CN112655137B - Rotary electric machine - Google Patents

Abstract

A rotary electric machine (10) is provided with an excitation element (40) and an armature (50) having a magnet portion (42) in which an easy axis is oriented in an arc shape, and in which arc-shaped magnets (91, 92) of a magnet magnetic path having an intrinsic coercive force of 400[ kA/m ] or more and a residual magnetic flux density of 1.0T or more are formed along the easy axis, the magnet magnetic path including a magnetic path on an Oriented Arc (OA) centered on a center point (0) set on a q-axis and passing through a first intersection point (P1) of a d-axis and an armature-side peripheral surface on the armature side among peripheral surfaces of the magnet, the excitation element including a soft magnetic body, namely an excitation element core member (43) at a position on the opposite side of the armature from the magnet portion, the excitation element core member and the magnets being laminated in a radial direction, and a part or all of the excitation element core member being arranged at a position on the armature side closer to the intersection point than a second intersection point (P2) of the q-axis and the orientation in the radial direction.

Description

Rotary electric machine

Citation of related application

The present application is based on Japanese patent application No. 2018-166444, filed on 5/9/2018, the contents of which are incorporated herein by reference.

Technical Field

The present disclosure relates to a rotating electrical machine.

Background

Conventionally, as a rotating electrical machine, a rotor has been widely used, in which a magnet housing hole is formed in a rotor core formed by laminating electromagnetic steel plates, and a magnet is inserted into an IPM (Interior Permanent Magnet: interior permanent magnet) of the magnet housing hole. For example, patent document 1 discloses a technique of improving the shape of a magnet housing hole, suppressing a magnetic field in a direction opposite to a magnetic flux from a rotor toward a stator, and thereby increasing a magnetic flux interlinking with the stator. In the above-described rotating electrical machine, a design optimizing the shape of the permanent magnet, rotor, stator, etc. is realized, and at the same time, an improvement in the capability of the rotating electrical machine and an improvement in the resistance of the permanent magnet to the demagnetizing field are realized.

Prior art literature

Patent literature

Patent document 1: japanese patent laid-open publication No. 2014-093859

Disclosure of Invention

As a rotor of the rotating electrical machine, a rotor of an SPM (Surface Permanent Magnet: surface permanent magnet) type different from the above-described IPM type is also proposed. In the same manner as in the SPM rotor, the shape of the permanent magnet, the rotor, the stator, and the like is designed to suppress a magnetic field in a direction opposite to a magnetic flux from the rotor toward the stator and to increase a magnetic flux interlinking with the stator.

In this case, in order to increase the coercive force of the permanent magnet at high temperature, terbium (Tb) and dysprosium (Dy), which are expensive heavy rare earths, are sometimes used for the magnet, but the magnet thickness tends to increase, and the magnet amount tends to increase. Therefore, from the viewpoint of cost, it is desirable to reduce the amount of magnets.

The present disclosure has been made in view of the above-described technical problems, and an object thereof is to provide a rotary electric machine capable of increasing magnetic flux density and reducing the amount of magnets.

The various modes disclosed in the present specification adopt mutually different technical means to achieve the respective purposes. The objects, features and effects disclosed in the present specification can be more clearly understood with reference to the following detailed description and the accompanying drawings.

One mode is a rotary electric machine,

comprises a field element having a magnet portion including a plurality of magnetic poles having polarities alternating in the circumferential direction, and an armature having a plurality of armature windings of phases, wherein either one of the field element and the armature is set as a rotor,

the magnet part has a magnet with an easy axis of magnetization oriented in an arc shape and an arc-shaped magnet magnetic circuit formed along the easy axis of magnetization,

in the magnet part, the intrinsic coercive force is 400[ kA/m ] or more, and the residual magnetic flux density is 1.0[ T ] or more,

The magnet magnetic circuit includes a magnetic circuit on an orientation circular arc which is centered on a center point set on a magnetic pole boundary of the magnet, i.e., q-axis, and passes through a magnetic pole center of the magnet, i.e., d-axis, and a first intersection point of an armature-side peripheral surface on the armature side of the peripheral surface of the magnet,

the exciting element includes a soft magnetic body, namely an exciting element core member, at a position on the opposite side of the armature from the magnet portion,

the field element core member and the magnet are laminated in a radial direction,

a part or all of the excitation element core member is arranged radially closer to the armature side than a second intersection of the q-axis and the orientation circular arc.

Since the magnetic flux density distribution is close to a sinusoidal wave, torque enhancement can be achieved, and since the magnetic flux change is gentle than that of the radial magnet, eddy current loss can be suppressed. In addition, torque ripple can be reduced. When the intrinsic coercive force of the magnet portion is 400[ ka/m ] or more and the residual magnetic flux density is 1.0[ t ] or more, it is preferable to use a magnet having an easy axis of magnetization oriented along an orientation circular arc and forming a circular arc-shaped magnet magnetic path along the easy axis of magnetization in order to provide the magnet portion having a magnetic flux density distribution close to a sine wave.

When the magnet portion is used, in order to suppress leakage of magnetic flux from the armature opposite side of the magnet portion, the thickness dimension of the magnet is preferably designed such that the magnet is disposed up to the second intersection point P2 of the q-axis and the orientation circular arc in the radial direction. However, when the intrinsic coercive force is 400[ kA/m ] or more and the residual magnetic flux density is 1.0[ T ] or more, expensive rare earth such as terbium (Tb) or dysprosium (Dy) needs to be used for the magnet, which causes a problem in terms of cost.

Therefore, the field element core member and the magnet are laminated in the radial direction, and a part or all of the field element core member is arranged at a position closer to the armature side (i.e., the magnet portion side) than the second intersection point of the q-axis and the orientation circular arc in the radial direction. That is, the thickness dimension of the magnet is reduced, and a soft magnetic material, that is, an excitation element core member is additionally disposed. Even if the magnet is thinned as described above, the magnetic flux passes through the soft magnetic material, that is, the exciting element core member, and therefore leakage of the magnetic flux can be suppressed. That is, the magnetic flux density is hard to decrease at the d-axis. As described above, the amount of magnets can be reduced without reducing the magnetic flux density.

In the second aspect, the orientation circular arc is set such that a tangent line at a first intersection point on the orientation circular arc is parallel to the d-axis.

When the orientation circular arc is set such that a tangent line at the first intersection point on the orientation circular arc is parallel to the d-axis, the easy axis is oriented along the orientation circular arc, and an arc-shaped magnetic circuit of the magnet is formed along the easy axis, the magnetic flux density is maximum at the d-axis. That is, since the magnetic circuit of the magnet is orthogonal to the armature-side circumferential surface at the first intersection, the magnetic flux density at the d-axis can be increased. Since the torque is related to the magnetic flux density at the d-axis, the torque can be increased by increasing the magnetic flux density at the d-axis.

In the third aspect, when the saturation magnetic flux density of the field element core member is greater than the residual magnetic flux density of the magnet, the thickness dimension in the radial direction of the field element core member is smaller than the dimension in the radial direction from a third intersection point of the q-axis and the armature opposite side circumferential surface on the armature opposite side of the circumferential surface of the magnet to the second intersection point.

When the saturation magnetic flux density of the exciting element core member is larger than the residual magnetic flux density of the magnet, the magnetic flux leakage can be appropriately suppressed even if the exciting element core member is replaced with a soft magnetic body having a thickness smaller than the magnet. Therefore, the leakage of magnetic flux can be appropriately suppressed while thinning. That is, the magnetic flux density at the d-axis can be increased, thereby improving the torque.

In any one of the first to third aspects, a magnet and an excitation element core member as follows are used: when the residual magnetic flux density of the magnet portion is Br, the saturation magnetic flux density of the field element core member is Bs, the distance from the center point to the first intersection point is Wh, and the thickness dimension of the field element core member in the radial direction is Wsc, the relationship of br×wh and bs×wsc is satisfied.

As long as the above-described relationship is satisfied, it is theoretically possible to prevent the magnetic flux from leaking from the opposite side of the armature. That is, the magnetic flux density at the d-axis can be increased, thereby improving the torque.

Mode five in any one of modes one to four,

the armature winding has a wire portion arranged at a position facing the exciting element at a predetermined interval in a circumferential direction,

in the case of the armature being a hollow,

a wire-to-wire member is provided between the wire portions in the circumferential direction, and as the wire-to-wire member, a magnetic material or a non-magnetic material satisfying the relationship of Wt×Bs.ltoreq.Wm×Br is used, where Wt is the width dimension in the circumferential direction of the wire-to-wire member of one magnetic pole, bs is the saturation magnetic flux density of the wire-to-wire member, wm is the width dimension in the circumferential direction of the magnet portion of one magnetic pole, and Br is the residual magnetic flux density of the magnet portion,

Alternatively, no inter-conductor member is provided between the conductor portions in the circumferential direction.

Thus, even if the magnetic flux density from the magnet portion becomes large, torque restriction due to magnetic saturation can be eliminated.

Mode six in any one of modes one to five,

the armature winding has a wire portion arranged at a position facing the exciting element at a predetermined interval in a circumferential direction,

each wire constituting the wire portion is a wire assembly as follows: bundling multiple strands of wire and having a resistance between the bundled wires greater than the resistance of the wire itself

Thus, even if the magnetic flux density from the magnet portion becomes large, eddy current loss at the wire portion can be suppressed. Further, since the magnetic flux density distribution from the magnet portion approximates to a sine wave, eddy current loss at the wire portion can be further suppressed.

In a seventh aspect, in any one of the first to sixth aspects, the armature winding includes a wire portion disposed at a position facing the exciting element at a predetermined interval in a circumferential direction,

the radial thickness dimension of the wire portion is smaller than the circumferential width dimension corresponding to one of the magnetic poles.

Thereby, torque can be increased and eddy current loss at the wire portion can be suppressed.

Drawings

The above objects, other objects, features and advantages of the present disclosure will become more apparent by reference to the accompanying drawings and the following detailed description. The drawings are as follows.

Fig. 1 is a longitudinal sectional perspective view of a rotary electric machine.

Fig. 2 is a longitudinal sectional view of the rotary electric machine.

Fig. 3 is a sectional view taken along line III-III of fig. 2.

Fig. 4 is a sectional view showing a part of fig. 3 in an enlarged manner.

Fig. 5 is an exploded view of the rotary electric machine.

Fig. 6 is an exploded view of the inverter unit.

Fig. 7 is a torque diagram showing a relationship between ampere-turns of a stator winding and torque density.

Fig. 8 is a cross-sectional view of a rotor and stator.

Fig. 9 is an enlarged view showing a part of fig. 8.

Fig. 10 is a cross-sectional view of a stator.

Fig. 11 is a longitudinal sectional view of the stator.

Fig. 12 is a perspective view of a stator winding.

Fig. 13 is a perspective view showing the structure of a wire.

Fig. 14 is a schematic view showing the structure of a wire rod.

Fig. 15 is a diagram showing the form of each wire of the n-th layer.

Fig. 16 is a side view showing the wires of the n-th layer and the n+1-th layer.

Fig. 17 is a diagram showing a relationship between an electric angle and a magnetic flux density of the magnet according to the embodiment.

Fig. 18 is a graph showing a relationship between an electric angle and a magnetic flux density of the magnet of the comparative example.

Fig. 19 is a circuit diagram of a control system of the rotary electric machine.

Fig. 20 is a functional block diagram showing a current feedback control process of the control device.

Fig. 21 is a functional block diagram showing torque feedback control processing of the control device.

Fig. 22 is a diagram schematically showing a comparative example of the magnet unit.

Fig. 23 is a diagram schematically showing a magnet unit.

Fig. 24 is a cross-sectional view showing the magnet unit.

Fig. 25 is a cross-sectional view of a rotor and a stator of the second embodiment.

Fig. 26 is an enlarged view showing a part of fig. 25.

Fig. 27 is a diagram specifically showing the flow of magnetic flux of the magnet unit.

Fig. 28 is a cross-sectional view of the stator of modification 1.

Fig. 29 is a sectional view of the stator of modification 1.

Fig. 30 is a cross-sectional view of a stator of modification 2.

Fig. 31 is a sectional view of a stator of modification 3.

Fig. 32 is a cross-sectional view of the stator of modification 4.

Fig. 33 is a cross-sectional view of a rotor and a stator of modification 7.

Fig. 34 is a functional block diagram showing a part of the processing of the operation signal generation unit in modification 8.

Fig. 35 is a flowchart showing steps of the carrier frequency changing process.

Fig. 36 is a diagram showing a connection method of each wire constituting the wire group in modification 9.

Fig. 37 is a diagram showing a structure in which four pairs of wires are stacked in modification 9.

Fig. 38 is a cross-sectional view of an inner rotor type rotor and stator according to modification 10.

Fig. 39 is an enlarged view showing a part of fig. 38.

Fig. 40 is a longitudinal sectional view of the inner rotor type rotary electric machine.

Fig. 41 is a longitudinal sectional view showing a schematic structure of an inner rotor type rotary electric machine.

Fig. 42 is a diagram showing a structure of a rotary electric machine having an inner rotor structure in modification 11.

Fig. 43 is a diagram showing a structure of a rotary electric machine having an inner rotor structure in modification 11.

Fig. 44 is a diagram showing a structure of a rotary armature type rotary electric machine according to modification 12.

Fig. 45 is a cross-sectional view showing the structure of the lead wire of modification 14.

Fig. 46 is a diagram showing the relationship among reluctance torque, magnet torque, and DM.

Fig. 47 is a view showing a tooth.

Fig. 48 is a cross-sectional view of a magnet unit showing another example.

Detailed Description

Hereinafter, a plurality of embodiments will be described with reference to the drawings. In various embodiments, functionally and/or structurally corresponding parts and/or associated parts may be denoted by the same reference numerals or by reference numerals differing by more than hundred. Reference may be made to the description of other embodiments for corresponding parts and/or associated parts.

The rotary electric machine in the present embodiment is used as a vehicle power source, for example. However, the rotating electric machine is widely used for industrial use, vehicles, home appliances, OA equipment, game machines, and the like. In the following embodiments, the same or equivalent portions are denoted by the same reference numerals in the drawings, and the description thereof is given to the portions with the same reference numerals.

(first embodiment)

The rotary electric machine 10 of the present embodiment is a synchronous multiphase ac motor, and has an outer rotor structure (outer rotor structure). Fig. 1 to 5 show an outline of a rotary electric machine 10. Fig. 1 is a longitudinal sectional perspective view of a rotary electric machine 10, fig. 2 is a longitudinal sectional view of the rotary electric machine 10 along a direction of a rotary shaft 11, fig. 3 is a cross sectional view (a sectional view taken along line III-III of fig. 2) of the rotary electric machine 10 in a direction orthogonal to the rotary shaft 11, fig. 4 is a sectional view showing a part of fig. 3 in an enlarged manner, and fig. 5 is an exploded view of the rotary electric machine 10. In fig. 3, for convenience of illustration, a hatching showing a cut surface is omitted except for the rotary shaft 11. In the following description, the direction in which the rotary shaft 11 extends is referred to as an axial direction, the direction in which the rotary shaft 11 radially extends from the center is referred to as a radial direction, and the direction in which the rotary shaft 11 circumferentially extends around the center is referred to as a circumferential direction.

The rotary electric machine 10 generally includes a bearing unit 20, a housing 30, a rotor 40, a stator 50, and an inverter unit 60. The above-described members are coaxially arranged together with the rotary shaft 11, and are assembled in the axial direction in a predetermined order to construct the rotary electric machine 10. The rotary electric machine 10 of the present embodiment has a structure including a rotor 40 serving as an "exciting element" and a stator 50 serving as an "armature", and is embodied as a rotary excitation type rotary electric machine.

The bearing unit 20 has: two bearings 21, 22 arranged apart from each other in the axial direction; and a holding member 23 for holding the bearings 21 and 22. The bearings 21, 22 are, for example, radial ball bearings, and each have: an outer race 25, an inner race 26, and a plurality of balls 27 disposed between the outer race 25 and the inner race 26. The holding member 23 is cylindrical, and the bearings 21 and 22 are assembled radially inside the holding member 23. The rotary shaft 11 and the rotor 40 are rotatably supported on the radial inner sides of the bearings 21 and 22. A set of bearings 21 and 22 rotatably support the rotary shaft 11.

In each of the bearings 21 and 22, the balls 27 are held by a stopper, not shown, and the distance between the balls is maintained in this state. The bearings 21, 22 have sealing members at the upper and lower axial portions of the stopper, and are filled with a nonconductive grease (for example, a nonconductive urea grease) inside thereof. Further, the position of the inner race 26 is mechanically held by the spacer, and constant pressure preload protruding in the up-down direction is applied from the inside.

The housing 30 has a cylindrical peripheral wall 31. The peripheral wall 31 has first and second ends opposite in the axial direction thereof. The peripheral wall 31 has an end face 32 at a first end and an opening 33 at a second end. The opening 33 is open at the entire second end. A circular hole 34 is formed in the center of the end surface 32, and the bearing unit 20 is fixed by a fixing member such as a screw or a rivet in a state of being inserted into the hole 34. In addition, a hollow cylindrical rotor 40 and a hollow cylindrical stator 50 are housed in the housing 30, that is, in an inner space defined by the peripheral wall 31 and the end face 32. In the present embodiment, the rotary electric machine 10 is of an outer rotor type, and a stator 50 is disposed inside the casing 30 in the radial direction of the cylindrical rotor 40. The rotor 40 is supported by the rotary shaft 11 in a cantilever manner on the end face 32 side in the axial direction.

The rotor 40 has: a magnet holder 41 formed in a hollow cylindrical shape; and an annular magnet unit 42 provided on the radially inner side of the magnet holder 41. The magnet holder 41 has a substantially cup-like shape and functions as a magnet holding member. The magnet holder 41 has: a cylindrical portion 43 having a cylindrical shape; a fixed portion (connecting portion) 44 which is also cylindrical and has a smaller diameter than the cylindrical portion 43; and an intermediate portion 45 as a portion connecting the cylindrical portion 43 and the fixing portion 44. A magnet unit 42 is attached to the inner peripheral surface of the cylindrical portion 43.

The magnet holder 41 is made of a cold rolled steel Sheet (SPCC), forged steel, carbon Fiber Reinforced Plastic (CFRP), or the like, which has sufficient mechanical strength.

The rotary shaft 11 is inserted through the through hole 44a of the fixing portion 44. The fixing portion 44 is fixed to the rotary shaft 11 provided in the through hole 44 a. That is, the magnet holder 41 is fixed to the rotary shaft 11 by the fixing portion 44. The fixing portion 44 is preferably fixed to the rotary shaft 11 by spline coupling using projections and depressions, key coupling, welding, caulking, or the like. Thereby, the rotor 40 and the rotary shaft 11 integrally rotate.

Further, the bearings 21, 22 of the bearing unit 20 are assembled to the radially outer side of the fixing portion 44. As described above, since the bearing unit 20 is fixed to the end face 32 of the housing 30, the rotary shaft 11 and the rotor 40 are rotatably supported by the housing 30. Thereby, the rotor 40 is rotatable in the housing 30.

The rotor 40 is provided with a fixing portion 44 at only one of the two axially opposite ends thereof, whereby the rotor 40 is cantilever-supported by the rotary shaft 11. Here, the fixing portion 44 of the rotor 40 is rotatably supported by the bearings 21, 22 of the bearing unit 20 at two different positions in the axial direction. That is, the rotor 40 is rotatably supported by the two bearings 21 and 22 spaced apart in the axial direction at one of the two end portions of the magnet holder 41 facing each other in the axial direction. Therefore, even in a structure in which the rotor 40 is cantilever-supported by the rotary shaft 11, stable rotation of the rotor 40 can be achieved. In this case, the rotor 40 is supported by the bearings 21, 22 at a position of the rotor 40 offset to one side with respect to the axial center position.

In the bearing unit 20, the gap sizes between the outer race 25, the inner race 26, and the balls 27 are different for the bearing 22 near the center (lower side in the figure) of the rotor 40 and the bearing 21 on the opposite side (upper side in the figure), and for example, the bearing 22 near the center of the rotor 40 is larger in the gap size than the bearing 21 on the opposite side. In this case, even if vibration due to vibration of the rotor 40 or unbalance due to component tolerance acts on the bearing unit 20 on the side close to the center of the rotor 40, the influence of the vibration or the vibration can be well absorbed. Specifically, by increasing the play size (gap size) by preload in the bearing 22 near the center (lower side in the drawing) of the rotor 40, the vibration generated in the cantilever structure can be absorbed by the above-described play portion. The pre-load may be any of a constant position pre-load or a constant pressure pre-load. In the case of constant-position preload, both the bearing 21 and the outer race 25 of the bearing 22 are joined to the holding member 23 using a press-in or bonding method or the like. Further, the bearing 21 and the inner race 26 of the bearing 22 are both joined to the rotary shaft 11 by press-fitting or bonding. Here, the preload can be generated by disposing the outer race 25 of the bearing 21 at a position different from the inner race 26 of the bearing 21 in the axial direction. The preload can also be generated by disposing the outer race 25 of the bearing 22 at a position different from the inner race 26 of the bearing 22 in the axial direction.

In the case of using a constant pressure preload, a preload spring such as a wave washer 24 is disposed in the region between the bearing 22 and the bearing 21 so as to generate preload in the axial direction from the region between the bearing 22 and the bearing 21 toward the outer race 25 of the bearing 22. In this case, the inner rings 26 of the bearing 21 and the bearing 22 are both joined to the rotary shaft 11 by press-fitting or bonding. The bearing 21 or the outer race 25 of the bearing 22 is disposed on the holding member 23 with a predetermined gap. With the above configuration, the elastic force of the preload spring acts on the outer race 25 of the bearing 22 in a direction away from the bearing 21. Then, the force is transmitted to the rotary shaft 11, and a force is applied to press the inner race 26 of the bearing 21 in the direction of the bearing 22. Thereby, the axial positions of the outer race 25 and the inner race 26 are offset for both bearings 21, 22, so that the preload can be applied to both bearings as in the constant position preload described above.

In addition, when the constant pressure preload is generated, it is not necessary to apply elastic force to the outer race 25 of the bearing 22 as shown in fig. 2. For example, the outer ring 25 of the bearing 21 may be elastically biased. The preload may be applied to both bearings by disposing the inner race 26 of either one of the bearings 21 and 22 on the rotary shaft 11 with a predetermined gap therebetween, and joining the outer race 25 of the bearing 21 and 22 to the holding member 23 by a press-fitting or bonding method.

In addition, when a force is applied to move the inner race 26 of the bearing 21 away from the bearing 22, it is preferable that a force is applied to move the inner race 26 of the bearing 22 away from the bearing 21 as well. Conversely, when force is applied to bring the inner race 26 of the bearing 21 closer to the bearing 22, it is preferable to apply force to bring the inner race 26 of the bearing 22 closer to the bearing 21.

In addition, when the rotary electric machine 10 is applied to a vehicle for the purpose of a vehicle power source or the like, there is a possibility that vibration having a component of the direction of occurrence of the preload is applied to a mechanism for generating the preload, and the direction of gravity of an object to which the preload is applied may vary. Therefore, in the case where the present rotary electric machine 10 is applied to a vehicle, it is desirable to employ constant position pre-pressing.

Further, the intermediate portion 45 has an annular inner shoulder 49a and an annular outer shoulder 49b. The outer shoulder 49b is located radially outward of the inner shoulder 49a in the intermediate portion 45. The inner shoulder 49a and the outer shoulder 49b are separated from each other in the axial direction of the intermediate portion 45. Thus, the cylindrical portion 43 and the fixing portion 44 partially overlap in the radial direction of the intermediate portion 45. That is, the cylindrical portion 43 protrudes axially outward from the base end portion (lower rear end portion in the figure) of the fixing portion 44. According to this configuration, the rotor 40 can be supported by the rotary shaft 11 at a position near the center of gravity of the rotor 40, and the stable operation of the rotor 40 can be achieved, as compared with the case where the intermediate portion 45 is formed in a flat plate shape without steps.

According to the structure of the intermediate portion 45, in the rotor 40, the bearing accommodating recess 46 that accommodates a part of the bearing unit 20 is formed annularly at a position that surrounds the fixing portion 44 in the radial direction and is close to the inside of the intermediate portion 45, and the coil accommodating recess 47 that surrounds the bearing accommodating recess 46 in the radial direction and is close to the outside of the intermediate portion 45 is formed to accommodate a coil side end portion 54 of the stator winding 51 of the stator 50 described later. The housing recesses 46 and 47 are disposed adjacent to each other in the radial direction. That is, a part of the bearing unit 20 and the coil side end 54 of the stator winding 51 are arranged to overlap each other radially inward and outward. This can shorten the axial length of the rotary electric machine 10.

The intermediate portion 45 is provided to protrude radially outward from the rotating shaft 11 side. The intermediate portion 45 is provided with a contact avoiding portion extending in the axial direction, which avoids contact with the coil side end portion 54 of the stator winding 51 of the stator 50. The intermediate portion 45 corresponds to a protruding portion.

The axial dimension of the coil side end 54 can be reduced by bending the coil side end 54 radially inward or outward, and the axial length of the stator 50 can be shortened. The bending direction of the coil side end 54 is preferably considered for assembly with the rotor 40. If the stator 50 is assembled to the radially inner side of the rotor 40, the coil side end 54 is preferably bent radially inward on the insertion tip side of the rotor 40. The bending direction of the coil side end portion on the opposite side to the coil side end portion 54 may be arbitrary, but a shape that is bent to the outside of a space margin is preferable for manufacturing reasons.

Further, the magnet unit 42 as a magnet portion is constituted by a plurality of permanent magnets arranged to alternately change the polarity in the circumferential direction on the inner side in the radial direction of the cylindrical portion 43. Thereby, the magnet unit 42 has a plurality of magnetic poles in the circumferential direction. The magnet unit 42 is described in detail later.

The stator 50 is provided radially inward of the rotor 40. The stator 50 includes a stator winding 51 wound in a substantially cylindrical shape (annular shape) and a stator core 52 as a base member disposed radially inward, and the stator winding 51 is disposed so as to face the annular magnet unit 42 with a predetermined air gap therebetween. The stator winding 51 is constituted by a plurality of phase windings. The phase windings are configured by connecting a plurality of wires arranged in the circumferential direction to each other at a predetermined pitch. In the present embodiment, the stator winding 51 is configured as a six-phase winding by using three-phase windings of U-phase, V-phase, and W-phase and three-phase windings of X-phase, Y-phase, and Z-phase and using two sets of the above-described three-phase windings.

The stator core 52 is formed in a circular ring shape from laminated steel plates in which electromagnetic steel plates, which are soft magnetic materials, are laminated, and is assembled radially inward of the stator winding 51. The electromagnetic steel sheet is, for example, a silicon steel sheet to which about several percent (e.g., 3%) of silicon is added to iron. The stator winding 51 corresponds to an armature winding, and the stator core 52 corresponds to an armature core.

The stator winding 51 has a coil side portion 53 and coil side end portions 54 and 55, the coil side portion 53 being a portion overlapping the stator core 52 in the radial direction and being located radially outside the stator core 52, the coil side end portions 54 and 55 protruding toward one end side and the other end side of the stator core 52 in the axial direction, respectively. The coil side portions 53 are radially opposed to the stator core 52 and the magnet units 42 of the rotor 40, respectively. In a state where the stator 50 is disposed inside the rotor 40, the coil side end 54 located on the bearing unit 20 side (upper side in the drawing) among the coil side end 54, 55 on both sides in the axial direction is accommodated in the coil accommodating recess 47 formed by the magnet holder 41 of the rotor 40. The stator 50 is described in detail later.

The inverter unit 60 has: a unit base 61 fixed to the housing 30 by a fastener such as a bolt; and a plurality of electrical components 62 assembled to the unit base 61. The unit base 61 is made of, for example, carbon Fiber Reinforced Plastic (CFRP). The unit base 61 includes: an end plate 63 fixed to an edge of the opening 33 of the housing 30; and a housing 64 integrally provided on the end plate 63 and extending in the axial direction. The end plate 63 has a circular opening 65 at its center, and a housing 64 is formed so as to stand up from a peripheral edge portion of the opening 65.

The stator 50 is assembled to the outer peripheral surface of the housing 64. That is, the outer diameter dimension of the housing 64 is the same as the inner diameter dimension of the stator core 52 or slightly smaller than the inner diameter dimension of the stator core 52. The stator 50 and the unit base 61 are integrated by assembling the stator core 52 to the outside of the housing 64. Further, since the unit base 61 is fixed to the housing 30, the stator 50 is in a state of being integrated with the housing 30 in a state where the stator core 52 is assembled to the case 64.

The stator core 52 is preferably assembled to the unit base 61 by bonding, press-fitting, or the like. Thereby, the stator core 52 is restrained from being displaced in the circumferential or axial direction with respect to the unit base 61 side.

The radially inner side of the housing 64 is a housing space for housing the electrical component 62, and the electrical component 62 is disposed in the housing space so as to surround the rotary shaft 11. The case 64 functions as an accommodating space forming portion. The electrical component 62 is configured to include: a semiconductor module 66, a control board 67, and a capacitor module 68 that constitute an inverter circuit.

The unit base 61 corresponds to a stator holder (armature holder) that is provided on the radially inner side of the stator 50 and holds the stator 50. The motor housing of the rotary electric machine 10 is constituted by the housing 30 and the unit base 61. In the motor case, the holding member 23 is fixed to the case 30 on one side in the axial direction with the rotor 40 interposed therebetween, and the case 30 and the unit base 61 are coupled to each other on the other side. For example, in an electric vehicle or the like as an electric vehicle, the rotating electrical machine 10 is mounted on the vehicle or the like by mounting a motor case on one side of the vehicle or the like.

Here, in addition to the above-described fig. 1 to 5, the structure of the inverter unit 60 will be further described with reference to fig. 6, which is an exploded view of the inverter unit 60.

In the unit base 61, the housing 64 has a cylindrical portion 71 and an end surface 72, and the end surface 72 is provided at one of the opposite ends (end portion on the bearing unit 20 side) in the axial direction. Opposite sides of the end face 72 in the axial both end portions of the cylindrical portion 71 are fully opened through the opening 65 of the end plate 63. A circular hole 73 is formed in the center of the end surface 72, and the rotation shaft 11 is inserted through the hole 73. A seal 171 is provided in the hole 73, and the seal 171 closes a gap between the hole 73 and the outer circumferential surface of the rotary shaft 11. The seal 171 is preferably a sliding seal made of, for example, a resin material.

The cylindrical portion 71 of the housing 64 is a partition portion that separates the rotor 40 and the stator 50 disposed radially outward from the electrical component 62 disposed radially inward, and the rotor 40, the stator 50, and the electrical component 62 are disposed so as to be aligned radially inward and outward with the cylindrical portion 71 interposed therebetween.

The electric component 62 is an electric component constituting an inverter circuit, and has a power running function of flowing electric current to each phase winding of the stator winding 51 in a predetermined order to rotate the rotor 40, and a power generating function of inputting three-phase alternating current flowing to the stator winding 51 as generated electric power along with rotation of the rotary shaft 11 and outputting the three-phase alternating current to the outside. The electrical component 62 may have only one of the power running function and the power generating function. For example, when the rotary electric machine 10 is used as a power source for a vehicle, the power generation function is a regeneration function that outputs regenerative electric power to the outside.

As a specific configuration of the electrical component 62, as shown in fig. 4, a hollow cylindrical capacitor module 68 is provided around the rotary shaft 11, and a plurality of semiconductor modules 66 are arranged in parallel in the circumferential direction on the outer peripheral surface of the capacitor module 68. The capacitor module 68 includes a plurality of smoothing capacitors 68a connected in parallel with each other. Specifically, the capacitor 68a is a stacked film capacitor in which a plurality of film capacitors are stacked, and has a trapezoidal cross section. The capacitor module 68 is constituted by arranging twelve capacitors 68a in a ring-like arrangement.

In the manufacturing process of the capacitor 68a, for example, a long film having a predetermined width is formed by laminating a plurality of films, the width direction of the film is set to be the trapezoid height direction, and the long film is cut into isosceles trapezoids so that the upper and lower bases of the trapezoids alternate, thereby manufacturing the capacitor element. Then, the capacitor 68a is fabricated by attaching an electrode or the like to the capacitor element.

The semiconductor module 66 has a semiconductor switching element such as a MOSFET (Metal-Oxide-Semiconductor Field-Effect Transistor: metal-Oxide semiconductor field effect transistor) or an IGBT (Insulated Gate Bipolar Transistor: insulated gate bipolar transistor), and is formed in a substantially plate shape. In the present embodiment, the rotary electric machine 10 includes two sets of three-phase windings, and since an inverter circuit is provided for each three-phase winding, a semiconductor module group 66A formed by annularly arranging twelve semiconductor modules 66 in total is provided in the electric component 62.

The semiconductor module 66 is disposed in a state of being sandwiched between the cylindrical portion 71 of the case 64 and the capacitor module 68. The outer peripheral surface of the semiconductor module group 66A abuts against the inner peripheral surface of the cylindrical portion 71, and the inner peripheral surface of the semiconductor module group 66A abuts against the outer peripheral surface of the capacitor module 68. In this case, heat generated in the semiconductor module 66 is transferred to the end plate 63 through the case 64 and is released from the end plate 63.

The semiconductor module group 66A preferably has a spacer 69 between the semiconductor module 66 and the cylindrical portion 71 on the outer peripheral surface side, i.e., in the radial direction. In this case, the cross-sectional shape of the cross-section orthogonal to the axial direction in the capacitor module 68 is a regular dodecagon, and the cross-sectional shape of the inner peripheral surface of the cylindrical portion 71 is a circle, and therefore the inner peripheral surface of the spacer 69 is a flat surface and the outer peripheral surface is a curved surface. The spacers 69 may be integrally provided on the radially outer side of the semiconductor module group 66A so as to be connected in an annular shape. The spacer 69 is preferably a good heat conductor, such as a metal like aluminum or a heat dissipating gel sheet. The cross-sectional shape of the inner peripheral surface of the cylindrical portion 71 may be a dodecagon similar to that of the capacitor module 68. In this case, both the inner peripheral surface and the outer peripheral surface of the spacer 69 are preferably flat surfaces.

In the present embodiment, a cooling water passage 74 through which cooling water flows is formed in the cylindrical portion 71 of the case 64, and heat generated in the semiconductor module 66 is also released to the cooling water flowing through the cooling water passage 74. That is, the housing 64 includes a water cooling mechanism. As shown in fig. 3 and 4, the cooling water passage 74 is formed in a ring shape to surround the electrical components 62 (the semiconductor modules 66 and the capacitor modules 68). The semiconductor modules 66 are arranged along the inner peripheral surface of the cylindrical portion 71, and cooling water passages 74 are provided at positions overlapping the semiconductor modules 66 radially inward and outward.

Since the stator 50 is disposed outside the cylindrical portion 71 and the electrical component 62 is disposed inside, heat of the stator 50 is transferred from the outside to the cylindrical portion 71 and heat of the electrical component 62 (for example, heat of the semiconductor module 66) is transferred from the inside to the cylindrical portion 71. In this case, the stator 50 and the semiconductor module 66 can be cooled at the same time, so that the heat of the heat generating member in the rotary electric machine 10 can be efficiently released.

At least a part of the semiconductor modules 66 constituting a part or all of the inverter circuit for operating the rotary electric machine by energizing the stator winding 51 is disposed in a region surrounded by the stator core 52 disposed radially outside the cylindrical portion 71 of the housing 64. Preferably, the entirety of one semiconductor module 66 is disposed in the area surrounded by the stator core 52. It is further preferable that the entire semiconductor module 66 is disposed in the region surrounded by the stator core 52.

In addition, at least a portion of the semiconductor module 66 is disposed within the area surrounded by the cooling water passage 74. Preferably, the entire semiconductor modules 66 are disposed in the region surrounded by the yoke 141.

Further, the electrical assembly 62 includes, in the axial direction: an insulating sheet 75 provided on one end surface of the capacitor module 68; and a wiring module 76 provided on the other end face. In this case, the capacitor module 68 has two end faces, i.e., a first end face and a second end face, which are opposite in the above-described axial direction. The first end face of the capacitor module 68 near the bearing unit 20 is opposed to the end face 72 of the housing 64, and coincides with the end face 72 in a state of sandwiching the insulating sheet 75. The wiring module 76 is assembled to the second end surface of the capacitor module 68 near the opening 65.

The wiring module 76 has a circular plate-shaped main body portion 76a made of a synthetic resin material and a plurality of bus bars 76b and 76c embedded therein, and is electrically connected to the semiconductor module 66 and the capacitor module 68 via the bus bars 76b and 76 c. Specifically, the semiconductor module 66 has a connection pin 66a extending from the above-described axial end face, and the connection pin 66a is connected to the bus bar 76b radially outside the main body portion 76 a. The bus bar 76c extends to the opposite side of the capacitor module 68 from the radial outside of the main body 76a, and is connected to a wiring member 79 (see fig. 2) at the tip end portion thereof.

As described above, according to the structure in which the insulating sheet 75 is provided on the first end face of the capacitor module 68 that faces in the axial direction and the wiring module 76 is provided on the second end face of the capacitor module 68, a path from the first end face and the second end face of the capacitor module 68 to the end face 72 and the cylindrical portion 71 is formed as the heat dissipation path of the capacitor module 68. That is, a path from the first end face to the end face 72 and a path from the second end face to the cylindrical portion 71 are formed. This allows heat to be dissipated from the end surface portion of the capacitor module 68 other than the outer peripheral surface of the semiconductor module 66. That is, heat can be radiated not only in the radial direction but also in the axial direction.

Further, since the capacitor module 68 has a hollow cylindrical shape and the rotary shaft 11 is disposed at the inner peripheral portion thereof with a predetermined gap therebetween, heat of the capacitor module 68 can be released from the hollow portion. In this case, since the air flow is generated by the rotation of the rotation shaft 11, the above-described cooling effect is improved.

The disk-shaped control board 67 is mounted on the wiring module 76. The control board 67 has a Printed Circuit Board (PCB) formed with a predetermined wiring pattern, and a control device 77 corresponding to a control section formed of various ICs, microcomputers, and the like is mounted on the board. The control board 67 is fixed to the wiring module 76 by a fixing member such as a screw. The control board 67 has an insertion hole 67a in its central portion through which the rotation shaft 11 is inserted.

In addition, the wiring module 76 has a first face and a second face that are opposite to each other in the axial direction, i.e., in the thickness direction thereof. The first face faces the capacitor module 68. A control board 67 is provided on the second surface of the wiring module 76. The bus bar 76c of the wiring module 76 extends from one side to the other side of both sides of the control board 67. In the above configuration, the control board 67 is preferably provided with a notch for avoiding interference with the bus bar 76 c. For example, a part of the outer edge portion of the control board 67 having a circular shape is preferably cut off.

As described above, according to the structure in which the electrical component 62 is housed in the space surrounded by the case 64, and the housing 30, the rotor 40, and the stator 50 are provided in layers on the outside thereof, electromagnetic noise generated in the inverter circuit can be desirably shielded. That is, in the inverter circuit, the switching control in each semiconductor module 66 is performed by PWM control based on a predetermined carrier frequency, and the electromagnetic noise is desirably shielded by the housing 30, the rotor 40, the stator 50, and the like on the outer side in the radial direction of the electric component 62, although the switching control is considered to generate electromagnetic noise.

Further, by disposing at least a part of the semiconductor module 66 in the region surrounded by the stator core 52 disposed radially outside the cylindrical portion 71 of the case 64, it is difficult to affect the stator winding 51 even if magnetic flux is generated from the semiconductor module 66, as compared with a structure in which the semiconductor module 66 and the stator winding 51 are not disposed via the stator core 52. Further, even if magnetic flux is generated from the stator winding 51, it is difficult to affect the semiconductor module 66. Further, it is more effective to dispose the entire semiconductor module 66 in the region surrounded by the stator core 52 disposed radially outside the cylindrical portion 71 of the case 64. Further, when at least a part of the semiconductor module 66 is surrounded by the cooling water passage 74, the effect that the heat generated from the stator winding 51 and the magnet unit 42 hardly reaches the semiconductor module 66 can be obtained.

In the cylindrical portion 71, a through hole 78 is formed in the vicinity of the end plate 63, and the through hole 78 is inserted with a wiring member 79 (see fig. 2) that electrically connects the stator 50 on the outside of the cylindrical portion 71 and the electrical component 62 on the inside. As shown in fig. 2, the wiring members 79 are respectively connected to the ends of the stator windings 51 and the bus bars 76c of the wiring module 76 by crimping, welding, or the like. The wiring member 79 is, for example, a bus bar, and the junction surface thereof is preferably flattened. The through-hole 78 is preferably provided at one location or a plurality of locations, and in the present embodiment, the through-hole 78 is provided at two locations. According to the configuration in which the through holes 78 are provided at two positions, the winding terminals extending from the three-phase windings of the two groups can be easily connected by the wiring members 79, respectively, and thus the configuration is suitable for multi-phase connection.

As described above, in the housing 30, as shown in fig. 4, the rotor 40 and the stator 50 are provided in order from the radial outside, and the inverter unit 60 is provided on the radial inside of the stator 50. Here, when the radius of the inner peripheral surface of the housing 30 is d, the rotor 40 and the stator 50 are disposed radially outward of a distance d×0.705 from the rotation center of the rotor 40. In this case, if a region radially inward from the inner peripheral surface of the stator 50 (i.e., the inner peripheral surface of the stator core 52) of the rotor 40 and the stator 50 is defined as a first region X1, and a region radially between the inner peripheral surface of the stator 50 and the housing 30 is defined as a second region X2, the cross-sectional area of the first region X1 is larger than the cross-sectional area of the second region X2. Furthermore, the volume of the first region X1 is larger than the volume of the second region X2 as viewed in a range where the magnet unit 42 of the rotor 40 and the stator winding 51 overlap in the radial direction.

In addition, when the rotor 40 and the stator 50 are configured as a magnetic circuit assembly, the volume of the first region X1 radially inward from the inner peripheral surface of the magnetic circuit assembly is larger than the volume of the second region X2 radially between the inner peripheral surface of the magnetic circuit assembly and the housing 30 in the housing 30.

Next, the structures of the rotor 40 and the stator 50 are described in more detail.

In general, as a structure of a stator of a rotating electrical machine, a structure is known in which a plurality of slots are provided in a circumferential direction in a stator core made of laminated steel plates and having a circular shape, and stator windings are wound in the slots. Specifically, the stator core has a plurality of pole teeth extending radially from a yoke at predetermined intervals, and slots are formed between circumferentially adjacent pole teeth. In the slot, a plurality of layers of wires are housed, for example, in the radial direction, and the stator winding is constituted by the wires.

However, according to the above stator structure, when the stator winding is energized, magnetic saturation occurs in the pole tooth portion of the stator core with an increase in magnetomotive force of the stator winding, which is considered to cause a limitation in torque density of the rotary electric machine. That is, it is considered that in the stator core, the rotating magnetic flux generated by energization of the stator winding concentrates on the pole teeth, and magnetic saturation occurs.

In general, as a structure of an IPM (Interior Permanent Magnet: interior permanent magnet) rotor of a rotating electrical machine, a structure is known in which a permanent magnet is arranged on a d-axis in a d-q coordinate system and a rotor core is arranged on a q-axis. In this case, by exciting the stator winding near the d-axis, the excitation magnetic flux flows from the stator into the q-axis of the rotor according to fleming's law. And thus it is considered that a wide range of magnetic saturation is generated in the q-axis core portion of the rotor.

Fig. 7 is a torque diagram showing a relationship between ampere-turns AT representing magnetomotive force of a stator winding and torque density Nm/L. The broken line indicates the characteristic of a general IPM rotor type rotating electrical machine. As shown in fig. 7, in a general rotating electrical machine, an increase in torque is limited because magnetomotive force is increased in a stator to magnetically saturate two portions, namely a tooth portion and a q-axis core portion, between slots. Thus, in the above-described general rotary electric machine, the ampere-turns pattern value is limited to A1.

Therefore, in the present embodiment, in order to overcome the limitation caused by the magnetic saturation, the following structure is provided in the rotary electric machine 10. That is, as a first aspect, in order to eliminate magnetic saturation generated at the pole teeth of the stator core in the stator, a slot-free structure is employed in the stator 50, and in order to eliminate magnetic saturation generated at the q-axis core portion of the IPM rotor, an SPM (Surface Permanent Magnet: surface permanent magnet) rotor is employed. According to the first aspect, although the portions of the two portions where magnetic saturation occurs can be eliminated, it is considered that the torque in the low current region is reduced (see the chain line of fig. 7). Therefore, as a second aspect, in order to recover torque reduction by achieving flux enhancement of the SPM rotor, a polar anisotropic structure that increases magnetic force by lengthening a magnetic circuit of a magnet is adopted in the magnet unit 42 of the rotor 40.

Further, as a third aspect, a flat wire structure that reduces the radial thickness in the stator 50 of the wire is employed in the coil side 53 of the stator winding 51 to recover the reduction in torque. Here, it is considered that by the above-described polar anisotropic structure that improves the magnetic force, a larger eddy current is generated in the stator winding 51 opposing the magnet unit 42. However, according to the third aspect, since the flat wire structure is thin in the radial direction, generation of eddy current in the radial direction in the stator winding 51 can be suppressed. In this way, according to the above-described respective first to third configurations, as shown by the solid lines in fig. 7, a magnet having a high magnetic force can be used to achieve a great improvement in torque characteristics, and the concern that the magnet having a high magnetic force will cause a large eddy current can be reduced.

Further, as a fourth aspect, a magnet unit using a polar anisotropic structure and having a magnetic flux density distribution similar to a sine wave is employed. In this way, the torque can be enhanced by improving the sine wave matching rate by pulse control or the like described later, and the eddy current loss (copper loss due to eddy current: eddy current loss) can be further suppressed because the magnetic flux change is more gentle than that of the radial magnet.

The sine wave matching rate will be described below. The actual measurement waveform of the surface magnetic flux density distribution measured by tracking the surface of the magnet or the like with the magnetic flux probe may be compared with the sine wave of the same period and peak value to obtain the sine wave matching rate. The ratio of the amplitude of the primary waveform, which is the fundamental wave of the rotating electric machine, to the amplitude of the actual measurement waveform, that is, the amplitude of the other harmonic component to the fundamental wave is increased, which corresponds to the sine wave matching ratio. When the sine wave matching ratio becomes high, the waveform of the surface magnetic flux density distribution gradually approaches the sine wave shape. Further, when a current of a primary sine wave is supplied from an inverter to a rotating electrical machine including a magnet having an improved sine wave matching rate, it is possible to approximate the waveform of the surface magnetic flux density distribution of the magnet to the sine wave shape and generate a large torque. The surface magnetic flux density distribution may be estimated by a method other than actual measurement, for example, electromagnetic field analysis using maxwell's equations.

In addition, as a fifth aspect, the stator winding 51 is provided as a wire conductor structure in which a plurality of wires are gathered and bundled. Thus, since the wires are connected in parallel, a large current can flow, and since the sectional areas of the wires each become small, compared to the third aspect of thinning in the radial direction, the flat wire structure can be suppressed more effectively and eddy current can be generated at the wires that expand in the circumferential direction of the stator 50. Further, since the multi-strand wire is twisted, eddy current corresponding to magnetic flux generated by the right-hand rule with respect to the current flowing direction can be canceled against magnetomotive force from the conductor.

In this way, when the fourth and fifth aspects are further increased, it is possible to further suppress eddy current loss caused by the above-described higher magnetic force and realize torque enhancement while adopting the magnet of the second aspect having the higher magnetic force.

Hereinafter, the structure of the stator 50 without the slots, the flat wire structure of the stator winding 51, and the polar anisotropic structure of the magnet unit 42 will be described in detail. Here, the non-slot structure of the stator 50 and the flat wire structure of the stator winding 51 will be described first. Fig. 8 is a cross-sectional view of the rotor 40 and the stator 50, and fig. 9 is an enlarged view showing a part of the rotor 40 and the stator 50 shown in fig. 8. Fig. 10 is a cross-sectional view showing a cross section of the stator 50 along the X-X line of fig. 11, and fig. 11 is a cross-sectional view showing a longitudinal section of the stator 50. Fig. 12 is a perspective view of the stator winding 51. In addition, in fig. 8 and 9, the magnetization direction of the magnets in the magnet unit 42 is indicated by arrows.

As shown in fig. 8 to 11, the stator core 52 has a cylindrical shape in which a plurality of electromagnetic steel plates are stacked in the axial direction and a predetermined thickness is provided in the radial direction, and the stator winding 51 is assembled to the radially outer side of the stator core 52 on the rotor 40 side. In the stator core 52, the outer peripheral surface on the rotor 40 side is a lead wire installation portion (conductor region). The outer peripheral surface of the stator core 52 has a curved surface shape without irregularities, and a plurality of lead groups 81 are arranged on the outer peripheral surface at predetermined intervals in the circumferential direction. The stator core 52 functions as a back yoke that becomes a part of a magnetic circuit for rotating the rotor 40. In this case, there is no tooth (i.e., iron core) made of soft magnetic material between each of the two wire groups 81 adjacent in the circumferential direction (i.e., a non-slot structure). In the present embodiment, the resin material constituting the sealing member 57 enters the gap 56 of each of the lead groups 81. That is, in the stator 50, the inter-conductor members provided between the conductor groups 81 in the circumferential direction are configured as the sealing members 57 of a non-magnetic material. In the state before sealing of the seal member 57, the stator 50 having the non-slot structure is configured by disposing the lead groups 81 at predetermined intervals in the circumferential direction on the radially outer side of the stator core 52 so as to separate the gaps 56, which are areas between the leads. In other words, as will be described later, each of the wire groups 81 is composed of two wires (conductors) 82, and only the non-magnetic material occupies between each of the two wire groups 81 adjacent in the circumferential direction of the stator 50. The nonmagnetic material includes nonmagnetic gas such as air, nonmagnetic liquid, and the like, in addition to the sealing member 57. In the following, the sealing member 57 is also referred to as an inter-conductor-to-conductor member (conductor-to-conductor member).

The structure in which the pole teeth are provided between the lead groups 81 arranged in the circumferential direction is a structure in which the pole teeth have a predetermined thickness in the radial direction and a predetermined width in the circumferential direction, and a magnet magnetic circuit, which is a part of a magnetic circuit, is formed between the lead groups 81. In this regard, the structure in which no tooth is provided between the wire groups 81 means a structure in which the above-described magnetic circuit is not formed.

As shown in fig. 10, the stator winding (i.e., armature winding) 51 is formed to have a prescribed thickness T2 (hereinafter also referred to as a first dimension) and a width W2 (hereinafter also referred to as a second dimension). The thickness T2 is the shortest distance between the outer side face and the inner side face of the stator winding 51 that are radially opposite to each other. The width W2 is a length in the circumferential direction of the stator winding 51 of a part of the stator winding 51 functioning as one of the phases (three phases of the U-phase, V-phase, and W-phase or three phases of the X-phase, Y-phase, and Z-phase in the embodiment) of the stator winding 51. Specifically, in fig. 10, when two wire groups 81 adjacent in the circumferential direction function as, for example, U-phase of one of the three phases, the width W2 is the width from one end to the other end of the two wire groups 81 in the circumferential direction. Also, the thickness T2 is smaller than the width W2.

In addition, the thickness T2 is preferably smaller than the total width dimension of the two wire groups 81 existing within the width W2. Further, when the cross-sectional shape of the stator winding 51 (more specifically, the lead 82) is a perfect circle, an ellipse, or a polygon, the maximum length in the radial direction of the stator 50 in the cross-section of the lead 82 along the radial direction of the stator 50 may be set to W12, and the maximum length in the circumferential direction of the stator 50 in the cross-section may be set to W11.

As shown in fig. 10 and 11, the stator winding 51 is sealed by a sealing member 57, and the sealing member 57 is composed of a synthetic resin material as a sealing material (molding material). That is, the stator winding 51 and the stator core 52 are molded together by a molding material. In addition, as a nonmagnetic substance or an equivalent of a nonmagnetic substance, the resin can be regarded as bs=0.

As seen in the cross section of fig. 10, the sealing member 57 is provided by filling the gaps 56, which are the gaps 56, between the wire groups 81, and the insulating member is interposed between the wire groups 81 by the sealing member 57. That is, the seal member 57 functions as an insulating member in the gap 56. The seal member 57 is provided on the radially outer side of the stator core 52 in a range including all the wire groups 81, that is, in a range in which the radial thickness dimension is larger than the radial thickness dimension of each wire group 81.

In addition, the seal member 57 is provided in a range including the turning portion 84 of the stator winding 51 as viewed in the longitudinal section of fig. 11. Inside the stator winding 51 in the radial direction, the seal member 57 is provided in a range including at least a part of the axially opposite end faces of the stator core 52. In this case, the stator winding 51 is resin-sealed in the entire end portion of the phase winding except for the end portion of the phase winding, that is, except for the connection terminal connected to the inverter circuit.

According to the structure in which the seal member 57 is provided in the range including the end face of the stator core 52, the laminated steel plates of the stator core 52 can be pressed inward in the axial direction by the seal member 57. This makes it possible to maintain the stacked state of the steel plates using the seal member 57. In the present embodiment, the inner peripheral surface of the stator core 52 is not resin-sealed, but the entire stator core 52 including the inner peripheral surface of the stator core 52 may be resin-sealed.

When the rotary electric machine 10 is used as a vehicle power source, the sealing member 57 is preferably composed of a highly heat-resistant fluororesin, an epoxy resin, PPS resin, PEEK resin, LCP resin, silicone resin, PAI resin, PI resin, or the like. In addition, when the linear expansion coefficient is considered from the viewpoint of suppressing cracking due to the expansion difference, the same material as the outer film of the wire of the stator winding 51 is desirable. That is, it is desirable to exclude silicone resins having a linear expansion coefficient generally more than a multiple of that of other resins. In addition, PP0 resin, phenolic resin, and FRP resin, which have heat resistance of about 180 ℃ are also candidates for electric products such as electric vehicles, which do not have a device using combustion. In the field where the ambient temperature of the rotating electrical machine is regarded as lower than 100 ℃, there is no limitation as described above.

The torque of the rotary electric machine 10 is proportional to the magnitude of the magnetic flux. Here, when the stator core has pole teeth, the maximum magnetic flux at the stator depends on and is limited to the saturation magnetic flux density at the pole teeth, but when the stator core does not have pole teeth, the maximum magnetic flux at the stator is not limited. Therefore, it is advantageous in that the current flowing to the stator winding 51 is increased to achieve an increase in torque of the rotary electric machine 10.

In the present embodiment, the inductance of the stator 50 is reduced by using a structure without pole teeth (a non-slot structure) in the stator 50. Specifically, in the stator of a general rotating electrical machine in which a wire is housed in each slot divided by a plurality of pole teeth, the inductance is, for example, about 1mH, whereas in the stator 50 of the present embodiment, the inductance is reduced to about 5 to 60 μh. In the present embodiment, the rotating electrical machine 10 having the outer rotor structure can be provided, and the mechanical time constant Tm can be reduced by reducing the inductance of the stator 50. That is, the mechanical time constant Tm can be reduced while achieving high torque. In addition, the mechanical time constant Tm is calculated by the following equation, where J is inertia, L is inductance, kt is torque constant, and Ke is back electromotive force constant.

Tm＝(J×L)/(Kt×Ke)

In this case, it can be confirmed that the mechanical time constant Tm is reduced by reducing the inductance L.

A plurality of wires 82 having a flat rectangular cross section are arranged in a radial direction of the stator core 52 so as to constitute respective wire groups 81 on the radially outer side of the stator core 52. The wires 82 are arranged in a direction of "radial dimension < circumferential dimension" in cross section. Thus, the radial thinning is achieved in each wire group 81. Further, the conductor region is flattened to extend to the region where the teeth are present in the past while achieving radial thinning, and the flat wire region structure is formed. Thus, the cross-sectional area of the conductor is increased by flattening in the circumferential direction, and an increase in the amount of heat generated by the wire due to the reduction in the cross-sectional area by thinning is suppressed. In addition, even in a structure in which a plurality of wires are arranged in the circumferential direction and the wires are connected in parallel, a decrease in the cross-sectional area of the conductor due to the portion of the conductor film occurs, but the same effect can be obtained for the same reason. In the following, each wire group 81 and each wire 82 are also referred to as a conductive member (conductive member: conductive member).

Since there is no notch, according to the stator winding 51 of the present embodiment, the conductor area occupied by the stator winding 51 of one circumferential direction thereof can be designed to be larger than the non-conductor occupied area where the stator winding 51 is not present. In a conventional rotating electrical machine for a vehicle, a conductor region/nonconductor occupied region around the circumference of a stator winding is naturally 1 or less. On the other hand, in the present embodiment, the conductor regions are made equal to or larger than the nonconductor occupation regions, and the respective wire groups 81 are provided. Here, as shown in fig. 10, if the wire region in which the wires 82 (i.e., the linear portions 83 described later) are arranged in the circumferential direction is denoted as WA, and the inter-wire region between adjacent wires 82 is denoted as WB, the wire region WA is larger than the inter-wire region WB in the circumferential direction.

As a structure of the wire group 81 of the stator winding 51, a radial thickness dimension of the wire group 81 is smaller than a circumferential width dimension corresponding to one in one magnetic pole. That is, in the configuration in which the lead wire group 81 is configured by two layers of lead wires 82 in the radial direction and two lead wire groups 81 are provided in the circumferential direction for one phase in one magnetic pole, when the radial thickness dimension of each lead wire 82 is Tc and the circumferential width dimension of each lead wire 82 is Wc, "tc×2< wc×2". In another configuration, in a configuration in which the lead group 81 is constituted by two layers of leads 82 and one lead group 81 is provided in the circumferential direction for one phase in one magnetic pole, it is preferable to construct a relationship of "tc×2< wc". In short, the radial thickness dimension of the wire portions (wire group 81) circumferentially arranged at predetermined intervals in the stator winding 51 is smaller than the circumferential width dimension corresponding to one of the one magnetic poles.

In other words, the radial thickness dimension Tc of each wire 82 is preferably smaller than the circumferential width dimension Wc. Further, the radial thickness dimension (2 Tc) of the wire group 81 constituted by the two-layer wires 82 in the radial direction, that is, the radial thickness dimension (2 Tc) of the wire group 81 is preferably smaller than the circumferential width dimension Wc.

The torque of the rotary electric machine 10 is approximately inversely proportional to the thickness of the stator core 52 of the wire group 81 in the radial direction. In this regard, the thickness of the lead wire group 81 is reduced on the outer side in the radial direction of the stator core 52, which is advantageous in terms of achieving an increase in torque of the rotary electric machine 10. The reason for this is that the distance from the magnet unit 42 of the rotor 40 to the stator core 52 (i.e., the distance of the portion where no iron is present) can be reduced, thereby reducing the magnetic resistance. This increases the interlinkage magnetic flux between the permanent magnets and the stator core 52, and thus increases the torque.

Further, by making the thickness of the conductor wire group 81 thin, even if magnetic flux leaks from the conductor wire group 81, the magnetic flux can be easily recovered to the stator core 52, so that the leakage of the magnetic flux to the outside can be suppressed without being effectively used for improving torque. That is, the magnetic force can be suppressed from decreasing due to the leakage of the magnetic flux, and the interlinkage magnetic flux between the permanent magnet and the stator core 52 can be increased to enhance the torque.

The conductor 82 (conductor) is composed of a covered conductor in which the surface of the conductor 82a is covered with the insulating film 82b, so that insulation is ensured between the conductors 82 overlapping each other in the radial direction and between the conductor 82 and the stator core 52, respectively. If the wire 86 described later is a self-fluxing coated wire, the insulating film 82b is a coating film thereof or is composed of an insulating member which is different from the coating film of the wire 86 and which is overlapped. In addition, each phase winding constituted by the wire 82 is kept insulated by the insulating film 82b except for the exposed portion for connection. The exposed portion is, for example, an input/output terminal portion or a neutral point portion when forming a star connection. In the wire group 81, the wires 82 adjacent to each other in the radial direction are fixed to each other using a resin fixation or a self-fluxing clad wire. Thereby, dielectric breakdown, vibration, and sound caused by mutual friction of the wires 82 are suppressed.

In the present embodiment, the conductor 82a is configured as an aggregate of a plurality of wires (wire) 86. Specifically, as shown in fig. 13, the conductor 82a is formed into a twisted wire shape by twisting the multi-strand wire 86. As shown in fig. 14, the wire 86 is formed as a composite in which a relatively thin fibrous conductive member 87 is bundled. For example, the wire 86 is a composite of CNT (carbon nanotube) fibers, and as the CNT fibers, fibers including boron-containing microfibers in which at least a part of carbon is replaced with boron are used. As the carbon microfibers, vapor phase growth carbon fibers (VGCF) and the like may be used in addition to CNT fibers, but CNT fibers are preferably used. The surface of the wire 86 is covered with a polymer insulating layer such as enamel. The surface of the wire 86 is preferably covered with a so-called enamel film composed of a polyimide film or an amidimide film.

The wires 82 constitute n-phase windings in the stator winding 51. Further, the respective wires 86 of the wires 82 (i.e., the conductors 82 a) are adjacent in a state of being in contact with each other. The wire 82 is a wire assembly in which the winding conductor has a portion formed by twisting a plurality of wires 86 at one or more positions in the phase, and the resistance value between the twisted wires 86 is larger than the resistance value of the wires 86 themselves. In other words, when each of the adjacent two strands 86 has a first resistivity in the above-described adjacent direction and each of the strands 86 has a second resistivity in the longitudinal direction thereof, the first resistivity is a value greater than the second resistivity. The lead 82 may be a wire assembly as follows: the multi-strand wire 86 is formed of the multi-strand wire 86 and covered with a first insulating member having extremely high resistivity. The conductor 82a of the wire 82 is formed of a twisted multi-strand wire 86.

Since the conductor 82a is formed by twisting the plurality of wires 86, the generation of eddy current at each wire 86 can be suppressed, and the eddy current of the conductor 82a can be reduced. Further, by twisting the wires 86, a portion where the magnetic fields are applied in opposite directions is generated in one wire 86, thereby canceling back electromotive force. Thus, eddy currents can still be reduced. In particular, by forming the wire 86 with the fibrous conductive member 87, the twisting frequency can be made finer and greatly increased, and the eddy current can be further desirably reduced.

The method of insulating the wires 86 from each other is not limited to the polymer insulating film described above, and a method of making it difficult for a current to flow between the twisted wires 86 by using a contact resistance may be used. That is, if the resistance value between the twisted wires 86 is greater than the resistance value of the wires 86 themselves, the above-described effect can be obtained by the potential difference generated by the difference in resistance values. For example, it is preferable that the contact resistance be increased by oxidizing the wire 86 according to the movement time, the operation interval, or the like by using a manufacturing apparatus for manufacturing the wire 86 and a manufacturing apparatus for manufacturing the stator 50 (armature) of the rotary electric machine 10 as separate, non-continuous apparatuses.

As described above, the wire 82 is a member having a flat rectangular cross section and a plurality of wires arranged side by side in the radial direction, and the shape is maintained by bringing together the multi-strand wires 86 covered with the self-fluxing covered wire including the fusion layer and the insulating layer in a twisted state and fusing the fusion layers to each other. In addition, the wire rod excluding the fusion layer and the wire rod of the self-fluxing coated wire may be firmly molded into a desired shape in a twisted state by using a synthetic resin or the like. When the thickness of the insulating film 82b of the wire 82 is set to, for example, 80 μm to 100 μm, and is thicker than the thickness (5 μm to 40 μm) of the wire that is generally used, insulation between the wire 82 and the stator core 52 can be ensured even without sandwiching insulating paper or the like between the wire 82 and the stator core 52.

Further, it is desirable that the insulating film 82b has a higher insulating property than the insulating layer of the wire 86, and can insulate between phases. For example, when the thickness of the polymer insulating layer of the wire 86 is, for example, about 5 μm, it is desirable that the thickness of the insulating film 82b of the wire 82 is about 80 μm to 100 μm, so that insulation between phases can be desirably performed.

The wire 82 may be a structure that bundles the plurality of wires 86 without twisting. That is, the wire 82 may be any one of a structure in which the multi-strand wire 86 is twisted over its entire length, a structure in which the multi-strand wire 86 is partially twisted over its entire length, and a structure in which the multi-strand wire 86 is bundled without being twisted over its entire length. In short, each wire 82 constituting the wire portion is a wire assembly as follows: the strands of wire 86 are bundled and the resistance between the bundled wires is greater than the resistance of the wire 86 itself.

The respective wires 82 are bent to be arranged in a predetermined arrangement pattern in the circumferential direction of the stator winding 51, whereby phase windings for the respective phases are formed as the stator winding 51. As shown in fig. 12, in the stator winding 51, the coil side portion 53 is formed by a straight portion 83 extending linearly in the axial direction of each wire 82, and the coil side end portions 54, 55 are formed by turning portions 84 protruding to both outer sides of the coil side portion 53 in the axial direction. The straight portions 83 and the bent portions 84 are alternately repeated, so that each wire 82 is configured as a series of wires of a waveform. The linear portions 83 are disposed at positions radially opposite to the magnet unit 42, and the linear portions 83 of the same phase disposed at positions axially outside the magnet unit 42 with a predetermined interval therebetween are connected to each other by the turning portions 84. The straight line portion 83 corresponds to a "magnet facing portion".

In the present embodiment, the stator winding 51 is formed in a circular shape by winding with distributed windings. In this case, the coil side portions 53 are arranged with the linear portions 83 at intervals corresponding to one pair of the magnet units 42 in the circumferential direction for each phase, and the linear portions 83 for each phase are connected to each other by the bent portions 84 formed in a substantially V-shape in the coil side end portions 54, 55. The directions of the currents of the straight portions 83 paired with the pair of poles are opposite to each other. Further, the combination of the pair of straight portions 83 connected by the turning portion 84 is different in one coil side end portion 54 and the other coil side end portion 55, and the stator winding 51 is formed in a substantially cylindrical shape by repeating the connection at the coil side end portions 54, 55 in the circumferential direction.

More specifically, in the stator winding 51, two pairs of wires 82 are used for each phase to form windings of each phase, and one three-phase winding (U-phase, V-phase, W-phase) and the other three-phase winding (X-phase, Y-phase, Z-phase) of the stator winding 51 are provided in two layers inside and outside in the radial direction. In this case, if the number of phases of the stator winding 51 is S (6 in the case of the embodiment) and the number of each phase of the wires 82 is m, 2×s×m=2 Sm wires 82 are formed for each pole pair. In the present embodiment, since the rotating electrical machine has the number of phases S of 6 and the number m of 4 and 8 pole pairs (16 poles), 6×4×8=192 wires 82 are arranged in the circumferential direction of the stator core 52.

In the stator winding 51 shown in fig. 12, the linear portions 83 are arranged in the coil side portion 53 so as to overlap in two layers adjacent to each other in the radial direction, and the turning portions 84 extend in the circumferential direction from the respective linear portions 83 overlapping in the radial direction in mutually opposite directions in the circumferential direction in the coil side end portions 54, 55. That is, in each of the wires 82 adjacent in the radial direction, the directions of the turns 84 are opposite to each other except for the end portions of the stator winding 51.

Here, a winding structure of the wire 82 in the stator winding 51 will be specifically described. In the present embodiment, a plurality of wires 82 formed by wave winding are stacked in a plurality of layers (for example, two layers) adjacent to each other in the radial direction. Fig. 15 (a) and 15 (b) are diagrams showing the form of each wire 82 in the n-th layer, where fig. 15 (a) shows the shape of the wire 82 as seen from the side of the stator winding 51, and fig. 15 (b) shows the shape of the wire 82 as seen from the axial side of the stator winding 51. In fig. 15 (a) and 15 (b), positions where the wire group 81 is arranged are denoted by D1, D2, and D3 …, respectively. For convenience of explanation, only three wires 82 are shown, and the wires are designated as a first wire 82_a, a second wire 82_b, and a third wire 82_c.

In each of the wires 82_a to 82_c, the straight portions 83 are arranged at the position of the nth layer, that is, at the same position in the radial direction, and the straight portions 83 separated at every 6 positions (corresponding to 3×m pairs) in the circumferential direction are connected to each other by the turning portions 84. In other words, two of the two ends of the seven linear portions 83 of each of the leads 82_a to 82_c, which are arranged adjacent to each other in the circumferential direction of the stator winding 51, are connected to each other by one bent portion 84 on the same circle centered on the axial center of the rotor 40. For example, in the first wire 82_a, a pair of straight portions 83 are disposed on D1, D7, respectively, and the pair of straight portions 83 are connected to each other by an inverted V-shaped turning portion 84. The other wires 82_b and 82_c are arranged in the same n-th layer so as to be shifted from each other in position in the circumferential direction. In this case, since the wires 82_a to 82_c are all arranged on the same layer, it is considered that the bent portions 84 interfere with each other. Therefore, in the present embodiment, an interference avoiding portion that biases a part in the radial direction is formed in the bent portion 84 of each of the wires 82_a to 82_c.

Specifically, the turning portion 84 of each of the wires 82_a to 82_c has: one inclined portion 84a as a portion extending in the circumferential direction on the same circle (first circle); a position offset from the inclined portion 84a toward the radially inner side (upper side in fig. 15 b) of the same circle, and reaching the top portion 84b of the other circle (second circle); a slope 84c extending circumferentially on the second circle; and a return portion 84d returning from the first circle to the second circle. The top 84b, the inclined portion 84c, and the return portion 84d correspond to interference avoidance portions. The inclined portion 84c may be offset radially outward from the inclined portion 84 a.

That is, the bent portions 84 of the leads 82_a to 82_c have one side inclined portion 84a and the other side inclined portion 84c on both sides thereof at the center position in the circumferential direction, that is, the top portion 84b, and the positions in the radial direction of the inclined portions 84a, 84c (the positions in the paper surface front-rear direction in fig. 15 (a), and the positions in the up-down direction in fig. 15 (b)) are different from each other. For example, the bent portion 84 of the first lead wire 82_a is configured to extend in the circumferential direction with the D1 position of the n layers as the start position, bend in the radial direction (for example, radially inward) at the top portion 84b which is the central position in the circumferential direction, bend again in the circumferential direction, extend again in the circumferential direction, bend again in the radial direction (for example, radially outward) at the return portion 84D, and reach the D7 position of the n layers which is the end position.

According to the above configuration, among the wires 82_a to 82_c, one of the inclined portions 84a is arranged vertically in the order of the first wire 82_a→the second wire 82_b→the third wire 82_c from above, and the upper and lower sides of the wires 82_a to 82_c are exchanged at the top portion 84b, and the other inclined portion 84c is arranged vertically in the order of the third wire 82_c→the second wire 82_b→the first wire 82_a from above. Accordingly, the wires 82_a to 82_c can be arranged in the circumferential direction without interfering with each other.

Here, in the structure in which the plurality of wires 82 are overlapped in the radial direction to form the wire group 81, the turning portion 84 connected to the radially inner side straight portion 83 and the turning portion 84 connected to the radially outer side straight portion 83 of the respective straight portions 83 of the plurality of layers are preferably arranged further away from each other in the radial direction than the respective straight portions 83. Further, when the multilayer wire 82 is bent to the same side in the radial direction in the vicinity of the edge of the bent portion 84, that is, the boundary portion with the straight portion 83, it is preferable that the insulation is not impaired by interference between the wires 82 of the adjacent layers.

For example, in D7 to D9 of fig. 15 (a) and 15 (b), the respective wires 82 overlapped in the radial direction are respectively bent in the radial direction at the return portion 84D of the turning portion 84. In this case, as shown in fig. 16, it is preferable that the curvature radius of the bent portion of the wire 82 of the n-th layer is different from that of the wire 82 of the n+1-th layer. Specifically, the radius of curvature R1 of the wire 82 on the radially inner side (n-th layer) is made smaller than the radius of curvature R2 of the wire 82 on the radially outer side (n+1-th layer).

Further, it is preferable that the radial offset amounts of the wires 82 of the n-th layer and the wires 82 of the n+1-th layer be different. Specifically, the shift amount S1 of the wire 82 on the radially inner side (n-th layer) is made larger than the shift amount S2 of the wire 82 on the radially outer side (n+1-th layer).

According to the above-described structure, even in the case where the respective wires 82 overlapped in the radial direction are bent in the same direction, the mutual interference of the respective wires 82 can be desirably avoided. Thus, good insulation can be obtained.

Next, the structure of the magnet unit 42 in the rotor 40 will be described. In the present embodiment, it is assumed that the magnet unit 42 is composed of a permanent magnet, and the residual magnetic flux density br=1.0 [ t ], and the intrinsic coercive force hcj=400 [ ka/m ] or more. In summary, the permanent magnet used in the present embodiment is a sintered magnet obtained by sintering a granular magnetic material to form and solidify, the intrinsic coercive force Hcj on the J-H curve is 400[ ka/m ] or more, and the residual magnetic flux density Br is 1.0[ t ] or more. When 5000 to 10000[ AT ] is applied by interphase excitation, if a permanent magnet having a magnetic length of one pair, i.e., N pole and S pole, in other words, a length of 25[ mm ] passing through the inside of the magnet in a path through which magnetic flux between the N pole and S pole flows is used, hcj=10000 [ A ] so as not to demagnetize.

In other words, regarding the magnet unit 42, the saturation magnetic flux density Js is 1.2[ t ] or more and the crystal grain diameter is 10[ μm ] or less, and js×α is 1.0[ t ] or more when the orientation ratio is α.

The magnet unit 42 is described in addition below. The magnet unit 42 (magnet) is characterized by 2.15[ T ]. Gtoreq.Js ]. Gtoreq.1.2 [ T ]. In other words, as the magnets used for the magnet unit 42, ndFe11TiN, nd2Fe14B, sm Fe17N3, feNi magnets having L10-type crystals, and the like are exemplified. Further, the structures of SmCo5, fePt, dy2Fe14B, coPt, etc., which are commonly called samarium-cobalt magnets (samarium-cobalt), cannot be used. Note that, although dysprosium, which is usually heavy rare earth, is used as in the case of compounds of the same type such as Dy2Fe14B and Nd2Fe14B, the higher Js characteristic of neodymium is slightly lost, and a magnet having a higher coercivity possessed by Dy can satisfy 2.15 t·js·1.2 t·t. In this case, it is called, for example, ([ Nd1-xDyx ]2Fe 14B). Further, a magnet composed of two or more kinds of materials having different compositions, for example, a magnet composed of two or more kinds of materials such as feni+sm2fe17n3 may be used, and a mixed magnet or the like in which a magnet of Js-rich Nd2Fe14B such as js=1.6 [ t ] is mixed with a small amount of Dy2Fe14B such as Js < 1[T ] to increase coercive force may be used.

In addition, for a rotating electrical machine that operates at a temperature outside the range of human activity, for example, at 60 ℃ or higher exceeding the temperature of a desert, for example, for a motor for a vehicle in which the temperature in a summer car is close to 80 ℃, it is particularly desirable to include components such as FeNi and Sm2Fe17N3 having a small temperature dependence coefficient. This is because, in the motor operation from the temperature state around-40 ℃ in northern europe, which is the range of human activity, to 60 ℃ or more exceeding the desert temperature, or to about 180 to 240 ℃ which is the heat-resistant temperature of the coil enamel film, the motor characteristics are greatly different depending on the temperature dependence coefficient, and therefore it is difficult to perform optimal control or the like by the same motor driver. If FeNi, sm2Fe17N3, or the like having the L10-form crystal is used, the burden on the motor driver can be desirably reduced by having a characteristic of a temperature dependence coefficient of half or less as compared with Nd2Fe 14B.

In addition, the magnet unit 42 is characterized in that the size of the particle diameter of the fine powder before orientation is 10 μm or less and the single magnetic domain particle diameter is set by using the above-mentioned magnet combination. In magnets, since the coercivity can be increased by micronizing particles of the powder to several hundred nm, micronized powder has been used as much as possible in recent years. However, when excessively thinned, the BH product of the magnet is reduced by oxidation or the like, and therefore, a single magnetic domain particle diameter or more is preferable. It is known that the coercivity increases due to miniaturization of particles having a particle diameter up to a single magnetic domain particle diameter. The particle size described here is the particle size of the fine powder in the orientation step in the production process of the magnet.

Further, the first magnet 91 and the second magnet 92 of the magnet unit 42 are sintered magnets formed by sintering magnetic powder at a high temperature, respectively. When the saturation magnetic flux density Js of the magnet unit 42 is 1.2T or more, the crystal grain sizes of the first magnet 91 and the second magnet 92 are 10 μm or less, and the orientation ratio is α, sintering is performed so as to satisfy the condition that jsxα is 1.0T (tesla) or more. Further, the first magnet 91 and the second magnet 92 are sintered in such a manner that the following conditions are satisfied, respectively. Next, the alignment is performed in the alignment step in the manufacturing step, and the alignment ratio (orientation ratio: alignment ratio) is set to be different from the definition of the magnetic force direction in the excitation step of the isotropic magnet. The saturation magnetic flux density Js of the magnet unit 42 of the present embodiment is 1.2T or more, and a high orientation ratio is set so that the orientation ratio α of the first magnet 91 and the second magnet 92 is Jr or more jsxα or more than 1.0[ T ]. Here, the orientation ratio α is, for example, when 6 magnets are provided in each of the first magnet 91 and the second magnet 92, 5 of the magnets are oriented in the same direction, that is, in the direction a10, and the remaining one is oriented in the direction B10 inclined by 90 degrees with respect to the direction a10, α=5/6, and when the remaining one is oriented in the direction B10 inclined by 45 degrees with respect to the direction a10, α= (5+0.707)/6 because the component of the remaining one oriented in the direction a10 is cos45 ° =0.707. In the present embodiment, the first magnet 91 and the second magnet 92 are formed by sintering, but the first magnet 91 and the second magnet 92 may be formed by other methods if the above conditions are satisfied. For example, a method of forming an MQ3 magnet or the like may be employed.

In the present embodiment, since the permanent magnet whose easy axis is controlled by orientation is used, the magnetic path length inside the magnet can be made longer than that of the conventional linear orientation magnet having a magnetic path length of 1.0[ t ] or more. That is, in addition to the magnetic path length per pole pair being realized with a smaller amount of magnet, the reversible demagnetization range can be maintained even when exposed to severe high temperature conditions, as compared with the design using the conventional linearly oriented magnet. Further, the inventors of the present application have found a structure that can obtain characteristics similar to those of a polar anisotropic magnet even if a magnet of the related art is used.

The easy axis of magnetization refers to a crystal orientation that is easily magnetized in a magnet. The direction of the easy axis in the magnet means a direction in which the orientation ratio indicating the degree of alignment of the direction of the easy axis is 50% or more, or a direction in which the orientation of the magnet is averaged.

As shown in fig. 8 and 9, the magnet unit 42 is annular and is provided inside the magnet holder 41 (more specifically, radially inside the cylindrical portion 43). The magnet unit 42 has a first magnet 91 and a second magnet 92 which are respectively polar anisotropic magnets and have polarities different from each other. The first magnets 91 and the second magnets 92 are alternately arranged in the circumferential direction. The first magnet 91 is a magnet having an N pole formed at a portion close to the stator winding 51, and the second magnet 92 is a magnet having an S pole formed at a portion close to the stator winding 51. The first magnet 91 and the second magnet 92 are permanent magnets made of rare earth magnets such as neodymium magnets.

In each of the magnets 91 and 92, as shown in fig. 9, in a known d-q coordinate system, the magnetization direction extends in an arc shape between the center of the magnetic pole, i.e., the d-axis (direct-axis), and the boundary between the magnetic poles of the N-pole and the S-pole (in other words, the magnetic flux density is 0 tesla), i.e., the q-axis (quadrature-axis). In each of the magnets 91 and 92, the magnetization direction is the radial direction of the annular magnet unit 42 on the d-axis side, and the magnetization direction of the annular magnet unit 42 is the circumferential direction on the q-axis side. The following is a further detailed description. As shown in fig. 9, the magnets 91, 92 each have: a first portion 250; and two second portions 260 located at both sides of the first portion 250 in the circumferential direction of the magnet unit 42. In other words, the first portion 250 is closer to the d-axis than the second portion 260, and the second portion 260 is closer to the q-axis than the first portion 250. Also, the magnet unit 42 is configured such that the direction of the easy axis 300 of the first portion 250 is more parallel to the d-axis than the direction of the easy axis 310 of the second portion 260. In other words, the magnet unit 42 is configured such that an angle θ11 formed by the easy axis 300 of the first portion 250 and the d-axis is smaller than an angle θ12 formed by the easy axis 310 of the second portion 260 and the q-axis.

In more detail, the angle θ11 is an angle formed by the d-axis and the easy magnetization axis 300 when the direction from the stator 50 (armature) toward the magnet unit 42 on the d-axis is set as positive. The angle θ12 is an angle formed by the q-axis and the easy axis 310 when the direction from the stator 50 (armature) toward the magnet unit 42 on the q-axis is set as positive. In the present embodiment, the angle θ11 and the angle θ12 are each 90 ° or less. The easy axes 300, 310 are defined herein as follows. In the respective portions of the magnets 91, 92, when one easy axis is oriented in the direction a11 and the other easy axis is oriented in the direction B11, the cosine absolute value of the angle θ (i cos θ i) formed by the direction a11 and the direction B11 is set as the easy axis 300 or the easy axis 310.

That is, the directions of the easy magnetization axes of the magnets 91 and 92 are different between the d-axis side (the portion near the d-axis) and the q-axis side (the portion near the q-axis), and the direction of the easy magnetization axes is the direction near the direction parallel to the d-axis on the d-axis side and the direction of the easy magnetization axes is the direction near the direction orthogonal to the q-axis on the q-axis side. Further, a circular arc-shaped magnetic circuit is formed according to the direction of the easy axis. In each of the magnets 91 and 92, the easy axis may be parallel to the d axis and the easy axis may be orthogonal to the q axis.

In the magnets 91 and 92, the stator-side outer surface of the circumferential surface of each magnet 91 and 92 on the stator 50 side (lower side in fig. 9) and the q-axis-side end surface in the circumferential direction are magnetic flux application surfaces, which are inflow and outflow surfaces of magnetic flux, and the magnet magnetic circuit is formed so as to connect the magnetic flux application surfaces (the stator-side outer surface and the q-axis-side end surface).

In the magnet unit 42, since the magnetic flux flows in an arc shape between the adjacent N pole and S pole by the respective magnets 91, 92, the magnet magnetic path is longer than, for example, a radial anisotropic magnet. Therefore, as shown in fig. 17, the magnetic flux density distribution approximates a sine wave. As a result, unlike the magnetic flux density distribution of the radially anisotropic magnet shown as a comparative example in fig. 18, the magnetic flux can be concentrated on the center side of the magnetic pole, and the torque of the rotating electrical machine 10 can be improved. In the magnet unit 42 of the present embodiment, it was confirmed that there was a difference in magnetic flux density distribution compared with the conventional halbach array magnet. In fig. 17 and 18, the horizontal axis represents the electrical angle, and the vertical axis represents the magnetic flux density. Further, in fig. 17 and 18, 90 ° of the horizontal axis represents the d-axis (i.e., the magnetic pole center), and 0 ° and 180 ° of the horizontal axis represent the q-axis.

That is, according to the respective magnets 91, 92 of the above-described configuration, the magnet flux in the d-axis is enhanced, and the magnetic flux variation in the vicinity of the q-axis is suppressed. Accordingly, the magnets 91 and 92 having gentle surface magnetic flux changes from the q axis to the d axis in the respective magnetic poles can be preferably realized.

The sine wave matching ratio of the magnetic flux density distribution is preferably, for example, 40% or more. In this way, the magnetic flux in the central portion of the waveform can be reliably increased as compared with the case of using a radially oriented magnet or a parallel oriented magnet having a sine wave matching rate of about 30%. Further, when the sine wave matching ratio is 60% or more, the magnetic flux in the central portion of the waveform can be reliably increased as compared with a magnetic flux concentration array such as halbach array.

In the radial anisotropic magnet shown in fig. 18, the magnetic flux density abruptly changes in the vicinity of the q-axis. The more abrupt the change in magnetic flux density, the more eddy current generated in the stator winding 51 increases. Further, the magnetic flux change on the stator winding 51 side also becomes abrupt. In contrast, in the present embodiment, the magnetic flux density distribution is a magnetic flux waveform close to a sine wave. Therefore, the change in magnetic flux density is smaller than that of the radial anisotropic magnet in the vicinity of the q-axis. Thus, generation of eddy current can be suppressed.

In the magnet unit 42, magnetic flux is generated in the vicinity of the d-axis (i.e., the magnetic pole center) of each of the magnets 91, 92 in a direction orthogonal to the magnetic flux application surface 280 on the stator 50 side, and the magnetic flux has an arc shape in which the distance from the magnetic flux application surface 280 on the stator 50 side increases. The more the magnetic flux orthogonal to the magnetic flux application surface, the stronger the magnetic flux. In this regard, in the rotary electric machine 10 of the present embodiment, since each of the lead wire groups 81 is thinned in the radial direction as described above, the radial center position of the lead wire group 81 is close to the magnetic flux application surface of the magnet unit 42, so that a strong magnet magnetic flux can be received from the rotor 40 in the stator 50.

In the stator 50, a cylindrical stator core 52 is provided on the radially inner side of the stator winding 51, that is, on the opposite side of the rotor 40 with the stator winding 51 interposed therebetween. Therefore, the magnetic flux extending from the magnetic flux application surface of each of the magnets 91, 92 is attracted by the stator core 52 and winds around the stator core 52 one turn while using the stator core 52 as a part of the magnetic circuit. In this case, the direction and path of the magnet flux can be optimized.

Hereinafter, as a method for manufacturing the rotary electric machine 10, an assembly sequence of the bearing unit 20, the housing 30, the rotor 40, the stator 50, and the inverter unit 60 shown in fig. 5 will be described. As shown in fig. 6, the inverter unit 60 includes a unit base 61 and an electric component 62, and each operation step including the assembly step of the unit base 61 and the electric component 62 is described. In the following description, an assembly composed of the stator 50 and the inverter unit 60 is referred to as a first unit, and an assembly composed of the bearing unit 20, the housing 30, and the rotor 40 is referred to as a second unit.

The manufacturing process comprises the following steps:

a first step of mounting the electrical component 62 on the radial inner side of the unit base 61;

a second step of manufacturing a first unit by attaching the unit base 61 to the radial inner side of the stator 50;

a third step of manufacturing a second unit by inserting the fixing portion 44 of the rotor 40 into the bearing unit 20 assembled to the housing 30;

a fourth step of attaching the first unit to the radially inner side of the second unit; and

fifth step of fastening and fixing the housing 30 and the unit base 61.

The steps are performed in the order of first step, second step, third step, fourth step, and fifth step.

According to the above manufacturing method, since the bearing unit 20, the housing 30, the rotor 40, the stator 50, and the inverter unit 60 are assembled into a plurality of assemblies (sub-assemblies), the assemblies are assembled with each other, and thus easy transportation, inspection completion of each unit, and the like can be achieved, and a reasonable assembly line can be constructed. Therefore, it is possible to easily cope with multi-variety production.

In the first step, it is preferable that a thermally conductive body having good heat conduction is attached to at least one of the radially inner side of the unit base 61 and the radially outer side of the electric component 62 by coating or adhesion, and in this state, the electric component 62 is mounted on the unit base 61. Thereby, the heat generated by the semiconductor module 66 can be efficiently transferred to the unit base 61.

In the third step, it is preferable to perform the insertion operation of the rotor 40 while maintaining the coaxiality of the housing 30 and the rotor 40. Specifically, for example, a jig for determining the position of the outer peripheral surface of the rotor 40 (the outer peripheral surface of the magnet holder 41) or the inner peripheral surface of the rotor 40 (the inner peripheral surface of the magnet unit 42) with respect to the inner peripheral surface of the housing 30 is used, and the housing 30 and the rotor 40 are assembled while either one of the housing 30 and the rotor 40 is slid along the jig. Thus, the weight members can be assembled without applying an unbalanced load to the bearing unit 20, thereby improving the reliability of the bearing unit 20.

In the fourth step, it is preferable to perform the assembly of the first unit and the second unit while maintaining the coaxiality of the two units. Specifically, for example, a jig for determining the position of the inner peripheral surface of the unit base 61 with reference to the inner peripheral surface of the fixing portion 44 of the rotor 40 is used, and the first unit and the second unit are assembled while one of them is slid along the jig. Accordingly, since the rotor 40 and the stator 50 can be assembled while preventing interference between the rotor and the stator with each other at a very small gap, damage to the stator winding 51, defective products due to assembly such as a notch of the permanent magnet, and the like can be eliminated.

The order of the above steps may be the second step, the third step, the fourth step, the fifth step, and the first step. In this case, the stress applied to the electrical component 62 in the assembly process can be minimized by finally assembling the precision electrical component 62.

Next, a configuration of a control system for controlling the rotary electric machine 10 will be described. Fig. 19 is a circuit diagram of a control system of the rotary electric machine 10, and fig. 20 is a functional block diagram showing a control process of the control device 110.

In fig. 19, two sets of three-phase windings 51a, 51b are shown as stator windings 51, the three-phase winding 51a being constituted by a U-phase winding, a V-phase winding, and a W-phase winding, and the three-phase winding 51b being constituted by an X-phase winding, a Y-phase winding, and a Z-phase winding. A first inverter 101 and a second inverter 102, which correspond to power converters, are provided for the three-phase windings 51a and 51b, respectively. The inverters 101 and 102 are configured by a full bridge circuit having upper and lower arms, the number of which is equal to the number of phases of the phase windings, and the current flowing through each phase winding of the stator winding 51 is adjusted by turning on/off a switch (semiconductor switching element) provided in each arm.

The dc power supply 103 and the smoothing capacitor 104 are connected in parallel to the inverters 101 and 102. The dc power supply 103 is constituted by a battery pack in which a plurality of single cells are connected in series, for example. The switches of the inverters 101 and 102 correspond to the semiconductor module 66 shown in fig. 1 and the like, and the capacitor 104 corresponds to the capacitor module 68 shown in fig. 1 and the like.

The control device 110 includes a microcomputer including a CPU and various memories, and performs energization control by turning on and off the respective switches of the inverters 101 and 102 based on various pieces of detection information in the rotating electrical machine 10, requests for power running driving and power generation. The control device 110 corresponds to the control device 77 shown in fig. 6. The detection information of the rotary electric machine 10 includes: for example, the rotation angle (electrical angle information) of the rotor 40 detected by a resolver or the like, the power supply voltage (inverter input voltage) detected by a voltage sensor, and the current supplied to each phase detected by a current sensor. The control device 110 generates and outputs operation signals for operating the respective switches of the inverters 101 and 102. In addition, when the rotary electric machine 10 is used as a power source for a vehicle, for example, the request for electric power generation is a request for regenerative drive.

The first inverter 101 includes a series connection of an upper arm switch Sp and a lower arm switch Sn, respectively, among three phases consisting of a U-phase, a V-phase, and a W-phase. The high-potential side terminal of the upper arm switch Sp of each phase is connected to the positive terminal of the dc power supply 103, and the low-potential side terminal of the lower arm switch Sn of each phase is connected to the negative terminal (ground) of the dc power supply 103. One end of a U-phase winding, a V-phase winding and a W-phase winding are respectively connected to the intermediate connection point between the upper arm switch Sp and the lower arm switch Sn of each phase. The above-described phase windings are star-connected (Y-connected), and the other ends of the phase windings are connected to each other at a neutral point.

The second inverter 102 has the same structure as the first inverter 101, and includes a series connection of an upper arm switch Sp and a lower arm switch Sn in three phases including an X phase, a Y phase, and a Z phase. The high-potential side terminal of the upper arm switch Sp of each phase is connected to the positive terminal of the dc power supply 103, and the low-potential side terminal of the lower arm switch Sn of each phase is connected to the negative terminal (ground) of the dc power supply 103. One end of an X-phase winding, a Y-phase winding and a Z-phase winding are respectively connected to the intermediate connection point between the upper arm switch Sp and the lower arm switch Sn of each phase. The above-described phase windings are star-connected (Y-connected), and the other ends of the phase windings are connected to each other at a neutral point.

Fig. 20 shows a current feedback control process of controlling the respective phase currents of the U-phase, V-phase, and W-phase, and a current feedback control process of controlling the respective phase currents of the X-phase, Y-phase, and Z-phase. Here, first, control processing on the U-phase, V-phase, and W-phase sides will be described.

In fig. 20, the current command value setting unit 111 uses torque-dq mapping to set a d-axis current command value and a q-axis current command value based on an electric angular velocity ω obtained by differentiating an electric angle θ with respect to a power running torque command value or a power generation torque command value of the rotating electric machine 10. The current command value setting unit 111 is provided in common to the U-phase, V-phase, and W-phase sides and the X-phase, Y-phase, and Z-phase sides. In addition, for example, when the rotating electrical machine 10 is used as a power source for a vehicle, the generated torque command value is a regenerative torque command value.

The dq conversion section 112 converts current detection values (three-phase currents) detected by current sensors provided for the respective phases into d-axis currents and q-axis currents which are components of an orthogonal two-dimensional rotation coordinate system having a field direction (direction of an axis of a magnetic field, or field direction: a magnetic field axis direction or a magnetic field direction) as a d-axis.

The d-axis current feedback control section 113 calculates a d-axis command voltage as an operation amount for feedback-controlling the d-axis current to a d-axis current command value. Further, the q-axis current feedback control section 114 calculates a q-axis command voltage as an operation amount for feedback-controlling the q-axis current to a q-axis current command value. In the feedback control units 113 and 114, the command voltage is calculated by using the PI feedback method based on the deviation between the d-axis current and the q-axis current and the current command value.

The three-phase conversion section 115 converts command voltages of d-axis and q-axis into command voltages of U-phase, V-phase, and W-phase. The units 111 to 115 are feedback control units that perform feedback control of the fundamental current based on the dq conversion theory, and command voltages of the U phase, V phase, and W phase are feedback control values.

Then, the operation signal generation unit 116 generates an operation signal of the first inverter 101 based on the three-phase command voltages using a well-known triangular wave carrier comparison method. Specifically, the operation signal generation unit 116 generates switching operation signals (duty signals) of the upper and lower arms of each phase by PWM control based on comparison of the magnitudes of the carrier signals such as the triangular wave signal and the signal normalized by the command voltages of the three phases with the power supply voltage.

The same configuration is also applied to the X-phase, Y-phase, and Z-phase sides, and the dq conversion section 122 converts current detection values (three-phase currents) detected by the current sensors provided for the respective phases into d-axis currents and q-axis currents which are components of an orthogonal two-dimensional rotation coordinate system having the excitation direction as the d-axis.

The d-axis current feedback control unit 123 calculates a d-axis command voltage, and the q-axis current feedback control unit 124 calculates a q-axis command voltage. The three-phase conversion section 125 converts command voltages of d-axis and q-axis into command voltages of X-phase, Y-phase, and Z-phase. Then, the operation signal generation section 126 generates an operation signal of the second inverter 102 based on the command voltages of the three phases. Specifically, the operation signal generation unit 126 generates switching operation signals (duty signals) of the upper and lower arms of each phase by PWM control based on comparison of the magnitudes of the carrier signals such as the triangular wave signal and the signal normalized by the command voltages of the three phases with the power supply voltage.

The driver 117 turns on and off the switches Sp, sn of the respective three phases in the respective inverters 101, 102 based on the switch operation signals generated by the operation signal generating sections 116, 126.

Next, a torque feedback control process will be described. The above-described processing is mainly used for the purpose of increasing the output and reducing the loss of the rotating electrical machine 10 under the operating conditions in which the output voltage of each inverter 101, 102 increases, for example, in the high rotation region and the high output region. The control device 110 selects and executes either one of the torque feedback control process and the current feedback control process based on the operation condition of the rotating electrical machine 10.

Fig. 21 shows torque feedback control processing corresponding to U-phase, V-phase, and W, and torque feedback control processing corresponding to X-phase, Y-phase, and Z. In fig. 21, the same components as those in fig. 20 are denoted by the same reference numerals, and description thereof is omitted. Here, first, control processing on the U-phase, V-phase, and W-phase sides will be described.

The voltage amplitude calculation unit 127 calculates a voltage amplitude command, which is a command value for the magnitude of the voltage vector, based on the electric angular velocity ω obtained by time differentiating the electric angle θ from the power running torque command value or the power generation torque command value of the rotating electric machine 10.

The torque estimation section 128a calculates torque estimation values corresponding to the U-phase, V-phase, and W based on the d-axis current and q-axis current converted by the dq conversion section 112. The torque estimating unit 128a may calculate the voltage amplitude command based on map information that sets the relationship between the d-axis current, the q-axis current, and the voltage amplitude command.

The torque feedback control unit 129a calculates a voltage phase command, which is a command value of the phase of the voltage vector, as an operation amount for feedback-controlling the torque estimated value to the power running torque command value or the power generation torque command value. The torque feedback control unit 129a calculates a voltage phase command by using a PI feedback method based on a deviation of the torque estimated value from the power running torque command value or the power generation torque command value.

The operation signal generation unit 130a generates an operation signal of the first inverter 101 based on the voltage amplitude command, the voltage phase command, and the electrical angle θ. Specifically, the operation signal generation unit 130a calculates the command voltages of the three phases based on the voltage amplitude command, the voltage phase command, and the electrical angle θ, and generates the switching operation signals of the upper and lower arms in each phase by PWM control based on a comparison of the magnitudes of the carrier signals such as the signal obtained by normalizing the calculated command voltages of the three phases with the power supply voltage.

The operation signal generation unit 130a may generate the switch operation signal based on pulse mode information, which is the map information that sets the relationship of the voltage amplitude command, the voltage phase command, the electrical angle θ, and the switch operation signal, the voltage amplitude command, the voltage phase command, and the electrical angle θ.

The same configuration is also applied to the X-phase, Y-phase, and Z-phase sides, and the torque estimation unit 128b calculates torque estimation values corresponding to the X-phase, Y-phase, and Z-phase based on the d-axis current and q-axis current converted by the dq conversion unit 122.

The torque feedback control unit 129b calculates a voltage phase command as an operation amount for feedback-controlling the torque estimated value to the power running torque command value or the power generation torque command value. The torque feedback control unit 129b calculates a voltage phase command by using a PI feedback method based on a deviation of the torque estimated value from the power running torque command value or the power generation torque command value.

The operation signal generation unit 130b generates an operation signal of the second inverter 102 based on the voltage amplitude command, the voltage phase command, and the electrical angle θ. Specifically, the operation signal generation unit 130b calculates the command voltages of the three phases based on the voltage amplitude command, the voltage phase command, and the electrical angle θ, and generates the switching operation signals of the upper and lower arms in each phase by PWM control based on a comparison of the magnitudes of the carrier signals such as the signal obtained by normalizing the calculated command voltages of the three phases with the power supply voltage. The driver 117 turns on and off the switches Sp, sn of the respective three phases in the respective inverters 101, 102 based on the switch operation signals generated by the operation signal generating sections 130a, 130 b.

The operation signal generation unit 130b may generate the switch operation signal based on pulse mode information, which is the map information that sets the relationship of the voltage amplitude command, the voltage phase command, the electrical angle θ, and the switch operation signal, the voltage amplitude command, the voltage phase command, and the electrical angle θ.

In the rotating electrical machine 10, there is a concern that the electric corrosion of the bearings 21, 22 occurs with the generation of the shaft current. For example, when the energization of the stator winding 51 is switched by a switch, distortion of magnetic flux occurs due to a minute deviation in the switching timing (imbalance of the switch), and thus, there is a concern that electric corrosion may occur in the bearings 21, 22 supporting the rotary shaft 11. Distortion of the magnetic flux is generated by inductance of the stator 50, and an electromotive force in the axial direction generated by the distortion of the magnetic flux causes dielectric breakdown in the bearings 21 and 22, thereby developing electrolytic corrosion.

In this regard, three countermeasures shown below are taken as the electrolytic corrosion countermeasures in the present embodiment. The first countermeasure against electric corrosion is the following countermeasure against electric corrosion: the inductance is reduced and the magnetic flux of the magnet unit 42 is smoothed in accordance with the coreless stator 50. The second countermeasure against electric corrosion is a countermeasure against electric corrosion in which the rotation shaft is a cantilever structure supported by bearings 21 and 22. The third countermeasure against electric corrosion is to mold the annular stator winding 51 together with the stator core 52 by a molding material. The above-described measures are described in detail below.

First, according to the first countermeasure against electric corrosion, in the stator 50, the seal members 57 made of a non-magnetic material are provided between the wire groups 81 in the circumferential direction instead of the pole teeth (iron cores) by providing the pole teeth between the wire groups 81 (see fig. 10). This can reduce the inductance of the stator 50. By reducing the inductance of the stator 50, even if the switching timing is deviated when the stator winding 51 is energized, the occurrence of magnetic flux distortion due to the deviation of the switching timing can be suppressed, and further, the electric corrosion of the bearings 21, 22 can be suppressed. The d-axis inductance is preferably equal to or less than the q-axis inductance.

In addition, in the magnets 91, 92, the direction of the easy magnetization axis on the d-axis side is oriented more parallel to the d-axis than on the q-axis side (see fig. 9). This enhances the magnetic flux of the magnet on the d-axis, and makes the surface magnetic flux change (increase or decrease of the magnetic flux) from the q-axis to the d-axis in each magnetic pole gentle. Therefore, abrupt voltage changes due to the uneven switching can be suppressed, and further, the electrolytic corrosion can be suppressed.

According to the second countermeasure against electric corrosion, in the rotating electrical machine 10, the bearings 21 and 22 are arranged so as to be offset to either side in the axial direction with respect to the axial center of the rotor 40 (see fig. 2). This reduces the influence of the electrolytic corrosion compared with a structure in which a plurality of bearings are provided on both sides of the rotor in the axial direction. That is, in the structure in which the rotor is supported by the plurality of bearing arms, a closed circuit is formed through the rotor, the stator, and the bearings (i.e., the bearings provided on both sides in the axial direction with the rotor interposed therebetween) with the generation of the high-frequency magnetic flux, and there is a concern that the bearing is electrically corroded by the shaft current. In contrast, in the structure in which the rotor 40 is cantilever-supported by the plurality of bearings 21 and 22, the closed circuit is not formed, and the electrolytic corrosion of the bearings is suppressed.

Further, regarding the structure for the one-sided arrangement of the bearings 21, 22, the rotary electric machine 10 has the following structure. In the magnet holder 41, a contact avoiding portion (see fig. 2) that extends in the axial direction and that avoids contact with the stator 50 is provided in the intermediate portion 45 that protrudes in the radial direction of the rotor 40. In this case, when a closed loop of the shaft current is formed via the magnet holder 41, the closed loop length can be made longer to increase the circuit resistance thereof. Thus, the electrolytic corrosion of the bearings 21 and 22 can be suppressed.

The holding member 23 of the bearing unit 20 is fixed to the housing 30 on one side in the axial direction of the rotor 40, and the housing 30 and the unit base 61 (stator holder) are coupled to each other on the other side (see fig. 2). According to this configuration, the bearings 21 and 22 can be preferably arranged so as to be offset to one side in the axial direction of the rotary shaft 11. Further, according to the present structure, since the unit base 61 is connected to the rotary shaft 11 via the housing 30, the unit base 61 can be disposed at a position electrically separated from the rotary shaft 11. Further, when an insulating member such as resin is interposed between the unit base 61 and the housing 30, the unit base 61 and the rotary shaft 11 are further electrically separated. Thus, the electrolytic corrosion of the bearings 21 and 22 can be appropriately suppressed.

In the rotating electrical machine 10 of the present embodiment, the shaft voltage applied to the bearings 21 and 22 is reduced by the single-sided arrangement of the bearings 21 and 22, or the like. Further, the potential difference between the rotor 40 and the stator 50 decreases. Therefore, even if the bearings 21 and 22 are not provided with conductive grease, the potential difference acting on the bearings 21 and 22 can be reduced. Since conductive grease generally contains fine particles such as carbon, noise is considered to be generated. In this regard, in the present embodiment, a nonconductive grease is used for the bearings 21 and 22. Accordingly, occurrence of noise in the bearings 21 and 22 can be suppressed. It is considered that noise countermeasures for the rotary electric machine 10 are necessary when applied to an electric vehicle such as an electric automobile, and the noise countermeasures described above can be desirably implemented.

According to the third electrolytic corrosion countermeasure, the stator winding 51 and the stator core 52 are molded together by the molding material, thereby suppressing the positional displacement of the stator winding 51 in the stator 50 (refer to fig. 11). In particular, in the rotary electric machine 10 of the present embodiment, since there is no inter-wire member (pole tooth) between the wire groups 81 in the circumferential direction of the stator winding 51, it is considered that there is a possibility that positional displacement occurs in the stator winding 51, but by molding the stator winding 51 and the stator core 52 together, occurrence of positional displacement of the wires of the stator winding 51 is suppressed. Therefore, the occurrence of magnetic flux distortion due to the positional displacement of the stator winding 51 and the occurrence of electric corrosion of the bearings 21 and 22 due to the magnetic flux distortion can be suppressed.

Further, since the unit base 61 as a housing member for fixing the stator core 52 is formed of Carbon Fiber Reinforced Plastic (CFRP), discharge to the unit base 61 can be suppressed as compared with a case of being formed of, for example, aluminum, and thus, an ideal countermeasure against electric corrosion can be achieved.

In addition, as measures against electric corrosion of the bearings 21 and 22, at least one of the outer ring 25 and the inner ring 26 may be formed of a ceramic material, or a structure in which an insulating sleeve or the like is provided on the outer side of the outer ring 25 may be used.

In the first embodiment, as the fourth embodiment described above, a magnet unit using a polar anisotropic structure and having a magnetic flux density distribution similar to a sine wave is adopted. In this way, the torque can be enhanced by increasing the sine wave matching rate by pulse control or the like described later, and eddy current loss can be further suppressed because of a more gradual magnetic flux change than that of the radial magnet.

Since the magnetic unit has a magnetic flux density distribution similar to a sine wave, a magnet having an easy axis of magnetization oriented in an arc shape and having an arc-shaped magnetic circuit of the magnet formed along the easy axis of magnetization is preferably used.

In addition, in the case of using the magnet as described above, as shown in fig. 22 (a), the thickness dimension of the magnet is preferably designed so that the magnet is disposed up to the second intersection point P2 of the q-axis and the orientation circular arc OA in the radial direction. This is to suppress leakage of magnetic flux from the opposite side of the stator of the magnet unit. The orientation circular arc OA is a circular arc which is centered on a center point O set on the q-axis, which is the magnetic pole boundary of the magnet, and passes through a first intersection point P1 of the d-axis, which is the magnetic pole center of the magnet, and the stator-side outer surface of the circumferential surface of the magnet.

That is, as shown in fig. 22 b, if the magnet is not formed to the second intersection point P2 of the q-axis and the orientation circular arc OA in the radial direction, the magnetic flux leaks from the opposite side of the stator of the magnet portion (indicated by the solid arrow). Fig. 22 shows a schematic view of linearly expanding a magnet, in which the lower side is the stator side and the upper side is the opposite side to the stator.

However, when a magnet having an intrinsic coercive force of 400[ kA/m ] or more and a residual magnetic flux density of 1.0[ T ] or more and having a circular arc-shaped magnetic circuit of the magnet formed along the easy axis, or a magnet having crystal particles of a main phase smaller than 10[ mu ] m and a saturation magnetic flux density Js larger than 1.2[ T ] is used, the manufacturing cost is generally high. That is, it is necessary to use neodymium magnets as the magnets or to use expensive rare earths such as terbium (Tb) and dysprosium (Dy) for the magnets. Therefore, if the thickness of the magnet is increased as shown in fig. 22 (a) to prevent leakage of magnetic flux, the amount of the magnet increases, which causes a problem in terms of cost. Therefore, the thickness dimension of the magnet in the radial direction is preferably as thin as possible. Therefore, in the first embodiment, the structure is as follows.

Fig. 23 shows the cylindrical portion 43 of the magnet holder 41 and the magnet unit 42 by being linearly expanded, and the lower side of the drawing is the stator side (armature side) and the upper side is the opposite side to the stator (opposite armature side). As shown in fig. 23, the magnet unit 42 includes magnets 91 and 92 having an easy axis of magnetization oriented in an arc shape and an arc-shaped magnet magnetic path formed along the easy axis of magnetization. The magnet magnetic paths of the magnets 91, 92 include magnetic paths on an orientation circular arc OA centered on a center point O set on the q-axis, which is the magnetic pole boundary of the magnets 91, 92, and passing through a first intersection point P1 of the d-axis, which is the magnetic pole center of the magnets 91, 92, and the stator-side outer surface 1001a (armature-side circumferential surface) of the circumferential surfaces of the magnets 91, 92.

The orientation circular arc OA is set such that a tangential line TA1 at a first intersection point P1 on the orientation circular arc OA is parallel to the d axis. In fig. 23, the center point O is the intersection of the q-axis and the stator-side outer surface 1001a, but the actual magnets 91, 92 are circular arcs along the circumferential direction of the rotor 40 as shown in fig. 24. Therefore, as shown in fig. 24, when the curvature of the magnets 91, 92 is taken into consideration, the center point O is arranged radially outside the stator-side outer surface 1001 a.

As shown in fig. 23, the magnet unit 42 is fixed to the inner peripheral surface of the cylindrical portion 43, which is a soft magnetic material. That is, the cylindrical portion 43 of the magnet holder 41 is stacked with the magnets 91, 92 in the radial direction. In the present embodiment, the entire cylindrical portion 43 is disposed radially closer to the stator than the second intersection point P2 of the q-axis and the orientation circular arc OA. That is, the thickness dimension in the radial direction is made thinner than the magnet shown in fig. 22 (a), and the cylindrical portion 43 is additionally disposed.

Further, in the first embodiment, the saturation magnetic flux density of the cylindrical portion 43 is about 2.0[ t ], and the residual magnetic flux density of the magnets 91, 92 is about 1.0[ t ]. That is, the saturation magnetic flux density of the cylindrical portion 43 is larger than the residual magnetic flux density of the magnets 91, 92. In this case, even if the thickness Wsc of the cylindrical portion 43 in the radial direction is smaller than the radial dimension L1 from the third intersection point P3 to the second intersection point P2 of the stator opposite side circumferential surface (armature opposite side circumferential surface) on the stator opposite side of the circumferential surfaces of the q-axis and magnets 91, 92, leakage of magnetic flux from the stator opposite side can be suppressed.

More specifically, the magnets 91 and 92 and the cylindrical portion 43 are designed so that the relationship of br×wh and bs×wsc is satisfied when the residual magnetic flux density of the magnet unit 42 is Br, the saturation magnetic flux density of the cylindrical portion 43 is Bs, the distance from the center point O to the first intersection point P1 is Wh, and the thickness dimension of the cylindrical portion 43 in the radial direction is Wsc. When the relationship of br×wh and bs×wsc is satisfied, even if the thickness dimension of the cylindrical portion 43 is smaller than the thickness dimension L1 in the radial direction from the third intersection point P3 to the second intersection point P2, the magnetic flux leakage can be appropriately suppressed. That is, in the first embodiment, the magnetic flux leakage can be appropriately suppressed as long as the thickness dimension of the cylindrical portion 43 is made to be half or more of the distance Wh from the center point O to the first intersection point P1. Therefore, in the first embodiment, the thickness Wsc of the cylindrical portion 43 is set to half the distance Wh from the center point O to the first intersection point P1. Thus, the thickness of the magnets 91, 92 and the cylindrical portion 43 can be reduced while appropriately suppressing leakage of magnetic flux.

The thickness dimension of the magnets 91 and 92 in the radial direction needs to be a thickness that aligns at least the easy axis and can form a circular arc-shaped magnetic circuit of the magnet. Further, it is preferable that the radial thickness dimension is a thickness that enables the magnets 91, 92 to be manufactured and takes into account the strength of the magnets 91, 92. In addition, in consideration of the strength of the cylindrical portion 43, the thickness dimension Wsc of the cylindrical portion 43 may be thicker than half the distance Wh.

According to the first embodiment, the following excellent effects are exhibited.

Since the magnet unit 42 having a magnetic flux density distribution close to a sine wave is provided, torque enhancement can be achieved, and since a change in magnetic flux is gentle than that of a radial magnet, eddy current loss can be suppressed. In addition, torque ripple can be reduced. When the intrinsic coercive force of the magnet unit 42 is 400[ ka/m ] or more and the residual magnetic flux density is 1.0[ t ] or more, it is preferable to use a magnet in which the easy axis of magnetization is oriented in an arc shape and an arc-shaped magnet magnetic path is formed along the easy axis of magnetization before the process of hardening the magnet at a high temperature (sintering process) in order to provide the magnet unit 42 having a magnetic flux density distribution close to a sine wave.

When the magnets 91 and 92 are used, in order to suppress leakage of magnetic flux from the opposite side of the stator of the magnet unit 42, as shown in fig. 22 (a), the thickness dimension of the magnets is preferably designed such that the magnets are disposed radially up to the second intersection point P2 of the q-axis and the orientation circular arc OA. However, when the magnet unit 42 having an intrinsic coercive force of 400[ kA/m ] or more and a residual magnetic flux density of 1.0[ T ] or more is used, expensive rare earth such as terbium (Tb) or dysprosium (Dy) needs to be used for the magnet, which causes a problem in terms of cost.

Therefore, in the first embodiment, as shown in fig. 23 and 24, the cylindrical portion 43 and the magnets 91 and 92 are stacked in the radial direction, and the entire cylindrical portion 43 is disposed at a position closer to the stator side (i.e., the magnet unit 42 side) than the second intersection point P2 of the q-axis and the orientation circular arc OA in the radial direction. That is, the thickness dimension of the magnets 91, 92 is made thinner than that of the magnet shown in fig. 22 (a), and the cylindrical portion 43 as the field element core member which is a soft magnetic material is disposed. Even if the magnets 91, 92 are thinned as described above, the magnetic flux passes through the soft magnetic material, that is, the cylindrical portion 43, so that the leakage of the magnetic flux can be suppressed. That is, the magnetic flux density is hard to decrease at the d-axis. As described above, the amount of magnets can be reduced without reducing the magnetic flux density.

The orientation circular arc OA is set such that the tangent TA1 at the first intersection point P1 on the orientation circular arc OA is parallel to the d-axis. Therefore, when the easy axis is oriented along the orientation circular arc OA and a circular arc-shaped magnet magnetic path is formed along the easy axis, the magnetic flux density is maximum at the d axis. That is, since the magnet magnetic circuit is orthogonal to the stator-side outer surface 1001a at the first intersection point P1, the magnetic flux density at the d-axis can be increased. Since the torque is related to the magnetic flux density at the d-axis, the torque can be increased by increasing the magnetic flux density at the d-axis.

In the first embodiment, the saturation magnetic flux density of the cylindrical portion 43 is larger than the residual magnetic flux density of the magnets 91, 92. In this case, even if the soft magnetic material having a smaller thickness than the magnets 91 and 92 is replaced, the leakage of magnetic flux can be appropriately suppressed. That is, the magnetic flux density at the d-axis can be increased, thereby improving the torque. Therefore, the thickness Wsc of the cylindrical portion 43 in the radial direction is made smaller than the radial dimension L1 from the third intersection P3 of the q-axis and the stator-opposite-side circumferential surface to the second intersection P2. Thereby, the total thickness of the cylindrical portion 43 and the magnet unit 42 can be made thin. By reducing the total thickness of the cylindrical portion 43 and the magnet unit 42, the rotary electric machine 10 can be miniaturized, for example. Alternatively, for example, the first region X1 can be increased from the inner peripheral surface of the stator 50 (i.e., the inner peripheral surface of the stator core 52) to the radially inner side. That is, the volume of the inner side of the rotor 40 can be increased.

The magnets 91 and 92 and the cylindrical portion 43 are designed to satisfy the relationship of Br×Wh. Ltoreq.Bs×Wsc. In the present embodiment, the thickness Wsc of the cylindrical portion 43 is set to half the distance Wh from the center point O to the first intersection point P1. Thus, it is theoretically possible to prevent leakage of magnetic flux from the opposite side of the stator and to make the thickness dimension thin. That is, the magnetic flux density at the d-axis can be increased, thereby improving the torque.

Hereinafter, other embodiments will be described focusing on differences from the first embodiment.

(second embodiment)

In the present embodiment, the polar anisotropic structure of the magnet unit 42 in the rotor 40 is changed, and the following description will be made in detail.

As shown in fig. 25 and 26, the magnet unit 42 is constituted using a magnet array called halbach array. That is, the magnet unit 42 has: the first magnets 131 having a radial magnetization direction (direction of magnetization vector) and the second magnets 132 having a circumferential magnetization direction (direction of magnetization vector) are arranged at predetermined intervals in the circumferential direction, and the second magnets 132 are arranged at positions between circumferentially adjacent first magnets 131. The first magnet 131 and the second magnet 132 are permanent magnets made of rare earth magnets such as neodymium magnets.

The first magnets 131 are arranged apart from each other in the circumferential direction so that poles on the side (radially inner side) opposite to the stator 50 alternate with N poles and S poles. Further, the second magnets 132 are arranged to alternate in polarity in the circumferential direction at positions adjacent to the respective first magnets 131. The cylindrical portion 43 surrounding the magnets 131 and 132 is preferably a soft magnetic iron core made of a soft magnetic material, and functions as a supporting iron core. In addition, the relationship between the easy axis and the d axis and the relationship between the easy axis and the q axis in the d-q coordinate system of the magnet unit 42 of the second embodiment are the same as those of the first embodiment.

Further, a magnetic body 133 made of a soft magnetic material is disposed radially outward of the first magnet 131, that is, on the cylindrical portion 43 side of the magnet holder 41. For example, the magnetic body 133 is preferably composed of an electromagnetic steel plate, soft iron, and a dust core material. In this case, the length of the magnetic body 133 in the circumferential direction is the same as the length of the first magnet 131 in the circumferential direction (in particular, the length of the outer peripheral portion of the first magnet 131 in the circumferential direction). The thickness of the integrated body in the radial direction in the state where the first magnet 131 and the magnetic body 133 are integrated is the same as the thickness of the second magnet 132 in the radial direction. In other words, the radial thickness of the first magnet 131 is thinner than the radial thickness of the second magnet 132 by the amount of the magnetic body 133. The magnets 131 and 132 and the magnetic body 133 are fixed to each other by, for example, an adhesive. The radially outer side of the first magnet 131 in the magnet unit 42 is the side opposite to the stator 50, and the magnetic body 133 is provided on the side opposite to the stator 50 (the side opposite to the stator) out of both sides of the first magnet 131 in the radial direction.

A key 134 is formed on the outer periphery of the magnetic body 133, and the key 134 is a convex portion protruding radially outward, that is, toward the cylindrical portion 43 side of the magnet holder 41. A key groove 135 is formed in the inner peripheral surface of the cylindrical portion 43, and the key groove 135 is a recess for accommodating the key 134 of the magnetic body 133. The protruding shape of the key 134 is the same as the groove shape of the key groove 135, and the same number of key grooves 135 as the key 134 are formed corresponding to the key 134 formed in each magnetic body 133. The positional displacement of the first magnet 131 and the second magnet 132 with the magnet holder 41 in the circumferential direction (rotational direction) is suppressed by the engagement of the key 134 and the key groove 135. In addition, the key 134 and the key groove 135 (convex and concave) may be provided in either the cylindrical portion 43 of the magnet holder 41 or the magnetic body 133, or in contrast to the above, the key groove 135 may be provided in the outer peripheral portion of the magnetic body 133 and the key 134 may be provided in the inner peripheral portion of the cylindrical portion 43 of the magnet holder 41.

Here, in the magnet unit 42, the magnetic flux density in the first magnet 131 can be increased by alternately arranging the first magnet 131 and the second magnet 132. Therefore, the magnetic flux can be concentrated on one surface of the magnet unit 42, and the magnetic flux can be intensified on the side close to the stator 50.

Further, by disposing the magnetic body 133 on the radially outer side of the first magnet 131, that is, on the opposite side of the stator, local magnetic saturation on the radially outer side of the first magnet 131 can be suppressed, and demagnetization of the first magnet 131 due to magnetic saturation can be suppressed. Thereby, the magnetic force of the magnet unit 42 can be increased eventually. In other words, the magnet unit 42 of the present embodiment is configured such that the portion of the first magnet 131 where demagnetization is likely to occur is replaced with the magnetic body 133.

Fig. 27 (a) and 27 (b) are diagrams specifically showing the flow of magnetic flux in the magnet unit 42, and fig. 27 (a) shows a case where a conventional structure having no magnetic body 133 in the magnet unit 42 is used, and fig. 27 (b) shows a case where the structure of the present embodiment having the magnetic body 133 in the magnet unit 42 is used. In fig. 27 (a) and 27 (b), the cylindrical portion 43 of the magnet holder 41 and the magnet unit 42 are linearly expanded, and the lower side of the drawing is the stator side, and the upper side is the opposite side to the stator.

In the structure of fig. 27 (a), the magnetic flux application surface of the first magnet 131 and the side surface of the second magnet 132 are in contact with the inner peripheral surface of the cylindrical portion 43, respectively. Further, the magnetic flux application surface of the second magnet 132 is in contact with the side surface of the first magnet 131. In this case, a resultant magnetic flux of the magnetic flux F1 entering the contact surface with the first magnet 131 through the outer path of the second magnet 132 and the magnetic flux of the magnetic flux F2 substantially parallel to the cylindrical portion 43 and attracting the second magnet 132 is generated in the cylindrical portion 43. Therefore, there is a concern that local magnetic saturation occurs in the vicinity of the contact surface of the first magnet 131 and the second magnet 132 in the cylindrical portion 43.

In contrast, in the configuration of fig. 27 (b), the magnetic body 133 is provided between the magnetic flux acting surface of the first magnet 131 and the inner peripheral surface of the cylindrical portion 43 on the opposite side of the first magnet 131 from the stator 50, so that the magnetic flux is allowed to pass through the magnetic body 133. Therefore, the magnetic saturation in the cylindrical portion 43 can be suppressed, and the resistance to demagnetization can be improved.

In addition, in the structure of fig. 27 (b), unlike fig. 27 (a), F2 that promotes magnetic saturation can be eliminated. Thus, the flux guide of the entire magnetic circuit can be effectively improved. With the above structure, the magnetic circuit characteristics can be maintained even under severe high temperature conditions.

In addition, the magnetic circuit of the magnet passing through the inside of the magnet is longer than that of the radial magnet in the existing SPM rotor. Therefore, the magnetic flux guide of the magnet can be increased to increase the magnetic force, thereby increasing the torque. In addition, the sine wave matching rate can be improved by concentrating the magnetic flux in the center of the d-axis. In particular, when the current waveform is made sinusoidal or trapezoidal by PWM control, or a 120-degree energized switching IC is used, the torque can be enhanced more effectively.

In addition, in the case where the stator core 52 is composed of an electromagnetic steel plate, the radial thickness of the stator core 52 is preferably greater than or equal to 1/2 of the radial thickness of the magnet unit 42. For example, the radial thickness of the stator core 52 is preferably 1/2 or more of the radial thickness of the first magnet 131 provided in the center of the magnetic pole in the magnet unit 42. Further, the radial thickness of the stator core 52 is preferably smaller than the radial thickness of the magnet unit 42. In this case, since the magnet magnetic flux is approximately 1[T and the saturation magnetic flux density of the stator core 52 is 2[T, it is possible to prevent leakage of magnetic flux to the inner peripheral side of the stator core 52 by setting the radial thickness of the stator core 52 to 1/2 or more of the radial thickness of the magnet unit 42.

In the magnet having the halbach structure and the polar anisotropic structure, the magnetic path is formed in an approximately circular arc shape, so that the magnetic flux can be increased in proportion to the thickness of the magnet receiving the magnetic flux in the circumferential direction. In the above-described configuration, it is considered that the magnetic flux flowing through the stator core 52 does not exceed the magnetic flux in the circumferential direction. That is, when the iron-based metal having the saturation magnetic flux density 2[T is used for the magnetic flux 1[T of the magnet, it is possible to desirably provide a small and lightweight rotating electrical machine without generating magnetic saturation by setting the thickness of the stator core 52 to be equal to or more than half the thickness of the magnet. Here, since the demagnetizing field from the stator 50 acts on the magnet flux, the magnet flux is generally 0.9[ t ] or less. Therefore, if the stator core has a half thickness of the magnet, the magnetic permeability can be desirably maintained high.

A modification in which a part of the above-described structure is changed will be described below.

Modification 1

In the above-described embodiment, the outer peripheral surface of the stator core 52 is formed in a curved surface shape having no irregularities, and the plurality of lead groups 81 are arranged side by side on the outer peripheral surface at predetermined intervals, but this may be modified. For example, as shown in fig. 28, the stator core 52 includes an annular yoke 141 and a protrusion 142, the yoke 141 being provided on the opposite side (lower side in the figure) of the stator winding 51 from the rotor 40 in the radial direction, and the protrusion 142 extending so as to protrude from the yoke 141 toward between the circumferentially adjacent linear portions 83. The protrusions 142 are provided at predetermined intervals on the radially outer side of the yoke 141, that is, on the rotor 40 side. The respective wire groups 81 and the protruding portions 142 of the stator winding 51 are engaged in the circumferential direction, and the protruding portions 142 are used as positioning portions of the wire groups 81 and are arranged side by side in the circumferential direction. The protruding portion 142 corresponds to "an inter-wire member".

The protrusion 142 is configured such that the radial thickness dimension from the yoke 141, in other words, as shown in fig. 28, the distance W from the inner surface 320 of the linear portion 83 adjacent to the yoke 141 to the apex of the protrusion 142 in the radial direction of the yoke 141 is smaller than 1/2 of the radial thickness dimension of the linear portion 83 adjacent to the yoke 141 in the radial direction among the radially inner and outer multi-layer linear portions 83 (H1 in the figure). In other words, the dimension (thickness) T1 (twice the thickness of the wire 82, in other words, the range of three-quarters of the shortest distance between the surface 320 of the wire group 81 in contact with the stator core 52 and the surface 330 of the wire group 81 facing the rotor 40) of the wire group 81 (conductive member) in the radial direction of the stator winding 51 (stator core 52) may be occupied by the nonmagnetic member (sealing member 57). Due to the thickness limitation of the protruding portion 142, the protruding portion 142 does not act as a pole tooth between the wire groups 81 (i.e., the straight portions 83) adjacent in the circumferential direction, and a magnetic circuit formed by the pole tooth cannot be formed. The protrusions 142 need not be provided for all of the wire groups 81 arranged in the circumferential direction, but the protrusions 142 may be provided between at least one group of the wire groups 81 adjacent in the circumferential direction. For example, the protruding portions 142 may be provided at regular intervals in the circumferential direction every predetermined number between the respective wire groups 81. The shape of the projection 142 may be any shape such as a rectangle or a circular arc.

Further, a single linear portion 83 may be provided on the outer peripheral surface of the stator core 52. Accordingly, in a broad sense, the radial thickness dimension of the protrusion 142 from the yoke 141 may be smaller than 1/2 of the radial thickness dimension of the straight portion 83.

Further, when a virtual circle is assumed that is centered on the axial center of the rotary shaft 11 and passes through the radial center position of the linear portion 83 adjacent to the yoke 141 in the radial direction, the protruding portion 142 preferably has a shape protruding from the yoke 141 within the range of the virtual circle, in other words, a shape not protruding radially outward (i.e., the rotor 40 side) of the virtual circle.

According to the above configuration, since the thickness dimension in the radial direction of the protruding portion 142 is limited and the protruding portion 142 does not act as a pole tooth between the linear portions 83 adjacent in the circumferential direction, each of the adjacent linear portions 83 can be pulled closer than in the case where a pole tooth is provided between each of the linear portions 83. This can increase the cross-sectional area of the conductor 82a, and can reduce heat generation associated with the energization of the stator winding 51. In the above configuration, the magnetic saturation can be eliminated by eliminating the pole teeth, and the current flowing to the stator winding 51 can be increased. In this case, it is possible to desirably cope with a case where the amount of heat generation increases with an increase in the energization current. In addition, in the stator winding 51, since the turning portions 84 have interference avoidance portions that are offset in the radial direction to avoid interference with other turning portions 84, different turning portions 84 can be disposed apart from each other in the radial direction. This can improve heat radiation performance even in the corner 84. In summary, the heat dissipation performance at the stator 50 can be rationalized.

If the yoke 141 of the stator core 52 and the magnet unit 42 (i.e., the magnets 91 and 92) of the rotor 40 are separated by a predetermined distance or more, the radial thickness dimension of the protrusion 142 is not limited to H1 of fig. 28. Specifically, if the yoke 141 and the magnet unit 42 are separated by 2mm or more, the radial thickness dimension of the protrusion 142 may be H1 or more in fig. 28. For example, when the radial thickness dimension of the straight portion 83 exceeds 2mm and the wire group 81 is constituted by the two-layer wires 82 in the radial direction, the protruding portion 142 may be provided in a range of half the position of the straight portion 83 not adjacent to the yoke 141, that is, the second-layer wire 82 from the yoke 141. In this case, as long as the radial thickness dimension of the protruding portion 142 does not exceed "h1×3/2", the above-described effect can be obtained greatly by increasing the conductor cross-sectional area in the wire group 81.

The stator core 52 may have the structure shown in fig. 29. In fig. 29, the seal member 57 is omitted, but the seal member 57 may be provided. In fig. 29, for convenience of explanation, the magnet unit 42 and the stator core 52 are shown linearly expanded.

In the structure of fig. 29, the stator 50 has the protruding portion 142 as the inter-conductor member between the circumferentially adjacent conductors 82 (i.e., the straight portions 83). The stator 50 has a portion 350, the portion 350 and one of the poles (N-pole or S-pole) of the magnet unit 42 exert a magnetic action together when the stator winding 51 is energized, and the portion 350 extends in the circumferential direction of the stator 50. When the length of the portion 350 in the circumferential direction of the stator 50 is Wn, the total width of the protrusions 142 existing in the length range Wn (i.e., the total dimension in the circumferential direction of the stator 50) is Wt, the saturation magnetic flux density of the protrusions 142 is Bs, the width dimension of the magnet unit 42 in the circumferential direction corresponding to one pole is Wm, and the residual magnetic flux density of the magnet unit 42 is Br, the protrusions 142 are made of a magnetic material satisfying the following formula.

Wt×Bs≤Wm×Br…(1)

In addition, the range Wn is set to include a plurality of wire groups 81 that are adjacent in the circumferential direction and the excitation times overlap. At this time, the center of the gap 56 of the lead group 81 is preferably set as a reference (boundary) in the set range Wn. For example, in the case of the structure illustrated in fig. 29, the lead group 81 having the shortest distance from the magnetic pole center of the N pole in the circumferential direction is sequentially equal to the plurality of lead groups 81 from the lead group 81 having the fourth shortest distance. Further, the range Wn is set to include the four wire groups 81. At this time, one end (start point and end point) of the range Wn is the center of the gap 56.

In fig. 29, since the range Wn includes half of the protruding portions 142 at both ends thereof, the range Wn includes a total of four protruding portions 142. Therefore, when the width of the protruding portion 142 (i.e., the size of the protruding portion 142 in the circumferential direction of the stator 50, in other words, the interval of the adjacent wire groups 81) is set to a, the total width of the protruding portions 142 included in the range Wn is wt=1/2a+a+a+1/2a=4a.

In detail, in the present embodiment, the three-phase winding of the stator winding 51 is a distributed winding, and in the stator winding 51, the number of the protrusions 142, that is, the number of the gaps 56 between the wire groups 81 is "the number of phases×q" for one pole of the magnet unit 42. Here Q refers to the number of contacts with the stator core 52 in the wires 82 of one phase. When the lead wire group 81 is configured such that the lead wires 82 are laminated in the radial direction of the rotor 40, Q is also the number of lead wires 82 on the inner peripheral side of the lead wire group 81 of one phase. In this case, when the three-phase windings of the stator winding 51 are energized in a predetermined order for each phase, the protrusions 142 corresponding to the two phases in one pole are excited. Accordingly, when the circumferential width dimension of the protruding portion 142 (i.e., the gap 56) is a, the total circumferential width dimension Wt of the protruding portion 142 excited by the energization of the stator winding 51 in the range corresponding to one pole of the magnet unit 42 is "the excited phase number×q×a=2×2×a".

Next, the total width Wt is defined in this way, and the protruding portion 142 is made of a magnetic material satisfying the relationship (1) in the stator core 52. The total width Wt is the circumferential dimension of a portion having a relative permeability of greater than 1 in one pole. In addition, the total width Wt may be set to the width of the protrusion 142 in one magnetic pole in the circumferential direction in consideration of the margin. Specifically, since the number of the protrusions 142 of one pole of the magnet unit 42 is "phase number×q", the width dimension (total width dimension Wt) of the protrusions 142 in one magnetic pole in the circumferential direction may be set to "phase number×q×a=3×2×a=6a".

The distributed winding as referred to herein means that there is one pole pair of the stator winding 51 at one pole pair period (N pole and S pole) of the magnetic pole. A pole pair of the stator winding 51 described herein is constituted by two straight portions 83 and the turning portion 84, in which currents flow in opposite directions to each other and are electrically connected at the turning portion 84. As long as the above conditions are satisfied, even the short-pitch winding (Short Pitch Winding) is regarded as equivalent to the distributed winding of the full-pitch winding (Full Pitch Winding).

Next, an example when winding is concentrated is shown. The concentrated winding as referred to herein means that the width of one pole pair of the magnetic poles is different from the width of one pole pair of the stator winding 51. As an example of the concentrated winding, the following relationship is given: 3 for one magnetic pole pair wire set 81, 3 for two magnetic pole pair wire sets 81, 9 for four magnetic pole pair wire sets 81, and 9 for five magnetic pole pair wire sets 81.

Here, when the stator winding 51 is a concentrated winding, when three-phase windings of the stator winding 51 are energized in a predetermined order, the stator winding 51 corresponding to two phases is excited. As a result, the protrusions 142 corresponding to the two phases are excited. Accordingly, in the range of the magnet unit 42 corresponding to one pole, the width dimension Wt in the circumferential direction of the protruding portion 142 excited by the energization of the stator winding 51 is "a×2". In addition, the width Wt is thus defined, and the protrusion 142 is made of a magnetic material satisfying the relationship (1) above. In the case of the concentrated winding as described above, the sum of the widths of the protrusions 142 located in the circumferential direction of the stator 50 in the region surrounded by the lead group 81 of the same phase is a. In addition, wm of the concentrated winding corresponds to "the entire circumference of the surface of the magnet unit 42 facing the air gap" × "the number of phases" + "the number of dispersions of the wire group 81".

BH product of 20[ MGOe (kJ/m) in neodymium magnet, samarium cobalt magnet, ferrite magnet ³ )]In the above magnet, bd is 1.0[ T ]]As above, br in iron is 2.0[ T ]]The above. Therefore, as a high-output motor, the protruding portion 142 may be a magnetic material satisfying the relationship of Wt < 1/2×wm in the stator core 52.

When the lead wires 82 include the outer layer 182, the lead wires 82 may be arranged in the circumferential direction of the stator core 52 so that the lead wires 82 contact the outer layer 182 of each other, as will be described later. In this case, wt may be regarded as 0, or the thickness of the outer layer film 182 of the two wires 82 in contact.

In the structure of fig. 28 and 29, the inter-conductor member (projection 142) is configured to have a small magnetic flux passing through the magnet on the rotor 40 side. In addition, the rotor 40 is a flat surface magnet type rotor having low inductance and no saliency in magnetic resistance. With the above configuration, the inductance of the stator 50 can be reduced, and occurrence of magnetic flux distortion due to the variation in the switching timing of the stator winding 51 can be suppressed, and further, the electric corrosion of the bearings 21 and 22 can be suppressed.

Modification 2

As the stator 50 using the inter-conductor member satisfying the relation of the above formula (1), the following structure may be adopted. In fig. 30, tooth-like portions 143 are provided as inter-conductor members on the outer peripheral surface side (upper surface side in the figure) of the stator core 52. The tooth portions 143 are provided at predetermined intervals in the circumferential direction so as to protrude from the yoke 141, and have the same thickness dimension as the wire group 81 in the radial direction. The side surfaces of the tooth 143 are in contact with the respective wires 82 of the wire group 81. However, a gap may be provided between the tooth 143 and each wire 82.

The tooth 143 imposes a limitation on the width dimension in the circumferential direction, and is a member including pole teeth (stator pole teeth) that are too thin with respect to the amount of the magnet. According to the above configuration, the tooth 143 can be reliably magnetically saturated by the magnetic flux of the magnet of 1.8T or more, and the inductance is lowered due to the decrease in the flux guide.

Here, in the magnet unit 42, when Sm is the surface area of each pole of the magnetic flux application surface on the stator side and Br is the residual magnetic flux density of the magnet unit 42, the magnetic flux on the magnet unit side is "sm×br", for example. When the surface area of the rotor side of each tooth 143 is St, the number of phases of the wire 82 is m, and the tooth 143 corresponding to two phases in one pole is excited by energization of the stator winding 51, the magnetic flux on the stator side is "st×m×2×bs", for example. In this case, by

St×m×2×Bs＜Sm×Br…(2)

The size of the tooth 143 is limited so that the relationship between them is established, thereby reducing the inductance.

When the axial dimensions of the magnet unit 42 and the tooth 143 are the same, the above formula (2) is replaced with the formula (3) when the width dimension in the circumferential direction of the magnet unit 42 corresponding to one pole is Wm and the width dimension in the circumferential direction of the tooth 143 is Wst.

Wst×m×2×Bs＜Wm×Br…(3)

More specifically, when bs= T, br =1t and m=2 are assumed, for example, the above formula (3) is a relationship of "Wst < Wm/8". In this case, the reduction of inductance is achieved by making the width dimension Wst of the tooth 143 smaller than 1/8 of the width dimension Wm of the magnet unit 42 corresponding to one pole. If the number m is 1, the width Wst of the tooth 143 is preferably smaller than 1/4 of the width Wm of the magnet unit 42 corresponding to one pole.

In the above formula (3), the "wst×m×2" corresponds to the width dimension in the circumferential direction of the tooth 143 excited by the energization of the stator winding 51 in the range corresponding to one pole of the magnet unit 42.

In the configuration of fig. 30, similarly to the configuration of fig. 28 and 29 described above, the inter-conductor member (tooth 143) having a small magnetic flux passing through the magnet on the rotor 40 side is provided. With the above configuration, the inductance of the stator 50 can be reduced, and occurrence of magnetic flux distortion due to the variation in the switching timing of the stator winding 51 can be suppressed, and further, the electric corrosion of the bearings 21 and 22 can be suppressed.

Modification 3

In the above embodiment, the seal member 57 covering the stator winding 51 is provided outside the stator core 52 in the radial direction in a range including all the wire groups 81, that is, in a range where the thickness dimension in the radial direction is larger than the thickness dimension in the radial direction of each wire group 81, but this may be modified. For example, as shown in fig. 31, the seal member 57 is configured to protrude a part of the lead 82. More specifically, the seal member 57 is configured to expose a portion of the lead 82 closest to the radially outer side of the lead group 81 to the radially outer side, that is, to the stator 50 side. In this case, the radial thickness dimension of the seal member 57 is preferably the same as or smaller than the radial thickness dimension of each wire group 81.

Modification 4

As shown in fig. 32, in the stator 50, each wire group 81 may not be sealed by the sealing member 57. That is, the sealing member 57 covering the stator winding 51 is not used. In this case, no inter-conductor member is provided between the conductor sets 81 arranged in the circumferential direction, leaving a gap. In other words, no inter-conductor member is provided between the conductor groups 81 arranged in the circumferential direction. Air may be regarded as bs=0 as a nonmagnetic material or a nonmagnetic material equivalent, and air may be disposed in the space.

Modification 5

When the inter-conductor member of the stator 50 is made of a nonmagnetic material, a material other than resin may be used as the nonmagnetic material. For example, a metal-based nonmagnetic material such as SUS304, which is austenitic stainless steel, may be used.

Modification 6

The stator 50 may be configured without the stator core 52. In this case, the stator 50 is constituted by the stator winding 51 shown in fig. 12. In addition, in the stator 50 not including the stator core 52, the stator winding 51 may be sealed by a sealing member. Alternatively, instead of the stator core 52 made of a soft magnetic material, the stator 50 may include an annular winding holding portion made of a non-magnetic material such as a synthetic resin.

Modification 7

In the first embodiment, the plurality of magnets 91 and 92 arranged in the circumferential direction are used as the magnet unit 42 of the rotor 40, but this may be modified to use annular permanent magnets, that is, annular magnets, as the magnet unit 42. Specifically, as shown in fig. 33, a ring magnet 95 is fixed to the inner side of the cylindrical portion 43 of the magnet holder 41 in the radial direction. The ring magnet 95 is provided with a plurality of magnetic poles having alternating polarities in the circumferential direction, and a magnet is integrally formed on either one of the d-axis and the q-axis. The annular magnet 95 has an arc-shaped magnet magnetic path oriented radially in the d-axis of each magnetic pole and circumferentially in the q-axis between each magnetic pole.

The annular magnet 95 may be oriented to form a circular arc-shaped magnet magnetic circuit as follows: the easy axis is a direction parallel to or nearly parallel to the d-axis in a portion near the d-axis, and the easy axis is a direction orthogonal to or nearly orthogonal to the q-axis in a portion near the q-axis.

Modification 8

In this modification, a part of the control method of the control device 110 is changed. In this modification, a part different from the structure described in the first embodiment will be mainly described.

First, the processing in the operation signal generating units 116 and 126 shown in fig. 20 and the operation signal generating units 130a and 130b shown in fig. 21 will be described with reference to fig. 34. The processing in the operation signal generating units 116, 126, 130a, and 130b is substantially the same. Therefore, the processing of the operation signal generation unit 116 will be described below as an example.

The operation signal generation section 116 includes a carrier generation section 116a, a U comparator 116bU, a V comparator 116bV, and a W comparator 116bW. In the present embodiment, the carrier generating unit 116a generates and outputs a triangular wave signal as the carrier signal SigC.

The U-phase comparator 116bU, the V-phase comparator 116bV, and the W-phase comparator 116bW receive the carrier signal SigC generated by the carrier generating unit 116a and the U-phase, V-phase, and W-phase command voltages calculated by the three-phase converting unit 115. The U-phase, V-phase, and W-phase command voltages are waveforms such as sine waves, and the phases are shifted every 120 ° in electrical angle.

The U-phase comparator 116bU, the V-phase comparator 116bV, and the W-phase comparator 116bW generate operation signals of the switches Sp, sn of the upper and lower arms of the U-phase, V-phase, and W-phase in the first inverter 101 by PWM (pulse width modulation: pulse width modulation) control based on the magnitude comparison of the U-phase, V-phase, and W-phase command voltages and the carrier signal SigC. Specifically, the operation signal generation unit 116 generates the operation signals of the switches Sp and Sn of the U-phase, V-phase, and W-phase by PWM control based on the comparison of the magnitudes of the signal obtained by normalizing the U-phase, V-phase, and W-phase command voltages with the power supply voltage. The driver 117 turns on and off the switches Sp, sn of the U-phase, V-phase, and W-phase in the first inverter 101 based on the operation signal generated by the operation signal generation section 116.

The control device 110 performs a process of changing the carrier frequency fc of the carrier signal SigC, that is, the switching frequency of each of the switches Sp and Sn. The carrier frequency fc is set higher in the low torque region or the high rotation region of the rotary electric machine 10, and is set lower in the high torque region of the rotary electric machine 10. The above setting is performed to suppress a decrease in the controllability of the current flowing through each phase winding.

That is, the inductance of the stator 50 can be reduced with the coreless stator 50. Here, when the inductance becomes low, the electrical time constant of the rotary electric machine 10 becomes small. As a result, ripple of current flowing through each phase winding increases, and the controllability of current flowing through the winding decreases, and there is a concern that current control may diverge. The above-described influence of the decrease in controllability when included in the low-current region is more remarkable than in the case where the current flowing through the winding (for example, the actual effective value of the current) is included in the high-current region. In order to solve the above-described problem, in the present modification, the control device 110 changes the carrier frequency fc.

The process of changing the carrier frequency fc will be described with reference to fig. 35. The above-described processing is repeatedly executed by the control device 110, for example, at a predetermined control cycle, as processing by the operation signal generation unit 116.

In step S10, it is determined whether or not the current flowing through the winding 51a of each phase is included in the low current region. The above-described processing is processing for determining that the current torque of the rotary electric machine 10 is in the low torque region. For example, the following first method and second method are cited as methods of judging whether or not they are included in the low current region.

< first method >)

A torque estimation value of the rotary electric machine 10 is calculated based on the d-axis current and the q-axis current converted by the dq conversion unit 112. When the calculated torque estimated value is determined to be lower than the torque threshold value, it is determined that the current flowing through the winding 51a is included in the low current region, and when the calculated torque estimated value is determined to be equal to or higher than the torque threshold value, it is determined to be included in the high current region. Here, the torque threshold value may be set to, for example, 1/2 of the starting torque (also referred to as a locked-rotor torque) of the rotary electric machine 10.

(second method)

When it is determined that the rotation angle of the rotor 40 detected by the angle detector is equal to or greater than the speed threshold, it is determined that the current flowing through the winding 51a is included in the low current region, that is, the high rotation region. Here, the speed threshold value may be set to, for example, the rotational speed at which the maximum torque of the rotary electric machine 10 is the torque threshold value.

When a negative determination is made in step S10, it is determined that the current is in the high current region, and the flow proceeds to step S11. In step S11, the carrier frequency fc is set to the first frequency fL.

When an affirmative determination is made in step S10, the flow proceeds to step S12, where the carrier frequency fc is set to a second frequency fH higher than the first frequency fL.

According to the present modification described above, the carrier frequency fc when the current flowing through each phase winding is contained in the low current region is set to be higher than when the current flowing through each phase winding is contained in the high current region. Therefore, in the low current region, the switching frequency of the switches Sp, sn can be increased, and an increase in current ripple can be suppressed. This can suppress a decrease in current controllability.

On the other hand, the carrier frequency fc when the current flowing through each phase winding is included in the high current region is set lower than when the current flowing through each phase winding is included in the low current region. In the high-current region, since the amplitude of the current flowing through the winding is larger than in the low-current region, the influence of the increase in current ripple caused by the inductance becoming low on the current controllability is small. Therefore, the carrier frequency fc in the high current region can be set lower than that in the low current region, and the switching loss of each inverter 101, 102 can be reduced.

In this modification, the following modes can be implemented.

When the carrier frequency fc is set to the first frequency fL, the carrier frequency fc may be gradually changed from the first frequency fL to the second frequency fH when an affirmative determination is made in step S10 in fig. 35.

In addition, when the carrier frequency fc is set to the second frequency fH, the carrier frequency fc may be gradually changed from the second frequency fH to the first frequency fL when a negative determination is made in step S10.

Instead of PWM control, the operation signal of the switch may also be generated by space vector modulation (SVM: space vector modulation) control. In this case, the above-described change of the switching frequency can also be applied.

Modification 9

In each of the above embodiments, as shown in fig. 36 (a), the wires of each pair of the phases constituting the wire group 81 are connected in parallel. Fig. 36 (a) is a diagram showing electrical connection between two pairs of wires, i.e., a first wire 88a and a second wire 88 b. Here, instead of the structure shown in fig. 36 (a), as shown in fig. 36 (b), the first wire 88a and the second wire 88b may be connected in series.

Further, three or more pairs of multilayer wires may be stacked in the radial direction. Fig. 37 shows a structure in which four pairs of wires, that is, first wire 88a to fourth wire 88d are stacked. The first to fourth wires 88a to 88d are arranged in the radial direction in the order of the first wire 88a, the second wire 88b, the third wire 88c, and the fourth wire 88d from the side close to the stator core 52.

Here, as shown in fig. 36 (c), the third and fourth wires 88c and 88d may be connected in parallel, and the first wire 88a may be connected to one end of the parallel connection body and the second wire 88b may be connected to the other end. When connected in parallel, the current density of the wires connected in parallel can be reduced, and heat generation during energization can be suppressed. Therefore, in the structure in which the cylindrical stator winding is assembled to the housing (unit base 61) in which the cooling water passage 74 is formed, the first and second wires 88a and 88b that are not connected in parallel are arranged on the side of the stator core 52 that is in contact with the unit base 61, and the third and fourth wires 88c and 88d that are connected in parallel are arranged on the opposite side of the stator core. This can equalize the cooling performance of the wires 88a to 88d in the multilayer wire structure.

The radial thickness of the lead group 81 formed of the first to fourth leads 88a to 88d may be smaller than the circumferential width of one magnetic pole.

Modification 10

The rotary electric machine 10 may be an inner rotor structure (inner rotor structure). In this case, for example, the stator 50 is preferably provided on the radially outer side and the rotor 40 is preferably provided on the radially inner side in the housing 30. Further, the inverter unit 60 is preferably provided on one or both of the axial ends of the stator 50 and the rotor 40. Fig. 38 is a cross-sectional view of the rotor 40 and the stator 50, and fig. 39 is an enlarged view showing a part of the rotor 40 and the stator 50 shown in fig. 38.

The structures of fig. 38 and 39 on the premise of the inner rotor structure are identical to the structures of fig. 8 and 9 on the premise of the outer rotor structure, except that the rotor 40 and the stator 50 are opposite to each other in the radial direction. In short, the stator 50 has a stator winding 51 of a flat wire structure and a stator core 52 having no pole teeth. The stator winding 51 is assembled radially inside the stator core 52. The stator core 52 has any of the following structures, as in the case of the outer rotor structure.

(A) In the stator 50, an inter-wire member is provided between each wire portion in the circumferential direction, and as the inter-wire member, a magnetic material satisfying the relationship of wt×bs and wm× Br is used, where Wt is the width dimension in the circumferential direction of the inter-wire member of one magnetic pole, bs is the saturation magnetic flux density of the inter-wire member, wm is the width dimension in the circumferential direction of the magnet unit of one magnetic pole, and Br is the residual magnetic flux density of the magnet unit.

(B) In the stator 50, an inter-wire member is provided between each wire portion in the circumferential direction, and a nonmagnetic material is used as the above-described inter-wire member.

(C) The stator 50 is configured such that no inter-conductor members are provided between the conductor portions in the circumferential direction.

The magnets 91 and 92 of the magnet unit 42 are also identical. That is, the magnet unit 42 is configured by using the magnets 91, 92, and the magnets 91, 92 are oriented such that the direction of the easy magnetization axis is parallel to the d-axis on the d-axis side, which is the magnetic pole center, as compared to the q-axis side, which is the magnetic pole boundary. Details of the magnetization directions and the like of the respective magnets 91, 92 are as described above. A ring magnet 95 (see fig. 33) can also be used for the magnet unit 42.

Fig. 40 is a longitudinal sectional view of the rotary electric machine 10 in the case of the inner rotor type, and corresponds to fig. 2. The difference from the structure of fig. 2 will be briefly described. In fig. 40, an annular stator 50 is fixed to the inside of a casing 30, and a rotor 40 is rotatably provided inside the stator 50 with a predetermined air gap therebetween. As in fig. 2, the bearings 21 and 22 are disposed so as to be offset to either side in the axial direction with respect to the axial center of the rotor 40, thereby cantilever-supporting the rotor 40. Further, an inverter unit 60 is provided inside the magnet holder 41 of the rotor 40.

Fig. 41 shows the rotary electric machine 10 of other structures as an inner rotor structure. In fig. 41, a rotary shaft 11 is rotatably supported by a housing 30 through bearings 21 and 22, and a rotor 40 is fixed to the rotary shaft 11. As in the configuration shown in fig. 2 and the like, the bearings 21 and 22 are disposed so as to be offset to either side in the axial direction with respect to the axial center of the rotor 40. The rotor 40 has a magnet holder 41 and a magnet unit 42.

In the rotary electric machine 10 of fig. 41, as a point different from the rotary electric machine 10 of fig. 40, the inverter unit 60 is not provided on the radially inner side of the rotor 40. The magnet holder 41 is coupled to the rotary shaft 11 at a position radially inward of the magnet unit 42. Further, the stator 50 has a stator winding 51 and a stator core 52, and is mounted to the housing 30.

Modification 11

Hereinafter, another configuration of the rotary electric machine having the inner rotor structure will be described. Fig. 42 is an exploded perspective view of rotary electric machine 200, and fig. 43 is a side sectional view of rotary electric machine 200. Here, the vertical direction is shown with reference to the states of fig. 42 and 43.

As shown in fig. 42 and 43, the rotary electric machine 200 includes: a stator 203 having an annular stator core 201 and a multi-phase stator winding 202; and a rotor 204 rotatably disposed inside the stator core 201. The stator 203 corresponds to an armature, and the rotor 204 corresponds to an excitation element. The stator core 201 is configured by laminating a plurality of silicon steel plates, and the stator winding 202 is mounted on the stator core 201. Although not shown, the rotor 204 has a rotor core and a plurality of permanent magnets as magnet units. A plurality of magnet insertion holes are provided at equal intervals in the circumferential direction in the rotor core. Permanent magnets are respectively arranged in the magnet insertion holes, and the permanent magnets are magnetized to be alternately changed in magnetization direction for adjacent magnetic poles. The permanent magnet of the magnet unit preferably has a halbach array as described in fig. 26 or a similar structure. Alternatively, the permanent magnet of the magnet unit preferably includes a characteristic of polar anisotropy in which the orientation direction (magnetization direction) extends in an arc shape between the d-axis, which is the center of the magnetic pole, and the q-axis, which is the boundary of the magnetic pole, as described with reference to fig. 9 and 33.

Here, the stator 203 is preferably of any one of the following structures.

(A) In the stator 203, an inter-wire member is provided between each wire portion in the circumferential direction, and as the inter-wire member, a magnetic material satisfying the relationship of wt×bs and wm× Br is used, where Wt is the width dimension in the circumferential direction of the inter-wire member of one magnetic pole, bs is the saturation magnetic flux density of the inter-wire member, wm is the width dimension in the circumferential direction of the magnet unit of one magnetic pole, and Br is the residual magnetic flux density of the magnet unit.

(B) In the stator 203, an inter-wire member is provided between each wire portion in the circumferential direction, and a nonmagnetic material is used as the above-described inter-wire member.

(C) The stator 203 is configured such that no inter-conductor member is provided between the conductor portions in the circumferential direction.

In addition, in the rotor 204, a plurality of magnets oriented such that the direction of the easy magnetization axis is more parallel to the d-axis on the d-axis side, which is the magnetic pole center, than on the q-axis side, which is the magnetic pole boundary, are used to constitute a magnet unit.

An annular inverter case 211 is provided at one end side in the axial direction of the rotary electric machine 200. The inverter housing 211 is configured such that a housing lower surface is in contact with an upper surface of the stator core 201. The inverter case 211 is provided with: a plurality of power supply modules 212 constituting an inverter circuit; a smoothing capacitor 213 for suppressing pulsation (ripple) of voltage and current generated by switching operation of the semiconductor switching element; a control board 214 having a control unit; a current sensor 215 that detects a phase current; and a resolver stator 216, which is a rotational speed sensor of the rotor 204. The power supply module 212 includes semiconductor switching elements, i.e., IGBTs and diodes.

The inverter case 211 is provided with: a power supply connector 217 connected to a direct current circuit of a battery mounted on the vehicle; and a signal connector 218 for transmitting various signals between the rotary electric machine 200 side and the vehicle side control device. The inverter case 211 is covered with a top cover 219. Direct current power from the vehicle battery is input via the power connector 217, converted to alternating current by the switch of the power module 212, and sent to the stator windings 202 of the respective phases.

A bearing unit 221 for rotatably holding the rotation shaft of the rotor 204 and an annular rear case 222 for accommodating the bearing unit 221 are provided on opposite sides of the inverter case 211 on both sides in the axial direction of the stator core 201. The bearing unit 221 includes, for example, two bearings in a group, and is disposed offset to either side in the axial direction with respect to the axial center of the rotor 204. However, a plurality of bearings in the bearing unit 221 may be provided so as to be distributed on both axial sides of the stator core 201, and the rotary shaft may be supported by the respective bearing arms. The rear case 222 is fastened and fixed to a mounting portion of a gear box, a transmission, or the like of the vehicle by bolts, and the rotary electric machine 200 is mounted on the vehicle side.

A cooling flow path 211a through which a refrigerant flows is formed in the inverter case 211. The cooling flow path 211a is formed by closing a space recessed annularly from the lower surface of the inverter case 211 with the upper surface of the stator core 201. The cooling flow path 211a is formed to surround the coil side end portion of the stator winding 202. A module case 212a of the power module 212 is inserted into the cooling flow path 211a. The rear case 222 also has a cooling flow path 222a formed so as to surround the coil side end portion of the stator winding 202. The cooling flow path 222a is formed by closing a space recessed annularly from the upper surface of the rear case 222 with the lower surface of the stator core 201.

Modification 12

Although the configuration of the rotary electric machine embodied as the rotary excitation type has been described, the rotary electric machine may be modified to be embodied as a rotary armature type rotary electric machine. Fig. 44 shows a structure of a rotary armature type rotary electric machine 230.

In the rotary electric machine 230 of fig. 44, bearings 232 are fixed to the housings 231a and 231b, respectively, and the rotary shaft 233 is rotatably supported by the bearings 232. The bearing 232 is an oil-containing bearing formed by, for example, containing oil in a porous metal. A rotor 234 serving as an armature is fixed to the rotary shaft 233. The rotor 234 includes a rotor core 235 and a multiphase rotor winding 236 fixed to an outer peripheral portion of the rotor core 235. In the rotor 234, the rotor core 235 has a non-slot structure, and the rotor winding 236 has a flat wire structure. That is, the rotor winding 236 is a flat structure in which the circumferential dimension of the region of each phase is larger than the radial dimension.

Further, a stator 237 as an excitation element is provided radially outside the rotor 234. The stator 237 has: a stator core 238 fixed to the housing 231 a; and a magnet unit 239 fixed to the inner peripheral side of the stator core 238. The magnet unit 239 includes a plurality of magnetic poles whose polarities alternate in the circumferential direction, and is configured to be oriented such that the direction of the easy magnetization axis is parallel to the d-axis side, which is the magnetic pole center, than the q-axis side, which is the magnetic pole boundary, in the same manner as the above-described magnet unit 42. The magnet unit 239 has an oriented sintered neodymium magnet having an intrinsic coercive force of 400[ kA/m ] or more and a residual magnetic flux density of 1.0[ T ] or more.

The rotating electric machine 230 of this example is a brushless coreless motor having three coils of two poles, the rotor winding 236 is divided into three, and the magnet unit 239 is two poles. The number of poles and the number of coils of the brush-equipped motor are various in accordance with the use thereof, such as 2:3, 4:10, and 4:21.

A commutator 241 is fixed to the rotation shaft 233, and a plurality of brushes 242 are arranged radially outward of the commutator 241. The rectifier 241 is electrically connected to the rotor winding 236 via a wire 243 buried in the rotary shaft 233. The inflow and outflow of the direct current to and from the rotor winding 236 is performed through the rectifier 241, the brush 242, and the wire 243. The rectifier 241 is configured to be divided appropriately in the circumferential direction according to the number of phases of the rotor winding 236. The brush 242 may be connected to a dc power source such as a battery directly via a harness or may be connected to a dc power source via a terminal block or the like.

A resin washer 244 is provided between the bearing 232 and the commutator 241 as a sealing member on the rotation shaft 233. The outflow of oil oozing out of the oil-impregnated bearing, i.e., bearing 232, to the rectifier 241 side is suppressed by the resin washer 244.

Modification 13

In the stator winding 51 of the rotating electric machine 10, each wire 82 may have a structure having a plurality of insulating films on the inside and the outside. For example, it is preferable to bundle a plurality of wires (wires) with an insulating film into a bundle, and cover the wires with an outer film to construct the wires 82. In this case, the insulating film of the wire rod constitutes an inner insulating film, and the outer layer constitutes an outer insulating film. In particular, the insulating ability of the outer insulating film among the plurality of insulating films in the wire 82 is preferably set higher than the insulating ability of the inner insulating film. Specifically, the thickness of the outer insulating film is set to be larger than that of the inner insulating film. For example, the thickness of the outer insulating film is set to 100 μm, and the thickness of the inner insulating film is set to 40 μm. Alternatively, as the insulating film on the outer side, a material having a lower dielectric constant than the insulating film on the inner side is preferably used. As long as at least any one of the above can be applied. The wire is preferably formed as an aggregate of a plurality of conductive members.

As described above, by enhancing the insulation of the outermost layer in the wire 82, it can be applied to the case of a vehicle system for high voltage. The rotary electric machine 10 can be appropriately driven even on a low-pressure altitude or the like.

Modification 14

In the lead 82 having a plurality of insulating films inside and outside, at least one of the linear expansion coefficient and the bonding strength may be different between the insulating film on the outside and the insulating film on the inside. Fig. 45 shows the structure of the lead 82 according to this modification.

In fig. 45, the wire 82 has: a plurality (4 strands in the figure) of wires 181; an outer layer 182 (outer insulating film) made of, for example, resin surrounding the multi-strand wire 181; and an intermediate layer 183 (intermediate insulating film) filled around each wire 181 in the outer layer 182. The wire 181 has a conductive portion 181a made of a copper material and a conductor film 181b (inner insulating film) made of an insulating material. When used as a stator winding, the outer layer 182 insulates the phases from each other. The wire 181 is preferably formed as an aggregate of a plurality of conductive members.

The intermediate layer 183 has a higher linear expansion coefficient than the conductor film 181b of the wire 181, and has a lower linear expansion coefficient than the outer layer film 182. That is, the linear expansion coefficient increases as the lead 82 approaches the outside. In general, the outer layer film 182 has a higher linear expansion coefficient than the conductor film 181b, but by providing the intermediate layer 183 having an intermediate linear expansion coefficient between the outer layer film 182 and the conductor film 181b, the intermediate layer 183 can be made to function as a buffer member, thereby preventing simultaneous cracking of the outer layer side and the inner layer side.

In the lead 82, the conductive portion 181a and the conductive film 181b are bonded to each other in the wire 181, and the conductive film 181b and the intermediate layer 183, and the outer layer 182 are bonded to each other, and the bonding strength is weaker as the bonding portion is located outside the lead 82. That is, the adhesive strength between the conductive portion 181a and the conductive film 181b is weaker than the adhesive strength between the conductive film 181b and the intermediate layer 183 and the adhesive strength between the intermediate layer 183 and the outer layer 182. When the adhesive strength of the conductor film 181b and the intermediate layer 183 is compared with the adhesive strength of the intermediate layer 183 and the outer layer 182, it is preferable that one (the outer side) of the latter is weaker or the same. The strength of adhesion between the films can be grasped, for example, by the tensile strength required for peeling the two films. By setting the bonding strength of the wire 82 as described above, even if a difference in temperature between the inside and outside occurs due to heat generation or cooling, cracking (co-cracking) can be suppressed in both the inner layer side and the outer layer side.

Here, the heat generation and the temperature change of the rotating electrical machine are mainly generated as copper loss generated from the conductive portion 181a of the wire 181 and iron loss generated from the inside of the iron core, but the two losses are transmitted from the conductive portion 181a in the wire 82 or the outside of the wire 82, and the heat generation source is not in the intermediate layer 183. In this case, the intermediate layer 183 has an adhesive force that can be buffered against both sides, so that the above-described simultaneous cracking can be prevented. Therefore, the present invention can be suitably used even in a field having a high pressure resistance or a large temperature change, such as a vehicle application.

Supplementary explanation is made below. The wire 181 may be, for example, an enamel wire, and in this case, has a resin film layer (conductor film 181 b) such as PA, PI, PAI. Further, it is preferable that the outer layer film 182 on the outer side of the wire 181 is composed of the same PA, PI, PAI and the like, and is thick. Thus, the film breakage due to the linear expansion coefficient difference is suppressed. In addition to the thickening of the material such as PA, PI, PAI, PPS, PEEK, fluorine, polycarbonate, silicon, epoxy resin, polyethylene naphthalate, LCP, and other materials having a dielectric constant smaller than PI and PAI are preferably used as the outer layer 182 to increase the conductor density of the rotating electrical machine. In the case of the resin, even if the PI or PAI film is thinner than the conductor film 181b or the thickness is the same as the conductor film 181b, the insulating ability is improved, and the occupancy of the conductive portion is thereby improved. In general, the above resin has better insulation than the insulating film of the enamel wire. Of course, there are also examples in which the dielectric constant is deteriorated by the molding state and the mixture. Among them, PPS and PEEK are suitable as the outer layer film of the second layer because their linear expansion coefficient is generally larger than that of enamel film but smaller than that of other resins.

Further, the adhesive strength between the two films (the intermediate insulating film, the outer insulating film) on the outer side of the wire 181 and the enamel film of the wire 181 is preferably weaker than the adhesive strength between the copper wire and the enamel film in the wire 181. This suppresses the phenomenon that the enamel film and the two films are destroyed at one time.

When a water cooling structure, a liquid cooling structure, or an air cooling structure is added to the stator, it is basically considered that thermal stress and impact stress are applied from the outer layer film 182. However, even when the insulating layer of the wire 181 and the two films are made of different resins, the thermal stress and impact stress can be reduced by providing a portion to which the films are not adhered. That is, the insulating structure is completed by providing a space between the insulating material and the wire (enamel wire), and disposing fluorine, polycarbonate, silicon, epoxy resin, polyethylene naphthalate, and LCP. In this case, it is preferable to bond the outer layer film and the inner layer film using an adhesive material composed of an epoxy resin or the like having a low dielectric constant and a low linear expansion coefficient. In this way, not only mechanical strength can be improved, but also damage to the film due to friction caused by vibration or the like of the conductive portion or damage to the outer layer film due to a difference in linear expansion coefficient can be suppressed.

As the outermost layer of the wire 82 having the above-described structure, which is subjected to mechanical strength, fixing, and the like, and is usually the final step of winding around the stator, a resin having good moldability such as epoxy resin and PPS, PEEK, LCP and having properties close to those of an enamel film such as dielectric constant and linear expansion coefficient is preferable.

Resin encapsulation is usually performed using polyurethane or silicon, but the linear expansion coefficient of the resin is nearly doubled as compared with other resins, and thermal stress which can shear the resin is generated. Therefore, it is not suitable for use in applications of 60V or more where strict insulation regulations are used internationally. In this regard, the above-described respective requirements can be satisfied by a final insulation step which is easily performed by injection molding or the like using an epoxy resin, PPS, PEEK, LCP, or the like.

Modifications other than the above will be described below.

The distance DM between the armature-side surface of the magnet unit 42 in the radial direction and the axial center of the rotor may be 50mm or more. Specifically, for example, the distance DM between the radially inner surface of the magnet unit 42 (specifically, the first magnet 91 and the second magnet 92) shown in fig. 4 and the axial center of the rotor 40 may be 50mm or more.

As a rotating electric machine having a grooving-free structure, a small rotating electric machine used for a model or the like whose output is several tens W to several hundreds W is known. Moreover, the present inventors have not found an example in which a grooving-free structure is generally employed in a large-sized rotary electric machine for industrial use, such as more than 10 kW. The inventors have studied for its reason.

In recent years, the mainstream rotary electric machines are roughly classified into the following four types. The rotating electric machine refers to a brushed motor, a cage-type induction motor, a permanent magnet synchronous motor, and a reluctance motor.

In a brushed motor, exciting current is supplied via brushes. Therefore, in the case of a brush motor of a large-sized device, the brushes become large, and maintenance becomes complicated. With the remarkable development of semiconductor technology, brushless motors such as induction motors have been gradually replaced. On the other hand, in the field of small motors, many coreless motors are also being supplied to the market from the viewpoint of low inertia and economical efficiency.

In a cage induction motor, the principle is as follows: torque is generated by receiving a magnetic field generated in a stator winding on a primary side with an iron core of a rotor on a secondary side to concentrate an induced current to a cage conductor to form a reaction magnetic field. Therefore, the removal of the iron core on both the stator side and the rotor side is not necessarily a good countermeasure from the viewpoint of downsizing and high efficiency of the apparatus.

Reluctance motors are motors that use changes in the reluctance of the core, and in principle, it is undesirable to eliminate the core.

In recent years, in permanent magnet synchronous motors, IPM (i.e., embedded magnet rotor) has become a mainstream, and in particular, in large-sized devices, IPM is generally used unless special conditions are present.

IPM has a characteristic of both magnet torque and reluctance torque, and operates while adjusting the proportion of the above torque in time by inverter control. Therefore, IPM is a small-sized motor with excellent controllability.

According to the analysis of the present inventors, when the radial distance DM between the armature-side surface in the radial direction of the magnet unit and the axial center of the rotor, that is, the radius of the stator core of the general inner rotor is plotted on the horizontal axis, the torque of the rotor surface generating the magnet torque and the reluctance torque is as shown in fig. 46.

The magnet torque is represented by the following formula (eq 1), and the potential position is determined by the intensity of the magnetic field generated by the permanent magnet, whereas the reluctance torque is represented by the following formula (eq 2), and the magnitude of the inductance, particularly the q-axis inductance, determines the potential position.

Magnet torque = k.ψ.iq.i.i.i.eq 1

Reluctance torque=k· (Lq-Ld) Iq. Id. Iq. Eq2

Here, the field strength of the permanent magnet and the magnitude of the inductance of the winding are compared with DM. The strength of the magnetic field produced by the permanent magnets, i.e. the magnetic flux ψ, is proportional to the total area of the permanent magnets on the side opposite the stator. If the rotor is cylindrical, the surface area of the cylinder is the surface area of the cylinder. Strictly speaking, the presence of the N-pole and S-pole is proportional to the exclusive area of half of the cylinder surface. The surface area of the cylinder is proportional to the radius of the cylinder and the length of the cylinder. That is, if the cylinder length is constant, it is proportional to the radius of the cylinder.

Another oneIn this respect, although the inductance Lq of the winding is affected by the shape of the core, the sensitivity is low, and since the inductance Lq of the winding is proportional to the square of the number of turns of the stator winding, it is highly correlated with the number of turns. When μ is the magnetic permeability of the magnetic circuit, N is the number of turns, S is the cross-sectional area of the magnetic circuit, and δ is the effective length of the magnetic circuit, inductance l=μ·n ² X S/delta. Since the number of turns of the winding depends on the size of the winding space, if a cylindrical motor, it depends on the winding space of the stator, i.e. the slot area. As shown in fig. 47, since the shape of the slit is substantially quadrangular, the slit area is proportional to the product a×b of the circumferential length dimension a and the radial length dimension b.

The circumferential length of the slit increases as the diameter of the cylinder increases, and is proportional to the diameter of the cylinder. The radial length dimension of the slot is proportional to the diameter of the cylinder. That is, the cutting area is proportional to the square of the diameter of the cylinder. Further, as can also be seen from the above equation (eq 2), the reluctance torque is proportional to the square of the stator current, and thus the performance of the rotating electrical machine is determined by how much large current flows, and the performance depends on the slot area of the stator. In summary, if the length of the cylinder is constant, the reluctance torque is proportional to the square of the diameter of the cylinder. Fig. 46 is a graph on the basis of which the relationship between the magnet torque, the reluctance torque, and DM is plotted.

As shown in fig. 46, the magnet torque increases linearly with respect to DM, and the reluctance torque increases as a quadratic function with respect to DM. It can be seen that when DM is relatively small, the magnet torque dominates and the reluctance torque dominates as the stator core radius becomes larger. The inventors have reached the following conclusion: under prescribed conditions, the intersection point of the magnet torque and the reluctance torque in fig. 46 is approximately in the vicinity of the stator core radius=50 mm. That is, in a 10 kW-level motor in which the radius of the stator core is sufficiently larger than 50mm, it is difficult to eliminate the core because the use of reluctance torque is currently the main stream, and it is estimated that this is one of the reasons why a slot-free structure is not adopted in the field of large-sized equipment.

In the case of a rotating electrical machine in which an iron core is used for a stator, magnetic saturation of the iron core is always a technical problem. In particular, in the radial gap type rotating electrical machine, the longitudinal cross-sectional shape of the rotating shaft is a sector shape for each magnetic pole, the width of the magnetic path is narrower toward the inner peripheral side of the machine, and the inner peripheral side dimension of the tooth portion forming the slot determines the performance limit of the rotating electrical machine. Whatever high-performance permanent magnet is used, the performance of the permanent magnet cannot be fully exerted when magnetic saturation occurs in the above-described portion. In order not to generate magnetic saturation in the above-described portion, the inner peripheral diameter is designed to be large, which leads to an increase in the size of the apparatus.

For example, in a rotary electric machine with distributed windings, if a three-phase winding is used, each magnetic pole shares and flows magnetic flux with three to six pole teeth, but since magnetic flux tends to concentrate on the pole teeth circumferentially forward, magnetic flux does not flow uniformly in the three to six pole teeth. In this case, the magnetic flux flows concentratedly to a part (for example, one or two) of the pole teeth, and the pole teeth that saturate the magnetism as the rotor rotates also move in the circumferential direction. This is also the main cause of the generation of the notch ripple.

In summary, in a rotating electrical machine having a non-slot structure in which DM is 50mm or more, it is desirable to remove pole teeth to eliminate magnetic saturation. However, when the pole teeth are removed, the reluctance of the magnetic circuits in the rotor and stator increases, resulting in a decrease in torque of the rotary electric machine. As a reason for the increase in magnetic resistance, for example, an air gap between the rotor and the stator may become large. Therefore, in the rotating electrical machine having the above-described non-slot structure in which DM is 50mm or more, there is room for improvement in enhancing torque. Therefore, the advantage of applying a structure capable of enhancing the torque is great in the rotating electrical machine having the non-slot structure in which the DM is 50mm or more.

In addition, the distance DM between the surface of the radial armature side of the magnet unit and the axial center of the rotor in the radial direction may be 50mm or more in the rotary electric machine of the inner rotor structure, not limited to the rotary electric machine of the outer rotor structure.

In the stator winding 51 of the rotating electrical machine 10, the linear portion 83 of the wire 82 may be formed as a single layer in the radial direction. In the case where the linear portion 83 is arranged in a plurality of layers in the radial direction, the number of layers may be arbitrary, and may be 3 layers, 4 layers, 5 layers, 6 layers, or the like.

For example, in the configuration of fig. 2, the rotation shaft 11 is provided to protrude in the axial direction to both one end side and the other end side of the rotary electric machine 10, but may be modified so as to protrude only to one end side. In this case, the rotary shaft 11 is preferably formed such that a portion cantilever-supported by the bearing unit 20 is an end portion and extends outward in the axial direction. In this configuration, since the rotation shaft 11 does not protrude into the inverter unit 60, the internal space of the inverter unit 60 can be used more, and in detail, the internal space of the cylindrical portion 71 can be used more.

In the rotating electrical machine 10 having the above-described structure, the non-conductive grease is used for the bearings 21 and 22, but the structure may be modified so that the conductive grease is used for the bearings 21 and 22. For example, a conductive grease containing metal particles, carbon particles, or the like is used.

As a configuration for rotatably supporting the rotary shaft 11, bearings may be provided at two positions on one end side and the other end side in the axial direction of the rotor 40. In this case, in the structure of fig. 1, the bearings are preferably provided at two locations on one end side and the other end side with the inverter unit 60 interposed therebetween.

In the rotating electrical machine 10 having the above-described structure, the intermediate portion 45 of the magnet holder 41 in the rotor 40 has the inner shoulder portion 49a and the annular outer shoulder portion 49b, but may be configured to have a flat surface without providing the shoulder portions 49a and 49 b.

In the rotating electric machine 10 having the above-described structure, the conductor 82a is an aggregate of the multi-strand wires 86 in the wire 82 of the stator winding 51, but this may be modified, or a rectangular wire having a rectangular cross section may be used as the wire 82. Further, a round wire having a circular cross section or an elliptical cross section may be used as the wire 82.

In the rotating electrical machine 10 having the above-described structure, the inverter unit 60 is provided on the radial inner side of the stator 50, but in addition to this, the inverter unit 60 may not be provided on the radial inner side of the stator 50. In this case, the radially inner region of the stator 50 can be taken as a space. In addition, a component different from the inverter unit 60 can be disposed in the internal region.

In the rotating electric machine 10 having the above-described structure, the casing 30 may not be provided. In this case, the rotor 40, the stator 50, and the like may be held in a part of, for example, a wheel or another vehicle component.

In the above-described embodiment, when the rotating electrical machine having the inner rotor structure is employed, as in the first embodiment, a part or all of the cylindrical portion 43 (field element core member) may be disposed at a position closer to the stator side (armature side) than the second intersection point P2 of the q-axis and the orientation circular arc OA in the radial direction. In the case of a rotating electrical machine having an inner rotor structure, in which the saturation magnetic flux density of the cylindrical portion 43 is larger than the residual magnetic flux density of the magnets 91 and 92, the radial thickness dimension of the cylindrical portion 43 may be smaller than the radial dimension L1 from the third intersection point P3 of the q-axis and the stator-opposite side circumferential surface to the second intersection point P2. In addition, when a rotating electrical machine of an inner rotor structure is employed, the magnet unit 42 and the cylindrical portion 43 satisfying the relationship of br×wh+.bs×wsc may also be used.

In the above embodiment, the entire cylindrical portion 43 is arranged on the stator side of the second intersection point P2 of the q-axis and the orientation circular arc OA in the radial direction, but as shown in fig. 48, only a part of the cylindrical portion 43 may be arranged.

In the above embodiment, when the saturation magnetic flux density of the cylindrical portion 43 is smaller than the residual magnetic flux density of the magnets 91, 92, the thickness dimension Wsc of the cylindrical portion 43 in the radial direction is thicker than the dimension L1 in the radial direction from the third intersection point P3 to the second intersection point P2 in order to suppress the leakage of the magnets. That is, as shown in fig. 48, the thickness dimension of the cylindrical portion 43 is preferably set.

The orientation circular arc OA in the above embodiment is set such that the tangential line TA1 at the first intersection point P1 on the orientation circular arc OA is parallel to the d-axis, but may be set at an angle close to parallel to the d-axis. For example, the angle between the direction of the d-axis and the tangential line TA1 may be set to be in the range of 80 degrees to 100 degrees.

The disclosure of the present specification is not limited to the illustrated embodiments. The present disclosure includes the illustrated embodiments and modifications based on them by a person skilled in the art. For example, the present disclosure is not limited to the combinations of parts and/or elements shown in the embodiments. The disclosure may be implemented in various combinations. The present disclosure may have an additional portion that can be added to the embodiment. The present disclosure includes embodiments in which components and/or elements of the embodiments are omitted. The present disclosure includes alternatives or combinations of parts and/or elements from one embodiment to another. The technical scope of the disclosure is not limited to the description of the embodiments. The technical scope of the disclosure is to be understood as indicated by the description of the claims, and also includes all modifications that are equivalent in meaning and scope to the description of the claims.

Although the present disclosure has been described based on the embodiments, it should be understood that the present disclosure is not limited to the above-described embodiments, constructions. The present disclosure also includes various modifications and modifications within the equivalent scope. In addition, various combinations and modes, including only one element, more than one or less than one other combinations and modes, are also within the scope and spirit of the present disclosure.

Claims (8)

1. A rotary electric machine, which is capable of rotating in a direction perpendicular to a rotation axis,

the rotary electric machine includes an exciting element having a magnet portion including a plurality of magnetic poles whose polarities alternate in a circumferential direction, and an armature having armature windings of a plurality of phases, either one of the exciting element and the armature being set as a rotor,

the magnet part has a magnet with an easy axis of magnetization oriented in an arc shape and an arc-shaped magnet magnetic circuit formed along the easy axis of magnetization,

the magnetic body has an intrinsic coercive force of 400 kA/m or more and a residual magnetic flux density of 1.0T or more,

the magnet magnetic circuit includes a magnetic circuit on an orientation circular arc which is centered on a center point set on a magnetic pole boundary of the magnet, i.e., q-axis, and passes through a magnetic pole center of the magnet, i.e., d-axis, and a first intersection point of an armature-side peripheral surface on the armature side of the peripheral surface of the magnet,

The exciting element includes a soft magnetic body, namely an exciting element core member, at a position on the opposite side of the armature from the magnet portion,

the field element core member and the magnet are laminated in a radial direction,

a part or all of the excitation element core member is arranged radially closer to the armature side than a second intersection of the q-axis and the orientation circular arc.

2. The rotating electrical machine according to claim 1, wherein,

the orientation circular arc is set such that a tangent line at a first intersection point on the orientation circular arc is parallel to the d-axis.

3. The rotating electrical machine according to claim 1, wherein,

when the saturation magnetic flux density of the field element core member is larger than the residual magnetic flux density of the magnet, the thickness dimension in the radial direction of the field element core member is thinner than the dimension in the radial direction from a third intersection point to the second intersection point, the third intersection point being an intersection point of the q-axis and an armature-opposite-side circumferential surface on the armature-opposite side of the circumferential surfaces of the magnet.

4. A rotary electric machine according to claim 2, wherein,

when the saturation magnetic flux density of the field element core member is larger than the residual magnetic flux density of the magnet, the thickness dimension in the radial direction of the field element core member is thinner than the dimension in the radial direction from a third intersection point to the second intersection point, the third intersection point being an intersection point of the q-axis and an armature-opposite-side circumferential surface on the armature-opposite side of the circumferential surfaces of the magnet.

5. A rotary electric machine according to any one of claim 1 to 4, wherein,

the following magnet and excitation element core member were used: when the residual magnetic flux density of the magnet portion is Br, the saturation magnetic flux density of the field element core member is Bs, the distance from the center point to the first intersection point is Wh, and the thickness dimension of the field element core member in the radial direction is Wsc, the relationship of br×wh and bs×wsc is satisfied.

6. A rotary electric machine according to any one of claim 1 to 4, wherein,

the armature winding has a wire portion arranged at a position facing the exciting element at a predetermined interval in a circumferential direction,

in the case of the armature being a hollow,

a wire-to-wire member is provided between the wire portions in the circumferential direction, and as the wire-to-wire member, a magnetic material or a non-magnetic material satisfying the relationship of Wt×Bs.ltoreq.Wm×Br is used, where Wt is the width dimension in the circumferential direction of the wire-to-wire member of one magnetic pole, bs is the saturation magnetic flux density of the wire-to-wire member, wm is the width dimension in the circumferential direction of the magnet portion of one magnetic pole, and Br is the residual magnetic flux density of the magnet portion,

Alternatively, no inter-conductor member is provided between the conductor portions in the circumferential direction.

7. A rotary electric machine according to any one of claim 1 to 4, wherein,

the armature winding has a wire portion arranged at a position facing the exciting element at a predetermined interval in a circumferential direction,

each wire constituting the wire portion is a wire assembly as follows: the strands of wire are bundled and the resistance between the bundled wires is greater than the resistance of the wire itself.

8. A rotary electric machine according to any one of claim 1 to 4, wherein,

the armature winding has a wire portion arranged at a position facing the exciting element at a predetermined interval in a circumferential direction,

the radial thickness dimension of the wire portion is smaller than the width dimension of a corresponding circumferential direction in a magnetic pole.