JPH1114599A - Flaw detecting apparatus by magnetic leakage flux detecting method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a flaw detecting apparatus which can detect a flaw signal in a wide range by a method wherein a threshold value according to a change in a noise signal is set. SOLUTION: The output signal of a sensor 8 at a detecting head is amplified by an amplifier 10, it is then divided into two signals, one signal (a) is passed through a low-pass filter 11 so as to be sent to an A/D converter 13, the other signal (b) is sent to the A/D converter 13 as it is, and data which is sampled by the A/D converter 13 is stored on a memory 14. Then, sampling data which is obtained through the low-pass filter 11 is sent to a computing means 18 via an arrangement means 16, it is multiplied and added, a thresholds value according to a change in a noise signal is found, the threshold value and data which is sent directly from the sensor 8 are compared by a comparison means 19, and whether the threshold value is good or not is judged by a judgment means 20.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a flaw detecting device for detecting a surface flaw of an object to be detected by a magnetic flux leakage inspection method.

[0002]

2. Description of the Related Art In the leakage magnetic flux detection method, a magnetic pole is brought close to an object to be detected to magnetize the object to be detected. This is a flaw detection method in which the presence / absence of a flaw is detected based on the strength of the output signal. According to this magnetic flux leakage inspection method, unlike magnetic powder inspection method in which fluorescent magnetic particles are adsorbed to the leakage magnetic flux and visually inspected under an ultraviolet lamp, flaws can be quantitatively detected, and inspection can be automated. The utility value will be high.

[0003]

By the way, the output waveform of the above-mentioned sensor greatly changes due to the influence of noise as shown in FIG. This is mainly due to the fluctuation of the gap between the object to be detected and the sensor, and it is very difficult to distinguish the flaw signal from this. Therefore, conventionally, as shown in FIG. 4, the threshold value L 'is set to a level sufficiently higher than the intensity level of the noise signal A. In this case, of course, a relatively large level of the flaw signal B
Although 1 can be detected, it is impossible to detect the flaw signal B2 having a relatively small intensity level, and a decrease in detection accuracy cannot be avoided.

For example, Japanese Patent Application Laid-Open No. 3-246463 discloses that a large number of sensors are arranged in the width direction of a steel sheet, divided into two groups, and a difference between peak values of detection signals for each group is obtained. It is described that a noise component generated in the width direction of the sensor is removed. However, according to this, a noise component generated on a scanning line of one sensor cannot be removed. Does not reach.

[0005] The present invention has been made in view of the above-mentioned conventional problems, and an object thereof is to set a threshold value in accordance with the fluctuation of a noise signal to detect a wide range of flaw signals. It is an object of the present invention to provide a flaw detection device using a magnetic flux leakage inspection method, which makes it possible to greatly improve the detection accuracy.

[0006]

SUMMARY OF THE INVENTION In order to solve the above problems, the present invention provides a detection head having a magnetic pole for magnetizing an object to be detected and a sensor for detecting magnetic flux leaking from the object to be detected, and a sensor for the detection head. Amplifier, a low-pass filter for extracting a low-frequency signal from the output signal of the amplifier, and an A / D for inputting a direct signal from the amplifier and a low-pass signal passed through the low-pass filter and performing A / D conversion. A converter, a memory for storing the data sampled by the A / D converter for each of the direct transmission signal and the low-pass signal, and calculating a threshold value based on a predetermined number of low-pass signal data stored in the memory And a comparing means for comparing the threshold value obtained by the calculating means with the data of the direct transmission signal stored in the memory. Characterized in that was.

In the flaw detection device thus configured, a large change (swell) of the noise signal can be mainly extracted by passing the output signal of the sensor through a low-pass filter, and the swell is digitally converted and sampled. Then, when this sampling data is sequentially included in the calculation formula, as shown in FIG. 2, a threshold value S corresponding to the fluctuation of the noise signal A is obtained, and the flaw signal B1 having a relatively large intensity level as well as the flaw signal B1 having a relatively large intensity level are obtained. Flaw signal B2 of small intensity level
Can also be detected.

[0008]

Embodiments of the present invention will be described below with reference to the accompanying drawings.

The present embodiment is applied to the detection of surface flaws of the plate-like ring 1 as shown in FIG. A plurality of extended rollers 3 are held in contact with the inner diameter of the plate-shaped ring 1 by being pinched by the roller pair 2, and the plate-shaped ring 1 is rotated using one of the extended rollers 3 as a driving wheel, and the plate surface is detected. The head 4 is moved closer. The detection head 4 includes a U-shaped yoke 6 around which a coil 5 is wound, and a sensor (magnetically sensitive sensor) 8 embedded in a resin layer 7 that fills an opening of the yoke 6. When a current is applied, both ends of the yoke 6 become magnetic poles, and a magnetic field is generated in the plate-like ring 1. At this time, if a flaw such as a crack is present on the surface of the plate-like ring 1, a leakage magnetic flux is generated, and the sensor 8 outputs a signal corresponding to the leakage magnetic flux.

The detection head 4 is incorporated in a flaw detection device according to the present invention, and as shown in FIG.
b, and one signal a is a low-pass filter L
The signal b is sent as a low-pass signal c through the PF 11 to a personal computer PC 12 described later, and the other signal b is sent to the PC 12 as it is. LPF11 is about 5H here
It has a function of extracting a signal with a low frequency of z or less, and the low-pass signal c mainly represents a large fluctuation (undulation) of the noise signal. Note that the low-pass signal c is delayed by a short time (about 0.1 sec) from the direct transmission signal b directly transmitted from the amplifier 10 to the PC 12.

The PC 12 has an A / D converter 13 for inputting the low-pass signal c and the direct transmission signal b and performing AD conversion.
It comprises a memory 14 for storing data sampled by the A / D converter 13 and a data processing unit 15 for processing the data stored in the memory 14. The A / D converter 13 has a function of sampling (sampling) in a predetermined time unit (for example, 0.6 msec unit), and the memory 14 stores a sampling for each of the direct transmission signal b and the low-pass signal c. The data will be stored.

On the other hand, the data processing section 15 includes a pair of arranging means 1 for arranging a predetermined number (for example, 2500) of data of the direct transmission signal b and the low-pass signal c stored in the memory 14.
6, 17 and an operation unit 18 for sequentially taking in the data of the low-pass signal c from one of the arrangement means 16 and obtaining a threshold value L by a calculation formula described later. Comparing means 19 for comparing the data of the direct transmission signal b sent from the other arranging means 17, and judging means 20 for judging pass / fail based on the comparison result by the comparing means 19. The data processing unit 15
Operate according to a program stored in a memory unit (not shown).

Here, the calculation formula used in the calculation unit 18 is a simple one including multiplication and addition, and is as shown in the following (1). L [i] = [i + d] × e + f (1) In the equation (1), L [i] is a threshold value corresponding to the array data i of the direct transmission signal b, e and f are coefficients, and d is An additional amount (approximately 150 data) of data corresponding to the time delay (0.1 sec) of the low-pass signal is shown. When the number of arrays by the array means 16 is 2500, an operation for changing i from 0 to 2350 Will be done.

In the present embodiment, the plate ring 1 is rotated one or more times in the mode shown in FIG. 3 and a part of the output signal of the sensor 8 is converted into the LPF 11 and the A / D converter 1 during this time.
3, sequentially pass through the memory 14 and the arranging means 16 and input to the calculating means 19 to determine the threshold value L. This threshold L
Is a value obtained by multiplying and adding data obtained by sampling the swell component of the noise signal extracted through the LPF 11 as described above. Therefore, as shown in FIG. fluctuate. In addition,
The fluctuation of the noise signal A is mainly caused by uneven rotation of the plate ring 1 due to slippage, eccentricity, distortion, and the like.
This is caused by the gap fluctuation between the sensor and the sensor 8.

The signal of the threshold value L obtained by the arithmetic means 19 is sent to the comparing means 19, and the comparing means 19
The threshold value L is compared with the sampling data (raw data) of the direct transmission signal b sent from the arrangement unit 17, and the comparison result is sent to the determination unit 20. Then, the judgment means 20
Counts the number of raw data exceeding the threshold value L. If the number is equal to or less than a predetermined number, it is judged as pass, and if the number exceeds the predetermined number, it is judged as defective. At this time, as shown in FIG. 2, since the threshold value L fluctuates in accordance with the fluctuation of the noise signal A, the flaw signal B having a relatively large intensity level is obtained.
1, the flaw signal B2 having a relatively small intensity level can be detected, and therefore, the surface flaw of the plate-like ring 1 can be detected with high accuracy.

In the above-described embodiment, an example in which the process is performed on the plate-like ring 1 has been described. However, the present invention can be applied to any arbitrary configuration, and for example, a flaw detection of a wide steel plate can also be performed. In this case, a plurality of the detection heads 4 are arranged in the width direction of the steel sheet, or a plurality of sensors 8 are arranged in the width direction of the steel sheet by sharing the magnetic poles (coils 5, yoke 6). The same processing as in the above embodiment is performed for the output signal of

[0017]

As described above, according to the flaw detection apparatus based on the leakage magnetic flux flaw detection method according to the present invention, a threshold value according to the fluctuation of the noise signal is set, and a wide range of flaw signals can be detected. Since the detection is enabled, the detection accuracy is remarkably improved, and the utility value is high.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a structure of a flaw detection device according to the present invention.

FIG. 2 is a graph showing a threshold value set according to the present invention superimposed on an output waveform of a sensor.

FIG. 3 is a perspective view schematically showing an execution mode of flaw detection according to the present invention.

FIG. 4 is a graph showing a threshold value according to a conventional setting method superimposed on an output waveform of a sensor.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Plate-shaped ring (detected body) 4 Detection head 5 Coil 6 Yoke 8 Sensor 10 Amplifier 11 Low-pass filter 13 A / D converter 14 Memory 18 Calculation means 19 Comparison means 20 Judgment means

Claims (1)

[Claims] 1. A detection head having a magnetic pole for magnetizing an object to be detected and a sensor for detecting leakage magnetic flux from the object, an amplifier for amplifying an output signal of the sensor of the detection head, and an output signal of the amplifier. A low-pass filter that extracts a low-frequency signal, an A / D converter that inputs a direct signal from the amplifier and a low-pass signal that has passed through the low-pass filter and performs AD conversion, and data sampled by the A / D converter. The direct transmission signal, a memory for storing each low-pass signal, calculation means for calculating a threshold based on a predetermined number of data of the low-pass signal stored in the memory, and a threshold calculated by the calculation means. A flaw detection device using a magnetic flux leakage inspection method, comprising: comparison means for comparing the data of the direct transmission signal stored in the memory.