JPH0578772B2 - - Google Patents

Description

Detailed Description of the Invention [Industrial Application Field] The present invention connects a two-pole resolver and a multi-pole resolver (not including two poles) directly to a rotating shaft, and uses two for rough position detection.
The present invention relates to a position detection method using a resolver, which uses a polar resolver and uses a multipolar resolver for fine position detection.

[Conventional technology]

FIG. 2 is a block diagram of a conventional position detection circuit that uses a resolver (two poles) to detect the rotation angle of an electric motor, and FIG. 3 is a time chart thereof.

Resolver excitation circuit 1 includes resolver excitation windings 2 and 3
Resolver excitation signal e〓=Esinω ₀ t, e〓=
Excite with Ecosω ₀ t. The resolver detection winding 4 detects a resolver detection signal of voltage e=E ₀ sin (ω ₀ t+θ). Here, E is the excitation voltage amplitude, E ₀ is the detection voltage amplitude, ω ₀ is the excitation angular frequency, and θ is the resolver rotation angle. The shaping circuit 5 receives the resolver detection signal e〓
Pulse 9 is generated at the zero crossing from negative to positive. The free run counter 6 counts the clock pulses of the clock oscillator 7 (the counting start point coincides with the zero-crossing point of the resolver excitation voltage e from positive to negative). The count-up period of this free run counter 6 is set to be the same as the period of the resolver excitation frequency. For example, if the period of the resolver excitation frequency is 1/4.5 ms and the frequency of the clock oscillator 7 is 18.432 MHz, the number of bits of the free run counter 6 is 12, and it repeats counting clock pulses from 0 to 4095 pulses. The latch circuit 8 latches the data of the free run counter 6 using a pulse 9 synchronized with the zero cross of the resolver detection signal e. That is, the latch circuit 8 outputs a pulse 9 which is a latch signal.
The data θ _d of the free run counter 6 at the times t ₁ and t ₂ inputted by the free run counter 6 are latched. The microprocessor 10 reads this data θ _d and calculates the resolver rotation angle θ using the following equation (1).

θ=2πθ _d /2 ^M ...(1) However, M is the number of bits of the free-run counter 6. In this case, the method to increase the position detection resolution (number of divisions per rotation) is to (1) increase the clock frequency. (2) There are two methods: lowering the resolver excitation frequency, but the former method requires the use of high-speed ICs as processing circuit elements, resulting in high costs.
The latter method has the disadvantage that the detection speed decreases and the stability of the servo system deteriorates.

Figure 4 is a configuration diagram of a position detection circuit that connects a two-pole resolver and a multi-pole resolver (number of poles n) directly to the motor rotating shaft and performs high-resolution position detection by combining the two resolvers. In addition to the position detection circuit using a two-pole resolver, the position detection circuit using a multi-pole resolver (shaping circuit 15, free run counter 16,
A latch circuit 18) is added.

The detection winding voltage of the two-pole resolver is e = = E ₀ sin (ω ₀ t
+θ), and the detection winding voltage of the multipolar resolver is
e〓 _o = Ensin (ω ₀ t+nθ). Therefore, when the motor rotates once, the phase of the detection winding voltage e〓 of the two-pole resolver changes by 2π (rad) with respect to the excitation voltage e〓, and the phase of the multi-pole resolver changes by 2πn (rad).

FIG. 5 shows the change in phase of the two-pole resolver and the multi-pole resolver with respect to the rotation angle θ in FIG. 4. That is, Fig. 1 shows a quantized change in the phase of the detection winding voltage e〓 with respect to the excitation voltage e〓 of a two-pole resolver (when resolution m = 6), and Fig. 2 shows a quantized change in the phase of the detection winding voltage e〓 with respect to the excitation voltage e〓 of a two-pole resolver. The change in the phase of the detected winding voltage _e〓o with respect to the excitation voltage e〓 is shown in a simulated manner as the free run counter 16 counts up (r is the resolution).

Now, the rotation angle θ when rotating from the θ ₀ point (origin) to the θ ₁ point can be found as follows.

First, the rotation angle (data of the latch circuit 8) Pm (O≦Pm≦m) of the two-pole resolver shown in FIG. In addition, the rotation angle of the multipolar resolver (latch circuit 18
data) Qr (O≦Pr≦r) is read into the microprocessor 10. Next, the microprocessor 10
The rotation angle θmr is calculated by calculating equation (2).

θmr=Pm×r×Qr (2) In this way, by combining the two resolvers, the position detection resolution becomes m×r per rotation. For example, if the resolution m of a bipolar resolver is 100 and the resolution r of a multipolar resolver is 4096, the resolution will be 100×4096=409600.

However, in order to obtain high resolution, if a two-pole resolver is used for coarse detection and a multi-pole resolver is used for fine detection, the lowest bit of the counter may flicker (on/off) due to variations in the mechanical precision of the resolvers. ) may occur, causing a shift in the detection position. In particular, if flickering occurs in the free run counter 6 of the two-pole resolver used for coarse detection, the detection value r (4096 in the above example) of the fine detection resolver will deviate, resulting in a high resolution The position cannot be detected.

[Problem that the invention seeks to solve]

An object of the present invention is to provide a position detection method using a resolver that can always perform stable high-resolution position detection even if rough position detection value data using a two-pole resolver flickers due to mechanical errors of the resolver.

[Means for solving problems]

The position detection method using a resolver of the present invention includes the following steps: calculating first speed information from temporal changes in position information obtained from resolver signals of two resolvers; a step of comparing the two speed information; and a step of comparing the two speed information; and if the two speeds do not match, calculating the second speed information so that the first speed information matches the second speed information;
and a step of correcting the position information by adding or subtracting the resolution of the multipolar resolver to the position information obtained from the resolver signals of the two resolvers.

[Operation] In this way, the position detection signal of the two-pole resolver is
Since both speed information is corrected to match, 2
Even if flickering occurs in the position detection signal from the polar resolver due to mechanical error of the resolver, stable high-resolution position detection is always possible.

〔Example〕

Embodiments of the present invention will be described with reference to the drawings.

FIG. 1 is a flowchart showing an embodiment of a position detection method using a resolver according to the present invention. The circuit configuration is the same as that shown in FIG. 4, except that the processing content of the microprocessor 10 is different.

1 Step 1: Read the phase difference Pm ^t of the two-pole resolver from the latch circuit 8.

2 Step 2: Read the phase difference ^Qrt of the multipolar resolver from the latch circuit 18.

3 Step 3: Calculate the position θmr ^t using equation (2).

4 Step 4: Find the speed Nmr using equation (3) from θmr ^t and the position θmr ^t-1 calculated one cycle before.

Nmr=K ₁ θmr ^t −θmr ^t-1 / Δt ...(3) However, K ₁ is a constant Δt is the cycle time 5 for acquiring position data Step 5: The position data θr of the multipolar resolver just read ^t and position data one cycle before θr ^t-1
Find the speed Nr from equation (4).

Nr=K ₂ θr ^t −θr ^t-1 / Δt (4) where K ₂ is a constant Δt is the cycle time for acquiring position data 6 Step 6: Compare speed Nmr and speed Nr.

7 Step 7: When Nmr=Nr, assuming that the detected position data is correct, set the detected position θ to θmr.

8. Step 8: When Nmr>Nr, calculate Pm-1 and repeat the calculation from step 3 assuming that the minimum bit error of the detected rough position data is +1.

9 Step 9: When Nmr<Nr, the minimum bit error of the detected coarse position data is -1, calculate Pm+1, and repeat the calculation from step 3.

In this way, the microprocessor 10 corrects the flickering of the least significant bit of the position data ^Pmt of the two-pole resolver using the speed information Nmr and Nr, so it is possible to eliminate errors in position data due to quantization. .

Note that in this embodiment, when calculating the speed information Nr from the multipolar resolver, a phase change of the multipolar resolver is used, but a frequency change can also be used.

〔Effect of the invention〕

As explained above, the present invention calculates speed information from the temporal change in position information obtained from two resolver signals and the frequency change or phase change of a multipolar resolver, and positions the position information so that both speed information coincides. By correcting the detection signal, the lowest bit of the position data of the two-pole resolver is corrected, and errors in the position data can be eliminated. High-resolution position detection is possible without increasing the processing speed, and position error correction can be performed by simple software processing, so the original motor control of the microprocessor is not affected.

[Brief explanation of drawings]

FIG. 1 is a flowchart showing an embodiment of the position detection method using a resolver according to the present invention, FIG. 2 is a configuration diagram of a conventional example of a position detection circuit using a two-pole resolver, FIG. 3 is a time chart thereof, and FIG. The diagram is 2
Figure 5 is a configuration diagram of a conventional example of a position detection circuit that combines a polar resolver and a multi-pole resolver.
FIG. 3 is a diagram showing changes in phase of a polar resolver and a multipolar resolver with respect to rotation angle θ. 1... Resolver excitation circuit, 2, 3... Resolver excitation winding, 4... Resolver detection winding, 5, 15...
...shaping circuit, 6,16...free run counter,
7... Clock oscillator, 8, 18... Latch circuit, 9, 19... Pulse, 10... Microprocessor.

Claims (1)

[Claims] 1. A position detection method using a resolver, in which a two-pole resolver and a multi-pole resolver are directly connected to a rotating shaft, the two-pole resolver is used for coarse position detection, and the multi-pole resolver is used for fine position detection, a step of calculating first speed information from temporal changes in position information obtained from resolver signals of the two resolvers; a step of calculating second speed information from frequency changes or phase changes of the multipolar resolver; comparing the two speed information; and if the two speeds do not match, comparing the second speed information so that the first speed information matches the second speed information;
1. A method for detecting a position using a resolver, comprising the step of correcting the position information obtained from the resolver signals of two resolvers by adding or subtracting the resolution of the multipolar resolver to the position information.