24
DYNAMIC PROPERTIES of FILLED RUBBER – Part V: Prediction of Cyclic Behavior of Elastomeric Components
Hamid Ahmadi**, Jan Kiok Chye Har*, Alan Muhr**& Khee Woon Wong*
* Kumpulan Jebco Sdn.Bhd.-Malaysia
**Tun Abdul Razak Research Centre-TARRC, UK
In parts I and II, a simple “viscoplastic” model was proposed, capable of implementation in existing commercial FEA packages, with the scope to capture those aspects of the stress-strain behaviour of filled rubber that are most significant in engineering application, in particular the Payne or Fletcher-Gent effect. Attention was given to assembling the model from separate physical contributions, namely viscoelasticity and elastoplasticity, so that not only is the number of parameters small but also they may be at least semi-quantitatively related to the formulation of the elastomer. It was confirmed that the proposed “viscoplastic” approach captured the essence of the behaviour when examined in several modes of deformation for filled NR and SBR.
Part III evaluated the Multi-linear Kinematic Hardening Plasticity (MKHP) rule recently implemented in the Abaqus FE code that could provide the appropriate stress-strain hysteresis loops for filled rubber. The model was shown to give similar results as that achieved by the overlay of several FE meshes of elastic-perfectly-plastic material, resulting in substantial reduction in computation time compared with the "overlay" approach.
Part IV explored the capability of the viscoplastic model to fit a comprehensive dynamic property data base, including effects of temperature, frequency and filler volume fraction, using small number of physically motivated.
This paper draws on the work published so far and examines the ability of the model to predict the cyclic behaviour of engineering component(s) with nonlinear strain distributions. The parameters of the viscoplastic material model are fitted to materials characterisation tests carried out either in quasi-static simple shear or to dynamic simple shear.
The influence of the pre-scragging on the predictions of the model is also discussed.