An Investigation of the Effect of Thermal Stratification on HCCI Combustion by using Rapid Compression Machine 2007-01-1870

A significant drawback to HCCI engines is the knocking caused by rapid increases in pressure. Such knocking limits the capacity for high-load operation. To solve this problem, thermal stratification in the combustion chamber has been suggested as possible solution. Thermal stratification has the potential to reduce the maximum value of the rate of pressure increase combustion by affecting the local combustion start time and extending the duration of combustion. The purpose of this study was to experimentally obtain fundamental knowledge about the effect of thermal stratification on the HCCI combustion process. Experiments were conducted in a rapid compression machine (RCM) equipped with a quartz window to provide optical access to the combustion chamber. The machine was fueled with DME, n-Butane, n-Heptane and iso-Octane, all of which are currently being investigated as alternative fuels and have different low temperature characteristics.

The two types of vertical direction thermal stratification used in this study were created using a buoyancy effect. The widths of the large and small thermal distributions were approximately 42K and 28K, respectively. Experiments were carried out using these two thermal distributions as initial conditions. Histories of in-cylinder gas pressure and piston lift were recorded to analyze the combustion characteristics. To investigate the local reaction of combustion, chemiluminescence images were obtained with a framing streak camera. Lastly, experimental results were compared with the computational results calculated with a multi-zone model.

The experimental results showed that each of the fuels tested had unique start temperatures at both the low and high temperature conditions, regardless of temperature gradient. It was also found that the combustion duration became shorter and the rate of pressure increase became smaller with the larger thermal distribution. Moreover, the experimental results corresponded to the results obtained by calculation. Chemiluminescence images indicated that with a small thermal distribution, luminescence appeared almost equally throughout the combustion chamber. Conversely, with large thermal distribution, luminescence occurred first in the upper part and secondly throughout the whole chamber. The time between the appearance of chemiluminescence in the upper and bottom parts of the combustion chamber was approximately 1.5ms, which almost corresponded to the time difference for the start of the high temperature reaction between the upper part and bottom parts, and was the result of a calculation performed for the two zones.