CC BY
54
y^K 539.2:539.374.
P.A. Petrov, J. Bast, M.A. Petrov
Moscow State Technical University - MAMI Technical University Mining Academy of Freiberg
EXPERIMENTAL AND NUMERICAL INVESTIGATION OF FRICTION DURING THE HOT ISOTHERMAL DEFORMATION OF ALUMINIUM ALLOY A95456
An investigation of friction during the hot isothermal deformation of ring specimens was carried out. The specimens were made from aluminum alloy A95456. Lubricants based on mineral and synthetic oils were used in the tests. Typical ring specimens were deformed at temperatures within the range of 200-450 °C. Numerical analysis of the experimental data allowed us to obtain the values of the friction factor, as well as the friction factor - temperature relationships, for both lubricants. The values of the friction factor within the investigated temperature range, as well as the calibration curves for their determination, were calculated by means of the QFORM simulation. Some practical recommendations were given.
Introduction. Hot isothermal forging is a specific type of bulk close-die forging technology. Forging under isothermal conditions implies the heating of both dies and work pieces to the same temperature.
Generally, isothermal forging requires special tooling materials and lubricants that can perform adequately at the forging temperature. In the case of isothermal forging of aluminium alloys, conventional die materials undergo no significant loss of strength. So, the major problem for any isothermal forging technology for Al alloys is the appropriate choice of lubricant.
A lubricant should guarantee the following characteristics: environmental safety, good heat-shielding properties and adhesion, and low tribological properties. The oil-based lubricants discussed in references [1, 2] meet these requirements.
To evaluate the friction conditions during metal-forming processes, numerous experimental techniques, including ring compression, forward rod-backward cup and double-cup extrusion tests, and the conic dies test, have been applied. The aim of these methods is to determine the proportionality coefficient in a friction model, namely Coulomb’s, Siebel’s, or the general friction model [3-5].
The ring-compression test is the most simple and widely used method for the quantitative estimation of the interfacial friction during bulk metal formation [6-8]. According to the ring-compression technique, a flat ring-shaped specimen is compressed by a certain amount. Then, the friction coefficient or factor can be calculated with the help of calibration curves or by means of the finite-element (FE) technique.
The calibration curves represent the relationship between the internal diameter of a ring specimen and its height or reduction in height after compression. Each of the curves corresponds to the definite value of the friction factor or coefficient.
It is known that the calibration curves can be defined by means of one of the following techniques: approximate stress analysis [7, 8], upper-bound analysis [9, 10], or FE modeling (FEM) [11].
In the present paper, the ring test and a numerical simulation were applied to estimate the tribological properties of lubricants [1, 2] that can be used for hot isothermal forging of alloy A95456. In particular, we investigated the effect of the material’s temperature, as well as that of the lubricant composition, on friction. Furthermore, calibration curves based on FEM results were obtained and compared with the experimental data.
Experimental procedure. The experimental investigation was carried out on a hydraulic press (nominal load = 2,5 MN). The ring samples were cut from a bar of aluminium alloy A95456 (Table 1).
Table 1
Chemical composition
Element Al Cu Mg Mn Fe Si Zn Ti
% Base 0,04 б,80 0,б3 0,22 0,1б 0,20 0,10
The sizes of the samples were as follows: inner diameter = 20 mm; outer diameter = 40 mm; height = 14 mm. The ring samples were heated to temperatures of 200, 300, 350, 390, 430 and 450 °C in the electric furnace. Deformation of the heated samples was carried out on flat dies that were warmed up with induction installation. Samples were compressed with lubrication. The lubricants used were as follows: 1) industrial oil in a combination with colloidal graphite (IO + G); 2) synthetic oil with graphite (SO + G). Samples were compressed to about 50 % of their initial height (Table 2). Die velocity was constant at V« 2 mm/s (hydraulic press = 2,5 MN), which corresponded to an initial strain rate of 0,14 s-1. This strain rate value belongs to the strain rate interval (10-4 - 10-1 s_1) within that the isothermal forging is usually carried out.
The values of height hexp and inner diameter cFxp were determined after compression of the ring samples. The inner diameter was measured in three locations along the height of the rings. Finally, the value of the inner diameter was determined as dexp = (dtop + dmid + dbot)/3, where dtop, dmid and dbot = inner diameter at the top, middle and bottom along the height of the ring, accordingly.
1. Mass loss of lubricant sample vs. temperature
The heat-shielding properties or thermal stability influence both the efficiency of the lubrication and the formation of the insulating lubricant layer at the deformed material interface, within the temperature range of 300-700 °C. The thermal-stability index relates to the mass loss of the lubricant sample heated to the investigated temperature. Figure 1 shows the curve of mass loss vs. temperature for both types of lubricant. It can be seen that the lubricant SO + G had better heat-shielding properties than IO + G.
Numerical simulation. A numerical simulation of the ring test was carried out by means of the FE system QFORM-2D (QuantorForm Ltd., Russia). The aims of the simulation were to determine the values of the friction factor based on the obtained experimental data for the lubricants under study, and to construct calibration curves. The details of the QFORM algorithm can be found in reference [12]. Levanov’s friction model [13] was implemented by the QFORM simulation software.
Ring compression is an axisymmetric process. To take this into account, the simulation in QFORM 2D was carried out for a quarter of the cross section of a specimen.
To determine the true value of the friction factor, the several trials of finite-element (FE) simulation of ring deformation were carried out. To identify the true value of the friction factor kn, the following criterion was used
S = dexp - dfem < 0,05, (1)
where dexp and dem = the inner (or outer) diameter of the ring sample obtained experimentally and by FEM, respectively.
The variable parameter in the simulation was the friction factor. The simulation was carried out for different forging temperature within the range of 200-450 °C. Stress-strain curves of A95456 alloy were taken from [14, 15, 16]. These curves were obtained under the following conditions: strain range = 0,1-0,9; strain-rate range = 0,01-0,4 s_1; temperature interval = 200-450 °C.
To perform the simulation, we assumed that the contact friction was constant at the defined temperature of deformation within the investigated range. In addition, we assumed that the deformation condition was isothermal, as observed in the experiments. This meant that the tools and the sample were at an equal temperature during the initial stage of deformation. Owing to the heat effect of plastic deformation, the sample temperature had increased by the end of deformation, whilst the temperature of the tools remained the same.
Results. The obtained results allowed us to determine the true value of the friction factor for both lubricants. According to equation (1), two parameters should be compared: the inner and outer diameters of the deformed ring sample. Comparing the volumes of a ring before and after the FE simulation showed that the loss in volume was approximately equal to 0,5 %. As a result, it is useless to control the outer diameter before and after each simulation trial. It was sufficient to compare the actual inner diameters of the experimental ring sample and its FE model after compression. The determined values of the friction factor kn are given in Table 2. Figure 2 shows the relationships between the friction factors and the temperatures for both of the lubricants.
Table 2
Friction factor kn values
Temperature, °C 200 300 350 390 430 450
Lubricant type IO+G
Friction factor 0,171 0,221 0,153 0,143 0,155 0,118
SO+G 0,245 0,259 0,206 0,180 0,150 0,120
An analysis of the curves in Figure 2 demonstrated the sensitivity of the friction factor kn to the temperature of deformation, as well as the lubricant composition. This underlined the well-known fact that friction is non-constant and depends on several parameters, such as the die velocity, the temperatures of the deformed material and the die, the thickness of the lubrication layer etc.
The temperature dependence of the friction factor can be described using the following equation:
kn = At + A1T0 + A2T0 , (2)
where A0, A1 and A2 - coefficients, and T0 - the temperature of the deformed material.
Temperature, °C
Fig. 2. Curve friction factor vs. temperature
The values of the coefficients in equation (2) for the investigated alloys are calculated with the help of least-squares method and given in Table 3.
Table 3
Coefficients of temperature dependence for factor kn
Lubricant type A0, °C Ai, 1/°C A2, 1/(°C)2
IO+G 0,155 0,000 88 -2,05-10"6
SO+G 0,160 0,000 37 -1,01-10"6
It can be seen that both lubricants guaranteed low interfacial friction under isothermal conditions within the investigated temperature range. By contrast, the lubricant based on industrial oil (IO + G) provided smaller values of the friction factor than the lubricant SO + G, while the heat-shielding properties of the former were worse, despite the fact that it could be used for hot bulk isothermal forging of the Al alloy A95456.
The numerical simulation also allowed us to construct calibration curves which are the keypoint of the determination of friction factor on the basis of ring-compression test. Figure 3 illustrates some charts of calibration curves. These charts were constructed for alloy A95456 and are valid if only this alloy is isothermally formed at the temperature of 430 or 450 °C.
Figure 3 also represents the experimental data which were obtained with the lubricant SO + G (circle points) and IO + G (square points) at a temperature of To = 430 °C or To = 450 °C.
k =1.0 / /<=0.8 k =0.7
FEM
k=0. 6
k =0.5
k =0.4
k =0.3
k =0.2
-—"
12 10 8
k =0.1
k=0.0
Height of ring sample, mm
a
Height of ring sample, mm
b
Fig. 3. Calibration curves for alloy A95456: a - T0 = 430 °C; b - To = 450 °C
During the earlier stage of ring-sample compression, the friction factor was constant and equal to about 0,1 (see Figure 3, b). The major change in the friction factor value occurred after a reduction in height of 28,57 %. At this height reduction, the friction factor tended to increase up to 0,12, which corresponded to the height of the ring sample at the end of its deformation (see Figure 3, b).
From our point of view, this observed discrepancy could be linked to the effect of sticking of some areas of the contact surface of the ring sample with the corresponding areas of the dies. This effect was observed at the end of the compression to a nominal height reduction of 50 %. Furthermore, in order for industrialist to choose an appropriate lubricant for either alloy A95456 hot isothermal forging at temperature of 430 or 450 °C, the obtained curves can be used for estimation of tribological properties of investigated lubricant.
Conclusions. In summary, the following conclusions can be drawn from the results presented here:
1) the obtained temperature equations for the friction factor can be used for FE simulation of the bulk isothermal forming processes of alloy A95456;
2) the calculated calibration curves for T0 = 430 °C and 450 °C can be applied to estimate the friction factor of a lubricant for use in the isothermal deformation of aluminium alloy A95456, at the temperature mentioned above.
The obtained calibration curves are valid only if the friction factor is constant during deformation; otherwise, the correct determination of the friction factor value is possible only during the earlier stage of the material deformation, up to a reduction in height of about 28 %.
List of references
1. Petrov A.N., Andrejchenko T.P., Sajranova T.A. Russian Patent № 2232797: Lubricant for metal forging, 2004.
2. Petrov P.A., Petrov M.A. Experimental and numerical investigation of friction in hot isothermal deformation of aluminium alloy AA3003 // Proc. of the 8th ESAFORM Conference on Material Forming; ed. D. Banabic, Romania, Cluj-Napoca, (2005) 511-514.
3. Grudev A.P., Zilberg U.V., Tilik V.T. Friction and lubrications in metal forming. Handbook. - M.: Metallurgia, 1982.
4. Siebel E. Resistance and deformation and the flow material during rolling. Stahl and Eisen. - 50 (1930). - P. 1769-1775.
5. Wanheim T. Friction at high normal pressures // Wear 25. - 1973. -P. 225-244.
6. Kunogi M. On Plastic Deformation of Hollow Cylinders Under Axial Compressive Loading // Rep. Sci. Res. Inst. (Tokyo) 2. - 1954. - P. 63-92.
7. Male A.T., Cocroft M.G. Method for the Determination of the Coefficient of Friction of Metals Under Conditions of Bulk Plastic Deformation // J. Inst.it. Metals. 93 - 1964-65. - P. 38-46.
8. Burgdorf M. Investigation of Friction Values for Metal Forming Processes by the Ring Compression Method // Industrie-Anzierger. 89. - 1967. - P. 799.
9. Van Tyne C.J., Avitzur B. Ring forming: an upper bound approach. Part 2: process analysis and characteristics // Trans. ASME J. Eng. Ind. 3. - 1982. -P. 238-247.
10. Kalpin U.G., Eliseev G.V. Determination of Friction Coefficient in Hot Isothermal Upsetting // Izv. vuzov. - M.: Mashinostroenie, 5. - 1976. -P.157-160.
11. Kobayashi S., Soo-Ik Oh, Altan T. Metal Forming and The Finite-Element Method. - Oxford University Press, New York-Oxford, 1989.
12. Biba N., Lishnij A., Stebounov S. The development of net shape forging technology by means of numerical modelling. Proc. of 22 Forging Industry Technical Conference, Lake Geneva, Wisconsin: FIERF, Sept. 20-22, 1999, USA.
13. Levanov A.N., Kolmogorov V.L., Burkin S.P., Kartak B.R., Ashpur U.V. and U.I. Spasskiy // Contact Friction in Metal Forging. - M.: Metallurgia, 1976.
14. Mikliev P.G. Mechanical properties of Light Alloys in Temperatures and Strain Rates of its Forging. - M.: Metallurgia, 1976.
15. Petrov P.A., Perfilov V.I. Research on hardening of A92618 aluminium alloy at the initial stage of flow stress-strain curve during hot deformation // Proceedings of scientific conference Modelling, software and science intensive technologies in metallurgy (Novokuzneck, 14-17 March 2006). - P. 205-212.
16. Petrov P.A., Perfilov V.I. Research on Softening Effect of Alloy A95456 // Izv. TuLGU. Mechanics of deformable solid body and metal forming. TulGU, Tula, 2 (2006).
Получено 13.09.2010
