20] Z. Jin, W.A. Cassada, C.M. Cady, G.T. Gray, Mechanical Response
of AA7075 Aluminium Alloy over a Wide Range of Temperatures
and Strain Rates, vol. 331–337, Material Science Forum, Switzer-
land, 2000, pp. 527–532.
[21] T.U. Seidel, A.P. Reynolds, A 2-D friction stir welding process model
based on fluid mechanics, Sci. Technol. Weld. Joining 8 (3) (2003)
175–183.
[22] K.C. Mills, Recommended Values of Thermophysical Properties for
Commercial Alloys, Woodhead Publishing Ltd., Cambridge, UK,
2002.
[23] J.R. Davis, ASM Speciality Handbook: Aluminium and Aluminium
Alloys, ASM International, 1993.
[24] J.P. Holman, Heat transfer, in: SI Units, seventh ed., McGraw-Hill
Book Company, London, 1990.
[25] Mechanical Desktop 5, Autodesk® , San Rafael, CA, 2000.
[26] GAMBIT, Release 2.0.4, Fluent Incorporated, Lebanon, NH,
2002.
[27] P.A. Colegrove, H.R. Shercliff, Development of the TrivexTM fric-
tion stir welding tool. Part II. 3-Dimensional flow modelling, Sci.
Technol. Weld. Joining 9 (4) (2004) 352–361.
[28] P.A. Colegrove, H.R. Shercliff, 2-Dimensional CFD modelling of
flow round profiled FSW tooling, Sci. Technol. Weld. Joining 9 (6)
(2004) 483–492.
[29] A.P. Reynolds, Visualisation of material flow in an autogenous
friction stir weld, Sci. Technol. Weld. Joining 5 (2) (2000) 120–
124.
[30] B. London, M. Mahoney, W. Bingel, M. Calabrese, D. Waldron,
Experimental methods for determining material flow in friction stir
welds, in: Proceedings of the Third International Symposium on
‘Friction Stir Welding’, TWI Ltd., Kobe, Japan, 2001.
[31] K.J. Colligan, Material flow behaviour during friction stir welding
of aluminium, Weld. J. 78 (7) (1999) 229–237.
[32] P. A. Colegrove, Modelling of Friction Stir Welding, Ph.D. Thesis,
The University of Cambridge, Cambridge, UK, 2003.
[33] P.A. Colegrove, H.R. Shercliff, Experimental and numerical analysis
of aluminium alloy 7075-T7351 friction stir welds, Sci. Technol.
Weld. Joining 8 (5) (2003) 360–368.
[34] M. Song, R. Kovacevic, Numerical and Experimental Study of the
Heat Transfer Process in Friction Stir Welding, Proc. Inst. Mech.
Eng. Part B: J. Eng. Manufacture 1 (2003) 73–85.
[35] A.P. Reynolds, W. Tang, Alloy, tool geometry and process parameter
effects on friction stir weld energies and resultant FSW joint proper-
ties, in: K.V. Jata, et al. (Eds.), Friction Stir Welding and Processing,
TMS, Warrendale, PA, 2001, pp. 15–23.
[36] J. Yan, M.A. Sutton, A.P. Reynolds, Process-structure-property rela-
tionships for nugget and HAZ regions of AA2524-T351 FSW Joints,
in: Proceedings of the Fifth International Symposium on ‘Friction
Stir Welding’, TWI Ltd., Metz, France, 2004.
[37] R. Johnson, Forces in friction stir welding of aluminium alloys—
further studies, TWI Core Research Report no. 716/2000, Cambridge,
UK, 2000.
[38] K.J. Bathe, Finite Element Procedures, Prentice Hall, Englewood
Cliffs, NJ, 1996.