RU2213336C2 - Method of ultrasonic test of antifriction bearings - Google Patents

Abstract

FIELD: ultrasonic diagnostics of antifriction bearings. SUBSTANCE: method of ultrasonic test of rotating antifriction bearings consists in excitation of ultrasonic oscillations by radiator put on shaft on which inner race of bearing is set, in reception of ultrasonic signals by detector-receiver mounted on outer race of bearing and in analysis of their propagation in article. Electric signal in the form of radio pulse is fed into radiator which filling frequency exceeds 200 kHz and is chosen equal to frequency of one of maxima of amplitude-frequency characteristic of line " radiator-bearing- receiver " generated under static condition by averaging of such individual amplitude-frequency characteristics for same angular position of receiver and radiator. Shaft is turned through some complete revolutions prior to measurement of each of them. Synchronization of exciting radio pulse is so executed that leading edge of synchronization pulse coincides in time with moment when angular displacement of radiator of ultrasonic signal relative to receiver is absent. EFFECT: potential for ultrasonic flaw detection in bearing in process of their rotation. 1 cl, 3 dwg

Description

 The invention relates to the field of ultrasonic diagnosis of rolling bearings.

 Known methods and devices for diagnosing bearings based on the analysis of acoustic noise and vibration that occur during rotation (Shubov I.G. Noise and vibration of electrical machines. 2nd ed., Revised and additional - L.: Energoatomizdat, 1986, 208 p.). The disadvantage of such methods is the identification of defects and wear by indirect signs, identified on the basis of the analysis of the amplitude-frequency spectrum of acoustic noise and vibration. At the same time, the results of such an analysis can be significantly affected by acoustic signals and vibrations excited by other parts of the rotating body or by mechanisms connected acoustically to the diagnosed node. The disadvantages of these methods should also include the dependence of the amplitude-frequency spectrum on the operating mode of the bearing (rotation speed, load).

 Methods and methods of active ultrasonic flaw detection are based on the analysis of time and amplitude characteristics of acoustic pulses excited and received by appropriate transducers (Devices for non-destructive testing of materials and products. Handbook. / Ed. By V.V. Klyuyev, Volume 2. - M .: Mechanical Engineering , 1986, 352 pp.). The use of these techniques for diagnosing bearings during their rotation (dynamic mode) is still difficult due to the fact that in dynamic mode the bearing is a source of acoustic noise, the level of which can significantly exceed the amplitude of the excited ultrasonic (ultrasonic) pulse. The main part of the bearing acoustic noise spectrum during rotation at a speed from 0 to 200 r / s falls on the frequency range 0 ÷ 100 kHz. To highlight a useful signal against the background of a high level of noise, radio engineering methods are used (Baskakov S.I. Radio engineering circuits and signals. - M.: Higher School, 1988, 448 p.). In particular, an ultrasonic pulse is excited, which is a radio pulse with a certain filling frequency. The envelope of this radio pulse is the corresponding video pulse, the selection of which for each implementation can be carried out by filtering, amplification, positive detection and demodulation.

 During rotation, the bearing is a dynamic system whose parameters change over time. Video pulses obtained on the basis of individual implementations differ significantly. Moreover, each individual implementation after detection and demodulation, in addition to the useful video pulse, contains significant distortions associated with noise components, the frequency of which is close to the frequency of filling the video pulse. Therefore, each of them individually cannot be the basis for determining the degree of wear of the bearing of the radio pulse.

 Closest to the proposed invention is a method of active ultrasonic testing of rotating parts, according to which an acoustic pulse with predetermined parameters is excited by an emitter mounted directly on the rotating part, and to which an electrical signal is supplied by electrical or radio engineering methods (A.I. Kondratiev, V.I. Rimland, AV Kazarbin. Method of ultrasonic testing of rotating products. RF Patent 2122728, 1998). The nature of the defect is judged by the change in the excited ultrasonic pulses. The disadvantages of this method include the impossibility of its application for the diagnosis of friction units, since it is assumed that the receiving transducer is also placed on a rotating part, which does not allow the ultrasonic signal to pass directly through the friction unit during rotation. This prototype also does not provide the necessary signal-to-noise ratio in order to highlight a useful signal against the background of significant bearing acoustic noise.

 The technical problem to which the present invention is directed is the development of a method that allows the application of methods of active ultrasonic inspection to bearings in the process of their rotation.

 This problem is solved in that in the proposed method for ultrasonic monitoring of rolling bearings by exciting the ultrasonic vibrations by a contact method by an emitter, to which an electric signal is supplied in the form of a radio pulse, receiving ultrasonic signals and analyzing their propagation in the product according to the invention, an ultrasonic pulse emitter is mounted on a rotating shaft near the internal bearing cages, and the receiver of ultrasonic signals is mounted on an external stationary cage.

 In this case, the excitation radio pulse is synchronized in such a way that the leading edge of the sync pulse coincides in time with the moment when the angular displacement of the emitter of ultrasonic signals relative to their receiver is absent (equal to zero).

 In addition, for each implementation of the ultrasonic signal from the ultrasound receiver, a useful signal is allocated in the form of a video pulse, which is recorded in the computer memory, and averaged over a large number of bearing revolutions, and average values of the amplitude and duration are determined for the video pulse obtained after averaging which are diagnostic parameters.

 Also, this problem is solved by the fact that the filling frequency of the exciting radio pulse exceeds 200 kHz and is chosen to be equal to the frequency of one of the maxima of the averaged amplitude-frequency characteristic of the emitter - bearing - receiver line obtained in the static mode.

 The fastening of the ultrasonic signal emitter on the rotating shaft and the ultrasound receiver on the stationary bearing cage ensures the passage of the acoustic signal along the line emitter - shaft - inner cage - balls (rollers) - external cage - receiver, that is, directly through the friction unit during rotation, which causes the possibility of using the proposed method for the diagnosis of bearings. In this case, the amplitude and duration of the transmitted ultrasonic signal depend on the quality of the contact of the balls (rollers) with the clips. Therefore, these parameters obtained in both static and dynamic modes can be accepted as diagnostic ones.

 Synchronization of the exciting radio pulse in dynamic mode increases the degree of repeatability of individual implementations and allows you to minimize the distance from the emitter to the receiver and thereby reduce signal attenuation and the influence of interference effects due to the superposition of ultrasonic waves transmitted through various balls (rollers) of the bearing.

 The amplitude and duration of the video pulse obtained by radio engineering methods, as the envelope passing through the bearing of the corresponding radio pulse, are also diagnostic parameters. Averaging the video pulse over a large number of implementations allows us to determine the average value of its amplitude and duration at a given rotation speed, as well as significantly reduce the components of acoustic noise.

 The exciting electric signal from the generator and the corresponding ultrasonic pulse are a radio pulse with a filling frequency of more than 200 kHz, that is, a frequency at which acoustic noise has a small amplitude. Setting the filling frequency of the exciting radio pulse to the frequency of one of the maxima of the averaged amplitude-frequency characteristic of the emitter – bearing – receiver line obtained in the static mode allows one to significantly increase the amplitude of the useful signal at the output of the ultrasound receiver in the dynamic mode.

 The figure 1 shows a diagram of a device for implementing the proposed method. A device for implementing the method comprises a shaft 1, a revolution sensor 2, a generator 3, a brush assembly 4, a piezoceramic ultrasound emitter 5, a bearing consisting of an internal rotating cage 6, balls 7, an external rotating cage 8, sound duct 9, piezoceramics of an ultrasound receiver 10, preliminary an amplifier 11, an electronic unit 12, an analog-to-digital converter 13, a computer 14. Sound duct 9, a piezoceramic of an ultrasound receiver 10 and a preamplifier 11 are assembled in the form of a single block representing an ultrasound receiver a.

 The method is as follows.

 The speed sensor, receiver and ultrasound transducer are installed in one angular position relative to the axis of rotation of the shaft. During rotation of the shaft 1, the speed sensor 2 generates a sync pulse (Fig. 2a), which starts the generator 3. The signal in the form of a radio pulse with a filling frequency ν (Fig. 2b) from the generator 2 is transmitted through a brush assembly 4 to the radiating converter 5. The converter 5 generates an ultrasonic pulse, similar in shape to a radio pulse, which propagates along the shaft 1, the inner race of the bearing 6, the balls of the bearing 7, the outer race 8, the sound duct 9, is recorded by the piezoceramic 10. The electrical signal from the piezoceramic 10 is transmitted to preamplifier 11, at the input of which a low-pass filter is installed. From the output of the amplifier 11, the signal is fed to the input of the electronic unit 12. In the electronic unit 12, the signal is filtered by a bandpass filter with a resonant frequency ν (Fig.2c shows the waveform of the signal after the filter), positive detection and demodulation. At the output of the electronic unit, the signal is a video pulse with superposition of the components of acoustic noise with a frequency close to ν (in Fig. 2c and 2d, the noise components are shown by arrows). The signal from the output of the electronic unit 12 is fed to the input of the analog-to-digital converter 13, the operation of which is synchronized by a pulse from the speed sensor 2. From the output of the analog-to-digital converter 13, the signal in the form of a separate digital implementation is sent to the computer 14, where it is subjected to software processing. The processing algorithm is the operation of averaging the recorded N realizations and determining the maximum amplitude A and duration τ of the averaged video pulse (Fig.2d). Parameters A and τ are diagnostic parameters.

 The frequency of filling the video pulse ν is determined previously in static mode. To do this, the radiating converter 5 is connected via a brush assembly to the output of the oscillator frequency generator of the spectrum analyzer, the output of the preliminary amplifier 11 is connected to the input of the spectrum analyzer. The output of the spectrum analyzer, the voltage at which corresponds to the measured frequency response, is connected via an analog-to-digital converter to a computer. The receiver and emitter of ultrasound are installed in one angular position. Measured the frequency response of the system emitter - bearing - receiver - pre-amplifier, which is recorded in the computer memory as a separate implementation of the frequency response (figa). This operation is repeated N 'times, with the shaft turning several full revolutions before each measurement. Then, the recorded N 'realizations are averaged, as a result of which the average frequency response is obtained (Fig. 3b), which determines the frequency ν exceeding 200 kHz and equal to the frequency of one of the frequency response maxima (shown by arrows in Fig. 3b).

 The set of technical solutions proposed in this invention provides the possibility of applying the methods of active ultrasonic diagnostics to rolling bearings in dynamic mode and, accordingly, to provide more reliable control of the degree of wear and condition of rolling bearings in the most critical friction units in comparison with traditional methods.

Claims (2)

 1. The method of ultrasonic testing of rotating rolling bearings by exciting ultrasonic vibrations by an emitter placed on a shaft on which the inner ring of the bearing is mounted, receiving ultrasonic signals by a sensor-receiver mounted on the outer ring of the bearing and analyzing their distribution in the product, characterized in that the emitter an electrical signal is supplied in the form of a radio pulse, the filling frequency of which exceeds 200 kHz and is chosen equal to the frequency of one of the amplitude-frequency maxima different characteristics of the emitter - bearing - receiver line, obtained in the static mode by averaging individual such amplitude-frequency characteristics for the same angular position of the receiver and emitter, before measuring each of which the shaft rotates several full revolutions.  2. The method according to p. 1, characterized in that the synchronization of the exciting radio pulse is such that the leading edge of the clock coincides in time with the moment when the angular displacement of the emitter of the ultrasonic signal relative to its receiver is absent (equal to zero).

Images