Prediction of retained residual stresses in laboratory fracture mechanics specimens extracted from welded components

The measurement of weld material fracture toughness properties is important for the structural integrity assessment of engineering components. However, welds can contain high levels of residual stress and these can be retained in fracture mechanics specimens, particularly when machined from non-stress relieved welds. Retained residual stresses can make the measurement of valid fracture toughness properties difficult. This paper describes the results of analytical work undertaken to investigate factors that can influence the magnitude and distribution of residual stresses retained in fracture mechanics specimen blanks extracted from as-welded ferritic and austenitic stainless steel plates. The results indicate that significant levels of residual stress can be retained in specimen blanks prior to notching, and that the magnitude and distribution of stress is dependent upon material properties, specimen geometry and size, and extraction location through the thickness of the weld. Finite element modelling is shown to provide a useful approach for estimating the level and distributions of retained residual stresses. A new stress partitioning approach has been developed to estimate retained stress levels and results compare favourably with FE analysis and available experimental data. The approach can help guide the selection of specimen geometry and machining strategies to minimise the level of residual stresses retained in fracture mechanics specimen blanks extracted from non stress-relieved welds and thus improve the measurement of weld fracture toughness properties.