Figure-of-eight trajectory detection program and eddy current flaw detection device and method

Abstract

To provide means that can quantitatively detect a "figure-of-eight trajectory" of the Lissajous curve, and can thereby determine the presence or absence of a flaw without individual differences.SOLUTION: A figure-of-eight trajectory detection program receives an X signal and a Y signal each periodically changing centered on the same definite point, and detects a figure-of-eight trajectory drawn by the Lissajous curve R obtained by overlapping the X signal and Y signal each other. An eddy current flaw detection device comprises a probe, an eddy current flaw detector, and a figure-of-eight trajectory detection device. An eddy current flaw detection method has a figure-of-eight detection step of detecting the figure-of-eight trajectory drawn by the Lissajous curve R obtained by overlapping the X signal and Y signal each other. In the figure-of-eight detection step, the figure-of-eight trajectory detection device calculates in real time a momentary phase difference Z that is the difference between the phase angle of the X signal and the phase angle of the Y signal, and when the difference between the maximum value and the minimum value of the momentary phase difference Z exceeds a predetermined threshold, detects the figure-of-eight trajectory drawn by the Lissajous curve R.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a figure-of-eight locus detection program and an eddy current flaw detection device and method for detecting a figure-of-eight locus of a Lissajous curve. [Background technology]

[0002] "Eddy current testing" can be applied to conductive materials such as steel, non-ferrous metals, and graphite, and is a test that detects cracks by utilizing the property that eddy currents induced in the material change due to cracks. "Eddy current testing equipment" is a device that performs eddy current testing by placing a coil on the material and passing an AC power through this coil.

[0003] In eddy current testing (ET), the input coil of the probe generates an eddy current in the test piece, and the output signal of the detection coil due to this eddy current is detected. The eddy current flaw detector also outputs ET signals (X and Y signals) that change periodically from the output signal, centered on the same fixed point. Furthermore, a Lissajous curve is drawn by superimposing the X and Y signals, and the presence or absence of flaws is judged based on the shape of this trajectory.

[0004] In eddy current testing, it is known that when the probe passes through a flaw, the Lissajous curve forms a "closed figure-eight curve." Hereinafter, this "closed curve in the shape of a figure of eight" will be referred to as a "figure of eight locus." The "figure of eight locus" in a Lissajous curve is disclosed in, for example, Patent Documents 1 and 2.

[0005] Paragraph 0029 of Patent Document 1 states that "When detection coil 19a reaches the position of vertical scratch 32, a change in eddy current due to vertical scratch 32 is detected, and a Lissajous waveform in the shape of an eight appears on the monitor."

[0006] Paragraph 0019 of Patent Document 2 states that "Since the ECT signal basically oscillates from the origin and returns to the origin, the waveform of the Lissajous 23 will be circular or have a figure-8 shape."

[0007] Moreover, an invention related to the present invention is disclosed in Patent Document 3. [Prior art documents] [Patent documents]

[0008] [Patent Document 1] JP 2008-292280 A [Patent Document 2] JP 2002-40000 A [Patent Document 3] Patent No. 5971425 Summary of the Invention [Problem to be solved by the invention]

[0009] As described above, in eddy current testing, when the probe passes over a flaw, the Lissajous curve traces a "figure-of-eight locus," and the presence or absence of a flaw can be determined from the "figure-of-eight locus."

[0010] However, such conventional flaw detection means have the following problems. (1) There are individual differences in the judgment of small flaws and flaws in parts with complex shapes. (2) Although automatic judgment methods based on pattern recognition are known, the judgment criteria are treated as a black box and cannot be explained. (3) There is no way to quantitatively judge the “figure-eight trajectory,” so the objectivity of the judgment cannot be guaranteed. (4) Visual judgment of the “figure-of-eight locus” places a burden on the inspector because it requires the inspector to frequently move his / her viewpoint between the probe, the test piece, and the Lissajous curve.

[0011] The present invention has been devised to solve the above-mentioned problems. That is, an object of the present invention is to provide a figure-of-eight locus detection program and an eddy current flaw detection device and method that can quantitatively detect the "figure-of-eight locus" of a Lissajous curve and thereby determine the presence or absence of flaws without individual differences. [Means for solving the problem]

[0012] According to the present invention, there is provided a figure-of-eight locus detection program for receiving an X signal and a Y signal that each change periodically around a fixed point, and detecting a figure-of-eight locus drawn by a Lissajous curve obtained by superimposing the X signal and the Y signal, comprising: On the computer, an instantaneous phase difference calculation means for calculating an instantaneous phase difference, which is a difference between a phase angle of the X signal and a phase angle of the Y signal, in real time; and a detection signal output means for outputting a detection signal of the figure-of-eight locus when the difference between the maximum and minimum values ​​of the instantaneous phase difference exceeds a predetermined threshold.

[0013] According to the present invention, there is also provided a probe for generating an eddy current in a test piece by an input coil and detecting an output signal of a detection coil due to the eddy current; an eddy current flaw detector that outputs an X signal and a Y signal that each periodically changes around a same fixed point from the output signal; a figure-of-eight locus detection device that detects a figure-of-eight locus drawn by a Lissajous curve obtained by superimposing the X signal and the Y signal; The figure-of-eight locus detection device comprises: Calculating an instantaneous phase difference, which is a difference between a phase angle of the X signal and a phase angle of the Y signal, in real time from the X signal and the Y signal; There is provided an eddy current flaw detector that outputs a detection signal of the figure-of-eight locus when the difference between the maximum value and the minimum value of the instantaneous phase difference exceeds a predetermined threshold value.

[0014] Further, according to the present invention, a first step of generating an eddy current in a test piece by an input coil of the probe and detecting an output signal of a detection coil due to the eddy current; a second step of outputting an X signal and a Y signal, each of which periodically changes around a same fixed point, from the output signal; and a third step of detecting a figure-of-eight locus formed by a Lissajous curve obtained by superimposing the X signal and the Y signal, In the third step, Calculating an instantaneous phase difference, which is a difference between a phase angle of the X signal and a phase angle of the Y signal, in real time from the X signal and the Y signal; There is provided an eddy current flaw detection method, which outputs a detection signal of the figure-of-eight locus when the difference between the maximum value and the minimum value of the instantaneous phase difference exceeds a predetermined threshold value. Effect of the Invention

[0015] It has been known that in eddy current testing, the Lissajous curve traces a "figure-of-eight" when the probe passes through a flaw. Furthermore, when the Lissajous curve traces a "figure-of-eight locus," the frequencies of the X and Y signals, which each change periodically about the same fixed point, have a relationship of approximately 2:1.

[0016] Furthermore, the inventors of the present invention have discovered a new finding that when the frequencies of the X signal and the Y signal have an approximately 2:1 relationship, the instantaneous phase difference, which is the difference between the phase angle of the X signal and the phase angle of the Y signal, becomes large at a certain point in time. The present invention is based on this new finding.

[0017] According to the present invention, an instantaneous phase difference, which is the difference between the phase angle of the X signal and the phase angle of the Y signal, is calculated in real time from the X signal and the Y signal, and when the difference between the maximum and minimum values ​​of the instantaneous phase difference exceeds a predetermined threshold, a figure-of-eight detection signal is output. Therefore, according to the present invention, the "figure-of-eight locus" of the Lissajous curve can be quantitatively detected, and the presence or absence of a flaw can be determined without individual differences. [Brief description of the drawings]

[0018] [Figure 1] FIG. 1 is a diagram illustrating the principle of the present invention. [Diagram 2] FIG. 1 is an explanatory diagram of the Synchro-PRIMO method. [Diagram 3] FIG. 13 is an explanatory diagram of a computer screen on which a post-analysis application is installed. [Figure 4]1 is an overall configuration diagram of an eddy current flaw detector according to the present invention; [Diagram 5] FIG. 2 is a circuit diagram of an eddy current flaw detector. [Figure 6] FIG. 13 is a diagram showing a specific example of an inspection image. [Figure 7] 1 is an overall flow diagram of an eddy current inspection method according to the present invention. [Figure 8] These are specific examples of ET signals, where (A) is an example with a flaw and (B) is an example without a flaw. [Figure 9] 9 is a specific example of an ET signal after removing the DC component corresponding to FIGS. 8(A) and (B). [Figure 10] 8(A) and (B) show Lissajous curves corresponding to FIG. [Figure 11] FIG. 10 is an enlarged view of FIG. 10(A)(B). [Figure 12] 8(A) and (B) show an example (calculation example) of extracting instantaneous amplitude using the SPRIMO method. [Figure 13] 8(A) and 8(B) show an example (calculation example) of extracting instantaneous frequency using the SPRIMO method. [Figure 14] 9 is an example (calculation example) of extracting an instantaneous phase difference using the SPRIMO method corresponding to FIGS. 8(A) and (B). [Figure 15] 13A and 13B are diagrams showing an ET signal when there is a flaw, an ET signal after removing a DC component, and an instantaneous phase difference. [Figure 16] FIG. 13 is a diagram showing an ET signal without a flaw, an ET signal after removing a DC component, and an instantaneous phase difference. [Figure 17] FIG. 13 is a diagram showing the difference in work flow during flaw detection. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0019] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. In the drawings, common parts are given the same reference numerals and duplicated explanations will be omitted.

[0020] The present invention is directed to a Lissajous curve obtained by superimposing an X signal and a Y signal that each vary periodically around the same fixed point. As described above, when the Lissajous curve draws a "figure-of-eight locus," the frequencies of the X and Y signals have a relationship of approximately 2:1.

[0021] (Principle of the Invention) FIG. 1 is a diagram illustrating the principle of the present invention. In the following, the amplitude and frequency of the X signal are defined as the first amplitude A and the first frequency ω X , the amplitude and frequency of the Y signal are the second amplitude B and the second frequency ω Y , the phase difference between them is φ. In this case, the X and Y signals of the ET signal at a certain time t can be expressed by the following equations. X(t) = A sin(ω X t+φ) (1) Y(t) = B sin(ω Y t) (2)

[0022] Figure 1(A) shows the phase difference φ=0, ω X :ω Y This shows the case where the ratio is 2:1. In this diagram, the horizontal axis represents time t, and the vertical axis represents the values ​​of the X and Y signals (voltage [V] in this example). In this figure, ω X :ω Y Since the relationship is 2:1, while the Y signal goes from 0 to 360° in one cycle, the X signal goes from 0 to 360° twice (i.e., two cycles). In this case, as described above, the Lissajous curve traces a "figure eight" shape.

[0023] FIG. 1B is a diagram showing the relationship between time t and phase angle [deg] in FIG. 1A, where the horizontal axis represents time t and the vertical axis represents the phase angle between the X signal and the Y signal. Since the phase angle repeats between 0 and 360° during one period, the X signal rises from 0° to 360° in the first period, and then rises again from 0° to 360° in the next period. In other words, the X signal reaches a maximum phase angle of 360° in the first period, and reaches a minimum phase angle of 0° at the next moment (the start of the next period). On the other hand, when the X signal reaches the maximum phase angle of 360° in the first period, the phase angle of the Y signal is 180°, and the X signal continues to rise to the maximum phase angle of 360°.

[0024] In FIG. 1(B), the phase angle of the X signal and the phase angle of the Y signal are the first phase angle θ X and the second phase angle θ Y Then, the first phase angle θ at a certain time t X and the second phase angle θ Y can be expressed by the following formula: θ X =ω X t1 (3) θ Y =ω Y t2 (4) Here, t1 and t2 are the times elapsed from the respective minimum phase angles of 0°.

[0025] FIG. 1C is a diagram showing the relationship between time t and instantaneous phase difference Z in FIG. 1B, where the horizontal axis is time t and the vertical axis is instantaneous phase difference Z. Note that the "instantaneous phase difference" is the difference between the phase angle of the X signal and the phase angle of the Y signal, that is, the first phase angle θ X and the second phase angle θ Y This means the difference between From this figure, it can be seen that in this example, during one period of the Y signal, which has a relatively long period (in this example, at the point of 0.5 period of the Y signal), the instantaneous phase difference Z becomes large, that is, reaches a maximum or minimum value. Therefore, by detecting that the difference between the maximum and minimum values ​​of the instantaneous phase difference Z exceeds a predetermined threshold, it is possible to detect a figure-of-eight locus.

[0026] In the above example, the values ​​of the X and Y signals at time t are represented as sine waves, but the same results can be obtained even if they are represented as cosine waves. In addition, although the phase difference φ=0, even if the phase difference φ≠0, the instantaneous phase difference Z becomes large during one period of the Y signal, which has a relatively long period. Also, ω X :ω Y We have shown an example of =2:1, but ω X :ω Y= 1:2. Furthermore, this ratio is not strict and may vary as long as the Lissajous curve traces a "figure eight".

[0027] (Synchro-PRIMO method) Figure 2 is an explanatory diagram of the Synchro-PRIMO method (hereinafter, SPRIMO method). The SPRIMO method is a calculation method that calculates the basic parameters of a sine wave composed of discrete data using simple mathematical expressions. If the frequency ω, amplitude A, and phase difference φ are known, a time series waveform can be easily synthesized as follows: X(t) = A·sin(ωt+φ). If this is considered a "forward problem," the mathematics of the SPRIMO method is an equation for finding the frequency ω, amplitude A, and phase difference φ from the time series waveform obtained by measurement (= solving the "inverse problem"). The mathematics of the SPRIMO method are disclosed in, for example, Patent Document 3.

[0028] The X and Y signals in Figure 1 are the values ​​at time t of the time series waveform obtained by measurement. Therefore, the frequency ω X ,ω Y , the amplitudes A and B, and the phase difference φ can be obtained. In addition, the first phase angle θ X and the second phase angle θ Y The change in the difference (i.e., the instantaneous phase difference Z) can be calculated in real time. Therefore, by using the SPRIMO method to detect whether the difference between the maximum and minimum values ​​of the instantaneous phase difference Z exceeds a predetermined threshold, a figure-of-eight locus can be detected.

[0029] (Figure 8 trajectory detection program) The figure-of-eight locus detection program of the present invention is a program that receives X and Y signals that each change periodically around the same fixed point, and detects a figure-of-eight locus drawn by a Lissajous curve obtained by superimposing the X and Y signals.

[0030] That is, the figure-of-eight locus detection program is installed in a storage medium of a computer and used, and is a program for causing the computer to execute the instantaneous phase difference calculation means and the detection signal output means.

[0031] The instantaneous phase difference calculation means calculates the phase angle of the X signal (first phase angle θ X ) and the phase angle of the Y signal (second phase angle θ Y ) and calculates the instantaneous phase difference Z in real time. Also, it is preferable that the instantaneous phase difference calculation means calculates in real time the change in the instantaneous phase difference Z during one period of the X signal or the Y signal, whichever has the longer period.

[0032] The detection signal output means outputs a figure-of-eight detection signal when the difference between the maximum and minimum values ​​of the instantaneous phase difference Z exceeds a predetermined threshold value. The predetermined threshold value is preferably set in advance and input into the computer.

[0033] The above-mentioned figure-of-eight trajectory detection program can be used as a post-analysis application (hereinafter referred to as the post-analysis app) to read previously acquired flaw detection data (i.e., X and Y signals), display them on a computer screen, and perform flaw determination.

[0034] FIG. 3 is an explanatory diagram of a computer screen on which a post-analysis application is installed. As shown in Figure 3(A), when a file of previously acquired flaw detection data is loaded, the loaded flaw signals (i.e., X and Y signals) and Lissajous curves are displayed on the screen. Furthermore, as shown in Figure 3(B), when the focus cursor on the screen is aligned with the flaw signal, the figure-eight locus is detected through internal processing, and the Lissajous locus within the cursor turns red. In addition, as shown in FIG. 3(C), the strength of preprocessing and the threshold settings for detecting the figure-of-eight locus can be freely changed.

[0035] (Eddy current flaw detection device) FIG. 4 is a diagram showing the overall configuration of an eddy current flaw detector 100 according to the present invention. In this figure, an eddy current flaw detector 100 includes a probe 10 , an eddy current flaw detector 20 , and a figure-of-eight locus detector 30 .

[0036] The probe 10 has an input coil 10a and a detection coil 10b. The input coil 10a generates an eddy current in the test piece and the detection coil 10b detects an output signal due to this eddy current.

[0037] The probe 10 is preferably a differential probe since non-differential probes will not provide a differential indication of flaws and will not produce a figure-of-eight Lissajous curve. Furthermore, by differentiating (subtracting) the output using digital signal processing, it is possible to create a pseudo-differential indication even with a single probe. Therefore, the probe 10 is not limited to a differential probe, and may be of another type as long as the eddy current flaw detector 20 can output an X signal and a Y signal.

[0038] The eddy current flaw detector 20 outputs ET signals (X signal and Y signal) that vary periodically from the output signal of the probe 10, each of which has the same fixed point as the center.

[0039] FIG. 5 is a circuit configuration diagram of the eddy current flaw detector 20. As shown in FIG. As shown in this figure, the eddy current flaw detector 20 is composed of an oscillator that supplies an AC current to the coil of the probe 10, a bridge for measuring the impedance of the coil, an amplifier, a synchronous detector, a phase shifter, and the like.

[0040] The signal processing in the eddy current flaw detector 20 is as follows. An AC signal (reference signal) generated by an oscillator is supplied to the bridge via a power amplifier. The bridge is adjusted so that output is minimal in the healthy part of the test specimen. When the probe passes over a flaw, the impedance of the coil changes with the change in eddy current, and a voltage is output that is amplitude-modulated by the flaw signal using the reference signal as the carrier wave. On the other hand, the phase shifter outputs reference signal 1, which is phase-shifted from the reference signal in the range of 0° to 360°, and the 90° phase shifter outputs reference signal 2, which is reference signal 1 further phase-shifted by 90°. The synchronous detector X detects (demodulates) the signal from the amplifier while synchronizing with the reference signal 1, and extracts the flaw signal X. Similarly, the synchronous detector Y detects (demodulates) the signal while synchronizing with the reference signal 2, and extracts the flaw signal Y. The filter removes noise with a different frequency component from the flaw signal. The flaw signals X and Y from which the noise has been removed are the ET signals (X signal and Y signal).

[0041] As mentioned above, the ET signal (X signal and Y signal) is based on the impedance change of the coil. Therefore, the Lissajous curve can be said to be the projection of the locus of the impedance change of the coil onto the XY plane, which is likened to an impedance plane.

[0042] In FIG. 4, the figure-of-eight locus detection device 30 detects a figure-of-eight locus drawn by a Lissajous curve obtained by superimposing the X signal and the Y signal output from the eddy current flaw detector 20.

[0043] In the example of FIG. 4, the probe 10 is electrically connected to an eddy current flaw detector 20 via a signal cable 12, and an output signal of the detection coil 10b is input to the eddy current flaw detector 20. In the example of FIG. Moreover, the figure-of-eight locus detection device 30 in this example is a control tablet 32, which has a built-in computer (PC) and receives X and Y signals from the eddy current flaw detector 20 via wire or wirelessly.

[0044] The computer of the figure-of-eight locus detection device 30 preferably has the above-mentioned figure-of-eight locus detection program installed. The figure-of-eight locus detection device 30 executes the following steps (1) and (2). (1) From the X and Y signals, the phase angle of the X signal (first phase angle θ X ) and the phase angle of the Y signal (second phase angle θ Y ) and calculates the instantaneous phase difference Z in real time. (2) When the difference between the maximum and minimum values ​​of the instantaneous phase difference Z exceeds a predetermined threshold, a figure-of-eight detection signal is output.

[0045] The control tablet 32 ​​is, for example, a tablet PC for control and display of inspection results, and displays an inspection image 34 based on the inspection results on the display screen. Moreover, the figure-of-eight locus detection device 30 (control tablet 32) has a detection sound output device 35 that outputs a detection sound in response to a detection signal.

[0046] FIG. 6 is a diagram showing a specific example of the inspection image 34. As shown in FIG. In this example, the figure-of-eight locus detection device 30 has a Lissajous curve display section 36 and a number of history display sections 37 on the display screen. The Lissajous curve display unit 36 ​​displays the Lissajous curve R in real time. A plurality of (six in this example) history display units 37 display the histories of a plurality of Lissajous curves R, respectively. In this figure, reference numeral 38 denotes an X signal display section, and reference numeral 39 denotes a Y signal display section.

[0047] (Eddy current testing method) FIG. 7 is an overall flow diagram of the eddy current flaw detection method according to the present invention. In this figure, the eddy current flaw detection method includes steps S1 to S6.

[0048] In step S1, the ET signal of the eddy current flaw detector 20 is acquired as digital data. At this time, A / D conversion is performed as necessary. Step S1 consists of a detection signal output step SA and an ET signal output step SB. In the detection signal output step SA, an eddy current is generated in the test piece by the input coil 10a of the probe 10, and an output signal of the detection coil 10b due to this eddy current is detected. In the ET signal output step SB, an X signal and a Y signal, each of which changes periodically around the same fixed point, are output from the output signal of the detection coil 10b.

[0049] FIG. 8 shows specific examples of ET signals, where (A) is an example with a flaw and (B) is an example without a flaw.

[0050] The acquired digital data is sampled at equal intervals. For example, the ET signal is preferably sampled at 1 KHz. Moreover, the sampling frequency is set to be sufficiently high relative to the frequency of the components of the analysis target (ET signal), which may be, for example, about 20 Hz. In addition, it is recommended to set the bit resolution during A / D conversion to a sufficiently large value, for example 12 bits.

[0051] In step S2 of FIG. 7, the DC component contained in the ET signal is removed by a high-pass filter or the like. This is because the SPRIMO method described above does not support analysis of signals that contain DC components. Note that if the DC components are removed in step S1 or if zero balance is achieved by calibration, step S2 can be omitted.

[0052] 9(A) and (B) are specific examples of ET signals after removing the DC components corresponding to FIGS. 8(A) and (B).

[0053] In the present invention, the Lissajous curve R is displayed in real time on the Lissajous curve display unit 36 ​​in FIG. 10(A) and (B) show Lissajous curves R corresponding to FIGS. 8(A) and (B), and FIGS. 11(A) and (B) are enlarged views of FIGS. 10(A) and (B). 10 and 11, it can be seen that when there is a flaw, the Lissajous curve R shows a figure-of-eight locus, but when there is no flaw, the Lissajous curve R does not show a figure-of-eight locus.

[0054] 7. The present invention also includes a figure-of-eight locus detection step SC for detecting a figure-of-eight locus drawn by a Lissajous curve R obtained by superimposing an X signal and the Y signal. The figure-of-eight locus detection step SC includes steps S3 to S5 in FIG.

[0055] In step S3, the instantaneous phase difference Z is calculated for the acquired ET signal by the SPRIMO method. That is, from the X and Y signals, the phase angle of the X signal (first phase angle θ X ) and the phase angle of the Y signal (second phase angle θ Y ) and calculates the instantaneous phase difference Z in real time. This calculation is valid if there are five or more sampling data points. Furthermore, by increasing the window width, the calculation accuracy is improved due to the averaging effect. The "window width" refers to the calculation interval for each calculation of the instantaneous phase difference along the time axis. Another example of the use of the term "window width" is the "window width of the time window in the short-time Fourier transform (STFT)." A graph of the instantaneous phase difference is obtained by plotting the instantaneous phase difference while shifting the window (window width). Note that if the window width is set to be greater than five points, the number of combinations of five points within the window increases, making it possible to calculate the instantaneous phase difference multiple times, and an averaging effect (noise resistance) can be obtained. Therefore, the window width should be at least five samples, and at most about 1 / 4 of the rough period of the ET signal.

[0056] Figures 12(A) and (B) are examples of extraction (calculation examples) of instantaneous amplitudes using the SPRIMO method corresponding to Figures 8(A) and (B). Note that instantaneous amplitudes refer to amplitudes at each time point t. Figures 13(A) and (B) are examples of extraction (calculation examples) of instantaneous frequency by the SPRIMO method corresponding to Figures 8(A) and (B). Note that instantaneous frequency means the frequency at each time point t. Furthermore, calculation of the instantaneous amplitude and instantaneous frequency is not essential and may be omitted.

[0057] In step S4, noise components of the features are removed based on the amplitude values ​​of the X and Y signals. This is to prevent miscalculation and overdetection by not performing calculations outside the section where the ET signal is generated.

[0058] In step S5, the presence or absence of a figure-of-eight locus is determined. For example, when the difference between the maximum and minimum values ​​of the instantaneous phase difference Z exceeds a predetermined threshold, a figure-of-eight detection signal is output. 14(A) and (B) are examples (calculation examples) of extracting the instantaneous phase difference Z by the SPRIMO method corresponding to FIGS. 8(A) and (B). In FIG. 14(A), the instantaneous phase difference Z is larger than in FIG. 14(B), and it can be seen that the presence or absence of a figure-8 locus can be detected by comparing the difference between the maximum and minimum values ​​of the instantaneous phase difference Z with a predetermined threshold value.

[0059] If there is no figure-of-eight locus (NO) in step S5, steps 1 to 5 are repeated.

[0060] If a figure-of-eight locus is present (YES) in step S5, a detection sound (beep sound) is output from the detection sound output device 35 in response to the detection signal in step S6. In addition, a plurality of Lissajous curves R at the time of outputting a detection signal with a figure-of-eight locus are displayed on the Lissajous curve display unit 36 ​​as history.

[0061] 15 and 16 are representative diagrams of examples with and without flaws. FIG. 15 shows the case with a flaw, and FIG. 16 shows the case without a flaw. (A) Original signal (ET signal), (B) Pre-processed (ET signal after removing DC component), and (C) Instantaneous phase difference. 15 and 16, it can be seen that the instantaneous phase difference Z in FIG. 15 is larger than that in FIG. 16, and the presence or absence of a figure-8 locus can be detected by comparing the difference between the maximum and minimum values ​​of the instantaneous phase difference Z with a predetermined threshold value.

[0062] 17 is a diagram showing the difference in work flows during flaw detection, in which (A) shows a conventional example and (B) shows the case of the present invention. In both (A) and (B), the “probe operation” is performed manually by the examiner.

[0063] In the conventional example of FIG. 17(A), the inspector visually checks the Lissajous figure (Lissajous curve) while operating the probe, and visually detects the "figure-of-eight locus." If the "figure-of-eight locus" is not found, the "probe operation" is repeated. Also, if the "figure-of-eight locus" is visually detected, the presence or absence of a flaw is judged. In the conventional method, visual judgment of the "figure-of-eight trajectory" requires frequent viewpoint shifting between the probe, the test piece, and the Lissajous curve, which places a burden on the inspector. In addition, individual differences occur in judgment of small flaws and flaws in parts with complex shapes.

[0064] In the present invention shown in Fig. 17(B), the "figure-of-eight locus" is automatically detected, and a sound indication (e.g., a beep) and its history are displayed on the screen. The inspector visually checks the Lissajous figure (Lissajous curve) based on the beep, for example, to determine whether there is a flaw. In the case of the present invention, there is no need to frequently move the viewpoint of the probe, the test object, and the Lissajous curve, and it is sufficient to check the Lissajous curve based on the beep sound. In addition, since the "figure-of-eight locus" is detected quantitatively, the judgment criteria are clear and individual differences can be eliminated.

[0065] As described above, the inventors of the present invention have found that when the frequencies of the X signal and the Y signal have a relationship of approximately 2:1, the phase angle of the X signal (first phase angle θ X ) and the phase angle of the Y signal (second phase angle θ Y We gained new knowledge that the instantaneous phase difference Z, which is the difference between the phase The present invention described above is based on this new finding.

[0066] According to the above-described embodiment of the present invention, the phase angle of the X signal (first phase angle θ X ) and the phase angle of the Y signal (second phase angle θ Y ) is calculated in real time, and when the difference between the maximum and minimum values ​​of the instantaneous phase difference Z exceeds a predetermined threshold, a figure-of-eight detection signal is output. Therefore, according to the present invention, the "figure-of-eight locus" of the Lissajous curve can be quantitatively detected, and the presence or absence of a flaw can be determined without individual differences.

[0067] It should be noted that the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit and scope of the present invention. [Explanation of symbols]

[0068] A 1st amplitude, B 2nd amplitude, R Lissajous curve, Z instantaneous phase difference, θ X First phase angle, θ Y second phase angle, ω X 1st frequency, ω Y Second frequency, φ phase difference, 10 probe, 10a input coil, 10b detection coil, 12 signal cable, 20 eddy current flaw detector, 30 figure-of-eight locus detection device, 32 control tablet, 34 inspection image, 35 detection sound output device, 36 Lissajous curve display unit, 37 history display unit, 38 X signal display unit, 39 Y signal display unit, 100 eddy current flaw detector

Claims

1. A figure-of-eight locus detection program for receiving an X signal and a Y signal that each change periodically around a same fixed point, and detecting a figure-of-eight locus drawn by a Lissajous curve obtained by superimposing the X signal and the Y signal, On the computer, an instantaneous phase difference calculation means for calculating an instantaneous phase difference, which is a difference between a phase angle of the X signal and a phase angle of the Y signal, in real time; a detection signal output means for outputting a detection signal of the figure-of-eight locus when a difference between a maximum value and a minimum value of the instantaneous phase difference exceeds a predetermined threshold value.

2. 2. The figure-of-eight locus detection program according to claim 1, wherein the instantaneous phase difference calculation means calculates in real time a change in the instantaneous phase difference Z during one period of the X signal or the Y signal which has a relatively longer period.

3. a probe for generating an eddy current in a test piece by an input coil and detecting an output signal of a detection coil due to the eddy current; an eddy current flaw detector that outputs an X signal and a Y signal that each periodically changes around a same fixed point from the output signal; a figure-of-eight locus detection device for detecting a figure-of-eight locus drawn by a Lissajous curve obtained by superimposing the X signal and the Y signal, The figure-of-eight locus detection device comprises: Calculating an instantaneous phase difference, which is a difference between a phase angle of the X signal and a phase angle of the Y signal, in real time from the X signal and the Y signal; An eddy current flaw detector that outputs a detection signal of the figure-of-eight locus when a difference between a maximum value and a minimum value of the instantaneous phase difference exceeds a predetermined threshold value.

4. The figure-of-eight locus detection device includes a Lissajous curve display unit that displays the Lissajous curve in real time; a detection sound output device that outputs a detection sound in response to the detection signal; The eddy current flaw detector according to claim 3 , further comprising a plurality of history display units that display the histories of a plurality of the Lissajous curves.

5. a detection signal output step of generating an eddy current in the test piece by an input coil of the probe and detecting an output signal of the detection coil due to the eddy current; an ET signal output step of outputting an X signal and a Y signal, each of which periodically changes around a same fixed point from the output signal; a figure-of-eight locus detection step of detecting a figure-of-eight locus drawn by a Lissajous curve obtained by superimposing the X signal and the Y signal; In the figure-of-eight locus detection step, Calculating an instantaneous phase difference, which is a difference between a phase angle of the X signal and a phase angle of the Y signal, in real time from the X signal and the Y signal; an eddy current flaw detection method, comprising: outputting a detection signal of the figure-of-eight locus when a difference between a maximum value and a minimum value of the instantaneous phase difference exceeds a predetermined threshold value.

6. The eddy current flaw detection method according to claim 5 , further comprising the step of outputting a detection sound in response to the detection signal.

7. The eddy current flaw detection method according to claim 5 , wherein the Lissajous curve is displayed in real time.

8. The eddy current flaw detection method according to claim 5 , further comprising displaying a plurality of the Lissajous curves at the time of output of the detection signal as a history.

Images