JP6261380B2 - Optical encoder - Google Patents

Description

  The present invention relates to an optical encoder, and more particularly to an optical encoder that enables origin detection.

  Optical encoders are widely used for measuring shapes such as the length, thickness, inner and outer diameters of objects to be measured, and displacements such as rotation angles or movement amounts, and their configurations are also widely known. Optical encoders include a transmission type that detects light transmitted through the optical scale and a reflection type that detects light reflected by the optical scale, and the reflection type is often used for miniaturization. Hereinafter, a reflective linear optical encoder will be described as an example. However, the present invention is not limited to this, and any optical encoder can be applied.

  As described in Patent Documents 1 and 2, reflective linear optical encoders in recent years have applied a semiconductor manufacturing technique to improve the accuracy, miniaturization / thinning, and ease of assembly. A sensor substrate on which a light receiving sensor is integrally formed is used. The sensor substrate is arranged so that the illumination light from the light source is reflected by the optical scale with respect to the reflective optical scale, and a light pattern corresponding to the pattern of the optical scale is formed on the light receiving sensor. The reflective linear optical encoder measures the relative movement amount of the optical scale and the sensor substrate.

  In order to reduce the influence of manufacturing variation of each part, the light receiving sensor has first to fourth sensor parts. The first to fourth sensor units are arranged at the same width and in the same direction as the arrangement direction of the light and shade pattern of the optical scale (hereinafter referred to as the main axis direction) at the sensor pitch, and one set of the first to fourth sensor units is set. Are repeatedly arranged in the main axis direction by a predetermined number of sets. Therefore, the first sensor units are arranged at a 4-sensor pitch, and the other second to fourth sensor units are also arranged at a 4-sensor pitch. The number of sets is 13, for example. The detection signals of a plurality of (for example, 13) first sensor units are added to form a + A signal. Similarly, the detection signals of the plurality of second to fourth sensor units are also added to form + B, −A, and −B signals, respectively. By adding signals from a plurality of sensor units, it is possible to reduce the influence of variations in sensitivity of sensor units, uneven illumination light, assembly errors, and the like.

  The optical encoder measures a relative movement amount of the optical scale and the sensor substrate by reading a light and shade pattern (having different reflection intensities) periodically formed on the optical scale. However, only the relative movement amount can be measured, and the absolute position cannot be measured.

If the absolute position can be measured by the optical encoder, it is advantageous in the following points as compared with the measurement of only the relative position.
First, consider a case where a miscount occurs due to some external cause. In this case, since the origin does not exist, that is, the measurement reference point does not exist, it is impossible for the mechanism to detect the miscount. For this reason, there is a risk that all measured values after the occurrence of miscounting will be incorrect values.
On the other hand, in the case of having an origin, a miscount can be detected by using the origin as a reference, so that an incorrect value can be corrected each time.

Also, consider a case where the count value is lost during measurement due to external factors such as power loss. When the origin does not exist, it is impossible to return to the original state using only the optical encoder. On the other hand, when the origin exists, the value of the origin position is determined in advance, and the original state can be restored by retracting to the origin.
The above is very important when an optical encoder is used for measurement during actual machining.

  Therefore, an optical encoder generally includes an origin detection mechanism separately from an optical scale for detecting a relative position.

JP 2010-223631 A JP2012-103230A

Patent Document 1 proposes an origin detection mechanism having two optical scales formed in parallel on the same scale substrate. However, the described origin detection mechanism is not sufficient in the following points.
Since the gradient of the amplitude change of the signal used for the origin detection is small, it is difficult to detect the origin by comparison with the threshold with high accuracy. In addition, there is a possibility that detection is performed by shifting the origin position due to the influence of the light amount fluctuation of the light source.

Therefore, in order to detect the origin with high accuracy, it is necessary to provide a circuit for correcting the origin detection mechanism in addition to the signal processing unit of the optical scale for detecting the relative position, which increases the size and cost.
Further, a Z-phase signal obtained by summing up the outputs of the four sensor units is generated, and an origin signal is generated in combination with the A-phase signal, and a deviation occurs with a reciprocal relative movement.

  An object of the present invention is to realize an optical encoder that uses two optical scales and can detect an origin with high accuracy with a simple arithmetic circuit.

  In the optical encoder of the present invention, one of two optical scales formed in parallel is a normal light / dark pattern in which the dark portion and the light portion have the same width, and the light / dark is repeated at the scale pitch. In the same manner, the gradation is repeated at the scale pitch, but at a predetermined position, the width of the dark portion or the light portion is wide in a range not exceeding 1/2 of the scale pitch. In other words, the phase is shifted at a predetermined position.

  That is, the optical encoder of the present invention includes a first optical system, a second optical system, and a signal processing unit. The first optical system includes a first optical scale having a first light and shade pattern that changes in a principal axis direction, a first light source that outputs the first illumination light of the first optical scale, and the first optical light by the first illumination light. A first light receiving sensor arranged to detect an image of a first gray pattern generated by illuminating the equation scale, and the first light receiving sensors are arranged in order in the main axis direction at a sensor pitch. It has a first sensor array in which a first sensor group composed of first to fourth main sensor units is repeatedly arranged a predetermined number of times in the main axis direction. The second optical system includes a second optical scale having a second light and shade pattern that changes in the principal axis direction, a second light source that outputs the second illumination light of the second optical scale, and a second optical light using the second illumination light. And a second light receiving sensor arranged to detect an image of the second gray pattern generated by illuminating the equation scale, and the second light receiving sensors are arranged in order in the main axis direction at a sensor pitch. A second sensor array in which a second sensor set including the first to fourth sub sensor units is repeatedly arranged a predetermined number of times in the main axis direction is provided. The signal processing unit performs arithmetic processing on the detection signal of the first light receiving sensor of the first optical system and the detection signal of the second light receiving sensor of the second optical system. The first optical scale and the second optical scale move relative to the first light source, the first light receiving sensor, the second light source, and the second light receiving sensor. In the first light and shade pattern of the first optical scale, the dark portion and the light portion have the same width, and the light and shade are repeated at the scale pitch. The second light and shade pattern of the second optical scale has the same width of the dark portion and the light portion, and repeats the light and shade at the scale pitch. At a predetermined position, the width of the dark portion or the light portion exceeds 1/2 of the scale pitch. It is wide in no range. The signal processing unit converts at least one of the outputs of the first to fourth main sensor units into a digital signal and outputs main position data, and the first to fourth sub sensor units. A sub-position data generation unit that converts at least one of the outputs into a digital signal and outputs sub-position data, a difference calculation unit that calculates difference data between main position data and sub-position data, and an absolute value of the difference data is predetermined An origin determination unit that determines whether the value is equal to or greater than the value and detects the origin from a change in the determination value.

  In the optical encoder of the present invention, the second gray scale pattern of the second optical scale is wide in a range where the width of the dark portion or the light portion does not exceed 1/2 of the scale pitch at a predetermined position. shift. Therefore, when the second light receiving sensor detects the second gray pattern on one side of the predetermined position, the main position data and the sub position data are in-phase signals, but the second gray level on the other side of the predetermined position. When a pattern is detected, it is a phase-shifted signal. Therefore, the movement position calculated from the detection signal of the first light receiving sensor and the phase difference (movement position difference) calculated from the detection signal of the second light receiving sensor are the same on one side of the predetermined position, and the phase difference on the other side. This is a shifted signal, and the phase difference gradually changes with the movement until the first to fourth sub-sensor units start detecting the portion at the predetermined position and stop detecting it. Therefore, by determining whether or not the phase difference exceeds a predetermined value, it can be determined whether the phase difference is on one side or the other side. When the phase difference exceeds the predetermined value, the predetermined position, that is, the origin Can be detected. Since the position can be accurately detected by the detection signals of the first to fourth main sensor units, it is only necessary to determine which scale pitch the origin detection is.

  The first optical scale and the second optical scale are adjacent to each other on a common scale substrate, and the first gray pattern and the second gray pattern are parallel to each other. Are preferably formed so as to match. The first light source, the first light receiving sensor, the second light source, and the second light receiving sensor are formed on a common sensor substrate, and the first light source and the second light source are formed as a common light source. In the two light receiving sensors, the first sensor array and the second sensor array are arranged in parallel on both sides of the common light source so that the positions of the first to fourth main sensor units and the first to fourth sub sensor units coincide with each other. It is desirable to be formed. Furthermore, the sensor substrate projects illumination light from the common light source reflected by the first light and shade pattern onto the first light receiving sensor and the illumination light from the common light source reflected by the second light and shade pattern on the scale substrate. It is desirable to arrange so as to be projected onto the second light receiving sensor.

  Thereby, the position adjustment of the sensor substrate and the scale substrate is facilitated, and the size and thickness can be reduced.

The scale pitch is twice the sensor pitch, and the sensor substrate is arranged such that the first density pattern and the second density pattern are projected on the scale board as a density pattern that is four times the sensor pitch. .
Thereby, it is possible to increase the amplitude of the detection signal and improve the resolution with a simple configuration.

  Similar to a general optical encoder, the signal processing unit outputs a phase within a cycle length of four times the sensor pitch as one cycle from the outputs of the first main difference calculation circuit and the second main difference calculation circuit. Is calculated. Therefore, origin detection only needs to be able to determine which period length the origin is in. Therefore, the signal processing unit detects in which period length the output change of the comparison circuit of the origin determination circuit has occurred, and determines in which period length the origin is in consideration of the moving direction.

  The optical encoder of the present invention uses only one of the two optical scales for movement detection and the other for origin detection, and simply adds a simple arithmetic circuit to a general circuit used for movement detection. There is an effect that the origin can be detected with high accuracy.

FIG. 1 is a diagram illustrating a main part of a reflective linear optical encoder according to an embodiment. FIG. 2 is a diagram illustrating a pattern of the light receiving unit. FIG. 3 is a diagram illustrating the positional relationship between the scale substrate and the sensor substrate. FIG. 4 is a diagram schematically illustrating a path in which light from the light source is reflected by the reflection pattern and projected onto the light receiving unit. FIG. 5 is a diagram illustrating a reflection pattern projected onto the light receiving unit. FIG. 6 is a diagram illustrating changes in detection signals accompanying changes in the relative positional relationship (length) between the scale substrate and the sensor substrate. FIG. 7 is a diagram illustrating a detection signal calculation circuit used in the optical encoder. FIG. 8 is a diagram illustrating changes in the A-phase signal and the B-phase signal accompanying changes in the relative positional relationship (length) between the scale substrate and the sensor substrate. FIG. 9 is a diagram illustrating a scale pattern in the reflective linear optical encoder according to the embodiment. FIG. 10 shows a reflection pattern projected on the light receiving unit when the sensor substrate is arranged as shown in FIG. 3 with respect to the scale substrate having the reflection pattern shown in FIG. FIG. FIG. 11 shows a case in which a signal obtained by combining the signals of 13 sets of cells of a light receiving unit on which a reflection pattern having a constant pitch is projected and a light receiving unit on which a reflection pattern is shifted is moved beyond a predetermined position. It is a figure which shows a change. FIG. 12 is a diagram illustrating a phase difference between signals of the light receiving unit according to the movement position in the embodiment. FIG. 13 is a diagram illustrating a configuration of a signal processing unit in the reflective linear optical encoder of the embodiment. FIG. 14 is a flowchart illustrating a processing flow of the signal processing unit in the reflective linear optical encoder of the embodiment. FIG. 15 is a diagram illustrating the relationship between the relative position signal calculated by the arithmetic circuit that calculates the signal of the light receiving unit and the origin signal generated by the origin detection described so far.

FIG. 1 is a diagram illustrating a main part of a reflective linear optical encoder according to an embodiment.
The reflective linear optical encoder of the embodiment includes a scale substrate 1 on which two reflection patterns 2-1 and 2-2 are formed, a light source 11, a light transmitting member 13 on which a grating 12 is formed, and two pieces The light receiving elements 14-1 and 14-2 are formed, and the sensor substrate 10 is covered with a transparent resin material 16. The scale substrate 1 is fixed to one of the two members relatively moving in the uniaxial direction, and the sensor substrate 10 is fixed to the other, and the relative movement distance of the two members is measured. Here, the moving direction is referred to as the main axis direction.

  The reflection patterns 2-1 and 2-2 are formed of a transparent glass substrate and are formed on the surface facing the sensor substrate 10. The reflection pattern 2-1 is a reflection pattern formed with a scale pitch p2 in the principal axis direction, and chromium is vapor-deposited in the reflection portion, and light is transmitted between the reflection portions, so that a reflection pattern with the pitch p2 is formed. The The reflection pattern 2-2 will be described later. The reflection pattern 2-1 functions as a main optical scale, and the reflection pattern 2-2 functions as a sub optical scale.

The light source 11 is an LED, and the grating 12 is an optical pattern having a grating pitch p1. Light receiving portions 15-1 and 15-2 having patterns as shown in FIG. 2 are formed on the surfaces of the light receiving elements 14-1 and 14-2, respectively.
The light source 11, the reflection pattern 2-1 and the light receiving element 14-1 form a first optical system, and the light source 11, the reflection pattern 2-2 and the light receiving element 14-2 form a second optical system, respectively.

  FIG. 2 is a diagram illustrating patterns of the light receiving units 15-1 and 15-2. As shown in the figure, sensor portions PS1 to PS4 having the same width are formed in order in the main axis direction at the sensor pitch sp, and are repeatedly formed. In other words, the sensor units PS1 to PS4 are set as one set, and a plurality of sets are formed at a repetition pitch p3 = 4sp. In the embodiment, for example, 13 sets are formed, so the total number of sensor units is 52. The sensor units PS1 to PS4 are formed of, for example, photodiodes and operate as independent light receiving elements. The outputs of a plurality (here, 13) of sensor units PS1 are connected in common and become a detection signal + A. Similarly, the outputs of the plurality of sensor units PS2 to PS4 are connected in common and become detection signals + B, -A, and -B.

  The pattern light receiving portions 15-1 and 15-2 are formed symmetrically with respect to the plane in the principal axis direction passing through the center of the light source 11.

FIG. 3 is a diagram showing an arrangement relationship between the scale substrate 1 and the sensor substrate 10.
As shown in FIG. 3, the light from the light source 11 passes through the grating 12, and in the figure, the light in the left direction is reflected by the reflection pattern 2-1, projected onto the light receiving unit 15-1, and light in the right direction. Is reflected by the reflection pattern 2-2 and projected onto the light receiving unit 15-2.

  Since details of the configuration, arrangement, and projected pattern in FIGS. 1 to 3 are described in Patent Document 1, further description thereof is omitted.

FIG. 4 is a diagram schematically illustrating a path in which light from the light source 11 is reflected by the reflection pattern 2 and projected onto the light receiving unit 15.
As shown in FIG. 4, the reflection pattern 2 is a reflection pattern having a scale pitch p2 (= 20 μm), and the duty is 50%. Therefore, the width of the reflection portion 2R is p2 / 2 (= 10 μm). In the embodiment, the reflective portion 2R is projected onto the light receiving unit 15 as a pattern having a width p2 (= 20 μm). Therefore, the pitch of the pattern projected on the light receiving unit 15 is 2p2 (= 40 μm).

FIG. 5 is a diagram illustrating a reflection pattern projected onto the light receiving unit 15.
As shown in FIG. 5, in the light receiving unit 15, four sensor units PS1 to PS4 are arranged at a pitch p2 / 2 (= 10 μm). On top of this, the projected shading pattern is a shading pattern with a pitch of 2p2 (= 40 μm), and the duty is 50%. Therefore, the width of the dark portion 4A and the light portion 4B is p2 (= 20 μm).

  As shown in FIG. 2, the outputs of the plurality of sensor units PS1 to PS4 are connected in common and become detection signals + A, + B, −A, and −B.

  FIG. 6 is a diagram illustrating changes in detection signals + A, + B, −A, and −B accompanying changes in the relative positional relationship (length) between the scale substrate 1 and the sensor substrate 10. As shown in FIG. 6, the detection signals + A, + B, −A, and −B are sinusoidal signals having a period of 20 μm and are shifted by ¼ phase (90 °). In other words, the detection signals + A and -A are signals that are shifted by 1/2 phase (180 °), and the detection signals + B and -B are signals that are shifted by 1/2 phase (180 °).

FIG. 7 is a diagram illustrating a detection signal calculation circuit used in the optical encoder.
The arithmetic circuit calculates the difference between the outputs of the amplifier circuits 31 to 34 that amplify the detection signals + A, -A, + B, and -B, and the amplifier circuits 31 and 32, and outputs the A-phase signal PHASEA. A circuit 35 and a second difference calculation circuit 36 that calculates a difference between outputs of the amplification circuits 33 and 34 and outputs a B-phase signal PHASEB.

  FIG. 8 is a diagram illustrating changes in the A-phase signal PHASEA and the B-phase signal PHASEB accompanying changes in the relative positional relationship (length) between the scale substrate 1 and the sensor substrate 10. As described above, the detection signals + A and −A are signals that are shifted by 1/2 phase (180 °), and the detection signals + B and −B are signals that are shifted by 1/2 phase (180 °). Even if the amplification factor is 1, the A-phase signal PHASEA and the B-phase signal PHASEB are signals having amplitudes twice that of the detection signals + A, -A, + B, and -B. Also. The A-phase signal PHASEA and the B-phase signal PHASEB are signals that are shifted by ¼ phase (90 °).

  Therefore, by detecting the ratio of the values of the A-phase signal PHASEA and the B-phase signal PHASEB, a phase of less than 90 ° can be detected. Even in an optical encoder that is actually used, the scale pitch or sensor pitch can be detected. The position (length) is detected with a resolution of several tenths. The same applies to the reflective linear optical encoder of the embodiment.

The configuration of the reflective linear optical encoder of the embodiment described above is described in Patent Document 1 and is widely known.
The reflective linear optical encoder of the embodiment described below is different in the reflective pattern 2-2 and the arithmetic circuit.

FIG. 9 is a diagram illustrating the reflection patterns 2-1 and 2-2 in the reflective linear optical encoder of the embodiment.
As described above, the reflection pattern 2-1 is a reflection pattern formed with a scale pitch p2 = 20 μm in the principal axis direction, and the reflection portion R and the transmission portion T have the same width of 10 μm.

  In the reflection pattern 2-2, the width of the transmission portion T at the predetermined position 5 extends to 14 μm, and the left portion of the predetermined position 5 has the same pattern as the reflection pattern 2-1, and the right portion thereof. The reflection pattern R has a reflection pattern in which the reflection portion R and the transmission portion T of the reflection pattern 2-1 are shifted by 4 μm. Here, the predetermined position 5 is the origin. Here, the width of the transmissive portion T at the predetermined position 5 is 14 μm, but may be smaller than 20 μm, and the width of the reflective portion R may be changed.

  FIG. 10 shows light reception when the sensor substrate 10 is arranged as shown in FIG. 3 so that the principal axis directions coincide with the scale substrate 1 having the reflection patterns 2-1 and 2-2 shown in FIG. It is a figure which shows the reflection pattern projected on the part 15-2.

  As shown in FIG. 10, in the light receiving unit 15-2, four sensor units PS1 to PS4 are arranged at a pitch p2 / 2 (= 10 μm). On top of this, the projected shading pattern is a shading pattern with a pitch of 2p2 (= 40 μm) except for a predetermined position, and the duty is 50%. Therefore, the dark portion (portion on which the transmissive portion is projected) 4A and light The width of the portion (the portion on which the reflection portion is projected) 4B is p2 (= 20 μm). The width of the dark portion (portion on which the transmission portion is projected) 4A ′ at the predetermined position is 28 μm.

As described above, the main and sub sensor units are arranged in a set of + A, + B, -A, and -B arranged at a pitch of 10 μm, 13 sets are continuously arranged, and a projected pattern is a pitch of 20 μm. It is twice the arrangement pitch of the sensors. Therefore, if the movement amount is x μm, the intensity of the output signal of the nth one + A cell is + A _n = sin (2π (x / 40)). The output signals of the 13 + A cells are combined to form a + A signal, but the shade pattern projected on the 13 cells is the same, and the intensity of the signal combined with the output signals of the 13 + A cells is As shown in FIG. 6, + A = 13 sin (2π (x / 40)). The same applies to the other cells + B, -A, and -B. As described above, the movement position is calculated from all or a part of these signals.

In the projection situation shown in FIG. 10, + A ₁ , + B ₁ , −A ₁ , −B ₁ to + A ₇ , + B ₇ , −A ₇ , −B ₇ on the left side of the light receiving unit 15-2 are included in the main light receiving unit 15. The same pattern as -1 is projected, but a wide dark portion having a width of 28 μm is projected on + A ₈ , + B ₈ , and −A ₈ on the right side, and −B ₈ , + A ₉ , + B ₉ , −A ₉ , and −B are projected. A pattern shifted by 8 μm is projected from ₉ to + A ₁₃ , + B ₁₃ , −A ₁₃ , and −B ₁₃ .

Therefore, the intensity of the output signal of one + A cell on the left side of the predetermined position is + A _n = sin (2π (x / 40)), whereas the intensity of the output signal of one right + A cell is + A _n = sin (2π (x / 40) −2π × 8/40) = sin (2π (x / 40−1 / 5)). The signals for 13 sets of the light receiving unit 15-2 are a combination of these signals. For example, in the case where a wide transmission portion having a width of 4 μm exists in the eighth set, + A = 8 sin (2π (x / 40)) + 5 sin (2π (x / 40−1 / 5)). If this is generalized as the nth wide pattern (shift pattern) in 13 sets, + A = nsin (2π (x / 40)) + (13−n) sin (2π (x / 40−1 / 5)).

  FIG. 11 shows the signals of 13 sets of cells of the light receiving unit 15-1 on which the reflection pattern 2-1 having a constant pitch is projected and the light receiving unit 15-2 on which the reflection pattern 2-2 having a pitch shift is projected. It is a figure which shows the change when the signal + A moves beyond a predetermined position. A solid line indicates + A of the light receiving unit 15-1, and a broken line indicates + A of the light receiving unit 15-2.

  As shown in FIG. 11, on the left side, the signal + A of the light receiving units 15-1 and 15-2 is the same, but when the end cell of 13 sets moves to a predetermined position, the signal + A of the light receiving unit 15-2 As the amplitude decreases, the phase starts to shift. When the cell of the seventh set moves to a predetermined position, the amplitude of the signal + A of the light receiving unit 15-2 decreases most, and then gradually returns to the original amplitude, but the phase difference further increases, and the opposite side of the 13th set When the end cell passes through a predetermined position, the amplitude is the same as the signal + A of the light receiving unit 15-1, and the phase is shifted by 8 μm.

  FIG. 12 is a diagram illustrating the phase difference between the signals + A of the light receiving units 15-1 and 15-2 according to the movement position in the embodiment. As shown in FIG. 12, the phase position is zero on the left side, the phase difference is 8 μm on the right side, and the width of 520 μm before and after the predetermined position (the portion having the shift pattern with different width) includes the phase difference of zero. The phase difference varies between 8 μm. It can be considered that the change of this part changes linearly.

  The change in the phase difference shown in FIG. 12 is the same between the signals -A, + B and -B of the light receiving units 15-1 and 15-2, and the phase signal PHASEA obtained by calculating the difference between + A and -A. The same applies to the phase signal PHASEB obtained by calculating the difference between + B and -B. Therefore, the difference between the main movement position calculated from the light receiving unit 15-1 and the sub movement position calculated from the light receiving unit 15-2 changes as shown in FIG.

FIG. 13 is a diagram illustrating a configuration of a signal processing unit in the reflective linear optical encoder of the embodiment.
Note that, in the reflective linear optical encoder of the embodiment, similarly to a general reflective linear optical encoder, the calculation for calculating the signal of the light receiving unit 15-1 that receives the reflected pattern of the reflective pattern 2-1. The circuit calculates the movement position with a resolution higher than the period of the scale pitch of the reflection pattern 2-1, but since this is widely known, a description thereof will be omitted. The reflective linear optical encoder of the embodiment is obtained by adding a different origin detection function to the position calculation function described above, and only the calculation processing related to the origin detection will be described.

  The signal processing unit includes a first reading head 41, a second reading head 42, a first interpolator 43, a second interpolator 44, a first counter 45, a second counter 46, and a common clock source. 47, a difference calculation circuit 48, an absolute value conversion circuit 49, a sum calculation circuit 50, a constant multiplication circuit 51, a comparator 52, a delay circuit 53, and an exclusive OR (XOR) gate 54. Have.

  The first reading head 41 and the second reading head 42 correspond to the light receiving units 15-1 and 15-2, and output analog signals + A, + B, -A, and -B. The first interpolator 43 and the second interpolator 44 convert the analog signals + A, + B, -A, -B into digital data and output them. Here, if only the origin detection process is performed, + A (or any of + B, -A, and -B) output from the light receiving units 15-1 and 15-2 may be output as a digital signal. Since the moving position is calculated with a resolution higher than the cycle of the scale pitch of −1, the difference between + A and −A and the difference between + B and −B are calculated and output by the analog circuit shown in FIG. The difference between + A and -A and the difference between + B and -B may be converted into digital data. Further, a change in the movement position calculated from these data may be output.

Therefore, the processes after the first interpolator 43 and the second interpolator 44 are all performed by digital signal processing, but in FIG. 13, in order to make the processing contents easy to understand, they are represented in the form of analog circuits. . In the following description, the first interpolator 43 and the second interpolator 44 output movement position change data, and 40 μm is divided into 2 ^m (for example, if m = 6, 64 equal parts, m = 8 If it is equal to 256, the divided data is output. The first interpolator 43 and the second interpolator 44 output a positive value when moving in the positive direction and a negative value when moving in the negative direction.

  The first counter 45 counts the main position data output from the first interpolator 43 in synchronization with the common clock from the common clock source 47. The second counter 46 counts the sub position data output from the second interpolator 44 in synchronization with the common clock from the common clock source 47. Since the first counter 45 and the second counter 46 perform the counting operation in synchronization with the common clock, there is no time lag between them. Since the first counter 45 and the second counter 46 count the movement position change data, the count values of the first counter 45 and the second counter 46 indicate the movement position.

The difference calculation circuit 48 calculates the difference between the count values of the first counter 45 and the second counter 46, that is, the difference in movement position. The absolute value circuit 49 calculates the difference absolute value of the count values.
The sum calculation circuit 50 and the constant multiplication circuit 51 perform a process of giving hysteresis characteristics to the difference absolute value output from the absolute value circuit 49 in order to prevent frequent determination value fluctuations. Specifically, the constant multiplication circuit 51 multiplies the binary data (zero or 1) output from the comparator 52 by a coefficient (about 10 times the resolution), and the sum calculation circuit 50 outputs the absolute value circuit 49. The output of the constant multiplication circuit 51 is added to the absolute value of the difference output, and the determination result is positively fed back. The range of the hysteresis is determined by the coefficient of the constant multiplication circuit 51. The origin signal is generated at a position that differs by the amount of hysteresis depending on the moving direction of the scale. As described above, the arithmetic circuit that calculates the signal of the light receiving unit 15-1 has a higher resolution than the cycle of the scale pitch of the reflection pattern 2-1. If the movement position is calculated and the position difference is within one scale pitch, there is no problem.

  The comparator 52 compares the output of the sum calculation circuit 50 with the threshold value Vth, and changes the output when the output of the sum calculation circuit 50 changes beyond the threshold value Vth. Specifically, the output of the comparator 52 changes from “1 (H: High)” to “0 (L: Low)” when the output of the sum calculation circuit 50 changes from a state smaller than the threshold value Vth to a larger state. When the output of the sum operation circuit 50 changes from a state larger than the threshold value Vth to a smaller state, the output changes from L to H. For example, when a difference of 8 μm occurs as shown in FIG. 12, Vth is set so that the output of the comparator 52 changes when the output of the sum calculation circuit 50 is a value corresponding to 4 μm.

  The delay circuit 53 delays the output of the comparator 52 for a short time. An exclusive OR (XOR) gate 54 calculates an exclusive OR of the output of the comparator 52 and the output of the delay circuit 53. As a result, when the output of the comparator 52 changes, the XOR gate 54 generates a short pulse. This is the origin detection signal.

  Based on the origin detection signal generated as described above, in the main position data output from the first interpolator 43, the scale pitch including the origin which is the first predetermined position is determined, and the origin is determined from the main position data. Determine.

FIG. 14 is a flowchart illustrating a processing flow of the signal processing unit in the reflective linear optical encoder of the embodiment.
In step S11, the loop number parameter N and the origin signal (one is zero for the origin and 1 for the other) S are initialized to zero.

In step S12, the count values count1 and count2 stored in the registers for the first counter 45 and the second counter 46 are read.
In step S13, a difference Δ between count1 and count2 is calculated.
In step S14, the difference Δ is converted into an absolute value.

In step S15, X is calculated by multiplying the parameter S by the coefficient of the constant multiplication circuit 51.
In step S16, X is added to the absolute value of Δ to calculate a new Δ.
In step S17, it is determined whether or not the new Δ is larger than the threshold value. If it is not larger, the process proceeds to step S18.

In step S18, S = 0.
In step S19, S = 1. Therefore, although there is a hysteresis component, if Δ is large, S = 1, and if Δ is small, S = 0.
In step S20, it is determined whether the parameter N is not zero. If it is zero, the process proceeds to step S21, and if it is not zero (if it is 1, the process proceeds to step S23.

In step S21, N = N + 1. That is, N is increased by 1.
In step S22, the variable Y = S is set.
In step S23, N = 0 is set.

In the above routine, when S20 is reached for the first time, N = 0, so the process proceeds to S21 and S22. After returning to S12, when S20 is reached for the second time, N = 1, the process proceeds to S23. It will be.
In step S24, an exclusive OR of S and Y is calculated, and the calculation result is set as a new S.

  In the above routine, if Δ is twice consecutively smaller than the threshold value or larger than the threshold value, the new S becomes zero and no origin signal is generated. On the other hand, if Δ differs between the two loops, the new S becomes zero and an origin signal is generated. In other words, regardless of the direction, when Δ changes beyond the threshold value, an origin signal is generated.

  As described above, in the reflective linear optical encoder of the embodiment, the signal of the light receiving unit 15-1 that receives the reflected pattern of the reflective pattern 2-1 is received as in the general reflective linear optical encoder. The calculation circuit for calculating calculates the movement position with a resolution higher than the cycle of the scale pitch of the reflection pattern 2-1.

  FIG. 15 is a diagram illustrating the relationship between the relative position signal calculated by the arithmetic circuit that calculates the signal of the light receiving unit 15-1 and the origin signal generated by the origin detection described so far.

  The arithmetic circuit that calculates the signal of the light receiving unit 15-1 uses one cycle when the sensor pitch is one cycle by utilizing the fact that the main A-phase signal indicates the sin component and the main B-phase signal indicates the -cos component. The phase within is calculated. In FIG. 15, phase zero in one cycle is when the main A-phase signal is zero and the main B-phase signal has a negative peak value. If the difference signal is generated near the center (180 degrees) of one cycle including the predetermined position that is the origin regardless of the moving direction, the pitch (cycle) of the scale corresponding to the origin can be determined, and the origin position Can be identified.

  Since the reflective linear optical encoder of the embodiment described above uses a count value after digital processing, it can be realized only by software without being mounted on a circuit. Since the origin signal is generated by detecting the count value in real time, it is desirable to complete the processing on software. Software processing may be executed by providing a dedicated chip, or may be executed on a device such as an external arithmetic unit, for example.

  For example, when an analog circuit compares signal voltages with a comparator, a change in the absolute value of the signal voltage itself affects the determination result. In particular, when the amount of light source varies, the maximum value of the signal voltage varies, so that an extra circuit such as normalization is required. On the other hand, if the reflective linear optical encoder of the embodiment correctly outputs the signal of the pattern projected by the first and second reading heads, the subsequent processing is performed by digital processing. Not affected.

In the embodiment described abnormally, a reflective scale is used, but a transmissive scale can also be used.
Furthermore, the calculation processing procedure is an example, and any processing procedure may be used as long as a change in misalignment can be detected.

  The present invention can be applied to various optical encoders.

DESCRIPTION OF SYMBOLS 1 Scale board | substrate 2-1, 2-2 Reflection pattern 5 Predetermined position 10 Sensor board | substrate 11 Light source 14-1, 14-2 Light receiving element 15-1, 15-2 Light receiving part 35 1st difference calculation circuit 36 2nd difference calculation circuit 41 First Reading Head 42 Second Reading Head 43 First Interpolator 44 Second Interpolator 45 First Counter 46 Second Counter 48 Difference Operation Circuit 50 Sum Operation Circuit 52 Comparator

Claims (3)

A first optical scale having a first light and shade pattern that changes in the main axis direction, a first light source that outputs the first illumination light of the first optical scale, and the first optical scale with the first illumination light. A first light receiving sensor arranged to detect an image of the first gray pattern generated by illuminating, wherein the first light receiving sensor is arranged in order at a sensor pitch in the main axis direction. A first optical system having a first sensor array in which a first sensor group including first to fourth main sensor units is repeatedly arranged in a predetermined number of sets in the main axis direction;
A second optical scale having a second shading pattern that changes in the principal axis direction; a second light source that outputs a second illumination light of the second optical scale; and the second optical scale using the second illumination light. And a second light receiving sensor arranged so as to detect an image of the second gray pattern generated by illuminating the light, and the second light receiving sensors are sequentially arranged at the sensor pitch in the main axis direction. A second optical system having a second sensor array in which a second sensor set including the first to fourth sub sensor units is repeatedly arranged a predetermined number of times in the main axis direction;
A signal processing unit that computes and processes a detection signal of the first light receiving sensor of the first optical system and a detection signal of the second light receiving sensor of the second optical system;
The first optical scale and the second optical scale move relative to the first light source, the first light receiving sensor, the second light source, and the second light receiving sensor;
The first light and shade pattern of the first optical scale has the same width as the dark portion and the light portion, and repeats light and shade at a scale pitch.
The second light and shade pattern of the second optical scale has the same width of the dark portion and the light portion, and repeats the light and shade at the scale pitch. / 2 is wide within a range not exceeding 2.
The signal processing unit
A main position data generation unit that converts at least one of the outputs of the first to fourth main sensor units into a digital signal and outputs main position data;
A sub position data generation unit that converts at least one of the outputs of the first to fourth sub sensor units into a digital signal and outputs sub position data;
A difference calculation unit for calculating difference data between the main position data and the sub position data;
An optical encoder comprising: an origin determination unit that determines whether an absolute value of the difference data is equal to or greater than a predetermined value and detects an origin from a change in the determination value. The first optical scale and the second optical scale are adjacent to each other on a common scale substrate, and the first gray pattern and the second gray pattern are parallel to each other, and are dark on one side of the predetermined position. And the position of the light part matches, and the other part is formed so that the dark part and the light part are shifted,
The first light source, the first light receiving sensor, the second light source, and the second light receiving sensor are formed on a common sensor substrate, and the first light source and the second light source are formed as a common light source, The first light receiving sensor and the second light receiving sensor have the first sensor array and the second sensor array in parallel on both sides of the common light source, the first to fourth main sensor units, and the first to fourth sensors. It is formed so that the position of the sub main sensor part matches,
The sensor substrate is configured such that illumination light from the common light source reflected by the first light and shade pattern is projected onto the first light receiving sensor and reflected by the second light and shade pattern with respect to the scale substrate. The optical encoder according to claim 1, wherein the optical encoder is disposed so that illumination light from the first light receiving sensor is projected onto the second light receiving sensor. The scale pitch is twice the sensor pitch,
The optical sensor according to claim 2, wherein the sensor substrate is arranged such that the first gray pattern and the second gray pattern are projected as a light / dark pattern four times the sensor pitch with respect to the scale substrate. Type encoder.

Images