Aluminium weld fatigue: From characterisation to design rules

Abstract In suspension design, traditionally subframes and links are welded structures, with one of the main design criteria being fatigue. Better understanding, correlation and correct design rules are vital to ensure safety and avoid over design that can lead to a reduction in competitiveness. A known mesh modelling T-joint SN method is used to design the welds and is mainly driven from standards used for thicker panels seen in the construction and ship building industry. To consider the thin sheet designs largely seen in automotive, Volvo & nCode have developed a new method based on structural stress and is known to be mesh insensitive. However, new difficulties emerged due to the clear separation between the membrane and bending behaviour needed for stress recovery. Specific tests need to be defined to ensure correct characterisation. Gestamp has generated a new standard of design rules to ensure correlation in Weld-Ends and in Mid-Weld cracks and correct toe position definition to ensure accurate stress recovery, and give more beneficial input to process by better defining the weld torch position. New test specimens have been developed and intensive testing carried out to fully characterise the bending and tension behaviour and reduce the gap in predictions between Lap Weld and Fillet Weld. This methodology has been successfully used in suspension designs with high confidence and correlation. These design rules have been used in conjunction with studying the effect of weld process and sequence on fatigue performance. Controlled distortion subframes, against high distortion subframes, have been tested to capture the benefit by reducing the fit-up gaps. Residual stresses are still a grey zone for Gestamp and further work is needed to better understand the effect.