GB2075678A

GB2075678A - A method of determining a fault expectancy zone for ultrasonic testing

Info

Publication number: GB2075678A
Application number: GB8109667A
Authority: GB
Original assignee: Krautkraemer GmbH and Co; Krautkraemer GmbH
Current assignee: Krautkraemer GmbH and Co; Krautkraemer GmbH
Priority date: 1980-05-09
Filing date: 1981-03-27
Publication date: 1981-11-18
Also published as: DE3017900A1; DE3017900C2; GB2075678B; JPS576357A

Abstract

In the ultrasonic testing of a pipe, for example, in order to exclude the pulse 5 and echoes 6 and 7' from consideration and concentrate evaluation on echo 7, a fault expectancy zone 10 is established. The method employed involves calculating the start time and duration of an approximate expectancy zone, subdividing said approximate zone into a plurality of shorter time intervals and, in each of a plurality of cycles of operation, transmitting an ultrasonic pulse into a test piece and recording the echo amplitude occurring during a respective one of such intervals, and thereafter locating the mid-point of the fault expectancy zone to be determined within the time interval wherein was recorded the maximum echo amplitude. <IMAGE>

Description

SPECIFICATION Improvements in ultrasonic testing In the non-destructive testing of workpieces in the form of pipes or rods, it is known to use a testing installation including test heads which rotate about the axis of the workpiece, or which remain stationary while the workpiece rotates about its axis. As the workpiece passes through the installation there is transmitted into it an ultrasonic signal in the form of a series of pulses.

Echoes are reflected from discontinuities, including the external surface of the workpiece, and the internal surface if the workpiece is in the form of a pipe, as well as from flaws.

The ultrasonic echoes are picked up by the test heads, each of which incorporates a transducer which is used to convert the ultrasonic signal into an electrical echo signal. This signal is evaluated and used to provide information concerning the workpiece. In carrying out this method it is necessary to establish at least one temporal fault expectancy zone. Echoes received within such a zone are taken to relate to flaws which need to be evaluated whereas echoes received before the beginning of a zone or after the end of a zone are disregarded.

Conveniently, there may be separate zones for internal and external flaws.

In the known pipe testing installation, the fault expectancy zones are determined manually by means of electronic windows. For this purpose, the ultrasonic tester compares the position of the test piece fault echoes on the screen of a monitor and establishes the zones which are also displayed on the screen in the form of a line.

It has already been proposed to establish a fault expectancy zone automatically in connection with an ultrasonic test installation for testing sheets of metal, see for example the Federal German Specification DT-OS 2321699, 2422439 and 2620590. In these known methods, the total sheet metal thickness is used as the basis for the fault expectancy zone and it is therefore only necessary to determine and store the test head transmission length which is the distance between the ultrasonic oscillator and the surface of the test piece, musing a test piece which has a known thickness and through which sound is propagated at a known velocity.The fault expectancy zone for a metal sheet under test can then be derived from the corresponding back wall echo, the test head transmission length being subtracted from the corresponding transit time value for that echo.

This method cannot be employed in pipe testing without difficulty. Firstly, because of the inclined angle of incidence of the ultrasonic wave on the curve surface of the pipe it is impossible to obtain a back wall echo comparable to that which arises from a sheet. Further, the angle of incidence and hence the fault expectancy zone vary depending upon the wall thickness to diameter ratio of the pipe. Secondly, the echoes which are mainly of interest in pipe testing are those arising from the external and internal surfaces of the pipe. Accordingly, fault expectancy zones should be so established that only echoes from these areas are evaluated.

Certain of these problems are also encountered in the testing of solid rods and bars.

In order to overcome these problems, there is proposed herein a method wherein, to determine the mid-point for a fault expectancy zone, a test piece with a test flaw is pulsed with ultrasonic energy and the magnitudes of the echoes received during each of a plurality of respective intervals of time are compared, and the time interval associated with the echo of greatest magnitude is selected, and the mid-point of the zone is located in that interval.

Conveniently, the operation is performed in a series of cycles, in each of which a pulse is transmitted into the test piece, and a time interval at a respective different time after the transmission of the pulse is examined. To exclude tirrie intervals other than time intervals in need of consideration, an approximate fault expectancy zone may be calculated from information concerning the test piece, and this approximate zone is divided into the time intervals which are used as described above.

In the drawings: Figures la and ib show cross-sections of a pipe with internal and external faults respectively to illustrate the transmission of an ultrasonic wave from a test head, Figures 2a and 2b depict the ultrasonic echoes arising from the pipes shown in Figs. 1 a and 1 b, and Figure 3 shows diagrammatically a circuit for performing the method proposed herein of establishing fault expectancy zones.

Referring to the drawings, each of Figs. 1 a and 1 b depicts a test head 1 of an ultrasonic installation for evaluating flaws in a pipe which rotates around its axis, although in an alternative arrangement, the test head may follow a circular path about the axis of the piipe under test. For the purpose of establishing the fault expectancy zones, use is made of a test pipe 2 having a test flaw 3 in its internal surface (see Fig. 1 a) and a test flaw 8 in its external surface (see Fig.

1 b). An ultrasonic wave emitted from the test head into the test pipe 2 shown in Fig. 1 a, is reflected from the test flaw 3 and the echo received by the head 1. Fig. 2a shows the forms of the pulses plotted against time, depicting the emitted pulse 5, the entry echo 6 produced by the ultrasonic wave striking the external surface of the pipe and the flaw echo 7 produced by reflection from the test flaw 3. Another echo 7' is obtained as a result of the rotation of the pipe allowing the ultrasonic wave 4 after reflection from the internal and external surfaces of the pipe to re-encounter the test flaw when the latter is in the position shown at 3'.

Fig. 1 b and Fig. 2b show the corresponding conditions for the external flaw 8 which is positioned at 8' when re-encountered by the wave 4. The corresponding echoes are indicated at 9 and 9' in Fig. 2b.

In order to exclude the pulse 5 and echoes 6, 7' and 9' from the evaluation, it is necessary to establish the fault expectancy zones 10 and 11 depicted in broken lines. In known testing equipment, these zones are displayed on the monitor screen in the form of window signals. The expectancy zones can then be adjusted by the ultrasonic tester to ensure that the test flaw echoes 7 and 9 which also appear on the screen fall within the zones.

Fig. 3 shows the test pipe 2 with the internal test flaw 3 and the external test flaw 8, together with parts of the installation showing a circuit for automatically establishing the fault expectancy zones. The installation comprises a transmitter 12 which in operation energises a transducer within the test head 1 to transmit a pulse into the pipe 2, a receiver amplifier 1 3 which amplifies the electrical signals generated by the transducer in response to ultrasonic echoes which it receives from the pipe, and an analogue,/-digital (A/D) converter 14, a window generator 1 5 and an evaluation unit 1 6. A position transmitter indicates the exact position of the pipe to the evaluation unit 1 6.

The A/D converter 1 8 is arranged to digitalise only those echo signals which come within the fault expectancy zones predetermined by the window generator 1 5. The digital echo values are then processed further in the evaluation unit 1 6. The installation may include a group of test heads which are employed in conjunction with one another to provide information concerning the condition of the pipe.

The window generator 15 comprises two reversible digital counters 151 and 152. Each counter has eight inputs, which are connected to the evaluation unit 1 6 via lines 1 7 and 18, while the ouputs are connected to the A/D converter via the lines 1 9 and 20.

The exact determination of the fault expectancy zones will be described in detail hereinafter in respect of the zone pertaining to internal flaws, i.e. longitudinal flaws, it being understood that the zone pertaining to external flaws may be determined in the same way. Initially, the test pipe thickness D, the average velocity C of sound within the material of the test pipe, and an estimated angle of incidence a (e.g. 45 ) are fed into the evaluation unit 1 6 which determines the duration of an approximate fault expectancy zone.For example, an approximation to the start time ts and end time tE for the zone may be made by assuming D ts s 2C cosa and 3D tE tEd , 2C cosa it being assumed that the ratio of the wall thickness to the outside diameter of the pipe is less than 0.2. The reversible counter 1 51 is automatically set to a value corresponding to the start time ts of the approximate fault expectancy zone so calculated. Counter 152, on the other hand, is set to the value tB of a partial fault expectancy zone much smaller than the duration (tE - ts) of the precalculated approximate fault expectancy zone.Thus, tB may be set to 1 /5 (tE - ts), this value of ts thus constituting the duration of a window used to form five time intervals into which the approximate expectancy zone is to be divided.

A trigger 22 connected to a synch generator 21 now triggers both the transmitter 12 and the counter 152 also connected to generator 21 and the following process takes place: One the preset counter has been counted down, a pulse is generated at the output of counter 1 51. This pulse releases the A/D converter via the line 1 9 so that the electrical echo signals now arriving are digitalised.

Counter 1 52 is also started via line 23. After the preset count of this counter has been counted down and the first cycle of operation completed, a signal appears at the output of counter 1 52 and stops the A/D converter. Subsequent echo signals are not digitalised. The maximum value of the echo signals falling within the window and first time interval thus predetermined is obtained in the evaluation unit. The first line of a table is then input to the memory 161, this table showing the maximum digital echo value obtained as a function of the associated window or time interval.The entire operation is then repeated during a second cycle on the next revolution of the pipe, the start time t's of the window being delayed by a time corresponding to the duration of the window tB. This sequence of operations is then repeated until the window reaches the end time tE for the approximate expectancy zone. The completed table is for example therefore as follows: Revolution tB t's Maximum echo amplitude 1 Constant ts 10 2 ,, to + to 30 3 ,, ts + 2ts 40 4 ,, ts + 3ts 20 5 ,, ts + 4tB = tE 7 When the above table has been completed, the maximum echo amplitudes are compared, e.g.

by means of computer 162, and the value t's used as the midpoint tM of the corrected fault expectancy zone is that at which the greatest maximum echo amplitude was measured. This is the case for ts + 2tB in this example.

Having thus determined its mid-point, the final expectancy zone may be established with a duration of suitable length, for example + D/2C relative to the mid-point.

To sum up, the procedure followed has been to create an approximate fault expectancy zone having a pre-set start time ts and a duration (tE - ts) depending upon the workpiece, and to subdivide this approximate zone into a plurality of different time intervals, each equal in length to the calculated partial fault expectancy zone. In each of a number of cycles of operation, a pulse is transmitted into the workpiece, and the maximum echo amplitude received during a respective one of these intervals is recorded in order to discover in which of these time intervals is displayed the maximum echo amplitude. The mid-point of the fault expectancy zone to be established for the testing of production workpieces is then located within this interval.In the preferred method, recording takes place sequentially, starting with the time interval nearest the start time ts and progressing towards the end time tE of the approximated zone but of course it is not essential for such an orderly sequence to be followed. Conveniently, the centre of the time interval is chosen to form the mid-point.

It has been found advantageous to repeat the entire operation several times in succession. By comparing the individual fault expectancy zone mid-points it is then possible to determine whether random faults (e.g. water bubbles between the test head and the testpiece) are falsifying the results. This may be assumed to be the case, for example, if the difference between two successively determined values for the mid-points of the fault expectancy zone is greater than a predetermined limit.

Of course the invention is not restricted to the testing of pipes, but can also be used equally successfully in testing rods or bars for external and core faults. In this case, a single fault expectancy zone may be established, instead of two zones as required for the testing of pipes. It may also be possible for the recording of echoes received in each of the successive intervals of length tB to be carried out during a single cycle provided the equipment is suitable. In other words, a single pulse may be transmitted into the test piece and the maximum echo amplitude received in each of a plurality of successive time intervals is recorded, that interval displaying the maximum amplitude being selected as the mid-point. While it is preferred for the time intervals to be discrete, it is of course possible for them to overlap.

Claims

1. A method of automatically establishing a fault expectancy zone starting at time ts and ending at time tE in an installation for the non-destructive testing of pipes or rods operating on the pulse echo principle, comprising transmitting pulses into a test piece having test faults, wherein approximate values for the times ts and tE are calculated from predetermined data concerning the test piece, a partial fault expectancy zone of width tB is formed wherein tB is substantially less than the calculated value of (tE - ts), the partial fault expectancy zone is relocated within the pre-calculated fault expectancy zone of width (tE - ts), in each of a plurality of cycles and the maximum echo amplitude within the partial fault expectancy zone in each cycle is determined and stored, the position of the partial fault expectancy zone relative to the precalculated zone at which the greatest maximum echo amplitude is measured is determined, and this partial fault expectancy zone is used as the mid-point of the fault expectancy zone to be established.

2. A method as claimed in claim 1, wherein the partial fault expectancy zone is move cyclically from the start ts to the end tE of the calculated expectancy zone during such re location.

3. A method according to claim 1 or claim 2, wherein one cycle corresponds to one relative revolution of the test piece and test head.

4. A method according to any preceding claim, wherein the same test piece is used and the mid-point of the fault expectancy zone is determined several times in succession and these midpoirits are compared with one another.

5. A method of establishing a fault expectancy zone substantially as hereinbefore described with reference to the drawings.

6. An installation substantially as iilustrated in Fig. 3.

7. A method of establishing the mid-point of a fault expectancy zone for use in ultrasonic testing, comprising calculating the start time and duration of an approximate expectancy zone, subdividing said approximate zone into a plurality of shorter time intervals and, in each of a plurality of cycles of operation, transmitting an ultrasonic pulse into the test piece and recording the echo amplitude occurring during a respective one of such intervals, and thereafter locating the mid-point of the fault expectancy zone to be determined within the time interval wherein was recorded the maximum echo amplitude.

8. The method wherein, to determine the mid-point for a fault expectancy zone, a test piece is pulsed with ultrasonic energy and the magnitudes of the echoes received during each of a plurality of respective intervals of time are compared, and the time interval associated with the echo of greatest magnitude is selected as the mid-point of the fault expectancy zone.

9. The method claimed in claim 8, wherein the echoes received in each time interval is produced by a respective pulse in a respective cycle of operation.