JPH0240186B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides an ultrasonic wave to a welded portion connecting members to be connected in reaction tubes for chemical plants such as reflow march tubes for hydrogen production, and tubes or steel plates for various other uses. This article relates to a method of flaw detection using a transmission method.

Conventionally, in the flaw detection method described above, as shown in FIG. Attached receiving probe 3b
The presence or absence of a defect in the welded portion 2 was determined from the attenuation state of the ultrasonic waves as they traverse the welded portion 2 in the direction of the tube axis P. However, the flaw detection accuracy was low due to the adverse effects of the macrostructure and surface roughness of the metal tubes 1a and 1b.

In order to eliminate the above conventional drawbacks, patent application No. 56-145118
No. (Special Publication No. 63-50661, Patent No. 1500674)
As shown in FIGS. 2 and 3, the tube 1
The transmitting probe 3a and the receiving probe 3b transmit ultrasonic waves in the direction of the tube axis P near the welds 2 of each of a and 1b, and transmit ultrasonic waves across the welds 2 as shown in FIG. The decibel value of the received ultrasonic wave is measured for each of the ultrasonic waves transmitted, and the difference between the arithmetic mean value of the transmitted decibel values for the pipes 1a and 1b and the transmitted decibel value for the welded part 2 is determined. A method for determining the presence or absence of defects in No. 2 has been proposed.

In the above newly proposed method, the macrostructure and surface roughness of the tubes 1a and 1b themselves can be eliminated by taking the deviation, and accurate flaw detection without the negative effects of the macrostructure and surface roughness of the tubes 1a and 1b is possible. However, it is only possible to detect whether or not there is a defect in the welded part 2, but it is not possible to quantitatively grasp the defective state of the welded part 2, and there is still room for improvement.

In view of the above circumstances, the present invention not only performs high-precision flaw detection without the adverse effects of macrostructures and surface roughness of connected members, but also quantitatively understands the defect status of welds and prevents leaks, etc. in welds. It is an object of the present invention to make it possible to prevent troubles reliably and satisfactorily, and to quantitatively grasp the defective state described above quickly and efficiently while being inexpensive and economical.

In order to achieve the above object, the present invention provides an ultrasonic flaw detection method for a welded part described above, in which ultrasonic waves are transmitted in the vicinity of the welded part of each of the connected members in a direction intersecting the longitudinal direction of the welded part. The decibel value of the received ultrasonic wave is continuously measured over the entire length or almost the entire length of the welded portion in the longitudinal direction for each ultrasonic wave transmitted across the welded portion in a direction intersecting the longitudinal direction of the welded portion, and The measurement results are recorded as graph information on a base paper, and after completing the measurement and recording of the entire front weld, the base paper is placed on a magnetic plate, and the graph is traced with a writing pen and transferred to the magnetic plate. The measurement results are input to an arithmetic processing unit using an A/D converter that reads the transferred information with a magnetic head, and the arithmetic mean of the transmitted decibel values T and B for the connected member at a specific position in the longitudinal direction of the welding part is calculated. Continuously determine the deviation between the transmission decibel value W and the transmission decibel value W for the welded portion by the arithmetic processing device over the longitudinal length or almost the entire length of the welded portion,
The defect state of the welded portion is determined based on the state of change of the deviation in the longitudinal direction of the welded portion.

In other words, the three types of decibel values T, B, and W are measured and recorded at the site, and after the measurement and recording are completed, the recorded base paper is collected, and a separate arithmetic processing device is installed at an appropriate location. The above-mentioned recorded information is input to calculate the arithmetic mean value, and from the deviation between the evaluation score obtained thereby and the decibel value for the weld, it is possible to identify defects along the entire longitudinal length or almost the entire length of the weld. It determines the state.

Therefore, similar to the new proposed method described above, high-precision flaw detection can be performed at each specific position in the longitudinal direction of the welded part without being adversely affected by the macrostructure or surface roughness of the members to be connected. is continuously determined from a large number of points over the entire length or almost the entire length of the weld, and the state of change in the deviation in the longitudinal direction of the weld, such as the width of each defect in the longitudinal direction of the weld, the distribution state of defects, etc. Since the defect state of the welded part is judged based on the defect status, it is possible to quantitatively understand the defect state, such as the degree of the defect, and re-inspection such as destructive inspection can be performed on specified parts to ensure that replacement is necessary or unnecessary. It has become possible to reliably improve safety and prevent troubles such as leaks from welds. Furthermore, in the present invention, since the arithmetic mean for the above-mentioned evaluation is performed using an arithmetic processing device, there is an effect that the defect state of the weld can be accurately determined in a short time. Moreover, in particular, the present invention uses an A/D converter composed of a magnetic plate, a writing pen, etc. as an input device, so conventional recording devices and recording media, such as raw paper, can be used as they are, and arithmetic processing devices can be used. It is possible to achieve the effect of inexpensively transitioning to analysis of the measurement results used. Furthermore, information input to the arithmetic processing unit,
This is done separately from the measurement and recording of the decibel values B, T, and W, so even when testing welds on a large number of connected members, there is no need to provide a calculation processing device for each. If a processing device is installed, transportation becomes troublesome during flaw detection, and a protective structure against humidity, temperature, or magnetism is required.In contrast, according to this proposal, a processing device can be installed at an appropriate location such as a control center. By being prepared, not only can inconveniences be avoided in terms of transportation, but the protective structure as described above is unnecessary, making it inexpensive and economical.

Examples of the method of the present invention will be described below.

Welding part 2 connecting the ends of the pipes 1a and 1b
2, the transmitting probe 3a and the receiving probe 3b are distributed and set on both sides as shown in FIGS. While transmitting sound waves, both probes 3a and 3b are integrally moved at a constant speed in the circumferential direction of the tube, and the flaw detection position changes in the circumferential direction and the maximum saturation decibel of the ultrasonic wave received by the receiving probe 3b is measured. Continuously measure the correlation with the change in value W (dB) as shown in Figure 5 A,
Automatically record on base paper 4.

A transmitting probe 3a and a receiving probe 3b are set on the upper tube 1a as shown in FIG. Both probes 3a and 3b are integrally moved at a constant speed in the tube circumferential direction, and the correlation between the change in the flaw detection position in the tube circumferential direction and the maximum saturation decibel value T (dB) of the received ultrasonic wave by the receiving probe 3b is determined. are measured continuously as shown in FIG. 5B and automatically recorded on the base paper 5.

A transmitting probe 3a and a receiving probe 3b are set on the lower tube 1b as shown in FIGS. 3 and 4, and ultrasonic waves are transmitted in the direction of the tube axis P in the vicinity of the welded part 2. While performing this, both probes 3a and 3b are integrally moved at a constant speed in the tube circumferential direction, and the flaw detection position in the tube circumferential direction and the maximum saturation decibel value B (dB ),
Measurements are taken continuously as shown in FIG. 5C and automatically recorded on the base paper 6.

Note that the above three types of flaw detection may be performed in any order or at the same time.

Collect the base papers 4, 5, and 6 that have been measured and recorded, and collect them at a suitable location such as the control center.
There, as shown in FIG. 6, the base paper 4,
5 and 6 on the magnetic plate 7, and tracing the recorded lines with the writing pen 8,
The information recorded on the base papers 4, 5, 6 is transferred to the magnetic plates 7... The magnetic plates 7 are configured so that different output voltages can be obtained in the width direction depending on the respective transfer positions during the transfer. There is.

The magnetic plate 7 after the transfer is placed on the feed roller 9 as shown in FIGS. 7 and 8, and while being moved by the feed roller 9, the digital signal is extracted by the magnetic head 10. It is input to the arithmetic processing unit 11.

In the arithmetic processing unit 11, the following formula is preprogrammed to obtain the evaluation point Xn from both decibel values T and B at specific positions in the circumferential direction of the tubes 1a and 1b.

Xn = Tn (dB) + Bn (dB) / 2-K n: Flaw detection position K: Constant (generally about 7.5) Then, from the three types of magnetic plates 7 based on the base paper 4, 5, 6... The recorded information is input to the arithmetic processing unit 11 to calculate the evaluation point Xn, and the evaluation line based on the evaluation point Xn and the decibel value W for the welded part 2 are output to the plotter 12.
A graph as shown in FIG. 9 is obtained.

Based on the above graph, it is determined that there is a defect in a location where the decibel value W is smaller than the evaluation line, and the defect location is determined. This involves quantitatively understanding the defect status, re-inspecting specified areas through destructive testing, determining whether or not the defect has reached the outer surface, and accurately determining whether or not replacement is necessary.

Incidentally, the plotter 12 may display the cross section of the welded portion 2 and also display the defect state thereon.

The present invention is applicable not only to the flaw detection of the welded part 2 between the ends of the pipes 1a and 1b, but also to the flaw detection of the welded part 2 between steel plates in a ship's hull, for example.
1b, steel plates, etc. are collectively referred to as connected members 1a, 1b.

[Brief explanation of drawings]

The drawings show an embodiment of the ultrasonic flaw detection method for a welded joint according to the present invention, and FIG. 1 is a schematic vertical cross-sectional view showing the flaw detection state for a welded joint, and FIGS. 2 and 3 are flaw detection for different pipes, respectively. Fig. 4 is a schematic perspective view showing the flaw detection state, A to C in Fig. 5 are graphs showing measurement data of decibel values W, T, and B, respectively, and Fig. 6 is the transfer state. , FIG. 7 is a plan view showing an overview of the defect determination device, FIG. 8 is a partially cutaway view taken along line A-A in FIG. 7, and FIG. 9 is a graph displayed on a plotter. It is. 1a, 1b...Connected member, 2...Welded part, 7
...Magnetic plate, 8...Writing pen, 10...Magnetic head, 11...Arithmetic processing unit.

Claims (1)

[Claims] 1 A method of detecting flaws in a welded part 2 in which connected members 1a and 1b are connected by ultrasonic transmission method, in which the welded part 2 is inspected in the longitudinal direction of the welded part 2 in the vicinity of the welded part 2 of each of the connected members 1a and 1b. The decibel value of the received ultrasonic wave is determined by setting the decibel value of the received ultrasonic wave to the entire longitudinal length of the welding part 2 or approximately to the longitudinal direction of the welding part 2. Continuously measure the entire length and record the measurement results as graph information on a base paper. After completing the measurement and recording of the entire welded part 2, the base paper is placed on the magnetic plate 7 and a writing pen 8 is used. After tracing the graph and transferring it to the magnetic plate 7,
A/D that reads the transferred information with the magnetic head 10
The measurement results are input to the processing unit 11 using a converter, and the arithmetic mean value of the transmitted decibel values T and B for the connected members 1a and 1b at a specific position in the longitudinal direction of the welding part 2 and the welding part 2 are calculated. The deviation from the transmitted decibel value W is continuously determined by the arithmetic processing unit 11 over the entire length or almost the entire length of the welded portion 2, and based on the state of change of the deviation in the longitudinal direction of the welded portion 2. An ultrasonic flaw detection method for a welded part characterized by determining the defect state of the front welded part 2.