JP2953301B2 - Ultrasonic flaw detection method and device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an ultrasonic flaw detection method and apparatus for scanning and flaw-detecting a weld or the like by an ultrasonic pulse reflection method.
It relates to a technique for improving the ratio.

[0002]

2. Description of the Related Art Inspection of welds for flaw detection is indispensable for quality assurance of welds in the production of welded steel pipes and building structures. Ultrasonic flaw detection is generally used because of its excellent detectability. The problem with flaw detection is noise. There are two types of noise, namely, electrical noise coming from other facilities in the factory and the like, and material noise resulting from grain boundary scattering at the welded portion. If there is such noise, the defective echo will be buried in the noise signal, making it difficult to identify the defective echo. Therefore, it is important to reduce the noise signal from the reflected echo signal and improve the SN ratio.

[0003] Synchronous averaging is effective for reducing noise signals. In this method, the average of the echo heights of the repeatedly arriving ultrasonic signals is calculated, and the noise intensity can be reduced to 1 / N ^1/2 (N = average number). -14169, JP-A-58-16
No. 9055).

[0004]

However, the prior art has the following problems. For example, in the case of flaw detection of a welding line while moving an object, such as in online flaw detection of a welded steel pipe, the time for a defect to cross an ultrasonic beam is limited.
For this reason, if the number of times of averaging is too large, the defect echo intensity after the averaging process is reduced, and defects are often overlooked. On the other hand, if the average number is reduced so as not to overlook the defect, the effect of improving the SN ratio is reduced. As described above, there is an optimum value for the average number, but in the related art, the optimum average number has to be obtained by trial and error. For this reason, every time flaw detection conditions such as the line speed and the probe are changed, trial and error is required each time, which is time-consuming and costly.

SUMMARY OF THE INVENTION The present invention has been made to solve the above problems, and has as its object to easily and promptly determine the optimum number of times of averaging for synchronous averaging processing without trial and error in ultrasonic flaw detection. And

[0006]

According to the ultrasonic flaw detection method of the present invention, an S / N ratio is improved by scanning flaw detection of an object by an ultrasonic pulse reflection method and synchronously averaging the detected reflected wave signals. In the method, the effective beam width D is detected in advance, the pulse density Pd is calculated from the pulse repetition frequency and the scanning speed, and the synchronous addition average of the reflected wave signal is calculated by the average number N calculated by N = D / Pd. It is characterized by performing.

The detection of the effective beam width D is performed by an ultrasonic flaw detection test for a fine defect hole such as a pinhole on a test piece which has been artificially prepared in advance.
An effective beam width D is obtained in advance. The effective beam width D _R of the actual subject is at least D _R ≧ D.

In the present invention, the effective beam width is 4
When the effective beam width is larger than 4 mm, the average number N is obtained from N = D / Pd.
From N = 4 / Pd.

The ultrasonic flaw detector used in the method of the present invention comprises:
Means for detecting a repetition frequency of a pulse transmitted to the ultrasonic probe (pulse repetition frequency detection means); and means for detecting a relative scanning speed between the ultrasonic probe and the subject (scanning speed detection means) ), Means for calculating the pulse density from the pulse repetition frequency and the scanning speed detected by both means (pulse density calculating means), and means for detecting the effective beam width of the ultrasonic probe (effective beam) Width detecting means) and the latter two means (pulse density calculating means and effective beam width detecting means) calculate the average number of times from the detected pulse density and effective beam width, and reflect the average number of times. Means for setting the wave signal synchronous addition and averaging processing means (average number calculation means).

[0010]

The drop of the defective echo after the synchronous averaging process occurs when the average number of times is set to be equal to or larger than the number of times the defective echo appears. Therefore, an optimal average number of times can be obtained by obtaining the number of times that a defect echo appears.

In the present invention, first, an effective beam width is obtained from an echo signal such as an artificially created minute defect hole by an effective beam width detecting means. The effective beam width is
This refers to the distribution width of the echo intensity, for example, the distribution width of the echo intensity when the reference level is set to a value lower by -3 dB than the maximum echo intensity. The reference effective beam width is obtained by an ultrasonic flaw detection test for a small defect hole artificially created as described above. Therefore, if there is a defect larger than the size of the minute defect hole in this range, a defect echo appears.

On the other hand, the pulse density Pd is calculated by the pulse density calculating means. Since the pulse density calculating means is composed of the pulse repetition frequency detecting means and the scanning speed detecting means, the pulse density Pd is calculated from Pd = V / PRF by the pulse repetition frequency PRF and the scanning speed V detected by these means. Ask. The value obtained by dividing the effective beam width by the pulse density is the number of times a defect echo appears. If the average number is set to this value, the defect echo after the averaging process does not decrease, and the SN ratio can be more appropriately improved.

In ultrasonic flaw detection of a welded portion performed online such as a welded steel pipe, according to experiments, the effective beam width is 4 mm.
It turned out that the optimal average number reached a plateau. This is because when the effective beam width becomes 4 mm or more, the beam path of the defect echo slightly changes at each point when the defect crosses the ultrasonic beam, and the average of the waveforms having the deviated beam path is calculated as the defect. This is because the echo is reduced. Therefore, when the effective beam width is 4 mm or more, the average number is 4 / pulse density. Although this shift differs depending on the wavelength, the nominal frequency is changed from 2 MHz to 1
The above result can be applied in the range of 0 MHz.

[0014]

FIG. 1 is a block diagram showing an ultrasonic flaw detector according to one embodiment of the present invention. In the figure, reference numeral 100 denotes a subject, and here, a case of an electric resistance welded tube is shown. 10
Reference numeral 1 denotes a weld, and reference numeral 102 denotes a defect due to insufficient penetration at the weld. The steel pipe 100 is traveling straight upward at a speed V in the figure.

The ultrasonic flaw detector 10 according to the present embodiment includes the following elements 11 to 20.

The ultrasonic probe 11 has a nominal frequency of 5 MHz.
z, a vibrator width of 10 mm is used. The ultrasonic wave is incident on the steel pipe 100 as a transverse wave having a refraction angle of 45 ° so that the welded portion 100 is detected.

The ultrasonic transmission / reception device 12 transmits an electric pulse to the probe 11 to generate an ultrasonic wave, captures the ultrasonic wave reflected from the welding portion 101 by the probe 11 and converts the ultrasonic wave into an electric signal. To detect.

The synchronous averaging processor 13 adds and averages ultrasonic electric signals (electric signals before averaging processing shown in FIG. 2) repeatedly arriving from the ultrasonic transmitting and receiving apparatus 12 with their phases aligned. FIG. 2 shows a case where the average number of times is four.

The defect judging device 14 notifies the detection of a defect when a defect having a certain strength or more appears.

The pulse repetition frequency detecting device 15 detects a set frequency of a clock signal for generating an ultrasonic pulse of the ultrasonic transmitting / receiving device 12.

The scanning speed detector 16 is, for example, a steel pipe 1
The scanning speed V is detected from the rotation speed of the sensor roll 17 that has come into contact with 00.

The pulse density calculating device 18 calculates the pulse density P
d is calculated from the equation Pd = V / PRF. Here, V is the scanning speed detected by the scanning speed detecting device 16, PRF
Is a pulse repetition frequency detected by the pulse repetition frequency detection device 15.

As shown in FIG. 3, the effective beam width detector 19 measures an echo train from a needle-like defect hole 111 provided in the test steel pipe 110 in advance, and, based on the maximum intensity of the defect echo, for example, -3 dB. Is determined as a reference level, and an echo intensity distribution width on the reference level is defined as an effective beam width. In this measurement, the average number is set to 1.

As will be described later, the average number calculation device 20 calculates the average number N from N = D / Pd when D <4 mm and N = 4 / Pd when D> 4 mm.

Next, the basis for calculating the average number of times will be described with experimental results. FIG. 4 shows the relationship between the optimum beam number and the effective beam width, with the number of times that the echo intensity decrease after averaging stays at -1 dB defined as the average number. Here, the pulse density Pd = 0.1 mm / p,
Cases of 0.25 mm / p and 0.5 mm / p are illustrated. The reason why the decrease in the echo intensity is defined as -1 dB is in the range of the reproducibility of the flaw detection, and such a decrease in the echo intensity does not lead to the overlooking of the defective echo.
The effective beam width was changed by changing the number of skips in oblique flaw detection.

As shown in FIG. 4, when the effective beam width is smaller than 4 mm, the optimum average number is reduced.
The relationship is approximately N = D / Pd. on the other hand,
When the effective beam width is larger than 4 mm, the optimum average number of times peaks at about 4 mm, and the relationship is about N = 4 / Pd. The reason for this is that at each point when the defect traverses the ultrasonic beam, the beam path of the defect echo changes slightly, and the average of the waveforms having the deviated beam path is averaged, so that the defect echo decreases. This is because, when the beam path is moved by 4 mm or more, the deviation of the beam path becomes remarkable.

FIG. 5 is an experimental result showing the relationship between the average number of times and the echo intensity reduction amount when the effective beam width is 4 mm or more. As shown in the figure, as the pulse density is smaller, the echo intensity does not decrease even if the average number is increased. The figure also shows the amount of decrease in the noise intensity. The greater the average number of times, the lower the noise intensity. From this figure, the average number of times that the amount of decrease in the echo intensity corresponds to -1 dB is found for each pulse density. Therefore, it has been proved that the SN ratio can be optimally improved by determining the average number of times by the above method.

In the ultrasonic flaw detection of this embodiment, first, the test steel pipe 110 is conveyed at the line speed V, and the effective beam width D is obtained from the reflected echo signal from the pinhole 111 by the method shown in FIG. Ask for.

Next, ultrasonic flaw detection is performed on the actual welded steel pipe 100. The pulse repetition frequency detector 15 detects the repetition frequency PRF of the ultrasonic pulse transmitted to the ultrasonic probe 11, and the scanning speed detector 16 detects the welding steel pipe based on the rotation speed sent from the sensor roll 17. The moving speed V of the motor 100 is detected and sent to the pulse density calculating device 18 to calculate the pulse density Pd by Pd = V / PRF. The obtained pulse density Pd is sent to the average number calculation device 20, where N = D / Pd when D <4 mm, and N = D / Pd when D> 4 mm.
The average number N is set in the synchronous averaging processor 13 as 4 / Pd. As described above, by performing the synchronous averaging process of the defective echo signal at the optimum average number N, the SN ratio can be improved to the highest state without a decrease in the defective echo. Therefore, if there is a defect in the welded portion, the defect determination device 14 notifies the defect without overlooking it.

Next, a description will be given of a specific example in which flaw detection is performed by changing the steel pipe forming speed, that is, the line speed, and simultaneously changing the average number of times. In the example shown in FIG.
2 kHz, steel pipe: outer diameter 219.1 mm, thickness 8.2 m
m, probe: 5Z10 × 10A45, skip number: 2.

First, FIG. 6A shows an example in which flaw detection was performed with the average number of times set to 1, and it is shown that echoes from a drill hole having an outer diameter of 3.2 mm were captured. It is also shown that noise is uniformly generated. Next, FIG. 6B is an example in which the average number of times is fixed at 8 and flaw detection is performed. Noise is about 1/3 of that of (a)
However, when the pulse density is large because the line speed is high, it can be seen that the defect echo is reduced.

FIG. 6C shows an embodiment of the present invention, wherein the average number is changed according to the line speed. For this reason,
Even if the line speed changes, the defect echo does not decrease, and it can be seen that the noise is optimally reduced.

The present invention is not limited to the detection of defects in a welded portion of a metal material, but can be applied to the detection of defects in a joint such as a composite material or a laminated steel plate. As the ultrasonic probe, a non-contact type such as an electromagnetic ultrasonic probe can be used, and a scanning type in which the subject is fixed and the probe is moved may be used. .

[0034]

As described above, according to the present invention, the number of times of averaging in the synchronous averaging process of the ultrasonic reflected wave signal is automatically and optimally set. Even if the conditions change, the average number can be set easily and quickly, and the highest SN ratio can be improved while preventing defects from being overlooked.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of an ultrasonic flaw detector according to an embodiment of the present invention.

FIG. 2 is a diagram illustrating an operation of a synchronous averaging processing device.

FIG. 3 is a diagram showing an operation of the effective beam width detecting device.

FIG. 4 is a diagram illustrating a relationship between an effective beam width and an optimum average number.

FIG. 5 is a diagram showing the relationship between the number of times of averaging and the signal intensity after averaging processing.

FIG. 6 is a diagram showing an example in which flaw detection is performed by changing the line speed.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Ultrasonic probe 12 Ultrasonic transmission / reception apparatus 13 Synchronous averaging processing apparatus 14 Defect determination apparatus 15 Pulse repetition frequency detection apparatus 16 Scanning speed detection apparatus 18 Pulse density calculation apparatus 19 Effective beam width detection apparatus 20 Average number calculation apparatus 100 Sample 101 Weldment 102 Defect

Claims (3)

(57) [Claims] 1. A method for scanning and flaw-detecting an object by an ultrasonic pulse reflection method and improving the SN ratio by synchronously averaging the detected reflected wave signals, wherein an effective beam width D is detected in advance. From the pulse repetition frequency and scanning speed, the pulse density Pd
An ultrasonic flaw detection method comprising: calculating synchronous average of the reflected wave signal at an average number N calculated by N = D / Pd. 2. When the effective beam width when scanning flaw detection is performed along a metal material inspection site is smaller than 4 mm, the average number N is obtained from N = D / Pd, and the effective beam width is 4 mm.
2. The method according to claim 1, wherein the average number of times N is obtained from N = 4 / Pd in a wider case. 3. A means for detecting a repetition frequency of a pulse transmitted to an ultrasonic probe; a means for detecting a relative scanning speed between the ultrasonic probe and a subject; From the detected pulse repetition frequency and scanning speed, a means for calculating a pulse density, a means for detecting an effective beam width of the ultrasonic probe, and a pulse calculated and detected by the latter two means, respectively. Means for calculating the average number of times from the density and the effective beam width, and setting the average number of times in the synchronous addition and average processing means of the reflected wave signal.