Experimental issues in tensile Hopkinson bar testing and a model of dynamic hardening

This paper is firstly aimed at the quantitative evaluation of the error levels achievable in the dynamic characterization of ductile metals via Hopkinson bar tensile tests, ran according to the classical strain-gauge-based experimental procedure and to the enhanced high-speed-camera-assisted procedure. Also the effect of the specimen slenderness is investigated for checking the sensitivity of both the above techniques to different specimen geometries. Secondly, a material model is proposed here for the dynamic hardening, based on the experimental true stress and on a dynamic postnecking correction derived from the MLR function, originally developed for the quasistatic necking. Finally, the necking onset is demonstrated to freeze the sensitivity of the stress-strain curve to the strain rate; this feature, supported here by both experiments and finite elements analyses, is a novelty according to the authors knowledge, and might have a considerable impact on the whole dynamic characterization of ductile metals undergoing large postnecking strains before failure