Finite element analysis of components subjected to ratchetting and creep
Many components in conventional and nuclear power plant, chemical plant and aero engines may be subjected to severe loading conditions, i.e. loads which cause reverse plasticity and/or incremental growth (Ratchetting). If operating temperatures are high, creep strains may also be significant and may exacerbate the ratchetting process. Also the residual stress fields associated with the cycling of load in the plastic region for a material will influence the accumulation of strain during the dwell periods between cycles when steady loading is sustained. Some analytical solutions for the cyclic behaviour of simple components and loadings are available, however very little information on the effects of stress concentrations and complex loading conditions on ratchetting is published. A better understanding of the mechanisms of ratchetting for complex components and loadings is essential in order to identify characteristic behaviours which can be used to aid the design process for components in potential ratchetting situations. A range of component geometries, uniform sections and stress concentrations, and loading conditions have been analysed by the finite element method to investigate ratchetting mechanisms and to obtain ratchet and dwell period strain data. The effects of stress concentrations, material behaviour models, loading conditions and stress redistribution due to creep on ratchetting mechanisms and strain accumulations are described. Dwell period creep effects are bounded by the 'no creep' (zero dwell period) condition on the one hand and by complete redistribution between cycles at the other extreme. The results of the analyses have been successfully used to 'extend existing approximate design rules for simple components to these more complex components and loadings. It has been shown that reasonable estimates (in some cases exact solutions) can be obtained from either a limited finite element analysis or by using approximate methods of solution. Comparisons between experimental ratchetting data for two components made from a lead alloy material and equivalent finite element predictions are presented. Simple material behaviour models are used and the results highlight both the benefits and shortfalls of these models. Improvements to modelling techniques for more accurate predictions are suggested although it is shown that, in certain circumstances, more realistic material behaviour modelling is unwarranted.