JP3451851B2 - Rotation speed detector - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a rotation speed detecting device, and more particularly to a motor speed control device, which is suitable for detecting a speed in a low speed region by using a sine wave encoder as a speed detector. The present invention relates to a rotation speed detection device.

[0002]

2. Description of the Related Art Conventionally, as a method of obtaining a rotation speed using a sine wave encoder, the number of pulses output by the sine wave encoder within a fixed period is counted and the position between the pulses is determined by the sine wave encoder output. A method is known in which the velocity is detected by finding the angle from the cosine wave.

As a method of obtaining the rotation angle, the sine wave and the cosine wave output from the sine wave encoder are detected at the same time point, the sine wave is divided by the cosine wave, and the angle at that time point is obtained from the quotient. Many methods are known, including Japanese Patent Publication No. 3-46530.

[0004]

In the above-mentioned prior art, the sine wave and the cosine wave output from the sine wave encoder,
And two pulses having the same phase for each of the sine wave and the cosine wave are used, and the quadrants of the sine wave and the cosine wave are discriminated by the two pulses. However, the two pulses have a phase delay with respect to the respective sine wave and cosine wave, and if the quadrant is determined while the phase delay is occurring, a determination error will occur and the position can be accurately determined. It becomes impossible to do so, and an error occurs in speed detection.

An object of the present invention is to provide a rotation speed detecting device capable of accurately detecting the rotation speed in a low speed region.

[0006]

[Means for Solving the Problems] First to achieve the above object
A feature of the invention is a sine wave encoder that outputs two-phase signals of a sine wave and a cosine wave that have both positive and negative voltage polarities and a phase difference of 90 degrees with each other by rotation of a rotating body, and the sine wave encoder. Pulse signal generating means for outputting a two-phase pulse signal having a phase difference of 90 degrees by comparing each of the wave signal and the cosine wave signal with a neutral point potential, and the sine wave signal as an absolute value and plus. An absolute value conversion means for outputting as an absolute value signal of only the side voltage and a rotation speed calculation device are provided, and the rotation speed calculation device is based on the output level of the two-phase pulse signal, and the sine wave signal or cosine wave signal. The quadrant is discriminated, the angle in the discriminated quadrant is obtained based on the absolute value signal, and the rotational position obtained by using this angle and the count value of the two-phase pulse signal are obtained. Zui and lies in that is configured to calculate the rotational speed.

The features of the second invention for achieving the above object are as follows:
The rotation speed calculation device has an analog / digital converter for converting the input absolute value signal into a digital signal, and uses the absolute value signal converted into a digital signal for calculation.

The features of the third invention for achieving the above object are as follows:
Due to the rotation of the rotating body, a two-phase signal of a sine wave and a cosine wave having both positive and negative voltage polarities and a phase difference of 90 degrees, and a first two-phase pulse signal having a phase difference of 90 degrees. And a pulse creating means for outputting a second two-phase pulse signal having a phase difference of 90 degrees by comparing each of the sine wave signal and the cosine wave signal with a neutral point potential. And a pulse switching means for outputting the first two-phase pulse signal in the high speed region of the rotating body and the second two-phase pulse signal in the low speed region of the rotating body, and output from the pulse switching means. Pulse counting means for counting the generated two-phase pulse signal,
A rotation speed calculation device, wherein the rotation speed calculation device calculates a rotation speed based on a count value of the first two-phase pulse signal in the high speed region, and the rotation speed calculation device in the low speed region. The rotation speed is calculated based on the count value of the two-phase pulse signal of No. 2.

The features of the fourth invention for achieving the above object are as follows:
Due to the rotation of the rotating body, a two-phase signal of a sine wave and a cosine wave having both positive and negative voltage polarities and a phase difference of 90 degrees, and a first two-phase pulse signal having a phase difference of 90 degrees. And a pulse creating means for outputting a second two-phase pulse signal having a phase difference of 90 degrees by comparing each of the sine wave signal and the cosine wave signal with a neutral point potential. An absolute value converting means for converting the sine wave signal into an absolute value and outputting it as an absolute value signal of only a plus side voltage; and in the high speed region of the rotating body, the first two-phase pulse signal, In the low speed region of, the pulse switching means for outputting the second two-phase pulse signal and the pulse switching means for outputting 2
The rotation speed calculation device includes pulse counting means for counting phase pulse signals, and a rotation speed calculation device. The rotation speed calculation device determines the rotation speed in the high speed region based on the count value of the first two-phase pulse signal. In the low speed region, the second
The quadrant of the sine wave signal is determined based on the output level of the two-phase pulse signal of, and based on the absolute value signal,
The rotation angle is calculated based on the determined angle in the quadrant and the rotation position obtained using this angle and the count value of the second two-phase pulse signal. It is in.

The features of the fifth invention for achieving the above object are as follows:
The absolute value signal output from the absolute value conversion means is provided with a means for detecting a peak value, the rotation speed calculation device, the sine wave specified peak value output by the sine wave encoder, and the absolute value signal Based on the peak value of
The correction coefficient is calculated, and the angle in the quadrant determined based on the absolute value signal is calculated as the angle in the quadrant determined based on the absolute value signal corrected by the correction coefficient. Is to be executed as the calculation of

The features of the sixth invention for achieving the above object are as follows:
When the absolute value signal is outside the dead band set for the absolute value signal, the rotation speed computing device performs quadrant discrimination of the sine wave signal based on the output level of the two-phase pulse signal. It is in.

The features of the seventh invention for achieving the above object are as follows:
The rotation speed calculation device is to set the dead band to a portion where the second two-phase pulse signal causes a phase delay with respect to the sine wave signal and the cosine wave signal.

The features of the eighth invention for achieving the above object are as follows:
The angle is to be calculated by Sin ^-1 calculation of the absolute value signal value.

[0014]

BEST MODE FOR CARRYING OUT THE INVENTION An embodiment of the rotation speed detecting device of the present invention will be described below. FIG. 2 shows the overall configuration of a motor control device that is a system to which the rotation speed detection device of this embodiment is applied.

The control device 6 comprises a computing device 7, PWM signal generators 8 and 9, and a rotation speed detecting device 10. The arithmetic unit 7 is composed of a unit having an arithmetic function such as a microcomputer, and inputs the rotational speed detection value detected by the rotational speed detection unit 10 at every predetermined sampling cycle. The rotation speed detection device 10 is composed of a one-chip microcomputer equipped with an analog / digital converter (A / D converter). The rotation speed detection device 10 includes a motor 1
The sine wave and the cosine wave output from the sine wave encoder 2 attached to the shaft of the above are input, and the pulse having the same phase as the sine wave and the cosine wave is input, and the rotation speed of the motor 1 is detected by executing the processing described later. . The arithmetic device 7 executes the rotational speed control calculation using the speed command value and the speed detection value output by the rotational speed detection device 10, and outputs the voltage command value to the PWM signal generation device 9. The PWM signal generator 9 outputs a PWM signal according to the voltage command to the converter 5. The converter 5 is
By converting DC power into AC power and supplying a predetermined AC power to the motor 1, the motor 1 is rotated at a rotation speed according to the speed command value. The converter 4 converts the AC power output from the AC power supply 3 into a predetermined DC power. The conversion in the converter 4 is performed by the PWM signal generator 8 outputting to the converter 4 a PWM signal corresponding to the voltage command value received from the arithmetic unit 7.

The detailed structure of the rotation speed detecting device 10 will be described with reference to FIG. The rotation speed detection device 10 includes a rotation speed calculation device 18 including a one-chip microcomputer that executes calculation of rotation speed. The rotation speed calculation device 18 is an A / D converter 1 that has a conversion capability only for positive voltage.
9, a rotation speed calculation means 20, a peak value correction coefficient calculation means 21, a minute position calculation means 22, and a timer 23 for generating rotation speed detection timing in a predetermined cycle. The rotation speed detection device 10 further includes an absolute value conversion means 11 for converting a sine wave having both positive and negative voltage polarities into an absolute value and converting the absolute value to a positive voltage, and a waveform value of the sine wave absolute valued at the rotation speed detection timing. Phase value holding means 12 for holding, pulse creating means 13 for creating a pulse by comparing sine wave, cosine wave and neutral point potential, pulses Psin, Pcos input from the sine wave encoder 2, and pulse creating means 13 Pulse switching means 15 for switching and outputting the two types of pulses CPsin and CPcos created in 1., edge extracting means 14 for outputting the peak detection timing of the sine wave to the peak value correction coefficient calculating means 21 and the peak value holding means 24, Quadrant holding means 16 for holding each output level of two pulses (for example, pulses CPsin, CPcos) output from the pulse switching means 15, a pulse Pulse time counting means for measuring the count and time of the pulse based on the output level of the relationship between the two pulses output from the switching means 15 1
7. A peak value holding means 24 for holding the waveform value of the peak of the sine wave that has been made into an absolute value is provided.

The sine wave encoder 2 is a sine wave having a frequency proportional to the rotation speed of the motor each time it rotates a certain amount, and having a phase difference of 90 degrees and both positive and negative voltage polarities.
Vsin, cosine wave Vcos, pulse Psin in-phase with sine wave Vsin, and pulse Pcos in-phase with cosine wave Vcos are output. The absolute value conversion means 11 outputs an absolute value sine wave Vabs obtained by converting a negative voltage of the sine wave Vsin into a positive voltage.
This absolute value sine wave Vabs is input to the A / D converter 19 capable of converting only the positive voltage, so that the sine wave Vsin output from the sine wave encoder 2 having both positive and negative voltage polarities is substantially Corresponds to being input to the A / D converter 19. The phase value holding means 12 is a timer 2
At the rotational speed detection timing output from 3, the detection phase value Vi is output as a signal holding the absolute value sine wave Vabs required for position calculation. Since the amplitude of the sine wave Vsin output from the sine wave encoder 2 varies depending on the rotation position of the motor, the detection accuracy is degraded. Therefore, it is necessary to correct the fluctuation of the amplitude, and the peak value hold means 24 causes the peak value of the absolute value sine wave Vabs at the peak detection timing output from the edge extraction means 14 described later.
Hold Vpeak. The pulses Psin and Pcos output from the sine wave encoder 2 are in-phase sine waves Vsi
There is a phase delay with respect to n and the cosine wave Vcos. In order to reduce this phase delay, the pulse creating means 13 uses a sine wave.
The pulses CPsin and CPcos are created by comparing Vsin and the cosine wave Vcos with the neutral point potential. Therefore, the pulse
CPsin and CPcos are pulses that rise / fall earlier than the pulses Psin and Pcos. The pulse switching means 15 is the rotation speed calculation means 2 of the rotation speed calculation device 18.
In accordance with the switching signal SW output by 0, one of the two types of pulse Psin and Pcos output from the sine wave encoder 2 and the pulses CPsin and CPcos created by the pulse creating means 13 is used as the pulse SPsin and Output as SPcos. That is, the pulse switching means 15 outputs the pulses Psin and Pcos as the pulses SPsin and SPcos in response to the switching signal SW indicating the high speed region described later. The pulse switching means 15 applies pulses CPsin and CPcos to a switching signal SW indicating a low speed region described later.
Are output as pulses SPsin and SPcos. The switching timing of this pulse is interlocked with the fall of the pulse Psin.

The edge extracting means 14 is an absolute value sine wave Vabs.
Since the peak timing of the pulse SPcos matches the rising edge and the falling edge of the pulse SPcos, the rising edge and the falling edge of the pulse SPcos are extracted and output as the peak detection timing. The pulse / time counting means 17 includes a pulse counter 17A and a time counter 17
Have B. The pulse counter 17A counts up or down according to the rotation direction at the rising and falling edges of the input pulse SPsin. Further, the time counter 17B always counts the reference clock having a constant cycle, and measures in synchronization with the time when the pulse counter 17A counts. Therefore, it is possible to accurately obtain the time required to count the pulses. However, up / down of the pulse counter 17A is
Determined from the relationship between the output levels of SPsin and SPcos. For example, if the rotation direction of the motor is forward rotation, up count,
If it is reversed, it will be counted down. Quadrant holding means 16
Holds the output levels of the pulses SPsin and SPcos necessary for the determination of four quadrants (see FIG. 7) that are separated by 90 degrees in the sine wave at the rotation speed detection timing output from the timer 23. A / D in the rotation speed calculation device 18
The converter 19, the peak value correction coefficient calculation means 21, and the minute position calculation means 22 are correction value calculation means for obtaining a correction value (position ΔM described later) for correcting the rotation speed in the low speed region.

The peak value correction coefficient calculation means 21 performs arithmetic processing based on the program executed at the peak detection timing output from the edge extraction means 14. The peak value correction coefficient calculation means 21 determines the level of the pulse SPsin at the peak detection timing and the peak value Vp of the absolute value sine wave Vabs.
The correction coefficient (αp, αn) for correcting the peak value Vpeak to the specified peak value Vo of the sine wave Vsin is calculated by taking eak and.

The minute position calculating means 22 performs arithmetic processing based on a program executed at the rotation speed detection timing output from the timer 24 at a predetermined rotation speed detection cycle. The minute position calculation means 22 detects the detection phase value Vi, the output levels of the pulses SPsin and SPcos held in the quadrant holding means 16, and the correction coefficient αp obtained by the peak value correction coefficient calculation means 21 at the rotation speed detection timing. Alternatively, αn is input, and the detected phase value Vi is subjected to peak correction and dead band processing described later to obtain a corrected phase value Vadj. Further, the minute position calculating means 22 calculates the position ΔM. That is, the minute position calculating means 22 uses the pulses SPsin and SPc.
The quadrant of the correction phase value Vadj is determined based on the output level of os, and the angle θ in this quadrant is calculated by using the correction phase value Vadj.
^{-Calculate by -1} calculation. The position ΔM is calculated based on this angle θ. The rotation speed calculation means 20 is the number of pulses output from the pulse / time counting means 17 (output of the pulse counter 17A), the time required for this (output of the time counter 17B), and the output of the minute position calculation means 22. The rotation speed is calculated and output based on the position ΔM. Regarding the calculation of the rotation speed in the rotation speed calculation means 20,
It will be described in detail later.

The individual processes described above will be described in detail. A correction method in the peak value correction coefficient calculation means 21 will be described with reference to FIG. The sine wave Vsin output from the sine wave encoder 2 causes a change in the amplitude of the sine wave Vsin depending on the rotational position of the motor and the like. The position ΔM is calculated from the angle θ of the phase value Vi detected at the speed detection timing. When the peak value fluctuation occurs in the sine wave Vsin, the accurate angle θ cannot be calculated, so it is necessary to correct the peak value fluctuation. This correction method is the peak value of sine wave Vsin
Vpeak is detected, a correction coefficient α for correcting the peak value Vpeak to the specified peak value Vo of the sine wave Vsin is calculated from the following formula, and the correction coefficient α is multiplied by the phase value Vi to add the correction. Since the magnitude of the peak value of the sine wave Vsin differs depending on the polarity of the plus / minus voltage, the correction coefficients αp and αn are calculated by the equations (1) and (2) for each polarity. However, since the peak value Vpeak is an absolute value, its polarity is determined by the output level of the pulse CPsin that is in phase with the sine wave Vsin.

Correction coefficient αp = specified peak value Vo / positive polarity detection peak value Vpeak (1) Correction coefficient αn = specified peak value Vo / negative polarity detection peak value Vpeak (2) For example, c cycle of FIG. The correction of the detected phase value Vi when the rotation speed detection timing occurs within is detected peak value Vpp (i) detected within the cycle a which is the previous cycle of the same polarity.
Is performed using the correction coefficient αp (i) calculated in (multiplication with the correction coefficient αp (i)). As a result of this correction, a corrected sine wave Vadj is obtained.

Actually, the peak value correction coefficient calculating means 21
Calculates the correction coefficients αp and αn, and the minute position calculating means 22 corrects the detected phase value Vi using the correction coefficients αp and αn.

The correction coefficient calculation processing executed by the peak value correction coefficient calculation means 21 at the peak detection timing supplied from the edge extraction means 14 will be described with reference to FIG. First, the detected peak value Vpeak output from the peak value holding means 24 is input via the A / D converter 108 (step 31). In step 32, a pulse CPsin having the same phase as the sine wave Vsin is fetched. In step 33, the polarity of the detected peak value Vpeak is judged by the output level of the pulse CPsin. If the polarity is positive, the correction coefficient αp is calculated in step 34 by the equation (1). If the polarity is negative, the correction coefficient αn is calculated in step 35 by the equation (2).

The dead band processing in the minute position calculating means 22 will be described with reference to FIGS. 5 and 6. The dead band processing uses the phase delay width of the pulse as the dead band,
The error due to the phase delay is prevented by not performing the quadrant determination and the angle θ calculation within the dead band. Pulses CPsin and CPco created by the pulse creating means 13
s has a slight phase delay Td at the rising or falling of the sine wave Vsin and the cosine wave Vcos. This phase delay
Due to Td, a count delay Cd corresponding to the phase delay Td is generated in the pulse counter 17A of the pulse / time counting means 17. Further, if the minute position calculation means 22 determines the quadrant while the phase delay Td is occurring, a quadrant determination error occurs. That is, in order to prevent the count delay Cd and the quadrant determination error, the dead band processing in the minute position calculating means 22 is performed by the absolute value sine wave Va as shown in FIG.
Set dead bands Dl and Dh to bs. Dead band d
l and Dh are set near the zero cross where the phase delay Td of the pulses CPsin and CPcos occurs and near the peak value. The widths of the dead bands Dl and Dh are the output values of the absolute value sine wave Vabs at the time of the phase delay Td. The judgment as to whether or not it is within the dead band is performed on the corrected sine wave Vadj obtained by adding the peak correction to the detected phase value Vi. If it is determined to be within the dead band, the quadrant is not determined, and if it is determined to be outside the dead band, the quadrant is determined.

The calculation of the position ΔM in the minute position calculating means 22 will be described. The rotation speed of the motor 1 is obtained from the number of pulses in the speed detection cycle. The number of pulses is counted by the pulse counter 17A. The pulse counter 17A counts up or down according to the rotation direction at the rising and falling edges of the input pulse SPsin. The position ΔM indicates a minute position within the pulse SPsin. That is, the position ΔM is the pulse counter 1
Amount from 0A count to the next count (0≤
ΔM <1). Then, the position ΔM is calculated using the absolute value sine wave Vabs in phase with the pulse SPsin used for counting. FIG. 7 shows the relationship between the position ΔM, the absolute value sine wave Vabs, and the pulse counter. Absolute sine wave Vabs is 90
It has four quadrants (I-IV) separated by degrees. pulse
The relationship between the output level of CPsin and the output level of pulse CPcos is as shown in FIG. The position ΔM has a value of 0 ≦ ΔM <1 if the motor 1 is rotating normally, and 0 to 1 are repeated in synchronization with the count of the pulse counter 17A. Similarly, even when the motor 1 rotates in the reverse direction, the position ΔM
Always takes a value of 0 ≦ ΔM <1 and repeats 1 to 0 in synchronization with the decrease of the pulse counter. That is, the value of the position ΔM during reverse rotation is in the range of 0 ≦ ΔM <1. However,
The pulse counter value when the value of the pulse counter 17A is positive, and the position ΔM equivalent after addition decreases by 1 to 0. Conversely, when the value of the pulse counter is negative, the pulse counter value and the position after addition The value of 0 to -1 is repeated for the amount corresponding to ΔM. Note that the method of obtaining the value of the position ΔM at the time of reverse rotation has the relationship with the position ΔM in each quadrant at the time of forward rotation,
It is replaced with the quadrant during reverse rotation.

Next, a method of calculating the position ΔM using the corrected sine wave Vadj will be described. The minute position calculating means 22 is an A / D
The detected phase value Vi is input at the rotational speed detection timing via the converter 19, and the detected phase value Vi is subjected to peak value correction processing and dead band processing to calculate a corrected sine wave Vadj. In addition, the quadrant is determined from the output levels of the pulses CPsin and CPcos, and based on this quadrant, the corrected sine wave Vadj
The angle θ is calculated by Sin ^-1 calculation. Next, the position ΔM can be calculated by dividing the obtained angle θ by 180 ° which is the pulse count point. The calculation of the position ΔM in the minute position calculation means 22 will be described with reference to the flowchart of FIG. In steps 41 to 43, pulse CPsin and CPcos
Determine the quadrant from the output level of. Next, step 44.
At 47, the angle θ in each quadrant is calculated by Sin ⁻¹ calculation using the corrected sine wave Vadj. However, since the angle calculated by the Sin ⁻¹ calculation is 0 ° ≦ θ <90 °, the angle θ in each quadrant is as follows. The angle θ in the I quadrant and the III quadrant can be obtained from the equation (3).

Θ = Sin ⁻¹ Vadj (3) In the quadrants II and IV, the angle θ is calculated by the equation (4).

Θ = 180 ° −Sin ⁻¹ Vadj (4) In step 48, the position ΔM is calculated by dividing the obtained angle θ by 180 °.

ΔM = θ / 180 (5) The calculation of the rotation speed executed by the rotation speed calculation means 20 will be described. The rotation speed of the motor 1 uses the pulses Psin and Pcos output from the sine wave encoder 2 at a frequency proportional to the rotation speed of the motor 1, the sine wave Vsin, and the pulses CPsin and CPcos created by the pulse creating means 13. Is calculated. The rotation speed of the motor 1 is obtained in the same manner as the formula (distance / time) for obtaining a general speed. That is, the distance is the pulse count number M, and the time is the time T required to count the pulse number that is the distance. The number of pulses counted and the time required for this counting are obtained by the pulse / time counting means 17.

When the rotation speed of the motor 1 becomes low, the cycles of the pulses Psin and Pcos output from the sine wave encoder 2 become long, and the number of pulses that can be counted within a predetermined rotation speed detection cycle also decreases. Therefore, it is necessary to compensate for the decrease in the number of pulses. Therefore, it is possible to compensate for the decrease in the number of pulses by using the position ΔM described above. Here, the number of pulses that can be counted for each rotation speed detection cycle is large, and there is no need to correct the decrease in the number of pulses, and the rotation speed region that is sufficiently accurate by calculation using the counted number of pulses is defined as the high speed region. , The sinusoidal wave Vsin from the decrease of the pulse number
The rotational speed region that needs to compensate for the decrease in the pulse number is defined as the low speed region. Switching between the high speed region and the low speed region can be arbitrarily set.

The rotation speed calculation means 20 calculates the rotation speed in the high speed region and the low speed region. First, a method of calculating the rotation speed in the high speed region will be described with reference to FIG. In the high speed region, it is not necessary to correct the decrease in the number of pulses, and the rotational speed is calculated using the counted number of pulses. Therefore, the pulses used for calculating the rotation speed are the pulses Psin and Pcos output from the sine wave encoder 2. In the high speed region, the pulse switching means 15 receives the input pulses Psin and Pcos, and the pulse Csin and Psin based on a signal SW indicating a high speed region described later.
Of Ccos, pulse Psin and Pcos are pulse SPsin and SPc
It is output to the pulse / time counting means 17 as os. Pulse counter 17A of the pulse / time counting means 17
Is up-counted at the rising and falling edges of the input pulse SPsin (motor 1 rotates forward) and the pulse number M
Is output. The time counter 17B counts the reference clock and measures the time T required for counting in synchronization with the count of the pulse counter 17A. The rotation speed calculation means 20 determines the pulse number M (i) from the pulse counter 17A at the rotation speed detection timing (i), and the time counter 17 as well.
Input time T (i) from B. Rotational speed calculation means 20
Uses the number of pulses M (i-1) and the time T (i-1) input at the previous rotation speed detection timing (i-1) to calculate the rotation speed in the high speed region by the equation (6). .

Rotational speed in high speed region = moving distance L / time required for moving distance L = (M (i) -M (i-1)) / (T (i) -T (i-1)) (6) In the high speed region, a large number of pulses from the encoder 2
Since Psin and Pcos are output and the influence of the pulse phase delay is reduced, the accuracy of the rotation speed in the high speed region obtained using the pulse number M (i) of the pulses Psin and Pcos is in a high state.

A method of calculating the rotation speed in the low speed region executed by the rotation speed calculation means 20 will be described with reference to FIG. In the low speed region where the number of pulses decreases, the rotational speed is calculated from the sine wave Vsin using the position ΔM (0 ≦ ΔM <1) between the pulses in order to correct the decrease in the number of pulses. The position ΔM is output from the minute position calculating means 22. In the low speed region, since the rotation speed is calculated using a sine wave, the phase delay of the pulses Psin and Pcos with respect to the sine wave Vsin and the cosine wave Vcos becomes a problem. Therefore, the pulses used for calculating the rotation speed in the low speed region are the pulses CPsin and CPcos with a small phase delay. The pulse switching means 15 receives the input pulse Psin on the basis of a signal SW indicating a low speed region described later in the low high speed region.
And Pcos, and the pulse CPsin of the pulses Csin and Ccos
It outputs n and CPcos to the pulse / time counting means 17 as pulses SPsin and SPcos. Pulse switching means 15
Pulse CPsi which is the pulse SPsin and SPcos output from
n and CPcos are edge extraction means 14 and quadrant holding means 1
It is also input to 6. Similar to the calculation of the rotation speed in the high speed region, the pulse number M (i) is changed from the pulse counter 17A to the time counter 17 at the rotation speed detection timing (i).
The time T (i) is input from B. In the calculation of the rotation speed in the low speed region, the position ΔM (i) is input from the minute position detection means 22 at the timing (i). Number of pulses M input at the previous rotation speed detection timing (i-1)
(i-1), time T (i-1) and position ΔM (i-1),
The rotational speed in the low speed region is calculated by the equation (7).

Rotational speed in low speed region = moving distance L / time required for moving distance L = M (i) + ΔM (i) − {M (i-1) + ΔM (i-1)} / (T ( i) -T (i-1)) (7) However, the time required for the moving distance L is not the time when the pulse counter 17A counts at high speed, but the time counter 17B at the speed detection timing. It is a measured value.

The contents of the above-described rotational speed calculation processing executed by the rotational speed calculation means 20 will be described with reference to FIG. The calculation for obtaining the rotation speed is executed in synchronization with the rotation speed detection timing output from the timer 23 in a predetermined rotation speed detection cycle. First, in step 51, the number of pulses M (i) from the pulse counter 17A is set to the time counter 1
Input time T (i) from 7B. The moving distance L (= M (i) -M (i-1)) within the rotation speed detection period is calculated (step 52). Based on the moving distance L, it is determined whether it is a high speed region or a low speed region (step 53). When the moving distance L calculated in step 52 is larger than the set moving distance L ₀ , it is determined to be in the high speed region. On the contrary, this moving distance L
Is less than the set movement distance L ₀ , it is determined to be a low speed region. The rotation speed calculation means 20 outputs either the signal SW indicating the high speed region or the signal SW indicating the low speed region based on the determination result of step 53. When the signal SW indicating the high speed region is output, the pulse switching means 15 outputs the pulses Psin and Pcos as the pulses SPsin and SPcos. When the signal SW indicating the low speed region is output, the pulse switching means 15 outputs the pulses CPsin and CPc.
Output os as pulses SPsin and SPcos. When the determination result of step 53 is in the high speed region, the calculation by the equation (6) is executed (step 63). In step 61,
The position ΔM calculated by the minute position calculating means 22 is input. When the determination result of step 53 is in the low speed region, the calculation by the equation (7) is executed (step 62).
Finally, the rotation speed obtained in step 62 or 63 is output to the arithmetic unit 7 (step 64). In step 64, the number of pulses M (i), time T (i), and position ΔM used in the rotation speed calculation at the rotation speed detection timing (i).
Each value of is updated as the previous value and stored as data to be used for the rotation speed calculation at the next rotation speed detection timing.
(Not shown).

According to this embodiment, since the method of calculating the rotation speed is switched according to the rotation speed of the rotating body, that is, in the high speed region and the low speed region, the number of pulses M and the time T are set particularly in the high speed region. Since it is performed by using the method, the processing time for obtaining the rotation speed can be reduced. In the low velocity region, the sine wave,
Pulse CP with little phase delay created based on cosine wave
Since the rotation speed is calculated using sin and CPcos, the error due to the pulse phase delay is reduced, and the accuracy of the rotation speed in the obtained low speed region is improved. Since the correction is performed using the position ΔM, the accuracy of the obtained rotation speed is further improved.

According to this embodiment, the pulses CPsin and CPcos are
Since the quadrant is determined by using, it is possible to prevent a quadrant determination error and appropriately determine the quadrant. Further, by providing a dead band and determining the quadrant, it is possible to eliminate the quadrant shift. These improve the accuracy of the rotational speed respectively obtained. In addition, by converting the sine wave having both positive and negative polarities into absolute values, the sine wave used to obtain the position between the pulses is built into the rotation speed calculation device 18 made of a one-chip microcomputer. A / D converter 1
Since the input can be made using 9, the rotation speed calculation device 18 can be downsized. This is the rotation speed detection device 10
It also leads to downsizing.

The contents of the above-described rotation speed calculation processing executed by the rotation speed calculation device 18 will be described with reference to FIG. Steps 51 to 53 and 62 to 64 are the same as the processing described in FIG. When it is determined in step 53 that it is in the high speed region, the calculation of step 63 is executed.

The case where it is determined in step 53 that the speed is in the low speed region will be described below. If it is determined to be in the low speed region, the low speed region rotational speed calculation processing in steps 54 to 62 is executed. First, the phase value Vi (i) held by the phase value holding means 12 is input via the A / D converter 19. Further, the output levels of the pulses SPsin and SPcos held by the quadrant holding means 16 are input (step 55). In step 56, the quadrant of the phase value Vi is determined from the relationship between the output levels of the pulses SPsin and SPcos.

Next, the peak value correction coefficient αp or αn obtained by calculating the phase value Vi (i) by the peak value correction coefficient calculation means 21.
Correct using. First, the polarity of the phase value Vi (i) is determined from the output level of the pulse SPsin (step 57). When it is determined that the polarity of the phase value Vi (i) is positive, the correction sine wave Vadj is calculated as the phase value Vi (i) and the correction coefficient αp (i-1).
It is determined by multiplication with and (step 58). Phase value Vi
If it is determined that the polarity of (i) is negative, the corrected sine wave Vadj is obtained by multiplying the phase value Vi (i) by the correction coefficient αn (i-1) (step 59). Then, it is checked whether the corrected sine wave Vadj is in the dead band (step 60). If it is in the dead band, step 51
The process from is repeated. This is for avoiding the calculation of the position ΔM based on the determination result of the quadrant in which the determination error occurs because the determination of the quadrant is performed in the dead band in step 56. If the determination in step 60 is No, the calculation in step 61A, that is, the angle θ from the correction sine Vadj and the position ΔM from the angle θ are calculated based on the obtained quadrant. The process of step 56 may be performed between step 60 and step 61 instead of between step 55 and step 57 described above. After the processing of step 61A, the processing of steps 62 and 64 is executed.

[0042]

According to the present invention, a sine wave and a cosine wave output from a sine wave encoder are used to generate a pulse with a small phase delay, and a sine wave is provided with a dead band for quadrant discrimination. Since it is possible to eliminate the deviation of, there is an effect that accurate speed detection can be realized. In addition, by converting the sine wave having both positive and negative polarities into an absolute value, the sine wave used when obtaining the position between the pulses can be input using the A / D converter incorporated in the one-chip microcomputer. Therefore, there is also an effect that the detection circuit can be downsized.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a rotation speed detection device that is a preferred embodiment of the present invention.

FIG. 2 is a configuration diagram of a motor control device to which the rotation speed detection device of FIG. 1 is applied.

FIG. 3 is an explanatory diagram showing an outline of a peak value variation correction method performed by a peak value correction coefficient calculation means.

FIG. 4 is a flowchart of a peak value fluctuation correction process performed by a peak value correction coefficient calculation means.

FIG. 5 is an explanatory diagram of pulse phase delay.

FIG. 6 is an explanatory diagram regarding installation of a dead band.

FIG. 7 is an explanatory diagram showing changes in the position ΔM during normal rotation (a) and reverse rotation (b).

FIG. 8 is a flowchart of the calculation of the position ΔM executed by the minute position calculation means.

FIG. 9 is an explanatory diagram showing an outline of calculation of a rotation speed in a high speed region.

FIG. 10 is an explanatory diagram showing an outline of calculation of a rotation speed in a low speed region.

FIG. 11 is an overall flowchart of rotation speed calculation executed by rotation speed calculation means.

FIG. 12 is a flowchart showing the entire processing of rotation speed calculation executed by the rotation speed calculation device.

[Explanation of symbols]

2 ... Sine wave encoder, 10 ... Rotational speed detection device, 11
... Absolute value conversion means, 12 ... Phase value holding means, 13 ...
Pulse creating means, 15 ... Pulse switching means, 17 ... Pulse / time counting means, 18 ... Rotation speed calculation device, 1
9 ... A / D converter, 20 ... Rotational speed calculation means, 21 ... Peak value correction coefficient calculation means, 22 ... Minute position calculation means, 2
4 ... Peak value holding means.

─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Junichi Takahashi 5-2-1 Omika-cho, Hitachi-shi, Ibaraki Hitachi Ltd. Omika factory (72) Inventor Toshihiko Matsuda 5-chome, Omika-cho, Hitachi-shi, Ibaraki No. 1 Inside the Omika Factory, Hitachi, Ltd. (72) Inventor, Kazuo Kawahara, 5-2-1, Omika-cho, Hitachi City, Hitachi City, Ibaraki Prefecture (56) Inside the Omika Factory, Hitachi, Ltd. (56) Reference JP-A-2-173523 (JP, A) JP-A-2-140616 (JP, A) JP-A-62-162968 (JP, A) JP-A-62-162917 (JP, A) JP-A-62-266468 (JP, A) JP-A-62 -228955 (JP, A) (58) Fields investigated (Int.Cl. ⁷ , DB name) G01P 3/489

Claims (5)

(57) [Claims] 1. A two-phase signal of a sine wave and a cosine wave having both positive and negative voltage polarities and a phase difference of 90 degrees with each other and a first phase having a phase difference of 90 degrees due to rotation of a rotating body. By comparing each of the sine wave signal and the cosine wave signal with the neutral point potential, a sine wave encoder that outputs a two-phase pulse signal of Pulse generation means to output,
In the high speed region of the rotating body, the first two-phase pulse signal is output, and in the low speed region of the rotating body, the second two-phase pulse signal is output, and the pulse switching unit outputs the pulse signals. The rotation speed calculation device comprises a pulse counting means for counting two-phase pulse signals and a rotation speed calculation device, wherein the rotation speed calculation device is based on the count value of the first two-phase pulse signal in the high speed region. The rotation speed detecting device is configured to calculate the rotation speed in the low speed region based on the count value of the second two-phase pulse signal. 2. A two-phase signal of a sine wave and a cosine wave having both positive and negative voltage polarities and having a phase difference of 90 degrees with each other and a first phase having a phase difference of 90 degrees due to rotation of a rotating body. By comparing each of the sine wave signal and the cosine wave signal with the neutral point potential, a sine wave encoder that outputs a two-phase pulse signal of Pulse generation means to output,
Absolute value conversion means for converting the sine wave signal into an absolute value and outputting it as an absolute value signal of only the positive side voltage; and in the high speed region of the rotating body, the first two-phase pulse signal is supplied to the rotating body of the rotating body. In the speed range, there are provided pulse switching means for outputting the second two-phase pulse signal, pulse counting means for counting the two-phase pulse signals output from the pulse switching means, and a rotation speed computing device, In the high speed region, the speed calculation device calculates the rotation speed based on the count value of the first two-phase pulse signal, and in the low speed region, based on the output level of the second two-phase pulse signal. The quadrant of the sine wave signal is determined, the angle in the determined quadrant is determined based on the absolute value signal, the rotational position obtained using this angle, and the second two-phase pulse Signal meter A rotation speed detection device configured to calculate a rotation speed based on a numerical value. 3. A rotation speed computing device, comprising: a means for detecting a peak value of the absolute value signal output from the absolute value converting means, wherein the rotation speed calculation device has a specified peak value of the sine wave output by the sine wave encoder. And a correction coefficient is calculated based on the peak value of the absolute value signal, and the calculation of the angle within the quadrant determined based on the absolute value signal is performed by the absolute value signal corrected by the correction coefficient. The rotation speed detection device according to claim 2 , which is executed as a calculation of an angle within the quadrant determined based on the above. 4. A plus and a minor depending on the rotation of the rotating body.
Have both voltage polarities and have a phase difference of 90 degrees from each other.
Sine wave encoder that outputs a two-phase signal of a sine wave and a cosine wave
And each of the sine wave signal and the cosine wave signal
Is compared with the neutral point potential, the phase difference of 90 degrees
Pulse generating means for outputting a two-phase pulse signal having
The sine wave signal is converted into an absolute value, and only the positive voltage
Absolute value conversion means that outputs as a value signal and rotation speed calculation
And a rotation speed calculation device , wherein the rotation speed calculation device outputs an output level of the two-phase pulse signal.
Quadrant of the sine or cosine signal based on Bell
Is determined based on the absolute value signal.
The angle in the quadrant
Based on the rotational position and the count value of the two-phase pulse signal.
There it is configured to calculate the rotational speed and the rotational speed calculating unit, when said absolute value signal is outside the deadband set for the absolute value signal, the output level of the two-phase pulse signal speed detecting apparatus for performing the quadrant determination of the sine wave signal based on. 5. The sine wave calculating device based on the output level of the second two-phase pulse signal, when the absolute value signal is outside a dead band set for the absolute value signal. The rotation speed detecting device according to claim 2 , which executes quadrant discrimination of a signal.