Analytical investigation of fretting wear with special emphasis on stress based models
Fretting refers to the minute oscillatory motion between two surfaces in contact under an applied normal load. It can cause either surface or subsurface initiated failure resulting in either fatigue or wear or both. Two distinct regimes – partial slip and gross slip are typically observed in fretting contacts. Due to the nature of contact, various factors such as wear debris, oxidation, surface roughness, humidity etc. effect failures caused due to fretting. A number of different techniques have been developed to quantify fretting damage and several numerical models are proposed to predict damage due to fretting. Fretting wear also depends on the inherent material properties of the two bodies in contact. In this investigation, material properties such as modulus of elasticity, tensile strength, hardness etc. are combined with damage quantification techniques such as damage mechanics and linear elastic fracture mechanics to propose novel fretting wear models. The wear models are independent of global wear coefficients obtained from experiments and therefore, could reduce the dependence on wear experiments to predict wear rates for different materials in contact. During fretting wear, the generated wear debris is often trapped in the contact. The contact parameters can be significantly affected due to the shape, size and properties of the third body layer. A finite element model is created to study the changes in fundamental contact parameters in presence of third body particles and effects of number of third bodies and their properties on fretting contact is investigated. In addition to material properties, other factors such as surface roughness of the bodies in contact, wear debris etc. also effect fretting wear rate. A detailed study of the effect of surface topography parameters such as RMS roughness, skewness and kurtosis on wear rate is evaluated. An analytical model is proposed which approximates a two-dimensional surface line profile as a collection of Hertzian point contacts. This method is computationally efficient and free of convergence issues compared to FE simulations of rough surfaces. As a result, very deep wear scars can be obtained and evolution of surface can be studied in greater detail. It is assumed that the fretting wear is in gross slip and therefore, this model can be extended to any reciprocating sliding condition. The models are validated by comparing them to published literature data. The study of the effect of wear debris, elastic plastic material property and variable coefficient of friction is proposed as an extension of the current work.
Sadeghi, Purdue University.
Off-Campus Purdue Users:
To access this dissertation, please log in to our