JPH01152335A - Abnormality diagnostic apparatus for roller bearing - Google Patents

Abstract

PURPOSE:To judge the degree and a part of damage to a bearing to be diagnosed with an automatic diagnosis thereof, by inputting and analyzing the name, shape and revolutions of the bearing being diagnosed to take in an arithmetic data. CONSTITUTION:A name, a geometrical shape and nominal inner ring revolutions and the like of a bearing to be diagnosed are inputted into a diagnosis conditions input circuit 22. According to a diagnosis algorithm stored in a memory circuit 25, a central processing circuit 24 controls a filter controlling circuit 4, a frequency analysis circuit 18, a spectrum control circuit 20 and a display circuit 28 in the apparatus and brings data in from a circuit 14 for computing a Q value expressed by the formula and a frequency analysis circuit 18 in the apparatus to perform a self-diagnosis of a bearing to be diagnosed. That is, a solid sound generated from the bearing being diagnosed during the rotation is detected with an acceleration detector 2 set near the bearing being diagnosed and a waveform processing and a computation are added to this operation to determine a Q value. On the other hand, a frequency spectrum is obtained to display the results of judgement of the Q value and a damaged part on a display circuit 28.

Description

[Detailed Description of the Invention] [Field of Industrial Application] This invention is for accurately diagnosing abnormalities in rolling bearings (hereinafter simply referred to as bearings), which are frequently used as bearings in rotating machines, while the machine is operating. Regarding the device.

[Conventional technology]

This type of device uses a piezo element to detect vibration acceleration, and calculates its root mean square value (hereinafter referred to as RMS value).
Alternatively, there are already commercially available products that use a peak value (hereinafter referred to as peak value) as an evaluation measure. However, since the waveform of vibration acceleration varies significantly depending on the type of bearing defect,
Accurate diagnosis cannot be made by displaying only a simple test value or peak value. In order to improve diagnostic accuracy, it is traditional and highly reliable to distinguish the types of defects using waveform processing, etc., and to diagnose based on the determination level for each type of defect, which has not yet been determined. Due to the necessity, it requires the intervention of a high-level engineer, and it is difficult to make it portable, making it unsuitable for on-site diagnosis. Another way to improve diagnostic accuracy is to display vibration levels corrected for differences in defect types, that is, differences in waveforms.
It is possible to diagnose using only this level display, but
Although it is suitable for on-site diagnosis, it requires know-how to determine the amount of correction, and that know-how is not theoretically clear, but is based on empirical rules for symptomatic treatment. I have not seen it realized as a device.

Therefore, by eliminating the drawbacks of these conventional techniques, (a) there is no need to perform waveform analysis, and therefore there is no need to provide a large-capacity information processing device, and as a result, a device that is portable is provided. thing.

(b) To provide a device that allows judgment of pass/fail based only on the value indicated by the meter, and therefore allows even a beginner engineer to perform diagnosis.

(c) To provide a device that can perform accurate diagnosis regardless of the type of defect.

In order to achieve this purpose, a rolling bearing abnormality diagnosis device has already been proposed, which is characterized by introducing a new diagnostic scale that takes into account the difference in evaluation level depending on the waveform from a large amount of this type of diagnostic data. (Special Publication No. 6l-57491).

[Problem that the invention seeks to solve]

However, as for this rolling bearing abnormality diagnosis device, l) the following device is generally known in the past.

That is, the vibration waveform generated from the bearing detected by the acceleration detector 2 installed near the bearing to be diagnosed is passed through a filter to detect solid sound of high frequency vibration of 1 kHz or more and IQkHz or less, and this is used to diagnose rolling bearing abnormalities. Therefore, if disturbance vibrations caused by the meshing frequency of the gears on the shaft supported by the rolling bearing to be diagnosed are mixed in as noise in this frequency band, this will cause errors in the diagnosis.

2) Fig. 2 is a 3° cross-sectional view of a rolling bearing to be diagnosed; (a) is a cross-sectional view of a ball bearing, and (b) is a cross-sectional view of a cylindrical roller bearing. Inner ring 32 of this bearing 30. Rolling element 34
．． The damage 38 occurring on the outer ring 36 is caused by the geometrical shape of the conventional bearing 30, that is, the diameter of the rolling elements 34: d,
Pitch diameter of rolling element 34: D.

The contact angle of the rolling elements 34: α (α 0'), the number Z of the rolling elements 34, the nominal rotation speed per second of the inner ring 32: fsn, and the result calculated by the following formula, and the envelope obtained by the envelope detection circuit. By comparing the waveform spectrum with the waveform spectrum, the damaged site of the flaw 38 is determined.

(1) Occurrence frequency f+(
2) Occurrence frequency fO (3) when there is damage to the outer ring raceway surface
) Occurrence frequency fb when the rolling element is damaged The above formula (
In 1), (2), and (3), the nominal inner ring rotation speed fs
Because n is measured in advance with this device and another device, or calculated using the value obtained from the value displayed on the motor,
The actual rotation speed fs of the inner ring may be slightly different from the actual rotation speed fs at the time of diagnosis using this device. For example, in an induction motor, slippage differs depending on the magnitude of the load, so the actual inner ring rotation speed fs may differ from the value fsn determined from the value displayed on the motor. For this reason, ft, fo.

There is a problem that fb cannot be calculated accurately and the damaged area may not be detected.

The present invention solves the above-mentioned problems of the conventional device, and 1) uses a variable filter in which the signal pass band of the signal to be passed is variable, so as to eliminate the solid sound of the bearing mixed in the electrical signal from the acceleration detector. 2) The actual rotational speed of the inner ring is measured according to the spectrum analysis of the envelope waveform, and this can be used to perform accurate calculations.
An object of the present invention is to provide a rolling bearing abnormality diagnosis device that can accurately detect a damaged area.

[Means for solving problems]

In order to solve the above problems, the present invention provides an acceleration detector that is installed in a rolling bearing and converts the solid sound generated by the rolling bearing into an electrical signal in the form of an acceleration signal and outputs it, and A filter that extracts only the frequency band from the electric signal component, an envelope detection circuit that obtains the envelope waveform of the waveform that has passed through this filter, and a root mean square value of the envelope waveform output from the envelope detection circuit. a root mean square value calculation circuit to obtain a peak value, a peak value calculation circuit to obtain a peak value of the envelope waveform, and from the root mean square value and the peak value, where Vrms ; root mean square value Vpeak ; peak value' a, b; A rolling bearing abnormality diagnosis device comprising a Q value calculation circuit for calculating a constant, a) a variable filter whose signal pass band of a signal to be passed is variable by control of a filter control circuit; ) a frequency analysis control circuit that switches and inputs the output from the variable filter and the output from the envelope detection circuit to the frequency analysis circuit according to a signal from a central processing circuit (to be described later), and changes the analysis band; c) a spectrum control circuit that selectively displays a spectrum to be displayed by controlling a display circuit using the output of the frequency analysis circuit and a signal from a central processing circuit described later; and 2) the name and geometric shape of the rolling bearing to be diagnosed.

A diagnostic condition input circuit for inputting the nominal inner ring rotation speed, etc., and e) Connected to the filter control circuit, Q value calculation circuit 2, frequency analysis control circuit 1, frequency analysis circuit, spectrum control circuit 1, disallowance condition input circuit, and display circuit. a) a storage circuit, a data output circuit, and a microprinter attached to the central processing circuit; It shall be possible to diagnose whether the rolling bearing is good or bad, and where scratches and defects occur from the displayed content of the parts.

[For production]

According to the present invention, by inputting the name, geometrical shape, nominal inner ring rotation speed, etc. of the bearing to be diagnosed into the diagnostic condition input circuit provided in the device, the central processing circuit also provided in the device can input the memory circuit,
Equipped with a data output circuit and a micro printer, the central processing circuit controls the device's filter control circuit, frequency analysis circuit, spectrum control circuit, and display circuit according to the diagnostic algorithm stored in the memory circuit, and calculates the Q value of the device. Circuit 9 Data from the frequency analysis circuit is taken in to perform automatic diagnosis of the target bearing. That is, the solid sound generated from the rotating target bearing is detected by an acceleration detector installed near the target bearing, and this is subjected to waveform processing and calculation to determine the Q value, while the frequency vector is obtained and the Q value is determined. As a result of the value determination, the damaged area is displayed on the display circuit, so that the degree of damage and the damaged area of the target bearing can be determined.

In the present invention, a variable filter is controlled by a filter control circuit to make the signal pass band variable, thereby removing disturbances unrelated to bearing abnormality diagnosis, such as gear mesh frequency, and further reducing the rotation speed of the bearing inner ring to the conventional nominal rotation speed. Instead of using a number, the actual rotational speed is determined from the measurement results and used in calculations, so the accuracy of the diagnostic results is improved compared to the conventional example.

〔Example〕

An embodiment of the invention is shown in FIG. Further, FIG. 2 is a sectional view showing a rolling bearing 30 that is a subject of diagnosis, and shows an inner ring 32 as a bearing component. Rolling element 34. It consists of an outer ring 36, (a)
3 shows a ball bearing, 38 shows a cylindrical roller bearing, and 38 shows a scraper.

The configuration and operation of the present invention will be explained with reference to FIG. The acceleration detector 2 is brought into contact with the vicinity of the bearing 30 to be diagnosed, and converts solid sound generated from the bearing 30 into a vibration acceleration CG.
) is detected as an electrical signal. The solid sound of the bearing 30 is a high frequency vibration of 1 kHz or more and 10 kHz or less, and a piezo element having good high frequency response characteristics in this band is used for the acceleration detector 2. The electrical signal detected by this acceleration detector 2 is the same as the conventional example (as it also contains vibration components generated from sources other than the bearing 30), so the signal pass band of the signal to be passed is controlled by the filter control circuit. The variable filter 6 allows the signal to pass through and removes unnecessary components and noise for diagnosis.

FIG. 3 is a rolling bearing abnormality diagnosing device according to the present invention, and various waveform diagrams for explaining the operation of the device. FIG. 3(A) shows, as a typical example, the original waveform detected by the acceleration detector 2 when the outer ring 36 of the bearing 30 has a flaw 38.
The horizontal axis shows time (unit: 10-'' seconds Hmsec), and the vertical axis shows vibration acceleration (G).

Figure 3 (CB) shows the frequency spectrum of the original waveform in Figure 3 (A), where the main component of vibration is 1.63 kHz.
It turns out that. On the other hand, Fig. 3(E) shows the frequency spectrum of the original wave of Fig. 3(A) as well as Fig. 3(B), and the meshing frequency of the gear is 3.78kHz (
30cps x 126 teeth), (cps is cycle p
er 5econd).

The electrical signal of the vibration detected by the acceleration detector 2 is sent to the bearing 3.
Since vibration components originating from sources other than 0 are also included, in order to remove them, the variable filter 6 is used to 1) usually remove vibrations below 1 kHz and above 10 kHz, and extract only high-frequency bearing vibrations. For this reason, the variable filter 6 is normally set to have a low pass filter (hereinafter referred to as LP filter) at 10 kHz and a bypass filter (hereinafter referred to as HPy filter) at 1 kHz by the filter control circuit 4.

2) In the display circuit 28 described later, if a predominant spectrum other than bearing high frequency vibration appears at 3.78 kHz as shown in FIG. Then, the filter control circuit 4 sets the variable filter 6 to a narrow band pass filter centered at 1.63 kHz.
Remove vibration components other than kHz components. Note that the band of this variable filter 6 is continuously variable.

The output from the variable filter 6 is input to an envelope detection circuit 8. Here, vibrations caused by damage included in the vibrations of the bearing 30 are extracted. That is, the main component of the vibration of the bearing 30 is high frequency vibration of 1 kHz or more, but when damage such as flaws 38 appears on the rolling surface of the bearing 30 or the rolling elements 34, this high frequency vibration becomes a low frequency (500 Hz or less) component. Modulated by Since the level and frequency of this low-frequency vibration changes depending on the type of damage, signal processing (demodulation) using an envelope detection method is performed and used for bearing diagnosis. FIG. 3(C) shows the envelope waveform of FIG. 3(A).

Next, in order to calculate the Q value representing the degree of damage to the bearing, a root mean square value calculation circuit 10 (hereinafter referred to as RMS value calculation circuit 1
0) is input from the envelope detection circuit 8, and here the RMS value which is the effective value of the envelope waveform for 0.1 seconds; V
Calculate rms. It is also input from the envelope detection circuit 8 to a peak value calculation circuit 12 (hereinafter referred to as peak value calculation circuit 12), where the peak value (hereinafter referred to as peak value) is the highest value of the envelope waveform for 0.1 seconds. ): Calculate Vpeak. For these RMS value calculation circuit 10 and peak value calculation circuit 12, commercially available IC dedicated arithmetic elements are used.

The Q value calculation circuit 14 receives Vr from the RMS value calculation circuit 10.
ms and Vpeak are input from the peak value calculation circuit 12, and the Q value of the diagnostic scale is calculated according to the following equation.

However, in equation 4), Vrms is the RMS of the envelope waveform.
The value Vpeak is the Peak value of the envelope waveform, and a and b are constants.

The Q value calculated by the Q value calculation circuit 14 is input to a central processing circuit 24, which will be described later.

On the other hand, a signal from the variable filter 6, a signal from the envelope detection circuit 8, and a central processing circuit 24 (described later)
A frequency analysis control circuit 16 controls a frequency analysis circuit 18 using signals from the frequency analysis circuit 18.

In this frequency analysis circuit 18, the next frequency analysis is performed according to the instructions from the central processing circuit 240.

1) The envelope line waveform shown in FIG. 3(C) from the envelope line detection circuit 8 is subjected to frequency analysis to obtain the frequency spectrum shown in FIG. 3(D). Figure 3 (D) is Figure 3 (C
) shows the frequency spectrum of the envelope waveform of the dominant component (78,156,211°233,38
9, 467 Hz) is the frequency spectrum caused by the flaw 38 on the outer ring 36. Further, the frequency spectrum resulting from the frequency analysis is input to the central processing circuit 241C, and dominant components of frequencies h, fo, and fb corresponding to the damaged area described in the prior art example are extracted.

2) The acceleration signal detected from the bearing 30 includes this bearing 3
It always includes a vibration component due to the unbalance of the rotor that is supported and rotates at 0, but this component is caused by the rotation speed fs of the inner ring
lc matches. The frequency analysis circuit 18 analyzes the frequency of the output waveform from the variable filter 6, and inputs the frequency spectrum to the central processing circuit 24, where the inner race 320 rotation speed fs is determined.

However, in this case, it is necessary to set fs to be the signal passband of the variable filter 6.

3) Normally, the variable filter 6 diagnoses the bearing 30 by setting the LP filter to 10 kHz and the HP filter to 1 kHz, but this band is the same as that of the conventional fixed band filter. , when a 3.78 kHz disturbance component is included in addition to the high frequency vibration (1.63 kHz component) of the bearing 30 as shown in FIG. must change the characteristics of For this purpose, the output waveform from the variable filter 6 is frequency analyzed by the frequency analysis circuit 18, and the frequency spectra of FIGS. 3(B) and 3(E) are displayed on the display circuit 28 (to be described later) to recognize the disturbance components. and change the characteristics of the variable filter 6.

The frequency analysis control circuit 16 switches the signal from the variable filter 6 and the signal from the envelope detection circuit 8 and outputs it to the frequency analysis circuit 18 according to the command from the central processing circuit 24.
At the same time, the analysis band of the frequency analysis circuit 18 is also changed. For example, the frequency fI, fo corresponding to the damaged area in 1) above
, fb, when performing an analysis to extract the prominent components of fb, switch to the signal from the envelope detection circuit 8 and set the analysis band to 500.
Set to Hz. When analyzing the inner ring 320 rotation speed fs in 2) above, the center frequency of the filter control circuit 4 is set by combining the nominal rotation speed fsnlc of the inner ring 32 of the target bearing 30 inputted in the diagnostic condition input circuit 22 described later with this nominal rotation speed. f
sn K is set, and the analysis band of the frequency analysis circuit 18 is set to 2fsn, which is twice the nominal rotational speed fsn. The signal from the variable filter 6 is also input to analyze the true rotational speed fs of the inner ring 32.

When analyzing the disturbance component in 3) above, input the signal from the variable filter 6 and set the analysis band to 10k.
Set to Hz. In this case, the frequency spectrum of the frequency analysis circuit 18 is determined by the waveform shown in FIG. 3(B) or FIG. 28
to be displayed.

The central processing circuit 24 controls the filter control circuit 4, the frequency analysis control circuit 16, the spectrum control circuit 20, and the display circuit 28 according to the diagnostic algorithm, and also controls the Q value data from the Q value calculation circuit 14 and the frequency analysis circuit 18. frequency spectrum data, diagnostic condition input circuit 22
Target bearing name and geometric shape, which is diagnosis target bearing data from .

The abnormality of the target bearing is automatically diagnosed using data such as the nominal inner ring rotation speed fsn. Furthermore, the data stored in the RAM of the storage circuit 25 is outputted from the data output circuit 26 via the central processing circuit 24 and can be input to an external large comviator system or the like. This data output circuit 2
6 is GP-IB (forming group) or R8232C (forming group)
Use LSIs such as In addition, the diagnostic results and frequency spectrum displayed on the display circuit 28 are displayed on the micro printer 27.
It can be printed by Furthermore, in the central processing circuit 24,
Commercially available microcenter processor unit (μCPU)
The diagnostic algorithm is stored in the ROM of the storage circuit 25. The functions of the central processing circuit 24 are as follows: 1) Before starting the diagnosis, the name of the bearing to be diagnosed input from the diagnosis condition input circuit 22, the bearing classification format according to JIS, and the diameter d of the rolling element 34 shown in FIG. , rolling element 34
Pitch diameter, number of rolling elements 34 2. Geometric shape data such as contact angle α of rolling element 34, nominal inner ring rotation speed fs
n is stored in the RAM of the storage circuit 25.

2) In order to measure the actual machine inner ring rotation speed fs, set the filter control circuit 4 and frequency analysis control circuit 16 according to the nominal inner ring rotation speed fsn, and input the frequency spectrum and torque from the frequency analysis circuit 18 to the central processing circuit 24. . Comparing this frequency spectrum and the nominal inner ring rotation speed fsn, f
A dominant spectrum near sn is detected, and the actual inner wheel 320 rotation speed fs is detected from the frequency and stored in the RAM of the storage circuit 25.

3) To perform diagnosis, connect filter control circuit 4 to LP filter A.
/Evening 10kHz K, HP filter 1kHz
The signal from the envelope detection circuit 8 is taken into the frequency analysis circuit 18 under the control of the frequency analysis control circuit 16, and the analysis frequency band is set to 500 Hz. Next, the Q value data from the Q value calculation circuit 14 is transferred to the central processing circuit 24.
The quality of the bearing 30 is determined by comparing it with the criteria stored in advance in the ROM of the storage circuit 25. This standard was determined through experiments, and a Q of 10 or less is considered good.
Scores from 10 to 50 are considered caution, and scores above 50 are considered dangerous.

At the same time, the frequency spectrum from the frequency analysis circuit 18 is taken into the central processing circuit 24, and the h corresponding to the damaged area is
, fo, and fb. This extraction is performed based on the actual rotational speed fs of the inner ring 32 detected in step 2) and the geometrical shape of the bearing 30 input in advance to solve the above-mentioned [problem to be solved by the invention]. Compare fI, fo, fb calculated by using fs instead of fsnO in equation (1) # (2) 5 (3) described in Section 1, and the frequency spectrum to find out the prominence corresponding to fi, fo, fb. Extract the resulting spectrum.

The determination result of the Q value and the damaged area are displayed on the display circuit 28, and the operator can determine whether the rolling bearing 30 is good or not based on the display contents.
Good, Caution, Danger), flaw 38° The location of the defect can be diagnosed. In this way, for example, (1) if the Q value is good and there are no fi, fo, or fb, the rolling bearing is diagnosed as being good.

(2) Even if the Q value is good, if any of h, fo, and fb appears, caution and diagnosis should be carried out, and the subsequent diagnosis cycle should be shortened.

(3) Even if the Q value is determined to be in a caution or replacement state,
If there is no fi, fo, or fb, the diagnosis is that the grease is running out or that normal, uniform wear is progressing.

(4) On the other hand, if the factor (3) above is not present and only the Q value is high, the rolling bearing is normal, but the diagnosis is that it is caused by disturbances other than the bearing.

(5) If the condition worsens from (2) above and the Q value is judged as caution or replacement, and any of fs, fo, or fb appears, if fI, it is the inner ring, if fo, it is the outer ring.
With fb, it is possible to diagnose the damage to rolling elements 5 to 5.

(6) On the other hand, as in (3) above, only the Q value is high and it is determined that the state requires caution or replacement, and f+, fo, f
If there is no b and it seems that the disturbance is due to some other disturbance, specify the frequency spectrum display of the original waveform to the spectrum control circuit 20, and display the frequency spectrum as shown in Fig. 3 (B) or Fig. 3 (E). can be displayed on the display circuit 28. If this display shows a normal spectrum and no disturbance as shown in FIG. 3(B), it can be confirmed that the grease has run out or normal wear is progressing as shown in (3) above. However, if the spectrum is as shown in WX3 (E), it can be concluded that the Q value is high due to the 3.78 kHz disturbance.
High frequency vibration of bearing 30 read from 1.63 kHz
z is input from the diagnostic condition input circuit 22, the characteristic of the variable filter 6 is set to narrow band in the filter control circuit 4, and 3
．． By removing the 78 kHz component and re-diagnosing the bearing, we were able to accurately diagnose the bearing 3 using only the vibration waveform data of the high-frequency vibration of the bearing.
0 abnormality diagnosis can be performed.

Based on the above diagnostic results, if the results are good, continue operation, if caution is required, shorten the diagnostic cycle, and if possible, replace the bearing parts or replace the bearing itself depending on the location of the damage. It is possible to take measures such as replacing the bearing, thereby preventing serious accidents in the machine due to damage to the bearing.

〔Effect of the invention〕

In the present invention, by inputting the name, shape, rotation speed, etc. of the bearing to be diagnosed into the diagnostic condition input circuit provided in the device, the central processing circuit also provided in the device controls the control circuit of the device, and performs analysis and calculations. The data is imported, the target bearing is automatically diagnosed, and the results are displayed on the display circuit.
Therefore, the degree of damage to the target bearing and the location of the damage can be determined.

Further, according to the present invention, l) the signal passband is made variable by a narrow band variable filter;
Since disturbances other than high-frequency bearing vibration can be removed, diagnostic accuracy can be improved and the range of models targeted for diagnosis can be expanded.

2) Since the inner ring rotation speed of an actual machine can be easily measured without the need for special preparations, it is possible to accurately detect the location of damage without making a mistake.

Depending on the above diagnostic results, replace the target bearing as necessary.
Since inspection and maintenance such as grease replenishment can be carried out appropriately and reliably, serious accidents of the rotating machine due to bearing damage can be prevented, and the reliability and operating rate of the rotating machine can be greatly improved. be able to.

[Brief explanation of the drawing]

Fig. 1 is a schematic diagram showing an embodiment of the present invention, Fig. 2 is a cross-sectional view of a rolling bearing to be diagnosed, (a) is a cross-sectional view of a ball bearing, (b) is a cross-sectional view of a cylindrical roller bearing, Figure 3 shows various waveform diagrams for explaining the operation of the device of the present invention.
A) is a diagram showing the original waveform detected by an acceleration detector when there is a scratch on the outer ring of a rolling bearing.
Figure 3 (C) shows the frequency spectrum of the original wave in A).
) is a diagram showing the envelope waveform of the original wave in Figure 3 (A),
Figure (D) is a diagram showing the frequency spectrum of the enveloping line waveform of Figure 3 (C), and Figure 3 (E) is a diagram showing the frequency spectrum of the envelope line waveform of Figure 3 (B).
It is a figure which shows the frequency spectrum of the original wave of figure (A). 2... Acceleration detector, 4... Filter control circuit, 6
...Variable filter, 8...Envelope detection circuit, 10 root mean square value calculation circuit (RMS value calculation times N), 12
...Peak value calculation circuit (peak value calculation circuit), 14...
・Q value calculation circuit, 16... Frequency analysis control circuit, 18
...Frequency analysis circuit, 2o...Spectrum control circuit, 22...Diagnostic condition input circuit, 24...Central processing circuit, 28...Display circuit, 3o...Rolling bearing (bearing). Figure 1 Figure 2

Claims (1)

[Scope of Claims] 1) An acceleration detector installed in a rolling bearing, which converts solid sound generated by the rolling bearing into an electrical signal in the form of an acceleration signal and outputs the electrical signal; A filter that extracts the components, an envelope detection circuit that obtains the envelope waveform of the waveform that has passed through this filter, and a root mean square value calculation circuit that obtains the root mean square value of the envelope waveform output from the envelope detection circuit. From the root mean square value and the peak value, Q=Vrms×Vpeak×[a|Vpeak/Vrm
s-5|+b] However, Vrms; root mean square value Vpeak; peak values a, b; Q value calculation circuit for calculating constants; a variable filter whose signal passband is variable under the control of a filter control circuit; and b) a frequency analysis circuit using a signal from a central processing circuit that outputs the output from the variable filter and the output from the envelope detection circuit as described below. a) a frequency analysis control circuit that switches input to input signals and changes the analysis band, and c) spectrum control that controls a display circuit to select and display spectra to be displayed using the output of the frequency analysis circuit and a signal from a central processing circuit described later. d) A diagnostic condition input circuit for inputting the name, geometrical shape, nominal inner ring rotation speed, etc. of the rolling bearing to be diagnosed, and e) The above-mentioned filter control circuit, Q value calculation circuit, frequency analysis control circuit, frequency A central processing circuit that connects with an analysis circuit, a spectrum control circuit, a diagnostic condition input circuit, and a display circuit to mutually transmit signals, f) A storage circuit, a data output circuit, and a microprinter attached to the central processing circuit. 1. A rolling bearing abnormality diagnosing device, which is capable of diagnosing the quality of a rolling bearing, and the location of a flaw or defect based on the determination result of a Q value displayed on a display circuit and the display content of a damaged location. 2) In the device according to claim 1, the frequency analysis circuit detects the actual rotational speed of the inner ring of the rolling bearing, and calculates frequencies caused by damage to the inner ring, outer ring, and rolling elements of the rolling bearing. A rolling bearing abnormality diagnosis device characterized in that the frequency spectrum caused by the damage is extracted using the actual rotation speed.