Finite element analysis of stress singularity in partial slip and gross sliding regimes in fretting wear

摘要

Fretting is a small oscillatory motion between two contact surfaces, which may cause wear or fatigue damage. Many parameters may affect fretting wear including normal load, applied displacement, material properties, surface roughness of the contact surfaces, frequency, etc. The design of engineering components subjected to fretting wear, such as couplings and splines, jointed structures, is still a challenge to engineers. This is because of the continuous change in the contact surfaces of component during fretting wear cycles. Therefore, a predictive technique that takes into account the wear progress during life cycle is desirable. Analytical solutions of wear problems are very difficult and limited to simple 2D configuration steady-state analysis. In contrast, numerical modelling techniques such as Finite Element Analysis (FEA) can be used for any type of structures in 3D configuration with many complicated details such as large deformation, material non-linearity, changes in geometry and time integration effect. In this article, we use FEA to find whether or not there exists a stress singularity at cylinder on flat contact according to different variables, such as applied displacement, coefficient of friction (COF) and fretting wear cycles. Based on a stress singularity signature method, it is found that stress singularity has close relation with fretting regime. There is no stress singularity neither in partial slip nor gross sliding after one-fourth of a fretting wear cycle for lower COF, but it exits for higher COF, in which condition the contact interface is almost stick. After 20,000 cycles, stress singularity exists in partial slip, while there is no stress singularity for gloss sliding condition, when COF is 0.8. Results reveal that more attention should be paid to the mesh size at contact interface, when the contact condition is under partial slip regime. (C) 2014 Elsevier B.V. All rights reserved.