JP7270903B2 - flat plate encoder - Google Patents

Description

The present invention relates to an encoder for detecting linear position or rotational angle, and more particularly to a plate-type encoder.

Optical encoders, resolvers, and the like are commonly used as angle detectors for detecting rotation angles among encoders. A general resolver has a configuration in which a plurality of slots are provided inside a ring-shaped stator, excitation windings are arranged in these slots, and a rotor consisting of an iron core is arranged at the center of the stator. . In this way, the conventional resolver, which consists of a stator and a rotor equipped with excitation windings, has been made thinner and lighter in order to solve the problem of unstable accuracy due to processing distortion of the iron core and variations in assembly and windings. At the same time, substrate-type resolvers have been proposed in order to improve and stabilize mass production accuracy.

For example, substrate type resolvers have been proposed as shown in Patent Documents 1 and 2.
A substrate type resolver is composed of a stator substrate and a rotor substrate arranged with a predetermined gap from the stator substrate. The stator substrate is formed with a sine winding pattern, a cosine winding pattern, and an excitation winding pattern arranged on the outer peripheral side of the sine winding pattern and the cosine winding pattern. A rotor winding pattern is formed on the rotor substrate.

The operation of the conventional substrate type resolver will be described below.
When an excitation current is applied to the excitation winding pattern of the stator substrate, magnetic flux is generated in the direction perpendicular to the stator substrate by the excitation winding pattern. When the rotor substrate is rotating, the rotor winding pattern crosses the magnetic flux generated by the excitation winding pattern, and the rotor winding pattern generates an eddy field proportional to the angle of rotation. A sin signal is output from the winding pattern, and a cos signal is output from the cos winding pattern. However, since the number of pole pairs is one or a low order, they have problems in accuracy and resolution, and lack redundant functions.

Further, Patent Document 3 discloses an electromagnetic induction type rotation sensor configured by a printed circuit board.
In the electromagnetic induction type rotation sensor of Patent Document 3, a sine receiving land pattern and a cosine receiving land pattern are formed as 16 pole-log patterns on the stator substrate side, respectively, and the corresponding rotor-side substrate is also divided into 16. A land pattern is formed.
On the inner peripheral side of the receiving land pattern on which the 16 pole pair pattern is formed on the stator substrate side, one turn of the sine receiving land pattern and one turn of the cos receiving land pattern are formed. Also one land pattern is formed.

Japanese Patent Application Laid-Open No. 2001-314069 Japanese Patent Application Laid-Open No. 2001-194183 Japanese Patent Application Laid-Open No. 2004-333478

In the case of an optical encoder as a conventional angle detector, it is necessary to use a code plate corresponding to the rotor, so there is a problem that it cannot be miniaturized (diameter of 30 mm or less).
In addition, the conventional substrate-type resolver as described above can be made smaller than a resolver equipped with an excitation winding, but if a redundant system is to be provided, the structure becomes double and the size increases. .

Further, according to the sensor disclosed in Patent Document 3, the number of sensor patterns on the rotor-side substrate and the number of sin-receiving sensor patterns and cosine-receiving sensor patterns on the stator-side substrate are the same. There is a problem that there is a lot of harmonic noise and the residual error is large and the accuracy is low.

SUMMARY OF THE INVENTION Accordingly, the present invention has been made to solve the above problems, and an object of the present invention is to provide a planar encoder which is smaller than the conventional one and has higher accuracy.

According to the flat plate encoder of the present invention, the stator plate is composed of a multi-layered plate, and the mover plate is arranged at a predetermined interval with respect to the stator plate, and the mover plate is A first sensor pattern in which M number of pole pairs of sensor patterns are discretely and evenly arranged is formed on the facing surface, and an exciting coil pattern is formed on the facing surface of the stator flat plate to the mover flat plate. In addition, a first receiving coil pattern facing the first sensor pattern of the mover plate is formed inside or outside the excitation coil pattern, and the first receiving coil pattern is formed on the surface or inner layer of the stator plate. and a first receiving coil pattern (B) formed on either the surface or the inner layer of the stator flat plate, and the The number of patterns N of the first receiving coil pattern (A) and the first receiving coil pattern (B) is such that N≠M with respect to the number of pole pairs M of the first sensor pattern, so that the amplitude envelope is It is characterized in that it is formed in a serpentine shape that is a periodic function with the same period and amplitude but out of phase.
By adopting this configuration, a uniform magnetic field is generated by the excitation coil pattern on the stator flat plate side, and the uniform magnetic field penetrates the first sensor pattern provided on the mover flat plate side to generate an eddy magnetic field. This eddy magnetic field is received by the first receiving coil pattern. At this time, the number of patterns N of the first receiving coil pattern with respect to the number of pole pairs M of the first sensor pattern is N≠M. For this reason, it is possible to detect the difference in the movable distance or the movable angle of the mover plate with respect to the stator plate, thereby achieving high accuracy due to the vernier effect.

Further, a second sensor pattern in which sensor patterns having a number of pole pairs of m are discretely and evenly arranged is formed at a position different from the first sensor pattern formed on the movable plate, and the excitation of the stator plate is performed. A second receiving coil pattern facing the second sensor pattern of the mover plate is formed inside or outside the coil pattern, and the second receiving coil pattern is formed on either the surface or the inner layer of the stator plate. and a second receiving coil pattern (B) formed on the other of the surface or inner layer of the stator flat plate, wherein the second receiving coil pattern The number of patterns n of (A) and the second receiving coil pattern (B) is such that the amplitude envelope has the same period and amplitude so that n≠m with respect to the number of pole pairs m of the second sensor pattern. may be formed in a serpentine shape that is a periodic function with a phase shift at .
According to this configuration, a uniform magnetic field penetrates through the second sensor pattern, an eddy magnetic field proportional to the rotation angle is generated, and this eddy magnetic field is received by the second receiving coil pattern. In this way, it is possible to provide a second receiving coil pattern to ensure a redundant system and to make the size smaller than before.

Further, the pole pair number M of the first sensor pattern and the pole pair number m of the second sensor pattern may be characterized by M≠m.
If the number of pole pairs of the first receiving coil pattern and the number of pole bodies of the second receiving coil pattern are made different in this way, the absolute angle of the rotor rotation angle based on the reference position can be calculated based on the signals obtained from the respective receiving coil patterns. can be detected (absolute type).

Further, the first receiving coil pattern (A) and the first receiving coil pattern (B) are composed of a conductor pattern whose envelope curve is the periodic function and a reverse polarity conductor whose polarity is opposite to the conductor pattern. The pattern may be formed in a unicursal state, and both ends of the conductor pattern may be formed so as to allow input to the positive and negative input terminals of the operational amplifier.
According to this configuration, each of the first receiving coil pattern (A) and the first receiving coil pattern (B) in the first receiving coil pattern is positive and negative symmetrical (area symmetrical) with respect to the zero cross point of each periodic function. Since it is formed so as to be such that noise is prevented from being mixed, highly accurate signal reception is possible. Further, it is possible to realize one-rotation absolute angle detection by the first receiving coil pattern and to achieve miniaturization.
Furthermore, according to a comparison between Patent Document 3 and the present application, in Patent Document 3, both a receiving coil pattern and a receiving coil pattern whose phase is shifted by π/2 from this receiving coil pattern are formed on the first layer of the stator-side substrate. However, it is necessary to electrically connect the second and lower layers through via holes where the two receiving coil patterns overlap, and there are a large number of via holes. If there are many via holes, there is a problem that the reliability is lowered and the electric resistance is increased. In this regard, in the present invention, the two receiving coil patterns are completely separated in the first layer and the second layer, so the number of locations for providing via holes is very small, which improves reliability and reduces electrical resistance. Improved performance can be achieved.

Also, the number of patterns N of the first receiving coil patterns may be N=Np*k (where Np is an arbitrary prime number and k is a positive integer equal to or greater than 1).
Furthermore, the number N of patterns of the first receiving coil pattern is N=Np*k (where Np is an arbitrary prime number and k is a positive integer equal to or greater than 1), and the number of patterns n of the second receiving coil pattern is , n=np*k (where np is an arbitrary prime number and k is a positive integer equal to or greater than 1).
The number of patterns of the unit phase and the number of poles is N/M (or M+1) = a/b, where a/b is an irreducible fraction. This is because all spatial harmonics other than ±1 can be eliminated. That is, with such a configuration, the superimposition of a large number of harmonic components that normally occur in the gap between the mover plate and the stator plate is eliminated, and only the number of spatial harmonic components corresponding to an arbitrary prime number is eliminated. Therefore, errors caused by spatial harmonics can be suppressed, and the S/N ratio can be increased.

Further, the length of the first sensor pattern of the mover plate in the direction orthogonal to the moving direction of the mover plate is the maximum amplitude of each pattern of the first receiving coil pattern of the stator plate. It may be characterized by being formed to be longer than twice as long.
Further, the length of the second sensor pattern of the mover plate in the direction orthogonal to the moving direction of the mover plate is the maximum amplitude of each pattern of the second receiving coil pattern of the stator plate. It may be characterized by being formed to be longer than twice as long.
By adopting this configuration, even if the mover plate is displaced in a direction orthogonal to the moving direction of the mover plate, the eddy current generated by the sensor pattern of the stator plate can be reliably received by the receiving coil pattern. Therefore, accurate position/angle detection is possible.

Further, the number of patterns N of the first receiving coil pattern (A) and the first receiving coil pattern (B) is N=M±1 or N=M±2 with respect to the pole pair number M of the first sensor pattern. It may be characterized by being formed so as to be
Furthermore, the number of patterns n of the second receiving coil pattern (A) and the second receiving coil pattern (B) is n=m±1 or n=m±2 with respect to the pole pair number m of the second sensor pattern. It may be characterized by being formed so as to be

Further, the mover flat plate may have a circular shape when the facing surface is viewed from the front.

Further, the stator flat plate may be formed of a printed circuit board.
Since N=M±1 or M±2, the electrical angle obtained by dividing one round 360° by M is detected by the electrical angle averaged by one round M±1 equal or M±2 equal. High precision can be achieved by the vernier effect.

According to the flat plate encoder of the present invention, high resolution can be achieved by the vernier effect, and miniaturization and weight reduction can be achieved.

1 is a side view of a printed circuit board type angle detector as an example of a flat plate type encoder; FIG. FIG. 4 is an explanatory diagram showing the facing surface of a stator printed circuit board as an example of a stator flat plate; FIG. 4 is an explanatory diagram showing the facing surface of a rotor printed circuit board as an example of a mover flat plate; FIG. 3 is an explanatory diagram in which an operational amplifier is connected in a straight line to make the sine receiving coil pattern easier to see; 4 is a graph for determining the shape of a sine receive coil pattern; FIG. 2 is a schematic cross-sectional view of a first layer and a second layer of a stator printed circuit board as an example of a stator flat plate; FIG. 10 is a graph showing a mechanical angle and an output curve of a pattern that is equally divided into 33 perimeters in the case of 31 pole pairs; FIG.

An embodiment of a printed circuit board type angle detector will be described below as an example of a flat plate type encoder.
FIG. 1 shows a side view of the printed circuit board type angle detector of this embodiment.
A printed circuit board type angle detector (hereinafter sometimes simply referred to as an angle detector) 10 includes a stator printed circuit board (stator flat plate) 12 and a rotor printed circuit board (mover flat plate) 14 as a basic configuration. . In the angle detector 10 of this embodiment, the stator printed circuit board 12 and the rotor printed circuit board 14 are arranged with a predetermined gap therebetween.

As the stator printed circuit board 12 and the rotor printed circuit board 14, general printed circuit boards made of an insulator can be used. As the material, for example, a glass epoxy board or a composite board generally used in a multilayer board can be adopted, but the material is not particularly limited.
An exciting coil pattern and a receiving coil pattern are formed on the stator printed circuit board 12, and a sensor pattern is formed on the rotor printed circuit board 14. These patterns are made of copper foil such as those formed on ordinary printed circuit boards. It should be formed by

(stator printed circuit board)
FIG. 2 shows a plan view of the printed shape formed on the surface (first layer) of the stator printed circuit board facing the rotor printed circuit board.
The diameter and thickness of the stator printed circuit board 12 of this embodiment are not limited to specific numerical values. The stator printed board 12 is composed of a multilayer board. In this embodiment, for example, the multilayer substrate is composed of six layers, but the number of layers is not particularly limited.

A triple excitation coil pattern is concentrically formed on the facing surface of the stator printed circuit board 12 . All of the triple excitation coil patterns are formed by one pattern. That is, an oscillator (not shown) is connected to the excitation coil pattern, and a high-frequency excitation current in the MHz range is input. It is formed with one pattern so as to return to the oscillator via the excitation coil pattern 20 of . A via hole 19 is formed at the end of each of the excitation coil patterns 16, 18, and 20, and the excitation coil patterns 16, 18, and 20 are connected through the via hole 19 in the second and subsequent layers. .

Each of the triple excitation coil patterns 16, 18 and 20 is also formed by a triple printed pattern.
Specifically, the outermost exciting coil pattern 16 is formed of three circular patterns so that the current flows three times clockwise when the facing surface is viewed in plan.
The excitation coil pattern 16 on the outermost circumference and the excitation coil pattern 18 on the intermediate circumference are connected by patterns on the second and subsequent layers of the printed circuit board.
The intermediate excitation coil pattern 18 is formed of three circular patterns so that the current flows three times counterclockwise when the facing surface is viewed in plan.
The excitation coil pattern 18 on the intermediate circumference and the excitation coil pattern 20 on the innermost circumference are connected by patterns on the second and subsequent layers of the printed circuit board.
The innermost exciting coil pattern 20 is formed of three circular patterns so that the current flows three times clockwise when the facing surface is viewed in plan.

A predetermined high-frequency excitation current from the oscillator first flows through the outermost excitation coil pattern 16 and then flows in the opposite direction through the intermediate excitation coil pattern 18 . Furthermore, from the intermediate circumference exciting coil pattern 18 , a high-frequency exciting current flows into the innermost exciting coil pattern 20 in the same direction as the outermost exciting coil pattern 16 .
Therefore, the magnetic fields between the outermost exciting coil pattern 16 and the intermediate exciting coil pattern 18 and between the intermediate exciting coil pattern 18 and the innermost exciting coil pattern 20 are canceled, resulting in triple A uniform magnetic field is generated over the entire exciting coil pattern.

Of the triple exciting coil patterns, the intermediate exciting coil pattern 18 may be formed on the second layer of the stator printed circuit board 12 . This is effective when it is difficult to secure a space on the facing surface.
Furthermore, the excitation coil pattern need not be formed as a triple excitation coil pattern if a single or double excitation coil pattern generates a sufficient magnetic field. That is, it is sufficient that only one or two of the outermost circumference, the intermediate circumference, or the innermost circumference are formed as the exciting coil pattern.

Further, in the present embodiment, between the intermediate excitation coil pattern 18 and the innermost excitation coil pattern 20 of the stator printed circuit board 12, a first sensor pattern facing a first sensor pattern of the rotor printed circuit board 14, which will be described later, is provided. A receiving coil pattern 25 is formed.
Between the outermost peripheral excitation coil pattern 16 and the intermediate peripheral excitation coil pattern 18 of the stator printed circuit board 12, a second receiving coil pattern 27 facing a second sensor pattern of the rotor printed circuit board 14, which will be described later, is formed. there is
The specific configurations of the first receiving coil pattern 25 and the second receiving coil pattern 27 will be described later.

(rotor printed circuit board)
FIG. 3 shows a plan view of a surface of the rotor printed circuit board 14 facing the stator printed circuit board 12. As shown in FIG.
The diameter and thickness of the rotor printed circuit board 14 of this embodiment are not limited to specific numerical values.
The rotor printed circuit board 14 may be a single layer, and may not be composed of a multi-layer circuit board.

On the facing surface of the rotor printed circuit board 14, fan-shaped sensor patterns are arranged in a predetermined number M (the number of pole pairs M) in the circumferential direction at equal intervals to form a first sensor pattern 22 . Further, on the outer peripheral side of the first sensor pattern 22, a predetermined number M±1 of fan-shaped sensor patterns (the number of pole pairs M±1) are arranged in the circumferential direction at equal intervals to form the second sensor pattern 24. are doing.
Each of the first sensor pattern 22 and the second sensor pattern 24 is configured to have the same area. The first sensor pattern 22 and the second sensor pattern 24 are each made of copper foil.
In this embodiment, the first sensor pattern 22 has 31 sensor patterns, that is, the number of pole pairs, and the second sensor pattern 24 has 32 sensor patterns, that is, the number of pole pairs.

Each sensor pattern of the first sensor pattern of the rotor printed circuit board 14 interlinks with the uniform magnetic field formed by the triple excitation coil pattern of the stator printed circuit board 12, eddy currents are generated on each sensor pattern, and the eddy currents generate magnetic flux. occurs. This magnetic flux is received by the first receiving coil pattern of the stator printed circuit board 12 to generate an output.
Similarly, each sensor pattern of the second sensor pattern of the rotor printed circuit board 14 interlinks with the uniform magnetic field formed by the triple excitation coil pattern of the stator printed circuit board 12, eddy currents are generated on each sensor pattern, and eddy currents are generated on each sensor pattern. A magnetic flux is generated by the electric current. This magnetic flux is received by the second receiving coil pattern of the stator printed circuit board 12 to generate an output.

(Structure of receiving coil pattern)
First, the first receiving coil pattern 25 will be described.
The first receiving coil pattern 25 is provided at a position facing the first sensor pattern 22 of the rotor printed circuit board 14 . In this embodiment, it is provided between the innermost exciting coil pattern 20 and the intermediate exciting coil pattern 18 .
The first receiving coil pattern 25 is composed of two patterns of a first sin receiving coil pattern (first receiving coil pattern (A)) 28 and a first cos receiving coil pattern (first receiving coil pattern (B)) 29. ing. The first sin receiver coil pattern will be mainly described below as the first receiver coil pattern.

In this embodiment, if the number of patterns of the first receiving coil pattern 25 (the number of zigzag-shaped divided detection units) is N, it is designed so that N=M±1 or N=M±2. Here, M is the number of pole pairs of the fan-shaped first sensor pattern 22 as described above. In this embodiment, M=31. Therefore, the number of patterns of the first receiving coil pattern 25 is N=32 or 33. The number of patterns N of the first receiving coil pattern 25 in this embodiment is N=33.

It is preferable that the number N of patterns of the first receiving coil pattern 25 is an integral multiple of a prime number, or that the number N of patterns is an integral multiple of the prime number ±1. This is to reduce spatial harmonics as described above.
By adopting such a configuration, superposition of a large number of harmonic components that normally occurs in the gap between the rotor-side printed circuit board 14 and the stator-side printed circuit board 12 is eliminated, and the number of spaces corresponding to an arbitrary prime number is eliminated. Since only harmonic components are used, errors caused by spatial harmonics can be suppressed and the S/N ratio can be increased.
Specifically, in this embodiment, as described above, the number of patterns N of the first receiving coil pattern 25 is N=33, so the prime number is 11×integer 3. By setting the number of pole pairs and the number of patterns based on a prime number as large as possible, it is possible to eliminate the superposition of many spatial harmonic components.

FIG. 4 shows a simplified drawing in which the first receiving coil pattern 25 of the stator-side printed circuit board 12, which is formed in a circular shape, is straightened to make it easier to see. Note that the number of patterns is not 33 because FIG. 4 is a simplified drawing.
The first receiving coil pattern 25 is formed so as to divide the zigzag-shaped divided detection section into one round by the number of patterns. An envelope connecting the vertices of the zigzag-shaped amplitude becomes a sinusoidal waveform. The first receiving coil pattern 25 is formed as a unicursal pattern so that both ends are connected to the positive and negative input terminals of the operational amplifier 30 . The sinusoidal waveform as the envelope curve with opposite polarities is formed so as to be symmetrical with respect to the zero-crossing point of the sinusoidal waveform.

In FIG. 4, the positive input terminal of the operational amplifier 30 is connected to the positive pattern 32 drawn by solid lines, and the negative input terminal of the operational amplifier 30 is connected to the negative pattern 33 drawn by broken lines. 33 is formed with a single pattern so as to be drawn in a single stroke. At the zero-cross point where the positive pattern 32 and the negative pattern 33 intersect, the patterns are connected in the second and lower layers through the via hole 19 . In this way, by arranging the receiving coil pattern unicursally and symmetrically on the positive electrode side and the negative electrode side, it is possible to prevent noise from being mixed in the output signal and to receive a highly accurate signal.
4, the excitation coil patterns 16, 18, and 20 are schematically shown so as to surround the first receiving coil pattern 25, and the excitation coil patterns 16, 18, and 20 have an oscillator 38 that generates a predetermined high-frequency current. is connected to apply

Next, based on FIG. 5, each amplitude (length in the radial direction) of the zigzag-shaped divided detection portion in the first receiving coil pattern 25 will be described.
FIG. 5 is a bar graph showing the height of the zigzag-shaped divided detection portion when the number of patterns of the first receiving coil pattern 25 is 33. In FIG. The height of the zigzag-shaped divided detection portion is calculated as follows.
First, one circumference of 360° is divided by the number of pole pairs of 33 to calculate the mechanical angles in the 33 divided detection units. The calculated mechanical angle is multiplied by the number of pole pairs 31 of the first sensor pattern on the rotor-side printed circuit board, and divided by 360° for one rotation to calculate the electrical angle at each divided detection unit. Then, the sine values of these electrical angles are arranged in order.

As for the first receiving coil pattern 25 described above, the first sin receiving coil pattern formed on the first layer (the facing surface of the rotor-side substrate) has been described.
As shown in FIG. 6, the first receiver coil pattern 25 has a first sin receiver coil pattern 28 formed on the first layer and a first cos receiver coil pattern 29 formed on the second layer. In order to ensure the detection of the magnetic flux in the first cosine receiving coil pattern 29 in the second layer, the thickness of the first layer should be made thinner than usual.
Note that FIG. 6 is a diagram showing a simplified configuration for showing the arrangement relationship of the receiving coil patterns, and actually has a laminated structure in which the first layer and the second layer are in complete contact.

The first cosine receiving coil pattern 29 has 33 patterns of zigzag-shaped divided detection portions whose phases are shifted by π/2 from the first sinus receiving coil pattern 28. is divided by the number of pole pairs, 33, to calculate the mechanical angle in the 33 divided detection units. The calculated mechanical angle is multiplied by the number of pole pairs 31 of the first sensor pattern on the rotor-side printed circuit board, and divided by 360° for one rotation to calculate the electrical angle at each divided detection unit. It is formed by arranging the cos values of the electrical angles in order.

Next, the second receiving coil pattern will be explained.
The second receiving coil pattern 27 is provided at a position facing the second sensor pattern 24 of the rotor printed circuit board 14 . In the present embodiment, it is provided between the intermediate excitation coil pattern 18 and the outermost excitation coil pattern 16 .
The second receiving coil pattern 27 is composed of two patterns, a second sin receiving coil pattern (second receiving coil pattern (A)) 32 and a second cos receiving coil pattern (second receiving coil pattern (B)) 33. ing. The second sin receiver coil pattern will be mainly described below as the second receiver coil pattern.

The number of pole pairs M of the second sensor pattern 24 is M=32. The number of patterns N of the second receiving coil pattern 27 facing this is N=33 and M+1.
It is preferable that the number N of patterns of the second receiving coil pattern 27 is an integral multiple of a prime number, or that the number N of patterns is an integral multiple of the prime number ±1. This is to reduce spatial harmonics.
By adopting such a configuration, superposition of a large number of harmonic components that normally occurs in the gap between the rotor-side printed circuit board 14 and the stator-side printed circuit board 12 is eliminated, and the number of spaces corresponding to an arbitrary prime number is eliminated. Since only harmonic components are used, errors caused by spatial harmonics can be suppressed and the S/N ratio can be increased.
Specifically, in this embodiment, similarly to the first receiving coil pattern 25, the number of patterns N of the second receiving coil pattern 27 is N=33, so the prime number is 11×3. By setting the number of pole pairs based on a prime number that is as large as possible, superimposition of many spatial harmonic components can be eliminated.

The pattern shape of the second receiving coil pattern 27 can also be explained with a simplified drawing as shown in FIG.
That is, the second receiving coil pattern 27 is formed so that the zigzag-shaped divided detection portions divide one round by the number of patterns. An envelope connecting the vertices of the zigzag-shaped amplitude becomes a sinusoidal waveform. The second receiving coil pattern 27 is formed as a unicursal pattern so that both ends are connected to the positive and negative input terminals of the operational amplifier 30 . The sinusoidal waveform as the envelope curve with opposite polarities is formed so as to be symmetrical with respect to the zero-crossing point of the sinusoidal waveform.

In FIG. 4, the positive input terminal of the operational amplifier 30 is connected to the positive pattern 32 drawn by solid lines, and the negative input terminal of the operational amplifier 30 is connected to the negative pattern 33 drawn by broken lines. 33 is formed with a single pattern so as to be drawn in a single stroke. At the zero-cross point where the positive pattern 32 and the negative pattern 33 intersect, the patterns are connected in the second and lower layers through the via hole 19 . In this way, by arranging the receiving coil pattern unicursally and symmetrically on the positive electrode side and the negative electrode side, it is possible to prevent noise from being mixed in the output signal and to receive a highly accurate signal.
In FIG. 4, the excitation coil patterns 16, 18 and 20 are schematically illustrated so as to surround the second receiving coil pattern 27, and the excitation coil patterns 16, 18 and 20 have an oscillator 38 that generates a predetermined high frequency current. is connected to apply

Further, each amplitude (length in the radial direction) of the zigzag-shaped divided detection portion in the second sin receiving coil pattern 32 will be described.
First, one circumference of 360° is divided by the number of patterns, 33, to calculate the mechanical angles in the 33 divided detection units. The calculated mechanical angle is multiplied by the number of pole pairs of 32 of the second sensor pattern on the rotor-side printed circuit board, and divided by 360° for one rotation to calculate the electrical angle at each split detector. Then, the sine values of these electrical angles are arranged in order.

As shown in FIG. 6, the second receiver coil pattern 27 is similar to the first receiver coil pattern 25 in that the second sin receiver coil pattern 32 formed on the first layer (the facing surface of the rotor-side substrate) It has a second cos receiving coil pattern 33 formed in two layers. In order to reliably detect the magnetic flux in the second cosine receiving coil pattern 32 in the second layer, the thickness of the first layer should be made thinner than usual.
Note that FIG. 6 is a diagram showing a simplified configuration for showing the arrangement relationship of the receiving coil patterns, and actually has a laminated structure in which the first layer and the second layer are in complete contact.

The second cosine reception coil pattern 33 has 33 patterns of zigzag-shaped split detectors that are out of phase with the second sinus reception coil pattern 32 by π/2. As for the magnitude of the amplitude of the zigzag-shaped divided detection section, first, one circumference of 360° is divided by the number of patterns of 33 to calculate the mechanical angle at the 33 divided detection sections. The electrical angle at each divided detection section is calculated by dividing the result obtained by multiplying the number of pole pairs of the second sensor pattern by 32 by 360° for one rotation. It is formed by arranging the cos values of the electrical angles in order.

As described above, the receiver coil pattern in this embodiment includes the first sin receiver coil pattern 28 and the second sin receiver coil pattern 32 formed on the first layer, and the first cos receiver coil pattern 29 and the second cos receiver coil pattern 33 formed on the first layer. It is formed in the second layer.
By adopting such a configuration, since the sine receiving coil pattern and the cosine receiving coil pattern in each receiving coil pattern are formed on the same plane, the number of via holes at the intersection of the mutual patterns can be reduced to improve reliability. It is possible to achieve an improvement in performance due to an improvement in resistance and a reduction in electrical resistance.

The radial length of the first sensor pattern 22 of the rotor printed circuit board 14 (the length in the direction perpendicular to the direction of rotation: x in FIG. 3) is It is formed so as to be longer than the sum of the amplitudes in the positive and negative directions of the maximum amplitude of each pattern (that is, twice the maximum amplitude of each pattern: x' in FIG. 2).
Similarly, the length in the radial direction of the second sensor pattern 24 of the rotor printed circuit board 14 (the length in the direction perpendicular to the direction of rotation: y in FIG. 3) is the second sin receiver coil pattern 32 of the stator printed circuit board 12 It is formed to be longer than the sum of the amplitudes in the positive direction and the negative direction at the maximum amplitude of each pattern (that is, the length twice the maximum amplitude of each pattern: y' in FIG. 2) .

In this way, each sensor pattern on the rotor printed circuit board 14 is made longer than the sum of the positive and negative patterns of the maximum amplitude of each pattern of the receiving coil patterns on the corresponding stator printed circuit board 12. Since the receiving coil pattern can reliably receive the vibrations of 14 rotations, it is possible to accurately detect the position and angle.

(About redundant system)
The angle detector of this embodiment has a vernier redundant system.
Specifically, 31 sensor patterns of the first sensor pattern 22 and 32 sensor patterns of the second sensor pattern 24 are provided on the rotor side, and the first receiving coil pattern and the second receiving coil pattern on the stator side are provided. The coil pattern has a zigzag-shaped divided detection portion that is divided into 33 parts.
Magnetic flux from the first sensor pattern 22 is received by the first receiving coil pattern 25 , and magnetic flux from the second sensor pattern 24 is received by the second receiving coil pattern 27 . First, in this way, a redundant system can be ensured by receiving respective magnetic fluxes with the two receiving coil patterns 25 and 27 .

Also, as described above, by changing the number of divisions of the first sensor pattern 22 and the first receiving coil pattern 25 and changing the number of divisions of the second sensor pattern 24 and the second receiving coil pattern 27, angle differences are created, I'm trying to improve resolution. This is the meaning of the vernier type in this embodiment.
In the first place, a vernier is a vernier scale formed on a vernier caliper or the like, and is used to subdivide and read a reference scale.

In the angle detector of the present embodiment, since magnetic flux detection is multipolarized, for example, one rotation generates an output cycle corresponding to the shaft angle multiple, and mechanical angle x shaft multiple angle = electrical angle. For example, when the rotor-side sensor pattern pole pair number (number) is 31 (31X), the electrical angle=mechanical angle×31. In the present embodiment, both cases of 31 and 32 sensor pattern pole pairs on the rotor side are adopted.
Then, with respect to the number of sensor patterns on the rotor side of the whole circumference M (MX), the number of patterns (the number of zigzag folds, the number of divided detection units) N of the zigzag receiving coil pattern (sin or cos) on the stator side is set to M + 1, By setting the number of patterns to an integral multiple of a prime number, the angle 360/Mi (i=1, 2, . . . M) per sensor pattern number on the rotor side and the angle 360/(Mi+1) per zigzag number on the stator side output a pattern of amplitude values proportional to sin (sine) and cos (cosine) of the angle difference between .

The sine output signal and cosine output signal output from the first receiving coil pattern and the sine output signal and cosine output signal output from the second receiving coil pattern of the printed circuit board type angle detector of the present embodiment are After being input to 30 and amplified, it is input from the resolver digital converter to the MPU and the angular position is calculated.

FIG. 7 shows the mechanical angle and the output curve of a pattern divided into 33 equal parts around the circumference when the number of pole pairs is 31. As shown in FIG. In this figure, the vertical axis represents the output voltage, and the vertical axis is set to 1 when the actual output of 30 mV is amplified 100 times to 3V.

It should be noted that the envelope of the amplitude of each divided detection section of each receiving coil pattern in the above-described embodiments is not limited to a sine wave (sin) or cosine wave.

Although an example of a printed circuit board type angle detector has been described in the above-described embodiments, the present invention is not limited to a printed circuit board for both the mover flat plate and the stator flat plate. The mover flat plate may be, for example, a simple metal plate or the like, and the stator flat plate may be a ceramic substrate or the like.
Further, in the above-described embodiments, both the mover flat plate and the stator flat plate are circular when viewed from the front. Although at least the armature plates are preferably circular for rotation, the stator plates may be non-circular, for example square in order to fix the stator plates in the motor.

In the above-described embodiments, a so-called rotary encoder for detecting an angle when the mover plate rotates with respect to the stator plate has been described. It is also possible to employ a linear encoder that detects the position when it is moved in parallel with respect to.

10 Angle detector 12 Stator printed circuit board 14 Rotor printed circuit board 16 Outermost excitation coil pattern 18 Intermediate excitation coil pattern 19 Via hole 20 Outermost excitation coil pattern 22 First sensor pattern 24 Second sensor pattern 25 First reception Coil pattern 27 Second receiver coil pattern 28 First sin receiver coil pattern 29 First cos receiver coil pattern 32 Second sin receiver coil pattern 33 Second cos receiver coil pattern

Claims (13)

a stator flat plate composed of multilayer plates;
a mover plate arranged at a predetermined interval with respect to the stator plate,
A first sensor pattern in which M pole pairs of sensor patterns are discretely and evenly arranged is formed on the facing surface of the mover flat plate,
An exciting coil pattern is formed on a surface of the stator plate facing the mover plate,
forming a first receiving coil pattern facing the first sensor pattern of the mover plate inside or outside the exciting coil pattern,
The first receiving coil pattern (A) is formed on either the surface or the inner layer of the stator flat plate, and is formed on the other of the surface or the inner layer of the stator flat plate. and a first receiving coil pattern (B),
The number of patterns N of the first receiving coil pattern (A) and the first receiving coil pattern (B) is such that the amplitude envelope is such that N≠M with respect to the number of pole pairs M of the first sensor pattern. , a flat plate encoder formed in a zigzag shape that is a periodic function having the same period and amplitude but out of phase. forming a second sensor pattern in which sensor patterns having a number of pole pairs of m are discretely and evenly arranged at a position different from the first sensor pattern formed on the mover flat plate;
forming a second receiving coil pattern facing the second sensor pattern of the mover plate inside or outside the excitation coil pattern of the stator plate,
The second receiving coil pattern (A) is formed on either the surface or the inner layer of the stator flat plate, and the second receiving coil pattern (A) is formed on the other of the surface or the inner layer of the stator flat plate. and a second receiving coil pattern (B),
The number of patterns n of the second receiving coil pattern (A) and the second receiving coil pattern (B) is such that the amplitude envelope curve is such that n≠m with respect to the number of pole pairs m of the second sensor pattern. 2. The flat-plate encoder according to claim 1, wherein the plate-type encoder is formed in a serpentine shape that is a periodic function whose period and amplitude are the same and whose phases are shifted. 3. The flat plate encoder according to claim 2, wherein the number of pole pairs M of the first sensor pattern and the number of pole pairs m of the second sensor pattern are M≠m. The first receiving coil pattern (A) and the first receiving coil pattern (B) are composed of a conductor pattern whose envelope is the periodic function and a reverse polarity conductor pattern whose polarity is opposite to the conductor pattern. is formed in a single stroke,
4. The planar encoder according to any one of claims 1 to 3, wherein both ends of said conductor pattern are formed so as to be able to be input to positive and negative input terminals of an operational amplifier. The second receiving coil pattern (A) and the second receiving coil pattern (B) are composed of a conductor pattern whose envelope is the periodic function and a reverse polarity conductor pattern whose polarity is opposite to the conductor pattern. is formed in a single stroke,
4. The flat-plate encoder according to claim 2, wherein both ends of said conductor pattern are formed so as to be able to be input to positive and negative input terminals of an operational amplifier. Among claims 1 to 5, the number of patterns N of the first receiving coil pattern is N=Np*k (where Np is an arbitrary prime number and k is a positive integer equal to or greater than 1). A flat plate encoder according to any one of the above. The number of patterns N of the first receiving coil patterns is N=Np*k (where Np is an arbitrary prime number and k is a positive integer equal to or greater than 1), and the number of patterns n of the second receiving coil patterns is n =np*k (where np is an arbitrary prime number and k is a positive integer equal to or greater than 1). In the first sensor pattern of the movable plate, the length in the direction orthogonal to the moving direction of the movable plate is
8. The length of each pattern of the first receiving coil pattern of the stator plate is longer than twice the maximum amplitude of each pattern. or the flat plate encoder according to item 1. In the second sensor pattern of the movable plate, the length in the direction orthogonal to the moving direction of the movable plate is
Claims 2, 3, and 5, wherein the length of each pattern of the second receiving coil pattern of the stator plate is longer than twice the maximum amplitude of each pattern. Or the planar encoder according to claim 7. The number of patterns N of the first receiving coil pattern (A) and the first receiving coil pattern (B) is N=M±1 or N=M±2 with respect to the pole pair number M of the first sensor pattern. 10. The planar encoder according to any one of claims 1 to 9, characterized in that it is formed as follows. The number of patterns n of the second receiving coil pattern (A) and the second receiving coil pattern (B) is n=m±1 or n=m±2 with respect to the pole pair number m of the second sensor pattern. 10. The planar encoder according to claim 2, 3, 5, 7 or 9, wherein the planar encoder is formed as follows. The flat plate encoder according to any one of claims 1 to 11, wherein the movable element flat plate has a circular shape when the facing surface is viewed from the front. The flat plate encoder according to any one of claims 1 to 12, wherein the stator plate is formed of a printed circuit board.

Images