JP2009014431A - Device for measuring physical quantity of rolling bearing unit - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To achieve a structure capable of acquiring an external force signal until reaching a high-speed traveling time, even when using an encoder in common with ABS or the like, and when not using an expensive computing unit. <P>SOLUTION: Each detection part of a pair of sensors A, B is faced to a surface to be detected of the encoder, having a phase of a characteristic change changed gradually in the width direction. Each output signal from both sensors A, B is processed successively by a phase difference ratio operation circuit 12, an adaptive filter 13, a low-pass filter 13 and a conversion processing circuit 15, to thereby determine an external force. The phase difference ratio operation circuit 12 calculates a phase difference ratio existing between each output signal from both sensors A, B in each of a plurality of periods relative to a change of each output signal from both sensors A, B. Hereby, even if the number of times of a characteristic change of the encoder is large, a load of the computing unit is suppressed, and the problem can be solved. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The physical quantity measuring device for a rolling bearing unit according to the present invention is used to measure a physical quantity such as an external force acting between an outer ring and a hub constituting the rolling bearing unit. Further, the obtained physical quantity is used for ensuring the running stability of a vehicle such as an automobile.

  For example, a vehicle wheel is rotatably supported by a suspension device such as a double-row angular type rolling bearing unit. Moreover, in order to ensure vehicle running stability, for example, anti-lock brake system (ABS), traction control system (TCS), and electronically controlled vehicle stability control system (ESC) etc. The device is in use. In order to control such various vehicle running stabilization devices, signals representing the rotational speed of the wheels, acceleration in each direction applied to the vehicle body, and the like are required. In order to perform higher-level control, it may be preferable to know the magnitude of a load (for example, one or both of a radial load and an axial load) applied to the rolling bearing unit via a wheel.

  In view of such circumstances, Patent Document 1 describes an invention in which a special encoder is used to measure the magnitude of a load applied to a rolling bearing unit. FIGS. 11 to 13 show a first example of a conventional structure related to a physical quantity measuring device for a rolling bearing unit, which employs the same load measurement principle as the structure described in Patent Document 1. FIG. In the first example of this conventional structure, a hub 2 that is a rotating side race ring that rotates together with the wheel in a state where the wheel is supported and fixed at the time of use is provided on the inner diameter side of the outer race 1 that does not rotate even when used. These are rotatably supported through a plurality of rolling elements 3 and 3. A preload is applied to each of the rolling elements 3 and 3 together with a contact angle of the rear combination type. In the illustrated example, balls are used as the rolling elements 3 and 3. However, in the case of a vehicle bearing unit that is heavy, tapered rollers may be used instead of balls.

  Further, the inner end of the hub 2 ("inner" in the axial direction means the center side in the width direction of the vehicle when assembled to the vehicle, and is the right side of FIGS. 11, 14, 15 and 17. On the contrary, to the vehicle. 11, 14, 15, and 17, which are outside in the width direction of the vehicle in the assembled state, are referred to as “outside” in the axial direction. The same applies to the entire specification). The hub 2 is supported and fixed concentrically. In addition, a sensor holder 7 in which a pair of sensors 6a and 6b are embedded is held inside a bottomed cylindrical cover 5 that closes the inner end opening of the outer ring 1, and a detection unit for both the sensors 6a and 6b. Is placed in close proximity to the outer peripheral surface which is the detected surface of the encoder 4.

  Of these, the encoder 4 is made of a magnetic metal plate. Through holes 8 and 8 and column portions 9 and 9 (see FIG. 12) are alternately arranged in the circumferential direction in the front half (axially inner half) of the outer circumferential surface of the encoder 4 which is a detected surface. And at equal intervals. The boundaries between the through holes 8 and 8 and the pillars 9 and 9 are inclined by the same angle with respect to the axial direction (width direction) of the detection surface, and the inclination direction with respect to the axial direction is determined by the detection target. The directions are opposite to each other with the axial middle portion of the surface as a boundary. Accordingly, each of the through holes 8 and 8 and each of the column portions 9 and 9 has a "<" shape with an axially intermediate portion protruding most in the circumferential direction. And among the axially outer half part and the axially inner half part of the detected surface, the boundary inclination directions are different from each other, the axially outer half part is defined as the first characteristic changing part 10, and the axially inner half part Is the second characteristic changing unit 11. In addition, each through-hole which comprises both these characteristic change parts 10 and 11 may be formed in the mutually continuous state like illustration, and it forms in the mutually independent state (each through-hole is made into "C" shape. Arrangement).

  The sensors 6a and 6b are supported and fixed inside the cover 5 by a sensor holder 7 made of synthetic resin. Each of the sensors 6a and 6b includes a permanent magnet and a Hall IC, Hall element, MR element, And a magnetic sensing element such as a GMR element. Such a pair of sensors 6a and 6b is embedded in a part of the sensor holder 7 (for example, the lower end), and the detection unit of one sensor 6a is used as the first characteristic change unit 10 and the other is The detection part of the sensor 6b is made to face and oppose the second characteristic change part 11 respectively. The positions at which the detection units of both the sensors 6a and 6b face both the characteristic change units 10 and 11 are the same as each other with respect to the phase of the characteristic change, such as the same position in the circumferential direction of the encoder 4. Further, in the neutral state where no axial load is applied between the outer ring 1 and the hub 2, the axially intermediate portions of the through holes 8, 8 and the column portions 9, 9 are projected most in the circumferential direction. The position of each member is regulated so that a portion (a portion where the tilt direction of the boundary changes) is located at the center position between the detection portions of the sensors 6a and 6b.

  In the case of the physical quantity measuring device for a rolling bearing unit configured as described above, when an axial load acts between the outer ring 1 and the hub 2 (the outer ring 1 and the hub 2 are relatively displaced in the axial direction (axial direction)). The phase at which the output signals of the sensors 6a and 6b change is shifted. That is, in the neutral state where an axial load is not applied between the outer ring 1 and the hub 2, the detecting portions of the sensors 6a and 6b are on the solid lines A and B in FIG. It faces a portion that is shifted by the same amount in the axial direction from the most protruding portion. Therefore, the phases of the output signals of the sensors 6a and 6b coincide as shown in FIG.

  On the other hand, when a downward axial load is applied to the hub 2 to which the encoder 4 is fixed as shown in FIG. 13A, the detecting portions of the sensors 6a and 6b are shown in FIG. , Opposite to the portions where the deviations in the axial direction from the most protruding portion are different from each other. In this state, the phases of the output signals of the sensors 6a and 6b are shifted as shown in FIG. Further, when an upward axial load is applied to the hub 2 to which the encoder 4 is fixed as shown in FIG. 13A, the detecting portions of both the sensors 6a and 6b are connected to the chain line hub shown in FIG. , C, that is, the deviation in the axial direction from the most projecting portion opposes different portions in the opposite direction. In this state, the phases of the output signals of the sensors 6a and 6b are shifted in the opposite direction to the case of (B), as shown in (D) of FIG.

  As described above, in the case of a conventionally known physical quantity measuring device for a rolling bearing unit as described in Patent Document 1, the phases of the output signals of both the sensors 6a and 6b are the same as those of the outer ring 1 and the hub. 2 is shifted in a direction corresponding to the direction of action of the axial load applied between the outer ring 1 and the axial direction of the outer ring 1 and the hub 2. Further, the degree of the phase shift of the output signals of the sensors 6a and 6b due to the axial load (relative displacement) increases as the axial load (relative displacement) increases. Therefore, based on the presence and absence of the phase shift of the output signals of the sensors 6a and 6b and the direction and magnitude of the shift, each is a physical quantity between the outer ring 1 and the hub 2. The direction and magnitude of the relative displacement in the axial direction, and the direction and magnitude of the axial load are obtained. Note that the processing for calculating such a physical quantity is performed by an arithmetic unit (CPU) (not shown). For this reason, the relation between the phase difference and the relative displacement or load in the axial direction, which has been examined in advance by theoretical calculation or experiment, is stored in the memory of this arithmetic unit as a relational expression or a map type. deep.

  In the case of the above-described first example of the conventional structure, the through holes 8 and 8 and the column portions 9 and 9 are alternately arranged on the detection surface of the encoder 4, and the characteristics of the detection surface are alternately set. And it is changed at equal intervals. On the other hand, as shown in FIG. 14, a physical quantity measuring device for a rolling bearing unit including an encoder 4a made of a permanent magnet, in which S poles and N poles are alternately arranged on the outer peripheral surface which is a detected surface, It is described in Patent Document 2 and the like and has been conventionally known. The basic structure and operation of the physical quantity measuring apparatus for the rolling bearing unit shown in FIG. 14 are the same as those in the first example of the conventional structure shown in FIGS. However, in the case of the second example of the conventional structure shown in FIG. 14, since a permanent magnet is provided on the encoder 4a side, in principle, if a magnetic detection element is provided on the sensor 6a, 6b side. Good, no permanent magnet is needed. In the case of the structure shown in FIG. 14, a caulking portion formed on the inner end in the axial direction of the hub body, instead of the nut shown in FIG. It is done by. However, such a structure has been conventionally known and is not related to the gist of the present invention.

  Furthermore, Patent Document 3 describes a physical quantity measuring device for a rolling bearing unit as shown in FIGS. First, in the case of the third example of the conventional structure shown in FIGS. 15 to 16, a slit-shaped transparent member is formed on the tip half of a cylindrical encoder 4 b made of a magnetic metal plate and fitted and fixed to the inner end of the hub 2. The holes 8a and 8a and the column portions 9a and 9a (see FIG. 16) are alternately arranged at equal intervals in the circumferential direction. The boundaries between the through holes 8a and 8a and the pillars 9a and 9a are linear shapes that are inclined by the same angle in the same direction with respect to the axial direction of the encoder 4b. Further, the detection portions of the pair of sensors 6a and 6b supported and fixed to the inner end portion of the outer ring 1 via the cover 5 and the sensor holder 7 are arranged on the upper and lower sides of the outer peripheral surface of the front half of the encoder 4b, which is the detection surface. The two positions (the portions in which the circumferential phases are different from each other by 180 degrees) are placed close to each other.

  In the case of a rolling bearing unit for supporting wheels of a vehicle, the axial load applied between the outer ring 1 and the hub 2 is input from the ground contact surface between the outer peripheral surface of the tire constituting the wheel coupled to the hub 2 and the road surface. Is done. Since this ground contact surface exists radially outward from the rotation center of the outer ring 1 and the hub 2, the axial load is not between the outer ring 1 and the hub 2 but as a pure axial load. 1 and the center axis of the hub 2 and the center of the grounding surface are applied with a moment in a virtual plane (in the vertical direction). When such a moment is applied between the outer ring 1 and the hub 2, the central axis of the hub 2 is inclined with respect to the central axis of the outer ring 1. Accordingly, the upper end of the encoder 4b is displaced in any direction with respect to the axial direction, and the lower end is similarly displaced in the opposite direction. As a result, the phases of the output signals of the sensors 6a and 6b, in which the detection units are placed close to and opposed to the upper and lower ends of the outer peripheral surface of the encoder 4b, are shifted in the opposite directions with respect to the neutral position. Therefore, the direction and magnitude of the axial load can be obtained based on the direction and magnitude of the phase shift between the output signals of both the sensors 6a and 6b.

  Further, in the case of the fourth example of the conventional structure shown in FIGS. 17 to 18, the through hole 8 b is formed in the tip half of the cylindrical encoder 4 c made of a magnetic metal plate and fitted and fixed to the inner end of the hub 2. 8b and column portions 9b and 9b (see FIG. 18) are alternately arranged at equal intervals in the circumferential direction. Each of these through holes 8b, 8b has a trapezoidal shape when viewed from the radial direction, and gradually changes the width dimension in the circumferential direction with respect to the axial direction. Further, the detection portion of one sensor 6 supported and fixed to the inner end portion of the outer ring 1 via the cover 5 and the sensor holder 7 is brought close to and opposed to the outer peripheral surface of the front half of the encoder 4c, which is the detected surface. ing. In the fourth example of the conventional structure configured as described above, when the outer ring 1 and the hub 2 are relatively displaced in the axial direction based on the axial load, the duty ratio of the output signal of the sensor 6 (high potential duration / 1 cycle) changes. Therefore, the magnitude of the relative displacement and further the magnitude of the axial load can be obtained based on the duty ratio.

  In the case of the first and second examples of the conventional structure shown in FIGS. 11 to 14 described above, a pair of sensors 6a in which the respective detection units are opposed to the first and second characteristic change units 10 and 11, respectively. , 6b, only one set is provided. In contrast, although not shown in the drawings, Patent Document 3 and Japanese Patent Application No. 2006-345849 each have a structure in which a plurality of sensor sets each including a pair of sensors are provided so that multidirectional displacement or external force can be obtained. Is described.

  Further, in Patent Document 4, regarding the physical quantity measuring device for a rolling bearing unit as described above, filtering processing by an adaptive filter, a notch filter, a low-pass filter, or the like is performed on the output signal of the sensor, thereby based on the whirling of the encoder. An invention for reducing errors is described. That is, if the geometric center and the rotation center of the encoder are deviated (eccentric), the encoder swings with the rotation of the rotating raceway such as the hub. The output signal of the sensor having the detection surface opposed to the detection surface of the encoder changes with the displacement of the detection surface based on the swinging motion. If no countermeasures are taken, the relative displacement between the stationary and rotating bearing wheels constituting the rolling bearing unit based on the output signal of the sensor, or between these bearing rings, is determined. An error occurs in the measured value of the applied external force. Therefore, in the case of the invention described in Patent Document 4, the output signal is subjected to a filtering process so as to prevent the deviation between the two centers from leading to an error in the measured value. Yes.

  The configuration and operation of a conventionally known physical quantity measuring device for a rolling bearing unit are as described above. When the invention relating to the physical quantity measuring device for a rolling bearing unit is implemented at a feasible cost, It is necessary to pay attention to various points. That is, as described above, the physical quantity measuring device for the rolling bearing unit obtains the external force applied between the wheel and the suspension device in order to improve the performance of the vehicle travel stabilization device such as ABS, TCS, ESC or the like. Is intended. For this reason, when an actual vehicle travel stabilization device is configured, a signal (rotational speed signal) indicating the rotational speed of the wheel is required for control of the ABS, TCS, ESC, and the like. And this rotational speed signal is the same as in the case of obtaining the signal representing the external force (external force signal), the encoder provided on the rotation side raceway such as a hub that rotates with the wheel, and the non-rotation part such as the stationary side raceway. Obtained in combination with a sensor supported by Therefore, the encoder for obtaining the rotational speed signal is preferably shared with the encoder constituting the physical quantity measuring device for the rolling bearing unit. Furthermore, it is possible to obtain the rotational speed signal on the basis of the output signal of any one or more of the sensors constituting the physical quantity measuring device of the rolling bearing unit. This is preferable from the viewpoint of effectively using the internal space of the bearing unit and reducing the cost of the entire vehicle travel stabilization device.

  From such a viewpoint, when it is intended to obtain the rotational speed signal in addition to the external force signal by one encoder (and the same sensor), if no countermeasure is taken, As an arithmetic unit (CPU) for performing calculation for obtaining an external force signal, an expensive one must be used. The reason for this is as follows.

  That is, the rotational speed signal needs to be obtained in real time compared to the external force signal. For this reason, as an encoder for obtaining this rotational speed signal, an encoder that has been generally used for ABS or the like and changes in about 48 cycles per revolution (48 pulses, in the case of an encoder made of a permanent magnet) It is preferable to use a 96-pole). The process for obtaining the rotational speed signal based on the output signal of the sensor is simple. When the vehicle travels at high speed, the above rotational speed signal is obtained even if the time of one cycle of the output signal is shortened. However, it is possible even without using a fast (expensive) computing unit.

  On the other hand, the process for obtaining the external force signal based on the output signal is much more troublesome than the case of obtaining the rotational speed signal, and the amount of calculation is increased. In particular, when a plurality of sensor groups are provided to perform the filtering process or to obtain external forces in multiple directions, the amount of calculation is considerably increased. For this reason, when the time of one cycle of the output signal is shortened during high-speed running, the external force signal may not be obtained unless an expensive arithmetic unit with a high processing speed is used. By limiting the maximum speed at which the traveling state stabilization process can be performed using an external force signal, the external force signal can be obtained as long as the traveling speed is within the setting without using an expensive calculator. However, such a countermeasure is not preferable in view of the necessity of high-accuracy traveling state stabilization processing at higher speeds.

JP 2006-1113017 A JP 2006-317420 A JP 2007-93580 A JP 2007-40954 A

  The physical quantity measuring device for a rolling bearing unit according to the present invention, in view of the above-described circumstances, when an encoder is shared with ABS or the like, outputs an external force signal until high speed running without using an expensive arithmetic unit. It was invented to realize the resulting structure.

A physical quantity measuring device for a rolling bearing unit that is an object of the present invention includes a rolling bearing unit and a physical quantity measuring device.
Of these, the rolling bearing unit includes a stationary-side bearing ring, a rotating-side bearing ring, and a plurality of rolling elements. Of these, the stationary-side raceway has a stationary-side raceway on the stationary-side peripheral surface and does not rotate during use. The rotation-side raceway has a rotation-side raceway on the rotation-side circumferential surface facing the stationary side circumferential surface in the radial direction, and rotates during use. Further, each of the rolling elements is provided between the rotating side track and the stationary side track so as to freely roll.
The physical quantity measuring device includes an encoder, at least one sensor, and a computing unit.
Of these, the encoder is supported and fixed to a part of the rotation side raceway, concentrically with the rotation side raceway, and has a detection surface concentric with the rotation side raceway. Are alternately changed in the circumferential direction.
The sensor is supported by a non-rotating portion such as the stationary raceway, or a housing or knuckle that supports the stationary raceway, with the detection unit facing the detection surface of the encoder. Therefore, the output signal is changed in response to the change in the characteristics of the detected surface.
Further, the computing unit is based on the output signal of the sensor, and at least of the relative displacement between the stationary and rotating side raceways and the external force acting between the raceways. One physical quantity is calculated.
In particular, in the physical quantity measuring apparatus for a rolling bearing unit according to the present invention, the computing unit calculates the physical quantity for each of a plurality of cycles with respect to a change in the output signal of the sensor.

  When the physical quantity measuring apparatus for a rolling bearing unit according to the present invention as described above is implemented, as described in claim 2, this calculation is used to obtain the rotational speed of the calculator or the rotating side race. The second arithmetic unit provided separately from the detector has the function of obtaining a signal representing the rotational speed of the rotating raceway by using the output signal of any one sensor for the entire period. Make it.

Further, when the physical quantity measuring device for a rolling bearing unit of the present invention as described above is implemented, for example, as described in claim 3, the position of the boundary where the characteristic of the detection surface of the encoder changes in the circumferential direction Are gradually changed in the width direction of the detected surface. Further, the detection portions of at least one pair of sensors are opposed to different portions of the detection surface. The computing unit calculates physical quantities such as load, force, and displacement based on a phase difference existing between output signals of the sensors.
And in order to calculate such a physical quantity, the said arithmetic unit adds up the phase difference which exists between the output signals of each said sensor regarding the continuous several period. Further, based on the total value and the total length of the plurality of cycles, the physical quantity between the stationary side and rotating side raceways is calculated.
When the invention described in claim 3 is carried out, as described in claim 4, for example, the sum of the phase differences belonging to a plurality of consecutive periods and the total length of the plurality of periods Filter the ratio. Thereafter, the physical quantity is calculated based on the ratio between the total value and the length.

Alternatively, as described in claim 5, the pitch at which the characteristic of the detected surface of the encoder changes is gradually changed with respect to the width direction of the detected surface. The computing unit calculates a physical quantity between the stationary side and the rotating side raceways based on the duty ratio (high potential duration / one cycle) of the output signal of the sensor.
In order to calculate such a physical quantity, the computing unit sums up the values (high potential duration) of the duty ratios of the output signals of the sensors with respect to a plurality of successive cycles, The physical quantity is calculated based on the value and the overall length of the plurality of periods.
When carrying out the invention described in claim 5, for example, as described in claim 6, a total value of values that become a numerator of a duty ratio belonging to a plurality of continuous cycles, A filtering process is applied to the ratio to the length of the entire period. Thereafter, the physical quantity is calculated based on the ratio of the length to the total value.

  Further, when the physical quantity measuring device for a rolling bearing unit of the present invention as described above is implemented, for example, as described in claim 7, the computing unit intermittently outputs information on an output signal of at least one sensor. Based on the obtained information, the physical quantity between the stationary side and the rotating side races is calculated.

  When the invention described in claim 7 is carried out, for example, as described in claim 8, the first to second examples of the conventional structure shown in FIGS. 11 to 14 described above, or FIGS. As in the third example, the position of the boundary where the characteristic of the detected surface of the encoder changes in the circumferential direction is gradually changed in the width direction of the detected surface. At the same time, the detection portions of at least one pair of sensors are opposed to different portions of the detected surface. The computing unit calculates a physical quantity based on the phase difference existing between the output signals of these sensors. In order to calculate the physical quantity, the computing unit intermittently calculates the phase difference existing between the output signals of the sensors, and calculates the phase difference and the length of the period to which the phase difference belongs. Based on this, the physical quantity between the stationary side and rotating side raceways is calculated.

  Alternatively, when the invention described in claim 7 is carried out, as described in claim 9, for example, as in the third example of the conventional structure shown in FIG. The pitch at which the characteristics change is gradually changed with respect to the width direction of the detected surface. Then, the computing unit calculates the physical quantity based on the duty ratio of the output signal of the sensor. In order to calculate the physical quantity, the computing unit intermittently obtains the duty ratio of the output signal of the sensor, and based on the obtained duty ratio, the physical quantity between the stationary side and the rotating side raceway rings. Is calculated.

  In carrying out the present invention, for example, as described in claim 10, the stationary side raceway is an outer ring that is supported and fixed to a suspension device of a vehicle when in use, and the stationary side raceway is defined as the outer race. A double row outer ring raceway provided on the inner peripheral surface. In addition, the rotating side raceway is a hub that rotates together with the wheel while the wheels are coupled and fixed in use, and the rotating side raceway is a double row outer ring raceway provided on the outer peripheral surface of the hub. A plurality of each rolling element is provided for each row.

According to the physical quantity measuring device of the rolling bearing unit of the present invention configured as described above, when the encoder is shared with ABS or the like, the external force signal is output until the high-speed running without using an expensive arithmetic unit. The resulting structure can be realized.
That is, in the case of the physical quantity measuring device for a rolling bearing unit according to the present invention, the computing unit calculates the physical quantity for each of a plurality of cycles of the output signal with respect to the change of the output signal of the sensor accompanying the rotation of the encoder. For this reason, even if the number of times per rotation (number of pulses / 1 rotation) at which the characteristics of the detected surface of the encoder change is increased in order to obtain a vehicle speed signal close to real time for the control of the ABS or the like, Since the physical quantity is calculated, the number of calculations that need to be performed during one rotation of the encoder can be suppressed. For this reason, it is possible to obtain an external force signal even when a relatively inexpensive arithmetic unit is used until the vehicle travels at a high speed.

[First example of embodiment]
A first example of an embodiment of the present invention corresponding to claims 1 to 4 and 10 will be described with reference to FIGS. The feature of this example is that the arithmetic unit constituting the physical quantity measuring device of the rolling bearing unit has a relative displacement amount in the axial direction between the outer ring 1 and the hub 2 (see, for example, FIG. 11), or these outer ring 1 and hub 2. The axial load acting during the period is calculated for each of a plurality of cycles with respect to changes in the output signals of the pair of sensors 6a and 6b. Since the structure and operation of the other parts are basically the same as those of the above-described conventional structure, overlapping illustrations and explanations of equivalent parts will be omitted or simplified, and the following description will focus on the features of this example. .

  In this example, as shown in FIG. 1, the width of the surface to be detected of the encoder E in which the boundary of the characteristic change (the boundary between the through hole and the column portion, or the boundary between the S pole and the N pole) is a “<” shape. Two positions in the direction are scanned by a pair of sensors A and B. Based on the phase difference existing between the output signals of both sensors A and B, a load applied to the rotating member that supports and fixes the encoder E is applied. The case where the present invention is applied to the required structure is shown. The mechanism for obtaining this load based on the phase difference is the same as the first and second examples of the conventional structure shown in FIGS. That is, when a load is applied in the vertical direction in FIG. 1, the encoder E is displaced in the vertical direction with respect to both the sensors A and B, and the phases of the output signals of both the sensors A and B are as shown in FIG. Shifts (changes). Therefore, the axial load (or the relative displacement in the axial direction) is obtained by processing the output signals of both sensors A and B by the circuit shown in FIG.

  In the circuit shown in FIG. 2, first, the phase difference ratio calculation circuit 12 obtains the phase difference existing between the output signals of the two sensors A and B, and further calculates the phase difference between the two sensors A and B. The phase difference ratio (= phase difference / 1 period) between the output signals of both the sensors A and B is obtained by dividing by the period of the output signal. The signal representing the phase difference ratio is then processed by the adaptive filter 13 to reduce detection errors based on geometric errors such as the shape error of the encoder E, the inclination at the time of attachment, and the eccentricity. This adaptive filter 13 is described in Patent Document 4 described above, and operates with, for example, an LMS algorithm or a synchronous LMS algorithm to reduce a detection error based on the geometric error. The signal representing the phase difference ratio that has been processed by the adaptive filter 13 is then passed through the low-pass filter 14 to remove noise. The low-pass filter 14 removes noise, but may remove even a necessary high-frequency component. Therefore, it is preferable to select a cut-off frequency or a cut-off order (rotation n-th order) as appropriate. Alternatively, the low-pass filter 14 itself may not be used.

  The signal representing the phase difference ratio that has undergone processing by the low-pass filter 14 (or processing by the adaptive filter 13 when the low-pass filter 14 is omitted) represents the axial load by the conversion processing circuit 15. Convert to signal. In order to enable this conversion, the conversion processing circuit 15 stores a relationship representing the phase difference ratio and the magnitude of the axial load, which has been obtained in advance by experiment or analysis, as a formula or a map. Record inside. Since each of the above-described processes has been conventionally known, for example, as described in Patent Document 4, detailed description thereof will be omitted.

  Among the processes performed by the circuit shown in FIG. 2, the phase difference ratio calculation circuit 12 obtains the phase difference ratio from the output signals of the sensors A and B, and the conversion processing circuit 15 calculates the phase difference ratio from the phase difference ratio. The processes for obtaining the axial load are all simple, and even if the number of processes increases somewhat, the possibility that the CPU capacity is insufficient is low. On the other hand, the processing by the adaptive filter 13 and the low-pass filter 14 has a large amount of calculation, and the processing by both the filters 13 and 14 each time a signal representing the phase difference ratio is sent to the conversion processing circuit 15. Need to do. As described above, when the encoder E is shared with the one for ABS control, there is a possibility that processing at high speed cannot be performed unless a high-speed CPU with high cost is used. In view of such circumstances, in the case of the present invention, in the processing circuit of FIG. 2, the number of times that the signal representing the phase difference ratio is sent after the adaptive filter 13 during one rotation of the encoder E is determined by the encoder. The number of period of characteristic change existing on the E detection surface is smaller than 1 (when N is a natural number of 2 or more, it is suppressed to 1 / N). That is, the phase difference ratio calculation circuit 12 is directed toward the adaptive filter 13, and the phase difference is directed toward the adaptive filter 13 every N periods (N: the same natural number as described above) with respect to the output signals of the sensors A and B. A signal representing the ratio is sent out. Hereinafter, the point of reducing the frequency of sending out a signal representing the phase difference ratio (set to 1 / N) will be described.

  First, a description will be given of a case where the processing known in the prior art described in Patent Document 4 and the like is performed by a general method (for each pulse) without applying the technique according to the present invention. In this case, as shown in FIG. 3, the phase difference x1 is from the edge of the sensor A {in the case of FIG. 3, the descending point (falling point)} to the edge of the sensor B. Then, a value obtained by dividing the phase difference x1 by the period y1 of the sensor A (same as the period of the sensor B), that is, x1 / y1 is a phase difference ratio, and a signal representing the phase difference ratio x1 / y1 is The data is sent to the conversion processing circuit 15 through the adaptive filter 13 and the low-pass filter 14, and the axial load is obtained. Hereinafter, signals representing the phase difference ratios x2 / y2 and x3 / y3 obtained one after another along with the rotation of the encoder E are sequentially converted through the adaptive filter 13 and the low-pass filter 14 into the conversion process. It sends to the circuit 15 and calculates | requires the said axial load.

  When such general processing is performed, as described above, the burden on the processing circuit shown in FIG. 2 may become excessive. That is, for example, when the number of characteristic change periods (pulses) during one rotation of the encoder E required for ABS control is 48, the encoder E is rotated as shown in FIG. The processing circuit needs to perform the processing 48 times including the adaptive filter 13 and the low-pass filter 14. The time required for one rotation of the encoder E during high-speed traveling is short, and an expensive CPU is required to perform processing 48 times during that time. On the other hand, even if traveling stability is taken into consideration, the signal indicating the axial load requires 48 times during one rotation of the encoder E, unless it is an extremely high-performance vehicle. It is considered unnecessary. Therefore, in the case of this example, the phase difference ratio calculation circuit 12 performs the following processing to reduce the load on the circuit shown in FIG.

  As the encoder E rotates, both the sensors A and B obtain information on boundary of characteristic change (pulse edge information) existing on the detected surface of the encoder E, that is, phase differences x1, x2, x3. And information indicating the periods y1, y2, y3... To which the phase differences x1, x2, x3. This point is the same as the case of the process described in Patent Document 4 and the like and conventionally known. However, in the case of this example, the phase difference ratio calculation circuit 12 does not change the phase difference ratios x1 / y1, x2 based on the phase differences x1, x2, x3... And the periods y1, y2, y3. / Y2, x3 / y3... Are not generated, but a signal indicating the phase difference ratio x / y is generated every N periods (N: natural number of 2 or more).

  For example, when a signal representing the phase difference ratio x / y is created every two periods, the output signals of the sensor A and the sensor B are added together with the phase differences x1 and x2 in the adjacent periods. Add together y1 and y2, which are the lengths of the periods to which both phase differences x1 and x2 belong. Then, with the length “y1 + y2” for the two cycles, the phase difference ratio “x1 + x2” for the two cycles is divided and the phase difference ratio for the two adjacent cycles (the average of the phase difference ratios for the two cycles). Value) “(x1 + x2) / (y1 + y2)” is obtained, and a signal representing this average value is sent to the conversion processing circuit 15 via the adaptive filter 13 and the low-pass filter 14 to obtain the axial load.

  By performing such processing, the amount of processing for the filtering processing by the adaptive filter 13 and the low-pass filter 14 and the processing for calculating the axial load by the conversion processing circuit 15 is processed for each pulse as described above. It becomes 1/2 compared with the case where it performs. Although the processing amount of the phase difference ratio calculation circuit 12 is slightly increased, the processing performed by the phase difference ratio calculation circuit 12 is simple and can be sufficiently handled without using an expensive CPU. In this example, the output signals (pulses) of the sensors A and B are not thinned out to generate a signal representing the phase difference ratio x / y, but as an average value of adjacent periods. . For this reason, even when an error exists in the pitch of the characteristic change of the detected surface of the encoder E, the influence of this error can be reduced by averaging. That is, regarding the phase difference ratio x / y, the amount of processing for calculating the axial load can be reduced while accurately obtaining the axial load without increasing an error.

  The above description is made when N = 2 and the phase differences x1 and x2 of adjacent periods and the lengths y1 and y2 of the periods are added, but N = 3 and three consecutive periods. The phase differences x1, x2, x3 and the cycle lengths y1, y2, y3 can be added together. That is, with respect to three consecutive cycles, the phase differences x1, x2, and x3 of the sensors A and B are added together, and the lengths y1, y2, and y3 of these three cycles are added together to obtain the phase difference in these three cycles. The average value “(x1 + x2 + x3) / (y1 + y2 + y3)” can also be obtained. Such a signal representing the average value “(x1 + x2 + x3) / (y1 + y2 + y3)” of the phase difference ratios for three periods is sent to the conversion processing circuit 15 via the adaptive filter 13 and the low-pass filter 14. If the axial load is obtained, the amount of processing in both the filters 13 and 14 becomes 1/3 compared to the case where processing is performed for each pulse.

  As described above, the amount of processing required to obtain the axial load can be reduced as the number N of consecutive periods in which the phase difference x and the period length y are added is increased, but the number N is increased. The time required to obtain a signal representing the axial load (the delay in load calculation output = the time required from the input of the output signals of the sensors A and B to the output of the signal representing the axial load). growing. In a situation where the axial load changes continuously, the magnitude of the signal indicating the axial load changes stepwise (a signal indicating the axial load obtained based on the output signals of the sensors A and B). The output representing this axial load becomes a stepped shape, as is thinned out). And if such a situation becomes remarkable (the step difference of the stairs becomes large), the accuracy of the control for stabilizing the running state in consideration of the axial load is deteriorated. Therefore, the number N cannot be increased. Therefore, the following discussion will be made on the consideration made to know the appropriate value of the number N.

  4 to 6 show that the axial load in both directions is applied while a vehicle equipped with a physical quantity measuring device for a rolling bearing unit incorporating an encoder E in which the characteristics of the surface to be detected change at 48 cycles per revolution is traveling at 80 km / h. The phase difference ratio x / y between the output signals of the pair of sensors A and B is from -1.0 to 1.0 (the output signals of both the sensors A and B coincide with each other from the state shifted by one cycle) In this case, the state is changed for 0.2 sec. Such a change is so rapid that it does not actually occur. The solid line in FIGS. 4-6 has shown the change condition of the said phase difference ratio x / y (axial load tied to). The broken line indicates that the phase difference ratio x / y is sent from the phase difference ratio calculation circuit 12 when N = 1 and the phase difference ratio x / y is obtained for every cycle of the sensors A and B. A signal representing y (a signal representing the axial load sent from the conversion processing circuit 15). The one-dot chain line indicates that the phase difference ratio x / y is sent from the phase difference ratio calculation circuit 12 when N = 2 and the phase difference ratio x / y is obtained every two cycles of the sensors A and B. A signal representing / y is represented. The two-dot chain line indicates that the phase difference ratio is sent from the phase difference ratio calculation circuit 12 when N = 3 and the phase difference ratio x / y is obtained every three cycles of the sensors A and B. It represents a signal representing x / y. Further, the three-dot chain line indicates that the phase difference ratio sent out from the phase difference ratio calculation circuit 12 when N = 4 and the phase difference ratio x / y is obtained every three cycles of the sensors A and B. It represents a signal representing x / y.

  As apparent from FIGS. 4 to 6, when the phase difference ratio x / y is obtained based on the output signals of the sensors A and B, the phase difference ratio x / y actually changes continuously. Even in this case, it is possible to obtain a signal such that the phase difference ratio x / y changes stepwise. Further, the step difference of the step-like change becomes more remarkable as the number N increases. As the step becomes larger, not only the error generated in the measured value of the axial load increases, but also the response delay until the axial load is obtained (the output signals of the sensors A and B are the phase difference ratio). The time required from when the signal is input to the arithmetic circuit 12 to when the signal representing the axial load is output from the conversion processing circuit 15 increases.

FIG. 6 shows the relationship between the number N and the response delay obtained by taking into account the processing time of the adaptive filter 13 and the low-pass filter 14. As shown in FIG. 6, this response delay stops at 1.4 msec when N = 1, whereas 3.5 msec when N = 2, 6 msec when N = 3, In the case of N = 4, it is 8.6 msec.
The above-mentioned simulation is for the case where the vehicle speed is 80 km / h. FIGS. 7 to 8 show the simulation results for the case where the vehicle speed is 40 km / h. As is apparent from a comparison between FIGS. 4 to 5 and FIGS. 7 to 8, the response delay increases further as the vehicle speed decreases. Specifically, the response delay when N = 1 is 2 msec, 5 msec when N = 2, 12 msec when N = 3, and 16 msec when N = 3.

FIG. 9 collectively shows the influence of the number of N and the vehicle speed on the response delay. As is apparent from FIG. 9, the response delay increases in proportion to N, which is the sum of the phase difference ratio and the length of the period, and is inversely proportional to the vehicle speed.
The absolute value of the response delay in the simulation whose result is shown in FIG. 6 varies depending on the filtering method using the adaptive filter 13 and the low-pass filter 14. Therefore, although the specific numerical value of the response delay shown in FIG. 6 is a reference value, even if the filtering method is changed, the tendency of the response delay based on the number of N and the vehicle speed is the same. In any case, in consideration of performing control for stabilizing the traveling state of the vehicle using the axial load, it is preferable from the aspect of performing highly accurate control to reduce the response delay. Considering these things, it is preferable to keep the number of N as small as possible within a range where the procurement cost of the CPU can be suppressed. In the case of N = 1, there is a high possibility that the procurement cost of the CPU will increase due to the reasons described above. Therefore, setting N = 2 can increase the rate at which the amount of computation can be reduced (N = 1 → N = 2 by setting N = 2), while reducing costs by reducing the amount of calculation. It is considered that the balance is the best from the viewpoint that the response delay can be minimized and the control for stabilizing the running state can be performed while ensuring sufficient accuracy.

[Second Example of Embodiment]
A second example of the embodiment corresponding to claims 1 to 3, 7, 8, and 10 of the present invention will be described with reference to FIG. In order to implement the present invention, as in the first example of the embodiment described above, the phase difference of N consecutive periods (preferably two periods) is divided by the length of N periods. Obtaining the average value of the phase difference ratios for N periods is preferable from the viewpoint of suppressing the pitch error related to the boundary of the characteristic change existing on the detected surface of the encoder E. However, if the accuracy of the pitch of the characteristic change is sufficiently high, the change (pulse) of the output signals of the sensors A and B is converted into the phase difference ratio calculation circuit 12 for the entire period (see FIG. 2) as in the first example. Even if the so-called frequency division is performed in which the pulses are thinned out and taken into the phase difference ratio calculation circuit 12, the same effect as in the case of obtaining the average value can be obtained. That is, the same as the above-mentioned average value is obtained by incorporating the pulse into the phase difference ratio calculation circuit 12 every other period (once every two periods) or every other period (once every three periods). In addition, it is possible to reduce the amount of processing for the filtering process by the adaptive filter 13 and the low-pass filter 14 and the axial load calculation by the conversion processing circuit 15 (see FIG. 2), and use a relatively inexpensive arithmetic unit. It is possible to obtain a signal representing the magnitude of an external force such as an axial load until traveling at a high speed.

  However, in the case of performing pulse division and frequency division, it is necessary that the pitch accuracy regarding the boundary of the characteristic change existing on the detection surface of the encoder E is sufficiently high. This point will be described with reference to FIG. Note that C and D shown in FIG. 10 each represent one cycle of characteristic change. In the case of frequency division, if the frequency division number is N, the frequency is divided into 1 / N. Considering the 1/2 frequency division, since one pulse is used by skipping, a boundary between the pair of sensors A and B (see FIG. 2) and the first pulse and period is C (for example, rotation). It is divided into two types: a case starting from the front edge in the direction) and a case starting from the boundary relating to D. This is the same as using a pair of encoders independent of each other between the sensors A and B. Therefore, the encoder E having a good accuracy of the surface to be detected (pitch and shape accuracy with respect to the boundary of the characteristic change) is used, and the pulse and the cycle have different periods regardless of whether they start from C or D. It is necessary not to be affected by the use. In other words, if the encoder E having a good detection surface accuracy is used, the amount of processing can be reduced by frequency division.

  The present invention can be implemented not only in the structure for obtaining the relative displacement amount and the axial load in the axial direction based on the phase difference existing between the output signals of the pair of sensors arranged apart in the axial direction. For example, as shown in FIGS. 15 to 16 described above, the inclination of the encoders of a pair of sensors arranged at two positions on the opposite side in the radial direction is calculated based on the phase difference existing between the output signals. It can be implemented with a structure for obtaining a load. Furthermore, the present invention can also be implemented with a structure for obtaining the relative displacement amount and the axial load in the axial direction based on the duty ratio of the output signal of one sensor as shown in FIGS.

The schematic diagram which shows the relationship between the to-be-detected surface of an encoder and the scanning position of a pair of sensor for demonstrating this invention. The block diagram which shows the circuit which processes the output signal of the said both sensors. The diagram which shows the output signal of both these sensors. Diagram showing the relationship between the sampling frequency of the output signals of both sensors and the response delay when the phase difference existing between the output signals of both sensors changes during traveling at 80 km / h. . The diagram which expands and shows the center part of FIG. FIG. 6 is a diagram corresponding to the central portion of FIG. 5 for explaining a response delay time until an output signal is obtained. A diagram showing the relationship between the sampling frequency of the output signals of both sensors and the response delay when the phase difference existing between the output signals of the two sensors changes during traveling at 40 km / h. . The diagram which expands and shows the center part of FIG. The diagram which shows the relationship between travel speed, the sampling frequency of the output signal of both said sensors, and a response delay. The schematic diagram which shows the to-be-detected surface of an encoder, in order to demonstrate the state which calculates | requires the information regarding the output signal of one sensor intermittently. Sectional drawing which shows the 1st example of a conventional structure. The figure which looked at a part of encoder incorporated in this 1st example from the diameter direction. The diagram for demonstrating the state from which the output signal of a pair of sensor changes based on an axial load. Sectional drawing which shows the 2nd example of a conventional structure. Sectional drawing which shows the 3rd example. The figure which looked at a part of encoder incorporated in this 3rd example from the diameter direction. Sectional drawing which shows the 4th example of a conventional structure. The figure which looked at a part of encoder incorporated in this 4th example from the diameter direction.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Outer ring 2 Hub 3 Rolling element 4, 4a-4c Encoder 5 Cover 6, 6a, 6b Sensor 7, 7a Sensor holder 8, 8a, 8b Through-hole 9 Column part 10 First characteristic change part 11 Second characteristic change part 12th place Phase difference ratio calculation circuit 13 Adaptive filter 14 Low pass filter 15 Conversion processing circuit A Sensor B Sensor E Encoder

Claims (10)

A rolling bearing unit and a physical quantity measuring device;
Of these, the rolling bearing unit has a stationary side raceway on the stationary side circumferential surface, and a stationary side raceway that does not rotate during use, and a rotational side raceway on the rotational side circumferential surface that faces the stationary side circumferential surface in the radial direction. A rotating side race ring that rotates when in use, and a plurality of rolling elements provided between the rotating side raceway and the stationary side raceway so as to be freely rollable.
The physical quantity measuring device includes an encoder, at least one sensor, and a computing unit,
Of these, the encoder is supported and fixed to a part of the rotation side raceway, concentrically with the rotation side raceway, and has a detection surface concentric with the rotation side raceway. Are alternately changed in the circumferential direction,
The sensor is supported by a portion that does not rotate with the detection unit facing the detection surface of the encoder, and changes an output signal in response to a change in the characteristics of the detection surface.
The computing unit is based on the output signal of the sensor, and at least one of the relative displacement between the stationary and rotating side raceways and the external force acting between the raceways. It has a function to calculate physical quantities.
In the physical quantity measuring device for rolling bearing units,
The apparatus for measuring a physical quantity of a rolling bearing unit, wherein the computing unit calculates the physical quantity for each of a plurality of cycles with respect to a change in an output signal of the sensor.   The second computing unit provided separately from this computing unit or the computing unit in order to obtain the rotational speed of the rotating side race ring uses the output signal of any one sensor for the entire period. Thus, the physical quantity measuring device for a rolling bearing unit according to claim 1, which has a function of obtaining a signal representing the rotational speed.   The position in the circumferential direction of the boundary where the characteristic of the detected surface of the encoder changes is gradually changed in the width direction of the detected surface, and the detection parts of at least one pair of sensors are changed to different parts of the detected surface. The computing unit calculates a physical quantity based on a phase difference existing between the output signals of each sensor. In order to calculate the physical quantity, the computing unit The phase difference existing between the output signals is summed for a plurality of consecutive periods, and the physical quantity between both stationary and rotating raceways is calculated based on the total value and the length of the entire periods. The physical quantity measuring device for a rolling bearing unit according to any one of claims 1 to 2, wherein   After filtering the ratio of the total value of phase differences belonging to a plurality of consecutive periods and the total length of the plurality of periods, based on the ratio of the length to these total values, the stationary side and the rotating side The physical quantity measuring device for a rolling bearing unit according to claim 3, wherein the physical quantity between the two bearing rings is calculated.   The pitch at which the characteristics of the detected surface of the encoder change gradually changes in the width direction of the detected surface, and the computing unit calculates a physical quantity based on the duty ratio of the output signal of the sensor. In order to calculate the above, the arithmetic unit sums up the values that become the numerator of the duty ratio among the output signals of the respective sensors with respect to a plurality of consecutive periods, and based on the total value and the length of the whole of the plurality of periods. The physical quantity measuring device for a rolling bearing unit according to any one of claims 1 to 2, wherein a physical quantity between both stationary and rotating side raceways is calculated.   Based on the ratio of the total value of the values that are the numerator of the duty ratio belonging to a plurality of consecutive periods and the ratio of the total length of the plurality of periods to the length of the total value. The physical quantity measuring device for a rolling bearing unit according to claim 5, wherein the physical quantity between the stationary side and rotating side raceways is calculated.   The computing unit intermittently obtains information related to an output signal of at least one sensor, and calculates a physical quantity between the stationary and rotating side raceways based on the obtained information. The physical-quantity measuring apparatus of the rolling bearing unit described in any one of these.   The position in the circumferential direction of the boundary where the characteristic of the detected surface of the encoder changes is gradually changed in the width direction of the detected surface, and the detection parts of at least one pair of sensors are changed to different parts of the detected surface. The computing unit calculates a physical quantity based on a phase difference existing between the output signals of each sensor. In order to calculate the physical quantity, the computing unit The phase difference existing between the output signals is intermittently obtained, and based on the obtained phase difference and the length of the period to which the phase difference belongs, the physical quantity between the stationary and rotating raceways is calculated. The physical quantity measuring device of a rolling bearing unit according to claim 7 for calculation.   The pitch at which the characteristics of the detected surface of the encoder change gradually changes in the width direction of the detected surface, and the computing unit calculates a physical quantity based on the duty ratio of the output signal of the sensor. In order to calculate the above, the computing unit intermittently obtains the duty ratio of the output signal of the sensor, and based on the obtained duty ratio, calculates a physical quantity between the stationary and rotating side raceways. The physical quantity measuring apparatus of the rolling bearing unit according to claim 7.   The stationary raceway is an outer ring that is supported and fixed to the suspension system of the vehicle when in use. Is a hub that rotates together with this wheel in a state of being coupled and fixed, the rotation side track is a double row outer ring raceway provided on the outer peripheral surface of the hub, and a plurality of rolling elements are provided for each row. The physical quantity measuring device for a rolling bearing unit according to any one of claims 1 to 9.

Images