JPS5883253A - Eddy current flaw detection equipment - Google Patents

Abstract

PURPOSE:To remove noises resembling to flaw signals effectively and to improve SN ratios by passing the flaw signal which is detected with a probe and is obtained by detection and frequency sepn. through a signal processing circuit consisting of a multiplier which operates said signal by self squaring and a low-pass filter which integrates the output signal of said multiplier. CONSTITUTION:The flaw signal detected with a probe is subjected to signal processing by detection and frequency sepn. with an amplifying detector 5, an LPF10, and an HPF11. The flaw signal SB outputted from the HPF11 is relatively larger than the noise components and is squared by a multiplier 21 having a function for product operation of the two signals inputted; therefore the flaw signal is amplified, the noises are compressed and the relative level difference therefrom is marked. The flaw signal contg. the higher harmonic components outputted from said multiplier 21 is integrated by an LPF22 of a long time constant, whereby the sharp flaw signal SC is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a flaw detection device, and particularly to an eddy current flaw detection device.

Eddy current flaw detection, also called electromagnetic induction flaw detection, is a method of inspecting for defects such as scratches (flaws) by using the electromagnetic induction effect between a coil through which alternating current is passed and the metal being tested, and is a high-speed inspection method. It is something that can be done.

The first example of a conventionally known eddy current flaw detection device is
The figure shows a self-comparison coil method, which allows self-comparison coils 38 to 3D to probe the surface of the object 1.
A probe 3 formed of is provided close to the subject 1. Excitation coils 3A and 3B of this self-comparison type coil are connected to an excitation oscillator 4, and pickup coils 3C and 3D are differentially connected to an amplifier/detector 5.
It is connected to the.

When searching for a flaw on the test object 1, while moving the test object 1 and the glove 3 relatively, the excitation oscillator 4 activates the excitation coils 3A and 3B at a high frequency (approximately IKHz to IKHz).
MH2) to generate an eddy current on the surface of the subject 1. Since the eddy current is disturbed by the flaw 2 or the like, the magnetic flux distribution becomes unbalanced, and this disturbance in the magnetic flux distribution is detected as a needle signal by the pickup coils 3C and 3D. Since the needle signal is outputted from the pickup coils 3C and 3D as an amplitude/phase modulated wave of the excitation high frequency, the amplification/detector 5 vectorizes, detects, extracts, and amplifies one phase change in the amplitude. The needle signal output from the amplification/detector 5 is like the needle signal waveform shown in FIG. 1(B) including noise, and the pickup coil 3C,
Since the 3D is differentially connected, a needle signal SA having a sinusoidal one-cycle waveform is obtained for one flaw 2. In this way, defects such as scratches are detected by sequentially moving and probing the surface of the object 1 with the probe.

However, as mentioned above, the needle signal output from the amplifier/detector 5 contains power supply induced noise (frequency 50/60H2).
, noise related to material changes and surface roughness of the object to be examined, noise due to changes in the distance between the probe and the object, noise such as lift-off, and random noise. Therefore, since it is difficult to detect and analyze flaws directly from this signal, various means have been devised to extract the true needle signal, and they have been combined to increase the S/N ratio. It is being said.

First, it is generally attempted to eliminate noise generation factors in the needle signal detection process and signal processing process of the flaw detection apparatus as a whole. For example, by using a self-comparison type coil method, noise components such as Shirifuto 7 can be removed, fluctuations in magnetic permeability due to magnetic saturation can be eliminated, needle signals can be extracted by a phase separation method, offsets can be eliminated by an automatic balance method, and low-frequency noise components can be eliminated. Removal is being carried out.

In addition, signal processing is usually performed to remove or reduce noise and increase the signal-to-noise ratio by performing frequency separation using a filter.

FIG. 2(4) shows a configuration diagram of a signal processing circuit based on frequency separation using an ordinary filter.

In FIG. 2, the amplifier/detector 5 has the same structure and function as that shown in FIG. 1, and receives a needle signal detected from a probe (not shown). The output flaw signal Sh of this amplifier/detector 5 is filtered through a low-pass filter (
(hereinafter abbreviated as LPF) 10, the corresponding LPFI
O removes noise at a higher frequency than the true needle signal.

This LPFIO output flaw signal is input to a high pass filter (hereinafter abbreviated as HPF) 11, and this IIPFI
I removes noise at frequencies lower than the true needle signal. As a result, the HPFII outputs a needle signal SB having a relatively clear waveform as shown in FIG. In this way, depending on the voltage level of the needle signal Sm, the flaw 2
The size of the image is determined.

However, in the method of performing flaw detection with a probe while transporting the test object, if the transport speed is slow, the relative velocity (flaw detection speed) between the probe and the test object becomes small, so that the frequency of the needle signal is, for example, 0. It may be as low as IHz to 5Hz. When the needle signal frequency becomes low in this way,
It becomes difficult to distinguish it from similar low-frequency noise, and it is not possible to clearly detect the true needle signal just by frequency separation using conventional LPF and HPF! There was a drawback. Moreover, when performing surface flaw detection on hot slabs, etc., the distance between the glove and the slab surface must be approximately 10'+o+ from the viewpoint of heat resistance, so the level of the detected needle signal is low. Since the noise signal is small and includes noise signals due to rough irregularities on the slab surface, the true needle signal may be hidden by the noise, making it difficult to identify defects on the hot slab surface using only the conventional frequency separation described above. was difficult.

The purpose of the present invention is to effectively remove noise similar to needle signals, greatly improve the signal-to-noise ratio, and to detect defects with high precision even on objects with rough surfaces such as hot slabs. The purpose of the present invention is to provide a current flaw detection device.

The present invention provides a signal processing circuit consisting of a multiplier that performs a self-square operation on a needle signal detected by a probe and obtained through detection and frequency separation, and a low-pass filter that integrates the output signal of the multiplier. This is intended to improve the S/N ratio by effectively removing noise similar to needle signals.

If the detected needle signal has a one-cycle waveform, a multiplier that multiplies the delayed flaw signal obtained by delaying the detected and frequency-separated needle signal by half a cycle and the non-delayed needle signal; - By processing the needle signal through an autocorrelation filter consisting of a low-pass filter that integrates the output signal of the multiplier, it is possible to effectively remove noise at a frequency different from the frequency of the needle signal and to generate a one-cycle waveform. The aim is to completely remove random noise that does not exist.

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below based on illustrated embodiments.

Embodiments of the present invention are shown in FIGS. 3 to 8, and the same reference numerals as those in the conventional example shown in FIGS. 1 to 2 have the same configurations and functions. It is.

FIG. 3 shows a first embodiment of the present invention.

As shown, the output terminal of HPFII is branched and connected to two input terminals of multiplier 21.

With this configuration, the needle signal detected by the probe (not shown) is transmitted to the amplifier/detector 5 and the LPFI.
O, HPFI 1 performs signal processing by detection and frequency separation, and the needle signal 8 is output from HPFII.
B is relatively larger than the noise component. This needle signal Ss is squared by the multiplier 21, which has the function of performing a product operation of two input signals, so the needle signal is amplified, noise is compressed, and the relative level difference between them is significant. Become something. The needle signal including the harmonic component outputted from the multiplier 21 is integrated by the LPF 22 with a long time constant to produce a clear needle signal Sc as shown in FIG. 3 (5).
is obtained.

Therefore, according to this embodiment, when the needle signal is at a level slightly higher than the noise due to frequency separation processing, by effectively increasing the level difference,
The SN ratio can be improved.

FIG. 4 shows a second embodiment of the present invention, which is suitable for a case where the needle signal detected by the glove has a waveform of one cycle.

In the figure, the difference from the first embodiment shown in FIG. The delay element 31 delays the phase of the needle signal by half the period τ of the needle signal.

The frequency of the needle signal detected by the probe is determined by the flaw detection speed of the probe, that is, if the probe is fixedly installed and probes while moving the object, the speed of movement of the object,
In the case where the object to be examined is fixed and the probe is moved while exploring, the moving speed of the probe, or in the case where the probe and the object are to be explored while moving together, the relative speed of the two, and , in the self-comparison type, two pickup coils (the 30° 3D
) is determined by the relationship with the interval. Further, when the probe is a self-comparison type coil as described above, the needle signal has a one-cycle and a half shape.

When such a one-cycle waveform needle signal is detected □, after amplifying and detecting it with the amplification/width/detector 5, the LPFI
By separating and extracting needle signal frequency components from O and HPFII, a needle signal having a waveform shown in FIG. 5(4) is output from HPFII. This needle signal is delayed by a half period .tau. by the delay element 31 and is input to the multiplier 21 as a delayed flaw signal 8 (T-.tau.) having a waveform shown in FIG.

The multiplier 21 performs a product operation of the delayed needle signal 5 (T-τ) and the non-delayed needle signal S (T),
The needle signal is converted to a negative polarity signal, and needle signals 8(T) and 5(T-τ) having the waveform shown in FIG. 5 (0) are output.

This needle signal is subjected to approximate integration of the following equation (1) using the LPF 22 having a long time constant, and is output as a clear needle signal SD as shown in FIG.

Sonia 8(T)・8(T-τ)dT...-
・(1) Here, if the integration shown in equation (1) is performed completely, it will be easily affected by offsets, drifts, etc. Therefore, it is not recommended to use the integration function of the low-pass filter LPF22 as in this embodiment. is suitable.

Therefore, according to the second embodiment, the needle signal is processed by an autocorrelation filter that takes the autocorrelation between the needle signal and the delayed flaw signal obtained by delaying the needle signal by half a period. Noise that does not have a cyclic waveform is completely removed, and periodic noise that has a frequency different from that of the needle signal is also removed.

Since the signal is greatly suppressed and the needle signal is amplified by multiplication, the signal-to-noise ratio is greatly improved, and flaws can be detected with high precision even on objects with rough surfaces such as hot slabs.

In the second embodiment described above, an example was explained in which the frequency of the needle signal is constant and the flaw detection speed is constant. Needless to say, the cutoff frequency of the filter and the delay time τ of the delay element must be controlled in accordance with the flaw detection speed.

Therefore, an embodiment applied to a flaw detection apparatus in which the flaw detection speed is arbitrarily controlled is shown in FIGS. 6 and 7.

FIG. 6 shows an overall configuration diagram of a flaw detection apparatus according to a third embodiment of the present invention, which performs flaw detection on a transported specimen.

As shown in the figure, the subject 1 is transferred to a transport roller 4.
It is installed on 1. A speed detector 42 is provided in engagement with the transport roller 41 so that the transport speed of the subject 1 can be detected.

This speed detector 42 calculates the frequency of the needle signal from the detected transport speed and based on this frequency.

The cutoff frequencies of LPFIo, 22 and HPFII and the delay time τ of delay element 31 are set.

As a result, the needle signal processing can be performed in the same manner as in the second embodiment shown in the 4° diagram.

Therefore, according to this embodiment, in addition to the effects of the second embodiment, needle signal processing corresponding to the actual flaw detection speed is performed even if the flaw detection speed is arbitrarily controlled or even if the flaw detection speed fluctuates. It has the effect of being able to

FIG. 7 shows a fourth embodiment of the present invention, which is a rotary flaw detection system in which the probe is rotated.

As shown, probes 3A and 3B
The drive motor 5 is fixedly attached to both ends of an arm 51 provided in parallel with the drive motor 5 at the center of the arm 51.
3 are engaged through the rotating shaft 51A. This drive motor 53 is connected to a motor control circuit 54, and the motor control circuit 54 is connected to a rotation speed detector 55.

Further, the glove 3A is connected to the amplifier/detector 5 via a signal transmitter 52. Probe 3B is also formed to undergo signal processing in the same way, although it is not shown.

With this configuration, by operating the drive motor 53 and rotating the arm 51, the probes 3A, 3
B is moved on a circular orbit and sequentially tests one surface of the object.

Further, the subject 1 is in a stopped state or is being transported at a sufficiently slow transport speed compared to the rotational speed of the probes 3A and 3B. In this way, the flaw signal detected by the rotating probe 3A is transmitted by the signal simulator 52 from the rotating section to the fixedly provided amplifier/detector 5. on the other hand.

The rotational speed detector 55 detects the rotational speed of the drive motor 53, calculates the frequency of the flaw signal corresponding to the flaw detection speed from this detected speed, and based on this frequency, LPFIo, 22
and cutoff frequency of HPFII.Thus, flaw signal processing can be performed in the same manner as in the second embodiment shown in FIG.

Therefore, according to the fifth embodiment, there is an effect similar to that of the third embodiment shown in FIG.

In addition, in this embodiment, if the delay time τ is set to the periodic time of one rotation of the probe and the same flaw is repeatedly detected, it is possible to perform autocorrelation for the same flaw more than once. As a result, the detection accuracy of flaw detection can be further improved.

FIG. 8 shows a configuration diagram of a fifth embodiment of the present invention, and shows an application example of the fourth embodiment shown in FIG.

As shown in FIG. 8, it is assumed that the probe 3A and the probe 3B are attached with a predetermined angular difference. By rotating the probes 3A and 3B, the flaw on the subject is moved in the direction of the arrow shown in the figure. The flaw signal detected by probe 3A is sent to amplifier/detector 5A and LPFI OA.
, detected through the HPFI IA and frequency separated by a predetermined frequency, and the probe 3A is detected by the delay element 31A.
The phase is delayed according to the mounting angle difference between the probe 3B and the rotational speed and is input to the multiplier 21. On the other hand, the flaw signal detected by the probe 3B is detected and frequency-separated by a predetermined frequency through the amplifier/detector 5B and the LPFIOB%HPF 11B, and is input to the multiplier 21 in the same manner as described above. The multiplier 21 multiplies these flaw signals, and the LPF 22 integrates the flaw signals. That is, this is a signal processing method that takes cross-correlation between the flaw signals detected by the probe 3A and the probe 3B, and basically performs noise removal similar to the autocorrelation filter described in the second embodiment shown in FIG. It is the law. Therefore, the same effect as the second embodiment can be obtained also by the fifth embodiment.

Mio, in the first to fifth embodiments described above, LPFIO
The frequency separation circuit formed from HPFII and HPFII has a function as a kind of band-pass filter, but when LPF22 is used as an integrator provided at the final stage, LPFlo is not necessarily necessary. do not have.

As explained above, according to the present invention, noise similar to flaw signals can be removed extremely effectively and the 8N ratio can be greatly improved. It has the remarkable effect of being able to detect flaws with high precision even if they occur.

[Brief explanation of drawings]

Figure 1 (3) is a configuration diagram showing the detection section of a conventional flaw detection device, ■ in the figure is a diagram of its output signal waveform, Figure 2 (a) is a configuration diagram of a signal processing circuit in the conventional example, and ■ in the same figure is a configuration diagram of the signal processing circuit of the conventional example. Its output signal waveform diagram,
FIG. 3 (4) is a configuration diagram of a signal processing circuit according to the first embodiment of the present invention, and 3 (4) in the same figure is an output signal waveform diagram. FIG. 4(A) is a configuration diagram of a signal processing circuit according to a second embodiment of the present invention, ■ in the same figure is a diagram of its output signal waveform, FIG. 5 is a signal waveform diagram of each part of the embodiment shown in FIG. 7 is a block diagram of a fourth embodiment of the present invention, and FIG. 8 is a block diagram of a signal processing circuit of an application example of the embodiment shown in FIG.・1...Object, 2...Flaw, 3...Probe, 4...Excitation oscillator, 5...Amplifier/detector,
1o...Low pass filter (L P F ), 11.
...High-pass filter (HPF), 21... Multiplier,
22...Low pass filter (LPF), 31...Delay element. Country I - Day L 7th

Claims (1)

[Scope of Claims] 1. A probe formed from an excitation coil excited by high frequency or the like and a pickup coil provided in association with the excitation coil is brought close to the surface of the object to be inspected to detect defects on the surface as desired. The flaw detection apparatus detects defects such as flaws on the object using a flaw signal having a frequency corresponding to the flaw detection speed obtained by detecting a signal induced in the pick-up coil. A filter that receives a signal as input and passes the frequency component of the flaw signal; a multiplier that performs a squaring operation on the output signal of the filter; a low-pass filter that integrates the output signal of the multiplier;
An eddy current flaw detection device comprising a signal processing circuit formed from. 2. A probe formed from an excitation coil excited by high frequency and a pickup coil provided in association with the excitation coil is brought close to the surface of the object to be inspected and relatively moved over the surface at a desired flaw detection speed, In an eddy current flaw detection apparatus that detects defects such as flaws on the object by using a cycle waveform flaw signal having a frequency corresponding to the flaw detection speed obtained by detecting a signal induced in the pickup coil, a filter that receives a signal and passes the frequency component of the flaw signal; a delay element that delays the output flaw signal of the filter by a time corresponding to a half cycle of the flaw signal; and a delayed flaw signal output from the delay element; a multiplier that performs a product operation with a flaw signal output from the filter;
A signal comprising: a low-pass filter that integrates the output signal of the multiplier; and means for setting the cut-off frequency of the filter and the low-pass filter and the delay time of the delay element according to the frequency of the flaw signal. An eddy current flaw detection device characterized by being equipped with a processing circuit. 3. An excitation coil excited by high frequency and the excitation coil,
A probe formed from a pickup coil provided in association with the pickup coil is brought close to the surface of the object to be inspected and relatively moved over the surface at a desired flaw detection speed, and a signal induced in the pickup coil is detected. In an eddy current flaw detection device that detects defects such as flaws on the object based on the obtained needle signal. a delay element configured so that the glove can output two or more needle signals having a predetermined period for one flaw, and delaying the phase of the needle signal by the predetermined period; a delay output from the delay element; An eddy current flaw detection device comprising: a multiplier that performs a product operation of a flaw signal and the needle signal; and a low-pass filter that integrates an output signal of the multiplier. .