JPS6356947B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an ultrasonic flaw detector for detecting defects in metal materials, welded parts thereof, etc., and particularly relates to improvements in processing means for reflected echo signals.

This type of ultrasonic flaw detection equipment, which has been known in the past, involves manually scanning a probe, amplifying and detecting the reflected echoes received by the probe using a flaw detector, and then
There is a manual flaw detection method that directly displays this and manually judges defects, but since this manual flaw detection method uses manual scanning, it lacks speed and accuracy, and it is difficult to detect true defects. This method has drawbacks such as difficulty in determining defects because it displays a mixture of bottom echoes and other pseudo-echoes in addition to echo signals.

Therefore, recently, automatic flaw detection methods have been adopted, which eliminate the drawbacks of the manual flaw detection methods described above. FIG. 1 shows an example of its configuration, in which the probe 1 is automatically scanned by the scanning mechanism 2, and the reflected echoes as shown in FIG. 2A received by the probe 1 are detected by the flaw detector 3. After amplification and detection, the threshold circuit 4 extracts only the signal that exceeds a certain level threshold L as shown in FIG. As shown in FIG. 2, the pulse signal is generated to indicate the peak value of each signal, and the magnitude of the pulse signal is displayed on the display 6 along with the time axis. According to the conventional automatic flaw detection method described above, since the probe 1 is automatically scanned and the reflected echo signals are automatically processed, one cross section can be automatically displayed with one one-dimensional scan. . Therefore, compared to the manual flaw detection method described above, defects can be determined relatively quickly and accurately.

However, in the conventional automatic flaw detection method described above, defects are judged by treating anything above a predetermined threshold value by the threshold circuit 4 as a defect echo signal, so there is no problem if the defect is very large. There is essentially no difference from the above-mentioned manual flaw detection method in that flaw detection and evaluation is performed simply based on the magnitude of the reflected echo signal, and there is a problem in the accuracy of discrimination between true defect echoes and pseudo echoes. Especially the first
As shown in the figure, when detecting a defect 9 in a weld 8 of a base metal 7, in addition to the true defect echo e1, a bottom echo e from the bottom of the base metal is detected.
2. Pseudo echoes such as a boundary echo e3 from the interface between the base material 7 and the weld 8, a back wave echo e4 generated by the back wave of the weld 8, and a forest echo e5 generated by particles in the weld. occurs frequently, so it is not easy to distinguish between the true defective echo e1 and the pseudo echoes e2 to e5 based on the size relationship of the reflected echoes. Furthermore, displaying peak values above the threshold value excludes defective echo signals below the threshold value, so there is also the drawback that minute defects cannot be detected. Furthermore, since detection is performed as a preprocessing of the reflected echo signal, for example, as shown in FIG. 3D, the defective echo signal e1 and the pseudo echo
When there is interference with en, the interference signal e
If 0 is detected, the waveform will be deformed as shown in E in the figure, and the presence of the defective echo signal e1 will no longer be noticeable. Moreover, when the deformed signal is subjected to differential processing, the phase of the signal showing the peak value as shown in FIG.

The present invention was developed in consideration of these circumstances, and its purpose is to be able to detect defects quickly and accurately even in flaw detection in locations where pseudo echoes occur frequently, such as near welds, and to also allow minute defects to be overlooked. It is an object of the present invention to provide an ultrasonic flaw detection device that can detect with high accuracy without any trouble, and can easily display the required cross section in a timely manner.

Hereinafter, the present invention will be explained in detail with reference to embodiments shown in the drawings.

FIG. 4 is a block diagram showing the configuration of an embodiment of the present invention, in which an ultrasonic probe 11 is driven by a scanning mechanism 12, and a flaw-detected member 1 having a welded portion 13a is
along the surface of 3, it is linearly scanned as x1, x2,... y1, y2...yn at predetermined intervals. In other words, it is scanned two-dimensionally. The flaw detector 14 applies a high voltage pulse to the probe 11 at a predetermined address scanned as described above to cause it to emit ultrasonic waves,
The reflected echoes received by the probe 11 at a predetermined address are amplified as a series of signals corresponding to the time axis. The scanning mechanism 12 and the flaw detector 14 are controlled by control signals from a controller 15 provided on, for example, an operating console. A sampling processor 16 samples a series of reflected echo signals amplified by the flaw detector 14 at a unit time of a predetermined width. The absolute value processor 17 obtains the absolute value signal of each sampled signal by inverting the polarity of all negative signals below zero level into positive signals for the sampled reflected echo signals. The three-dimensional memory 18 stores the absolute value signal output from the absolute value processor 17 at a predetermined address according to the address designation signal from the controller 15.

FIG. 6 is a perspective view schematically showing the three-dimensional memory. As is clear from this figure, the three-dimensional memory 18 stores the X-axis indicating the width direction positions x1, x2, . Along three axes in which the Y-axis indicating the longitudinal positions y1, y2...yn and the Z-axis indicating the time axis of the reflected echo signal (=thickness direction position of the tested member 13) z1, z2...zn are orthogonal to each other. Each memory element M is arranged three-dimensionally, and the series of reflected echo signals is sequentially stored in each memory element z1, z2, . . . zn at a predetermined address, for example, x1, y1.

The classification processor 19 is the three-dimensional memory 18
The reflected echo information read out from the absolute value processor 17 or the signal outputted every moment from the absolute value processor 17 is classified into classes according to the magnitude of the absolute value. The color display 20 is composed of, for example, a cathode ray tube, and performs predetermined color modulation on each of the classified signals, and displays a cross-sectional image determined by, for example, the X-axis and the Z-axis at each position y1 on the Y-axis. ,y2
...Display about yn. The color printer 21 prints out the content displayed on the color display 20.

Next, the operation of the apparatus configured as described above will be explained. First, when the scanning mechanism 12 and the flaw detector 14 are activated by a control signal from the controller 15, the probe 11 moves along the surface of the member to be tested 13 at y1, x.
1, y1, x2, . The received reflected echoes for the first addresses y1 and x1 are amplified by the flaw detector 14 as a series of reflected echo signals corresponding to the time axis as shown in FIG. 5G.
This amplified signal is sent to a sampling processor 16
is sampled at a predetermined time width t, and then the absolute value processor 17 inverts the polarity of all negative signal portions indicated by diagonal lines in the waveform of FIG. 5G, as shown in FIG. 5H. It becomes an absolute value signal with positive polarity. The above absolute value signal is sequentially stored in each memory element on the Z axis at a predetermined address of the three-dimensional memory 18 specified by the controller 15, that is, x1, y1. At the same time, the absolute value signal is classified by a classification processor 19 according to its size range L1, L2, . . . and provided to a color display 20. Thus color display 2
0, predetermined color modulation is performed on each classified signal, and the signals are sequentially plotted on the display screen corresponding to the probe position x1 as shown in FIG.
A similar operation is performed for the next addresses y1 and x2, and the absolute value signal is stored in the three-dimensional memory 18, and the color-modulated signal for each class is plotted on the x2 axis of the color display 20. . When the above operation is repeated and the first one-dimensional scan of the probe 11 is completed, the reflected echo information of the member 13 at the position y1 is stored in the three-dimensional memory 18 as an absolute value, and the color display At 20, a cross-sectional image based on classified and color-modulated signals is displayed as shown in FIG. This cross-sectional image is printed by a color printer 21.

Next, when the scanning mechanism 12 moves the probe 11 by Δy in the Y-axis direction and performs one-dimensional scanning in the X-axis direction at the position y2, the same operation as described above is performed, and the three-dimensional memory 18 stores y2. Reflected echo information of the member 13 at a certain position is stored as an absolute value, and a cross-sectional image based on classified and color-modulated signals is displayed on a color display 20, and this is printed by a color printer 21.

As described above, at the same time as the probe 11 scans, the color display 20 displays the cross section and the color printer 2
1 can be printed, but if necessary, if the reflected echo information stored in the three-dimensional memory 18 is read out at a timely manner, it can be displayed in the same manner as in the above case, or
You can print it out.

Thus, normally, from each cross-sectional image at each position y1, y2...yn on the Y axis shown in FIG. 7, E1 is a defect echo, E2 and E3 are back wave echoes of the welded part,
E4 is easily determined to be a forest echo. If it is difficult to distinguish only from the above-mentioned cross-sectional image, the controller 15 can be
gives a command to display a cross-sectional image at a specific position on the Z-axis or a specific position on the X-axis to the three-dimensional memory 18. Then, among the reflected echo information stored in the three-dimensional memory 18, for example, information from a group of memory elements arranged two-dimensionally along the X-axis and the Y-axis at a specific position Zi on the Z-axis is selectively selected. Defect information and the like when the member 13 is horizontally cut is displayed or printed as a cross-sectional image. Therefore, in this case, a cross-sectional image that is perpendicular to the cross-sectional image shown in FIG. 7 is obtained, so that the situation of defective echoes can be determined from a different angle.
Similarly, by extracting and displaying information from a group of memory elements arranged two-dimensionally along the Y and Z axes at a specific position xi on the is obtained. Therefore, a more precise determination of defective echoes can be made.

In this way, this device does not simply display a cross-section of reflected echoes based on their size, but instead displays and prints the reflected echoes that are color-modulated according to the magnitude of the absolute value-processed signal, which eliminates defective echoes. Pseudo echoes can also be understood as signals at the same level.
Therefore, even minute defect echoes can be detected accurately. In addition, it is possible to understand defect echoes as spatial spreads due to color distribution, and it is also possible to obtain cross-sectional images that are 90° different from each other with respect to any two axes of X, Y, and Z, so defect echoes and pseudo echoes can be distinguished. Regarding changes in vibration waveforms caused by interference, defective echoes and pseudo echoes can be clearly distinguished. Furthermore, since the reflected echo information stored in the three-dimensional memory 18 can be output for display, printing, etc. whenever necessary, there is also the advantage that defect analysis can be performed at an appropriate time.

Note that the present invention is not limited to the above-mentioned embodiment. For example, in the embodiment described above, the reflected echo signal is sampled and then subjected to absolute value processing, but sampling may be performed after absolute value processing. Further, in the above embodiment, the case where the reflected echo signal is sampled on a fixed time axis is illustrated, but after absolute value processing of the signal shown in FIG. 5G is performed, only its maximum value is extracted, and this maximum value is stored and displayed. You may also do the following. In addition, in the above embodiment, the case where each classed signal is color-modulated and displayed in color and printed in color is exemplified.
Display with display elements such as a combination of different types of marks,
Printing etc. may also be performed.

As explained above, according to the present invention, reflected echo signals subjected to absolute value processing are classified into classes according to their magnitudes, and these are displayed using different types of display elements, so that defective echoes etc. can be spread out spatially. As a result, defects can be detected quickly and accurately even when detecting defects near welds where there are many false echoes, and even if the defects are minute, they can be clearly displayed using specific marking elements. This allows it to be recognized as a signal at the same level as other parts, making it possible to accurately detect minute defects without overlooking them, and to retrieve and display reflected echo information stored in three-dimensional memory in a timely manner. It is possible to provide an ultrasonic flaw detection device that can display accurate cross sections at any time. Furthermore, the three-dimensional memory of the present invention receives reflected echo information from a group of memory elements two-dimensionally arranged along the other two axes at a specific position on any one of the three axes, the X, Y, and Z axes. Since it can be selectively extracted, cross-sectional images viewed from different angles can be obtained, and defective echoes can be determined with high accuracy.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a schematic configuration of a conventional device, FIGS. 2 and 3 are waveform diagrams for explaining the operation of the device in FIG. 1, and FIGS. 4 to 7 are one embodiment of the present invention. Figure 4 is a block diagram showing the overall configuration, Figure 5 is a waveform diagram for explaining operation, Figure 6 is a perspective view schematically showing the configuration of the three-dimensional memory, and Figure 7 is a color display. It is a figure showing the display state of a device. 11... Ultrasonic probe, 12... Scanning mechanism, 13...
Flaw-detected member, 13a... Welded part, 14... Flaw detector, 1
5...Controller, 16...Sampling processor, 17...
Absolute value processor, 18... Three-dimensional memory, 19... Classification processor, 20... Color display, 21... Color printer.

Claims (1)

[Claims] 1. An ultrasonic probe, a scanning mechanism that scans the ultrasonic probe two-dimensionally along the surface of the member to be inspected, and a scanning mechanism for scanning the probe at a predetermined address scanned by the scanning mechanism. a flaw detector that applies a high-voltage pulse to emit ultrasonic waves and amplifies the reflected echoes received by the ultrasonic probe at the predetermined address as a series of signals corresponding to the time axis; an absolute value processor that processes the absolute value of the reflected echo signal, and a main scanning direction of the probe is set as the X axis so that the signal from the absolute value processor is stored in correspondence with the position of the probe. The sub-scanning direction perpendicular to this axis is the Y axis, and the thickness direction of the member perpendicular to each axis is the Z axis, and memory elements are arranged along each axis, and any of the three axes X, Y, and Z is arranged. A three-dimensional memory configured to selectively extract reflected echo information from a group of memory elements arranged two-dimensionally along the other two axes at a specific position on one axis; means for classifying the reflected echo information and the output signal from the absolute value processor according to magnitude;
An ultrasonic flaw detection apparatus comprising display means for displaying signals classified by this means using different display elements.