Misfire and Partial Burn Detection based on Ion Current Measurement 2011-24-0142

The paper presents the implementation of a combustion diagnosis system that integrates crankshaft speed oscillations analysis with ion current signal processing, for V8 and V12 high performance engines.

Ion current sensing has been introduced in the last V8 and V12 Ferrari models in order to improve combustion control by implementing ion current based closed-loop spark-advance control systems, both under knocking and non-knocking conditions (respectively based on measured knocking level, and on ion current peak position control).

Another area where ion current sensing can improve the engine controller performance is related to the ability of detecting and isolating missing and partial burn combustions. The typical approach to misfire detection (based on engine speed oscillation measurement) is in fact particularly critical for engines with a large number of cylinders, and ion current sensing provides additional information not only about presence (or absence) of combustion, but also about the causes that generated the fault. Moreover, the paper shows that real-time analysis of specific ion current signal features allows isolating incomplete and inefficient combustion events, thus providing extremely useful information to the engine control system, which can for example be used to activate multi-spark discharge ignition mode.

The first part of the paper shows the main critical aspects of speed-measurement based misfire detection, and introduces the ion current signal main features during regular engine operation. Then, ion current signal is analyzed during abnormal combustion events: absence of combustion (both due to missing injections and missing ignitions) and partial burn cycles. It is shown how it is possible to isolate missing and incomplete combustions in a relatively straightforward way, and also how the causes that induced the fault may be isolated by integrating standard diagnostic functions with specific ion current signal processing algorithms. Finally, the performance of the diagnostic system that integrates engine speed oscillation analysis and information extracted from the ion signal has been evaluated during on-board tests, and the main results are presented at the end of the paper.