Wheel bearing device

The dual-sensor system with phase-difference detection and filter processing in the wheel bearing device corrects high-order component errors, enhancing the accuracy of load determination on the inner member by addressing manufacturing-induced whirling issues.

JP2024157111APending Publication Date: 2024-11-07JTEKT CORP
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
JP2023071245
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Filing Date
2023-04-25
Publication Date
2024-11-07

AI Technical Summary

Technical Problem

Existing wheel bearing devices suffer from manufacturing and assembly errors that cause the inner member to eccentrically rotate, leading to whirling, which results in high-order component errors in sensor detection signals, affecting the accuracy of load determination on the inner member.

Method used

The wheel bearing device employs a dual-sensor system with sensors positioned 180 degrees apart to detect the phase difference in the sensor rotor's rotation, coupled with a processing device that performs angle detection, displacement calculation, and filter processing to correct high-order components based on angular velocity and number of teeth, allowing for accurate load determination.

Benefits of technology

This configuration enables precise determination of the load on the inner member by reducing high-order component errors, ensuring accurate load calculation regardless of dynamic changes in the applied load.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024157111000001_ABST
    Figure 2024157111000001_ABST
Patent Text Reader

Abstract

To obtain load acting on an inner member.SOLUTION: A processor 18 which a wheel bearing device 10 has can perform the following: angle detection processing which obtains a rotation angle at a first detection position of a first sensor 7A and a second detection position of a second sensor 7B with respect to a sensor rotor 17 on the basis of a waveform signal; displacement arithmetic processing obtaining a displacement quantity of the sensor rotor 17 on the basis of a difference in the rotation angle at the first detection position and the second detection position; angular velocity arithmetic processing which obtains angular velocity of the sensor rotor 17 from the rotation angle; property setting processing which sets filter property for attenuating a high-order component included in the signal in accordance with the angular velocity; high-order component correction processing which obtains a displacement quantity after correction by performing the filter processing of the signal indicating the displacement quantity by the filter property; and load arithmetic processing obtaining load acting on an inner member 11 from the corrected displacement quantity.SELECTED DRAWING: Figure 5
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present invention relates to a wheel bearing device. [Background technology]

[0002] Patent Document 1 discloses a wheel bearing device equipped with a sensor that detects the displacement of an inner member. This wheel bearing device is capable of determining the load acting on the wheel from the displacement of the inner member.

[0003] Inevitably, errors occur in the manufacture and assembly of components of a wheel bearing device. Due to these errors, the inner member whirls (rotates eccentrically) slightly regardless of the load. In the case of the sensor structure disclosed in Patent Document 1, the whirling of the inner member appears as a periodic fluctuation in the detection signal of the sensor. The periodic fluctuation becomes an unwanted component in the detection signal of the sensor. There are cases where the amount of whirling accounts for a high proportion of the target detection amount by the sensor. The invention disclosed in Patent Document 1 makes it possible to accurately determine the displacement of the inner member and the acting load by correcting the sensor output, which includes the above-mentioned whirling in the detection result. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2020-85840 A Summary of the Invention [Problem to be solved by the invention]

[0005] In the case of the invention disclosed in Patent Document 1, the amount of whirling per revolution of the wheel bearing device in an unloaded state is obtained in advance, and a correction is performed in which the amount of whirling is subtracted from the detection signal of the sensor. In the case of the sensor configuration disclosed in Patent Document 1, an error in a high-order component occurs in the detection signal of the sensor due to the number of teeth of the sensor rotor, etc., and as a result, an error occurs in the calculation result of the load acting on the inner member. Moreover, the error in the high-order component changes depending on the load acting on the inner member. For this reason, it is not possible to deal with this by a means of generating correction data in advance and subtracting it, as in the invention disclosed in Patent Document 1.

[0006] SUMMARY OF THE PRESENT DISCLOSURE An object of the present invention is to provide a wheel bearing device that makes it possible to more accurately determine the load acting on the inner member. [Means for solving the problem]

[0007] (1) The wheel bearing device of the present invention comprises an outer member fixed to a vehicle body, an inner member rotating around the central axis of the outer member, a rolling element arranged to be free to roll between the outer member and the inner member, a sensor rotor rotating integrally with the inner member, a sensor device having first and second sensors arranged on the outer member and detecting the sensor rotor, and a processing device which processes signals output from the sensor device. When the inner member is displaced relative to the outer member, the first sensor outputs a periodic waveform signal with a delayed phase, and the second sensor outputs a periodic waveform signal with an advanced phase.

[0008] The processing device is capable of executing the following operations: angle detection processing to determine the rotational angles of the sensor rotor at each of a first detection position of the first sensor and a second detection position of the second sensor based on the waveform signal; displacement calculation processing to determine the displacement amount of the sensor rotor based on the difference between the rotational angles at the first detection position and the second detection position; angular velocity calculation processing to determine the angular velocity of the sensor rotor from the rotation angle; characteristic setting processing to set filter characteristics for attenuating high-order components contained in a signal according to the angular velocity; high-order component correction processing to perform filter processing on a signal indicating the displacement amount using the filter characteristics to determine the corrected displacement amount; and load calculation processing to determine the load acting on the inner member from the corrected displacement amount.

[0009] When a load acts on the wheel support bearing device having the above-mentioned configuration, for example, a signal with a distorted waveform while changing in amplitude is acquired as the waveform signal. In the case of an actual wheel support bearing device, the load acting changes dynamically, and this changes as a high-order component error and appears in the waveform signal, affecting the calculation for determining the load acting on the inner member. Therefore, the wheel bearing device having the above-mentioned configuration performs filtering to attenuate the high-order components, and furthermore, sets the filter characteristics according to the angular velocity of the sensor rotor. Therefore, even if the load changes dynamically, it is possible to reduce the error of the high-order components corresponding to the change. As a result, it is possible to more accurately determine the load acting on the inner member.

[0010] (2) Preferably, the device further includes a memory device that stores first-order component correction data indicating an amount of whirling of the inner member within one rotation of the inner member, and the processing device is capable of executing a first-order component correction process to subtract the amount of whirling from the amount of displacement obtained by the displacement calculation process and obtain an amount of displacement after first-order component correction. In the load calculation process, when the rotational speed of the inner member is less than a predetermined value, the load acting on the inner member is determined from the displacement amount after the first-order component correction, and when the rotational speed of the inner member is equal to or greater than a predetermined value, the load acting on the inner member is determined from the displacement amount after higher-order component correction obtained by performing the filter process on a signal indicating the displacement amount after the first-order component correction.

[0011] According to the above configuration, when the rotational speed of the inner member is low, the angular velocity is small, and it is possible to relatively accurately determine the load acting on the inner member without reducing the error in the high-order components. On the other hand, when the rotational speed of the inner member is high, the angular velocity is large, and it is possible to more accurately determine the load acting on the inner member by reducing the error in the high-order components through the filtering process.

[0012] (3) Preferably, the sensor rotor is an annular member having a gear shape with a plurality of teeth provided on an outer circumferential surface thereof at equal intervals in the circumferential direction. In the characteristic setting process, a high-order frequency band to be attenuated is set as the filter characteristic in accordance with the angular velocity and the number of the plurality of teeth. The number of teeth of the sensor rotor (annular member) affects the error of the high-order components as described above. Therefore, according to the above configuration, the high-order frequency band to be attenuated is set according to the angular velocity and the number of teeth of the sensor rotor, so that the error of the high-order components can be more effectively reduced. Effect of the Invention

[0013] According to the wheel bearing device of the present invention, it is possible to more accurately determine the load acting on the inner member. [Brief description of the drawings]

[0014] [Figure 1] 1 is a cross-sectional view showing an example of a wheel bearing device of the present invention. [Diagram 2] 1 is a schematic view of a wheel bearing device as viewed from the inner side of a vehicle. [Diagram 3]4 is an explanatory diagram of a waveform signal output from a sensor. FIG. [Figure 4] FIG. 4 is an explanatory diagram of an angle signal obtained by converting a waveform signal. [Diagram 5] FIG. 2 is a block diagram illustrating a process executed by a processing device. [Figure 6] FIG. 11 is an explanatory diagram showing an example of first-order component correction data. [Figure 7] FIG. 11 is a waveform diagram of a signal indicating the amount of displacement of the inner member before correction. [Figure 8] FIG. 11 is a waveform diagram of a signal indicating the amount of displacement of an inner member after first-order component correction. [Figure 9] FIG. 11 is a waveform diagram of a signal indicating the amount of displacement of the inner member before correction. [Figure 10] FIG. 11 is a waveform diagram of a signal indicating the amount of displacement of an inner member after first-order component correction. [Figure 11] FIG. 11 is a waveform diagram showing a portion of the signal shown in FIG. 10 with the range of the vertical axis being detailed. [Figure 12] FIG. 11 is a waveform diagram of a signal indicating the displacement amount of the inner member after high-order component correction. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] [Configuration of the wheel bearing device 10] Fig. 1 is a cross-sectional view showing an example of a wheel bearing device of the present invention. The wheel bearing device 10 is a bearing device for a wheel used in an automobile, and is also called a hub unit. The wheel bearing device 10 includes an inner member (hub axle) 11, a cylindrical outer member (outer ring) 12 provided radially outward from a portion of the inner member 11, and balls 13 which are rolling elements.

[0016] In the wheel bearing device 10, the direction along the central axis C of the outer member 12 is defined as the "axial direction." The "axial direction" also includes a direction parallel to the central axis C. The direction perpendicular to the central axis C is the "radial direction." In this embodiment, the inner member 11 rotates around the central axis C. The direction along a circle centered on the central axis C is the "circumferential direction." The direction along the central axis C is the Y-axis direction. The horizontal direction perpendicular to the central axis C is the X-axis direction. The vertical direction perpendicular to the central axis C is the Z-axis direction. In each figure, an orthogonal coordinate system is defined by the X-axis, Y-axis, and Z-axis.

[0017] With the wheel bearing device 10 fixed to the body of an automobile, the front-rear direction of the automobile coincides with the direction along the X-axis, the left-right direction coincides with the direction along the Y-axis, and the up-down direction coincides with the direction along the Z-axis. In Fig. 1, one side in the axial direction, which is the right side, is the vehicle outer side, and the other side in the axial direction, which is the left side, is the vehicle inner side.

[0018] The outer member 12 has a cylindrical outer ring main body 21 and a fixing flange 22 extending radially outward from the outer ring main body 21. The outer member 12 is attached to a knuckle (not shown) that is a vehicle body member by the flange 22, whereby the wheel bearing device 10 including the outer member 12 is fixed to the vehicle. In this embodiment, the outer member 12 serves as a fixed ring, and the inner member 11 serves as a rotating ring.

[0019] The inner member 11 has an inner shaft 23 and an inner ring 24 attached to the vehicle inner end of the inner shaft 23. The inner shaft 23 has an inner shaft main body portion 26 provided radially inward of the outer member 12, and a flange portion 27 provided on the vehicle outer side of the inner shaft main body portion 26. Although not shown, a wheel and a brake rotor are attached to the vehicle outer side surface of the flange portion 27.

[0020] The inner ring 24 is an annular member, and is attached to the end of the inner shaft body 26 on the vehicle inner side. An inner raceway surface with which the balls 13 roll and make contact is provided on the outer peripheral side of each of the inner shaft body 26 and the inner ring 24. An outer raceway surface with which the balls 13 roll and make contact is provided on the inner peripheral side of the outer ring body 21. The balls 13 are provided rollably between the outer raceway surface of the outer member 12 and the inner raceway surface of the inner member 11. The multiple balls 13 included in each row are held by a cage (not shown).

[0021] The wheel bearing device 10 further includes a sensor device 6, a processing device 18 that processes signals output from the sensor device 6, and a storage device 19. The storage device 19 may be configured by a memory or the like included in the processing device 18, but may also be a memory or the like provided separately from the processing device 18.

[0022] The sensor device 6 has a sensor rotor 17 that rotates integrally with the inner member 11, and a plurality of sensors 7 provided on the outer member 12. The sensor rotor 17 is attached to the vehicle inner side of the inner member 11 around the central axis C. In the embodiment shown in FIG. 1, the sensor rotor 17 is fitted and fixed to a part of the inner ring 24. The sensor rotor 17 is an annular member, and has a gear shape with a plurality of teeth provided on its outer circumferential surface at equal intervals in the circumferential direction (see FIG. 2). FIG. 2 is a schematic diagram of the wheel bearing device 10 as viewed from the vehicle inner side. In this embodiment, the sensor rotor 17 is made of a ferromagnetic material.

[0023] The sensors 7 are provided around the sensor rotor 17, and detect the sensor rotor 17 (the teeth which are a part of it). In the present embodiment, four sensors 7 are provided at equal intervals along the circumferential direction. The suffixes (A, B, C, D) attached to the symbols of the sensors 7 will be explained. The suffixes A, B, C, D indicate the arrangement with respect to the central axis C, and in FIG. 2, they indicate the front, rear, top, and bottom with respect to the central axis C.

[0024] For example, the sensor located in the front is designated by the symbol "7A" and the sensor located in the back is designated by the symbol "7B". The configuration of each sensor 7 is the same regardless of whether it is in the front, back, top, or bottom. In common explanations regarding the sensors 7, the suffixes (A, B, C, D) are omitted. The first sensor 7A and the second sensor 7B are provided 180 degrees apart around the central axis C. These front and rear sensors 7A and 7B are considered to be one sensor group. The third sensor 7C and the fourth sensor 7D are provided 180 degrees apart around the central axis C. These upper and lower sensors 7C and 7D are considered to be another sensor group. The sensor 7A and the sensor 7C are provided 90 degrees apart around the central axis C.

[0025] The sensor 7 is a magnetic sensor and functions as a rotation sensor (displacement sensor) using a magnetoresistance element. The sensor 7 has an A-phase magnetoresistance element 31 and a B-phase magnetoresistance element 32. The A-phase magnetoresistance element 31 and the B-phase magnetoresistance element 32 each detect a change in a magnetic field accompanying the movement of the sensor rotor 17. The A-phase magnetoresistance element 31 and the B-phase magnetoresistance element 32 are arranged close to each other in the circumferential direction but at different positions. Therefore, the A-phase magnetoresistance element 31 and the B-phase magnetoresistance element 32 output periodic waveform signals of two phases (A-phase and B-phase) having a phase difference.

[0026] The sensor 7 detects the outer periphery of the sensor rotor 17, which is provided with teeth. The sensor rotor 17 rotates about the central axis C, causing a change in the magnetic field in the sensor 7. Based on this change in the magnetic field, the sensor 7 outputs, as detection signals, waveform signals of phase A and phase B that are out of phase with each other.

[0027] The output signal of the sensor 7 will be further explained. As described above, the sensor rotor 17 has a gear shape. Therefore, the sensor 7 repeatedly outputs a waveform signal with one period extending from one tooth of the sensor rotor 17 (through a recess) to another tooth adjacent thereto in the circumferential direction. The number of teeth on the sensor rotor 17 is "N." While the inner member 11, which is integral with the sensor rotor 17, makes one rotation (360 degrees), each sensor 7 outputs one period of the waveform signal N times. Each sensor 7 outputs a periodic waveform signal for multiple periods while the inner member 11 makes one rotation together with the sensor rotor 17.

[0028] As described above, each sensor 7 outputs a periodic waveform signal of two phases (A phase and B phase) having a phase difference. Specifically, as the sensor rotor 17 moves, a SIN signal is output from the A phase magnetic resistance element 31 and a COS signal is output from the B phase magnetic resistance element 32. When the inner member 11 rotates once (360 degrees), each sensor 7 outputs one period of the SIN signal N times and also outputs one period of the COS signal N times. As will be described later, the rotation angle, angular velocity, displacement amount, etc. of the inner member 11 are determined based on these SIN and COS signals. The process for determining these parameters is performed by the processing device 18 (see FIG. 1).

[0029] The processing device 18 may be provided outside the wheel bearing device 10. In this embodiment, the processing device 18 is configured by an ECU (Electronic Control Unit) mounted on the automobile. The processing device 18 acquires two-phase waveform signals from each of the four sensors 7 and executes various processes.

[0030] [Sensor signal processing] A description will be given of the waveform signal output by the sensor 7 and various processes executed by the processing device 18. Fig. 5 is a block diagram for explaining the processes executed by the processing device 18.

[0031] Processing device 18 has a processor that performs various types of calculation processing, a memory in which a computer program is stored, etc. Through processing executed by the processor in accordance with the computer program, processing device 18 has a phase difference / displacement detection unit 41, a precision angle calculation unit 42, a first correction unit 43, a precision angular velocity calculation unit 44, a second correction unit 45, and a calculation unit 46 as the respective functional units shown in Fig. 5. The angle detection units shown in Fig. 5 correspond to the four sensors 7A, 7B, 7C, and 7D (see Fig. 2) on the top, bottom, left, and right.

[0032] The phase difference / displacement detection unit 41 executes displacement calculation processing to determine the displacement of the inner member 11. The precise angle calculation unit 42 executes angle detection processing to determine the rotational angle of the sensor rotor 17. The first correction unit 43 executes primary component correction processing. The precise angular velocity calculation unit 44 executes angular velocity calculation processing to determine the angular velocity of the sensor rotor 17. The second correction unit 45 executes characteristic setting processing to set filter characteristics, and higher order component correction processing to correct signals. The calculation unit 46 executes load calculation processing acting on the inner member 11. Specific examples of the processing executed by each of the functional units of the processing device 18 will be described later.

[0033] In the wheel bearing device 10, the inner member 11 whirls slightly relative to the outer member 12 due to unavoidable manufacturing errors. The whirling of the inner member 11 appears as a periodic fluctuation in the detection signal of the sensor 7. This periodic fluctuation is an unwanted component in the waveform signal output from the sensor 7. This unwanted component is also called a "rotation fluctuation component." In other words, the waveform signal acquired from the sensor device 6 includes a signal of the rotation fluctuation component. When a load acts on the inner member 11, if the processing device 18 determines the load by using the waveform signal as is without correcting it, an accurate value cannot be obtained. Therefore, as will be described later, a first-order component correction process is executed to reduce periodic fluctuations in the detection signal of the sensor 7 caused by the whirling of the inner member 11.

[0034] In the case of the sensor device 6 shown in FIG. 2, when a load acts on the inner member 11, the gap between the sensor 7 and the sensor rotor 17 changes in the direction in which the load acts. For example, when a downward load acts on the inner member 11, the gap between the sensor 7C and the sensor rotor 17 and the gap between the sensor 7D and the sensor rotor 17 change. Then, the sensor signal such as a sine wave output by the sensor 7 (7C, 7D) is accompanied by a change in offset, amplitude, or waveform distortion. In particular, when the gap changes, the amplitude and waveform distortion fluctuate. That is, the waveform is distorted while the amplitude changes, and a dynamic deviation occurs from the ideal waveform. The dynamic displacement of the inner member 11 from the initial position becomes the errors of the first harmonic, the second harmonic, and the third harmonic per electrical angle after the angle calculation, and is superimposed on the sensor signal. Therefore, as will be described later, a high-order component correction process is executed to attenuate the high-order component signals.

[0035] [Sensor 7 output signal] A case will be described where a downward load along the Z axis acts on the inner member 11 (see Figs. 1 and 2) while the inner member 11 is rotating. In this case, the inner member 11 and the sensor rotor 17 are slightly displaced downward along the Z axis relative to the outer member 12. This displacement can be considered to be equivalent to a rotational movement of the outer circumferential surface of the sensor rotor 17.

[0036] In Fig. 2, the rotation direction of the sensor rotor 17 is the direction of arrow R. On the sensor 7A side, the direction in which the sensor rotor 17 is displaced by the load is opposite to the rotation of the inner member 11 (to the tangent direction of the rotation trajectory). For this reason, compared to when no load is applied, the sensor 7A outputs waveform signals with a delayed phase for both phase A and phase B, as shown in the explanatory diagram at the top of Fig. 3. In other words, the sensor 7A detects the displacement of the sensor rotor 17 as a delay angle. In Fig. 3, the signal from the sensor 7 before the load is applied is shown by a solid line, and the signal from the sensor 7 after the load is applied is shown by a dashed line.

[0037] 2, on the sensor 7B side, the direction in which the inner member 11 and the sensor rotor 17 are displaced by the load is the forward direction (same direction) with respect to the rotation of the inner member 11 (with respect to the tangent direction of the rotation trajectory). Therefore, compared to when no load is applied, the sensor 7B outputs waveform signals with phases leading for both A and B, as shown in the explanatory diagram at the bottom of Fig. 3. In other words, the sensor 7B detects the displacement of the sensor rotor 17 as a leading angle.

[0038] In this way, when the inner member 11 is displaced in one radial direction (in this embodiment, downward along the Z-axis) relative to the outer member 12, the sensor 7A outputs a waveform signal with a delayed phase for both phase A and phase B, and the sensor 7B outputs a waveform signal with an advanced phase for both phase A and phase B.

[0039] Conversely, when the load direction is upward along the Z axis, sensor 7A detects the displacement as a lead angle, and sensor 7B detects the displacement as a lag angle. As described above, the waveform signal output from sensor 7 contains, in addition to the component due to the load, a signal of an unwanted rotation fluctuation component and a signal of a higher order component.

[0040] 3, when a load acts on the inner member 11, a phase difference Q occurs in the signals output from a pair of sensors 7A and 7B that are provided 180 degrees apart about the central axis C. This phase difference Q is obtained as a time difference between waveforms.

[0041] [Angle detection process] Therefore, in order to detect the (instantaneous) phase difference Q at a specific timing, angle detection processing is performed on the waveform signals obtained from the sensors 7A and 7B. A specific example of the angle detection processing will be described below.

[0042] The waveform signals from the sensors 7A and 7B are subjected to arctangent calculation as a predetermined arithmetic process, and are converted into angle signals (electrical angle signals) having monotonicity for each period of the waveform signals (see FIG. 4). The angle signal here is a signal of electrical angle (unit [rad]). The monotonicity means that the signal increases or decreases without having a maximum or minimum in the middle. In this embodiment, a triangular angle signal having monotonicity is obtained. The upper side of FIG. 4 is a waveform diagram of the angle signal from the sensor 7A, and the lower side of FIG. 4 is a waveform diagram of the angle signal from the sensor 7B. The angle (angle signal) shown on the vertical axis of the waveform diagram in FIG. 4 is the rotation angle (electrical angle [rad]) of the sensor rotor 17 at the first detection position by the sensor 7A and the second detection position by the sensor 7B.

[0043] In this way, based on the rotation angle of the sensor rotor 17 obtained from the waveform signal from the sensor device 6, the rotation angle (electrical angle) θA of the sensor rotor 17 at the first detection position by the sensor 7A at a specific timing (specific time) t is obtained, and the rotation angle (electrical angle) θB of the sensor rotor 17 at the second detection position by the sensor 7B is obtained.

[0044] Furthermore, at the first detection position and the second detection position, the difference in rotation angle (θAB=θA-θB) between the sensor 7A side and the sensor 7B side at the same timing (same time) t is calculated. The unit of the electrical angles θA and θB here is [rad], and the difference in electrical angle (θAB [rad]) is calculated as the difference in rotation angle θAB.

[0045] [Displacement calculation processing] The displacement amount ΔZ of the inner member 11 (sensor rotor 17) is proportional to the difference θAB between the electrical angle θA on the sensor 7A side and the electrical angle θB on the sensor 7B side. Therefore, when the difference in electrical angle (θAB) is calculated, the displacement amount ΔZ of the inner member 11 caused by the load can be calculated by the following equation (1). This calculation process is a displacement calculation process, and the displacement amount of the sensor rotor 17 is found based on the difference in rotation angle found between the first detection position of sensor 7A and the second detection position of sensor 7B.

[0046] ΔZ=2×r×θAB / N [mm] ···(1)

[0047] In formula (1), "r" is the distance (design value) from the rotation center (central axis C) of the inner member 11 to the outer periphery (teeth) of the sensor rotor 17, that is, the radius [mm] of a virtual circle passing through the outer periphery of the sensor rotor 17. "N" is the number of evenly spaced teeth (detection targets) on the sensor rotor 17.

[0048] In this way, the waveform signal output from the sensor device 6 is used to determine the amount of displacement of the inner member 11 (sensor rotor 17). The sensor device 6 has a function as a displacement sensor for determining the amount of displacement of the inner member 11.

[0049] [Load calculation processing] Before describing the first-order component correction process and the high-order component correction process, the load calculation process executed by the calculation unit 46 will be described. Once the displacement ΔZ of the inner member 11 has been found, the displacement ΔZ is then converted into a load acting on the inner member 11 (load Fz acting downward along the Z axis) by equation (2). The applied load Fz is calculated by multiplying the displacement ΔZ by the bearing stiffness Kz [N / mm].

[0050] Fz = Kz × ΔZ [N] (2)

[0051] However, as described above, the inner member 11 whirls slightly relative to the outer member 12. The whirling appears as a periodic fluctuation in the detection signals of the sensors 7A and 7B. In other words, the waveform signals of the sensors 7A and 7B contain signals of unnecessary rotation fluctuation components. Furthermore, the waveform signals contain signals of unnecessary higher-order components. For this reason, it is considered that an error occurs in the calculated displacement amount ΔZ.

[0052] Therefore, in this embodiment, before calculating the load Fz, a correction process for removing unnecessary components is performed by the first correction unit 43 and the second correction unit 45. Then, the load Fz is calculated by the above formula (2) based on the displacement amount corrected by the first correction unit 43 or by both the first correction unit 43 and the second correction unit 45. In this way, the load calculation process in this embodiment is a process for determining the load Fz acting on the inner member 11 from the corrected displacement amount.

[0053] [1st-order component correction processing] The first correction unit 43 executes a primary component correction process using the primary component correction data stored in the storage device 19. The primary component correction data is generated in advance using the wheel bearing device 10 that is the detection target for the load Fz. First, the generation process of the primary component correction data will be described.

[0054] The inner member 11 is rotated with no load applied to the wheel bearing device 10 (see FIG. 1), and the waveform signals obtained by each sensor 7 are acquired by the processing device 18. When the inner member 11 whirls and rotates with a rotation fluctuation component, the distance between the sensor 7 and the sensor rotor 17 fluctuates. For this reason, the waveform signal output from the sensor 7 does not include a load component, but does include a rotation fluctuation component.

[0055] When no load is applied, the signals obtained from each sensor 7 are waveform signals as shown by solid lines in Fig. 3. The phase difference / displacement detection unit 41 executes the angle detection process using the waveform signals from each sensor 7 as input data. That is, the waveform signals from the sensors 7 are converted into angle signals having monotonicity for each period of the waveform signal as shown in Fig. 4. The angle signals here are in electrical angles [rad].

[0056] The precise angle calculation unit 42 constantly acquires the count number k of teeth (detection targets) from the starting point when the inner member 11 is rotated once. For example, if the total number N of teeth is "30" and the inner member 11 has rotated half a turn (180 degrees), the count number k becomes "15." This count number is obtained from the number of periods of the waveform signal from the sensor 7.

[0057] From the angle signal θ [rad] obtained by the angle detection processing at a certain timing, the number of teeth N, and the count number k at the timing, the rotation angle at the timing within one rotation of the inner member 11 from the starting point is calculated by the following formula (3). The rotation angle is a mechanical angle α [deg]. This calculation is performed by the precision angle calculation unit 42.

[0058] α=k×(360 degrees / N)+θ / N (3)

[0059] Using this formula (3), the rotation angle (mechanical angle (0≦α≦360) [deg]) of the inner member 11 at a predetermined timing within one rotation from the start of rotation (the starting point) can be found. The angle signal θ may be a value from one of the four sensors 7, or may be an average value of the values ​​from these sensors 7. In this embodiment, the angle signal θ is an average value of values ​​from a pair of sensors 7 (e.g., sensor 7A and sensor 7B) that are provided 180 degrees apart.

[0060] The amount of whirling G [mm] of the inner member 11 corresponding to a predetermined rotation angle within one rotation from the start of rotation (the starting point) is calculated by the following formula (4). This calculation is performed using the same sensor 7 and based on data at the same timing, and the calculated value is stored. Therefore, it is not affected by timing errors or systematic errors with the external measuring device (angle sensor). This calculation is performed by the processing device 18 (phase difference / displacement detection unit 41 or precision angle calculation unit 42).

[0061] G = G(ΔZ(θAB,α)) (4)

[0062] The rotation angle difference θAB [rad] in equation (4) is a value at the timing, and is a value found by the phase difference / displacement detection unit 41 through angle detection processing similar to that described above, based on the waveform signal acquired from the sensor device 6, when no load is applied to the wheel bearing device 10 (see FIG. 1). Because the sensor rotor 17, which rotates integrally with the inner member 11, whirls, a difference (θAB) occurs between the rotation angle (electrical angle) θA on the sensor 7A side and the rotation angle (electrical angle) θB on the sensor 7B side, even when no load is applied.

[0063] From the above, data is generated that indicates the amount of whirling G for each rotation angle (mechanical angle α [deg]) within one rotation of the inner member 11, as shown in Fig. 6. This data is data on the amount of whirling G within one rotation (rotation angle from 0 degrees to 360 degrees) of the inner member 11. In other words, the data shown in Fig. 6 becomes first-order component correction data that indicates the relationship between the rotation angle within one rotation of the inner member 11 and the amount of whirling of the inner member 11 at that rotation angle.

[0064] The first-order component correction data may be in the form of a graph as shown in Fig. 6, or may be in the form of a table in which the rotation angle (mechanical angle α) corresponds to the amount of whirling G at that rotation angle α. This first-order component correction data is stored in the storage device 19.

[0065] Next, the first-order component correction process executed by the first corrector 43 using the first-order component correction data will be described. 7 is a waveform diagram showing the result of calculating the displacement amount ΔZ of the inner member 11 (using the above-mentioned formula (1)) as described above based on the waveform signal obtained from the sensor 7 when the whirling wheel bearing device 10 is rotated once while the load is increased (monotonically increasing) on ​​the wheel bearing device 10. In other words, FIG. 7 is a waveform diagram of a signal showing the displacement amount ΔZ of the inner member 11 before correction.

[0066] As the primary component correction process, the first correction unit 43 subtracts the amount of whirling G of the primary component correction data (see FIG. 6) for each rotation angle (mechanical angle α) from the amount of displacement ΔZ before correction shown in FIG. 7. The result of this subtraction is shown in FIG. 8, and the amount of displacement ΔZ' after correction for the primary component is obtained. The amount of displacement ΔZ' shown in FIG. 8 has the rotation fluctuation component removed. In this way, the primary component correction process is a process for subtracting the amount of whirling G of the inner member 11 from the amount of displacement ΔZ of the sensor rotor 17 obtained by the displacement calculation process as described above, to obtain the amount of displacement ΔZ' after the primary component correction. The load acting on the bearing device 10 is calculated using the above-mentioned formula (2) based on the amount of displacement ΔZ′ after correction for this first-order component, thereby obtaining a load from which the whirling component has been removed.

[0067] [About each variable] Here, the variables and the like used in each of the above processes will be collectively explained. ΔZ: the displacement of the inner member 11, in millimeters. θA: the angle of the sensor rotor 17 at the position facing the sensor 7A, expressed in electrical angle (rad). θB: the angle of the sensor rotor 17 at the position facing the sensor 7B, expressed in electrical angle (rad). θAB: the angular difference (angular phase difference) between the opposing position of sensor 7A and the opposing position of sensor 7B, expressed in electrical angle (rad). The electrical angle is a virtual angle in which one period of a periodic waveform is defined as 360 degrees (2π radians). Fz: the load acting on the inner member 11, expressed in newtons [N]. Kz: A constant, which is the bearing stiffness along the Z axis, in Newtons per millimeter [N / mm].

[0068] [Higher-order component correction processing] The high-order component correction process is a process of performing a filter process on a signal indicating the obtained displacement amount of the inner member 11 to obtain a corrected displacement amount. The filter process is a process of attenuating high-order frequency bands. The "displacement amount of the inner member 11" to be corrected may be the displacement amount of the inner member 11 including a whirling component obtained by the phase difference / displacement detection unit 41 (FIG. 9), but in this embodiment, it is the displacement amount ΔZ' (FIG. 10) after first-order component correction obtained by the phase difference / displacement detection unit 41 and further corrected by the first correction unit 43.

[0069] In this embodiment, a notch filter is used as the filtering process. In this embodiment, a characteristic setting process is executed as a process for setting the filter characteristics of the notch filter. Specifically, in the characteristic setting process, a transfer function of the notch filter is set. In order to set the transfer function, an angular velocity calculation process is executed.

[0070] The transfer function of the notch filter is expressed by the following formula (5). In formula (5), ωn is the center frequency [rad / s] of the notch (high-order frequency band to be attenuated). ζ is a parameter that determines the width of the notch, and d is a parameter that determines the depth of the notch, and these can be set arbitrarily.

[0071] G(s)=(s^2+d·2ζωns+ωn^2) / (s^2+2ζωns+ωn^2) ···(5)

[0072] The angular velocity calculation process and the characteristic setting process will be described below, and an example of the high-order component correction process will be described.

[0073] [Angular velocity calculation processing] The angular velocity calculation process is a process for determining the angular velocity ω (electrical angle [rad / s]) of the sensor rotor 17 from the rotation angle θ (electrical angle [rad]) of the sensor rotor 17 determined by the same process as the angle detection process. The rotation angle θ used here may be a representative one of the multiple rotation angles θ determined from the waveform signals of the multiple (four) sensors 7, or may be an average value of the multiple rotation angles θ. The angular velocity ω is calculated by the time differentiation of the rotation angle θ (electrical angle), and in terms of digital processing, the angular velocity ω is calculated by the following equation (6).

[0074] ω=(θ[n]-θ[nm]) / Δt...Equation (6) Here, Δt=m×st, and st is the sample time or control period. [nm] represents m samples before n.

[0075] [Characteristics setting process] The characteristic setting process is a process for setting filter characteristics for attenuating high-order components contained in the waveform signal output by the sensor device 6. A high-order frequency band to be attenuated is set as the filter characteristic according to the angular velocity ω of the sensor rotor 17 obtained by the angular velocity calculation process. In other words, the transfer function of the notch filter is set according to the angular velocity ω. In this embodiment, the center frequency ωn of the high-order frequency band to be attenuated is set to a value obtained by multiplying the angular velocity ω obtained by the angular velocity calculation process by a natural number, which is the number of teeth N of the sensor rotor 17 (N×M). An example of this is as follows.

[0076] ·Angular velocity ω[rad / s]×Number of teeth N×1 ·Angular velocity ω[rad / s]×Number of teeth N×2 ·Angular velocity ω[rad / s]×Number of teeth N×3 ·Angular velocity ω[rad / s]×Number of teeth N×···

[0077] In this manner, in the case of this embodiment, the characteristic setting process is a process for setting a high-order frequency band to be attenuated as a filter characteristic in accordance with the angular velocity ω and the number N of the teeth of the sensor rotor 17.

[0078] [Example of high-order component correction processing] 9 is a waveform diagram of a signal indicating the displacement amount ΔZ of the inner member 11 obtained by the phase difference / displacement detection unit 41. The signal indicating the displacement amount ΔZ contains a first-order component signal due to the whirling of the inner member 11 as well as higher-order component signals. FIG. 10 is a waveform diagram of a signal indicating the amount of displacement ΔZ′ after the first-order component correction obtained by the phase difference / displacement detection section 41 and further corrected by the first correction section 43. As shown in FIG. Fig. 11 is a waveform diagram showing a part of the signal shown in Fig. 10, with the range of the vertical axis being detailed. In the signal showing the displacement amount ΔZ' shown in Fig. 11, the first-order component due to the whirling of the inner member 11 has been removed, but a signal of higher-order components is included.

[0079] The second correction unit 45 performs filtering on the signal indicating this displacement amount ΔZ' to attenuate higher-order frequency bands, thereby obtaining a corrected displacement amount ΔZ". FIG. 12 shows a signal indicating the displacement amount ΔZ" after correcting for higher-order components obtained by performing this filtering. From the signal indicating this corrected displacement amount ΔZ", the first-order component due to whirling of the inner member 11 has been removed, and the signal of the higher-order component has been attenuated. The vertical axis range shown in FIG. 12 is the same as that in FIG. 11.

[0080] When the filtered signal of the displacement amount ΔZ'' of the inner member 11 is obtained, the displacement amount ΔZ'' is converted into the load acting on the inner member 11 by the above-mentioned equation (2) (load calculation process by the calculation unit 46). In other words, the applied load is calculated by multiplying the displacement amount by the bearing stiffness Kz [N / mm].

[0081] In this embodiment, the load calculation process is executed by dividing the process into the following "Condition 1" and "Condition 2". Condition 1 is a case where the rotational speed of the inner member 11 is less than a predetermined value. Condition 2 is a case where the rotation speed of the inner member 11 is equal to or higher than a predetermined value. It should be noted that the "rotational speed is less than a predetermined value" in Condition 1 also includes a state in which the inner member 11 is not rotating and is at rest.

[0082] In the case of condition 1, the calculation unit 46 determines the load acting on the inner member 11 from the displacement amount ΔZ' after the first-order component correction, as the load calculation process. That is, in the case of condition 1, the high-order component correction process by the second correction unit 45 is skipped. In the case of condition 2, the calculation unit 46 performs load calculation processing by performing the filter processing on a signal indicating the displacement amount ΔZ' after the first-order component correction to obtain the displacement amount ΔZ'' after the higher-order component correction, thereby obtaining the load acting on the inner member 11 from the displacement amount ΔZ'' after the higher-order component correction.

[0083] In the case of condition 1, the angular velocity of the sensor rotor 17 is small, and it is possible to obtain the load acting on the inner member 11 relatively accurately without reducing the error of high-order components by filtering. In contrast, in the case of condition 2, the angular velocity of the sensor rotor 17 is large, and by reducing the error of higher-order components by the above-mentioned filtering process, it becomes possible to more accurately determine the load acting on the inner member 11.

[0084] [Regarding the wheel bearing device 10 of this embodiment] As described above, the wheel bearing device 10 of this embodiment has the outer member 12 fixed to the vehicle body, the inner member 11 that rotates together with the wheel, the balls (rolling elements) 13 provided so as to be able to roll freely between the outer member 12 and the inner member 11, the sensor device 6, and the processing device 18. The sensor device 6 has a sensor rotor 17 that rotates integrally with the inner member 11, and a first sensor 7A and a second sensor 7B that are provided on the outer member 12 and detect the sensor rotor 17. The processing device 18 processes a signal output from the sensor device 6.

[0085] When the inner member 11 is displaced downward relative to the outer member 12, the first sensor 7A outputs a periodic waveform signal with a delayed phase, and the second sensor 7B outputs a periodic waveform signal with an advanced phase.

[0086] The processing device 18 is capable of executing angle detection processing, displacement calculation processing, angular velocity calculation processing, characteristic setting processing, high-order component correction processing, and load calculation processing. In the angle detection process, the rotation angles θA and θB of the sensor rotor 17 at the first detection position of the first sensor 7A and the second detection position of the second sensor 7B are determined based on the waveform signal. In the displacement calculation process, the amount of displacement of the sensor rotor 17 is found based on the difference (θAB) in the rotation angle between the first detection position and the second detection position.

[0087] In the angular velocity calculation process, the angular velocity ω of the sensor rotor 17 is found from the rotation angle. In the characteristic setting process, a filter characteristic for attenuating high-order components contained in a signal is set in accordance with the angular velocity ω. In the high-order component correction process, a signal indicating the amount of displacement is subjected to filtering using the filter characteristics described above, and a corrected amount of displacement ΔZ″ is obtained. In the load calculation process, the load acting on the inner member 11 is found from the corrected displacement amount ΔZ″.

[0088] When a downward load acts on the wheel bearing device 10 having the above-mentioned configuration, for example, a signal with a distorted waveform while changing in amplitude is acquired as the waveform signal. In reality, the acting load changes dynamically, and this changes into a high-order component error that appears in the waveform signal and affects the calculation for determining the load acting on the inner member 11. Therefore, the wheel support bearing device 10 having the above-mentioned configuration performs filtering to attenuate the high-order components, and furthermore, sets the filter characteristics according to the angular velocity ω of the sensor rotor 17. Therefore, even if the load changes dynamically, it is possible to reduce errors in the high-order components corresponding to the change. As a result, it is possible to more accurately determine the load acting on the inner member 11.

[0089] In this embodiment, the sensor rotor 17 has a gear shape. In the characteristic setting process, a high-order frequency band to be attenuated is set as the filter characteristic according to the angular velocity ω and the number of teeth N of the multiple teeth. The number of teeth N of the sensor rotor 17 affects errors in high-order components. Therefore, by setting the high-order frequency band to be attenuated according to the angular velocity ω and the number of teeth N of the sensor rotor 17, it becomes possible to more effectively reduce errors in high-order components.

[0090] In the case of this embodiment, first-order component correction data (see FIG. 6) indicating the amount of whirling G of the inner member 11 within one rotation of the inner member 11 is stored in the storage device 19. The processing device 18 is capable of executing a first-order component correction process for subtracting the amount of whirling G from the amount of displacement ΔZ obtained by the displacement calculation process, and obtaining an amount of displacement ΔZ′ after the first-order component correction.

[0091] In the load calculation process, when the rotation speed of the inner member 11 is less than a predetermined value (in the case of the above-mentioned condition 1), the load acting on the inner member 11 is found from the displacement amount ΔZ' after first-order component correction. When the rotation speed of the inner member 11 is equal to or greater than a predetermined value (in the case of the above-mentioned condition 2), the load acting on the inner member 11 is found from the displacement amount ΔZ'' after higher-order component correction found by performing the above-mentioned filter process on a signal indicating the displacement amount ΔZ' after first-order component correction.

[0092] When the rotational speed of the inner member 11 is low, the angular velocity ω is small, and it is possible to relatively accurately determine the load acting on the inner member 11 without reducing the error in the higher-order components. In contrast, when the rotational speed of the inner member 11 is high, the angular velocity ω is large, and it is possible to more accurately determine the load acting on the inner member 11 by reducing the error in the higher-order components through the filtering process.

[0093] 〔others〕 In the above embodiment, the load acts along the Z axis, but when the load acts along the X axis, the same processing as described above is executed using the output signals of the third sensor 7C and the fourth sensor 7D, thereby making it possible to detect the displacement of the inner member 11 in the direction along the X axis and the load acting on the wheel bearing device 10.

[0094] Although the case where the sensor rotor 17 is separate from the inner member 11 has been described, a part of the inner member 11 may be configured to function as the sensor rotor. The sensor rotor 17 may have a shape other than a gear shape, and may be made of a magnet, for example, with north and south poles arranged alternately along the circumferential direction. The north poles (south poles) are evenly distributed. In this case, each sensor 7 repeatedly outputs a waveform signal in which one cycle is from one north pole of the sensor rotor 17 through the south pole to the next north pole.

[0095] The sensor 7 may be of any type as long as it outputs two-phase periodic waveform signals having a phase difference by the rotation of the sensor rotor 17, and the type may be changed. The sensor 7 may be a Hall sensor other than an MR sensor. Alternatively, the sensor may be an optical type such as a transmission type photoelectric type or a reflection type photoelectric type. The type of the sensor rotor 17 is changed accordingly.

[0096] The embodiments disclosed herein are illustrative and not restrictive in all respects. The scope of the present invention is not limited to the above-described embodiments, but includes all modifications within the scope of the equivalents of the configurations described in the claims. [Explanation of symbols]

[0097] 6 Sensor device 7A First Sensor 7B Second Sensor 10 Wheel bearing device 11 Inner member 12 Outer member 13 Ball (rolling element) 17 Sensor rotor 18 Processing Equipment 19 Storage device

Claims

1. An outer member fixed to a vehicle body side; an inner member that rotates about a central axis of the outer member; a rolling element provided between the outer member and the inner member so as to be able to roll freely; a sensor device including a sensor rotor that rotates integrally with the inner member, and a first sensor and a second sensor that are provided on the outer member and detect the sensor rotor; A processing device for processing a signal output from the sensor device, When the inner member is displaced relative to the outer member, the first sensor outputs a periodic waveform signal having a delayed phase, and the second sensor outputs a periodic waveform signal having an advanced phase, The processing device includes: an angle detection process for determining a rotation angle of the sensor rotor at each of a first detection position of the first sensor and a second detection position of the second sensor based on the waveform signal; a displacement calculation process for determining a displacement amount of the sensor rotor based on a difference between the rotation angle at the first detection position and the rotation angle at the second detection position; an angular velocity calculation process for calculating an angular velocity of the sensor rotor from the rotation angle; a characteristic setting process for setting a filter characteristic for attenuating high-order components included in the signal in accordance with the angular velocity; a high-order component correction process for performing a filter process on the signal indicating the displacement amount using the filter characteristics to obtain a corrected displacement amount; a load calculation process for determining a load acting on the inner member from the corrected displacement amount; The wheel bearing device is capable of carrying out the above.

2. a storage device that stores first-order component correction data indicating an amount of whirling of the inner member within one rotation of the inner member, The processing device includes: A first-order component correction process is executed to subtract the amount of whirling from the amount of displacement calculated by the displacement calculation process, and to calculate a displacement amount after the first-order component correction. In the load calculation process, when the rotation speed of the inner member is less than a predetermined value, a load acting on the inner member is calculated from the displacement amount after the first-order component correction; when the rotation speed of the inner member is equal to or higher than a predetermined value, a load acting on the inner member is calculated from a displacement amount after high-order component correction calculated by subjecting a signal indicating the displacement amount after the first-order component correction to the filtering process.

2. A wheel bearing device according to claim 1.

3. the sensor rotor is an annular member having a gear shape with a plurality of teeth provided on an outer circumferential surface thereof at equal intervals in a circumferential direction; In the characteristic setting process, a high-order frequency band to be attenuated is set as the filter characteristic in accordance with the angular velocity and the number of the plurality of teeth.

3. A wheel bearing device according to claim 1 or 2.

Citation Information

Patent Citations

  • Rotation device

    JP2020085840A