How sensitivity of metals to strain, strain rate and temperature affects necking onset and hardening in dynamic tests

The material characterization at high strain rates still has room for improvements, especially regarding the possible interactions between strain, strain rate and temperature effects. Some known approximations are still common practice, such as the use of simplified stress–strain calculations at post-necking strains, based on the specimen length, instead of the formulation based on the specimen cross-section. In this work, the behaviour of A2-70 stainless steel at different strain rates and temperature is analyzed, with particular attention to how strain rate and temperature sensitivities affect necking onset and hardening. At the strain rates of interest for this work, the inertia effect is shown to play a secondary role in anticipating or delaying the necking onset. Therefore, the instability onset is firstly analytically derived by just explicitly accounting for the coupling between strain, strain rate and temperature variables within a general multiplicative hardening function, without the need of recalling bifurcation or perturbation analyses. The strains at necking onset, obtained from experiments at different temperatures and strain rates (by hydraulic machine and by tensile Hopkinson bar – SHTB), allow to check the validity of the proposed relationships expressing the combined effect of the relevant variables on the instability condition. Video recordings and image analysis measurements of the deforming round specimens allow comparing the diameter-based effective true stresses, true strains and true strain rates to their nominal length-based counterparts, enabling to quantify the approximation intrinsic in the simplified calculations. The fracture strains and the shapes of necked specimens at incipient failure finally highlighted that the elongation-based approach spoils the accuracy of stress–strain measurements in remarkably different ways for same material, depending on the temperature – strain rate history at hand, thus preventing the use of the latter approach also for merely comparative purposes.