JP4457701B2 - Rolling bearing unit with rolling element revolution speed detector - Google Patents

Description

  A rolling bearing unit with a revolution speed detecting device for a rolling element according to the present invention relates to an improvement of a rolling bearing unit for supporting wheels of a moving body such as an automobile, a railway vehicle, and various transport vehicles, and constitutes the rolling bearing unit. The revolution speed of the rolling element and the load applied to the rolling bearing unit (one or both of the radial load and the axial load) are measured and used to ensure the stability of the operation of the moving body.

  For example, an automobile wheel is rotatably supported by a double row angular rolling bearing unit with respect to a suspension device. In order to ensure the running stability of automobiles, vehicle running stabilizers such as an antilock brake system (ABS), a traction control system (TCS), and a vehicle stability control system (VSC) are used. . In order to control such various vehicle travel stabilizers, 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 (one or both of a radial load and an axial load) applied to the rolling bearing unit via the wheel.

  In view of such circumstances, Patent Document 1 describes a rolling bearing unit with a load measuring device capable of measuring a radial load. The conventional rolling bearing unit with a load measuring device of the first example measures a radial load and is configured as shown in FIG. A hub 2, which is a rotating wheel, is connected to the inner diameter side of the outer ring 1, which is a stationary wheel, and is supported by the suspension device. The hub 2 includes a hub body 4 having a rotation-side flange 3 for fixing a wheel at an outer end thereof (an end on the outer side in the width direction when assembled to a vehicle), and an inner end of the hub body 4. And an inner ring 6 that is externally fitted to the end (on the widthwise center side in the assembled state in the vehicle) and held down by a nut 5. And the double row outer ring raceways 7 and 7 each formed on the inner peripheral surface of the outer ring 1 and each of which is a stationary side track, and the double row each formed on the outer peripheral surface of the hub 2 and each of which is a rotation side track. A plurality of rolling elements 9 a and 9 b are arranged between the inner ring raceways 8 and 8, respectively, so that the hub 2 can freely rotate on the inner diameter side of the outer ring 1.

  A mounting hole 10 that diametrically penetrates the outer ring 1 is formed in a substantially vertical direction at an upper end portion of the outer ring 1 in a portion between the double row outer ring raceways 7 and 7 at an intermediate portion in the axial direction of the outer ring 1. Yes. In the mounting hole 10, a circular (rod-shaped) displacement sensor 11, which is a load measuring sensor, is mounted. This displacement sensor 11 is a non-contact type, and the detection surface provided on the front end surface (lower end surface) is closely opposed to the outer peripheral surface of the sensor ring 12 fitted and fixed to the intermediate portion in the axial direction of the hub 2. When the distance between the detection surface and the outer peripheral surface of the sensor ring 12 changes, the displacement sensor 11 outputs a signal corresponding to the amount of change.

  In the case of the conventional rolling bearing unit with a load measuring device configured as described above, the load applied to the rolling bearing unit can be obtained based on the detection signal of the displacement sensor 11. That is, the outer ring 1 supported by the vehicle suspension device is pushed downward by the weight of the vehicle, whereas the hub 2 supporting and fixing the wheel tends to stop at the same position. For this reason, the greater the weight, the greater the deviation between the center of the outer ring 1 and the center of the hub 2 based on the elastic deformation of the outer ring 1, the hub 2, and the rolling elements 9a, 9b. The distance between the detection surface of the displacement sensor 11 and the outer peripheral surface of the sensor ring 12 provided at the upper end of the outer ring 1 becomes shorter as the weight increases. Therefore, if the detection signal of the displacement sensor 11 is sent to the controller, the radial load applied to the rolling bearing unit in which the displacement sensor 11 is incorporated can be obtained from a relational expression or a map obtained beforehand through experiments or the like. Based on the radial load applied to each rolling bearing unit thus obtained, the ABS is properly controlled, and the driver is informed of the poor loading state.

  In the conventional structure shown in FIG. 17, in addition to the radial load applied to the rolling bearing unit, the rotational speed of the hub 2 can also be detected. For this purpose, the rotational speed detecting encoder 13 is fitted and fixed to the inner end of the inner ring 6, and the rotational speed detecting sensor 15 is supported on the cover 14 attached to the inner end opening of the outer ring 1. Yes. And the detection part of this rotational speed detection sensor 15 is made to oppose the detected part of the said rotational speed detection encoder 13 through a detection gap.

  When the rolling bearing unit incorporating the rotational speed detecting device as described above is used, the rotational speed detecting encoder 13 is rotated together with the hub 2 to which the wheel is fixed, and the detected portion of the rotational speed detecting encoder 13 is rotated as described above. When traveling in the vicinity of the detection unit of the speed detection sensor 15, the output of the rotation speed detection sensor 15 changes. The frequency at which the output of the rotational speed detection sensor 15 changes in this way is proportional to the rotational speed of the wheel. Therefore, if the output signal of the rotational speed detection sensor 15 is sent to a controller (not shown), ABS and TCS can be controlled appropriately.

  The rolling bearing unit with a load measuring device of the first example of the conventional structure as described above is for measuring the radial load applied to the rolling bearing unit, but the structure for measuring the axial load applied to the rolling bearing unit is also, It is described in Patent Document 2 and the like and has been conventionally known. FIG. 18 shows a rolling bearing unit with a load measuring device described in Patent Document 2 for measuring an axial load. In the case of the second example of the conventional structure, the rotation side flange 3a for supporting the wheel is fixed to the outer peripheral surface of the outer end portion of the hub 2a which is a rotating wheel. In addition, a fixed-side flange 17 for fixing the outer ring 1a to the knuckle 16 constituting the suspension device is fixed to the outer peripheral surface of the outer ring 1a which is a stationary wheel. A plurality of rolling rings are respectively provided between the double row outer ring raceways 7 and 7 formed on the inner peripheral surface of the outer ring 1a and the double row inner ring raceways 8 and 8 formed on the outer peripheral surface of the hub 2a. By providing the moving bodies 9a and 9b so as to be able to roll, the hub 2a is rotatably supported on the inner diameter side of the outer ring 1a.

  Further, a load sensor 20 is attached to a part surrounding the screw hole 19 for screwing the bolt 18 for connecting the fixed side flange 17 to the knuckle 16 at a plurality of positions on the inner side surface of the fixed side flange 17. Has been established. Each load sensor 20 is sandwiched between the outer side surface of the knuckle 16 and the inner side surface of the fixed-side flange 17 in a state where the outer ring 1 a is supported and fixed to the knuckle 16.

  In the case of the load measuring device of the rolling bearing unit of the second example having such a conventional structure, when an axial load is applied between a wheel (not shown) and the knuckle 16, the outer surface of the knuckle 16 and the fixed side flange 17 The inner surface strongly presses the load sensors 20 from both sides in the axial direction. Therefore, the axial load applied between the wheel and the knuckle 16 can be obtained by summing up the measured values of the load sensors 20. Although not shown, Patent Document 3 describes a method of obtaining the revolution speed of the rolling element from the vibration frequency of a member corresponding to an outer ring having a reduced rigidity, and measuring the axial load applied to the rolling bearing. ing.

  In the case of the first example of the conventional structure shown in FIG. 17 described above, the radial load applied to the rolling bearing unit is measured by measuring the displacement in the radial direction between the outer ring 1 and the hub 2 by the displacement sensor 11. However, since the displacement amount in the radial direction is small, it is necessary to use a highly accurate displacement sensor 11 in order to obtain this load with high accuracy. Since high-precision non-contact sensors are expensive, it is inevitable that the cost of the entire rolling bearing unit with a load measuring device increases.

  In the case of the second example of the conventional structure shown in FIG. 18 described above, it is necessary to provide as many load sensors 20 as bolts 18 for supporting and fixing the outer ring 1a to the knuckle 16. For this reason, coupled with the fact that the load sensor 20 itself is expensive, it is inevitable that the cost of the entire load measuring device of the rolling bearing unit is considerably increased. Further, the method described in Patent Document 3 requires that the rigidity of a part of the outer ring equivalent member be lowered, and it may be difficult to ensure the durability of the outer ring equivalent member.

  In view of such circumstances, the present inventors have previously described this rolling bearing unit based on the revolution speed of a pair of rolling elements (balls) constituting a rolling bearing unit which is a double row angular ball bearing. An invention relating to a load measuring device for a rolling bearing unit that measures a radial load or an axial load applied to the bearing is performed (Japanese Patent Application Nos. 2003-171715 and 172483). In the case of the load measuring device of the rolling bearing unit according to the present invention, in order to obtain the revolution speed of the rolling elements in each row, it is possible to detect the rotation speed of the cage holding the rolling elements in each row. Is effective in terms of obtaining high resolution. However, between the rolling surfaces of the rolling elements in each row and the inner surfaces of the pockets of the cages in each row, the rolling elements are allowed to roll and to the rolling surfaces of these rolling elements. Since there is a gap (pocket gap) for allowing the grease to adhere, the cages may rotate while being displaced in the radial direction as they rotate. When such a whirling occurs, the revolution center of the rolling elements in each row and the rotation center of the cage in each row shift, and the rotational speed of the cage in each row, and thus each row It becomes impossible to accurately measure the revolution speed of the rolling element.

The cages are not only displaced at the frequency of the revolving primary component based on the pocket gaps, but also the diameter differences of the rolling elements 9a, 9b, the outer ring raceways 7, 7 and the inner ring raceways. Based on the undulations of 8 and 8 and the like, the displacement is made even at higher frequencies. The displacement based on such a diameter difference or the like is a vibration phenomenon called 1/2 fractional vibration, where n is a positive integer, and Z is a rolling element existing between the inner ring raceway 8 and the outer ring raceway 7 facing each other. 9a, the number of 9a (or 9b, 9b), when the f _c and revolution frequency of the rolling elements 9a, 9a (or 9b, 9b), represented by (n / 2) · Z · f c Frequency component. For example, when the number Z of the rolling elements 9a, 9a (or 9b, 9b) is nine, the cage is diametrically oriented at a frequency of 4.5th revolution and 9th revolution based on the vibration phenomenon. It is displaced to. The radial displacement based on such a vibration phenomenon is smaller than the displacement of the revolution primary component based on the pocket gap and has a high frequency. For example, when ensuring the stability of the operation of the moving object at a high level, Is a problem. That is, unless the revolution speed error based on the radial displacement is corrected, it is not possible to secure the stability of the operation at a high level. In addition, for example, the radial displacement at the frequency of the revolution 4.5th order is a higher frequency than the revolution primary component, but is still a low frequency. Therefore, when it is removed by a low-pass filter, the response cannot be ignored. There is a delay.

Among the above-mentioned causes, in order to prevent deterioration in accuracy of rotational speed detection based on the swing of the revolution primary component, detection by a pair of rotation detection sensors disposed at two positions on the opposite side in the radial direction of the encoder By adding the signals together, it is possible to eliminate the influence of the deviation between the two centers. However, in this case, two rotation detection sensors are required, which may increase the cost and installation space, and may be difficult to adopt. Further, in such a method, according to the vibration phenomenon, (n / 2) · Z · f it is impossible to correct the error based on the frequency component represented _c. In

  As a technique for removing a relatively low-frequency noise component, an adaptive filter that is described in Non-Patent Document 1 and operates according to an LMS algorithm is known. Further, the outline of the adaptive filter is conventionally known in Non-Patent Documents 2 to 4 and the like. A synchronous adaptive filter, which is a kind of adaptive filter, has been conventionally known, for example, as described in Non-Patent Document 5. Further, a technique for suppressing engine vibration by a synchronous LMS algorithm has been conventionally known, for example, as described in Non-Patent Document 6. Conventionally, however, the adaptive filter as described above mainly uses so-called active noise control that reduces low-frequency noise by emitting sound waves having a phase opposite to that of low-frequency noise. That is, conventionally, the adaptive filter is used to reduce low-frequency noise that enters the room from the duct of the air conditioner, or low-frequency exhaust or traveling sound that enters the passenger car's room, and further enters from outside the headphones. It was only used to reduce low-frequency noise, such as reducing external frequency noise. The technique described in Non-Patent Document 6 is also intended to suppress engine vibration. In other words, the adaptive filter technique known from the past, as described in Non-Patent Document 1 above, can improve the accuracy of rotational speed detection using this encoder, regardless of the whirling motion of the encoder. , Was not considered at all.

  A notch filter is also known as a technique for removing a relatively low-frequency noise component without causing a large response delay. However, the use of such a notch filter for removing the noise component contained in the sensor signal for detecting the rotational speed of the cage rotating with the revolution of the rolling element has not been considered conventionally. Even if the notch filter known from No. 1 is used for obtaining the revolution speed of the rolling element as it is, data processing cannot be performed.

JP 2001-21577 A Japanese Patent Laid-Open No. 3-209016 Japanese Patent Publication No.62-3365 Haruo Hamada, “Basics of Adaptive Filter (Part 2)”, Journal of the Acoustical Society of Japan, Vol. 45, No. 9, The Acoustical Society of Japan, 1989, p. 731-738 Chuo University, Department of Electrical, Electronic, Information and Communication Engineering, “Adaptive Filter”, “online”, [searched on August 29, 2003], Internet, <URL: http: //www.elect.chuo-u. ac.jp/chao/forB3/dsp/volterra/filter.html> The MathWorks, Inc., "Adaptive Filter Overview and Applications", [online], [Search August 29, 2003], Internet, <URL: http://www.mathworks.ch/access/helpdesk/jhelp /toolbox/filterdesign/adaptiv2.shtml> The MathWorks, Inc., “Example of Adaptive Filter Using LMS Algorithm”, [online], [searched on August 29, 2003], Internet, <URL: http://www.mathworks.ch/access/ helpdesk / jhelp / toolbox / filterdesign / adaptiv9.shtml> Haruo Hamada, 3 others, “Synchronous adaptive filter and its application to active noise / vibration control”, Proceedings of the Acoustical Society of Japan, 3-5-13, Acoustical Society of Japan, March 1992, p. 515-516 Shigeki Sato, 4 others, “Development of Active Mount”, Automotive Technology, Japan Society for Automotive Technology, Vol. 53, no. 2, February 1999, p. 62-66

  In view of the circumstances as described above, the present invention conforms to the revolution speed of a plurality of rolling elements by using a fluctuation removal filter technology for removing the influence of fluctuations in detection signals, such as an adaptive filter and a notch filter. It is intended to be used for removing noise components contained in the detection signal of the rotation detection sensor (revolution speed detection sensor) for detecting the rotation speed of the cage. At the same time, it is intended to ensure that data processing can be performed without adopting a complicated configuration by making the relationship between the number of characteristic changes of the encoder and the number of rolling elements appropriate. These make it possible to configure the rolling element at low cost, without causing problems in durability and installation space, and measuring the rotational speed of the rotating member while ensuring the accuracy required for control. A rolling bearing unit with a revolution speed detecting device is realized.

A rolling bearing unit with a revolution speed detection device for a rolling element according to the present invention includes an inner ring equivalent member, an outer ring equivalent member, a plurality of rolling elements, an encoder, a rotation detection sensor, and a calculator.
Of these, the inner ring equivalent member has an inner ring raceway on the outer peripheral surface.
The outer ring equivalent member has an outer ring raceway on the inner peripheral surface.
Each rolling element is provided between the outer ring raceway and the inner ring raceway so as to roll freely.
The encoder is supported and fixed concentrically with the retainer on a part of the retainer that retains the rolling elements, and rotates together with the retainer. The characteristics are alternately changed in the circumferential direction. .
Further, the rotation detection sensor is provided in a state where the detection portion faces the detection surface of the encoder.
The computing unit calculates a rotational speed of the rotating wheel based on a signal sent from the rotation detection sensor.
The computing unit includes a fluctuation removal filter for removing the influence of fluctuations in the detection signal of the rotation detection sensor.
Further, the number of times that the characteristic per rotation of the encoder changes is an integral multiple of the number of rolling elements existing between the inner ring raceway and the outer ring raceway facing each other.

The rolling bearing unit with the revolution speed detection device of the rolling element of the present invention configured as described above is held based on the pocket clearance or the diameter difference of each rolling element, and further, the undulation of the inner ring raceway and the outer ring raceway. vibration or rotation to a vessel (revolution) primary component, (n / 2) · vibration frequency component represented by Z · f _c is generated, varies during the detection signal of the rotation sensor on the basis of these vibrations Even if this occurs, this variation can be canceled. For this reason, it is possible to accurately grasp various states based on the rotational speed of the cage, and thus based on the rotational speed, and perform a prompt and appropriate treatment.

When the present invention is implemented, preferably, as described in claim 2, the number of times the characteristic per one revolution of the encoder changes is determined by the number of rolling elements existing between the inner ring raceway and the outer ring raceway facing each other. Is an even multiple of.
If it does in this way, the vibration of the frequency of the order (n * Z / 2 order) proportional to 1/2 of the number (Z) of rolling elements can be removed using a notch filter.

Preferably, when carrying out the present invention, preferably, as described in claim 3, the encoder includes a permanent magnet having S poles and N poles alternately arranged at equal intervals in the circumferential direction on the surface to be detected. To do. Then, the sum of the S pole and the N pole per circumference of the detected surface is set to an integer multiple of the number of rolling elements existing between the inner ring raceway and the outer ring raceway facing each other.
If an encoder made of a permanent magnet is used, a rotation detection sensor having a simple structure can be used, and the rotation speed of the cage can be obtained accurately from the time of low rotation.

Further, when implementing the present invention, preferably, as described in claim 4, the fluctuation removing filter is an adaptive filter. When n is a positive integer, Z is the number of rolling elements existing between the inner ring raceway and the outer ring raceway facing each other, and f _c is the revolution frequency of each rolling element, the adaptive filter is has among the signals representing the revolution speeds of the rolling elements, variation of the revolution primary frequency component, and its ability to reject variations in frequency component represented by (n / 2) · Z · f c Shall.
Using an adaptive filter having such function, for generating the cage, the rolling elements based on the diameter difference, etc., (n / 2) · Z · f c by the rotation detection based on the oscillation frequency component represented Sensor output fluctuations can be eliminated. Of course, the adaptive filter can also suppress fluctuations in the output of the rotation detection sensor based on the vibration of the rotation (revolution) primary component based on the presence of the pocket gap.

In this case, preferably, as described in claim 5, the adaptive filter is operated by an LMS algorithm .

Further, when the present invention is implemented, preferably, the fluctuation removal filter is a fixed-order notch filter as described in claim 6 .
When such a notch filter is used, there is no time delay at all based on the process for removing the output fluctuation of the rotation detection sensor based on the low frequency vibration as in the case of using the above adaptive filter. Although this is not possible, this time delay can be minimized to such an extent that no problems arise due to various controls.
For this reason, various practical control devices can be configured at a relatively low cost.

In carrying out the invention according to claim 6 as described above, preferably, as described in claim 7 , n is a positive integer and Z exists between the inner ring raceway and the outer ring raceway facing each other. rolling the number of the moving object, when the f _c was revolution frequency of the rolling elements, notch filter, from among the signals representing the revolution speeds of the rolling elements, revolving primary variation in the frequency components, and, (n / 2) · a Z · variation of the frequency component represented by f _c and has a function of removing.
Using notch filter having such function, for generating the cage, the rolling diameter difference and the inner ring of the moving object, based on the undulation of the outer ring each orbit, represented by (n / 2) · Z · f c The output fluctuation of the rotation detection sensor based on the vibration of the frequency component can be removed. Of course, the notch filter suppresses fluctuations in the output of the rotation detection sensor based on the vibration of the primary rotation component based on the presence of the pocket gap.

Further, preferably when practicing the present invention, as set forth in claim 8, using a rolling bearing unit, in order to rotatably support the wheel suspension system. That is, the wheel is supported and fixed to one of the outer ring equivalent member and the inner ring equivalent member (rotating wheel), and the other race ring (stationary ring) is fixed to the suspension device.
Preferably, as described in claim 9 , a plurality of rotation detection sensors are provided. The plurality of rotation detection sensors measure the rotation speeds of a plurality of rotating portions generated during operation of a single rolling bearing unit, including the rotation speed of at least one cage. And the signal showing the rotational speed of the several places which the said several rotation detection sensor detects is utilized in order to obtain | require the load added between an inner ring equivalent member and an outer ring equivalent member.
If the present invention is implemented with such a structure, the load applied to the wheel can be obtained and used for control for ensuring the running stability of the vehicle.

  1 to 5 show Example 1 of the present invention. This embodiment is a rolling bearing that measures the load (radial load and axial load) applied to a rolling bearing unit for supporting the driven wheels of an automobile (FR wheel, RR wheel, front wheel of MD vehicle, rear wheel of FF vehicle). A case where the present invention is applied to a unit load measuring apparatus is shown. Since the configuration and operation of the rolling bearing unit itself are the same as those of the conventional structure shown in FIG. 17 described above, the same parts are denoted by the same reference numerals and redundant description is omitted or simplified. The description will focus on the characteristic part.

Formed on the outer peripheral surface of the hub 2 that is a rotating wheel, each formed on the inner peripheral surface of the double-row angular inner ring races 8 and 8 that are rotating side tracks, and the outer ring 1 that is a stationary ring, respectively. The rolling elements (balls) 9a and 9b are divided into double rows (two rows) between the double row angular outer ring races 7 and 7, which are tracks, and a plurality of cages 21a are provided for each row. The hub 2 is rotatably supported on the inner diameter side of the outer ring 1 by being provided so as to be able to roll while being held by the spring 21b. In this state, the rolling elements 9a and 9b in each row are provided with contact angles α _a and α _b (see FIG. 2) in opposite directions and of the same size, so Configures an angular ball bearing. Sufficient preload is applied to the rolling elements 9a and 9b in each row so as not to be lost due to an axial load applied during use. When such a rolling bearing unit is used, the outer ring 1 is supported and fixed to a suspension device, and a braking disk and a wheel of a wheel are supported and fixed to the rotation side flange 3 of the hub 2.

  A mounting hole 10a is formed in the axially intermediate portion of the outer ring 1 constituting the rolling bearing unit as described above between the double row outer ring raceways 7 and 7 so as to penetrate the outer ring 1 in the radial direction. ing. The sensor unit 22 is inserted into the mounting hole 10 a from the radially outer side of the outer ring 1 to the inner side, and the tip 23 of the sensor unit 22 is projected from the inner peripheral surface of the outer ring 1. The distal end portion 23 is provided with a pair of revolution speed detection sensors 24a and 24b, each of which is a rotation detection sensor, and one rotation speed detection sensor 15a.

  Among these, the revolution speed detection sensors 24a and 24b are for measuring the revolution speed of the rolling elements 9a and 9b arranged in the double row, and the axial direction of the hub 2 in the tip portion 23. Each detection surface is arrange | positioned on the both side surfaces regarding (the left-right direction of FIGS. 1-2). In this embodiment, the revolution speed detection sensors 24a and 24b detect the revolution speeds of the rolling elements 9a and 9b arranged in the double row as the rotation speeds of the cages 21a and 21b. For this reason, in the case of the present embodiment, the rim portions 25, 25 constituting the retainers 21a, 21b are arranged on the sides facing each other. Then, revolving speed detection encoders 26a and 26b each having a ring shape are attached and supported on the surfaces of the rim portions 25 and 25 facing each other over the entire circumference. The characteristics of the detected surfaces of the encoders 26a and 26b are changed alternately and at equal intervals in the circumferential direction, and the rotational speeds of the cages 21a and 21b are changed by the revolution speed detection sensors 24a and 24b. Detectable.

  For this purpose, the detection surfaces of the revolution speed detection sensors 24a and 24b are made to face each other and face to face, which are the detection surfaces of the revolution speed detection encoders 26a and 26b. The distances (detection gaps) between the detected surfaces of the revolution speed detection encoders 26a and 26b and the detection surfaces of the revolution speed detection sensors 24a and 24b are the inner surfaces of the pockets of the retainers 21a and 21b. And larger than a pocket gap, which is a gap between the rolling elements 9a and 9b, and preferably 2 mm or less. If the detection gap is equal to or less than the pocket gap, it is not preferable because the detected surface and the detection surface may rub against each other when the cages 21a and 21b are displaced by the pocket gap. On the contrary, if the detection gap exceeds 2 mm, it becomes difficult to accurately measure the rotations of the revolution speed detecting encoders 26a and 26b by the revolution speed detecting sensors 24a and 24b.

  On the other hand, the rotational speed detection sensor 15a is for measuring the rotational speed of the hub 2, which is a rotating wheel, and is provided on the distal end surface of the distal end portion 23, that is, on the radially inner end surface of the outer ring 1. The detection surface is arranged. A cylindrical rotational speed detecting encoder 13a is externally fitted and fixed between the double-row inner ring raceways 8 and 8 at the intermediate portion of the hub 2. The detection surface of the rotational speed detection sensor 15a is opposed to the outer peripheral surface, which is the detected surface of the rotational speed detection encoder 13a. The characteristics of the surface to be detected of the rotational speed detecting encoder 13a are changed alternately and at equal intervals in the circumferential direction so that the rotational speed of the hub 2 can be detected by the rotational speed detecting sensor 15a. The measurement gap between the outer peripheral surface of the rotational speed detection encoder 13a and the detection surface of the rotational speed detection sensor 15a is also suppressed to 2 mm or less.

  As the encoders 26a, 26b, and 13a, those of various structures that have been conventionally used for detecting the rotational speed of the wheels can be used in order to obtain ABS and TCS control signals. For example, as the encoders 26a, 26b, and 13a, multipole magnets in which N poles and S poles are alternately arranged at equal intervals on the detection surface (side surface or outer peripheral surface) can be preferably used. However, a simple encoder made of a magnetic material or one whose optical characteristics are changed alternately and at equal intervals in the circumferential direction can be used (in combination with an optical rotation speed detection sensor). is there.

  In the case of the present embodiment, as each of the revolution speed detecting encoders 26a and 26b, an annular permanent magnet in which S poles and N poles are alternately arranged at equal intervals on the side surface in the axial direction as the detected surface. Is used. Such revolving speed detecting encoders 26a and 26b are bonded and fixed to the side surfaces of the rim portions 25 and 25 of the holders 21a and 21b, which are separately manufactured, or these holders 21a and 21b. When injection molding is performed, insert molding is performed by setting each of the revolution speed detecting encoders 26a and 26b in the cavity. Which method is adopted is selected according to the cost, the required bond strength, and the like.

  In addition, as each of the revolution speed detection sensors 24a and 24b and the rotation speed detection sensor 15a, which are sensors for detecting the rotation speed, magnetic rotation speed detection sensors can be preferably used. As the magnetic rotational speed detection sensor, an active sensor incorporating a magnetic detection element such as a Hall element, Hall IC, magnetoresistive element (MR element, GMR element), or MI element can be preferably used. . In order to construct an active type rotational speed detection sensor incorporating such a magnetic detection element, for example, one side surface of the magnetic detection element is directly or via a stator made of a magnetic material, the magnetization direction of the permanent magnet. Abut against one end surface (when using a magnetic material encoder), and connect the other side surface of the magnetic detection element directly or via a magnetic material stator to the detected surface of each encoder 26a, 26b, 13a. Make them face each other. In this embodiment, a permanent magnet encoder is used, so that no permanent magnet on the sensor side is necessary.

In the case of the load measuring device of the rolling bearing unit of the present embodiment, the detection signals of the sensors 24a, 24b and 15a are input to a calculator (not shown). Based on the detection signals sent from the sensors 24a, 24b, and 15a, the computing unit applies one or both of the radial load and the axial load applied between the outer ring 1 and the hub 2. Is calculated. For example, when the radial load is obtained, the computing unit obtains the sum of the revolution speeds of the rolling elements 9a and 9b in each row detected by the revolution speed detection sensors 24a and 24b, and the sum and the rotation speed. The radial load is calculated based on the ratio to the rotational speed of the hub 2 detected by the detection sensor 15a. The axial load is obtained by calculating the difference between the revolution speeds of the rolling elements 9a and 9b in each row detected by the revolution speed detection sensors 24a and 24b, and the difference and the rotational speed detection sensor 15a are detected. Calculation is based on the ratio to the rotational speed of the hub 2. This point will be described with reference to FIG. In the following description, it is assumed that the contact angles α _a and α _b of the rolling elements 9 a and 9 b in each row are the same in a state where the axial load F _a is not applied.

FIG. 4 schematically shows the rolling bearing unit for supporting the wheel shown in FIG. 1 and shows the action state of the load. Preloads F ₀ and F ₀ are applied to the rolling elements 9 a and 9 b arranged in a double row between the double row inner ring raceways 8 and 8 and the double row outer ring raceways 7 and 7. Further, a radial load F _r is applied to the rolling bearing unit during use due to the weight of the vehicle body or the like. Further, by the centrifugal force or the like applied during cornering, applied is the axial load F _a. These preloads F ₀ , F ₀ , radial load F _r , and axial load F _a all affect the contact angles α (α _a , α _b ) of the rolling elements 9a, 9b. Then, the contact angle alpha _a, the alpha _b is changed, these rolling elements 9a, the revolution speed n _c of 9b changes. The diameter of the pitch circle of each of these rolling elements 9a, 9b is D, the diameter of each of these rolling elements 9a, 9b is d, the rotational speed of the hub 2 provided with each of the inner ring raceways 8, 8 is n _i , When the rotational speed of the outer race 1 provided with the outer ring raceway 7, 7 and n _o, the revolution speed n _c is expressed by the following equation (1).
n _c = {1− (d · cos α / D) · (n _i / 2)} + {1+ (d · cos α / D) · (n _o / 2)} --- (1)

As is clear from this equation (1), the rolling elements 9a, the revolution speed n _c of 9b, these rolling elements 9a, the contact angle α (α _a, α _b) of 9b varies in response to changes in The contact angles α _a and α _b vary according to the radial load F _r and the axial load F _a . Thus the revolution speed n _c is changed according to these radial load F _r and axial load F _a. In this embodiment, the hub 2 is rotated, since the outer ring 1 is not rotated, specifically, with respect to the radial load F _r, the revolution speed n _c is slow enough to increase. Further, with respect to the axial load F _a, the revolution speed n _c columns for supporting the axial load F _a is faster, the revolution speed n _c of the columns that do not support this axial load F _a is delayed. Therefore, the radial load F _r and the axial load F _a can be obtained based on the revolution speed n _c .

However, the contact angle α which leads to a change in the revolution speed n _c, as well as the radial load F _r and the axial load F _a is changed while associated with each other, also varies the preload F _0, F ₀ To do. Also, the revolution speed n _c is changed in proportion to the rotational speed n _i of the hub 2. Therefore, if the radial load F _r , the axial load F _a , the preload F ₀ , F ₀ , and the rotational speed n _i of the hub 2 are not all considered, the desired load can be accurately _calculated from the revolution speed n _c. I can't ask for it. Of these, the preloads F ₀ and F ₀ do not change according to the operating state, so it is easy to eliminate the influence by initial setting or the like. On the other hand, the radial load F _r , the axial load F _a , and the rotational speed n _i of the hub 2 constantly change according to the operating state, so that the influence cannot be eliminated by initial setting or the like.

In view of such circumstances, in this embodiment, as described above, when the radial load F _r is obtained, the revolutions of the rolling elements 9a, 9b of the respective rows detected by the respective revolution speed detection sensors 24a, 24b are detected. by obtaining the sum of the rates, and reduce the influence of the axial load F _a. Further, when determining the axial load F _a , the influence of the radial load F _r is reduced by determining the difference in revolution speed between the rolling elements 9 a and 9 b in each row. Further calculation, in any case, and the sum or difference, the ratio to the radial load F _r or the axial load F _a on the basis of the rotational speed n _i of the hub 2 to the rotational speed detecting sensor 15a detects by and by eliminating the influence of the rotational speed n _i of the hub 2.

There are various other methods for calculating one or both of the radial load F _r and the axial load F _a based on the signals of the revolution speed detection sensors 24a and 24b. Such a method is described in detail in the above-mentioned Japanese Patent Application Nos. 2003-171715 and 172484, and is not related to the gist of the present invention.
However, even if the seek any load by any method, is accurately determined the revolution speed n _c of the rolling elements 9a, 9b of each column based on the respective revolution speed detecting sensor 24a, 24b detect signal This is important for increasing the load measurement accuracy.

On the other hand, the detection signals of the revolution speed detection sensors 24a and 24b are based on an error in the magnetization pitch of the surface to be detected (the pitch between the S pole and the N pole adjacent in the circumferential direction). Relatively high frequency fluctuations, relatively low frequency fluctuations as described above due to the swinging motion of the cages 21a and 21b, and the diameter difference between the rolling elements 9a and 9b, or the rolling elements 9a, The fluctuation of the medium frequency based on the undulations of the outer ring raceways 7 and 7 and the inner ring raceways 8 and 8 with which the rolling contact surface 9b comes into rolling contact. Without handle such variation (reduction), the rolling elements 9a of the column, not be accurately determined the revolution speed n _c of 9b, the measurement accuracy of the radial load F _r and the axial load F _a is deteriorated. Therefore, in the case of the present embodiment, the adaptive filter 27 as shown in FIG. 5 reduces the relatively low frequency fluctuation based on the whirling motion and the moderate frequency fluctuation based on the diameter difference and the waviness of the raceway surface. In addition to reduction, a low-pass filter such as an averaging filter (not shown) reduces relatively high-frequency fluctuations based on the magnetization pitch error.

First, the reason why the above three types of fluctuations occur will be described with reference to FIGS. The inner surface of the pocket of the holder 21a (21b) holding the revolution speed detecting encoder 26a (26b) (or having the function as an encoder) and the rolling surface of each rolling element 9a (9b) There is a gap between these rolling elements 9a (9b) so as to be able to roll freely. Therefore, no matter how much the assembly accuracy of each component is increased, the center of the pitch circle (rotation center of the hub 2) O _{2 of the} rolling elements 9a (9b) and the cage 21a ( There is a possibility that the rotation center O _{21 of} 21b) is shifted by δ as shown in an exaggerated manner in FIG. Then, the revolution speed detecting encoder 26a on the basis of the deviation (26b) performs a rotation motion blur around the rotational center O _21. As a result of the swinging motion, the detected surface of the revolution speed detecting encoder 26a (26b) has a moving speed in addition to the rotating direction. Then, the moving speed other than the rotational direction, for example, the lateral moving speed in FIG. 6 is added to or subtracted from the rotational speed. On the other hand, the revolution speed detection sensor 24a (24b) detects the revolution speed of each rolling element 9a (9b) based on the moving speed of the detected surface of the revolution speed detection encoder 26a (26b). The eccentricity of δ affects the detection signal of the revolution speed detection sensor 24a (24b) in which the detection surface faces the side surface of the revolution speed detection encoder 26a (26b).

When the detection surface of the revolution speed detection sensor 24a (24b) faces the side surface of the revolution speed detection encoder 26a (26b), the detection signal of the revolution speed detection sensor 24a (24b) is as shown in FIG. As shown by the chain line α in FIG. That is, even when the revolution speed of each rolling element 9a (9b) is constant, the revolution speed represented by the output signal of the revolution speed detection sensor 24a (24b) is sinusoidally as shown by the chain line α. Change. Specifically, when the moving speed in the left-right direction in FIG. 6 is added to the moving speed in the rotation direction, the output signal is a signal corresponding to a speed faster than the actual revolution speed. On the other hand, when the moving speed in the left-right direction in FIG. 6 is subtracted from the moving speed in the rotational direction, the output signal is a signal corresponding to a speed slower than the actual revolution speed. FIG. 6 shows the eccentricity δ exaggerated as compared with the actual case. For example, the radial load F _r and the axial load F _a applied to the rolling bearing unit are further increased in order to more strictly control for vehicle stability. In the case of obtaining accurately, it is necessary to eliminate the error due to the eccentricity.

The so-called bearing vibration of the (n / 2) Z order frequency based on the undulations of the rolling elements 9a and 9b, the outer ring races 7 and 7 and the inner ring races 8 and 8 is also the revolution speed detecting sensor 24a ( It appears as the fluctuation of the revolution speed represented by the output signal of 24b). Therefore, in order to more accurately determine the loads F _r and F _a in order to perform the control for stabilizing the vehicle more precisely, it is necessary to eliminate the error associated with the bearing vibration. Of the fluctuations due to the bearing vibration, the fluctuations in the revolution speed represented by the output signal based on the bearing vibration having a frequency where n is large (for example, n is 3 or more) generally have a small amplitude. Not only is the influence on the fluctuation small, but even if it is removed by a low-pass filter, there is no time delay that causes a practical problem. On the other hand, when a fluctuation corresponding to n of 1 or 2 is to be removed by a low-pass filter (especially when n = 1), there is a possibility of causing a time delay that causes a practical problem. .

  Furthermore, the pitches of the S poles and N poles arranged on the side surface of the revolution speed detecting encoder 26a (26b) should be essentially the same, but are little by little due to magnetization errors that occur during manufacturing. May be different from each other. Based on this error, the detection signal of the revolution speed detection sensor 24a (24b) varies. The period of fluctuation based on such an error in the magnetizing pitch is much shorter than the period of fluctuation based on the above-mentioned swing motion. For example, when the characteristic (repetition of S pole and N pole) of the side surface (detected surface) of the revolution speed detecting encoder 26a (26b) changes 60 times on the entire circumference of the detected surface, the magnetization is performed. The fluctuation period based on the pitch error is about 1/60 of the fluctuation period based on the swing motion.

The detection signal output from the revolution speed detection encoder 26a (26b) is as shown by a solid line β in FIG. 7 in which the above three types of fluctuations are added (superimposed). Incidentally, there are various vibration frequencies based on the bearing vibration in accordance with the value of n, but the frequency when the value of n is small is the revolution primary corresponding to the eccentricity δ shown by the broken line α. It is higher than the frequency and lower than the frequency of fluctuation based on the magnetization pitch error. In order to accurately obtain the radial load F _r and the axial load F _a , it is necessary to reduce the above three types of fluctuations. Therefore, in the case of the present embodiment, the relatively low frequency fluctuations accompanying the whirling motion and the bearing vibration are reduced by the adaptive filter 27 shown in FIG. High-frequency fluctuations are reduced by a low-pass filter such as an averaging filter (not shown). As the adaptive algorithm, an LMS (least mean square) algorithm (an operation rule for minimizing the mean square error based on the steepest descent method) using an FIR filter described later as the adaptive filter is preferable.

First, the low frequency fluctuation reduction by the adaptive filter 27 shown in FIG. 5 will be described. The displacement speed of the revolution speed detection encoder 26a (26b) at the part where the detection part of the revolution speed detection sensor 24a (24b) faces is based on the actual rotational speed d _d and the eccentricity of δ. The primary rotation component due to the swirl and the apparent speed fluctuation d _n based on the bearing vibration are superimposed. Therefore, the output signal d of the revolution speed detecting sensor 24a (24b) is a signal representing the speed (d _d + d _n ) obtained by adding the actual rotational speed d _d and the variation d _n . By subtracting the variation amount d _n from the output signal d by the adaptive filter 27 (Genzure In) will be asked to the actual speed d _d.

On the other hand, in order to operate the adaptive filter 27, the whirling and the reference signal x having a correlation with the variation d _n based on the bearing vibration is required. If this reference signal x can be obtained, the adaptive filter 27 performs self-learning and has the same characteristic as the transfer characteristic of the actual signal flow “d _n → d”. The FIR (finite impulse response) filter (impulse response time) A finite filter = a filter whose impulse response becomes 0 within a finite time). Then, if the cancel signal y {= y (k)} to be described later] obtained as a result of the calculation by the adaptive filter 27 is subtracted from the output signal d of the revolution speed detection sensor 24a (24b), the revolution speed detection is performed. This is equivalent to (d−d _n ) in which the fluctuation d _n based on the swing and the bearing vibration is removed from the output signal d of the sensor 24a (24b). In this way, when removing this variation d _n , the adaptive filter 27 does not filter the main route (the upper half of FIG. 5) of the signal with respect to the output signal d to be sent, based on the reference signal x is fed to the lower half portion) of FIG. 5 calculates the cancellation signal y for removing the variation amount d _n. Since the cancel signal y is simply subtracted from the output signal d that is the main route, a response delay of the output signal d is not caused.

  In the case of the present embodiment, the reference signal x is determined based on the number of characteristic changes during one revolution of the revolution speed detection encoder 26a (26b) and the revolution speed opposed to the revolution speed detection encoder 26a (26b). It is self-generated by the processing circuit for the output signal of the speed detecting sensor 24a (24b) or the processing circuit for calculating the revolution speed of each rolling element 9a (9b) based on this output signal. Therefore, the cost required for generating the reference signal x can be reduced. That is, when the structure of active noise control, which has been generally known as an application of an adaptive filter, is applied as it is to the structure for accurately determining the revolution speed of each rolling element 9a (9b), the structure for detecting the revolution speed is used. The swing of the encoder 26a (26b) is detected by a separately provided sensor such as a displacement sensor or a rotation speed sensor, and the detection signal of this sensor is correlated with the fluctuation caused by the swing and bearing vibration, so that It is used as the reference signal x of the filter 27. Of course, although it is possible to carry out with such a structure, cost and installation space are required as much as a separate sensor is provided.

On the other hand, in this embodiment, the reference signal x is obtained without using the detection signal of such a separately provided sensor, and the revolving speed detection encoder 26a ( The fluctuation d _n of the output signal d of the revolution speed detection sensor 24a (24b) based on the swinging motion of the bearing 26b) and the bearing vibration is reduced. That is, the number of characteristic changes (number of S poles and N poles) during one revolution of the revolution speed detecting encoder 26a (26b) is known in advance. Therefore, by observing the number of pulses for one revolution of the revolution speed detecting encoder 26a (26b), there is a correlation with the fluctuation d _n even without providing a separate sensor such as a displacement sensor or a rotational speed sensor. The reference signal x can be generated. Specifically, the influence of the swing of the revolution speed detection encoder 26a (26b) is a waveform whose main component is the rotation primary. For example, the revolution speed detection encoder 26a (26b) has a waveform of 60 per revolution. In the case of a pulse, it can be self-generated as a sine wave, a triangular wave, a sawtooth wave, a rectangular wave, a pulse wave, etc. that makes one cycle with 60 data. Among the fluctuations d _n , the reference signal x for reducing the revolution (n / 2) · Z-order frequency component can be similarly set by appropriately setting the number of data in one cycle. Can self-generate.

Such a waveform of the reference signal x can be generated by a processing circuit (CPU) for calculating the revolution speed of each rolling element 9a (9b) or attached to the revolution speed detection sensor 24a (24b). It can also be generated by an electronic circuit unit (IC). Anyway, the resulting canceled signal y calculated based on the reference signal x, is subtracted from the output signal d of the revolution speed detecting sensor 24a (24b), modification representing the actual speed d _d A signal e {= e (k)} described later is obtained. The correction signal e obtained in this way is sent to the processing circuit for calculating the revolution speed of each rolling element 9a (9b) and used for obtaining the revolution speed, and the adaptive filter 27 is self-learning. Also used as information to do.

In the adaptive filter 27, the cancel signal y is obtained, and the cancel signal y is further subtracted from the output signal d of the revolution speed detection sensor 24a (24b) to obtain the correction signal e. , Based on the following equations (2) to (4).

In the above equations (2), (3), and (4), k is a data number in time series, and N is the number of taps of the FIR filter used as the adaptive filter 27. Further, w represents a filter coefficient of the FIR filter, w _k is a filter coefficient used when k-th data processing is performed, and w _{k + 1} is used when a next data series (k + 1-th) is processed. The filter coefficient i represents the number from 0 to N-1. In the case of the present embodiment, the FIR filter is an adaptive filter in which the filter coefficient is sequentially updated appropriately according to the above equation (4). The filter coefficient w _k used at the start of the calculation may be self-adapted if it begins to move even if zero is substituted. However, a desired filter characteristic is obtained in advance and the value is substituted. You can keep it. Furthermore, the filter coefficient last used in the previous process may be stored in a storage unit such as an EEPROM and used at the time of restart.

尚、この（５）式中のαも、フィルタ係数を自己適正化させていく為の更新量を決定するパラメータとなるが、０＜α＜１の範囲であれば良く、上記μよりも設定が容易である。又、本実施例の場合には、前記参照信号ｘを自己生成するので、上記（５）式中の分母の値は既知であり、μの最適値を事前に算出しておく事もできる。計算量削減の観点からは、予め（５）式でこのμを算出しておき、このμを定数として上記（４）式でフィルタ係数を自己適正化させるのが望ましい。 Also, μ in the above equation (4) is a value called a step parameter that determines the update amount when the filter coefficient is self-optimized, and is usually about 0.01 to 0.001. However, in practice, the validity of the adaptive operation can be examined and set in advance, or it can be updated sequentially using the following equation (5).

Note that α in the equation (5) is also a parameter for determining the update amount for self-optimizing the filter coefficient, but it may be in the range of 0 <α <1, and is set more than the above μ. Is easy. In this embodiment, since the reference signal x is self-generated, the denominator value in the above equation (5) is known, and the optimum value of μ can be calculated in advance. From the viewpoint of reducing the amount of calculation, it is desirable to calculate this μ in advance using equation (5), and to self-optimize the filter coefficient using equation (4) using this μ as a constant.

As described above, the output signal from d of the revolution speed detecting sensor 24a (24b), By subtracting the cancellation signal y, wherein the adaptive filter 27 is calculated to obtain the modified signal e representative of the actual rotational speed d _d It is done. And based on the correction signal e calculated | required in this way, the revolution speed of each said rolling element 9a (9b) can be calculated | required correctly. In an actual case, the output signal d of the revolution speed detection sensor 24a (24b) includes a fluctuation based on the swing speed of the revolution speed detection sensor 24a (24b) based on the magnetization pitch error. There are fluctuations with a shorter period. Therefore, a low-pass filter such as an averaging filter for averaging the fluctuations is provided before or after the adaptive filter 27, so that the revolution speeds of the rolling elements 9a and 9b can be accurately obtained regardless of the fluctuations. To be able to. Since the structure and operation of a low-pass filter such as an averaging filter are well known in the art, detailed description thereof is omitted.

FIG. 8 shows an example of a simulation on the action of suppressing the fluctuation based on the swing of the encoder using the adaptive filter 27. FIG. 8 shows a case where the rotational speed of the rotating member rotating at a constant speed of 100 min ⁻¹ is measured by a 60 pulse / 1 rotation encoder. The solid line A is the result (corresponding to the output signal d) obtained by performing only the moving average process with the number of taps = 15 (providing only the averaging filter) on the detection result of the rotation speed detection sensor. In this case, the calculated value of the rotational speed fluctuates between about 70 to 130 min ⁻¹ due to the swing of the encoder. Note that the amount of swirling of the encoder was set to be considerably larger than the actually generated value.

On the other hand, the broken line B shows the result (corresponding to the correction signal e) obtained by correcting the data after moving average, which is indicated by the solid line A, using an adaptive filter. As apparent from the broken line (b), although the calculated value fluctuates immediately after the start of the adaptive filter, the filter coefficient self-adapts after a short time, and the calculation result converges to a constant value of approximately 100 min- ¹ . For this reason, by using an averaging filter and an adaptive filter together, even if an encoder with a large magnetization pitch error or a large deviation between the rotation center and the geometric center (running motion) is used, It can be seen that the rotation speed can be obtained accurately.
In order to obtain the two lines A and B shown in FIG. 8, the reference signal x is a sine wave having one cycle of 60 pulses while counting the number of pulses in the speed calculation device. Suppose you are self-generating. The step parameters of the adaptive filter are μ = 0.002 and the number of taps N = 30.

  The above description has been given for the case of removing the error component based on the first rotation of the revolution based on the pocket gap. The error component based on the bearing vibration is also basically removed in the same manner. However, in order to perform data processing at low cost and smoothly, the revolution speed detecting encoder 26a (26b) corresponds to one rotation. The number of pulses per hit needs to be regulated in relation to the number of rolling elements 9a, 9a (9b, 9b) in each row. For example, when the number of rolling elements 9a, 9a (9b, 9b) in each row is 9, 60 data / 9 = 6.67 data is used to correct an error component based on the 9th-order vibration. Therefore, it is necessary to use a sine wave having one cycle as the reference signal x shown in FIG. In this way, when the reference signal x is not an integer (when a fraction after the decimal point appears), it is impossible to use the reference signal x previously stored in the memory in the computing unit. In such a case, every time data is sampled, that is, every time when the output signal d of the revolution speed detecting sensor 24a (24b) is received, a sine wave that becomes one cycle for each cycle of the bearing vibration. Must be calculated. Such calculation must be continuously performed while the revolution speed detection of the rolling elements 9a, 9a (9b, 9b) of each row is performed by the revolution speed detection sensor 24a (24b). It becomes a huge amount of calculation. In order to process such an enormous amount of calculation, it is necessary to use a relatively expensive CPU with a high processing speed.

  In view of such circumstances, in the case of the present embodiment, the number of pulses per revolution of the revolution speed detecting encoder 26a (26b) is set to the number of rolling elements 9a, 9a (9b, 9b) in each row. It is an integer multiple of the number Z. For example, as described above, when the number Z of rolling elements 9a, 9a (9b, 9b) in each row is 9, the number of pulses per rotation of the revolution speed detecting encoder 26a (26b) For example, 63 pulses (7 times 9). The 9th-order revolution error component based on the bearing vibration contained in the output signal d of the revolution speed detection sensor 24a (24b) for detecting the rotational speed of the revolution speed detection encoder 26a (26b) is obtained. In order to correct, it is only necessary to store a sine wave having one cycle of 7 data (= 63 data / 9) in the memory, so that the amount of calculation is drastically reduced. The same applies when correcting the 4.5th-order error component of revolution, and as a reference signal x shown in FIG. 5, a sine wave having one cycle of 14 data (= 63 data / 4.5) is stored in the memory. Just keep it.

  9 to 12 show Example 2 of the present invention. The feature of the present embodiment is that the number of calculation processes required for the detection signal of the rotation detection sensor for each pulse of the encoder is greatly reduced, and the calculation speed is not particularly fast. ) Is possible. For this reason, in the case of the present embodiment, the amount of calculation can be greatly reduced by using a synchronous LMS algorithm. However, even when the synchronous LMS algorithm is used, the number of pulses per rotation of the encoder needs to be an integral multiple of the number of rolling elements. If the synchronous LMS algorithm is simply used, the primary rotation component that is the encoder swing and the revolution (n / 2) / Z-order component based on the bearing vibration are corrected (cancelled) and simultaneously detected. Even the DC level representing the rotational speed is corrected (cancelled). In this case, since the original function of the revolution speed detecting device is lost, the zero point of the filter coefficient is monitored, and zero point correction is performed to prevent the cancellation of the DC level. The characteristics of the present embodiment considered from such a viewpoint will be described below. Although the level is not a problem in practical use, the DC level may be slightly deviated even in the example shown in FIG. Therefore, in order to perform more accurate control, it is preferable to perform zero point correction also in this case. In the following description, the reference numerals of the rolling bearing unit are shown in FIG.

  Each of the above formulas (2), (3), and (4) used for optimizing the adaptive filter in the first embodiment described above is a simple formula, but the amount of calculation is problematic in actual application. It may be possible. For example, if the number of taps of the adaptive filter is N = 60, the multiplication is performed 60 times by the above formula (2), the subtraction is performed by the above formula (3), the multiplication is 120 times by the above formula (4), and the addition is performed 60 times. 180 times, a total of 241 arithmetic operations must be performed for each pulse of the encoder. Therefore, the amount of calculation required to obtain the revolution speed of the double row rolling elements provided in one rolling bearing unit is 482 times / 1 pulse. This calculation amount (number of operations) is not physically unprocessable, but it is necessary to use a relatively expensive CPU with a high processing speed. For example, when detecting the rotational speed of automobile wheels (four wheels) for the control of a vehicle running stabilization device such as ABS, TCS, VSC, etc., four expensive CPUs (or every pulse) are used. It is necessary to use a CPU that is high enough to perform 241 times × 2 × 4 = 1920 times of four arithmetic operations, which is not preferable because it causes an increase in the cost of the vehicle travel stabilization device.

  In view of such circumstances, in the case of the present embodiment, it is intended to use a synchronous LMS algorithm to greatly reduce the amount of calculation and enable the use of a low-cost CPU. However, when the adaptive filter is operated by the synchronous LMS algorithm, the adaptive filter can be used only with the primary revolution component due to the swing of the encoder and the revolution (n / 2) / Z-order component based on bearing vibration. In addition, the DC component representing the rotation speed is also canceled. This phenomenon of canceling the DC component is remarkable when the synchronous LMS algorithm is used. Therefore, in the case of the present embodiment, the DC level representing the rotational speed can be accurately detected by providing a function for setting the output value of the adaptive filter to zero.

First, the operation principle of the synchronous LMS algorithm will be described. In the block diagram shown in FIG. 5 described above, the reference signal x input to the adaptive filter 27 is the primary revolution component of the encoder caused by the encoder swing, and the revolution (n / 2) / Z order component based on the bearing vibration. As long as the signal is correlated with Of these, the revolution primary component may be one impulse signal per one revolution of the encoder, and the revolution (n / 2) · Z order component may be one impulse signal per vibration period. However, when the synchronous LMS is used, it is assumed that the tap number N of the adaptive filter 27 is equal to the number of pulses per vibration period. It is assumed that the reference signal x is one impulse signal per vibration period and the number of taps N of the adaptive filter 27 is equal to the number of pulses per rotation of the encoder. In this case, the reference signal x used for calculation at the instant of time series k is expressed by the following equation (6).

In this equation (6), the position j at which the reference signal x becomes an impulse having the value 1 is shifted one by one to the right as the time series k advances, up to the “N−1” th position on the rightmost side. If shifted, a new impulse value appears at the leftmost 0th in the next time series. That is, the reference signal x is a data string obtained by circulating the position of the impulse having the value 1 from the 0th to the (N-1) th. When this equation (6) is applied to the aforementioned equations (2) and (4), the following equations (7) and (8) can be obtained.

  When the adaptive filter 27 is operated by a normal LMS algorithm that is not synchronous, it is necessary to repeat the calculations shown in the equations (2), (3), and (4) as described above. When the adaptive filter 27 is operated by the equation LMS algorithm, it is only necessary to perform the calculations shown in the equations (7), (8) and (3). For example, when the number of taps N of the adaptive filter 27 is 60, when the adaptive filter 27 is operated by a normal LMS algorithm, the total number of calculations for each encoder pitch is 241 as described above. On the other hand, when the adaptive filter 27 is operated by the synchronous LMS algorithm, the above formula (7) is not calculated only by data replacement, is subtracted once by the above formula (3), and is multiplied by the above formula (8). It is only necessary to perform four arithmetic operations in total, three times for each pulse of the encoder, two times, once and one addition. In other words, by adopting a synchronous method as the LMS algorithm, the number of operations can be reduced to approximately 1/80 compared to a case where the LMS algorithm is not employed.

  In this way, the amount of calculation can be reduced by adopting the synchronous LMS algorithm. However, as described above, calculation cannot be performed unless the number of taps N of the adaptive filter is equal to the number of pulses per vibration period. . Naturally, the number of taps N must be an integer. Therefore, if the number of pulses per vibration cycle is not an integer, the calculation by the synchronous LMS algorithm cannot be performed. Since the period of the primary revolution of the revolution based on the pocket gap coincides with the period of one revolution of the revolution speed detecting encoder 26a (26b), the pulses per one revolution of the revolution speed detecting encoder 26a (26b). There is no need to devise a number. On the other hand, with respect to the revolution (n / 2) / Zth order component based on the bearing vibration, the number of pulses per one rotation of the revolution speed detecting encoder 26a (26b) is set to each rolling element 9a. , 9a (9b, 9b), if not devised in relation to the number Z, the number of pulses per vibration cycle will not be an integer, and calculation by the synchronous LMS algorithm cannot be performed. Therefore, also in the present embodiment, as in the first embodiment, the number of pulses per revolution of the revolution speed detecting encoder 26a (26b) is set to the rolling elements 9a, 9a (9b) of each row. 9b) is an integer multiple of the number Z. Accordingly, the number of pulses per revolution of the revolution speed detecting encoder 26a (26b) is an integer even for the revolution (n / 2) / Z order component based on the bearing vibration.

Further, when the synchronous LMS algorithm is employed to operate the adaptive filter 27, the zero point of the adaptive filter 27 is set to prevent cancellation of even the DC component that is a signal representing the rotational speed. It is necessary to correct. The zero point correction will be described below. As a specific example of the phenomenon that requires this zero point correction, FIG. 9 shows an example of a rotational speed detection error caused by an encoder swing. The diagram shown in FIG. 9 is the same as the case of FIG. 8 described above when the rotational speed of a rotating member rotating at a constant speed of 100 min ⁻¹ is measured by a 60 pulse / 1 rotation encoder. It shows. A solid line A is a result (corresponding to the output signal d in FIG. 10) obtained by performing only the moving average process with the number of taps = 15 (providing only the averaging filter) on the detection result of the rotation speed detection sensor. In this case, the calculated value of the rotational speed fluctuates between about 70 to 130 min ⁻¹ due to the swing of the encoder. Note that the amount of swirling of the encoder is set to be considerably larger than the actually generated value.

When the measurement data relating to the rotational speed as indicated by the solid line A in FIG. 9 is processed using the adaptive filter 27 as shown in FIG. Depending on the set value of 27, in addition to the error component based on this swing, there is a possibility that the DC level of the rotational speed that is the detection target (a signal representing 100 min ^-1 indicated by a broken line in FIG. 9) is also canceled. There is. Thus, the phenomenon of canceling to the required DC level is remarkable when a synchronous system is adopted as the LMS algorithm for operating the adaptive filter 27. A chain line C shown in FIG. 9 is a specific example.

  When a synchronous system is adopted as the LMS algorithm for operating the adaptive filter 27 and no particular measures are taken, it represents not only the fluctuation component based on the swing of the encoder but also the rotational speed as indicated by the chain line c. Even the DC component is canceled and the output value becomes zero. This is a phenomenon that occurs because the filter coefficient w of the adaptive filter 27 has a DC level due to the adaptive operation, and as a result, the output signal y of the adaptive filter 27 has a DC level. In order to solve this problem, in this embodiment, as shown in FIG. 10, the DC level is calculated from the average value of the filter coefficient w, and this DC level is multiplied by the impulse value of the reference signal x. A DC signal is calculated (when the impulse value is 1, no multiplication is required). Then, by adding the DC signal (average value of w described below) calculated as described above to the signal e whose error has been canceled by the adaptive filter 27, a DC level representing an accurate rotational speed can be obtained. Like.

  Next, a method for calculating the DC level from the average value of the filter coefficient w will be described. By operating the adaptive filter 27 using the synchronous LMS algorithm, the error component included in the signal representing the rotational speed obtained from the output signal of the revolution speed detecting sensor 24a (24b) is canceled, and the chain line As shown in FIG. 11, the filter coefficient of the adaptive filter 27 when the output value becomes zero varies as shown in FIG. In the example shown in FIG. 9, since the number N of taps of the adaptive filter 27 is 60, the filter coefficient w shown in FIG. 11 is composed of 60 values. The average value of the filter coefficient w, that is, the DC level representing the rotational speed to be obtained can be obtained by summing up the 60 values and dividing by 60. However, if such a calculation is performed, the number of operations increases, and it is not possible to sufficiently reduce the cost of the CPU, which is the object of this embodiment.

  By the way, the object of error cancellation, that is, the undulation based on the shake of the encoder is a rotation n-order component mainly composed of the rotation first order. In this embodiment, since the tap number N of the adaptive filter 27 is equal to the number of pulses per encoder rotation, the filter coefficient w is a periodic function having a period of N (= 60). Become. In the example shown in FIG. 11, the rotation is a first-order periodic function. Therefore, the average value of any two points set with an interval of N / 2 (= 30) is equivalent to the average value of all N (= 60) points. Thus, if the average value of these two points is obtained and set to the DC level representing the rotational speed, the number of calculations can be greatly reduced, which is advantageous from the viewpoint of the cost reduction of the CPU. If reliability remains uncertain with the average of only two points, select any two points with an interval of N / 2 (= 30) apart from the above two points, and average the total of four points Is calculated. Although not shown, even when the filter coefficient w is a rotation n-th order periodic function, the average value is similarly set by appropriately increasing the number of points for obtaining the average point and appropriately setting the interval. Desired.

FIG. 12 shows an example of a simulation for the effect of suppressing the fluctuation based on the swing of the encoder by the structure of the present embodiment. FIG. 12 shows the case where the rotational speed of a rotating member rotating at a constant speed of 100 min ⁻¹ is measured by a 60 pulse / 1 rotation encoder. The solid line A is the result (corresponding to the output signal d) obtained by performing only the moving average process with the number of taps = 15 (providing only the averaging filter) on the detection result of the rotation speed detection sensor. In this case, the calculated value of the rotational speed fluctuates between about 70 to 130 min ⁻¹ due to the swing of the encoder. The broken line b shows the revolution speed detection sensor 24a (24b) by using the adaptive filter 27 that operates according to the above-mentioned synchronous LMS algorithm shown in FIG. 10 and correcting the DC component by the filter coefficient w described above. This is a result of canceling the error component included in the signal representing the rotation speed obtained from the output signal d. As is apparent from the broken line (b), although the data fluctuates immediately after the adaptive filter 27 is started, the filter coefficient w is self-adapted after a short time, and the calculation result converges to a constant value of about 100 min ^−1. did. It should be noted that the speed detection error based on the bearing vibration can be processed in the same manner to prevent the DC component from being canceled.

  FIGS. 13 to 15 show a case where a notch filter is used as a fluctuation removal filter as a third embodiment of the present invention. If an adaptive filter that operates with the LMS algorithm is used as in the first embodiment, the amount of calculation increases and the cost of the arithmetic unit increases. If the synchronous LMS algorithm is used as in the second embodiment described above, the amount of calculation can be greatly reduced. However, when the synchronous LMS algorithm is used, the revolution (n / 2) If different fluctuation components exist in the vicinity of the Z-order component, this revolution (n / 2) .Z-order component may not be sufficiently removed.

  In this embodiment, in view of such circumstances, by using a notch filter as a fluctuation removal filter, it is possible to reliably remove the revolution (n / 2) / Z order component based on the bearing vibration. Is. Even when the revolution (n / 2) and the Z-order component based on the bearing vibration are large, the rotational speed of the revolution speed detection encoder 26a (26b), that is, the rolling elements 9a, 9a (9b, 9b) in each row. The revolving speed is obtained with high accuracy, and the load applied to the rolling bearing unit can be detected with high accuracy. At the same time, the response delay is suppressed to a level where there is no practical problem. In the following description, the reference numerals of the rolling bearing unit are shown in FIG.

The notch filter used in the present embodiment is a comb notch filter as shown in FIG. 13 and removes revolution (n / 2) · Z order component based on the bearing vibration by Z conversion. In FIG. 13, N is the number of pulses per revolution existing on the detected surface of the revolution speed detecting encoder 26a (26b), A is a constant that defines the notch frequency, and α is the steepness of the notch. This is a constant to be determined (which also affects convergence). Further, x {x (k)} is input data to the notch filter, and is obtained from the output signal of the revolution speed detection sensor 24a (24b) in which the detection portion is opposed to the revolution speed detection encoder 26a (26b). This is the required pulse period or its inverse (pulse speed). The calculation formulas used in the processing performed in the flowchart shown in FIG. 13 are the following formulas (9) and (10).

  In the above equation (9), y ′ (k−N / A) means y ′ calculated in the past by N / A from the current time (processing number k). In order to calculate the current y ′ (k) in the above equation (9), a value obtained by multiplying y ′ (k−N / A) by α is subtracted from the input data x (k). The past y ′ (k−N / A) is stored in a memory or the like. In the initial state of calculation, there is no past y ′ (k−N / A). However, calculation may be started by substituting zero, or an appropriate value is stored in the memory in advance as an initial value. Also good.

  The output y {y (k)} of the comb notch filter as described above is based on the input data x (k), as shown in the above equation (10), the latest y ′ (k) and the past y It calculates from '(k-N / A). As described above, A is a constant that defines the notch frequency. The constant A is combined with the number of pulses N per revolution, and the output y (k) is calculated based on the input data x (k). By using it for the calculation for obtaining, it is possible to construct a so-called fixed-order notch filter that follows a frequency that changes as the rotational speed increases or decreases. In this notch filter, the order of the notch is determined by the value of the filter constant A, and the notch order component is fixed even if the frequency corresponding to the target order changes due to the change in revolution speed. For example, if A = 2, it becomes a notch filter that removes the primary error component of revolution, if A = 9, it becomes a notch filter that removes the error component of 4.5th revolution, and if A = 18, it becomes revolution 9 The notch filter removes the next error component. FIG. 14 shows the attenuation characteristics of the notch filter with A = 18, and FIG. 15 also shows the phase lag characteristics. 14 and 15, by considering the number Z of the rolling elements 9 a, 9 a (9 b, 9 b) in each row and appropriately setting the constant A, the revolution based on the bearing vibration (n / 2 It can be seen that the removal of the Z-order component can be performed while suppressing the response delay to a level that does not cause a problem in practice. That is, it can be seen from FIG. 14 that the revolution Z-order component (revolution 9th-order component) of the number of rolling elements Z (= 9) has been removed, and from FIG. It can be seen that there is almost no phase delay in the first-order band or lower.

  However, in order to obtain the output y (k) based on the input data x (k) using both equations (9) and (10), N / A in these equations is not an integer. Don't be. That is, the notch filter performs calculation using y ′ calculated in the past by N / A from the current time (processing number k), that is, y ′ (k−N / A). If is a value having a decimal, the process cannot be executed. Therefore, when the constant A that defines the notch frequency is 18, which is a premise for obtaining FIGS. 14 and 15, the number of pulses N must be an integral multiple of 18. That is, the number N of pulses per revolution of the revolution speed detecting encoder 26a (26b) must be an even multiple (2 × integer multiple) of the order of fluctuation to be removed.

  Therefore, the number of pulses N is an integral multiple of the number Z of rolling elements 9a, 9a (9b, 9b) in each row when the error component to be removed is 0.5Z order, and the removal target is 1.0Z order. In this case, it must be an even multiple of this number Z. Similarly, when the removal target is a 1.5Z order component, the pulse number N is an integer multiple of 3 × Z, and when the removal target is a 2Z order component, the pulse number N is an integer multiple of 4 × Z. Things are desirable. However, since the fluctuation amount (amplitude) of the high frequency error component such as the 1.5Z order component or more is small as described above, it can be ignored in many cases. Even if the error component is removed by the low-pass filter, the phase delay can be suppressed to a level that does not cause a problem in practice. Therefore, it is not necessary to determine the number of pulses N in consideration of removal of high-frequency error components of 1.5Z order or higher. Therefore, it is not necessary to increase the number N of pulses.

  The present embodiment is effective from the aspect of removing the revolution (n / 2) / Z order component even when different fluctuation components exist in the vicinity of the revolution (n / 2) / Z order component based on bearing vibration. There is. Therefore, the revolution primary component (if there is no different fluctuation component in the vicinity) may be removed by an adaptive filter as shown in the first and second embodiments. However, if the revolution primary component and the revolution (n / 2) / Z order component are processed by separate filters, the amount of calculation increases and the processing becomes troublesome. Therefore, when applying this embodiment, It is preferable to remove the primary revolution component by a notch filter.

  Even if the number of pulses N is not set as described above, even if the data sampled at each pulse interval is resampled in the CPU (arithmetic unit), the above error component is removed by the notch filter. it can. However, not only the data time delay occurs due to the resampling, but also the calculation amount becomes enormous. Therefore, when obtaining the load applied to the rolling bearing unit for ensuring the running stability of the automobile, it is impractical to configure the notch filter by the method as described above.

FIG. 16 shows a fourth embodiment of the present invention. In the case of the first embodiment shown in FIG. 1 described above, a pair of revolution speed detection sensors 24a and 24b and one rotation speed detection sensor 15a are incorporated at the tip of a single sensor unit 22. . On the other hand, in the case of the present embodiment, only a pair of revolution speed detection sensors 24a and 24b are incorporated at the tip of the sensor unit 22a. The rotational speed detection sensor 15 b is supported by a cover 14 that closes the inner end opening of the outer ring 1. In accordance with this, the rotational speed detection encoder 13 b is externally fitted and fixed to the inner end portion of the inner ring 6 constituting the hub 2. The number of rolling elements 9a, 9a (9b, 9b) is, for example, 12 for each row (24 in total). In this case, each of the detected surfaces of the revolution speed detecting encoder 26a (26b) made of a permanent magnet has 72 pulses, which is 6 times 12 pulses per revolution, A total of 72 poles are provided, 36 poles each. When the rotation of the detection surface of the revolution speed detection encoder 26a (26b) is measured by the revolution speed detection sensor 24a (24b) incorporating a magnetic detection element such as a Hall IC, the rotation speed is 36 times per revolution. Although a periodic signal is obtained, this is doubled to detect a signal of 72 pulses per revolution. Alternatively, even when 72 pulses per revolution are detected in the same manner, 72 poles of S and N poles are provided on the detection surface of the revolution speed detecting encoder 26a (26b), 144 poles in total. The rising edge or only the falling edge of the periodic signal measured 72 times per revolution may be detected by using the multipolar magnet encoder. This is the same as in the case of the first embodiment.

  Regarding the configuration and operation of the fluctuation removing filter for removing the fluctuation component by processing the detection signals of the three sensors 24a, 24b, and 15b, the above-described first and second embodiments or the above-described implementation are used. Since it is the same as that of Example 3, the overlapping description is abbreviate | omitted. Each encoder and sensor is not limited to a multipole magnet and a Hall IC. An encoder made of a simple magnetic material can be used, and another magnetic detection element such as an MR element can be used instead of the Hall IC. Or what is called a passive type which combined a permanent magnet, a pole piece, and a coil as a sensor can also be used. In short, the operation and effect of using the fluctuation elimination filter that is a feature of the present invention is that the number of pulses to be detected is an integral multiple of the number of rolling elements, regardless of the type of sensor or encoder, or the pulse detection method. It is obtained by doing.

  The rolling bearing unit with rolling speed detection device of the rolling element of the present invention is not limited to the load measuring device of the rolling bearing unit for measuring the load applied to the rolling bearing unit that supports the wheel of the automobile as shown in the embodiment. First, it can be used to detect the revolution speed of the rolling elements constituting the rotation support part of various rotary machine devices. In this case, it is possible to accurately obtain the revolution speed of each rolling element without particularly increasing the mounting accuracy of the revolution speed detecting encoder and the shape and dimensional accuracy of each component of the rolling bearing unit. Cost can be reduced.

BRIEF DESCRIPTION OF THE DRAWINGS Sectional drawing of the rolling bearing unit incorporating the rotation detection apparatus for load measurement which shows Example 1 of this invention. The A section enlarged view of FIG. FIG. 3 is a schematic diagram showing a cage, rolling elements, an encoder, and a rotation detection sensor as seen from above in FIG. 2. The schematic diagram of a rolling bearing unit for demonstrating the reason which can measure a load based on rotational speed. The block diagram which shows the circuit for reducing the fluctuation | variation of the output signal of the rotational speed detection sensor based on the whirling of a cage | basket, and a bearing vibration from an adaptive filter. FIG. 4 is a schematic diagram showing the cage and the encoder as viewed from the side of FIGS. 1 to 3 in order to explain the reason why the output signal of the rotation speed detection sensor varies based on the swing of the cage. The diagram which shows the state from which the output signal of a rotational speed sensor fluctuates based on the whirling of a cage | basket, a bearing vibration, and the error of a magnetization pitch. The diagram which shows the state which reduces the fluctuation | variation of the output signal of a rotational speed sensor by an adaptive filter. In order to explain the necessity of the second embodiment of the present invention, a diagram showing a fluctuation state of a signal representing a rotation speed when the adaptive filter is operated by the synchronous LMS algorithm and correction for the DC level is not performed. . The figure similar to FIG. 5 which shows Example 2 of this invention. The graph which shows the state which samples a filter coefficient in order to correct | amend regarding DC level. In order to show the effect of Example 2, the diagram which shows the fluctuation | variation state of the signal showing a rotational speed in the case where the adaptive filter is operated by the synchronous LMS algorithm and the correction related to the DC level is performed. The flowchart which shows the structure of the comb notch filter which shows Example 3 of this invention. The diagram which shows an example of the attenuation characteristic of a notch filter. Similarly, the diagram which shows one example of the characteristic of a phase delay. Sectional drawing of the rolling bearing unit incorporating the rotation detection apparatus for load measurement which shows Example 4 of this invention. Sectional drawing of the rolling bearing unit which incorporated the sensor for radial load measurement known conventionally. Sectional drawing of the rolling bearing unit which incorporated the sensor for axial load measurement conventionally known.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1, 1a Outer ring 2, 2a Hub 3, 3a Rotation side flange 4 Hub body 5 Nut 6 Inner ring 7 Outer ring raceway 8 Inner ring raceway 9a, 9b Rolling element 10, 10a Mounting hole 11 Displacement sensor 12 Sensor ring 13, 13a, 13b Rotational speed Detection encoder 14 Cover 15, 15a, 15b Rotational speed detection sensor 16 Knuckle 17 Fixed flange 18 Bolt 19 Screw hole 20 Load sensor 21a, 21b Cage 22 Sensor unit 23 Tip 24a, 24b Revolution speed detection sensor 25 Rim 26a, 26b Revolution speed detection encoder 27 Adaptive filter

Claims (9)

  An inner ring equivalent member having an inner ring raceway on an outer peripheral surface, an outer ring equivalent member having an outer ring raceway on an inner peripheral surface, a plurality of rolling elements provided in a freely rollable manner between the inner ring raceway and the outer ring raceway, and An encoder whose characteristics are alternately changed with respect to the circumferential direction, which is supported and fixed concentrically with the retainer on a part of the retainer that holds each rolling element, and rotates together with the retainer, and a detection unit of the encoder. Based on a rotation detection sensor provided in a state of being opposed to the detection surface and a signal sent from the rotation detection sensor, an operation for calculating the rotation speed of the cage that matches the revolution speed of each rolling element The arithmetic unit is provided with a fluctuation removing filter for removing the influence of fluctuations in the detection signal of the rotation detection sensor, and the number of times the characteristic per rotation of the encoder is changed is calculated with respect to each other. The inner ring raceway and the revolution speed sensing rolling bearing unit of the rolling elements, characterized in that the integral multiple number of rolling elements existing between the outer ring raceway opposed.   The revolution speed detecting device for a rolling element according to claim 1, wherein the number of times the characteristic per revolution of the encoder changes is an even multiple of the number of rolling elements existing between the inner ring raceway and the outer ring raceway facing each other. Rolling bearing unit.   The encoder is a permanent magnet in which S poles and N poles are alternately arranged at equal intervals in the circumferential direction on the detected surface, and the sum of the S poles and N poles per one rotation of the detected surface is The rolling bearing unit with a revolution speed detecting device for a rolling element according to claim 1, wherein the rolling bearing unit is an integral multiple of the number of rolling elements existing between an inner ring raceway and an outer ring raceway facing each other. A rejection filter is an adaptive filter, and n is a positive integer, the number of rolling elements existing between the inner ring raceway and the outer ring raceway opposing the Z together was a f _c and revolution frequency of the rolling elements If, the adaptive filter, from among the signals representing the revolution speeds of the rolling elements, revolving primary variation in the frequency components, and, of the frequency component represented by (n / 2) · Z · f c The rolling bearing unit with a revolution speed detection device for a rolling element according to any one of claims 1 to 3, which has a function of removing fluctuations. The rolling bearing unit with a rotational speed detection device according to claim 4, wherein the adaptive filter is operated by an LMS algorithm.   The rolling bearing unit with a revolution speed detecting device for a rolling element according to claim 1, wherein the fluctuation removing filter is a fixed-order notch filter. and n is a positive integer, the number of rolling elements existing between the inner ring raceway and the outer ring raceway opposing the Z each other, when the f _c and revolution frequency of the rolling elements, notch filter, each of these from among the signals rolling representing the revolution speed of the moving object, variation of the revolution primary frequency component, and has a function of removing the variation of the frequency component represented by (n / 2) · Z · f c A rolling bearing unit with a revolution speed detecting device for a rolling element according to claim 6 . The rolling bearing unit with a rolling speed detection device for a rolling element according to any one of claims 1 to 7 , which is used for rotatably supporting a wheel on a suspension device. A plurality of rotation detection sensors are provided, and the plurality of rotation detection sensors measure the rotation speeds of a plurality of rotation parts generated during operation of a single rolling bearing unit, including the rotation speed of at least one cage. is intended to utilize signals representative of the rotational speed of the plurality of locations of the plurality of rotation detection sensor detects, for determining the load applied between the inner ring equivalent member and the outer ring member, according to claim 8 A rolling bearing unit with a revolution speed detection device for a rolling element described in any of the above.

Images