An experimental study of the mixture response performance of
novel, port-fuel injection strategies upon combustion stability in
a gasoline engine was undertaken at low engine load and speed
conditions in the range of 1.0 bar to 1.8 bar GIMEP and 1000 rpm to
1800 rpm. The aim was to improve the thermal efficiency of the
engine, by extending the lean limit of combustion stability,
through promotion of stable charge stratification. The
investigation was carried out using a modified 4-valve
single-cylinder head, derived from a 4-cylinder, pent-roof,
production, gasoline engine. The cylinder head was modified by
dividing the intake tract into two, separate and isolated passages;
each incorporating a production fuel injector. The fuel injection
timing and duration were controlled independently for each
injector. The performance effects of a single or multiple fuel
injection event on a single-sided injector were compared to
simultaneous and phased fuel injection for the pair of injectors,
with both open valve or closed valve fuel injection timings. A
model of the engine, implemented in the Ricardo WAVE software and
refined using in-cylinder pressure data, was used to support the
findings. The initial experimental results showed good agreement
with the model's prediction and baseline data obtained in a
previous study. Analysis of the experimental results for the
alternative injection strategies showed that the engine could be
operated with far leaner mixtures at low speeds and loads.
Combustion stability, defined for a single-cylinder engine as 10%
CoVGIMEP, was improved for each engine condition tested. At
1000 rpm and 1.0 bar GIMEP, the lean combustion limit was extended
from 14:1 air-to-fuel ratio (AFR) to 17.5:1. At 1500 rpm and 1.5
bar GIMEP, the lean combustion limit was extended from 17.5:1 to
approximately 21:1 AFR. At 1800 rpm and 1.8 bar GIMEP, the lean
combustion limit was improved from 21:1 AFR to 22:1. The improved
tolerance of the combustion system to charge dilution, due to the
optimized injection strategies, was evaluated for high levels of
trapped residuals. The relevance to conditions required for
controlled auto-ignition combustion is discussed. Finally, the
influence of the new strategies upon the rates of heat release and
the combustion duration were evaluated and compared to
cycle-resolved measurements of the concentration of unburnt
hydrocarbons in the exhaust gases.