Comportement en fatigue de joints d'aluminium 7075-t6 soudes par friction-malaxage et paracheves

In consideration of the surface characteristics of the joint after friction stir welding (FSW), post-weld treatments that mitigate the severity of the geometrical defects (i.e. striations, burrs) and/or the weld distortions would be key enablers for widespread application of FSW.;The present study was targeted to determine the mechanical response (tensile and fatigue) of friction stir butt welded AA7075-T6 in the as-welded condition, as well as after different post-weld treatments, including brushing, deep rolling, machining, and solution heat treatment and precipitation hardening. Each post-weld treatment was deliberated to address the concerns in regard to FSW of AA7075-T6. For instance, brushing was considered for minimizing the surface defects (striations and burrs). Deep rolling was intended for reducing distortion in the welds. Machining of the weld upper surface would allow the understanding of the mechanical response without the surface defects. Finally, post-weld heat treatment was aimed at recovering the strength loss from precipitate dissolution during FSW.;Inevitably to investigate the effect of the post-weld treatments on the fatigue performance of the friction stir welded AA7075-T6, process conditions allowing integral welds, i.e. without defects such as wormholes or lack of penetration, were identified. Specifically, using an optimal vertical force of 11 kN rendered integral welding of 2.3 mm thick AA7075-T6. In this initial phase, it was remarked that the burrs on either side of the joint can be reduced in size by decreasing the vertical force, but, in due course, this is at the expense of the 180° bend performance of the weld.;For the as-welded condition, residual stress measurements showed tensile stresses in the direction parallel to the welded joint. The maximum measured stress was +193 MPa in the middle of the joint, which corresponds to 37.8% of the yield strength of AA7075-T6. The residual stress in the direction perpendicular to the welded joint varied between -38 MPa and +20 MPa in the weld zone. However, the residual stress measured on the welded coupons was between -49 and -87 MPa, which may lead to an overestimation of the fatigue strength.;Evaluation of the tensile mechanical properties revealed reductions of 25% and 12% for the yield and ultimate tensile strengths of the welds as compared to that of the base material. Post-weld treatment with brushing and/or deep rolling appeared to have no effect on the tensile strength of the welded joints. Machining was observed to slightly decrease (by 2%) the tensile strength of the weld, which may be a remnant effect of reducing the gage thickness. A combination of brushing and post-weld heat treatment was most effective; the joint efficiency was nearly 95%.;The fatigue strength in the as-welded condition was lower than that of the base metal. Specifically, the stress amplitude corresponding to 106 cycles was 76.5% of that of the base metal. The fatigue cracks initiated at the weld striations on the upper surface of the weld. Brushing improved the fatigue strength to about 90% of that of the base metal. The fatigue cracks initiated in the low hardness location of the heat affected zone (HAZ) on the lower (root) surface of the weld. This suggests that the localized softening effect becomes predominant over the mechanical defects (i.e. striations) for this post-weld treatment.;Deep rolling reduced weld distortion by 50%. Despite no influence of deep rolling on the tensile strength, a marked reduction (40%) in the fatigue strength of the weld was determined as compared to that of the base metal. This reduction was attributed to the stress concentration resulting from the burrs crushing into the underlying regions, i.e. just below the weld edges. Combinations of brushing and deep rolling showed similar fatigue strengths and fracture modes, as the former post-weld treatment did not sufficiently diminish the size of the burrs on either side of the weld. Hence, the severity of the stress concentration after deep rolling remained similar with or without brushing.;Machining of the welds to remove the striations on the upper surface gave similar fatigue results to that obtained by brushing. Specifically, the fatigue strength of the machined welds was 90% of that of the base metal and failure initiated on the lower surface of the weld in the softened region of the HAZ.;Heat treatment after brushing improved the fatigue strength to 96.2% of that of the base metal. The fatigue crack initiated in the middle of the weld or at the lower (root) surface in the HAZ. This combination addressed two shortcomings of the FSW process: brushing eliminated the effect of the geometrical defects on the upper surface and the heat treatment mitigated the softening in the HAZ, although not entirely to the initial T6 condition. (Abstract shortened by UMI.).