Experimental investigation of the influence of the specimen geometry on Hopkinson tensile test results

Split Hopkinson bar set-ups are considered to be valuable, if not indispensable, to study the strain rate dependent mechanical behaviour of materials. In recent years results of SHB experiments on a wide range of materials have been reported in literature. Given their importance for crash relevant applications, special attention has been paid to the tensile properties of steel sheets used in the automotive industry. For tensile experiments most often dogbone-shaped geometries are used, however with widely divergent dimensions. In this contribution results of an extensive experimental study on the influence of the specimen geometry are presented. Based on data found in literature, seven geometries are defined and subjected to strain rates ranging from 650/s to 1250/s. The material considered is a TRIP (TRansformation Induced Plasticity) steel sheet. An advanced measurement technique is used to monitor the true distribution of the deformation along the length of the specimen. It is shown that the influence of the specimen geometry on the test results cannot be neglected, and that this can mainly be attributed to the fact that the true distribution of the deformation deviates from the assumed distribution.