RU2722089C1 - Noncontact ultrasonic flaw detection using a doppler effect - Google Patents

Abstract

FIELD: defectoscopy.SUBSTANCE: invention can be used for non-contact ultrasonic flaw detection using Doppler effect. Essence of the invention consists in the fact that during the relative motion of the contactless acoustic transducer and the controlled article at a given frequency, ultrasonic vibrations are emitted into the article with a beam pattern symmetric in the probing plane with known angles, receiving reflected echo signals, at known speed of movement calculating Doppler frequency shift of echo signals, determining instantaneous frequencies of echo signals received in limits of opposite directed main lobes of the beam pattern of the converter, in direction and values of their change and by mutual time position of groups of echo signals received by said lobes of the beam pattern, presence and position of the defect are determined.EFFECT: high reliability of detecting defects at considerable scanning speeds (up to 120 km/h and higher).1 cl, 4 dwg

Description

The invention relates to the field of ultrasonic (US) non-destructive non-destructive testing of products, can mainly be used to detect and evaluate defects in long products, in particular, in rails of railway transport, underground, in pipes, in rods for various purposes at high speeds of relative motion between radiating receiving system and controlled product. The claimed technical solution is considered on the example of control of railway rails with an inclined input of ultrasonic vibrations.

To detect dangerous defects in the form of cracks of different orientations in long products, an inclined input of ultrasonic vibrations is used. When monitoring with high scanning speeds (close to train speeds of up to 120 km / h), it is necessary to ensure contactless input and reception of ultrasonic vibrations. With the non-contact input of ultrasonic vibrations using electromagnetic-acoustic (EMA) excitation, laser generation or a pulsed beam of charged particles in the controlled product, radiation patterns are formed, which are two oppositely directed symmetrical petals (with EMA with a maximum radiation angle of about ± 30 ° [ 1], with thermoelastic excitation - with a maximum at an angle of ± 55 °) [2].

When scanning a product with a double radiation pattern, certain problems arise with signal selection related to the complexity of defect localization. For example, there is a known method of EMA defectoscopy of metal products [3], which consists in the fact that an EMA flaw detector is installed on the product, emitting acoustic sounding signals directed in the plane of sounding deep into the product at known angles, the flaw detector is moved around the product at a known speed, signals are received, reflected from possible defects determine the depth of their occurrence.

The disadvantages of the method [3] are the low accuracy and performance of flaw detection. Low accuracy is due to the following. In the static position, the EMA flaw detector excites bi-directional ultrasonic vibrations in the product. Signals reflected from possible defects are received in the form of single radio pulses. As a result, it turns out to be impossible to determine which direction of the ultrasonic probe signals detected the defect. To solve this problem in the known method, it is assumed that when moving the EMA of the flaw detector, the amplitude of the probing signals changes. Then one of the directions of ultrasonic sounding, the “back” lobe [3], is considered not significant and is excluded from consideration.

This approach does not take into account the interference associated with the detection of defects by the “back” lobe, which reduces the accuracy of flaw detection.

The low performance of flaw detection is due to the fact that unidirectional ultrasound probing in the method [3] is proposed to be carried out with reverse movement of the EMA of the flaw detector, which increases the time of flaw detection by 2 times.

At significant scanning speeds, the well-known Doppler effect is manifested very clearly, when the frequency of the echo signal f _e received from the defect differs from the radiation frequency of ultrasonic vibrations f ₀ by the value of the Doppler shift F _d :

Where

v is the scanning speed;

C is the propagation velocity of ultrasonic vibrations;

α is the angle of entry of the ultrasound beam into the controlled product.

The choice of the sign between the terms in (1) depends on the direction of motion: with the mutual approximation of the probe and the defect (+), with the removal (-).

The feasibility of using the Doppler effect in relation to flaw detection was shown in [4]. To further increase the noise immunity at a variable scanning speed, it was proposed in [5] to adjust the receiver passband (more precisely, the Doppler frequency filter band) of the flaw detector depending on the speed.

Further development of this direction of increasing the efficiency of ultrasonic inspection of products at high scanning speeds was obtained in technical solutions [6], [7] and [8], where the independence of the number of periods of the Doppler frequency in the echo signal of a specific defect on speed [6] and the possibility of highlighting using narrow-band filters of the extreme sections of the Doppler spectrum [7] to determine the parameters of detected defects [8]. A common disadvantage of the known technical solutions [4-8] is a narrow scope, focused only on their use in continuous radiation of ultrasonic vibrations and the contact method of input / reception of ultrasonic vibrations.

Such a solution is a more universal technical solution aimed at the contactless input and reception of ultrasonic vibrations with high-speed control of products. The main idea of the method [9] is to isolate signals having a Doppler frequency shift. In the process of relative movement of the BAP and the controlled product, ultrasonic vibrations are emitted into the product at a given frequency at a certain angle, at a known scanning speed, reflected echoes are received and echoes from defects are taken into account taking into account the Doppler frequency shift. Moreover, on the example of control of railway rails, it is shown that an increase in the scanning speed (of a flaw detector car) leads to a more pronounced Doppler effect and better efficiency.

The disadvantage of the method [9], adopted as a prototype, is the low noise immunity, and hence the low reliability of the control results caused by the incomplete use of informative parameters of the echo signals obtained by high-speed scanning. In addition, in the known method it was not taken into account that with the considered non-contact excitation methods, the most effective oblique input of ultrasonic vibrations is carried out with the formation of a two-leaf (symmetrical) radiation pattern. Ignoring this feature further reduces the reliability of the control.

The aim of the invention is to increase the reliability of detection of defects in the presence of relative high-speed movement between the radiation-receiving system and the controlled product through the use of a fine structure of ultrasonic echo signals from defects, taking into account the Doppler effect.

The technical result of the implementation of the proposed method is to increase the reliability of detection of defects in non-contact ultrasonic testing at significant scanning speeds.

To achieve this result, in the method of non-contact ultrasonic inspection using the Doppler effect, which consists in the fact that in the process of relative movement of the BAP and the controlled product at a given frequency, ultrasonic vibrations with a radiation pattern symmetrical in the sounding plane with known angles are emitted into the product, the reflected echo signals, at a known scanning speed, the Doppler frequency shift of the echo signals is calculated, additionally, with the necessary discreteness, the instantaneous frequencies of the echo signals received within the opposite directional main lobes of the radiation pattern of the BAP are determined by the direction and magnitude of their change and the relative position of the echo groups taken by the indicated lobes of the radiation pattern determine the presence and position of the defect.

Significant differences of the proposed method compared to the prototype is to account for changes in the Doppler shift (echo frequency) in the process of sounding the defect and taking into account the multidirectional deviation of this frequency with successive sounding of the defect by the front (“moving in”) and back (“moving away”) lobes of the radiation pattern of the BAP . In addition, according to the time (spatial) interval between groups (packets) of signals received by the indicated petals, it becomes possible to determine the position of the desired defect within the controlled product, which, together with the frequency signs noted above, is an additional sign of signals from the defect.

The method [9] does not take into account that the frequency of the echo signal depends not only on the scanning speed, but also on the variable (within the radiation pattern), the angle of entry / reception of ultrasonic vibrations, which changes during the sounding of the defect. Also in [9], it is assumed that, upon contactless excitation of ultrasonic vibrations, a unidirectional pattern is formed in the product. It is known that, in practical implementation, the most effective and stable excitation of ultrasonic vibrations in solids can be achieved with the formation of a bidirectional beam symmetric in the sounding plane [1, 2]. Lack of accounting for these important factors in the processing of echo signals received during scanning leads to a decrease in the control reliability.

Note that some differences in the expression for determining the Doppler displacement in the prototype from expression (2) in this text are explained by the peculiarities of counting the angle of entry of ultrasonic vibrations into the controlled product. In domestic practice, the angle of input and vibration is measured from the normal to the scanning surface.

In real situations, with any method of excitation of ultrasonic vibrations (by a non-contact laser, electromagnetic-acoustic radiation or a pulsed beam of charged particles) when receiving echoes reflected from an internal defect, a symmetrical resulting radiation pattern with central input angles + α ₀ and - α ₀ , and, as a rule, with identical opening angles 2ϕ _p . In the process of continuous scanning of a controlled product with a speed of v _c , when approaching a local internal defect at the initial moment of time, the defect is voiced by the edge of the incident petal at an angle α _n = (α ₀ + ϕ _p ). Moreover, in accordance with (1) and (2), where α = α _n = (α ₀ + ϕ _p ), the ultrasonic echo signal has a frequency f _ns (come near, start) component

As the BAP moves in the location zone ΔL of the defect, a gradual “hit” of MDs on the defect occurs, with the predominant sounding of the defect occurring at an angle α ₀ , and the instantaneous frequency of the echo signal is determined by expressions (1 and 2), where α = α ₀ .

In the future, the "congress" of the incident petal of the bottom beam of the ultrasonic beam from the plane of the reflector begins and sounding of the defect occurs under the fragility α _to = (α ₀ -ϕ _p ),

As a result, the frequency of filling the echo signals during the scanning process, defined as the difference (4) and (3), changes by

This leads to the obvious conclusion that the wider the DN (value ϕ _p ) of the BAP, the greater the change in the frequency (deviation) Δf _{e of the} frequency of the echo signal in the process of location of the defect.

Similarly, the formation of echo signals and a corresponding change in their filling frequency during subsequent sounding of the defect by the back, "departing" lobe of the bottom of the BAP. However, in contrast to the sounding by the incident petal of the beam, the sign (-) is taken between the terms in the expression (1), i.e. the echoes generated by the moving lobe of the beam have a frequency lower than the radiation frequency f ₀ by an amount determined by the Doppler shift. Moreover, the change in this frequency within the main lobe of the beam occurs in the opposite direction, because sounding of the defect at the initial moment occurs at an angle of - (α ₀ -ϕ _p ), and subsequently, at angles - α ₀ and - (α ₀ + ϕ _p ). Under the condition of symmetry of the MD, consisting of two oppositely directed main lobes, the deviation Δf _{E of the} frequency of the echo signal and when voiced by the moving lobe of the MD is also determined by expression (5).

The inventive method is illustrated by the following graphic materials:

FIG. 1 - An enlarged version of the functional diagram of the flaw detector for the implementation of the proposed method, where:

1. The path sensor (odometer) of the relative displacement of the BAP 4 and the controlled product.

2. The calculator.

3. Shaper of probing ultrasonic signals (generator of probing vibrations of ultrasonic frequency).

4. Contactless acoustic transducer (BAP).

5. Echo receiver.

6. The output of the calculator.

FIG. 2a. The ultrasound scheme of sounding a local defect with an inclined BAP, forming a two-beam radiation pattern, where:

7. Controlled product (for example, rail).

8. The scan surface of the product.

9. Two-leaf beam pattern of the BAP:

10. Local defect in the product.

11. Surface damage to the product in the direction of the BAP.

12. Surface damage to the product against the direction of movement of the BAP.

FIG. 2b. Displaying the filling frequencies of the echo signals received by the incoming and outgoing petals of the bottom line of the UPS, on the coordinate plane “frequency f is the time t of the movement of the UPS”, where the time axis is correlated with the path of the UPS along the surface of the controlled product (see Fig. 2a).

FIG. 3. Display of the two-petal radiation pattern of the UAV in the Cartesian coordinate system (Fig. 3a) and the placement of the frequency spectra of ultrasonic echo signals and interference on the frequency axis (Fig. 3b), where:

S _n and S _aw - zones of the spectral components of the frequencies of ultrasonic echo signals from internal defects;

And _n and A _aw are the spectral components of the frequencies of the ultrasonic signals from possible interference (irregularities and microcracks 11 and 12 on the scanning surface) received by the side lobes of the beam at angles α close to 90 °;

B - frequency spectrum of ultrasonic signals from possible irregularities of the opposite surface of the product, adopted by the side lobes of the beam at angles α close to 0 °.

FIG. 4 - Change in the filling frequencies of ultrasonic echo signals (Fig. 4a) and changes in the temporal positions of echo signals on a scan of type B (Fig. 4b). For clarity, the position of the signals in FIG. 4a are displayed on the axis of the Doppler frequency module.

Consider the process of implementing the proposed method of non-contact ultrasonic inspection using the Doppler effect.

A device that implements the method consists of series-connected (Fig. 1) displacement sensor 1, calculator 2, probing oscillator 3, contactless acoustic transducer (BAP) 4, an echo signal receiver 5, the output of which is connected to the input of the calculator 2. Parameter control (amplification, selection of the frequency band) of the receiver 5 is performed by the calculator 2. The signals of the calculator 2 processed according to the considered algorithms are supplied to the output 6 for further documentation and activation of the flaw detector (not shown in Fig.). The construction of the main components of the device that implements the proposed method, obviously and does not cause difficulties in the implementation.

As a method of contactless input and reception of ultrasonic vibrations in the known solution [9] several implementation options are proposed. The most practical of which, according to the author of this application, are two methods: laser (opto-thermal) excitation of ultrasonic vibrations and reception using a contactless electromagnetic-acoustic (EMA) receiver and the second option - EMA excitation and EMA reception of ultrasonic echo signals from possible internal defects.

The EMA method has been relatively widely used [1, 2, and 10]. This is due to its relatively large functionality: reversibility, the ability to generate various types of acoustic waves in materials — longitudinal, shear, and Rayleigh waves.

Two main structural elements of BAP 4 can be distinguished - an EMA converter [10]: a magnetic system consisting of a magnet (a set of permanent magnets or an electromagnet) and a magnetic circuit forming a magnetization field; inductor, usually representing an elliptical (or any other configuration) flat inductor (or several coils).

At high scanning speeds, well-known technical solutions [11–13] can be used as a magnetization system in EMA converters, when the wheels of a trolley of a moving unit serve as poles of electromagnets, and the inductor (flat coil) is located near or directly under the wheel in the annular recesses of the wheel.

The greater degree of distance is possessed by the methods of excitation of acoustic vibrations by opto-thermal (laser) and radiation-acoustic methods [14-16]. In this case, echo signals reflected from ultrasonic defects can be detected by EMA receivers and processed by the method proposed above, taking into account the Doppler frequency shift and intra-signal frequency modulation.

The choice of specific methods of non-contact excitation / reception of ultrasonic vibrations depends on the control conditions, the realized relative displacement velocities and available energy resources (especially for laser radiation).

BAP 4, moving at a speed of v _c above the scanned surface 8 of product 7, using a laser, EMA, or other method, excites ultrasonic vibrations of frequency f ₀ in the controlled product 7 with a symmetrical radiation pattern of 9 _n and 9 _aw with central angles α _0n and α _0аw with angles disclosure 2ϕ _p (Fig. 2 and 3). In the process of moving the BAP 4 in the zone of location of the defect 10 (with the direction of movement indicated in Fig. 2), the defect 10 is sequentially voiced first - by the incoming petal ДН 9 _n (come near) of the BAP 4, then - by the moving petal 9 _aw (move away) . In this case, in accordance with expressions (2), (1) and (3), the frequency of the echo signal changes. Moreover, when voiced by the incident petal, the frequency of the echo signal is greater than the radiation frequency f ₀ and varies within the echo signals from f _ns (start) to f _ne (end).

After a certain time equal to τ _d , the defect 10 is voiced by the moving lobe of the ДН 9 _aw , and the signal frequency, in accordance with (1) and (2), will be less than the radiation frequency f ₀ . During the movement of the BAP 4 with the scanning speed v _{c the} instantaneous frequency of the echo signals voiced by the departing lobe varies from f _aw.s to f _aw.е (Fig. 2b).

Thus, in the process of sounding the defect 10 of a two-beam symmetrical beam (9 _n and 9 _aw ) of BAP 4, two groups (packs) of echo signals with different filling frequencies are formed.

Changes in the frequencies of ultrasonic echo signals (Fig. 2) within the main lobes of DN 9 is one of the characteristic features, taking into account which it is possible to reliably identify useful signals from defects against a background of various interference.

For example, for real values of the input angles (α ₀ = 30 °) during EMA excitation / reception of ultrasonic vibrations, radiation frequency f ₀ = 2.5 MHz and half the radiation pattern width (ϕ _p = 10 °), at practically realized scanning speeds (v _c = 25 m / s = 90 km / h) and the propagation velocity of ultrasonic vibrations in the controlled product (c = 3260 m / s), the value Δf _E , in accordance with expression (5), is about 11.43 kHz. Those. the deviation of the filling frequencies of the pulses in the packet is almost 60% of the average value of the Doppler shift F _d (about 20.0 kHz), which is quite enough for its practical consideration in order to increase the reliability of control.

The location (reception) zone ΔL of the packets of these signals is determined by the width of the main lobes of the beam, and the duration x depends on the scanning speed (Fig. 2):

Measuring the time distance τ _d between the centers of these two signal packets, from geometric considerations, we can also estimate the depth h of defect 10 (Fig. 2):

We note the fundamental differences in the values of the coordinates of defects determined by the proposed method according to expression (7) and the traditional method for pulsed radiation (Fig. 3.11 [17]) by measuring the travel time of ultrasonic vibrations to the defect and vice versa. In the traditional calculation of the depth, the travel time of ultrasonic vibrations to the reflecting point (plane) of the defect and back to the transducer is determined. With the proposed method, the location of the center of the defect is estimated (the geometrical location of the successive intersection of the axes of the beam in the region where the defect is located).

In most cases, especially when detecting local defects of small sizes, these values are approximately equal (within the measurement error). But when defects of considerable size are detected, the values of the defect h determined by the considered methods can noticeably differ. From the point of view of repair work, information on the coordinate of the center of the defect obtained by the proposed method is preferable to the coordinates of two reflective points of the surfaces of the defect. In any case, the addition of traditional defect data with new information about the coordinate of the center of the defect further increases the reliability of the product control.

If we present the radiation pattern of the BAP 4 in the Cartesian coordinate system (Fig. 3a) and display the frequency spectrum of the echo signals (Fig. 3b), correlated to the angular parameters of the beam, it is clearly seen that the spectra of the expected signals from defects on the frequency axis are concentrated symmetrically with respect to the studied frequency f ₀ with center frequencies (f ₀ + F _D ) and (f ₀ -F _D ). The effective width of each burst of signals Δf _n = (f _ns -f _ne ) and Δf _aw = (f _aw.s -f _aw.e ) is determined by the width 2ϕ _{p of the} corresponding lobe of the beam (Fig. 3). As a rule, with contactless excitation of ultrasonic vibrations Δf _n = Δf _aw .

Two band-pass filters (not shown in FIG.), Bandwidth Δf _E defined by expression (5), and center frequencies (f ₀ + F _D ) and (f ₀ -F _D ), all expected signals from potential defects can be distinguished. According to the law of change (deviation) of the instantaneous frequency of the signals, it is possible to extract ultrasonic echo signals from internal defects against the background of possible interference. When converting signals into digital form, all these operations are performed in the calculator 2 (Fig. 1).

An interesting fact, which additionally confirms the above conclusions on multidirectional deviation of the signal filling frequency in the process of scoring the internal defect of a BAP with a symmetrical two-beam beam, is the identity of those presented in FIG. 4 patterns of frequency (Fig. 4A) and temporary (Fig. 4c) echo signals during scanning. Changes in the instantaneous echo frequencies due to the Doppler effect being voiced by the incident petal during the burst duration τ _n occur from the maximum value to the minimum (in Fig. 4a, the arrow shows the direction of decrease in the Doppler frequency shift). After a certain time, approximately equal to the time τ _d , sounding of the defect occurs by the moving petal of the beam and the Doppler frequency modulus of the instantaneous frequencies of the echo signals in the packet (duration τ _aw ) increases from minimum to maximum values determined by expressions (1) and (2).

A similar dependence, now already in the time domain, also occurs when a defect is voiced by contact oblique piezoelectric transducers (NEC) in and against the scanning direction (Fig. 4b) (see, for example, Fig. 3.17 [17], Fig. 3.4 and 3.10 [18 ]). In this case, two identical, but multidirectional probes with identical MDs are used. On a traditional defectogram with the display of signals on a type B scan (on the coordinate plane "propagation time of ultrasonic vibrations - time of PEP movement" (or, if there is an odometer, the scanning path)), usually two oblique lines (or groups of points representing the echo signals in the B-scan). These two lines with decreasing (relative to the probe pulse) echo delay time for incoming probes and increasing delay time for departing probes have been the main sign of defects in the controlled product for many years [18].

A peculiar frequency deviation of the ultrasonic echo signals from defects during the scanning process is characteristic of both the specific and very rare continuous radiation in flaw detection, and the traditional pulsed radiation of ultrasonic vibrations (Fig. 2b).

With pulsed ultrasonic radiation in the location zone ΔL _n (Fig. 1), during scanning from defect 10, a group (pack) of echo pulses is received (Fig. 2b) [10]. These pulses are formed sequentially in the process of sounding the reflector. Traditionally, the duration of a single echo pulse is only a few microseconds (about 10-12 periods of ultrasonic frequency), which is much less than one period of the Doppler frequency. Therefore, the frequency deviation caused by the change in the defect location angle during scanning is proposed to be estimated by determining the instantaneous frequencies of the echo pulses in the packet.

In contrast to technical solutions [4-8], using continuous radiation of ultrasonic vibrations, the transition to pulsed radiation with the separation of the Doppler frequency of the echo signals can significantly simplify the process of determining the occurrence depth of the desired defects in the traditional way (by the time delay of the echo pulses relative to the probe pulse) .

Various methods are known for determining the effective (instantaneous) frequency of an undetected radio frequency echo signal. In particular, in accordance with [19], the effective frequency is determined by the maximum frequency spectrum of the echo signal using the window Fourier transform. However, this method has low noise immunity.

The most practical determination of the effective frequency can be performed in a known manner [20], which proposes an algorithm for estimating the instantaneous frequency of an undetected radio frequency echo signal based on the use of wavelet transform. Moreover, within the duration of the radio pulse (echo signal), a certain time window is selected with a width comparable to the period of the carrier frequency. It is advisable to choose a time window in the area of the echo pulse corresponding to the maximum amplitude. A wavelet function is a function of two variables, localized in time and frequency, from which, by stretching and compressing along the time axis, one can obtain a basis for the analysis of functions. The essence of wavelet analysis is the decomposition of the analyzed signal into basic functions, followed by visualization of the frequency distribution of the investigated process in the time or frequency domain.

All transformations of the wavelet analysis are performed in the transmitter 2 of the flaw detector (Fig. 1). By determining the instantaneous frequencies of all the echo signals included in the packet of signals from the desired defect 10 (Fig. 2b and Fig. 4a), it is possible to trace the frequency deviation in the packs of echo signals during the scanning process.

The discreteness of the readout of the frequency of echo pulses in a signal packet during pulsed radiation of ultrasonic vibrations is determined in a natural way - the discreteness corresponds to the frequency of the probe pulses (in particular, when the probe frequency of the probe pulses is 4 kHz, which is implemented in existing flaw detectors, the instantaneous frequency is counted every 250 μs) . With continuous emission, the instantaneous frequencies of a frequency-modulated echo can be performed with the same discreteness or continuously (in each period of the Doppler frequency). The choice of sampling resolution depends on the desired accuracy of tracking the frequency deviation of the echo signals.

Thus, even with pulsed radiation of ultrasonic vibrations, one can trace the frequency deviation in the received signal packets from defects. According to the law of change (deviation) of the instantaneous signal frequencies, it is possible to isolate ultrasonic echo signals from internal defects against the background of possible interference.

Examples of such interference can be reflections from surface microcracks 11 and 12 (Fig. 1), which are not dangerous from the point of view of breaking the strength of the controlled product, with ultrasonic fixation of which the scoring angle remains constant.

It is known that any radiation pattern 9 also has side lobes. And in this case, possible reflections from minor damage 11 and 12 on the scanning surface 2 will also form ultrasonic echo signals. However, since the sound angle does not change relative to the normal and will be close to α = ± 90 °, the signals from them, although they have a Doppler shift (moreover, the maximum value), will not have a frequency deviation during scanning (zone A _n and A _aw in Fig. 3b).

Similarly, possible reflections from the opposite surface of the controlled product (rail) received at an angle close to α = 0 ° will be concentrated on the frequency axis near the radiation frequency f ₀ , and they will also lack both Doppler shift and frequency deviation (zone B in Fig. 3b). Only signals from internal defects, voiced by the main lobes of the DN 9 _n and 9 _aw with certain opening angles (2 _ϕp ), will form ultrasonic echo signals (Fig. 2b), the filling frequency of which will vary depending on the current scoring angle (Fig. 3 - zones S _n and S _aw ). Thus, by tracking the frequency deviation of ultrasonic echo signals, it becomes possible to isolate signals from internal defects against all kinds of interference. It is unlikely, also in practice unavoidable that control electromagnetic interference affecting the BAP 4 and incident on the input of the receiver 5 will be expected at the current scan velocity and Doppler shift corresponding NAM 9 frequency deviation Δf _E.

Summing up, it can be noted that when implementing the proposed method of non-contact ultrasonic inspection, in addition to the available ones (amplitude, delay time of the echo signal relative to the probe pulse), the following informative signs of signals from defects are used:

- the presence of a Doppler shift of the echo signals;

- frequency deviation within the bursts of echoes from defects;

- multidirectional change in the frequency of the echo signals in packs of echo pulses received by the incoming and outgoing petals of the bottom of the contactless acoustic transducer;

- the coordinate of the center of the defect, determined by the implementation of the two-beam pattern.

The last three signs are introduced for the first time for flaw detection with pulsed radiation from ultrasonic vibrations. Moreover, the possibility of comparing the coordinates of the defect determined in the traditional way (according to the travel time of ultrasonic vibrations) and the proposed method (according to the interval between packets of signals received by a symmetrical beam), further increases the reliability of detection of defects with non-contact ultrasonic testing.

Used new informative features, the deviation of the frequency of the echo signals, their direction, depending on the direction of sounding the defect, are not correlated with the amplitude of the signals, which provides an additional increase in the reliability and accuracy of the proposed method. The results obtained by the proposed method do not depend on the quality of the controlled surface 8 and the stability of the gap z between the BAP 4 and the scanned surface 8 (Fig. 2a), which is very important when implementing high-speed ultrasonic inspection.

Thus, between the set of essential features of the proposed method and the achieved technical result, there is a causal relationship, namely, the determination of the instantaneous frequencies of the echo signals received within the opposite direction of the main lobes of the radiation pattern of the BAP, taking into account the direction of their change and the relative temporal position of the echo groups -signals received by the indicated lobes of the radiation pattern, increases the reliability of detection of defects in non-contact ultrasonic testing at significant scanning speeds.

Naturally, the given specific values of the relative motion speed, the angles of the radiation pattern of the BAP, the emitted and received frequencies of the signals are given only to understand the principle of operation of the claimed technical solution and can have values in a wide range.

The operability of the above proposals was verified by mathematical modeling and experimental studies and does not raise doubts about the implementation with high-speed non-destructive testing of products.

The inventive method of defectoscopy can be used not only for non-contact inspection of rails, but also for defectoscopy of many lengthy products: pipes, rods and rods, drill rods, conveyor belts, contact wires of electric traction composition (electric locomotives, trams and trolleybuses), etc., in including the contact method of excitation / reception of ultrasonic vibrations.

Thus, the inventive method can be implemented, provides increased reliability of detection of defects at significant scanning speeds.

Sources of information:

1. Chabanov V.E., Zhukov V.A. Features of ultrasonic testing using electromagnetic-acoustic transducers // In the world of NK. 2013 No. 1 (59), p. 36-43.

2. Muzhikbaev A.S., Skvortsov M.P. Analysis of non-contact ultrasonic methods for monitoring welded joints // Actual problems of aviation and astronautics. - 2015.Vol. 1. S. 266-268.

3. Patent RU 187834

4. US 3,321,959.

5. SU 896550.

6. SU 998943.

7. SU 1429013.

8. SU 1527573.

9. US 6324912.

10. Ants V.V., Strizhak V.A., Balobanov E.N. To the calculation of the parameters of the magnetization system of an electromagnetic-acoustic transducer // Measuring equipment, 2011, No. 1 (17), p. 197-205.

11. RU 2680103.

12. RU 2653663.

13. RU 2661312.

14. Budenkov G.A. Excitation of elastic waves in solids by a laser beam due to the thermoelastic effect // Defectoscopy. 1979. No. 2. S. 75-81.

15. RU 2381496.

16. Simanchuk VI, Yanisov VV, Generation of ultrasonic vibrations in solids by pulsed beams of charged particles // Defectoscopy. 1984. No. 3. S. 30-38.

17. Markov A.A., Kuznetsova E.K. Defectoscopy of rails. The formation and analysis of signals. Book 1. Basics. SPb: KultInformPress, 2010 .-- 292 p.

18. Markov A.A., Kuznetsova E.K. Defectoscopy of rails. The formation and analysis of signals. Book 2. Explanation of defectograms. - St. Petersburg: Ultra Print, 2014 .-- 326 p.

19. GOST 26266-90.

20. RU 2524451.

Claims (1)

A method of non-contact ultrasonic flaw detection using the Doppler effect, which consists in the fact that during the relative motion of the non-contact acoustic transducer and the controlled product at a given frequency, ultrasonic vibrations with a radiation pattern symmetrical in the sounding plane with known angles are emitted into the product, reflected echo signals are received, when the known speed of motion, the Doppler shift of the frequency of the echo signals is calculated, characterized in that the instantaneous frequencies of the echo signals received within the opposite directional main lobes of the radiation pattern of the transducer are determined by the direction and magnitude of their change and by the relative temporal position of the groups of echo signals received the indicated petals of the radiation pattern determine the presence and position of the defect.

Images