An analysis of stress uniformity for concrete-like specimens during SHPB tests

As far as concrete-like brittle materials are concerned, whether stresses have been uniformly distributed along the specimen thickness before fracture occurs is a key problem with regard to the validity of SHPB data. By using characteristics method of wave propagation, the SHPB tests for cement mortar which is respectively regarded as the so-called viscoelastic material and the elastic material are numerically studied in the present paper, in order to see how the stress distribution is during both loading and unloading processes. It is found that different rise-times τs of incident waves evidently influence the stress uniformity, and the optimum rise-time looks like τs/tL = 2, either shorter or longer rise-time will result in worse situation in stress uniformity; the stress amplitude has no influence on the stress uniformity, but markedly influences the value of strain and average strain rate in specimens. The error found in the determination of dynamic stress–strain curves of rate-dependent brittle materials by the traditional SHPB technique, either by three-wave method or two-wave method, is actually unacceptable, although the dynamic fracture stress can be correctly determined if the requirement of stress uniformity is satisfied before the specimen fails.