Characterization of an Electrical Sensor for Combustion Diagnostics

University dissertation from Combustion Physics, Lund Institute of Technology, P.O. Box 118, SE-22100 Lund, Sweden

Abstract: The ionization sensor is an electrical probe for diagnostics in internal combustion engines. The combustion process affects the electrical properties of the gas in the cylinder. Thus the sensor signal contains copious information about the conditions in the combustion chamber. A thorough characterization of the sensor makes it possible to take advantage of a larger portion of this information for feedback control of the engine. The present work focuses on the identification of the basic mechanisms governing the functioning of the ionization sensor and their interaction. Optical diagnostics, equilibrium analysis and an elementary model have been employed to characterize the sensor. It was found that the contact between flame front and cathode as well as the mixture composition along the main current path governs the sensor signal during early combustion. On the basis of these findings, a zone-based model for the sensor was suggested. Imaging of the flame propagation revealed that turbulence distorts the shape of the first current peak by affecting the contact between the cathode and the wrinkled flame front. Experimental data and an analysis of the ionization equilibrium in the post-flame gas showed that traces of alkali metals in the atmosphere make a major contribution to thermal ionization at temperatures characteristic of the combustion of diluted mixtures. An investigation of the relationship between in-cylinder pressure and ionization sensor signal under various gas flow conditions indicated that high gas flow impairs this relationship. Imaging of nitric oxide and hydroxyl radicals in the post-flame gas supplied experimental evidence that the flow of cold, possibly unburned gas from the edge of the combustion chamber to the region of the electrode gap can explain this impaired relationship. The knowledge obtained will hopefully help to improve algorithms to derive information from the sensor signal and to use this information to monitor and optimize the combustion process.

  This dissertation MIGHT be available in PDF-format. Check this page to see if it is available for download.