Fast thermo-elasto-hydrodynamic modeling approach for mixed lubricated journal bearings in internal combustion engines

摘要

Journal bearings are key components in internal combustion engines. Their reliability, durability, and economy are of highest importance. Especially, the accurate prediction of friction loss power and wear are essential for the development of an engine. For an appropriate representation of the hydrodynamic load-carrying capacity and also the friction behavior, both the dynamics of the contacting components, the shape of the contacting component surfaces, the amount of the available lubricant, and the properties of the lubricant itself are of importance. A Reynolds-averaged equation with laminar flow conditions in combination with an asperity contact model is a typical modeling approach for that purpose. The lubricant properties are in particular influenced by its thermal conditions on one hand. On the other hand, the thermal conditions are influenced by the mixed lubricated contact conditions as well. These interactions require a coupled modeling approach, which combines the component flexibility and its interaction with load-carrying capacity as well as the thermal behavior of the lubricant and the component surfaces. In this work, a thermo-elasto-hydrodynamic contact model is presented, which computes the oil film temperature using a 2D energy equation. The 2D equation is derived from the equivalent 3D energy equation by integration over the clearance gap height. Besides component material properties such as specific heat capacity, density, heat conductivity for lubricant and structures, also heat transfer through mixed lubricated regimes and partly filled clearance gaps, as implied in cavitation regions, are being considered. The presented method is applied for a typical engineering task of a sensitivity analysis for oils with different viscosity index (VI) improvers in a main bearing of a four-cylinder inline diesel engine. The influence of the oil film temperature on the oil film viscosity and therefore on the load-carrying capacity is shown. Furthermore, the simplified 2D approach is compared with a 3D approach both in terms of obtained result data and in terms of elapsed calculation times. The presented results show similar accuracy of the 2D approach with significantly reduced simulation time compared to the equivalent 3D case.