THIS investigation was carried out in the aeronautical-engine laboratory at the Massachusetts Institute of Technology to determine the practical value of the use of a fuel-injection system in place of a carbureter on an Otto-cycle engine using spark ignition. Gasoline was used for most of the investigation, but comparative tests were also made using fuel oil. The equipment used was a single-cylinder laboratory-test engine and such other apparatus as was necessary for a complete performance test. A Diesel-engine injection-pump was used. A Diesel-type injection valve was used for injection into the inlet manifold, and a valve of special design, giving fine spray and little penetration, was used for the tests in which the injection was into the cylinder.
A series of tests was run to compare the performance of an engine having (a) injection of fuel into the inlet pipe and (b) injection into the cylinder, on the one hand, with the performance obtainable with the conventional carbureter, on the other. The results of this investigation are herein presented by means of graphs of engine performance as determined from the tests and a discussion of the graphs pointing out significant facts indicated and suggesting an explanation of each.
Fuel injection, either into the inlet pipe or into the cylinder, was found to be superior in performance to the usual type of carburetion. The available power was increased by over 10 per cent with injection into the cylinder, and substantially lower fuel consumption was obtained. The mechanical problems of injection were found simpler than those of a similar system for Diesel engines. Operation with fuel oil compared very favorably with gasoline operation at a compression ratio at which no detonation occurred with either fuel. Unfortunately, this compression ratio had to be quite low on account of the poor antiknock-characteristics of the fuel oil. The influence of injection instead of carburetion of the fuel, either gasoline or fuel oil, upon the highest useful compression-ratio was insignificant.
Discussion* of the paper brought out a number of suggested directions in which further research might be profitable. Among them were stratification of the charge, to produce a good mixture at the spark-plug while the total mixture-ratio is poor, and injection after the inlet valve is closed.
Reduction of the volume that must pass the inlet valve when fuel is injected into the cylinder was advanced as a reason for the greater volumetric efficiency with such injection. Vaporization of gasoline injected into the cylinder was said to be almost instantaneous, because of the high temperature that it encounters. Metering and distribution with pumps were said to be more accurate than with a carbureter at low throttle.