Dissimilar friction stir welds in AA5083-AA6082. Part II: Process parameter effects on microstructure
- PDF / 566,022 Bytes
- 12 Pages / 576 x 720 pts Page_size
- 93 Downloads / 210 Views
AS a relatively new joining technique, few systematic studies of friction stir (FS) welding have been performed. The relationships between the various weld parameters and the resulting weld properties have not been identified, and there are many remaining questions. The aim of this study is to investigate the bounds of the so-called processing window—the range of welding speeds (rotation and traverse) within which good-quality welds will be produced—for the FS welding of AA5083 to AA6082. In part I,[2] the impact of changes in the rotation and traverse speed on the global process parameters (forces, torque, and power input) and the thermal excursion and macrostructure were determined. A thermal model was presented that was calibrated against thermocouple data in the weldments and in the supporting backing plate. This model indicated that the temperature under the tool is more strongly dependent on the rotation speed than the traverse speed. This conclusion was supported by the measured torque data and the extent of material flow around the tool. In this paper, the microstructure and local mechanical properties are examined in greater detail. In addition, the outputs of the thermal model are coupled to analytic models for predicting the hardness variation of the two alloys across the dissimilar welds to elucidate the relationship between the welding parameters and the subsequent weld properties. The variation in hardness in welded AA6xxx aluminum alloy series has been previously linked with changes in the precipitate distribution due to the imposed thermal cycle.[3,4,5] Far from the weld line, the lower temperatures cause the b0 precipitates to coarsen or transform to the nonhardening b9 phase, reducing the hardness. At higher temperatures the precipitates dissolve, with the reverted fraction increasing as one nears the weld line. During cooling, some of this solute may reprecipitate as stable, nonhardening phases such as b9. The remaining solute results in some strength recovery via natural aging in the days and weeks following welding. [1]
M.J. PEEL, Post-Doctoral Fellow, A. STEUWER, Post-Doctoral Fellow, and P.J. WITHERS, Professor, are with the Materials Science Centre, Manchester University, Manchester, U.K. M.J. PEEL and A. STEUWER are with FaME38 at the ESRF-ILL, Grenoble, France. Contact e-mail: [email protected] Manuscript submitted September 29, 2005. METALLURGICAL AND MATERIALS TRANSACTIONS A
The AA5xxx series of alloys may be used in the annealed or strain hardened condition. If the alloy has been annealed, then the microstructure is stable and no softening will occur in the heat-affected zone (HAZ). In contrast, a worked structure will readily recover or recrystallize during welding, so softening may occur.[6] The AA5083 material used in the current study is in a highly cold-worked condition, and the welding process has been shown to result in significant softening through recovery and recrystallization.[7] II.
EXPERIMENTAL PROCEDURE
Two materials form the basis of this study: AA5083 (cold-rolled) and AA 6083
Data Loading...
