Computational Analysis of Heat Transfer and Fluid Flow with Relevance for IC-Engine Cooling

Abstract: Modern IC-engines produce more power per displaced volume and operate at higher combustion pressures and temperatures. In addition, engine structure materials such as aluminium and magnesium alloys, which are more sensitive to thermal loads, are more commonly used. These two parameters might lead to problems such as thermal fatigue, knock, high fuel consumption and emissions, performance loss, bore distortion, oil consumption, noise and vibration. These may, however, be limited by careful design of well-aimed precision engine cooling strategies which provide efficient and sufficient cooling of the extremely hot regions while avoiding over-cooling of the other regions. To enable precision cooling, an accurate and quick prediction tool is needed. CFD has been used during the last decade as a tool for optimization of flow and pressure drop in the IC-engine coolant jackets. There is, however, a need to improve the CFD-methods for more accurate simulations and to extend its benefit to the prediction of the heat transfer. The goal of this work is to contribute to the understanding of the fluid flow and heat transfer mechanisms relevant to IC-engine cooling and to develop CFD methods for realistic predictions of the heat transfer coefficient for this kind of application. In this work the flow and heat transfer processes have been investigated for well known generic cases relevant to the engine cooling applications. These cases consist of confined flow in curved ducts with various cross section shapes. The performance of several turbulence models including different versions of high- and low-Reynolds number two-equation models, mostly k-epsilon models, with linear and non-linear Reynolds stress formulations and also V2F and RSM have been investigated. Wall functions, damping functions and a two-layer technique were used for wall treatment. Available experimental data in the literature have been used for validation of these studies. The strengths and weaknesses of these models are discussed in the thesis. The focus has been on forced convection heat transfer. The impact of parameters such as inlet boundary conditions and duct cross-section shape on the flow and heat transfer has also been studied. It was found that these parameters, especially the upstream boundary conditions have significant influence on the pressure-driven secondary flow and, thereby, the heat transfer mechanism in curved ducts. The convergence problem of V2F and the RSM and the physical shortcomings of linear eddy viscosity models as well as the robustness of models like Suga's cubic low-Re k-epsilon model and Chen's k-epsilon model have been discussed. Finally full scale IC-engine cooling analyses utilizing some of the gained experiences from the more fundamental work have been presented.