metals

Article
Effect of the Tool Tilt Angle on the Heat Generation
and the Material Flow in Friction Stir Welding
Narges Dialami * , Miguel Cervera and Michele Chiumenti
International Center for Numerical Methods in Engineering (CIMNE), Technical University of Catalonia,
Campus Norte UPC, 08034 Barcelona, Spain; [email protected] (M.C.); [email protected] (M.C.)
* Correspondence: [email protected]; Tel.: +34-93-401-6529




Received: 5 December 2018; Accepted: 24 December 2018; Published: 29 December 2018


Abstract: This work studies the effect of the tool tilt angle on the generated heat and the material flow
in the work pieces joint by Friction Stir Welding (FSW). An apropos kinematic framework together
with a two-stage speed-up strategy is adopted to simulate the FSW problem. The effect of tilt angle
on the FSWelds is modeled through the contact condition by modifying an enhanced friction model.
A rotated friction shear stress is proposed, the angle of rotation depending on the process parameters
and the tilt angle. The proposed rotation angle is calibrated by the experimental data provided for a
tilt angle 2.5◦ . The differences of generated heat and material flow for the cases of tool with tilt angle
of 0◦ and 2.5◦ are discussed. It is concluded that due to the higher temperature, softer material and
greater frictional force in the trailing side of the tool, the material flow in the rear side of the FSW tool
with the title angle is considerably enhanced, which assists to prevent the generation of defect.

Keywords: FSW; tilt angle; friction; material flow

1. Introduction
Friction Stir Welding (FSW) uses a tool with a high rotating speed which moves forward between
the pieces to be joined and generates heat. The main function of the tool (consisting of pin and
shoulder) is to mix the work piece material and to generate heat by friction. The final properties of
friction stir welds depend on factors such as the process parameters (advancing and rotating speed),
the tool design and the tool tilt angle [1–3]. In previous works, the authors have studied the effects of
the tool velocity [4] and the tool design [5]. In this work, the effect of the tool tilt angle is addressed.
Figure 1 presents a cross-sectional view of an (exaggeratedly) tilted tool inside the work piece.
Typical tilt angles used in practice are between 0◦ and 3◦ , where a zero value signifies that the tool
is perpendicular to the work piece. The tool tilt angle affects the material flow during the weld and
thus the heat generation. In FSW, the heat is generated by friction and plastic dissipation. As the
mechanical properties are notably temperature-dependent, material flow and heat generation are
dependent on each other, making FSW a strongly coupled thermo-mechanical problem. The tool tilt
angle has a fundamental importance for the weld quality in FSW. On the one hand, a non-zero tilt
angle ensures the contact among the tool shoulder and the work piece; moreover, it facilitates the flow
of the material around the tool. On the other hand, an inadequately large tilt angle raises the pin from
the weld root, resulting in damaged welds. Consequently, it is essential to properly choose the tool tilt
angle. An optimal tool tilt angle guarantees that the tool shoulder imprisons the deformed material
and transports it proficiently from the front edge to the rear side of the pin [6].
The tilt angle of the tool and its noticeable effect on the final post-weld quality has been studied
by several investigators [7–9]. These studies show that the tool tilt angle has a significant effect on the
formation of defects during the weld. The optimal tool tilt angle facilitates the material flow around
the tool and avoids the formation of defects in the weld zone. Several experimental tests have to be

Metals 2019, 9, 28; doi:10.3390/met9010028 www.mdpi.com/journal/metals

 Metals 2018, 8, x FOR PEER REVIEW 2 of 17

The tilt angle of the tool and its noticeable effect on the final post-weld quality has been studied
by several investigators [7–9]. These studies show that the tool tilt angle has a significant effect on the
Metals 2019, 9, of
formation 28 defects during the weld. The optimal tool tilt angle facilitates the material flow around
2 of 17
the tool and avoids the formation of defects in the weld zone. Several experimental tests have to be
performed to obtain the optimal tilt angle. However, the fundamental mechanism of the tilt effect on
performed to obtain the optimal tilt angle. However, the fundamental mechanism of the tilt effect on
heat generation and material flow is yet to be understood.
heat generation and material flow is yet to be understood.

Figure1.1.Cross-sectional
Figure Cross-sectionalview
viewof
ofan
an(exaggeratedly)
(exaggeratedly)tilted
tiltedtool
toolinside
insidethe
thework
workpiece.
piece.

Reshad
Reshadetetal. al.[10]
[10]study
studythe theeffects
effectsof ofthe
thetool
tooltilttiltangle
angleon onFSW
FSWof ofpure
puretitanium.
titanium.They Theyconsidered
considered
several
severaltesttestcases
caseswhere
wherethe thetooltooltilt
tiltangle
angleisisvaried
variedand andthe theeffect
effectofofthis
thisvariation
variationon onthethepost-weld
post-weld
properties ◦
propertiesisisexamined.
examined. They They obtained
obtained 11° as as the
the best
best tilt
tiltangle
anglefor foraadefect-free
defect-freewelding
weldingwith withhighhigh
mechanical
mechanicalproperties.
properties.
Banik
Baniketetal.al.[11]
[11]examine
examineweld weldqualities
qualitiesof ofFSW
FSWAA6061-T6
AA6061-T6from fromthe thepoint
pointof ofview
viewof ofthe
thefinal
final
mechanical
mechanicalproperties
propertiesofofthe theworkworkpiecepieceby bychanging
changingthe thetool
tooltilt
tiltangles
anglesfor fortaper
taperfeatureless
featurelessand andtaper
taper
threaded
threadedtools.tools.They
Theyobserve
observethat thatananincrease
increase ofofthethetool
tooltilttilt
angle
angle increases
increases thetheforces
forcesandand thethe
torque
torqueat
the tool/work piece interface.
at the tool/work piece interface.
Elyasi
Elyasietetal.
al.[12]
[12]study
studythe theeffect
effectof ofthe
thetilt
tiltangle
angleon onFSWFSWof ofdissimilar
dissimilaralloys
alloys (aluminum
(aluminum to to steel).
steel).
Tilt anglesofof1◦1°,
Tiltangles , 2◦2°andand 3◦3°arearechosen.
chosen. They
They observe
observe thatthat
a larger
a largertilt angle increases
tilt angle the axial
increases force force
the axial and
the
andinteraction between
the interaction betweenaluminum aluminum and steel.
and steel.
Hamid
Hamid and and Roslee
Roslee [13] [13] investigate
investigate the the tilttilt angle
angle effect
effect on onmicrostructural
microstructural and andmechanical
mechanical
characteristics
characteristics of of FSWelded
FSWelded dissimilar aluminum aluminum alloys. alloys.They Theyobserve
observe that
that thethe
tilttilt angle
angle affects
affects the
the mechanical
mechanical properties
properties of ofthetheFSW FSWjointsjointsconsiderably.
considerably. Microstructure
Microstructure of of the weld
weld alsoalso changes
changes
significantly
significantlyby byvarying
varyingthe thetilttiltangle,
angle,specifically
specificallyininthe thearea
areaofofweldweldnugget
nuggetand andheatheataffected
affectedzone.zone.
Meshram
Meshram and Reddy [14] study the role of the tilt angle on defects generation and materialflow
and Reddy [14] study the role of the tilt angle on defects generation and material flow
ininFSW.
FSW.They They observe
observe thatthat
the the
variation
variationof toolof tilt
toolangle changes
tilt angle the thermo-mechanical
changes the thermo-mechanical results during
results
FSW
during andFSWtherefore alters thealters
and therefore material flow in the
the material flowweld andweld
in the controls
and the weldthe
controls defects.
weld defects.
InInspite
spiteof ofthe
theimportance
importanceofofthe theeffect
effectthat
thatthe thetool
tooltilt
tiltangle
anglehas hasononthe
thefinal
finalquality
qualityof ofthe
thewelded
welded
work
workpiece,
piece,there
thereare areonly
onlyaafew fewcomputational
computationalstudies studiesofofthis thisphenomenon.
phenomenon.Numerical Numericalsimulations
simulations
may
mayprovide
providedetailed
detailedknowledge
knowledgeofofthe theprocess
processfrom fromboth boththermal
thermaland andmechanical
mechanicalpoint pointofofviews.
views.
Long
Longetetal.al.[15]
[15]present
presentaa3D 3Dthermo-mechanical
thermo-mechanicalmodel modelwith withaanon-zero
non-zerotilt tiltangle
angleand andstudy
studyits its
effect
effectonon thethe
final joint.joint.
final TheyThey use DEFORM-3D
use DEFORM-3D to simulate the FSW the
to simulate process
FSWin process
a Lagrangianin a framework.
Lagrangian
In their work,In
framework. the tilt angle
their work,isthe considered
tilt angleinside of the geometrical
is considered inside of the model. They testmodel.
geometrical two casesThey 0◦ and
oftest two
◦
2cases
tilt angle. Wormhole defects are observed in case ◦
of 0 , while the weld ◦
of 0° and 2° tilt angle. Wormhole defects are observed in case of 0°,in casethe
while of 2weldis defect
in case free.
of 2°
Chauhan
is defect free. et al. [16] investigate the effect of three tilt angles (0◦ , 1◦ and 2◦ ) on the formation
of defectsChauhan in FSWet al. applying
[16] investigate a Coupledthe effectEulerian
of three andtilt Lagrangian
angles (0°, 1°(CEL) and 2°) method. They use
on the formation of
ABAQUS/Explicit
defects in FSW to model FSW
applying process with
a Coupled a cylindrical
Eulerian pin. In order(CEL)
and Lagrangian to avoid the serious
method. Theymesh use
distortion encountered
ABAQUS/Explicit when modeling
to model FSW process FSW,with the work piece is defined
a cylindrical as a Eulerian
pin. In order to avoidbody. Their model
the serious mesh
predicts thatencountered
a tilt angle of ◦
2 produces a defect
distortion when modeling FSW, free the weld.
work piece is defined as a Eulerian body. Their
model Aghajani
predicts Derazkola
that a tiltand angle Simchi
of 2° [17] present
produces experimental
a defect free weld. and numerical analysis of friction stir
welding of poly (methyl methacrylate) work pieces. They study the effect of process parameters such
as tilt angle to define the appropriate conditions for seeking defect-free joints. They observe that the
Metals 2019, 9, 28 3 of 17

tool tilt angle affects the material flow around the tool. The applied downward forging force needs to
be increased for increasing tilt angles, and this results in more frictional heat generation.
There are forecast models based on Artificial Neural Network (ANN). In these cases, both
experimental and numerical data are collected to correlate process parameters with technical features
of the welded joint. Hamilton et al. [18] integrate differential scanning calorimetry curves for 2017A
and 7075 in an existing computational model of the FSW process for heat generation and material
flow to create the phase transformations maps occurring in the weld zone. The tool tilt angle of 1.5 is
considered during the process. They observe that close to the weld tool, the processing temperatures
dissolve fully the equilibrium phase in 7075 and partially in 2017A. Casalino et al. [19] implement ANN
in order to investigate the effects of process parameters on the laser welding process quality. Using
statistical estimation, the relevance of the process parameters with the weld geometry is studied. It is
demonstrated that ANN modeling is beneficial for optimizing the quality of manufacturing processes.
Pathak and Jaiswal [20] provide a review on the applications of ANN in FSW. They consider the
tilt angle as one of the controlling factor. They conclude that ANN results are matching with the
experimental data.
From the previous works, it can be concluded that the tilt angle has a significant effect on the heat
generation and material flow and is a controlling parameter to produce a defect free joint.
In previous works devoted to the numerical modeling of the effect of the tilt angle, this angle
was considered in the geometrical setting, but not in the contact condition at the tool/work piece
interface. Reference [21] is one of the few works, both experimental and numerical, to address heat
and mass transfer due to the tilt angle. They use an Eulerian framework for an axisymmetric pin and
an incomplete contact boundary condition that applies frictional tangential force on a contact area
defined based on the tilt angle (α) and an in plane rotating angle (β) of the contact area. From the
experimental evidence, they conclude that this in plane rotating angle is 45◦ and they use it in the
numerical analysis.
In this work, we address the numerical analysis of the effect of the tool tilt angle on FSW from
the computational approach developed previously by the authors [5]. It allows obtaining the steady
state rapidly at the speed-up phase of the simulation. This is followed by a periodic stage simulation,
assuming the first stage as the initial condition. An apropos kinematic system is used by mixing
Arbitrary Lagrangian Eulerian (ALE), Eulerian and Lagrangian schemes for different areas of the
computational model. The framework can accommodate any pin shapes.
The influence of the tilting is to be represented by the enhanced friction model accounting for the
effect of non-uniform pressure distribution under the tool and tilting. The friction model is modified by
introducing an in plane rotating angle (β) which depends on the tool tilt angle (α) and the advancing
and rotating velocities. In the current study, this parameter is calibrated from the temperature field
obtained experimentally for the tilt angle 2.5◦ presented in reference [21]. Alternatively, the rotating
angle β can be obtained experimentally from the relationship between the longitudinal and the
transversal forces exerted on the tool.
The outline of this paper is as follows. In Section 2, the general solution strategy used in this
work is explained. In Section 3 the modified friction model considering the effect of the tilt angle is
presented and discussed. The last section is devoted to the analysis of tool tilt angle effect on the
thermo-mechanical behavior in FSW. Mechanical results including the material flow are presented and
compared for the no tilt (α = 0◦ , β = 0◦ ) and with tilt (α = 2.5◦ , β = 25◦ ) cases. Lastly, some conclusions
are drawn.
Metals 2019, 9, 28 4 of 17

2. Solution Strategy
The simulation of FSW can be performed in different kinematic frameworks: Lagrangian, Eulerian
and ALE.
In a Lagrangian framework, material moves together with the reference system. Therefore,
the material flow during the weld is the direct solution of the problem. However, due to the
large material deformation in the stir zone of FSW, the mesh used in this area requires continuous
re-meshing during the simulation. Re-meshing introduces a significant computational overhead and
re-interpolation errors. Thus, the application of other kinematic frameworks is more attractive.
In an Eulerian framework the movement of the material is defined on a fixed configuration.
Therefore, no re-meshing is needed. This framework presents limitations when non axisymmetric tool
pin shapes are modeled. In these cases, the boundaries of the model are constantly changing by the
rotation of the tool pin. Thus re-characterization of the integration domain at every time step of the
analysis is indispensable.
The alternative to Lagrangian and Eulerian approaches is an ALE framework where the reference
system is not fixed and allowed to move independently from the material movement. An ALE
framework permits to treat arbitrary pin geometries and re-meshing can be avoided using a mesh
around the tool that rotates rigidly together with the tool.
In a Lagrangian framework, the tool tilt angle can be directly included in the geometrical modeling.
In the Eulerian and ALE frameworks, the geometrical model can include the tool tilt angle directly
only if the pin shape is axisymmetric. Tilted non-axisymmetric pin shapes require specific ALE
approaches [22] as the rotation of the tool is not synchronized with the rotation of the mesh around
the tool.
In this work a feasible kinematic framework and a two-stage (speed-up and periodic stages)
strategy are adopted for the solution of the overall problem [5,23] (Figure 2). The strategy uses a fully
coupled thermo-mechanical framework at both stages. The solution of the coupled thermo-mechanical
problem is acquired by performing a staggered time-stepping algorithm solving the thermal and
mechanical sub-problems sequentially for each time step.
The speed-up stage aims at obtaining the steady state rapidly by modifying the thermal inertia
term in the energy balance equation. At this stage an Eulerian formulation is used.
The periodic stage considers the results obtained at the first stage as an initial condition. At this
stage an apropos kinematic framework is used [23]. The choice of this framework is for combining the
benefits of ALE, Eulerian and Lagrangian formulations by applying them in the stir zone, the remains
of the work piece and the pin-tool, respectively.
The main effect of the tilt angle on the process behavior is the heat generation and its influence on
the material flow during FSW. As the sources of heat generation in FSW are plastic dissipation and
friction, the suitable modification of the friction law is the strategy proposed here in order to include
the effect of the tilt angle.
All the implementations used for this work are done in the in-house finite element code
COMET [24] developed by the authors. Details on the technical and computational aspects of the
formulation are given in the references [5,23,25].
The resulting model incorporates a two-stage strategy that can speed up the transient stage to
obtain the periodic stage with 50 times reduced computational costs comparing with the standard
models [5]. Moreover, the model is enriched with an enhanced friction model that considers the real
process behavior for generating the frictional heat and can consider the effect of the tilt angle in the
heat generation and material flow.
Metals 2019, 9, 28 5 of 17
Metals 2018, 8, x FOR PEER REVIEW 5 of 17

Figure2.2.Two-stage
Figure Two-stagestrategy
strategyconcept.
concept.

3.3.Friction
FrictionModel
ModelIncluding
Includingthe
theTilt
TiltAngle
Angle
The
Thefriction
frictionlaw
lawdescribes
describesthethecontact
contactcondition
conditionatatthe
theinterface
interfacebetween
betweenthethetool
tooland
andthethework
work
piece as indicated by their relative sliding velocities. Coulomb’s [26–29] and Norton’s [22]
piece as indicated by their relative sliding velocities. Coulomb’s [26–29] and Norton’s [22] friction friction laws
are
lawsregularly utilized
are regularly in FSW
utilized insimulation.
FSW simulation.
In
In the previous work ofthe
the previous work of theauthors
authors[30],
[30],aamodified
modifiedNorton’s
Norton’slawlawisisproposed
proposedconsidering
consideringthethe
non-uniform
non-uniform pressure distribution that is generally found under the tool during FSW.The
pressure distribution that is generally found under the tool during FSW. Theenhanced
enhanced
friction
frictionmodel
modeldefines
definesthe
thefriction
frictionshear
shearstress
stressatateach
eachpoint
pointatatthe
thecontact
contactsurface
surfaceasas
Metals 2018, 8, x FOR PEER REVIEW 6 of 17
 τmax + τmin + (τmax − τmin ) tanhx x k∆vqT−1kq−1 ∆vT ,
 
τ max + τ min + (τ max −τ min ) tanh
τ T τ=T 0=.50.5 
 Δv T Δv T ,
R/6 (1)
(1)
−1 Fx R 6
 β = tan  F  (3)
where τT is the friction shear stress, ∆vT is the sliding 
yvelocity, 0 ≤ q ≤ 1 is the sensitivity parameter, x
iswhere
theThe τdetailed
T isofthe
location friction
numerical
each shear
point atinvestigation
the tool/work Δpiece
stress, andvTitsisexperimental
the sliding
contact surface velocity,
validation 0ofthe
relative to ≤the
q rotation
≤dependence
1 is the
axissensitivity
of angle
projected
parameter,
βononthe
thewelding x is the
process locationand
parameters
direction of
areeachis point
Rout of
thethe at the tool/work
scope
shoulder ofradius. piece
this work.
τmaxWecontact
willτmin
and surface
focus relative
areonthe
the to theof
influence
maximum rotation
andangle
the
βminimum
on the
axis thermo-mechanical
projected on the behavior
welding of
directionFSW, by
and comparing
R is the two cases:
shoulder without
radius.
friction tractions. Figure 3 presents a schematic view of this distribution tilt (α =
and0°,τ
τ max of frictionβ = 0°)
areand
the
mintraction
with
where tiltthe
maximum (α average
=and
2.5°,the
β value
=minimum
25°).of thefriction
frictiontractions.
is at the Figure
center of the tool a(xschematic
3 presents = 0). Noteview
that of the
this
τmax and τmin
distribution
values are attained
of friction traction at the leading
where and trailing
the average value ofedges of the shoulder.
the friction is at the center of the tool (x = 0). Note that
the τ max and τ min values are attained at the leading and trailing edges of the shoulder.
This friction law in Equation (1) does not take into account the effect of the tilt angle. In case of
having a tilt angle α (backwards), as the FSW tool advances in the weld direction, the contact surface
that has a maximum friction value in front of the tool and minimum friction in the rear side rotates a
certain angle β (counterclockwise) around the (counterclockwise) rotating axis due to the tilt
influence [21]. This is detected from the experimental evidence in the reference [21] where the effect
of tilting appears on the rotation of the contact print. Figure 4 shows schematically what is observed
in the experiments. Therefore, the maximum friction is not at the front side but it is rotated by an
angle β. The tilting of the tool results in the subsequent rotation of the average friction line. Figure 5
shows schematically how the distribution of the friction under the shoulder is affected by the tool tilt.
The x-axis is along the welding direction with or without tilt angle; it is perpendicular to the average
friction line (y-axis) when no tilt angle exists. For a tilt angle α, the average friction line (y’-axis) is
rotated an angle β in the horizontal plane.
The angle β of the friction shear stress distribution depends on the welding parameters such as
tilt angle α, rotating speed ω and advancing speed vadv:
Figure 3. Friction shear traction distribution under the tool (0◦ tilt angle).
β = f (α , ω , vadv ) (2)
This frictionFigure
law in3.Equation
Friction shear traction
(1) does notdistribution
take into under thethe
account tooleffect
(0° tiltof
angle).
the tilt angle. In case
In this
of having work,
a tilt angle
angle β is obtainedasbythe
α (backwards), calibration
FSW toolfrom the temperature
advances in the weldfield of the the
direction, experiment
contact
presented in [21]. However, the rotating angle β can be obtained experimentally from the relationship
between the longitudinal (Fx) and the transversal (Fy) forces exerted on the tool.
Metals 2019, 9, 28 6 of 17

surface that has a maximum friction value in front of the tool and minimum friction in the rear side
rotates a certain angle β (counterclockwise) around the (counterclockwise) rotating axis due to the tilt
influence [21]. This is detected from the experimental evidence in the reference [21] where the effect
of tilting appears on the rotation of the contact print. Figure 4 shows schematically what is observed
in the experiments. Therefore, the maximum friction is not at the front side but it is rotated by an
angle β. The tilting of the tool results in the subsequent rotation of the average friction line. Figure 5
shows schematically how the distribution of the friction under the shoulder is affected by the tool tilt.
The x-axis is along the welding direction with or without tilt angle; it is perpendicular to the average
friction line (y-axis) when no tilt angle exists. For a tilt angle α, the average friction line (y’-axis) is
rotated an angle β Figure 3. Friction shear
in the horizontal traction distribution under the tool (0° tilt angle).
plane.

Figure4.4.Schematic
Figure Schematiccontact
contact print
print observed
observed in in
thethe experiment
experiment of FSW.
of FSW. Without
Without tilt angle
tilt angle (top);(top);
withwith
tilt
Metals 2018, 8, x FOR PEER REVIEW 7 of 17
tilt angle
angle (bottom).
(bottom).

Figure 5. Contact condition under the (counterclockwise) rotating tool. Without tilt angle (left); with
Figure 5. Contact condition under the (counterclockwise) rotating tool. Without tilt angle (left); with
tilt angle (right).
tilt angle (right).
The angle β of the friction shear stress distribution depends on the welding parameters such as
Considering the effect of the tilt and the rotation of the contact shear stress between the tool and
tilt angle α, rotating speed ω and advancing speed vadv :
the work piece, the reference axes x and y are rotated to the new position x’ and y’.

 x′   cos
β= β f (α, ω, β ) x 
sinvadv (2)
=
 y ′ − sin β cos β   y 
, (4)
  
where β is the rotating angle of the contact surface. Therefore, Equation (1) can be rewritten as

 x cos β + y sin β 
τ T = 0.5 τ max + τ min + (τ max − τ min ) tanh
q −1
 Δv T Δv T , (5)
 R6 
Metals 2019, 9, 28 7 of 17

In this work, angle β is obtained by calibration from the temperature field of the experiment
presented in [21]. However, the rotating angle β can be obtained experimentally from the relationship
between the longitudinal (Fx ) and the transversal (Fy ) forces exerted on the tool.
 
Fx
β = tan−1 (3)
Fy

Figure
The 5. Contact
detailed condition
numerical under the (counterclockwise)
investigation rotating
and its experimental tool. Without
validation tiltdependence
of the angle (left); with
of angle
tilt angle (right).
β on the process parameters are out of the scope of this work. We will focus on the influence of angle β
on the thermo-mechanical behavior of FSW, by comparing two cases: without tilt (α = 0◦ , β = 0◦ ) and
Considering the effect of the tilt and the rotation of the contact shear stress between the tool and
with tilt (α = 2.5◦ , β = 25◦ ).
the work piece, the reference axes x and y are rotated to the new position x’ and y’.
Considering the effect of the tilt and the rotation of the contact shear stress between the tool and
the work piece, the reference axes  x′x and y areβrotated
 cos sintoβthe
  xnew
 position x’ and y’.
 "y ′ = #− sin β cos β   y , # (4)
 x  sinβ  x
" #"
0 cos β
0 = , (4)
where β is the rotating angle of theycontact surface. − sin βTherefore,
cos β y
Equation (1) can be rewritten as

 angle of the contact surface. Therefore,

where β is the rotating sin β  (1) can
x cos β + yEquation q −1 be rewritten as
τ T = 0.5 τ max + τ min + (τ max − τ min ) tanh  Δv T Δv T , (5)

 R 6β + y sinβ 
x cos
τT = 0.5 τmax + τmin + (τmax − τmin ) tanh k∆vT kq−1 ∆vT , (5)
R/6
Figure 6 presents the distribution of friction law in case of having a tilt angle. The average value
of theFigure
friction is rotated
6 presents thearound the center
distribution of the law
of friction tool.in case of having a tilt angle. The average value
of the friction is rotated around the center of the tool.

Figure 6. Friction shear traction distribution under the tool (2.5◦ tilt angle).

Figure
4. Analysis of the 6. of
Effect Friction shear
the Tilt traction distribution under the tool (2.5° tilt angle).
Angle
In this section,
4. Analysis the effect
of the Effect of the
of the Tilttool tilt angle on the thermo-mechanical results of a FSW simulation
Angle
is studied. The thermal effects are studied through the temperature field. The mechanical effects are
analyzed via velocity, stress and strain rate fields and material flow around the tool.
Two cases are considered: 0◦ and 2.5◦ tilting angle. The two cases are identical from the point of
view of material, processing parameters, geometry and they only differ in tilt angle.
The temperature field in case of having tilt angle obtained from numerical analysis is
compared with experiment [21] in order to obtain the corresponding angle of rotation β. Then the
thermo-mechanical results in both cases of with and without tilt angle are compared against each other.
The material selected is aluminum alloy AA2024-T4. The chemical composition (wt%) of the
aluminum alloy AA2024-T4 is Cu = 4.53, Mg = 1.62, Mn = 0.65, Si = 0.066, Fe = 0.21 and Al = Bal. [31].
 Thetemperature
The temperaturefield fieldin
incase
caseofofhaving
havingtilt
tiltangle
angleobtained
obtainedfromfromnumerical
numericalanalysis
analysisisiscompared
compared
with experiment
with experiment [21] [21] in
in order
order toto obtain
obtain the
the corresponding
corresponding angleangle of of rotation
rotation β.β. Then
Then the
the thermo-
thermo-
mechanicalresults
mechanical resultsin inboth
bothcases
casesofofwith
withand
andwithout
withouttilttiltangle
angleare
arecompared
comparedagainst
againsteach
eachother.
other.
Thematerial
The materialselected
selectedisisaluminum
aluminumalloy alloyAA2024-T4.
AA2024-T4.The Thechemical
chemicalcomposition
composition(wt%)(wt%)of ofthe
the
Metals
aluminum2019,
aluminumalloy 9, 28
alloyAA2024-T4
AA2024-T4isisCu Cu==4.53,
4.53,Mg
Mg==1.62,
1.62,Mn
Mn==0.65,
0.65,SiSi==0.066,
0.066,FeFe==0.21
0.21andandAlAl==Bal. 8 of
Bal.[31].17
[31].
Thedimension
The dimensionof ofthethework
workpiecepieceisis300
300××75
75××55mm mm3.3.Figure
Figure77shows
showsthe thegeometry
geometrymodelmodelincluding
including
tool,stir
tool, stirzone
zoneand andthe therest
restof
ofthe
thework
workpiece.
piece.The
Thetool
toolhas
hasaaflat
flatshoulder
shoulderof of16
16mm mmdiameter
diameterandandaa
The dimension of the work piece is 300 × 75 × 5 mm3 . Figure 7 shows the geometry model including
featurelessconical
featureless conicalpin.pin.The
Thetoptopand
andbottom
bottomdiameter
diameterof ofthe
thepin
pinare
are66mmmmand and44mm,
mm,respectively.
respectively.TheThe
tool, stir zone and the rest of the work piece. The tool has a flat shoulder of 16 mm diameter and a
heightof
height ofthe
thepin pinisis4.8
4.8mm.
mm.Figure
Figure88shows
showsthethedimension
dimensionof ofthe
thetool
toolused.
used.
featureless conical pin. The top and bottom diameter of the pin are 6 mm and 4 mm, respectively.
The height of the pin is 4.8 mm. Figure 8 shows the dimension of the tool used.

Figure7.
Figure
Figure 7.The
7. Thegeometry
The geometrymodel.
geometry model.
model.

Figure 8. Conical tool geometry.

Figure8.8.Conical
Figure Conicaltool
toolgeometry.
geometry.
The rotating and advancing velocities are 800 rpm and 20 mm/min, respectively.
The process parameters are selected as such to compare the numerical results obtained in this
work with the experimental data published in [21].
The computational model consists of 380,000 tetrahedral elements and 60,000 points approximately.
Figure 9 shows the corresponding mesh used to discretize the model.
Heat generated through both plastic dissipation and frictional contact is considered.
The visco-plastic dissipation (Dmech ) is defined as
.
Dmech = γs : εε, (6)
.
where γ is the Taylor–Quinney coefficient, s is the deviatoric stress and ε is the strain rate.
 Metals 2019, 9, 28 9 of 17
Metals 2018, 8, x FOR PEER REVIEW 9 of 17

(a) (b)
Meshresolution.
Figure9.9.Mesh
Figure resolution.(a)
(a)entire
entiredomain;
domain;(b)
(b)zoom
zoomon
onthe
thestir
stirzone.
zone.

Therotating
The analysisandconsiders
advancingthe velocities
minimumare and800 maximum
rpm andfriction tractions
20 mm/min, as τ min = 1.55 × 107 and
respectively.
= 3.1
τ maxThe × 108 at
process tool/wor-kpiece
parameters contact
are selected assurface,
such torespectively.
compare theThey are obtained
numerical resultsfrom the calibration
obtained in this
from the temperature field [30]. The in plane rotating angle β = 25 ◦ is calibrated from the temperature
work with the experimental data published in [21].
fieldThe
obtained experimentally
computational model forconsists
the tilt angle 2.5◦ andtetrahedral
of 380,000 presented inelements
reference and
[21]. 60,000
The qualitative
points
mechanical results including the material flow are presented
approximately. Figure 9 shows the corresponding mesh used to discretize the model. and compared for the cases (α = 0◦ ,
◦
β = Heat
0 ) and ◦
(α = 2.5 through ◦
, β = 25 both
). plastic dissipation and frictional contact is considered.
generated
◦ , the longitudinal and the transversal forces are
The visco-plastic dissipationrotating
To verify the choice of the ( Dmech )angle β = 25as
is defined
evaluated in both cases, without and with tilt angle. In the first case (without tilt angle), the forces are
Fx = 170 N and Fy = 28000 N. Thus according γ s : ε , (3), the in plane rotation angle is β ∼
D mechto=Equation ◦
(6)0 .
=
In the second case (with tilt angle), the forces are Fx = 12000 N and Fy = 25000 N. Thus the in plane
where
rotation γ angle
is theisTaylor–Quinney
effectively β = 25 coefficient,
◦. s is the deviatoric stress and ε is the strain rate.
The analysis considers the minimum and maximum friction tractions as τmin = 1.55 × 107 and τmax
=4.1.
3.1 Thermal
× 108 at Effects
tool/wor-kpiece contact surface, respectively. They are obtained from the calibration
from theThetemperature fieldcaused
thermal effects [30]. The byinthe
plane
toolrotating angle
tilt angle areβpresented
= 25° is calibrated from the
in this section intemperature
terms of the
field obtained experimentally
temperature field at the steady state. for the tilt angle 2.5° and presented in reference [21]. The qualitative
mechanical results including the material flow are presented and compared for the cases (α = 0°, β=
0°) and (α = 2.5°, β = 25°).
Temperature
To verify the choice of the rotating angle β = 25°, the longitudinal and the◦transversal forces are
Figure 10 shows the computed temperature field for the tilt angles of 0 and 2.5◦ . The results
evaluated in both cases, without and with tilt angle. In the first case (without tilt angle), the forces are
are also shown on a vertical section at the center of the tool and the leading side in order to see the
Fx = 170 N and Fy = 28000 N. Thus according to Equation (3), the in plane rotation angle is β ≅ 0°. In
temperature field on the top surface and within the depth of the work piece.
the second case (with tilt angle), the forces are Fx = 12000 N and Fy = 25000 N. Thus the in plane rotation
The difference caused by the tool tilt angle on the temperature distribution can be clearly seen.
angle is effectively β = 25°.
In both cases, the maximum temperature is on the advancing side (AS) due to the non-uniform
distribution of the friction at the contact surface between tool and work piece. The tool tilt angle causes
4.1. Thermal Effects
a slight rotation of the temperature field. It increases the temperature in the neighboring zone of the
FSWThe toolthermal effects
in the rear caused side.
advancing by the tool
This tilt angle are
observation is inpresented
accordance inwith
this the
section in terms finding
experimental of the
temperature
in [32]. field at the steady state.
The difference of the temperature in the advancing (AS) and the retreating sides (RS) diminishes
Temperature
with the distance from the tool center. The computed maximum temperatures are compared with the
measured ◦ at different distances from the welding line on both
Figure ones presented
10 shows in [21] fortemperature
the computed tilt angle 2.5field for the tilt angles of 0° and 2.5°. The results are
advancing and retreating side, see Figures 11 and 12.
also shown on a vertical section at the center of the tool and the leading side in order to see the
temperature field on the top surface and within the depth of the work piece.
The difference caused by the tool tilt angle on the temperature distribution can be clearly seen.
In both cases, the maximum temperature is on the advancing side (AS) due to the non-uniform
distribution of the friction at the contact surface between tool and work piece. The tool tilt angle
measured
measuredones
onespresented
presentedinin[21]
[21]for
fortilt
tiltangle
angle2.5°
2.5°atatdifferent
differentdistances
distancesfrom
fromthe
thewelding
weldingline
lineon
onboth
both
advancing and retreating side, see Figures 11 and
advancing and retreating side, see Figures 11 and 12. 12.
ItItcan
canbe
beseen
seenfrom
fromthethefigures
figuresthatthataagood
goodagreement
agreementisisachieved
achievedbetween
betweennumerical
numericaland and
experimental results at both retreating and advancing sides of the FSWelded work
experimental results at both retreating and advancing sides of the FSWelded work piece. piece.
Metals 2019, 9, 28 10 of 17

(a)
(a) (b)
(b)

(c)
(c) (d)
(d)
Figure 10.
Figure Temperature distribution. (a,c) 00°◦ tilt
10. Temperature tilt angle;
angle; (b,d) 2.5◦ tilt
(b,d) 2.5° tiltangle.
angle.
Figure 10. Temperature distribution. (a,c) 0° tilt angle; (b,d) 2.5° tilt angle.

Figure 11. Temperature comparison between numerical results and experimental data on the advancing
side (AS) at different distances from the weld line (2.5◦ tilt angle).
Metals 2018, 8, x FOR PEER REVIEW 11 of 17

Figure 11. Temperature comparison between numerical results and experimental data on the
Metals 2019, 9, 28 11 of 17
advancing side (AS) at different distances from the weld line (2.5° tilt angle).

Figure12.
Figure 12.Temperature
Temperature comparison
comparison between
between numerical
numerical resultsresults and experimental
and experimental data
data on the on the
retreating
side (RS) atside
retreating different distances
(RS) at differentfrom the weld
distances fromline
the(2.5 ◦ tilt
weld angle).
line (2.5° tilt angle).

ItTable
can be seen from
1 compares thethe figures that
differences a good
between theagreement
temperatures is achieved between
at retreating numerical sides
and advancing and
experimental
obtained from results at both retreating
experimental and advancing
and numerical sidesaoftiltthe
analyses when FSWelded
angle work
of 2.5° is piece.
used. This comparison
Table 1 compares
is performed the differences
for distances of 10, 14, between
20 and 28 themm
temperatures
from the weld at retreating
line. Theand advancing
agreement sides
between
obtained from experimental and numerical ◦
numerical results and experimental data isanalyses when a tilt angle of 2.5 is used. This comparison
remarkable.
is performed for distances of 10, 14, 20 and 28 mm from the weld line. The agreement between
numerical
Tableresults and experimental
1. Difference data is remarkable.
between temperatures on the advancing and retreating sides at different locations
from the weld line.
Table 1. Difference between temperatures on the advancing and retreating sides at different locations
from the weld Temperature Difference of AS and RS (Tilt Angle of 2.5°)
Distance fromline.
the Welding Line (mm)
Experimental Data Numerical Analysis
10 Temperature Difference
35 of AS and RS (Tilt36Angle of 2.5◦ )
Distance from the Welding Line (mm)
14 Experimental
15 Data Numerical15 Analysis
10 20 3513 1136
14 28 1511 1015
20 13 11
28 11 10
4.2. Mechanical Effects
The mechanical
4.2. Mechanical Effects effects caused by the tool tilt angle are presented in this section in terms of
velocity, stresses and strain rate fields and material flow around the tool.
The mechanical effects caused by the tool tilt angle are presented in this section in terms of velocity,
stresses and strain
4.2.1. Velocity, rateand
Stress fields andRate
Strain material flow around the tool.

4.2.1. The computed

Velocity, Stress velocity fields
and Strain Ratefor α = 0° and α = 2.5° under the shoulder are depicted in Figure
13.
The computed velocity fields for α = 0◦ and α = 2.5◦ under the shoulder are depicted in Figure 13.
The velocity field in case of tilt angle α = 2.5° is rotated β = 25° (obtained from the calibration of
The velocity field in case of tilt angle α = 2.5◦ is rotated β = 25◦ (obtained from the calibration of
the rotating angle β from the temperature field presented in [21]). As expected, the maximum velocity
the rotating angle β from the temperature field presented in [21]). As expected, the maximum velocity
is at the border of the shoulder and in case of 0° tilt angle at the leading edge.
is at the border of the shoulder and in case of 0◦ tilt angle at the leading edge.
Metals 2018, 8, x FOR PEER REVIEW 12 of 17

Metals 2019, 9, 28 12 of 17

(a) (b)

(a) (b)

(c) (d)

Figure 13. Velocity field. (a,c) 0° tilt angle (b,d) 2.5° tilt angle.
(c) (d)
The J2 stress distributions under the shoulder◦ in both studied cases ◦ are presented in Figure 14.
Figure 13. Velocity
Figure 13. Velocityfield.
field. (a,c)
(a,c) 00° tilt angle (b,d) 2.5
2.5° tilt angle.
The effect of the tilt angle is the rotation of this distribution and the increase of the stresses in the
retreating
The side of distributions
the leading front.under Asthe the tool tilt anglestudied
increases theare
temperature onFigure
the rear
The J2
J2 stress
stress distributions under the shoulder
shoulder in both
in both cases
studied cases presented in
are presented in Figure 14.14.
advancing
The effect ofside
theoftiltthe toolis(Figure
angle the 10), the
rotation of material
this flow stress
distribution and decreases
the increasecorrespondingly
of the stresses in
in this
the
The effect of the tilt angle is the rotation of this distribution and the increase of the stresses in the
region andside
retreating consequent softening of theAsmaterial toolfacilitates
anglethe flow. The thecomputed behavior agrees
retreating side of of the
the leading
leading front.
front. As the the tool tilt angle
tilt increases
increases the temperature
temperature on the
on the rear
rear
with the finding
advancing side ofinthe[32].
tool (Figure 10), the material flow stress decreases correspondingly in this region
advancing side of the tool (Figure 10), the material flow stress decreases correspondingly in this
and consequent
region softening
and consequent of the material
softening facilitates
of the material the flow.the
facilitates Theflow.
computed behaviorbehavior
The computed agrees with the
agrees
finding
with theinfinding
[32]. in [32].

(a) (b)
Figure 14. Cont.
(a) (b)
Metals 2018, 8, x FOR PEER REVIEW 13 of 17

Metals 2019, 9, 28 13 of 17
Metals 2018, 8, x FOR PEER REVIEW 13 of 17

(c) (d)

Figure 14. J2 stress field. (a,c) 0° tilt angle; (b,d) 2.5° tilt angle.

The strain rate distribution

(c) is shown in Figure 15 for both tilt angles. (d) As the strain rate defines
the stirring action in FSW [33], the distribution of the strain rate under the tool can give an insight to
J2 stress
Figure 14.Figure 14. field.
J2 stress 0◦ tilt
(a,c)field. angle;
(a,c) (b,d)
0° tilt 2.5◦(b,d)
angle; tilt angle.
2.5° tilt angle.
the material stirring. In case of having tilt angle, the stirring effect increases on the rear edge of the
tool on
Thethe advancing
strain side. Therefore
rate distribution is shown theintiltFigure
angle15 can forstrengthen the material
both tilt angles. As the stirring action
strain rate at this
defines the
The strain rate distribution is shown in Figure 15 for both tilt angles. As the strain rate defines
zone.
stirring action in FSW [33], the distribution of the strain rate under the tool can give an insight to the
the stirring action in FSW [33], the distribution of the strain rate under the tool can give an insight to
material stirring. In case of having tilt angle, the stirring effect increases on the rear edge of the tool on
the material stirring. In case of having tilt angle, the stirring effect increases on the rear edge of the
the advancing side. Therefore the tilt angle can strengthen the material stirring action at this zone.
tool on the advancing side. Therefore the tilt angle can strengthen the material stirring action at this
zone.

(a) (b)

(a) (b)

(c) (d)
0◦ tilt
Figure 15. J2 strain rate field. (a,c) 0° tilt angle;
angle; (b,d) 2.5◦tilt
(b,d) 2.5° tiltangle.
angle.

4.2.2. Material Flow

4.2.2. Material Flow (c) (d)
In order to visualize the differences on the material flow around the tool, stream lines are shown
Figure 15. J2 strain rate field. (a,c) 0° tilt angle; (b,d) 2.5° tilt angle.
in Figures 16 and 17 for both tilting cases at different depths (1 mm and 3 mm). In the xy plane, the
effect of tilting manifests in the rotation of the streamlines (Figures 16b and 17b).
4.2.2. Material Flow
Metals 2018, 8, x FOR PEER REVIEW 14 of 17

In order to visualize the differences on the material flow around the tool, stream lines are shown
in Figures 16 and 17 for both tilting cases at different depths (1 mm and 3 mm). In the xy plane,
Metals 2019, 9, 28
the
14 of 17
effect of tilting manifests in the rotation of the streamlines (Figures 16b and 17b).
In the case of no tilt angle and 1 mm depth, the material moves extensively around the tool. It is
In both
visible the case
in theof no
xy tilt angle
plane and 116a)
(Figure mmanddepth,the the material
xz plane movesthat
showing extensively
the materialaround the tool.
passes It is
through
visible both in the xy plane
all the depth of the tool (Figure 16c).(Figure 16a) and the xz plane showing that the material passes through all
the depth of the tool
The material (Figure 16c).
movement at the depth of 1 mm is very much affected by the shoulder movement
The16c,d).
(Figure material Due movement
to the tiltat the depth
angle effect,ofmaterial
1 mm isisvery muchbyaffected
trapped by the shoulder
the shoulder movement
on the trailing edge
(Figure 16d).
16c,d).It Due to the tilt angle effect, material is trapped by the shoulder
can be seen from Figure 16d that tilting induces a considerable accelerating effect on the trailing edge
(Figure the
behind 16d).pinIttool
canwhich
be seen from Figure
is difficult 16d that
to achieve tilting
in case of induces
having no a considerable
tilt angle. Theaccelerating
reason for thiseffect
is
behind the pin tool which is difficult to achieve in case of having no tilt angle.
that a higher friction force exists on the rear side of the tool at the interface between the tool and the The reason for this is
that apiece.
work higherThis friction force exists
stimulating onofthe
effect therear
toolside of the material
tilt on tool at the
flowinterface
helps to between
avoid thetheformation
tool and theof
work piece.
defects in theThis
weld,stimulating effect of
as low material the toolmight
velocity tilt onlead
the tomaterial
defect flow helps [34].
formation to avoid the formation of
defects in thefrom
Further weld, asshoulder,
the low material velocity
at the depth might
of 3mm, lead thetoeffect
defectofformation
tilting is less[34].evident. In both cases
of 0° Further
and 2.5°from the shoulder,
tilt angles, at thethe
apart from depth of 3mm,
rotation of thethestreamlines,
effect of tilting
the is less evident.
material movementIn both cases
around
◦ and 2.5◦ tilt angles, apart from the rotation of the streamlines, the material movement around
of 0tool
the is similar (Figure 17a,b). Without tilt angle, the material around the tool goes through all the
the tool
depth ofisthesimilar (Figure 17a,b).
pin resembling Without
the case with tilt angle, the material
angle (Figure around the tool goes through all the
17c,d).
depth It of
can thebepin
seenresembling
from Figure the case
17d with
that tilt
the angle (Figure
material flows 17c,d).
around the pin and then accumulates
It can be seen from
behind the tool on the weld. Figure 17d that the material flows around the pin and then accumulates behind
the tool
Dueon tothe
theweld.
higher temperature, softer material and greater frictional force in the trailing side of
Due the
the tool, to the higher flow
material temperature,
in the rearsofter material
side of the and FSW greater frictional
tool with forceangle
the title in theistrailing side of
considerably
the tool, the material flow in the rear side of
enhanced, which assists to prevent the generation of defect. the FSW tool with the title angle is considerably enhanced,
which assists to prevent the generation of defect.

(a) (b)

(c) (d)
Figure 16. Material flow paths in horizontal and vertical view at depth of 1 mm. (a) 0◦ tilt angle, depth
of 1 mm, xy plane; (b) 2.5◦ tilt angle, depth of 1 mm, xy plane; (c) 0◦ tilt angle, depth of 1 mm, xz plane;
(d) 2.5◦ tilt angle, depth of 1 mm, xz plane.
 Figure 16. Material flow paths in horizontal and vertical view at depth of 1 mm. (a) 0° tilt angle, depth
of 1 mm, xy plane; (b) 2.5° tilt angle, depth of 1 mm, xy plane; (c) 0° tilt angle, depth of 1 mm, xz plane;
(d) 2.5° tilt angle, depth of 1 mm, xz plane.
Metals 2019, 9, 28 15 of 17

(a) (b)

(c) (d)
Figure 17.
Figure Material flow
17. Material flow paths
paths in in horizontal
horizontal and vertical view at depth of 3 mm. (a) 0° 0◦ tilt
tiltangle,
angle,depth
depth
of 3 mm, xy plane; (b) 2.5 ◦ tilt angle, depth of 3 mm, xy plane; (c) 0 ◦ tilt angle, depth of 3 mm, xz plane;
of 3 mm, xy plane; (b) 2.5° tilt angle, depth of 3 mm, xy plane; (c) 0° tilt angle, depth of 3 mm, xz plane;
(d) 2.5 ◦ tilt angle, depth of 3 mm, xz plane.
(d) 2.5° tilt angle, depth of 3 mm, xz plane.

5. Summary and Conclusions

5. Summary and Conclusion
In this work, the effect of tool tilt angle on the thermo-mechanical results (heat generation and
In this work, the effect of tool tilt angle on the thermo-mechanical results (heat generation and
material flow) in FSW process is studied. The thermo-mechanical results are presented for two tilt
material flow) in FSW process is studied. The thermo-mechanical results are presented for two tilt
angle cases of α = 0◦ and α = 2.5◦ .
angle cases of α = 0° and α = 2.5°.
The friction model is modified by introducing an in plane rotating angle β of the friction shear
The friction model is modified by introducing an in plane rotating angle β of the friction shear
stress in order to account for the effect of tilting. This rotation angle is calibrated from the temperature
stress in order to account for the effect of tilting. This rotation angle is calibrated from the temperature
field obtained experimentally for the tilt angle 2.5◦ . The qualitative mechanical results including the
field obtained experimentally for the tilt angle 2.5°. The qualitative mechanical results including the
material flow are presented and compared for the cases (α = 0◦ , β= 0◦ ) and (α = 2.5◦ , β = 25◦ ). It is
material flow are presented and compared for the cases (α = 0°, β= 0°) and (α = 2.5°, β = 25°). It is
verified that the rotating angle β can be defined through the relationship between the longitudinal and
verified that the rotating angle β can be defined through the relationship between the longitudinal
the transversal forces exerted on the tool.
and the transversal forces exerted on the tool.
It is observed that a non-zero tilt angle results in the rotation of the friction distribution under the
It is observed that a non-zero tilt angle results in the rotation of the friction distribution under
shoulder. The computed temperature is compared with the experimental data and good agreement is
the shoulder. The computed temperature is compared with the experimental data and good
obtained at both advancing and retreating sides. Differences between the temperatures at retreating
agreement is obtained at both advancing and retreating sides. Differences between the temperatures
and advancing sides are also compared with the experimental measurements.
at retreating and advancing sides are also compared with the experimental measurements.
It is observed that the tool tilt angle:
It is observed that the tool tilt angle:
• increases stresses at the leading edge of the tool on the work piece.
• increases the temperature in the neighboring zone of the FSW tool in the rear advancing side.
• decreases the material flow stress in the rear advancing side.
• strengthens the material stirring action at trailing edge on the advancing side.
Metals 2019, 9, 28 16 of 17

• facilitates the material flow behind the tool.

To sum up, the effect of the tool tilt angle can be represented by the in plane rotation of the
corresponding thermo mechanical fields. Owing to this, the material flow behind the FSW tool is
considerably improved contributing to the prevention of the defect formation. Future work will
be addressed to extend the experimental evidence in order to correlate the tilt angle and process
parameters with obtained in plane rotation of the friction shear traction.

Author Contributions: N.D., M.C. (Miguel Cervera) and M.C. (Michele Chiumenti) developed the numerical
model; N.D. implemented the model; N.D. and M.C. (Miguel Cervera) discussed and analyzed the data.
Funding: This research received no external funding.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Langari, J.; Kolahan, F.; Aliakbari, K. Effect of tool speed on axial force, mechanical properties and weld
morphology of friction stir welded joints of a7075-t651. Int. J. Eng. 2016, 29, 403–410.
2. Mehta, K.P.; Badheka, V.J. Effects of tilt angle on properties of dissimilar friction stir welding copper to
aluminum. Mater. Manuf. Process. 2016, 31, 255–263. [CrossRef]
3. Jaiganesh, V.; Maruthu, B.; Gopinath, E. Optimization of process parameters on friction stir welding of high
density polypropylene plate. Procedia Eng. 2014, 97, 1957–1965. [CrossRef]
4. Dialami, N.; Chiumenti, M.; Cervera, M.; Agelet de Saracibar, C.; Ponthot, J.-P. Numerical simulation and
visualization of material flow in friction stir welding via particle tracing. In Numerical Simulations of Coupled
Problems in Engineering; Springer International Publishing: Cham, Switzerland, 2014; pp. 157–169.
5. Dialami, N.; Chiumenti, M.; Cervera, M.; de Saracibar, C.A. A fast and accurate two-stage strategy to evaluate
the effect of the pin tool profile on metal flow, torque and forces during friction stir welding. Int. J. Mech. Sci.
2017, 122, 215–227. [CrossRef]
6. Chien, C.H.; Lin, W.B.; Chen, T. Optimal FSW process parameters for aluminum alloys AA5083. J. Chin.
Inst. Eng. 2011, 34, 99–105. [CrossRef]
7. Arici, A.; Selale, S. Effects of tool tilt angle on tensile strength and fracture locations of friction stir welding
of polyethylene. Sci. Technol. Weld. Join. 2007, 12, 536–539. [CrossRef]
8. Payganeh, G.H.; Arab, N.B.M.; Asl, Y.D.; Ghasemi, F.A.; Boroujeni, M.S. Effects of friction stir welding rocess
parameters on appearance and strength of polypropylene composite welds. Int. J. Phys. Sci. 2011, 6, 595–601.
9. Shazly, M.; El-raey, M. Friction stir welding of polycarbonate sheets. In Characterization of Minerals, Metals,
and Materials; Wiley: Hoboken, NJ, USA, 2014; pp. 555–563.
10. Reshad Seighalani, K.; Besharati Givi, M.K.; Nasiri, A.M.; Bahemmat, P. Investigations on the Effects of the
Tool Material, Geometry, and Tilt Angle on Friction Stir Welding of Pure Titanium. J. Mater. Eng. Perform.
2010, 19, 955. [CrossRef]
11. Banik, A.; Roy, B.S.; Barma, J.D.; Saha, S.C. An experimental investigation of torque and force generation
forvarying tool tilt angles and their effects on microstructure and mechanical properties: Friction stir welding
of AA 6061-T6. J. Manuf. Process. 2018, 31, 395. [CrossRef]
12. Elyasi, M.; Aghajani Derazkola, H.; Hosseinzadeh, M. Investigations of tool tilt angle on properties friction
stir welding of A441 AISI to AA1100 aluminium. Proc. Inst. Mech. Eng. Part B J. Eng. Manuf. 2016, 230,
1234–1241. [CrossRef]
13. Hamid, H.A.D.; Roslee, A.A. Study the Role of Friction Stir Welding Tilt Angle on Microstructure and
Hardness. Appl. Mech. Mater. 2015, 799–800, 434–438. [CrossRef]
14. Meshram, S.D.; Reddy, G.M. Influence of Tool Tilt Angle on Material Flow and Defect Generation in Friction
Stir Welding of AA2219. Def. Sci. J. 2018, 68, 512–518.
15. Long, L.; Chen, G.; Zhang, S.; Liu, T.; Shi, Q. Finite-element analysis of the tool tilt angle effect on the
formation of friction stir welds. J. Manuf. Process. 2017, 30, 562–569. [CrossRef]
16. Chauhan, P.; Jain, R.; Pal, S.K.; Singh, S.B. Modeling of defects in friction stir welding using coupled Eulerian
and Lagrangian method. J. Manuf. Process. 2018, 34, 158–166. [CrossRef]
Metals 2019, 9, 28 17 of 17

17. Aghajani Derazkola, H.; Simchi, A. Experimental and thermomechanical analysis of friction stir welding of
poly(methyl methacrylate) sheets. Sci. Technol. Weld. Join. 2018, 23, 209–218. [CrossRef]
18. Hamilton, C.; Dymek, S.; Kopyscianski, M.; Weglowska, A.; Pietras, A. Numerically Based Phase
Transformation Maps for Dissimilar Aluminum Alloys Joined by Friction Stir-Welding. Metals 2018, 8,
324. [CrossRef]
19. Casalino, G.; Facchini, F.; Mortello, M.; Mummolo, G. ANN modelling to optimize manufacturing processes:
The case of laser welding, IFAC Proceedings. IFAC-PapersOnline 2016, 49, 378–383. [CrossRef]
20. Pathak, M.; Jaiswal, D. Application of Artificial Neural Network in Friction Stir Welding: A Review. Int. J.
Technol. Explor. Learn. 2014, 3, 513–517.
21. Zhang, S.; Shi, Q.; Liu, Q.; Xie, R.; Zhang, G.; Chen, G. Effects of tool tilt angle on the in-process heat transfer
and mass transfer during friction stir welding. Int. J. Heat Mass Transf. 2018, 125, 32–42. [CrossRef]
22. Guerdoux, S. Numerical Simulation of the Friction Stir Welding Process. Ph.D. Thesis, École Nationale
Supérieure des Mines de Paris, Paris, France, 2007.
23. Dialami, N.; Chiumenti, M.; Cervera, M.; de Saracibar, C.A. An apropos kinematic framework for the
numerical modeling of friction stir welding. Comput. Struct. 2013, 117, 48–57. [CrossRef]
24. Cervera, M.; Agelet de Saracibar, C.; Chiumenti, M. COMET: Coupled Mechanical and Thermal Analysis,
Data Input Manual, Version 5.0, Technical Report IT-308. 2002. Available online: http://www.cimne.upc.es
(accessed on June 2002).
25. Dialami, N.; Cervera, M.; Chiumenti, M.; Agelet de Saracibar, C. Local-global strategy for the prediction of
residual stresses in FSW processes. Int. J. Adv. Manuf. Technol. 2016, 88, 3099–3111. [CrossRef]
26. Veljic, D.M.; Rakin, M.P.; Perovic, M.M. Heat generation during plunge stage in friction stir welding.
Therm. Sci. 2013, 17, 489–496. [CrossRef]
27. Zhang, Z.; Zhang, H.W. Effect of contact model on numerical simulation of friction stir welding.
Acta Metall. Sin. 2008, 44, 85–90.
28. Chao, Y.J.; Qi, X.; Tang, W. Heat transfer in friction stir welding: Experimental and numerical studies.
J. Manuf. Sci. Eng. 2003, 125, 138–145. [CrossRef]
29. Schmidt, H.; Hattel, J. A local model for the thermomechanical conditions in friction stir welding.
Model. Simul. Mater. Sci. Eng. 2005, 13, 77–93. [CrossRef]
30. Dialami, N.; Chiumenti, M.; Cervera, M.; Segatori, A.; Osikowicz, W. Enhanced friction model for Friction
Stir Welding (FSW) analysis: Simulation and experimental validation. Int. J. Mech. Sci. 2017, 133, 555–567.
[CrossRef]
31. Jung, J.; Oak, J.-J.; Kim, Y.-H.; Cho, Y.J.; Park, Y.H. Wear Behaviors of Pure Aluminum and Extruded
Aluminum Alloy (AA2024-T4) Under Variable Vertical Loads and Linear Speeds. Met. Mater. Int. 2017, 23,
1097–1105. [CrossRef]
32. Sheppard, T.; Wright, D.S. Determination of flow-stress constitutive equation for aluminum-alloys at
elevated-temperatures. Met. Technol. 1979, 6, 215–223. [CrossRef]
33. Pashazadeh, H.; Teimournezhad, J.; Masoumi, A. Numerical investigation on the mechanical, thermal,
metallurgical and material flow characteristics in friction stir welding of copper sheets with experimental
verification. Mater. Des. 2016, 55, 619–632. [CrossRef]
34. Zhu, Y.; Chen, G.; Chen, Q.; Zhang, G.; Shi, Q. Simulation of material plastic flow driven by non-uniform
friction force during friction stir welding and related defect prediction. Mater. Des. 2016, 108, 400–410.
[CrossRef]

© 2018 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/).