JPH0246106B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to an ultrasonic flaw detection method for welded steel pipes. Welded steel pipes, such as electric resistance welded steel pipes, are generally known to have a welded part (hereinafter referred to as a seam) that is gently twisted due to the manufacturing method.Furthermore, the welded part is prone to cracks, penetrators, non-metallic inclusions, etc. Welding defects such as lamination occur. These welding defects significantly deteriorate welding reliability, and an effective method for detecting them is the ultrasonic pulse reflection method, which projects an ultrasonic beam and detects the ultrasonic waves that bounce back from the defect. It is. However, 1~ which is usually used in ultrasonic flaw detection method.
For 5MHz ultrasound, the effective ultrasound beam width is 5
It is normal that it is only about ~1 mm. Therefore, even if the seam is twisted, the relative positional relationship between the ultrasonic flaw detection probe and the steel pipe seam position, which is the material to be inspected, must not be disrupted. To solve this problem, there is a technique of arranging a large number of probes, but it is not uncommon for the tube to twist by about 1°/1 meter of tube length, and in order to detect flaws in a long tube with a tube length of over 20 meters, a large number of ultrasonic probes are required. is necessary and undesirable. The present invention advantageously solves the above-mentioned drawbacks, and its gist is to provide weld detectors at both ends of a steel pipe, detect the welds at both ends of the steel pipe while rotating the steel pipe, and collect the detection signal. This is an ultrasonic flaw detection method for welded steel pipes, which is characterized in that the torsion rate of the welded part is determined from the following equation (1) based on the torsion rate of the welded part, and then flaw detection is performed by following the twisted welded part based on the torsion rate of the welded part. Welded part torsion rate = πD×dp/PT/L (mm/m)
...Formula (1) Next, the present invention will be explained in detail with reference to FIGS. 1 to 7. The electric resistance welded steel pipe 1, which is the material to be inspected, is transported horizontally to the skid 2.
and sent to an ultrasonic flaw detection device. The ERW steel pipe 1 is mounted on ultrasonic flaw detection carts 3a, 3b (hereinafter
When the automatic seam detector carts 4a, 4a', 4b, 4b' (hereinafter referred to as seam carts) at both ends of the pipe are reached directly below the steel pipe turning device 5 disposed under the cross-feeding skid 2; The steel pipe 1 is raised by the roller 6 of the turning device 5.
The steel pipe 1 is rotated by lifting it up and rotating the roller 6. In synchronization with the lifting of the turning device 5, the seam carts 4a, 4a', 4b, 4b' and the UST carts 3a, 3b
Each sensor descends and connects with the pipe, and immediately begins preparatory operations for seam detection and ultrasonic flaw detection. Sensor 7 that automatically detects seams at both pipe ends,
As 7', we adopted an electromagnetic sensor that detects the seam by detecting abnormal propagation of electromagnetic ultrasonic waves (transverse waves) at the seam. Although the sensors 7 and 7' have a very small tolerance for variation in the distance between the detection end and the steel pipe (hereinafter simply referred to as gap), they utilize magnetic flux (electromagnetic attraction) to generate electromagnetic ultrasonic waves. By reducing gap fluctuations using the vertically movable magnetic poles 8 and contact bearings 9, 9', the detection accuracy is higher than those using already known eddy current sensors or ultrasonic wall thickness measurement methods. A pulse oscillator is directly connected to the motor driving the roller 6 of the turning device 5.
A steel pipe rotation pulse 10 is outputted every moment. On the other hand, the seam twist amount is determined from the pulse difference between the seam signals 11 and 11' obtained from the seam detection sensors 7 and 7' obtained from both pipe ends and the outer diameter and length of the steel pipe 1 using the following formula. Seam torsion rate = πD×dp/PT/L (mm/m)……(1) π: 3.1415 D: Steel pipe outer diameter (mm) L: Steel pipe length (m) PT: Number of pulses generated during one rotation of the steel pipe (k) dP: Number of pulse differences between the seam signals at both pipe ends (k) The seam signals obtained by the seam detectors 7 and 7' at both pipe ends are sent to the turning device 5, which controls the rotation of the steel pipe and seams the steel pipe. The rotation of the steel pipe is stopped so that the steel pipe is at a specified position (for example, the seam at either end of the pipe is directly above the seam). When the alignment is completed, the seam carts 4a, 4a', 4b, 4b' at both ends of the tube are temporarily retracted from both ends while flipping up the detection ends to both sides. In synchronization with this, the UST detection end mechanism section 12 and the UST carts 3a and 3b, which have completed preparations for flaw detection, perform supplementary flaw detection on the tube end once in order to reduce the tube end flaw detection zone. After completing the tube end capture, the UST carts 3a and 3b gradually start moving from one tube end to the other, and accelerate when the UST detection end mechanism 12 is completely on top of the material to be inspected. Once it reaches the maximum speed, it maintains that speed until it reaches just before the end of the tube, and then decelerates from just before the end of the tube. The above-mentioned UST carts 3a and 3b drive and control the torsion following motor 13 based on the seam torsion rate determined by equation (1) to follow the estimated seam line 14 connecting the seam positions of both pipe ends with a straight line. This estimated seam line 14 has an error of 1 to 2 mm at most compared to the actual seam line 15. Incidentally, large bends and warps in the steel pipe are tracked by the vertical and horizontal gimbal mechanisms that make up the UST detection end mechanism section 12. When the flaw detection is completed, the steel pipe turning device 5 descends, starts the traverse skid 2, and transfers the steel pipe. During this time, the UST carts 3a, 3b and both tube end seam carts 4a, 4a', 4b, 4b' return to the flaw detection preparation position and seam detection position, respectively. The flaw detection procedure for the next steel pipe is the same, but
UST carts 3a and 3b will be tested for flaws from the opposite direction from the previous steel pipe, but if there is enough cycle time, there is no problem in performing flaw detection from a fixed side. In addition, to explain in detail the seam detecting operation of the present invention, the seam carts 4a, 4a', 4
From b and 4b', the electromagnetic seam sensor section is lowered and connected to the pipe. After the sensor contacts the pipe, the rollers 6 of the turning device 5 are driven to start gently rotating the steel pipe. The seam detection sensor enters the seam detection state immediately after the pipe is connected, and outputs a seam detection pulse when the seam passes directly below it. If the steel pipe continues to rotate, a seam detection pulse will be output every time the steel pipe rotates, as shown in FIG. In order to accurately understand seam twist when the steel pipe is not rotating at a constant speed, roller 6
A pulse oscillator is directly connected to the motor that drives the. Although the pulse intervals are different during acceleration/deceleration and constant speed rotation, the number of pulses is proportional to the outer circumferential length of the outer surface of the steel pipe because the rollers 6 and the steel pipe are operated so as to hardly cause slippage. Therefore, if the number of pulse differences dp between both sensors (append a + sign based on one of them) is known, the amount of seam twist (mm) at both ends of the steel pipe can be determined from πD×dp/PT. If it is determined that the seam has passed directly under the seam detection sensor, the steel pipe is usually decelerated immediately and the seam position of the steel pipe is moved to a specified position (for example,
(Position where the seam of either steel pipe is directly above)
The roller 6 is controlled so that the rotation is stopped. Next, the operation of detecting flaws along the seam based on the seam torsion rate will be explained. UST trolley 3a,
3b and seam carts 4a, 4a', 4b, and 4b' are mounted on a portal frame, and the motor that drives the carts is equipped with a pulse oscillator in order to know the position in the tube axis direction. Seam trolley 4a, 4
The pulse oscillators a', 4b, and 4b' are reset to zero based on the limit switch signal of the specified standby position, so that the position of the truck can be known at all times. On the other hand, the UST carts 3a and 3b are blocks (hereinafter referred to as
It has a sensor at the end (simply called the UST saddle) that detects the end of the steel pipe. This tube end signal
The pulse oscillators of the UST trucks 3a, 3b are reset to zero so that it is possible to know exactly where the trucks are. In order for the UST carts 3a and 3b, which have the above-described position confirmation mechanism in the tube axis direction, to perform flaw detection along the seam twist, the following procedure is performed. First, the seam position is determined, and when the steel pipe slowly stops so that the seam is at the specified seam position, the UST trolley 3
The UST saddle comes down from a and 3b and connects to the steel pipe. Start pouring the flaw detection water and prepare for flaw detection. Next, gradually move through the UST saddle toward the end of the tube. During this movement, the seam following motor 13 is driven to match the amount of seam twist at the tube end.
Twist the UST saddle. Based on the signal from the tube end sensor, the signal position of the UST carts 3a, 3b is reset to zero, and is set as the origin of the position in the tube axis direction. The UST carts 3a and 3b then begin flaw detection, but since the pipe length L has been input into the UST cart control system in advance, high-speed flaw detection is performed at the center of the tube, and a pattern during acceleration and deceleration is adopted at the tube ends. However, even under such a speed pattern,
Since the amount of seam twist at the position xm from the above-mentioned origin can be easily determined by xτ from the seam twist rate τ (mm/m) described above, the seam of the UST carts 3a and 3b is determined based on the position signals from the UST carts 3a and 3b. If the tracking motor 13 is driven, the estimated seam line 1 in Fig. 5 is obtained by interpolating a straight line between the seam positions determined at both pipe ends.
4 enables ultrasonic flaw detection. The above example has been described with two sets each of the UST carts 3a, 3b and both pipe end seam detectors 4a, 4a', 4b, 4b', but there is no difference even if there is only one set each. Figure 7 shows a time analysis for each of the above flaw detection steps. Further, Table 1 shows the required time and operation contents in steps .about. in FIG. 7.

【table】

[Table] According to the present invention, conventionally, an operator marks the seam, aligns it using a remote ITV, etc., and visually follows the seam twist while the UST detection end or the material to be inspected is traveling. In some cases, the seam twist is measured at the beginning of the lot as disclosed in Japanese Patent Application Laid-Open No. 54-54691.
Compared to technology that applies this value to all parts in a lot and automatically follows it, it has become possible to perform ultrasonic flaw detection in and around welds with much higher productivity and reliability. By the way, in the present invention, an ultrasonic flaw detection probe is used in the sensor section, but even if sensors such as leakage magnetic flux flaw detection method, magnetic flaw detection method, eddy current flaw detection method, and optical flaw detection method are installed in the sensor section, high reliability can be achieved. , it is clear that productivity can be gained.

[Brief explanation of the drawing]

Fig. 1 is an explanatory diagram of an apparatus for carrying out the method of the present invention;
The figures are partially enlarged views of the turning device, Figures 3 and 4.
The figure is an enlarged view of the electromagnetic ultrasonic sensor, Figure 5 is a conceptual diagram of seam detection according to the present invention, Figure 6 is an explanatory diagram for determining the seam torsion rate according to the present invention, and Figure 7 is the ultrasonic flaw detection according to the present invention. It is an explanatory diagram of a cycle. 1: ERW steel pipe, 2: Cross feed skid, 3a, 3
b: Ultrasonic flaw detection trolley, 4a, 4a', 4b, 4b':
Seam truck, 5: Steel pipe turning device, 6: Roller, 7, 7': Sensor, 8: Vertical movable magnetic pole,
9, 9': contact bearing, 10: steel pipe rotation pulse, 11, 11': seam signal.

Claims (1)

[Scope of Claims] 1 Weld detectors are provided at both ends of the steel pipe, the welds at both ends of the steel pipe are detected while the steel pipe is rotated, and the torsion rate of the weld is determined from the following equation (1) based on the detection signal. 1. An ultrasonic flaw detection method for welded steel pipes, characterized in that flaw detection is then performed following a twisted weld based on the torsion rate of the weld. Welded part torsion rate = πD×dp/PT/L (mm/m)
...(1) Equation π: 3.1415 D: Steel pipe outer diameter (mm) L: Steel pipe length (m) PT: Number of pulses generated during one rotation of the steel pipe (ke) dp: Appearance pulse difference between welding signals at both ends of the pipe Number (ke)