This work has investigated the potential for information contained in the residual transverse strain and microhardness profiles to predict the fatigue performance ranking for a series of 12 MIG welds made in RQT701 steel under different conditions. The experimental matrix included two plate thicknesses, two heat inputs and three types of filler metal. Predictive modelling considered the use of a number of parameters that might characterise the effect of residual strain and microhardness profiles on fatigue performance, including maximum range experienced at the toe, gradients associated with the weld toe, and changes in these parameters across the weld toe. It proved possible to predict the fatigue ranking at a life of 105 cycles for 10/12 of the weld conditions tested, based on weld toe hot spot strain and a combination of residual strain range at 1 mm depth and the hardness range at mid-depth.
Relationships were observed between the measured parameters and fatigue performance at lives of 2x106 cycles but additional discriminating parameters are required to predict fatigue ranking.