JPH05248889A - Detecting method for angle of rotation - Google Patents

Abstract

PURPOSE:To obtain the title method capable of markedly enhancing resolving power, capable of smoothing the steady operation of a rotary body when a detected value is used as a feedback value for position control and capable of enhancing positioning accuracy. CONSTITUTION:When an encoder is connected to the shaft of a rotary body 1 to detect the angle of rotation of the rotary body, as the encoder, a sine wave encoder 10 outputting sine wave signals of Aand B-phases is used. The output of the sine wave encoder 10 is shaped into a pulse waveform by a waveform shaping device 11 to be multiplied by four by a direction discriminating and multiplying circuit 12 and the pulse count value of this multiplied pulse is latched at a predetermined cycle by a latch 13 while the A-and B-phase outputs of the sine wave encoder 10 are subjected to A/D conversion and the tan<-1> operation of A-and B-phase digital values is performed at the predetermined cycle by a tan operator. The tan<-1> operation value is added to the pulse count value to detect the angle of rotation of a rotary body.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for detecting a rotation angle of a rotating body such as a servomotor.

[0002]

2. Description of the Related Art FIG. 5 shows a conventional rotation angle detecting device. In the figure, 1 is a motor, 2 is a motor 1
It is a two-phase pulse encoder connected to the. This two-phase pulse encoder generates an A-phase pulse and a B-phase pulse whose phases are different from each other by 90 ^° . Reference numeral 3 denotes a direction discriminating and multiplying circuit (which generates a quadruple multiplying pulse), into which the A phase pulse and the B phase pulse are input, and as shown in FIG. When it is in the lead phase, it is assumed that the motor 1 is rotating normally, and the pulse train U is output. As shown in FIG. 6B, when the B-phase pulse is in the lead phase with respect to the A-phase pulse, , The pulse train D is output assuming that the motor 1 is rotating in the reverse direction. 4 is an up-down counter, which receives a pulse train U at its up-up terminal,
The down DOWN terminal receives the pulse train UD. The count value of the up / down counter 4 becomes a rotation angle detection signal.

[0003]

Since the resolution of this rotation angle detection device is determined by the pulse resolution, the resolution is low, and when the count value of the up / down counter 4 is fed back to perform position control, the steady state is obtained. In, the vibration for one pulse is generated. Therefore, when the vibration cycle is high, noise is generated and the positioning accuracy is lowered. The present invention has been made to solve this problem, it is possible to significantly increase the resolution compared to the conventional, when using the detected value as a feedback value for position control, compared to the conventional, An object of the present invention is to provide a rotation angle detecting method capable of smoothing a steady operation of a rotating body and improving positioning accuracy.

[0004]

In order to achieve the above-mentioned object, the present invention uses an A-phase or B-phase encoder as the encoder in the case where an encoder is connected to a rotating body to detect the rotation angle of the rotating body. A sine wave encoder that outputs a sine wave signal is used.The output of this sine wave encoder is shaped into a pulse waveform, then multiplied, and the pulse count value of this multiplied pulse is latched at a specified cycle. After the A-phase and B-phase outputs of the encoder are A / D converted, tan ⁻¹ of the A-phase digital value and the B-phase digital value is calculated at the above-mentioned predetermined cycle
The calculation is performed and the calculated tan ⁻¹ value is added to the pulse count value to detect the rotation angle of the rotating body.

According to a second aspect of the present invention, the change amount of the current latch value of the pulse count value with respect to the previous latch value is calculated, and the calculated value of the tan ^-1 calculation is added to the change amount integrated value to determine the rotation angle of the rotating body. I tried to detect it.

In the third aspect, the rotation direction of the rotating body is discriminated from the phase relationship between the A-phase signal and the B-phase signal.
The combination of the A-phase digital value and the B-phase digital value, which is the calculation parameter of the tan ⁻¹ calculation, is changed based on the positive / negative of the A-phase and B-phase signals.

[0007]

In the present invention, the rotation angle is not calculated only from the pulse count value of the multiplied pulse, but the digital conversion value of the A-phase and B-phase signals of the encoder is used for this rotation angle.
Since the rotation angle is detected by adding the rotation angle obtained in a predetermined calculation cycle by n ⁻¹ calculation, the resolution is the pulse resolution ×
A / D resolution is achieved, and it is possible to detect a rotation angle with a remarkably high resolution as compared with the conventional method.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

In FIG. 1, the encoder axially connected to the motor 1 is a two-phase sine wave encoder 10, and this sine wave encoder 10 has an A phase sine wave signal whose phase is different from each other by 90 ^° (see (a) in FIG. 4). And a B-phase sine wave signal (shown by a broken line in FIG. 4A). The A-phase sine wave signal and the B-phase sine wave signal are converted into rectangular wave signals S _A and S _B by a waveform shaper 11, and then input to a direction discriminating and multiplying circuit (which generates a quadruple multiplying pulse) 12. .. Similarly to the direction discriminating and multiplying circuit 3 in FIG. 5, the direction discriminating and multiplying circuit 12 determines that the motor 1 is rotating normally when the A-phase sine wave signal is in the advance phase with respect to the B-phase sine wave signal pulse. ,
When the pulse train U is output and the B-phase sine wave signal is in the advance phase with respect to the A-phase sine wave signal pulse, it is assumed that the motor 1 is in reverse rotation, the pulse train D is output, and it is synchronized with the clock c. The A-phase rectangular wave pulse S _A1 , the B-phase rectangular wave pulse S _B1 (both are shown in FIG. 3) and the direction determination signal FR that becomes “L” at motor forward rotation “H” are output. The direction discrimination signal FR is generated using the flip-flop circuit shown in FIG. Direction discrimination multiplication circuit 1
The pulse trains U and D output by 2 are up / down counters 4.
_Is input to the rotation angle signal n _aN which is the output of the up / down counter 4 (N: indicates the Nth output).
Are supplied to the first latch 13. Also, the A-phase rectangular wave pulse S _A1 , the B-phase rectangular wave pulse S _B1 and the direction determination signal F
R is supplied to the second latch 14. The first latch 13 and the second latch 14 receive an interrupt signal oscillated by the oscillator 30 in a predetermined cycle T as a latch command. The rotation angle signal n _aN output from the first latch 13 is taken out from the latch 13 by the adder 15 at the (N−1) th time and
Of the rotation angle signal n _{a (N-1)} stored in the memory M1 of This deviation Δn _a = n _aN −n _{a (N−1)} is the amount of change between the periods of the interrupt signal and is cumulatively added by the integrator 16. This cumulative value is n _bN . 17 is an adder, M
2 is a second memory for storing n _{b (N-1)} .

The A-phase sine wave signal and the B-phase sine wave signal output from the sine wave encoder 10 are also converted into digital signals A ^* and B ^* by the A / D converter 18, and then θ _a = ta.
It is given to the n ⁻¹ calculator 19. However, -90 ^O <θ _a
It is a value in the range of <90 ^O. This tan ^-1 calculator 19
Is the A-phase rectangular wave pulse S _A1 ^* , the B-phase rectangular wave pulse S _B1 ^* (shown in (b) and (c) of FIG. 4, respectively) output from the second latch 14 in response to the latch command, and the direction determination signal. F
R ^* (shown in (d) of FIG. 4) is fetched, and based on the positive / negative of the A-phase rectangular wave pulse S _A1 ^* , the positive / negative of the B-phase rectangular wave pulse S _B1 ^* , and H / L of the direction determination signal FR ^* , The tan ^-1 operation from (a) to (h) is executed.

In this embodiment, this tan ^-1 operation,
The calculation of Δn _a and its cumulative calculation are executed by the software of the microcomputer with the cycle T of the interrupt signal (which is considerably smaller than the cycle of the multiplied pulse) as the calculation cycle.

(A) When S _A1 ^* > 0, S _B1 ^* <0, FR ^* = H, θ _a = tan ⁻¹ (A ^* / − B ^* ), which is the straight line in (e) of FIG. It means that 1 is calculated.

(B) When S _A1 ^* > 0, S _B1 ^* <0, FR ^* = L, θ _a = tan ⁻¹ (B ^* / A ^* ), which is shown in FIG.
This means that the straight line 2 of is calculated.

(C) When S _A1 ^* > 0, S _B1 ^* > 0, and FR ^* = H, θ _a = tan ⁻¹ (B ^* / A ^* ), which is shown in FIG.
This means that the straight line 2 of is calculated.

(D) When S _A1 ^* > 0, S _B1 ^* <0, FR ^* = L θ _a = tan ⁻¹ (−A ^* / B ^* ), which is the straight line of (e) in FIG. It means that 3 is calculated.

(E) When S _A1 ^* <0, S _B1 ^* > 0 and FR ^* = H, θ _a = tan ⁻¹ (−A ^* / B ^* ), which is the straight line in (e) of FIG. It means that 3 is calculated.

(F) When S _A1 ^* <0, S _B1 ^* > 0, and FR ^* = L, θ _a = tan ⁻¹ (−B ^* / − A ^* ), which corresponds to (e) of FIG. This means that the straight line 4 is being calculated.

(G) When S _A1 ^* <0, S _B1 ^* <0, FR ^* = H, θ _a = tan ⁻¹ (−B ^* / − A ^* ), which corresponds to (e) of FIG. This means that the straight line 4 is being calculated.

(H) When S _A1 ^* <0, S _B1 ^* <0, FR ^* = L, θ _a = tan ⁻¹ (−A ^* / − B ^* ), which corresponds to (e) of FIG. This means that the straight line 1 is calculated.

The combination of A ^* and B ^* in (a) to (h) is shown in FIG. 4 (f).

The angle θ _a calculated in this way is added to the n _{bN in the} adder 20, and θ _f = (n _bN +
θ _a ) is the rotation angle detection value.

In the present embodiment, the rotation angle θ _a of the motor 1 is calculated by tan ⁻¹ calculation at _a predetermined calculation cycle T, and basically, this tan ⁻¹ calculated value is the pulse count value n of the multiplied pulse. _Since it is added to _aN , the resolution becomes the above-mentioned pulse resolution × A / D resolution, which is significantly higher than the conventional resolution.

Further, in this embodiment, the amount of change Δn _a of the rotation angle in the calculation cycle T is cumulatively calculated, and the cumulative calculation value n
_The value obtained by adding the above rotation angle θ _a to _bN is the rotation angle detection value θ _f
Therefore, the detected rotation angle θ _f changes smoothly.

Therefore, if the rotation angle detection value θ _{f is} fed back to perform the position control, the above-mentioned conventional vibration is suppressed and the positioning accuracy is improved.

The above n _aN and n _bN are integrated values of the input pulse, but the number of bits (digits) of the soft counter for counting n _bN can be increased by adopting the method of this embodiment. .. On the contrary, if the maximum value of the change amount (= Δn _a ) of the up / down counter in one operation cycle is Δn _amax , the bit number (digit number) K of the up / down counter should satisfy the following equation. Become.

2 × Δn _amax <2 ^K (1) For example, if Δn _amax = 3, then K should satisfy K > .

Therefore, for example, since the up / down counter can be 3 bits and the soft counter can be 16 bits, the cost and circuit scale of the up / down counter can be saved. That is, once Δn _a is obtained, the number of digits of the counters before and after the counter has a degree of freedom.

[0028]

As described above, the present invention does not obtain the rotation angle from only the pulse count value of the multiplied pulse, but
The rotation angle is detected by adding the rotation angle obtained in a predetermined calculation cycle by tan ^-1 calculation using the digitally converted value of the A-phase and B-phase signals of the encoder as the calculation parameter, and the rotation angle is detected. The rotation angle can be detected with extremely high resolution.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is a diagram showing an example of a circuit that generates a direction determination signal used in the above embodiment.

FIG. 3 is a waveform time chart for explaining the operation of the above embodiment.

FIG. 4 is a waveform time chart for explaining the operation of the above embodiment.

FIG. 5 is a block diagram showing a conventional rotation angle detection device.

FIG. 6 is a waveform time chart for explaining the operation of the conventional example.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 motor 4 up-down counter 10 2-phase sine wave encoder 11 waveform shaper 12 direction discriminating and multiplying circuit 13, 14 latches 15, 17, 20 adder 16 integrator 18 A / D converter 19 tan ^-1 calculator

Claims (3)

[Claims] 1. A sine wave encoder that outputs sine wave signals of A-phase and B-phase is used as the encoder when the rotation angle of the rotator is detected by connecting an encoder to the rotator. The output of the wave encoder is shaped into a pulse waveform and then multiplied, and the pulse count value of this multiplied pulse is latched at a predetermined cycle, while the A phase and B phase outputs of the sine wave encoder are A / D converted. After that, at the above-mentioned predetermined cycle, tan of the A-phase digital value and the B-phase digital value
^-1 calculation is performed and the calculated tan ^-1 value is added to the pulse count value to detect the rotation angle of the rotating body. 2. A change amount of a current latch value of a pulse count value with respect to a previous latch value is calculated, and the calculated value of the tan ⁻¹ calculation is added to the change amount integrated value to detect a rotation angle of a rotating body. The rotation angle detecting method according to claim 1, wherein 3. A tan ^-1 operation parameter is determined based on the result of the determination and the positive / negative of the A-phase and B-phase signals by determining the rotation direction of the rotating body from the phase relationship between the A-phase and B-phase signals. 3. The rotation angle detecting method according to claim 1, wherein the combination of the A-phase digital value and the B-phase digital value is changed.