JPS61294214A - Abnormality monitor for rolling bearing - Google Patents

Abstract

PURPOSE:To enable to detect abnormality of a bearing in an early stage, by detecting the abnormality with using jointly a detection by means of acoustic emission and a detection power by means of vibration acceleration. CONSTITUTION:An acoustic emission conversion element 1 and an acceleration detector 2 are installed on an outer ring, and signals detected respectively by these acoustic emission conversion element 1 and acceleration detector 2 are processed with discriminators 7, 8. At the discriminators 7, 8, the signals are counted by counters 9, 10 and stored in memories 11, 12 every unit time, only in case that the signals go beyond a preset threshold value and a duration t is longer than a set period. Therefore, a control unit 13 makes a diagnosis from a counted value of the counters 9, 10 and a stored content of the memories 11, 12, so that an occurrence of abnormality of a bearing can be detected in an early stage.

Description

DETAILED DESCRIPTION OF THE INVENTION "Field of Industrial Application" The present invention relates to a rolling bearing abnormality monitoring device that detects abnormalities in rolling bearings.

``Prior Art'' Conventionally, vibration acceleration has been mainly used as a method for detecting abnormalities in bearings, and many diagnostic instruments for bearings that utilize this have been released.

Recently, research has been carried out to utilize the crack stand @ (hereinafter referred to as AE), which is used for inspections of pressure vessels, etc., for diagnosis of bearings, and this has actually been made into a product. There is also. (For example, PHYSI
CAL ACOυ5TIC5C0IIPORATION
Acouste ° Ik Emin Silane Bearing Monitor 6110.6120).

``Problem that the invention aims to solve'' Detecting abnormalities in equipment earlier is possible by predicting the lifespan of equipment earlier and stopping it in a planned manner.
This not only reduces downtime losses, but also prevents sudden major accidents, which can be said to be extremely effective.

Experiments show that the timing of abnormality detection of fatigue peeling on the rolling surface and rolling element of a rolling bearing is earlier with the detection ability of AE than with detection using vibration acceleration, which is used for abnormality diagnosis of bearings. People who get results.

However, although AE can detect peeling, cracking, etc. at an early stage, experimental results have shown that it has lower detection sensitivity than acceleration when it comes to poor lubrication such as oil loss.

"Means for Solving Problems" This invention uses detection by AE and detection power by vibration acceleration in combination to take advantage of the strengths of each in order to detect any abnormalities in rolling bearings at an early stage. be.

Cracking, peeling, etc. can be detected early by detecting the toe, hole, and AE, and lubrication failure can be detected early by detecting acceleration.

AE occurs as the separation grows, but if the growth stops, it will no longer occur. Therefore, AE is good for checking the progress of a wound, but to check the extent of the wound, you need to check the number of past occurrences! The only way to diagnose is to multiply A.

In this respect, detection based on acceleration is more suitable than detection based on AE for determining the current degree of damage.

In other words, it is better to use AE to discover the occurrence of scratches, and to monitor the subsequent state using both vibration acceleration and AE.

"Embodiment" FIG. 1 shows an embodiment of the rolling bearing abnormality monitoring device according to the present invention.
A transducer l and an acceleration detector 2 are installed, and these AE
The signals detected by the IRIA element 1 and the acceleration detector 2 are each amplified by the respective amplifiers 3.4, and the respective amplified outputs are limited to a predetermined frequency band by the respective P wave amplifiers 5 and 6. , whose p-wave output is processed by a discriminator 7.8.

In the discriminator 7.8, only when a preset threshold value is exceeded and the duration t of the waveform is longer than the set time is counted by the counters 9 and 10 using the event counting method. These counts are memo IJII, 1 for each unit time.
2. The control is performed by the control unit 13, and the control unit 13 stores the count values of the counters 9 and 10, and the memory 11.1.
Diagnosis is performed based on the memory contents in step 2 according to the following.

a) Counter 9. It is normal when neither IO is counted.

b) When only counter 9 (AE output) starts counting and counter 10 (acceleration signal) does not start counting, it is very small! Be careful as All separation occurs.

C) On the other hand, if only the counter 10 (acceleration signal) starts counting, this is a disconnection stage due to poor lubrication, so be careful.

d) both counters9. An abnormality occurs when both IOs start counting.

However, AE occurs with the growth of MIIM,
one time? Even after , 11 distances occur, the growth of jυ1 distances stops, and if it does not grow, it will not be counted. Therefore, AI by counter 9! Once the signal count has started, the warning will not be canceled unless it is manually reset.

The control unit 13 displays the diagnosis results based on the above a) to d) on the display 14.

ll1 Accelerated life test results were conducted on a thrust ball bearing (151105) at a rotation speed of 1500 rpm and a thrust load of 600 ktr. The results are shown in FIGS. 2 and 3. In these figures, solid lines 16 and 17 indicate the count value of the counter 9, that is, the AE signal, and dotted lines 18 and 19 indicate the count value of the counter 10, that is, the acceleration signal, respectively. FIG. 2 shows a case where rolling fatigue peeling occurs on the outer ring surface, and FIG. 3 shows a case where the oil runs out. Both were confirmed by observing the bearing under test during the experiment. The pass frequency band of the p wave generator 5 for the AE signal when obtaining each result is 100kHz-1 former 1z
The pass frequency band of the p-wave device 6 for the acceleration signal is set to 1.
Set amplifier 3 to 0 to 30 k II z, respectively.
．． The amplification degrees of 4 and 4 were both set to 40 dB, and the thresholds of discriminator 7.8 were both set to 0.2 V.

Particularly good results were obtained under these conditions.

As shown in Part 2, the human E signal is 1. Counting started at point t, but the acceleration signal started counting from time t, about 70 minutes later. Then, at time 0 t8, where microcracks were discovered to have occurred at time t1, peeling has occurred, and at time 1. In this case, the AE flaw signal count value is zero and the growth of the flaking has stopped, but the acceleration signal count value is obtained and at time 0 L9, when it is understood that flaking exists, there is a large amount on the orbital surface. It had peeled off. In this way, rolling fatigue! For lji[, the AE flaw signal is detected earlier than the acceleration signal.

Figure 3 shows the case where the oil runs out. The acceleration signal starts counting at time t, and is delayed by approximately 609 seconds from time t.
AE flaw signal counting was started at &. In other words, when oil was running out, the acceleration signal was able to detect it earlier.

``Effects of 9 inventions'' The lifespan of rolling bearings varies widely, but for example,
Assuming ooo time, in the case of Figure 2, the AE flaw signal detects peeling approximately 600 hours (25 days) earlier than the acceleration signal, and in the case of Figure 3, the acceleration signal detects peeling approximately 9 hours earlier than the acceleration signal.
00 hours (38 days) It can be said that the oil shortage was detected earlier than by the E signal. As described above, according to the present invention, by using the AE flaw signal and acceleration signal detection together, it is possible to detect any abnormalities as early as about one month. Note that memory 11.12
does not necessarily have to be provided.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an example of a device according to the present invention;
Figure 2 shows the event rate of the AE flaw signal acceleration signal when rolling fatigue flaking occurs, and Figure 3 shows the A during oil failure.
It is a figure which shows the event rate of an E-flaw signal acceleration signal. l: AE converter, 2: acceleration detector, 3, 4: amplifier,
5.6: band p-wave device, 7, 8: discriminator, q, to: counter, 13: control unit, 14: display. Patent applicant Asahi Kasei Industries Co., Ltd. Agent Takashi Kusano No. 1
Figure 1o 12 3IFp2 Figure 3

Claims (1)

[Claims] (1) An AE conversion element that is attached to the rolling bearing to be monitored and detects its crack sound as an electrical signal, an acceleration detector that is attached to the rolling bearing and detects its vibration acceleration as an electrical signal, and the AE conversion element a first discriminator that discriminates the occurrence of crack sound from the detection signal of the acceleration detector; a second discriminator that discriminates the occurrence of vibration from the detection signal of the acceleration detector; and a discriminator output of each of the first and second discriminators. first and second counters that count the number at regular intervals; a control unit that determines whether there is an abnormality in the rolling bearing based on the respective counted values of the first and second counters and the counting start time; and a control unit that displays the determination results. A rolling bearing abnormality monitoring device consisting of an indicator.