JP5191446B2 - Eddy current flaw detection method and apparatus - Google Patents

Description

  The present invention relates to an eddy current flaw detection method and apparatus for measuring the length of a defect imparted to a specimen by an eddy current flaw detection method.

  In the eddy current flaw detection method, an AC voltage is applied to the excitation coil while scanning the surface of the test object with an eddy current probe to generate an eddy current in the surface layer of the test object, which is a conductor, This is a nondestructive inspection method for determining the presence or absence of a defect by detecting current disturbance with a detection coil.

  There are many types of eddy current probes due to differences in coil types and excitation methods. Among these, the mutual induction type standard comparison type top-mounted probe outputs a strong signal when the eddy current probe floats, so it is not only suitable for checking the contact state in actual machine inspection, but also has a certain length. If there is a defect, a strong signal peak can be obtained at the end of the defect opening surface, so that there is an advantage that defect length sizing is easy. Further, by arranging the coils in an arbitrary direction and making them multi-numbered, signal information of a plurality of detection coils included in each detection mode can be obtained by a single flaw detection. Patent Document 1 discloses a known flaw detection technique using this format.

  There is also a so-called self-guided standard comparison method, and an example of this is introduced in Patent Document 2.

JP 2009-19909 Japanese Laid-Open Patent Publication No. 9-178710

  As disclosed in Patent Document 1, it is effective to perform a flaw detection using a plurality of coils because a wide area can be inspected by a single scan. When inspecting the base material part or welded part of the test body, when the defect propagation direction is parallel to the eddy current probe, the number of detection coils indicating the signal feature point of the defect end is generally limited to one, The defect length can be easily evaluated by the defect length determination method. In addition, when the defect propagation direction is orthogonal to the eddy current probe, the signal feature point at the defect end is placed on the same line on the flat display, so the defect length can be easily calculated by the defect length determination method. .

  However, in the case of a defect that has an inclination with respect to the traveling direction of the eddy current probe, the signal generated due to the defect edge straddles the signal waveforms of multiple coils, and which detection depends on the inclination of the defect. Since it is not known whether the peak of the detection signal can be obtained by the detector, it is necessary to extract the feature point of the defect end in detail while comparing a plurality of signal waveforms at each detector.

Therefore, in the eddy current flaw detection method of the present invention, an eddy current flaw is generated in the test specimen by the exciting coil, and a change in the eddy current is detected by a plurality of detection coils to determine the defect length of the test specimen. In the method
A probe equipped with an excitation coil and a detection coil is scanned over the specimen to detect eddy current change signals from multiple detection coils.
Select one signal that changed when approaching a defect from multiple detection coil signals,
Select one signal that changed when leaving the defect from multiple detection coil signals,
Specify the change start point position and change end point position from each selected signal,
The length of the defect given to the specimen is determined as the difference between the start point position and the change end point position.

In the eddy current flaw detector according to the present invention, a probe equipped with a plurality of sets of excitation coils and a plurality of detection coils is scanned on the test body, and the eddy current is generated in the test body by the excitation coil, and a plurality of eddy current changes are detected. Flaw detection mechanical device that detects with a detection coil, eddy current flaw detector that obtains signals from multiple detection coils and flaw detection position signals, and a computer that performs arithmetic processing using signals from the eddy current flaw detector And an eddy current flaw detector comprising a display unit for displaying the output from the computer,
The display unit has a plurality of detection coil signals when approaching the defect and a plurality of detection coil signals when moving away from the defect, with the horizontal axis representing the flaw detection position signal on the specimen and the vertical axis superimposed as an amplitude value. I decided to display it.

  According to the method and apparatus of the present invention, a wide range can be detected by a single probe scan, and the length of the defect can be easily determined.

The figure which shows the processing flow of the eddy current flaw detection method by this invention Diagram showing an example of element probe Diagram showing an example of a multi-probe Plan view of normally used biaxial scanning flaw detector Side view of commonly used biaxial scanning flaw detector The figure which shows the probe attachment of the flaw detector of the biaxial scanning normally used The figure which shows an example of the processing apparatus which processes the signal from the flaw detection apparatus of a biaxial scanning Diagram showing the flaw detection method using a flat specimen Plan view of signal strength Diagram for explaining signal waveforms The figure which shows the flaw detection result with the flat specimen Diagram showing the processing flow of the eddy current flaw detection method that is normally performed The figure which shows the relation of the detection coil which responds when flawing the defect of the oblique direction with respect to the probe Diagram of channel signal waveform at the time of flaw detection superimposed on the display Diagram showing two selected waveforms on the display

  The present invention will be described in detail below.

  First, FIG. 2 shows an outline of a top probe of a mutual induction type standard comparison system as an example of a probe applicable to the present invention. This probe is the same as that introduced in FIG. 2 of Patent Document 1 mentioned above. As shown in this figure, the multi-probe 1 shown in FIG. 2B is obtained by arranging a plurality (2a, 2b, 2c) of element probes 2 shown in FIG. 2A in the probe length direction. Since the element probe 2 is composed of one excitation coil 8 and three detection coils 9a, 9b, 9c as shown in FIG. 1A, three sets of detection signals are obtained with only the element probe. Furthermore, in the case of the multi-probe 1, a detection signal multiplied by the number of loaded element probes is handled and signal processing is performed.

  In this element probe, the direction in which the excitation coil 8 and the detection coil 9a are arranged is d1, the direction in which the excitation coil 8 and the detection coil 9b are arranged is d2, and the direction in which the excitation coil 8 and the detection coil 9c are arranged is d3. Then, it can be said that the detection coils 9b and 9c are arranged in a direction intersecting with the direction d1.

Multi probe 1 shown in FIG. 2 (b) is obtained by arranging the elements probe 2 shown in FIG. 2 (a) a plurality of sets d1 direction, is composed of three elements probes 2a, 2b, from 2c in the example of FIG. The multi-probe 1 has the number of channels corresponding to the number of arrayed detection coils 8.
Since channel switching is possible in a short time by switching control, it is possible to carry out a wide range of inspections with a single flaw detection.

  In both the element probe 2 and the multi-probe 1, the eddy current generated by the excitation coil 8 is detected by the three detection coils 9a, 9b, 9c placed at different positions. It has a detection mode, and data in each detection mode can be acquired by a single flaw detection.

  As mentioned above, the example of the top probe of the mutual induction type standard comparison system is shown as the probe applicable to the present invention. However, the present invention is not limited to this system, and a plurality of detection signals are obtained from any of them. Any device that detects a defect is applicable.

  The probe is usually used by being attached to a two-axis scanning flaw detector. As a two-axis scanning flaw detection apparatus, a known apparatus can be used, and an example thereof is shown in FIG.

  FIG. 3 shows an example of a two-axis scanning eddy current flaw detector when the multi-probe 1 is used. FIG. 4A is a plan view of the flaw detection scanning mechanism, and FIG. 4B is a side view thereof. As shown in FIGS. 4A and 4B, the outer shape of the flaw detection scanning mechanism is constituted by flaw detection mechanical frames 14a, 14b, 14c, and 14d, and a rail 16 is fixedly disposed between the frames 14a and 14b, and the rail The traveling vehicle 15 is installed to be movable. The traveling vehicle 15 on which the multi-probe 1 is mounted travels one axis in the X direction using the traveling rail 16, the traveling wheel 17, and the X-axis drive motor 18.

  The flaw detection scanning mechanism itself can be moved in the Y-axis direction by a Y-axis drive motor (not shown). Further, the traveling rail 16 rotates around the axis and can swing in the Y direction, and any finder position can be taken by these mechanics.

  FIG. 3C shows a pattern in which the probe positioned as described above is in contact with the surface to be tested for flaw detection. The element probe 2 or the multi-probe 1 (multi-probe in the example in the figure) is attached to the traveling vehicle 15 and is brought into close contact with the surface to be tested for flaw detection.

  The multi-probe 1 is pressed from above by air supplied by, for example, a compressor, and is in close contact with the surface to be tested. At this time, the excitation coil 8 and the detection coil 9 are pressed from above by air and the elastic material 20, but are protected by a buffer material 21 such as a sponge, for example, so that the multi-probe 1 is in close contact with the surface of the specimen. Can be flawed.

  FIG. 4 shows a general processing apparatus configuration in which a flaw detection signal is input using the flaw detection scanning mechanism 22 of FIG. 3 and flaw detection is performed by data processing. The apparatus performs multi-probe 1 or element probe 2, flaw detection mechanical device 22 equipped with position encoder 19, eddy current flaw detector 11, and calculates signal amplitude value and phase angle from detection signals at each coil and position encoder. The calculation unit 5 includes a computer 3 including a memory 4 for storing calculated signals and position information, an input unit 6 of the computer 3, a display unit 7, and the like.

  Next, flaw detection performed using the well-known flaw detection apparatus and arithmetic processing apparatus as described above will be described.

  First, as shown in FIG. 5, flaw detection is performed, for example, while driving the probe 1 in the direction of the defects 12 and 13 on the flat specimen 23. Here, the defect 12 is a defect parallel to the traveling direction, and the defect 13 is a defect orthogonal to the traveling direction.

  Here, the excitation / detection method in the probe length direction (d1 direction in FIG. 2A) by the combination of the excitation coil 8 and the detection coil 9a is the V detection mode, and the probe by the combination of the excitation coil 8 and the detection coil 9b. The excitation / detection method in the width direction (d2 direction in FIG. 2 (a)) is the H1 detection mode, and the excitation / detection method in the probe width direction (d3 direction in FIG. 2 (a)) is the combination of the excitation coil 8 and the detection coil 9c. The detection method is called the H2 detection mode.

  The positional relationship between the excitation coil 8 and the detection coil 9 is that the eddy current is most disturbed and the impedance change becomes large when the defect enters at a right angle with respect to the excitation direction. It is suitable for detecting the defect 12, and the above-described H1 and H2 detection modes are suitable for detecting the defect 13 in the direction orthogonal to the probe.

  FIG. 6 is a signal obtained when the detection coil 9b or 9c in the H detection mode in FIG. 5 detects the defect 13. FIG. 6A is a plan view of the signal intensity, and FIG. . The horizontal axis of the signal waveform in (b) represents the flaw detection position obtained until the probe approaches the defect, reaches the defect, and then passes, and at the left end portion of the defect 13 that appears as a plan view of the signal intensity. The maximum signal amplitude value 24a, the minimum signal amplitude value 25a, and the signal vanishing point 26a are extracted, and the maximum signal amplitude value 24b, the minimum signal amplitude value 25b, and the signal vanishing point 26b are extracted at the right end portion of the defect 13. In FIG. 6, as an example, a signal obtained when the detection coil 9 b or 9 c in the H detection mode detects the defect 13 is used.

  Thus, when flaw detection is performed while moving the probe, the signal begins to change as it approaches the left end of the defect 13, and then attenuates from 26a to 25a as the probe moves, and reaches a maximum at 24a. . Although it changes in the vicinity of the maximum value for a while, the signal attenuates from 24b to 25b when approaching the right end of the defect 13, and after increasing to 26b, the signal does not change.

  FIG. 7 shows the flaw detection result of the flat specimen 23. If the defect 12 is in the direction parallel to the probe, the signal amplitude maximum value 24 and the signal amplitude minimum value 25 appear in the signal waveform in the V detection mode and in the case of the defect 13 orthogonal to the probe in the H1 or H2 detection mode. Furthermore, the peak point is generally limited to the signal waveform of one channel.

  Next, the procedure of a general length measuring method in eddy current flaw detection will be described. FIG. 8 is a flowchart showing a well-known normal processing procedure when the defect length is measured from the flaw detection signal taken into the processing apparatus of FIG. 4 using the biaxial flaw detection apparatus of FIG. Hereinafter, a general method for measuring the length of the processing when the signal of FIG. 6 is obtained will be described.

  As a procedure for measuring the length, first, a plane display indicating the signal intensity is created in step 101 based on the result of flaw detection using each probe. For example, the signal waveform in each detection coil 9 is displayed. Next, a plurality of signal waveforms created and displayed in step 103 are compared, and in step 104, signal feature points resulting from the end of the defect opening surface are extracted. The signal feature points are the signal amplitude maximum value 24, the signal amplitude minimum value 25, the signal vanishing point 26, and the like described above with reference to FIG.

  Further, in step 105, position data of each signal feature point is extracted from the information, and in steps 106a and 107a, the signal vanishing point is extracted and the signal vanishing length is calculated. In steps 106b and 107b, the maximum signal amplitude value, the minimum signal amplitude value, and the -12 dB drop point are extracted, and the -12 dB instruction length is calculated. As the position data, a signal obtained from the position encoder 19 of the flaw detection apparatus shown in FIG. 4 is used.

  The above length measurement will be described in more detail. When the length is -12 dB, the maximum signal amplitude value 24a (or 24b) and the minimum signal amplitude value 25a (or 25b) are shown at both ends of the defect in FIG. When the signal amplitude difference is set to 1, a point that is 3/4 down from the maximum signal amplitude value 24a (or 24b) is set as a -12 dB drop point 27a (or 27b), and is calculated from the coordinate difference between the left end and the right end of the defect. (107). Similarly, the signal disappearance length is calculated from the coordinate difference between the signal disappearance points 26a and 26b. As these signal evaluation methods, generally well-known methods can be used, and the evaluator can easily extract the signal peak point and measure the defect length.

  In the above description, the case where the defect is parallel or perpendicular is described, but in general, the defect is in an arbitrary direction and is not necessarily formed parallel or perpendicular to the scanning direction of the probe (X and Y axis directions). Not only.

  FIG. 9 shows the relationship of the detection coil that responds when flaw detection is performed on the defect 18 in the oblique direction with respect to the probe, and FIG. 10 shows the flaw detection result. In the case of the flaw detection shown in FIG. 5, since the defects are parallel or orthogonal, even if the multi-probe 1 is used, any one probe represents the characteristic signal. Represents a feature signal. For this reason, since it is necessary to evaluate the signals of a plurality of element probes, each probe or its signal is called a channel.

  FIG. 9 shows the detection coil 9 which generates a signal in response to the multi-probe 1 passing through the oblique defect 18 in a bold notation. The channels ch1 to ch6 in the V detection mode and the H detection are shown. The mode channels ch7 to ch18 correspond to this.

  FIG. 10 is a waveform diagram in which the signal waveform obtained from each channel at this time is processed by the computer of FIG. 4 and displayed on the display unit 7. The horizontal axis of each waveform diagram indicates the position signal detected by the position encoder 19. The vertical axis indicates the amplitude value of the waveform. In particular, the present embodiment is characterized in that signals of a plurality of channels responding to defects are superimposed and described.

  According to the present invention, as a channel signal in response to the region 1 caused by the end 18a of the oblique defect 18, the V detection mode waveforms ch1, ch2, ch3 and the H detection mode waveform ch7 are shown on the left side of FIG. ch8, ch9, ch10, ch11, and ch12 are displayed. Further, as a channel signal in response to the region 2 caused by the end 18b of the oblique defect 18, waveforms V4, ch5, ch6 in the V detection mode and waveforms ch13, ch14, ch15, ch16 in the H detection mode are shown on the right side of FIG. , Ch17, and ch18 are displayed.

  In the example of the waveform diagram of FIG. 10, as a channel responding to the region 1 caused by the end 18a of the oblique defect 18, the V detection mode responds more significantly than the response in the H detection mode, and It can be seen that, among the signals in the V detection mode, ch1 having the detection coil at the position closest to the end responds quickly. Further, even in the channel responding to the region caused by the end 18b of the oblique defect 18, the V detection mode responds more significantly than the response in the H detection mode, and among the signals in the V detection mode. It can be seen that ch6 having the detection coil at the position closest to the end responds last.

  In this way, when all the waveforms are displayed in an overlapping manner, it becomes clear whether the V detection mode should be selected or the H detection mode should be selected. By the way, the point of view when selecting the detection mode is to make the detection mode that changes the signal more prominently. The larger the signal waveform, the more accurately the position can be specified and the more effective for the defect length measurement. is there.

  In addition, the planar display of the signal intensity representing the defect is also displayed on the display unit 7, but at the position where the signal intensity starts to increase (end 18a), the channel ch1 that caused the signal change first is the end 18a. Since it is close, it is understood that this signal may be used for detecting the position of the end portion 18a. For the same reason, at the position where the signal intensity starts to disappear (end portion 18b), the channel ch6 that finally caused the signal change is closest to the end portion 18b, so this signal may be used for position detection.

  FIG. 11 shows only two waveforms selected in this way (waveforms from ch1 and ch6) on the display unit 7. In ch1, a defect is detected from the information on the rising point with respect to the change in position on the horizontal axis. The position of the end 18a is evaluated. In ch6, the position of the defect end 18b is evaluated from the information on the falling part. As a result, if the position difference between the 12 dB drop points is derived, the length of the defect can be obtained.

  FIG. 1 is a flowchart showing a method for measuring the length of an oblique defect displayed as described above, and part of the concept is the same as the concept of FIG. 8 described above.

  First, in step 201, a planar display of signal intensity in each detection mode (H, V1, V2) is created. Next, in step 202, the signal waveforms (ch1 to ch18) of the respective detection coils 9 representing the signal change are displayed on the display unit 7 in FIG. The display screen displayed at this time is as shown in FIG. 10. The response signal in the region 1 of the end 18a is divided into the V detection mode and the H detection mode, the horizontal axis is the position, the vertical axis is the waveform length. Overlaid on the screen. Similarly, the response signal in the region 2 of the end 18b is divided and displayed on the screen with the horizontal axis as the position and the vertical axis as the waveform length, divided into the V detection mode and the H detection mode.

  Looking at this superimposed display, when the probe moves in the direction of FIG. 9, the detection coils on the left side of FIG. 10 (ch1, ch2, ch3, ch7 to ch12) respond first, and then the detection coils on the right side (ch4, It can be seen that ch5, ch6, ch13 to ch18) responded. Since the horizontal axis is the probe position, considering the probe scanning direction, the horizontal axis can also be called the time axis. Therefore, it can be said that the waveform group on the left side of FIG. 10 is the waveform group that responded first when approaching the defect, and the waveform group on the right side is the waveform group that responded last when leaving the defect.

  Next, in step 203, the responsiveness for each end is evaluated. Since the processing in this is basically the same, step 203 (a) on the end 18a side will be described as an example here. In step 204, one of the end portions is selected first, and only all signal waveforms whose signals have changed at the end portion selected in 205 are extracted and displayed. The display at this time may be considered to be that the screen shown in FIG. 10 was displayed but only the screen on the left side of FIG. 10 was extracted and displayed in an enlarged manner.

  Next, at step 206, either the V detection mode screen displayed at the top or the H detection mode displayed at the bottom is selected from the screen on the left side of FIG. In the example of this figure, the signal in the V detection mode in which the signal change is remarkable is selected. Similarly, a signal in the V detection mode is also selected on the step 206 (b) side.

  In step 207, one channel signal used for defect inspection is selected from the extracted three signals in the V detection mode. The concept of selection at this time is the signal ch1 of the channel that first approaches the defect and starts to respond. Similarly, in the selection at 207 (b), the signal ch6 of the channel that responded last when moving away from the defect is used. In step 208, the maximum amplitude point, minimum amplitude point, and signal vanishing point are calculated for the signal of the channel ch1 selected as shown in FIG.

  When the series of processing in step 203 is completed for each end, the screen display is then switched to FIG. 11 and the waveforms of channel ch1 and channel ch6 are displayed together with feature points (maximum amplitude point, minimum amplitude point, and signal vanishing point). Is done. Then, a 12 dB drop position is calculated at the rising point for ch1, which is the first response signal when approaching a defect. Similarly, a 12 dB drop position is calculated at the falling point for ch6 which is the last response signal when leaving the defect. Furthermore, the position of the signal vanishing point is calculated.

  In step 209, it is selected whether the defect size is calculated at the 12 dB drop position or the signal vanishing point position. In step 210, the defect length is determined as the difference between the coordinate positions of ch1 and ch6. To do.

  As described in detail above, according to the method and apparatus of the present invention, a wide range of flaw detection can be performed by a single probe scan, and an optimum combination can be extracted from a plurality of response signals. The length of the defect can be easily measured.

  The present invention can be applied to, for example, eddy current flaw inspection in a reactor internal structure of a nuclear power plant, and can perform length measurement on a wide surface in a short time without depending on the direction of defects.

DESCRIPTION OF SYMBOLS 1 Multi-probe 2 Element probe 3 Computer 4 including a memory 4 Memory 5 Calculation part 6 Computer input part 7 Computer display part 8 Excitation coil 9 Detection coil (a, b, c)
DESCRIPTION OF SYMBOLS 11 Eddy current flaw detector 12 Defect in the direction parallel to the probe 13 Defect in the direction orthogonal to the probe 14 Flaw detection mechanical frame 15 Traveling vehicle 16 Traveling rail 17 Traveling wheel 18 Drive motor 19 Position encoder 20 Elastic material
21 Buffer material 22 Flaw detection mechanical device 23 Flat plate test body 24 Signal amplitude maximum value (a, b)
25 Minimum signal amplitude (a, b)
26 Signal vanishing point (a, b)
27 -12 dB drop point (a, b)

Claims (5)

An eddy current flaw detection method in which an eddy current is generated in a test body by an excitation coil and a change in eddy current is detected by a plurality of detection coils,
A probe mounted with the excitation coil and the plurality of detection coils is scanned on a test body to detect eddy current change signals from the plurality of detection coils together with the position signals of the probes ,
With the position signal of the probe as a reference, the plurality of detection coil signals at the position are superimposed,
Select the detection coil signal that first detected the start of change when approaching the defect from the plurality of superimposed detection coil signals and set the position at this time as the change start point position ,
And selects the signal from the detection coil which detects the last change ends when away from the defect from the superimposed plurality of detection coils signals the position at this time a change ending point position,
An eddy current flaw detection method, comprising: determining a defect length of a specimen as a difference between the change start point position and the change end point position. An exciting coil and element probe with a plurality of second detection coils in the direction in which the first detection coil intersecting the arranged direction with a probe in which a plurality of sets mounted, resulting eddy currents in the test specimen by the exciting coil And an eddy current flaw detection method for detecting a change in eddy current with the plurality of detection coils,
The probe mounted with the plurality of excitation coils and the plurality of detection coils is scanned on a test body to detect eddy current change signals from the plurality of detection coils together with the position signals of the probes ,
With the position signal of the probe as a reference, the plurality of detection coil signals at the position are superimposed,
Select the detection coil signal that first detected the start of change when approaching the defect from among the plurality of detection coil signals superimposed, the position at this time as the change start point position ,
And selects the signal from the detection coil which detects the last change ends when away from the defect from the superimposed plurality of detection coils signals the position at this time a change ending point position,
An eddy current flaw detection method, comprising: determining a defect length of a specimen as a difference between the change start point position and the change end point position. The eddy current flaw detection method according to claim 2 ,
When both the first detection coil group composed of the plurality of first detection coils and the second detection coil group composed of the plurality of second detection coils detect a change in eddy current, Select the detection coil group signal,
An eddy current flaw detection method comprising: selecting one signal from a plurality of detection coil signals with respect to the signal from the selected detection coil group. The eddy current flaw detection method according to claim 2 ,
When both the first detection coil group composed of the plurality of first detection coils and the second detection coil group composed of the plurality of second detection coils detect a change in eddy current, An eddy current flaw detection method characterized in that, when selecting a signal of a detection coil group, the selection is performed individually when approaching a defect and when moving away from the defect. A flaw detection mechanical device that scans a test body mounted with a plurality of excitation coils and a plurality of detection coils on the test body, causes eddy currents to be generated in the test body by the excitation coils, and detects changes in eddy currents using the plurality of detection coils. An eddy current flaw detector that obtains signals from the plurality of detection coils and a flaw detection position signal on the specimen, a computer that executes arithmetic processing using signals from the eddy current flaw detector, and an output from the computer An eddy current flaw detector comprising a display unit for displaying
In the display unit, the plurality of detection coil signals when approaching the defect and the plurality of detection coil signals when moving away from the defect take the flaw detection position signal on the specimen on the horizontal axis and the amplitude on the vertical axis. It ’s overlaid as a value ,
From the plurality of detection coil signals when approaching a defect, the signal from the detection coil that first changed, and from the detection coil that last changed among the plurality of detection coil signals when leaving the defect The eddy current flaw detection apparatus is selected in which only the signal is selected, the flaw detection position signal on the specimen is taken on the horizontal axis, and the amplitude value is displayed on the vertical axis .

Images