Investigation of Automotive Fuel Tank Filling and Venting

This research was conducted in collaboration with Kautex Textron through an Natural Sciences and Engineering (NSERC) Collaborative Research and Development (CRD) Grant. It involved a study of the automotive fuel tank filling process to predict the complicated two-phase liquid fuel and vent flow rates as well as tank dome pressure. Of particular interest was the prediction of the occurrence of premature shut-off and spill-back. An experimental investigation on an instrumented, clear fuel tank, using gasoline of various grades of volatility and flow rate, formed the basis of the M.A.Sc. thesis of Maurizio Mastroianni. A simplified, lumped-parameter model of the entire tank filling process was developed by Shelagh Fackrell as her M.A.Sc. thesis. The model included the effects of vapour generation, fuel volatility, compressibility of the vapour vent flows , fuel dispensing flow rate and venting conditions. It was capable of predicting the occurrence of premature shut-off for the experiments to which it was compared and fair agreement with the predicted pressures. A more accurate numerical model of the tank filling process has been developed by Allan Zhao, a Ph.D. student who successfully defended his Ph.D. dissertation in December 2002. A combination of a computational fluid dynamic (CFD) model of the complicated two-phase flow in the filler tube with lumped parameter models of the vent tube flows and tank vapour and liquid was utilized. This combination is shown to result in a reduction in computational time compared to the CFD solution. This work has resulted in a better understanding of the complicated fuel tank filling process and a practical tool for fuel tank designers.