Numerical study of the influence of the specimen geometry on split Hopkinson tensile test results

摘要

In recent years numerous studies on the high strain rate tensile properties of sheet materials using Split Hopkinson Tensile Bar (SHTB) experiments have been reported in literature. For SHTB experiments no consensus exists on the specimen geometry to be used and its influence on the observed behaviour. However, previous studies have revealed that changes in the specimen geometry give rise to distinct differences in established mechanical behaviour. In this contribution results are presented of finite element simulations of SHTB experiments using different specimen geometries. These simulations not only confirm previously obtained experimental results, but also give complimentary and detailed information on the true distribution of the stress and the strain in the specimen, including the non-axial stresses. Attention is paid to the basic assumptions of Hopkinson experiments: the uniaxiality of the stress state and the homogeneity of the strain. It is shown that the validity of these assumption is highly geometry dependent. The influence of the deviation from these assumptions on the material behaviour extracted from a Hopkinson experiment will be discussed.