JP6611672B2 - Rotary encoder - Google Patents

Description

  The present invention relates to a rotary encoder that detects the relative rotation of a scale and outputs an absolute value.

An encoder that detects an absolute pattern provided on a scale and outputs an absolute value is described in Patent Document 1. In Patent Document 1, an absolute pattern is used in which one 0 is inserted into a portion where n-1 zeros continue in an M-sequence pattern of n bits and the period is ²ⁿ . In Patent Document 1, the scale extends linearly.

JP-A-9-280892

  Here, at present, there is no low encoder provided with the absolute pattern disclosed in Patent Document 1 and outputting an absolute value with high accuracy.

  In view of this point, an object of the present invention is to provide a rotary encoder that includes an absolute pattern based on an M-sequence pattern in a scale and can output an absolute value with high accuracy.

In order to solve the above-described problem, the rotary encoder of the present invention inserts one 0 into a portion where n-1 zeros are consecutive in an M-sequence pattern of n bits, and sets the period of the M sequence to 2 ⁿ . A scale having an annular absolute track having an absolute pattern, and an absolute value output unit that reads the absolute track of the scale that rotates relative to each other and outputs an absolute value, and each bit of the absolute pattern includes are arranged at a constant pitch along the track, the number of the bit n is Ri number der divisible when dividing 360 by 2 ^n, each bit of the absolute pattern is a Chaku磁領area or Muchaku magnetized zone There, the pitch is the angle obtained by dividing 360 ° by 2 ^n, said Absolue The track includes: a first track having the absolute pattern; and a second track extending in a relative rotational direction in parallel with the first track and positioned on an inner peripheral side of the first track, In the track, a magnetized area and a non-magnetized area are arranged at the same pitch as the first track in the circumferential direction of the scale, and the arrangement of the magnetized area and the non-magnetized area is opposite to that of the absolute pattern. A magnetic field of the first track and a non-magnetized region of the second track are adjacent to each other in the radial direction, and a non-magnetized region of the first track and a magnetized region of the second track Are adjacent in the radial direction .

In the present invention, as an absolute pattern, an M-sequence pattern of n bits is provided with a period of ²ⁿ by inserting one 0 into a portion where n-1 zeros are consecutive. Since such an absolute pattern has a period that is a power of 2, it is suitable for outputting an absolute value using an arithmetic circuit that performs an arithmetic operation using a binary number. In the present invention, the number of bits n is a number that is divisible when 360 is divided by ²ⁿ . In this way, when the M-sequence bits are arranged at a constant pitch in the circumferential direction of the scale, an absolute pattern for one period can be provided on the annular absolute track without excess or deficiency. As a result, an absolute pattern for one cycle can be accurately provided on the scale, and the absolute value can be accurately output by detecting the absolute pattern.

  In the present invention, the number of bits n may be 7.

  In the present invention, the absolute value output unit reads the absolute pattern of the first track and outputs a first signal, and reads the magnetization pattern of the second track. A second signal output unit that outputs two signals, and outputs an absolute value based on a differential signal of the first signal and the second signal.

  According to the present invention, the absolute track includes a first track having an absolute pattern and a second track having a magnetized pattern having a logical value opposite to that of the absolute pattern. Here, when the first track is read by the first signal output unit, the magnetic field is detected in the portion where the magnetized region and the non-magnetized region are adjacent to each other and in the portion close to the magnetized region in the non-magnetized region. The Therefore, the first signal output unit outputs a signal even in the non-magnetized region. On the other hand, when the first signal output unit is reading the non-magnetized region of the first track, the second signal output unit is reading the magnetized region of the second track. A second signal larger than the first signal is output. Therefore, in the differential signal of the first signal output from the first signal output unit and the second signal output from the second signal output unit, the non-magnetization is performed in the non-magnetized region in the absolute pattern. It is possible to remove the influence of the magnetic field generated in the region near the magnetized region. As a result, as a differential signal, a waveform in which the waveform does not invert at the boundary between the magnetized region and the non-magnetized region can be obtained. Therefore, the absolute value can be accurately obtained using the threshold value.

  In the present invention, in the first track, the magnetized regions adjacent in the circumferential direction face the same pole, and in the second track, the magnetized regions adjacent in the circumferential direction are the same. It is desirable that the poles face each other. In this way, self-demagnetization works between adjacent magnetized regions in the first track. In addition, self-demagnetization works between adjacent magnetized regions in the second track. Therefore, when a magnetic field is detected in a portion close to the magnetized region in the non-magnetized region due to an overshoot of the magnetic field in the magnetized region in the portion where the magnetized region and non-magnetized region of each track are adjacent to each other. However, compared with the case where self-demagnetization is not working, the output of the signal resulting from the magnetic field is small.

According to the rotary encoder of the present invention, an absolute pattern in which an M-sequence period is 2 ⁿ by inserting one 0 into a portion where n-1 zeros are consecutive in an M-sequence pattern of n bits is an annular absolute When used in a track, the absolute value can be accurately output by reading the absolute track.

It is explanatory drawing of the rotary encoder to which this invention is applied. It is explanatory drawing of a magnetic track and a magnetoresistive element. It is explanatory drawing of an example of an absolute pattern. It is a block diagram of the control system of a rotary encoder. It is explanatory drawing of each signal which a magnetic sensor apparatus outputs. It is explanatory drawing of the 1st signal output from a 1st signal output part, the 2nd signal output from a 2nd signal output part, and the differential signal of a 1st signal and a 2nd signal. It is description of the magnetic field and its detection signal in the boundary part of a magnetization area | region and a non-magnetization area | region. It is explanatory drawing of the rotary encoder of a modification.

  A rotary encoder according to an embodiment to which the present invention is applied will be described below with reference to the drawings.

(overall structure)
FIG. 1 is an explanatory diagram of a rotary encoder to which the present invention is applied. As shown in FIG. 1, the rotary encoder 1 of this example includes a disk-shaped magnetic scale 2 and a magnetic sensor device 3 that reads the magnetic scale 2. The magnetic scale 2 includes an annular magnetic track 4 provided on a circular surface of the magnetic scale 2. The magnetic track 4 is formed coaxially with the axis L of the magnetic scale 2. The magnetic sensor device 3 is disposed at a position facing the magnetic track 4 of the magnetic scale 2 in the direction of the axis L. The magnetic sensor device 3 detects a change in the magnetic field formed on the surface of the magnetic scale 2 when the magnetic sensor device 3 and the magnetic scale 2 are relatively rotated about the axis L, and the magnetic scale device 2 or the magnetic sensor device 3 is detected. The absolute angular position of is output.

  The magnetic sensor device 3 includes a sensor substrate 12 such as a silicon substrate or a ceramic grace substrate, and a plurality of magnetoresistive elements (first magnetoresistive elements 37 for detecting an incremental signal) formed on a surface of the sensor substrate 12 facing the magnetic scale 2. , A second magnetoresistive element 38 for detecting an incremental signal, a first magnetoresistive element 45 for detecting an absolute value, and a second magnetoresistive element 46 for detecting an absolute value. The magnetoresistive elements 37, 38, 45, and 46 include a permalloy film as a magnetosensitive film. The magnetoresistive elements 37, 38, 45, 46 and the magnetic scale 2 face each other with a predetermined gap.

  In the rotary encoder 1, one of the magnetic scale 2 and the magnetic sensor device 3 is arranged on the fixed body side, and the other is arranged on the rotating body side. In this example, the magnetic scale 2 is disposed on the rotating body side, and the magnetic sensor device 3 is disposed on the fixed body side.

(Magnetic scale)
FIG. 2 is an explanatory view of the magnetic track 4 and the magnetoresistive elements 37, 38, 45, 46 provided on the magnetic scale 2. FIG. 2 is an enlarged view of a T portion surrounded by a dotted line in FIG. As shown in FIG. 1, the magnetic track 4 includes a first incremental track 21, a second incremental track 22 extending in the circumferential direction X of the magnetic scale 2 (the relative rotational direction of the magnetic scale 2 and the magnetic sensor device 3), and An absolute track 23 is provided. The first incremental track 21 is located on the outer peripheral side of the second incremental track 22, and the absolute track 23 is located on the outer peripheral side of the first incremental track 21.

  The absolute track 23 includes a first track 24 and a second track 25 that are arranged in parallel and extend in the circumferential direction X. The first track 24 is located on the outer peripheral side of the second track 25. The first track 24 and the second track 25 are provided coaxially with no gap in the radial direction Y. The first track 24 and the second track 25 of the first incremental track 21, the second incremental track 22, and the absolute track 23 are coaxial. Note that the absolute track 23 may be provided inside the first incremental track 21 and the second incremental track 22.

As shown in FIG. 2, the first incremental track 21 has a first incremental pattern 21a formed with a first pitch P1. First incremental pattern 21a
Is one in which N poles and S poles are alternately magnetized in the circumferential direction X at the first pitch P1.

  The second incremental track 22 has a second incremental pattern 22a formed at a second pitch P2 having a pitch length longer than the first pitch P1. The second incremental pattern 22a is obtained by alternately magnetizing N poles and S poles in the circumferential direction X at the second pitch P2. The first incremental track 21 is positioned between the absolute track 23 and the second incremental track 22 in the radial direction Y. The first incremental pattern 21 a and the second incremental pattern 22 a form a strong and weak magnetic field in which the strength of the magnetic field appears perpendicular to the surface of the magnetic scale 2.

  The first track 24 of the absolute track 23 has an absolute pattern 24a formed with a third pitch P3 having a pitch length longer than the first pitch P1 and the second pitch P2. The absolute pattern 24a is a pattern in which a magnetized magnetized region and a non-magnetized non-magnetized region are arranged in the circumferential direction X with a pseudo-random pattern with a third pitch P3. Each magnetized region has an N pole and an S pole in the circumferential direction X. In the first track 24, the magnetized regions adjacent in the circumferential direction X have the same poles facing each other. The absolute pattern 24 a forms a strong and weak magnetic field in which the strength of the magnetic field appears perpendicular to the surface of the magnetic scale 2.

  The second track 25 of the absolute track 23 includes a magnetized pattern 25a in which a magnetized region and a non-magnetized region are arranged at the third pitch P3 and opposite to the absolute pattern 24a. Therefore, in the absolute track 23, the non-magnetized region of the second track 25 is positioned next to the magnetized region of the first track 24 in the radial direction Y. In addition, the magnetized region of the second track 25 is located next to the non-magnetized region of the first track 24 in the radial direction Y. The magnetized pattern 25 a forms a strong and weak magnetic field in which the strength of the magnetic field appears perpendicular to the surface of the magnetic scale 2. In the second track 25, each magnetized region has an N pole and an S pole in the circumferential direction X. In the second track 25, the magnetized regions adjacent in the circumferential direction X have the same poles facing each other.

  The third pitch P3, which is the formation pitch of the absolute pattern 24a and the magnetized pattern 25a, is an integral multiple of the first pitch P1 and the second pitch P2.

(Absolute pattern)
The absolute pattern 24a in this example is based on a 7-bit M-sequence pattern. That is, the absolute pattern 24a represents the absolute angular position of the magnetic scale 2 by the arrangement of the magnetized region and the non-magnetized region in seven consecutive regions. More specifically, the absolute angle on the magnetic scale 2 is determined by the arrangement of 1 and 0 in seven consecutive regions when the magnetized region has a logical value of 1 and the non-magnetized region has a logical value of 0. The position is indicated by a 7-bit value.

FIG. 3 is an explanatory diagram of an example of an absolute pattern 24 a provided in the first track 24. First, the M series is a numerical sequence represented by the following linear recurrence formula (1).
Xn = Xn-p + Xn-q (p> q) (1)
n is the number of bits, the value of each term is 0 or 1, and the + sign is an exclusive logical ring (XOR). A number sequence obtained by dividing the M sequence by n bits is the M sequence pattern. Here, since the M series pattern is a pseudo-random number, the absolute angular position of the magnetic scale 2 can be specified by acquiring the M series pattern.

Here, in this example, an absolute pattern 24a is obtained by inserting one zero into a portion where n-1 zeros are consecutive in a 7-bit (n = 7) M-sequence pattern. Further, in this example, by inserting one 0 into a portion where n-1 0s are consecutive in a 7-bit M-sequence pattern, the period of the M sequence until the same M-sequence pattern appears next is 2 ⁿ .

  When creating the absolute pattern 24a of this example, first, an M series is created with n = 7, p = 7, q = 1, and an initial value of “1100001” (arbitrary). Next, attention is paid to an M sequence pattern obtained by dividing the M sequence by 7 bits. Here, as shown in FIG. 3, in the M sequence pattern starting from the 56th bit of the M sequence, there are 6 (n−1) 0s in succession. Accordingly, one 0 is inserted as a new bit at the 57th position of the M series, and this is used as the absolute pattern 24a.

Note that there is no M sequence pattern of “0000000” in the M sequence (normal M sequence) in which 0 is not inserted. Therefore, the period of the M sequence until the next same M sequence pattern appears is 2 ⁿ −1 (2 ⁷ −1). On the other hand, the absolute pattern 24a of this example in which 0 is inserted as the 57th new bit has an M-sequence pattern of “0000000”. Thereby, the period of the M sequence until the next same M sequence pattern appears is 2 ⁿ (2 ⁷ ).

Here, the third pitch P3 which is the formation pitch of each region (magnetized region or non-magnetized region) of the absolute pattern 24a is an angle (2.8125 °) obtained by dividing 360 ° by ²ⁿ . In other words, each bit of the absolute pattern 24a is arranged along the first track 24 in the circumferential direction X of the magnetic scale 2 at an angle (third pitch P3) obtained by dividing 360 ° by ²ⁿ .

(Magnetic sensor)
FIG. 4 is a schematic block diagram showing a control system of the rotary encoder 1. FIG. 5 is an explanatory diagram of each signal acquired by the magnetic sensor device 3 by reading the magnetic scale 2. FIG. 5 schematically shows the arrangement of the absolute value detection first magnetoresistive element 45 and the absolute value detection second magnetoresistive element 46. FIG. 6 is an explanatory diagram of a first signal output from the first signal output unit, a second signal output from the second signal output unit, and a differential signal between the first signal and the second signal. FIG. 6A is an explanatory diagram schematically showing the absolute track 23, the first signal output unit, and the second signal output unit. 6B is a graph of the first signal output from the first signal output unit, FIG. 6C is a graph of the second signal output from the second signal output unit, and FIG. ) Is a graph of the differential signal, and FIG. 6E is an absolute value. In FIG. 6A, a gap is provided between the first track 24 and the second track 25 in order to explain the magnetic field in the magnetized region. FIG. 7A is an explanatory diagram of the magnetic field at the boundary portion between the magnetized region and the non-magnetized region in the absolute pattern, and FIG. 7B shows the first signal at the boundary portion between the magnetized region and the non-magnetized region. It is a graph of.

  As shown in FIG. 4, the magnetic sensor device 3 includes a first incremental signal output unit 31, a second incremental signal output unit 32, an incremental signal calculation unit 33, an absolute value output unit 34, and an absolute position acquisition unit 35. .

  As shown in FIG. 2 and FIG. 4, the first incremental signal output unit 31 includes a first magnetoresistive element 37 for detecting an incremental signal disposed opposite to the first incremental track 21. The first magnetoresistive element 37 for detecting an incremental signal has a magnetosensitive direction in the circumferential direction X. As shown in FIG. 5, the first incremental signal output unit 31 has a first incremental signal θA having a first wavelength λ1 having a length corresponding to the first pitch P1 of the first incremental pattern 21a as the magnetic scale 2 moves. Is output. The first incremental signal θA is a periodic signal whose phase changes from 0 to 2π every time the magnetic scale 2 rotates by the first pitch P1.

As shown in FIGS. 2 and 4, the second incremental signal output unit 32 includes a second magnetoresistive element 38 for detecting an incremental signal disposed opposite to the second incremental track 22. The second magnetoresistive element 38 for detecting the incremental signal has a magnetic sensitive direction in the circumferential direction X. As shown in FIG. 5, the second incremental signal output unit 32 generates a second incremental signal having a second wavelength λ <b> 2 having a length corresponding to the second pitch P <b> 2 of the second incremental pattern 22 a as the magnetic scale 2 moves. Output θB. The second incremental signal θB is a periodic signal whose phase changes from 0 to 2π every time the magnetic scale 2 rotates by the second pitch P2.

  The incremental signal calculation unit 33 calculates the third incremental signal θC having the third wavelength λ3 based on the first incremental signal θA and the second incremental signal θB. The third incremental signal θC is a vernier signal obtained by subtracting the phase of the second incremental signal θB from the phase of the first incremental signal θA.

  Here, the third wavelength λ3 is an integral multiple of the first wavelength λ1 of the first incremental signal θA, and is an integral multiple of the second wavelength λ2 of the second incremental signal θB. The third wavelength λ3 is a length corresponding to the third pitch P3 that is the pitch length of the absolute pattern 24a. The third incremental signal θC is a periodic signal whose phase changes from 0 to 2π every third pitch P3.

  Next, as shown in FIGS. 2 and 4, the absolute value output unit 34 reads a first track 24 (absolute pattern 24a) and outputs a first signal E1, and a second track. 25 (magnetization pattern 25a) and a second signal output unit 42 that outputs a second signal E2. The absolute value output unit 34 outputs an absolute value ABS based on the differential signal D between the first signal E1 and the second signal E2.

  Here, as shown in FIG. 5, the first signal output unit 41 includes a plurality of absolute value detection first magnetoresistive elements 45 facing the first track 24 at the third pitch P3. Each of the plurality of absolute value detecting first magnetoresistive elements 45 has a magnetic sensitive direction in the circumferential direction X. The first signal output unit 41 detects the respective magnetic fields of the plurality of regions of the absolute pattern 24a continuous in the circumferential direction X by the plurality of absolute value detection first magnetoresistive elements 45 and outputs the first signal E1. To do. In this example, in order to obtain a 7-bit absolute value ABS, the first signal output unit 41 includes seven absolute value detection first magnetoresistive elements 45, and magnetic fields in seven consecutive regions of the first track 24. Is detected. FIG. 6B is a graph of the first signal E1 output from the first signal output unit 41 when the magnetic fields of the seven regions of the first track 24 are detected.

  The second signal output unit 42 includes a plurality of absolute value detecting second magnetoresistive elements 46 facing the second track 25 at the third pitch P3. Each of the plurality of absolute value detecting second magnetoresistive elements 46 has a magnetosensitive direction in the circumferential direction X. The second signal output unit 42 detects the magnetic field of each of the plurality of regions of the magnetization pattern 25a continuous in the circumferential direction X by the plurality of absolute value detection second magnetoresistive elements 46, and outputs the second signal E2. Output. In this example, in order to obtain a 7-bit absolute value ABS, the second signal output unit 42 includes seven absolute value detecting second magnetoresistive elements 46, and magnetic fields in seven consecutive regions of the second track 25. Is detected. FIG. 6C is a graph of the second signal E2 output from the second signal output unit 42 when the magnetic fields of the seven regions of the second track 25 are detected.

Here, the seven absolute value detecting first magnetoresistive elements 45 and the seven absolute value detecting second magnetoresistive elements 46 are in the same position in the circumferential direction X (positions overlapping when viewed from the radial direction Y). The arranged absolute value detecting first magnetoresistive element 45 and absolute value detecting second magnetoresistive element 46 are configured as one set (a pair). Also, each set of absolute value detecting first magnetoresistive element 45 and absolute value detecting second magnetoresistive element 46 are connected in series between the voltage input terminal Vcc and the ground terminal GND to form a series circuit (first Series circuit) 47 is formed.

  The absolute value output unit 34 outputs a differential signal D (middle) output from a midpoint 48 between the absolute value detecting first magnetoresistive element 45 and the absolute value detecting second magnetoresistive element 46 in the series circuit 47. The absolute value ABS is output based on the point voltage.

  FIG. 6D is a graph of the differential signal D (midpoint voltage) when the magnetic field of seven consecutive areas of the absolute track 23 is detected. As shown in FIG. 6D, the differential between the first signal E1 and the second signal E2 is output as a differential signal D (midpoint voltage) from the middle point 48 of the series circuit 47. Therefore, when the absolute value detecting first magnetoresistive element 45 detects the magnetic field in the magnetized region and the absolute value detecting second magnetoresistive element 46 detects the magnetic field in the non-magnetized region, the differential A voltage signal having a midpoint potential E0 or higher is output as the signal D. On the other hand, when the absolute value detecting first magnetoresistive element 45 detects the magnetic field in the non-magnetized region and the absolute value detecting second magnetoresistive element 46 detects the magnetic field in the magnetized region, the differential A voltage signal lower than the midpoint potential E0 is output as the signal D.

  Therefore, the absolute value output unit 34 uses the midpoint potential E0 as a threshold value, sets 1 as an output from a group in which the differential signal D is greater than or equal to the threshold value, and 0 as an output from a group in which the differential signal D is lower than the threshold value The absolute value ABS (M series pattern) of the bit is output. As a result, the absolute value ABS is as shown in FIG. The midpoint potential E0 is a voltage signal output from the midpoint 48 in a state in which neither the absolute value detection first magnetoresistance element 45 nor the absolute value detection second magnetoresistance element 46 detects a magnetic field. It is.

  Here, when the absolute track 23 includes only the first track 24, when the first signal output unit 41 reads the first track 24, as shown in FIG. The first signal E1 is output at a portion where the region R1 and the non-magnetized region R0 are adjacent to each other and at a portion near the magnetized region R1 in the non-magnetized region R0. That is, as shown in FIG. 7A, at the boundary position R between the magnetized region R1 and the non-magnetized region R0, a magnetic field overshoots from the magnetized region R1 to the non-magnetized region R0 and returns to the magnetized region R1. Since F is generated, the absolute value detecting first magnetoresistive element 45 detects the magnetic field F and outputs the first signal E1. Therefore, if the threshold value for obtaining the absolute value ABS is not set appropriately, the output may exceed the threshold value even in the non-magnetized region R0, and the absolute value ABS may not be obtained accurately. .

  On the other hand, in this example, as shown in FIGS. 6B and 6C, when the first signal output unit 41 reads the non-magnetized region of the first track 24, the second signal Since the output unit 42 reads the magnetized region of the second track 25, a signal larger than the first signal output unit 41 is output from the second signal output unit 42. Further, as shown in FIG. 7B, at the boundary position R between the magnetized region R1 and the non-magnetized region R0, the magnetic flux density of the magnetized region R1 is greater than the magnetic flux density (on the non-magnetized region R0 side). Therefore, the inclination angle θ1 of the signal on the magnetization region R1 side with respect to the boundary position R in the first signal E1 output from the first signal output unit 41 is greater than the boundary position R. Also, the angle is larger than the inclination angle θ2 of the signal on the non-magnetized region R0 side. Therefore, if the differential between the first signal E1 from the first signal output unit 41 and the second signal E2 from the second signal output unit 42 is obtained, the boundary position R between the magnetized region R1 and the non-magnetized region R0. , A signal having a waveform that does not have an inversion portion can be obtained. That is, the influence of the magnetic field generated in the non-magnetized region R0 near the magnetized region R1 can be removed.

Further, if the midpoint potential E0 is a threshold value, a logical value of 1 is greater than or equal to the threshold value, and a logical value of 0 is smaller than the threshold value, a 7-bit absolute value at the third wavelength λ3 corresponding to the third pitch P3. ABS can be acquired accurately. In other words, in the differential output, the center of the amplitude (midpoint potential E0) of the signal oscillating in the plus and minus directions is used as the threshold value, so that the logical value of the code length of the third wavelength λ3 can be obtained accurately. Thereby, the code length of each logical value is the same as the pitch of the arrangement of the magnetized region and the non-magnetized region, and is constant. Therefore, the cycle of the absolute value ABS output from the absolute value output unit 34 does not deviate from the cycle of the third incremental signal θC.

  In addition, if the magnetized region and the non-magnetized region are arranged at a constant pitch in the absolute track 23, the length of the magnetized portion in the circumferential direction X within each pitch is set to the third pitch P3. A 7-bit absolute value ABS can be accurately obtained at the corresponding third wavelength λ3. Therefore, the degree of freedom of magnetization with respect to the magnetization region increases.

  Next, the absolute position acquisition unit 35 acquires the absolute angular position of the magnetic scale 2 based on the absolute value ABS, the phase of the third incremental signal θC, and the phase of the first incremental signal θA. The absolute position acquisition unit 35 includes an arithmetic circuit that calculates the absolute value ABS as a binary number when acquiring the absolute angular position.

(Absolute angular position detection operation)
When the magnetic scale 2 arranged on the rotating body side rotates, as shown in FIG. 5, the first incremental signal output unit 31 outputs the first incremental signal θA having the first wavelength λ1, and the second incremental signal output unit 32 A second incremental signal θB having a second wavelength λ2 longer than the first wavelength λ1 is output. In parallel with this, the incremental signal calculation unit 33 acquires the third incremental signal θC of the third wavelength λ3 based on the first incremental signal θA and the second incremental signal θB.

  The absolute value output unit 34 outputs the absolute value ABS every time the magnetic scale 2 rotates by the third pitch P3. That is, the absolute value output unit 34 gives the absolute value ABS for each cycle of the third incremental signal θC. Therefore, the absolute position acquisition unit 35 can acquire the absolute angular position of the magnetic scale 2 based on the absolute value ABS of the absolute value ABS, the phase of the third incremental signal θC, and the phase of the first incremental signal θA.

  In this example, the absolute value ABS is acquired based on the differential signal D of the first signal E1 from the first signal output unit 41 and the second signal E2 from the second signal output unit 42. Since the differential signal D has a waveform that does not invert at the boundary between the magnetized region and the non-magnetized region of the absolute pattern 24a, the absolute value ABS is accurately determined based on the threshold value (midpoint potential E0). Can be obtained. In addition, since the center of the amplitude (midpoint potential E0) of the differential signal D that swings between plus and minus is used as a threshold value, the absolute value ABS is accurately the logical value of the code length of the third wavelength λ3. Can be obtained.

Further, in this example, the absolute pattern 24a is provided with a cycle of ²ⁿ by inserting one 0 into a portion where n-1 zeros continue in the M-sequence pattern of n bits. Here, the period of the absolute pattern 24a is a power of two. Therefore, it is suitable for calculating the absolute value ABS in the calculation circuit of the absolute position acquisition unit 35 that performs calculation in binary numbers.

Further, in this example, the number of bits n is a number that is divisible when 360 is divided by ²ⁿ . Accordingly, when the angle obtained by dividing 360 ° by ²ⁿ is the pitch for arranging the bits of the absolute pattern 24a (the third pitch P3), the absolute pattern 24a for one cycle is excessively or insufficient in the annular first track 24. Can be provided. As a result, the absolute pattern 24a for one cycle can be accurately provided on the magnetic scale 2, so that the absolute value ABS can be accurately output by detecting the absolute pattern 24a.

(Other embodiments)
In the above example, the absolute track 23 includes the absolute pattern 24a representing the absolute angular position with 7 bits, but the number of bits n is not limited to 7. For example,
The absolute track is an absolute pattern 24a representing an absolute angular position with 3 bits.
Can be provided. In this case as well, the absolute pattern may be set to a cycle of 2 ⁿ by inserting one 0 into a portion where two (n−1) zeros are consecutive in the M-sequence pattern of 3 bits. it can. In this case, the third pitch P3 for arranging each bit along the absolute track is an angle (45 °) obtained by dividing 360 ° by 2 ⁿ (n = 3).

  In the above example, the first incremental track 21 and the second incremental track 22 are provided as the incremental tracks, and the first incremental signal θA and the second incremental track 22 output by reading the first incremental track 21 are read. Although the third incremental signal θC having the third wavelength λ3 is calculated based on the output second incremental signal θB, the number of incremental tracks may be one. In this case, the incremental track may be provided with a magnetized pattern that can output an incremental signal of the third wavelength λ3 when the incremental track is read. Further, the absolute position acquisition unit 35 can acquire the absolute angular position of the magnetic scale 2 based on the absolute value ABS and the incremental signal of the third wavelength λ3.

In the rotary encoder 1, the magnetic track 4 extending in the circumferential direction X is provided on the circular surface of the disk-shaped magnetic scale 2, but the magnetic track 4 is provided on the annular outer peripheral surface of the disk-shaped magnetic scale 2. May be. FIG. 8 is an explanatory diagram of a rotary encoder according to a modification. In the rotary encoder 1A of this example, a magnetic track 4 extending in the circumferential direction X (the relative rotational direction of the magnetic scale 2 and the magnetic sensor device 3) is provided on the annular outer peripheral surface of the disk-shaped magnetic scale 2.

  The magnetic track 4 includes a first incremental track 21, a second incremental track 22, and an absolute track 23 that extend in the circumferential direction X of the magnetic scale 2. The first incremental track 21 is positioned next to the second incremental track 22 in the direction of the axis L. Further, the absolute track 23 is located on the opposite side of the second incremental track 22 with the first incremental track 21 in between in the axis L direction. The absolute track 23 includes a first track 24 and a second track 25 that are arranged in parallel and extend in the circumferential direction X. The first track 24 and the second track 25 are provided at positions adjacent to each other in the axis L direction. The second track 25 is located next to the first incremental track 21 in the axis L direction. Note that the arrangement of the tracks 21, 22, and 23 constituting the magnetic track 4 is not limited to this.

The magnetic sensor device 3 faces the magnetic scale 2 from a direction (radial direction Y) orthogonal to the axis L. The magnetic sensor device 3 detects a change in the magnetic field formed on the surface of the magnetic scale 2 when the magnetic sensor device 3 and the magnetic scale 2 are relatively rotated about the axis L, and the magnetic scale device 2 or the magnetic sensor device 3 is detected. The absolute angular position of is output. Since the rotary encoder 1A of the present example has a configuration corresponding to that of the rotary encoder 1, the same reference numerals are given to the corresponding portions, and description thereof is omitted.

  Further, the rotary encoder provided with the absolute pattern 24a of this example may be an optical type. In this case, only the first track 24 is provided as the absolute track 23. In the absolute pattern 24a, it is assumed that a light-shielding portion and a light-transmitting portion are arranged on the disk-shaped scale 2 along the absolute track 23 at the third pitch P3. Alternatively, in the absolute pattern 24a, it is assumed that a reflective portion and a non-reflective portion are arranged on the disk-shaped scale 2 along the absolute track 23 at the third pitch P3. In this case, instead of the magnetic sensor device 3, a reading device that optically reads seven regions of the absolute pattern 24a is provided. As the reading device, one having a light emitting part and a light receiving part can be used.

1.1A ... Rotary encoder 2 ... Magnetic scale (scale)
23 ... Absolute track 24 ... First track 24a ... Absolute pattern 25 ... Second track 25a ... Magnetized pattern 34 ... Absolute value output unit 41 ... First signal output unit 42 ... Second signal output unit P3 ... Third pitch (constant) Pitch)
D ... Differential signal ABS ... Absolute value X ... Circumferential direction

Claims (4)

A scale provided with an annular absolute track having an absolute pattern in which an M-sequence period is 2 ⁿ by inserting one 0 into a portion where n-1 zeros continue in an M-sequence pattern of n bits;
An absolute value output unit that reads the absolute track of the scale that rotates relative to each other and outputs an absolute value;
Each bit of the absolute pattern is arranged at a constant pitch along the absolute track,
It said number of bits n is Ri number der divisible when dividing 360 by 2 ^n,
Each bit of the absolute pattern is a magnetized region or a non-magnetized region,
The pitch is an angle obtained by dividing 360 ° by 2 ⁿ ,
The absolute track includes a first track having the absolute pattern, and a second track extending in a relative rotational direction in parallel with the first track and positioned on an inner peripheral side of the first track,
In the second track, a magnetized region and a non-magnetized region are arranged at the same pitch as the first track in the circumferential direction of the scale, and the arrangement of the magnetized region and the non-magnetized region is opposite to the absolute pattern. With a magnetized pattern of
The magnetized region of the first track and the non-magnetized region of the second track are adjacent in the radial direction, and the non-magnetized region of the first track and the magnetized region of the second track are in the radial direction. A rotary encoder characterized by being adjacent to each other . In claim 1 ,
The rotary encoder is characterized in that the number of bits n is 7. In claim 1 ,
The absolute value output unit reads the absolute pattern of the first track and outputs a first signal, and reads the magnetization pattern of the second track and outputs a second signal. A rotary encoder that outputs an absolute value based on a differential signal of the first signal and the second signal. In claim 1 or 3 ,
In the first track, the magnetized regions adjacent in the circumferential direction have the same poles facing each other,
In the second track, the magnetized regions adjacent in the circumferential direction have the same poles opposed to each other.