Materials Transactions, Vol. 49, No. 10 (2008) pp.

2179 to 2184
Special Issue on Advanced Light Metals and Processing in Asia 2007
#2008 The Japan Institute of Light Metals

Influence of Welding Parameters on Mechanical Properties

of Friction Stir Welded 6061-T6 Launch Box
Hsing-Ta Hsieh1;2; * and Jahau Lewis Chen1
1
Department of Mechanical Engineering, National Cheng Kung University, Tainan, Taiwan 701, R. O. China
2
Metal Industry Research & Development Centre, Kaohsiung, Taiwan 811, R. O. China

Friction stir welding (FSW) is a welding process which deals with joining parts in a solid state at low temperature to result in welded parts
with excellent mechanical performance, such as low distortion and high tensile strength. Additionally friction stir welding is applicable to
aluminum alloy products with precision dimensions. By using friction stir welding parameters, this research studies the tensile strength,
hardness, elongation rates, and shrinkage of extruded 6061-T6 alloy. Results indicate that the joining strength of the extruded 6061-T6 alloy can
reach 78% of the base metal after friction stir welding. Meanwhile, welding parameters can accurately predict and control the welding distortion
of welded products. This research applies these results in the manufacturing of launch boxes to arrive at a technology that can be directly applied
to welded products without expensive as-welded modifications. [doi:10.2320/matertrans.L-MRA2008829]

(Received December 25, 2007; Accepted May 13, 2008; Published September 25, 2008)
Keywords: friction stir welding, shrinkage, 6061-T6 alloy, launch box

1. Introduction Therefore, FSW is classified as a low temperature/solid state

joining process.1) The resulting weld has low distortion, low
Missile and rocket systems play vital role in the modern shrinkage, without porosity and cracking that are commonly
military equipment. The systems have strong mobility and occurred in traditional arc welding. FSW is especially
high reliability. Additionally these missile systems are well effective in the use of joining aluminum alloys.2,3) Basic
equipped with air seals, pressure resistance, heat resistance study and industrial applications of FSW are gradually and
and corrosion resistance to facilitate daily storage and long rapidly developed. For the last few years, many in-depth
distance transportation. The box must be able to sustain the FSW studies have been conducted about materials, welding
missile weight, recoil and high temperature after launching. parameters, microscopic structure, mechanical property4–12)
Aluminum alloy has a high strength/weight ratio and and the design of stir tools.13–16) Comparison of the related
enough stability to effectively reduce the weight of a launch studies indicates that differences in materials, experimental
box. Additionally, aluminum alloys can be repeatedly used equipment and the designs of stir tools will change the
and recycled. relationship between operational parameters and mechanical
Since the design of a launch box has to take internal properties. However it is difficult for above mentioned
dimensions and housing strength into account, the feasible researches in the practical application.
manufacturing methods include machining, extrusion and 6061-T6 aluminum alloy is chosen as material for launch
assembly. (1) Machining: Machining has limitations of box, this research conducts the study of necessary welding
processing equipment and cost. (2) Aluminum extrusion: parameters, including rotation speed, traveling speed, shoul-
The launch box can be easily and rapidly formed using der diameter, pin diameter and mechanical properties, such as
extrusion, but the required extrusion equipment and molds tensile strength, weld hardness, elongation rate and shrink-
are very expensive. After launch box is formed, it is very age. Additionally this research hopes to locate properly
difficult to modify the inner dimension and distortion. (3) applied parameter combinations to achieve desired mechan-
Welding assembly: The four sides of a launch box are ical properties and to control joining shrinkage change to
separately processed using welding methods and assembly. reach a net joining of the inner box without as-welded
Material distortion produced by high fusion temperatures in treatment.
the weld is hard to modify and predict in advance. However
above mentioned methods are not economical or practical. 2. Experimental Procedures
As a result, this research introduces FSW to demonstrate a
new manufacturing method for launch boxes. The research adopts extruded 6061-T6 aluminum alloy
FSW is a joining process and was developed by TWI of the with a composition analysis of 0.59%Si, 0.12%Fe, 0.23%Cu,
United Kingdom in 1991. FSW uses spinning of a tool to 0.009%Mn, 0.99%Mg, 0.027%Zn, 0.007%Ni, 0.09%Cr,
produce heat on the surface of the work piece along with 0.007%Ti, 0.007%Pb and balance Al(in mass%).
mechanical stir to generate material plasticization and to The FSW experiment uses a Makino Milling Machine with
achieve joining through extrusion and a forging mechanism. mandrel power of 3.75 KW. Figure 1 displays experimental
Since alloy materials have low heat input, the joining positioning and sampling to show a welding specimen with a
temperature of a work piece does not reach melting point. size of 250
80
9 mm, while the specimen is fixed on one
side and the other side of the specimen is applied with a fixed
*Graduate Student, National Cheng Kung University, Corresponding torque to make two pieces of specimen to be tightly held
author, E-mail: [email protected] together. At both sides of weld, flat blocks are used to
2180 H.-T. Hsieh and J. L. Chen

(a) (b)

Fig. 3 The diagrams of (a) cross section of box and (b) individual part
before welding.

that are shown in Fig. 2. At the beginning of the welding

process, the pause of 8 seconds is set in the boundary between
Fig. 1 The tensile specimens are sampled in friction stir welding tests. the leading plate and the work piece.
This research analyzes the influence of tensile strength,
hardness, shrinkage and elongation rate on the weld when
welding parameters change. According to ASTM E8M,17)
Table 1 The experimental parameters of friction stir welding.
tensile strength analysis adopts the MTS 810 system
Basic parameters1 1000KNC with testing conditions at 0.2% initial speed of
Changed Variable Setting Values 2.0 mm/min and a secondary speed 6.0 mm/min are used.
A Shoulder (mm) 16 20 24 28 — — — The measuring distance of the elongation rate is 50 mm.
2 2 Micro-hardness distribution tests on the weld using 100 g
B Pin (mm) 3 4 6 8 10 — —
load, the testing location is a 3 mm distance under surface and
C Translation (mm/min) 50 150 250 350 450 550 650
hardness data is taken every 1 mm. In terms of shrinkage
D Rotation (rpm) 560 700 900 1100 1400 1800 — measurement, linear scale is used to measure the dimensional
E Fixed force (Nm) 0 2.825 5.65 8.475 11.3 14.125 — change between pre-weld and as-welded at the same location
1
The basic parameters are 1100 rpm, 250 mm/min, pin-8 mm, shoulder- on the experimental machine. In the welding process, the
20 mm, fixed force-5.65 Nm and tilt angle-2 . work piece is limited by normal grip; the research does not
2
The pin broke during the friction stir welding process. consider the influence of as-welded warp.
This research results apply feasible welding parameters to
the joining of aluminum alloy launch box parts. Through
prevent the specimen from warping. The leading plates are material preparation, machining and joining, the dimension
added at the front and rear ends of the specimen to serve as and performance of the finished product is also tested. In
beginning and ending of the weld and to maintain specimen the actual manufacturing of a launch box, the box is divided
stability during the welding process. into four units based on cross-sectional shape, as shown in
Using a selected material, Table 1 shows the values of the Fig. 3. Each unit adopts a single cross-sectional extruded
welding parameters in the FSW which may affect the 6061-T6 alloy. Through machining, the extruded unit is
mechanical properties of the work piece. In the beginning, made to conform to pre-assembly dimensions. Afterwards
the research sets basic reference parameters that are this research conducts assembly welding after obtaining
established through experience. Each set of experiment feasible welding parameters. Finally, this research carries
changes one parameter; the rest remains at the original basic out welding quality testing and precision measurements of
settings among which pins use columns with standard threads important dimensions to verify the feasibility of FSW.

Fig. 2 The photograph of all kind friction stir tools.

 Influence of Welding Parameters on Mechanical Properties of Friction Stir Welded 6061-T6 Launch Box 2181

Strength-Traveling Speed
(a) 350
UTS-PM (c) Hardness - Translation Speed
316 YS-PM
130 50mm /min
UTS
300 150mm /min
YS 120
250mm /min
Strength (MPa)

253.8 110 350mm /min

450mm /min

Hv (100g)
250
100 550mm /min
242.1 247.9 240.6
231.3 90 650mm /min
222
205.3 218.5
200
80
70
150 161.2
150.9 156.3 162.7 162.7 164.6 60 RS AS
137.7
50
100 20 16 12 8 4 0 -4 -8 -12 -16
50 150 250 350 450 550 650
Distance to Nugget Center(mm)
Traveling Speed (mm /min)
(b) (d)
Hardness - Shoulder Diameter Elongation - Translation Speed
18
130 EL-PM
shoulder-16mm 16.2
shoulder-20mm
16 EL

Elongation Rate (%)

120
shoulder-24mm 14
110 shoulder-28mm
12
100
10
Hv (100g)

90 8 7.3 6.4
5.6 6.3 6.2
80 5.7
6
70 4
3.1
60 2
RS AS 50 150 250 350 450 550 650
50
20 16 12 8 4 0 -4 -8 -12 -16 -20 Translation Speed (mm /min)
Distance to Nugget Center(mm)

Fig. 4 The mechanical behaviors are relative to the welding parameters (a) strength to travelling speed, (b) hardness distribution to
shoulder diameter, (c) hardness distribution to travelling speed, (d) elongation rate to travelling speed.

3. Results and Discussion (a)

Research results indicate that the sampling points are the

average of data taken from three sets of specimens. For the
base metal of the extruded 6061T-6 alloy, the average tensile
strength is 316 MPa, yielding strength is 253.8 MPa, elonga-
tion rate is 16.2% and hardness is Hv 120 respectively.

3.1 Tensile strength

These research results indicate when shoulder diameter
progressively increases from 16 to 28 mm, the tensile
strength decreases from 247.9 MPa to 225.4 MPa. As soon as
pin diameter gradually increases from 6 to 10 mm (the pin
diameter 3 mm & 4 mm broke in experiment), tensile
strength decreases from 235.2 MPa to 227.4 MPa. However,
applied side force from 0 to 14.125 Nm does not remarkably (b)
affect tensile strength. Its value remains around 230:3 
4:9 MPa.
Stir rotation speed changes from 560 to 1800 rpm; the
result indicates that a peak of 235.7 MPa is shown at 900 rpm.
However at 900 rpm, tensile strength shows a slow decrease.
Change of traveling speed has significant influence on
tensile strength. The corresponding changes are shown in
Fig. 4(a). The result demonstrates that a translation scope of
350550 mm/min has better tensile strength. In Fig. 4(a), a
maximum tensile strength value of 247.9 MPa is shown
around 450 mm/min.
Figure 5 exhibited the average grain size for low traveling
speed was significantly bigger than high traveling condition.
It was implied the ultimate tensile strength decreased at low
traveling speed. But the distribution of precipitates for these Fig. 5 Microstructure of Thermo Mechanically Affected Zone, 100X
specimens displayed very much alike. (a) traveling speed 50 mm/min (b) traveling speed 550 mm/min.
2182 H.-T. Hsieh and J. L. Chen

decreasing, the hardness decreases and hardness changing

area expands.

3.3 Elongation
In terms of elongation rate, the results indicate that
changes of pin diameter, shoulder diameter, rotation speed
and side force do not affect elongation rates too much. The
average changes fall between 5.5 and 6.8%. When stir
traveling speed is slow, the elongation increases. However
650 mm/min speed shows a tremendous elongation decrease
of 3.1%, the result shows in Fig. 4(d). Weld microscopic
structure indicates that unfilled porosity is occurred in the
Fig. 6 The macrograph of porosity exists in advancing side (travelling advancing side of the weld at a speed of 650 mm/min,
speed is 650 mm/min).
shown as Fig. 6.

On the other way, the linear heat input is small for high 3.4 Shrinkage
traveling speed; it will cause insufficient fusion and create With regard to side shrinkage, when shoulder diameter
some defects in nugget (Fig. 6). If the defect is occurred on increases, the sidewise shrinkage produced by as-welded
the fusion zone, it also reduces the tensile strength. specimens increases at the same time. When pin diameter
Result indicates that traveling speed is increasing, linear changes, the same situation is also occurred as shown in
heat input is then decreasing. Furthermore more fine grain Fig. 6(a) and (b). However welding traveling speed changes,
size on welded zone and get higher strength are the results, an increase of traveling speed produces smaller shrinkage as
yet a peak is occurred at different traveling speeds. shown in Fig. 7(c). The higher rotation speeds generate more
shrinkage, as Fig. 7(d). From Fig. 7(c)–(d), traveling speed
3.2 Micro hardness is much more sensitivity than rotation speed on the effect
Measurements of weld hardness indicate that changes of width shrinkage. The side force does not show significant
in rotation speed, pin diameter and side force have no influence on shrinkage.
significant difference on weld hardness distribution. However Figure 7(a)–(d) shows specimen shrinkage under different
when shoulder diameter enlarges, hardness changing area parameters relating to different slope. The slopes are nearly
expands as well. In Fig. 4(b), the expanding area is in direct linear, increase at beginning stage and gradually reach stable
proportion to changes of diameter. conditions. For welding procedure, the specimen temperature
The influence of traveling speed on hardness distribu- will increase from room temperature to a high and stable
tion is shown in Fig. 4(c). As soon as traveling speed is temperature situation. The experimental specimen length

Wid Shrinkage - Shoulder Diameter Width Shrinkage - Travelling speed

(a) 0.8
(c)
shoulder 16 0.8
50 mm /min
0.75 shoulder 20
0.75 150 mm /min
0.7 shoulder 24
0.7 250 mm /min
shoulder 24
0.65 0.65 350 mm /min
shoulder 28
Shrinkage (mm)

450 mm /min
Shrinkage (mm)

0.6 0.6 550 mm /min

0.55 0.55 650 mm /min
0.5 0.5
0.45 0.45
0.4 0.4
0.35 0.35
0.3 0.3
0.25 0.25
0.2 0.2
5 45 85 125 165 205 245 5 45 85 125 165 205 245
Distance from Beginning(mm) Distance from Beginning(mm)

(b) Widef Shrinkage - Pin Diameter (d) Width Shrimkage - Rotation Speed
0.65 pin 6mm 0.65
pin 8mm 560 rpm
0.6 0.6
pin 10mm 700 rpm
0.55 900 rpm
0.55
1100 rpm
Shrinkage (mm)

Shrinkage(mm)

0.5 0.5 1400 rpm

0.45 0.45 1800 rpm

0.4 0.4

0.35 0.35

0.3 0.3

0.25 0.25

0.2
0.2
5 45 85 125 165 205 245
5 45 85 125 165 205 245
Distance from Beginning(mm) Distance from Beginning(mm)

Fig. 7 The shrinkage distribution relates to the welding parameters (a) width shrinkage to shoulder diameter, (b) width shrinkage to pin
diameter, (c) width shrinkage to travelling speed, (d) width shrinkage to rotation speed.
 Influence of Welding Parameters on Mechanical Properties of Friction Stir Welded 6061-T6 Launch Box 2183

Width Shrinkage- Approved Test

0.45
250mm /min
0.4 650mm /min
Shrinkage (mm)

0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
20 100 180 260 340 420 500 580
Distance from Beginning (mm)

Fig. 8 The width shrinkage distribution of approved test.

Fig. 11 The precision fixed position of finished box.

Table 2 Examinations of as-welded box.

Test Items Designed Valve As Welded Value

Nugget X-ray — Accept
Main Circle 252 þ 0:3 mm 252:164 mm
90 90 90
Fig. 9 The appearance of each as-machined part before FSW.
Corner 90 89.90 90.16
 
Angle/m 90 89.97 89.91
90 90.13 89.93

completed launch box. After external milling, the welded

launch box is completed. Furthermore this research conducts
an important internal dimension measurement and an X-ray
Welded inspection of the weld. The inspection unit is every meter.
position The checking positions are shown as Fig. 11 and the results
are displayed on Table 2.
Weld at each point in the assembly goes through X-ray
non-destruction tests so that corner angles of as-welded
distortion within the launch box and as-welded central
diameters can be precisely estimated and controlled to
Fig. 10 The appearance of launch box after FSW. achieve the expected results without post machining on
internal launch box.
cannot indicate the complete shrinkage situation. To further
confirm the shrinkage-free area, a shrinkage test of the 4. Conclusions
specimen at 600 mm in length is conducted at two separate
traveling speeds. The result indicates when welding starts and (1) In FSW, changes in shoulder diameters affect weld
the process moves to a certain distance, the shrinkage quality than those in pin diameters. Hardness distribu-
gradually stabilizes and fluctuation remains within 0.1 mm tion and cross-sectional microscopic structure observa-
as shown in Fig. 8. tion indicate that enlargement of shoulder diameters
will expand the heat affected zone of welds and enlarge
3.5 Parameters application as-welded side shrinkage.
To summarize the relationship between welding parame- (2) Proper grip is a necessary process for FSW. Research
ters and mechanical properties, this research selects a feasible indicates as long as a grip mechanism or fixture
and stable parameter combination and applies in the welding constraint stiffness is sufficient, it is not necessary to
process. Figure 9 shows the work pieces that are machined apply additional force.
extruded parts before joining. This research conducts the (3) Changes in traveling speeds demonstrate significant
welding is conducted after necessary assembly in a Cincin- influence on the mechanical properties of welds. At
nati Milacron milling machine. Figure 10 illustrates the low speeds, unit weld lengths have high input heat
2184 H.-T. Hsieh and J. L. Chen

which enlarges the heat affected zone, affect the on Friction Stir Welding, (2001).
3) M. Wood, S. Larsson and H. Dahlstrom: 3rd International Symposium
changes in hardness distribution, produce bigger grain,
on Friction Stir Welding, (2001).
reduce joining strength and increase as-welded shrink- 4) S. Lim, S. Kim, C.-G. Lee and S. Kim: Metall. Mater. Trans. A 35A
age. The high speed area has higher tensile strength and (2004) 2829–2835.
low shrinkage. 5) T. J. Lienert and R. J. Grylls: 1st International Symposium on Friction
(4) Research and application indicate that FSW can Stir Welding, (1999).
precisely control the quality and dimension of finished 6) A. von Strombeck, J. F. dos Santos, F. Torster, P. Laureano and M. K.
Ocak: 1st International Symposium on Friction Stir Welding, (1999).
products in the welding process to achieve precision 7) H. J. Liu, H. Fujii, M. Maeda and K. Nogi: 4th International
joining without post machining. Symposium on Friction Stir Welding, (2003).
8) Y. S. Sato, M. Orata and H. Kokawa: Metall. Mater. Trans. 33A (2002)
Acknowledgments 625–635.
9) L. Liu, H. Nakayama, S. Fukumoto, A. Yamamoto and H. Tsubakino:
Mater. Trans. 45 (2004) 285–291.
This research is supported by the Infrastructure Project of 10) S. Kumai, G. Wakana, M. Yamawaki, K. J. Lee and A. Sato: J. Japan
Mechanical And Automation Technology of MIRDC 94-EC- Inst. Light Metals 53 (2003) 542–547.
17-A-08-R7-0707, which is sponsored by the Ministry of 11) R. W. Fonda and J. F. Bingert: Metall. Mater. Trans. A 35A (2004)
Economic Affairs, R.O.C. The author wishes to express his 1487–1499.
gratitude to Dr. Tze-Ching Yang and Mr. Wo-Zone Chang 12) K. Katon, H. Tokisue and G. Ito: J. Light Metal Welding &
Construction 41 (2003) 28–36.
of MIRDC for their assistance on experiments, testing and 13) M. Boz and A. Kurt: Materials and Design 25 (2004) 343–347.
modifying manuscript. 14) Y.-H. Zhao, S.-B. Lin, L. Wu and F.-X. Qu: Mater. Lett. 59 (2005)
2948–2952.
15) H. Fujii, L. Cui, M. Maeda and K. Nogi: Mater. Sci. Eng. A 419 (2006)
REFERENCES 25–31.
16) K. J. Colligan, J. Xu and J. R. Pickens: Friction Stir Welding and
1) C. J. Dawes and W. M. Thomas: Weld. J. March, (1996) 41–46. Processing Volume II (2003) 181–190.
2) J. Mononen, M. Siren and H. Hanninen: 3rd International Symposium 17) Annual book of ASTM Standards, Volume 03.01, (2006) pp. 88–111.