Date of Award


Document Type

Open Access Master's Report

Degree Name

Master of Science in Mechanical Engineering (MS)

Administrative Home Department

Department of Mechanical Engineering-Engineering Mechanics

Advisor 1

Scott A. Miers

Committee Member 1

Darrell L. Robinette

Committee Member 2

Scott W. Wagner


The purpose of this project was to explore the emissions, combustion, and performance effects of running a gasoline/ethanol fuel mixture of 20 percent by volume (E20) in a fuel-injected, two-stroke engine. The engine was operated at five engine speeds that corresponded with the EPA 5-mode emissions test for snowmobile engines. Single parameter sweeps were conducted along with a preliminary recalibration of the test engine at two E0 target values (lambda and mid-pipe temperature) using E20 fuel.

Baseline testing showed that running E20 fuel produced a leaner air/fuel mixture compared to E0, resulting in higher lambda values for all modes and higher mid-pipe temperatures in modes 1 and 2. The increase in lambda resulted in lower CO and THC emissions at all modes and an increase in formaldehyde and acetaldehyde emissions. An increase in CO2 and NO emissions followed the trend of increasing mid-pipe temperature at modes 1 and 2.

Single parameter sweeps were performed by changing one engine parameter at a time and sweeping over a range of predetermined values. Engine parameters included injection time (duration), injection end angle, and ignition timing. Increasing the amount of fuel injected into the combustion chamber decreased lambda values, decreased mid-pipe temperatures, increased CO and THC emissions, and decreased CO2, NO, formaldehyde, and acetaldehyde emissions. Advancing the ignition timing decreased mid-pipe temperatures which decreased CO2, NO, formaldehyde, and acetaldehyde emissions. CO and THC emissions were increased with the advancement of ignition timing. Opposite trends could be seen with retarding ignition timing, except with NO emissions where retarding ignition timing also resulted in a reduction in NO emissions. Adjusting the injection end angle showed little effect on performance, but increases in CO2, NO, formaldehyde and acetaldehyde emissions were seen at large advances of degrees.

Recalibration of injection parameters for E20 fuel to meet E0 baseline lambda values was performed by increasing the injection timing values in the ECU. This created a richer mixture at all modes when compared to the E20 baseline test, while some modes were still leaner than stoichiometric. Matching lambda values resulted in mid-pipe temperatures that were still higher than the E0 baseline test in modes 1 and 2. CO emissions were still lower in all modes except in mode 3 as well as THC emissions except for an increase of two percent in mode 1. CO2 and NO emissions saw a decrease in mode 1 although both values were still higher than the baseline E0 test.

Meeting E0 mid-pipe temperatures with E20 fuel resulted in a higher lambda value at modes 1, 4 and 5. CO emissions followed these trends with higher values in modes 2 and 3 when compared to the E0 baseline test. CO2 emissions were opposite CO emissions with increases at modes 1, 4 and 5. NO and THC emissions saw an increase at mode 1 and decreases in modes 2 through 4.