A system was developed for generating a premixed, homogeneous charge using liquid gasoline fuel which is part of an ongoing effort to determine the effect of mixture preparation on the emissions and performance of utility engines. The mixture system developed for this work was designed with the following parameters in mind:

1) Atomization of the liquid fuel
2) Vaporization of the liquid fuel
3) Mixing of the vaporized fuel with the intake air
4) Safety

Liquid fuel atomization was accomplished by forcing fuel through a sapphire orifice at moderate pressures. Vaporization was aided by heating the intake air and increasing intake charge residence time. Large and small scale mixing of the fuel and air were enhanced by inducing turbulence and increasing residence time. Safety mechanisms were included to prevent unplanned mixture ignition, relieve high pressures in system components, protect the experimenter, and enable rapid shutdown if necessary. The ability of the mixture system to produce homogenous air/fuel mixtures was verified.

The engine used for this project was a Briggs and Stratton Model 977. It was a four-stroke, spark-ignited, single-cylinder, overhead valve-type utility engine with fixed spark timing and a rated speed of 3600 RPM. The engine speed was maintained at 3060 RPM (85% of the rated speed) for all tests. Engine tests were run at modes 8 and 10 of the SAE J1088 Test Procedure over a wide range of A/F's using the stock carburetor and the homogeneous mixture system (HMS). Engine testing provided one means of evaluating the effectiveness of the mixture preparation system.

The HMS allowed adjustment of intake air temperature, system wall temperatures, fuel pressure, engine inlet and exhaust pressures, A/F, fuel injector orifice size, and induced flow turbulence.

The work covered in this thesis provides a preliminary look at the effect of mixture preparation on emissions and performance by comparing the emissions and performance of a stock, carbureted engine with that of an engine provided with a homogeneous air/fuel mixture.

Engine testing was performed at modes 8 and 10 of the SAE J1088 test procedure (5) using the stock carburetor and HMS induction systems. Testing covered a range of A/F's extending from fuel-rich to the lean burn limit.

Test results showed that emissions levels are a strong function of A/F, and that there is a relationship between mixture preparation, exhaust emissions, and engine performance. The importance of mixture preparation increases as operation leaner than the stoichiometric A/F is desired.

Mixture homogeneity and A/F consistency extended the lean burn limit for the 50% load case. Hydrocarbon emissions were reduced by an average of 2000ppmC1 for the 10% load case at A/F's lean of the stoichiometric value. The 50% load case showed even more dramatic HC reductions than the 10% load case. Peak NO emissions levels increased approximately twofold as a result of improved mixture homogeneity for the 50% load test case.

Carbon monoxide levels decreased for the HMS-equipped engine in the region of approximately +/-1.5 A/F and +/-1 A/F from the stoichiometric value for the 50% load and 10% load cases, respectively. The CO levels of the HMS-equipped engine at the stoichiometric A/F were reduced by factors of 3 and 1.5 for the 50% load and 10% load cases, respectively when compared to the carbureted engine.

Variations in HC, CO, and CO emissions levels decreased for both load cases with the HMS-equipped engine when compared to the carbureted engine. Also, power output averaged 0.2BHP higher for the HMS-equipped engine at the 50% load test case at A/F's leaner than the stoichiometric A/F.