Gasoline Engine HCCI Combustion - Extending the high load limit

Doctoral thesis, 2012

There is an increasing global focus on reducing emissions of greenhouse gases. For the automotive industry this means reducing CO2 emissions of the vehicles manufactured, which is synonymous with reducing their fuel consumption or adapting them for using renewable fuels. This thesis is based on a project aimed at improving the efficiency of gasoline engines in the lower load/speed region. The focus was mainly on a combustion strategy called homogeneous charge compression ignition (HCCI), but also on homogeneous lean and stratified lean spark-ignited combustion. In contrast to traditional stoichiometric spark-ignited combustion, HCCI can operate with diluted mixtures, which leads to better cycle efficiency, smaller pumping losses and smaller heat losses. However, at relatively high loads, HCCI combustion becomes excessively rapid, generating in-cylinder pressure oscillations (ringing), which are perceived as noise by the human ear. The main objective of the project was to identify ways to avoid this ringing behaviour in order to increase the upper load limit of HCCI. This is vital to avoid the need for mode switches to spark-ignited combustion at higher loads and to operate the engine as much as possible in the more effective HCCI mode. The strategy for reducing ringing investigated most extensively in the project was charge stratification, achieved by injecting part of the fuel late in the compression stroke. Available literature on effects of this strategy gave conflicting indications, both positive and negative effects have been reported, depending on the type of fuel and engine used. It was soon found that the strategy is effective for reducing ringing, but with resulting increases of NOX emissions. Further, in order for the strategy to be effective, global air/fuel ratios must not be much leaner than stoichiometric. The increases in NOX emissions were countered by shifting the ratio towards stoichiometric using exhaust gas recirculation (EGR), allowing a three-way catalyst to reduce the excess NOX. Intake air boosting was also experimented on and is discussed as an alternative method or as a method to use in combination with charge stratification. During the project, experiments have been conducted with a productionlike multi-cylinder engine and a single-cylinder research engine to investigate the potential of various strategies for raising the high load limit of HCCI when using gasoline or gasoline-like fuels. To explain observed phenomena, optical experiments were conducted in which high-speed video was used to capture light from the combustion and the residuals. A method was developed to extract pressure oscillations from these measurements and to correlate them to the combustion. Laser-based experiments were further used to analyse fuel and temperature distributions before the combustion to investigate their effects on combustion and pressure oscillations. Based on the acquired data, plausible reasons why charge stratification can reduce ringing, and the circumstances in which it can do so, are presented. The thesis also shows the extent to which the load can be increased using the strategy, and the resulting efficiency penalties, observed in both the production-like gasoline engine and single-cylinder research engine. Finally, the various strategies for load extension using combinations of charge stratification, EGR and boosting were compared to operating the engine in two-stroke HCCI mode. Although two-stroke operation was investigated very briefly, in an engine not designed for it, indications were obtained that this might be a much better alternative, since it provided higher loads, more stable combustion, less ringing, low NOX levels and higher efficiency than any of the other tested load extension strategies.