JPH0287210A - Method and instrument for measuring rotational speed - Google Patents

Abstract

PURPOSE:To prevent a rotational speed from being erroneously operated by switching a mode to a both edges detecting mode or a one-side edge detecting mode and operating the rotational speed of a unit arithmetic period from a passing time between effective pulse edges, which appear in the head or end of the respective present and preceding unit operating periods, and the number of the effective pulse edges. CONSTITUTION:When the rotational speed of a rotating body is changed from the unit arithmetic period <= a first prescribed speed (<=30Km/h) to the next unit arithmetic period, in case two conditions for the pulse to be kept at H are satisfied, the mode can be switched to the both edges detecting mode. When the rotational speed of the rotating body is >= a second prescribed speed (>=50Km/h), the mode can be switched to the one-side edge detecting mode. The rotational speed of the unit arithmetic period is operated from the passing time between the effective pulse edges, which appear in the head or end of the respective present and preceding unit arithmetic periods, and the number of the effective pulse edges. Thus, erroneous arithmetic is not executed and smooth speed control can be executed.

Description

[Detailed Description of the Invention] [Industrial Application Field] This invention provides a rotational speed system that outputs a pulse signal at a period proportional to the rotational speed of a rotating body and calculates the rotational speed of the rotating body from the interval of the pulse signals. This invention relates to improvements in measuring methods and devices.

[Prior art] As shown in FIG. 9, the tooth tips of the rotating body R are electromagnetically detected,
It is well known that a pulse signal is generated at a period proportional to the rotational speed of the rotating body R, and the rotational speed of the rotating body is calculated from the number of pulses within a unit time or the pulse interval. The error becomes large at some point. For this reason, a method is known that uses the number N of pulses within the unit calculation period Tm and the elapsed time T between the first pulse or the last pulse of the current and previous unit calculation periods, and the radius of the rotating body is calculated as R. , the number of pulses generated per rotation is Z, the rotational speed V is given by V=2πRN/ZT=αN/T. Figure 1O shows the unit operation period Tm in this case.
, is a diagram showing the relationship between the elapsed time T and the number of pulses N. Since the elapsed time T, which varies from T to T4 depending on the pulse detection time, is used instead of the fixed calculation period Tm, According to this method, calculation accuracy can be improved.

However, in the conventional example, calculation processing was performed using only either the rising part or the falling part of the pulse, so it was impossible to calculate accurate rotational speed due to lack of information, especially at low speeds. Ta.

In order to solve this problem, it is conceivable to always perform calculation processing using both the rising and falling parts of the pulse, but as the speed increases, the eccentricity of the rotating body causes the rotating body to tremble little by little, causing erroneous errors. Due to the output of the signal, there is a possibility that accurate rotational speed calculation may not be possible even with this method.

[Problems to be Solved by the Invention] From the above, at high speeds, either the rising or falling portion of the pulse, that is, one edge of the pulse, is used, and at low speeds, both the rising or falling portion of the pulse, that is, the pulse It is understood that it is sufficient to use edges on both sides, but specifically, under what conditions should the mode be switched to detecting edges from one side to both sides, or from both sides to one side?

This invention focuses on the above-mentioned problem, and provides a rotation speed switching condition that prevents the rotation speed from being calculated incorrectly during the unit calculation period in which the switching is performed and in any unit calculation period before or after the switching. The purpose of this work was to provide a method for measuring the amount of , and a device for the same.

[Means for Solving the Problem] In order to achieve this object, in the present invention, the rotational speed calculated within a certain unit calculation period is equal to or lower than the first predetermined speed, and the rotation speed calculated in the next unit from the unit calculation period is If the pulse continues to exist at the time when the calculation period changes,
Switching to the double edge detection mode, which detects both the rising and falling edges of a pulse as valid pulse edges, if the rotation speed calculated within a certain unit calculation period is equal to or higher than the second predetermined speed, the rising or falling edges of the pulse are detected as valid pulse edges. Switching to single edge detection mode in which either one of the falling edges is detected as a valid pulse edge, the elapsed time between the first valid pulse edges appearing at the beginning or between the last valid pulse edges appearing in each of the current and previous unit calculation periods is detected. The rotation speed (V) of the unit calculation period (Tm) is calculated from the time ('rd) and the number of effective pulse edges (N) within the elapsed time.

[Function] The present invention is used, for example, for speed control of a vehicle, and the rotating body is connected to wheels. The rotational speed of the rotating body is equal to or lower than the first predetermined speed, for example, when the wheel speed is 30 kn+/h.
In the following, when a pulse continues to exist at the time when one unit operation period changes to the next unit operation period, it means that the pulse is maintained at H (high state) at that time.

If these two conditions are satisfied, the mode is switched to the double edge detection mode, and highly accurate speed detection is performed.

Further, the rotational speed of the rotating body being equal to or higher than the second predetermined speed means, for example, that the wheel speed is equal to or higher than 501 on/h, and if this condition is satisfied, the mode is switched to the single edge detection mode.

By setting the switching conditions as described above, smooth speed control is possible without performing erroneous calculations during the unit calculation period (also called cycle) during which mode switching is performed and the subsequent unit calculation period. .

[Example] Next, an illustrated embodiment will be described.

FIG. 1 shows a schematic diagram of a device according to the present invention, in which a sine wave ωl (FIG. 2) from a sine wave output device 2 is shaped into a rectangular wave ω2 (FIG. 2) by a waveform shaping circuit 4. . For example, as shown in FIG. 8, the sine wave output device 2 is composed of a gear that transmits the rotation of a rotating body and an electromagnet placed near the tip of the tooth, and when the tooth passes the tip of the electromagnet, the magnetic flux density changes. occurs, and a sine wave is generated.

The rectangular wave ω2 from the waveform shaping circuit 4 is sent to the one-shot circuit 6, which outputs a pulse Ql that responds to the rising edge on one side of the rectangular wave ω2, and a pulse Q2 that responds to the rising edge and falling edge on both sides of the rectangular wave ω2. Ql is applied to one input of OR gate 10, and Q2 is applied to one input of AND gate 8. An output signal from a calculation circuit 22, which will be described later, is applied to the other input terminal of the AND gate 8, and an output signal from the calculation circuit 22, which will be described later, is applied to the other input terminal of the OR gate 1O.
The output signal of D gate 8 is added. The output signal of OR gate 10 is applied to calculation circuit 22.

The timer 12 is F n C (Free Running).
The timer 2 outputs the time like a clock, and the timer 2 measures a certain unit calculation period Tm, and also outputs various timing signals (for example, unit calculation period Tm).
m, outputs 32 interrupt signals Tin). Read circuit 16 receives pulse Q! from OR gate 10. or Q2
In response to this, the time is read, and the read content to is
It is stored in the memory Ml. The content to of the memory Ml is overwritten with a new time each time it is read, and in response to a timing signal at or near the time when the unit operation period To+ has elapsed (for example, a timing signal at the time when the unit operation period Tm has elapsed), Shifted to memory M2. Immediately, t-
In the memory M2, the time of the valid pulse edge appearing at the end of the unit calculation period Tm is stored as ti(i=1, 2, .
).

Next, the operation performed by the calculation circuit (CPLI) 22 will be explained according to the flowcharts shown in FIGS. 3 to 5.

FIG. 3 shows the main flow, in which subroutine processing A and subroutine processing B are processed in parallel.

On the other hand, subroutine processing A is executed, after which a unit calculation period Tm has elapsed, it is determined whether or not there are 1, and after period Tm has elapsed, processing A is executed again. On the other hand, subroutine processing B is repeatedly executed, and the period Tm
At the elapsed time point, the time ti of the valid pulse edge that appeared at the end of the period Tm is shifted from the memory M1 to M2, and the content number of the counter that counts the number of valid pulse edges is shifted.
is shifted to the memory N2, the elapsed time tc of the period Tm is shifted to the memory Tc, and the counter content No. is reset.

Before explaining process A, process B will be explained.

In the initial state, FI=I and F2=2, and in FIG. 5, it is determined whether the flag PI is 1 or 0 (step Sl).
. If the flag F1=0, that is, the wheel speed is, for example, 5
If the pulse edge reaches 0kn+/h or more and only one edge of the rising edge of the pulse is counted as a valid pulse edge, the process moves to step S2, and it is determined whether or not there is a rising edge of the pulse. that time to
is stored in memory Ml, and the count value of counter No. is set to 1.
(Step S5). Also, if the flag F1=1, that is, the wheel speed drops to, for example, 30 km/h or less,
Furthermore, if the predetermined condition (step #5 in Figure 4) is satisfied and both the rising and falling edges of the pulse are counted as valid pulse edges, the process moves to step S3, and it is determined whether there is a rising edge or a falling edge of the pulse. If yes, the current time to is stored in the memory M1 in the same manner as described above, and 1 is added to the count value of the counter No.

This process B is executed every time there is an interrupt signal.

FIG. 4 shows the contents of process A, which is executed by a timing signal (TX) such as an interrupt signal immediately before the unit calculation period Tm has elapsed.

In step #1, time 1. of the valid pulse edge that appeared at the end of the previous period Tm. (contents of memory M2) and
The difference Td between the effective pulse edge appearing at the end of the current period Tm and time t2 (contents of memory M+) is calculated.

In step #2, the difference T between the elapsed time tc of the current period Tm (the time obtained by adding the period Tm to the elapsed time tc of the previous period Trn) and the time of the current final effective pulse edge is T.
Calculate e. In step #3, the time Td is divided by the current number of effective pulse edges NO to calculate the average value Tf of one pulse period. In step #4, it is determined whether the current effective pulse edge number No is less than a predetermined value Na, that is, whether the current wheel speed is less than a constant speed V3, for example 30 km/h. Move on to 5.

In step #5, it is determined whether the pulse continues to exist at the time when the current unit calculation period Tm changes to the next unit calculation period TI+1, that is, whether the pulse is TI+1 at this time. This judgment is based on the fact that the time Te from the rising edge of the pulse to the end of the period Tm is 50%, preferably 25%, of the average period 'r'.
This is done by determining whether or not the following is true.

Expressing this as a formula, Te≦Tr*Ca −(1), and C
a is a value smaller than 0.5, preferably 0.25.

If Ca is selected to be 0.25, it is determined whether Te is smaller than half of the pulse. If Te is smaller, even if the average pulse period Tf changes greatly due to a rapid speed change, the period Tl11 is Until the final point, the pulse ω
This is because it is expected that it is almost certain that 2 is kept at H.

If the judgment result in step #5 is YES, step #6
Then, the flag F1 is set to 1, indicating that the mode has been switched to a mode in which both edges of a pulse are counted as valid pulse edges (both edge detection mode).

Note that this flag Fl=1 means AND in FIG.
Since gate 8 is manually operated, in both edge detection mode,
The pulse signal Q2 corresponding to both edges of the pulse is applied to the calculating circuit 22 via the AND gate 8 and the OR gate 10, as well as to the reading circuit 16.

Next, in step #7, the calculation formula switching flag F2 is set to 0.
If F2=O, it is changed to F2=1 (step #8), and if F'2≠0, the process goes to step #9 (.

In step #9, the calculation formula switching flag F2 is set to 0.
1.2.3.4. F2-
If it is 0 or 1, that is, one edge detection mode (Fl=
O), proceed to step #lO and calculate V(i)"'(N2/Td)*Cb ・-(2)c
b calculates the rotational speed, that is, the wheel speed (V i) using a constant.

In the case of F2-2, that is, both edge detection mode (FI=1
), proceed to step #11 and calculate V(i)−(
N2/Td)*Cb*'...The wheel speed is calculated by (3).

In the case of F2-3 or 4, that is, when the period is T+n immediately after switching from both edge detection mode to single edge detection mode (period T+a starting from switching to single edge detection mode (F1=0), and , in the next period 1'm,
F2 = 3 (step #20) and F2 = 4 (step #22), respectively), proceed to step #12 and obtain V(i) - V(i-1) + A(i-1) * Td ・・
- Calculate the wheel speed by (4) (A(i-1) is the previously determined wheel acceleration and is calculated in the previous step #13).

Next, in step #13, A(i)=(V(i)-V(i-1))/Td・
...The wheel acceleration is calculated by (5).

In step #14, it is determined whether the calculation formula switching flag F2 is 0, and in step #15, the flag F2 is determined to be 0.
It is determined whether 2 is 1 or not.

If F2=1, set F2-2 in step #16.

However, if -danF2=1, in the next cycle,
F2=2. In step #17, it is determined whether the flag is F2 or not. If F2-2, the current wheel speed or constant speed V is determined in step #18. , for example 5
It is determined whether or not it is greater than 0km/h, and if it is not, the flag Fl is set! The flag F2 is kept at 2.

Therefore, the both edge detection mode is maintained, and the wheel speed calculation formula used in step #11 is used.

If the wheel speed becomes greater than 50 km/h, step #I9. Proceed to #20, set F1=0 and set F
2-3. As a result, the double edge detection mode is switched to the single edge detection mode, and the wheel speed calculation formula, etc.
In the next cycle, the estimation formula in step #12 will be used.

In the latter half of the cycle, step #21 and step #22 are executed, and flag F2 becomes 4. Therefore, in the next cycle, the wheel speed is calculated again using the estimation formula in step #12. Then, in the next cycle, F2=0 (step #23), the one edge detection mode is maintained, and the wheel speed calculation formula is based on the actual measured value (step #lO). Ru.

In addition, in the case of a vehicle having an anti-lock system, steps #30, #31. shown in FIG. 6 are performed instead of steps #19, #20. Just add #32. As a result, the switching from the double edge detection mode to the single edge detection mode is performed under the conditions that the speed is at least a predetermined speed (50 km/h) and that the anti-lock control is not in progress.

FIG. 7 shows a modification of FIG. 4. In the case of FIG. 4, in the period Tm immediately after switching from the single edge detection mode to the double edge detection mode and in the period Tm next, step #
The wheel speed V(i) was estimated using the equations in step #11 and step #12, but in the case of the modified example shown in FIG.
) is configured to calculate.

In FIG. 7, steps #I to #7 are the same as those in FIG. If the result of determination in step #7 is F2-0, that is, if it is determined that only switching from double edge detection mode to single edge detection was performed in the previous unit calculation period Tm, step #l ! Then, it is determined whether FI=0 or not.

In the case of Fl-0, that is, in the single edge detection mode,
Proceeding to step #12 (corresponding to step #!0 in FIG. 4), the wheel speed is calculated using equation (2), and the flow is returned.

When FI=1, that is, in both edge detection mode,
Proceeding to step #13 (corresponding to step #II in FIG. 4), the wheel speed is calculated using equation (3).

Next, it is determined whether the current wheel speed V(i) is greater than the constant speed VO, and if it is not greater, the process returns as is, but if it is greater, the pulse 1 is
An intermediate average value Tg of the period is calculated.

For example, at time ts shown in FIG. 2, step # of FIG.
15 is executed, the time to of the pulse Ql that is before ts and closest to ts, and the time 11 (in Fig. 2,
The intermediate average value Tg is calculated by dividing the difference from the time point indicated by t to time t by the number of pulses Q1 generated after the time t. Next, in step #16, the time Tn which is the difference between the time ts and the time to of the pulse Q1 which is before ts and is closest to ts is calculated, and Tn is 50%, preferably 25% of the average period Tg. Determine whether the following is true.

Expressing this as a formula, Th≦Tg*Cc...(6)Cc<
1 (for example, 0.5. or 0.25). This determination is based on the same reason as the determination made using equation (1).

If Th is small and it is expected that the pulse ω2 is almost certainly kept at H or L, the time to of the pulse Ql which is ts@ and is closest to ts is stored as t3 (step #18) In addition, the number N of pulses Ql detected so far in the current unit calculation period Tm
o is stored as N3 (step #19). Next, the flag F1 is set to 0 to switch to the single edge detection mode (step #20), and the flag F2 is set to 1 to indicate that the double edge detection mode has been switched to the single edge detection mode in the current unit operation period of 1 mm.

Incidentally, steps #15 to #22 are repeated until the unit operation period Tm ends. If process A ends without executing steps #I5 to #22, as is clear from the flowchart in FIG.
Executed only once.

Next, in the next unit operation period Tn+, after proceeding to step #7 in FIG. 7, since P2=1, the flow proceeds to step #8, at times t, l and time L1 (in FIG. ,
From the above explanation, since the unit operation period Tm has changed, the difference Ti between the time t and the point indicated as tl is calculated, and in step #9, V(i) - (N3/Ti )*Cb*−=(7),
Calculate wheel speed based on actual measurements.

Next, in step #10, F2-0 is set to indicate that the single edge detection mode is in effect.

According to the modification of FIG. 7, in the unit calculation period immediately after switching to the one-edge detection mode, at a certain point (ts) within the current unit calculation period, the effective value that appeared at the end of the previous unit calculation period is From the elapsed time (T i) between the pulse edge and the last effective pulse edge that appeared before the time (ts), and the number of effective pulse edges (N,) within the elapsed time, the rotation speed for the current unit calculation period (Tm) is calculated. Therefore, it is possible to calculate a more accurate wheel speed based on the actual measurement value even during the unit calculation period Tm immediately after the mode is switched from the double edge detection mode to the single edge detection mode.

[Effect of the invention] Since this invention is configured as explained above,
This produces the effects described below.

The conditions for switching from the single edge detection mode to the double edge detection mode are (1) when the wheel speed reaches the first predetermined speed (for example, 30
km/h), and (II) the pulse continues to exist at the time when the current unit calculation period changes to the next unit calculation period (steps #4 and #5). In particular, by satisfying the latter condition (11), it is possible to measure the rotational speed more accurately.

Figure 8a shows that if condition (+1) is not met, A
, and the waveform diagram for case B is shown. Assume that the single edge detection mode is switched to the double edge detection mode at the time indicated by the dotted line. Furthermore, in the waveform diagram, detected effective pulse edges are indicated by arrows. In each of the current (Tmt) and previous (Tm,) unit calculation periods, the elapsed time (Td) between the valid pulse edges that finally appear, and the number of valid pulses (N) within the elapsed time.

The number of rotations for the current unit calculation period is calculated from the first effective pulse (not counting), but if condition (11) is satisfied (waveform B) or not (waveform A), the mode Elapsed time immediately after switching (Td
) are not the same. This difference is evaluated and determined as follows.

Since the unit calculation period (Tm3) immediately after mode switching is in both edge detection mode, the number of effective pulse edges within the elapsed time (Td) must include both rising and falling edges. In this case, the falling edge remains uncounted, resulting in an incorrect calculation. On the other hand, in the case of waveform B, valid pulse edges are correctly counted in any period.

Therefore, in the present invention, when one unit calculation period changes to the next unit calculation period, after confirming that the pulse continues to exist, the single edge detection mode is switched to the double edge detection mode.

Such confirmation is performed by determining whether the period ('re) from the final rising edge of a certain unit operation period to the final point of that unit operation period is 50%, preferably 25% or less of the pulse period. There is. Therefore, mode switching can be performed without erroneously calculating the rotational speed.

Also, the conditions for switching from both edge detection mode to single edge detection mode are (1) when the wheel speed reaches the second predetermined speed (50
km/h) or more. Further, as shown in FIG. 6, if the vehicle is equipped with an anti-lock system, in addition to the condition (1), there is also a condition (11) that anti-lock control is not in progress. In this case, since double edge detection and single edge detection coexist in the unit calculation period immediately after the mode switch, correct values may not always be obtained in the 1.2 unit calculation period after the mode switch. . FIG. 8b and FIG. 8C show waveform charts at that time. Comparing waveforms A and B in Fig. 7b, we can see that even though waveform A is in single edge mode, it is measured between the rising edge and the falling edge, so the elapsed time (T
d) are not the same. Similarly, the elapsed times of waveforms A and B in FIG. 8C are not the same.

Therefore, in this invention, the rotation speed calculation in the two unit calculation periods after switching from the double edge detection mode to the single edge detection mode is performed using the speed (vl-1, + and acceleration (At-) calculated in the previous unit calculation period. -1) (step #12), a value close to the actual measurement value can always be obtained.

[Brief explanation of the drawing]

FIG. 1 is a schematic diagram of a rotational speed measuring device according to the present invention, FIG. 2 is a waveform diagram of the signals appearing in FIG. 1, and FIGS. 3, 4, 5, and 6 are , a flowchart showing the operation of the apparatus shown in FIG. 1, FIG. 7 a flowchart showing a modification of FIG. 4, and FIGS. FIG. 9 is a schematic diagram of a device for detecting rotation, and FIG. 1O is an operational waveform diagram of a conventional rotation speed measuring device. 2... Sine wave output circuit, 4... Waveform shaping circuit, 6
・One yacht circuit, 12.14...Timer, 16...Reading circuit, 2
2...Calculating device.

Claims (1)

[Claims] 1. Detecting pulses (ω2) output at a period proportional to the rotational speed of the rotating body (R), and calculating the rotational speed (Vi) of the rotating body for each unit calculation period (Tm). A method for measuring rotational speed in which the rotational speed (Vi) calculated within a certain unit calculation period is equal to or lower than a first predetermined speed and continues at the time when the unit calculation period changes to the next unit calculation period. If the pulse continues, the mode is switched to a double edge detection mode in which both the rising and falling edges of the pulse are detected as valid pulse edges, while the rotational speed calculated within a certain unit calculation period is equal to or higher than the second predetermined speed. If so, switch to single-edge detection mode that detects either the rising or falling edge of the pulse as a valid pulse edge, and detect the gap between the first valid pulse edges or the final edge in the current and previous unit calculation periods. A method for measuring rotational speed, characterized in that the rotational speed of the unit calculation period (Tm) is calculated from the elapsed time (Td) between effective pulse edges appearing in the elapsed time and the number of effective pulse edges within the elapsed time. 2. The rotational speed measuring method according to claim 1, in which the rotational speed (Vi-_1) calculated in the previous unit calculation period is calculated in the two unit calculation periods immediately after switching to the single edge detection mode. and acceleration (Ai-_1) to calculate the rotation speed for the unit calculation period. 3. The rotational speed measuring method according to claim 1, in which the unit calculation period immediately after switching to the single edge detection mode,
At a certain point in time (ts) within the current unit operation period, the elapsed time (Ti ) and the number of effective pulse edges (N_3) within the elapsed time to calculate the rotation speed for the current unit calculation period (Tm). 4. Pulse (
ω2), and a calculation means (22) that calculates and outputs the rotational speed (Vi) of a rotating body for each unit calculation period (Tm). The apparatus is characterized in that the rotational speed (Vi) calculated within a certain unit calculation period is equal to or less than a first predetermined speed, and the pulse continues to exist at the time when the unit calculation period changes to the next unit calculation period. In this case, means for switching to a double edge detection mode for detecting both the rising and falling edges of a pulse as valid pulse edges, and when the rotational speed calculated within a certain unit calculation period is equal to or higher than a second predetermined speed, A means for switching to a single edge detection mode in which either the rising or falling edge of a pulse is detected as a valid pulse edge, and the means for switching to a single edge detection mode in which either the rising or falling edge of a pulse is detected as a valid pulse edge, and the means for switching between the valid pulse edges that appear at the beginning or between the valid pulse edges that appear at the end in each of the current and previous unit calculation periods. means for calculating the elapsed time (Td) between effective pulse edges that are passed and the number of effective pulse edges within the elapsed time, and using the obtained results to calculate the rotation speed for the unit operation period (Tm). A rotation speed measuring device characterized by: 5. The rotational speed measuring device according to claim 4, furthermore, in the two unit calculation periods immediately after switching to the single edge detection mode, the rotational speed (Vi- 1) and acceleration (Ai-_1), a rotational speed measuring device comprising means for calculating the rotational speed in a unit calculation period from the acceleration (Ai-_1). 6. The rotational speed measuring device according to claim 4, furthermore, in the unit calculation period immediately after switching to the single edge detection mode, at a certain point (ts) within the current unit calculation period, the previous unit calculation is performed. Means for calculating the elapsed time (Ti) between the valid pulse edge that appeared at the end of the period and the last valid pulse edge that appeared before the time (ts), and the number of valid pulse edges (N_3) within the elapsed time. ).