JPH04120419A - High resolution absolute value encoder - Google Patents

Abstract

PURPOSE:To enable high resolution detection by using first and second patterns, interpolating respective absolute value data of rough angles and two pseudo-sine waves to obtain absolute value data of a precise angle and coupling absolute data of the rough and precise angles. CONSTITUTION:Photo sensors 41a to 41f are provided per track of a gray code reading slit, while photo sensors 41w to 41z are provided corresponding to four sets of slits. Current signals proportional to an amount of received light are obtained from the photo sensors, and they are converted 21 into voltage to permit voltage signals obtained corresponding to a rotating slit plate. Signals 21a to 21f corresponding to gray regions of these signals are binarized by a comparator 22 to be gray codes 22a to 22f and made to be gray-binary coded 23, which are input to data coupling unit 29 as rough angle data. Signals 21w to 21a corresponding to fringe regions are made into pseudo-sine waves 24a, 24b via a differential amplifier 24, which are interpolated 28 and input to the coupling unit as precise angle data then to be coupled with the rough angle data.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a high-resolution absolute value encoder suitable for use in detecting the angle and position of an object to be measured.

[Conventional technology]

FIG. 7 is a schematic diagram showing an example of a conventional absolute value type encoder for measuring the absolute amount of the displacement state of an object to be measured. In the figure, reference numeral 11 denotes a slit plate on the rotating side, and a plurality of tracks 13 are arranged concentrically on the rotating slit plate 11.
is provided.

Each track 13 is provided with an absolute value pattern, for example, a pattern in which a gray code is formed by a plurality of slits 15 each consisting of a light passing area and a non-light passing area. 18 is a fixed slit plate. 17 is a fixed slit row provided on the fixed slit plate 18, and selectively selected from the slit pattern of each track 16.
A chair 19 to the right of the frontage of the seat meter is a light emitting section having a plurality of light emitting elements, and a reference numeral 21 is a light receiving section having a plurality of light receiving elements.

The light emitting section 19 and the light receiving section 21 are arranged with the rotating slit plate 11 and the fixed slit plate 18 in between, such that each light emitting element and each light receiving element correspond to each opening of the fixed slit row 17, respectively.

In the configuration shown in the figure, the light flux emitted from each light emitting element of the light emitting section 19 passes through the slit 15 and the fixed slit array 17 and enters each light receiving element of the light receiving section 21 . Each light receiving element generates an output signal corresponding to the amount of incident light, and the absolute rotational position of the rotary slit plate 11 is determined from a combination of the output signals.

[Problem to be solved by the invention]

However, if an attempt is made to construct a high-resolution absolute value encoder using only binary codes such as Gray codes, it is necessary to create all the code patterns corresponding to the resolution, making it very difficult to create the patterns. Furthermore, since the number of tracks in the pattern increases, the number of light emitting elements and light receiving elements corresponding to the number of tracks becomes necessary, which increases the cost. Furthermore, due to the phenomenon of light diffraction, there is a limit to the slit width that can be realized.
The minimum diameter of the encoder is determined by the resolution. For example, if the minimum slit width is 10 μm, 100
To realize an encoder with 10,000 divisions, 10 μm x 1
From the calculation of 10,000,100 m, the circumference is 10 m and the diameter is approximately 3 m, far exceeding the dimensions that can be put to practical use.

In addition, in order to obtain correct absolute accuracy, the rotating slit plate 11 and the fixed slit plate 18 must be arranged so that the LSB intervals are equal.
must be adjusted. However, such adjustment is difficult and the LSB spacing becomes uneven, so
The problem is that it is difficult to obtain high absolute accuracy.

The purpose of the present invention is to solve these problems, and to provide a high-resolution absolute value encoder with high absolute accuracy without increasing the size of the entire device.

[Means to solve the problem]

For this reason, the high-resolution absolute value encoder of the present invention forms a first pattern consisting of a binary code with a small number of tracks and a second pattern consisting of a striped code on a moving slit plate. From the first pattern, absolute value data of the rough angle is detected, and from the second pattern, the absolute value data of the coarse angle is detected, which has a cycle that is an integer multiple of twice or more than the LSB of the absolute value data of the rough angle, and has a phase shift of 90 degrees. Detect two pseudo sine waves. Electrically interpolated coarse angle absolute value data is obtained from the latter two pseudo sine waves, and is combined with the coarse angle absolute value data to obtain continuous absolute value data.

In addition, if the positional relationship between the movable slit plate and the fixed slit plate is not accurately matched, the error in reading information between the two most distant tracks in each trunk will be the largest. Therefore, a second pattern is arranged to sandwich the first pattern, two pseudo sine waves with a phase shift of 90 degrees are extracted, and the surge of the two pseudo sine waves is
The movable slit plate and the fixed slit plate are adjusted based on the waveform and +1 can be removed.

Since the movable slit plate and the fixed slit plate are adjusted using the pseudo sine wave obtained from the second pattern as described above, errors in reading the tracks between them are also removed. In this way, since the positional relationship between the movable slit plate and the fixed slit can be accurately grasped based on the Lissajous waveform, the movable slit plate and the fixed slit can be accurately positioned.

By the above method, the number of tracks of the code pattern can be reduced and the moving slit and fixed slit can be accurately positioned, so that a high-resolution absolute value encoder that is compact and has high absolute accuracy can be obtained.

[Example] Hereinafter, an example of a high-resolution absolute value encoder according to the present invention will be described with reference to the drawings.

FIG. 1 is a block diagram of a signal processing system of an absolute value encoder to which the present invention is applied, FIG. Figure 3 is a wired diagram of the standard slit pattern of the encoder, Figure 4 is a diagram of the output voltage signal from the light receiving element, and Figure 5 is the relationship between coarse angle data and fine angle data. It is a diagram.

As shown in FIG. 2, on the rotating slit plate 12, there are striped areas 31, 35 as a second pattern for detecting a pseudo sine wave, and a first pattern for obtaining absolute value data of rough angles. A gray area 36 is provided. A pattern indicating a gray code is formed in the gray area 66. In the striped area 31.35, a plurality of striped patterns are formed at equal intervals with an angle that is half the minimum slit width Wg of the Gray code. The finest putter 7(r) is arranged around the change line.

As shown in FIG. 3, on the fixed slit plate 16, in an area corresponding to the gray area 6'6 of the rotating slit plate 12,
A gray code reading slit 46 having a width less than half the minimum slit width Wg of the gray area 36 is formed. In addition, in the area corresponding to the striped area 61.35,
Slit groups 42, 46, 44, and 45 having the same width and pitch as the pattern of the striped area 6165 are formed. The slit groups 46, 44, and 45 are formed at positions shifted from the slit group 42 by 1/4.2/4.3/4 cycles, respectively.

One photosensor 412 to 41f is arranged for each track of the gray code reading slit 46.

In addition, a total of four photosensors 41W to 41z, one each corresponding to the 4Mi slit groups 42, 43, 44, and 45, are installed.
is located.

A current signal proportional to the amount of light received is obtained from the photosensor group, and as shown in FIG.
is converted to voltage by As a result, the rotating slit plate 1
2 (FIG. 2), the voltage signal shown in FIG. 4 is obtained.

Among these signals, signals 213 to 21f corresponding to the gray region 33 (FIG. 2) are binarized by the comparator 22 to become gray codes 22a to 22f, and further binary encoded by the gray binary converter 26. be done. The gray binary conversion method is described in detail in Japanese Patent Application No. 1-233977 previously filed by the applicant, and according to this method, high-speed conversion processing is possible with a simple configuration.

On the other hand, signal 2 corresponding to striped area 61.35 (FIG. 2)
1W to 21z are input to two sets of differential amplifiers 24 (FIG. 1), and generate pseudo sine wave signals 24a and 2 with a phase shift of 90 degrees, each having one cycle of the bright and dark pattern of the striped area 31.35.
It becomes 4b. These two pseudo sine wave signals 24a, 24
b is subjected to interpolation processing by the interpolation processing unit 28, inputted to the data combination unit 29, and combined with the rough angle data.

(Combining fine and coarse data) From the striped area 61.35 (Fig. 2), absolute value data of fine angles obtained by interpolating the pseudo sine wave signals 24a and 24b are obtained, and from the gray area 33 (Fig. ) gives the absolute value data of the coarse angle. By establishing a certain relationship between these two signals and connecting them, it becomes possible to detect the absolute angle within the -rotation with high resolution. To do this, you can perform the following processing. The relationship between the change point of the absolute value data of the coarse angle and the absolute value data of the fine angle is clear from Figure 4.
This will be explained in more detail with reference to FIG. From this figure, it can be seen that the LSB of the coarse angle data and the MSB of the fine angle data have the same period and are out of phase. Therefore, when the absolute value data of the fine angle increases monotonically, at the point where the maximum value changes to 0, the absolute value data of the coarse angle is always either an odd number or an even number. This is determined by design. Incidentally, when the absolute value data of the fine angle is monotonically decreasing, the same thing can be said in that O changes to the maximum value. Here, for example, if the rough angle data is an odd number at the point where the rough angle data changes from the maximum value to 0 as shown in Figure 5, then if the value of the rough angle data is n, then n is an even number. Then, the coarse angle data is n/2, ■ If n is an odd number and the MSB of the fine angle data is 1, the coarse angle data is (n - 1) / 2, ■ n is an odd number and the MSB of the fine angle data is is 0, the coarse angle data is (n +
1) / 2, it can be seen that it is sufficient to connect it with precise angle data.

Also, if the coarse angle data is an even number at the point where the fine angle data changes from the maximum value to 0, then if the value of the coarse angle data is n, then if n is an odd number, the coarse angle data is n/2. , ■ If n is an even number and the MSB of the fine angle data is 1, the coarse angle data is (n-1)/2, ■ If n is an even number and the MSB of the fine angle data is O, the coarse angle data is ( n+1
) / 2 and take the quotient.

FIG. 6 shows an example of the node configuration of the data coupling section 29. 71 is an adder, and 76 is an AND gate. Also,
This figure shows a configuration in which the coarse angle data is an odd number in that the fine angle data changes from the maximum value to 0 or from 0 to the maximum value.

IPO to IP7 are outputs of precision angle data from the interpolation processing unit 28, where IPO represents the LSB and IF5 represents the MSB. BINO-BIN5 is the coarse angle data output from the gray binary converter 26, where BINO represents LSB and BT N5 represents hi SB. Adder 7
1 sequentially repeats the calculation k1+Bl+CI=St4-C++1 from i=Q to 5 and outputs the addition result. Here, A is the flag pit indicating the relationship between IF5 and BINO, B+ is the input pit to the adder of BINO to BIN5, S is the output bit from adder 71, and C3 and C1+ are the carry bits. It means a pit. Also, let BIN be the rough angle data before input to the adder, and S be the output data from the adder. ABS 0-ABS
12 is absolute position data obtained as a result of the correction calculation.

To explain this operation, ■ The LSB (BINO
) is 0, the output of the AND gate 76 is 0, and A
, ~A, are all 0. Since Ao is also O, BINO-BIN5 or B. By adding O to each of ~B, an output car of S=B I N is obtained.

■ Gray binary converter 23 LSB (BINO
) is 1, and the MSB (IF5) of the bracket interpolation processing unit 28 is 1.
When , the output of the AND gate 73 becomes 1 and A1~
A is all 1.

Since Ao is also 1, BINo to BIN5, that is, B.

3F (hexadecimal notation) is added to -B, respectively. This is the same as subtracting 1 from B0 to B, and an output of 5=BIN-1 is obtained.

■ Gray binary converter 23 LSB (BINO
) is 1, and the MSB (IF5) of the bracket interpolation processing unit 28 is O.
At this time, the output of the AND gate 73 becomes 0, and A-
A is all 0.

Since Ao is l, BINO-BIN5, that is Bo
-B5 and l will be added to each, thereby obtaining an output of 5=BIN+1.

By the above calculation, we obtained the solution S = B IN + α (α--1, O, +1), but if we do not use the LSB (So) of S and use the upper pit from S as the output, A halved solution is obtained.

Similarly, by inserting an inverter before the BINO signal is input to the AND gate 73, the fine angle data changes from the maximum value to 0, or from O to the maximum value. A circuit can be constructed.

In this way, the LSB of coarse angle data and the MS of fine angle data
By making B and B represent the same resolution, even if a deviation occurs within the range where the LSB of the corresponding coarse angle data does not change at the point where the fine angle data changes from the maximum value to 0, the error can be absorbed. can do.

The obtained angle data can be directly output in parallel, but if it is subjected to P/S conversion processing by the P/S converter 30 and transmitted as serial data as shown in Fig. 1, it can be output on the signal line. The number can be reduced.

(Positioning of the rotating slit plate and the fixed slit plate) As shown in FIG. Four pseudo sine waves with different phases can be extracted from the corners. For example, since the signal outputs of the photosensors 41W and 41X have phases different from each other by 90 degrees, if the output signals are sine waves, the Lissajous waveform will be circular. However, if the positional relationship between the rotating slit plate 12 (FIG. 2) and the fixed slit plate 16 is not accurate, the Lissajous waveform will be distorted. That is, if the positions of the rotating slit plate 12 and the fixed slit plate 16 are determined so that the Lissajous waveform is circular, the positional relationship of the slit group 42 and the slit group 43 with respect to the rotating slit plate is determined accurately. Similarly, the rotating slit plate 12 is set so that the signal output surge waveforms of the photosensors 41W and 41y are linear, and the signal output surge waveforms of the photosensors 41W and 412 are circular.
By adjusting the fixed slit plate 16, the rotating slit plate 1
The position of the slit group 42.46.44.45 with respect to 2,
That is, since the positional relationship between the four corners of the fixed slit plate 16 is determined accurately, high absolute accuracy can be achieved. Naturally, the position of the gray code reading slit 46 is also determined accurately.

Although the above explanation has been made for a rotary encoder, it goes without saying that this can be extended to a linear encoder and applied to a linear encoder.

〔Effect of the invention〕

As detailed above, according to the high-resolution absolute value encoder that adopts the detection method of the present invention, high-resolution detection is possible without increasing the number of tracks of the binarized code, so the entire device is No need to make it large. Further, when positioning the rotating slit plate and the fixed slit plate, high absolute accuracy can be achieved because the positioning can be performed based on the Lissajous waveform. From the above, it is possible to realize a high-resolution absolute value encoder that is small and has high absolute accuracy.

Further, since there are slit groups at the four corners of the fixed slit, the position of the photosensor can be visually confirmed directly through the fixed slit, so that positioning of the photosensor can be facilitated.

[Brief explanation of the drawing]

Fig. 1 is a block diagram of a signal processing system of an embodiment of an absolute value detection device to which the present invention is applied, Fig. 2 is a diagram showing a pattern formed on a rotating slit plate, and Fig. 3 is a diagram showing a signal formed on a fixed slit plate. 4 is a diagram showing the relationship between the output signals of the photosensor group and the absolute value data obtained from them; FIG. 5 is a diagram showing the relationship between coarse angle data and fine angle data; FIG. 6 is a diagram showing a hardware configuration for combining fine angle data and coarse angle data, and FIG. 7 is a perspective view of a conventional absolute value detection type encoder. 11.12... Rotating slit plate, 16.18.
...Fixed slit plate, 21...Current/voltage converter, 2...Comparator, 3...Gray binary 4...Difference Dynamic amplifier, 8... Interpolation processing unit, 9... Data combination unit, 1.35... Striped area, 6... Gray area. -Converter, Fig. 2, Fig. 3, Fig. 4, Fig. 7

Claims (2)

[Claims] (1) Using a moving slit plate on which a first pattern consisting of a binary code and a second pattern consisting of a striped code are formed, and the first pattern, absolute value data of a coarse angle is obtained. means for obtaining, using the second pattern, a first pseudo sine wave having a period that is twice or an integral multiple of the LSB of the absolute value data of the rough angle; means for obtaining a second pseudo sine wave of the same period and 90 degrees out of phase with the second pseudo sine wave; an interpolation processing section for interpolating the two pseudo sine waves to obtain absolute value data of a fine angle; A high-resolution absolute value encoder comprising: a data combining unit that combines absolute value data of an angle and absolute value data of the fine angle. (2) The high-resolution absolute value encoder according to claim 1, wherein the second pattern is arranged to sandwich the first pattern.