ABSTRACT
Examination of Harris’ results (1987), and others has led to the suggestion that a dissipative process is responsible for both the amplitude effect and the hysteresis in filled rubber (Ahmadi, Gough, Muhr & Thomas 1999 and Gough 2000). Therefore any generalized 3 dimensional model for filled rubber should exhibit a dissipative process that follows the “retraction rule” observed during Harris’ uniaxial simple shear tests. The exact nature of the micro-mechanical processes responsible for this dissipative behavior in filled rubber is not well understood. However, in the absence of this knowledge the computational mechanics community can either model filled rubber using a phenomenological approach or use models based on physical processes that exhibit such behavior. Plasticity models are one possible candidate material model that do exhibit a “retraction rule” which may provide an approach to modeling the phenomena studied by Fletcher & Gent, Payne and Harris mentioned above. The two major advantages of using such models are a) the existence of extensive scientific theories, developed primarily for steel industry, in the form of plasticity literature and b) the
ABSTRACT: The paper describes the desired hardening rule for a proposed viscoplastic model for filled rubber. The suitability of plasticity models currently available in a commercial FE code for this purpose is then examined. It is concluded that a recently implemented Multilinear Kinematic Hardening Plasticity (MKHP) material model is capable of providing the desired Masing rule for filled rubber and is indistinguishable from a previously proposed overlay approach by Austrell, leading to a significant reduction in computation time. In addition, results are presented for the overlaid model where a newly offered finite strain plasticity model is used to calculate the response of hyperelastic-perfectly-plastic materials. The model uses multiplicative decomposition of deformation gradient and has been used to show that finite yield strains may also be used in the viscoplastic model with additive decomposition after all without significant problem.
