JPS6270753A - Method for inspecting pulse body current - Google Patents

Abstract

PURPOSE:To obtain good quality detection output reduced in wave-form distortion, by storing the output of a detection coil at a flawless time and taking the differential voltage between memory output and the output of a detection coil at the time of actual inspection to amplify the same. CONSTITUTION:In a flawless state, switches 42, 44 are closed to an '1' side and a memory order switch 40 is closed. Subsequently, the differential output of detection coils 14, 16 is applied to an A/D converter 26 which in turn samples differential voltage at every clock input of 1,024kHz to write the same in a memory 28. After the storing of the differential voltage in the memory 28 was finished, the switches 42, 44 are changed over to a '2' side to perform the inspection of an actual object to be inspected. The differential voltage of the coils 14, 16 is applied to one input terminal to a differential amplifier 38 and the output voltage of a D/C converteer 30 read from the memory 28 is applied to the other input terminal of the amplifier 38 to calculates the difference between both of them. This difference is almost zero if the object 10 to be inspected is flawless and, if there is a flaw, comes to differential voltage determined by said flaw and S/N is enhanced to a large extent and good amplification can be performed and good quality detection output free from wave-form distortion is obtained.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an eddy current inspection method for continuously inspecting metal materials in a non-destructive and non-contact manner, in particular, a pulsed magnetic field is applied to an object to be inspected.
The present invention relates to a pulse eddy current inspection method that processes the detection coil voltage generated by the eddy current flowing through the test object to inspect the properties of the test object.

[Conventional technology]

When the magnetic field acting on a metal body is changed, eddy currents are generated in the metal body, and the distribution state of the eddy currents changes depending on the properties of the metal body, so by detecting the distribution state, it is possible to perform flaw detection on the metal body. . A magnetic field is generated by passing a current through an excitation coil, and an alternating current (continuous wave) is often used as the current.

Pulsed currents may also be used, and compared to eddy current testing that uses continuous waves, eddy current testing that uses pulsed waves can apply a strong magnetic field to the local area of the subject, so it is possible to obtain deep information about the subject. (possible to detect internal defects and measure the dimensions of thick plates); ■If the object to be examined is a ferromagnetic material, it is possible to detect information about the characteristics of the magnetic saturation region due to the effect of a strong magnetic field (the advantages of high carbon steel) (Quality determination); Since the pulse wave contains frequency components in a wide range, it is possible to simultaneously investigate eddy current reactions to multiple frequencies (selective detection of harmful defects).
, and other characteristics. Figure 4 shows the general method of the pulsed eddy current method.

In this figure, 10 is the subject, 12 is the excitation coil, and 14.16
20 is a clock generator, 24 is a pulse generator, 38 is an amplifier, 51 to 5n are delay times t +
−t n delay circuits, and 61 to 6n are sample and hold circuits. To explain the operation with reference to the waveform diagram in FIG. 5, the clock generator 20 generates a timing signal (
Using the clock as a trigger, the pulse generator 24 generates a pulse (excitation current), and the coil 12 is excited with the excitation signal having a pulse width τ. Coil 12 generates a magnetic field, which causes eddy currents to flow in subject 10 and induces a voltage in detection coils 14,16.

The detection coils 14 and 16 are arranged on both sides of the excitation coil 12, so that the induced voltages due to the magnetic flux generated by the excitation coil 12 are equal, and because they are differentially connected, they cancel each other out. The induced voltage caused by the eddy current also cancels each other out, but there is a flaw in the test object and the eddy current distribution state is different from that directly below the coil 14 to that of the coil 1.
6, there is a difference in the induced voltage, and this difference is amplified by the amplifier 38 and becomes the output voltage. The amplifier output voltage is applied to sample and hold circuits 61 to 6n, and each sample and hold circuit 61 to 6n samples the amplifier output voltage at a timing delayed by t1 to tn from the clock output from the clock generator 20. Therefore, the outputs Ul to Un of the sample and hold circuits 61 to 6n are as shown in the figure. From these outputs U1 to On, object characteristic information ranging from the surface layer to the deep part of the object can be obtained. In detail, the short delay time t1
The output U1 sampled at 1 includes information from the surface to a relatively shallow part, and the output Un sampled at a long delay time tn includes information from the surface to the deep part,
Output (U
i) contains information from the surface to medium depth. If these outputs U1 to Un are mutually calculated appropriately, it is possible to obtain independent information for each layer from the surface to the deep part of the object. When applied to defect detection, once the form of a harmful defect is known, it is possible to obtain information that is independent of each layer. It is also possible to selectively extract.

[Problem that the invention seeks to solve]

By the way, in the pulsed eddy current method, it is possible to generate a strong magnetic field by passing a large current through the excitation coil, and the use of a strong magnetic field is a feature of the pulsed eddy current method. This means that it is significant, and the problem is how to cancel it.

Since the detection coil is placed close to the excitation coil, there is magnetic coupling between the two coils, and if there is a change in the current flowing through the excitation coil, a voltage is induced in the detection coil. This induced voltage (noise voltage) is much larger than the voltage (signal voltage) generated in the detection coil by the eddy current flowing through the test object, and since weak signal voltages are greatly amplified and used for inspection, the above induced voltage must be strictly controlled. If they are not canceled out, a small signal voltage will be superimposed on a large-amplitude noise voltage, and since amplifiers generally have saturation characteristics, amplifying a composite wave will distort the waveform and make it impossible to extract the signal component. In order to perform effective amplification, it is necessary to suppress and eliminate the noise voltage below the signal voltage.

To eliminate the induced voltage caused by the direct coupling between the excitation coil and the detection coil, two detection coils are placed at equal intervals around the excitation coil as shown in Figure 3 (al) (this is the same as in Figure 4). , by differentially connecting these detection coils 4 and 16, or by connecting the excitation coil 1 as shown in FIG. 3(b).
2 is electromagnetically shielded with a conductive material and/or magnetic material 18, leaving the surface facing the subject 10, or as shown in FIG.
As shown in C1, two excitation coils and two detection coils are used (12a and 12b, 14a and 14b are used), with one set facing the subject 10 and the other set facing the reference material 10A.
There is a method in which the detection coils 14a and 14b are differentially connected.

These are commonly used in continuous wave eddy current testing, but are insufficient for pulsed wave eddy current testing. In other words, since the pulse method uses a magnetic field that is 10 to 1000 times stronger than the continuous wave method, the induced voltage due to direct coupling is also 10 to 1000 times larger.On the other hand, the magnitude of the weak signal voltage is not much different between the two methods. In order to implement the pulse method, induced voltage suppression performance that is 10 to 100 times higher than that of the continuous wave method is required. Although the induced voltage can be suppressed to a considerable extent by the methods shown in FIG. 3, it is difficult to completely cancel it out. The reason for this is that the two test coils used differentially are made with exactly the same dimensions and arrangement.
For example, it is impossible to assemble it due to the limits of machining accuracy.

The present invention provides a pulsed eddy current inspection method that can sufficiently suppress the induced voltage due to direct coupling in a strong magnetic field, effectively extract weak signal components, and fully utilize the effects of a strong magnetic field. This is what I am trying to do.

[Means for solving problems]

The present invention uses an excitation coil and a detection coil, applies a pulsed magnetic field to a subject by supplying a pulsed current that repeats at a constant cycle to the excitation coil, and converts the voltage of the detection coil generated by the eddy current generated in the subject into an amplifier. In the pulsed eddy current inspection method, in which the amplified output is sampled and held at small intervals and the properties of the specimen are inspected using the multiple sampled and held outputs, the detection coil output when there are no defects is synchronized with the pulsed current. Sample it at small intervals, digitize it, and store it in a memory, and for inspection, read out the memory in synchronization with the pulse current, convert the continuous output into an analog, and find the difference between this and the detection coil output. , the difference is amplified by the amplifier.

[Effect]

In the present invention, the detection coil output obtained in a state where there is no defect and is not facing the object to be inspected, or even if it is facing the object, it is known that there is no defect) is captured in the memory, Obtained by facing a possible specimen (during examination)
Take the difference between the detection coil output and the memory readout output,
The induced voltage is suppressed. According to this method, the induced voltage can be sufficiently suppressed, and the voltage can be amplified and tested without causing waveform distortion.

〔Example〕

FIG. 1 shows an embodiment of the invention. The same parts as in FIG. 4 are given the same reference numerals. In the present invention, the detection coil 14.1
An A/D converter 26 that samples and digitizes the differential output (analog) of 6, a memory 28 that stores the output of the converter, and a D/A converter 30 that converts the read output of the memory into analog. establish. 36 is an address counter that outputs the access address of the memory 28, and 34 is an A/D.
It is an AND gate that clocks converter 26.

The clock generator 20 generates a 1024 KHz clock, which becomes an I KHz trigger pulse in the frequency divider 22, which is input to the pulse generator 24 and outputs the coil 12.
It generates an excitation pulse current of 1021IK Hz.
The clock is also input to the address counter 36, and when the memory command switch 4o is pressed, the start logic circuit 3
2 produces an H level output, it is input to the A/D converter 26 through the AND gate 34. The IKHz l/rega signal is input to delay circuits 51 to 5n and the delay timing t is
It becomes a reference signal for generating l % t n, and also inputs to the address counter 36 and start logic circuit 32 and becomes a reset signal.

As shown in FIG. 2, the pulse generator 24 is connected to a DC power source 7B.
, a choke coil 72, a cyrisk 74, a pulse transformer 76, and an LC distributed constant circuit. DC power supply 78
Current flows from the choke coil 72 to the excitation coil 12 through the LC circuit, and in this state, when an IKHz pulse from the frequency divider 22 is applied to the transformer 76, the thyristor 7
4 turns on and shorts the power supply circuit. As a result, the current flowing through the coil 12 suddenly decreases, causing a change in the subject excitation magnetic field. The width of the pulse thus generated is determined by E and the number of stages N of the LC circuit, and the pulse amplitude is determined by the voltage of the power supply 78. Note that if the values of L and C are determined so that 17Σ〒- is equal to the impedance of the excitation coil, a current pulse with an efficient and smooth waveform can be generated.

To explain the operation in Fig. 1, initially there is no defect (the coil is not facing the object, or even if it is facing the object, the object is normal and has no defects), and the switches 42 and 44 are set to 1.
side and close the memory command switch 40. In this state, the differential output of the detection coil 14°16 is output to the A/D converter 2.
6, the converter performs sampling and A/D processing of the sample every time a 1024KHz clock is input.
Perform the conversions and write them to memory 28. Each sample is represented by 10 bits in this example, and each 10-bit sample is stored in memory 28 at an address output by counter 36. If the detection coil difference voltage generated at a frequency of IKHz is A/D converted at 1024KHz, 1024 samples are obtained each time, and these are stored in the memory 28.
By storing the data in the data, reading it out, and performing D/A conversion using the converter 30, the above detection coil difference voltage can be reproduced with high precision.

The 1024 samples stored in the memory 28 are IKH
It is that during one particular period of z.

The start logic circuit 32 is for this purpose, for example, the second IKHz signal after the memory command switch 40 is pressed.
A pulse opens AND gate 34, and a third pulse closes it.

Since the address counter 36 is also reset by the IKHz trigger signal, the count value repeats 0 to 1023. A/
When writing the output of the D converter 26, the memo IJ 28 is placed in write mode (the write enable signal is
6, etc.), and then goes into read mode.
Therefore, the D/A converter 30 repeatedly outputs a voltage approximate to the detection coil difference voltage.

When the storage of the detection coil difference voltage in the memory 28 is completed, the present apparatus is ready, and the next step is to inspect the actual object 10 (which may have a defect). At this time, switch 42.
44 switches to the 2 side. The voltage difference between the detection coils 14 and 16 is applied to one input of the differential amplifier 38, and also to the D/A
The voltage output by converter 30 is applied to the other input of the differential amplifier, and the difference between the two is determined. This difference is \0 if the test object 10 has no defects, and if there is a defect, it can be expected that the voltage difference will be determined by the defect, and the S/N
will be significantly improved. Since the small signal voltage is not superimposed on the large amplitude noise voltage, high gain amplification is possible.

The output voltage of the amplifier 38 is sampled and held as in FIG. 4, and an output UI"Un is produced. Delay circuits 51 to 5
n can be easily configured using one-shot multicasting or the like.

The coil arrangement in Fig. 1 is equivalent to Fig. 3 (a), but it may also be the same as Fig. 3 (C1).
In the case of 1, sampling to the memory 28 is performed with both coil sets facing the reference material.

The sample and hold timing tl-1n is the A/D
It is not the same as the sampling timing of converter 26.

It is also possible to keep this the same and digitize the detection coil difference voltage during actual measurement and find the difference with the readout output of the memory 28 (digitally), but with this method, the detection coil difference voltage is displayed on the display, It is difficult to examine visually, intuitively, and physically (in the case of Figure 1, the output of the amplifier 38 can be input to the display), it requires a CPU with high calculation speed, and it uses analog amplification means. This requires a high resolution A/D converter (eg 16 bits instead of 10 bits).

〔Effect of the invention〕

As explained above, according to the present invention, the detection coil output when there is no defect is stored, the difference between this and the detection coil output during actual inspection is taken, and this difference is amplified. Amplification with good S/N ratio can be achieved, and high quality detection output with little waveform distortion can be obtained.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a circuit diagram of a pulse generator, FIG. 3 is a layout diagram of an excitation coil and a detection coil, and FIG. 4 is a block diagram showing a conventional example. FIG. 5 is a waveform diagram for explaining the operation. In the drawing, 12 is an excitation coil, 14.16 is a detection coil, and I
0 is a subject, 28 is a memory, and 38 is an amplifier. Applicant: Sumitomo Metal Industries Co., Ltd. Applicant: Hara Denshi Sokki Co., Ltd. Representative Patent Attorney Minoru Aoyagi

Claims (1)

[Claims] Using an excitation coil and a detection coil, a pulsed current is supplied to the excitation coil repeatedly at a constant cycle to apply a pulsed magnetic field to the subject, and the voltage of the detection coil is generated by the eddy current generated in the subject. In the pulsed eddy current inspection method, in which the amplified output is amplified by an amplifier, the amplified output is sampled and held at small intervals, and the properties of the specimen are inspected using the multiple sampled and held outputs, the detection coil output when there are no defects is synchronized with the pulsed current. The pulse current is sampled at small intervals, digitized, and stored in a memory. During inspection, the memory is read in synchronization with the pulse current, the read output is converted into an analog, and the difference between this and the detection coil output is calculated. A pulsed eddy current inspection method characterized in that the difference is amplified by the amplifier.