Engineering Applications of Multi-Dimensional CFD Analysis of Lubrication System

This paper reports on engineering insights that can be gained from a rigorous, transient, three-dimensional CFD analysis of the complete lubrication system of automotive internal combustion engines. Building such a model is a formidable task because the computational domain of such a model is vast and includes scores of bearings as well as components such as the pump, pressure relief valve, oil filter, oil cooler, piston cooling jets etc. Thus far, the only publication on 3D CFD analysis of an engine lubrication system was for a 16-cylinder engine in which the feasibility and the potential opportunities of such a model were demonstrated. The aim of this work is to cover four engineering topics of interest in a lubrication system: 1. Showcase the capability of the CFD tool to accurately, robustly and reliably predict the engine lube system performance of a wide variety of automotive engines with no reliance on tuning the inputs. Best analysis practices were developed in order to achieve this goal by compensating for the variabilities that arise in a complex system level CFD model that connects scores of components across far flung locations in the engine lube system. The predicted pressures were validated at different locations in the lube system. Three different engines were considered, two of which were 4-cylinder and the third was 6-cylinder. 2. The paper also addresses the ability of such a model to help study dependencies of the pressure and flow rate predictions, at different components in the system, on the pump and bearing clearances. 3. One of the other points of interest addressed in this paper is the capability of such models to decide the pump to be chosen for a lubrication system. The study compares the pressures and efficiencies obtained from a system by attaching four different oil pumps and then it ranks them in the order of best to worst for that particular system. 4. Finally, the cold start priming analysis is performed on a lubrication system to estimate oil filling times and predict pressures in the system. The paper also discusses in detail the relatively similar timelines to setting up and running such a complex CFD model to that of a 1-D model. The paper illustrates the potential of 3D CFD that can be used as stand-alone alternatives or in conjunction with 1D CFD in performing engine lube system design and development.