In this paper a computational fluid dynamics (CFD) approach for solving elastohydrodynamic lubrication using the freely available package OPENFOAM is introduced. The full Navier–Stokes equations are solved, which enables the entire flow domain to be modeled and all gradients inside the lubricated contact to be resolved. The phenomenon of cavitation is taken into account by employing a homogenous equilibrium cavitation model, which maintains a specified cavitation pressure inside the cavitating region. The energy equation used considers the effects of heat conduction and convection, viscous heating, and the heat of evaporation. The developed method has been applied to a series of cases of lubricated metal-on-metal line contact with an entrainment velocity of uent=2.5m∕s, viscosities η0=[0.01,1]Pas, and slide-to-roll ratios SRR=[0,1,2] under both thermal and isothermal conditions. The isothermal results are compared to the Reynolds theory and most results agree very well. Only the high-viscosity pure rolling case shows small differences. The combined effects of temperature, pressure, and shear-thinning are studied for the thermal cases. A temperature-induced shear band occurs in the case of sliding combined with very large viscosity compared to the isothermal case, which results in significant pressure variations across the thickness of the film. The impact of temperature on the friction force is discussed, showing differences of up to −88.5% compared to the isothermal case. The developed method is capable of giving new insights into the physics of elastohydrodynamic lubrication, especially in cases where the usual assumptions of the Reynolds theory break down.

This paper describes a new computational fluid dynamics methodology for modelling elastohydrodynamic contacts. A finite-volume technique is implemented in the ‘OpenFOAM’ package to solve the Navier-Stokes equations and resolve all gradients in a lubricated rolling-sliding contact. The method fully accounts for fluid-solid interactions and is stable over a wide range of contact conditions, including pressures representative of practical rolling bearing and gear applications. The elastic deformation of the solid, fluid cavitation and compressibility, as well as thermal effects are accounted for. Results are presented for rolling-sliding line contacts of an elastic cylinder on a rigid flat to validate the model predictions, illustrate its capabilities, and identify some example conditions under which the traditional Reynolds-based predictions deviate from the full CFD solution

A commercial CFD code has been applied to model lubricant flow behaviour within 2D and 3D linear pad bearings having closed pockets or recesses. The study indicates that the presence of closed pockets can result in a significant reduction in bearing friction coefficient. At high convergence ratios, no cavitation is predicted within the pockets. This means that suitably-positioned pockets in the high-pressure region of the bearing result in a much greater reduction in local shear stress than in local pressure, so that there is an overall reduction in friction coefficient. At low convergence ratios, cavitation occurs in the inlet to the pockets so that each pocket acts an effectively-independent step bearing. This results in the overall bearing supporting a higher load and thus having lower friction coefficient than is the case without pockets.

The onset of smearing damage was studied under controlled conditions in a custom test rig that simulates the passage of a rolling element through loaded and unloaded zones of a rolling bearing. The set-up comprises a spherical roller which is intermittently loaded between two bearing raceways driven at a prescribed speed. The roller is free to accelerate during the loading phase. Contact load, roller speed and acceleration and electrical contact resistance are recorded during the test. Contact shear stress, friction coefficient, frictional power intensity and elastohydrodynamic film thickness are calculated from the recorded kinematics data. Results suggest that the first onset of smearing occurs early in the loading phase where the roller is nearstationary and the frictional power intensity is high. The raceway speed at the onset of damage decreases with increasing load and increasing lubricant supply temperature. The maximum frictional power intensity is found to be relatively constant at all contact conditions that led to 2 smearing. An existing thermo-mechanical contact model is used to estimate the contact temperature distribution under smearing conditions and the potential for EHL film thickness reduction due to forward heat conduction.

This article presents a novel approach for modelling rolling contact fatigue cracks in the presence of lubricants. The proposed formulation captures the interaction between fluid pressure and solid deflections both at the contact interface and along the crack faces using a fully coupled finite volume/boundary element solver. This enables shedding light on the mechanisms which govern crack propagation in various loading conditions and geometrical configurations. It is shown that by linking the fluid behaviour and the elastic deflections within the crack to the film formed at the contact interface it is possible to overcome one of the main limitations of classical models available in the literature, which consists in having to prescribe pressure and/or pressure gradient at the crack mouth during the each loading cycle. The application of linear elastic fracture mechanics principles for the determination of crack stress intensity factors suggests that the results obtained using the approach developed by the authors produce a more realistic characterisation of the crack tip behaviour and it is capable of producing an improved estimate of crack propagation rates. Implications of these findings for the development of rolling contact fatigue lifing tools and potential extensions of the technique are also discussed.