System and method for determining bearing preload by vibration measurement

Abstract

A method for determining bearing preload by vibration measurement includes: mounting the bearing on a vibration tester; clamping the outer ring to prevent rotation; mounting a sensor near the outer ring to measure vibration; rotating the inner ring of the bearing to uncover the eigenfrequency; transmitting the measured vibration data from the sensor to a computer workstation; the computer workstation performing a numerical FFT that transforms the data to the frequency domain; analyzing the spectral data from the FFT by the computer workstation; a peak detection algorithm determining the peaks as the various modes of the outer ring; classifying the modes and identifying the dominant mode; obtaining a numerical relationship for each mode between resonance and preload; comparing the mode relationship with one or more benchmarks; a match between the numerical relationship of each mode and the numerical relationship of the ideal benchmark's graphical mode indicates that the correct preload has been determined. Additionally, a system for performing this method.

Description

Technical Field

[0001] The present invention relates to a system and method for determining a bearing preload by vibration measurement. More particularly, the present invention relates to a system and method for determining a bearing preload by sensing vibration on an outer ring of a bearing and comparing the sensed vibration data with ideal reference data. Background Art

[0002] The internal preload of a wheel bearing has an important impact on several key performances of the final product:

[0003] - Friction

[0004] - Life

[0005] - Stiffness

[0006] - Noise (NVH)

[0007] However, the internal preload cannot be directly measured, and extracting the product preload is challenging. In addition, for quality purposes, preload measurement should be performed within the cycle time of the production line to avoid additional costs.

[0008] Today, the industry is using two methods to check the preload. One method is based on spacers (intrusive, long) inserted into the product, and the second method is based on the stiffness measurement of components under an axial load associated with a numerical model. The second method is faster than the original method. However, the industry faces many challenges in making the device repeatable and fast enough to be integrated into the production line.

[0009] Preload affects the stiffness of the bearing product and thus its natural vibration frequency. Summary of the Invention

[0010] The concept disclosed in the present invention is to use the vibration response of the product and compare the vibration response with a calibrated chart and a numerical model to identify the internal preload. Brief Description of the Drawings

[0011] The present invention and its advantages will be better understood by studying the detailed description of the specific embodiments given by way of non-limiting examples and illustrated in the drawings, in which:

[0012] Figure 1 is a schematic diagram of a system for determining a bearing preload by vibration measurement according to a first embodiment of the present invention;

[0013] Figure 2 is based on Figure 1 A side view of a profile sensor of the system in contact with the outer race of the bearing and mounted to the spindle;

[0014] Figure 3 is based on Figure 1 A side view of a laser vibrometer of the system sensing the outer race of a bearing and mounted to a spindle;

[0015] Figure 4 is based on Figure 1 A cross-sectional view of a wheel hub bearing assembly of the system;

[0016] Figure 5 is based on Figure 1 A frequency versus preload graph of the system;

[0017] Figure 6 is based on Figure 1 A graph of vibration modes of a bearing under preload according to an embodiment of the present invention; and

[0018] Figure 7 shows the method for executing Figure 1 Method steps for implementing the functionality of the system. DETAILED DESCRIPTION

[0019] The following detailed description is merely exemplary in nature and is not intended to limit the described embodiments or the applications and uses of the described embodiments. As used herein, the words "exemplary" or "illustrative" mean "used as an example, instance, or illustration." Any embodiment described herein as "exemplary" or "illustrative" is not necessarily to be construed as being preferred or advantageous over other embodiments. All embodiments described below are exemplary embodiments provided to enable those skilled in the art to make or use embodiments of the present disclosure and are not intended to limit the scope of the present disclosure as defined by the claims.

[0020] For the purposes of this description, the terms "inwardly", "outwardly", "upwardly", "downwardly", "left", "rearwardly", "right", "front", "vertical", "horizontally" and their derivatives shall refer to the same. Figure 1 The present invention is directed to the invention in the following embodiments. Furthermore, it is not intended to be bound by any expressed or implied theory presented in the foregoing technical field, background technology, summary of the invention, or the following detailed description. It should also be understood that the specific devices and processes shown in the drawings and described in the following description are merely exemplary embodiments of the inventive concepts defined in the appended claims. Therefore, unless the claims explicitly state otherwise, the specific dimensions and other physical characteristics related to the embodiments disclosed herein should not be considered limiting.

[0021] system

[0022] In Figure 1 is shown a system 100 for determining a bearing preload by vibration measurement. The system 100 provides a bearing 10 which is mounted on a mounting fixture 20 or a clamping system. The bearing 10 includes an outer ring 15, an inner ring 17 and at least two rows of rolling elements 19. Also see Figure 2 and Figure 3 .

[0023] The mounting fixture 20 is arranged on a bearing noise and vibration tester 30. The bearing noise and vibration tester 30 includes a rotating device 32 and a control 40 having associated electronics for controlling the functions of the tester 30. The rotating device 32 includes a rotating arbor 12 attached thereto. The outer ring 15 of the bearing 10 is rigidly clamped on the clamping system 20. This is done to prevent the outer ring 15 from rotating when the inner ring 17 of the bearing 10 rotates. That is, the clamping load on the outer ring is preferably less than or greater than the internal preload to be measured. This will make the relationship between the preload / eigenfrequency more linear. The outer ring (OR) anti-rotation system 20 is provided to ensure that the outer ring remains stationary when the inner ring rotates.

[0024] The rotating arbor 12 is provided to rotate the inner ring 17 of the bearing 10. The inner ring 17 of the bearing 10 rotates at a predetermined rate.

[0025] When the inner ring 17 of the bearing 10 rotates at a predetermined rate, the bearing rotation generates vibration noise on the outer ring 15 of the bearing 10. As will be further disclosed, the preload is determined using the frequency of the generated noise.

[0026] The sensor 35 is arranged near the outer ring 15 of the bearing 10 to measure and record the vibration on the outer ring 15. The vibration of the outer ring 15 can be set to be at least one of position / displacement, velocity, and acceleration motion at a predetermined rotational rate. In this way, the sensor can be set to be one of a laser vibrometer (velocity), an accelerometer (acceleration), and a coil vibrometer. Both the coil vibrometer and the laser vibrometer measure velocity. However, the coil vibrometer needs to be in contact with the surface of the product and there must be enough space for the coil vibrometer to reach the surface of the product. The laser vibrometer does not need to contact the bearing, only the laser beam needs to contact the bearing. This enables measurement even when the surface of the outer ring is difficult to reach. In addition, the laser vibrometer can remain stationary between measurements, while the coil vibrometer cannot remain stationary between measurements. Then, the system for moving the coil vibrometer is more complex and has a higher risk of problems.

[0027] Figure 2 The sensor 35a providing a coil vibrometer is shown, which has a stylus in physical contact with the outer ring 15 of the bearing 10. Figure 3 The sensor 35b providing a laser vibrometer is shown, which does not contact the outer ring 15 of the bearing 10. Figure 4 The bearing 10 is shown mounted in the wheel hub 70 and the inner ring 17 rotates via mounting studs 74. The outer ring is clamped by at least one bolt screwed through the mounting holes 72. To optimize the signal quality, measurements are made on the machined surface and on the non-rotating elements. For the wheel bearing unit, measurements are made on the outer ring (on the inboard surface or on the outboard surface). Even with complex loading and clamping devices, these surfaces allow easy access to the surface using a remote laser unit.

[0028] Preferably, the data is acquired at a sampling rate of at least twice the maximum frequency. As an example, the bearing is rotated in the range from 100 rpm to 2000 rpm. Here, the bearing rotates at a constant speed of 700 rpm. At least 5 revolutions at the steady-state speed are used to record the data.

[0029] It should be noted that the force clamping the outer ring is preferably configured not to be equal to the internal bearing preload. As previously disclosed, the clamping load on the outer ring is preferably less than or greater than (but not equal to) the internal preload to be measured. This makes the relationship between the preload and the eigenfrequency more linear when plotted.

[0030] Here, the measured vibration data is transmitted from the laser (sensor) 35 to the computer workstation 50. The computer workstation 50 provides a software analysis package 52. The software analysis package 52 is capable of performing a numerical Fast Fourier Transform (FFT) on the data to transform the data into the Frequency Domain.

[0031] The computer workstation 50 is used to analyze the spectral data from the FFT. The software analysis package 5 uses a peak detection algorithm to determine the peaks of the amplitude data in the frequency domain. Figure 5 The peaks and modes determined via the generated algorithm are shown. ( Figure 5 Micro on the left means "micro". The modes are classified, and the peak detection algorithm is used to identify the main or dominant modes. The number of modes selected is based on the design choice, but at least one mode needs to be selected. Here, peaks or modes 1, 2, 3, and 4 are various modes of the outer ring with respect to amplitude versus frequency. The frequencies of the OR (outer ring) are measured and analyzed, and at these frequencies, the mode shapes of the OR are observed for optimization, tilting, etc.

[0032] In Figure 6 the numerical relationships obtained for each mode between resonance and preload are shown. Here, a polynomial curve is fitted to the data points or a look-up table. The mode relationship or the polynomial curve represented as a graph is compared with one or more benchmarks / references.

[0033] One or more benchmarks for the bearing are ideal benchmark curves. The data for the ideal benchmark curves can be set by simulation or by measuring known bearings (where the preload has been measured by existing methods). The ideal benchmark curves can also be set by at least one of a calibration chart stored in the memory on board the computer workstation, by simulation, and by the "Gold Set" of known bearings (whose data is stored on the remote server 60).

[0034] Therefore, the match between the numerical relationship for each mode and the ideal benchmark curve modes indicates that the correct preload has been determined. For quality control purposes, the bearing preload is desired. That is, whether to retain or reject the bearing and to evaluate variations in the production process.

[0035] Method

[0036] Figure 7 Method 200 for determining bearing preload by vibration measurement is shown.

[0037] The method includes a first step 210: providing a bearing having an inner ring, an outer ring, and at least two rows of rolling elements.

[0038] The second step 220 of method 200 includes mounting the bearing on a mounting fixture, which is arranged on a bearing noise and vibration tester. The mounting fixture has a rotating spindle attached thereto. The rotating spindle is set to rotate the inner ring of the bearing at a predetermined rate. The rotation of the bearing generates vibration noise on the outer ring.

[0039] In step 230, the method provides: mounting a sensor near the outer ring of the bearing so as to measure and record the vibration of the outer ring at a predetermined rotational rate.

[0040] In step 240, method 200 provides: rotating the bearing on the spindle at a constant speed of 700 rpm configured to excite the eigen - frequency of the outer ring. In step 250, the method provides: measuring the noise or vibration data emitted from the outer ring of the bearing arranged on the spindle.

[0041] Method 200 provides step 260: transmitting the measured vibration data from the sensor to a computer workstation. The computer workstation has a software analysis package. The software analysis package performs a numerical Fast Fourier Transform (FFT) on the data to transform the data into the frequency domain.

[0042] In step 270, method 200 provides: analyzing the spectral data from the FFT using the computer workstation. The software analysis package employs a peak detection algorithm to determine the peaks of the amplitude data in the frequency domain. Here, the peaks are the various modes of the outer ring.

[0043] In step 280, method 200 provides: classifying the modes and identifying the main or dominant modes using the peak detection algorithm. Here, the number of modes selected is based on the design choice. Again, at least 1 mode is required for an effective determination of the pre - load.

[0044] Method 200 provides step 290: obtaining a numerical relationship for each mode between resonance and pre - load. Here, a polynomial curve is fitted to the data points or a look - up table.

[0045] Method 200 provides step 300: comparing the mode relationship or the graphed polynomial curve with one or more benchmarks. Here, one or more benchmarks of the bearing are ideal benchmark curves.

[0046] In step 310, method 200 finally stipulates that the matching between the numerical relationship of each modality and the numerical relationship of the modality of the ideal reference curve indicates that the correct preload has been determined.

[0047] Since many modifications, variations and changes in detail can be made to the described preferred embodiments and methods of the present invention, it is intended that all matter shown in the foregoing description and in the accompanying drawings be interpreted as illustrative and not in a limiting sense. Accordingly, the scope of the present invention is to be determined by the appended claims and their legal equivalents.

Claims

1. A method for determining bearing preload by vibration measurement, the method comprising the following steps: Providing a bearing having an inner ring, an outer ring and at least two rows of rolling elements; Mounting the bearing on a mounting fixture, the mounting fixture being disposed on a bearing noise and vibration tester, the mounting fixture having a rotating spindle attached thereto, the rotating spindle being configured to rotate the inner ring of the bearing at a predetermined rate, wherein the rotation of the bearing generates vibration noise on the outer ring; Mounting the outer ring of the bearing using a clamping and anti-rotation device to prevent the outer ring from rotating when the inner ring of the bearing rotates; Mounting a sensor near the outer ring of the bearing to measure and record the vibration of the outer ring at a predetermined rotational rate; Rotating the bearing on the spindle at a steady-state speed, the steady-state speed being configured to excite the natural frequency of the outer ring; Transmitting the measured vibration data from the sensor to a computer workstation, the computer workstation having a software analysis package, the software analysis package performing a numerical fast Fourier transform (FFT) on the data to transform the data into the frequency domain; Analyzing the spectral data from the fast Fourier transform using the computer workstation, the software analysis package employing a peak detection algorithm to determine the peaks of the amplitude data in the frequency domain, wherein the peaks are the various modes of the outer ring; Classifying the modes and identifying the main or dominant modes using the peak detection algorithm, the number of selected modes being selected according to the design; Obtaining a numerical relationship for each mode between resonance and the preload, wherein a polynomial curve is fitted to the data points or a look-up table; Comparing the mode relationship or the polynomial curve represented as a graph with one or more benchmarks, wherein one or more benchmarks for the bearing are ideal benchmark graphs; wherein a match between the numerical relationship for each mode and the numerical relationship of the ideal benchmark graph mode indicates that the correct preload has been determined.

2. The method for determining bearing preload by vibration measurement according to claim 1, wherein, Data is acquired at a sampling rate of at least twice the maximum frequency.

3. The method for determining bearing preload by vibration measurement according to claim 1, wherein, It further includes rotating the bearing at a constant speed of 700 rpm.

4. The method for determining bearing preload by vibration measurement according to claim 1, wherein, It further includes rotating the bearing in the range from 100 rpm to 2000 rpm.

5. The method for determining bearing preload by vibration measurement according to claim 1, wherein, It further includes providing a clamping force and ideally achieving a more linear relationship for the mode between the bearing preload and the frequency of bearing rotation.

6. The method for determining bearing preload by vibration measurement according to claim 1, wherein, It further includes measuring the vibration of the outer ring using a laser, wherein the laser does not physically contact the outer ring.

7. The method for determining bearing preload by vibration measurement according to claim 1, wherein, it further includes setting the sensor as one of a laser vibrometer, an accelerometer, and a coil vibrometer.

8. The method for determining bearing preload by vibration measurement according to claim 1, wherein, it further includes identifying at least one mode to ensure the minimum accuracy of bearing preload determination.

9. The method for determining bearing preload by vibration measurement according to claim 1, wherein, the ideal reference curve graph is provided by at least one of a calibration chart stored in the memory on the computer workstation, by simulation, by measuring known bearings, and by a transfer function.

10. The method for determining bearing preload by vibration measurement according to claim 1, wherein, optimized signal quality is obtained by measuring vibrations on the machined surface and on the non-rotating element.

11. The method for determining bearing preload by vibration measurement according to claim 1, wherein, the outer ring vibration at a predetermined rotational speed is completed by measuring and recording at least one of position, velocity, and acceleration motion.

12. The method for determining bearing preload by vibration measurement according to claim 1, wherein, measuring the wheel bearing unit is completed by a laser on one of the inner surface or the outer surface of the outer ring.

13. The method for determining bearing preload by vibration measurement according to claim 12, wherein, even with complex loads and clamping devices, these surfaces allow easy access to the surfaces by the distance laser unit.

14. The method for determining bearing preload by vibration measurement according to claim 1, wherein, the force for clamping the outer ring is configured not to be equal to the bearing preload.

15. The method for determining bearing preload by vibration measurement according to claim 5, wherein, the clamping force is less than or greater than the bearing preload.

16. A system for determining bearing preload by vibration measurement, the system comprises: a bearing having an inner ring, an outer ring, and at least two rows of rolling elements, the bearing is mounted on a mounting fixture, the mounting fixture is arranged on a bearing noise and vibration tester, the mounting fixture has a rotating spindle attached thereto, the rotating spindle is set to rotate the inner ring of the bearing at a predetermined rate, wherein, the rotation of the bearing generates vibration noise on the outer ring, and the outer ring is fixed by a clamping and anti-rotation system to prevent the outer ring from rotating when the inner ring of the bearing rotates, a sensor arranged near the outer ring of the bearing to measure and record the outer ring vibration, wherein, the outer ring vibration is at least one of position, velocity, and acceleration motion at a predetermined rotational speed, the bearing rotates on the spindle at a speed of 700 rpm, and the speed of 700 rpm is configured to excite the natural frequency of the outer ring. The measured vibration data is transmitted from the sensor to a computer workstation, which provides a software analysis package that performs a numerical fast Fourier transform (FFT) on the data to transform the data into the frequency domain. The spectral data from the fast Fourier transform is analyzed using the computer workstation, and the software analysis package employs a peak detection algorithm to determine the peaks of the amplitude data in the frequency domain, where the peaks are the various modes of the outer ring. The modes are classified, and the peak detection algorithm is used to identify the main or dominant modes, and the number of selected modes is selected according to the design. A numerical relationship is obtained for each mode between resonance and preload, where a polynomial curve is fitted to the data points or a look-up table. The mode relationship or the polynomial curve represented as a graph is compared with one or more benchmarks, where one or more benchmarks for the bearing are ideal benchmark graphs. A match between the numerical relationship for each mode and the numerical relationship of the ideal benchmark graph mode indicates that the correct preload has been determined.

17. The system for determining bearing preload by vibration measurement according to claim 16, wherein, further comprising: The bearing is installed in a wheel hub, and the inner ring rotates via mounting studs.

18. The system for determining bearing preload by vibration measurement according to claim 16, wherein, The force clamping the outer ring is configured to be not equal to the bearing preload.

Images