JP2018048874A - Rotary encoder and method of detecting absolute angle position of rotary encoder - Google Patents

Abstract

PROBLEM TO BE SOLVED: To propose a method of detecting the absolute angle position of a rotary encoder with which it is possible to suppress detection degradation caused by a relative positional deviation between a first sensor unit and a second sensor unit.SOLUTION: As a preprocess of absolute angle position detection operation, first absolute angle data abs-1 is acquired by a first sensor unit 1a of a rotary encoder 1, incremental angle data INC is acquired by a second sensor unit 1b, and a phase difference Δp of these is acquired. Next, calculation is made of phase-corrected first absolute angle data abs-1p in which the phase of the first absolute angle data is corrected by the phase difference Δp. Then, a correction value Δq is obtained on the basis of a difference between incremental signal conversion absolute angle data INC-abs in which the incremental angle data is converted into absolute angle data of one rotation and the phase-corrected first absolute angle abs-1p, and error-corrected first absolute angle data abs-1c in which the phase-corrected first absolute angle data abs-1p is corrected by the correction value Δq is obtained.SELECTED DRAWING: Figure 13

Description

  The present invention relates to a rotary encoder that detects the absolute angular position of a rotating body at the moment, and an absolute angular position detection method of the rotary encoder.

  A rotary encoder that detects the rotation of the rotating body with respect to the fixed body is described in Patent Document 1. The rotary encoder of the same document includes a first sensor unit and a second sensor unit, and based on the detection result of the first sensor unit and the detection result of the second sensor unit, the moment of rotation of the rotating body is determined. Detect absolute angular position. The first sensor unit includes a first magnet having one N pole and one S pole, a first magnetoresistive element facing the first magnet, a first Hall element facing the first magnet, And a second Hall element disposed opposite to the first magnet and disposed at a position shifted by 90 ° in mechanical angle around the rotation center axis with respect to the first Hall element. The second sensor unit includes a second magnet having a plurality of pairs of poles disposed around the rotation center axis, and a second magnetoresistive element facing the second magnet. In the rotary encoder, the instantaneous angular position of the rotating body is determined based on absolute angle data for one rotation and one cycle in the first sensor unit and incremental angle data for one rotation and N cycles in the second sensor unit.

Japanese Patent No. 5666886

  As in Patent Document 1, in a rotary encoder including a first sensor unit and a second sensor unit, a relative positional shift may occur between the first sensor unit and the second sensor unit. Further, when a positional deviation of each sensor unit occurs, a phase deviation or an error occurs between the absolute angle data in the first sensor unit and the incremental angle data in the second sensor unit, so that the detection accuracy decreases. A problem occurs.

  In view of such problems, the problem of the present invention is that even when the absolute angular position of the rotating body is detected based on the detection result of the first sensor unit and the detection result of the second sensor unit, An object of the present invention is to provide a rotary encoder that can suppress detection degradation caused by a relative positional shift with the second sensor unit, and an absolute angular position detection method of the rotary encoder.

In order to solve the above-described problem, the present invention includes a first sensor unit and a second sensor unit, wherein the first sensor unit acquires first absolute angle data for one rotation and one cycle, and the second sensor unit. In the sensor unit, when N is a positive integer equal to or greater than 2, incremental angle data of one rotation N period is acquired, the first detection result from the first sensor unit, and the second detection from the second sensor unit In a rotary encoder that detects an absolute angle position based on a result, a phase difference acquisition unit that acquires a phase difference between the first absolute angle data and the incremental angle data, and the incremental angle data for N cycles are rotated once. A converted absolute angle data calculating unit for calculating incremental signal converted absolute angle data converted into absolute angle data; and the first absolute angle based on the phase difference. A first phase correction unit that corrects data and generates phase corrected first absolute angle data in which the phase of the first absolute angle data is matched with the phase of the incremental signal converted absolute angle data; and the incremental signal converted absolute angle Correction value acquisition unit for acquiring a correction value based on a difference between the data and the phase correction first absolute angle data; and error correction first absolute angle data obtained by correcting the phase correction first absolute angle data with the correction value. Relative error correction unit to be generated, and absolute angle acquisition for acquiring an absolute angle based on the first detection result, the second detection result, the phase difference, the error correction first absolute angle data, and the incremental angle data And a section.

  In the present invention, based on the first detection result from the first sensor unit (first absolute angle data in one rotation and one cycle) and the second detection result from the second sensor unit (incremental angle data in one rotation and N cycles). The absolute angular position at the moment of the rotating body is detected. Accordingly, it is possible to detect the absolute angular position at the moment of the rotating body with high resolution. In the present invention, the phase difference between the first absolute angle data and the incremental angle data is acquired, and the phase corrected first absolute angle data is calculated by matching the phase of the first absolute angle data with the phase of the incremental angle data. In addition, a correction value is obtained based on the difference between the incremental signal converted absolute angle data obtained by converting the incremental angle data into absolute rotation angle data and the phase correction first absolute angle data, and the phase correction first absolute angle is calculated based on the correction value. Error corrected first absolute angle data obtained by correcting the data is obtained. Here, the error-corrected first absolute angle data is obtained by adding a relative error between the first absolute angle data and the N-cycle incremental angle data to the first absolute angle data. Therefore, the error correction first absolute angle data and the incremental angle data have no or no relative error between them. Therefore, if the absolute angle position is determined based on the error-corrected first absolute angle data and the incremental angle data, it is possible to suppress a decrease in detection due to a relative positional deviation or the like between the first sensor unit and the second sensor unit. .

  In the present invention, in order to acquire error correction first absolute angle data, the correction value acquisition unit acquires the correction value by subtracting the phase correction first absolute angle data from the incremental signal conversion absolute angle data. The relative error correction unit may add the correction value to the phase correction first absolute angle data to obtain the error correction first absolute angle data.

  In the present invention, in order to acquire error correction first absolute angle data, the correction value acquisition unit subtracts the incremental signal conversion absolute angle data from the phase correction first absolute angle data, thereby correcting the correction value. The relative error correction unit may subtract the correction value from the phase correction first absolute angle data to obtain the error correction first absolute angle data.

  In the present invention, the image processing apparatus further includes a storage unit, and the correction value acquisition unit acquires the correction values at a plurality of angular positions in one rotation and one cycle, and obtains each angular position and the correction value acquired at the angular position. The relative error correction unit is stored in the storage unit in association with each other, and the relative error correction unit is an intermediate intermediate angular position between the two angular positions based on the correction value acquired at each of the two adjacent angular positions. It is desirable to calculate a correction value and correct the phase correction first absolute angle data using the correction value and the intermediate correction value stored and held in the storage unit as the correction value. In this way, the phase correction first absolute angle data can be corrected by the correction value stored and held in the storage unit and the intermediate correction value complemented between the correction values stored and held in the storage unit. An error between the angle data and the incremental angle data can be further suppressed. In addition, since the intermediate correction value can be obtained by calculation, the capacity for storing and holding the correction value can be suppressed.

In the present invention, the absolute angle acquisition unit includes a second absolute angle data generation unit that generates second absolute angle data obtained by interpolating the error-corrected first absolute angle data into N pieces, and the second absolute angle A phase comparison unit that compares the phase of the data and the phase of the incremental angle data, and the phase of the second absolute angle data and the phase of the incremental angle data are shifted in the comparison result of the phase comparison unit A phase correction unit for correcting the second absolute angle data; a phase correction first detection result obtained by correcting the first detection result by the phase difference; the second detection result; the second absolute angle data; and the incremental. It is desirable to include an angular position determination unit that determines an absolute angular position of the rotating body based on the angle data. In this way, even in the error-corrected first absolute angle data, when there is an error such as a phase shift with the incremental angle data, these errors can be suppressed by correction. Moreover, since the determination of the absolute angle position is performed using the phase difference phase correction first detection result obtained by correcting the first inspection result from the first sensor unit with the phase difference, the first sensor unit and the second sensor unit It is possible to further suppress a decrease in detection accuracy due to a relative positional deviation between the two.

  In the present invention, the first sensor unit includes a first magnetoresistor that is opposed to the first magnet in which one N pole and one S pole are arranged around the rotation center axis in the rotation center axis direction of the first magnet. An element, a first Hall element facing the first magnet, and a position opposed to the first magnet and shifted by 90 ° in mechanical angle around the rotation center axis with respect to the first Hall element A second Hall element, and the second sensor unit includes a second magnet having a plurality of pole pairs arranged around the rotation center axis, and a second magnetoresistive element facing the second magnet. Can be.

  Next, the present invention includes a first sensor unit and a second sensor unit, wherein the first sensor unit acquires first absolute angle data of one rotation and one cycle, and the second sensor unit includes N Incremental angle data of one rotation N period is acquired when the value is a positive integer of 2 or more, and based on the first detection result from the first sensor unit and the second detection result from the second sensor unit. In the absolute angle position detecting method of the rotary encoder for detecting the absolute angle position, a phase difference obtaining step for obtaining a phase difference between the first absolute angle data and the incremental angle data, and the incremental angle data for N cycles are obtained. A converted absolute angle data calculating step of calculating incremental signal converted absolute angle data converted into absolute angle data of one rotation, and a previous step based on the phase difference A first phase correction step for generating phase corrected first absolute angle data in which the first absolute angle data is corrected to make the phase of the first absolute angle data coincide with the phase of the incremental signal conversion absolute angle data; and the incremental A correction value acquisition step of acquiring a correction value based on a difference between the signal conversion absolute angle data and the phase correction first absolute angle data; and error correction first in which the phase correction first absolute angle data is corrected with the correction value. An absolute angle is acquired based on a relative error correction step for generating absolute angle data, and the first detection result, the second detection result, the phase difference, the error correction first absolute angle data, and the incremental angle data. And an absolute angle acquisition step.

  In the present invention, based on the first detection result from the first sensor unit (first absolute angle data in one rotation and one cycle) and the second detection result from the second sensor unit (incremental angle data in one rotation and N cycles). The absolute angular position at the moment of the rotating body is detected. Accordingly, it is possible to detect the absolute angular position at the moment of the rotating body with high resolution. In the present invention, the phase difference between the first absolute angle data and the incremental angle data is acquired, and the phase corrected first absolute angle data is calculated by matching the phase of the first absolute angle data with the phase of the incremental angle data. In addition, a correction value is obtained based on the difference between the incremental signal converted absolute angle data obtained by converting the incremental angle data into absolute rotation angle data and the phase correction first absolute angle data, and the phase correction first absolute angle is calculated based on the correction value. Error corrected first absolute angle data obtained by correcting the data is obtained. Here, the error-corrected first absolute angle data is obtained by adding a relative error between the first absolute angle data and the N-cycle incremental angle data to the first absolute angle data. Accordingly, the error correction first absolute angle data and the incremental angle data have no or no relative error between them. Therefore, if the absolute angle position is determined based on the error-corrected first absolute angle data and the incremental angle data, it is possible to suppress a decrease in detection due to a relative positional deviation or the like between the first sensor unit and the second sensor unit. .

  In the present invention, in order to acquire error correction first absolute angle data, in the correction value acquisition step, the phase correction first absolute angle data is subtracted from the incremental signal conversion absolute angle data to acquire the correction value. In the relative error correction step, the error correction first absolute angle data can be obtained by adding the correction value to the phase correction first absolute angle data.

  In the present invention, in order to obtain error correction first absolute angle data, in the correction value acquisition step, the incremental signal conversion absolute angle data is subtracted from the phase correction first absolute angle data to obtain the correction value. In the relative error correction step, the error correction first absolute angle data can be obtained by subtracting the correction value from the phase correction first absolute angle data.

  In the present invention, the correction value acquisition step acquires the correction values at a plurality of angular positions in one rotation and one cycle, stores the angular positions and the correction values acquired at the angular positions in association with each other and stores them in the storage unit. The relative error correction step calculates an intermediate correction value of an intermediate angular position between the two angular positions based on the correction value acquired at each of the two adjacent angular positions, and It is desirable that the phase correction first absolute angle data is corrected using the correction value and the intermediate correction value stored and held in the storage unit as a correction value. In this way, the phase correction first absolute angle data can be corrected by the correction value stored and held in the storage unit and the intermediate correction value complemented between the correction values stored and held in the storage unit. An error between the angle data and the incremental angle data can be further suppressed. Further, since the intermediate correction value can be obtained by calculation, the capacity for storing and holding the correction value can be suppressed.

  In the present invention, the absolute angle acquisition step includes a second absolute angle data generation step of generating second absolute angle data obtained by interpolating the error-corrected first absolute angle data into N pieces, and the second absolute angle A phase comparison step of comparing the phase of the data with the phase of the incremental angle data, and the phase of the second absolute angle data and the phase of the incremental angle data are shifted in the comparison result of the phase comparison step A second phase correction step of correcting the second absolute angle data, a phase correction first detection result obtained by correcting the first detection result by the phase difference, the second detection result, the second absolute angle data, and And an angular position determination step of determining an absolute angular position of the rotating body based on the incremental angle data. In this way, even in the error-corrected first absolute angle data, when there is an error such as a phase shift with the incremental angle data, these errors can be suppressed by correction. Moreover, since the determination of the absolute angle position is performed using the phase difference phase correction first detection result obtained by correcting the first inspection result from the first sensor unit with the phase difference, the first sensor unit and the second sensor unit It is possible to further suppress a decrease in detection accuracy due to a relative positional deviation between the two.

  In the present invention, the first sensor unit includes a first magnetoresistor that is opposed to the first magnet in which one N pole and one S pole are arranged around the rotation center axis in the rotation center axis direction of the first magnet. An element, a first Hall element facing the first magnet, and a position opposed to the first magnet and shifted by 90 ° in mechanical angle around the rotation center axis with respect to the first Hall element A second Hall element, and the second sensor unit includes a second magnet having a plurality of pole pairs arranged around the rotation center axis, and a second magnetoresistive element facing the second magnet. It is desirable.

In the present invention, error corrected first absolute angle data obtained by adding a relative error between the first absolute angle data and the incremental angle data from the second sensor unit to the first absolute angle data from the first sensor unit is acquired. Then, the absolute angle position is detected based on the error correction first absolute angle data and the incremental angle data. Here, there is no relative error between the first absolute angle data and the incremental angle data for N cycles or the relative error between the error-corrected first absolute angle data and the incremental angle data. Is suppressed. Therefore, if the absolute angle position is acquired based on the error-corrected first absolute angle data and the incremental angle data, it is possible to suppress a decrease in detection due to a relative positional deviation or the like.

It is explanatory drawing which shows the external appearance etc. of the rotary encoder to which this invention is applied. It is a side view which notches and shows a part of fixing body of the rotary encoder of FIG. It is explanatory drawing which shows the structure of the sensor part etc. of a rotary encoder. It is explanatory drawing which shows the detection principle of the angular position of a rotary encoder. It is explanatory drawing which shows the basic composition of the determination method of the angular position of a rotary encoder. It is explanatory drawing of the phase difference of 1st absolute angle data and incremental angle data, and phase correction 1st absolute angle data. It is explanatory drawing of conversion absolute angle data. It is explanatory drawing of a correction value. It is explanatory drawing of error correction 1st absolute angle data. It is explanatory drawing which shows the specific structure of the determination method of the angular position of a rotary encoder. It is explanatory drawing in case the phase of 2nd absolute angle data has advanced. It is explanatory drawing when the phase of 2nd absolute angle data is behind. It is a flowchart of the absolute angle position detection operation | movement which detects an absolute angle position.

  Embodiments of a rotary encoder to which the present invention is applied will be described below with reference to the drawings. In the following description, a magnetic rotary encoder in which a sensor unit is configured by a magnet and a magnetosensitive element (magnetoresistance element, Hall element) will be mainly described as a rotary encoder. In this case, either a configuration in which a magnet is provided on the fixed body and a magnetic sensing element is provided on the rotating body, or a configuration in which a magnetic sensing element is provided on the fixed body and a magnet is provided on the rotating body may be adopted. In the following description, the description will focus on a configuration in which a magnetosensitive element is provided on a fixed body and a magnet is provided on a rotating body. In the drawings referred to below, the configuration of the magnet, the magnetic sensing element, and the like are schematically shown, and the number of magnetic poles in the second magnet is schematically shown by reducing the number thereof. In addition, the configuration of the magnetoresistive pattern in the magnetoresistive element (magnetic sensitive element) is also schematically shown with the positions shifted from each other.

(overall structure)
FIG. 1 is an explanatory view showing the appearance and the like of a rotary encoder to which the present invention is applied. 1A is a perspective view when the rotary encoder is viewed from one side in the rotational axis direction and from an oblique direction, and FIG. 1B is a plan view when viewed from one side in the rotational axis direction. . FIG. 2 is a side view of the rotary encoder to which the present invention is applied with a part cut away.

  The rotary encoder 1 is a device that magnetically detects rotation around the rotation axis of the rotating body 2 with respect to the fixed body 10. The fixed body 10 is fixed to a frame or the like of the motor device, and the rotating body 2 is used in a state of being connected to a rotation output shaft or the like of the motor device. As shown in FIGS. 1 and 2, the fixed body 10 includes a sensor substrate 15 and a plurality of support members 11 that support the sensor substrate 15. In this example, the support member 11 includes a base body 12 including a bottom plate portion 121 in which a circular opening 122 is formed, and a sensor support plate 13 fixed to the base body 12.

The sensor support plate 13 is fixed to the substantially cylindrical body 123 protruding from the edge portion of the opening 122 in the base body 12 in the first direction L1 in the rotation center axis direction L by screws 191 and 192. . A plurality of terminals 16 protrude from the sensor support plate 13 in the first direction L1 in the rotation center axial direction L. A protrusion 124, a hole 125, and the like are formed on an end surface of the body portion 123 that is positioned in the first direction L1 of the rotation center axis direction L. The sensor substrate 15 is formed in the body portion 123 using the hole 125 and the like. Is fixed by a screw 193 or the like. At that time, the sensor substrate 15 is accurately fixed in a state where the sensor substrate 15 is positioned at a predetermined position by the protrusions 124 and the like. On the sensor substrate 15, a connector 17 is provided on the surface in the first direction L <b> 1 in the rotation center axis direction L. The rotating body 2 is a cylindrical member disposed inside the trunk portion 123, and a rotation output shaft (not shown) of the motor is connected to the inside by a method such as fitting. Accordingly, the rotating body 2 can rotate around the rotation axis.

(Layout of magnets and magneto-sensitive elements)
FIG. 3 is an explanatory diagram showing the configuration of the sensor unit and the like of the rotary encoder 1 to which the present invention is applied. In FIG. 3, the data processing unit 90 includes a CPU that operates based on a program stored in advance. The configuration of the data processing unit 90 is shown in a functional block diagram.

  The first sensor unit 1a includes a first magnet 20 on the rotating body 2 side. The first magnet 20 includes a magnetized surface 21 in which an N pole and an S pole are magnetized one by one in the circumferential direction. The magnetized surface 21 faces the first direction L1 of the rotation center axis direction L. The first sensor unit 1a includes a first magnetoresistive element 40 facing the magnetized surface 21 of the first magnet 20 in the first direction L1 in the rotation center axis direction L on the fixed body 10 side, A first Hall element 51 that faces the magnetized surface 21 of one magnet 20 in a first direction L1 in the rotation center axis direction L, and a first Hall element 51 in the rotation center axis direction L with respect to the magnetized surface 21 of the first magnet 20. And a second Hall element 52 disposed opposite to the first Hall element 51 at a mechanical angle of 90 ° around the rotation axis.

  The 2nd sensor part 1b has the 2nd magnet 30 on the rotating body 2 side. The second magnet includes an annular magnetized surface 31 in which a plurality of N poles and S poles are alternately magnetized in the circumferential direction at positions spaced radially outward from the first magnet 20. The magnetized surface 31 faces the first direction L1 of the rotation center axis direction L. In this example, a plurality of tracks 310 in which N poles and S poles are alternately magnetized in the circumferential direction are arranged in parallel on the magnetized surface 31 of the second magnet 30 in the radial direction. In this example, two tracks 310 are formed. In this example, when N is a positive integer, the second magnet 30 has a total of N pairs of N and S poles. In this example, N is 128, for example. The positions of the N pole and the S pole are shifted in the circumferential direction between the two tracks 310. In this example, the N pole and the S pole are shifted by one pole in the circumferential direction between the two tracks 310. The second sensor unit 1 b includes a second magnetoresistive element 60 that faces the magnetized surface 31 of the second magnet 30 in the first direction L1 in the rotation center axial direction L on the fixed body 10 side.

  The first magnet 20 and the second magnet 30 rotate around the rotation axis integrally with the rotating body 2. The first magnet 20 is made of a disk-shaped permanent magnet. The second magnet 30 has a cylindrical shape and is disposed at a position spaced apart from the first magnet 20 on the outer side in the radial direction. The 1st magnet 20 and the 2nd magnet 30 consist of bond magnets.

The first magnetoresistive element 40 includes an A phase (SIN) magnetoresistive pattern and a B phase (COS) magnetoresistive pattern having a phase difference of 90 ° with respect to the phase of the first magnet 20. 1 magnetoresistive element. In the first magnetoresistive element 40, the A-phase magnetoresistive pattern includes the + a-phase (SIN +) magnetoresistive pattern 43 and the -a-phase (SIN-) magnetoresistive pattern 43 that detect the movement of the rotating body 2 with a phase difference of 180 °. A magnetoresistive pattern 41 is provided. The B phase magnetoresistive pattern includes a + b phase (COS +) magnetoresistive pattern 44 and a −b phase (COS−) magnetoresistive pattern 42 that detect movement of the rotating body 2 with a phase difference of 180 °. Here, the + a-phase magnetoresistive pattern 43 and the -a-phase magnetoresistive pattern 41 constitute a bridge circuit, and the + b-phase magnetoresistive pattern 44 and the -b-phase magnetoresistive pattern 42 are also of + a-phase. Similar to the magnetoresistive pattern 43 and the -a phase magnetoresistive pattern 41, a bridge circuit is configured.

  The second magnetoresistive element 60 includes an A phase (SIN) magnetoresistive pattern and a B phase (COS) magnetoresistive pattern having a phase difference of 90 ° with respect to the phase of the second magnet 30. In the second magnetoresistive element 60, the A-phase magnetoresistive pattern includes the + a-phase (SIN +) magnetoresistive pattern 64 and the -a-phase (SIN-) that detect movement of the rotating body 2 with a phase difference of 180 °. A magnetoresistive pattern 62 is provided. The B phase magnetoresistive pattern includes a + b phase (COS +) magnetoresistive pattern 63 and a −b phase (COS−) magnetoresistive pattern 61 that detect the movement of the rotating body 2 with a phase difference of 180 °. Here, the + a phase magnetoresistive pattern 64 and the −a phase magnetoresistive pattern 62 constitute a bridge circuit, like the first magnetoresistive element 40, and the + b phase magnetoresistive pattern 63 and the −b phase magnetoresistive pattern 63. The magnetoresistive pattern 61 forms a bridge circuit as shown in the same manner as the + a phase magnetoresistive pattern 64 and the −a phase magnetoresistive pattern 62.

  In this example, the first magnetoresistive element 40, the first Hall element 51, the second Hall element 52, and the second magnetoresistive element 60 are all in the second direction L2 of the rotation center axis direction L of the sensor substrate 15. The first surface 151 is provided. On the second surface 152 opposite to the first surface 151 in the sensor substrate 15, a position overlapping the first magnetoresistive element 40 in plan view is through a through hole (not shown) penetrating the sensor substrate 15. A first amplifier 91 electrically connected to the first magnetoresistive element 40 is provided. A second amplifier electrically connected to the second magnetoresistive element 60 through a through hole penetrating the sensor substrate 15 at a position overlapping the second magnetoresistive element 60 in plan view on the second surface 152. 92 is provided. The first Hall element 51 and the second Hall element 52 are electrically connected to the first amplifier 91 through a through hole that penetrates the sensor substrate 15.

(Detection principle)
FIG. 4 is an explanatory diagram showing the detection principle in the rotary encoder 1 to which the present invention is applied. FIG. 4A is an explanatory diagram of signals and the like output from the magnetoresistive element 4, and FIG. 4B shows the signal shown in FIG. 4A and the angular position (electrical angle) of the rotating body 2. It is explanatory drawing which shows a relationship. FIG. 5 is an explanatory diagram showing the basic configuration of the method for determining the angular position in the rotary encoder.

  As shown in FIG. 3, in the rotary encoder 1 of this example, the outputs of the first magnetoresistive element 40, the first Hall element 51, the second Hall element 52, and the second magnetoresistive element 60 are the first amplifier 91, The data is output to the data processing unit 90 via the second amplifier 92 and AD converters 93a, 93b, and 94. Based on the outputs from the first magnetoresistive element 40, the first hall element 51, the second hall element 52, and the second magnetoresistive element 60, the data processing unit 90 calculates the absolute angular position of the rotating body 2 with respect to the fixed body 10. Ask for.

More specifically, in the rotary encoder 1, when the rotating body 2 makes one revolution, the first magnet 20 makes one revolution. Therefore, from the first magnetoresistive element 40 of the first sensor unit 1a, FIG. The sine wave signals sin and cos shown are output for two cycles. Therefore, if the data processing unit 90 obtains θ = tan ⁻¹ (sin / cos) from the sine wave signals sin and cos as shown in FIG. 4B, the angular position θ of the rotating body 2 can be obtained. Further, in the present example, the first sensor element 1 a is provided with the first hall element 51 and the second hall element 52 at a position shifted by 90 ° from the center of the first magnet 20. For this reason, since it can be known in which section of the sine wave signals sin and cos the current position is located, the absolute angular position of the rotating body 2 can be known.

In the rotary encoder 1 of the present example, the second magnet 30 having the annular magnetized surface 31 in which a plurality of N poles and S poles are alternately magnetized in the circumferential direction is used for the second sensor portion 1b. From the second magnetoresistive element 60 facing the second magnet 30, the rotating body 2 is moved to the second magnet 30.
The sine wave signals sin and cos are output each time the magnetic poles of one cycle are rotated. Accordingly, for the sine wave signals sin and cos output from the second magnetoresistive element 60, θ = tan ⁻¹ (sin / cos) can be obtained from the sine wave signals sin and cos as shown in FIG. For example, the angular position θ of the rotating body 2 within an angle corresponding to one period of the magnetic poles of the second magnet 30 is known.

  Therefore, in this example, the first absolute angle data abs-1 (see FIG. 5A) of one rotation and one cycle, which is the first detection result from the first sensor unit 1a, and the second detection from the second sensor unit 1b. The instantaneous angular position of the rotator 2 is detected based on the 1-rotation N-cycle incremental angle data INC (see FIG. 5B), which is the 2 detection result. Therefore, even when the resolution of the first absolute angle data abs-1 is low, absolute angle data with high resolution can be obtained.

  In adopting such a detection method, in the rotary encoder 1 of the present example, the relative displacement between the first sensor unit 1a and the second sensor unit 1b, the first sensor unit 1a and the second sensor unit 1b are configured. The first absolute angle data abs-1 (see FIG. 5A) from the first sensor unit 1a due to the influence of the member characteristic error, the sampling time difference between the first sensor unit 1a and the second sensor unit 1b, etc. In some cases, a phase shift may occur between the second sensor unit 1b and the incremental angle data INC (see FIG. 5B) of one rotation N cycle. Further, depending on the displacement of the first sensor unit 1a from the rotation axis, the displacement of the second sensor unit 1b from the rotation axis, the characteristics of the first sensor unit 1a, and the characteristics of the second sensor unit 1b, In some cases, an error may occur with respect to the true value (angular position). Here, if there is a relative error including a phase shift between the first absolute angle data abs-1 and the incremental angle data INC, the detection accuracy is lowered.

  Therefore, in this example, in order to detect the absolute angle position, first, the phase difference Δp between the first absolute angle data abs-1 and the incremental angle data INC is acquired, and the first absolute angle data abs-1 and the incremental angle are acquired. A pre-correction step is performed to suppress the relative error and the phase difference Δp generated between the data INC and the error correction first absolute angle data abs-1c.

  Thereafter, a first detection result that is the first absolute angle data abs-1 output from the first sensor unit 1a, and a second detection that is the instantaneous angle data INC output from the second sensor unit 1b. As a result, an absolute angle acquisition step of acquiring an absolute angle position based on the phase difference Δp acquired in the previous correction step, the error correction first absolute angle data abs-1c, and the incremental angle data INC is performed.

(Control system)
FIG. 6 is an explanatory diagram of the phase difference between the first absolute angle data abs-1 and the incremental angle data INC and the phase-corrected first absolute angle data. FIG. 7 is an explanatory diagram of converted absolute angle data. FIG. 8 is an explanatory diagram of correction values. FIG. 9 is an explanatory diagram of the error correction first absolute angle data abs-1c. FIG. 10 is an explanatory diagram showing a specific configuration of the method for determining the angular position in the rotary encoder. In FIG. 10, reference numerals 1, 2,..., N-1, n, n + 1,... N indicating the period of each position of the second absolute angle data abs-2 with respect to the true angular position are indicated. In addition, reference numerals 1, 2,..., M−1, m, m + 1,... N indicating the position of each period of the incremental angle data INC with respect to the true angle position are attached. FIG. 11 is an explanatory diagram when the phase of the second absolute angle data is advanced in the rotary encoder. FIG. 12 is an explanatory diagram when the phase of the second absolute angle data is delayed in the rotary encoder.

As shown in FIG. 3, the data processing unit 90 includes a pre-correction processing unit 100 that performs a pre-correction step and an absolute angle acquisition unit 101 that performs an absolute angle acquisition step.

(Pre-correction processing unit)
The pre-correction processing unit 100 includes a memory (storage unit) 102, a phase difference acquisition unit 103, a converted absolute angle data calculation unit 104, a first phase correction unit 105, a correction value acquisition unit 106, and a relative error correction unit 107. .

  The phase difference acquisition unit 103 acquires a phase difference Δp between the first absolute angle data abs-1 from the first sensor unit 1a and the incremental angle data INC from the second sensor unit 1b. That is, as shown in FIGS. 6A and 6B, the phase difference acquisition unit 103 receives the first absolute angle data abs-1 of one rotation and one cycle from the first sensor unit 1a and the second sensor unit 1b. Assuming that there is a phase shift with the incremental angle data INC of one rotation N cycle, the point of the angular position 0 ° of the first absolute angle data abs-1 from the first sensor unit 1a, An angle difference from the point at the angular position 0 ° of the incremental angle data INC closest to the point is acquired as a phase difference Δp. The ideal first absolute angle data abs-1 is linear data as shown in FIG. However, the first absolute angle data abs-1 from the first sensor unit 1a is deviated from the ideal data by the included error, as shown in FIG. 6A, for example. The ideal incremental angle data INC is linear data as shown in FIG. However, the incremental angle data INC from the second sensor unit 1b is deviated from the ideal data by the included error, as shown in FIG. 6B, for example.

  As shown in FIG. 7, the converted absolute angle data calculation unit 104 converts the N-cycle incremental angle data INC (see FIGS. 6B and 7A) from the second sensor unit 1 b into one rotation 1 Periodic incremental signal conversion absolute angle data INC-abs (see FIG. 7B). That is, the converted absolute angle data calculation unit 104 sequentially accumulates the incremental angle data INC from the point of the angle position 0 ° of the incremental angle data INC closest to the point of the angle position 0 ° in the first absolute angle data abs−1. Finally, the absolute angle data INC (incremental signal conversion absolute angle data INC-abs) corresponding to the first absolute angle data is calculated by integrating the incremental angle data INC for N cycles.

  The first phase correction unit 105 corrects the first absolute angle data based on the phase difference Δp acquired by the phase difference acquisition unit 103, and converts the phase of the first absolute angle data to the incremental signal conversion absolute angle data INC-abs. Phase corrected first absolute angle data abs-1p matched with the phase is generated. That is, as shown in FIG. 6C, the first phase correction unit 105 generates phase corrected first absolute angle data abs-1p obtained by offsetting the first absolute angle data by the phase difference Δp. Thereby, the angle position 0 ° of the phase correction first absolute angle data abs-1p matches the angle position 0 ° of the incremental signal conversion absolute angle data INC-abs.

  The correction value acquisition unit 106 acquires the correction value Δq based on the difference between the incremental signal conversion absolute angle data INC-abs and the phase correction first absolute angle data abs-1p. In this example, the correction value acquisition unit 106 performs correction by subtracting the phase correction first absolute angle data abs-1p shown in FIG. 8B from the incremental signal conversion absolute angle data INC-abs shown in FIG. The value Δq1 is acquired. The correction value Δq1 shown in FIG. 8C is a relative error component between the incremental signal conversion absolute angle data INC-abs and the phase correction first absolute angle data abs-1p.

Here, the correction value acquisition unit 106 acquires, as the correction value Δq1, the first correction value Δq1 at a plurality of angular positions in one rotation and one cycle. Then, the correction value acquisition unit 106 stores and holds in the memory 102 in the form of a table in which the angular position and the first correction value Δq1 are associated with each other. In this example, the plurality of angular positions at which the correction value Δq1 is acquired are the respective angular positions when one rotation is divided at equal intervals by the number N of the magnetic pole pairs of the second magnet 30. Note that the plurality of angular positions from which the correction value Δq1 is acquired can be, for example, each angular position when one rotation is divided at equal intervals by twice the number of magnetic pole pairs of the second magnet 30.

  The relative error correction unit 107 refers to the memory 102 and calculates an intermediate correction value Δq2 of an intermediate angular position between the two angular positions based on the correction value Δq1 acquired at each of the two adjacent angular positions. To do. In this example, the intermediate angular position is the central angular position of two adjacent angular positions, and the correction value acquisition unit 106 linearly calculates the intermediate correction value Δq2 at the intermediate angular position and the correction value Δq1 of the two adjacent angular positions. Complementary calculation.

  Further, as shown in FIG. 9, the relative error correction unit 107 corrects the phase correction first absolute angle data abs-1p with the correction value Δq1 and the intermediate correction value Δq2 recorded and held in the memory 102. Absolute angle data abs-1c is generated. In this example, the relative error correction unit 107 adds the correction value Δq1 and the intermediate correction value Δq2 shown in FIG. 9B to the phase correction first absolute angle data abs-1p shown in FIG. Thereby, the relative error correction unit 107 acquires error correction first absolute angle data abs-1c shown in FIG.

  Here, since the correction value Δq1 is an error component of the incremental signal conversion absolute angle data INC-abs (incremental angle data INC for N cycles) with respect to the phase correction first absolute angle data abs-1p, the phase correction first absolute angle data. Error corrected first absolute angle data abs-1c obtained by correcting abs-1p with the correction value Δq1 and the intermediate correction value Δq2 is substantially the same error component as the incremental signal conversion absolute angle data INC-abs (see FIG. 7B). It will be equipped with. In other words, the error correction first absolute angle data abs-1c includes almost the same error component as the N-cycle incremental angle data INC (see FIG. 7A). Therefore, the relative error is eliminated or suppressed between the error correction first absolute angle data abs-1c and the incremental angle data INC. Therefore, if the absolute angle position is acquired based on the error-corrected first absolute angle data abs-1c and the incremental angle data INC, the relative positional deviation between the first sensor unit 1a and the second sensor unit 1b, etc. It is possible to suppress a decrease in detection due to.

  The correction value acquisition unit 106 may acquire the correction value by subtracting the incremental signal conversion absolute angle data INC-abs from the phase correction first absolute angle data abs-1p. In this case, the relative error correction unit 107 subtracts the correction value from the phase correction first absolute angle data abs-1p to obtain error correction first absolute angle data abs-1c.

(Absolute angle acquisition unit)
Next, as shown in FIG. 3, the absolute angle acquisition unit 101 includes a second absolute angle data generation unit 110, a first memory 111, a second memory 112, and an angular position determination unit 113. The absolute angle acquisition unit 101 includes a phase comparison unit 114 and a second phase correction unit 115.

  As shown in FIG. 10A, the second absolute angle data generation unit 110 uses the error correction first absolute angle data abs-1c obtained by the preprocessing step as the number of magnetic pole pairs (N : 2nd absolute angle data abs-2 which is divided by interpolation into a positive integer of 2 or more. Then, the second absolute angle data generation unit 110 stores and holds the second absolute angle data abs-2 in the first memory 111. Here, the second memory 112 stores and holds incremental angle data INC.

The angular position determination unit 113 outputs the first phase difference Δp at the moment of output from the first sensor unit 1a.
First detection result that is absolute angle data, second detection result that is instantaneous angle data from the second sensor unit, second absolute angle data abs-2 stored in the first memory 111, and second memory Based on the incremental angle data INC stored in 112, the absolute angular position of the rotating body 2 at the moment is determined.

  More specifically, when the angular position determination unit 113 acquires the first detection result (first absolute angle data abs-1 at the moment) from the first sensor unit 1a, the first detection result is expressed by the phase difference Δp. A corrected phase correction first detection result (instantaneous phase correction first absolute angle data abs-1p) is calculated. Then, the angle position determination unit 113 uses the period of the second absolute angle data abs-2 stored and held in the first memory 111 as the higher-order data of the digital data, as the first phase correction detection result. Which position of the incremental angle data INC stored and held in the second memory 112 the second detection result from the second sensor unit 1b (instantaneous incremental angle data INC) corresponds to the lower data of the digital data, The absolute angular position of the rotating body 2 at the moment is determined.

  Here, the phase comparison unit 114 compares the phase of the second absolute angle data abs-2 with the phase of the incremental angle data INC at a predetermined timing. The second phase correction unit 115 also determines that the phase of the second absolute angle data abs-2 and the incremental angle data INC are shifted from each other by the phase comparison unit 114. And the phase of the second absolute angle data abs-2 and the phase of the incremental angle data INC are matched, and the corrected second absolute angle data abs-2 is stored and retained (overwritten) in the first memory 111. The predetermined timing is, for example, the time when the rotary encoder 1 is turned on.

  More specifically, the phase comparison unit 114 includes a third absolute angle data generation unit 116, a first determination unit 117, a second determination unit 118, and a third memory 119, as shown in FIG.

  As shown in FIGS. 11B and 12B, the third absolute angle data generation unit 116 is data in which the error-corrected first absolute angle data abs-1c is interpolated into (2 × N) pieces. The third absolute angle data abs-3 corresponding to is generated and stored in the third memory 119.

  The first determination unit 117 determines whether the phase of the second absolute angle data abs-2 has advanced with respect to the incremental angle data INC based on the third absolute angle data abs-3. Specifically, as illustrated in FIGS. 11A and 11B, the first determination unit 117 determines that the phase detection first detection result is an odd number (for example, i-th) in the third absolute angle data abs-3. And the phase of the second absolute angle data abs-2 advances from the phase of the incremental angle data INC when the second detection result by the second sensor unit 1b is equal to or greater than the first threshold value TH1 in the incremental angle data INC. It is determined that That is, when the phase of the second absolute angle data abs-2 and the phase of the incremental angle data INC match, the phase correction first detection result is an odd-numbered cycle in the third absolute angle data abs-3. In some cases, the second detection result by the second sensor unit 1b is less than the first threshold value TH1 in the incremental angle data INC. Therefore, according to the above processing, the phase of the second absolute angle data abs-2 is the incremental angle data. It is possible to detect that the phase is ahead of the INC phase. In the present embodiment, the first threshold value TH1 is 270 degrees in electrical angle.

The second determination unit 118 determines whether there is a phase delay of the second absolute angle data abs-2 with respect to the incremental angle data INC based on the third absolute angle data abs-3. Specifically, as illustrated in FIGS. 12A and 12B, the second determination unit 118 corrects the first detection result obtained by the first sensor unit 1 a with the phase difference Δp (first phase correction detection result ( The instantaneous phase correction first absolute angle data abs-1p) is an even-numbered (for example, (i + 1) th) period in the third absolute angle data abs-3, and the second detection result by the second sensor unit 1b. Is equal to or smaller than the second threshold value TH2 in the incremental angle data INC, it is determined that the phase of the second absolute angle data abs-2 is delayed from the phase of the incremental angle data INC. That is, when the phase of the second absolute angle data abs-2 and the phase of the incremental angle data INC match, the phase correction first detection result is the even-numbered cycle in the third absolute angle data abs-3. Since the second detection result by the second sensor unit 1b exceeds the second threshold value TH2 in the incremental angle data INC, according to the above processing, the phase of the second absolute angle data abs-2 is the same as that of the incremental angle data INC. It can be detected that the phase is delayed. In the present embodiment, the second threshold TH2 is 90 degrees in electrical angle.

  In the second phase correction unit 115, the first phase correction detection result is the i-th (odd number) period in the third absolute angle data abs-3, and the second detection result by the second sensor unit 1b is the incremental angle data. For a period that is equal to or greater than the first threshold value TH1 in INC, as shown in FIG. 11C, the (((i + 1) / 2) -1) th cycle (n-1) in the second absolute angle data abs-2. The second absolute angle data abs-2 is corrected so that the second cycle). Accordingly, the phase of the incremental angle data INC and the corrected second absolute angle data abs-2 are the same. Then, the second phase correction unit 115 stores and holds (overwrites) the corrected second absolute angle data abs-2 in the first memory 111.

  In addition, the second phase correction unit 115 has the (i + 1) -th (even-numbered) period in the third absolute angle data abs-3, and the second detection result by the second sensor unit 1b. Is a second threshold TH2 or less in the incremental angle data INC, as shown in FIG. 12C, the (((i + 1) / 2) +1) -th cycle in the second absolute angle data abs-2 ( The second absolute angle data abs-2 is corrected so as to be (n + 1) th cycle). Accordingly, the phase of the incremental angle data INC and the corrected second absolute angle data abs-2 are the same. Then, the second phase correction unit 115 stores and holds (overwrites) the corrected second absolute angle data abs-2 in the first memory 111.

  The angle position determination unit 113 refers to the second absolute angle data abs-2 stored and held in the first memory 111 when determining the absolute angle position. Accordingly, when the second absolute angle data abs-2 is corrected by the second phase correction unit 115 and the corrected second absolute angle data abs-2c is stored and held in the first memory 111, the angular position is determined thereafter. The unit 113 determines the absolute angle position based on the corrected second absolute angle data abs-2c.

(Absolute angular position acquisition operation)
Next, the absolute angular position acquisition operation will be described with reference to FIG. The absolute angle position acquisition operation includes a pre-correction step (step ST1) and an absolute angle position acquisition step (step ST2).

  In the pre-correction step (step ST1), the rotating body 2 is rotated to acquire the first absolute angle data abs-1 for one rotation and one cycle based on the output from the first sensor unit 1a, and the second sensor unit Based on the output from 1b, incremental angle data INC for one rotation N period is acquired (step ST11).

Next, the phase difference acquisition unit 103 acquires the phase difference Δp between the first absolute angle data abs-1 and the incremental angle data INC (step ST12: phase difference acquisition step). When the phase difference Δp is acquired, the converted absolute angle data calculating unit 104 calculates incremental signal converted absolute angle data INC-abs obtained by converting incremental angle data INC for N cycles into absolute angle data of one rotation, and The first phase correction unit 105 corrects the first absolute angle data abs-1 based on the phase difference Δp to match the phases of the first absolute angle data abs-1 and the incremental signal conversion absolute angle data INC-abs. Phase correction first absolute angle data abs-1p is generated (step ST13: conversion absolute angle data calculation step, first phase correction step).

  Thereafter, the correction value acquisition unit 106 acquires the correction value Δq1 based on the difference between the incremental signal conversion absolute angle data INC-abs and the phase correction first absolute angle data abs-1p. Further, the correction value acquisition unit 106 stores and holds the correction value Δq1 in the memory 102 in the form of a table associated with the angular position. (Step ST14: Correction value acquisition step).

  When the correction value Δq1 is acquired, the relative error correction unit 107 generates error correction first absolute angle data abs-1c obtained by correcting the phase correction first absolute angle data abs-1p with the correction value Δq1 and the intermediate correction value Δq2. (Step ST15: Relative error correction step).

  Here, since the correction value Δq1 is an error component of the N angle incremental angle data INC with respect to the phase correction first absolute angle data abs-1p, the phase correction first absolute angle data abs-1p is used as the correction value Δq1 and the intermediate correction value. The error-corrected first absolute angle data abs-1c corrected by Δq2 has substantially the same error component as the incremental signal conversion absolute angle data INC-abs (N-cycle incremental angle data INC). Thereby, the relative error between the error correction first absolute angle data abs-1c and the incremental angle data INC is eliminated or suppressed. Therefore, in the next absolute angle acquisition step, if the absolute angle position is determined based on the error-corrected first absolute angle data and the incremental angle data, the relative positional deviation between the first sensor unit and the second sensor unit, etc. It is possible to suppress a decrease in detection due to. Note that the pre-correction step is not for correcting the outputs from the two sensor units 1a and 1b with respect to the true absolute angle, but for correcting the outputs from the two sensor units 1a and 1b. It can be said.

(Absolute angle acquisition process)
In the absolute angle acquisition step (step ST2), the second absolute angle data generation unit 110 uses the error-corrected first absolute angle data abs-1c as the number of magnetic pole pairs of the second magnet 30 (N: a positive integer greater than or equal to 2). The second absolute angle data abs-2 interpolated into () is created, and the second absolute angle data abs-2 is stored and held in the first memory 111 (step ST21: second absolute angle data generation step). Thereafter, the rotating body 2 is rotated, the first detection result from the first sensor unit 1a (the first absolute angle data at the moment output from the first sensor unit 1a) and the second detection from the second sensor unit 1b. A detection result (incremental angle data at the moment output from the second sensor unit 1b) is obtained. Further, the angle position determination unit 113 corrects the first detection result (first absolute angle data at the moment output from the first sensor unit 1a) with the phase difference Δp (phase correction at the moment). Corrected first absolute angle data) is calculated.

  Here, when it is not determined by the first determination unit 117 and the second determination unit 118 that the phase of the second absolute angle data abs-2 and the phase of the incremental angle data INC are shifted (second absolute angle data If the phase of abs-2 and the phase of the incremental angle data INC match (step 22: phase comparison step, Yes), the angular position determination unit 113 indicates that the first phase correction detection result is the first. Which period of the second absolute angle data abs-2 stored and held in the memory 111 is used as the higher order data of the digital data, and which of the incremental angle data INC stored in the first memory 111 is the second detection result. The absolute angular position of the rotating body 2 at the moment is determined by using the low-order data of the digital data as a position corresponding to the position (step ST). 3: angular position determining step).

On the other hand, when the first determination unit 117 determines that the phase of the second absolute angle data abs-2 is ahead of the phase of the incremental angle data INC, or the second determination unit 118 determines the second absolute angle data abs. -2 is delayed from the phase of the incremental angle data INC (step 22: phase comparison step, No), the second phase correction unit 115 corrects the second absolute angle data abs-2. Thus, the phase of the second absolute angle data abs-2 is made to coincide with the phase of the incremental angle data INC. Then, the corrected second absolute angle data abs-2 is stored and held in the first memory 111 as new second absolute angle data abs-2 (step ST24: second phase correction step).

  After that, the angular position determination unit 113 uses the period of the second absolute angle data abs-2 stored and held in the first memory 111 as the higher order data of the digital data, and the first phase correction detection result is the first data. 2 The absolute angle position of the rotator 2 at the moment is determined using the position of the incremental angle data INC stored and held in the first memory 111 as the lower data of the digital data, which corresponds to the position of the detection result (step ST23: Angular position determination step).

(Modification)
Note that a post-correction step of further correcting the absolute angle position detected through the absolute angle acquisition step may be provided after the absolute angle acquisition step. In this case, the error between the absolute angle position detected through the absolute angle acquisition step and the reference absolute angle position acquired by the reference encoder is acquired in advance and stored in the memory as position error data. The absolute angle position detected through the angle acquisition step is corrected using the position error data to match the reference absolute angle position acquired by the reference encoder. The post-correction unit that performs the post-correction step acquires error between the absolute angle position detected through the memory (storage unit) and the absolute angle acquisition step and the reference absolute angle position acquired by the reference encoder, and position error data As a reference absolute angle position obtained by correcting the absolute angle position detected through the absolute angle acquisition process using the position error data and storing it in the memory (storage part) And an absolute angle position correction unit that matches the above.

(Other embodiments)
In the magnetic rotary encoder of the above embodiment, a magnet and a magnetoresistive element are used as the second detection result of the first sensor unit 1a and the second sensor unit 1b. However, the first sensor unit 1a and the second sensor unit 1b are used. The present invention may be applied when one or both of the second detection results are configured by a resolver.

  Although the rotary encoder of the above embodiment is magnetic, the present invention may be applied to an optical rotary encoder.

DESCRIPTION OF SYMBOLS 1 ... Rotary encoder 1a ... 1st sensor part 1b ... 2nd sensor part 20 ... 1st magnet 30 ... 2nd magnet 40 ... 1st magnetoresistive element 51 ... 1st 1 Hall element 52 ... 2nd Hall element 60 ... 2nd magnetoresistive element 101 ... Absolute angle acquisition part 102 ... Memory (storage part)
103 ... Phase difference acquisition unit 104 ... Conversion absolute angle data calculation unit 105 ... First phase correction unit 106 ... Correction value acquisition unit 107 ... Relative error correction unit 110 ... Second absolute Angle data generation unit 113 ... angular position determination unit 114 ... phase comparison unit 115 ... phase correction unit abs-1 ... first absolute angle data INC ... incremental angle data INC-abs ... Incremental signal conversion absolute angle data L: rotation center axis direction Δp: phase difference Δq1: correction value Δq2: intermediate correction value

Claims (12)

A first sensor unit and a second sensor unit, wherein the first sensor unit acquires first absolute angle data of one rotation and one period, and in the second sensor unit, N is a positive integer of 2 or more A rotary encoder that acquires incremental angle data of one rotation N period and detects an absolute angle position based on a first detection result from the first sensor unit and a second detection result from the second sensor unit In
A phase difference acquisition unit for acquiring a phase difference between the first absolute angle data and the incremental angle data;
A converted absolute angle data calculating unit for calculating incremental signal converted absolute angle data obtained by converting the incremental angle data for N cycles into absolute angle data for one rotation;
The first absolute angle data is corrected based on the phase difference to generate phase corrected first absolute angle data in which the phase of the first absolute angle data is matched with the phase of the incremental signal conversion absolute angle data. A phase correction unit;
A correction value acquisition unit that acquires a correction value based on a difference between the incremental signal conversion absolute angle data and the phase correction first absolute angle data;
A relative error correction unit that generates error correction first absolute angle data obtained by correcting the phase correction first absolute angle data with the correction value;
An absolute angle acquisition unit that acquires an absolute angle based on the first detection result, the second detection result, the phase difference, the error correction first absolute angle data, and the incremental angle data;
A rotary encoder comprising: In claim 1,
The correction value acquisition unit acquires the correction value by subtracting the phase correction first absolute angle data from the incremental signal conversion absolute angle data,
The relative error correction unit adds the correction value to the phase correction first absolute angle data to obtain the error correction first absolute angle data. In claim 1,
The correction value acquisition unit acquires the correction value by subtracting the incremental signal conversion absolute angle data from the phase correction first absolute angle data,
The relative error correction unit subtracts the correction value from the phase correction first absolute angle data to obtain the error correction first absolute angle data. In any one of claims 1 to 3,
Furthermore, a storage unit is provided,
The correction value acquisition unit acquires the correction values at a plurality of angular positions in one rotation and one cycle, associates each angular position with the correction values acquired at the angular position, and stores and holds them in the storage unit.
The relative error correction unit calculates an intermediate correction value of an intermediate angular position between the two angular positions based on the correction value acquired at each of the two adjacent angular positions, and the correction value A rotary encoder that corrects the phase correction first absolute angle data using the correction value and the intermediate correction value that are stored and held in the storage unit. In any one of claims 1 to 4,
The absolute angle acquisition unit
A second absolute angle data generating unit that generates second absolute angle data in which the error-corrected first absolute angle data is interpolated into N pieces;
A phase comparator that compares the phase of the second absolute angle data with the phase of the incremental angle data;
A phase correction unit that corrects the second absolute angle data when the phase of the second absolute angle data is shifted from the phase of the incremental angle data in the comparison result of the phase comparison unit;
The absolute angle position of the rotating body is determined based on the phase detection first detection result obtained by correcting the first detection result with the phase difference, the second detection result, the second absolute angle data, and the incremental angle data. An angular position determination unit to perform,
A rotary encoder comprising: In any one of claims 1 to 5,
The first sensor unit includes a first magnet in which one N pole and one S pole are arranged around a rotation center axis, a first magnetoresistive element facing in the rotation center axis direction of the first magnet, A first Hall element that faces the first magnet, and a second Hall element that faces the first magnet and is positioned at a mechanical angle of 90 ° around the rotation center axis with respect to the first Hall element. And comprising
The second sensor unit includes a second magnet having a plurality of pole pairs arranged around the rotation center axis, and a second magnetoresistive element facing the second magnet. A first sensor unit and a second sensor unit, wherein the first sensor unit acquires first absolute angle data of one rotation and one period, and in the second sensor unit, N is a positive integer of 2 or more Incremental angle data of one rotation N period is acquired, and the absolute angle position is detected based on the first detection result from the first sensor unit and the second detection result from the second sensor unit. In the absolute angular position detection method of the rotary encoder,
A phase difference acquisition step of acquiring a phase difference between the first absolute angle data and the incremental angle data;
A converted absolute angle data calculating step of calculating incremental signal converted absolute angle data obtained by converting the incremental angle data of N cycles into absolute angle data of one rotation;
The first absolute angle data is corrected based on the phase difference to generate phase corrected first absolute angle data in which the phase of the first absolute angle data is matched with the phase of the incremental signal conversion absolute angle data. A phase correction process;
A correction value acquisition step of acquiring a correction value based on a difference between the incremental signal conversion absolute angle data and the phase correction first absolute angle data;
A relative error correction step of generating error correction first absolute angle data obtained by correcting the phase correction first absolute angle data with the correction value;
An absolute angle acquisition step of acquiring an absolute angle based on the first detection result, the second detection result, the phase difference, the error correction first absolute angle data, and the incremental angle data;
An absolute angular position detection method for a rotary encoder, comprising: In claim 7,
In the correction value acquisition step, the correction value is acquired by subtracting the phase correction first absolute angle data from the incremental signal conversion absolute angle data,
In the relative error correction step, the absolute value of the rotary encoder is detected by adding the correction value to the phase correction first absolute angle data to obtain the error correction first absolute angle data. In claim 7,
In the correction value acquiring step, the correction value is acquired by subtracting the incremental signal conversion absolute angle data from the phase correction first absolute angle data;
In the relative error correction step, an absolute angle position detection method for a rotary encoder, wherein the correction value is subtracted from the phase correction first absolute angle data to obtain the error correction first absolute angle data. In any one of claims 7 to 9,
The correction value acquisition step acquires the correction values at a plurality of angular positions in one rotation and one cycle, associates each angular position with the correction values acquired at the angular position, and stores and holds them in the storage unit;
The relative error correction step calculates an intermediate correction value of an intermediate angular position between the two angular positions based on the correction value acquired at each of the two adjacent angular positions, and the correction value A method of detecting an absolute angular position of a rotary encoder, wherein the phase correction first absolute angle data is corrected using the correction value and the intermediate correction value stored and held in the storage unit. In any one of claims 7 to 10,
The absolute angle acquisition step includes
A second absolute angle data generation step of generating second absolute angle data in which the error correction first absolute angle data is interpolated into N pieces;
A phase comparison step of comparing the phase of the second absolute angle data with the phase of the incremental angle data;
A second phase correction step of correcting the second absolute angle data when the phase of the second absolute angle data is shifted from the phase of the incremental angle data in the comparison result in the phase comparison step;
The absolute angle position of the rotating body is determined based on the phase detection first detection result obtained by correcting the first detection result with the phase difference, the second detection result, the second absolute angle data, and the incremental angle data. An angular position determination step,
An absolute angular position detection method for a rotary encoder, comprising: In any one of claims 7 to 11,
The first sensor unit includes a first magnet in which one N pole and one S pole are arranged around a rotation center axis, a first magnetoresistive element facing in the rotation center axis direction of the first magnet, A first Hall element that faces the first magnet, and a second Hall element that faces the first magnet and is positioned at a mechanical angle of 90 ° around the rotation center axis with respect to the first Hall element. And comprising
The absolute encoder of the rotary encoder, wherein the second sensor unit includes: a second magnet having a plurality of pole pairs arranged around the rotation center axis; and a second magnetoresistive element facing the second magnet. Angular position detection method.

Images