Rubber-unloading-behaviour evaluation using product-orientated specimen based on a resilience test

In industry, the important design parameters for rubber products are currently almost always based on only the loading part of loading-unloading histories, i.e., load-deflection and fatigue requirements. Rubber-like materials experience different strain energy levels and stress values during loading and unloading for the same load value. Hence, the performance of rubber products may be substantially different during loading and unloading, which can lead to unexpected effects, including the Mullins effect. Herein, a new approach is proposed to account for the Mullins effect. Existing elastomeric models, which are based on the strain energy density, are modified during loading and unloading. A key engineering parameter, the rebound resilience (the ratio between the rebound energy and the initial loading energy), is introduced in this approach. A typical rubber-to-metal bonded component, which is widely used in engine installation, is selected to validate the proposed approach. It has been shown that the predictions offered by the new approach are consistent with the load-deflection histories yielded by loading-unloading experiments. In addition, if the unloading characteristics are not considered, the results obtained from the stress calculations can show an error margin 30%.