JPH0618284A - Method of evaluating output signal of revolution-number sensor - Google Patents

Abstract

PURPOSE: To identify a reference mark with high reliability even when there are many markings by setting a large time interval as a numerator, setting a small time interval as a denominator, forming a quotient, and comparing the quotient with a threshold value. CONSTITUTION: An incremental disk 10 having many marks 11 of the same type on the surface is fixedly mounted at a shaft 12. The marks 11 are all disposed at the same interval. In the case of a cam shaft, reference marks are formed at synchronous defects 13a, 13b. When the disk 10 is rotated, it is scanned by a number-of-revolutions sensor 14, its output signal is formed in a rectangular pulse by molding stage 15, and evaluated by a calculator 16. A pulse sequence U15 is plotted with respect to a time (t). The times t and t have the same lengths, and the lapse time t in the case of passing the reference mark via the sensor 14 is longer than it. Large time interval is always used as a numerator, the small time interval is used as a denominator to form a quotient, and which is compared with a threshold value.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for evaluating the output signal of a rotational speed sensor, which scans a rotary element having a plurality of markings of the same kind and at least one distinguishable marking and outputs the output signal. Then, a pulse sequence depending on the rotation speed is formed from the output signal, the time interval between the pulses or the time interval between the same kind of edges of the pulse is obtained by the computer, and each quotient of the two time intervals is formed.
It relates to a method for evaluating the output signal of a speed sensor, which is used for comparison in order to identify distinguishable markings.

[0002]

Methods and devices for evaluating the output signal of a speed sensor are already known, for example in connection with the evaluation of speed sensors in motor vehicles. At this time, for example, the disc is attached and fixed to the shaft whose rotational speed is to be detected. The surface of this disc has various markings. The disk is scanned by a sensor and its output signal is processed into a pulse sequence which depends on the speed of rotation.

The number of revolutions is determined by comparing the time intervals of the individual pulses. If the disc has one distinguishable marking in addition to a plurality of similar markings, it is also possible to detect fiducial marks that can unambiguously set the position of the shaft.

A method of this kind which makes it possible to measure the rotational speed and to identify the reference marks is described, for example, in the publication Electroni.
quee Applications, 12 December 1982
From January to January 1983, No. 27, Amsterdam,
Netherlands, "Allumage electroniq"
ue et microprocesure ", the disk to be scanned has, in addition to a plurality of notches which are equally spaced from one another, additional notches which have a relatively short spacing. Used as a mark.

In order to increase the reliability of the reference mark identification, this known method identifies the reference mark when the time interval between two pulses characteristically differs from the preceding time interval. Here, “characteristically different” means that the time intervals can be clearly distinguished even at the time of maximum acceleration.

For evaluation purposes, two consecutive successive time intervals are converted into a ratio on the one hand when the rotational speed is constant and on the other hand at various maximum possible accelerations due to the system. In doing so, consider the geometrical configuration of the notch or fiducial mark.

The known method for disks having a total of eight markings on their surface has the disadvantage that it cannot be easily applied in the case of a disk having a markedly large number of markings. This is because the time relationship is relatively disadvantageous in that case.

Another drawback of this known method is that erroneous fiducial mark identification can occur if the sensor malfunctions or if a fault occurs. Furthermore, the lack of any sensor monitoring function is a serious drawback.

[0009]

SUMMARY OF THE INVENTION An object of the present invention is to construct a reference mark which is more reliable even when there are a large number of markings such as an increment disc.

[0010]

According to the present invention, the above object is to form a quotient from each of two time intervals, compare this quotient with a limit value, and from the comparison the quotient differs from the limit value in a predetermined manner. In some cases, a fiducial mark can be identified and configured so as to be solved.

Since the quotient is always formed with the same time interval, preferably a relatively large time interval, always being the numerator and a relatively small time interval being the denominator, a unique fiducial mark identification is obtained. One limit is sufficient to be feasible.

Particularly advantageously, the method according to the invention allows monitoring of the speed sensor for static and dynamic validity and complete failure. This is done by comparing the quotient formed in each case from the two time intervals with another limit value. Error identification is possible by exceeding or falling below this further limit value.

Further advantages of the invention result from the dependent claims.

[0014]

1 shows an apparatus suitable for carrying out the method of the present invention. Incremental disc 10 having a large number of similar marks 11 on its surface
It is attached and fixed to 2. The number of rotations of this shaft is detected. This shaft may be, for example, a camshaft or a crankshaft of an internal combustion engine (not shown).

Mark 11 on the increment disk 10
Are all spaced at the same distance. When the shaft 12 is a cam shaft, a reference mark is provided for each cylinder of the internal combustion engine. In the embodiment, the reference mark is formed as a mark lacking portion, and is hereinafter referred to as a synchronous notch portion 13a, 13b.

The structure of the increment disk is usually the same in the length of the mark and the length of the intermediate space, and one increment is associated with the angle (camshaft angle). That is, the angle between two consecutive marks 11 is 3 ° NW.

The synchronous notch has a spacing of 12 ° NW between the two trailing edges of the mark. The angle between the two synchronizing cutouts is called a segment. The segment is 180 ° NW in the case of the illustrated embodiment with two synchronous notches
Is.

For example, the increment disk 10 rotating in the direction of the arrow is scanned by a rotation speed sensor 14, for example, an inductive sensor. The rotational speed sensor delivers an output signal which depends on the rotational speed. This output signal is shaped in the pulse shaping stage 15 into a rectangular pulse, as is known, and is evaluated again in the calculator 16.

The pulse sequence U15 to be evaluated is shown in FIG.
At time t. The time intervals to be evaluated are from the trailing edge of the pulse sequence U15 to the respective trailing edge. After this, the edge triggers the interrupt signal INT.

The illustration of FIG. 2 is applicable when the number of revolutions is constant, and the times t1 and t3 have the same length. The time t2 that elapses when the reference mark passes through the rotation speed sensor 14 is significantly longer than that. In the example of FIG. 2, it is twice as long. If two marks are missing, the time will be four times longer.

Block 17 is shown in FIG. This block is a component of the computing device 16 in which preprocessing and monitoring of the rotational speed sensor 14 is performed. Here, the signal of the rotational speed sensor 14 is supplied via the input 18 and the interrupt signal INT reaches the block 17 via the same input.

On the output side 19 of the block 17, the transit time tm for the increment is output. The transit time is used in block 20, for example, as a rotational speed basis for the angle extrapolation method and in block 21 as a rotational speed basis for the pump characteristic field calculation. In this case, the block 21 determines the number of revolutions averaged over the length of one segment. Therefore, the signal tj is output to the output side of the block 20, and the rotation speed signal nNW is output to the output side of the block 21. The above embodiments relate to known EDC devices.

The other output 22 of the block 17 is connected to an increment counter 23, and the output 24 outputs the identified state of the speed sensor.

From the output side 25 of the block 17 the connection path A
An interrupt signal is supplied to the AND block, which leads to the ND block 26. AN by AND connection
On the output side of the D block 26, a synchronizing pulse, the shaft 12
If is a camshaft, a NW sync pulse is generated.

FIG. 4 shows a detailed block circuit diagram for signal processing. In this device configuration, which is a component of a calculation device (not shown), the rotation speed sensor signal is the first evaluation circuit 2
7 is supplied. A further processed pulse sequence is generated at the output of the evaluation circuit 27. This pulse sequence reaches the counter 28. The output side of the counter is a block 29 for detecting the instantaneous rotation speed, the average rotation speed nN.
It is connected to a block 30 for forming W, a block 31 for monitoring the maximum time and a block 32 for performing dynamic plausibility checking.

In block 33, defect identification is performed which outputs an error condition to output 34. Another block 35 operates as an increment counter. Block 36
A static plausibility check is performed at and a synchronization pulse is formed at block 37. At that time, the block 37 is further supplied with a signal from the threshold switch 38.

Description of the signal processing In the computing device 16, the pulse sequence U15 or the associated interrupt pulse INT forms the time interval required for another evaluation. Hereinafter, this time interval will be referred to as increment passing times tm1 to tmn. The detection of the instantaneous speed is indicated in blocks 20 to 29 and takes place in a known manner.

The average rotation speed of the camshaft is obtained by totaling the increment passage times belonging to one segment and obtaining the average rotation speed from the sum. The averaging is performed over one segment as already described. The end of a segment is identified when the increment counter 23, which counts the increment, counts a number corresponding to the segment length.

In order to perform fiducial mark identification or notch identification and dynamic plausibility checking, blocks 17 to 27 form the quotient of two immediately preceding successive increment transit times. At that time, a relatively large increment passage time tmg is set as a numerator and a relatively small increment passage time tmk is set as a denominator. Therefore, the following equation applies.

Tr, j = tmg / tmk The quotient tr, j thus determined is compared with a previously determined threshold. Therefore, depending on the threshold value or depending on the comparison result, a notch identification or a guess about the validity is derived. What is important is that the same type of time always stands at the same point in the above formula.

The comparison is made by the calculation device 16 or block 31,
One of 32 and 37 passes. Thresholds S1, S2, S3
And optionally a further threshold value is stored in the computing device 16 or is formed by a threshold switch 38 and is supplied to the computing device 16 therefrom.

The quotient is formed in this way. This reduces the number of thresholds for later evaluation. The threshold is usually set below this threshold even during maximum camshaft acceleration. The under-detection will be described in detail later.

The relationship between the value of the quotient tr, j and the possible identifications is shown in the following table.

Identification tr, j tmg tmk tr, j <1 Program error 1 <tr, j <S1 increment increment S1 <tr, j <S2 Impossible S2 <tr, j <S3 notch increment or increment notch S3 <Tr, j Impossible Here, S1, S2, and S3 are threshold values, tr and j are quotients of two passage times, and the passage time is an increment passage time or a notch passage time.

As can be seen from the above table, there are five different cases for the quotient, program error, transition from increment to increment, invalid state, and notch to increment or increment to notch. Transitions are identified.

The values listed in the following table are used as typical values for the thresholds S1, S2, S3.

Threshold value nNW <400 min to ¹ nNW> 400 min to ¹ S1 2.4 1.4 S2 2.4 3.3 S3 8.5 4.7 The setting of the threshold value is selected as follows. That is, the maximum possible camshaft acceleration is selected so that it cannot lead to error identification. In doing so, the following assumptions for the acceleration of the camshaft should be considered.

For the order of notch-increment or increment-notch, b = + /-15000 m.
An acceleration of in ~ ¹ s ~ ¹ is assumed. This corresponds to the superposition of the maximum average engine acceleration and the rotational uniformity on the camshaft.

Increment-The increment order is evaluated experimentally. The following values were obtained during the individual measurements.

NNW / min ~ ¹ 500 1000 2200 b> 0: tr, j 1.22 1.24 1.23 1.33 b> 0: tr, j 1.15 1.27 1.27 1.30 400min For camshaft speeds below ~ ¹ ,
b = 160000 min ~ ¹ s ~ ¹ is calculated by the acceleration. For b <0, the limit value is tr, j≈2.4. This corresponds to the stationary state of the engine after the camshaft angle of 6 °.

If the increment transit time is dynamically identified as invalid during the evaluation, two cases are distinguished.

1. If no further statically valid segments are identified after the notch identification or after a complete failure of the camshaft speed sensor, the state "dynamically invalid" is set and the next increment The increment transit time for is started when it is zero. This allows for a dynamically invalid increment transit time after a dynamically invalid increment, or when tr, j> tr, max (which especially occurs during the first or second detected increment). Later, the increment transit time is prevented from continuing to the counter overflow. In this case, tr, j can only be dynamically within a valid region after a counter overflow if the increment transit time has elapsed.

2. Once the statically valid segment has been identified at least once, the state "dynamically invalid" is set and the increment transit time is further counted until the next dynamically valid increment transit time. this is,
It occurs even if the camshaft speed sensor is temporarily or eventually identified as a failure. By this means very few disturbing pulses occur.

After initializing the microcomputer (reset at the time of making connection), the increment passing time is selected as follows. That is, the transit time is selected to be valid within a relatively large area for the first detected increment. This largely avoids dynamic errors on the first increment.

Another validation is done using an increment counter and static validation. At that time, the increment counters 23 and 25 are set to zero when the reference mark or the synchronous cutouts 13 and 13b are identified. The increment counter increments by one each time the speed pulse is dynamically valid. If no synchronous notch is identified, then static validity monitoring is also performed.

Static validity identification is performed by monitoring increment counters 23, 25 in association with pulse identification from dynamic validity. The following table shows this relationship.

Identification from Dynamic Validity If it is not statically valid Other error response tm, j-1 tm, j increment increment j ≧ jmax increment counter reset increment notch j <jmax increment counter reset where tm, j -1 is the immediately preceding increment passage time, tm, j is the instantaneous increment passage time, j is the increment counter, and jmax is the number of increments between the two synchronous notches.

If the method according to the invention is used in already known EDCs, the first injection will only take place after a statically valid segment. Unless statically valid segments are identified, it is not possible to average the camshaft speed nNW over the segments and another evaluation,
For example, the calculation of the pump characteristic field does not make sense.

Another monitoring is carried out by checking whether at least one rpm pulse has occurred within a maximum time of 120 ms. The maximum time is then defined as a multiple of 10 ms. This causes an uncertainty of 10 ms in the monitoring time.

All of the dynamic validation monitoring performed in block 32, the static validation monitoring performed in block 36, and the maximum time monitoring performed in block 31 all perform defect identification in block 33. Trigger. Defect identification causes an error indication at the output 34 in the presence of an error.

[0051]

According to the present invention, highly reliable reference mark identification is possible even when there are many markings such as in the case of an increment disc.

[Brief description of drawings]

FIG. 1 is a schematic diagram of known rotation speed detection.

FIG. 2 is a diagram for explaining rotation speed detection.

FIG. 3 is a block diagram of the present invention.

FIG. 4 is a detailed block diagram of an embodiment of the present invention.

[Explanation of symbols]

 10 Increment Disc 11 Mark 12 Shaft 14 Revolution Sensor 15 Pulse Forming Stage 16 Calculator

Front page continued (72) Inventor Peter Schmitz Ludwigsburg-Osweil Schottenberger Weg 10 (72) Inventor Dietbelt Schoenfelder Germany Stuttgart 1 Baldi Reweg 14 (72) Inventor Joachim Berger Federal Republic of Germany Winterbach Falkenstraße 11 (72) Inventor Peter Lutz Federal Republic of Germany Weinsberg Opdem Thiefenweg 27-1

Claims (8)

[Claims] 1. A method for evaluating an output signal of a rotation speed sensor, which scans a rotating element having a plurality of markings of the same kind and at least one distinguishable marking and outputs an output signal, the method comprising: A pulse sequence depending on the number of revolutions is formed, a time interval between pulses or a time interval between the same kind of edges of the pulse is obtained by a calculation device, each quotient of two time intervals is formed, and distinguishable markings are identified. In the method for evaluating the output signal of the rotational speed sensor, the same time interval, preferably a relatively large time interval (tm
g) is always a numerator, and a relatively small time interval (tm
A method for evaluating an output signal of a rotation speed sensor, which comprises forming a quotient (tr, j) such that k) is a denominator and comparing the quotient with at least one threshold value. 2. The rotating element is an increment disc (10) associated with a rotating shaft of an internal combustion engine, preferably a crankshaft (12), said at least one distinguishable marking (13a, 13b) The method according to claim 1, which is a synchronous notch. 3. The method according to claim 1, wherein the number of rotations is obtained from the time interval. 4. The method according to claim 1, wherein the quotient of the two time intervals is compared with another threshold value and the result of the comparison is used for error identification and error response. 5. The method according to claim 4, wherein static error identification and error response and / or dynamic error identification and error response are processed. 6. The method according to claim 1, wherein the threshold values (S1, S2, S3) are set in consideration of the maximum allowable rotational speed change. 7. A counter (2) depending on said synchronous notch.
8) is started, the counter increments by one for each increment, and when it reaches a counting state that is greater than the number of increments (11) between the two synchronization notches (13, 13b) and for each synchronization. The method according to claim 5, wherein when the notch is identified, the counter is reset, and the error identification and the error response are performed depending on the counting state of the counter. 8. The counting state has two synchronous notches (13,
Method according to claim 7, wherein an error is identified when it is greater than the number of increments (11) during 13b).