Fatigue strength of welded aluminum structures – A re-evaluation of fatigue data of welded aluminum joints with nominal stress approach

• The 2nd gen. EC9 standard is well applicable for friction stir welded aluminum joints. • Both the 2nd gen. EC9 and IIW recommendations were unconservative for butt-welds. • Friction stir welded 5xxx and 6xxx alloys exhibited differing sensitivities to mean stress, impacting their fatigue performance. • Walker’s mean stress correction method proved to be the most effective for arc welded aluminum joints. This work analyzes the applicability of the nominal stress approach to welded aluminum joints, with respect to the second generation of Eurocode 9: Design of Aluminium Structures and recommendations of International Institute of Welding, utilizing experimental fatigue test data extracted from existing relevant literature. The influence of the aluminum alloy series and the mean stress effects on the scatter in the data were also studied. The extracted dataset included data from 5xxx and 6xxx aluminum alloys welded with friction stir (FS) welding, gas metal arc (MIG) welding and tungsten inert gas (TIG) processes. The fatigue data collection was limited to tests performed under constant amplitude loading, and the extracted datasets included series with applied stress ratio varying from −3 to 0.5. The mean stress was found to have a considerable effect on the scatter of fatigue data in small-scale specimens. Consequently, various mean stress correction approaches were tested, of which the Smith-Watson-Topper (SWT) approach was found to be the most suitable. However, the SWT approach had no discernible beneficial effect in 6xxx series aluminum FS specimens.