JP2023019809A - Absolute encoder and device equipped with the same - Google Patents

Abstract

To provide a compact encoder that corrects an eccentric component after predicting eccentricity and eccentric angle.SOLUTION: The encoder comprises a scale which is equipped with a first and a second track provided in different diameter, a sensor which is capable of moving relative to the scale and reads the first and second tracks, and a processing unit for acquiring an absolute position signal that indicates the absolute position of one of the scale and the sensor. The processing unit acquires a position signal using a first and a second cyclic signal based on the signal obtained by the sensor by reading the first and second tracks, acquires eccentricity and an eccentric angle using a third cyclic signal that is generated when acquiring the position signal, acquires using the eccentricity and eccentric angle, a first and a second eccentricity correction signal that correspond to the first and second tracks, and acquires an absolute position signal using a first eccentricity correction cyclic signal based on the first cyclic signal and first eccentricity correction signal and a second eccentricity correction cyclic signal based on the second cyclic signal and second eccentricity correction signal.SELECTED DRAWING: Figure 1

Description

The present invention relates to an absolute rotary encoder and an apparatus provided therewith.

In an absolute rotary encoder that detects the position of a movable member according to the relative movement of the scale and sensor, in order to detect the position of the movable member with high accuracy, it is necessary to take measures to compensate for the eccentricity of the scale with respect to the axis of rotation of the scale. Is required.

Patent Document 1 discloses an absolute rotary actuator that corrects an eccentric component by averaging signals obtained from two sensors arranged to face each other in the radial direction of a scale. An encoder is disclosed.

Further, Patent Literature 2 discloses a position detection device that corrects an eccentric component by subtracting an error component due to a difference between the center of rotation of the scale and a predetermined point on the scale from the position information.

JP 2020-091104 A JP 2014-178227 A

However, in the absolute rotary encoder of Patent Document 1, since the two sensors are arranged so as to face each other in the radial direction of the scale, the size of the holding member for the sensors becomes large, which prevents downsizing of the device.

Further, the position detection device of Patent Document 2 can perform eccentricity correction, but cannot determine the amount of eccentricity that indicates the degree of eccentricity and the eccentric angle that indicates at what angle the eccentricity occurs. If the amount of eccentricity and the angle of eccentricity can be determined, it is also possible to manually adjust the eccentricity component.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a compact absolute rotary encoder that predicts the amount of eccentricity and the angle of eccentricity and corrects the eccentricity component.

An absolute rotary encoder as one aspect of the present invention includes a scale having a first track and a second track provided with mutually different diameters; and a processing unit that acquires an absolute position signal indicating the absolute position of either the scale or the sensor. A position signal is obtained by using the signal and a second periodic signal based on a signal obtained by reading the second track by the sensor, and a third periodic signal generated when obtaining the position signal is used to determine the amount of eccentricity and eccentricity. A first eccentricity correction signal corresponding to the first track and a second eccentricity correction signal corresponding to the second track are obtained using the eccentricity amount and the eccentricity angle, and a first periodic signal and a second eccentricity correction signal are obtained. The absolute position signal is obtained by using a first eccentricity correction periodic signal based on the first eccentricity correction signal and a second eccentricity correction periodic signal based on the second periodic signal and the second eccentricity correction signal. .

According to the present invention, it is possible to provide a compact absolute rotary encoder that predicts the amount of eccentricity and the angle of eccentricity and corrects the eccentric component.

3 is a diagram showing the configuration of an encoder of Example 1. FIG. 4 is a partially enlarged view of the scale of Example 1. FIG. 4 is a diagram showing the configuration of the sensor of Example 1. FIG. FIG. 10 is an explanatory diagram of switching of a detection cycle according to the first embodiment; 3 is a diagram showing the configuration of a processing unit of Example 1; FIG. 4 is a flow chart showing absolute position acquisition processing and eccentricity correction processing of Embodiment 1. FIG. FIG. 10 is a diagram showing rounding errors when an absolute position is correctly acquired; FIG. 10 is a diagram showing rounding errors in a state where there is a high possibility that an absolute position cannot be acquired correctly; FIG. 10 is a diagram showing rounding errors in a state where eccentricity occurs; FIG. 4 illustrates rounding errors with wraparound; 10 is a flowchart showing loopback processing; FIG. 10 is a diagram showing the configuration of an encoder of Example 2; It is a figure which expands a part of scale of Example 2. FIG. FIG. 10 is a diagram showing a reading area of a scale in the sensor of Example 2; 10 is a flow chart showing absolute position acquisition processing and eccentricity correction processing of Embodiment 2. FIG. FIG. 10 is a diagram illustrating an example of signal processing in Example 2; FIG. 12 is a diagram showing a robot arm of Example 5;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In each figure, the same reference numerals are given to the same members, and overlapping descriptions are omitted.

FIG. 1 is a diagram showing the configuration of an encoder 1a of this embodiment. The encoder 1 a is a reflective optical absolute rotary encoder having a scale 10 , a sensor 20 and a processing section 30 . In this embodiment, a case will be described in which the rotational position (rotational angle) of a rotating shaft (not shown) of a movable member is detected by acquiring the absolute position of either the scale 10 or the sensor 20 .

In this embodiment, a reflective optical absolute rotary encoder will be described as an example of the encoder 1a, but the present invention is not limited to this. The present invention can also be applied to encoders with different detection methods, such as transmission optical absolute rotary encoders, magnetic absolute rotary encoders, and electromagnetic induction magnetic absolute rotary encoders.

The scale 10 is attached so as to rotate integrally with the rotating shaft of the movable member. The sensor 20 is attached to a fixed member (not shown) and is movable relative to the scale 10 . Alternatively, the scale 10 may be attached to the fixed member and the sensor 20 may be attached to the rotating shaft of the movable member.

FIG. 2 is an enlarged view of a part of the scale 10. A plurality of (two in this embodiment) tracks (first tracks) 11 and tracks (second tracks) provided on the scale 10 with different diameters. ) 12. Each track is provided with a periodic pattern including reflective portions (black portions in the drawing) and non-reflective portions (white portions in the drawing) alternately arranged at a constant cycle (pitch). Specifically, the track 11 is provided with a periodic pattern with a pitch of P1, and the track 12 is provided with a periodic pattern with a pitch of Q1. The numbers of gratings of the periodic patterns with pitches P1 and Q1 are 97 and 24, respectively. Note that the reflective portion and the non-reflective portion of each periodic pattern may be reversed.

FIG. 3 is a diagram showing the configuration of the sensor 20. As shown in FIG. The sensor 20 includes a light source 21 and a plurality of (two in this embodiment) light receiving portions 22 and 23 . The light source 21, the light receiving section (first detecting section) 22, and the light receiving section (second detecting section) 23 are arranged on the same plane. The light source 21 is composed of a light-emitting element such as an LED. The light receiving section 22 is composed of a plurality of photoelectric conversion elements (light receiving elements) that photoelectrically convert light emitted from the light source 21 and reflected by the reflecting section of the track 11 . The light receiving section 23 is composed of a plurality of photoelectric conversion elements (light receiving elements) that photoelectrically convert light emitted from the light source 21 and reflected by the reflecting section of the track 12 . When the scale 10 and the sensor 20 are displaced relative to each other, the intensity of the reflected light (received light intensity) received by each light receiving element in the light receiving sections 22 and 23 changes according to the amount of relative displacement. The sensor 20 outputs a sinusoidal signal corresponding to changes in the intensity of light received by the light receiving section 22 and outputs a sinusoidal signal corresponding to changes in the intensity of light received by the light receiving section 23 .

Next, switching of the detection period will be described with reference to FIG. FIG. 4 is an explanatory diagram of switching of the detection cycle. The detection cycle is switched by changing the arrangement of the light receiving elements that output the signals A(+), B(+), A(-), B(-).

In the light receiving section 22, as shown in FIG. 4A, light receiving elements 20a, 20b, 20c, and 20d are cyclically arranged along the position detection direction. The outputs of the light receiving elements 20a, 20b, 20c and 20d are treated as signals A(+), B(+), A(-) and B(-), respectively. The detection cycle of the light-receiving unit 22 is always constant, and is set to a detection cycle P1 that matches or is sufficiently close (substantially matches) to the pitch P1.

In the light receiving section 23, as shown in FIG. 4B, four adjacent light receiving elements among the 16 light receiving elements arranged along the position detection direction form one set, and each set of light receiving elements 20a and 20b , 20c and 20d are configured to have one output. Each set of outputs is treated as signals A(+), B(+), A(-), B(-), respectively. The detection cycle of the light-receiving unit 23 is always constant and is set to 4×P1, which matches or is sufficiently close (substantially matches) to the pitch Q1.

For the signals A(+), B(+), A(-), B(-), the equations A=A(+)-A(-) and B=B(+)-B(-) are expressed. By performing the processing described above, two-phase quasi-sine wave signals A and B having different phases are obtained.

FIG. 4C shows the reading area of the scale 10 in the sensor 20. FIG. When the detection period is set to P1, the light receiving section 22 reads the periodic pattern 11a with the pitch P1. In this case, two-phase pseudo sine wave signals (two-phase signals of P1) having a phase difference of about 90 degrees corresponding to the pitch P1 are output. When the detection period is set to 4×P1, the light receiving section 23 reads the periodic pattern 12a with the pitch Q1. In this case, a two-phase pseudo sine wave signal (Q1 two-phase signal) having a phase difference of about 90 degrees corresponding to the pitch Q1 is output.

A reading area 120 on the track by the light receiving section 22 and a reading area 130 on the track by the light receiving section 23 are ranges in which each light receiving section receives light that is irradiated from the light source 21 and reflected by each periodic pattern.

In this embodiment, the sensor outputs two-phase signals whose phases are different by about 90 degrees, but a three-phase signal, a triangular wave signal, or the like may be output as long as the phase can be detected.

In this embodiment, the same sensor element whose detection period can be switched is used to detect a plurality of periodic patterns, but the present invention is not limited to this, and separate sensor elements having different detection periods are used. A sensor element may be used.

FIG. 5 is a diagram showing the configuration of the processing unit 30. As shown in FIG. The processing unit 30 includes an AD converter 31 , a phase calculator 32 , an absolute position calculator 33 and an eccentricity corrector 34 . The AD converter 31 converts the two-phase signals (the P1 two-phase signal and the Q1 two-phase signal) output from the light receiving portions 22 and 23 of the sensor 20 into digital signals. The phase calculator 32 is composed of elements such as a CPU and circuits such as FPGA and ASIC, and obtains a phase using a two-phase signal converted into a digital signal by the AD converter 31 . The absolute position calculator 33 is composed of elements such as a CPU and circuits such as FPGA and ASIC, and obtains the absolute position using the phase obtained by the phase calculator 32 . The eccentricity correction unit 34 is composed of elements such as a CPU and circuits such as FPGA and ASIC, and acquires the amount of eccentricity of the scale 10 with respect to the rotation axis. Note that when the sensor 20 is attached to the rotating shaft, the eccentricity of the sensor 20 with respect to the rotating shaft is acquired.

FIG. 6 is a flow chart showing absolute position acquisition processing and eccentricity correction processing of this embodiment.

In step S101, the AD converter 31 converts the two-phase signals (the P1 two-phase signal and the Q1 two-phase signal) output from the light receiving sections 22 and 23 of the sensor 20 into digital signals.

In step S102, the phase calculator 32 acquires the phase using the two-phase signal converted into the digital signal in step S101. As described above, a two-phase signal is a signal having a phase difference of about 90 degrees, so the phase is obtained by performing the arctan operation. Hereinafter, the phase (first periodic signal) obtained from the two-phase signal of P1 is assumed to be θP97, and the phase (second periodic signal) obtained from the two-phase signal of Q1 is assumed to be θQ24. As described above, the numbers of gratings of the periodic patterns with pitches P1 and Q1 are 97 and 24, respectively. Therefore, the phases .theta.P97 and .theta.Q24 are respectively a 97-cycle signal and a 24-cycle signal. Further, the accuracy within one cycle of the phase θP97 is high, and the accuracy within one cycle of the phase θQ24 is low. In this embodiment, the phase is obtained by the arctan calculation, but parameters other than the phase may be obtained as long as they represent the position within a specific range.

In step S103, the absolute position calculator 33 obtains the absolute position signal x1 by multiplying the phases θP97 and θQ24 by an integer using the following equation (1). MOD(x, y) represents the remainder when x is the dividend and y is the divisor.

In step S104, the absolute position calculator 33 obtains a highly accurate absolute position signal x97_1 by processing the absolute position signal x1. Specifically, the absolute position calculator 33 synthesizes the absolute position signal x1 and the phase θP97 using the following equations (2) and (3). As a result, the absolute position signal x97_1 having the number of cycles m97 of the phase θP97 and the accuracy of the phase θP97 is obtained. However, ROUND(x) represents an integer value obtained by rounding off the first decimal place of x.

The eccentricity acquisition process and the eccentricity correction process in this embodiment will be described below. First, the rounding error d representing the reliability of the absolute position used in the eccentricity correction process will be described.

A rounding error is a difference between values before and after rounding in rounding processing when obtaining the number of cycles. Taking the above equation (2) as an example, the number of cycles m97 of the phase θP97 is obtained using ROUND(x). That is, since the rounded value is the number of cycles m97, the rounding error in Equation (2) can be obtained by subtracting the value before rounding (before rounding) from the number of cycles m97. The rounding error is expressed within the range of ±0.5, and if it is close to +0.5 or -0.5, it can be said that the rounding process is not performed correctly. That is, there is a high possibility that the number of cycles is shifted and the absolute position cannot be acquired correctly.

FIG. 7 is a diagram showing rounding errors when the absolute position is correctly obtained. FIG. 8 is a diagram showing rounding errors when there is a high possibility that the absolute position cannot be acquired correctly. 7 and 8, (A) to (C) respectively show the number of cycles after rounding, the number of cycles before rounding, and the difference between the number of cycles after rounding and the number of cycles before rounding (rounding error). there is

As for the rounding error, when the scale 10 is eccentric with respect to the rotation axis, a signal of one cycle is output as shown in FIG. 9A to 9C show rounding errors when the scale 10 is eccentric by 100 μm, 200 μm, and 300 μm, respectively. In the eccentricity correction process of this embodiment, eccentricity correction is performed using the rounding error.

In step S105, the eccentricity correction unit 34 uses the rounding error d to obtain an eccentricity amount A representing the amount of deviation between the center of the rotation axis and the center of the scale. In this embodiment, as shown in FIG. 9, the amount of eccentricity is obtained using the feature that the difference (peak-to-peak value) dp_p between the maximum value and the minimum value of the rounding error increases as the amount of eccentricity increases. The eccentricity correction unit 34 first rotates the scale 10 once and obtains the rounding error of the entire circumference of the scale 10 . Next, the eccentricity correction unit 34 multiplies the rounding error by 100 so that it is expressed as ±50 because the rounding error is expressed as ±0.5. After that, the eccentricity correction unit 34 acquires the peak-to-peak value dp_p of the rounding error, and acquires the eccentricity amount A using the following equation (4).

Formula (4) is an approximation formula obtained by actual measurement, but it is desirable to change it appropriately according to the environment using the relationship between the eccentricity A and the peak-to-peak value dp_p of the rounding error d.

It should be noted that the rounding error is multiplied by 100 to make the notation of ±50, but this is done for ease of handling when processing, and is not an essential process.

Further, although the rounding error d is obtained by rotating the scale 10 once here, the peak-to-peak value dp_p of the rounding error may be obtained from the difference between the maximum value and the minimum value of the moving average value by performing moving average processing. By executing the moving average process, the influence of noise and the like can be reduced, and the eccentricity A can be obtained more accurately.

In step S106, the eccentricity correction unit 34 first acquires the eccentricity angle θ, which is the absolute position signal (angle) when the rounding error d is maximized, using the rounding error d. Next, the eccentricity correction unit 34 acquires an eccentricity correction signal that is a signal in phase opposite to the eccentricity component, and calculates the phase shift amount Δθ, which is the difference between the eccentricity angle θ and the angle at which the eccentricity correction signal is the minimum value, as follows. It is obtained using Equation (5).

In this embodiment, the angle at which the eccentricity correction signal has the minimum value is 180 degrees.

As the eccentricity A increases, the rounding error d may have a folded waveform as shown in FIG. 10(A). In this case, in the ±0.5 notation, the maximum value of the peak-to-peak value dp_p is 1, and an accurate amount of eccentricity A cannot be obtained. Also, the maximum value of the rounding error d, the eccentric angle θ, and the phase shift amount Δθ cannot be obtained accurately. Therefore, it is necessary to express the notation of the rounding error as ±0.5 or more. Hereinafter, the processing performed to express ±0.5 or more will be referred to as folding processing.

FIG. 11 is a flow chart showing loopback processing.

In step S201, the eccentricity correction unit 34 acquires the rounding error of the entire circumference of the scale 10 at regular intervals.

In step S202, the eccentricity correction unit 34 determines whether the absolute value of the difference between the acquired rounding error and the previously acquired rounding error exceeds 0.5. When it is determined that the absolute value exceeds 0.5, that is, when it is determined that folding has occurred, the process proceeds to step S203, and when it is determined otherwise, the process proceeds to step S204.

In step S203, the eccentricity correction unit 34 adds 1 to or subtracts 1 from the acquired rounding error, and acquires the rounding error within the range of ±1.5 as shown in FIG. 10B. For example, if the rounding error obtained last time is the rounding error d1 at the rotation angle of 30 degrees in FIG. 10A, and the obtained rounding error is the rounding error d2 at the rotation angle of 60 degrees in FIG. is greater than 0.5. Therefore, it is determined that folding occurs at a rotation angle of 60 degrees, and 1 is added to the rounding error d2. By performing such processing, the waveform shown in FIG. 10B can be acquired.

In step S204, the eccentricity correction unit 34 acquires the average value dave of the rounding error d using the equation (6) because the acquired rounding error d may be offset, and uses the equation (7) to acquire the rounding error d is corrected to obtain the corrected rounding error d'. Note that Average(x) represents the average value of x.

The corrected rounding error d' has a waveform shown in FIG. 10(c).

As described above, by expressing the rounding error d as ±0.5 or more and correcting the offset, even if the amount of eccentricity is large and the rounding error d is folded back, the peak-to-peak value dp_p, the eccentricity angle θ, and the phase shift amount Δθ can be obtained accurately.

In step S107, the eccentricity correction unit 34 uses the eccentricity amount A and the phase shift amount Δθ to obtain a first eccentricity correction signal (first eccentricity correction periodic signal) Msig corresponding to the track 11 and a second eccentricity correction signal Msig corresponding to the track 12. A correction signal (second eccentricity correction periodic signal) Ssig is acquired.

A method for acquiring the first eccentricity correction signal Msig will be described below. First, the eccentricity correction unit 34 obtains the position Mpos where the light receiving unit 22 is arranged using the following formula (8). Note that r is the radius of the scale 10 and D1 is the distance between the light source 21 and the light receiving section 22 .

Next, the eccentricity correction unit 34 acquires the angle angle of the eccentricity correction signal using the following equation (9).

Next, the eccentricity correction section 34 acquires the first eccentricity correction signal Msig using the following equation (10).

A method of acquiring the second eccentricity correction signal Ssig will be described below. First, the eccentricity correction unit 34 obtains the position Spos at which the light receiving unit 23 is arranged using the following formula (11). Note that r is the radius of the scale 10 and D2 is the distance between the light source 21 and the light receiving section 23 .

As for the angle angle of the eccentricity correction signal, the angle obtained when the first eccentricity correction signal Msig is obtained is used.

Next, the eccentricity correction section 34 acquires the second eccentricity correction signal Ssig using the following equation (12).

Incidentally, in this embodiment, the radius of the scale 10 and the boundaries between the tracks 11 and 12 match.

In step S108, the absolute position calculator 33 adds the first eccentricity correction signal Msig to the phase θP97 to acquire the phase θP97′ from which the eccentric component is removed, as shown in the following equation (13). Further, the absolute position calculator 33 adds the second eccentricity correction signal Ssig to the phase θQ24 to acquire the phase θQ24′ from which the eccentric component is removed, as shown in the following equation (14).

In the present embodiment, the eccentricity component is reduced by adding the eccentricity correction signal, which is a signal of opposite phase to the eccentricity component, to the eccentricity component. By doing so, the eccentricity component may be reduced.

In step S109, the absolute position calculator 33 acquires the absolute position signal x1' using the following equation (15).

In step S110, the absolute position calculation unit 33 synthesizes the absolute position signal x1′ and the phase θP97′ using the following equations (16) and (17) to obtain the number of cycles m97′ and phase θP97′ of the phase θP97′. An absolute position signal x97_1' having an accuracy of θP97' is obtained.

As described above, according to this embodiment, it is possible to provide a compact absolute rotary encoder that predicts the amount of eccentricity and the angle of eccentricity and corrects the eccentric component.

In the present embodiment, the scale 10 is rotated in a fixed direction, but the eccentricity component may be predicted in a shorter interval by rotating the scale 10 toward the maximum or minimum value of the rounding error. Alternatively, the maximum and minimum values of the rounding error may be predicted from the slope of the rounding error near the zero cross without measuring the maximum or minimum value.

In this embodiment, a case of using a scale having a configuration different from that of the first embodiment will be described.

FIG. 12 is a diagram showing the configuration of the encoder 1b of this embodiment. The encoder 1b has a scale 100, a sensor 200, and a processing section 300. FIG. The encoder 1b is a reflective optical absolute rotary encoder that detects the absolute position of the movable member. In this embodiment, a case will be described in which the rotational position (rotational angle) of a rotating shaft (not shown) of a movable member is detected by acquiring the absolute position of either the scale 10 or the sensor 20 .

In this embodiment, a reflective optical absolute rotary encoder will be described as an example of the encoder 1b, but the present invention is not limited to this. The present invention can also be applied to encoders with different detection methods, such as transmission optical absolute rotary encoders, magnetic absolute rotary encoders, and electromagnetic induction magnetic absolute rotary encoders.

Since the mounting of the scale 100 and the sensor 200 is the same as in the first embodiment, the description is omitted. In the following description, the parts different from the first embodiment will be mainly described.

FIG. 13 is an enlarged view of a part of the scale 100. A plurality of (two in this embodiment) tracks (first track) 111 and a track (second track) provided on the scale 100 with different diameters. ) 112. Each track is alternately provided with a plurality of periodic patterns having different pitches along a direction perpendicular to the detection direction (scale width direction). Specifically, the track 111 has a periodic pattern with a pitch of T1 and a periodic pattern with a pitch of T2 alternately along the scale width direction. The numbers of gratings of the periodic patterns with pitches T1 and T2 are 1649 and 388, respectively. Also, on the track 112, the periodic pattern with the pitch U1 and the periodic pattern with the pitch U2 are alternately provided along the scale width direction. The numbers of gratings of the periodic patterns of the pitches U1 and U2 are 1632 and 384, respectively. The number of gratings of the periodic patterns of the pitch T1 and the pitch U1 and the pitch T2 and the pitch U2 are almost equal.

Since the configurations of the sensor 200 and the processing unit 300 are the same as those of the first embodiment, description thereof is omitted. However, the sensor 200 can switch the detection cycles of the light receiving units 22 and 23 between the pitches T1 and U1 and the pitches T2 and U2 in accordance with the detection cycle switching signal from the processing unit 300 . The switching of the detection period is the same as in FIGS. 4A and 4B, so the explanation is omitted.

FIG. 14 is a diagram showing the reading area of the scale 100 in the sensor 200. As shown in FIG. When the detection period is set to the pitch T1, the light receiving section 22 reads the periodic pattern 111a of the pitch T1. In this case, two-phase pseudo sine wave signals (two-phase signals of T1) having a phase difference of about 90 degrees corresponding to the pitch T1 are output. When the detection period is set to the pitch T2, the light receiving section 22 reads the periodic pattern 111b of the pitch T2. In this case, two-phase quasi-sine wave signals (two-phase signals of T2) having a phase difference of about 90 degrees corresponding to the pitch T2 are output. When the detection period is set to the pitch U1, the light receiving section 23 reads the periodic pattern 112a of the pitch U1. In this case, two-phase pseudo sine wave signals (two-phase signals of U1) having a phase difference of about 90 degrees corresponding to the pitch U1 are output. When the detection period is set to the pitch U2, the light receiving section 23 reads the periodic pattern 112b of the pitch U2. In this case, two-phase pseudo sine wave signals (two-phase signals of U2) having a phase difference of about 90 degrees corresponding to the pitch U2 are output.

A reading area 121 on the track by the light receiving section 22 and a reading area 131 on the track by the light receiving section 23 are ranges in which each light receiving section receives light that is irradiated from the light source 21 and reflected by each periodic pattern. Each reading area is set to include a plurality of combinations of two periodic patterns alternately arranged along the scale width direction on the track.

In this embodiment, the sensor outputs two-phase signals whose phases are different by about 90 degrees, but a three-phase signal, a triangular wave signal, or the like may be output as long as the phase can be detected.

Further, in the present embodiment, the switching of the detection period of the light receiving section 22 and the light receiving section 23 is performed by the processing section 300, but it may be performed by a signal from an external device.

FIG. 15 is a flow chart showing absolute position acquisition processing and eccentricity correction processing of this embodiment.

In step S301, the processing unit 300 switches the detection cycles of the light receiving units 22 and 23 to the pitches T1 and U1. As a result, the light receiving portion 22 outputs a two-phase signal of T1, and the light receiving portion 23 outputs a two-phase signal of U1.

In step S302, the A/D converter 31 converts the T1 two-phase signal and the U1 two-phase signal into digital signals.

In step S303, the processing unit 300 switches the detection cycles of the light receiving units 22 and 23 to the pitches T2 and U2. As a result, the light receiving section 22 outputs a two-phase signal of T2, and the light receiving section 23 outputs a two-phase signal of U2.

In step S304, the A/D converter 31 converts the T2 two-phase signal and the U2 two-phase signal into digital signals.

In step S<b>305 , the phase calculator 32 acquires phases from the two sets of two-phase signals converted into digital signals by the AD converter 31 . As described above, a two-phase signal is a signal having a phase difference of about 90 degrees, so the phase is obtained by performing the arctan operation. In the following description, the phase obtained from the two-phase signal T1 is θT1, the phase obtained from the two-phase signal T2 is θT2, the phase obtained from the two-phase signal U1 is θU1, and the phase obtained from the two-phase signal U2 is θT2. Let the obtained phase be θU2. As described above, the numbers of gratings of the periodic patterns with pitches T1, T2, U1 and U2 are 1649, 388, 1632 and 384, respectively. Therefore, the phases .theta.T1, .theta.T2, .theta.U1 and .theta.U2 are respectively a signal of 1649 cycles, a signal of 388 cycles, a signal of 1632 cycles and a signal of 384 cycles.

In step S306, the absolute position calculator 33 acquires a phase θT97 of 97 cycles, a phase θT388 of 388 cycles, and a phase θT1649 of 1649 cycles using the following equations (18) to (20).

Note that the phase θT1649 has the highest precision within one cycle, and the phase θT97 has the lowest precision within one cycle. Therefore, in this embodiment, phases of 97 cycles with an accuracy of phase θT1649 are acquired.

Specifically, as shown in FIG. 16, the phase θT388 changes 388 times from 0 to 2π along the entire circumference of the track, while the phase θT97 changes 97 times from 0 to 2π along the circumference of the track. That is, the amount of change in phase θT97 is ¼ of phase θT388.

In step S307, the absolute position calculator 33 obtains the number of cycles m200_4 of the phase θT388 and the signal y388_97 of 97 cycles having the accuracy of the phase θT388 using the following equations (21) and (22).

In step S308, the absolute position calculator 33 acquires the number of cycles m200_17 of the phase θT1649 and the signal y1649_97 of 97 cycles having the accuracy of the phase θT1649 using the following equations (23) and (24).

In step S309, the absolute position calculator 33 acquires the phase θU96 of 96 cycles, the phase θU384 of 384 cycles, and the phase θU1632 of 1632 cycles using the following equations (25) to (27).

In step S310, the absolute position calculator 33 acquires the number of cycles n200_4 of the phase θU384 and the signal y384_96 of 96 cycles having the accuracy of the phase θU384 using the following equations (28) and (29).

In step S311, the absolute position calculator 33 acquires the number of cycles n200_17 of the phase θU1632 and the signal y1632_96 of 96 cycles having the accuracy of the phase θU1632 using the following equations (30) and (31).

In step S312, the absolute position calculator 33 acquires the absolute position signal y1 from the sensor 200, which has one cycle for the entire circumference of the track, using the following equation (32).

In step S313, the absolute position calculator 33 acquires a highly accurate absolute position signal y1649_1 by processing the absolute position signal y1. Specifically, the absolute position calculator 33 synthesizes the absolute position signal y1 and the signal y1649_97 using the following equations (33) and (34). As a result, the absolute position signal y1649_1 having the number of cycles m200_97 of the signal y1649_97 and the accuracy of the signal y1649_97 is acquired.

In step S314, the eccentricity correction unit 34 performs eccentricity correction and acquires an absolute position signal after eccentricity correction. A rounding error that can be obtained when obtaining the number of cycles m200_97 is used as the rounding error used for eccentricity correction. Since the processing after eccentricity correction is the same as that of the first embodiment, a detailed description thereof will be omitted.

As described above, according to this embodiment, it is possible to provide a compact absolute rotary encoder that predicts the amount of eccentricity and the angle of eccentricity and corrects the eccentric component.

In this embodiment, a method of obtaining the eccentricity A and the eccentricity angle θ by a method different from that of the first and second embodiments will be described. Descriptions of the same parts as those in the first and second embodiments are omitted.

In Embodiments 1 and 2, the eccentricity amount A is obtained by applying the rounding error peak-to-peak value dp_p to the eccentricity obtaining formula (e.g., formula (4)). Acquire the amount of eccentricity A. Specifically, when the peak-to-peak value dp_p of the rounding error is less than 30, the eccentricity is set to the first eccentricity (for example, 0 μm), and when the peak-to-peak value dp_p is 30 or more and less than 55, the eccentricity is set to A second amount of eccentricity (for example, 50 μm) is set. As a result, the eccentricity can be obtained without applying the eccentricity obtaining formula.

However, since the threshold shown here is an example, it may be changed as appropriate. At that time, it may be derived from the relationship between the eccentricity amount A and the peak-to-peak value dp_p of the rounding error.

In Examples 1 and 2, the angle at which the rounding error value is maximized is the eccentricity angle θ, but in this example, the angle at which the rounding error value is minimized is used to obtain the eccentricity angle θ. Specifically, when the angle at which the value of the rounding error is minimized is θdmin, the eccentric angle θ is obtained using the following equation (35).

Since subsequent processes are the same as those in the first and second embodiments, the description thereof is omitted.

As described above, according to this embodiment, it is possible to provide a compact absolute rotary encoder that predicts the amount of eccentricity and the angle of eccentricity and corrects the eccentric component.

In this embodiment, a method of obtaining the eccentricity A and the eccentricity angle θ by a method different from that of the first to third embodiments will be described. Specifically, in Examples 1 to 3, the scale is rotated once to acquire the rounding error, but in this example, the scale is rotated until either the maximum value or the minimum value of the rounding error is acquired. This makes it possible to correct eccentricity even when the scale cannot make one rotation. Descriptions of the same portions as those of Examples 1 to 3 are omitted.

In Examples 1 to 3, the eccentricity A can be obtained when the peak-to-peak value dp_p of the rounding error is obtained. In this embodiment, the rounding error is a signal of one cycle for one rotation, so the peak-to-peak value dp_p is either the maximum value or the minimum value of the rounding error by using the following equations (36) and (37). By doing so, the eccentricity A can be obtained. Note that dmax is the maximum value of rounding error, and dmin is the minimum value of rounding error. Also, ABS(x) is the absolute value of x.
(if you know the maximum value)

(if you know the minimum value)

In subsequent processing, the eccentricity amount A is obtained using the peak-to-peak value dp_p in the same manner as in the first to third embodiments. The eccentric angle θ can also be obtained by grasping the maximum value or minimum value of the rounding error as described in the first to third embodiments.

As described above, according to this embodiment, it is possible to provide a compact absolute rotary encoder that predicts the amount of eccentricity and the angle of eccentricity and corrects the eccentric component.

In this embodiment, an example of a device equipped with the encoder described in the first to fourth embodiments will be described.

FIG. 17 is a diagram showing a robot arm 1000 equipped with the encoder 1a described in the first embodiment. In the robot arm 1000, an encoder 1a is attached to each axis and used to detect the absolute rotational position of each axis. The scale 10 is attached to a movable member that rotates on each axis. The movable member is rotated by an actuator (not shown). When the movable member rotates to drive the robot arm 1000, the absolute rotational position of each axis of the robot arm 1000 is detected by the encoder 1a, and the position information is output to a CPU (not shown). The CPU drives the actuator based on the position information to move the robot arm 1000 to the target position.

Note that the encoder of each embodiment is not limited to the robot arm 1000, and can be used in various applications such as position detection of a print head or paper feed roller of a printer (optical device), detection of the rotational position of a photosensitive drum of a copier (optical device), and the like. It can be used in various applications of the device.

Although preferred embodiments of the present invention have been described above, the present invention is not limited to these embodiments, and various modifications and changes are possible within the scope of the gist.

1a, 1b encoder (absolute rotary encoder)
10,100 scale 11,111 track (first track)
12,112 tracks (2nd track)
20, 200 sensors 30, 300 processing unit

Claims (11)

a scale comprising a first track and a second track provided with mutually different diameters;
a sensor movable relative to the scale and reading the first track and the second track;
a processing unit that acquires an absolute position signal indicating an absolute position of one of the scale and the sensor;
The processing unit is
A position signal is acquired using a first periodic signal based on a signal obtained by reading the first track by the sensor and a second periodic signal based on a signal obtained by reading the second track by the sensor. ,
Acquiring an eccentricity amount and an eccentric angle using a third periodic signal generated when acquiring the position signal,
obtaining a first eccentricity correction signal corresponding to the first track and a second eccentricity correction signal corresponding to the second track using the eccentricity amount and the eccentricity angle;
The absolute An absolute rotary encoder characterized by acquiring a position signal. the first track includes a first periodic pattern and a second periodic pattern having different periods;
the second track includes a third periodic pattern having a different period from the first and second periodic patterns, and a fourth periodic pattern having a different period from the first to third periodic patterns;
The processing unit obtains the first periodic signal by processing signals corresponding to the first and second periodic patterns, and processes signals corresponding to the third and fourth periodic patterns. 2. The absolute rotary encoder according to claim 1, wherein the second periodic signal is obtained by: 3. The absolute rotary encoder according to claim 1, wherein the sensor includes a first detection section and a second detection section having different detection cycles. 4. The absolute rotary encoder according to any one of claims 1 to 3, wherein the sensor can switch the detection cycle. 5. The absolute rotary encoder according to claim 1, wherein the processing unit acquires the eccentricity using a difference between the maximum value and the minimum value of the third periodic signal. The processing unit sets the amount of eccentricity to a first amount of eccentricity when the difference between the maximum value and the minimum value of the third periodic signal is greater than a threshold, and sets the amount of eccentricity when the difference is less than the threshold. is set as the second eccentricity amount. The absolute rotary encoder according to any one of claims 1 to 6, wherein the processing unit acquires the position signal at which the value of the third periodic signal is maximum as the eccentric angle. 7. The absolute rotary encoder according to any one of claims 1 to 6, wherein the processing unit acquires the position signal at which the third periodic signal has a minimum value as the eccentric angle. The absolute rotary encoder according to any one of claims 1 to 8, wherein the processing unit performs moving average processing on the third periodic signal when acquiring the eccentricity amount and the eccentric angle. . 10. The scale according to any one of claims 1 to 9, wherein when acquiring the eccentricity amount and the eccentric angle, the scale rotates until at least one of a maximum value and a minimum value of the third periodic signal is acquired. 1. The absolute rotary encoder according to claim 1. a rotating movable part;
11. A device, comprising: the absolute rotary encoder according to any one of claims 1 to 10, capable of detecting the rotational position of said movable part.

Images