GB2259148A

GB2259148A - Device for determining a rotational speed gradient dn/dt of an internal combustion engine

Info

Publication number: GB2259148A
Application number: GB9217684A
Authority: GB
Inventors: Vi-Minh Tu
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 1991-08-21
Filing date: 1992-08-20
Publication date: 1993-03-03
Anticipated expiration: 2012-08-20
Also published as: DE4127576C2; JPH05203658A; JP3300044B2; GB2259148B; DE4127576A1; GB9217684D0

Abstract

A disc 10 having markers 11 is coupled to engine shaft 12 and cooperates with a fixed sensor 13. The resultant square-wave signal, Fig. 2, is processed to derive the speed gradient by a calculation made initially in the inverse plane. The inverse gradient dt/dn is calculated from the quotient of a current time interval divided by the speed difference between the current and previous intervals i.e. dt DIVIDED dn = Time interval (Tn2) DIVIDED Speed difference in successive intervals (n2-n1) The desired speed gradient is found by dividing the result into a constant. <IMAGE>

Description

2259143

DESCRIPTION

A DEVICE FOR DETERMINING THE ROTATIONAL SPEED GRADIENT dn/dt OF AN INTERNAL COMBUSTION ENGINE The invention relates to a device for determining the rotational speed gradient of an internal combustion engine.

Devices for determining the rotational speed gradient dn/dt of an internal combustion engine are known. for example from DE-PS 34 21 640 in which the rotational speed is calculated from the intervals between the pulse pairs dependent upon the rotational speed and the rotational speed gradient is determined from at least two rotational speeds calculated in this way.

In this particular method the rotational speed gradient is calculated directly, and this has the disadvantage that it is costly to carry out the calculations and it requires a large sample range, moreover when carrying out the calculation using integer numbers, small rotational speed gradients or small fluctuations in the rotational speed cannot be reliably recorded.

In accordance with the present invention there is provided a device for determining the rotational speed gradient dn/dt of an internal combustion engine comprising a device for producing a pulse sequence, 1 -2dependent upon the rotational speed, and a calculating device in which the rotational speeds are calculated from the intervals of time between the pulses or between similar sides of pulses, in which the reciprocal value of the change in rotational speed is determined from the quotient an interval of time over the difference between the rotational speed value of said interval of time and the rotational speed value of a previous interval of time.

This has, in contrast to the devices known from the prior art, the advantage that it is less costly to carry out the calculations and less memory is required, which is achieved by carrying out the calculation in the inverse plane and subsequently inverting. Furthermore, it is advantageous that the evaluation is carried out by means of rotational speed values which are in any case to be calculated.

It is particularly advantageous that the inverse rotational speed gradient is calculated in such a way that a more accurate evaluation is achieved in the region of extremely small rotational speed gradients than in the region of larger gradients; carrying out the calculation in the inverse plane therefore leads to a gradient which is not linearly quantised.

The invention will be described further hereinafter, by way of example only, with reference to 1 t- F h.

-3the accompanying drawings, in which:- Fig.1 is a highly schematic view of a device for determining rotational speeds, the external construction of which is already known from, for example, DE-OS 34 23 664, but it can also be used to realise the present invention; Fig.2 is a graph, illustrating the wave voltage U14 against time t; and Fig 3 is a graph illustrating the non-linear quantisation of the rotational speed gradient.

Fig.1 illustrates a transmitting disk 10, which is attached to a shaft 12, whose rotational speed is to be determined. This shaft can, for example, be the crank or camshaft of an internal combustion engine.

A number of markers 11 are arranged on the upper surface of the transmitting disk 10. the markers 11 are equidistant apart and are of identical length. The number of markers is normally closely related to the number of the cylinders of the internal combustion engine and, for example, is equal to half the number of cylinders.

The transmitting disk 10 is scanned by a sensor 13, for example an inductive sensor. The signal produced by the sensor is then processed downstream by circuitry 14. where a square-wave voltage U14 is produced from the signal voltage of the sensor 13, whose time duration Tn is evaluated in a control unit 15.

This square wave voltage is plotted in Fig.2 as the voltage U14 over the time t, the times Tnl and Tn2 are entered as fundamental durations of time, where the duration of time Tnl is the elapsed time for the shaft 12 to rotate 180. i.e. the time which elapses between the passing of the front side of the first marker 11 and the front side of the next marker 11 on the sensor 13.

As the distance between the angular markers is known and the durations of time or rather the intervals of time Tnl and Tn2 are measured, it is possible to determine the rotational speed in a known manner as shown in equation (1).

const. const.

n = ------ and nl = ----- (1) zTn ZTnl As the number of markers 11 on the transmitting disk 10 corresponds to the number of cylinders, the number of cylinders z must appear in the denominator of equation (1).

The rotational speed gradient dn/dt can be calculated as shown in equation (2) wherein:

n2-nl dn/dt = Tn2 (2) A dn/dt = C Tnl-Tn2 (3) Tn1Tn2Tn2 In formulae (2) and (3), nl and n2 represent two rotational speeds measured consecutively and Tn1 and Tn2 represent the associated durations of time and C is a constant, which is also to be divided by the number of cylinders z.

In order to determine the rotational speed gradient from equation (3), the term (Tnl-Tn2)/TnlTI42TN2 must be calculated. As the term Tn1Tn2Tn2 is much larger than Tnl-Tn2, it is not possible to carry out the calculation using integer numbers. So that it is possible to carry out a calculation, a calculation could be made in the inverse plane, in accordance with the equation:

TnlTn2Tn2 l/dn/dt = C(Tnl-Tn2) (4) It would also be extremely costly to carry out this calculation and a large number range would be required, as the term Tn1Tn2M can be very large. In order that it would be possible to carry out such calculation, it must be ensured that after a 16-bit multiplication, only the 16 high value bits are used. Therefore extremely small rotational speed gradients would no longer be determined. If on the other hand, a rotational speed gradient is calculated in -6accordance with equation (2) in the inverse plane, this can be substantially simplified and moreover the calculation can be more accurate.

Therefore, in accordance with the invention. the reciprocal value of this rotational speed gradient is calculated in accordance with equation (5):

Tn2 l/dn/dt = ----- (5) n2-nl The value, obtained in this way, is inverted and the rotational speed gradient dn/dt which is to be calculated, is obtained directly.

It is evident from equation (5) that the larger the rotational speed gradient, the smaller the numerical result. As the rotational speeds n2 and nl are in any case calculated and therefore also available for further evaluations, it is only necessary, for calculating the gradient in the inverse plane, to store per calculation only the old rotational speeds, to subtract the two rotational speed values and to carry out a division.

As already mentioned, the rotational speeds n and the rotational speed gradient are calculated digitally in the control unit 15. For this purpose it is necessary to determine the duration of time Tn of the square-wave signal U14. In order that it is also possible to determine small gradients. the rotational 1 h speed should be quantised with a rotation per minute.

In order to achieve sufficient insensitivity to interference and that the rotational speed gradient associated with the algebraic symbols can be easily represented, a time Tn' = 1/2 Tn should be used for the calculation.

The method for calculating a rotational speed gradient is now described by reference to the following Example. With a rotational speed quantisation of 0.78 1/min. (revolutions per minute) and a Tnquantisation of 1.6 microseconds and a cylinder number of z = 4, under the condition that Tn2' = 1/2 Tn, for the inverse rotational speed gradient it results in:

21.6Tn2' l/dn/dt With the abbreviation:

Tn2' 0.781000000(n2-n1) INT = ------- (6) (7) n2-n1 it results in the rotational speed gradient dn/dt to be determined:

243750 dn/dt = -------- (8) INT and the rotational speed gradient is obtained in 1/min/sec.(revolutions per minute and per second).

This interrelation in accordance with equation -8(8) is plotted in Fig.3 and it is evident that the gradient determined in the inverse plane is not linearly quantised. Rather. the gradient is extremely finely quantised in the lower region, which enables the hunting oscillations to be particularly advantageously determined, where small rotational speed gradients must be determined.

With the larger rotational speed gradients, the quantisation is more approximate, so that it is possible to represent a greater rotational speed range, wherein, for example, it is possible to represent the recognition of switching up or down when engaging the gears.

1 11

Claims

-9CLAIMS

1. A device for determining the rotational speed gradient dn/dt of an internal combustion engine comprising a device for producing a pulse sequence, dependent upon the rotational speed, and a calculating device in which the rotational speeds are calculated from the intervals of time between the pulses or between similar sides of pulses, in which the reciprocal value of the change in rotational speed is determined from the quotient an interval of time over the difference between the rotational speed value of said interval of time and the rotational speed value of a previous interval of time.

2. A device as claimed in claim 1, in which the rotational speed n is calculated in accordance with the equation const. n = ------ zTn where const. is a constant, z is the number of cylinders in the internal combustion engine and Tn is an interval of time.

3. A device as claimed in claim 1 or 2, in which the device for producing the pulse sequence dependent upon the rotational speed comprises a transmitting disk having a number of markers on its outer surface, and wherein the markers are scanned by a sensor, whose 11 -10output signal is prepared and processed in the calculating device.

4. A device as claimed in one of the aforementioned claims, in which the markers and the gaps between the markers are of equal length in each case.

5. A device as claimed in one of claims 3 or 4, in which the number of markers is related to the number of cylinders of the internal combustion engine.

6. A device as claimed in claim 1, in which the number of markers is equal to half the number of cylinders.

7. A device for determining the rotational speed gradient of an internal combustion engine. substantially as hereinbefore described with reference to and as illustrated in the accompanying drawings.

1 9