JP6611688B2 - Resolver and rotating electric machine - Google Patents

Description

  The present invention relates to a resolver and a rotating electric machine.

In general, a rotating electrical machine for a vehicle is used as a synchronous motor when the engine is started, and as an AC generator or a synchronous motor while the engine is operating. When used as a synchronous motor, it is necessary to control the energization timing to the coils wound around the stator core and the rotor core. Therefore, a rotation angle detector is arranged on the rotation shaft on which the rotor core is mounted so as to detect the rotation angle of the rotation shaft.
Here, when using a change in magnetism as the rotation angle detector, for example, a resolver or a Hall element, a part of the magnetic flux generated by energizing the rotor coil wound around the rotor core is passed through the rotation shaft. There is a risk of leaking into the rotation angle detector and lowering the angle detection accuracy. The case where a resolver is used as a rotation angle detector will be described. The resolver detects the angle of the resolver rotor by utilizing the magnetic permeance change between the resolver stator and the resolver rotor. Therefore, when leakage magnetic flux flows through the resolver, there is a problem that noise components are superimposed on the output waveform and rotation angle detection accuracy is lowered.
Thus, in the prior art, Patent Document 1 proposes a configuration in which a flux barrier is arranged on the rotor core for the purpose of reducing the leakage flux of the motor interlinked with the resolver stator.

JP 2013-223388 A

  However, the improvement in angle detection accuracy is insufficient only by arranging the flux barrier on the rotor core as in Patent Document 1. In order to improve the angle detection accuracy, it is necessary to cancel the leakage flux of the motor. In the prior art document 1, since the flux barrier is arranged on the entire circumference and the spatial order and phase of the permeance are not considered, it is difficult to cancel the leakage magnetic flux.

  The present invention has been made in view of the above, and an object of the present invention is to provide a resolver that can suppress a decrease in angle detection accuracy due to a leakage flux of a motor.

  In order to achieve the above-described object, the resolver of the present invention is fixed to a rotating shaft to which a conductor that generates a magnetomotive force of the space Np (Np ≠ N, N: number of salient poles of the resolver rotor) in order by energization is fixed. The resolver includes a resolver rotor, and a zero-order magnetomotive force is generated on the surface of the resolver rotor by a rotor coil connected to the conductor. In the resolver, the resolver rotor generates Np-order magnetic flux generated by a current flowing through the conductor. It has a spatial Np-order permeance arranged in a phase to be reduced.

  According to the present invention, the angle detection accuracy can be improved by canceling the leakage flux of the motor.

It is a sectional side view which shows the structure of the rotary electric machine for vehicles incorporating the resolver in Embodiment 1 of this invention. It is a top view of the resolver in Embodiment 1 of this invention. It is a top view of the resolver (with a bridge) in Embodiment 1 of this invention. It is sectional drawing which shows the flow of the magnetic flux resulting from the conductor in the rotary electric machine for vehicles incorporating the resolver in Embodiment 1 of this invention. It is a sectional side view which shows the structure of the rotary electric machine for vehicles incorporating the resolver in Embodiment 2 of this invention. It is a top view of the resolver in Embodiment 2 of this invention. It is an example of the winding number pattern of the exciting winding in Embodiment 2 of this invention. It is an example of the number pattern of sin windings in Embodiment 2 of this invention. It is an example of the winding number pattern of the cos winding in Embodiment 2 of this invention. It is a top view of the resolver in Example 2 of Embodiment 2 of this invention. It is a top view of the resolver in Example 3 of Embodiment 2 of this invention. It is a top view of the resolver in Example 1 of Embodiment 3 of this invention. It is a waveform of the gap length of the resolver in Example 1 of Embodiment 3 of this invention. This is dependency of resolver angle error in (Embodiment 1) of Embodiment 3 of the present invention (number of turns of rotor coil × Np-order amplitude of gap length / minimum gap length before adding Np-order). This is the dependency of the resolver angle error in Example 1 of Embodiment 3 of the present invention on the direction in which the Np-th order gap length is minimized. It is a top view of the resolver in Example 2 of Embodiment 3 of this invention. It is a waveform of the gap length of the resolver in Example 2 of Embodiment 3 of this invention. It is a top view of the resolver in Example 3 of Embodiment 3 of this invention. This is the dependence of the angle error of the resolver in the second embodiment of the present invention on the flux barrier range (electrical angle) (flux barrier center: -90 degrees). This is the dependency of the angle error of the resolver in the second embodiment of the present invention on (radial length of the flux barrier) / (minimum gap length). This is the dependence of the angle error of the resolver in the second embodiment of the present invention on the flux barrier range (electrical angle) (flux barrier center: 90 degrees).

  Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. In the drawings, the same reference numerals indicate the same or corresponding parts. The following embodiment relates to a variable reluctance resolver with improved angle detection accuracy and a rotating electrical machine using the resolver.

Embodiment 1 FIG.
1 is a side sectional view of a rotating electrical machine for a vehicle incorporating a resolver according to Embodiment 1 of the present invention. FIG. 2 is a top view of the resolver in the first embodiment of the present invention. FIG. 3 is a top view of the resolver (with a bridge) according to Embodiment 1 of the present invention. FIG. 4 is a cross-sectional view showing the flow of magnetic flux caused by the conductor in the vehicular rotating electrical machine incorporating the resolver according to the first embodiment of the present invention.

  In FIG. 1, the rotating electrical machine 1 includes a rotor 1 a, a stator 1 b, and a housing 2. The stator covers the outer periphery of the rotor via a gap and is disposed to face the rotor. The housing 2 is a housing of the vehicular rotating electrical machine 1 and is configured by fixing a pair of left and right brackets 3 and 4 with screws 5. At this time, the stator core 24 is sandwiched and fixed between the brackets 3 and 4.

  Reference numeral 7 is a rotating shaft, and reference numerals 8 and 9 are bearings that rotatably support the rotating shaft 7 on the housing 2. These bearings 8 and 9 are attached to brackets 3 and 4, respectively. Reference numeral 10 is a pulley around which a belt (not shown) is suspended, and this pulley 10 is fixed to the rotary shaft 7 by a nut 11.

  The rotor core 12 is configured by integrating a pair of core members 16 and 17, and is press-fitted and fixed to the rotary shaft 7. Each of the core members 16 and 17 has claw-shaped magnetic pole portions 16b and 17b extending from the cylindrical portions 16a and 17a in which the bobbin around which the rotor coil 13 is wound is accommodated to the position where they cover the rotor coil 13 and intersect each other. Each is extended. The claw-shaped magnetic pole portions 16b and 17b have a shape arranged at a constant pitch with a predetermined interval along the circumferential direction.

  Cooling fans 18 and 19 are attached to the axial end surfaces of the core members 16 and 17, respectively.

  A stator coil 25 is wound around the stator core 24, and the stator coil 25 is connected to a three-phase inverter circuit (not shown). Reference numeral 26 denotes a brush that contacts the slip ring 22 to form an energization path, and the slip ring 22 is fixed to the rotary shaft 7.

  A pulley 10 is attached to one shaft end of the rotating shaft 7, and a resolver 31 as a rotation angle detector of the rotating shaft 7 is disposed on the opposite shaft end.

  In the resolver 31, a rotation shaft mounting hole (hereinafter referred to as “mounting hole”) 102 of the resolver rotor 32 is fitted and fixed to the shaft end portion of the rotation shaft 7, and a resolver stator 34 is fixed to the bracket 3. A resolver coil 34a is wound and configured. The resolver coil 34a is composed of an excitation winding and an output winding. However, each winding may have 0 turns.

The above configuration is already known, and detailed description is omitted.
In the rotating electrical machine, the coil disposed so as to surround the rotating shaft 7 is only the rotor coil 13, and a magnetic field current is passed through the rotor coil 13 via the brush 26 and the slip ring 22 to magnetize the rotor core 12. ing. The product of the number of turns of the rotor coil and the field current flowing therethrough is a magnetomotive force source that generates magnetic flux in the axial direction of the rotating shaft 7.

  Most of the magnetic flux generated by energizing the rotor coil 13 with a field current follows the path A flowing from the rotor to the stator as shown in FIG. It leaks to the resolver 31 through the path B that passes. When the leakage magnetic flux flows into the resolver stator 34, a noise voltage component is induced in the resolver coil 34a by the magnetic flux and is superimposed on the signal component, which causes a problem that the angle detection accuracy is lowered.

  Two conductors for supplying current from the slip ring to the rotor coil are arranged in the axial direction near the resolver. Since the noise voltage component is induced by the current flowing through the conductor and is superimposed on the signal component, there is a problem that the angle detection accuracy is lowered. In the conventional method of disposing a flux barrier in Patent Document 1 or the like, the magnetic flux interlinked with the resolver stator is somewhat reduced by increasing the magnetic resistance of the resolver rotor, but the magnetic fluxes cannot be canceled out. Therefore, in the first embodiment, the method described below is used in order to further prevent a decrease in angle detection accuracy.

  In the first embodiment, as shown in FIG. 2, an uneven outer peripheral surface (hereinafter referred to as “mountain convex portion” is referred to as “mountain portion 32a”, “ A flux barrier 101 (gap) is opened in the axial direction in a resolver rotor 32 having a “crest-like concave portion” (referred to as “groove portion 32 b”) along the outer peripheral surface of the rotary shaft 7 (so as to surround the mounting hole 102). ing. Here, the reason why the magnetic fluxes cancel each other will be described. First, the resolver, the relative position of the rotating electrical machine, the winding direction of the rotor coil, and the electrical connection between the rotor coil and the conductor will be described. When looking at the cross section of the resolver and the conductor in the vicinity of the resolver, the rotor coil is viewed in the same direction as the paper surface is viewed from above. When the right screw is turned so that the right screw advances in the axial direction from the rotor coil to the resolver, if the current direction flows through the rotor coil in the same direction as the right screw turns, current flows in the traveling direction of the right screw. The position of any flowing conductor is 0 degree, the center at that time is the center of rotation, and the direction opposite to the current of the rotor coil at that time is positive. That is, when FIG. 2 is viewed from above, the direction from the center of rotation to the conductor (current direction in FIG. 2) through which current flows downward is defined as 0 degrees and the counterclockwise direction is positive. The center of the flux barrier is arranged in the direction of 270 degrees.

  There are two magnetic fluxes that are the main cause of the decrease in angle detection accuracy. Magnetic flux can be expressed by the product of magnetomotive force and permeance. First, the magnetomotive force can be expressed by the product of the number of turns and the current. Permeance is the reciprocal of magnetoresistance. The magnetic resistance is usually the gap (between the stator and the rotor) and the flux barrier, which is more dominant than the stator core and rotor core, which are usually made of electromagnetic steel plates (not limited to iron). Focus on the permeance. Magnetomotive force and permeance are usually a combination of multiple periodic components. Magnetic flux can be expressed by superposition of products of magnetomotive force and permeance components. Since each component of magnetomotive force and permeance is expressed by a trigonometric function, each component of magnetic flux is expressed by a product of trigonometric functions. Therefore, using the trigonometric product-sum formula, the order of the magnetic flux is | the order of the magnetomotive force ± the order of the permeance |. In the conductor arrangement of FIG. 2, the magnetomotive force of the conductor is primary in space. In addition, unless otherwise noted, the order is a spatial order (one pulsation in one circumferential direction is the primary). The first main magnetic flux that is unnecessary for angle detection is the primary magnetic flux generated by the magnetomotive force (primary component) and permeance (zero-order component) of the conductor shown in FIG. As shown in the cross section including the axis of FIG. 4, the direction of the magnetic flux is upward in the portion where the conductor is present and downward in the resolver stator. The direction of the magnetic flux differs depending on whether or not the conductor passes through the resolver. The second is the primary magnetic flux generated by the magnetomotive force (zero-order component) of the rotor coil and the permeance (primary component) of the flux barrier of FIG. The main component of the permeance of the flux barrier in FIG. 2 is the spatial primary component, because the core has higher permeability than the flux barrier (air gap), so the permeance is larger, and the radial direction per angle of the part with the flux barrier This is because the permeance is small, the radial permeance per angle of the portion without the flux barrier is large, and it is once in the circumferential direction. As described above, the second magnetic flux is generated by adding the first-order permeance to the conventional resolver by the flux barrier. Since this magnetic flux has the same order as that of the first magnetic flux, adjusting the radial length, width, and center position of the flux barrier can adjust the amplitude and phase of the permeance primary to cancel the two magnetic fluxes. . Thereby, the fall of angle detection accuracy can be prevented. If the main component of the permeance is first order, the strength of the rotor core can be improved by changing a part of the flux barrier to a core as shown in FIG. 3 (the changed core part is called a bridge). it can. The main component of permeance can be grasped, for example, by performing frequency analysis using the reciprocal of (gap length + flux barrier radial direction length) as a function of the circumferential direction.

  Note that the core of the resolver stator 34 and the core of the resolver rotor 32 having a gap in the radial direction are generally configured by laminating thin steel plates. This is to prevent loss caused by the generation of eddy currents in the core due to the magnetic flux flowing in the radial direction. At this time, the respective thin plates constituting the core are often processed by pressing. The flux barrier 101 in each embodiment can be formed at the same time as pressing a thin plate, and no separate parts other than the iron core material are required. Therefore, the cost does not increase when the present invention is applied to the resolver. .

  That is, the first embodiment includes a resolver rotor fixed to a rotating shaft to which a conductor that generates the next magnetomotive force in space Np (Np ≠ N, N: the number of salient poles of the resolver rotor) by energization is fixed, In a variable reluctance resolver in which a zero-order magnetomotive force is generated on the surface of the resolver rotor by a rotor coil connected to the conductor, the resolver rotor has a phase that reduces Np-order magnetic flux generated by a current flowing through the conductor. The space Np-order permeance arranged at. As a result, Np-order magnetic flux generated by the magnetomotive force (Np-order component) and permeance (zero-order component) of the conductor is changed to Np generated by the magnetomotive force (zero-order component) of the rotor coil and permeance (Np-order component) of the flux barrier. By canceling with the next magnetic flux, a decrease in angle detection accuracy can be prevented.

  In addition, the distribution of the permeance of the space Np order with respect to the straight line connecting the circumferential position where the space Np order magnetic flux generated by the conductor generating the magnetomotive force of the order of space Np (Np ≠ N) is maximum and the rotation center is obtained. In the case of line symmetry, the effect of canceling the Np-order permeance amplitude of the space Np-order magnetic flux generated by each of the conductor and the rotor coil is larger than in the case of non-line symmetry. The amplitude of the Np-order permeance can be small. Thereby, since the radial direction length of the rotor core of a part with a flux barrier can be thickened, intensity | strength can be improved.

  Further, when a flux barrier (gap) is arranged as Np-order permeance in the rotor core, no separate parts other than the iron core material are required, which increases the cost when applying the present invention to the resolver. Therefore, it is possible to prevent the angle detection accuracy from being lowered.

Embodiment 2. FIG.
Next, a second embodiment of the present invention will be described. The second embodiment is the same as the first embodiment described above except for the contents described below. FIG. 5 is a side sectional view showing a configuration of a vehicular rotating electrical machine incorporating the resolver according to the second embodiment. FIG. 6 is a top view of the resolver in the second embodiment. FIG. 7 is an example of the number pattern of excitation windings in the second embodiment. FIG. 8 is an example of the number pattern of sin windings in the second embodiment. FIG. 9 is an example of the number of turns pattern of the cos winding according to the second embodiment. FIG. 10 is a top view of the resolver in Example 2 of the second embodiment. FIG. 11 is a top view of the resolver in the third example of the second embodiment.

  First, the difference in arrangement around the resolver between the first embodiment and the second embodiment will be described. In the first embodiment, the conductor disposed near the inner periphery of the resolver rotor 32 does not penetrate the inner periphery. On the other hand, the second embodiment is different in that a conductor penetrates the inner periphery of the resolver rotor 32. However, the effect that the two magnetic fluxes described in the first embodiment cancel each other is also in the second embodiment, and a decrease in the angle detection accuracy of the resolver can be prevented. Details are described below.

Below, it demonstrates including the structure of a rotary electric machine.
FIG. 1 shows an example in which the resolver 31 is arranged at the shaft end portion of the rotating shaft 7. The resolver rotor 32 at the shaft end portion is arranged on the rotating shaft in the order of the slip ring 22, the bearing 8, and the rotor core 12. Yes. These are members necessary for configuring the rotating electrical machine for the vehicle, but the arrangement is not limited to this order. For example, the arrangement is in the order of slip ring, resolver rotor, bearing, rotor core, bearing, slip ring. , A resolver rotor and a rotor core can also be arranged in this order.

The second embodiment will be described below with reference to FIG.
FIG. 5 is an example in which the bearing 8, the slip ring 22, the resolver 31, and the rotor core 12 are arranged in this order from the shaft end of the rotating shaft 7.
At this time, since the slip ring 22 is a component for supplying electric power to the rotor coil 13 disposed in the rotor core 12, a conductor for energizing between the slip ring 22 and the coil terminal of the rotor coil 13. At least two 35 are arranged. In FIG. 5, the current flowing through the conductor 35 is indicated by an arrow C. Although only one is visible in FIG. 5, at least two conductors are arranged. Here, as in the first embodiment, the product of the number of turns of the rotor coil and the field current flowing therethrough is a magnetomotive force source that generates magnetic flux in the axial direction of the rotating shaft 7.

  As shown in FIG. 5, when the resolver 31 is disposed on the rotational axis between the slip ring 22 and the rotor core 12, the conductor 35 is disposed along the axial direction of the rotational shaft 7 on the inner periphery of the resolver rotor 32. become.

FIG. 6 is a top view of the resolver according to the second embodiment of the present invention. In FIG. 6, the rotating shaft 7 and the conductor 35 are disposed on the inner periphery of the resolver rotor 32. Two conductors 35 are arranged in parallel to the rotation axis, and a current flows through the conductor 35 on the left side of FIG. 6 toward the back of the paper surface and toward the front of the paper surface on the right side.
Compared to the first embodiment, in the second embodiment, since the conductor penetrates the inner circumference of the resolver rotor, the direction of the primary magnetic flux generated by the magnetomotive force (primary component) and permeance (zero-order component) of the conductor Differ by 180 degrees. In the first and second embodiments, the angle reference is changed so that the angle of the magnetic flux is the same in the first and second embodiments when Np = 1, which is the order of the main spatial component of the magnetomotive force of the conductor.

  In FIG. 6, two conductors 35 are arranged on the inner periphery of the resolver rotor 32. The conductor 35 on the left side in FIG. 6 is energized in the back direction toward the paper surface, and the right conductor 35 is energized in the front direction. However, the arrangement is not limited to the form shown in FIG. 6 as long as it is a pair of conductors. Even when the center angle of the two conductors is other than 180 degrees or when two conductors are arranged, the effect of the second embodiment can be obtained in the same manner.

  Here, the resolver, the relative position of the rotating electrical machine, the winding direction of the rotor coil, and the electrical connection between the rotor coil and the conductor will be described. When looking at the cross section of the resolver and the conductor in the vicinity of the resolver, the rotor coil is viewed in the same direction as the paper surface is viewed from above. When the right-hand screw is turned so that the right-hand screw advances in the axial direction from the rotor coil to the resolver, when the current direction flows through the rotor coil in the same direction as the right-hand screw turns, the direction of the right-hand screw travels in the opposite direction. The position of any conductor through which current flows is 0 degrees, the center at that time is the center of rotation, and the direction opposite to the current of the rotor coil at that time is positive. That is, when FIG. 6 is viewed from above, the direction from the center of rotation to the conductor in which the current flows upward on the paper (right direction in FIG. 6) is 0 degree, and the counterclockwise direction is positive. The center of the flux barrier is arranged in the direction of 270 degrees.

  When the current direction of the rotor coil is reversed, the current direction of the conductor is also reversed, and the direction of the magnetic flux is also reversed, so that the arrangement of the flux barriers is the same and the same canceling effect as in the second embodiment is obtained. It is done. In addition, it is the same in all the embodiments that the magnetic flux canceling effect is the same regardless of whether the current direction is positive or negative.

  In the above, the number of phases of the winding is 3, an excitation winding for applying an AC voltage, and a sin winding and a cosine winding for outputting a voltage necessary for angle detection (sin winding and cos winding are Both are output windings). In order to enable angle detection, the phase difference between the sin winding and the cos winding is about 90 degrees.

  In the following, it is the same as above that the number of phases of the winding is 3, and the phase difference between the sin winding and the cos winding is about 90 degrees, but the spatial order of the main component of the output winding is The case where it is limited to N0 satisfying the following will be described. When the spatial order of the main component of the output winding is N0, the number of slots of the resolver stator is S, and arbitrary integers are n1 and n2, n1 satisfying at least | N0 + n1 × S | = | Np |, or | N0− N0 is determined so that there exists n2 satisfying n2 × S | = | Np |. When N0 is determined by this method, more space Np-order magnetic fluxes are interlinked with the output winding than when N0 is determined by other methods, so that the effect of reducing the angle error is further increased. large. The number of turns of each tooth is calculated by, for example, a trigonometric function so as to have the above-mentioned order, rounded off, and in some cases, the number of turns within about several turns may be adjusted for each tooth so that the angle error is reduced. Good.

Here, the angle error will be described. The angle error is a value obtained by subtracting the true angle from the detected angle of the resolver. The smaller the angle error, the better the angle detection accuracy. The angle error increases when a voltage component other than the voltage component necessary for angle detection is superimposed on the voltage of the sin winding and the cos winding. The voltage is determined by the flux linkage. In the case of Np = 1 in the first and second embodiments, the flux linkage is the primary component. When this becomes a voltage, it is a voltage primary component. However, all of the voltage orders are time orders in which one pulsation is the primary in one rotation of the rotor. As described above, the primary voltage is generated in each of the sin winding and the cos winding due to the generation of the linkage flux primary. Since the interlinkage magnetic flux required for angle detection is the same quaternary component as that of the shaft angle multiplier, the voltage generated by this interlinkage magnetic flux is fourth order. Therefore, since the amplitude and phase of the first-order permeance change depending on the range and center angle of the flux barrier, the voltage-first-order canceling ratio is changed and the angle error is also different.
FIG. 19 shows the dependence of the angle error of the resolver in the second embodiment of the present invention on the flux barrier range (electrical angle) (flux barrier center: −90 degrees). 19 to 21, the angle reference is the same as that in FIG. 6. In FIG. 6, the center of the flux barrier is −90 degrees. (Flux barrier radial direction length) / (minimum gap length) was set to 1.75. Here, 100% of the angle error is an angle error before the flux barrier is disposed (same as the flux barrier range of 0 degree). When the horizontal axis is 90 degrees or more and 260 degrees or less, the angle error is 40% or less. The angle error is 10% or less when the horizontal axis is 110 degrees or more and 245 degrees or less.
FIG. 20 shows the dependence of the angle error of the resolver in the second embodiment of the present invention on (the radial length of the flux barrier) / (minimum gap length). The center of the flux barrier is -90 degrees. The range of the flux barrier is 180 degrees. Here, 100% of the angle error is defined as the angle error before the flux barrier is disposed (same as when (the radial direction length of the flux barrier) / (the minimum gap length) is 0). The angle errors when the flux barrier is arranged are all 10% or less. Therefore, the angle error can be reduced by arranging the flux barrier without depending on (the radial length of the flux barrier) / (minimum gap length).
FIG. 21 shows the dependence of the angle error of the resolver in the second embodiment of the present invention on the flux barrier range (electrical angle) (flux barrier center: 90 degrees). (Flux barrier radial direction length) / (minimum gap length) was set to 1.75. The angle error is 30% or less when the horizontal axis is 150 degrees or more and 200 degrees or less.

  7 to 9 show an example of the number pattern of the excitation winding and the number pattern of the output winding (sin winding and cos winding) when N0 is determined by the above method. The teeth number is a number in which an arbitrary tooth is numbered 1 and assigned one by one from the number 1 to the number of slots in order in either the positive or negative direction in the circumferential direction. Regarding the number of turns, one represents right-handed and the other represents left-handed.

  Even when the flux barrier is divided into two or more layers in the radial direction (the case of two layers is shown in FIG. 11 as Example 3), the total radial length is the same as the radial length of one layer. As with certain cases, there is an effect of reducing the angle error.

  In the above, the magnetomotive force of the conductor is primary, and the permeance by the flux barrier is primary. However, these orders are not limited to the first order. FIG. 10 shows a case where the magnetomotive force of the conductor is secondary and the permeance by the flux barrier is secondary (Example 2). In the case of Np = 2, the same effect as in the case of Np = 1 can be obtained. When the magnetomotive force of the conductor is Np order, if the permeance by the flux barrier is Np order, it is possible to prevent the angle detection accuracy from being lowered as described above. However, when the shaft angle multiplier is the same as Np, the angle detection accuracy decreases, so Np is set to a value different from the shaft angle multiplier.

  In FIG. 6, the space Np-order permeance is axisymmetric with respect to a straight line connecting the circumferential position where the space Np-order magnetic flux generated by the conductor penetrating the inside of the resolver rotor is maximum and the rotation center. If the Np-order permeance has the same amplitude, the effect of canceling the magnetic flux of the space Np-order generated by the conductor and the rotor coil is greater when the line-symmetry is greater than when the line-symmetry is not line-symmetric. The required Np-order permeance amplitude is small. Thereby, since the radial direction length of a flux barrier can be made small, intensity | strength can be improved. In addition, when the outer periphery of the rotor is changed so as to superimpose the Np-order component on the gap length or permeance as shown in Embodiment 3 instead of the flux barrier, the minimum gap length is more in the case of line symmetry than in the case of non-line symmetry. Since the tolerance can be increased, the manufacturing tolerance can be increased and the cost can be reduced. If it is line symmetrical, this effect is not limited to FIG.

  In the second embodiment, the following functions and effects can be obtained in addition to the functions and effects corresponding to those of the first embodiment. In the second embodiment, when the spatial order of the main component of the output winding is N0, the number of slots of the resolver stator is S, and arbitrary integers are n1 and n2, at least | N0 + n1 × S | = | Np | N0 is determined such that n1 that satisfies or n2 that satisfies | N0−n2 × S | = | Np | exists. When N0 is determined by this method, more space Np-order magnetic fluxes are interlinked with the output winding than when N0 is determined by other methods, so that the effect of reducing the angle error is further increased. large.

  Further, when the reciprocal of the gap length is used as a function of the circumferential position, if the component of the function is configured to include only the space Nth order, the space 0th order, and the space Npth order that are the same as the axial multiple angle, When the gap length is added, permeances other than the Np order are not superimposed, and the angle detection accuracy does not decrease.

Embodiment 3 FIG.
Next, a third embodiment of the present invention will be described. The third embodiment is the same as the first or second embodiment described above except for the contents described below. FIG. 12 is a top view of the resolver in the first example of the third embodiment. FIG. 13 is a waveform of the resolver gap length in Example 1 of the third embodiment. FIG. 14 shows the dependency of the resolver angle error (the number of turns of the rotor coil × the Np-order amplitude of the gap length / the minimum gap length before adding the Np-order) in Example 1 of the third embodiment. FIG. 15 shows the dependency of the angle error of the resolver in Example 1 of the third embodiment on the direction in which the Np-th order gap length is minimized. FIG. 16 is a top view of the resolver in the second example of the third embodiment. FIG. 17 is a waveform of the resolver gap length in Example 2 of the third embodiment. FIG. 18 is a top view of the resolver in the third example of the third embodiment.

  In the above description, the flux barrier is used as the spatial Np-order permeance. However, in this embodiment, the space Np-order permeance is changed by changing the outer periphery of the rotor (gap length or the inverse of the gap length) instead of the flux barrier. A configuration having the above will be described.

  In the conventional outer periphery of the rotor, the spatial order of the reciprocal of the gap length includes the same order as the shaft angle multiplier and the 0th order (DC component) order.

  FIG. 12 (Example 1) shows a case where Np = 1. An Np order gap length is added to the outer periphery of the conventional rotor. The position of the center of rotation and the center of the conductor is the same as in FIG. FIG. 13 shows a waveform of the gap length after the addition. That is, when frequency analysis is performed on the reciprocal of the gap length to which the Np-th order gap length is added, the main spatial orders are the same component, the zeroth-order component, and the Np-th order component as the shaft angle multiplier. Even if a flux barrier is not disposed, since there is a permeance primary due to the outer periphery of the rotor, a decrease in angle detection accuracy can be prevented as in the second embodiment.

  FIG. 14 shows the dependence of the angle error on (number of turns of the rotor coil) × (Np-order amplitude of the gap length) / (minimum gap length when there is no Np-order in the gap length). Here, 100% of the angle error is the angle error before adding the Np-order component to the conventional gap length. The standard of 100% is the same in the following graph. When the horizontal axis is 0 or more and 85 or less, the angle error is 100% or less. The angle error is 50% or less when the horizontal axis is 30 or more and 75 or less.

  The angle error causes the torque ripple of the motor and the difference between the command value of the average torque and the actual value. When the horizontal axis is 0 or more and 85 or less, the torque ripple can be reduced as compared with the conventional case, and the average torque can be brought close to the command value. Further, when the horizontal axis is 30 or more and 75 or less, the angle error is 50% or less, so that the torque ripple can be reduced to less than half of the conventional value, and the average torque can be made much closer to the command value than the conventional value. . If the conductor does not penetrate the resolver, the same effect as described above can be obtained only by changing the angle reference. Here, the reason why the number of turns of the rotor coil is included on the horizontal axis will be described. In the present embodiment, the magnetic flux generated by the space Np order magnetomotive force and the permeance space 0th order of the conductor is canceled out by the magnetic flux generated by the space coil 0th order magnetomotive force and the permeance space Np order. The former does not depend on the number of turns of the rotor coil, whereas the latter depends on the number of turns of the rotor coil. Therefore, the number of turns of the rotor coil is included on the horizontal axis.

  FIG. 15 shows the dependence of the angle error on the direction in which the Np-th order gap length is minimized. If this direction is an electrical angle (space Np first period is 360 degrees) and is not less than 10 degrees and not more than 170 degrees, the angle error is 100% or less. Further, when this direction is 70 degrees or more and 110 degrees or less (preferably 80 degrees or more and 100 degrees or less, and more preferably 90 degrees), the angle error is near the minimum, and therefore the change in the angle error with respect to this direction is small. It is more preferable. Therefore, even if this direction varies depending on the product due to manufacturing variations, the change in the angle error is small, so that the variation between the torque ripple and the command value of the average torque and the actual value is small for each product. That is, the quality is stabilized.

  When an angle such as a direction is represented by an electrical angle, a value obtained by dividing the circumference of the rotor by Np is 360 degrees. In the case of Np = 1, this angle is the same as the mechanical angle. 0 degrees in FIG. 12 is the right direction from the center of rotation. The method for determining the center is the same as in the second embodiment. If the conductor does not penetrate the resolver, the same effect as described above can be obtained only by changing the angle reference.

  As described in the second embodiment, Np is not limited to 1, and any number other than a shaft multiple angle can be used. An example of Np = 2 (Example 2) is shown in FIG. FIG. 17 shows a waveform of the gap length. Of course, the same effect can be obtained even when Np = 2.

  In the above, the outer periphery of the conventional rotor is changed so that the space Np-order component is superimposed on the gap length or the reciprocal of the gap length. The method of superimposing the space Np-order component on the gap length or the reciprocal of the gap length is not limited to this. In the case of Np = 1, the same effect as described above can be obtained even if the rotor whose conductor or rotor outer periphery is symmetrical from the center of the shape is dynamically eccentric. However, when the dynamic eccentricity is performed, the center of the line segment connecting the opposing conductors is different from the center of rotation, and therefore the effect of reducing the angle error is greater when the outer periphery of the conventional rotor is changed.

  As described above, when the flux barrier is not arranged, a pressing step for arranging the flux barrier becomes unnecessary. Since the same pressing process as that of a resolver that does not include a conventional Np-order component in the gap length or permeance is sufficient, it can be produced in the same time.

  In the above description, an example in which a space Np-th order gap length is added to the outer periphery of a conventional rotor has been described. However, the same effect as described above can be obtained even in a configuration in which a space Np-order permeance is added to the permeance instead of the gap length. It is done. FIG. 18 shows a resolver (Example 3) when Np = 1 and an Np-order component is superimposed on the permeance. When permeance to which Np-order permeance (proportional to the reciprocal of the gap length) is added is subjected to frequency analysis, the spatial order is only the N-order component, the 0th-order component, and the Np-order component that are the same as the shaft angle multiplier. On the other hand, when an Np-th order gap length is added, the permeance is proportional to the reciprocal of the gap length, so that not only the Np-order component but also permeances other than the Np-order are superimposed on the N-order component and the zero-order component. Depending on the conditions of the external magnetic field, the magnetic flux becomes other than the Np order, which may slightly reduce the angle detection accuracy. However, when only the Np-th order permeance is added to the permeance, the permeances other than the Np-th order do not overlap, and the angle detection accuracy does not deteriorate.

  In the third embodiment, the following operational effects can be obtained in addition to the above-described corresponding operational effects of the first and second embodiments. In the third embodiment, when the gap length is a function of the circumferential position, the component of the function includes the same spatial order as the axial multiple angle and the spatial Np order. According to this, since a flux barrier is unnecessary, the press process for arrange | positioning a flux barrier becomes unnecessary. Since the same pressing process as that of a resolver that does not include a conventional Np-order component in the gap length or permeance is sufficient, it can be produced in the same time.

  Further, in the above, when D = (the number of turns of the rotor coil) × (Np-order amplitude of gap length) / (minimum gap length before adding Np-order), when D is 0 or more and 85 or less, the gap length If the conventional angle error without adding the Np-th order component is 100%, the angle error is 100% or less. That is, torque ripple can be reduced as compared with the conventional technique, and the average torque can be made closer to the command value.

  In the above, when the rotor coil and the conductor are electrically connected, the conductor does not penetrate the resolver, and the right screw is turned so that the right screw advances in the axial direction from the rotor coil to the resolver. When the current direction flows through the rotor coil in the same direction as the right-hand screw turns, the position of any conductor through which the current flows in the right-hand screw traveling direction is 0 degree, and the center at that time is the rotation center, and the rotor at that time When the direction opposite to the coil current is positive, the direction in which the Np-th order gap length is minimum is 10 degrees or more and 170 degrees or less as a value obtained by dividing the mechanical angle by Np. Assuming that the conventional angle error without adding the next component is 100%, the angle error is 100% or less, the torque ripple can be reduced as compared with the conventional method, and the average torque can be made closer to the command value.

  Alternatively, in the above, when the rotor coil and the conductor are electrically connected, the conductor passes through the inside of the resolver, and the right screw is turned so that the right screw advances in the axial direction from the rotor coil to the resolver, the right side When the current direction flows through the rotor coil in the same direction as the screw turns, the position of any conductor through which the current flows in the direction opposite to the direction of the right-handed screw is 0 degrees, and the center is the center of rotation. When the direction opposite to the rotor coil current is positive, the direction in which the Np-th order gap length is minimum is 10 degrees or more and 170 degrees or less as a value obtained by dividing the mechanical angle by Np. If the conventional angle error without adding the Np-order component to 100% is 100%, the angle error becomes 100% or less, the torque ripple can be reduced as compared with the conventional case, and the average torque can be made closer to the command value.

  In addition, the distribution of permeance of the space Np order is a line with respect to the straight line connecting the circumferential position where the space Np order magnetic flux generated by the conductor generating the magnetomotive force of the order of space Np (Np ≠ N) is maximum and the rotation center. If it is symmetric, the minimum gap length can be made larger in the case of line symmetry than in the case of non-line symmetry when changing the outer periphery of the rotor so that the Np-order component is superimposed on the gap length or permeance instead of the flux barrier. , Manufacturing tolerances can be increased and costs can be reduced.

  Although the contents of the present invention have been specifically described with reference to the preferred embodiments, various modifications can be made by those skilled in the art based on the basic technical idea and teachings of the present invention. It is self-explanatory.

  DESCRIPTION OF SYMBOLS 1 Rotating electrical machine for vehicles, 2 Housing, 3, 4 Bracket, 5 Screw, 7 Rotating shaft, 8, 9 Bearing, 10 Pulley, 11 Nut, 12 Rotor core, 13 Rotor coil, 16, 17 Core member, 16a Cylindrical part, 16b Claw-shaped magnetic pole part, 17a Cylinder-shaped part, 17b Claw-shaped magnetic pole part, 18, 19 Cooling fan, 22 Slip ring, 24 Stator core, 25 Stator coil, 26 Brush, 31 Resolver, 32 Resolver rotor, 32a Mountain part (mountain shape Convex part), 32b groove part (mountain-shaped concave part), 32c bridge part, 32d communication path, 34 resolver stator, 34a resolver coil, 35 conductor, 101 flux barrier (gap), 102 attachment hole (rotary shaft attachment hole).

Claims (12)

A resolver rotor fixed to a rotating shaft to which a conductor that generates the next magnetomotive force in space Np (Np ≠ N, N: the number of salient poles of the resolver rotor) by energization is fixed;
In a resolver in which a zero-order magnetomotive force is generated on the surface of the resolver rotor by a rotor coil connected to the conductor,
The resolver rotor has a spatial Np-order permeance arranged in a phase for reducing an Np-order magnetic flux generated by a current flowing through the conductor.
Resolver. Spatial Np-order permeance distribution is axisymmetric with respect to a straight line connecting a circumferential position where the space Np-order magnetic flux generated by the conductor generating the magnetomotive force of space Np (Np ≠ N) order is maximum and the rotation center. Is,
The resolver of claim 1. When the spatial order of the main component of the output winding is N0, the number of slots of the resolver stator is S, and arbitrary integers are n1 and n2,
N0 is determined such that at least n1 satisfying | N0 + n1 × S | = | Np | or n2 satisfying | N0−n2 × S | = | Np | exists.
The resolver according to claim 1 or 2. A flux barrier is arranged as Np-order permeance in the core of the rotor.
The resolver according to any one of claims 1 to 3. When the gap length is a function of the circumferential position, the component of the function includes the same spatial order as the axial double angle and the spatial Np order.
The resolver according to any one of claims 1 to 3. When D = (number of rotor coil turns) × (Np-order amplitude of gap length) / (minimum gap length before adding Np-order), D is 0 or more and 85 or less.
The resolver according to claim 5. The rotor coil and conductor are electrically connected,
The conductor does not penetrate the resolver,
When the right-hand screw is turned so that the right-hand screw advances in the axial direction from the rotor coil to the resolver, if the current direction flows through the rotor coil in the same direction as the right-hand screw turns, current flows in the traveling direction of the right-hand screw. When the position of any flowing conductor is 0 degree, the center at that time is the center of rotation, and the direction opposite to the current of the rotor coil at that time is positive,
The direction in which the Np-th order gap length is minimum is 10 degrees or more and 170 degrees or less as a value obtained by dividing the mechanical angle by Np.
The resolver according to claim 5. The rotor coil and conductor are electrically connected,
The conductor penetrates the inside of the resolver,
When the right-hand screw is turned so that the right-hand screw advances in the axial direction from the rotor coil to the resolver, when the current direction flows through the rotor coil in the same direction as the right-hand screw turns, the direction of the right-hand screw travels in the opposite direction. When the position of any conductor through which current flows is 0 degrees, the center at that time is the center of rotation, and the direction opposite to the current of the rotor coil at that time is positive,
The direction in which the Np-th order gap length is minimum is 10 degrees or more and 170 degrees or less as a value obtained by dividing the mechanical angle by Np.
The resolver according to claim 5. When the reciprocal of the gap length is a function of the circumferential position, the component of the function includes only the space Nth order, the space 0th order, and the space Npth order that are the same as the axial multiple angle.
The resolver according to any one of claims 1 to 3. The rotor coil and conductor are electrically connected,
The conductor does not penetrate the resolver,
When the right-hand screw is turned so that the right-hand screw advances in the axial direction from the rotor coil to the resolver, if the current direction flows through the rotor coil in the same direction as the right-hand screw turns, current flows in the traveling direction of the right-hand screw. When the position of any flowing conductor is 0 degree, the center at that time is the center of rotation, and the direction opposite to the current of the rotor coil at that time is positive,
The center of the flux barrier is −90 degrees, and the range of the flux barrier is 90 degrees or more and 260 degrees or less.
The resolver according to any one of claims 1 to 4. The rotor coil and conductor are electrically connected,
The conductor penetrates the inside of the resolver,
When the right-hand screw is turned so that the right-hand screw advances in the axial direction from the rotor coil to the resolver, when the current direction flows through the rotor coil in the same direction as the right-hand screw turns, the direction of the right-hand screw travels in the opposite direction. When the position of any conductor through which current flows is 0 degrees, the center at that time is the center of rotation, and the direction opposite to the current of the rotor coil at that time is positive,
The center of the flux barrier is −90 degrees, and the range of the flux barrier is 90 degrees or more and 260 degrees or less.
The resolver according to any one of claims 1 to 4. The resolver according to any one of claims 1 to 11,
A rotor having the conductor and the rotor coil, and the resolver attached to the rotating shaft;
A stator disposed opposite to the rotor via a gap,
Rotating electric machine.

Images