RU2756933C1 - Method for high-speed ultrasonic flaw detection of long-dimensional objects - Google Patents

Abstract

FIELD: quality control.

SUBSTANCE: invention can be used for high-speed ultrasonic flaw detection of long objects. The essence of the invention consists in the fact that the ultrasonic probing signals are periodically radiated into the object of control when the electroacoustic transducer is moved along the object, the reflected ultrasonic signals are received, their parameters are measured, according to the results of which the presence of defects is judged, while a convex-concave piezo plate with a wide directional pattern is used.

EFFECT: increasing the reliability and validity of defect location during high-speed ultrasonic flaw detection.

1 cl, 5 dwg

Description

The method of high-speed ultrasonic (US) flaw detection of long objects refers to non-destructive testing of materials and products and can be used for ultrasonic testing of railway rails and other extended products.

The requirement to increase the throughput of rail transport leaves less time for monitoring the condition of the rails and eliminating the defects identified in them. Therefore, the creation of reliable methods for high-speed control of rails is an urgent task.

High V scanning speeds lead to a reduction in the residence time Ax of the search system, consisting of one or several electroacoustic transducers (EAC), in the defect location zone. In the terminology of ultrasonic flaw detection, the defect location zone, in accordance with the current GOST 18576 - 96 and other regulatory documents [1, 2], is considered as the conditional length of the defect ΔL and is one of the main parameters of the recorded signals. Indeed, at limiting scanning speeds (for example, 160 km / h or 44.4 m / s), the time interval Δτ = ΔL / V of the EAT location in the defect location zone, at real values of the conditional length (ΔL = 50 mm), is only 5 times more than the time t _{p of the} travel (propagation) of ultrasonic vibrations to the defect and back close to 200 μs (for example, a defect in the rail foot of the P65 type). All this leads to the reception of a small number of pulses from the sought defects and to a decrease in the reliability of the control.

The number of pulses received from a defect could be increased by increasing the frequency F of probing pulse transmissions, however, this value is also limited by the basic principle of the echo method of ultrasonic testing with pulsed radiation: a subsequent probe pulse can be injected only after receiving all expected echo signals from potential defects in the controlled item. When monitoring rails, as follows from the given example and shown in the sources below, the frequency F cannot be higher than 5.0 kHz (F ≤ 5.0 kHz). Those. the period T between the sequences of probing pulses cannot be less than T = 1 / F = 200 μs.

To increase the location area of the desired defect at high scanning speeds, various methods are used. For example, there is a known method [3] of ultrasonic sensing of the area of holes in a bolted joint of a rail, which consists in feeding ultrasonic sounding signals into the rail by an electro-acoustic transducer, receiving by receiving electro-acoustic transducers from the rail of reflected ultrasonic signals, their joint assessment, according to the results of which, by means of processing, the presence of a defect in cracks around the bolt hole. The essence of the method is that the alleged area of the defect is probed at different angles using an EAT with two piezoplates.

The disadvantage of this method [3] is unsuitability for high-speed flaw detection.

A known method for detecting defects in rails [4], which consists in the fact that to the main direct EAT, emitting longitudinal ultrasonic vibrations, within the reach of the probing ultrasonic signals of the first electroacoustic transducer, reflected from the rail foot, install additional EAT, which receive reflected from the rail foot and possible defects ultrasonic signals, and joint analysis make a decision on the presence and position of a defect in the rail.

The disadvantage of this method [4] is the unsuitability for high-speed flaw detection when testing inclined EAT with excitation and reception of transverse ultrasonic vibrations. It is known [2, 5] that most of the defects in rails are detected with the help of inclined EAT, emitting transverse ultrasonic vibrations.

There is a method of high-speed ultrasonic control [6], which consists in feeding ultrasonic sounding signals into the rail by several unidirectional oblique electro-acoustic transducers, receiving reflected ultrasonic signals by them, jointly evaluating them by means of processing, according to the results of which the presence of a defect is judged. The disadvantage of this method is limited use caused by the detection of defects only in a certain area of the controlled object.

The closest to the claimed method is a method of continuous high-speed ultrasound control of rails [7], which consists in feeding ultrasonic sounding signals into the rail by several emitting unidirectional oblique electro-acoustic transducers, receiving reflected ultrasonic signals by them, jointly evaluating them by means of processing, according to the results of which the presence of a defect is judged.

The idea of the known method [7] is that ultrasonic sounding is carried out simultaneously by several n oblique (main and additional) EATs. The distance between them is selected from a certain ratio, taking into account the required scanning step.

The disadvantages of the method [7] are in the following circumstances:

- The complexity of the design. At large values of n, the EATs turn out to be separated by considerable distances (up to a meter or more). The possibility of constructive execution of such an ultrasonic search unit with a given distance between the EAT seems to be doubtful.

- Low measurement accuracy due to two circumstances:

- the required discreteness of sounding at a fixed relative position of the main and additional EAT and simultaneous sounding is realizable only at fixed speeds of their movement. Accurate maintenance of such speeds is difficult to implement;

- for the simplicity of the technical implementation of the search unit, it is advisable to place the main and additional EATs next to each other, for example, at n = 3. In this case, the problem of mutual influence of the probing and reflected signals comes to the fore. Cross-connections appear (main - additional, additional - main EAT), these signals are difficult to evaluate correctly.

- Complexity of implementation, low reliability and reliability of control. The use of several EATs for sounding characteristic defects at a certain angle (optimal for specific types of defects) provides for the presence of additional flaw detection channels for each additional EAT (probing pulse generators, amplifiers, timing, amplitude and frequency selection circuits) with appropriate signal processing, which significantly complicates flaw detection equipment and reduces the reliability of its operation;

- in the known flaw detection systems for detecting characteristic defects in rails, several basic EATs are used (at typical angles α = 42 °, 50 °, 58 °, 70 °) [8]. When implementing the method taken as a prototype [7], due to the use of additional unidirectional with the main EAT, the number of working EAT increases n times (at least 2 times), which further reduces the reliability and reliability of control due to the need to ensure reliable acoustic control. And this, with high-speed control, is the most difficult and, not yet fully resolved, problem. In addition, the flow rate of the contacting liquid also increases.

Thus, the known method [7] has a low reliability and reliability of control, its implementation is possible only with a significant complication of flaw detection equipment and the design of the search system.

The objective of the proposed method is to improve the reliability of high-speed ultrasonic flaw detection without significantly complicating the equipment of the flaw detection complex.

The technical result of the implementation of the proposed method is to increase the reliability and reliability of detection of defects in high-speed ultrasonic flaw detection.

To achieve this result in the method of high-speed ultrasonic testing of long objects, which consists in periodic radiation of ultrasonic sounding signals into the control object when moving the electro-acoustic transducer along the object, receiving reflected ultrasonic signals, measuring their parameters, according to the results of which the presence of defects is judged, according to the claimed method, use electro-acoustic transducer with a wide directivity pattern.

Moreover, in a particular case, obtaining a wide directivity pattern of an electro-acoustic transducer is achieved through the use of a convex-concave piezoelectric plate.

In another particular case, in order to provide contactless control of an object, an electro-acoustic transducer with a wide directional pattern is performed on the basis of an electromagnetic-acoustic conversion.

Hereinafter, a wide pattern is understood as a pattern whose value is k times greater than the width 2ϕ _{n of a} typical EAT. Moreover

where ϕ _{w max} is the maximum possible value of half of the opening angle of the MD for the excited type of ultrasonic vibrations (suggestions for its definition are given below, in the paragraph before expression (3)).

A distinctive feature of the proposed method is the use of EAT with a wide directional pattern for high-speed ultrasonic inspection of long objects, and obtaining a wide DP with contact input of ultrasonic vibrations is achieved through the use of a convex-concave piezoelectric plate EAP. With non-contact input / reception of ultrasonic vibrations, a wide DP is formed through the use of electromagnetic-acoustic (EMA) conversion.

The inventive method is illustrated by the following graphical materials: FIG. 1 - the formation of the amplitude envelope (Fig. 1a) of echo signals during the movement of the EAT at a speed V along the scanned surface of the controlled object in the defect location zone when sounding its EAT with a typical (narrow) - 2ϕ _n and an expanded (wide) 2ϕ _w directivity pattern (Fig. 1b), and the corresponding representation of these signals on a type B scan (Fig. 1c), where:

1. Object of control (for example, a railroad rail).

2. EAP.

3. Scanning surface of the object of control.

4. Defect.

5. Threshold level.

6. Echo impulses on a type B scan.

FIG. 2 - dependence of the value of Ee relative change in the conditional length on the scanning speed V at different values of the opening angle ϕ of the EAT directivity pattern due to a noticeable EAT shift during the propagation of ultrasonic vibrations to the defect and back.

FIG. 3 - to change the zone of interaction with the scanned surface of the test object, where:

7. Convex-concave piezoplate (shaded in Fig. 3).

8. Flat piezoplate (shown with dashed lines).

9. Protector.

Fig. 4 - Fragments of the passport data of a typical and experimental EAT: a) DN of a typical EAT with a flat piezoplate with a diameter of 12 mm; b) DN EAP with a convex-concave piezoplate.

FIG. 5 - Experimental defectograms obtained from the defect model in the rail: a) using a typical (with a narrow directional pattern) EAT; b) an experimental EAT with a wide BP. Bursts of red signals were received by the "driving" on the defect, and blue ones - by the "driving away" EAT.

Let's consider the possibility of implementation and the advantages of the proposed method. The EAT 2, installed on the scanning surface 3 of the object, is moved along the long test object 1 at a speed V (Fig. 1b), periodically (with a set frequency T) emitting ultrasonic pulses. In the presence of a local defect 4 in object 1 and the impact of the EAT 1 on the defect, ultrasonic vibrations are reflected and recorded by EAT 1. In the general case, the EAT can operate in combined or in separate mode.

Echo signals from a defect with a gradually increasing and, subsequently, falling amplitude (amplitude envelope) can be recorded on a type A scan (Fig. 1a). During the time Δτ (Fig. 1c) of the EAT staying in the location zone ΔL, a defect group (pack) of N echo signals is formed:

When registering a large flow of information, typical for high-speed control, it is advisable to register signals on a type B scan (Fig. 1c) [5], where all echo signals 6 are recorded, the amplitudes of which exceed a predetermined threshold 5, taking into account their temporal position relative to the probing pulse on the coordinate plane "Time of propagation of ultrasonic vibrations t _p - Time of movement of EAT /" (Fig. 1c).

With traditional methods of radiation / reception of ultrasonic vibrations using standard EAT (see Fig. 1b for a typical EAT), for example, based on the use of flat piezoelectric plates in a controlled object (in a railway rail), a rather narrow directional pattern (DP) of a transducer with a width of 2ϕ is formed _n , which, depending on the angle α of the ultrasound beam input, is only 6 °… 8 ° [9]. Accordingly, the zone of location of the defect ΔL _n will be insignificant (Fig. 1), which causes the reception of a small number N echo signals from the defect 4. Naturally, when using an EAT with an extended BP with a width of 2ϕ _{w, the} number of received echo signals N in in accordance with (1) will increase in proportion to the increase in ΔL _w (Fig. 1a), which contributes to an increase in the reliability of control.

In [10], it was shown that, in addition to the indicated factor, the value of the conditional length ΔL, due to a noticeable shift of the EAT during the propagation of ultrasonic vibrations to the defect and back, is also very noticeably affected by the scanning speed V, decreasing the value of ΔL (a kind of compression of the defect location zone) ... When the EAT moves towards the reflector, the relative change ε _{ΔL of the} conditional length ΔL _dyn measured in the dynamic mode (V >> 0), compared with the length ΔL _st , determined in the stationary mode (V ≈ 0), can be estimated with sufficient accuracy for practice using the expression

where c is the velocity of propagation of ultrasonic vibrations in a controlled object (for transverse ultrasonic vibrations in a metal rail, c = 3260 m / s). Note that, in accordance with [10], for a “departing” EAT, the “-” sign in the numerator before <p of expression (2) must be replaced by the “+” sign, which, however, insignificantly changes the value of ε _ΔL .

The relative change in ε _{ΔL of the} nominal length depends on the angular parameters of the EAT (the angle of entry α and the width of the radiation pattern 2ϕ). From the expression it follows that the larger the directional pattern (BP), the smaller the decrease in ΔL _dyn , which favorably affects the reliability of high-speed control.

As seen in FIG. 2, constructed according to (2) for the values of the entry angle α = 50 ° at a typical opening angle ϕ _n = 4 ° and a wide angle ϕ _w = 12 °, with an increase in the width of the BP 2ϕ by 3 times, it is possible to achieve a decrease in the relative change in the nominal length at large speeds more than 2.5 times. Naturally, other things being equal, in the location zone ΔL of greater length, it is possible to obtain a greater number of pulses (see Fig. 1), which contributes to an increase in the reliability of control, especially with high-speed control.

It should be noted that the width of the pattern can be increased by known methods only within certain limits. For example, the effective angles for the excitation of transverse ultrasonic vibrations in a metal are limited by the first α _cr1 and the second α _cr2 critical angles [2, 9]. It can be assumed that for rail steel the sector of angles possible for the excitation of these waves is limited to angles from 38 ° to 70 °. Those. the maximum possible value of 2 _ϕwmax should not exceed 32 °. In addition, with a symmetrical DN, the angle of entry a must satisfy the condition

which guarantees that the entire beam of ultrasonic waves will propagate within the sector of angles of effective excitation of the required types of waves.

In the known methods of high-speed control, the search for defects in the required cross-section of the controlled object is carried out using a search system consisting of several EATs with fixed angles of entry and with narrow patterns. For example, to search for defects in rails along the entire height (for R65 rails - 180 mm) using transverse ultrasonic vibrations, four EATs are used with typical entry angles 42 °, 50 °, 58 ° and 70 ° with a width of 6 ° ... 8 ° [2, 8 and 9]. In this case, naturally, sectors in which defects can be missed remain sounded in a non-optimal way. To detect certain types of defects, additional EATs are introduced, which sound the interbeam spaces of the rails (see, for example, [11]). However, this complicates the design of the search system and forces the introduction of additional emitting / receiving channels into the equipment of the flaw detection complex, to a certain extent reducing the reliability of its operation. In this case, the problem of increasing the reliability of control is only partially solved.

Searching for defects using the proposed EAT method with a wide BP, covering the maximum controlled section of the object, excludes the appearance of uncontrolled sectors and increases the reliability of control without increasing the flaw detection channels.

Thus, the search for defects using an EAT with an extended DN allows obtaining the expected technical result due to the four positive factors discussed above:

- sounding the desired defect in a section (in the location zone ΔL) of greater length, which allows you to get a larger number of impulses from defects at high speeds and more information about the defect;

- reducing the influence of high scanning speeds on the compression (compression) of control signals;

- expansion of the zone of active interaction of the EAT with the scanned surface of the controlled object with an increase in the time of "contact" with the defect;

- elimination of "blind" zones in the controlled section of the object, typical for sounding schemes, implemented with the help of several EATs with discrete angles of input / reception of ultrasonic vibrations and with narrow DP.

A significant difference of the proposed method is the implementation of the receiving-emitting EAT with a wide directivity pattern.

As an EAT in the claimed method, various methods of excitation / reception of acoustic vibrations using contact methods of water / reception of ultrasonic vibrations based on the use of piezoplates, including using phased antenna arrays (PAR) [12], and contactless - based on electromagnetically -acoustic (EMA) [13] and even laser excitation and EMA reception.

Let us consider the possibilities of expanding the DN when using ultrasonic transducers based on piezoplates as an EAT. As is known [2], the opening angle ϕ of the DN (Fig. 1) depends on the wavelength in the test object, the radius of the piezoplate, the angle of the prism, and the velocities of ultrasonic waves. A wide DN with a large opening angle has advantages at the stage of searching for defects, since it allows sounding significant volumes of material. Narrow MD allows one to measure the coordinates of defects, to estimate their shape and size with a smaller error.

In high-speed inspection, of course, the main goal is to find defects. And the assessment of coordinates, shape and size of defects is carried out with the specification of manual control, provided for by the current technologies of flaw detection [1].

One of the simple ways to increase the width of the pattern is to reduce the size of the piezoelectric plate, but this also reduces the energy of acoustic vibrations propagated in the controlled object, which will inevitably lead to a decrease in the sensitivity of testing, which is unacceptable for high-speed flaw detection.

An obvious way to expand the pattern is to add an acoustically scattering lens to the EAT. However, this will also lead to a decrease in the radiated energy and weakening of the received ultrasonic vibrations due to the damping of vibrations in the lens.

In the particular case of the implementation of the method for expanding the DN, it is advisable to use a convex-concave piezoplate (Fig. 3). In this case, it is possible to achieve a multiple increase in the width of the EAT pattern without complicating the equipment and design of the search system of the flaw detection complex.

Additionally, the use of an EAT with a wide BP, obtained using a convex-concave piezoplate, makes it possible to increase the zone Z _{w of} active interaction of an inclined EAT with the scanned surface (compared to Z _{n with a} flat piezoplate of the same size) by 30% or more (Fig. 3). The latter depends on the design parameters of the plate: the radius R of curvature of the convex piezoelectric plate, size 2 a and the distance δ of the piezoelectric plate from the scanning surface of the test object (for the tread thickness) and can be calculated based on geometric considerations and the law of refraction (Snell) of acoustic vibrations [2, 9 ].

When using contactless input / reception of ultrasonic vibrations using EMA conversion [13, 14], an increase in the width of the BP by the required value can be achieved by selecting the configuration of the inductor coils (emitted and receiving) located under the concentrator of the magnetization system.

The efficiency of the above proposals has been verified by mathematical modeling and experimental research (Figs. 4 and 5) and there is no doubt about the implementation in high-speed non-destructive testing of products. As seen in FIG. 4, the width of the BP with a convex-concave piezoplate (Fig. 4b), measured on a standard CO-2 sample [15], is more than three times wider than a typical EAT with a flat piezoplate (Fig. 4a). As a consequence (see Fig. 5), the size of the signal packets from the same defect in the rail (cross cut in the rail foot 5 mm high) is noticeably (2-3 times) larger when scanning with an EAT with an extended pattern ( Fig. 5b) than when scoring a typical EAT.

The method of ultrasonic testing, implemented in accordance with the invention, provides an increase in the reliability of testing in comparison with other similar methods. It can be used with high operational efficiency for ultrasonic flaw detection of various objects, including rails. The use of this method is especially effective in mobile diagnostic tools for ultrasonic testing of railway infrastructure, allowing reliable detection of rail defects at a high speed of movement of diagnostic tools. This, in turn, makes it possible to carry out diagnostics of the rail track without interrupting regular freight and passenger transportation, by including a mobile diagnostic tool, made, for example, in the form of a flaw detector car, into the corresponding train. The specified technical result is provided within the framework of the implementation of the appointment by the entire set of essential features presented in the formula.

The inventive flaw detection method can be used not only for the control of rails, but also for flaw detection of many long products made of ferromagnetic materials: pipes, rods and rods, drill rods, conveyor belts, contact wires of an electric traction train (electric locomotives, trams and trolleybuses), etc.

Thus, due to new essential features (EAP with an extended DN and the use of a convex-concave piezoplate or EMA excitation to expand the DN), the claimed technical result is achieved: an increase in the efficiency and reliability of control during high-speed ultrasonic testing. Moreover, in contrast to analogs and the prototype [3, 4, 6, 7] - this is achieved without a significant increase in the design dimensions of the EAT and the use of additional flaw detection channels.

The inventive method can be implemented, provides an increase in the reliability and reliability of the detection of defects in the objects of control at significant scanning speeds.

Sources of information

1. Regulations on the system of non-destructive testing of rails and the operation of means of rail flaw detection in the track facilities of the railways of JSC Russian Railways (approved by JSC Russian Railways dated July 26, 2017, No. 1471 / r).

2. Markov A.A., Shpagin D.A. Ultrasonic flaw detection of rails. St. Petersburg, Education-Culture, 2006, 206 p.

3. RU 2052808.

4. RU 2668941.

5. Markov A.A., Koz'yakov A.B., Kuznetsova E.A. Deciphering of defect-grams of ultrasonic testing of rails. St. Petersburg, Education-Culture, 2013, 284 p.

6. RU 2715885.

7. RU 2 440 568.

8. Markov A.A., Kuznetsova E.A. Flaw detection of rails. Formation and analysis of signals. Book 2. Decoding of defectograms (a practical guide in two books. - St. Petersburg: Ultra Print, 2014.-332 p.

9. Gurvich A.K., Kuzmina L.I. Reference directional diagrams of ultrasonic flaw detectors finders. Kiev: Technique, 1980. 101 p.

10. Markov A.A. Peculiarities of evaluating the conditional sizes of defects at significant scanning speeds. Defektoskopiya, 1988, No. 3. S. 8-11.

11. RU 2725705.

12. RU 2 682 983.

13. RU 2489713.

14. Patent D 19544217 A1.

15. RU 2415388.

Claims (1)

The method of high-speed ultrasonic flaw detection of long objects, which consists in periodically emitting ultrasonic sounding signals into the test object while moving the electroacoustic transducer along the object, receiving reflected ultrasonic signals, measuring their parameters, the results of which are used to judge the presence of defects, characterized in that for the emission and reception of ultrasonic signals use a convex-concave piezoplate with a wide directivity pattern.

Images