JPS6144245B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a detection error correction device when a resolver is used as a position detector of a numerical control device.

Conventional numerical control devices use an electromagnetic detector (e.g. resolver, inductosyn) or an optical detector (e.g. encoder) as their position detector.
are commonly used, but most of these detectors are of the rotating type.

One of the typical rotary position detectors is a resolver, and its position detection method generally uses a phase difference detection method. 1st
The figure shows a conventional device using this method, where 2 is a resolver, 1 is an excitation circuit for the resolver,
3 is the stator winding of the resolver 2, 4 is the rotor winding of the resolver 2 having two electromagnetically orthogonal windings, and 5 is the output of the two windings of the rotor winding 4 as shown in FIG. A known phase shifter synthesizes the phase using a resistor 21 and a capacitor 22 and converts it into an electrical phase shift amount proportional to the mechanical rotation angle of the resolver 2. 6 is a phase shifter that converts the output of the excitation circuit 1 into a movement command input 10 digitally or analogously. A known phase modulator capable of modulation; 7 is a phase discriminator that discriminates the phase difference between the outputs of the phase shifter 5 and the phase modulator 6; 8 is a servo amplifier driven by the output of the phase discriminator 7; 9 indicates a servo motor.

In the device shown in FIG. 1, first, a movement command is given to the phase modulator 6 by a movement command input 10, and as a result, a phase difference is generated between it and the phase shifter 5, and the difference is as follows.
The servo motor 9 is rotated via the discriminator 7 and the amplifier 8, and the rotor winding 4 is rotated via the mechanical coupling mechanism 11.

As a result, the movement continues until the output phases of the phase shifter 5 and the discriminator 6 match, and positioning operation becomes possible. By the way, the problem with this method is the mechanical accuracy of the resolver itself and the residual voltage error of the rotating winding 4 caused by harmonics included in the excitation waveform of the excitation circuit 1. This is a deviation between a theoretical mechanical angle, that is, an electrical phase angle, and an actual detected mechanical angle in response to a command input that normally occurs in one cycle or several cycles during rotation, and the situation is normally as shown in FIG.

This error is usually not a very large value and does not affect repeatability, so it is less of a problem in terms of dimensions and accuracy. It has a considerable effect on surface roughness and becomes a problem. That is, the undulations that occur once or several times per one rotation of the detector as shown in FIG. 3 are formed on the cutting surface, and often become a problem especially during turning.

The present invention has been made to solve these drawbacks with a simple configuration, and by approximately eliminating the undulations shown in Fig. 3, the above-mentioned problems can be solved and a good cutting surface can be obtained. It was made so that it could be obtained.

FIG. 4 shows an embodiment of the present invention.
11 are the same as in FIG. 12 is a ring counter operated by the movement command input 10; 13;
14 indicates an error correction amount command device which generates a correction phase shift command or a correction pulse command as shown in FIG. 5, for example, according to the count value of the ring counter 12;

FIG. 6 shows the relationship between the movement command input 10 and the phase modulator 6 shown in FIG. The relationship between the phase shift amount of the modulator 6 corresponding to 10 is as follows: [phase angle corresponding to one electrical angle cycle of the resolver, that is, 360 degrees] = [phase shift amount corresponding to the minimum movement command input] x [integer] ing. This is generally an electrical angle of 1
The mechanical movement amount corresponding to a cycle is 2 mm, and the minimum movement command is 10 μ or 1 μ. In this case, the phase shift amount corresponding to the minimum movement command input is 1/1 cycle of electrical angle in Fig. 6. The phase amount is 200 or 1/2000, that is, 1.2 degrees or 0.12 degrees.

Next, the operation of the present invention will be explained.

The basic detection and drive operations in Figure 4 are as follows:
As shown in FIGS. 1 to 11, the operation will first be described assuming that only the minimum movement unit shown in FIG. 6 is commanded by the movement command input 10.

In response to the command input 10, the phase of the phase modulator 6 is shifted by the unit movement amount, and the phase discriminator 7 generates an output, which activates the servo amplifier 8 and the servo motor 9, thereby creating a mechanical connection. The resolver rotor winding 4 moves via the mechanism 11 until it matches the phase of the phase modulator 6, and then stops.

On the other hand, input 10 is also applied to ring counter 12 . This ring counter 12 in FIG. 4 is constructed of a ring counter that can count the same number of divisions in the minimum movement command shift amount for one cycle of the resolver electrical angle in FIG. 6, and the shift input or pulse corresponding to the minimum movement amount is When commanded,
Depending on the direction of movement, it is possible to add or subtract one count. The ring counter 12 is reset to 0 by a resolver rotation origin detection circuit 15 which is constructed in a suitable manner and outputs an output only at one point in one electrical angle cycle of the resolver.

If the ring counter 1 is detected by the resolver rotation origin detection circuit 15 immediately before the input 10 is given,
2 is reset to 0, and input 10 is a movement command with a positive count, the ring counter 12
becomes the count “1”. As a result, this output is given as an error correction shift input E to the detection circuit 14 by the error correction command unit 13 as shown in FIG. The excitation output is given to the resolver 2.

As a result, the output of the rotor winding 4 of the resolver is corrected by the movement amount corresponding to ±E compared to the case without the ring counter 12 and the error correction command unit 13, and the mechanical movement amount due to the input 10 is calculated by the ring counter 12. , the amount of movement in the conventional method without the error correction command unit 13 and the phase modulator 14 is ±E. Therefore, if the direction and amount for correcting this ±E are set in accordance with the error corresponding to the count "1" position of the resolver in Figure 3, it is possible to cancel the error caused by the resolver itself. Become.

From the above, this error correction is performed by specifying the count number of the ring counter 12 as described above, and setting the correction amount with the error correction command unit 13 for each count value as described above. This makes it possible to eliminate the fluctuation of the error shown in FIG. It is also effective to use an approximation method to correct the count number of the ring counter 12 by dividing the number of points by an appropriate multiple of the minimum movement unit (for example, by setting 8 points as shown in Fig. 7). 12 and error correction command unit 1
3 can be simplified.

The above explanation was given for the case where only the input corresponding to the minimum movement amount was given as the movement command input 10, but
It goes without saying that even when the movement command input 10 is continuously applied, the same process continues and the correction operation is performed normally.

On the other hand, the correction by the device of the present invention can also be applied to a speed detection circuit that applies the phase output of a resolver.
In the conventional device, the error curve shown in FIG. 3 of the resolver causes rotational unevenness, but by using this device, it becomes possible to eliminate rotational unevenness.

Further, the series of operations shown in FIG. 4 can of course be processed by software using a microcomputer.

As described above, the present invention can greatly contribute to improving the cutting performance of a numerically controlled device that uses a resolver as a phase detector, especially in improving surface roughness, and can also significantly improve the cutting performance of a numerical control device that uses a resolver as a phase detector, and can also significantly improve the cutting performance of a numerical control device that uses a resolver as a phase detector. This can also be useful in preventing uneven rotation.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram for explaining a conventional device;
FIG. 2 is a supplementary explanatory diagram of FIG. 1, FIG. 3 is a resolver error curve diagram generated in a conventional device, FIG. 4 is an explanatory diagram showing an embodiment of the present invention, and FIGS.
The figure is a supplementary explanatory diagram for explaining FIG. 4, and FIG. 7 is an explanatory diagram showing a simple example of the present invention. The same reference numerals in the figures indicate the same or corresponding parts, 2
is a resolver, 5 is a phase shifter, 6 and 14 are phase modulators, 7 is a phase discriminator, 8 is a servo amplifier, 9 is a servo motor, 10 is a movement command input, 12 is a ring counter, 13 is an error correction command 15 is the resolver rotation origin detection circuit, X is the phase shift amount corresponding to the minimum movement command input, T is the resolver excitation cycle period, n is an integer, _l1 is the correction approximate curve based on the correction shift amount, _l2 is the resolver This is the error curve.

Claims (1)

[Claims] 1. A device for correcting detection errors of a resolver that performs position detection, speed detection, etc. using a phase shift difference, comprising a counter means for counting movement command inputs, an error correction amount command means for weighting the output of the counter means, It is equipped with a phase shift modulation means that shifts the excitation phase shift using this output, and corrects errors periodically generated by the resolver itself or its peripheral circuits corresponding to one cycle of electrical angle of the resolver by inputting a command. A resolver detection error correction device characterized by: