Refine Your Search

Search Results

Author:
Viewing 1 to 2 of 2
Technical Paper

A Study of Crevice Flow in a Gas Engine using Laser-induced Fluorescence

2001-03-05
2001-01-0913
We visualized behavior of crevice flows in a spark ignition gas engine by laser-induced fluorescence (LIF) of acetone and OH. The fuel used in this work was methane with 10 vol. % acetone vapor. Acetone acts as a LIF tracer for unburned fuel, and OH radical which exists naturally in hydrocarbon flames is used as a marker species of high temperature zones. This technique enables simultaneous tomographic observations of high temperature zones and unburned zones in engine cylinders, and the technique gives abundant information on oxidation process of fuel. A transparent engine whose cylinder wall was made of fused silica for observations and laser access was used for experiments. A crevice flow released from a crevice between a piston and a cylinder wall was visualized. In this work, we investigated effects of valve timing, back-pressure, and ignition timing. We found that behavior of the crevice flow changes suddenly at moment of exhaust valve opening.
Technical Paper

Air-to-Fuel Ratio Control of Gas Engines Using Response Characteristics of Three-Way Catalysts under Dynamic Operation

1991-10-01
912362
This paper presents a new method of controlling the air-to-fuel ratio (A/F) in gas engines with three-way catalysts. In the method under discussion, the center of the λ-window is detected directly through the dynamic response of an O2 sensor installed at the rear of the catalysts. As the result, the A/F remains stable at the center of the λ-window for long periods of time, even in cases where static response characteristics of the O2 sensor change due to the deterioration of the O2 sensor, or in instances where the λ-window shifts or decreases due to the aging of the catalysts. The responses of the O2 sensor outputs have delay times in both the step changes of lean-to-rich and rich-to-lean A/F. Varying delay times were observed, and it was determined that the differences depended on the mean value of the A/F of the step change. Moreover, the delay times equalized when the mean A/F was at the center of the λ-window.
X