Professional Documents
Culture Documents
Abstract
Purpose – The purpose of this paper is to evaluate temperature and properties at interface of AISI 1040 steels joined by friction welding.
Design/methodology/approach – In this study, AISI 1040 medium carbon steel was used in the experiments. Firstly, optimum parameters of the
friction welding were obtained by using a statistical analysis. Later, the microstructures of the heat-affected zone are presented along with micro
hardness profiles for the joints. Then, the temperature distributions are experimentally obtained in the interface of the joints that is formed during the
friction welding of 1040 steels with the same geometry. This study was carried out by using thermocouples at different locations of the joint-interface.
The results obtained were compared with previous studies and some comments were made about them.
Findings – It was discovered that temperature had a substantial effect on the mechanical and metallurgical properties of the material.
Research limitations/implications – The maximum temperature in the joint during frictional heating depends not only on the pressure, but also on
the temperature gradient which depends on the rotational speed in particular. It is important to note that the measurement process was successfully
accomplished in this study although it was particularly difficult to obtain temperature due to the large deformations at the interface. Future work could
be concentrated on the temperature measurement of the joined materials.
Practical implications – Temperature is one of the most important of all physical quantities in industry. Its measurement plays a key part in industrial
quality and process control, in the efficient use of energy and other resources, in condition monitoring and in health and safety. This paper contributes to
the literature about temperature measurement in welded, brazed and soldered materials.
Originality/value – The main value of this paper is to contribute and fulfill the influence of the interface temperature on properties in welding of
various materials that is being studied so far in the literature.
Keywords Friction welding, Temperature measurement, Steels
Paper type Research paper
308
Evaluation of temperature and properties Assembly Automation
Hilmi Kuscu, Ismail Becenen and Mumin Sahin Volume 28 · Number 4 · 2008 · 308 –316
Figure 2 Parameters on continuous drive friction welding measurements were made at different locations using
thermocouples. A finite element model was used to
Friction Time (tf) Braking determine the appropriate coefficient of the friction to fit
the experimental data. The predictions of the coefficient of
Waiting Upset Time(tu) the friction were in close agreement with the experimental
results obtained from the direct drive friction welding trials on
Upset Pressure (Pu) the 1045 steel. The current results suggest that the new
Rotational Speed (s) friction law may be used to determine the effects of friction
welding parameters on friction coefficients in other material
Friction Pressure (Pf) systems. Sahin (2001) investigated the effects of the work-
piece dimensions and the plastic deformation on the friction
Torque (T) welding method. Then, Sahin and Akata(2001) examined an
experimental study on the application of friction welding for
Starting parts with different diameters and width. D’Alvise et al.
Shortening
Time (2002) studied the finite element modelling of the inertia
friction welding process between dissimilar materials. Sahin
and Akata (2003) carried out the joining of plastically
ambient temperature chuck ends. They discussed the deformed steel with friction welding by using a statistical
establishment of the boundary conditions using the analysis. Akata and Sahin (2003) directed research on the
continuous drive method to join AISI 4140 steel tubes. effect of dimensional differences in the friction welding of the
Imshennik (1971) examined the heating properties in friction AISI 1040 specimens. Lambrakos et al. (2003) directed a
welding. Healy et al. (1976) carried out an analysis of study on analysis of friction stir welds using thermocouple
frictional phenomena in the friction welding of mild steel. measurements. The friction stir welding process was analysed
Kinley (1979) directed a study on the friction welding set up. via an inverse problem approach, using experimentally
Sluzalec (1990) developed a finite element model to simulate obtained thermocouple information to constrain the thermal
this process and to represent the work pieces and surface field of the model in their studies. Triantafyllidis et al. (2003)
contact conditions. The predictions of the temperature investigated comparison of high power diode laser and
distribution, thermal expansion and thermo-plastic stresses Nd:YAG laser microwelding of k-type thermocouples. Ferro
were obtained from this model. The comparison of the et al. (2005) carried out investigation of electron-beam
analytic results to the test data were presented and discussed welding in wrought Inconel 706 – experimental and numerical
by the author. Nentwig (1996) investigated the effect on cross analysis. Moreira et al. (2007) examined the temperature field
section differences of components on the joint quality of acquisition during gas metal arc welding using
friction welding and stated that friction pressure, upset thermocouples. The paper presents and compares
pressure and rotation speed must be changed in the friction measurements made in welded plates of aluminium alloy
welding of the different cross-sections. Bendzsak et al. (1997) 6082-T6. Tests were performed in both plate surfaces and a
investigated a numerical model in friction welding. Fu and good agreement between the three techniques was found.
Duan (1998) carried out an analysis of the coupled thermo- In this study, AISI 1040 medium carbon steel was used in
mechanical problem during friction welding by using a finite the experiments. Firstly, the optimum parameters of friction
element method, according to the constitutive relation of a welding were obtained by using a statistical analysis. Later,
large elasto-plastic deformation and the principle of the the microstructures of the HAZ are presented along with the
virtual work in their studies. Then, the heat flow and stress- micro hardness profiles for the joints. Then, the temperature
strain process at the heating stage of the friction welding were distributions are experimentally obtained in the interface of
simulated, and the law of the variation of temperature, stress the joints that is formed during the friction welding of the
and the strain fields during friction welding were 1040 steels with the same geometry. This study was carried
systematically investigated by the authors. Balasubramanian out by using thermocouples at different locations of the joint-
et al. (1999) presented the results of a combined experimental interface. The results obtained were compared with previous
and a numerical study of the continuous drive friction welding studies and comments were made.
of the 1045 steel. A new friction law was proposed for the
estimation of the apparent coefficient of the friction during
2. The experimental procedure
direct drive friction welding. Temperature distributions were
empirically and numerically predicted in the heat-affected 2.1 The experiment set up
zone (HAZ) that is formed during the direct drive friction The friction welding set up designed and constructed for
welding of 1045 steel to 1045 steel. The temperature experimental part of the present study is shown in Figure 3.
309
Evaluation of temperature and properties Assembly Automation
Hilmi Kuscu, Ismail Becenen and Mumin Sahin Volume 28 · Number 4 · 2008 · 308 –316
Figure 3 Continuous drive friction welding set-up The basis of this approach is the assumption of a simplified
linear model for the optimization parameter h given by
Drive Work-pieces Hydraulic h ¼ b0 þ b1x1 þ b2x2 þ . . . , where x1, x2 . . . , etc. are the
Sliding Guide
Group
factors which h depends on and b0, b1, b2 . . . , etc. represent
1 the “true” values of the corresponding unknowns. From the
results of an experiment comprising a finite number of trials,
one can arrive at sample estimates of the coefficients, b, which
2
are then usually fitted into a linear regression equation of the
Chuck Clamps
Clutch type y ¼ b0 þ b1x1 þ b2x2 þ . . . , where y is the response
Bearings
Friction T
function and the bs are the “estimated” values of the bs. In
3 simple terms, each coefficient represents the influence of the
Pressure P T
Control corresponding factor on the quality of the weld expressed by
M: Motor Upset the optimization parameter.
P: Pressure Pressure The statistical analysis involves two steps. The first step is
T: Tank exit Control the adequacy of the model tested. A suitable method is based
P T
1: Return Valve on the Fischer or “F” ratio, which can be used to confirm if
2: Pressure Indicator the terms in the assumed linear function are statistically
3: Solenoid Valve M
significant. The second step is to obtain optimal estimates of
the regression coefficients for the significant factors, which
may be carried out by using the method of the fewest squares
Oil Tank
(Sahin and Akata, 2003; Draper and Smith, 1981). The
parameter optimization was carried out by using the factorial
The set-up was designed and constructed according to the design of the experiments. In the present study, the friction
principals of continuous drive welding machines. A drive motor time and friction pressure were chosen as the two factors. The
with 4 kW power and 1410 rpm was selected as adequate for the other parameters such as upset time, upset pressure and
torque capacity in the friction welding of the steel bars within rotational speed were kept constant. Experimental results are
10 mm diameter taking into account the friction and the upset given in Table II.
pressures. The friction and the upsetting pressures can be seen Firstly, the optimal estimates of the regression coefficients
on number 2 pressure indicator, and the stages of the welding were obtained by using the method of the Fisher ratio. The
sequences are controlled by number 3 solenoid valve driven by
resulting equation is also given below:
an external timer.
y ¼ 14; 146 2 32; 533x1 þ 45; 067x2
2.2 Test parts and geometry of parts
The chemical composition of the AISI 1040 steel is given in
However, if the correlation coefficient is examined in the
Table I (Stahlschlüssel, 1995).
resulting equation, it is quantitatively shown that the effects of
The experiment specimens were machined from AISI 1040
friction time and friction pressure on the tensile strength are
steel on the geometry below. The geometry of parts is given in
as significant as expected.
Figure 4.
Later, the parameters having the least error by using the
It is thought that friction time and friction pressure has a
method of fewest squares were taken as the optimum welding
direct effect on the tensile strength of the joints. Therefore, in
parameters. Hence, the optimum parameters were found
the last years, the statistical analysis which has been used in
order to discover the effects of the parameters that have a (30 MPa) for friction pressure and (5 sec) for friction time.
significant role on the results in such studies (Draper and The parameters used in the friction welding experiments are
Smith, 1981; Akata and Sahin, 2003; Sahin and Akata, 2003) given in Table III.
was also used in this study. Besides this, on thinking that the The effects of friction time and friction pressure on the
parameters affect the results directly, the linear style was used welding strength of the joints were examined in the welding of
in the statistical analysis. The statistical analysis in this study the parts with an equal diameter. Results of the two sets of the
is given in section 2.3. welding experiments, keeping the upset time and the upset
pressure constant, respectively, as 20 sec and 110 MPa, are
2.3 Selection of the optimum parameters shown in the diagrams of Figures 5 and 6.
Firstly, tests were conducted to determine the optimum As seen in Figures 5 and 6, the tensile strengths of the joints
parameters for a convenient joint. The optimum parameters increase as friction time and friction pressure increase.
were determined by using statistical analysis with 10 mm However, after a value where the maximum tensile strength is
specimens having an equal diameter. The welding obtained, the joints become over-deformed and the joint-
experiments were directed to obtain an optimum friction interface loses its property. Consequently, those friction time
time and friction pressures by using the upset time (20 sec) and friction pressure values which increase after that
and the upset pressure (110 MPa). maximum value decrease the tensile strengths of the joints.
310
Evaluation of temperature and properties Assembly Automation
Hilmi Kuscu, Ismail Becenen and Mumin Sahin Volume 28 · Number 4 · 2008 · 308 –316
Figure 4 Part dimensions used in the experiments Figure 5 The effect of friction time on the tensile strength
D = 10 (mm) d =10 (mm)
s (rpm) AISI 1040 D = 10mm. , d = 10mm.
Pf (MPa), (Pf = 30 MPa. Pu = 110 MPa. tu = 20sec)
tf (sec) 1,000
900
311
Evaluation of temperature and properties Assembly Automation
Hilmi Kuscu, Ismail Becenen and Mumin Sahin Volume 28 · Number 4 · 2008 · 308 –316
Figure 6 The effect of friction pressure on the tensile strength acquired through a data-logger. After the motor stopped,
pressure was increased to 110 MPa. The latter pressure was
AISI 1040 D = 10mm., d = 10mm. maintained until the parts cooled down (20 sec). The
(tf = 5sec. tu = 20sec. Pu = 110 MPa.)
1,000 temperature changes in the interface of joints were
monitored until the temperature decreased to the levels
below room temperature.
900
R3
R2
Locations of thermocouples
at front-face of parts
Pf (MPa),
tf (sec)
R4 (Centre of Part)
R1
312
Evaluation of temperature and properties Assembly Automation
Hilmi Kuscu, Ismail Becenen and Mumin Sahin Volume 28 · Number 4 · 2008 · 308 –316
Figure 8 The photo of GreenLine HOBO U12 data-logger Figure 10 Data-logger connected to computer and thermocouple in
welded parts
(a)
(b)
313
Evaluation of temperature and properties Assembly Automation
Hilmi Kuscu, Ismail Becenen and Mumin Sahin Volume 28 · Number 4 · 2008 · 308 –316
650
600 Horizontal
Distances (x)
550
500
Vickers Hardness (HV)
450 Weld-Center
400
350
300
250
200
150
100
50
0
–4 –3 –2 –1 0 1 2 3 4
Horizontal Distance (mm)
314
Evaluation of temperature and properties Assembly Automation
Hilmi Kuscu, Ismail Becenen and Mumin Sahin Volume 28 · Number 4 · 2008 · 308 –316
1,000
800
600
400
200
0
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Time (s)
315
Evaluation of temperature and properties Assembly Automation
Hilmi Kuscu, Ismail Becenen and Mumin Sahin Volume 28 · Number 4 · 2008 · 308 –316
.
In this study, the effect of the welding optimum values Healy, J., McMullan, D.J. and Bahrani, A.S. (1976),
achieved with the help of the statistical analysis on the “Analysis of frictional phenomena in friction welding of
joining quality was important, and helped in the case of mild steel”, Wear, Vol. 37, pp. 279-89.
carrying out experiments in a short time and obtaining Imshennik, K.P. (1971), “Heating in friction welding”, Weld
results. Prod, Vol. 76, p. 79.
.
As it can be seen from the microstructures, although the Kinley, W. (1979), “Inertia welding: simple in principle and
sizes and forms of the pearlite and ferrite grains were application”, Weld and Met. Fab., Vol. 10, pp. 585-9.
changed, all the structures consisted of pearlite and ferrite. Lambrakos, S.G., Fonda, R.W., Milewski, J.O. and Mitchell, J.E.
For that reason, the strength of the joints is nearly the (2003), “Analysis of friction stir welds using thermocouple
same as that of the original material strength. measurements”, Science and Technology of Welding and Joining,
.
Subsequently, the maximum hardness values are obtained Vol. 8 No. 5, pp. 385-90.
from the weld interface because of the increasing Moreira, P.M.G.P., Frazão, O., Tavares, S.M.O.,
temperature and the rapid cooling at the joint-interface De Figueiredo, M.A.V., Restivo, M.T., Santos, J.L. and
as seen from the temperature measurement. Then, the De Castro, P.M.S.T. (2007), “Temperature field acquisition
deformation at the interface causes a strong decrease in during gas metal arc welding using thermocouples,
the grain size, which leads to a hardening in the region of thermography and fibre bragg grating sensors”,
the interface. Therefore, it can be observed that hardness Measurement Science and Technology, Vol. 18 No. 3,
at the weld interface is increased by deformation. pp. 877-83.
.
It is important to note that the measurement process was Nentwig, A.W.E. (1996), “Friction welding of cross section of
successfully accomplished in this study although it was different size”, Schweissen und Schneiden/Welding & Cutting,
particularly difficult to obtain a temperature measurement Vol. 48 No. 12, pp. 236-7.
due to the large deformations at the interface. Rich, T. and Roberts, R. (1971), “Thermal analysis for basic
friction welding”, Met. Const. and British Weld J., Vol. 3,
References pp. 93-8.
Sahin, M. (2001), “Investigation of effects of work-piece
Akata, H.E. and Sahin, M. (2003), “An investigation on the dimensions and plastic deformation on friction welding
effect of dimensional differences in friction welding of AISI method”, doctoral dissertation, Graduate School of Natural
1040 specimens”, Ind. Lub. & Tribo., Vol. 55 No. 5, Applied Sciences of Trakya University, Edirne.
pp. 223-32. Sahin, M. and Akata, H.E. (2001), “An experimental study
Balasubramanian, V., Li, Y., Stotler, T., Crompton, J., on the application of friction welding on parts with different
Soboyejo, A., Katsube, N. and Soboyejo, W. (1999), diameters and width”, The Third International Congress
“A new friction law for the modelling of continuous drive Mechanical Engineering Technologies, Sofia, 24-26 June,
friction welding: applications to 1045 steel welds”, Mat. and pp. 499-501, 01.
Man. Proc., Vol. 14 No. 6, pp. 845-60. Sahin, M. and Akata, H.E. (2003), “Joining with friction
Bendzsak, G.J., North, T.H. and Li, Z. (1997), “Numerical welding of plastically deformed steel”, J. Mat. Proc. Tech.,
model for steady-state flow in friction welding”, Acta Vol. 142, pp. 239-46.
Matellurgica, pp. 1735-45. Sluzalec, A. (1990), “Thermal effects in friction welding”,
D’Alvise, L., Massoni, E. and Walloe, S.J. (2002), “Finite I. J. Mech. Sci., Vol. 467, p. 478.
element modelling of the inertia friction welding Stahlschlüssel (1995), Verlag Stahlschlüssel Weg Str. GmbH.
process between dissimilar materials”, J. Mat. Proc. Tech., Triantafyllidis, D., Schmidt, M.J.J. and Li, L. (2003),
Vol. 125-126, pp. 387-91. “Comparison of high power diode laser and Nd:YAG
Draper, N.R. and Smith, H. (1981), Applied Regression laser microwelding of K-type thermocouples”, J. Mat. Proc.
Analysis, 2nd ed., Wiley, New York, NY. Tech., Vol. 138, pp. 102-8.
Ferro, P., Zambon, A. and Bonollo, F. (2005), “Investigation Vill, V.I. (1962), Friction Welding of Metals, AWS,
of electron-beam welding in wrought Inconel 706 – New York, NY.
experimental and numerical analysis”, Mat. Sci. Eng. A,
Vol. 392, pp. 94-105.
Fu, ByL and Duan, L. (1998), “The coupled deformation
Corresponding author
and heat flow analysis by finite element method during
friction welding”, Weld Res. Supp., Vol. 5, pp. 202-7. Mumin Sahin can be contacted at: mumins@trakya.edu.tr
316