1

Friction Stir Welding Laboratory

Casting Forming and Welding Lab (ME39007)
Experiment No. 4: Friction Stir Welding
Objectives:
To perform friction stir welding (FSW) on two similar aluminum sheets of same
gauge by varying the FSW tool rotational and worktable translational speeds.
To analyse the influence of spindle tilt on surface defects (e.g. amount of flash
formation, root flaw, etc.) by visual examination.
To study the influence of FSW tool rotational speed and worktable translation
speed (i.e. welding speed), on welding force (F
x
), axial force (F
z
) and power
consumption.
Instruments required:
Numerically controlled FSW machine, equipped with various sensors (to measure
change in speeds, forces, and torque during welding),
Fixture for clamping the workpiece on the machine table,
Al sheets of dimensions (100 40 2.5 mm
3
),
FSW tool (having flat shoulder surface and straight cylindrical pin),
Few others necessary equipments for tightening and loosing the desired nuts and
bolts.
Hurdles in welding of aluminum:
Differences of hydrogen solubility in liquid and solid states of aluminium alloys,
leads to hydrogen trapping inside the weld; which causes formation of defects like
blow holes, porosity etc.
Formation of aluminium oxide on faying surfaces initially exposed to atmosphere,
which is non-conductive, hard and has high melting point (approx 2072
o
C).
High thermal conductivity as compared to steel, which causes rapid heat
distribution due to conduction within the material
High thermal expansion (almost twice of steel) and high shrinkage volume
(approx 6 % of volume) increases distortion and weld crater size.
Though various techniques like MIG, TIG, rotary friction welding and few others are
used for welding of aluminum alloys; but none of them provides high weld strength
required for aerospace industries, due to significant softening at heat affect zone (HAZ),
distortion etc. Therefore, to achieve high weld strength FSW was developed at The
Welding Institute (TWI), UK in 1991 [1].


2

Theory of FSW:
In friction stir welding a rotating non-consumable tool with specially designed
shoulder and pin is plunged into the abutting edges of firmly clamped work-piece, and
then translated along the weld line to form weld. Thereafter, the tool is finally plunged
out from the work-piece (as shown in Fig. 1). The tool serves two purposes, firstly
softening of material by heat generation due to friction and plastic deformation, and
secondly mixing of material through stirring of pin to form weld.









The side of the work-piece where the tool rotation and tool translation vectors are
same, is called advancing side (AS); and the side where the tool rotation and translation
vectors remain opposite is called retreating side (RS). Furthermore, a typical friction stir
welding cross-section consists of four micro-structural zones; namely stirred zone or
nugget zone (NZ), thermo-mechanically affected zone (TMAZ), heat affected zone
(HAZ) and unaffected or base metal zone (BM). Threadgill [2] was the first who
nomenclatured these FSW zones. The stirred zone is the region where the work-piece
material gets stirred due to influence of both rotary shoulder and pin. Obviously,
materials at this zone remain subjected to high strain, high strain rate and higher
temperature (due to heating by both friction and large plastic deformation). TMAZ is
subjected to heat and some inertia of the materials from the NZ (which causes grain
orientation) and material in HAZ is subjected to heat. Temperature in HAZ of FSW
remains less as compared to HAZ of fusion welding. After HAZ the region which
remains unaffected is called BM. There are a number of parameters which affect the FSW
process. Appendix-I provides some explore on these parameters, up to the level suited for
undergraduate studies.
Mechanics of material flow:
Nature of material flow during FSW is extreme complex and very difficult to
clearly understood. However, most of the researchers agreed that the material flows from
the leading edge of the tool to the trailing edge through retreating side (RS); and at last,

Fig. 1 Basic steps involved in a FSW (left side), and various micro-structural
zones in a typical FSW cross-section (right side); [2]

3

centrifugally forged into the advancing side (AS). They justify this concept by reporting
that, whenever there occurs some significant material loss (either by excessive flash or by
some other means) the total amount of material required to be forged into the AS finds
lacking; and hence, defects like groove and tunnel formation always takes place only at
the AS.
Mechanism of heat generation and stabilization during FSW:
The mechanical work-done by the FSW tool on the work-piece (both by the
frictional rubbing and materials stirring) gets converted into heat; according to the law of
conservation of energy. This heat generation raises the work-piece temperature. With
increase in work-piece temperature, the yield strength of the material decreases.
Therefore, less work is required for further plasticizing of material, and hence, rate of
heat generation decreases. This results in decrease in temperature; and thus, this
mechanism stabilizes the temperature during FSW and avoids melting of the material.
Mechanics of defect formation:
Surface flash, root flaws and worm holes are the defects which occur most often
in FSW; however, few other defects like surface galling, nugget collapse, kissing bond
etc. are also occur time to time. Appendix-II provides detail on these defects. In general,
as the FSW tool is plunged into the material; a key hole is formed, and work-piece
material starts stirring around it. When, the tool starts to travel along the welding line; the
hot plasticized stirring material is forged into the key hole generated due to pin travel.
Thus, improper choice of FSW parameters and/or lack of axial pressure (beneath of the
FSW tool shoulder) leads to defect formation. Zhang et al. [3] proposed a mathematical
model and formulated a relation among axial pressure, tool rotational speed and work-
table translational speed, in order to get a defect free weld. The relation is given below,
(for complete derivation of the relation, please see the appendix-II).

Where, P is the axial pressure,

is the inverse of welding rate, V is the volume of the

key hole, k is a constant depends on the material properties and weld temperature,

and

represent radius of tool shoulder and pin, respectively. According to

this relation for a particular FSW tool, the weld speed must be less than a specific value,
when the axial pressure and tool rotational speed are kept constant; and/or if welding
speed is kept constant, then the axial pressure and tool rotational speed must be larger
than a specific value, in order to achieve a defect free weld.
Experimental procedure:
1. Switch on the MCB switch (electric input to the machine), On/Off switch on the
machine body at the left side and start the computer monitor attached with the
machine.
4

2. Then, double click on the stir welding icon on the desktop in order to start the
software. Once the software gets started; it will ask for pressing hydraulic knob for
switching it on, and then for spindle home.
3. Fix the FSW tool into the spindle through proper dimension of collet, and then tight
the adapter nut using semicircle spanner.
4. Place the work-piece on the anvil of machine work-table. Make sure the sheets are
being placed in such a manner that the joining line is parallel to the machine X-axis,
and then clamp the work-piece using bolts.
5. Once it will be done, choose the manual mode option from the available options.
Insert suitable values for work-table movement, and move the work-table against
fixed FSW tool for obtaining the X
start
, X
end
and Z
end position
values. Here, X
start
, X
end

denotes start and end point of welding on the work-piece clamped, and Z
end position

denotes the point up to which the FSW tool will reach during plunging.
6. Come out from the manual mode and then go to the parameter for specifying the
parameters for weld to be conducted. If required change the spindle tilt angle.
7. After specifying the parameters go to the auto mode and click on icon continue,
to start the weld. But, before pressing cycle on switch, make sure that the coolant
pump for spindle is switched on.
8. After finishing of the weld, switch off the coolant pump, and hydraulic knob.
OBSERVATION TABLE:
















Fig. 2 Work-piece with process parameter details

5

Table 1 Work-piece and tool dimensions
Work-piece dimension (length width,
mm
2
)

Tool shoulder diameter (mm)
Tool pin profile
Tool pin diameter (mm)
Tool pin height (mm)
Work-piece sheet thickness (mm)
Plunge depth (mm)

Table 2 Welding runs with corresponding parameters for the respective weld span
Sr.
No.
Tool
rotational
speed
(rpm)
Welding
speed
(mm/min)
Average
axial force F
z

(N)
Average
spindle torque
(Nm)
Average
welding force
F
x
(N)
1.
Rpm1
Feed rate 1
2. Feed rate 2
3.
Rpm 2
Feed rate 1
4. Feed rate 2

Results and Discussion:
Experimental data will be mailed to the student and they have to do the following
Calculate average value of axial force, welding force and power (= spindle
torque* tool rotational speed).
Fill the observation table with respective values as calculated in above step.
Plot F
z
, F
x
and power with respect to rpm and feed rate. While plotting graphs
keep one parameter constant (e.g. rpm) and vary other parameter (e.g. feed rate),
so that effect of process parameters could be studied on forces and power. Total
three graphs should be plotted. Sample graph is shown below. Interpret the result.








Fig. 3 Sample plot for output parameters versus process parameters

6

Conclusion:
Here, students have to write (very briefly) about the overall effect of spindle tilt
and other varied parameters on weld defects (mainly surface flash and root flaws, by
visual inspection), and on forces and power consumed during welding. Spindle tilt by 1 -
2 (depending upon the work-piece sheet thickness) may reduce the flash formation when
compared to 0 spindle tilt case.
Precautions:
1. Before clamping the work-piece on the anvil the edges of the faying surfaces
should be straight, so that there should not be any gap between faying surfaces; in
order to achieve good weld.
2. The joining line should be parallel to the machine X-axis, and the anvil should be
placed in such a position so that the FSW tool pin centre should not be off-set
from the expected joining line.
3. When coming out from the manual mode, move the spindle up in the Z-axis;
otherwise while selecting auto mode the spindle will directly move to the last Z-
value the spindle had in the manual mode.
4. Before pressing the cycle on knob in the monitor panel the work-piece should be
rigidly clamped on the anvil, and the allen bolts (placed at the right bottom side of
the anvil) must be tightened in order to prevent the vibration of anvil along Y-
axis.
Benefits of using FSW:
There are a number of benefits of using FSW [4]; however, only few important
benefits are addressed here.
1. No distortion of the work-piece takes place; since the process occurs at the solid
state. Also, better dimensional accuracy, better surface finish, and less softening at
the HAZ are achieved; which results in better weld strength.
2. Due to continuous dynamic recrystallization, fine equiaxed grains are developed
in the stirred region, which further improves the mechanical and metallurgical
properties of the weld.
3. Since no consumable is used, so the weld composition remain same as the base
metal; hence, problems associated with the compositional change remain
completely absent here.
4. Since no fumes and no harmful rays are generated, and also no shielding gas is
required; so the process is environment as well as operator friendly.
5. The process is energy efficient, because relatively less energy is required in FSW
when compared to other welding processes (like laser, plasma and electron beam
welding).
6. It reduces the fuel consumption, because the process is widely being used in
aerospace, automotive, shipbuilding and marine industries to fabricate less weight
component, which reduce the fuel consumption.

7

7. Very good scope for repeatability, as the process is easy to automate.
Limitations of FSW:
1. At the end of welding a pin hole remains present in the region where the tool is
plunge out. This leads to some amount of material loss in form of scrap.
2. Significant thinning takes place at the weld region due to formation of ribbon
flash.
3. The process requires adequate reaction force from the bottom side for producing
sound weld; so, use of backing plate is must.
4. Sufficient clamping force is also required in order to oppose the abutting force.
Because of that completely elimination of residual stress is still difficult.
Applications of FSW:
There are a number of fields where friction stir welding is being utilized for
fabrication of various parts [5]. However, few main areas are listed below.
Aerospace industry
1. Wings, fuselages
2. Cryogenic fuel tanks for space vehicles
3. Aviation fuel tanks
4. Repair of faulty MIG welds
Railway industry
1. Rolling stock of railways, underground carriages, trams
2. Railway tankers and goods wagons
Automotive industries
1. Engine and chassis cradles, Body-in-white structure, Tailor welded blanks
2. Wheel rims
3. Truck bodies, Tail lifts for lorries, Armour plate vehicles
4. Fuel tankers
Shipbuilding and marine Industries
1. Panels for decks, sides, bulkheads and floors
2. Boat sections
3. Helicopter landing platforms
4. Marine structures
Construction industries
1. Aluminium bridge
2. Window frames
3. Aluminium pipelines
4. Heat exchangers and air conditioners, using Friction Stir Channelling
8

Electrical industries:
1. Electric motor housings
2. Bush bars
3. Electrical connectors
Electronics industries:
1. Apple next generation iMac [6]
Questions:
Q1. What is the significance of axial force in FSW?
Q2. How does the temperature stabilization occur in FSW?
Q3. What is the role of welding speed on defect formation in FSW?
Q4. Why does the weld power increases with increase in tool rotational speed?

References:
[1] W. M. Thomas, E. D. Nicholas, J. C. Needham, M. G. Murch, S. P. Temple, and C.
J. Dawes, Friction stir butt welding, PCT/GB92/022031991.
[2] H. Jin, C. Ko, S. Saimoto, and P. L. Threadgill, Microstructure of Friction Stir
Welded Joints in AA5182, Mater. Sci. Forum, vol. 331337, pp. 17251730,
2000.
[3] H. Zhang, S. B. Lin, L. Wu, J. C. Feng, and S. L. Ma, Defects formation
procedure and mathematic model for defect free friction stir welding of magnesium
alloy, Mater. Des., vol. 27, no. 9, pp. 805809, 2006.
[4] R. S. Mishra and Z. Y. Ma, Friction stir welding and processing, Mater. Sci.
Eng. R Reports, vol. 50, no. 12, pp. 178, 2005.
[5] R. Nandan, T. Debroy, and H. Bhadeshia, Recent advances in friction-stir welding
Process, weldment structure and properties, Prog. Mater. Sci., vol. 53, no. 6, pp.
9801023, Aug. 2008.
[6] Apple, Website addresses, 2013. [Online]. Available:
https://www.apple.com/in/imac/design/

I

Appendix-I
Parameters affecting Friction Stir Weld
Simply knowing physics of the process and mechanics of weld formation are not
enough; but, the proper understanding of effects of various parameters affecting FSW
process, and various defects occurs in a FSW, their causes and remedies are also essential in
order to produce good quality welds. Here (in appendix-I) a little explore is being given on
those various influencing parameters. Causes and remedies of various FSW defects will be
explored in the next appendix.
There are a number of parameters which affect the FSW process. Broadly they can be
classified as design parameters, process parameters and material parameters. These are
tabulated in Table-A1.
Table A1 Various parameters affecting FSW
Process parameters Design parameters Material parameters
Tool rotational speed Tool shoulder shape and size Workpiece material
Worktable translational
speed
Tool pin shape and size Anvil material
Tool tilt angle (towards the
trailing edge)
Joint geometry Tool material
Plunge depth

Effect of process parameters:
Each parameters do not find direct relationship with all the output
characteristic, rather have influence on only few characteristic; like tool rotational speed has
direct influence on tool torque, but has indirect influence on the welding force and axial
force. Similarly, welding speed has direct influence on the welding force, but has indirect
influence on the axial force; whereas, plunge depth finds direct influence on the axial force
and indirect influence on the welding force. Further, if the tool is tilted towards the trailing
edge; then some part at the leading edge of the tool does not keep contact with the work-
piece. As a result, reduction in flash formation takes place. This also influence the heat
generation, and prevent excessive softening at the HAZ.
Effect of design parameters:
Tool shoulder diameter, its shape (either flat with scrolled, or concave), and
tool pin profile also influence the weld quality up to great extent. Tool shoulder diameter has
direct relationship with the axial force, and tool pin profile has direct relationship with the
welding force. Also, increase in the shoulder diameter reduces the flash formation (up to
some extent) without tilting the spindle. A constant ratio (2.5 3) of tool shoulder diameter to
the pin diameter produces good weld.
II

Effect of material parameters:
Tool material, anvil material and few other environmental conditions decide
the rate of heat dissipation. Use of a good heat conductive material for anvil, makes the
welding of some alloys (titanium and its alloys) easier; whereas for welding of aluminum and
copper alloys, use of low heat conductive materials for anvil reduces the unnecessary heat
loss during welding, and improves the weld quality.
III

Appendix-II
Friction Stir Welding Defects
As it has been seen, that the whole FSW process is carried out by the machine itself
with least human intervention; so defect free weld is quite obvious. However, it has already
been mentioned in the previous section that distinct property of work-piece material and
wrongly chosen parameters can lead to defect formation. Hence, deciding suitable parameters
(e.g. designing of FSW tool shoulder and pin, selection of tool rpm, worktable feed and
plunge force etc.) for a particular work-piece material is a rigorous task. Proper
understanding of various FSW defects (their causes and remedies) can make this rigorous
task a little simpler. That`s why, here an attempt has been made to classify various FSW
defects, and then, on their causes and remedies.
There could be different perspective to classify various FSW defects. One perspective
could be based on the geometrical shape of the defect, like point, line or volume defect.
Another perspective could be based on the location of the defects; like surface defects, and
internal defects. Similarly one more perspective could be based on few certain conditions in
which few specific kinds of defects are formed; like defects from, either too cold weld
condition or too hot weld condition, and defects due to fault in designing of tool and/or
fixture etc. Here, the defects are being classified (Table-A2) considering the latter
perspective.
Table A2 Classification of various FSW defects
Defects due to too cold
condition
Defects due to too hot
condition
Defects due to fault in
design
Lack of fill Ribbon flash Lack of penetration
Lack of fusion (kissing bond) Nugget collapse Excessive indentation
Wormhole Surface galling Oxide entrapment
Root flaws

Lack of fill:
This defect is observed just like a wide crack at the advancing side (shown in Fig.
A1a). It is caused due to insufficient forging action. It can be avoided by either increasing the
plunge force or by tilting the spindle towards the trailing edge of the tool.
Lack of fusion:
It occurs due to insufficient weld temperature. It is also called kissing bond (shown
in Fig. A1b). It can be avoided by either increasing the tool rpm or by doing some
modification in the pin profile.
IV

Wormhole:
It occurs due to either insufficient material to be filled into the cavity at the AS, or due
to too high welding speed (shown in Fig. A1c). It can be avoided by either applying sufficient
plunge force on the tool shoulder or by decreasing the welding speed.
Ribbon flash:
It occurs either due to relatively large plunge force and/or due to mismatch in the
work-piece sheet thickness (shown in Fig. A1j). Sometimes if the sheet thicknesses are same;
yet, slight flash formation takes place only at the RS. If it is so, then it can be avoided by
tilting the spindle towards the trailing edge of the tool.












(Note: All the figures are only for representation of defects, and are not to scale)
Fig. A1 Various FSW defects; (a) Lack of fill, (b) Lack of fusion, (c) Wormhole, (d) Oxide
entrapment, (e) Surface galling, (f) Lack of penetration, (g) Nugget collapse, (h) Root flaws, (i)
Excessive indentation, (j) Ribbon flash.
Nugget collapse:
It occurs due to too high tool rotational speed (shown in Fig. A1g). It can be avoided
by reducing the tool rotational speed.
Surface galling:
It occurs due to sticking of the work-piece material with the tool shoulder and pin
surfaces (Shown in Fig. A1e). Generally, few work-piece materials (like aluminum and few
others) easily attain the temperature where sticking starts to occur. It can be avoided by
regulating parameters which affect the rate of heat generation and dissipation.

V

Root flaws:
It occurs due to sticking of the workpiece material on the backing plate or anvil
(shown in Fig A1h). This condition happens, when the pin height remains slightly larger than
the desired limit. It can be avoided by reducing the tool pin height to the desired limit.
Lack of penetration:
It can be seen from Fig. A1f. It occurs due to slightly smaller pin length. It can be
avoided by increasing the tool pin length to the desired limit.
Excessive indentation:
It is shown in Fig. A1i. Sometimes operator gives relatively larger plunge force in
order to avoid the defect occurs from lack of penetration, without increasing the pin length. It
results in excessive flash formation (also called excessive indentation). It can be avoided by
reducing the plunge force.
Oxide entrapment:
This occurs due to presence of oxides on faying surfaces. Hence, when the sheets are
welded without removing the oxides; it got entrapped into the weld zone, and appears as a
zigzag line (shown in Fig. A1d), when micro-structural analysis is done. In order to prevent
this defect to occur, the oxides layers should be removed from the faying surfaces before
welding.
A mathematical model for defect free welds:










Fig. A2 (a) Creation of cavity due to pin travel (top view), (b) FSW tool subjected to
plunge force (side view)

AS
RS
P

(a)
(b)
VI

As already mentioned that; when FSW tool is inserted into the workpiece, a pin hole
is formed, and when the tool starts to translate along the joining line the material stirring
around the pin hole (i.e. around the tool pin) is centrifugally forged into the cavity created
due to pin translation. So, the FSW process can be considered as the micro-hole formation
and filling process (as shown in the Fig. A2) [3].

Let us consider that;

= Welding speed

= Tool rotational speed

= Welding pressure
= Welding tool linear movement per revolution
= Workpiece sheet thickness

= Radius of tool shoulder

= Radius of tool pin

= Welding rate

= Inverse of the welding rate

= Volume of the micro hole created due to pin travel per revolution
= Volume of the materials centrifugally forged into the hole per second
= Time that the shoulder play effect on the microhole
Now, for a constant welding pressure P and constant tool rotation speed

; the time for

which the shoulder play effect on the micro-hole would be,
=

(1)
If the micro-hole gets completely filled, then the fill metal must be equal to the volume of the
micro-hole. If the time of filling of metal into the micro-hole is
1
, then
=
1


1
=

(2)

Obviously, the time for which the shoulder play effect on the micro-hole should be greater
than the time for filling the hole, in order to avoid any void or hole to remain present. Hence,
VII


1
(3)
Now, combining Eqs. (1) and (2), gives the following equation.

(4)
The volume of the material centrifugally forged into the micro-hole per second depends upon
the tool rotational speed, if the other parameters are kept constant. So,

Similarly, the frictional force under the shoulder will increase with the increase in the
welding pressure (i.e. vertical force on the FSW tool). According to the momentum
theory
0
= ; so the forging speed (i.e. speed with which the material is being
centrifugally forged into the micro-hole) will also increase with the FSW tool axial force.
Hence;

The above relationship between S and the welding pressure P, and the rotational speed

can
be combinly written as;
=

(5)
In the above equation, k is the proportionality constant, and it depends upon the material
properties and the welding temperature.

(6)

Now, if the tool pin size is also considered, then the Eq. (6) can be modified as;

)
(7)
For a particular FSW tool and work-piece material Eq. (7) can further be simplified as;

(8)
Where, would be constant for a particular FSW tool and work-piece material combination.
Thus, if the welding pressure and the tool rotational speed are kept constant, then the welding
speed must be less than a specific value, and/or if the welding speed is kept constant, then the
welding pressure and tool rotational speed must be larger than a specific value; in order to
avoid any voids. Hence, these three welding parameters are dependent on each other, if rest
other parameters are kept fixed.