Resolver

Abstract

This resolver (1) comprises a coil (13) provided to a sheet-shaped substrate (31). The coil (13) has a first coil (16) and a second coil (17) formed by a winding (W13) wound in different directions in the circumferential direction in each of a first layer (Lf2) and a second layer (Lb2) of the substrate (31). The coil (13) forms magnetic poles in mutually different directions between an outer peripheral part (16o) of the first coil (16) and an inner peripheral part (17i) of the second coil (17), and between an inner peripheral part (16i) of the first coil (16) and an outer peripheral part (17o) of the second coil (17). The outer peripheral parts (16o, 17o) and the inner peripheral parts (16i, 17i) are each formed of a plurality of windings (Wfo, Wfi, Wbo, Wbi) extending in both circumferential directions with reference to the center of each magnetic pole within a conductive wire (W13), and the number of windings (Wfo, Wfi, Wbo, Wbi) decreases as the distance from the center of the magnetic pole increases.

Description

resolver

[0001] The present invention relates to a resolver that detects the rotation angle of a rotor relative to a stator.

[0002] Conventionally, resolvers that detect the rotation angle of a rotor relative to a stator include coils arranged on a sheet-like substrate, the coils having magnetic properties that form a magnetic flux distribution with little distortion (no large magnetic flux changes) in the circumferential direction. For example, Patent Document 1 discloses a resolver with an axial angle multiplier of 1X, in which a resolver stator coil is arranged on an insulating sheet (substrate). In Patent Document 1, the resolver stator coil is composed of two spiral excitation coils that are formed long over half of a circumference. Each excitation coil has a folded pattern portion whose pattern width is wider than the pattern widths of the inner and outer pattern portions. This reduces the concentration of magnetic flux changes near the folded pattern portions, and is said to enable the magnetic flux distribution generated by the resolver stator coil to approach a sinusoidal distribution.

[0003] Patent No. 5342963

[0004] One of the objects of the present invention is to provide a novel resolver having magnetic characteristics that form a magnetic flux distribution with little distortion in the circumferential direction. However, in addition to this object, another object of the present invention is to achieve effects that cannot be obtained by conventional techniques, which are derived from the configurations shown in the below-described embodiments of the invention.

[0005] The disclosed resolver can be realized as the following disclosed aspects (application examples), which solve at least some of the above problems.

[0006] The resolver disclosed herein is a resolver that detects a rotation angle of a rotor relative to a stator, and includes: a sheet-like substrate provided on the stator or the rotor, the sheet-like substrate having a first layer and a second layer stacked in the axial direction; and a coil provided on the substrate. The coil includes a substantially annular first coil formed by a conductor wound in a first circumferential direction in the first layer, and a substantially annular second coil formed by a conductor wound in a second direction opposite to the first direction in the second layer. Each of the first coil and the second coil has an outer circumferential portion extending in the circumferential direction and an inner circumferential portion extending in the circumferential direction radially inward from the outer circumferential portion, and the coils form magnetic poles that are oriented in different directions between the outer circumferential portion of the first coil and the inner circumferential portion of the second coil, and between the inner circumferential portion of the first coil and the outer circumferential portion of the second coil. Each of the outer circumferential portion and the inner circumferential portion is formed by a plurality of windings of the conductor that extend in the first direction and the second direction based on the center of the magnetic pole formed by each of the outer circumferential portion and the inner circumferential portion, and the number of windings decreases as the distance from the center of the magnetic pole increases.

[0007] According to the disclosed resolver, a coil having magnetic properties that form a magnetic flux distribution with little distortion in the circumferential direction can be formed by combining a substantially annular first coil formed by a conductor wound in a first circumferential direction on a first layer of a substrate and a substantially annular second coil formed by a conductor wound in a second direction opposite to the first direction on a second layer, thereby providing a novel resolver.

[0008] Fig. 5 is a schematic diagram showing the structure of a resolver according to an embodiment. Fig. 6 is a plan view showing a first substrate of a stator and a second substrate of a rotor provided in the resolver of Fig. 1, arranged side by side. Fig. 7 is a plan view showing a first upper layer and a first lower layer of the stator of Fig. 2, arranged side by side. Fig. 8 is a graph defining the number of turns of a first portion and a second portion of each of two excitation coils arranged in the stator of Fig. 3. Fig. 9 is a plan view showing a second upper layer and a second lower layer of the rotor of Fig. 2, arranged side by side. Fig. 10 is a graph defining the number of windings of an outer circumferential portion and an inner circumferential portion of each of the upper layer coil and the lower layer coil arranged in the rotor of Fig. 5.

[0009] A resolver according to an embodiment will be described with reference to the drawings. The following embodiment is merely an example, and is not intended to exclude various modifications and applications of techniques not explicitly described in the following embodiment. The configurations of the present embodiment can be modified in various ways without departing from the spirit of the invention.

[0010] The resolver of this embodiment is a detector (sensor) that detects the rotation angle of a rotor that rotates around a center line C relative to a stator. In the following description, the direction in which the center line C extends is defined as the axial direction, and the direction perpendicular to the axial direction, away from the center line C and toward the center line C, is defined as the radial direction. In addition, in the radial direction, the side of the center line C is defined as the radially inner side, and the opposite side (the side away from the center line C) is defined as the radially outer side. The direction perpendicular to the axial direction and going around the center line C is defined as the circumferential direction.

[0011] [1. Overall Configuration] Fig. 1 is a schematic diagram showing the configuration of a resolver 1 according to this embodiment. The resolver 1 includes a stator 2 and a rotor 3, and is a detector (sensor) that detects the rotation angle of the rotor 3 relative to the stator 2. The resolver 1 is applied to, for example, a servo motor (not shown). The stator 2 is a component that is fixed to a casing (not shown) of a device to which the resolver 1 is applied (for example, a servo motor or a device to which a servo motor is attached). The rotor 3 is a component that is rotatably supported (around a center line C) with respect to the stator 2, integrally with a shaft (not shown). The rotor 3 is disposed with a small gap G in a first axial direction Da1 relative to the stator 2. Hereinafter, the axial direction opposite to the first axial direction Da1 will be referred to as a second axial direction Da2.

[0012] The stator 2 and the rotor 3 each include a sheet-like substrate 21, 31. Each substrate 21, 31 is a thin sheet made of insulating resin (e.g., polyimide) and has a disk shape unfolded about a center line C, as shown in Fig. 2. Fig. 2 is a diagram showing the first substrate 21 of the stator 2 and the second substrate 31 of the rotor 3 side by side, as viewed from the gap G side between the stator 2 and the rotor 3.

[0013] In this embodiment, each of the substrates 21, 31 has two layers, an upper layer Lf and a lower layer Lb, stacked in the axial direction, as shown in Fig. 3 and Fig. 5. The first substrate 21 provided on the stator 2 has a first upper layer Lf1 located on the rotor 3 side (i.e., the first axial direction Da1 side) and a first lower layer Lb1 located on the side farther from the rotor 3 than the first upper layer Lf1 (i.e., the second axial direction Da2 side), as shown in Fig. 5. The second substrate 31 provided on the rotor 3 has a second upper layer Lf2 located on the stator 2 side (i.e., the second axial direction Da2 side) and a second lower layer Lb2 located on the side farther from the stator 2 than the second upper layer Lf2 (i.e., the first axial direction Da1 side), as shown in Fig. 5.

[0014] The upper layer Lf and the lower layer Lb are respectively provided on the surface of each substrate 21, 31 facing the gap G and on the surface facing away from the gap G. In Figures 3 and 5, the lower layer Lb is shown as seen through from the upper layer Lf (i.e., the gap G side).

[0015] 1, the stator 2 may be provided with a core 22 stacked on the second axial direction Da2 side of the first substrate 21, and a non-magnetic metal plate 23 covering the first substrate 21 and the core 22 from the second axial direction Da2 side. Similarly, the rotor 3 may be provided with a core 32 stacked on the first axial direction Da1 side of the second substrate 31, and a non-magnetic metal plate 33 covering the second substrate 31 and the core 32 from the first axial direction Da1 side.

[0016] As shown in Fig. 2, a plurality of coils 11 to 15 (sheet coils) are formed on each of the substrates 21 and 31. In other words, the resolver 1 is a sheet-type resolver having a plurality of sheet-shaped coils 11 to 15. These coils 11 to 15 are made of, for example, copper foil. By using sheet coils, it is possible to reduce the thickness of each of the coils 11 to 15, and ultimately to reduce the thickness of the resolver 1.

[0017] In this embodiment, a two-input, one-output (two-phase excitation, single-phase output) resolver 1 having an axial multiplier of 1X is exemplified. The resolver 1 has two excitation coils 11, 12, a detection coil 13 (coil), a transmission coil 14, and a reception coil 15. The resolver 1 also has, as the two excitation coils 11, 12, a sine excitation coil 11 and a cosine excitation coil 12 that transmit AC signals whose electrical angle phases differ from each other by 90 degrees. The sine excitation coil 11, cosine excitation coil 12, and detection coil 13 are all coils having an axial multiplier of 1X, and have magnetic properties that form a magnetic pole pair (north and south poles) when a current flows through them.

[0018] 1 and 2, the two excitation coils 11, 12 and the receiving coil 15 are provided on a first substrate 21 and electrically connected to the signal processing circuit 4. The detecting coil 13 and the transmitting coil 14 are provided on a second substrate 31 and electrically connected to each other (connected in series).

[0019] On the first substrate 21, the two excitation coils 11, 12 are arranged, for example, in the shape of a ring centered on a center line C and overlapped in the axial direction. Each excitation coil 11, 12 is configured to form two magnetic poles (one magnetic pole pair) facing in different directions when a current flows in the direction of the white arrows shown in Figures 2 and 3. The two excitation coils 11, 12 are also configured so that the centers of the magnetic poles are positioned alternately and equidistantly in the circumferential direction. Note that in Figures 2 and 3, for convenience, the sine excitation coil 11 is shown by a solid line and the cosine excitation coil 12 is shown by a dashed line.

[0020] 2, when a current flows in the direction of the solid white arrow, the sine excitation coil 11 is configured to form a magnetic pole facing the back of the paper (a magnetic pole marked with a cross (x) surrounded by a thin solid line, hereinafter referred to as an "S pole") and a magnetic pole facing the front of the paper (a magnetic pole marked with a dot (·) surrounded by a thin solid line, hereinafter referred to as an "N pole"), with the center line C interposed therebetween. In the following description of the two excitation coils 11 and 12, the 3 o'clock position on the paper of the stator 2 shown in FIGS. 2 and 3 is defined as the origin (0°) in the circumferential direction of the stator 2. The sine excitation coil 11 forms, for example, an S pole centered at the 90° position (12 o'clock on the paper) and an N pole centered at the 270° position (6 o'clock on the paper).

[0021] The cosine excitation coil 12 is configured such that, when a current flows in the direction of the dashed white arrow, it forms an S pole (a magnetic pole marked with an x ​​and surrounded by a thin dashed line) and an N pole (a magnetic pole marked with a dot and surrounded by a thin dashed line) on either side of a center line C. For example, the cosine excitation coil 12 forms an S pole centered at the 0° position (3 o'clock on the paper) and an N pole centered at the 180° position (9 o'clock on the paper).

[0022] In other words, the two excitation coils 11, 12 are configured so that, counterclockwise from the origin in the circumferential direction of the stator 2, the centers of the south pole of the cosine excitation coil 12, the south pole of the sine excitation coil 11, the north pole of the cosine excitation coil 12, and the north pole of the sine excitation coil 11 are positioned with a phase shift of 90° in this order. Note that, since each excitation coil 11, 12 is actually excited by an AC signal input from the signal processing circuit 4, the magnetic pole orientation (polarity) of each excitation coil 11, 12 is not constant but changes over time when the rotation angle of the rotor 3 is detected. The detailed configuration of the two excitation coils 11, 12 will be described later.

[0023] On the first substrate 21, the receiving coil 15 is formed, for example, radially inward of the two exciting coils 11, 12, with a conducting wire (conductor) arranged to spiral around the center line C.

[0024] On the second substrate 31, the detection coil 13 is, for example, in the shape of a ring centered on a center line C, and is disposed opposite the two excitation coils 11 and 12. The detection coil 13 is configured to form a magnetic pole pair of an south pole (magnetic pole marked with an x) and an north pole (magnetic pole marked with a dot) on either side of the center line C when a current flows in the direction of the black arrows shown in Figures 2 and 5. Note that the detection coil 13 actually generates an induced current when the magnetic fluxes of the excitation coils 11 and 12 interlink. Therefore, when detecting the rotation angle of the rotor 3, the detection coil 13 does not spontaneously form magnetic poles.

[0025] Therefore, strictly speaking, the detection coil 13 is configured to have magnetic properties that form south and north poles on either side of the center line C when a current flows in the direction of the black arrows shown in Figures 2 and 5. In the following description of the detection coil 13, unless otherwise specified, it is assumed that a current flows in the direction of the black arrows shown in Figures 2 and 5. Furthermore, in the following description of the detection coil 13, the 6 o'clock position of the rotor 3 shown in Figures 2 and 5 is defined as the origin (0°) in the circumferential direction of the rotor 3. The detection coil 13 forms, for example, a south pole centered at the 180° position (12 o'clock on the page) and a north pole centered at the 0° position (6 o'clock on the page).

[0026] As shown in Fig. 5, the detection coil 13 has an upper layer coil 16 (first coil) arranged in the second upper layer Lf2 (first layer) and a lower layer coil 17 (second coil) arranged in the second lower layer Lb2 (second layer). The detection coil 13 is configured so that, when a current flows, the magnetic fluxes formed by the upper layer coil 16 and the lower layer coil 17 interact to form the above-mentioned single magnetic pole pair. For convenience, in Fig. 2, the upper layer coil 16 is indicated by a solid line and the lower layer coil 17 is indicated by a dashed line. The detailed configuration of the detection coil 13 will be described later.

[0027] On the second substrate 31, the transmitting coil 14 is disposed opposite the receiving coil 15. The transmitting coil 14 is formed, for example, radially inward of the detecting coil 13, with a conductor wire arranged to spiral around the center line C.

[0028] The signal processing circuit 4 is an electric circuit provided at a position different from the stator 2 and the rotor 3, and as shown in Fig. 1, it incorporates a signal generation circuit 5 and a signal detection circuit 6. The signal generation circuit 5 generates an amplitude-modulated AC signal and inputs (supplies) it to the excitation coils 11 and 12. The signal detection circuit 6 outputs angle information corresponding to the rotation angle based on the AC signal returned from the receiving coil 15.

[0029] AC signals whose electrical angle phases differ by 90 degrees are input to the sine excitation coil 11 and the cosine excitation coil 12 from the signal generating circuit 5. More specifically, a cosine wave AC signal is input to the sine excitation coil 11, and a sine wave AC signal is input to the cosine excitation coil 12.

[0030] In the resolver 1, when an AC signal is input from the signal processing circuit 4 to the excitation coils 11 and 12, the excitation coils 11 and 12 are excited and generate magnetic flux (see the open arrows in FIG. 1 ). This magnetic flux interlinks with the detection coil 13 on the rotor 3 side, generating an induced voltage. As a result, the detection coil 13 generates a signal that is phase-modulated according to the rotation angle (relative angle) of the rotor 3 with respect to the stator 2. The signal transmitted from the detection coil 13 then excites the transmission coil 14 and generates magnetic flux (see the filled arrows in FIG. 1 ). This magnetic flux interlinks with the reception coil 15 on the stator 2 side, generating an induced voltage. The output waveform of this induced voltage is output to the signal processing circuit 4, and the rotation angle of the rotor 3 is calculated based on the phase change of the output waveform. In other words, the resolver 1 is a modulated-wave resolver that receives an amplitude-modulated AC signal and uses it to detect the rotation angle from the phase-modulated signal generated in the detection coil 13.

[0031] 2 to 4, the configuration of the two excitation coils 11, 12 will be described in detail. As described above, the two excitation coils 11, 12 are arranged on the first substrate 21, overlapping each other in the axial direction. This allows the excitation coils 11, 12 to be widely arranged (deployed) in the circumferential direction, and the excitation coils 11, 12 can be configured to form a magnetic flux distribution with little distortion over a wide range in the circumferential direction. Therefore, a detection signal (phase-modulated signal) corresponding to the rotation angle of the rotor 3 is obtained by the detection coil 13, improving the detection accuracy of the resolver 1. Note that a magnetic flux distribution with little distortion here means a magnetic flux distribution in which there are few locations in the circumferential direction where magnetic coupling between the excitation coils 11, 12 and the detection coil 13 is lost and where there are no sudden changes in magnetic flux.

[0032] Here, the axial positions of the first upper layer Lf1 and the first lower layer Lb1 are slightly different on the first substrate 21. For this reason, it is assumed that the two excitation coils 11, 12 are stacked on the first substrate 21, for example, with the sine excitation coil 11 arranged on the first upper layer Lf1 and the cosine excitation coil 12 arranged on the first lower layer Lb1. In this case, a slight difference in the axial positions of the layers Lf1, Lb1 causes a magnetic resistance difference (impedance difference) between the sine excitation coil 11 and the cosine excitation coil 12 with respect to the detection coil 13, and the detection accuracy of the resolver 1 may be reduced.

[0033] Therefore, each of the two excitation coils 11, 12 is provided with a configuration that suppresses the impedance difference between the two excitation coils 11, 12. Specifically, as shown in Fig. 3, each of the two excitation coils 11, 12 is provided with a first portion 11f, 12f arranged on a first upper layer Lf1 (common layer, upper layer) and a second portion 11b, 12b arranged on a first lower layer Lb1 (layer excluding the common layer, lower layer), and the first portion 11f and the second portion 11b, and the first portion 12f and the second portion 12b are combined to form one magnetic pole. Each of the first portions 11f, 12f extends in a first circumferential direction Dc1 (e.g., clockwise direction, first direction) circumferentially around the center of the corresponding magnetic pole. Furthermore, each second portion 11b, 12b extends in a second circumferential direction Dc2 (e.g., counterclockwise direction, second direction) opposite to the first circumferential direction Dc1 with respect to the center of the corresponding magnetic pole, and is stacked on the first portion 12f, 11f of the excitation coil 12, 11 of a different (other) phase from the excitation coil 11, 12 that forms the corresponding magnetic pole. This makes the axial positional relationship (axial distance) between the two excitation coils 11, 12 and the detection coil 13 equal, thereby suppressing the impedance difference between the two excitation coils 11, 12.

[0034] In this embodiment, each excitation coil 11, 12 forms two magnetic poles, an S pole and an N pole. Therefore, each excitation coil 11, 12 has two first portions 11f, 12f and two second portions 11b, 12b. The sine excitation coil 11 has an S-pole first portion 11fs and an S-pole second portion 11bs as portions that form the S pole, and an N-pole first portion 11fn and an N-pole second portion 11bn as portions that form the N pole. The cosine excitation coil 12 has an S-pole first portion 12fs and an S-pole second portion 12bs as portions that form the S pole, and an N-pole first portion 12fn and an N-pole second portion 12bn as portions that form the N pole.

[0035] Hereinafter, the center position of the south pole of the sine excitation coil 11 (90° position) will be referred to as the first south pole center position Pcs1, and the center position of the north pole of the sine excitation coil 11 (270° position) will be referred to as the first north pole center position Pcn1. The first south pole center position Pcs1 and the first north pole center position Pcn1 will also be collectively referred to as the first center position Pc1. The center position of the south pole of the cosine excitation coil 12 (0° position) will be referred to as the second south pole center position Pcs2, and the center position of the north pole of the cosine excitation coil 12 (180° position) will be referred to as the second north pole center position Pcn2. The second south pole center position Pcs2 and the second north pole center position Pcn2 will also be collectively referred to as the second center position Pc2.

[0036] The south pole first portion 11fs, which forms the south pole of the sine excitation coil 11, has a quadrant ring shape (partial ring shape) extending from the first south pole center position Pcs1 to, for example, near a position (0°) where the phase differs by 90° in the first circumferential direction Dc1 (shorter than this position). The south pole second portion 11bs has a quadrant ring shape extending from the first south pole center position Pcs1 to, for example, near a position (180°) where the phase differs by 90° in the second circumferential direction Dc2 (same position). The south pole second portion 11bs is stacked on the north pole first portion 12fn of the cosine excitation coil 12.

[0037] The N-pole first portion 11fn, which forms the N-pole of the sine excitation coil 11, has a quadrant ring shape extending, for example, to a position (180°) where the phase differs by 90° in the first circumferential direction Dc1 based on the first N-pole center position Pcn1. The N-pole second portion 11bn has a quadrant ring shape extending, for example, to a position (0°) where the phase differs by 90° in the second circumferential direction Dc2 based on the position Pcn1, and is stacked on the S-pole first portion 12fs of the cosine excitation coil 12.

[0038] Similarly, the S-pole first portion 12fs, which forms the S-pole of the cosine excitation coil 12, has a quadrant ring shape extending, for example, to near a position (270°) where the phase differs by 90° in the first circumferential direction Dc1, based on the second S-pole center position Pcs2. The S-pole second portion 12bs has a quadrant ring shape extending, for example, to near a position (90°) where the phase differs by 90° in the second circumferential direction Dc2, based on the position Pcs2, and is stacked on the S-pole first portion 11fs of the sine excitation coil 11.

[0039] The N-pole first portion 12fn, which forms the N-pole of the cosine excitation coil 12, has a quadrant ring shape extending, for example, to a position (90°) where the phase differs by 90° in the first circumferential direction Dc1 based on the second N-pole center position Pcn2. The N-pole second portion 12bn has a quadrant ring shape extending, for example, to a position (270°) where the phase differs by 90° in the second circumferential direction Dc2 based on the position Pcn2, and is stacked on the N-pole first portion 11fn of the sine excitation coil 11.

[0040] When a portion forming one magnetic pole is composed of two portions provided on different layers Lf1 and Lb1 as described above, the portion must have a connection point that penetrates the first substrate 21 and electrically connects the portions 11f and 12f provided on the first upper layer Lf1 with the portions 11b and 12b provided on the first lower layer Lb1. However, such connection points are prone to variations in the resistance of the conductor due to manufacturing reasons (e.g., etching and plating processes). Such variations in the resistance of the connection points are caused by variations in the wiring resistance of the excitation coils 11 and 12, which can easily lead to differences in wiring resistance between the two excitation coils 11 and 12.

[0041] Therefore, each of the two excitation coils 11, 12 is provided with a configuration that suppresses the above-mentioned wiring resistance difference. Specifically, the first portions 11f, 12f and the second portions 11b, 12b of each excitation coil 11, 12 are formed into a substantially closed shape by winding conductor wires W11, W12 around the center positions Pc1, Pc2 of the respective magnetic poles, and the first portions 11f, 12f and the second portions 11b, 12b that form one magnetic pole are electrically connected via single through holes 11h, 12h (connection points). This minimizes the number of connection points for the parts that form one magnetic pole, thereby suppressing wiring resistance differences caused by variations in resistance values ​​at the connection points. Note that the conductor wire W11 that forms the portions 11f, 11b of the sine excitation coil 11 and the conductor wire W12 that forms the portions 12f, 12b of the cosine excitation coil 12 are separate conductors and are not electrically connected to each other.

[0042] A detailed description will be given focusing on the sine excitation coil 11. The first portion 11f and the second portion 11b of the sine excitation coil 11 are formed, for example, by routing the conductor W11 so as to surround (circumnavigate) the first central position Pc1, thereby forming a substantially closed shape. Note that the term "substantially closed shape" as used here means a shape in which the conductor W11 is present all around the first central position Pc1 of the magnetic pole, and does not mean a shape in which the conductor W11 is absent in part of the periphery of the first central position Pc1, such as a U-shape or a C-shape.

[0043] Each of the portions 11f, 11b of the sine excitation coil 11 may be formed, for example, as shown in the figure, by winding a single conductor W11 in a spiral (snake pattern) around the first center position Pc1 in each of the layers Lf1, Lb1. In this way, each of the portions 11f, 11b may be formed by the conductor W11 that is continuously arranged without interruption in a single stroke in each of the layers Lf1, Lb1. Each of the portions 11f, 11b forms a magnetic pole in an inner region of the winding of the conductor W11 (the region indicated by dotted lines in FIG. 3).

[0044] Furthermore, in this embodiment, the first portion 11 f and the second portion 11 b are wound so that the number of turns decreases as the distance from the first center position Pc1 increases. That is, the conducting wire W11 forming each of the portions 11 f and 11 b is wound in one direction in a partially annular shape from a starting point (through hole 11 h) near the first center position Pc1, and is wound so that the ending point is on the outer side of the winding, and the turned-back portion on the side opposite to the first center position Pc1 is sparser than the turned-back portion on the side closer to the first center position Pc1.

[0045] This results in a magnetic flux distribution in which the magnitude of the magnetic flux peaks at the first central position Pc1 and decreases as the distance from the first central position Pc1 increases, thereby suppressing distortion of the magnetic flux distribution (particularly, abrupt changes in the magnetic flux). The above description has been focused on the sine excitation coil 11, but for the cosine excitation coil 12 as well, the conductor wire W12 is similarly arranged around the second central position Pc2 to form a first portion 12f and a second portion 12b.

[0046] The conductors W11 and W12 are arranged in the first portions 11f and 12f and the second portions 11b and 12b, respectively, so that the number of turns thereof satisfies the relationship shown in the graph of Fig. 4. Fig. 4 is a graph defining the number of turns M of each of the portions 11f, 11b, 12f, and 12b in the circumferential direction, with the horizontal axis corresponding to the circumferential angle shown in Fig. 3 and the vertical axis corresponding to the number of turns M of each of the portions 11f, 11b, 12f, and 12b. On the vertical axis, the "(S)" side corresponds to the number of turns M of the portions 11fs, 12fs, 11bs, and 12bs that form the south pole, and the "(N)" side corresponds to the number of turns M of the portions 11fn, 12fn, 11bn, and 12bn that form the north pole.

[0047] 4, for convenience, the portions 11fs, 11fn, 12fs, and 12fn arranged on the first upper layer Lf1 are indicated by thicker lines than the portions 11bs, 11bn, 12bs, and 12bn arranged on the first lower layer Lb1. Furthermore, the portions 11fs, 11fn, 11bs, and 11bn of the sine excitation coil 11 are indicated by solid lines, and the portions 12fs, 12fn, 12bs, and 12bn of the cosine excitation coil 12 are indicated by dashed lines.

[0048] 4, the number of turns M of the S-pole first portion 11fs of the sine excitation coil 11 increases from near 0° to near 90°, and is set to a maximum value (5 in this case) near 90°, i.e., near the first S-pole center position Pcs1. The number of turns M of the S-pole second portion 11bs of the sine excitation coil 11 is set symmetrically to the S-pole first portion 11fs, decreasing from near 90° to near 180°.

[0049] The N-pole first portion 11fn of the sine excitation coil 11 is set so that the number of turns M increases from around 180° to around 270° and reaches a maximum value (5 in this case) around 270°, i.e., around the first N-pole center position Pcn1. The N-pole second portion 11bn of the sine excitation coil 11 is set so that the number of turns M decreases from around 270° to around 360°, symmetrically with the N-pole first portion 11fn.

[0050] In the sinusoidal excitation coil 11, the number of turns M of each of the portions 11fs, 11bs, 11fn, and 11bn is set so as to approximate a sinusoidal waveform with one cycle per circumferential revolution, as described above. As a result, when the sinusoidal excitation coil 11 is excited, a magnetic flux distribution is formed in accordance with the change in the number of turns M of each of the portions 11fs, 11bs, 11fn, and 11bn, i.e., a sinusoidal magnetic flux distribution, and therefore a magnetic flux distribution with little distortion is obtained.

[0051] The number of turns M of the S-pole second portion 12bs of the cosine excitation coil 12 is set to be maximum (5 in this case) near 0°, i.e., near the second S-pole center position Pcs2, and decreases from near 0° to near 90°. The number of turns M of the S-pole first portion 12fs of the cosine excitation coil 12 is set symmetrically to the S-pole second portion 12bs, and decreases from near 0° to near −90° (270°).

[0052] The N-pole first portion 12fn of the cosine excitation coil 12 is set so that the number of turns M increases from around 90° to around 180° and reaches a maximum value (5 in this case) around 180°, i.e., around the second N-pole center position Pcn2. The N-pole second portion 12bn of the cosine excitation coil 12 is set so that the number of turns M decreases from around 180° to around 270°, symmetrically with the N-pole first portion 12fn.

[0053] In the cosine excitation coil 12, the number of turns M of each of the portions 12bs, 12fn, 12bn, and 12fs is set so as to approximate a cosine waveform in which one cycle corresponds to one circumferential revolution, as described above. As a result, when the cosine excitation coil 12 is excited, a magnetic flux distribution is formed in accordance with the change in the number of turns M of each of the portions 12bs, 12fn, 12bn, and 12fs, i.e., a cosine waveform magnetic flux distribution, and therefore a magnetic flux distribution with little distortion is obtained.

[0054] In this embodiment, the first portions 11f, 12f and the second portions 11b, 12b, which respectively form one magnetic pole, have the same number of turns (maximum number of turns). However, the first portions 11f, 12f and the second portions 11b, 12b may have different numbers of turns. For example, the number of turns of the first portions 11f, 12f may be set to be less than the number of turns of the second portions 11b, 12b. This suppresses the impedance difference between the first portions 11f, 12f and the second portions 11b, 12b, which occurs due to the difference in the axial positions of the layers Lf1, Lb1.

[0055] 2 and 3, the radial widths of the first portions 11f, 12f and the second portions 11b, 12b, which respectively form one magnetic pole, are approximately equal. However, the radial widths of the first portions 11f, 12f and the second portions 11b, 12b may be different. For example, the area of ​​the region where the magnetic pole is formed in the first portions 11f, 12f may be smaller than the area of ​​the region where the magnetic pole is formed in the second portions 11b, 12b. This suppresses the impedance difference between the first portions 11f, 12f and the second portions 11b, 12b, which occurs due to the difference in the axial positions of the layers Lf1, Lb1.

[0056] 2, 5, and 6, the configuration of the detector coil 13 will be described in detail below. As described above, the detector coil 13 has an upper layer coil 16 and a lower layer coil 17, and the magnetic fluxes of the upper layer coil 16 and the lower layer coil 17 interact with each other to form magnetic poles (south pole and north pole) oriented in different directions. Hereinafter, the center position of the south pole of the detector coil 13 (the position at 180° of the rotor 3 shown in FIGS. 2 and 5) will be referred to as a third south pole center position Pcs3, and the center position of the north pole of the detector coil 13 (the position at 0° of the rotor 3 shown in FIGS. 2 and 5) will be referred to as a third north pole center position Pcn3.

[0057] 5, the detection coil 13 is formed by a conductor W13 (conductor) arranged in the second upper layer Lf2 and the second lower layer Lb2. The conductor W13 forming the detection coil 13 is a conductor of a different system from the conductors W11 and W12 forming the excitation coils 11 and 12, and is not electrically connected to these conductors W11 and W12.

[0058] The upper layer coil 16 is a generally annular coil formed by winding the upper layer conductor Wf, which is disposed in the second upper layer Lf2 (first layer) of the conductor W13, in a first circumferential direction Dc1 (first direction) starting from a radially outer end point. The lower layer coil 17 is a generally annular coil formed by winding the lower layer conductor Wb, which is disposed in the second lower layer Lb2 (second layer) of the conductor W13, in a second circumferential direction Dc2 (second direction) starting from a radially inner end point. The upper layer conductor Wf and the lower layer conductor Wb are wound around the center line C at least two or more times (multiple times) without turning back in the direction opposite to their respective winding directions (the first circumferential direction Dc1 and the second circumferential direction Dc2). The upper layer conductor Wf and the lower layer conductor Wb are electrically connected to each other via the through holes 13h, thereby being continuously provided.

[0059] 5 indicates the routing direction of the conductor W13 when the upper layer conductor Wf and the lower layer conductor Wb are routed in this order, starting from the outer end point of the winding of the upper layer conductor Wf (i.e., the position of the black arrow in FIG. 5). This thin arrow can also be said to be the direction of current flow when current flows in the direction of the black arrow in FIG.

[0060] Each of the upper coil 16 and the lower coil 17 has an outer circumferential portion 16o, 17o and an inner circumferential portion 16i, 17i, which form two magnetic poles oriented in different directions. Each of the outer circumferential portions 16o, 17o extends in the circumferential direction near the outer circumferential edge of the upper coil 16 and the lower coil 17, which are substantially annular in shape. Each of the inner circumferential portions 16i, 17i extends in the circumferential direction radially inward of each of the outer circumferential portions 16o, 17o.

[0061] The upper-layer outer circumferential portion 16o of the upper-layer coil 16 is arranged so as not to overlap partially with the upper-layer inner circumferential portion 16i of the upper-layer coil 16 in the circumferential direction, and is arranged at approximately the same circumferential position as the lower-layer inner circumferential portion 17i of the lower-layer coil 17. Similarly, the lower-layer outer circumferential portion 17o of the lower-layer coil 17 is arranged so as not to overlap partially with the lower-layer inner circumferential portion 17i in the circumferential direction, and is arranged at approximately the same circumferential position as the upper-layer inner circumferential portion 16i.

[0062] In this embodiment, the upper layer outer peripheral portion 16o extends in both circumferential directions (first circumferential direction Dc1 and second circumferential direction Dc2) based on the third N pole center position Pcn3, except for the area around the third S pole center position Pcs3 in the circumferential direction. The lower layer inner peripheral portion 17i is located adjacent to the radially inward side of the upper layer outer peripheral portion 16o. The upper layer inner peripheral portion 16i extends in both circumferential directions based on the third S pole center position Pcs3, except for the area around the third N pole center position Pcn3 in the circumferential direction. As a result, the upper layer outer peripheral portion 16o and the upper layer inner peripheral portion 16i are arranged so as not to overlap partially in the circumferential direction around the third N pole center position Pcn3 and the third S pole center position Pcs3.

[0063] In this way, since upper layer outer peripheral portion 16o is provided on the third N-pole central position Pcn3 side and upper layer inner peripheral portion 16i is provided on the third S-pole central position Pcs3 side, upper layer coil 16 can also be said to be eccentric toward the third N-pole central position Pcn3 side with respect to center line C. When current flows in the direction of the black arrow shown in Figure 5, upper layer coil 16 generates an N-pole magnetic flux in a substantially circular upper layer region Rf (the region shown by dark dots in Figure 5) that is developed around a position closer to the third N-pole central position Pcn3 than the center line C.

[0064] The same applies to the lower layer coil 17. The lower layer outer peripheral portion 17o extends in both circumferential directions with the third south pole center position Pcs3 as the reference, except for the area around the third north pole center position Pcn3 in the circumferential direction. The upper layer inner peripheral portion 16i is located adjacent to the radially inward side of the lower layer outer peripheral portion 17o. The lower layer inner peripheral portion 17i extends in both circumferential directions with the third north pole center position Pcn3 as the reference, except for the area around the third south pole center position Pcs3 in the circumferential direction. As a result, the lower layer outer peripheral portion 17o and the lower layer inner peripheral portion 17i are arranged so as not to overlap partially in the circumferential direction around the third north pole center position Pcn3 and the third south pole center position Pcs3.

[0065] In this way, since lower layer outer peripheral portion 17o is provided on the third S-pole central position Pcs3 side and lower layer inner peripheral portion 17i is provided on the third N-pole central position Pcn3 side, lower layer coil 17 can be said to be eccentric toward the third S-pole central position Pcs3 side with respect to center line C. When a current flows in the direction of the black arrow shown in Figure 5, lower layer coil 17 generates an S-pole magnetic flux in a substantially circular lower layer region Rb (the region shown by light dotted lines in Figure 5) that is centered at a position closer to the third S-pole central position Pcs3 than the center line C.

[0066] In the detection coil 13, when a current flows in the direction of the black arrows shown in Figures 2 and 5, the magnetic fluxes of the upper layer coil 16 and the lower layer coil 17 interact with each other. As a result, in the region radially inward of the inner circumferential portions 16i and 17i, the magnetic fluxes of the upper layer coil 16 and the lower layer coil 17 cancel each other out, and as shown in Figure 2, magnetic poles of different orientations are formed between the upper outer circumferential portion 16o and the lower inner circumferential portion 17i, and between the upper inner circumferential portion 16i and the lower outer circumferential portion 17o. More specifically, an N pole is formed in a semicircular ring-shaped region (indicated by thick dots in Figure 2) centered on the third N pole central position Pcn3 between the upper outer circumferential portion 16o and the lower inner circumferential portion 17i, and an S pole is formed in a semicircular ring-shaped region (indicated by light dots in Figure 2) centered on the third S pole central position Pcs3 between the upper inner circumferential portion 16i and the lower outer circumferential portion 17o.

[0067] 5, the upper-layer outer peripheral portion 16o is formed by a plurality of outer windings Wfo of the upper-layer conductor Wf, which extend in both circumferential directions based on the third north-pole center position Pcn3. The upper-layer inner peripheral portion 16i is formed by a plurality of inner windings Wfi of the upper-layer conductor Wf, which extend in both circumferential directions based on the third south-pole center position Pcs3. The number of windings Wfo, Wfi forming each of the upper-layer outer peripheral portion 16o and the upper-layer inner peripheral portion 16i decreases with increasing distance from the respective magnetic pole centers Pcn3, Pcs3.

[0068] Furthermore, in this embodiment, the upper layer conductor Wf is provided with a plurality of connection wires Wfc that connect the plurality of outer windings Wfo and the plurality of inner windings Wfi. The upper layer conductor Wf is arranged so that, for each turn around the center line C, an outer winding Wfo, an inner winding Wfi, and two connection wires Wfc that connect the ends of these windings Wfo, Wfi are formed.

[0069] For example, the upper layer conductor Wf is routed starting from a position near the third north-pole center position Pcn3 and radially outward of the position where the upper layer outer peripheral portion 16o is formed, and extending in the first circumferential direction Dc1 to form the outer winding Wfo. Then, just before reaching the third south-pole center position Pcs3, the upper layer conductor Wf is bent radially inward to form one connection line Wfc, the inner winding Wfi, and the other connection line Wfc, in that order. More specifically, the upper layer conductor Wf is routed radially inward to form one connection line Wfc, then in the first circumferential direction Dc1 to form the inner winding Wfi, and then in the radially outward to form the other connection line Wfc. The upper layer conductor Wf is thus routed around the third south-pole center position Pcs3, cutting out a portion of the outer circle along which the outer winding Wfo extends. In this embodiment, the connection wires Wfc extend in the radial direction as shown in the figure, but the connection wires Wfc may extend in a direction inclined relative to the radial direction. For example, one of the connection wires Wfc may be arranged to extend radially inward in the first circumferential direction Dc1, and the other connection wire Wfc may be arranged to extend radially outward in the first circumferential direction Dc1. The upper-layer conductor Wf is then arranged again in the first circumferential direction Dc1 to form the outer winding wire Wfo. When passing near the third N-pole center position Pcn3, the upper-layer conductor Wf is arranged slightly radially inward of the previous outer winding wire Wfo, and this process is repeated.

[0070] The upper layer conductor Wf is arranged so that the circumferential width of the notch formed by the pair of connecting wires Wfc and the inner winding Wfi increases with each turn around the center line C. As a result, both the upper layer outer peripheral portion 16o and the upper layer inner peripheral portion 16i are configured so that the number of windings Wfo, Wfi decreases as they move away from the center positions Pcn3, Pcs3 of the respective magnetic poles.

[0071] In this embodiment, to shift the phase between the circumferential position of the connection wire Wfc of the upper-layer conductor Wf and the circumferential position of the connection wire Wbc of the lower-layer conductor Wb (described later), the entire upper coil 16 is tilted at a predetermined angle in the first circumferential direction Dc1 with respect to a diameter line connecting the third south-pole center position Pcs3 and the third north-pole center position Pcn3. Specifically, the outer winding Wfo extends on both sides of the circumferential direction, with a first offset position P1 offset by a predetermined angle in the first circumferential direction Dc1 from the third north-pole center position Pcn3 as its center. The inner winding Wfi extends on both sides of the circumferential direction, with a second offset position P2 offset by a predetermined angle in the first circumferential direction Dc1 from the third south-pole center position Pcs3 as its center. The predetermined angle may be set depending on the maximum number of windings Wfo and Wfi.

[0072] The lower-layer outer peripheral portion 17o is formed by a plurality of outer windings Wbo of the lower-layer conductor Wb that extend in both circumferential directions based on the third south-pole center position Pcs3. The lower-layer inner peripheral portion 17i is formed by a plurality of inner windings Wbi of the lower-layer conductor Wb that extend in both circumferential directions based on the third north-pole center position Pcn3. The plurality of windings Wbo, Wbi that form the lower-layer outer peripheral portion 17o and the lower-layer inner peripheral portion 17i, respectively, are set so that the number of windings Wbo, Wbi decreases with increasing distance from the respective magnetic pole center positions Pcs3, Pcn3, similar to the windings Wfo, Wfi of the upper-layer outer peripheral portion 16o and the upper-layer inner peripheral portion 16i.

[0073] Furthermore, in this embodiment, the lower conductor Wb is provided with a plurality of connection wires Wbc connecting the plurality of outer windings Wbo and the plurality of inner windings Wbi, similar to the upper conductor Wf. The lower conductor Wb is arranged so that, for each turn around the center line C, an outer winding Wbo, an inner winding Wbi, and two connection wires Wbc connecting the ends of these windings Wbo and Wbi are formed.

[0074] The lower-layer conductor Wb is routed, for example, from a position (the position of the through-hole 13h) near the third north-pole center position Pcn3 and radially inward of the position where the lower-layer inner circumferential portion 17i is formed, toward the second circumferential direction Dc2 to form the inner winding Wbi. Then, just before reaching the third south-pole center position Pcs3, the lower-layer conductor Wb is bent radially outward to form one connection line Wbc, the outer winding Wbo, and the other connection line Wbc in this order. More specifically, the lower-layer conductor Wb is routed radially outward to form one connection line Wbc, then in the second circumferential direction Dc2 to form the outer winding Wbo, and finally inward to form the other connection line Wbc. The lower-layer conductor Wb is thus routed so as to protrude from a portion of the inner circle around the third north-pole center position Pcn3, along which the inner winding Wbi extends. In this embodiment, the connection wires Wbc extend in the radial direction as shown in the figure, but the connection wires Wbc may extend in a direction inclined relative to the radial direction. For example, one of the connection wires Wbc may be arranged to extend radially outward in the second circumferential direction Dc2, and the other connection wire Wbc may be arranged to extend radially inward in the second circumferential direction Dc2. The lower-layer conductor Wb is then arranged again in the second circumferential direction Dc2 to form the inner winding Wbi. When passing near the third N-pole center position Pcn3, the lower-layer conductor Wb is arranged slightly radially outward of the previous inner winding Wbi, and this process is repeated.

[0075] The lower layer conductor Wb is arranged so that the circumferential width of the protruding portion, which is made up of the pair of connecting wires Wbc and the outer winding Wbo, increases with each turn around the center line C. As a result, both the lower layer outer peripheral portion 17o and the lower layer inner peripheral portion 17i are configured so that the number of windings Wbo, Wbi decreases with increasing distance from the center positions Pcs3, Pcn3 of the magnetic poles.

[0076] In this embodiment, the entire lower layer coil 17 is disposed opposite the upper layer coil 16, tilted at a predetermined angle in the second circumferential direction Dc2 with respect to a diameter line connecting the third south pole center position Pcs3 and the third north pole center position Pcn3. Specifically, the outer winding Wbo extends on both sides of the circumferential direction around a third offset position P3, which is offset by a predetermined angle in the second circumferential direction Dc2 from the third south pole center position Pcs3. The inner winding Wbi extends on both sides of the circumferential direction around a fourth offset position P4, which is offset by a predetermined angle in the second circumferential direction Dc2 from the third north pole center position Pcn3. As a result, the circumferential positions of the connection lines Wfc of the upper layer conductors Wf and the connection lines Wbc of the lower layer conductors Wb are arranged with a phase shift so that they alternate in the circumferential direction, as shown in FIG. 2 . The predetermined angle of the lower layer coil 17 may be set according to the maximum number of windings Wbo, Wbi. The predetermined angle of the upper layer coil 16 may be the same as or different from the predetermined angle of the lower layer coil 17.

[0077] The numbers of windings Wfo, Wfi, Wbo, and Wbi in each of the outer circumferential portions 16o, 17o and the inner circumferential portions 16i, 17i in the circumferential direction are set to satisfy the relationship of the numbers shown in the graph of Fig. 6. Fig. 6 is a graph that defines the number m of windings Wfo, Wfi, Wbo, and Wbi in each of the portions 16o, 16i, 17o, and 17i in the circumferential direction.

[0078] In the graph of Fig. 6, the horizontal axis corresponds to the angle in the circumferential direction shown in Fig. 5. The vertical axis on the left corresponds to the number m of the windings Wfo, Wfi, Wbo, and Wbi. On the vertical axis on the left, the "(S)" side corresponds to the number m of the windings Wfi and Wbo that form the south pole, and the "(N)" side corresponds to the number m of the windings Wfo and Wbi that form the north pole. For convenience, in Fig. 6, the outer windings Wfo and Wbo are indicated by thicker lines than the inner windings Wfi and Wbi. The windings Wfo and Wfi of the upper conductor Wf are indicated by solid lines, and the windings Wbo and Wbi of the lower conductor Wb are indicated by dashed lines.

[0079] 6, the vertical axis on the right side corresponds to the number of turns M of the south pole and north pole formed in the detection coil 13 by the interaction of the portions 16o, 16i, 17o, and 17i. On the vertical axis on the right side, the "(S)" side corresponds to the number of turns M of the south pole, and the "(N)" side corresponds to the number of turns M of the north pole.

[0080] 6, the number m of the inner winding Wfi of the upper-layer conductor Wf and the outer winding Wbo of the lower-layer conductor Wb is set to a maximum value (5 in this case) near 180°, i.e., near the third S-pole center position Pcs3. The number m decreases as the distance from the 180° position increases, and the number m is set to a minimum value (0 in this case) near 0° (360°).

[0081] Conversely, the number m of the outer winding Wfo of the upper-layer conductor Wf and the inner winding Wbi of the lower-layer conductor Wb is set to a maximum value (5 in this case) near 0° (360°), i.e., near the third north-pole center position Pcn3. The number m decreases as the angle moves away from the 0° position, and is set to a minimum value (0 in this case) near 180°.

[0082] By setting the number m of each of the windings Wfo, Wfi, Wbo, and Wbi as described above, the number of turns M changes in a cosine waveform, with one cycle per circumferential revolution, as shown by the dotted waveform, in the detection coil 13. In other words, the detection coil 13 changes in the number of turns in a sine waveform, with one cycle occurring between the S and N poles, starting from the 90° position, which is the boundary between the S and N poles.

[0083] As a result, when a current flows through the detection coil 13, a magnetic flux distribution corresponding to the change in the number of turns of the detection coil 13, i.e., a sinusoidal magnetic flux distribution, is formed, thereby obtaining a magnetic flux distribution with little distortion. In other words, by setting the number m of each of the windings Wfo, Wfi, Wbo, and Wbi so that the dotted waveform, which is a composite wave of the lines (waveforms) of the windings Wfo, Wfi, Wbo, and Wbi, approximates a sinusoidal waveform with one cycle per circumferential turn (between the south and north poles), the detection coil 13 can be provided with magnetic properties that form a sinusoidal magnetic flux distribution with little distortion between the south and north poles.

[0084] In this embodiment, since the connecting wires Wfc, Wbc extend in the radial direction as described above, the step positions of the waveforms representing the windings Wfo, Wfi, Wbo, and Wbi correspond to the positions of the connecting wires Wfc, Wbc. As such, the graph in Figure 6 also shows the positions of the connecting wires Wfc, Wbc, and can be said to be a graph that defines the positional relationship of the connecting wires Wfc, Wbc in the circumferential direction.

[0085] The detection coil 13 has a plurality of connecting wires Wfc, Wbc set (routed) so as to pass through circumferential positions that form a sinusoidal magnetic flux distribution that forms one cycle between the south pole and the north pole, that is, the step positions of the waveforms of the windings Wfo, Wfi, Wbo, and Wbi shown in FIG. 6. This results in a change in the number of turns of the waveform shown by dotted lines in FIG. 6, and suppresses distortion of the magnetic flux distribution when current flows.

[0086] The number of turns M of the detection coil 13 changes at the step positions, and the magnitude of the magnetic flux changes in accordance with this change in the number of turns M. Therefore, the step positions can also be referred to as magnetic flux change positions. Even when the connection wires Wfc, Wbc extend at an inclination relative to the radial direction, the detection coil 13 is configured so that the connection wires Wfc, Wbc pass through the magnetic flux change positions, thereby suppressing distortion of the magnetic flux distribution when a current flows. The magnetic flux change positions may be set based on the circumferential length between a pair of magnetic poles and the maximum number m of the windings Wfo, Wfi, Wbo, and Wbi. When the connection wires Wfc, Wbc extend at an inclination relative to the radial direction, the connection wires Wfc, Wbc are preferably configured to pass through the magnetic flux change positions and at an inclination that makes the step-like steps in FIG. 6 slope. This allows the dotted waveform, which is a composite wave of the waveforms of the windings Wfo, Wfi, Wbo, and Wbi, to change more smoothly, resulting in a waveform that more closely resembles a sine waveform, thereby suppressing distortion of the magnetic flux distribution (especially sudden changes in the magnetic flux).

[0087] Furthermore, in this embodiment, due to the arrangement (inclination) relationship between the upper layer coil 16 and the lower layer coil 17, the inner winding Wfi of the upper layer conductor Wf and the outer winding Wbo of the lower layer conductor Wb are shifted in phase by the difference between the second shift position P2 and the third shift position P3. The outer winding Wfo of the upper layer conductor Wf and the inner winding Wbi of the lower layer conductor Wb are shifted in phase by the difference between the first shift position P1 and the fourth shift position P4.

[0088] Therefore, the positions (step positions, magnetic flux change positions) of the connecting wires Wfc of the upper conductor Wf and the positions (step positions, magnetic flux change positions) of the connecting wires Wbc of the lower conductor Wb are shifted from each other by the above-mentioned phase difference. Because the circumferential positions through which the connecting wires Wfc of the upper conductor Wf pass and the circumferential positions through which the connecting wires Wbc of the lower conductor Wb pass are shifted in phase, the steps of the dotted waveform become finer. Therefore, a waveform that more closely resembles a sinusoidal waveform is obtained, thereby suppressing distortion of the magnetic flux distribution (especially sudden changes in magnetic flux).

[0089] [4. Actions and Effects] (1) In the resolver 1 described above, the detector coil 13 disposed on the second substrate 31 includes an upper layer coil 16 disposed on the second upper layer Lf2 of the second substrate 31 and a lower layer coil 17 disposed on the second lower layer Lb2. The upper layer coil 16 and the lower layer coil 17 are formed of conductor wires Wf, Wb wound in different circumferential directions. As described above, unlike conventional resolvers in which one coil (detector coil or excitation coil) is formed of two spiral coils formed over half a circumference in one layer, the detector coil 13 in the resolver 1 described above is formed of the upper layer coil 16 and the lower layer coil 17 formed of conductor wire W13 wound in different circumferential directions in different layers Lf2, Lb2. Therefore, a novel resolver 1 can be provided.

[0090] In the resolver 1 described above, the upper layer coil 16 and the lower layer coil 17 have outer peripheral portions 16o, 17o and inner peripheral portions 16i, 17i. When a current flows through the detection coil 13, magnetic poles of different orientations are formed between the upper outer peripheral portion 16o and the lower inner peripheral portion 17i, and between the upper inner peripheral portion 16i and the lower outer peripheral portion 17o. Each of the outer peripheral portions 16o, 17o and the inner peripheral portions 16i, 17i of the detection coil 13 is formed by a plurality of windings Wfo, Wfi, Wbo, Wbi extending in both circumferential directions based on a center position Pcs3, Pcn3 of each magnetic pole, and the number of windings Wfo, Wfi, Wbo, Wbi decreases with increasing distance from the center position Pcs3, Pcn3 of each magnetic pole. This results in a magnetic flux distribution in which the magnitude of the magnetic flux peaks at the magnetic pole center positions Pcs3, Pcn3 and gradually decreases as the distance from the magnetic pole center positions Pcs3, Pcn3 increases. Therefore, the resolver 1 described above can provide the detection coil 13, which has a different structure from conventional coils, with magnetic characteristics that form a magnetic flux distribution with little distortion in the circumferential direction. Therefore, an appropriate detection signal according to the rotation angle of the rotor 3 can be obtained, thereby improving the detection accuracy of the resolver 1.

[0091] (2) In the resolver 1 described above, the multiple connection wires Wfc, Wbc of each of the upper layer coil 16 and the lower layer coil 17 are set to pass through circumferential positions (magnetic flux change positions) where a sinusoidal magnetic flux distribution that forms one cycle between the south pole and the north pole is formed. This allows the detection coil 13 to have magnetic characteristics that form a sinusoidal magnetic flux distribution with little distortion between the south pole and the north pole. Therefore, a more appropriate detection signal according to the rotation angle of the rotor 3 can be obtained, thereby further improving the detection accuracy of the resolver 1.

[0092] (3) In the resolver 1 described above, the circumferential positions through which the multiple connection wires Wfc of the upper layer coil 16 pass are out of phase with each other, and the circumferential positions through which the multiple connection wires Wbc of the lower layer coil 17 pass are out of phase with each other. This allows the magnetic flux of the detection coil 13 to change more finely (in more steps) in the circumferential direction, and therefore allows the detection coil 13 to have magnetic characteristics that form a magnetic flux distribution with less distortion.

[0093] The resolver 1 described above is configured by combining two excitation coils 11, 12 with little distortion in the magnetic flux distribution and suppressed impedance difference, and a detection coil 13 with magnetic characteristics that form a magnetic flux distribution with little distortion, so that when an AC signal is input from the signal processing circuit 4, a change in interlinkage magnetic flux (a change in induced voltage) according to the rotation angle of the rotor 3 is obtained on the detection coil 13 side. As a result, an appropriate detection signal according to the rotation angle of the rotor 3 can be obtained, thereby improving the detection accuracy of the resolver 1.

[0094] [5. Other] The configuration of the resolver 1 described above is an example and is not limited to the above configuration. While the resolver 1 described above has an axial multiplier of 1X, the resolver 1 may have an axial multiplier of kX (k is a natural number greater than or equal to 2). In other words, the detection coil 13 may not be a coil having an axial multiplier of 1X with magnetic properties that form one magnetic pole pair, but may be a coil having an axial multiplier of kX with magnetic properties that form k magnetic pole pairs. In this case, each of the upper layer coil 16 and the lower layer coil 17 may have k outer peripheral portions 16o, 17o and k inner peripheral portions 16i, 17i.

[0095] The upper layer conductor Wf forming the upper layer coil 16 may be wound in the second circumferential direction Dc2 in the second upper layer Lf2, and the lower layer conductor Wb forming the lower layer coil 17 may be wound in the first circumferential direction Dc1 in the second lower layer Lb2. Furthermore, the first circumferential direction Dc1 and the second circumferential direction Dc2 in the resolver 1 described above are merely examples, and these directions may be reversed. In other words, the "first coil" and the "second coil" described in the claims only need to be formed by conductors that are arranged in different layers and wound in different circumferential directions.

[0096] The second substrate 31 on which the detection coil 13 is arranged may have three or more layers stacked in the axial direction instead of a two-layer structure. Also, the substrate on which the detection coil 13 is arranged may be the first substrate 21 of the stator 2 instead of the second substrate 31 of the rotor 3. A well-known excitation coil configuration may be applied to the excitation coil provided in the resolver 1 instead of the configuration of the excitation coils 11 and 12 described above.

[0097] The "coil" recited in the claims is not limited to a detection coil and may also be an excitation coil. The "resolver" recited in the claims may not be a two-input one-output type resolver but may be a one-input two-output type resolver. Furthermore, the "resolver" recited in the claims may be one in which the rotor does not have a coil (a so-called inductive sensor). In the case of this resolver, both an excitation coil and a detection coil are provided in the stator, and the rotor may be provided with a conductor that generates a demagnetizing field in a direction that cancels out the magnetic field of the excitation coil, with a magnitude corresponding to the rotation angle. The "coil" recited in the claims may be applied as either the excitation coil or the detection coil of such a resolver.

[0098] REFERENCE SIGNS LIST 1 resolver 2 stator 3 rotor 13 detection coil (coil) 16 upper layer coil (first coil) 16i upper layer inner circumferential portion (inner circumferential portion) 16o upper layer outer circumferential portion (outer circumferential portion) 17 lower layer coil (second coil) 17i lower layer inner circumferential portion (inner circumferential portion) 17o lower layer outer circumferential portion (outer circumferential portion) 31 second substrate (substrate) Dc1 first circumferential direction (first direction) Dc2 second circumferential direction (second direction) Lb2 second lower layer (second layer) Lf2 second upper layer (first layer) m number (number of windings) W13 conducting wire (conducting wire) Wbc connecting wire Wbi inner winding (winding) Wbo outer winding (winding) Wfc connecting wire Wfi inner winding (winding) Wfo outer winding (winding)

Claims

1. A resolver that detects a rotation angle of a rotor relative to a stator, comprising: a sheet-like substrate provided on the stator or the rotor, having a first layer and a second layer stacked in the axial direction; and a coil provided on the substrate, wherein the coil has a substantially annular first coil formed by a conductor wound in a first circumferential direction in the first layer, and a substantially annular second coil formed by the conductor wound in a second direction opposite to the first direction in the second layer, each of the first coil and the second coil having an outer circumferential portion extending in the circumferential direction and an inner circumferential portion extending in the circumferential direction radially inward of the outer circumferential portion, and the coil forms magnetic poles of mutually different orientations between the outer circumferential portion of the first coil and the inner circumferential portion of the second coil, and between the inner circumferential portion of the first coil and the outer circumferential portion of the second coil, each of the outer peripheral portion and the inner peripheral portion is formed by a plurality of windings of the conductor that extend in the first direction and the second direction based on the center of a magnetic pole formed by each of the outer peripheral portion and the inner peripheral portion, and the number of windings decreases as the distance from the center of the magnetic pole increases.

2. The resolver according to claim 1, wherein each of the first coil and the second coil is provided with a plurality of connection wires that connect the plurality of windings in the outer circumferential portion with the plurality of windings in the inner circumferential portion, and the plurality of connection wires are set to pass through circumferential positions that form a sinusoidal magnetic flux distribution that forms one cycle between the magnetic poles that are oriented in different directions.

3. The resolver according to claim 2, wherein the positions through which the plurality of connecting wires of the first coil pass and the positions through which the plurality of connecting wires of the second coil pass are out of phase with each other.

Images