Search

The relationship between joint strength and deformation heat input at the upset stage and the relationship between joint strength and upset bum-off length were examined on friction welded joints of 1050 pure aluminum to C1100 tough pitch copper. Joint quality was evaluated from the tensile strength and fatigue strength. It was found that both the deformation heat input at the upset stage and the upset burn-off length correlated well with joint strength. Additionally, when the deformation heat input at the upset stage or upset burn-off length exceeded a certain value, stable tensile strength was obtained. The mixing layer, intermetallic compounds of Al-Cu were interpolated into A1050, formed at the weld interface, and joints having a thick mixing layer fractured at the weld interface. Joint efficiencies of sound joints for the fatigue limit of the A1050 base metal were 88-100%. Judging from the fatigue limit, sound joints could be produced when either the deformation heat input at the upset stage or the upset burn-off length exceeded a certain value.

Similar and dissimilar joints of 2017 and 6061 aluminum alloys were friction welded. Fatigue strength and fatigue life distribution of these joints were examined. The S-N curve showed that the fatigue strength of A2017/A2017 was almost the same as that of A2017 base metal, while that of A6061/A6061 was lower than that of A6061 base metal. The fatigue strength of A2017/A6061 was lower than that of A6061/A6061. The fatigue limit at 108 cycles of A2017 base metal, A2017/A2017, A6061 base metal, A6061/A6061 and A2017/A6061 were 120MPa, 120MPa, 70MPa, 60MPa and 55MPa, respectively. The shape parameter m and the location parameter γ in Weibull distribution function of the joints were lower than those of the base metals. These results suggest that the scatter in fatigue life distribution is larger and the smallest fatigue life is shorter for the present joints compared to the base metals.

Aluminum alloys to S25C carbon steel were friction welded, and the friction weldability and intermetallic formations at the weld interface of joints were investigated. Joints of A1050, A5052, A6061 and A6063 to S25C showed good-weldability, and A5056/S25C joint was somewhat poor than former one. An intermetallic compound of Al13Fe4 was formed at the weld interface of A5056/S25C and A6061/S25C joints, and their intermetallic formations were tend to increase with friction pressure, friction time and rotation speed. The joint strength of these joints was high when thin instermetallic conpound layer was formed under high upset pressure. Somewhat poor-weldability of A5056/S25C joint compared with A6061/S25C joint seemed a cause of the intermetallic formation accelated by high content of magnesium in A5056. Joints of A2011, A2017, A2024 and A7075 to S25C showed poor-weldability. Brittle intermetallic compounds of CuAl2 and Al7Cu2Fe were formed at the weld interface of A2011/S25C joint containing a copper.

Friction welding of aluminum alloys to copper was carried out in order to examine the weldability with particular reference to the effect of the intermetallic formation at the weld interface. The maximum joint efficiencies of A1050, 5000 and 6000 series alloys to copper were more than 80 %, while those of 2000 series and A7075 alloys joints were less than 50 % and scattered widely. For joints of A2024 and 5000 series alloys which contain rather high magnesium content, intermetallic compounds of Al-Cu-Mg system were formed at the weld interface in addition to that of Al-Cu system. In the other joints, only intermetallic compounds of Al-Cu system were formed. A mixing layer, in which pieces of copper and intermetallic compounds were incorporated into aluminum alloy, was formed near the weld interface. The mixing was very intense, in the joints with low efficiency, namely joints of A2011, A2024 and A7075 to copper, and these joints were fractured in this mixing layer and at the weld interface. On the other hand, the joints with high efficiency, namely joints of A5052 and A6061 to copper, had thin and even mixing layer. The mixing layers observed in the joints of A1050, A2017, A5056 and A6063 to copper were thinner than those of the other joints.

In present study, various aluminum alloys pipe to steels pipe were friction welded, and the mean value and scatter of welded joints were investigated. That is, thirty specimens for each combination of three aluminum alloys pipe to five steels pipe were friction welded under the optimum friction welding condition and the joint efficiency of welded joints were evaluated by tensile test. After performing statistical analysis and Weibull analysis on the joint efficiency, the following results were obtained, (1) The joint efficiency of A6061/steel showed a normal type distribution, but the A5052/steel and A6063/steel showed a distortional type and a double peak type distribution, respectively. (2) A difference was found in the joint efficiency of A5052/carbon steel and A5052/stainless steel showing high joint efficiency in the A5052/S45C and A5052/SUS304. (3) In the A606l/carbon steel and A6061/stainless steel, no difference was found in joint efficiency. (4) In the case of A6063/carbon steel, any difference of the joint efficiency was not found, while a difference was found in the case of A6063/stainless steel. (5) A reference table on the joint efficiency and joint strength for three aluminum alloys pipe to five steels pipe was presented.

Friction welding of various aluminum alloy pipes to SUS304 stainless steel pipe was carried out making clear weldability at dissimilar material pipe. A1050/SUS304, A5052/SUS304, A5056/SUS304, A6061/SUS304 and A6063/SUS304 joints showed good weldability and the maximum joint efficiencies of these joints in tensile test were about 91%, 71%, 60%, 79% and 97%, respectively. In these cases, the formation of Al13Fe4 intermetallic compound was induced by the interdiffusion through the weld interface. On the other hand, other joints had poor weldability. In the cases of A2011/SUS304, A2017/SUS304 and A2024/SUS304 joints, CuAl2 and Al7Cu2Fe were observed on the friction surface, and the maximum joint efficiencies of these joints were about 26%, 39% and 32%, respectively. For the case of A7075/SUS304 joint, Al23CuFe4 was observed on the friction surface, and the maximum joint efficiency was about 37%.

Comparatively thick pipes of 6061 aluminum alloy and S25C carbon steel were friction welded under a preferable welding condition and the strength distribution on the weld interface was investigated. The results obtained are as follows; (1) According to a decrease of the thickness ratio 0.30, 0.16, 0.15, 0.14, 0.13 and 0.12, the joint efficiency decreased 73.9%, 68.1%, 63.8%, 55.8%, 39.9% and 21.6%, respectively. (2) Friction welding of thin pipe having less thickness ratio t/D0 than 0.11 was difficult to produce a sound weld (D0: outer diameter, t: thickness of pipe). (3) The tensile strength of each 11.1% area from inner and outer edges of the weld interface was low, and the joint efficiencies were 30∼50%. (4) The joint efficiency of the joint removed inner and outer areas of the weld interface was about 95%.

This paper discussed the posibility of selection of optimum welding condition by adopting the estimating equation for welding consisting of some welding behaviours to responce surface method using a 12 mm diameter SUS304 stainless steel. First, by analysing the relation between the welding behaviour and the tensile strength of welded joint using multiple regression analysis, an estimating equation for welding capable to estimate non-destructively was obtained. Next, by setting up the joining evaluation value obtained by the estimating equation for welding as a dependent variable and the welding condition as an independent variable, the optimum welding condition was selected using the responce surface method, namely, the posibility of selection of optimum welding condition without tensile strength was confirmed.The optimum welding condition obtained was the friction pressure P1=30 MPa, the upset pressure P2=170 MPa, the friction time t1=3.5 s and the rotating speed N=2425 rpm, and the joint efficiency was 102% with a satisfaction.

In order to measure a heat input (quantity of heat) generated during friction welding, a new type calorimeter of acryl resin was developed. The heat input during friction welding of a stainless steel was measured using the new calorimeter. Furthermore, the transient temperature distribution was estimated by finite-element analysis. The heat input measured during friction welding was in good agreement with the value calculated from friction torque, burn-off length, rotation speed and pressure. The friction heat input was much greater than the deformation heat input through the friction and upset stages in the brake-type friction welding.