JP3345559B2 - High division encoder device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an encoder for detecting the position of a moving body, and more particularly to a high division encoder device capable of realizing a high division of a conventional encoder.

[0002]

2. Description of the Related Art In a system such as an industrial robot used in a factory, a rotational position of a servo motor or a moving position of a linear motor that moves linearly is detected to obtain position data used for the control. An encoder is used. In a typical photoelectric rotary encoder, four photodiodes that receive light through 8192 slits provided on a rotating disk are arranged so as to be shifted by an electrical angle of 90 degrees, respectively, and have an electrical angle of 90 degrees as shown in FIG. The shifted A ', B', A '* and B' * signals are obtained. Based on the A, B, A * and B * signals (FIG. 3) obtained by amplifying this current signal, performing I / V conversion, and further operating and amplifying, two types of pulse trains 180 ° out of phase with each other. The absolute position data obtained by obtaining an encoder signal consisting of the following, and further converting the encoder signal into a pulse train signal of a quadruple frequency using the rising and falling timings of the encoder signal, and dividing 360 degrees of the mechanical angle into 8192 × 4 = 32768 Can be obtained.

[0003]

At present, when a workpiece with higher precision is required, there is a demand for an encoder capable of outputting position data with a higher division. Simply thinking, it is sufficient to increase the number of slits of the rotating disk. However, there is a limit to increasing the number of slits and optically dividing the photoelectric encoder into a high resolution. This is because there is a limit in reducing the slit width of the rotating disk due to the light diffraction phenomenon.

An object of the present invention is to provide an encoder device capable of realizing high-precision and high-resolution division with a simple configuration by performing electrical signal processing and operation on signals without using an optical method. To provide.

Another object of the present invention is to provide an encoder device capable of quickly obtaining high-divided position data with a small-scale arithmetic device by relatively simple arithmetic operations.

[0006]

SUMMARY OF THE INVENTION The present invention provides a method for measuring an electric angle of 90.
A signal, B signal, A * signal (inverted signal of signal) and B * signal (inverted signal of B signal) of sinusoidal wave shape with different phase
An encoder for outputting absolute position data based on an encoder signal including: and high division absolute position data output means for further dividing between absolute position data continuously output from the encoder and outputting high division absolute position data The high-division encoder device provided with In the present invention, the high division absolute position data output means receives the A signal, the B signal, the A * signal, and the B * signal, compares the signals, and determines the intersection of the A signal and the B signal, and the B signal and the A * signal. , An intersection of the A * signal and the B * signal, and a comparator circuit that outputs a signal switching signal at the intersection of the B * signal and the A signal. The signal switching circuit outputs A based on the signal switching signal output from the comparator circuit.
A signal, a B signal, an A * signal, and a B * signal are selectively switched to output a reference voltage signal that repeatedly appears in the order of A signal → B * signal → A * signal → B signal. The reference voltage signal output from the signal switching circuit is converted into a predetermined number of bits (a predetermined sampling frequency) using an A / D conversion circuit.
A / D conversion is performed. The memory means latches and stores the switching voltage value of the A / D converted reference voltage signal when the signal switching signal is output. The arithmetic means includes: a voltage value of an A / D-converted reference voltage signal when the request signal is input from the control unit side to the encoder device; and a type of signal selected by the signal switching circuit;
The relative position data at the time of inputting the request signal is calculated based on the immediately preceding (one cycle before) four switching voltage values stored in the memory means when the request signal is input, and the relative position data is calculated. The target position data is added to the absolute position data output from the encoder before the request signal is input, and output as high division absolute position data.

More specifically, the encoder is preferably a slit transmission type photoelectric encoder, and includes a rotary encoder for detecting rotation of a rotating body and a linear encoder for detecting movement of a linear moving body. The encoder outputs absolute position data based on an encoder signal including an A signal, a B signal, an A * signal, and a B * signal that differ in phase by 90 degrees in electrical angle in accordance with the rotation of a rotating body as an object to be detected. . In order to obtain the A signal, the B signal, the A * signal, and the B * signal, typically, an LED is used as a light emitting element and a photodiode is used as a light receiving element. The four photodiodes A, B, A *, and B * are arranged so as to be out of phase by 90 degrees in electrical angle. The four photodiodes A, B, A *, B * are arranged, for example, in the rotation direction,
Alternatively, two sets may be arranged in the rotational direction, and two sets may be arranged in the radial direction. In the case of a rotary encoder, the number of slits is usually 8192. Four photodiodes A,
The waveforms of the A signal, B signal, A * signal, and B * signal obtained from B, A *, and B * are ideally triangular waves, but the peak portion is actually distorted as shown in FIG. It becomes sinusoidal.

The high division absolute position data output means includes a comparator circuit, a signal switching circuit, an A / D conversion circuit,
It has a memory means and an arithmetic means. The comparator circuit receives the A signal, the B signal, the A * signal, and the B * signal as inputs and compares the signals. Since the signals are out of phase by 90 degrees in electrical angle, the voltage of each signal fluctuates relative to each other when the moving body moves and changes its position. The comparator circuit determines the intersection of the A signal and the B signal,
A signal switching signal is output at the intersection of the B signal and the A * signal, the intersection of the A * signal and the B * signal, and the intersection of the B * signal and the A signal. Here, the “intersection” can be rephrased as signal coincidence.

The signal switching circuit outputs a reference voltage signal based on the signal switching signal output from the comparator circuit. This reference voltage signal is basically configured by partially combining the A signal, the B signal, the A * signal, and the B * signal. Therefore, the A signal, the B signal, the A * signal, and the B * signal are selectively switched. That is, the reference voltage signal repeatedly appears in the order of A signal → B * signal → A * signal → B signal. The timing of each switch is
It is based on the signal switching signal, and is switched from the A signal to the B * signal at the intersection of the A signal and the B * signal.
At the intersection with the * signal, the signal switches from the B * signal to the A * signal.

The reference voltage signal is output to an A / D conversion circuit, A / D converted with a predetermined number of bits, and converted into a digital signal. If the predetermined number of bits is, for example, 8 bits,
The reference voltage signal will be divided by 256.

The memory means latches and stores the switching voltage value of the A / D converted reference voltage signal when the signal switching signal is output from the comparator circuit. This latch may use a latch circuit.

When a request signal is input from outside, the calculating means calculates high division absolute position data at that time. First, the arithmetic means determines the type of signal selected by the signal switching circuit when the request signal is input, that is, the signal switching signal at that time, and sets the reference voltage signal to A
A signal, a B * signal, an A * signal, or a B signal is selected. By this determination, how many times the request signal is generated in the electrical angle (for example, 90 degrees to 1 degree)
(Request signal is generated in the 80 degree section).
As a result, the voltage value of the reference voltage signal cannot be used to determine
The generation section of the request signal can be specified. Then, the voltage value of the reference voltage signal when the request signal is input and the four switching voltages immediately before (up to 360 degrees before the electrical angle one cycle before) stored in the memory means when the request signal is input. Depending on the value, for example, 9 in electrical angle
At which position between 0 ° and 180 ° the request signal is generated, the result is used as relative position data.

In this operation, for example, the total amount of change in the voltage of the reference voltage signal from the start of the cycle to the input of the request signal with respect to the total amount of change in the voltage of the reference voltage signal in one cycle before one cycle is calculated. Is obtained, and the ratio is multiplied by the number of divisions (the number determined by the number of bits of the A / D converter: 256 × 4 for 8 bits) to obtain relative position data. Next, the relative position data is added to the latest absolute position data output from the encoder before the request signal is input to obtain high division absolute position data. The high division absolute position data is output to an external control device that has transmitted the request signal, for example, a servo amplifier or the like, and is used for controlling the moving body.

In the calculation by the calculation means, it is assumed that the voltage between the switching voltage values changes linearly, and the relative position indicating the relative position from the position indicated by the previously input absolute position data is assumed. Computing the target position data,
Calculation becomes easy.

The A signal, B signal, A * signal and B * signal are ideally triangular waves. However, actually, the peak portion of each waveform is distorted, and even if the triangular wave A, A *, B, and B * signals are used as they are to achieve high division, the accuracy is not high. Therefore, in the present invention, only the portion of each triangular wave having good linearity is used.

That is, the reference voltage signals are A signal, B signal, A * signal and B signal whose phases are shifted by 90 electrical degrees.
* Since the signals are switched at the intersections,
The peak portions of the A, B, A *, and B * signals where the distortion is large are truncated, and each switching voltage value becomes a peak, and a triangular waveform changes almost linearly between the peaks. Therefore, even if the calculation is performed by regarding each of the peaks as a straight line, a large error does not occur. As a result, relative position data can be quickly calculated by a simple calculation.

[0017]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. FIG. 1 is a circuit block diagram showing a main part of one embodiment of a high division encoder device according to the present invention. In the figure, the high division encoder device comprises an encoder 1 and a high division absolute position data output device 2.

The encoder 1 is a rotary slit transmission type optical rotary absolute encoder,
It has a function of calculating and outputting absolute position data based on these signals in addition to encoder signals of so-called A, B, and C channels. The encoder 1 includes a photodiode unit 11 for position detection and an operation amplification circuit 12 for operationally amplifying an output from the photodiode unit 11. In addition, a photodiode for detecting the origin (for the C channel) and the like are provided, but the description is omitted here.

The photodiode unit 11 is composed of four photodiodes A, B, A *, and B * which are arranged so as to be out of phase by 90 degrees in electrical angle. These four photodiodes A, B, A *, and B * are, for example, arranged along a plurality of slits provided on a rotating plate.
They may be juxtaposed in the order of A *, B, B *. Similarly to the photodiodes A and A * arranged along the circumferential direction of the rotating plate, two sets of photodiodes B and B * arranged along the circumferential direction of the rotating plate are arranged in the radial direction. It may be configured to be long so as to face the two sets. In this example, the number of slits is usually 8192 in each case. Each photodiode A, B, A *, B
* Generates A 'signal, B' signal, A '* signal, and B' * signal which are current signals 90 degrees out of phase as shown in FIG.

The differential amplifying circuit 12 performs I / V conversion on the A 'signal and the A' * signal, respectively, and further differentially amplifies them, and performs I / V conversion on the B 'signal and the B' * signal, respectively. The signal is further differentially amplified, and the A signal, the B signal, and the A signal shown in FIG.
Create * signal and B * signal. That is, the A 'signal, A' * signal or B 'signal and B' * signal output from the four photodiodes of the photodiode unit 11 are currents having phases shifted by 90 degrees as shown in FIG. Signal. However, these current signals are
It is as small as 0 μA and cannot be used as it is without amplification. It fluctuates under the influence of temperature change. Therefore, the signals A'-A '* = A signal, A' *-A '= A * signal, B'-B' * = The B signal and the B ′ * − B ′ = B * signal are obtained.

According to the conventional encoder device, as described above, four signals are generated based on the A signal, the B signal, the A * signal, and the B * signal.
A double frequency pulse signal was created to obtain absolute position data. However, in the present embodiment, in order to obtain position data with a higher division, the A signal, B signal, A * signal, and B * signal are subjected to arithmetic processing by the high division absolute position data output device 2, and first, It is characterized in that highly divided relative position data is obtained, and the relative position data is added to the absolute position data.

The high division absolute position data output device 2 comprises a comparator circuit 21, a signal switching circuit 22, an A / D conversion circuit 23, a latch circuit 24, an operation / command circuit 25, a memory circuit 26, and a signal input / output circuit 28. . The absolute position data 27 is obtained from the encoder 1 by, for example, the method described above. The latest absolute position data 27 may be stored in a dedicated memory, or may be stored in the memory circuit 26.

The output from the differential amplifier circuit 12 is ideally a triangular wave A signal, A * signal, B signal, and B * signal. However, in actuality, each waveform has a distorted peak portion, and has a sine wave shape as shown in FIG. For this reason, even if the triangular wave A signal, A * signal, B signal and B * signal are used as they are and high resolution is obtained, the precision is not easily obtained. It gets complicated. Therefore, in the present invention, as shown in FIG. 4A, only the portions having good linearity of the A signal, the A * signal, the B signal, and the B * signal are used.

First, the comparator circuit 21 compares the A signal, the B signal, the A * signal, and the B * signal input from the encoder 1 and determines the intersection of the A signal and the B signal (point of coincidence of levels). , B signal and A * signal (level coincidence point), A * signal and B * signal intersection (level coincidence point), and B * signal and A signal intersection (level coincidence point) Outputs a signal switching signal. Specifically, in the comparator circuit 21, four A signals, A * signals, B signals,
Among the B * signals, the A signal is compared with the B signal, the A signal is compared with the B * signal, the A * signal is compared with the B * signal,
The A * signal and the B signal are compared to obtain signals as shown in FIGS. The “A> B” signal in FIG. 4B is a high period during which the A signal is larger than the B signal.
And the “A> B *” signal in FIG.
The signal rises high during the period when the signal is larger than the B * signal, and the “A *> B *” signal in FIG.
The * signal rises high during the period when the signal is larger than the B * signal, and the “A *> B” signal in FIG.
The signal rises high while the signal is larger than the B signal. To compare these, the comparator circuit 21 includes four comparing means.

Based on the rising positions of the four signals shown in FIGS. 4B to 4E, four-divided signal switching signals are obtained. The comparator circuit 21 is configured as shown in FIGS.
A signal synchronized with the rise of each signal shown in (E) is output as a signal switching signal. In this example, FIGS.
Since the four rising portions (the portions marked with circles in the drawing) shown in (E) are used, the accuracy of the signal switching signal is increased.

The signal switching circuit 22 selectively switches the A signal, the B signal, the A * signal and the B * signal based on the signal switching signal output from the comparator circuit 21, and the A signal → B * signal → A A reference voltage signal RVS [FIG. 4 (A)] that appears repeatedly and repeatedly in the order of signal → signal B is created. More specifically, the signal switching circuit 22 switches the A signal, the A * signal, the B signal, and the B * signal based on the switching signal output from the comparator circuit 21 to generate the voltage reference signal shown in FIG. Generate signal RVS. FIG.
The rise of the “A> B” signal in FIG. 4B is a timing signal indicating the stop of use of the B signal and the start of use of the A signal, and the rise of the “A> B *” signal in FIG. The timing signal indicates the stop and the start of use of the B * signal, and the rise of the “A *> B *” signal in FIG.
The timing signal indicates the stop of the use of the * signal and the start of the use of the A * signal. The rise of the "A *>B" signal in FIG. 4E indicates the timing signal indicating the stop of the use of the A * signal and the start of the use of the B signal. Becomes

The voltage reference signal RVS generated by switching the A signal, A * signal, B signal, and B * signal based on these switching signals is Va, Vb *,
It has four peaks, Va * and Vb, and occurs at the same cycle as one cycle of the A signal, A * signal, B signal, and B * signal.
That is, the signal is generated when one slit moves by 360 degrees in electrical angle. By using such a voltage reference signal RVS, it is possible to avoid using peak portions of the A signal, the A * signal, the B signal, and the B * signal.

The A / D conversion circuit 23 converts the reference voltage signal RVS into an A signal for digital processing (microcomputer processing).
/ D conversion. Specifically, the A / D conversion circuit 23 converts the reference voltage signal RVS output from the signal switching circuit 22 into a digital signal with 8 bits. That is, the A / D conversion circuit 23 outputs the reference voltage signal RVS for 1/4 cycle to 8
The digital signal is divided by a bit, that is, divided by 256. Therefore, the number of divisions for one cycle is 256 × 4
Becomes

The latch circuit 24 converts Va, Vb *, Va *, V * from the voltage reference signal RVS based on the switching signal.
The data of the four peak voltages b are sequentially latched. The latched value is stored in the memory circuit 26. The memory circuit 26 stores at least the four peak voltages immediately before, that is, one cycle before, and the value to be latched in the current cycle. Then, the previous stored data is updated with the latched value. The value obtained by A / D conversion of the reference voltage signal RVS may be sequentially stored in the memory circuit 26 as it is. Actually, the measurement data used one cycle before (one cycle: 360 degrees in electrical angle), that is, one slit before is used as the basis of the calculation.

When the request signal X is input from the signal input / output circuit 3 of the servo amplifier via the signal input / output circuit 28, the operation / command circuit 25 calculates and creates the high division absolute position data at that time. The signal is further converted into angle data and output to the signal input / output circuit 3 of the servo amplifier through the signal input / output circuit 28.

That is, the signal input / output circuit 3 of the servo amplifier
When the request signal X is input from the CPU, the operation / command circuit 25 calculates and outputs the rotational position at the position where the request signal X is input. First, the absolute position data Dab1 is input from the encoder 1. The absolute position data 27 is output from the encoder 1, and the encoder 1 adds or subtracts the absolute position data 27 according to the rotation direction every time one slit rotates by 360 electrical degrees. In FIG. 4A, Dab1 and Dab2 are the update positions of the absolute position data, and one slit between Dab1 and Dab2 has an electrical angle of 360.
One rotation or one cycle. In the present embodiment, this one cycle is further divided by the division number 256 × 4 to achieve high division.

The operation / command circuit 25 first generates an A signal, a B * signal, and an A * signal using the reference voltage signal RVS depending on the type of signal selected by the signal switching circuit 22 when the request signal X is input. , B signal is selected. By this specification, it is determined how many sections in the electrical angle the request signal X has been generated. As a result, the request signal X generation section [(1) in FIG. 4 (A)] cannot be determined only by the voltage value of the reference voltage signal RVS.
To (4) section] can be specified. Then, the voltage value of the reference voltage signal RVS when the request signal X is input and the voltage value immediately before being stored in the memory unit 26 when the request signal X is input (up to 360 degrees before the electrical angle one cycle before). Four switching voltage values (Vb, Va, Vb
*, Va *), at which divisional position in the period of one cycle the request signal was generated is calculated, and the calculated position is used as relative position data.

More specifically, the operation / command circuit 25 adds the absolutely divided relative data to the absolute position data 27 to obtain the request signal X.
Calculates the high division absolute position data at the time when is output.
Where the request signal X occurs in one slit depends on which of the signals in FIGS. 4B to 4E is used as the signal switching signal and the voltage Vx when the request signal X is input. (The signal selected from the phase relationship of the switching signal at that time, for example, the voltage of the A signal in the example of FIG. 4A) and four arithmetic expressions prepared in advance.

Assuming that the request signal X is input at the position shown in FIG. 4A, since the rising portion of the "A>B" signal is used as a switching signal, the period (1)
That is, it is possible to specify that the request signal has been generated (in a section of 0 to 90 degrees in electrical angle). From this specification, the relative position data is calculated from the voltage value of the reference voltage signal to which the request signal is input and the four switching voltage values one cycle before using the following equation (1). Note that four types of arithmetic expressions are prepared according to the type of the selected signal and the section in which the request signal X is input. That is, the equation (1) used when the request signal is input at a position between X0 degrees and 90 degrees,
Expression (2) used when input is made at a position of 80 degrees, Expression (3) used when the request signal is input at a position of X180 degrees to 270 degrees, and a position where the request signal is 270 degrees to 360 degrees Expression (4) used when the input is made as follows.

[0035]

(Equation 1) In the case of FIG. 4A, the above equation (1) is selected, the actual measured voltage Vx is substituted, and the relative position data DX is calculated. These arithmetic expressions are expressions for determining how far the position where the request signal X has been input is from the absolute position Dab1, and obtaining the number of divisions according to the position. The basis of this equation is that the sum of the voltages from Dab1 to Dab2, that is, the total change (denominator), and the sum of the voltages from Dab1 to the position where Vx is input, that is, the total change (numerator)
Is obtained by multiplying this ratio by the number of divisions determined by the number of bits of the A / D converter. That is, in this calculation method, the peaks Va and Vb of the triangular wave are calculated.
The voltage change to each of *, Va *, and Vb is considered to be linear. As described above, the triangular wave A signal, A
The * signal, the B signal, and the B * signal have a sine wave shape with a distorted peak portion, but the other portions are almost linear. Therefore, the comparator circuit 21 and the signal switching circuit 22 cut off a portion having a large amount of distortion to obtain a reference voltage signal RVS that leaves only a straight line portion that is easy to calculate. Thus, relative position data is obtained.

In this device, the division accuracy is determined by the A / D conversion circuit 2
3 depending on the number of bits. In other words, if the data is divided by 8 bits, it is possible to considerably increase the number of divisions into 256 × 4 during one cycle, and 8192 × 256 × 4 = 8388608 in one rotation. If the number of bits of the A / D conversion circuit 23 is further increased, further division can be achieved.

In the operation / command circuit 25, the absolute position data 27 is added to the relative position data DX obtained based on the operation expression to obtain high-divided absolute position data, which is further converted into angle data. Output to the input / output circuit 28. In this example, the operation / command circuit 25 constitutes operation means.

In the present embodiment, the present invention is applied to an optical absolute position encoder. However, the present invention can be applied to other types of encoders.

[0039]

As described above, according to the high-division encoder apparatus of the present invention, it is possible to realize high-precision high-division with a simple configuration by electric signal processing and calculation. .

[Brief description of the drawings]

FIG. 1 is a circuit block diagram showing a main configuration of an embodiment of a high division encoder device according to the present invention.

FIG. 2 is a diagram illustrating outputs of four photodiodes according to the embodiment of FIG. 1;

FIG. 3 is a diagram showing a signal obtained by performing I / V conversion on an output of the photodiode in FIG. 2;

FIG. 4A is a diagram showing a reference voltage signal RVS obtained by using an A signal, a B signal, an A * signal, and a B * signal for explaining the principle of the present invention, and FIGS. (E) is a diagram showing four types of signal switching signals, respectively.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Encoder 11 Photodiode unit 12 Differential amplifier circuit 2 High division absolute value data output device 21 Comparator circuit 22 Signal switching circuit 23 A / D conversion circuit 24 Latch circuit 25 Arithmetic / command circuit 26 Memory circuit 27 Absolute position data 28 Signal Input / output circuit 3 Signal input / output circuit of servo amplifier

──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Tomoharu Sekine 1-15-1 Kita-Otsuka, Toshima-ku, Tokyo Inside Sanyo Electric Co., Ltd. (72) Inventor Shoji Ito 1-1-55 Kita-Otsuka, Toshima-ku, Tokyo No. 1 Sanyo Electric Co., Ltd. (56) References JP-A-1-163615 (JP, A) JP-A-62-249008 (JP, A) JP-A-6-74917 (JP, U) (58) Field surveyed (Int.Cl. ⁷ , DB name) G01D 5/00-5/62 G01B ⁷ /00-7/34 G01B 11/00-11/30 G01P 1/00-3/80

Claims (3)

(57) [Claims] An encoder that outputs absolute position data based on an encoder signal including a sine wave A signal, a B signal, an A * signal, and a B * signal that differ in phase by 90 degrees in electrical angle; A high division absolute position data output unit that further divides the absolute position data to be output to output high division absolute position data, wherein the high division absolute position data output unit includes: The A signal, the B signal, the A * signal, and the B * signal are input and each signal is compared, and the intersection of the A signal and the B signal is calculated.
A comparator circuit for outputting a signal switching signal at an intersection of the B signal and the A * signal, an intersection of the A * signal and the B * signal, and an intersection of the B * signal and the A signal; Based on the signal switching signal output from the circuit, the A signal, B signal, A * signal and B signal
* The signal is selectively switched and the A signal → the B *
A signal switching circuit for outputting a reference voltage signal which repeatedly appears in the order of a signal → the A * signal → the B signal; and an A / D converter for A / D converting the reference voltage signal with a predetermined number of bits.
/ D conversion circuit; memory means for latching and storing the switching voltage value of the reference voltage signal which has been A / D converted when the signal switching signal is output; and the reference when the request signal is input. A relative value when the request signal is input based on the voltage value of the voltage signal and the signal switching signal and the four immediately preceding switching voltage values stored in the memory means when the request signal is input. Calculating means for calculating target position data, adding the relative position data to the absolute position data output from the encoder before the request signal is input, and outputting the resultant as the high division absolute position data. High division encoder device. 2. The calculation means, assuming that the voltage between the switching voltage values changes linearly, and indicating the relative position from the position indicated by the absolute position data previously input. The high division encoder device according to claim 1, wherein the high division encoder device is configured to calculate relative position data. 3. An optical absolute encoder that outputs absolute position data based on an encoder signal including a sine wave A signal, a B signal, an A * signal and a B * signal having phases different by 90 degrees in electrical angle, and A high division absolute position data output means for further dividing the absolute position data continuously output from the encoder to output high division absolute position data, wherein the high division absolute position data The output means receives the A signal, the B signal, the A * signal, and the B * signal as inputs and compares the signals. The signal levels of the A signal and the B signal match, and the output signals of the B signal and the A * signal are compared. A signal switching signal is output each time the signal level matches, the signal level of the A * signal matches the signal level of the B * signal, and the signal level of the B * signal matches the signal level of the A signal. A signal, a B signal, an A * signal and a B signal based on the signal switching signal output from the comparator circuit.
* The signal is selectively switched and the A signal → the B *
A signal switching circuit for outputting a reference voltage signal which repeatedly appears in the order of a signal → the A * signal → the B signal; and an A / D converter for A / D converting the reference voltage signal with a predetermined number of bits.
/ D conversion circuit, a latch circuit for latching a switching voltage value of the reference voltage signal which has been A / D converted when the signal switching signal is output, and a memory means for storing the voltage value latched by the latch circuit An A / D-converted voltage value of the reference voltage signal when the request signal is input, and the signal switching signal;
When the request signal is input, relative position data is calculated based on the four switching voltages one cycle before stored in the memory means, and the relative position data is input to the request signal. Calculating means for adding the absolute position data to the latest absolute position data output from the encoder before output as the high-divided absolute position data, wherein the calculating means outputs a signal switching signal when a request signal is input. When the request signal is generated, the signal selected by the signal switching circuit is specified, and then the request signal is started after the cycle is started with respect to the total change in the voltage of the reference voltage signal in one cycle one cycle before. The ratio of the reference voltage signal to the total amount of change in the voltage up to the time when is input is determined, and this ratio is determined by the number of bits. A high division encoder device wherein the relative position data is obtained by multiplying the division number.