JPH01219556A - Ultrasonic wave flow detecting method for weld zone - Google Patents

Abstract

PURPOSE:To enable accurate flaw detection and to shorten flaw detection time by setting a gate start point and a gate end point where only a defect echo can be extracted previously by using a the sound nondefective weld zone of a test material for gate setting. CONSTITUTION:A probe scanning device 20 is mounted on the test material 1 for gate setting and the probe 4a is scanned in parallel to the center line 2a of the weld zone 2 while a scanning reference line O at a proper distance from the nondefective weld zone 2 is regarded as the start point. This operation is repeated with respect to plural probe distances X. Simultaneously, a computer 24 inputs the signal outputted by a position reader 22 to calculate a distance X and a weld-line direction distance Y. An ultrasonic flaw detector 23, on the other hand, catches the echo from the weld zone 2 at constant pitch Y and inputs it to the computer 24. The computer 24 finds the mean value and standard deviation of data of a beam path W which are considered to be an echo group of the same kind for the same distance X. The gate start point WS and gate end point WE where only a defective echo can be detected are found from said values. The points WS and WE are set in the flaw detector 23 previously and the flaw detection of the weld zone of a material to be inspected is performed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention relates to an ultrasonic flaw detection method for welded parts such as butt-welded steel pipes.

[Prior art]

Ultrasonic flaw detection of welded joints such as butt-welded steel pipes is usually performed by angle angle flaw detection, and the detected echoes include bead echoes generated by reflection from beads and defect echoes generated by reflection from weld defects. Therefore, it is necessary to remove the bead echo and extract only the defective echo. The following are the main two conventionally used
We will discuss two extraction methods. FIG. 7 is a schematic diagram showing the first method, in which the probe 4a is placed on the aromatic material l for flaw detection near the welded part 2, and the shoulder part 3 on the lower side of the bead is When an ultrasonic beam is incident, it is reflected by the shoulder 3 and a bead echo is generated. Here, depending on the value of the horizontal distance (hereinafter referred to as probe distance) X between the center line of the welding part 2 and the beam incidence point of the probe 4a, the beam is The beam path W to the point where the beam is reflected by the bead is calculated in advance geometrically, and a gate is set to extract the monitoring signal so that the bead echo is not included in the detected echo according to the probe distance X. Perform flaw detection.

On the other hand, in the second method, both bead echoes and defective echoes are put into a gate and detected as detection echoes, the result is converted into an image, and bead echoes and defective echoes are discriminated based on the displayed position and shape. This method extracts only defective echoes.

[Problem to be solved by the invention]

In the first method, the reflection position of the bead echo changes due to a slight change in the shape of the weld 2, and even if the shape is the same, if the sound speed of the beam passing through the base material changes, the round trip propagation time of the beam changes. On the other hand, there is a problem in that the beam path W fluctuates and the setting of the monitor gate becomes uncertain. On the other hand, when testing welded parts of large-diameter steel pipes, even if formed to the same size, the strength of the base metal varies and the roundness will differ, so geometric calculations must be performed in advance assuming a perfect circle. There is a problem in that an error occurs between the beam path where the bead echo occurs and the actual beam path, making it difficult to accurately set the gate. In addition, as shown in Figure 8, an ultrasonic beam incident as a transverse wave hits a bead, converts the mode into a longitudinal wave, is reflected by the bead on the opposite side of the irradiation bead, and is then converted back into the original transverse wave mode and received. There is a problem in that an echo (hereinafter referred to as a mode conversion echo) may enter a set gate depending on the bead shape and be determined as a defective echo.

On the other hand, in the second method, the operator can distinguish between bead echoes and defect echoes from the images of detected echoes obtained, but if the operator is not skilled, it takes a long time for flaw detection, and There was a problem of erroneous judgments being made.

The present invention has been made in view of the above circumstances, and its purpose is to accurately set a gate in advance by injecting ultrasonic waves into a healthy weld, and to remove the weld within the gate. It is an object of the present invention to provide an ultrasonic flaw detection method for a welded part that accurately extracts only defect echoes by performing flaw detection of defects, shortens flaw detection time, and eliminates misjudgment.

[Means to solve the problem]

The ultrasonic flaw detection method for welds according to the present invention involves injecting ultrasonic waves into a weld formed by butt welding, applying a gate to the reflected signal, and extracting a required signal to detect flaws in the weld. In this method, ultrasonic waves are applied to a substantially defect-free weld that has substantially the same bead shape and base material as the weld to be inspected, and the reflected signals identify the ultrasonic incident point and the weld bead. In order to detect the entire volume of the weld, measure the path between the ultrasonic reflection point and the weld, and calculate the average value and standard deviation of the multiple measured values of the path with the same separation distance between the probe and the weld. The method is characterized in that calculations are made for each of different separation distances in a necessary range, and a gate is set using a predetermined calculation formula based on the relationship between these and the separation distance.

[Effect]

In the method of the present invention, a monitor gate is set in the flaw detection device by injecting an ultrasonic beam into a defect-free welded part of the same shape and the same base material in advance, and the inspection target for flaw detection is set within the monitor gate. Perform ultrasonic flaw detection.

〔Example〕

Hereinafter, the method of the present invention will be specifically explained based on drawings showing examples thereof. FIG. 1 is a perspective view of a probe scanning device 20 for ultrasonic flaw detection used in carrying out the method of the present invention.
legs attached to its four corners 15.15.15.15
At the lower end of the pedestal 15a+15a+15a, 15
A is provided. In addition, each of the mounts 13 has opposing long sides 2
0a, 20b and short sides 20c, 20d, and the short sides 20c, 20d are arranged parallel to the center line 2a with the welding part 2 in the center. Long side 20a, 20
A guide rod 9 and a threaded rod 10 (only one of which is visible) are provided inside b.

The guide rod 9 and the threaded rod 10 pass through and support the sliders 7, 7, and the slider 7.7 has a long side 20a as the threaded rod 10 is rotated by a drive device such as a pulse motor (not shown). , 20b, respectively, while sliding against the guide rod 9 and spiraling against the threaded rod 10. Also, the short sides 20c, 20d
A guide rod 11 and a threaded rod 12 (only one of which is visible) are provided inside. The guide rod 11 and the threaded rod 12 pass through and support the sliders 8, 8, and the sliders 8, 8 rotate the short side 20c as the threaded rod 12 is rotated by a drive device such as a pulse motor (not shown). , 20
It is arranged to move along the inside of d while sliding with respect to the guide rod 11 and screwing with respect to the screw stud 12, respectively.

A guide rod 5.5 is installed in parallel between the opposing surfaces of the slider 7.7, and a guide rod 6.6 is installed in parallel between the opposing surfaces of the slider 8.8. Both guide lofts are movably inserted into the block-shaped probe scanning block 40.

Further, a digital scale (not shown) is attached to the long side 20a and the short side 20c in the longitudinal direction thereof, and the head thereof is provided on the movable member 7.8. - On the other hand, a shaft 40a is provided which passes through the probe scanning section 40 in a vertically movable manner, and the probe holder 4 is connected to the lower end of the shaft 40a.

Further, the shaft 40a is urged downward by a pressure applying device (not shown). The probe holder 4 is fixed by a connecting jig 30 connected to a shaft 40a and a pin 31, and is rotatable so that the probe holder 4 follows a specimen having a curvature like a large diameter tube. A roller 14 is attached to the probe holder 4 to smoothly move the probe 4a and the metal material 1 to be inspected for flaw detection.

FIG. 2 is a schematic block diagram showing the configuration of the entire device when carrying out the method of the present invention using the probe scanning device 20 described above. The probe scanning device controller 21 drives the movers 7, 7 and 8, 8 of the probe scanning device 20 by a pulse motor (not shown). Thereby, the position information of the probe 4a moving on the flavor material 1 to be tested for flaw detection can be obtained from the mover 7,
The signal is inputted to the position reader 22 as an output from the head of the digital scale attached to the position sensor 8. This is output to the computer 24 as probe distance data. On the other hand, probe 4a
A flaw detector 23 connected to the probe 4a makes an ultrasonic wave/beam incident on the welding part 2, detects an echo generated thereby, and inputs the detected echo to a computer 24.

The method of the present invention is carried out as follows using the apparatus having the above configuration.

First, ultrasonic waves are applied to a healthy weld of a test material for gate setting that has the same material, wall thickness, and weld shape as the aromatic material to be tested for flaw detection, such as a large-diameter steel pipe that has a weld that is to be ultrasonically detected. The beam is made incident on the probe holder 4 while scanning it as shown by the two-dot chain line in FIG. 1, and the generated echo is detected. That is, the probe scanning device 20 is placed on the gate setting test material in parallel with the welded part 2. Then, as shown in FIG. 3, the probe 4 is moved parallel to the center line 2a of the defect-free weld 2 with the scanning reference point O, which is an appropriate distance away from the defect-free weld 2, as the starting point.
Scan a. This is repeated for multiple probe distances X. During this time, the computer 24 receives the signal output from the position reader 22 and calculates the probe distance X and the weld line direction distance Y.

On the other hand, the ultrasonic flaw detector 23 excites the probe 4a, captures the echoes from the welded part 2 at every fixed pitch Y, and inputs the echoes to the computer 24. The computer 24 calculates the beam path length W from this. After the scanning is completed, the computer 24 performs statistical calculations using the beam path length W read for the same X, but considering the spread of the ultrasonic beam, there are bead echoes that hit the lower side of the bead and are reflected even for the same X. Three types of echo groups can be detected: a bead echo that hits the upper side of the bead and is reflected (hereinafter referred to as an external bead echo), and the mode conversion echo mentioned above. The average value and standard deviation of a plurality of beam path length W data that can be considered as a group are determined.

The processing procedure will be explained below according to the flowchart shown in FIG. First, initialize the beam path separation distance δW for grouping the beam paths into three groups at most.
Variables M, N, P used for data counting within a group
Set 0 to 0 (Step 1) Set the 0th W data to cent.Step 2), if it is the first data (Step 3), set W up to the (M+1)th of the 6th group of the 1st group.
Average value W□1° Average value W of up to (N+1)th W in the second group 812 + Average value W of up to (P+1)th W in the third group, 3. Set the first W in (step 4), and set the X for the second and subsequent data.
is the same (step 5), and then l
Compare W - W - Il and δW (Step 6), and if l is smaller than δW, amplify the count number M of the first group (Step 7)
, find the average value W1 up to the (M+1)th group of the first group (step 8), and if IW-W,..., l is greater than or equal to δW, this W is data belonging to the second group or the third group. If it is the first data of the second group (step 9), the initial value of W,2 is set again (step 10).

Next, Iw-w. 1 and δW, 1w−w
, tz l is smaller than δW (step 11), increase the count number N of the second group (step 12)
, the average value W1□ of the second group up to (N+1) is determined (step 13). Also, IW-W! □ If 1 is greater than or equal to δW, this W belongs to the third group, and if it is the first data of the third group (step 14), the initial value of W, 3° is set again (
Step 15).

Next, the count number P of the third group is amplified (step 16), and the average value W of the third group up to the (P+1)th
, 3. (Step 17). After passing through one of the average value calculation flows for each group described above, return to the set of data W for fire (step 2).Next, loop as long as X is the same in the same flow, When different Xs are read, the average value W1 and standard deviation W,l of W for the entire first group are calculated (step 18), and the average value W, ! of W for the entire second group is calculated. and the standard deviation W, 2 (step 19), the average value of W for the entire third group, W, 3
and the standard deviation W, , are calculated (step 20).

After that, return to the initial settings and perform the above-mentioned processing for the different X. The beam path separation distance δW used in performing the above-mentioned calculation is within the tolerance range for separating the beam path into three groups. .

When performing ultrasonic flaw detection on a welded part of a thick-walled steel pipe, the group of beam path W that can be detected for the same probe distance In many cases, there are only processes corresponding to the second and third groups in the flowchart shown in Figure 4.
Step 9) - (Step 17), (Step 19)
, (step 20) are often not performed, but when performing ultrasonic flaw detection on a welded part of a thin-walled steel pipe, the group of beam path W that can be detected for the same probe preload MX is It is quite common for three groups corresponding to the three types of echoes to exist simultaneously, so the fourth
All the processes in the flowchart shown in the figure will be performed.

Mean value W of W from the first group to the third group obtained in this way, , W, , W, , Standard deviation W, ,
From the actual measured values of W, tW, and X, a gate starting point W and a gate ending point W at which only defective echoes can be detected are determined. First, for the same X, the average value W ai and W,
5 (where i = 1.2.3), and the relationship between the X of each bead echo and the i-th Wsi can be regarded as continuous, and the regression line approximation is performed between the minimum and maximum values of X in the same group. Then, the following formula (1) is obtained.

W@! = AAX + B t (t = L2
．． 3)...(1) A5, B5: constants Here, for the same X, W4 to W1. A maximum of three regression lines can be detected, and if we select the one with the minimum W and i from them, and then let i be k and the gate end point W, Wt is expressed by the following equation (2). It will be done.

WE = A@X + Bk −nWa e II”
'(2) However, W, , is Wow (k=1.2.3
), and n is preferably 3 based on experience.

A large-diameter steel pipe with an outer diameter of 56'' and a wall thickness of 18 mm was used as a testing material for flaw detection, and the values of W were measured with respect to a sound welded portion while varying X. The results are shown in FIG.

In the figure, the dots are actually measured values, which were divided into three groups depending on the range of X. In descending order of W value, inner bead echo, mode conversion echo, and outer bead echo are respectively supported.0 In this example, since the pipe is a thick-walled steel pipe,
Only one regression line obtained for the same
).

Wl=AIX+Bl 3W@g+...
・(3) If the regression lines obtained corresponding to X are not continuous as in this example, the end points corresponding to the maximum value of X of one regression line are respectively By connecting the end point to the starting point of the regression line corresponding to the high level of X with a straight line, the gate end point W
I'm looking for something.

X k= X kllia + 1.3(X ha□−
X1aifi) ... (4) Xks□: Maximum value of X of the regression line detected in th X11a: Minimum value of X in the regression line detected in th: 1 in this example. 1.3 is a value determined from experience.

Furthermore, when the straight line extended by equation (4) overlaps the regression line corresponding to a high level of X for the same X, the regression line is adopted.

On the other hand, the gate starting point W is determined by the following equation (5).

Here, θ is the actual refraction angle of the beam in the base material as shown in FIG. 6, and B. is the width of the bead. These values are entered in advance.

Gate starting point W obtained as described above.

As shown in FIG. 6, is the distance to the vicinity where the beam starts to enter the welding part, and the gate end point W is the distance slightly before the beam irradiates the bead shoulder 3. It goes without saying that the gate starting point W needs to be set in a range where the welded part can be sufficiently detected and is not affected by reverberant echoes within the probe.

Gate starting point W and gate ending point W obtained in this way
, is set in advance on the ultrasonic flaw detector, and the ultrasonic waves are applied to the welded part of the flaw detection inspection material and scanned as shown in Figure 3.
According to the value of , a required signal is extracted while setting a gate between a gate starting point W and a gate ending point W for the reflected signal, and the welded portion of the inspection material for flaw detection is detected.

The large-diameter steel pipe used in this example with an outer diameter of 56" and a wall thickness of 18 m was used as a test material, and when ultrasonic flaw detection was performed on 72 defective parts of 50 welds, no bead echo was detected within the gate. , only defective echoes were detected.

〔effect〕

As detailed above, in the method of the present invention, a healthy defect-free weld of a test material for gate setting having the same material, wall thickness, and weld shape as the inspection material for flaw detection is used, and only defect echoes are detected in advance. Since the gate start point and gate end point are set to extract the flaws, bead echoes are reliably removed when detecting weld defects on the test object, and only the defect echoes are extracted, allowing accurate flaw detection. In addition, since gates can be set using a computer, the flaw detection time can be greatly shortened, and erroneous judgments can be eliminated.

[Brief explanation of the drawing]

Fig. 1 is a three-dimensional view of a probe scanning device according to the present invention, Fig. 2 is a schematic block diagram showing the configuration of the entire device according to the present invention, and Fig. 3 is a probe scanning device for a sound welded part of a test material. FIG. 4 is a flowchart showing the process of determining the detected echo for gate setting and determining the average value and standard deviation of the beam path, and FIG. 5 is a schematic diagram showing the scanning direction of the beam path. A graph showing the results of gate setting, Fig. 6 is a schematic diagram specifically showing the gate start point and gate end point, Fig. 7 is a schematic diagram explaining the conventional flaw detection method, and Fig. 8 shows the generation of mode conversion echo. It is a schematic diagram for explaining. 1... Test object for flaw detection 2... Bead 4a...
Probe Holder Applicant Sumitomo Metal Industries Co., Ltd. Company Representative Patent Attorney Norio Kono Figure 2 Figure 5 Figure 6 Shoulder Figure 7 Figure 8 Procedural Amendment (Voluntary) March 1988 1st 2, Name of the invention Ultrasonic flaw detection method for welded parts 3, Relationship with the case of the person making the amendment Address of patent applicant 5-15 Kitahama, Higashi-ku, Osaka Name (21)
1) Sumitomo Metal Industries Co., Ltd. Representative Yasuo Shingu 4, Agent address: 207-6 Nisshin Building, 1-14-22 Shitennoji, Tennoji-ku, Osaka 543, page 18, line 16 of the statement of contents of the amendment "Sumitomo Metal Industries, Ltd." is corrected to "Sumitomo Metal Industries, Ltd."

Claims (1)

[Claims] 1. In a method of detecting welds by injecting ultrasonic waves into a weld formed by butt welding and applying a gate to the reflected signal to extract the required signal, the weld to be inspected and the bead shape are detected. Ultrasonic waves are applied to a welded part with substantially the same base material and substantially defect-free, and the reflected signal creates a gap between the ultrasonic incidence point and the ultrasonic reflection point in the defect-free welded part. The distance between the probe and the defect-free weld zone is determined by measuring the average value and standard deviation of multiple measured values of the distance between the probe and the defect-free weld zone. An ultrasonic flaw detection method for a welded part, characterized in that each distance is calculated, and a gate is set using a predetermined calculation formula based on the relationship between these distances and the separation distance.