Numerical Simulation of Contacts Working Under Mixed Lubricating Conditions.

Abstract

Numerical simulation of lubrication is a powerful and cost-effective tool for designing machinery that is both efficient and durable compared to experimental methods. It helps to understand and optimise the behaviours of lubricants under various operating conditions. Additionally, it also allows engineers to predict the major consequences that can arise on the machine components working under various regions of lubrication including hydrodynamic, boundary and mixed lubrication. This thesis aims to conduct numerical simulations using a deterministic model of contacting surfaces working under the mixed region of lubrication. Mixed lubrication (ML) indicates a state in which hydrodynamic lubrication (HL) and boundary lubrication (BL) coexist. In this state, contacting surfaces experience both the presence of a lubricating oil film and direct solid-to-solid contact between asperities on the surfaces. This duality introduces significant complexities in defining and modelling the ML region when simulating the behaviour of the tribological system. A substantial aspect of the ML regime is the asperity-asperity interface, whose tribological performance is significantly influenced by contact mechanisms, flash temperature and tribo-chemistry reactions. However, in this study, the focus has been solely on numerically simulating the fluid flow around asperity contacts without considering the contact mechanics or the tribofilms forming between asperities The Reynolds equation (RE) is the fundamental equation used to model fluid film lubrication. However, in the ML regime, where both BL and HL coexist, the accurate definition of boundary condition (BC) becomes challenging. The problem of simulating ML arises from the fact that this is a complex lubrication state with unique characteristics. In real-world applications, ML is often encountered in situations where the load on the machine component is high, and the speed of operation is slow. This situation is common in various machine components, for instance, gears, rolling element bearings, journal bearings, cam and follower mechanisms, piston ring lubrication etc. This is the crucial lubrication state where the transition occurs from BL to HL. In BL the external forces (load) are entirely supported by the asperity contact, whereas in HL , the external load is entirely maintained (supported) with the help of oil film separating the contacting surfaces. In contrast to HLFigure 1.1 The Stribeck curve [3]., the interface for the mixed region of lubrication is not fully separated. Therefore, some external load is sustained by interrupted lubricant film, and some by direct contact with solid asperities. As a result, when trying to apply the Reynolds equation for ML simulation, determining appropriate mathematical boundary conditions becomes problematic. Inaccurate boundary conditions can lead to unrealistic simulation results, and it becomes challenging to predict the behaviour of lubricant film accurately in the ML regime. Direct metal contact within ML can significantly increase the risk of high-stress concentration, friction, heating, and energy losses. These factors collectively contribute to irreversible harm like wear, scuffing, and fatigue failure. Traditional approaches, such as the Reynolds model and inappropriate boundary conditions, cannot precisely predict such outcomes, especially within conditions of ML. Therefore, it is imperative to accurately predict the behaviour of lubricants in ML regimes in order to minimize the hazards of failure. The Reynold equation, designed for full film lubrication or HL isn’t applicable when direct solid asperity contact occurs. This is due to the significant changes in lubrication dynamics in these regions. Without proper lubrication, unpredictable energy losses and increased friction can impair system efficiency. A more systematic theoretical and numerical simulation analysis is required in ML to effectively tackle the fluid-solid interaction inherent in this ML. Furthermore, there is a need to understand the flow behaviour around the asperity contact. Although a significant volume of research has been conducted in modelling lubricated contacts, no model can still adequately and efficiently solve the ML operating conditions. This is in part because the mathematical formulation of ML problem is incomplete and some of the assumptions that are frequently utilised in HL may not be applicable when dealing with ML. The consequences of these limitations are profound, including the need to develop robust designs that consider mixed lubrication scenarios to prevent premature failures. This thesis introduces a Mixed Lubrication theory that has been scientifically validated and is distinct in its approach to incorporating roughness interactions with fluid that occurs in the ML region. The theory is founded on the conventional RE. The precise prediction of pressure distribution in the lubricant film provided by the proposed model will help industrial engineers, researchers, and academic experts to better understand the behaviour of mixed lubrication under severe loading conditions.