GB2343955A - Detecting tooth spacing errors of a toothed wheel for an internal combustion engine - Google Patents

Abstract

Tooth 4,6 spacing errors of a toothed wheel 1 eg crankshaft/camshaft timing wheel of an internal combustion engine are determined from fuel correction factors derived from a fuel balancing algorithm. Fuel correction factors are provided by the cyclical variations in the amount of fuel required to be added to or subtracted from the quantity required for a particular cylinder in order to correct surges in engine speed and have a period equal to the period of one full revolution of the toothed wheel. Cyclical positive and negative values of fuel correction factors (Fig.2), or factor values above a given threshold eg 50% of normal amount of fuel required, may indicate tooth spacing error. Tooth spacing error is independent of, but its magnitude varies with, engine speed. An engine control system may be disabled to ensure that the quantities of fuel supplied to the engine are independent of the output of the fuel balancing algorithm to allow tooth spacing errors to be searched for in isolation.

Description

TOOTH SPACING ERROR DETECTION METHOD This invention relates to a method for use in detecting tooth spacing errors, the invention being particularly suitable for use in the calibration or control of an internal combustion engine.

It is known to use a fuel balancing algorithm in controlling the quantity of fuel to be supplied to each cylinder or combustion space of an internal combustion engine. The algorithm may compare the engine speed after the firing stroke of consecutive cylinders have been completed, the algorithm calculating fuelling correction factors for use in controlling the quantity of fuel supplied to each cylinder to reduce the engine speed variations.

The engine speed is conveniently measured using a toothed wheel which is driven at a speed associated with engine speed, for example camshaft or crankshaft speed, the speed being measured by using a sensor to monitor the movement of the teeth past a predetermined location, and measuring the time taken for successive teeth to pass the predetermined location.

Clearly, in order to allow the algorithm to operate correctly, the engine speed must be measured accurately. One area which introduces errors into the measurement of the engine speed is manufacturing inaccuracies in the production of the toothed wheels, and in particular in the location of the teeth. For example, if the teeth are unequally spaced, then if the engine is running at a given constant speed, the measured speed will vary in a cyclical manner, the measured speed being faster than the actual speed where the teeth are closer together than they should be, and slower than the actual speed where the teeth are spaced too far apart from one another.

It is an object of the invention to provide a method whereby the presence of tooth spacing errors can be sensed.

According to a first aspect of the invention there is provided a method of sensing the presence of tooth spacing errors comprising the steps of: (a) using a fuel balancing algorithm to calculate fuel correction factors for each cylinder of an internal combustion engine; and (b) using the calculated fuel correction factors to determine whether or not tooth spacing errors are present.

Conveniently, the method inclues an initial step of disabling the engine control system in such a manner as to ensure that the quantities of fuel supplied to the engine cylinders are independent of the output of the fuel balancing algorithm. As a result, tooth spacing errors can be searched for in isolation.

The step of using the calculated fuel correction factors may comprise the step of determining whether the factors vary in a cyclical pattern, the cyclical pattern having a period equivalent to the period of one full revolution of the toothed wheel. It will be appreciated that the number of cycles of the pattern which are equivalent to one revolution of the engine will depend upon the speed of rotation of the wheel relative to engine speed.

The method is conveniently repeated over a range of engine speeds and may include the additional step of determining whether the variations in the calculated factors vary in a manner related to the engine speed.

Alternatively or additionally, the step of using the calculated factors may include the step of determining whether one or more of the fuel correction factors derived using the algorithm is greater than a predetermined level or greater than a predetermined proportion of the quantity of fuel injected.

The presence of such a cyclical output, the fuel correction factor exceeding a predetermined level, or the fuel correction factor being related to engine speed may indicate the presence of a tooth spacing error.

The invention will further be described, by way of example, with reference to the accompanying drawings, in which: Figure 1 is a diagram illustrating a rotary position and speed transducer suitable for use with a four cylinder engine; Figure 2 is a diagram illustrating the fuel balancing algorithm output for each cylinder over a range of engine speeds, indicative of the presence of a tooth spacing error; Figure 3 is a graph illustrating the output of the fuel balancing algorithm for an engine cylinder over a range of engine speeds due to a tooth spacing error of 0.5 and showing the effect of injecting an additional 1 mm3 of fuel to the cylinder ; and Figure 4 is a graph illustrating the additional quantity of fuel which would have to be supplied to the cylinder to cause the algorithm output to match that achieved where a 0.5 tooth spacing error is present as shown in Figure 3.

Figure 1 illustrates a rotary position transducer which comprises a wheel arranged to be rotated at a speed associated with the speed of operation of an associated engine, for example at the crankshaft speed or the camshaft speed. A wheel 1 arranged to be rotated at the crankshaft speed of a four cylinder engine is located adjacent a variable reluctance sensor 2 which is arranged to sense the movement of teeth 4,6 provided on the wheel 1 past a detection location of the sensor 2. The teeth 4,6 are intended to be spaced apart from one another by 180 , thus if the engine is operating at a consistent, uniform speed, then the wheel 1 will rotate at a constant speed and the time intervals between each tooth passing the sensor 2 and the successive tooth passing the sensor 2 will be equal. It will be appreciated, however, that if the teeth 4,6 are not equi-angularly spaced apart from one another, then if the engine and the wheel 1 are rotating at constant, uniform speeds, then the time intervals between the teeth passing the sensor 2 wi I I not be equal, but rather wi I I fol low a cycl ical pattern, the period of the cyclical pattern being equivalent to one full revolution of the wheel 1. As the wheel 1 is arranged to rotate at crankshaft speed, it will be appreciated that two full revolutions of the wheel 1 and hence two cycles of the pattern occur during each engine cycle.

The time intervals between the teeth 4,6 passing the sensor 2 are used in a fuel balancing algorithm which is used to calculate a fuel correction factor for use in adjusting the quantity of fuel supplied to each cylinder of an engine in order to minimise engine speed variations which occur between the firing strokes of consecutive cylinders of the engine, in use.

As mentioned hereinbefore, in order to allow the algorithm to function correctly, it is important that the engine speed inputs to the algorithm are accurate. One cause of inaccuracies in the inputs to the algorithm arise from errors in the teeth locations upon the toothed wheel. It is thus important to be able to detect the presence of such errors to permit appropriate compensation factors to be incorporated into the algorithm.

In accordance with an embodiment of the invention, the output of the algorithm is monitored in order to determine whether or not the output follows a cyclical pattern, the period of which is equivalent to one full rotation of the toothed wheel 1. Figure 2 illustrates, in simplifie form, the output of the algorithm for a four cylinder engine over a range of engine speeds. The engine speed is monitor using a transducer of the type illustrated diagrammatically in Figure 1, the wheel 1 of which rotates at crankshaft speed. It will be appreciated from Figure 2 that for a given engine speed, for example at an engine speed of 1000 rpm, the algorithm indicates that the engine speed for cylinders 1 and 3 is greater than a mean value by approximately 10 rpm and that the engine speed after the firing strokes of cylinders 2 and 4 is less than the mean value by approximately 10 rpm. It is clear from Figure 2 that the algorithm output follows a cyclical pattern, the period of the cyclical pattern being two firing strokes, and as this is equivalent to one full rotation of the toothed wheel, it is likely that the variations in the algorithm output are attributable to tooth spacing errors rather than a fundamental problem with the engine.

The determination of whether or not the algorithm output follows a cyclical pattern may be achieved using a number of techniques. For example, in the embodiment described hereinbefore, it is known that the algorithm output values for cylinders 1 and 3 should either both be positive or both be negative, the output values for cylinders 2 and 4 being the reverse sign to cylinders 1 and 3. One simple technique for determining, roughly, whether the algorithm output follows a cyclical pattern is to monitor the sign of the output values in order to determine whether or not the above described sign pattern is present. If the sign pattern is not present, then no tooth spacing error has been detected. If the sign pattern is present, then a tooth spacing error may be present.

The technique may be enhanced by comparing the magnitudes of the output values at a given speed, as these values should be substantially equal. If the magnitudes of the values for cylinders 1 and 3 are substantially equal to one another and those for cylinders 2 and 4 are substantially equal to one another, and if the sign pattern described hereinbefore is present, then it is likely that a tooth spacing error is present.

As shown in Figure 2, the cyclical characteristic of the algorithm output is independent of engine speed, the magnitude of the algorithm output varying depending upon engine speed and at a rate proportional to the change in engine speed, the period of the cyclical pattern remaining equivalent to one revolution of the wheel. This is another characteristic of a tooth spacing error.

Once it has been determined that a tooth spacing error is present, the presence of such an error can be flagged, and the magnitude of the tooth spacing error can be calculated, using an appropriate technique for example as described in EP 0769134, and used in the fuel balancing algorithm to compensate for the tooth spacing error.

In practice, it is envisaged that whilst the method of determining whether or not a tooth spacing error is present is being used, although the output of the fuel balancing algorithm is used to determine whether or not a tooth spacing error is present, the fuel balancing process should be disabled so that the quantity of fuel delivered is independent of the output of the algorithm, thus allowing the tooth spacing error to be searched for in isolation.

Although as described hereinbefore the method of determining whether or not a tooth spacing error is present commences with determining whether or not the output of the fuel balancing algorithm is cyclical, and following this with determining whether or not the output is related to engine speed, it will be appreciated that, if desired, either step could be omitted or the two steps of the method could be performed in the reverse order.

In accordance with a further embodiment of the invention, the presence of a tooth spacing error may be determined by using the output of the fuel balancing algorithm to determine a fuel correction factor i. e., a quantity of fuel which should be added to or subtracted from the quantity of fuel next injected to a particular cylinder to correct for engine speed surges or decelerations between consecutive firing strokes, and determining whether or not the correction factor falls within a predetermined range.

Figure 3 illustrates the output of the fuel balancing algorithm or the fuel correction factor calculated where a tooth spacing error of 0.5 is present, over a range of engine speeds, and also illustrates the effect on the output of the fuel balancing algorithm of supplying 1 mm3 of additional fuel to a cylinder relative to that delivered during a previous injection. It is clear from Figure 3 that the presence of a tooth spacing error can have a much greater effect upon the output of the fuel balancing algorithm than an error in the quantity of fuel injected, particularly at high engine speeds. Figure 4 illustrates the quantity of additional fuel which would have to be supplied to a cylinder in order to generate an algorithm output equivalent to that shown in Figure 3 due to the presence of a 0.5 tooth spacing error.

The quantities of additional fuel necessary to cause an effect equivalent to a 0.5 tooth spacing error are very large compare to the quantity of fuel normally delivered to a cylinder. It is very unlikely that such a large additional quantity of fuel would be required to compensate for engine speed variations between cylinders, and so if the algorithm indicates that such a large quantity of additional fuel should be supplied to a cylinder, then it is assumed that a tooth spacing error is present, the tooth spacing error giving rise to inaccurate speed measurements which are causing the algorithm to produce such high fuel correction factors.

In practice, the quantity of fuel supplied to each cylinder for each firing stroke depends upon the capacity of each cylinder, and the engine speed and load. For example, for a turbocharged engine of approximately 0.5 t/cylinder capacity, the fuel delivery level varies between about 5 and 50 mm3 for each firing stroke, depending upon the engine speed or load.

If the output of the fuel balancing algorithm indicates that the additional quantity of fuel which must be supplied to the cylinder in order to balance the engine speeds of consecutive cylinders falls outside of a predetermined range, for example 50% of the intended fuelling level for a given engine speed or load, then this may be used to flag that a tooth spacing error may be present. Thus, in this example, if the fuel correction factor exceeds 2.5 mm3 at idling or 25 mm3 at full output, it is assumed that a tooth spacing error is present. It will be appreciated that the 50% threshold is an arbitrarily chosen value, and that in practice, an alternative value may be chosen.

Although in the description hereinbefore a 50% threshold is used to determine whether or not a tooth spacing error is present, it will be appreciated that other thresholds may be used. The threshold may, if desired, vary with engine speed and load conditions, or may be a fixed volume, for example 25 mm3 The method described hereinbefore may be used in conjunction with the method of determining whether or not the algorithm output follows a cyclical pattern and whether the output varies in a manner rotated to the engine speed.

Although the methods are described hereinbefore with reference to a transducer including a wheel having two teeth, it will be appreciated that the invention is also applicable to arrangements in which the wheel has a greater number of teeth-.' ! The number of teeth used is conveniently related to the number of cylinders of the engine, the speed of rotation of the wheel relative to engine speed, and the nature of the fuel balancing algorithm.

Claims (9)

1. CLAIMS 1. A method of sensing the presence of tooth spacing errors comprising the steps of : (a) using a fuel balancing algorithm to calculate fuel correction factors for each cylinder of an internal combustion engine; and (b) using the calculated fuel correction factors to determine whether or not tooth spacing errors are present.
2. 2. The method as claimed in Claim 1, including an initial step of disabling the engine control system in such a manner as to ensure that the quantities of fuel supplied to the engine cylinders are independent of the output of the fuel balancing algorithm.
3. 3. The method as claimed in Claim 1 or Claim 2, wherein the step of using the calculated fuel correction factors comprises the step of determining whether the fuel correction factors vary in a cyclical pattern, the cyclical pattern having a period equivalent to the period of one full revolution of a toothed wheel for which the presence or not of tooth spacing errors is to be sensed.
4. 4. The method as claimed in any of Claims 1 to 3, wherein the method is repeated over a range of engine speeds.
5. 5. The method as claimed in Claim 4, and including the additional step of determining whether the variations in the calculated factors vary in a manner related to the engine speed.
6. 6. The method as claimed in any of Claims 1 to 5, wherein the step of using the calculated factors includes the step of determining whether one or more of the fuel correction factors derived using the algorithm is greater than a predetermined level or greater than a predetermined proportion of the quantity of fuel injected.
7. 7. The method as claimed in Claim 6, and including the step of varying the predetermined level or the predetermined proportion of the quantity of fuel injected with engine speed.
8. 8. The method as claimed in Claim 6, and including the step of varying the predetermined level or the predetermined proportion of the quantity of fuel injected with engine load conditions.
9. 9. A method of sensing the presence of tooth spacing errors substantially as hereinbefore described with reference to the accompanying figures.