JP2006118869A - Defect detector for rolling bearing, and defect detection method for the rolling bearing - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To detect a flaw or the like, even when a pulse based on a defect, such as flaw existing in one part of a vibration signal and within a measurement time, by detecting the vibration generated from a rolling bearing. <P>SOLUTION: A thrust load is applied to the rolling bearing to make an outer ring and an inner ring relatively rotate at a prescribed number of revolution, and the vibration is detected by a vibration sensor 101 and is converted into a digital signal by an A/D converter 104, and to be converted thereafter into an envelope signal by an envelope processing part 105. The number of peaks, appearing in the envelope signal exceeding a set value obtained by multiplying an effective value of the envelope signal with a dB magnification, is integrated by a peak counter 111, and the presence of the defect, such as flaws in the rolling bearing is determined by a defect determination part 112, based on whether an integrated value exceeds a prescribed threshold value. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention detects a vibration generated from a rolling bearing rotating with a thrust load applied, and detects a defect such as a flaw in the rolling bearing based on the detected vibration, and a defect in the rolling bearing. It relates to a detection method.

  Conventionally, in order to detect defects such as flaws in a rolling bearing, a defect detection apparatus having the configuration shown in FIG. 10 has been proposed. In this device, vibration generated from the radial direction or the axial direction of the rolling bearing is detected by a pickup 301 including a speed sensor or an acceleration sensor, and a signal is amplified by an amplifier 302, and only a predetermined frequency band is band-limited from the amplified vibration signal. Extracted by the filter 303 and converted into a digital signal by the A / D converter 304. Next, the digital signal is subjected to envelope processing by an envelope processing unit 305, and frequency analysis is performed by a frequency analysis unit 306 by FFT (Fast Fourier Transformation) to convert it into frequency spectrum data. The determination unit 307 determines whether or not there is a defect such as a flaw in the rolling bearing by comparing and evaluating the amplitude of the frequency spectrum data with a defect frequency such as an inner ring flaw and a defect frequency such as an outer ring flaw. A method of performing envelope processing on the detected digital signal of vibration and performing frequency analysis by FFT calculation to determine the presence or absence of defects such as flaws is called an envelope FFT method.

  For example, when detecting a vibration generated from a rolling bearing having a defect such as a flaw in the outer ring, the defect detection apparatus using the envelope FFT method can detect all flaws within the measurement time as shown in FIG. When a pulse based on the defect is generated, frequency analysis is performed by the frequency analysis unit 306, so that the maximum amplitude indicating that there is a defect such as a flaw in the frequency spectrum is clearly shown in FIG. 11B. appear. Therefore, it can be easily determined that the outer ring has a defect such as a flaw.

However, for example, when a vibration generated from a rolling bearing having a defect such as a flaw in the inner ring is detected, a pulse based on the defect such as a flaw is generated only in a part of the measurement time as shown in FIG. May come out. In such a case, even if frequency analysis is performed by FFT calculation, a clear maximum amplitude does not appear as shown in FIG. 12B, and even if there is a defect such as a flaw in the rolling bearing, it cannot be detected. There is a fear.
JP2003-232674 (page 3-5, FIG. 1)

  The present invention has been made in view of the above-described circumstances, and its purpose is to detect a vibration generated from a rolling bearing and determine a part of a vibration signal within a measurement time when determining the presence or absence of defects such as flaws. It is an object of the present invention to provide a rolling bearing defect detection device and a rolling bearing defect detection method capable of detecting a defect such as a flaw even when a pulse waveform based on a defect such as a flaw is present. .

In order to achieve the above-described object, a rolling bearing defect detection apparatus according to the present invention is characterized by the following (1) to (4).
(1) A predetermined thrust load is applied to the rolling bearing, the outer ring and the inner ring are rotated relative to each other at a predetermined number of rotations to detect vibrations generated from the rolling bearings, and defects in the rolling bearings are detected based on the detected vibrations. A rolling bearing defect detection device for detecting
Vibration detecting means for detecting vibration generated from the rolling bearing and generating a vibration signal;
Frequency filter means for extracting a vibration signal in a predetermined frequency band from the vibration signal;
Analog-digital conversion means for converting the extracted vibration signal of a predetermined frequency band into a digital signal at a predetermined sampling frequency;
An envelope signal processing means for generating an envelope signal by performing envelope processing on the vibration signal converted into the digital signal;
When the number of peaks in the envelope signal that is greater than or equal to the threshold value exceeds a predetermined set value, any of the members constituting the rolling bearing has a defect and is generated from the rolling bearing other than the defect. First defect determination means that tentatively determines that there is a foreign object sound to be detected;
An autocorrelation function for the envelope signal when it is determined by the first defect determination means that any of the members constituting the rolling bearing is defective and there is a foreign object sound generated from the rolling bearing other than the defect Autocorrelation function computing means for computing
When the position of the time delay element (lag) where the peak intensity indicating that the correlation is strong in the calculation result of the autocorrelation function appears and the member constituting the rolling bearing is defective, from the bearing specifications of the rolling bearing By collating the position of the time delay element for each member predicted by performing the autocorrelation function calculation based on the defect frequency specific to each member to be calculated and the predetermined sampling frequency, A second defect determining means for identifying the member having a rolling bearing defect;
Be provided.
(2) In the defect detection device for a rolling bearing configured as described in (1) above, when the second defect determination unit cannot identify a defective member, vibration detected by the vibration detection unit is generated from the rolling bearing. It is determined that the foreign object sound is generated.
(3) In the defect detection device for a rolling bearing having the configuration (1) or (2), the threshold value is a value obtained by multiplying an effective value of the envelope signal by a predetermined dB magnification.
(4) In the defect detection device for a rolling bearing having the structure according to any one of (1) to (3), the analog / digital conversion unit, the envelope signal processing unit, the first defect determination unit, Processing in at least one of the autocorrelation function calculating means and the second defect determining means is executed by a program of a microcomputer unit.

In order to achieve the above-described object, a rolling bearing defect detection method according to the present invention is characterized by the following (5) and (6).
(5) A predetermined thrust load is applied to the rolling bearing, the outer ring and the inner ring are rotated relative to each other at a predetermined number of rotations, vibrations generated from the rolling bearings are detected, and the rolling bearings are detected based on the detected vibrations. A rolling bearing defect detection method for detecting defects, comprising:
Detecting vibration generated from the rolling bearing to generate a vibration signal;
Extracting a vibration signal of a predetermined frequency band from the vibration signal;
The vibration signal limited in the frequency band is converted into a digital signal at a predetermined sampling frequency,
An envelope signal is generated by performing envelope processing on the vibration signal converted into the digital signal,
When the number of peaks in the envelope signal that is greater than or equal to the threshold value exceeds a predetermined set value, any of the members constituting the rolling bearing has a defect and is generated from the rolling bearing other than the defect. Tentatively determined that there is a foreign object sound,
An autocorrelation function is calculated for the envelope signal when it is determined that any of the members constituting the rolling bearing has a defect and there is a foreign object sound generated from the rolling bearing other than the defect,
When the position of the time delay element (lag) where the peak intensity indicating that the correlation is strong in the calculation result of the autocorrelation function appears and the member constituting the rolling bearing is defective, from the bearing specifications of the rolling bearing By collating the position of the time delay element for each member predicted by performing the autocorrelation function calculation based on the defect frequency specific to each member to be calculated and the predetermined sampling frequency, Identifying said member with a defective rolling bearing.
(6) If the defective member cannot be identified by collating the position of the time delay element, it is determined that the detected vibration is a foreign object sound generated from the rolling bearing.

According to the defect detection device for a rolling bearing configured as described in (1) above, it is possible to determine the presence or absence of defects such as failure based on vibration generated from the rolling bearing, and in particular, bearing vibration measurement provided with a mechanism that causes misalignment. Even when a pulse waveform based on a defect such as a flaw appears in only a part of the vibration signal waveform within the measurement time, which occurs frequently when detecting a defect such as a flaw in a rolling bearing using the device Such defects can be detected effectively.
According to the defect detection device for a rolling bearing configured as described in (2) above, defects such as flaws in the rolling bearing and foreign sounds generated from the rolling bearing can be separated and detected, so that defects such as flaws in the rolling bearing can be detected. The presence / absence can be accurately detected.
According to the defect detection device for a rolling bearing having the configuration (3) above, the determination of the presence or absence of defects such as flaws in the rolling bearing is determined based on the effective value of the vibration signal. Can be detected.
According to the defect detection device for a rolling bearing having the configuration (4), the hardware configuration can be simplified, and the device can be reduced in size and cost.

Further, according to the rolling bearing defect detection method having the above step (5), it is possible to determine the presence or absence of defects such as failure based on vibrations generated from the rolling bearing, and in particular, a mechanism for causing misalignment is provided. When a defect such as a flaw in a rolling bearing is detected using a bearing vibration measuring device, a pulse waveform based on a flaw such as a flaw appears only in a part of the vibration signal waveform within the measurement time. However, defects such as scratches can be detected effectively.
Further, according to the defect detection method for a rolling bearing having the above step (6), defects such as a flaw in the rolling bearing and foreign matter sound generated from the rolling bearing can be separately detected, so that a flaw in the rolling bearing, etc. It is possible to accurately detect the presence or absence of defects.

  According to the present invention, when detecting the vibration generated from the rolling bearing and determining the presence or absence of defects such as flaws, a pulse waveform based on the defects such as flaws appears only in part of the vibration signal within the measurement time. Even if it is, defects such as flaws can be detected.

  The present invention has been briefly described above. Furthermore, the details of the present invention will be further clarified by reading through the best mode for carrying out the invention described below with reference to the accompanying drawings.

  A rolling bearing defect detection device according to an embodiment of the present invention is mounted on a bearing vibration measuring device, loaded with a predetermined thrust load, and driven by vibration generated from the rolling bearing rotating at a predetermined rotational speed. It detects defects such as scratches. DESCRIPTION OF EMBODIMENTS Hereinafter, embodiments of a defect detection device for a rolling bearing according to the present invention will be described with reference to the drawings. FIG. 1 is a diagram illustrating a functional schematic configuration of a defect detection device for a rolling bearing.

  In FIG. 1, a defect detection device 100 for a rolling bearing includes a vibration sensor 101 that functions as vibration detection means, a signal amplification circuit 102, a low-pass filter (LPF) 103 that functions as frequency filter means, and analog / digital conversion means. A functioning A / D converter 104, an envelope processing unit 105 functioning as an envelope signal processing unit, an autocorrelation function computing unit 106 functioning as an autocorrelation function computing unit, an effective value computing unit 107, and a dB magnification The configuration includes a setting unit 108, a multiplier 109, a comparator 110, a peak counter 111, and a defect determination unit 112 that functions as first and second defect determination means.

  The rolling bearing defect detection apparatus 100 shown in FIG. 1 can be entirely configured by hardware, but the range for performing digital signal processing surrounded by a broken line in the figure is a CPU (Central Processing Unit). ) And a microcomputer unit having a memory, and can be processed by a program.

  The vibration sensor 101 is composed of a vibration measuring element such as a piezoelectric element or an acceleration sensor, and is attached to a bearing preload adding mechanism 220 of a bearing vibration measuring apparatus 200 described later. The defect detection device 100 for a rolling bearing detects, for example, axial vibration generated when the inner ring of the rolling bearing 10 is rotated at 1800 rpm for a predetermined measurement time, for example, 0.8 seconds. Convert to signal.

  The signal amplifying circuit 102 is a preamplifier composed of an OP amplifier or the like, and amplifies the vibration signal detected by the vibration sensor 101 into a voltage signal of a predetermined level and outputs it with a low impedance.

  The low-pass filter 103 removes unnecessary high-frequency components such as noise contained in the vibration signal and limits the frequency band to ½ or less of the sampling frequency of the A / D converter 104, and an OP amplifier is incorporated. Active filter.

  The A / D converter 104 converts the analog vibration signal output from the low-pass filter 103 into a digital signal for a measurement time of 0.8 seconds at a predetermined sampling frequency, for example, 24 kHz. Thereby, 19200 pieces of data are obtained.

  The envelope processing unit 105 performs an envelope processing necessary for analyzing a periodic attenuation signal generated when the rolling bearing has a defect such as a flaw. The A / D converter 104 By performing amplitude modulation using the Hilbert transform on the digital signal output from, an envelope signal on the plus side only is obtained.

  The autocorrelation function calculation unit 106 calculates an autocorrelation function (ACF: Autocorrelation Function) representing the temporal characteristics of the envelope signal obtained by the envelope processing unit 105. When the time function x (n) is used, the calculation is performed by the following formula defined as the function of the delay time m, Rxx (m).

但し、Ｎはデータ数である。

N is the number of data.

  The result of calculating the autocorrelation function represents the degree of coincidence with the original signal for each delay time m called lag, and the intensity increases at the position of lag where the correlation is strong. Hereinafter, the lag position where the intensity increases is referred to as a pitch period.

  Hereinafter, the reason why it is possible to determine which of the members constituting the rolling bearing has a defect such as a flaw by calculating the autocorrelation function for the envelope signal will be described.

  If each member of the rolling bearing to be measured has, for example, a flaw (that is, a defect), the flaw frequency (that is, the flaw frequency) that is expected to show a peak in the envelope signal is calculated from the bearing specifications according to the following equation. Can be calculated.

ｆｒ：内輪回転速度（Ｈｚ）
Ｚ ：転動体の数
α ：接触角（度）
Ｄａ：転動体直径（ｍｍ）
ｄｍ：ピッチ円直径（ｍｍ）

fr: Inner ring rotational speed (Hz)
Z: Number of rolling elements α: Contact angle (degrees)
Da: Rolling element diameter (mm)
dm: pitch circle diameter (mm)

  By calculating the autocorrelation function of the envelope signal for each flawed member according to the above calculation formula, obtaining the predicted pitch period, and matching with the output result of the autocorrelation function calculation unit 106 It can be determined that there is a flaw in a member constituting the rolling bearing. In addition, in the output result of the autocorrelation function calculation unit 106, when there are strengths in a plurality of pitch periods, it can be determined that there are defects in a plurality of different members.

  However, since the sampling frequency is finite, the inner ring rotational speed fluctuates, and there are tolerances in the bearing specifications, it is necessary to allow a certain range for confirmation of the pitch period.

  Next, an embodiment in which it is determined that there is a flaw in a member constituting the rolling bearing by using the above autocorrelation function will be described with reference to the drawings. FIGS. 4 to 8 show the case where the bearing inner ring is measured at a rotational speed of 1800 rpm, when the rolling bearing has no flaw, the outer ring has a flaw, the inner ring has a flaw, The figure shows the waveform of the vibration signal and the waveform obtained as a result of the autocorrelation function calculation when there are flaws in the rolling element) and foreign matter noise is generated from the rolling bearing.

  FIG. 4A shows a waveform of a vibration signal generated from a rolling bearing which is considered to be a non-defective product without any flaws, and a characteristic pulse peak does not appear when there is a flaw in the member. In the waveform obtained as a result of performing the autocorrelation function calculation shown in FIG. 4B, a characteristic pitch period does not appear, indicating that the autocorrelation is low. The waveform resulting from the autocorrelation function calculation is calculated on both the positive and negative sides centering on the position where lag is 0, but the pitch period used for the flaw determination may be on either the positive or negative side.

  FIG. 5A shows a vibration signal waveform in the case where there is a flaw in the outer ring, and a peculiar peak appears when there is a flaw at the beginning of the measurement time of 0.8 seconds. As a result of performing the autocorrelation function calculation on the waveform of the envelope signal based on the vibration signal shown in FIG. 5A, a waveform whose intensity increases near the pitch period 250 as shown in FIG. 5B. was gotten.

  On the other hand, when the autocorrelation function is calculated based on the outer ring flaw frequency of 94.6 Hz obtained by the equation (2) and the sampling frequency of 24 kHz of the A / D converter 104, the pitch period is 254. FIG. A good agreement with the peak in the vicinity of the pitch period 250 of the waveform in (b) was observed.

  In the vibration signal waveform in the case where there is a flaw in the inner ring shown in FIG. 6A, a peak waveform showing a flaw appears at the end of the measurement time of 0.8 seconds, and the result of performing the autocorrelation function calculation on the envelope signal The waveform shown in FIG. 6B was obtained. The intensity increases in the vicinity of the pitch period 160, which is in good agreement with the pitch period 163 of the autocorrelation function calculated based on the flaw frequency 146.9 Hz of the inner ring obtained by the equation (3).

  Similarly, as a result of performing the autocorrelation function operation on the envelope signal based on the vibration signal waveform shown in FIG. 7A when the ball has a flaw, the waveform shown in FIG. The strength increases in the vicinity of the pitch period. This is a value approximated to the pitch period 195 of the autocorrelation function predicted from the ball flaw frequency of 123.2 Hz calculated by Expression (4).

  FIGS. 8A and 8B show the waveforms of the vibration signal and autocorrelation function calculation results when the foreign object noise is generated from the rolling bearing. Although there is a small peak over the entire measurement time, no peak indicating intensity is seen in the output result of the autocorrelation function calculation as in the case of non-defective products. In such a case, as described later, together with the result of counting the number of peaks of the envelope signal, it is determined that the sound is a foreign object sound generated from the rolling bearing.

  Thus, when a pulse-like peak waveform indicating that there is a flaw in a part of the vibration signal at the measurement time appears, the pitch period in which the intensity increases in the waveform obtained by performing the autocorrelation function calculation on the envelope signal, By checking the coincidence with the pitch period predicted by calculation from the flaw frequency of each member constituting the rolling bearing, it is possible to identify the flawed member.

  Returning to FIG. 1, the description of the configuration of the rolling bearing defect detection apparatus will be continued.

  In FIG. 1, the effective value calculator 107 sets a magnification by which the effective value of the envelope signal obtained by the above is multiplied by a dB value.

  The effective value calculator 107 calculates an effective value of the envelope signal obtained by the envelope processing unit 105 according to the following equation (5).

  The dB magnification setting unit 108 sets the magnification by which the effective value of the envelope signal obtained by the effective value calculator 107 is multiplied by the dB value.

  The multiplier 109 multiplies the effective value of the envelope signal obtained by the effective value calculator 107 by the dB magnification set by the dB magnification setting unit 108. The multiplied value becomes a threshold value for peak extraction in the comparator 110.

  The comparator 110 compares the envelope signal obtained by the envelope processing unit 105 with a threshold value obtained by multiplying the effective value of the envelope signal by dB magnification, and in the envelope signal waveform larger than the threshold value. A peak (amplitude indicating the maximum) is extracted.

  FIG. 9 illustrates the envelope signal obtained by the envelope processing unit 105, the effective value obtained by the effective value calculator 107, and the threshold value for peak extraction that is the output result of the multiplier 109, respectively. FIG.

  The peak counter 111 counts the peaks extracted by the comparator 110, and counts peaks that are equal to or higher than the peak count level shown in FIG. This is called a peak count method and is suitably used for determining defects such as minute flaws.

  If the number of peaks counted by the peak counter 111 exceeds a predetermined NG limit value, the defect determination unit 112 determines that the target rolling bearing has a defect such as a flaw. When it is determined that there is a defect such as a flaw, as described above, it is determined from the calculation result by the autocorrelation function calculator 106 which of the members constituting the rolling bearing has a defect such as a flaw. Further, when it is determined that there is a defect such as a flaw in the count result of the peak counter 111, when the peak does not appear in the waveform of the calculation result by the autocorrelation function calculator 106, the defect determination unit 112 sets the rolling bearing. It is determined that the sound of foreign matter generated inside the rolling bearing other than defects such as scratches.

  The determination result is sent to a display device such as a liquid crystal display (not shown) to be viewed by an operator, or transmitted to a personal computer or the like and used as data in the production management system.

  Next, when a defect such as a flaw in a rolling bearing is detected by the rolling bearing defect detection apparatus according to the present embodiment described above, a rolling bearing to be detected is mounted, a predetermined thrust load is applied, and a predetermined thrust load is applied. A bearing vibration measuring apparatus that rotates at a rotational speed of 2 will be described. FIG. 2 is a diagram showing a schematic configuration of a bearing vibration measuring apparatus for applying a thrust load that causes misalignment in a rolling bearing.

  In FIG. 2, the bearing vibration measuring apparatus 200 includes a spindle arbor 210, a bearing preload adding mechanism 220, a drive mechanism 230, and a piezo control circuit 240, and the vibration sensor 101 is a disc portion 220 b of the bearing preload adding mechanism 220. It is the structure fixed to the center part. The alternate long and short dash line A indicates a center line that passes through the shaft centers of the bearing 10, the spindle arbor 210, the bearing preload mechanism 220, and the drive mechanism 230 when a positive thrust load is applied to the rolling bearing 10.

  A rolling bearing 10 which is a detection object of defects such as scratches is, for example, a radial deep groove ball bearing, and is disposed between the inner ring 11, the outer ring 12, and the inner and outer rings 11, 12 so as to be rotatable in the circumferential direction. A plurality of balls (that is, rolling elements) 13 are provided.

  The vibration sensor 101 is fixed to the central portion of the disc portion 220 b of the bearing preload adding mechanism 220, and detects vibration generated in the inner ring 11, the outer ring 12, and the ball 13 of the rolling bearing 10 via the drive mechanism 230. The vibration signal detected by the vibration sensor 101 is input to the signal amplification circuit 102 of the rolling bearing defect detection device 100 via the cable 250.

  The spindle arbor 210 has a first shaft portion 210 a that contacts the one end surface 11 a of the inner ring 11, and a second shaft portion 210 b that is concentrically projected from the center portion of the first shaft portion 210 a and is fitted into the inner ring 11. Then, it is connected to a rotating shaft of a motor (not shown) driven by an inverter and rotates at a predetermined rotational speed.

  The drive mechanism 230 is arranged such that one end is fixed on the disc portion 220b of the bearing preload adding mechanism 220 and the other end is in contact with the one end surface 12b of the outer ring 12 of the rolling bearing 10 with an interval of 120 °. The three piezo-electric elements extend and contract in the axial direction by a piezo drive signal supplied from the piezo control circuit 240.

  The bearing preload adding mechanism 220 is arranged opposite to the spindle arbor 210, and the disc portion 220b formed concentrically with the shaft portion 220a is used for the outer ring 12 of the bearing 10 through the cylindrical piezo element of the drive mechanism 230. A predetermined preload is applied to the end face 102b.

  The piezo control circuit 240 includes a CPU (central processing unit) (not shown), generates a three-phase piezo drive signal having a phase difference of 120 ° in accordance with a processing program, and sends it to three piezo elements. Accordingly, the three piezoelectric elements can be expanded and contracted with a phase difference of 120 °, respectively, and a thrust load that causes misalignment with respect to the rolling bearing 10 is applied.

  If a piezo control circuit 240 applies a predetermined level of DC voltage or in-phase alternating voltage to the three piezo elements, a static positive thrust load and an alternating thrust load can be applied to the rolling bearing 10 respectively. it can. Further, a static positive thrust load can be applied to the rolling bearing 10 by using a rod-like rigid body or elastic body instead of the piezo element (in this case, the piezo necessary for driving the piezo element). Needless to say, the control circuit 240 is not necessary. A known example of a mechanism for applying a static positive thrust load to the rolling bearing 10 is shown in FIG. 4 of JP-A-2002-350289.

  In this way, by using the bearing vibration measuring device, it is possible to appropriately select the type of thrust load according to the measurement conditions when detecting defects such as flaws in the rolling bearing.

  Next, the operation of the rolling bearing defect detection apparatus 100 configured as described above will be described. FIG. 3 is a flowchart for explaining a processing procedure in the defect detection device for a rolling bearing according to the present embodiment. The detected defect will be described by taking a flaw as an example.

  First, the vibration sensor 101 detects the vibration generated from the rolling bearing 10 that is loaded with a thrust load so as to cause misalignment by the bearing preload application mechanism 220 of the bearing vibration measuring device 200 and rotates at a predetermined rotational speed, and the vibration signal is detected. Is generated (step S101).

  The vibration signal is sent to the signal amplifier 102 via the cable 250 and amplified to a level in consideration of the dynamic range of the A / D converter 104 (step S102).

  The amplified vibration signal is passed through the low-pass filter 103, so that the frequency band is limited to ½ or less of the sampling frequency of the A / D converter 104 (step S103).

  In step S104, the analog vibration signal band-limited by the low-pass filter 103 is converted into a digital signal of 0.8 seconds, which is a measurement time, for example.

  The vibration signal converted into the digital signal is subjected to envelope processing by the envelope processing unit 105 to be converted into an envelope signal (step S105), and an effective value calculator 107 calculates an effective value of the envelope signal (step S105). S106).

  Subsequently, the effective value of the calculated envelope signal is multiplied by a dB magnification predetermined by the dB magnification setting unit 108 by the multiplier 109 to generate a threshold value for extracting the peak of the envelope signal (step S107). ).

  The comparator 110 compares the threshold value with the envelope signal, and extracts the peak of the envelope signal exceeding the threshold level as shown in FIG. 9 (step S108).

  The extracted peaks are integrated by the peak counter 111 (step S109), and the defect determination unit 112 determines whether or not the integrated value is greater than or equal to a predetermined set value (step S110).

  As a result, when the number of integrated values is less than the set value, the defect determination unit 112 determines that the target rolling bearing is a non-defective product (step S111), and the operation of the bearing vibration measuring device 200 is stopped. The process ends.

  On the other hand, when it is determined in step S110 that the integrated value is greater than or equal to the set value, the autocorrelation function calculation unit 106 performs autocorrelation function calculation on the envelope-processed vibration signal (step S112).

  Next, the defect determination unit 112 determines whether or not the pitch period in the autocorrelation function calculation result matches the predicted outer ring pitch period predicted from the outer ring flaw frequency (step S113). As a result, when it is determined that they match, it is determined that there is a scratch on the outer ring 12 of the rolling bearing 10 (step S114).

  If the pitch period value in the calculation result of the autocorrelation function does not match the predicted pitch period value when there is a flaw in the outer ring in step S113, the pitch period in the calculation result of the autocorrelation function is calculated from the inner ring flaw frequency in step S115. It is determined whether or not it matches the predicted inner ring predicted pitch period.

  As a result, when it is determined that the pitch period value in the calculation result of the autocorrelation function matches the predicted pitch period when there is a flaw in the inner ring, it is determined that there is a flaw in the inner ring 11 (step S116).

  If the pitch period value in the calculation result of the autocorrelation function does not match the predicted pitch period value when there is a flaw in the inner ring in step S115, the pitch prediction in the calculation result of the autocorrelation function is predicted from the ball flaw frequency. It is determined whether or not it coincides with the pitch period (step S117). As a result, when it is determined that the pitch period value in the calculation result of the autocorrelation function matches the predicted pitch period when there is a defect in the ball, it is determined that there is a defect in the ball 13 (step S118).

On the other hand, if the pitch period value in the calculation result of the autocorrelation function does not coincide with the predicted pitch period value when there is a flaw in the ball in step S117, the vibration sensor 101 detects the foreign object sound generated from the rolling bearing 10. It is determined that the bearing has been picked up (step S119), the bearing vibration measuring device 200 is stopped, and the process is terminated.
As described above, step S110, step S111, step S113, step S114, step S115, step S116, step S117, step S118, and step S119 are executed by the defect determination unit 112. In particular, Step S110 and Step S111 are executed when the defect determination unit 112 functions as the first defect determination means, and Step S113, Step S114, Step S115, Step S116, Step S117, Step S118, and Step S119 are performed. Is executed when the defect determination unit 112 functions as the second defect determination means.

  As described above, according to the embodiment of the defect detection device for a rolling bearing according to the present invention, a predetermined thrust load is applied to the rolling bearing 10 mounted on the bearing vibration measuring device 200 to cause the inner ring to rotate at a predetermined rotational speed. The vibration generated from the rolling bearing 10 is detected, and the detected vibration signal is converted into a digital signal and envelope processing is performed. Then, a peak at a predetermined level or more is extracted from the obtained envelope signal, the number of peaks is counted, and a defect such as a flaw in the rolling bearing 10 according to whether or not the peak number exceeds a predetermined threshold and other than the defect Thus, the presence or absence of foreign matter noise generated from the rolling bearing 10 is tentatively determined.

  When it is temporarily determined that the rolling bearing 10 has a defect such as a flaw and there is a foreign object sound generated from the rolling bearing 10 other than the defect, an autocorrelation function calculation is performed on the envelope signal, and the strength is obtained in the calculation result. By determining whether the increased pitch period matches each pitch period predicted based on the defect frequencies of the outer ring, the inner ring, and the ball calculated from the specifications of the rolling bearing 10, the rolling bearing 10 A member having a defect such as a scratch is identified. Even in this case, when a member having a defect such as a flaw in the rolling bearing 10 cannot be specified, it is determined that the vibration detected by the vibration sensor 101 is a foreign object sound generated from the rolling bearing 10.

  Thereby, even when a pulse based on a defect such as a flaw appears only in a part of the vibration signal within the measurement time, the defect such as a flaw can be detected effectively.

  The present invention is a rolling bearing defect detection device capable of detecting a defect such as a flaw even when a pulse based on a defect such as a flaw appears only in a part of a vibration signal within a measurement time. This is useful for a rolling bearing defect detection device and a detection method for detecting defects such as flaws by detecting vibrations generated from the rolling bearing.

  In addition, this invention is not limited to embodiment mentioned above, A deformation | transformation, improvement, etc. are possible suitably. In addition, the material, shape, dimension, numerical value, form, number, arrangement location, and the like of each component in the above-described embodiment are arbitrary and are not limited as long as the present invention can be achieved.

It is a figure which shows schematic structure of the defect detection apparatus of the rolling bearing which is embodiment which concerns on this invention. It is a figure which shows schematic structure of the bearing vibration measuring apparatus used in embodiment which concerns on this invention. It is a flowchart for demonstrating the process sequence in the defect detection apparatus of the rolling bearing which is embodiment which concerns on this invention. (A) is a figure which shows the waveform of the vibration signal detected from the rolling bearing without a flaw, and (b) is a figure which shows the waveform at the time of performing autocorrelation function calculation about the envelope signal of a rolling bearing without a flaw It is. (A) is a figure which shows the waveform of the vibration signal detected from the rolling bearing with a crack in an outer ring, and (b) is when autocorrelation function calculation is performed about the envelope signal of the rolling bearing with a crack in an outer ring. It is a figure which shows a waveform. (A) is a diagram showing a waveform of a vibration signal detected from a rolling bearing with a flaw in the inner ring, and (b) is a case where autocorrelation function calculation is performed on the envelope signal of the rolling bearing with a flaw in the inner ring. It is a figure which shows a waveform. (A) is a figure which shows the waveform of the vibration signal detected from the rolling bearing with a ball flaw, and (b) is the case where autocorrelation function calculation is performed about the envelope signal of the rolling bearing with a ball flaw. It is a figure which shows a waveform. (A) is a figure which shows the waveform of the vibration signal at the time of detecting the foreign material sound which has generate | occur | produced from the rolling bearing, and (b) is the waveform at the time of performing autocorrelation function calculation about the envelope signal of foreign material sound FIG. It is a figure for demonstrating a peak count system. It is a figure which shows schematic structure of the defect detection apparatus of the conventional rolling bearing. (A) is a diagram showing a waveform of a vibration signal having a peak indicating that the rolling bearing has a flaw all over the measurement time, and (b) shows a flaw in the rolling bearing over the whole measurement time. It is a figure which shows the waveform which performed the envelope FFT process on the vibration signal which has the peak which shows that there exists. (A) is a figure which shows the waveform of the vibration signal which the pulse which shows that there is a crack in a rolling bearing in a part in measurement time, and (b) is a rolling bearing in part in a measurement time. It is a figure which shows the waveform which performed the envelope FFT process on the vibration signal which has the peak which shows that there exists a defect.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Rolling bearing 11 Inner ring 12 Outer ring 13 Ball 100 Defect detection apparatus of rolling bearing 101 Vibration sensor (vibration detection means)
102 signal amplification circuit 103 low-pass filter (frequency filter means)
104 A / D converter (analog / digital conversion means)
105 Envelope processing unit (envelope signal processing means)
106 Autocorrelation function calculator (autocorrelation function calculation means)
DESCRIPTION OF SYMBOLS 107 Effective value calculator 108 dB magnification setting part 109 Multiplier 110 Comparator 111 Peak counter 112 Defect determination part (1st defect determination means, 2nd defect determination means)
200 Bearing vibration measuring device 210 Spindle arbor 220 Bearing preload application mechanism 230 Drive mechanism (piezo element)
240 Piezo control circuit

Claims (6)

A predetermined thrust load is applied to the rolling bearing, the outer ring and the inner ring are relatively rotated at a predetermined number of rotations to detect vibrations generated from the rolling bearings, and defects in the rolling bearings are detected based on the detected vibrations. A defect detection device for a rolling bearing,
Vibration detecting means for detecting vibration generated from the rolling bearing and generating a vibration signal;
Frequency filter means for extracting a vibration signal in a predetermined frequency band from the vibration signal;
Analog-digital conversion means for converting the extracted vibration signal of a predetermined frequency band into a digital signal at a predetermined sampling frequency;
An envelope signal processing means for generating an envelope signal by performing envelope processing on the vibration signal converted into the digital signal;
When the number of peaks in the envelope signal that is greater than or equal to the threshold value exceeds a predetermined set value, any of the members constituting the rolling bearing has a defect and is generated from the rolling bearing other than the defect. First defect determination means that tentatively determines that there is a foreign object sound to be detected;
An autocorrelation function for the envelope signal when it is determined by the first defect determination means that any of the members constituting the rolling bearing is defective and there is a foreign object sound generated from the rolling bearing other than the defect Autocorrelation function computing means for computing
When the position of the time delay element (lag) where the peak intensity indicating that the correlation is strong in the calculation result of the autocorrelation function appears and the member constituting the rolling bearing is defective, from the bearing specifications of the rolling bearing By collating the position of the time delay element for each member predicted by performing the autocorrelation function calculation based on the defect frequency specific to each member to be calculated and the predetermined sampling frequency, A second defect determining means for identifying the member having a rolling bearing defect;
A defect detection device for a rolling bearing, comprising:   The said 2nd defect determination means determines with the vibration detected by the said vibration detection means being the foreign material sound which generate | occur | produces from the said rolling bearing, when a member with a defect cannot be specified. Defect detection device for rolling bearing as described.   3. The rolling bearing defect detection apparatus according to claim 1, wherein the threshold value is a value obtained by multiplying an effective value of the envelope signal by a predetermined dB magnification.   The microcomputer unit performs processing in at least one of the analog / digital conversion means, the envelope signal processing means, the first defect determination means, the autocorrelation function calculation means, and the second defect determination means. The rolling bearing defect detection device according to any one of claims 1 to 3, wherein the defect detection device is executed by the program. A predetermined thrust load is applied to the rolling bearing, the outer ring and the inner ring are rotated relative to each other at a predetermined number of rotations, and vibrations generated from the rolling bearings are detected, and defects in the rolling bearings are detected based on the detected vibrations. A rolling bearing defect detection method comprising:
Detecting vibration generated from the rolling bearing to generate a vibration signal;
Extracting a vibration signal of a predetermined frequency band from the vibration signal;
The vibration signal limited in the frequency band is converted into a digital signal at a predetermined sampling frequency,
An envelope signal is generated by performing envelope processing on the vibration signal converted into the digital signal,
When the number of peaks in the envelope signal that is greater than or equal to the threshold value exceeds a predetermined set value, any of the members constituting the rolling bearing has a defect and is generated from the rolling bearing other than the defect. Tentatively determined that there is a foreign object sound,
An autocorrelation function is calculated for the envelope signal when it is determined that any of the members constituting the rolling bearing has a defect and there is a foreign object sound generated from the rolling bearing other than the defect,
When the position of the time delay element (lag) where the peak intensity indicating that the correlation is strong in the calculation result of the autocorrelation function appears and the member constituting the rolling bearing is defective, from the bearing specifications of the rolling bearing By collating the position of the time delay element for each member predicted by performing the autocorrelation function calculation based on the defect frequency specific to each member to be calculated and the predetermined sampling frequency, Identifying said member with a defective rolling bearing,
A defect detection method for a rolling bearing characterized by the above.   The rolling according to claim 5, wherein if the member having a defect cannot be identified by collating the position of the time delay element, the detected vibration is determined to be a foreign object sound generated from the rolling bearing. Bearing defect detection method.

Images