JP5217679B2 - VR type resolver, motor and linear motion linear motor - Google Patents

Description

The present invention relates to a phase change according to a change in reluctance between a plurality of pole piece teeth formed on a pole piece provided on a stator and each tooth of a rotor having a dentition facing the pole piece tooth. The present invention relates to a VR resolver , a motor, and a linear motion linear motor that output a plurality of resolver signals having different values .

Conventionally, for example, a resolver, an encoder, or the like is used as a position detection sensor for detecting the rotation angle position of the rotor or the movement position of the mover. In particular, the resolver does not use precision parts or electronic parts in the sensor body, and therefore has better environmental resistance than an encoder or the like, and is often used for a long period of time.
For example, a VR (variable reluctance) type resolver is rotatably attached to a rotating shaft of an actuator such as a motor, and the reluctance between the rotor and the stator varies depending on the position of the rotor, and a voltage corresponding to the change. The resolver signal is output. A resolver is provided in the motor unit, and a multiphase output (for example, three phases) from the resolver is taken into a drive unit that drives and controls the motor. In the drive unit, the captured multi-phase output signal is converted into a two-phase output (SIN, COS) signal by a phase conversion circuit, and this analog signal is converted into a digital signal by an R / D converter (RDC). The angle data obtained by the conversion is corrected by the correction data to obtain a final digital position signal, and the actuator is controlled based on this.

  By the way, in the case of a VR type resolver having no windings in the rotor, the resolver signal corresponding to the position is detected using the change in gap permeance due to the tooth shape of the stator and the tooth shape of the rotor. Therefore, the harmonic component that is a noise component included in the resolver signal of each phase also greatly changes. The larger the harmonic component, the greater the influence on the position detection accuracy. Therefore, in order to improve the position detection accuracy, it is necessary to correct the error due to the harmonic component using correction data or the like.

As a technique for solving this problem, for example, there is a position detection device described in Patent Document 1.
This position detection device rotates the excitation winding and output winding with different numbers of poles in the stator core slot, the number of excitation winding pole pairs is p1, and the number of output winding pole pairs is p2. The child is an iron core having M salient poles and no winding is provided. In the structure, p1 + p2 = M or p1−p2 = ± M, the excitation winding is a single phase, and the output winding is two or three phases. In this case, two phases or three phases of a sine wave with a period of 1 / M of the entire circumference of the rotor is induced in the output winding. When the output winding has a single phase, the voltage induced in the output winding is a sinusoidal voltage whose phase changes by 2π when the rotor moves 1 / M of the entire circumference. To detect the position. Furthermore, when the space angle representing the position of the outer periphery of the rotor with the center of the salient pole as the origin is θ ₂ , the rotor has a rotor shape such that the gap permeance pulsation due to the rotor shape becomes cos (Mθ ₂ ), Or it is comprised in the rotor shape approximated to this.
JP-A-6-213614

However, in the prior art disclosed in Patent Document 1, as a configuration for reducing the harmonic component included in the induced voltage waveform, the rotor shape is changed to a rotor shape such that the gap permeance pulsation becomes cos (Mθ ₂ ). In addition, the excitation winding and the output winding are provided separately, and these two types of windings are wound around one pole, thus having a complicated winding structure.

Therefore, the present invention has been made paying attention to such an unsolved problem of the conventional technology, and is used to reduce the harmonic component contained in the detection signal with a relatively simple winding structure. It is an object of the present invention to provide a suitable VR resolver , motor, and linear motion linear motor .

  [Invention 1] In order to achieve the above object, the VR type resolver according to Invention 1 is fixedly supported by a plurality of pole pieces having a plurality of pole piece teeth at the front end portion, and each pole piece is fixedly supported. And an annular stator formed by winding a coil having N phase excitation and signal output in series for each phase, and formed in a circumferential direction so as to face the pole piece teeth of the stator with a gap. In a VR type resolver comprising an annular rotor that is arranged concentrically with a stator and is supported so as to be relatively rotatable, each pole piece tooth of the stator and each tooth of the rotor R of a magnitude that relatively reduces the harmonic component contained in the resolver signal output from the coil of each phase due to the gap permeance variation between the first and second poles at the tip of each pole piece tooth of the stator At both ends in the circumferential direction and at the tip of each tooth in the dentition of the rotor That provided in the circumferential direction of the both end portions.

  With such a configuration, variation in gap permeance can be reduced by R (roundness) provided at both ends in the circumferential direction at the tip of the stator pole piece and the rotor tooth. Compared to a configuration in which Rs are not provided at both ends in the circumferential direction at the tip of the pole piece of the child and the teeth of the rotor, the harmonic component contained in the resolver signal output from the coil of each phase is reduced. Can do. That is, the resolver signal can be approximated to an ideal waveform (SIN waveform) signal by providing R of an appropriate size at both ends of the tooth tip in the circumferential direction.

This eliminates the need for correction processing for the detection signal without changing the winding structure. Therefore, the cost can be reduced by reducing the number of correction processing circuits with a simpler configuration compared to the prior art, and space saving by circuit reduction. Alternatively, it is possible to obtain an effect such as a reduction in calculation load on the processor.
Further, since the structure is such that R is provided at the tip of both the pole piece teeth of the stator and the teeth of the rotor, for example, when there is a need to limit the size of one R, the other R By adjusting the size, it is possible to obtain a structure that can further reduce the harmonic components as compared with the case where R is provided only on one side with a limited size of R.

[Invention 2] Further, the VR resolver of Invention 2 is the VR resolver according to Invention 1, wherein the tooth width of the pole piece teeth of the stator and the tooth width of the teeth of the rotor are the same.
In such a configuration, compared with the case where the tooth width of the pole piece teeth of the stator is different from the tooth width of the rotor teeth, the stator pole piece teeth and the rotor teeth are more appropriate. The magnitude of R can be set easily (for example, it can be set to the same value).

  [Invention 3] Further, the VR resolver according to Invention 3 is the VR resolver according to Invention 2, wherein each of the pole pieces adjacent to each other in the circumferential width W1 of each pole piece tooth of the stator. The distance L1 between the pole piece teeth is configured to have a width and length of W1: L1 = 1: 2, and the depth H1 of the recess between the adjacent pole piece teeth is set to W1: H1 = 1: 2. In the dentition of the rotor, the circumferential width W2 of each tooth and the distance L2 between the adjacent teeth are set to a width and length that satisfy W2: L2 = 1: 2. The depth H2 of the recess between the adjacent teeth is set to a depth of W2: H2 = 1: 2.

  In such a configuration, W1 = W2, L1 = L2, and H1 = H2, and therefore, an error with respect to position detection accuracy with the size of R, as compared with the case where the size is not configured with these relationships. Therefore, it is possible to more easily set an appropriate R size for the pole piece teeth of the stator and the teeth of the rotor.

  [Invention 4] The VR resolver according to Invention 4 is the VR resolver according to Invention 3, wherein the R size R1 and W1 provided at the both ends of the pole piece teeth of the stator are Is divided within the range of “13 [%] ± 2 [%]”, and the R size R2 and the W2 provided at the both ends of the teeth of the rotor, The R2 was divided by W2 so that the result was within the range of “13 [%] ± 2 [%]”.

With such a configuration, the variation in gap permeance can be reduced more reliably, and the harmonic component contained in the resolver signal output from the coil of each phase can be more reliably reduced.
The above “13 [%] ± 2 [%]” is a numerical value based on the knowledge obtained by the present inventors through experiments, and R1 that satisfies R1 / W1 = 13 [%] and R2 / W2 = 13 [%]. , R2, W1, and W2, harmonic components can be made substantially zero. Further, by setting the value within the range of 13 [%] ± 2 [%], the error with respect to the position detection accuracy can be suppressed within ± 20 [seconds].

  [Invention 5] Further, the VR resolver according to Invention 5 is the VR resolver according to Invention 3, in which R size R1 and W1 provided at both ends of the pole piece teeth of the stator and the rotor An average of a result obtained by dividing R1 by W1 and a result obtained by dividing R2 by W2 with respect to R size R2 and W2 provided at both ends of the tooth is “13 [%] ± 2 [% ] ”In the range.

With such a configuration, the variation in gap permeance can be reduced more reliably, and the harmonic component contained in the resolver signal output from the coil of each phase can be more reliably reduced.
The above “13 [%] ± 2 [%]” is a numerical value based on the knowledge obtained by the present inventors through experiments, and R1 in which the average value of R1 / W1 and R2 / W2 is 13 [%], By setting R2, W1, and W2, the harmonic component can be substantially “0”. Further, by setting the value within the range of 13 [%] ± 2 [%], the error with respect to the position detection accuracy can be suppressed within ± 20 [seconds].

  [Invention 6] In order to achieve the above object, the VR type resolver of Invention 6 is fixedly supported by having a plurality of pole pieces having a plurality of pole piece teeth at the front end portion, and being fixedly supported. An annular stator in which a coil having both N-phase excitation and signal output is wound in series for each phase and a pole piece is formed in a circumferential direction facing the pole piece teeth of this stator. A VR resolver comprising an annular rotor that is arranged concentrically with the stator and is supported so as to be relatively rotatable, the tooth width in the circumferential direction of the pole piece of the stator; The circumferential tooth width of the rotor teeth is configured to have the same width W, and the tooth width W of each pole piece tooth and the distance L1 between each pole piece tooth adjacent to each pole piece, The width and length of W: L1 = 1: 2, and the tooth width W of each tooth of the rotor and the distance L2 between the adjacent teeth are expressed as W: L2 = : Configured to have a width and length of 2 and included in the resolver signal output from the coil of each phase due to gap permeance variation between each pole piece tooth of the stator and each tooth of the rotor R of a magnitude that relatively reduces the harmonic component is set to the circumferential direction at both ends in the circumferential direction at the tip of each pole piece tooth of the stator and at the tip of each tooth in the dentition of the rotor. It was provided in either one of both ends.

  With such a configuration, variation in gap permeance can be reduced by R provided at both ends in the circumferential direction at the tip of one of the pole piece teeth of the stator and the teeth of the rotor, Compared with the configuration in which R is not provided at both ends in the circumferential direction at the tip of the stator pole piece teeth and the rotor teeth, the harmonic component contained in the resolver signal output from the coil of each phase is reduced. be able to. That is, even if R is not provided on both the pole piece teeth of the stator and the rotor teeth, the gap permeance variation can be reduced by providing it on either one.

[Invention 7] The VR resolver according to Invention 7 is the VR resolver according to Invention 6, wherein R is provided at both end portions of one of the pole piece teeth of the stator and the teeth of the rotor. The result of dividing the size of R by the circumferential width W of one of the stator pole teeth and the rotor teeth is “13 [%] ± 2 [%]”. The size was within the range.
With such a configuration, the variation in gap permeance can be reduced more reliably, and the harmonic component contained in the resolver signal output from the coil of each phase can be more reliably reduced.

  The above-mentioned “13 [%] ± 2 [%]” is a numerical value based on knowledge obtained by the present inventors through experiments, and R1 that satisfies R1 / W1 = 13 [%] or R2 / W2 = 13 [%]. And W1 or R2 and W2, the harmonic component can be made substantially zero. Further, by setting the value within the range of 13 [%] ± 2 [%], the error with respect to the position detection accuracy can be suppressed within ± 20 [seconds].

[Invention 8] The VR resolver according to Invention 8 is the VR resolver according to any one of Inventions 1 to 7, wherein the stator has any one of 3, 4 and 6 as the phase number N. A coil having the number of phases is wound in series for each phase.
With such a configuration, it is possible to obtain the same operations and effects as the VR resolvers of the first to seventh inventions with respect to the three-phase, four-phase, or six-phase VR resolvers.

  [Invention 9] The VR resolver according to Invention 9 is the VR resolver according to any one of Inventions 1 to 8, wherein the stator is configured with an even number of phases with the number of phases N being 4 or more. A differential signal generating means for generating a differential signal of the resolver signal for each two-phase combination in the N-phase resolver signal output from the coil of each phase when the differential signal generating means Based on the generated two sets of differential signals, a two-phase signal of a SIN signal and a COS signal is generated.

  With such a configuration, it is possible to generate a differential signal of each two resolver signals output from a set of two-phase coils of even-numbered phases of four or more phases, and based on the differential signals, SIN A two-phase signal of a signal and a COS signal can be generated. As a result, a part of the harmonic components (even harmonic components) included in the resolver signal can be removed. Therefore, for example, R increases due to wear of a die for manufacturing the stator and the rotor. When the height changes, the influence of the harmonic component on the position detection accuracy can be reduced.

[Invention 10] On the other hand, in order to achieve the above object, the VR type resolver of Invention 10 has a plurality of pole pieces having a plurality of pole piece teeth at the tip portion and is fixedly supported in a straight line. Each pole piece is formed by winding a coil having both N-phase excitation and signal output in series for each phase, and facing the pole piece teeth of this stator with a gap. In a VR resolver comprising a stator and a linear mover supported in a relatively movable manner, each pole piece tooth of the stator and each tooth of the linear mover R of a magnitude that relatively reduces the harmonic component included in the resolver signal output from the coil of each phase due to the gap permeance variation between the first and second poles at the tip of each pole piece tooth of the stator and opposite end portions, each tooth in the teeth of the linear mover Provided on the both end portions of the linear direction at the tip.

With such a configuration, even in a linear VR resolver constituted by a linear stator and a linear movable element , the same operation and effect as those of the first aspect can be obtained.

  As described above, according to the VR resolver of the first aspect, both circumferential ends of the tip of the pole piece tooth of the stator and the circumference of the tip of each tooth in the dentition of the rotor Since R of a magnitude that can relatively reduce the harmonic component due to the gap permeance is provided at both ends of the direction, the harmonic component contained in the resolver signal of each phase is compared with the case where R is not provided. The effect that it can reduce is acquired.

Furthermore, according to the VR type resolver of the invention 2, in addition to the effect of the invention 1, the tooth width of the pole piece tooth of the stator and the tooth width of the tooth of the rotor are the same or substantially the same width. As compared with the case where the tooth widths are different, an effect that an appropriate size of R can be easily set is obtained.
Furthermore, according to the VR type resolver of the invention 3, in addition to the effect of the invention 2, the tooth width W1 of each pole piece tooth and the distance L1 between the pole piece teeth of each pole piece are expressed as W1: L1. = 1: 2 width and length, and the tooth width W2 (= W1) and inter-tooth distance L2 (= L1) of the teeth of the rotor are W2: L2 = 1: 2. The depth H1 of the recess between the pole piece teeth is set to a depth of W1: H1 = 1: 2, and the depth H2 (= H1) of the recess between the teeth is set to W2: H2 = Since the depth is set to 1: 2, the gap permeance variation can be made to depend only on the shape of the tooth tip (the size of R), so that the appropriate size of R can be set more easily. The effect of being able to be obtained.

Furthermore, according to the VR resolver of the invention 4, in addition to the above-described effect of the invention 3, the R size R1 and W1 provided at the tip of the pole piece teeth of the stator are set to R1 / W1 = 13 [%] ± The size and width are in the range of 2 [%], and the R size R2 and W2 provided at the tip of the rotor teeth are within the range of R2 / W2 = 13 [%] ± 2 [%]. Therefore, an effect that the error with respect to the position detection accuracy can be suppressed within 20 [seconds] can be obtained.
Further, according to the VR resolver of the invention 5, in addition to the above-mentioned effects of the invention 3, the average of R1 / W1 and R2 / W2 is 13 [%] ± 2 [%] for R1, R2, W1, and W2. Since the size and width are within the range, an effect that the error with respect to the position detection accuracy can be suppressed within 20 [seconds] can be obtained.

  Furthermore, according to the VR type resolver of the invention 6, both ends in the circumferential direction of the tip of the pole piece tooth of the stator and both ends in the circumferential direction of the tip of each tooth in the dentition of the rotor Since R of a magnitude that can relatively reduce the harmonic component due to the gap permeance is provided in either of them, the harmonic component contained in the resolver signal of each phase is compared with the case where R is not provided. The effect that it can reduce is acquired.

  Further, according to the VR resolver of the invention 7, in addition to the effect of the invention 6, the magnitudes R1 and W1 of R provided at the tip of the pole piece teeth of the stator are set to R1 / W1 = 13 [%] R2 and W2 = 13 [%] ± 2 [%] R2 and W2 are set to the size and width of ± 2 [%] or the R size R2 and W2 provided at the tip of the rotor teeth. Therefore, an effect that the error with respect to the position detection accuracy can be suppressed within 20 [seconds] can be obtained.

  Furthermore, according to the VR resolver of the ninth aspect, a differential signal of each two resolver signals is generated for each two-phase set of four or more even phases, and a SIN signal and a COS are generated based on the generated differential signals. Since a two-phase signal is generated, a part of the harmonic component included in the resolver signal can be removed, so that R increases due to wear of a mold for manufacturing the stator and the rotor. Even if a change occurs, the effect of the change on the position detection accuracy can be reduced.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. 1 to 13 are views showing an embodiment of a VR type resolver according to the present invention.
First, the structure of the three-phase VR resolver according to the present invention will be described with reference to FIG. FIG. 1 is a plan view showing the structure of a three-phase VR resolver according to the present invention.
As shown in FIG. 1, the VR type resolver according to the present embodiment is an inner rotor type in which an annular resolver rotor 20 is combined inside an annular resolver stator 10. A plurality of pole pieces (poles) 12 having pole piece teeth 11 are provided at equal intervals in the circumferential direction, and coils 13 are wound around each pole piece 12 and fixedly supported. The rotor 20 has a large number of teeth 21 formed at the same pitch as the pole piece teeth in the circumferential direction so as to face the pole piece teeth 11 of the stator 10, and is arranged concentrically with the stator 10 and supported in a relatively rotatable manner. ing. Each pole piece tooth 11 of the resolver stator 10 is an integral multiple of the pitch of the teeth 21 of the resolver rotor 20 so that the phases of the electrically adjacent pole pieces 12 of the resolver stator 10 are 120 [°]. 1/3 pitch.

The coil windings 13 wound around the pole pieces 12 of the resolver stator 10 are respectively connected to the three-phase alternating current A phase, B phase, and C phase with 120 [°] phases being different.
Now, when the resolver rotor 20 rotates synchronously with a motor rotor (not shown), the reluctance in the gap (air gap) between the teeth 21 of the resolver rotor 20 and the pole piece teeth 11 of the resolver stator 10 changes the position of the resolver rotor 20. The current corresponding to the change flows through the coil winding 13 of the resolver stator 10. The rotational angle position or rotational speed is detected by detecting the current as a modulation signal. When the resolver rotor 20 rotates by one tooth, a modulation signal corresponding to an electrical angle of 360 [°] is detected on the resolver stator 10 side.

  For example, when the number of teeth of the resolver rotor 20 is 150, 150 cycles of the A, B, and C phase signals can be obtained in one rotation. The obtained three-phase signal is converted into a two-phase signal by an electric circuit, and a COS signal and a SIN signal are input to an arithmetic unit composed of an RDC or the like to obtain a position signal, and this is fed back to the motor. The rotation speed and rotation angle (position) can be accurately controlled.

FIG. 1 is a diagram showing an example of a three-phase VR resolver. In the case of four phases, each pole piece of the resolver stator 10 is set so that the respective phases of the pole pieces 12 are 90 [°]. The teeth 11 are shifted by 1/4 pitch from an integral multiple of the pitch of the teeth 21 of the resolver rotor 20. Similarly, in the case of 6 phases, the phase is shifted by 1/6 pitch so that the phase becomes 60 [°].
Here, FIG. 2A is a waveform diagram showing an ideal detection signal (ignoring the excitation signal) and a detection signal indicating a harmonic component, and FIG. 2B is a harmonic in the waveform of FIG. It is a wave form diagram which shows only a component. 2A and 2B, the horizontal axis indicates the electrical angle, and the vertical axis indicates the signal level.

  Ideally, the signal detected by the resolver should have an ideal SIN waveform as shown by the solid line waveform in FIG. However, since the VR resolver detects a change in reluctance in the air gap between the tooth 21 and the pole piece tooth 11 as a signal, the air gap changes according to the shape of the tooth 21 and the pole piece tooth 11. With this change, the gap permeance (permeance coefficient) of the coil winding 13 changes. When this gap permeance change is large, a large amount of harmonic components are included in the detection signal. Therefore, the detected signal deviates from the ideal waveform as shown by the dotted line waveform in FIG. Waveform. That is, due to the change in gap permeance, the ideal SIN signal includes the harmonic component shown in FIG. Since the deviation of the detection signal from the ideal waveform becomes an error as it is and the position detection accuracy is deteriorated, it is desirable that the gap permeance change be as small as possible.

On the other hand, the change in gap permeance is greatly related to the shapes of the teeth 21 and the pole piece teeth 11 facing each other across the air gap. Therefore, the gap permeance change is minimized or the detection error is ignored. By configuring the shape within a possible range, harmonic components can be canceled or reduced.
Here, FIG. 3A is a diagram illustrating the configuration of the pole piece teeth 11, and FIG. 3B is a diagram illustrating the configuration of the teeth 21. FIGS. 4A and 4B are side views showing a configuration when R is provided only on the pole piece teeth 11 and only the teeth 21. FIG. 5A is a diagram showing the relationship between R1 / W1 and the magnitude of the harmonic component when R is provided only on the pole piece tooth 11, and FIG. It is a figure which shows the relationship between R2 / W2 when providing and the magnitude | size of a harmonic component.

  In this embodiment, in order to cancel or reduce the harmonic component contained in the detection signal, first, the pole piece teeth 11 in each pole piece 12 of the resolver stator 10 and the teeth 21 of the resolver rotor 20 are provided. Were made the same size. Further, R is provided at both ends in the circumferential direction of the tip portion of the pole piece tooth 11, and R is provided at both ends in the circumferential direction of the tip portion of the tooth 21.

Here, as shown in FIG. 3A, the tooth width of the pole piece teeth 11 is W1, the length between the pole piece teeth 11 in the pole piece 12 (hereinafter referred to as interdental distance) is L1, and the pole piece teeth. The depth of the recess between 11 is H1. Furthermore, as shown in FIG. 3 (b), the tooth width of the tooth 21 is W2, the length between the teeth 21 (hereinafter referred to as the interdental distance) is L2, and the depth of the recess between the teeth 21 is H2. To do.
In the present embodiment, specifically, first, the pole piece teeth 11 and the teeth 21 are configured such that the tooth width W1 and the tooth width W2 are the same width (W1 = W2). Further, the pole piece tooth 11 is configured such that the tooth width W1 and the inter-tooth distance L1 are such that W1: L1 = 1: 2 and the depth H1 is the tooth width W1 and the depth H1. Is formed to a depth where W1: H1 = 1: 2.

On the other hand, the tooth 21 is configured such that the tooth width W2 and the inter-tooth distance L2 have a width and length such that W2: L2 = 1: 2, and the depth H2 is defined by the tooth width W1 and the depth H2. , W1: H2 = 1: 2.
3 (a) and 3 (b), R (roundness) of size R1 is provided at both ends in the circumferential direction at the tip of the pole piece tooth 11, and the circumference at the tip of the tooth 21 is provided. R of size R2 is provided at both ends in the direction.

As a method of providing R at the tip, generally, in mass production, a thin plate such as a resolver stator or resolver rotor constituting a resolver is produced using a punching die. In the case of this method, a die is manufactured in advance after determining R1 and R2 having a size appropriate for canceling or reducing the harmonic component contained in the detection signal.
Here, as shown in FIG. 4 (a), when R of magnitude R1 is provided only on the tip of the pole piece tooth 11 on the resolver stator 10, the harmonic component in the detection signal and “R1 / W1”. "[%]" Is as shown in FIG. In FIG. 9A, the tooth width W1 is set to about 0.78 [mm], the R size R1 of the tip of the pole piece tooth 11 is changed, and “R1 / W1” is 8 to 20 [%]. ] Was measured. As can be seen from FIG. 9A, the harmonic component increases at 13 [%] (R1≈0.1) as it becomes smaller or larger. That is, the harmonic component cannot be made zero even if “R1 / W1” is too large or too small.

  On the other hand, as shown in FIG. 4B, when R of the size R2 is provided only at the tooth tip of the tooth 21 on the resolver rotor 20 side, the harmonic component in the detection signal and “R2 / W2” [% ] Is as shown in FIG. In FIG. 5B, the tooth width W2 is set to about 0.78 [mm], the magnitude R2 of R is changed, and the harmonic component when “R2 / W2” is 8 to 20 [%] is shown. It was measured. As can be seen from FIG. 5B, the harmonic component increases at 13 [%] (R2≈0.1) as it becomes smaller or larger. That is, the harmonic component cannot be made zero even if “R2 / W2” is too large or too small.

  From the above, it can be seen that the values of “R1 / W1” and “R2 / W2”, that is, the magnitudes of R1 and R2, affect the harmonic components in the detection signal. This is because the SIN wave excitation signal is supplied to the resolver and the gap permeance between the pole piece teeth 11 and the teeth 21 changes depending on the magnitude of R. In other words, by providing R with a magnitude of “R1 / W1 = R2 / W2” of 13 [%] on either one of the tip of the pole piece tooth 11 and the tooth 21, the VR resolver is made to have gap permeance. It can be configured such that the change in the frequency just cancels out the harmonic component.

In addition, as shown in FIG. 6, it is also possible to provide R that satisfies “R1 / W1 = R2 / W2 = RA / WA (RA = R1 = R2)” on both the tip of the pole piece tooth 11 and the tooth 21. Good.
Here, FIG. 6 is a side view showing a configuration when R is provided on both the pole piece teeth 11 and the tooth tips of the teeth 21. FIG. 7 is a view showing the relationship between RA / WA and detection error when R is provided on both the pole piece teeth 11 and the tooth tips of the teeth 21.

  When both the pole piece teeth 11 and the tooth tips of the teeth 21 are provided with R, the relationship between the error component in the detection signal and “RA / WA” is as shown in FIG. In this figure, when the tooth widths W1 and W2 are about 0.78 [mm] and the sizes of R1 and R2 (R1 = R2) are changed and “RA / WA” is 8 to 20 [%] The harmonic component of was measured. In the same figure, as in the case where R is provided only on one of the tip of the pole piece tooth 11 and the tooth 21, even if it becomes smaller than 13 [%] (RA≈0.1) as a boundary, It can be seen that the harmonic component is increasing even if it increases. That is, the harmonic component cannot be made zero even if “RA / WA” is too large or too small.

Accordingly, the sizes R1 and R2 (R1 = R2≈0.0) are applied to both the pole piece teeth 11 and the tooth tips of the teeth 21 so that both “R1 / W1” and “R2 / W2” are 13%. It can be seen that the detection error can be made substantially zero by providing R of 1).
Hereinafter, considering the application of the VR resolver of the present embodiment to a direct drive motor (hereinafter referred to as a DD motor), the size of R provided at the tooth tip is set.

In general, the position detection accuracy of a DD motor is put into practical use in 90 seconds (± 45 seconds). As shown in FIG. 6, when RA / WA is in the range of 13 ± 2 [%], the detection error is within ± 20 seconds, and the position detection accuracy sufficiently satisfactory for the detection error of ± 45 seconds can be obtained. A VR type resolver can be manufactured.
Therefore, in the present embodiment, at least one of the pole piece teeth 11 and the tooth tips of the teeth 21 so that at least one of “R1 / W1” and “R2 / W2” is within a range of 13 ± 2 [%]. R is provided with R of size R1 or R2.

  Next, based on FIG. 8 to FIG. 13, at least one of R1 / W1 and R2 / W2 is within a range of 13 ± 2 [%] on at least one of the tip of the pole piece tooth 11 and the tooth 21. Specific actions and effects when an N-phase resolver having an R of size R1 or R2 (R1 = R2 when both are provided) are applied to various resolver control circuits will be described. Note that the N-phase resolver is attached to a rotating shaft of a DD motor (not shown).

  Here, FIG. 8, FIG. 9 and FIG. 10 are diagrams showing an example of a resolver control circuit when the above configuration is applied to a three-phase, four-phase and six-phase resolver. FIG. 11 is a diagram illustrating an example of a resolver control circuit when the above configuration is applied to a configuration in which a three-phase ABS resolver and a three-phase INC resolver are combined. FIG. 12 is a diagram illustrating an example of a resolver control circuit when the above configuration is applied to a four-phase resolver and an angle is detected without using an RDC. FIG. 13 is a diagram illustrating the relationship between the electrical angle after the angle calculation process and the angle calculation value in the calculation unit.

First, the operation and effect on the resolver control circuit when using the resolver device to which the above configuration is applied to the pole piece teeth 11 and the teeth 21 of the three-phase resolver having the A phase, the B phase, and the C phase will be described.
As shown in FIG. 8, the resolver control circuit includes an excitation circuit 40 including an oscillator 40a and an amplifier (Amp) 40b, an A-phase detection resistor Ra corresponding to each phase A to C, and _a B-phase detection resistor. A current / voltage conversion circuit 41 composed of R _b and a C-phase detection resistor R _c , a 3/2 conversion circuit 42, a calculation unit 43, and a phase shifter 44 are configured.

  Then, an excitation signal sin (ωt) is supplied from the excitation circuit 40 to a three-phase resolver. On the other hand, when the rotating shaft of the DD motor is rotated, the reluctance in the air gap between the tooth pieces of the pole piece teeth 11 and the teeth 21 facing each other of the three-phase resolver is changed. An analog resolver signal corresponding to the output is output. The analog three-phase resolver signal is converted from a three-phase signal current into a three-phase signal voltage in the current / voltage conversion circuit 41 and output to the 3/2 conversion circuit 42. In the 3/2 conversion circuit 42, the three-phase resolver signal is converted into a two-phase signal of the SIN signal and the COS signal and output to the calculation unit 43.

  The calculation unit 43 includes an RDC, an ASIC (Application Specific Integrated Circuit), and the like. In the RDC, the analog two-phase signal (SIN, COS) from the 3/2 conversion circuit is converted into a digital angle signal. The angle is converted to φ and output to the ASIC, and the rotation angle position of the motor is calculated from the digital angle signal φ in the ASIC. This rotation angle signal is output to a host control circuit that controls the motor.

The phase shifter 44 generates a Ref signal (sin (ωt)) that is synchronized with the phase of the carrier signal of the two-phase signal by delaying the phase of the excitation signal output from the oscillator 40a. Generates an analog speed signal after synchronous rectification based on the oscillation angular frequency of the oscillator 40a. This speed signal is also output to the upper control circuit.
Here, if the detection signals of each phase of the three-phase resolver are A (θ), B (θ), and C (θ), each detection signal considering the harmonic components up to the fifth order is expressed by the following equation (1) ) To (3).

A (θ) = Adc ₀ + Aac ₁ cosθ + Aac ₂ cos2θ + Aac ₃ cos3θ + Aac ₄ cos4θ + Aac ₅ cos5θ
... (1)
B (θ) = Bdc ₀ + Bac ₁ cos (θ-120 °) + Bac ₂ cos2 (θ-120 °) + Bac ₃ cos3 (θ-120 °) + Bac ₄ cos4 (θ-120 °) + Bac ₅ cos5 (θ-120 °)
... (2)
C (θ) = Cdc ₀ + Cac ₁ cos (θ + 120 °) + Cac ₂ cos2 (θ + 120 °) + Cac ₃ cos3 (θ + 120 °) + Cac ₄ cos4 (θ + 120 °) + Cac ₅ cos5 (θ + 120 °)
... (3)

In the above formula (1) ~ (3), Adc 0, Bdc 0, Cdc 0 is, A-phase, B-phase, a direct current component included in the detection signal of the C _{phase, Aac 1 cosθ~Aac 5 cos5θ, Bac} 1 cos (θ-120 °) to Bac ₅ cos5 (θ-120 °), Cac ₁ cos (θ + 120 °) to Cac ₅ cos5 (θ + 120 °) are detection signals for A phase, B phase and C phase Are the first to fifth order AC components.

Here, for convenience of explanation, the values of Adc _{0 to} Cdc ₀ , Aac _{1 to} Cac ₁ , Aac _{2 to} Cac ₂ , Aac _{3 to} Cac ₃ , Aac _{4 to} Cac ₄ , and Aac _{5 to} Cac ₅ are the same order. Each shall be equal.
Therefore, when A (θ), B (θ), and C (θ) shown in the above equations (1) to (3) are converted into two-phase signals fc (θ) and fs (θ), It can be represented by the following formulas (4) and (5).

fc (θ) = 3 (Aac ₁ cos θ + Aac ₂ cos 2θ + Aac ₄ cos 4θ) / 2 (4)
fs (θ) = 3 (Aac ₁ sinθ + Aac ₂ sin2θ + Aac ₄ sin4θ) / 2 (5)

From the above equations (4) and (5), second-order and fourth-order harmonic components remain in the detection signal converted into two phases.

Further, the arithmetic unit calculates a digital angle signal φ shown in the following equation (6) from the above equations (4) and (5).

φ = θ + Δθ (6)

In the above equation (6), Δθ is represented by the following equation (7).

Δθ = tan ^-1 [(Aac ₂ -Aac ₄ ) / sin3θ / {Aac ₁ + (Aac ₂ + Aac ₄ ) cos3θ}]
... (7)

Δθ shown in the above equation (7) is an error component caused by the second and fourth harmonic components. In particular, the second harmonic component is dominant in this error component.

Originally, correction processing is required to remove this error component. In the present embodiment, however, the tooth tip of the pole piece tooth 11 of the resolver stator 10 of the three-phase resolver and the tooth 21 of the resolver rotor 20 are used. Since at least one of the tooth tips of R1 / W1 and R2 / W2 is provided with R in which at least one of R1 / W1 and R2 / W2 is within a range of 13 ± 2 [%], an error within ± 20 seconds with respect to absolute accuracy is ignored. In the above equation (7), the error component due to the second and higher harmonics can be made substantially zero. As a result, the above equations (4) and (5) can be approximated to the following equations (8) and (9).

fc (θ) = 3 (Aac ₁ cosθ) / 2 (8)
fs (θ) = 3 (Aac ₁ sinθ) / 2 (9)

As a result, Δθ represented by the above equation (7) can be significantly reduced, so that it is not necessary to provide a memory, a circuit, or the like for correcting the error in the calculation unit 43, or the error in the calculation unit 43 can be reduced. There is no need to perform correction calculations. Accordingly, it is possible to reduce costs by eliminating the need to provide a memory, a circuit, or the like, or it is possible to omit correction processing computations, so that an effect such as a reduction in computation load can be obtained.

Next, based on FIG. 9, a resolver control circuit in the case of using a resolver device to which the above configuration is applied to the pole piece teeth 11 and teeth 21 of a four-phase resolver having an A phase, a B phase, a C phase, and a D phase. The effect | action and effect with respect to are demonstrated.
As described above, the four-phase resolver is configured such that each pole piece tooth 11 of the resolver stator 10 is an integer of the pitch of the teeth 21 of the resolver rotor 20 so that the respective phases of the pole pieces 12 are 90 [°]. The structure is formed by shifting from the double by 1/4 pitch. The four-phase resolver is the same as the three-phase resolver, because the new component is not added to the three-phase resolver, only the tooth pitch is changed. .

The resolver control circuit shown in FIG. 9 has a configuration in which two differential circuits 45a and 45b are added in place of the 3/2 conversion circuit 42 in the resolver control circuit of FIG. Further, due to the increase in the D phase, a D phase detection resistor R _d is added to the current / voltage conversion circuit 41.
Therefore, when the excitation signal sin (ωt) is supplied from the excitation circuit 40 to the four-phase resolver and the rotation shaft of the DD motor rotates, first, an analog resolver signal corresponding to the rotation position from each phase of the A to D phases. Is output. The analog four-phase resolver signal is converted from a four-phase signal current into a four-phase signal voltage by a current / voltage conversion circuit 41. Then, the differential circuit 45a generates these differential signals (SIN signals) from the B-phase and D-phase signals, and the differential circuit 45b generates these differential signals (SIN signals) from the A-phase and C-phase signals. COS signal) is generated. These SIN signal and COS signal are output to the arithmetic unit 43.

The operations of the arithmetic unit 43 and the phase shifter 44 are the same as those of the resolver control circuit shown in FIG.
Here, if the detection signals of each phase of the four-phase resolver are A (θ), B (θ), C (θ), and D (θ), each detection signal considering the harmonic components up to the fifth order is And can be represented by the following formulas (10) to (13).

A (θ) = Adc ₀ + Aac ₁ cosθ + Aac ₂ cos2θ + Aac ₃ cos3θ + Aac ₄ cos4θ + Aac ₅ cos5θ
... (10)
B (θ) = Bdc ₀ + Bac ₁ cos (θ + 90 °) + Bac ₂ cos2 (θ + 90 °) + Bac ₃ cos3 (θ + 90 °) + Bac ₄ cos4 (θ + 90 °) + Bac ₅ cos5 (θ + 90 °)
(11)
C (θ) = Cdc ₀ + Cac ₁ cos (θ + 180 °) + Cac ₂ cos2 (θ + 180 °) + Cac ₃ cos3 (θ + 180 °) + Cac ₄ cos4 (θ + 180 °) + Cac ₅ cos5 (θ + 180 °)
(12)
D (θ) = Ddc ₀ + Dac ₁ cos (θ + 270 °) + Dac ₂ cos2 (θ + 270 °) + Dac ₃ cos3 (θ + 270 °) + Dac ₄ cos4 (θ + 270 °) + Dac ₅ cos5 (θ + 270 °)
... (13)

In the above equation (10) ~ (13), Adc 0, Bdc 0, Cdc 0, Ddc 0 is, A-phase, B-phase, C phase, a direct current component included in the detection signal of the D phase, Aac ₁ cosθ~ Aac ₅ cos5θ, Bac ₁ cos (θ + 90 °) to Bac ₅ cos5 (θ + 90 °), Cac ₁ cos (θ + 180 °) to Cac ₅ cos5 (θ + 180 °), Dac ₁ cos (θ + 270 °) to Dac ₅ cos5 (θ + 270 °) are primary to fifth order AC components included in the detection signals of the A phase, the B phase, the C phase, and the D phase.

Here, for convenience of explanation, the values of Adc _{0 to} Ddc ₀ , Aac _{1 to} Dac ₁ , Aac _{2 to} Dac ₂ , Aac _{3 to} Dac ₃ , Aac _{4 to} Dac ₄ , Aac _{5 to} Dac ₅ are the same order. Each shall be equal.
Accordingly, A (θ) and B (θ), C (θ), and D (θ) shown in the above equations (10) to (13) are used to calculate A (θ) and C (θ) in the differential circuit 45a. When the differential signal of B (θ) and D (θ) is generated in the differential circuit 45b of the differential signal 45b, the two-phase signals fc (θ) and fs (θ) composed of this differential signal are Equations (14) and (15) are obtained.

fc (θ) = A (θ) -C (θ) = 2Aac ₁ cosθ + 2Aac ₃ cos3θ + 2Aac ₅ cos5θ
(14)
fs (θ) = B (θ) -D (θ) = 2Aac ₁ cos (θ-180 °) + 2Aac ₃ cos3 (θ-180 °) + 2Aac ₅ cos5 (θ-180 °)
... (15)

From the above equations (14) and (15), third-order and fifth-order harmonic components remain in the detection signal converted into two phases. Therefore, when the error calculated by the angle is converted into a waveform, an error waveform in which the third order is dominant is obtained.

In the present embodiment, correction processing is necessary to remove this error component. In the present embodiment, the tooth tip of the pole piece tooth 11 of the resolver stator 10 of the four-phase resolver and the tooth 21 of the resolver rotor 20 are used. Since at least one of the tooth tips of R1 / W1 and R2 / W2 is provided with R in which at least one of R1 / W1 and R2 / W2 is within a range of 13 ± 2 [%], an error within ± 20 seconds with respect to absolute accuracy is ignored. In the above equations (14) and (15), the error component due to the third and higher harmonics can be made substantially zero. As a result, the above equations (14) and (15) can be approximated to the following equations (16) and (17).

fc (θ) = 2Aac ₁ cosθ (16)
fs (θ) = 2Aac ₁ cos (θ-180 °) (17)

As a result, the harmonic component that causes the error can be greatly reduced, so that it is not necessary to provide a memory or a circuit for correcting the error in the calculation unit 43, or the error correction is performed in the calculation unit 43. There is no need to perform an operation. Therefore, even for a four-phase resolver, it is possible to reduce costs by eliminating the need to provide a memory, a circuit, or the like, or it is possible to omit correction processing computations, so that the effect of reducing the computation load can be obtained.

Next, based on FIG. 10, a resolver device in which the above configuration is applied to the pole piece teeth 11 and teeth 21 of a 6-phase resolver having an A phase, a B phase, a C phase, a D phase, an E phase, and an F phase is used. The operation and effect on the resolver control circuit in the case of being present will be described.
As described above, in the six-phase resolver, each pole piece tooth 11 of the resolver stator 10 is set to an integer of the pitch of the teeth 21 of the resolver rotor 20 so that the respective phases of the pole pieces 12 are 60 [°]. The structure is formed by shifting the pitch by 1/6 pitch. The 6-phase resolver is the same as the 3-phase resolver in that only the tooth pitch is changed and a new component is not added. .

The resolver control circuit shown in FIG. 10 adds a differential circuit 45c to the resolver control circuit shown in FIG. 9 and also adds an E phase and an F phase, thereby adding an E phase detection resistor R to the current / voltage conversion circuit 41. _e and an F-phase detection resistor R _f are added. Further, a 3/2 conversion circuit 42 is added to convert three signals from the three differential circuits 45a to 45c into two-phase signals.

  Therefore, when the excitation signal sin (ωt) is supplied from the excitation circuit 40 to the 6-phase resolver and the rotation shaft of the DD motor rotates, an analog resolver signal corresponding to the rotational position is generated from each phase of the A phase to the F phase. Is output. The 6-phase analog resolver signal is converted from a 6-phase signal current to a 6-phase signal voltage in the current / voltage conversion circuit 41. The differential circuit 45a generates these differential signals from the F-phase and C-phase signals, and the differential circuit 45b generates these differential signals from the E-phase and B-phase signals. In the dynamic circuit 45c, these differential signals are generated from the D-phase and A-phase signals. These three signals are output to the 3/2 conversion circuit 42, where they are converted into analog two-phase signals (SIN, COS) and output to the arithmetic unit 43.

The operations of the arithmetic unit 43 and the phase shifter 44 are the same as those of the resolver control circuit shown in FIG.
Here, if the detection signals of each phase of the 6-phase resolver are A (θ), B (θ), C (θ), D (θ), E (θ), and F (θ), up to the fifth order Each detection signal in consideration of the harmonic component can be expressed by the following equations (18) to (23).

A (θ) = Adc ₀ + Aac ₁ cosθ + Aac ₂ cos2θ + Aac ₃ cos3θ + Aac ₄ cos4θ + Aac ₅ cos5θ
... (18)
B (θ) = Bdc ₀ + Bac ₁ cos (θ + 60 °) + Bac ₂ cos2 (θ + 60 °) + Bac ₃ cos3 (θ + 60 °) + Bac ₄ cos4 (θ + 60 °) + Bac ₅ cos5 (θ + 60 °)
... (19)
C (θ) = Cdc ₀ + Cac ₁ cos (θ + 120 °) + Cac ₂ cos2 (θ + 120 °) + Cac ₃ cos3 (θ + 120 °) + Cac ₄ cos4 (θ + 120 °) + Cac ₅ cos5 (θ + 120 °)
... (20)
D (θ) = Ddc ₀ + Dac ₁ cos (θ + 180 °) + Dac ₂ cos2 (θ + 180 °) + Dac ₃ cos3 (θ + 180 °) + Dac ₄ cos4 (θ + 180 °) + Dac ₅ cos5 (θ + 180 °)
... (21)
E (θ) = Edc ₀ + Eac ₁ cos (θ + 240 °) + Eac ₂ cos2 (θ + 240 °) + Eac ₃ cos3 (θ + 240 °) + Eac ₄ cos4 (θ + 240 °) + Eac ₅ cos5 (θ + 240 °)
(22)
F (θ) = Fdc ₀ + Fac ₁ cos (θ + 300 °) + Fac ₂ cos2 (θ + 300 °) + Fac ₃ cos3 (θ + 300 °) + Fac ₄ cos4 (θ + 300 °) + Fac ₅ cos5 (θ + 300 °)
(23)

In the above equation (18) ~ (23), Adc 0, Bdc 0, Cdc 0, Ddc 0, Edc 0, Fdc 0 is, A-phase, B-phase, C phase, D phase, E phase, F phase detection signal a direct current component included _{in, Aac 1 cosθ~Aac 5 cos5θ, Bac} 1 cos (θ + 60 °) ~Bac 5 cos5 (θ + 60 °), Cac 1 cos (θ + 120 °) ~Cac 5 cos5 ( θ + 120 °), Dac ₁ cos (θ + 180 °) to Dac ₅ cos5 (θ + 180 °), Eac ₁ cos (θ + 240 °) to Eac ₅ cos5 (θ + 240 °), Fac ₁ cos ( θ + 300 °) to Fac ₅ cos5 (θ + 300 °) are primary to fifth order AC components included in the detection signals of A phase, B phase, C phase, D phase, E phase, and F phase. .

Here, for convenience of explanation, the values of Adc _{0 to} Fdc ₀ , Aac _{1 to} Fac ₁ , Aac _{2 to} Fac ₂ , Aac _{3 to} Fac ₃ , Aac _{4 to} Fac ₄ , Aac _{5 to} Fac ₅ are the same order. Each shall be equal.
Accordingly, from A (θ), B (θ), C (θ), D (θ), E (θ), and F (θ) shown in the above equations (18) to (23), in the differential circuit, When a differential signal of A (θ) and D (θ), a differential signal of B (θ) and E (θ), and a differential signal of C (θ) and F (θ) are generated, These three-phase signals da, db, and dc are expressed by the following equations (24) to (26).

da = A (θ) -D (θ) = 2Aac ₁ cosθ + 2Aac ₃ cos3θ + 2Aac ₅ cos5θ
... (24)
db (θ) = B (θ) -E (θ) = 2Aac ₁ cos (θ-120 °) + 2Aac ₃ cos3 (θ-120 °) + 2Aac ₅ cos5 (θ-120 °)
... (25)
dc (θ) = C (θ) -F (θ) = 2Aac ₁ cos (θ + 120 °) + 2Aac ₃ cos3 (θ + 120 °) + 2Aac ₅ cos5 (θ + 120 °)
... (26)

From the above equations (24) to (26), it can be seen that the third-order and fifth-order harmonic components remain in the differential signal. That is, even-order harmonic components are removed by generating a differential signal.

Further, when the signals expressed by the above equations (24) to (26) are converted into two-phase signals by the 3/2 conversion circuit 42, the two-phase signals are expressed by the following equations (27) and (28). be able to.

fc (θ) = 3 (Aac ₁ cos θ + Aac ₅ cos 5θ) / 2 (27)
fs (θ) = 3 (Aac ₁ sinθ + Aac ₅ sin5θ) / 2 (28)

Accordingly, when the error obtained by calculating the angle is converted into a waveform, an error waveform in which the fifth harmonic component is dominant is obtained.

In the present embodiment, correction processing is necessary to remove this error component. In this embodiment, however, the tooth tip of the pole piece tooth 11 of the resolver stator 10 of the six-phase resolver and the tooth 21 of the resolver rotor 20 are used. Since at least one of the tooth tips of R1 / W1 and R2 / W2 is provided with R in which at least one of R1 / W1 and R2 / W2 is within a range of 13 ± 2 [%], an error within ± 20 seconds with respect to absolute accuracy is ignored. In the above equations (27) and (28), the error component due to the fifth and higher harmonics can be made substantially zero. As a result, the above equations (27) and (28) can be approximated to the following equations (29) and (30).

fc (θ) = 3 (Aac ₁ cosθ) / 2 (29)
fs (θ) = 3 (Aac ₁ sinθ) / 2 (30)

As a result, the harmonic component that causes the error can be greatly reduced, so that it is not necessary to provide a memory or a circuit for correcting the error in the calculation unit 43, or the error correction is performed in the calculation unit 43. There is no need to perform an operation. Therefore, even for a 6-phase resolver, it is possible to reduce costs by eliminating the need to provide a memory, a circuit, or the like, or it is possible to omit correction processing calculations, so that the effect of reducing the calculation load can be obtained.

Next, based on FIG. 11, a composite type of a three-phase ABS (Absolute) resolver having an A phase, a B phase, and a C phase and a three-phase INC (Increment) resolver having an A phase, a B phase, and a C phase. The operation and effect of the resolver control circuit when using the resolver device to which the above configuration is applied to the pole piece teeth 11 and teeth 21 of the resolver will be described.
The composite type resolver uses two three-phase resolvers in combination, and the configuration of each three-phase resolver is the same as that of the three-phase resolver shown in FIG. Therefore, the reference numerals of the three-phase resolver and each component are the same.

The resolver control circuit shown in FIG. 11 is an excitation changeover switch 48 for selectively supplying an excitation signal from the excitation circuit 40 to one of two three-phase resolvers with respect to the resolver control circuit shown in FIG. It has a configuration that includes. Furthermore, in order to process the signals of the two 3-phase resolver, and 41a a current / voltage conversion circuit consisting of the 3-phase ABS resolver A phase ~C phase detection resistor R _a to R _c, the 3-phase INC resolver A current / voltage conversion circuit 41b composed of A-phase to C-phase detection resistors R _{1 to} R ₃ , an ABS resolver 3/2 conversion circuit 42a, an INC resolver 3/2 conversion circuit 42b, and these three The structure includes an ABS / INC changeover switch 49 that selectively switches the output of the / 2 conversion circuits 42a and 42b to the arithmetic unit 43.

Both the excitation changeover switch 48 and the ABS / INC changeover switch 49 perform a switching operation based on a control signal from the ASIC of the calculation unit 43.
Therefore, when the excitation changeover switch 48 is activated by the control signal from the ASIC and the output terminal of the excitation circuit 40 is connected to the three-phase ABS resolver, the excitation signal sin is first sent from the excitation circuit 40 to the three-phase ABS resolver. (Ωt) is supplied. On the other hand, when the rotating shaft of the DD motor rotates, the resolver rotor 20 rotates thereby, and an analog resolver signal is output from each phase of the three-phase ABS resolver. The analog three-phase resolver signal is converted from a three-phase signal current into a three-phase signal voltage in the current / voltage conversion circuit 41a and output to the 3/2 conversion circuit 42a. In the 3/2 conversion circuit 42a, the three-phase resolver signal is converted into a two-phase ABS signal of a SIN signal and a COS signal. On the other hand, the ABS / INC changeover switch 49 is actuated by a control signal from the ASIC, and the output terminal of the 3/2 conversion circuit 42 a is connected to the input terminal of the calculation unit 43. Therefore, the two-phase ABS signals (SIN signal and COS signal) converted and output by the 3/2 conversion circuit 42 a are input to the calculation unit 43.

  Further, when the excitation changeover switch 48 is actuated by a control signal from the ASIC and the output terminal of the excitation circuit 40 is connected to the three-phase INC resolver, the excitation signal sin ( ωt) is supplied. As a result, an analog resolver signal is output from each phase of the three-phase INC resolver. The analog three-phase resolver signal is converted from a three-phase signal current into a three-phase signal voltage in the current / voltage conversion circuit 41b and output to the 3/2 conversion circuit 42b. The 3/2 conversion circuit 42b converts a three-phase resolver signal into a two-phase INC signal of a SIN signal and a COS signal. On the other hand, the ABS / INC changeover switch 49 is actuated by a control signal from the ASIC, and the output terminal of the 3/2 conversion circuit 42 b is connected to the input terminal of the calculation unit 43. Therefore, the two-phase INC signal (SIN signal and COS signal) converted and output by the 3/2 conversion circuit 42 b is input to the calculation unit 43.

The arithmetic unit 43 converts the ABS signal from the 3/2 conversion circuit 42a and the INC signal from the 3/2 conversion circuit 42b into a digital angle signal φ in the RDC, and these digital angles in the ASIC. The rotational angle position is calculated from the signal φ. Then, this rotation angle signal is output to a host control circuit that controls the motor.
The operation of the phase shifter 44 is the same as that of the resolver control circuit of FIG.

Here, the detection signals A (θ) to C (θ) of the respective phases of the three-phase ABS resolver and the three-phase INC resolver are the same as those of the three-phase resolver applied to the resolver control circuit shown in FIG. (1) to (3).
Therefore, also in the present embodiment, R1 / W1 and R2 / W2 are provided on at least one of the tooth tip of the pole piece tooth 11 of the resolver stator 10 of each three-phase resolver and the tooth tip of the tooth 21 of the resolver rotor 20. Since at least one of these is provided with R within a range of 13 ± 2 [%], the same operation and effect as the resolver control circuit of FIG. 8 can be obtained.

Next, based on FIG. 12 and FIG. 13, when a resolver device in which the above configuration is applied to the pole piece teeth 11 and teeth 21 of a four-phase resolver having an A phase, a B phase, a C phase, and a D phase is used. The operation and effect of the resolver control circuit configured to detect the angle without using the RDC will be described.
Since this four-phase resolver has the same configuration as the four-phase resolver in FIG. 9, the reference numerals of the components are the same.

In the resolver control circuit shown in FIG. 12, A / D conversion circuits 46a and 46b for converting analog output signals of the differential circuits 45a and 45b into digital signals are added to the resolver control circuit shown in FIG. It becomes the composition. Further, the calculation unit 43 is configured only by a processor such as an ASIC and does not have an RDC.
As shown in FIG. 13, the arithmetic unit 43 uses an ATAN2 function that returns arc tangent to the two digital values of SIN and COS from the A / D conversion circuits 46a and 46b, as shown in FIG. For each [°], the angular position information is converted to −180 to 180 [°].

  The resolver control circuit shown in FIG. 12 is different from the resolver control circuit shown in FIG. 9 in that the calculation unit 43 calculates the rotational angle position without using the RDC, and the other configuration is the same. The detection signals A (θ) to D (θ) of the respective phases of the four-phase resolver are the same as those in the resolver control circuit to which the four-phase resolver shown in FIG. 9 is applied, and all are represented by the above equations (10) to (13). .

Therefore, also in this embodiment, R1 / W1 and R2 / W2 are provided on at least one of the tooth tip of the pole piece tooth 11 of the resolver stator 10 of the four-phase resolver and the tooth tip 21 of the resolver rotor 20. Since at least one of R is in the range of 13 ± 2 [%], the same operation and effect as the resolver control circuit of FIG. 9 can be obtained.
As described above, the VR type resolver of the present embodiment has at least one of the tooth tip of the pole piece tooth 11 of the resolver stator 10 and the tooth tip of the tooth 21 of the resolver rotor 20 at least one of R1 / W1 and R2 / W2. Since one of them is provided with R within a range of 13 ± 2 [%], variation in gap permeance due to variation in the air gap between the tooth tips can be reduced. As a result, harmonic components included in the detection signal can be canceled or reduced, so that an error in position detection accuracy can be reduced.

In addition, since the detection error is reduced by adjusting the tooth width, inter-tooth distance (tooth pitch), depth, and R size of the pole piece teeth 11 and teeth 21, the configuration of the winding coil, etc. As it is, the detection error can be easily reduced.
In the said embodiment, the resolver stator 10 respond | corresponds to the stator of any one of invention 1-4 and 6-9, and the resolver rotor 20 is any one of invention 1-4, 6, 7, and 10. The differential circuits 45, 45a, and 45b correspond to the differential signal generation circuit according to the ninth aspect.

In the above embodiment, the rotary VR resolver of the resolver stator 10 and the resolver rotor 20 has been described as an example. However, the present invention is not limited to this configuration, and the resolver stator 10 and the resolver rotor 20 are linear. The present invention may be applied to a linear VR resolver.
In the case of this configuration, linear stator corresponds to the linear stator of the invention 10, a linear moving element corresponds to the linear movable member of the invention 10.

In the above embodiment, at least one of R1 / W1 and R2 / W2 is within a range of 13 ± 2 [%] on at least one of the pole piece teeth 11 of the resolver stator 10 and the teeth 21 of the resolver rotor 20. However, the present invention is not limited to this configuration, and a configuration may be employed in which R is such that the average value of R1 / W1 and R2 / W2 is within a range of 13 ± 2 [%].
Thus, for example, even when the upper limit of R2 is limited to 10% or less at the tip of the tooth 21 on the resolver rotor side, R2 / W2 = 10 [%] at the tip of the tooth 21. And R which becomes R1 / W1 = 16 [%] is provided at the tooth tip of the pole piece tooth 11 on the resolver stator 10 side, so that “{(R1 / W1) + (R2 / W2)} / 2 = (10 + 16) / 2 = 13 [%] ", and the detection error is reduced in the same manner as in the case of providing R in which at least one of R1 / W1 and R2 / W2 is within the range of 13 ± 2 [%]. be able to.

In the case of this configuration, the resolver stator 10 corresponds to the stator of the fifth aspect, and the resolver rotor 20 corresponds to the rotor of the fifth aspect.
Further, in the above embodiment, the tooth width W1 of the pole piece tooth 11 and the tooth width W2 of the tooth 21 are configured to be the same width (W1 = W2), the inter-tooth distance L1 in the resolver stator 10 and the resolver rotor 20. The inter-tooth distance L2 is configured with the same length (L1 = L2) and W1: L1 = W2: L2 = 1: 2. However, the configuration is not limited to this, and the lengths of the tooth widths W1 and W2 May be configured to have different lengths.

For example, as shown in FIG. 14A, W2 may be configured larger than W1. In this case, L2 is shortened by increasing W2.
In this way, when W1 and W2 are different, the relationship between W2 / L2 and the detection error is a non-linear relationship as shown in FIG. In the above embodiment, the ratio between the tooth width W2 and the interdental distance L2 is 1: 2. This means that W2 is 1/3 of the length (width) of L2 (1 / 3≈33.33 [%]).

  On the other hand, when the tooth width W2 is widened, the detection error increases as the tooth width W2 is widened at 33.33 [%] where W2: L2 = 1: 2. This is the same when W2 is narrowed, and the detection error increases as the width is narrowed. However, in FIG.14 (b), R of the tooth tip of the pole piece tooth 11 is set to the magnitude | size which becomes R1 / W1 = 13 [%], and R of the tooth tip of the tooth | gear 21 is set to R2 / W2 = 13 [ %].

Further, the relationship between R2 / W2 and the detection error in the resolver rotor 20 increases as shown in FIG. 14 (c), even if R2 is too large or too small with 13% as a boundary. To do.
From the above relationship, it is understood that there is R2 / W2 (for example, 13 [%]) that cancels the detection error when W2 / L2 is set to a fixed value (for example, 33.33 [%]).
Therefore, for example, when the tooth width W2 of the tooth 21 is set to a width satisfying W2 / L2 = 31 [%], the detection error at 31 [%] is +16 [seconds] from FIG. Therefore, it can be understood that it is sufficient to set the value of R2 so that this +16 [seconds] is canceled out by an error component (high frequency component) caused by the magnitude of R of the tooth tip.

  From FIG. 14C, since R2 / W2 = 9 [%] and the detection error is −16 [seconds], the R of the tooth tip of the tooth 21 is large so that R2 / W2 = 9 [%]. By providing the length R2, the detection error of +16 [seconds] can be canceled. Since it is only necessary to cancel the detection error, the detection error can be similarly obtained by providing the tip R of the pole piece tooth 11 with a size R1 that satisfies R1 / W1 = 9 [%] instead of the tooth 21 side. Can be countered. Further, the same effect can be obtained by providing R with a size of R1 / W1 = R2 / W2 = 9 [%] at the tips of both the teeth 21 and the pole piece teeth 11.

For example, when the tooth width W2 of the tooth 21 is set to a width that satisfies W2 / L2 = 35 [%], the detection error at 35 [%] is −24 [seconds] from FIG. Therefore, the value of R2 is set so that −24 [seconds] is canceled by an error component caused by the magnitude of R of the tooth tip.
In this case, from FIG. 14C, since R2 / W2 = 19 [%] and the detection error is +24 [seconds], the R of the tooth tip of the tooth 21 is set to R2 / W2 = 19 [%]. The detection error of −24 [seconds] can be canceled out by providing the size R2. Since it is only necessary to cancel the detection error, the detection error can be similarly obtained by providing R of the tip of the pole piece tooth 11 with a size R1 that satisfies R1 / W1 = 19 [%] instead of the tooth 21 side. Can be countered.

  In addition, although it described centering on the tooth | gear 21 of the resolver rotor 20, when changing the tooth width W1 by the side of the pole piece tooth | gear 11, it becomes the same as that of the tooth | gear 21. FIG.

It is a top view which shows the structure of the three-phase VR type | mold resolver which concerns on this invention. (A) is a waveform diagram which shows an ideal detection signal and the detection signal which shows a harmonic component, (b) is a waveform diagram which shows only the harmonic component in the waveform of (a). (A) is a figure explaining the structure of the pole piece tooth | gear 11, (b) is a figure explaining the structure of the tooth | gear 21. FIG. (A) And (b) is a figure which shows a structure when R is provided only in the pole piece tooth 11 and only the tooth | gear 21. FIG. (A) is a figure which shows the relationship between R1 / W1 when R is provided only in the pole piece tooth 11, and the magnitude | size of a harmonic component, (b) is when R is provided only in the tooth 21 It is a figure which shows the relationship between R2 / W2 and the magnitude | size of a harmonic component. It is a side view which shows a structure when R is provided in both the pole tip teeth 11 and the tooth tips of the teeth 21. It is a figure which shows the relationship between RA / WA and detection error when R is provided in both the tip of the pole piece tooth | gear 11 and the tooth | gear 21. FIG. It is a figure which shows an example of the resolver control circuit at the time of providing R which becomes in the range of R / W = 13 +/- 2 [%] in a three-phase resolver. It is a figure which shows an example of the resolver control circuit at the time of providing R which becomes in the range of R / W = 13 +/- 2 [%] in a 4-phase resolver. It is a figure which shows an example of the resolver control circuit at the time of providing R which becomes in the range of R / W = 13 +/- 2 [%] in a 6 phase resolver. It is a figure which shows an example of the resolver control circuit at the time of applying the said structure to the structure which combined the monopolar 3 phase resolver and the multipolar 3 phase resolver. It is a figure which shows an example of the resolver control circuit in the case of detecting an angle without applying RDC while applying the said structure to a 4-phase resolver. It is a figure which shows the relationship between the electrical angle after an angle calculation process, and an angle calculation value in a calculating part. (A) is a side view which shows the structure of the pole piece tooth | gear 11 and the tooth | gear 21 at the time of comprising W2 larger than tooth width W1, (b) shows the relationship between tooth width W2 and a detection error. (C) is a figure which shows the relationship between R2 and a detection error.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Resolver stator 11 Pole piece 12 Pole piece 20 Resolver rotor 21 Teeth 40 Excitation circuit 41, 41a, 41b Current / voltage conversion circuit 42, 42a, 42b 3/2 conversion circuit 43 Operation part 44 Phase shifter 45, 45a-45c Differential circuit 46a, 46b A / D conversion circuit

Claims (12)

A plurality of pole pieces having a plurality of pole piece teeth at the front end portion are fixedly supported by a plurality of circumferentially equal pieces, and coils having both N-phase excitation and signal output are serially connected to each pole piece for each phase. An annular stator wound around the stator and a tooth row formed in a circumferential direction facing the pole piece teeth of the stator with a gap and arranged concentrically with the stator to support relative rotation. A VR resolver including an annular rotor,
R having a magnitude that relatively reduces the harmonic component contained in the resolver signal output from the coils of each phase due to the variation in gap permeance between each pole piece tooth of the stator and each tooth of the rotor. Are provided at both ends in the circumferential direction at the tip of each pole piece tooth of the stator and at both ends in the circumferential direction at the tip of each tooth in the dentition of the rotor. VR type resolver.   2. The VR resolver according to claim 1, wherein a tooth width of the pole piece teeth of the stator is equal to a tooth width of the teeth of the rotor.   The circumferential tooth width W1 of each pole piece tooth of the stator and the distance L1 between each pole piece tooth adjacent to each pole piece are set to a width and length such that W1: L1 = 1: 2. And the depth H1 of the recess between the adjacent pole piece teeth is set to a depth of W1: H1 = 1: 2, and the tooth width in the circumferential direction of each tooth in the dentition of the rotor The distance L2 between each tooth adjacent to W2 is configured to have a width and length such that W2: L2 = 1: 2, and the depth H2 of the recess between each adjacent tooth is set to W2: H2 = 1. The VR resolver according to claim 2, wherein the VR resolver is configured to have a depth of 2.   The result obtained by dividing R1 R1 and W1 provided at the both ends of the pole piece teeth of the stator by W1 is within the range of “13 [%] ± 2 [%]”. The R size R2 and W2 provided at the both ends of the rotor teeth are divided by R2 and the result obtained by dividing R2 by W2 is within the range of “13 [%] ± 2 [%]”. The VR resolver according to claim 3, wherein the size is such that   R size R1 and W1 provided at the both end portions of the pole piece teeth of the stator and R size R2 and W2 provided at the both end portions of the rotor teeth, and R1 as W1 4. The VR type according to claim 3, wherein an average of a result obtained by dividing and a result obtained by dividing R <b> 2 by W <b> 2 is within a range of “13 [%] ± 2 [%]”. Resolver. A plurality of pole pieces having a plurality of pole piece teeth at the front end portion are fixedly supported by a plurality of circumferentially equal pieces, and coils having both N-phase excitation and signal output are serially connected to each pole piece for each phase. An annular stator wound around the stator and a tooth row formed in a circumferential direction facing the pole piece teeth of the stator and arranged concentrically with the stator and supported in a relatively rotatable manner. In a VR type resolver provided with an annular rotor,
The tooth width in the circumferential direction of the pole piece teeth of the stator and the tooth width in the circumferential direction of the teeth of the rotor are configured to the same width W,
The tooth width W of each pole piece tooth and the distance L1 between each pole piece tooth adjacent to each pole piece are configured to have a width and length such that W: L1 = 1: 2.
The tooth width W of each tooth of the rotor and the distance L2 between the adjacent teeth are configured to have a width and length of W: L2 = 1: 2.
R having a magnitude that relatively reduces the harmonic component contained in the resolver signal output from the coils of each phase due to the variation in gap permeance between each pole piece tooth of the stator and each tooth of the rotor. At both ends in the circumferential direction at the tip of each pole piece tooth of the stator and at both ends in the circumferential direction at the tip of each tooth in the dentition of the rotor. VR type resolver.   The magnitude of R provided at either end of either one of the pole piece teeth of the stator and the rotor teeth is set to the magnitude of R, which is either the pole piece teeth of the stator or the teeth of the rotor. 7. The VR resolver according to claim 6, wherein a result obtained by dividing by one circumferential width W is in a range of “13 [%] ± 2 [%]”.   8. The stator according to claim 1, wherein the stator is formed by winding a coil of any one of 3, 4 and 6 as the number of phases N in series for each phase. The VR resolver according to item 1.   When the stator is configured with an even number of phases having a phase number N of 4 or more, the difference in resolver signal with respect to each two-phase combination in the N-phase resolver signal output from the coil of each phase A differential signal generating means for generating a dynamic signal, and generating two-phase signals of a SIN signal and a COS signal based on the two sets of differential signals generated by the differential signal generating means, The VR resolver according to any one of claims 1 to 8. Each pole piece is provided with a plurality of pole pieces having a plurality of pole piece teeth in a straight line and is fixedly supported, and each pole piece has both N-phase excitation and signal output for each phase. A linear stator that is wound in series, and has a tooth row that is formed so as to face the pole piece teeth of the stator with a gap, and is linearly movable supported relative to the stator. In a VR resolver including a child,
The harmonic component contained in the resolver signal output from the coil of each phase due to the variation in the gap permeance between each pole piece tooth of the stator and each tooth of the linear mover is relatively reduced. R is provided at both ends in the linear direction at the tip of each pole piece tooth of the stator and at both ends in the linear direction at the tip of each tooth in the dentition of the linear mover. VR type resolver.   A motor comprising the VR resolver according to claim 9.   A linear motion linear motor comprising the VR resolver according to claim 10.

Images