Multi-spot welded joints are routinely used in the automotive and other industries to produce structurally critical components. Unfortunately the size and complexity of the joints often limits the practicality of testing their mechanical properties, particularly when long term testing is required, such as in the case of determining fatigue properties.

Conversely, single spot welded joints are easily produced and can be tested on existing equipment with little or no modification. Clearly it would be extremely advantageous to be able to predict the fatigue properties of a multi-spot welded component based on the data obtained from testing single spot welded joints.

Currently there are several models that propose methods for predicting the fatigue life of a given joint of simple geometry, the majority of which are based on a fracture mechanics approach. However, as more spot welds are added and the complexity of the joint increases, these existing models tend to lose their accuracy.

This project intends to determine whether it is possible to use a statistical approach to fatigue strength characteristics to divide a complex multi-spot welded joint into its individual welds and determine the overall fatigue properties of the joint as a whole by understanding the fatigue properties of the individual welds. If this process proves possible then the accuracy of the existing models could still be maintained even for complex joints, provided the joint is first divided into its constituent welds.