Search Results

A numerical study which compares between the diffusion limit (DL) and the well mixed (WM) droplet heating models under diesel combustion conditions is presented. The CFD code KIVA-II is used to perform the computations. Predictions are compared with available experimental data in n-decane and n-heptane sprays. For the n-decane spray measured gas temperatures support employing the DL-model, and as compared with the DL-model the WM-model clearly yields underpredicted vapor concentrations. In the n-heptane spray vapor concentration predictions show a strong independency of the employed heating model, and generally are in a good agreement with the measurements. Based on this independency, in subsequent comparisons between the two models under different spray operating conditions only the fuel with the lower volatility (n-decane) is considered.

The impact of two models representing two limiting cases of the droplet heat-up process, namely the infinite diffusivity (ID) and the diffusion limit (DL) model, on the evaporation, self-ignition and subsequent combustion in a diesel spray is investigated. The simulation results show that, as compared with the DL model, the ID model leads to an over-prediction of the length of the ignition delay period by about 20 %. This is attributed mainly to the under-prediction of the evaporated mass predicted in the case of the ID model, during the early stages after injection. Accordingly, the time evolution of the combustion processes in the case of the ID model is found to lag that predicted when the DL model is employed, by roughly 0.3 ms.

The vaporization of n-dodecane sprays during the early stages of the vaporization process under engine-like conditions is studied numerically using Well-mixed (W-M) and diffusion-limit (DL) evaporation models. A spray with Sauter mean radius (SMR) of 12 mm at injection is considered. The initial droplet size distribution is obtained from a χ-squared distribution using a Monte Carlo sampling technique. The initial ambient air pressure and temperature are taken to be 45 bar and 800 K respectively. The effect of the injection velocity on the vaporization process is studied by considering two initial injection velocities 100 and 200 m/s. A modified version of the finite volume code Kiva-II is used to solve the governing equations for both the gas and the liquid phase.