A Three-Dimensional Deterministic Model for Rough Surface Line-Contact EHL Problems

Abstract

This paper reports the development of a novel three-dimensional (3D) deterministic model (3D L-EHL) for rough surface line-contact mixed-elastohydrodynamic lubrication (EHL) problems. This model is highly demanded because line contacts are found between many mechanical components, such as various gears, roller and needle bearings, cams and followers, and work rolls and backup rolls in metal-forming equipment. The macro aspects of a line-contact problem can be simplified into a two-dimensional (2D) model; however, the topography of contacting rough surfaces, microasperity contacts, and lubricant flows around asperities are often three-dimensional. The present model is based on Hu and Zhu’s unified 3D mixed-EHL model (Hu and Zhu, 2000, “Full Numerical Solution to the Mixed Lubrication in Point Contacts,” ASME J. Tribol., 122(1), pp. 1–9) originally developed for point contacts and the mixed fast Fourier transform (FFT)-based approach for deformation calculation formulated by Chen et al. (2008, “Fast Fourier Transform Based Numerical Methods for Elasto-Plastic Contacts With Normally Flat Surface,” ASME J. Appl. Mech., 75(1), 011022-1-11). It is numerically verified through comparisons with results from the line-contact Hertzian theory and the conventional 2D line-contact smooth-surface EHL formulas. Numerical examples involving 3D sinusoidal and digitized machined surfaces are also analyzed. Sample cases indicate that transverse roughness may yield greater film thickness than longitudinal roughness. This observation is qualitatively in agreement with the trend predicted by Patir and Cheng’s stochastic model (1978, “Effect of Surface Roughness on the Central Film Thickness in EHL Contacts,” Proceedings of the Fifth Leeds-Lyon Symposium on Tribology, London, pp. 15–21). However, the roughness orientation effect does not appear to be quantitatively as great as that shown in the work of Patir and Cheng for the same range of λ ratio.