Options
2013
Conference Paper
Title
Optimization of FSW clamping systems by utilizing numerical and experimental approaches
Abstract
In the present work the influence of different clamping scenarios for optimizing FSW processing was investigated by applying a sequential thermo-mechanical FE-Model. Since friction stir welding is a solid-state welding process, residua! stresses are expected to be lower than those in fusion welding. However, as these stresses are still present they will lead to distortions in the final assembly of the welded structure. For planar structures, these final distortions are clearly visible in terms of out-of-plane distortion. Indeed, even small changes of the geometrical appearance from one clamping scenario to another can be detected. For that reason all numerical calculations and experimental approaches have been examined on small planar sheets to study the effect of varying clamping conditions during friction stir welding. Although the influence of varying clamping distances was found out to be small, it has been made visible by means of the numerical approach. The comparisons between numerical and experimental results show, that the results of the FE-Model are in good agreement with measured distortions of performed friction stir welding experiments. Therefore the model marks a useful start point for more complex analysis of clamping influences. Already the tendencies identified by the study of small coupons help to assess clamping strategies. However, further model enhancements are necessary in regard to more realistic slip stick effects or buckling. It was shown that the relative transversal position of the clamps to the weld line has a small but noticeable influence on the global distortion behavior of the welded sheets. Increasing transversal distances between continuous clamps results in higher out-of-plane distortion after welding. An introduction of discontinuous clamps will generally result in higher distortions of the welded plates. This effect was significantly stronger than changes to the transversal distance only. Interestingly, for wide gaps the distortion can be partly compensated by increasing the transversal distance between the clamps. If the longitudinal distances are narrow between the clamps, this effect is not seen. Although this paper deals with small coupons only, the effect of varying transversal distances on the final distortion is expected to rise with the dimension of the parts to be welded. If the results of the investigated coupons are generalized it can be concluded that especially for 3D-structures made of sheets, where discontinuous clamps are necessary, clamps should be positioned as close as possible to each other in longitudinal direction. Furthermore, although a close distance of clamps to the weld line is favorably, this value is less important in comparison to the previously mentioned longitudinal clamp distance.
Author(s)