GB2453573A

GB2453573A - Monitoring cylinder performance of an engine by measuring output torque from the crankshaft

Info

Publication number: GB2453573A
Application number: GB0719888A
Authority: GB
Inventors: Raymond David Lohr
Original assignee: Transense Technologies PLC
Current assignee: Transense Technologies PLC
Priority date: 2007-10-11
Filing date: 2007-10-11
Publication date: 2009-04-15
Also published as: GB0719888D0

Abstract

A method of monitoring the cylinder performance in an internal combustion engine comprises measuring the output torque from the crankshaft of the engine at intervals during the engine cycle, and comparing readings for indications of changes in performance of the engine. Readings may be averaged over an integer number of engine cycles to provide an average torque output for the engine, or the torque at a particular point during or over a particular part of the engine cycle may be measured and averaged over a number of cycles to provide an average torque output for a particular cylinder of the engine. Torque characteristics measured may include peak torque, instantaneous torque versus angle, and average torque during the firing stroke; and these may be analysed to find the firing point and torque generating period of each cylinder. The sampling interval may be 0.8 to 8 KHz and measurements may be taken by a surface acoustic wave (SAW) sensor attached to an engine flexplate.

Description

I

Method for Monitoring Cylinder Performance in Internal Combustion Engines The present invention relates to methods for monitoring the overall and in-cylinder performance of internal combustion engines.

Modern internal combustion (ic) engines are controlled via an engine control unit (ECU).

The control of a specific engine type is based on the dynamometer testing of representative sample engines, prior to production, which establishes the multi-dimensional map of target air flow, fuel flow and ignition advance for different engine rpm and throttle settings over the operating range of ambient temperature and pressure.

As emissions legislation gets ever tighter, especially in the US, Europe and Japan, various means for improved engine control are being utilised. One technique involves the real time measurement and feedback of in-cylinder pressure for combustion control by placement of a pressure sensor directly in each cylinder, which communicates back to a central processor. This approach is technically promising but very expensive since each cylinder requires its own pressure sensor which must survive the harsh in-cylinder pressure and temperature conditions over the operating life of the vehicle -typically 10 years or 100,000 miles.

According to the present invention there is provided a method of monitoring the cylinder performance in an internal combustion engine comprising measuring the torque from the crankshaft of the engine at defined intervals, and comparing readings for indications of changes in performance of the engine.

A method in accordance with the invention has the advantage that it enables effective monitoring of cylinder performance at a fraction of the cost of prior art approaches due to the removal of the need for expensive sensors.

The output torque measured at any instant is the torque contribution of the whole engine, i.e. from all the cylinders, at an instant in time. In one embodiment, the output torque is measured at sampling intervals and then an average calculated over a defined period, for example an integer number of engine cycles, so that all cylinders will have fired the same number of times. In a four stroke engine, this might be 2 crank shaft revolutions (720 degrees) or an integer multiple thereof. This has the advantage that compensation is made for the fact that individual cylinders are not identical and the power output will not, therefore, be the same from each cylinder. This approach then provides information on the overall performance of all the cylinders but not on individual ones, so that it will identify a drop in performance but will not identify if this is due to a particular cylinder.

In a preferred embodiment, however, the measurements taken for each engine cycle are used to identify the firing point and subsequent torque generating period of each cylinder during the cycle (that is torque measurements may be taken at a point during or over a particular part of the engine cycle), from which information can be derived regarding the torque output of each cylinder. This may optionally be averaged over a number of engine cycles, for example by taking the average of corresponding torque peaks in each cycle corresponding to a particular cylinder so as to provide an average torque output for each cylinder. This has the advantage of providing information not only of the general engine performance but also on the performance of individual cylinders.

Just a single measurement can be taken for each cylinder during the cycle so as to get just a single peak torque measurement for the firing of each cylinder. For example, in the case of a four cylinder engine, four measurements can be taken at even intervals in order to get the torque in the output shaft corresponding to peak drive from each cylinder.

Preferably, however, sampling intervals typically in the range of 0.8 to 8 KHz are used, both for monitoring individual cylinder performance as well as general engine performance.

S

Accordingly, the method of the present invention is able to determine not only the average torque output of the engine but also the average torque contribution from individual cylinders. It offers the potential benefit of individual in-cylinder pressure sensing but at much lower cost.

Average engine output torque monitoring provides an improved signal for determining the optimum change point for automatic transmission gear ratio selection. It also will enable or assist the optimisation of overall engine fuelling and ignition advance to benefit efficiency and reduce emissions.

Individual cylinder torque monitoring will enable comparison of the torque contributions from each cylinder, which should in turn enable optimisation of fuelling and ignition timing on an individual cylinder basis, where this is possible. This approach should benefit combustion efficiency and thereby provide overall improvements in fuel Consumption and C02 emissions. In addition, individual cylinder torque monitoring should assist in the detection of misfire and knocking, both undesirable occurrences which can increase emissions, reduce engine running smoothness and even damage the engine permanently.

Torque measurements are preferably taken by means of a sensor such as a SAW based sensor attached to an engine flexplate situated between the crankshaft and the torque converter in an automatic transmission equipped vehicle.

In order that the invention may be well understood, there will now be described some embodiments thereof, given by way of example.

In a first embodiment of the invention, engine torque in a four cylinder four stroke engine is monitored using a RF SAW based sensor which is attached to an axial face of an engine flexplate, in particular in a recess in a radially extending torque transmitting rib, the sensor optimally being located on or proximate to the neutral axis of the flexplate. A sampling interval in the range 0.8 to 8KHz is used to obtain instantaneous measurements of the torque output of the engine. Because the outputs from individual cylinders are not identical, the measurements are averaged over some period, in particular 2 crankshaft rotations (720°) or some multiple (n) thereof, since in this period all the cylinders will have fired once or n' times precisely. The resulting average reading can then be used to provide information on the performance of the engine.

In an alternative embodiment of the method of the invention, engine torque in a four cylinder four stroke engine is again monitored using a RF SAW based sensor which is attached to an axial face of an engine flexplate, in particular in a recess in a radially extending torque transmitting rib, the sensor optimally being located on or proximate to the neutral axis of the flexplate. A sampling interval in the range 0.8 to 8KHz is used to obtain instantaneous measurements of the torque output of the engine. When each cylinder fires in an engine it generates a torque pulse between top dead centre (tdc) and bottom dead centre (bdc). it is therefore possible by analysis of the torque measurements taken over a cycle, e.g. counting pulses and/or by referring to a crank and/or cam sensor (typically fitted as standard to ic engines) either to identify which torque pulse relates to which specific cylinder or to time measurements to coincide with peak torque output for each cylinder. In this way the torque output for each cylinder is monitored.

The torque pulses for each individual cylinder is then averaged over multiple (m) firings, so as to determine the mean peak torque, average torque and average shape of the torque versus crank angle characteristic. A moving average algorithm can beneficially be used in this context. This information can then be used to compare the torque contributions from the individual cylinders and adjust the fuelling and ignition timing for the individual cylinders to improve combustion efficiency and engine smoothness. It should also enable or assist detection of misfiring and knocking within individual cylinders.

The number (m) of firings averaged generally depends on the engine rpm and the torque sampling rate. Sampling strategies may be time based, ie x' samples per second or linked to crankshaft angle, ie y' samples per crankshaft revolution.

It may be especially useful to capture the torque versus crank angle characteristic over the 1800 firing stroke, since differences in this characteristic between individual cylinders should enable or assist in the identification of misfire or engine knocking.

Claims

Claims 1. A method of monitoring the cylinder performance in an internal combustion engine comprising the step of measuring the instantaneous output torque from the crankshaft of the engine at intervals during the engine cycle, and comparing readings for indications of changes in performance of the engine.
2. A method according to claim 1, comprising the further step of averaging the instantaneous torque readings over one engine cycle (2 complete revolutions for a 4 stroke engine) in order to compute the average torque output of the engine.
3. A method according to claim 2, comprising the further step of averaging over an integer number of engine cycles the average torque measured during each engine cycle, adding the newest value and dropping the oldest value, so as to produce a smoothed moving average engine torque output signal.
4. A method according to claim 1, comprising the further step of identifying the torque output measurement corresponding to the interval during which each cylinder is firing and using the measurements corresponding to each cylinder to monitor the performance of each cylinder individually.
5. A method according to claim 4, comprising the further step of averaging the instantaneous torque measurements corresponding to each cylinder over a number of engine cycles in order to obtain and compare average torque characteristics for each cylinder.
6. A method according to claim 5, where the average torque characteristics may include (but not exclusively): peak torque, instantaneous torque versus angle, and average torque during the firing stroke. i1
7. A method according to claim 4 or claim 5, wherein the measurements are analysed to identify the firing point and subsequent torque generating period of each cylinder during the cycle.
8. A method according to any of the preceding claims, wherein the measurement sampling interval is in the range of 0.8 to 8 KHz
9. A method according to any of the preceding claims, wherein torque measurements are taken by means of one or two torque sensors attached to an engine flexplate in the output power train of the engine.
10. A method according to any of the preceding claims, wherein the torque sensor is a non-contacting Surface Acoustic Wave (SAW) sensor comprising a one port SAW resonator.
11. A method substantially as herein described.