High strain rate forming (HSRF) is promising to break through the conventional forming limit of materials and thus to form hard-to-deform components. During HSRF process, long-lasting micro-defects before fracture, i.e. adiabatic shear bands (ASBs) in compression and voids in tension, are significant characteristics existing in a large strain range. These defects bring about the non-destructive softening of stress and continuous exertion of ductility, in return, the stress responses affect the evolution of micro-defects. However, the interaction between stress responses and defects evolution has not yet been reflected in the existing models, resulting in a limited prediction accuracy. Aiming at this issue, the competition between hardening caused by thermal activated dislocation movements and the flow softening brought by micro-defects evolution was well modeled in this work. During the modeling, the relation between the normal strain and effect zone of ASBs was established via the modification of Bai–Dodd model by considering the geometric features of ASBs. Moreover, by introducing a rate-dependent ASB trace angle, the half width of ASBs was expressed as a function of maximum shear strain and critical instability strain. The effect of strain and strain rate to the evolution of voids under tensile conditions was taken into account by combining Hollomon hardening law with Johnson–Mehl–Avrami–Kolmogorov (JMAK) equation. Then, the interactions between micro-defects and structure-related athermal stress were characterized by connecting volume fraction of voids and effective bearing area in tension and the intensity of flow localization with ASBs width in compression. As a consequence, a unified model of constitutive behaviors coupled with micro-defects evolution was established with considering rate-dependent hardening and softening. Applied to aluminum alloys, this model predicts the stress responses, evolution of ASBs and voids, and low or negative strain rate sensitivity with high precision in large ranges of strain (0–0.6) and strain rate (0.001–5000 s−1). The proposed model is thus believed to be with a successful application in precise prediction and optimization of HSRF processes.
Read full abstract