Measurements and Multidimensional Modeling of Gas-Wall Heat Transfer in a S.I. Engine

Abstract

<div class="htmlview paragraph">The computational fluid dynamics codes, which help to predict the behaviour of combusting gas in reciprocating engines, need, as boundary conditions for the momentum and energy equations, to approximate wall frictions and heat transfer between gas and walls. The purpose of this work is to outline the physical parameters which affect the heat transfer in spark ignited engines, and then to validate a multidimensional model which takes these parameters into account.</div> <div class="htmlview paragraph">In an experimental study, measurements were made on a test-engine instrumented with fast-response surface heat flux gages. Each gage was made of a steel cylinder, containing two thermocouples. A laser doppler velocimeter was used to analyze the influence of fluid dynamics on heat transfer. Up to ten data inputs could be simultaneously recorded at each crank — angle, including the heat flux at four locations of the combustion chamber, two velocity components and the cylinder pressure.</div> <div class="htmlview paragraph">A parametric analysis of the engine conditions including spark timing, volumetric efficiency, engine speed, equivalence ratio and swirl number was carried out. It revealed that: <ul class="list disc"> <li class="list-item"> <div class="htmlview paragraph">the gas density variation greatly affects the heat transfer when the spark timing or the load are changed. It also plays a role when the equivalence ratio is varied.</div></li> <li class="list-item"> <div class="htmlview paragraph">turbulence intensity is the main parameter influencing the heat transfer during variations of the engine speed.</div></li></ul></div> <div class="htmlview paragraph">In a computational study, a heat transfer model was introduced in a 3-dimensionai code. Because of prohibitive computing time, the boundary layer could not be meshed, so the model relies on a wall function. Coupled with a k-ε formulation of the turbulence, the response of this mode! to the density and turbulence was studied, analysing the influence of spark timing and engine speed. A comparison between measurements and computational predictions was done, showing as good agreement in terms of instantaneous and local heat-flux as in terms of global heat balance.</div>