A multi-scale model integrating both global ring pack behavior and local oil transport in internal combustion engines

Abstract

Improving fuel economy of internal combustion engines is one of the major focuses in automotive industry. The piston ring friction contributes as much as 25% of total mechanical loss in internal combustion engines [1] and is an area of great interests to the automotive industry in their overall effort to improve engine efficiency. However, typical methods to reduce friction loss from piston ring pack, such as ring tension reduction, may cause additional oil consumption. A compromise between reduction of friction loss and control of gas leakage and oil consumption needs to be made, which requires a deep understanding of oil transport mechanism. This compromise gives rise to the interest in modeling work. Both experimental results and previous experience showed that oil film distribution on the piston varies significantly along the circumference and the oil leakage occurs locally. Therefore to predict oil transfer across different ring pack regions, one needs to integrate both global and local processes. This work is aimed at establishing an enduring framework for all the cycle-based processes at different length scales. As a first step, a multi-scale multi-physics piston ring pack model was developed by coupling the structural dynamics of the piston rings with gas flows and local interactions at ring-groove and ring-liner interfaces. A curved beam finite element method was adopted to calculate the ring structural response to interaction between the ring and the liner as well as the ring and the groove. Compared to a traditional straight beam finite element method, the curved beam separates the structural mesh and contact grid by utilizing the shape functions. In this work, a multi-length-scale ring pack model was, for the first time, successfully assembled. This model bears its fundamental values to truly reflect the integral results of all the relevant mechanisms. The significance of the current work is that it demonstrated such an integration of all the length scales is possible for a cycle model with a reasonable computation cost. With the current model, one can realistically investigate the effects of all kinds of piston and liner distortion, piston secondary motion, bridging, and lube-oil dilution on gas sealing, oil transport and lubrication. As a result, optimization of the ring designs and the part of block design contributing to bore distortion can be coordinated to reduce development costs.