JPS62172218A - Method for detecting position by resolver/digital converter - Google Patents

Abstract

PURPOSE:To enable highly accurate positional detection by a simple constitution by enabling high frequency response, by converting only the position signal of a resolver to a digital value with an A/D converter and processing all of residual operations with a microprocessor. CONSTITUTION:The exciting signal of a resolver 1 is imparted from a microprocessor 3 through a D/A converter 4 and the conversion start signal ST of an A/D converter 2 applying A/D conversion to the position signal of the resolver 1 is outputted from an address decoder 35. At the same time, an address value is latched by an address latch circuit 36 so as to enable an operational processor 37 to identify the phase angle of an exciting signal REF and taken in the processor 37. Operation calculating a digital position from thus obtained A/D conversion value of the resolver position signal and the address value identifying the phase of the signal REF is performed in the processor 37 according to a Runge-Kutta method and the position signal Dp of the resolver 1 is outputted from an output port.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention is directed to converting an analog output signal of a resolver into a tracking type resolver/digital converter (hereinafter referred to as rR).
/D converter) to obtain parallel digital position signals.

[Conventional technology]

Tracking method R/ using conventional analog processing circuit
The D converter is a high-precision trigonometric function multiplier.

It consists of an error amplifier, a phase detector, a waveform shaping circuit, a wideband voltage controlled oscillator, an addition/subtraction counter, etc.

In order to create a position detector with high precision and quick response, it is important not only to perform high-precision signal processing in each circuit, but also to increase the loop gain of the converter and extend the phase characteristics to high frequencies.

[Problems that the invention attempts to solve]

However, before]! In the conventional analog circuit configuration, filter characteristics including integration are unavoidable, and it is difficult to increase the gain and speed up the response. Furthermore, in terms of accuracy, there are factors that reduce accuracy everywhere.

<≧That is, the disadvantage of the conventional example is that the following speed is slow (compared to the change of the party leader).

In addition, the conventional circuit is mostly composed of an analog signal circuit, and the A/D conversion is composed of a V/f converter and a counter. requires precision ζf for all multipliers, V/f converters, amplifiers, etc.

Moreover, if the conventional configuration is replaced with a digital circuit, the tracking speed of position changes will be slower.
Accuracy is also not good.

However, the frequency characteristics of the conventional tracking type R/'D converter, which performs phase lead/delay compensation in the control loop, are 300 to 4001-1 z Pil.
Maro is the limit.

Furthermore, since analog estuary processing is performed, highly accurate component parts are required, and the circuit configuration is complicated.

Here, the present invention overcomes the drawbacks of the conventional means and uses a high-precision Iα digital position signal with a large number of pins.
Its purpose is to provide a position detection method using a D converter.

(Means for Solving the Problems) The present invention provides a part for directly converting an alilog signal, which is a position detection signal of a resolver, into a digital signal using a △/D-] converter and a D/△ supplying an excitation signal for the resolver. Except for the converter section, all remaining circuit operations are processed digitally by a microprocessor.Therefore, the accuracy of the R/D converter is solely due to the accumulation of electricity in the Δ/D converter, and the accuracy of the R/D converter is limited to 1KHz or more. Frequency I, i5 answer R/
This is a position detection method using a D converter.

[For production]

Synchronizing with the phase of the resolver's excitation signal, the high amplitude points of the resolver's output signal are summed to improve the detection accuracy, and furthermore, the signal Brollelli is used to perform four arithmetic operations such as trigonometric functions, and the Runge-Kutta method is used. The position angle is determined using the zero-position method, and the average value of several measurements is used as a digital position signal.

[Actual example]

A block diagram showing the circuit +i4 configuration in one embodiment of the present invention is shown in FIG.

The excitation signal (REF) of the resolver (RES) 1 is supplied to the clock generator (O20) 31 of the microprocessor 3.
The address counter (A
ROM table (SiNR) 33 that drives DRCNT>32 and stores sine waveform data for each small constant angle.
Pay out the contents of and convert it to D/△ using D/△ converter 4.
It is created by converting.

Among the outputs of the address counter 32, some are input to the address decoder (△I)R1)EC)35, and the position signal g of the resolver 1 is converted to A with a predetermined address value.
Conversion start signal (S
-”r).

At the same time, the address latch circuit (8D RL CI-1) 36 latches an appropriate address value and calculates it, so that the operation propi tree (SIGMPU) 37 identifies the phase angle of the excitation signal (RE I= ). Propituri 3
7 input boat.

Figure 2 shows the timing variations of the excitation voltage (RIEF), resolver position signal (SIN) (CO8), and Δ/D converter conversion start signal (ST).1 Figure 3 shows the lean-pull timing of the resolver signal. Indicates -C.

ω is the angular frequency of the excitation signal and t is the time. The sample index in FIG. 3 is =1.2.・・・・・・、6
At time 0, the address decoder 35 outputs an ST low signal, the conversion of the Δ/L)-1 inverter 2 starts, and the AT1 no latch circuit 3G creates EIJ.

As shown in Fig. 2, the resolver 10 position signal is the result of the excitation signal being modulated in the iE sinusoidal and cosine shapes depending on the position of the resolver 1, and is detected at the location where the amplitude of this transmission 8 is large. , the detected viscosity and detection sensitivity are set to 1.

This J/D of the resolved resolver position signal
FIG. 4 shows a flowchart of the calculation p':l algorithm that outputs the digital position from address 1, which identifies the phase of conversion 1 and excitation signal ¥3, to 0.

The basic calculation for determining the position angle is to obtain the phase angle of detection. , (here, ΔP is a small value determined by the error in digital calculation and the error in A/D conversion.) All of these steps 401 to 412 are executed in the calculation block 37, and the output port is is output as the position signal -D of the resolver 1. In addition, θ is the resolver position angle, Eo is the amplitude of the excitation voltage, K is the detection voltage constant, and Δ
P is the judgment limit of the calculation method, 5GN(sin(ωt)
・P) kt sign function when 5in(ωt) -P>O 1. sin (ωshi) ・When []<○, it is -1.

(Effect of the invention) Thus, according to the present invention, only the resolver position signal is
Since the digital conversion is performed by the A/D converter and all remaining calculations are processed by the microprocessor, the accuracy of this method depends only on the A/D converter, and the frequency response is b I K +- (Z or more This makes it possible to perform highly accurate position detection with a simple circuit configuration, and to reduce construction costs.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing a circuit configuration according to an embodiment of the present invention, FIG. 2 is a timing explanatory diagram showing each signal voltage waveform, FIG. 3 is a timing explanatory diagram of resolver signal sampling, and FIG. The figure shows flowcharts 1 to 1 of arithmetic processing in an arithmetic processor. 1... Resolver 2... A/D converter 3... Microphone I coprocessor 4... D/Δ converter 31... Clock generator 32... Address counter 33... Sine waveform data storage ROM 35...Address decoder 36...Address latch circuit 37...Arithmetic processor. Applicant's agent Sato - male duck 1 figure ST' yell tool t (same gate) rod 2 figure

Claims (1)

[Claims] In a resolver/digital converter that converts an angle signal from a resolver into a digital value, a phase angle that increases the S/N ratio of the resolver's angle signal in synchronization with the phase of the resolver's excitation signal. Sampling is performed by an A/D converter, and the position angle of the resolver is determined by the Rungekutta method using the address value corresponding to the phase of the excitation voltage of the resolver and the angle signal of the resolver digitally sampled by the previous method. A position detection method using a resolver/digital converter, which is characterized by finding using the zero position method.