JPH05172586A - Absolute encoder - Google Patents

Abstract

PURPOSE:To enlarge the gaps between a detection element for the absolute and a detection element for the increment and a mark plate and well protect those detection elements. CONSTITUTION:In between a mark plate 1 and a detection part 4 where an absolute track 2 and incremental track 3 are formed, an index scale 10 forming an index patterns 12, 13 corresponding to the pattern of the incremental track 3 are arranged. The detection part 4 directly reads the pattern of the absolute track 2 and at the same time, reads the pattern of the incremental track 3 via the index patterns 12, 13.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an absolute encoder suitable for use as, for example, a linear encoder or a rotary encoder, and more particularly to an absolute encoder having one track of an absolute pattern.

[0002]

2. Description of the Related Art As an absolute encoder capable of detecting a relative displacement between two members which are relatively displaced as an absolute position, for example, an encoder having a gray code on a large number of parallel tracks is known. In order to increase the resolution or widen the measurement range, it is necessary to significantly increase the number of tracks, and the scale becomes large.
On the other hand, an absolute encoder of an optical type, a magnetic type, a capacitance type, or the like that can detect an absolute position with one absolute track is disclosed in, for example, Japanese Patent Laid-Open No. 2-168115.
It is disclosed in the publication.

The absolute encoder having one absolute track reads the two patterns in synchronization with the code plate in which the track of the absolute pattern having the pitch λ and the track of the incremental pattern are formed in the relative movement direction. The detector is arranged so as to be movable relative to the code plate in the longitudinal direction of the two tracks. In this detector, 2n (n is an integer of 3 or more) absolute detection elements and incremental detection elements having a pitch λ / 2 are formed, and 2n absolute detection elements arranged at the pitch λ / 2 are arranged. The absolute position is detected by n detection elements selected from the elements. The selection is made by the output of the incremental detection element that also functions as an index scale formed on the surface of the detector.

FIG. 7 shows a conventional absolute encoder disclosed in Japanese Patent Laid-Open No. 2-168115, in which the value of n is 4, and in FIG. 7, an absolute track 2 is provided on a code plate 1 made of a transparent substrate. And the incremental track 3 are formed in parallel. In the absolute track 2, for example, the following pattern "00" called a full-cycle array when the minimum reading unit λ is n = 4
00101100111101 "is formed. Assuming that one pulse is generated each time the code plate 1 is relatively displaced by λ, the number P of pulses in the entire range in which the absolute position can be detected is 2 ⁴ (= 16).

In the pattern, the transparent portion (white portion) is, for example, a pattern having a value of “0”, and the opaque portion (hatched portion) formed by a metal deposition film or the like is a pattern having a value of, for example, “1”. On the other hand, the incremental track 3 is a grid pattern having a pitch λ, and the width of the transparent portion and the width of the opaque portion are both λ / within one pitch.
It is 2. Further, the boundary of each pitch of the incremental track 3 is λ / 4 (90 ° in phase) with respect to the boundary of each minimum reading unit of the absolute track 2.
Just shifted.

Reference numeral 4 denotes a detector, and a photodiode array consisting of eight photodiodes 5-1 to 6-4 having a pitch of λ / 2 is arranged in a region of the detector 4 facing the absolute track 2. A single photodiode 7 having a width of λ / 2 is arranged in a region of the detector 4 facing the incremental track 3. The position of the photodiode 7 is equal to the position of any photodiode in the photodiode array with respect to the longitudinal direction of the photodiode array. The photodiode array consists of a set of photodiodes 5-1 and 5-for each λ pitch.
2, 5-3 and 5-4 and another set of photodiodes 6-1, 6-2, 6-3 and 6-4 for each λ pitch are grouped. Then, for example, when the output signal of the photodiode 7 is high level, the absolute position is detected by the output signals of the four photodiodes of the former group, and when the output signal of the photodiode 7 is low level, the latter position is detected. The absolute position is detected by the output signals of the four photodiodes in the set.

FIG. 8 is a side view when the encoder of FIG. 7 is used. In FIG. 8, the illumination light IL emitted from the light source 8 such as a light emitting diode is converted into a parallel light flux by the collimator lens 9. The converted code plate 1 is illuminated from the back side. Within the illumination light IL, the light fluxes that have passed through the absolute track 2 and the incremental track 3 formed on the light-shielding film 1a on the surface of the code plate 1 enter the photodiode array and the photodiode 7 of the detection unit 4, respectively.

In this case, since the illumination light IL is not a perfect parallel light beam and there is an image blur due to the diffraction effect, a shadow image of the pattern of the absolute track 2 and the incremental track 3 of the code plate 1 (hereinafter simply referred to as " The image is referred to as "image"), and the image becomes blurred as it gets away from the code plate 1. Generally, as the pattern pitch becomes smaller, the contrast of the image of the pattern due to the illumination light IL deteriorates rapidly. In particular, the deterioration of contrast due to the diffraction effect is proportional to the square of the pitch of the pattern.

In the example of FIG. 7, the minimum pitch of the pattern of the absolute track 2 is 2λ, and the minimum pitch of the pattern of the incremental track 3 is λ. Therefore, if the contrast of the image of the pattern having the narrowest line width of the absolute track 2 becomes about 1/2 at a predetermined distance D from the light shielding film 1a of the code plate 1, the distance D from the light shielding film 1a of the code plate 1 becomes. At / 2, the contrast of the image of the pattern on the incremental track 3 is reduced to about 1/2 or less as early as possible.

[0010]

However, in the conventional encoder, the photodiode array as the absolute detecting element and the photodiode 7 as the incremental detecting element also serving as the index scale are formed on the same plane. There is. Therefore, FIG.
As shown in, the gap G between the surface of the code plate 1 on which the light-shielding film 1a is formed and the surface of the photodiode array and the photodiode 7 of the detection unit 4 is such that the contrast of the image of the pattern of the incremental track 3 deteriorates too much. It is set to a range that does not. The gap G is about 1/2 or less of the gap defined under the condition that the contrast of the image of the pattern of the absolute track 2 is not significantly deteriorated, which is a disadvantage that it is considerably narrow.

Therefore, it is difficult to add a protective transparent substrate or the like to the surfaces of the photodiode array and the photodiode 7 of the detecting section 4, and the photodiode array and the like are easily deteriorated depending on the use environment. Further, there is a possibility that the photodiode array or the photodiode 7 may come into contact with the code plate 1 and break down in the assembly process, the adjustment process, or the like. Furthermore, in the conventional encoder, the phases of the absolute signal, which is the signal obtained from the absolute track 2, and the incremental signal are mechanically determined by the pattern arrangement of the code plate 1 and the photodiode arrangement of the detection unit 4. There is almost no room for adjustment. Therefore, even if the arrangement of the photodiodes of the detection unit 4 is slightly deviated due to the manufacturing lot, the adjustment is difficult, and the product yield may be deteriorated.

In view of the above, the present invention makes it possible to increase the absolute detection element and the gap between the incremental detection element and the code plate without deteriorating the SN ratio of the incremental signal, and these detection elements are well protected. It is an object of the present invention to provide an absolute encoder. It is another object of the present invention to facilitate adjustment of the phase relationship between the absolute signal obtained from the absolute track and the incremental metal signal.

[0013]

In a first absolute encoder according to the present invention, as shown in FIG. 1, for example, an absolute pattern track (2) and an incremental pattern track (3) are formed in a relative movement direction. Since the code plate (1) and the two patterns are read synchronously, detection is possible relative to the code plate (1) in the longitudinal direction (X direction) of the two tracks (2, 3). (4), and the detection data obtained by reading the absolute pattern is processed by the signal of the incremental pattern to correspond to the absolute position of the detector (4) with respect to the code plate (1). In an absolute encoder that generates a generated output signal, the incremental encoder is provided between the code plate (1) and the detector (4). Index scale reference pattern (12, 13) are formed corresponding to Le pattern (1
0) is arranged, and the detector (4) directly reads the absolute pattern and the incremental pattern through the reference patterns (12, 13).

In the second absolute encoder according to the present invention, as shown in FIG. 1, for example, P pulses (2 ^n-1 <P ≦ 2 ⁿ , n is 3) with an absolute position detection range of λ pitch.
A code plate (1) in which a track (2) of an absolute pattern and a track (3) of an incremental pattern of (the above integers) are formed in the relative movement direction, and
Code plate for synchronously reading individual patterns (1)
2n or more absolute detection elements (5-1 to 6-4) and incremental detection elements which are arranged so as to be relatively movable in the longitudinal direction (X direction) of these two tracks with a pitch λ / 2. (14, 15) is formed on the code plate (1) by processing the detection data obtained by reading the absolute pattern by the signal of the incremental pattern. In an absolute encoder for generating an output signal corresponding to the absolute position of a detector (4), a reference pattern (12) arranged between the code plate (1) and the detector (4) and corresponding to the incremental pattern (12). , 13) on which the index scale (10) is formed, and the incremental detection element (1).
4, 15) processes a signal read from the incremental pattern via the reference pattern (12, 13) to obtain a plurality of incremental signals (for example, signals SA, SB, SA *, SB * in FIG. 4). ) And a selecting means (for example, the rotary switch 34 of FIG. 6) for selecting a certain specific incremental signal from among the 2) or more absolute detection elements among the 2n or more absolute detecting elements. The output signals of n or more detection elements for each λ pitch are selected and used for absolute position detection.

[0015]

As described above, in general, as the distance from the code plate increases, the contrast of the incremental metal pattern image becomes faster than that of the absolute pattern image. Therefore, it is desirable to set the substantial gap between the incremental pattern and the detector smaller than the substantial gap between the absolute pattern and the detector. Therefore, in the first absolute encoder of the present invention, between the detector (4) and the code plate (1),
Reference pattern (1 corresponding to the incremental pattern
Index scale (10) formed with 2, 13)
Are arranged.

The image of the absolute pattern (shadow image) of the code plate (1) passes through the index scale (10) as it is and reaches the detector (4), where the image of the incremental pattern of the code plate (1). Is the reference pattern (12, 1) of the index scale (10).
The detector (4) is reached via 3). Therefore, the substantial gap between the incremental pattern and the detector (4) is equal to the gap between the reference pattern (12, 13) of the index scale (10) and the incremental pattern of the code plate (1), and the absolute pattern Is smaller than the substantial gap between the detector and the detector (4). Therefore, the contrast of the image of the incremental pattern is not deteriorated and the SN ratio of the incremental signal is maintained well, and the index scale (10)
Protects the surface of the detector (4).

In the second absolute encoder of the present invention, the absolute detection element (5-1
To 6-4) and the incremental detection element (14, 1)
5) is formed between the detector (4) and the code plate (1), the reference pattern (12, 1) for the incremental signal.
There is an index scale (10) formed with 3). Unlike the conventional example, the incremental detection element (14, 15) itself does not also serve as a reference scale for incremental signals.

At this time, the reference scale (12, 13) for the incremental signal and the absolute detection elements (5-1 to 6-4) in the longitudinal direction of the track (X direction).
The relative position in is not fixed, and the signals obtained from the absolute detection elements (5-1 to 6-4) cannot be selected using the signals obtained from the incremental detection elements (14, 15) as they are. Therefore, in the present invention, the incremental detection element (14, 15) is used.
From the absolute detection elements (5-1 to 6-4) to obtain a plurality of incremental signals having different phases from each other. The incremental signal whose phase difference is closest to λ / 4 (90 °) is selected. From the absolute detection elements (5-1 to 6-4), n or more detection elements are sequentially selected for each λ pitch by the selected signal,
The absolute position can be detected by processing the output signal of the selected detection element.

Absolute pattern track (2)
Since the minimum reading unit of λ is λ, and the pitch of the absolute detection elements (5-1 to 6-4) is λ / 2, one set of detection elements for each λ pitch (for example, 5-1 to 5-4). )
, The output signal of the other set of detection elements (for example, 6-1 to 6-4) for each λ pitch varies in the opposite phase with the pitch λ. Therefore, by switching the output signals of the one set of detection elements and the output signal of the other set at the pitch λ / 2, it is possible to always detect an accurate absolute position with the resolution λ. On the other hand, if the switching is not performed, the detection elements (5-1 to 6)
-4) The absolute position cannot be detected in the area where the output signal changes.

As described above, in the present invention, the absolute detecting element (5-
Signals having a phase difference of 1 to 6-4) with respect to the change point of the output signal within a predetermined range are selected. Therefore,
Even if the phase relationship between the reference patterns (12, 13) of the index scale (10) and the absolute detection elements (5-1 to 6-4) is varied at the time of assembly adjustment, etc., the selection method should be selected according to the variation. By changing it, the accurate absolute position can be detected. Further, an absolute detection element (5-1 to 6) in the detector (4)
-4) and the incremental detection element (14, 15)
Even if the positional relationship between and does not affect the detection of the absolute position, the detector (4) can be easily manufactured.

In the present invention, a plurality of phases of incremental signals are generated, but the incremental detection element (1
4 and 15), it is only necessary to be able to detect a two-phase incremental signal (having a period of λ) with a phase difference of 90 ° at the minimum. These two-phase signals are converted into A-phase signal SA and B-phase signal SB.
Then, by inverting the signals SA and SB, the inverted A-phase signal SA * and the inverted B-phase signal SB * are obtained. These four-phase signals SA, SB, SA * and SB
The phases of * are at 90 ° intervals, and the maximum phase difference between the signal of arbitrary period λ and the signals of these four phases is within ± 45 °.

Therefore, as the timing signal for selecting n or more of λ pitches from the 2n or more absolute detection elements (5-1 to 6-4), the above-mentioned four-phase signal S is selected.
Of the A, SB, SA * and SB *, the difference from the change point of the output signal of the absolute detection element is ± λ / 8 (= ±
It is sufficient to select a signal within 45 °). As a result, the deviation from the absolute signal falls within the allowable range. The selection means is, for example, the rotary switch (34) shown in FIG. 6, and the selection by the selection means can be easily performed at the time of signal adjustment, for example.

Furthermore, if the gap between the index scale (10) and the detector (4) is, for example, the height of wire bonding, it is not necessary to extend the signal lead-out portion to the outside of the detector (4) and the detector (4) ) Can be made small.

[0024]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of an absolute encoder according to the present invention will be described below with reference to FIGS. This example is an improvement of the conventional example shown in FIGS. 7 and 8.
1 to 3, parts corresponding to those in FIGS. 7 and 8 are designated by the same reference numerals, and detailed description thereof will be omitted. FIG. 1 shows a state in which the absolute encoder of this example is disassembled. In FIG. 1, in the longitudinal direction (X direction) of the code plate 1 as a scale made of a transparent substrate, the detection range is λ · in the minimum reading unit λ. An absolute track 2 composed of 2 ⁴ absolute patterns and an incremental track 3 composed of a grid-like incremental pattern having a pitch λ are formed in parallel.

The absolute pattern and the incremental pattern are formed of, for example, a chromium vapor-deposited film, and the shaded portion is the light shielding portion, and the other portions are the light transmitting portions. Further, the positional relationship in the X direction between the absolute track 2 and the incremental track 3 is arbitrary,
The code plate 1 is easier to manufacture than the conventional example.

Reference numeral 4 denotes a detection unit. Also in this example, the detection unit 4 has eight photodiodes 5-1 with a pitch of λ / 2.
A photodiode array formed by arranging 6 to 4 is formed. The photodiode array is divided into two groups, one set of photodiodes 5-1 to 5-4 for each λ pitch and the other set of photodiodes 6-1 to 6-4 for each λ pitch. .. Further, the detector 4 has two photodiodes 14 in parallel with the photodiode array.
And 15 are formed. There is no pattern on the surface of these photodiodes 14 and 15. Note that, in FIG. 1, the photodiode array and the like are illustrated as being arranged on the front side with respect to the paper surface for convenience, but are actually arranged on the opposite side with respect to the paper surface.

Reference numeral 10 indicates an index scale, and the absolute track 2 of this index scale 10
A light-transmissive window 11 is formed at a position corresponding to. In addition, two sets of index patterns 12 and 1 each made of a grid pattern having a pitch λ are provided at positions corresponding to the incremental tracks 3 of the index scale 10.
3 is formed. The index patterns 12 and 13 are also formed by, for example, chromium vapor deposition, the shaded portion is a light shielding portion, and the other portions are light transmitting portions. In this case, the phase difference between the index pattern 12 and the index pattern 13 is set to (m + 1/4) λ (5λ / 4 in FIG. 1). Therefore, since the phase difference between the signals obtained through the index patterns 12 and 13 is 90 °, the pattern 13 will be referred to as the A-phase index pattern and the pattern 12 will be referred to as the B-phase index pattern below.

FIG. 2 shows a state in which the parts shown in FIG. 1 are assembled. In FIG. 2, a base 17 is arranged so as to be relatively movable in the X direction with respect to the code plate 1, and detection is performed on the base 17. Fix the part 4. In addition, spacers 17 and 18 are fixed on both sides of the detection unit 4 on the base 17, and the spacers 1
Fix the index scale 10 on 7 and 18,
The index scale 10 is opposed to the scale forming surface of the code plate 1.

At this time, the index scale 10 is attached to the absolute track 2 on the code plate 1 shown in FIG.
Window 11 and photodiode 5-1 of the detector 4
The photodiode arrays of 6-4 are made to overlap so that the index patterns 12, 13 of the index scale 10 and the photodiodes 14, 15 of the detector 4 overlap the incremental track 3. Further, the photodiodes 14 and 15 are made to overlap the index patterns 12 and 13, respectively.

FIG. 3 is a side view of FIG. 2. As shown in FIG. 3, the light shielding film 1a of the code plate 1 and the light shielding film 10a of the index scale 10 face each other with a gap G1. Further, the refractive index of the index scale 10 is α, the thickness is t, and the gap between the back surface of the index scale 10 and the front surface of the photodiode array including the photodiodes 5-1 to 6-4 is G3. In this case, when the code plate 1 is illuminated with the parallel illumination light IL, the image (shadow image) of the absolute track 2 of the code plate 1 passes through the index scale 10 and is projected on the surface of the photodiode array of the detector 4 as it is. It On the other hand, the image of the incremental track 3 of the code plate 1 is projected onto the index patterns 12 and 13 of the index scale 10, and the photodiodes 14 and 15 have the light amounts corresponding to the degree of overlap between the track 3 and the scales 12 and 13, respectively. Is detected.

Therefore, in this example, the substantial gap G _inc regarding the incremental pattern and the substantial gap G _abs regarding the absolute pattern are respectively expressed by the following equations, and the gap G _inc <gap G _abs is established.

## EQU1 ## G _inc = G1 G _abs = G1 + (t / α) + G2

In this example, since the minimum pitch of the absolute pattern is 2λ and the pitch of the incremental pattern is λ, the contrast of the image of the incremental track 3 is larger than that of the image of the absolute track 2 from the code plate 1. It gets worse rapidly with distance. Therefore, according to the configuration of this example capable of reducing the substantial gap G _inc related to the incremental pattern, the S / N ratio of the incremental signal is good, and the photodiode on the detector 4 is protected by the index scale 10. There is.

Further, in FIG. 3, the photodiode 1
Ignoring the thicknesses such as 4 and the like, by ensuring the gap G2 between the index scale 10 and the detection unit 4 by the height of the wire bonding, the signal extraction unit of the detection unit 4 can be accommodated in the gap G2. .. This has an advantage that the detection unit 4 can be downsized. Further, it is desirable that the substantial gap G _abs regarding the absolute pattern is also small, but for that purpose (Equation 1)
Therefore, the refractive index α of the index scale 10 may be set as large as possible (the thickness t is set to about 1 mm, for example).

Returning to FIG. 1, during the measurement of this example, the detector 4
An absolute signal corresponding to the pattern of the absolute track 2 is obtained from the photodiode array including the photodiodes 5-1 to 6-4. As an example, absolute signals output from the photodiodes 6-4 and 5-4 are shown in FIGS. 4A and 4B, respectively. Further, from the photodiode 15 of the detector 4, the A-phase signal SA corresponding to the degree of overlap between the incremental track 3 and the A-phase index pattern 13 (see FIG. 4).
(C)) is output, and the B-phase signal SB (FIG. 4) corresponding to the degree of overlap between the incremental track 3 and the B-phase index pattern 12 is output from the photodiode 14.
(D)) is output. The phase difference between the A-phase signal SA and the B-phase signal SB is 90 °.

Further, there is variation in the relative positional relationship in the X direction between the index scale 10 and the detector 4, which is determined during assembly and manufacturing, and as a result, the index patterns 12 and 13 and the photodiode 5 on the detector 4 are different.
The positional relationship in the X direction with the photodiode array composed of -1 to 6-4 is indefinite. However, ideally, for example, the rising point of the A-phase signal SA output from the photodiode 14 and, for example, one set of photodiodes 6-1 to 6
It is desirable to adjust the fixed position of the index scale 10 so that the changing point of the absolute signal obtained from -4 matches. However, such adjustment is actually difficult, and further, in the present example, since the adjusting mechanism is provided as described later, it is not necessary to perform such adjustment.

The adjusting mechanism will be described below.
First, in this example, an inversion signal SA * (FIG. 4 (e)) of the A-phase signal SA and an inversion signal SB * (FIG. 4 (f)) of the B-phase signal SB are generated by an inversion circuit described later. Then, one of the four-phase signals SA, SB, SA * and SB * is selected by a selective throw such as a rotary switch. In the example of FIG. 4, the inverted B-phase signal SB * is most suitable. That is, as apparent from FIG. 4, when the inverted B-phase signal SB * is at the high level "1", the photodiodes 5-1 to 5-5 are provided.
-4 is stable and the inverted B-phase signal SB * is at low level "0", the photodiodes 6-1 to 6-1
The output signal of 6-4 is stable. Therefore, when the inverted B-phase signal SB * is at high level, the photodiode 5-1
5-4 output signals are selected and decoded, and when the inverted B-phase signal SB * is at low level, the photodiode 6-
By selecting and decoding the output signals 1 to 6-4, the accurate absolute position is always detected.

Next, the conditions for an ideal incremental signal will be considered with reference to FIG. FIG. 5A shows, for example, one set of photodiodes 6-1 to 6 for each λ pitch.
Absolute position x, x obtained by decoding the output signal of -4
5B shows absolute positions x, x + 1, x + 2 obtained by decoding the output signals of the other photodiodes 5-1 to 5-4 for each λ pitch.
Show ... The absolute position of FIG. 5 (a) and the absolute position of FIG. 5 (b) are out of phase by 180 ° (= λ / 2).
Then, before and after the position where the absolute position in FIG.
Within the range of λ / 4, the absolute position of FIG.
Conversely, ± λ / before and after the position where the absolute position in Fig. 5 (b) changes
Within the range of 4, the absolute position of FIG. 5 (a) is used.

Therefore, the level of the ideal incremental signal is inverted at the center between the change point of FIG. 5 (a) and the change point of FIG. 5 (b), like the signal 19 shown in FIG. 5 (c). The signal has a period λ. However, instead of the signal 19, a signal whose level is inverted between a signal 20 having a phase advanced by 45 ° (= λ / 8) and a signal 21 having a phase delayed by 45 ° can be used. When a four-phase signal with a 90 ° phase difference is used as the incremental signal as in the case of FIG. 4, there is always a signal whose level is inverted between the signal 20 and the signal 21, so that The incremental signal may be selected by the selection means.

The phase margin before and after the ideal incremental signal 19 shown in FIG. 5C is, for example, ± 180 °.
If / N (N is an integer of 4 or more), the width of the phase margin is 360 ° / N (= λ / N). In this case, if there are N-phase incremental signals with equal phases,
There is always an incremental signal that matches the ideal incremental signal 19 at the phase margin. Therefore, by further increasing the number N of incremental signals, a signal closer to the ideal incremental signal 19 can be selected and a large noise margin can be secured.

As an incremental signal, as shown in FIG.
A signal of one phase is selected from the four-phase incremental signals of (f), and the output signals of the photodiodes 5-1 to 5-4 and the photodiode 6-1 are selected by the selected signal.
6 shows a configuration example of a circuit for alternately selecting the output signals of 6-4. In FIG. 6, the input terminal 22-1
22 to 22-4 are the photodiodes 5-1 of FIG.
To 5-4 output signals and input terminals 22-1 to 22
-4 are connected to one input terminals of the AND circuits 23-1 to 23-4, respectively. Also, the input terminals 24-1 to 24-
4 are photodiodes 6-1 to 6-4 of FIG. 1, respectively.
Of the AND circuits 25-1 to 25-4 are connected to the input terminals 24-1 to 24-4, respectively.

The output terminals of the AND circuits 23-1 to 23-4 are connected to one input terminals of the OR circuits 26-1 to 26-4, respectively, and the output terminals of the AND circuits 25-1 to 25-4 are connected. The OR circuits 26-1 to 26-4 are connected to the other input terminals, and the output terminals of the OR circuits 26-1 to 26-4 are connected to the 4-bit address input terminals of the read only memory (ROM) 27. 4 of that ROM27
Bit data output terminals are output terminals 28-1 to 28-4
Connect to.

Reference numerals 29 and 30 denote input terminals for the incremental signal. The A-phase signal SA output from the photodiode 15 of FIG. 1 is supplied to the input terminal 29, and the B-phase signal SB output from the photodiode 14 is supplied. Input terminal 3
Supply to 0. Reference numerals 31-1 to 31-4 respectively denote three-state output buffer circuits, and input terminal 2
9 is connected to the input terminal of the buffer circuit 31-1 and the input terminal of the inverter 32, the input terminal 30 is connected to the input terminal of the buffer circuit 31-2 and the input terminal of the inverter 33, and the output terminals of the inverters 32 and 33 are connected. They are connected to the input terminal of the buffer circuit 31-3 and the input terminal of the buffer circuit 31-4, respectively.

Reference numeral 34 denotes a four-step change-over type rotary switch as selection means. The movable contact of the rotary switch 34 is grounded, and the four fixed contacts of the rotary switch 34 are respectively connected to the resistors 35-1 to 35-4. Through to the high level side. Further, the rotary switch 34
The four fixed contacts of the buffer circuits 31-1 to 31-1
31-4 is connected to a negative logic output control terminal. Therefore,
Among the buffer circuits 31-1 to 31-4, only the output signal of the buffer circuit corresponding to the fixed contact that is in contact with the movable contact of the rotary switch 34 becomes the high level or the low level, and the output signals of the other buffer circuits have the high impedance. become.

These four buffer circuits 31-1 to 31-3
The output terminals of 1-4 are collectively connected to the check terminal 37, and the check terminal 37 is connected to the AND circuits 23-1 to 23-1.
23-4 and the other input terminal of each and the inverter 3
6 input terminals are commonly connected. And the inverter 3
The output terminal of 6 is connected to the other input terminal of each of the AND circuits 25-1 to 25-4. As a result, when the incremental signal generated at the check terminal 37 is at the high level, the output signals of the photodiodes 5-1 to 5-4 are selected and supplied to the address input terminal of the ROM 27,
When the incremental signal generated at the check terminal 37 is at low level, the output signals of the photodiodes 6-1 to 6-4 are selected and supplied to the address input terminal of the ROM 27. Therefore, every time the level of the incremental signal generated at the check terminal 37 is inverted, the decoded data of the output signals of the photodiodes 5-1 to 5-4 and the output signals of the photodiodes 6-1 to 6-4 are decoded. Data is alternately output from the data output terminal of the ROM 27.

The check terminal 38 is connected to the input terminal 24-1 for adjustment. Then, when adjusting the rotary switch 34, the operator brings the two probes of the oscilloscope into contact with the check terminals 37 and 38. In this state, the operator turns the movable contact of the rotary switch 34 while looking at the oscilloscope, and the center of the rising and falling edges of the signal at the check terminal 38 is closest to the center of the high level signal at the check terminal 37. It suffices to select such a fixed contact.

As described above, according to this example, the optimum incremental signal can be easily selected only by operating the rotary switch 34. An electronic switch may be used instead of the rotary switch 34, and the selection may be automatically performed.
It should be noted that the present invention is not limited to the above-described embodiments, and it is needless to say that the present invention can be applied to a rotary encoder and various other configurations without departing from the scope of the present invention.

[0047]

According to the first and second absolute encoders of the present invention, by using the index scale, the absolute signal can be detected with a wide gap,
Since the incremental signal can be detected with a narrow gap, there is an advantage that the detector can be protected by the index scale and at the same time, the SN ratio of the incremental signal can be kept good.

Further, according to the second absolute encoder, since one phase is selected from the plurality of incremental signals having a phase shift and the absolute signal is switched, the absolute signal is manufactured when the detector is manufactured. The positional relationship between the use detection element and the incremental detection element may be indefinite, and the accuracy of the position where the index scale reference pattern is arranged may be rough. Furthermore, since the positional relationship between the absolute pattern and the incremental pattern of the code plate may be indefinite, there is an advantage that each component can be easily manufactured.

[Brief description of drawings]

FIG. 1 is a partially cutaway plan view showing respective constituent parts of an embodiment of an absolute encoder according to the present invention.

FIG. 2 is a perspective view showing a state where the encoder of the embodiment is assembled.

FIG. 3 is a side view of the assembled encoder of the embodiment.

4 is a timing chart showing signals output from the photodiodes of FIG. 1. FIG.

FIG. 5 is a waveform diagram showing a relationship between an ideal incremental signal and an absolute signal.

6 is a circuit diagram showing an example of a processing circuit of an output signal of each photodiode of FIG.

FIG. 7 is a partially cutaway plan view showing each component of a conventional absolute encoder.

8 is a side view showing a state where the encoder of FIG. 7 is assembled.

[Explanation of symbols]

1 Code Plate 2 Absolute Track 3 Incremental Track 4 Detector 5-1 to 5-4 Photodiode 6-1 to 6-4 Photodiode 10 Index Scale 11 Drilled Hole 12 A-Phase Index Pattern 13 B-Phase Index Pattern 14, 15 Photodiode 27 ROM 31-1 to 31-4 Three-state output buffer circuit 34 Rotary switch

Claims (2)

[Claims] 1. A code plate in which a track of an absolute pattern and a track of an incremental pattern are formed in a relative movement direction, and a track of the two tracks with respect to the code plate for synchronously reading the two patterns. A detector movable relative to the longitudinal direction is provided, and the detection data obtained by reading the absolute pattern is subjected to signal processing by the signal of the incremental pattern to correspond to the absolute position of the detector with respect to the code plate. In an absolute encoder generating an output signal, an index scale having a reference pattern corresponding to the incremental pattern is arranged between the code plate and the detector, and the detector directly reads the absolute pattern. , The incremental power An absolute encoder characterized in that a turn is read through the reference pattern. 2. A track of an absolute pattern and a track of an incremental pattern of P pulses (2 ^n-1 <P ≦ 2 ⁿ , n is an integer of 3 or more) with an absolute position detection range of pitch λ are formed in relative movement directions. Code plate and the two patterns are synchronously read, and are arranged so as to be relatively movable in the longitudinal direction of the two tracks with respect to the code plate, and for the absolute number of 2n or more with a pitch λ / 2. A detector formed with a detection element and an incremental detection element, the detection data obtained by reading the absolute pattern, by signal processing by the signal of the incremental pattern, the absolute of the detector with respect to the code plate In an absolute encoder that generates an output signal corresponding to a position, between the code plate and the detector, An index scale that is arranged and has a reference pattern formed corresponding to the incremental pattern, and a plurality of incremental signals obtained by processing the signal read from the incremental pattern by the incremental detection element via the reference pattern. And a selection means for selecting a certain specific incremental signal from among the 2n or more absolute detection elements, and n or more detection elements for each λ pitch are selected from the 2n or more absolute detection elements. An absolute encoder characterized in that the output signal of is selected and used for absolute position detection.