Damage-mechanics insights into the relationship between upper-shelf Charpy testing and J-integral testing considering varying tensile test properties

Abstract

Requirements on the yield-to-tensile strength ratio σy/σu, fracture elongation A and the Charpy energy Cv are used together as part of an indirect method of ensuring sufficient ductility at localised areas of stress and strain concentration in the design of steel structures. Recent studies have found that these indirect requirements could be inadequate in certain situations involving cracks or manufacturing defects. Furthermore, requirements on the σy/σu which are enforced regardless of the structural context and other material properties may unnecessarily constrain the use of steels which nonetheless have high strength, fracture toughness and ductility. In contrast to the σy/σu, A, and Cv, a more direct measurement of a structure’s ability to resist fracture is given by fracture toughness testing, such as J-integral testing, but this is less frequently used, because these tests are significantly costlier than tension and Charpy tests. More often, Charpy tests are performed and correlations between upper-shelf Cv and J values are used to estimate the fracture toughness of the material. However, the existing correlations are predominantly based on empirical findings and have not systematically accounted for the effect of variations in the σy/σu, which has been shown in recent studies to affect the fracture toughness. Using a previously validated coupled damage-mechanics model with rate- and temperature-dependent plasticity and damage softening, this paper investigates the correlation between Cv and JQ (the critical J) numerically, including how it is affected by other material certificate properties such as the σy/σu and A. First, a correlation based on regression between the damage parameters and the mechanical properties from mill test certificates is found by calibrating the damage parameters for a large database of these steels. Then, the correlation between Cv and JQ is assessed by simulating the single-edge-notch bending test for a range of varying mill test certificate properties, taking into account how the damage parameters vary with these mechanical properties. The results are analysed to give better insight into how the notch toughness correlates to the fracture toughness, taking the σy/σu and A into account. It is seen that although varying σy/σu and A has some effect on how the total notch energy Cv is correlated to JQ, it does not reflect a significant effect on the ductile fracture initiation toughness but is rather associated with the fact that the Cv includes a significant portion of energy for stable ductile propagation and fracture occurring at the specimen’s free surface, while JQ primarily concerns the onset stage of stable ductile tunnelling behaviour at the centre of the specimen. The σy/σu and A are seen to have an even smaller effect on the correlation between JQ and the energy (Cvm) dissipated up to the occurrence of the peak force in the instrumented Charpy test, in comparison with the Cv–to–JQ correlation, especially for low Cvm.