DE2825588A1 - PROCEDURE FOR THE AUTOMATIC GENERATION OF TEST RESULTS DURING NON-DESTRUCTION-FREE MATERIAL TESTING WITH ULTRASONIC IMPULSES - Google Patents

Description

24 ·. May 1978 IG / vh K 103 Krautkrämer GmbH

Luxemburger Str. 44-9, 5000 Cologne 4-1

PROCEDURE FOR THE AUTOMATIC GENERATION OF TEST RESULTS IN THE NON-DESTRUCTIVE FABRIC TESTING WITH ULTRASONIC IMPULSES

The invention relates to a method for automatic formation of a test result in the non-destructive testing of materials with ultrasonic pulses by evaluating two or more consecutive ones dxgxtalxsierten Α images that are obtained by changing the position of the probe arrangement, preferably from several probes with different radiation characteristics consists.

It is known that the sound pulse from the transmitter probe under a to reflect the specified angle to the surface normal in the test piece. On a reflector, which only then as a fault can be described if it is "harmful"

909850/0450

' , the sound pulse is reflected and converted into electrical signals in the receiver probe. The transmitter or receiver test head can also consist of a unit in which, in addition, one and the same electroacoustic transducer is used for transmitting and receiving the ultrasonic pulses. From the

; Travel time of the ultrasonic pulse is about the for the; Material of the test piece known speed of sound the distance between the probe and reflector is determined.

The echo sound is a measure under certain conditions for the reflector size.

If the echo height is dependent on the

'Time of flight measured and on the timeline of a cathode ray oscillograph en mapped, with between the term

'and the reflector distance via the speed of sound If there is proportionality, one speaks of an A-picture. (see Krautkrämer, materials testing with ultrasound, 3rd edition,

! P. 253). This term is also used when the

Echo height as a function of the transit time not on a cathode

beam oscillograph, but an evaluator apparatus, e.g. a computer, is supplied in digitized form.

; To look at an unknown reflector optimally, it is known, the sound at different angles of incidence in eirtschallen the test piece, whereby it can also be expedient

9O98B0 / OAB0

can use separate transmitter and receiver probes. Want one cannot determine the depth of the reflector from the beam geometry and To determine the running time, you can use several receiver probes one after the other arrange so that with fixed distances between the probes within the probe arrangement each probe reflectors in a certain depth zone. This process is known as the tandem technique, see J. and II. Krautkrämer, materials testing with ultrasound, Springer-Verlag, 3rd edition 1975 »p. 473 and 527.

It is also known that a useful statement about the type of defect, e.g. whether it is slag lines, ßindefehler, pores etc., it is possible that the examiner constantly observes the Α images on the cathode ray oscilloscope and the Probe arrangement on the test piece both in the direction of the reflector as well as moves parallel to the reflector. The change in the Α image as a function of the movement of the probe arrangement is called "echo dynamics".

If, for example, when testing a weld seam, the echo indicators fluctuate between a maximum value and a threshold value, c§U ' ^s · the zero line, when the test head arrangement is shifted parallel to the weld seam, these are separate reflectors. Fluctuations in the amplitude when moving the test head arrangement parallel to the weld seam give the tester indications of interruptions in the reflector. Furthermore, the test person receives information on the geometric shape of the reflector from the echodyiiamic envelope, which results from the displacement of the test head arrangement.

909850/0 4 50

2825585

- υ -

Fcichoanzeigen remain in the displacement dei ¹ probe arrangement parallel to .Jchweiiinaht a Langeren Hereich in amplitude obtained within a fluctuation, and show 3ie in t3ewegung to the weld seam and away from the away Jchweiiinaht a sharp rise and fall again
the hchohuLLkurve, e3 is a f Laughing PehLer m iichweLinahtrichtung.

Because of the fact that the echo heights change when the probe assembly is shifted parallel to the weld iti one of them, i.e. go the displays do not reach a threshold value, if necessary. the zero line back, there will be an accumulation of reflectors in the direction of the weld.

By its very nature, this manual check is very subjective and
places high demands on the auditor. One can try to schematize these examinations to a large extent, see Krautkrämer- ■ special print 1o'7 and Voest / Alpine report Ph 12 / υ9 ·

h.3, however, remains a disadvantage that

909850 / 0A50

the examiner as a subjective element is an essential link in the Identifying and evaluating reflector displays, creating fluctuations and uncertainties in the interpretation of the Echo indications are inevitable. The high test capacity required and the associated test capacity are disadvantageous for automation high costs, the high level of training required by the testing staff and the lack of comparability in the assessment of the findings. Under these circumstances, a repeat test is required not readily comparable with a previous test, i.e. the tests are partially reproducible. The time required for such manually performed tests is also significantly greater than the test effort for machine tests Tests in which one has to forego the advantageous statements of the echo dynamics.

The aim of the invention is to provide an ultrasonic testing method which avoids disadvantageous subjective influences on the part of the tester and thus making the test itself automatable, with the movement the test head arrangement is either computer-controlled or a manually operated manipulator. From the

809850 / (HSO

-S-

Change of the Ortskooixlinaten the Pi'üfkopf ano3. * Dnung and the assigned A-figures should be obtained objective findings, how an Echodynaraik is able to deliver. Even the reproducibility of the tests should be guaranteed. According to the invention, this object is achieved by the subject matter of Claim 1 solved.

It is also advantageous that the computer as a decision aid to be able to enter further data of the test piece to form the findings.

Embodiments of the invention are shown in the drawing and are explained in more detail below. Show it:

Fig. La a plate-shaped test piece, on the surface

as a test head arrangement

a plurality of probes erP'rrilt different Positions and "respective sound beam path, put on and which contains a spherical yeast deflector

Fig. 1b the same door flat reflectors,

Fig./2a a graphic representation, wherein, each upper half J2b echo signal values (Α picture) as a function of the Specifies the position of a certain test head for Yeflector types in weld seams

3 is a graphical representation of an echo display sequence at a test piece with spherical reflectors,]

Figure 4- is a graphical representation of an echo display sequence at a test piece with flat reflectors

5 shows a block diagram of a device for executing the method with the computer.

609860/0460

First of all, the connection between the Α images that occur during the "dynamic test" and the movement will be discussed the test head arrangement, the following abbreviations being chosen:

EK "single probe

TK »Tandem arrangement of the probes A» Echo height (echo amplitude)

The indices indicate the angle of incidence to the perpendicular of the surface, eg EK ₇ »7o angle probe.

Example 1. If a reflector 2 in the workpiece 1 is sounded on by test heads EK ₇ etc. at different sound beam angles, Fig. Ia ^ and the echo amplitudes A of the different test heads - assuming uniformly set display sensitivity - are approximately the same and are in the echo display there is also no essential difference, if the reflector is sounded on from the opposite side, a reflector must have a spherical limitation. It can be said that it is a spherical reflector. The statement that the amplitude heights are approximately the same for the different sound angles,

809850/0450

282558a

can be represented symbolically as follows:

^A EK7o ^A EK45 ^{: A} EKo ^{; A} TK45

This applies to sounding from both the right and the left.

Example 2. In the case of an extensive reflector 3, Fig. 1b., The various Probes, i.e. the probes that are operated under different Sound in angles, also receive echoes of different heights, represented symbolically:

^A EK7o £ ^A EK45 / ^ ^A EKo Φ ^Α ΤΚ45

This applies to sounding from the right as well as from the left.

As can be easily seen, different angles of incidence are selected to avoid errors in every angular position to be able to find.

5 is a block diagram for carrying out the method specified. The test heads or the test head arrangement Io is moved by the manipulator 11 over the surface of the test piece

909850/0450

1 taking into account the usual coupling techniques The manipulator is controlled by the computer 13 via a control device 12 according to a predetermined test program is, so that the coordination of the probe assembly to the Computers are known. In manual operation, the manipulator reports its actual coordination to the control unit Computer. From the ultrasonic device part 14, the pulses for the computer and for the probes generated. Amplifier for the received pulses, evaluation units and the A / D conversion for Digitization of the Α image are also part of the ultrasound device part and are not drawn separately. The computer has the usual peripheral units, such as a printer 15, a data storage device (tape recorder) 16 and an input and display device 17 assigned and coupled.

If, in example 1, when the probe arrangement is displaced in both the X and Y directions, a single echo is obtained which during the displacement emerges from the threshold value bohe or η

grows out of the zero line and then subsides again, A is a single pore (area a in FIG. 2a).

Obtains a F "-e separate individual echoes, ie during the displacement of the probe in a coordinate direction increases an echo disappears within a predetermined displacement path ^ after a further displacement of the probe arrangement creates a new \

909850/0450

Echo and disappears again within a displacement path, etc., it is an accumulation of several pores (row of pores) (area b in FIG. 2a). In both cases, the echo dynamics have a maximum over a predetermined area A t of the time line. The echo dynamics correspond to the echo amplitudes. Echodynamics is understood to mean the range of fluctuation of the envelope of the successive echo amplitudes.

If the A-picture is a superimposition of many individual echoes that are not are more resolved (area c of FIG. 2a), this is classified as a pore nest according to predetermined criteria. The echo dynamics within the area / it is much smaller than in Area a or b.

This is obtained in example 2 when the probe arrangement is moved both in the X and in the Y direction a single echo (area a in Fig. 2b), which is only within a range ^ t is present, it is an approximately flat reflector surface with a linear expansion comparable to or smaller than that Sound beam cross-section. However, this display remains approximately the same in height when it is shifted in the X-direction (Area b of FIG. 2b), it is a reflector that is extended in the X direction. Are when moving in the X direction If a sequence of separate individual echoes (area c, FIG. 2b) is displayed, there are several individual reflectors.

009850/0460

If you get a display when moving in the X direction with significantly reduced dynamics in one area. t, as shown in area d, is a cluster planar reflectors.

You can also use reflectors lying one on top of the other, i.e. in different Classify the depth in the test piece by moving a test head in / -direction.

Decision values obtained from the above knowledge are given to a computer which, using the Coordinate values for the movement of the probe and the echo height in the A-picture are classified by the reflectors. Necessary What is involved here is that the A-picture is fed to the computer in digitized form using methods known per se will. In addition, a memory that is connected to the computer can store the data of the respective areas— coordinates of the probe arrangement, the respective echo height and save the successful classification of the reflectors. In the case of a repeat test, the findings can be compared with one another and using the addresses of the surface coordinates Changes in findings can be determined automatically.

The test is thus reproducible, provided that the other relevant data of the test are also saved.

809850/0450

An example of a procedure int iie classi f izierunt; in a single poro, toroiianhäiiTurin, tOrunn ^ fsfc, flat, short Eiruoe! reflector (Re flek to! 'Extension smaller than üchal lbündel <iuurijchnitt), Flat long single reflective cores and an accumulation of flat reflectors.

First of all, the calculator determines whether the condition

* = ^{: a} eko - ^ ^{1 A} TK70 "^ ^A TK45" ¹ ^ ^etc *

or the condition

■■ / ^A TK70 / ^A TK45 / ^etc. * is fulfilled. The computer has already made the assignment ^

1st class "spherical reflector" or

2nd class "flat reflector."

1. Spherical reflector: If it is a spherical reflector, a base point width must be specified on the time line for a further decision, preferably not in time units but in length units of the coordinate system. This dot width, e.g. B. £ _Xj depends on the test equipment used. It must be slightly larger than the base point width of an echo from a small test reflector and can be determined by measuring it on a reflector. So a test reflector, / around a borehole, is looked at. In the

909850/0450

Shifting the probe arrangement in the X direction now becomes the time line? a leno emerge, grow and subside again (area a uei Fig. 2a). This region on the time line was referred to as Δ t in the previous discussions. Here this area should now be called A x, since it relates to the displacement of the probe arrangement, where -A χ «c .At with c ■ is the speed of sound. A parameter k is established with the condition that k is a little larger than ^ d χ (Fig. 2). In principle, the same applies to the displacement of the test head arrangement in the Y direction. In this example, the same factor k is used for the Y direction. Whether the same factor is used as a decision aid for the computer in both directions also depends on the radiation characteristics of the probes and considerations of expediency. It is not the level of this decision factor that is important for the innovation, but that it is formed.

Now fluctuate with manual or computer-controlled shifting of the probe assembly the amplitudes of each other following echoes by more than one dynamic value w (area b of FIG. 2a) measured in dB, is by definition a Pore accumulation present. If the echo occurs when the probe is displaced for the distance k but only once when the w-value is exceeded, it is a single pore (area a Fig. 2a). The size of w becomes according to the test problem defined as a dynamic measure. The size w as another

8O98S0 / G4SO

The decision factor is also not for the innovation It is important, rather, that this quantity is formed.

This is explained in more detail in FIG. 3.

When the test head arrangement is shifted in the X direction, echoes from position X to X occur, the displays of which repeatedly return to the zero line within the path k, ie the dynamic value w is exceeded within the range k. This is to be classified as a row of pores. There is _{no display between positions X 2} and X ₃ when shifted by k. This area is reflector-free. An echo is present at position X ^ which disappears again within the range k. A single pore can be assigned to this echo. There is no display up to pos. X-. The area is reflector-free. From X ₄ to X _{5 there} are echo displays, the fluctuation of which is smaller than the value w, but the displacement range of the test head arrangement in which echoes occur is greater than k, so that this is to be classified as a pore nest. There is no display from Xr to X _& a distance that is greater than the value k. So this is to be assessed as a reflector-free area, in other words as a separation to the next Iorennest from X, to Xy, there with the Eos. X ^ to Xj again displays with a Dynairik 1 · leiner w are available. If the ad-free Sliccic from X _t his. X, Vleiriei as the quantity k, the Ifif'di Nf tier should read a 2Uf arrrr enhänqeride ε from the I ί eimer air

fi for £ 98 »0/0 / .b0

be classified. The shifts of the probe arrangement in each case in the i-direction q provide information about the Expansion of the reflector areas with respect to the Laqe in the Weld width. It is advantageous to abet on a case-by-case basis it is not absolutely necessary to use the same decision factors k, w for the / -direction as for the X-direction.

Mari also does not have to be classified as: "uniform Pore nest "/" separated pore ridge "use the factor k obtained from the base point width, but one can use one specify another factor.

2. Flat reflector: If the computer has decided that flat reflectors are involved (Fig. 4-), the computer has to decide according to the following criteria: The test head arrangement is moved again from, for example, left to right according to FIG X direction. At pos. X. an echo occurs with the characteristic that it is only present for a probe displacement smaller than the factor k. This display can therefore be assigned to a flat individual reflector that is smaller than the sound beam cross-section. There is no display up to pos. X _n. This area is reflector-free. At X ₀ an echo occurs which remains approximately constant in its display up to position X, ie the amplitude dynamics are smaller

909850/0 4 50

-IH-

as a preqeqebener «tfert ν, which, similar to the / tfert w .ms, is stipulated in the performance conditions and for clarification in the Fiq. 4 is indicated by two horizontal lines as a dynamic range. The computer classifies this display as an extended long reflector. From Pos. X to Pos. X, there are no indications. This area is reflector-free. At X ₁ up to pos. X ^ displays appear whose height fluctuates more than it corresponds to the dynamic value w, so that this area is classified by the computer as an array of reflectors. From X ₁ - to X, no display occurs. The area is evaluated as reflector-free if the shift is greater than k. A display appears from X. to X-7, the location-dependent amplitude fluctuation of which is smaller than the value w but larger than ν; the computer classifies this as a reflector accumulation.

There is _{no display from X 7} to X ^. Of X ₈ to Xg occurs Am; on eiqe, _m ifc the same characteristics as in the movement of the test head assembly of Xg to X ₁ - ,, so that it is classified as another lektoranhäufunq ran if the Verschietunqsstrecke Kj to X , q is greater than k. If the range of verses Xy to K ₁ -, is smaller than the i / ert k, then the computer has to? av / eite reflector accumulation with the first reflective cluster as a "closed larger reflector cluster" from H- to X.-. / erten.

0 0 0 0 α 0/0 h 50

For the displacements of the test head arrangement in the Y direction; the same decision criteria can apply or other decision parameters can be selected according to the test problem will.

Table 1

^A EK 70 - ^ ^A EK 45 '^^ ^A EK 0 -AA coincidence: spherical reflectors

Single pore

Row of pores in the X direction

Pore nest

70 4 * ^A EI- ^ 5 ^ ^A EK 0 No coincidence: flat reflectors

A Ax) w; zlx <lc = >> Imrzer flat single reflector in x-direction

Anieihuiif by ilfichi ic lic 1 t oi (i.

8098BOVOASO

r = 5> long flat reflector in x-direction

v </ 4 Ax <w; <^ x> k accumulation flat. Reflectors

^A EK 70 ^ ^A EK 45 ^ ^A EK 0 No uniqueness

Might classifiable disorder

Check of the system, if necessary, manual or re-examination

In Table 1, the decision-making features for the computer are summarized again. So the computer only has after Plan of table 1 alternatively to decide whether a first condition is met or not, in order to then follow the planned logic Check the following conditions and print out the findings from the results. The classification of the reflector location and the depth of a reflector in the test piece and the specification of the reflector size takes place from the coordinates of the manipulator, the transit time and the echo height in a manner known per se.

A further embodiment of the invention provides that the echo amplitudes are compared with default values according to j of a known type after the depth compensation has taken place, and the method is only used for

809850/0450

EcIi ο amp Ii tud en becomes effective that exceed the "default value, whereby the reflectors generating these echoes are automatically classified as faults.

Reflectors that generate smaller echo amplitudes than this default value are not subjected to the evaluation, i.e. not rated as an error.

The reflectors are located from the transit time according to the known ones Methods.

To further expand the automation, the testing of Welds the computer the position of the weld, represented by their coordination, can be entered, so that this only Evaluates echoes from this entered area, v / eiterhin can the liahtform is fed to the computer in the form of the corresponding data can be used to evaluate certain reflector classes differently, e.g. to decide whether a reflector is classified as Whether or not an error is to be seen is a relation of the size of the error can be formed for the IT seam geometry and used as a decision variable. The computer can also send data about the welding process must be entered in order not to evaluate reflectors, which may be typical of the process, as errors.

The method allows the advantages of the error dynamics in manual testing to be applied to machine or computer-controlled, i.e. automated test processes,

909850/0450

to apply with ultrasonic pulses in such a way that signal values are fed to the computer as decision criteria so that he is forced to perform the reflector classifying steps and to be classified according to error or non-error.

909850/0450

Blank page

Claims (4)

May 24, 1978 IG / vh KRAUTKRAMER GMBH K 103 PATENT CLAIMS m) A method for the automatic generation of a test result in the non-destructive testing of materials with ultrasonic pulses by evaluating two or more successive digitized Α-images obtained by changing the position of the test head arrangement, which preferably consists of several test heads with different radiation characteristics, characterized in that for one position , which is specified in coordinates of the test piece surface, created a digitized A-picture by means of this probe arrangement in a known manner and in an electronic or the like. Memory is stored, then at least one change in position of this probe arrangement in the X and / or Y direction, depending on these coordinates, is made and thus additional Α images are created and stored, the changes in the echo amplitudes (ΔΑ) of the successive Α images are separated in the X and Y directions with comparative values of the same dimensions, for example electrical voltages, for the echo amplitudes (echo dynamics) and for the change in position or the displacement path of the test head arrangement in a computer according to a predetermined sequence, for example after 8098S0 / 0450 Table 1, compared and assigned to specified error classes and displayed, registered and / or used as manipulated variables for a subsequent work stage 2. The method according to claim 1, characterized in that one or more evaluation gates are present in the signal path of the echo pulses are that allow only echoes with a minimum amplitude to be classified, so that the computer according to predetermined Decision variables only reflectors whose echoes exceed the specified value are classified as errors. 3. The method according to claim 1 and 2 for the classification of reflectors only within a test piece area, characterized in that that the coordinates of the area to be tested are given to the computer as characteristic values. 4. The method according to claim 1 and 2 for the evaluation of classified Reflectors, characterized in that decision values are given to the computer in such a way that one or several reflector classes can be assessed as faults. 909350/0450