JPS6162352A - Phase error compensating method of resolver - Google Patents

Abstract

PURPOSE:To compensate the phase error of harmonic wave degree of waves erased including basic wave by controlling an exciting current at equal time interval in every cycle of every step over twice of phase error harmonic wave degree to be erased. CONSTITUTION:A ROM41 in a phase error compensator 4 stores to produce a 8-bit output 32: ideltakcos(kdelta) in response to a 4-bit input kdelta from th K-th phase error kdelta in advance by the measurement. A ROM42 stores to produce 8-bit output 32: ideltaksin(Kdelta) in response to a 4-bit input kdelta. The maximum degree of the phase error to be erased is 8, and an exciting current vector is corrected at 16 equal time points. Thus, the phase error of harmonic wave degree to be erased including a basic wave is compensated.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a method for compensating phase errors in a resolver that detects the rotational phase of a rotating body, particularly in an inductor type resolver.

[Problems with conventional technology]

Simple position detectors with a small number of grooves and conductors on each pole, such as inductor-type Solpaya Inductosin, contain many phase errors, and removing them would complicate the structure of the machine. It's not easy either.

These errors have good reproducibility, that is, repeatability, so
Although it is possible to compensate for this on the detection side, there is a drawback that the compensation circuit must be extremely complex in order to achieve high resolution and high coverage.

[Purpose of the invention]

Here, the present invention overcomes the difficulties of the conventional arrangement, can compensate for phase errors of all harmonic orders to be canceled including the fundamental wave, has a simple circuit, and can easily set the amount of compensation. To provide a resolver phase error compensation method,
That purpose.

[Criteria of invention]

The present invention eliminates the phase error caused by the structure of the resolver body.
Using the excitation current waveform as a secondary axis? In the l'Ii compensation method, when the mutual inductance of the detection winding wy and the excitation windings Wα and Wβ is Mα1Mβ, and the excitation currents are 1α and 1β, 1”a + i”/ = conet (constant) 1β /
This is a resolver phase error compensation method in which the side legs of 1α = Mβ/Mα are performed at equal time intervals in each cycle with a step number at least twice the phase error harmonic order to be canceled.

[Principle of the invention]

The principle of the present invention is as follows.

The machine constants of the current-excited resolver are M and Mβ(θ)
It is expressed as

The magnetic flux σ interlinking with the detection winding is H=1α・Mα (eaves 1β・M〆0) ・・・・・・・・・
・(1) Detection voltage eI5 is ea = a3'l/at (Since t is 1a', Cg = 11α/A t - Mα (θt←d1β/at
-M, 6i θ) ...(2) From this, when the vector (11α/it, i1β/it) is orthogonal to the vector (Mα(g)1Mβ(θ)), gu=O.

By the way, in the phase modulation method, the time during which the detection voltage θ=0 is measured.

1 = i e ' - constant f 6 Root, when σ = o, the vector (1α, 1β) is the vector (Mα (σ)
.

M path becomes parallel to θ)).

Therefore, 1β/1α=M〆θVytα(丙 ・・・・・・・・・
Group... Old... (3) Here, the phase error is tan CM β(6)/M a (σ)]-〇, but θ found from the phase of the current is tan(iβ/
1α], so tan [:Mβ(v)/M a (σ):l
-tan Ciβ/4α) = .

becomes.

〔Example〕

A block diagram showing a circuit configuration in an example of the present invention is shown in the figure.

l is an oscillator that generates a frequency of 16f8, 2 is an l/16 frequency divider that creates a reference time axis, and 3 is an output stage with 4 pits.
Counter consisting of 6 steps, 5=0゜■,...
15, the ROM is stored in advance so that when δ=2π/16, an output of 8 pits 15iak δ is output corresponding to δ in response to the 4 pits human power from the counter 5.

6 is 4 bits human power and δ corresponds to 8 bits 1 cosk
ROM pre-stored to output δ.

7 and 8 are D/A converters, 9 and 10 are adders/subtractors, 11 and 12 are current amplifiers, and 13 is a resolver.
β is β-phase excitation current, 1α is α-phase excitation current, θ is detection voltage, 14Fi bandpass filter, 15Vil/16
A synchronous rectifier synchronously rectifies the output of the band pass filter 14 in accordance with the reference time axis from the frequency divider 2, 16 is a phase output, and 4 is a phase error compensation circuit.

This compensation circuit 4 consists of the following elements.

41 is the output 32·1 of 8 pits corresponding to δ from the 2nd phase error δk obtained by measurement in advance.
RoM stored to output δkcos (km)
, 42 is a 4-pit input so as to output an 8-bit output 32·1 δksin (km) corresponding to δ; 9 ROM; 43 and a shrine are D/A converters;
is an attenuator that attenuates the input to l/32.

The highest order of the phase error to be erased is assumed to be 8, and the excitation current vector is corrected at equal time intervals of 16 points.

For this purpose, first measure Mα(θ), M7yF),
For σ = for δ, Mα (km) + jMβ (km) = M (km) = M (:
GXI))"kδ+δk) By the way, M is a real number, a+T
= δ and ω is the average excitation current angular frequency.

That is, the current vector has a good deviation in addition to a direction that is shifted by the phase error δ from the phase detected at each time point.

δk (Since δ = ωτ, 1 [e: rp] j ('+ δk ) = 1 [e xp ]
j km f (1+jδk)
'd 1(exp)""", jδ becomes.

Here, assuming that δk is 1°=−1−Io, 016 rad, if this is to be reduced by compensation, a resolution of is required.

8 bits for setting kδ, 1/3 8 bits for correcting δ
If used with attenuation to 2, the resolution of 13 pits will be better,
It is sufficient to add Δiα=−1δksin (kδ) to α and Δ1β=1δkcO5(kδ) to 1β.

A portion 14 surrounded by a broken line in the figure is this compensation circuit.

While M changes continuously, 1 is added only in a stepwise manner, and the 16 times higher harmonic generated is removed by inserting a bandpass filter 14 on the detection side.

As another embodiment of the present invention, in order to compensate for unbalance, offset, etc. of the excitation circuit and signal processing circuit, the ROM4
1. By including this amount in the ROM 42 and adding +Δk to δ, the damage is good (Δk is the compensation amount).

This can also be automatically measured and tuned.

The following means can be considered as another embodiment of the present invention.

■ The figure is a diagram of the principle, so it is complicated software, but if you process it with software using a microcomputer, it will become extremely complex.

(2) Similar compensation can be performed in the amplitude modulation method.

@ The figure shows current excitation, but a similar method is possible with voltage excitation.

〔Effect of the invention〕

Thus, according to the present invention, a number of effects can be obtained as described below, and the present invention will be of great benefit to the field concerned.

[Co., Ltd.] Detectors such as inductor resolvers and inductosin can be manufactured easily and with high accuracy at low cost.

■ Since the structure of the detector is simple and good, it is possible to make it extremely small. In addition, a linear type can be easily realized.

■ Temperature correction is not necessary because the structure is not affected by the temperature change of inductor type resol/resistance.

■ The circuit is simple and simple, and the cost is low.

[Brief explanation of drawings]

The figure is a block diagram showing a circuit configuration in an embodiment of the present invention. 1... Oscillator, 2... Frequency divider, 3... Counter, 4
...Compensation circuit, 41, 42...FtOM, 43.
・14...D/A converter, 45, 46...Attenuator, 5.6...ROM, 7.8...D/A converter, 9
．． 10... Adjustment armor, 11, 12... Current Anne 7
', 13... Resolver, 14... Band pass filter, 15... Synchronous rectifier, 16... Output.

Claims (1)

[Claims] 1. In a method of compensating the phase error caused by the structure of the resolver body by controlling the excitation current waveform, the mutual inductance of the detection winding wθ and the excitation windings wα and wβ is set to M
When α and Mβ are excitation currents and iα and iβ are excitation currents, i^2α+i^2β=const (constant) i^2α/iα=Mβ/Mα. A phase error compensation method for a resolver characterized by performing the steps at equal time intervals every cycle at least twice the number of steps.