Motorola Corporation

Gregory Steinl

Marvin L Lynch

Michael A McClish

Margaret A Selfe

Donald J Remboski

A modified trimmed mean filtering method and apparatus for a misfire detection includes acquiring a time-continuous series of acceleration data samples (601) representative of acceleration behavior of a reciprocating engine. And, providing a filtered acceleration data sample (607) dependent on an average acceleration magnitude of a portion of the acceleration data samples bounded within a range of amplitudes and, preferably, a time-centered acceleration data sample (606) of the fixed length time-continuous series of acceleration data samples. A misfire condition can be indicated (409) dependent on a magnitude of the filtered acceleration data sample.

Modified trimmed mean filtering for misfire detection

A method for detrending engine positional data includes acquiring positional encoder data over a plurality of consecutive engine revolutions as the engine is decelerating. Then, a trend (207) in the acquired positional encoder data consistent with behavior occurring at less than a frequency of one cycle per engine revolution is identified. Next, corrected positional encoder data (208) is generated dependent on removing the identified trend (207).

Detrending engine positional data for rotating position encoders

A misfire detection system and method measures fluctuations of air charge ingested into an engine 303, preferably as fluctuations of the engine's intake air pressure or as fluctuations of the mass air flow. A misfire indication 337 is provided dependent on a behavior of the fluctuations of air charge. Preferably, to eliminate errors associated with engine transient operating conditions, such as acceleration, the fluctuations of air charge are differentiated 325 before a misfire detection mechanism 333 is used to determine misfire behavior.

Growing community of inventors

Royal Oak, MI, United States of America

Gregory Steinl