Engine misfire detection system

• US 7,031,828 B1
• Filed: 08/28/2003
• Issued: 04/18/2006
• Est. Priority Date: 08/28/2003
• Status: Active Grant

• Current Assignee: Thompson Automotive Labs, LLC
• Original Assignee: John M. Thompson
• Inventors: Thompson, John M., Auer, Kenneth W.
• Primary Examiner(s): Wolfe, Willis R.
• Assistant Examiner(s): Hoang, Johnny H.
• Application Number:
  US10/651,054
• Time in Patent Office: 964 Days
• Field of Search: 701/114, 701/100, 701/101, 701/109, 123/673
• US Class Current: 701/114
• CPC Class Codes: F02D 2200/1015 Engines misfires
• CPC Class Codes: F02D 41/008 Controlling each cylinder individually
• CPC Class Codes: F02D 41/1443 with one sensor per cylinder or group of cylinders
• CPC Class Codes: F02D 41/1456 with sensor output signal being linear or quasi-linear with the concentration of oxygen
• CPC Class Codes: F02D 41/1459 the characteristics being a hydrocarbon content or concentration
• CPC Class Codes: G01M 15/11 by detecting misfire

Abstract

An engine misfire detection system includes a database of engine fingerprints, each fingerprint corresponding to a known misfire condition. A technician uses software to compare the fingerprints in the database to an engine with a combustion inefficiency to determine which cylinder in the engine has the combustion inefficiency. The fingerprints are generated by evaluating an output from a lambda sensor and a timing reference. An alternate embodiment associates oxygen sensors with each of the cylinders in the engine and infers combustion inefficiency when a given sensor detects a peak in the amount of oxygen.

Claims

1. A method of detecting combustion inefficiency in an engine having multiple cylinders, comprising:

detecting oxygen levels in an exhaust stream with a sensor;

connecting the sensor to an external probe;

comparing sensor readings to a database of fingerprints associated with the external probe, wherein the database of fingerprints comprises fingerprints of engines having combustion inefficiencies; and

providing an indication as to which cylinder has the combustion inefficiency.

2. The method of claim 1 wherein detecting the oxygen levels with the sensor comprises using a lambda sensor.

3. The method of claim 2 wherein using the lambda sensor comprises using a lambda sensor positioned in an exhaust manifold.

4. The method of claim 2 wherein using the lambda sensor comprises using a lambda sensor positioned proximate a catalytic converter.

5. The method of claim 1 wherein comparing sensor readings to the database of fingerprints comprises linking a peak in the oxygen levels to a particular cylinder in the engine so as to identify a misfire in the particular cylinder.

6. The method of claim 1, wherein the database of fingerprints is derived empirically.

7. The method of claim 1 wherein detecting the peak in the oxygen level comprises directly detecting the oxygen levels.

8. The method of claim 1 wherein detecting the peak in the oxygen levels comprises inferentially detecting the oxygen levels.

9. The method of claim 1 further comprising generating a timing reference associated with the engine.

10. The method of claim 9 further comprising linking the timing reference to the database of fingerprints.

11. The method of claim 6 wherein the database of fingerprints has different fingerprints for different types of engines.

12. The method of claim 6 wherein the database of fingerprints is derived empirically by introducing known combustion inefficiencies and recording oxygen levels for the known combustion inefficiencies.

13. The method of claim 1 further comprising using a known firing order of the multiple cylinders to assist in determining which cylinder released exhaust gases detected by the sensor.

14. A computer readable medium having software stored thereon, said software adapted to detect combustion inefficiency in an engine having multiple cylinders by:

accepting data from a sensor through an external probe;

comparing sensed oxygen levels from the sensor coupled with a timing reference to a database of fingerprints, wherein the database of fingerprints comprises fingerprints of engines having combustion inefficiencies; and

providing an indication as to which cylinder has the combustion inefficiency.

15. The computer readable medium of claim 14 wherein said software accepting data from the sensor through the external probe is adapted to operate with an external probe connected to a lambda sensor and a timing reference generator.

16. The computer readable medium of claim 14 wherein said software is adapted to identify that the cylinder is misfiring.

17. The computer readable medium of claim 14 wherein said software is adapted to generate the timing reference for engines selected from a group consisting of those using a distributorless ignition system (DIS) and those using a distributor system through the calculation of an offset.

18. The computer readable medium of claim 14 wherein said software is adapted to accept data from the sensor through the external probe is adapted to work with a sensor that senses oxygen levels indirectly.

19. The computer readable medium of claim 14 wherein the database of fingerprints comprises a database of fingerprints relating to empirically derived oxygen levels.

20. The computer readable medium of claim 19 wherein the empirically derived oxygen levels are derived from introducing a known combustion inefficiency into a normally operating engine and wherein said fingerprints comprise data related to oxygen levels detected in an exhaust path associated with the normally operating engine.

21. A method of detecting a misfire in an engine having multiple cylinders comprising:

connecting an external probe to a sensor associated with an exhaust path of the engine;

collecting readings from the sensor relating to oxygen levels within the exhaust path;

comparing, through software, the oxygen levels collected from the sensor to a database of fingerprints; and

providing an indication as to which of the multiple cylinders is misfiring.

22. The method of claim 21 further comprising forming the database of fingerprints empirically.

23. The method of claim 22 wherein forming the database empirically comprises, for a given model of engine, sequentially introducing a known misfire into each cylinder of the given model of engine and recording oxygen levels associated with each introduced known misfire for each cylinder and associating each recording of oxygen levels with a timing reference.

24. The method of claim 23 wherein comparing, through software, the oxygen levels collected from the sensor to the database comprises identifying to the software a type of engine being tested and comparing the type of engine to a corresponding type of engine in the database of fingerprints.

25. The method of claim 23 further comprising identifying a cylinder within the engine that has a combustion inefficiency short of a misfire instead of providing the indication as to which of the multiple cylinders is misfiring.