JP2511020B2 - Pipe defect inspection device - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a crack detection technique for detecting the shape of a crack that has occurred in a metal structural member, and particularly for accurately detecting the shape of a surface crack on the inner surface of a pipe. It relates to a suitable device.

[Conventional technology]

As a conventional crack detection method, there is an ultrasonic flaw detection method. There are various ultrasonic flaw detection methods, including the edge peak echo method, aperture synthesis method, and holography method. Although each of these methods has its own features, it may not be possible to obtain an echo from the crack tip, which is particularly important for crack detection, in which case there is the problem that the shape of the crack cannot be determined, and in particular, the progressive Regarding cracks (for example, fatigue cracks), there is a drawback that the growth state of cracks cannot be grasped.

Further, regarding the crack shape detection by the potential method related to the present invention, Advances in Crack Growth Measurement (1981) pp. 159 to 174 (Advances in Crack Growth Measurement, (1981) pp.
159-174). According to this document, the potential difference distribution is measured along the surface crack placed in the center of the flat plate-shaped test piece, and the correspondence between the potential difference distribution and the surface crack shape is investigated, but the measuring device is a flat plate-shaped test piece at the measuring terminal. The surface can be scanned, and the crack shape detection from the potential difference distribution remains qualitative.

[Problems to be solved by the invention]

The above-mentioned conventional technique has a physical problem that a reflection echo from the vicinity of the crack tip cannot be obtained in the case of the ultrasonic method, and in the case of the potential method, it is not possible to accurately grasp the turbulence of the peculiar electric field generated around the crack. In addition, there is a problem that the crack shape cannot be detected accurately. The object of the present invention is to scan the power supply terminal and the measuring terminal along the inner surface of the pipe in order to detect the shape of the surface crack generated on the inner surface of the pipe by the potential method, and especially to measure the potential difference distribution by allowing vertical pipes and bent pipes to run. However, it is another object of the present invention to provide an apparatus capable of determining a crack shape by analyzing the measured potential difference distribution by an original method.

[Means for solving problems]

As a result of investigating the detection of surface cracks occurring in structural members by the DC potentiometric method, the potential difference distribution measured around the cracks and the electric field analysis by the FEM of the cracked member corresponded, or a standard crack that was analyzed in advance. The surface crack shape can be judged with high accuracy by a simple surface crack shape judgment method based on the electric field analysis result of the member by FEM. Therefore, the above object can be achieved by creating an apparatus capable of measuring the potential difference distribution around the surface crack on the inner surface of the pipe. However, there are vertical pipes and bent pipes in actual equipment, and it is necessary to devise a scanning device that can pass through these pipes, and cracks occur not only in the axial direction of the pipe but also in the circumferential direction. Since it may occur with an inclination in both directions, it is necessary to create a measurement head capable of simultaneously measuring the potential difference distribution in both directions. Therefore, we created a mechanism that can be fixed to the inner surface of the pipe by itself and movable in the axial direction as well, and devised the arrangement of the power supply terminal and the measurement terminal to provide a measurement head that can simultaneously measure the potential difference distribution in the axial and circumferential directions. By creating it, it was possible to achieve the above purpose.

[Action]

The gist of the present invention configured as described above is not to scan the inner wall surface of the pipe in one direction at one point, but at least in the circumferential direction and the axial direction (the circumferential direction and the axial center direction of the pipe to be inspected). Scanning) to obtain a two-dimensional defect image. Thereby, the crack on the surface of the pipe can be determined with high accuracy.

〔Example〕

An embodiment of the present invention will be described below. FIG. 1 is a left side view of a pipe defect inspection apparatus which is an embodiment of the present invention, and FIG.
Similarly, FIG. 3 is a rear front view of the pipe defect inspection apparatus, and FIG. 3 is an external view of an embodiment different from the above. (See FIG. 1) The main body of the pipe defect inspection apparatus is divided into a front drive section 30 and a rear drive section 40, which are connected to an axial drive air cylinder 20 and a large bellows universal joint 33. The front drive unit 30 and the rear drive unit 40 respectively include two blocks, a front block and a rear block.
It is divided into 31,31 ', and 41,41'. 2 front drive
The individual blocks 31, 31 'are connected by a thin-walled cylindrical member 32, and the two blocks 41, 41' of the rear drive unit are connected by a central shaft 42 which also serves as a guide surface of a measuring head scanning unit described later. As shown in FIG. 2, each block is provided with four double-acting air cylinders 18 and 22 in radial directions at 90 ° intervals. (Four air cylinders 22 are shown in FIG. 2). Legs 33 in which hard rubber 34 having a large friction coefficient is fitted are attached to the shaft ends of the air cylinders 18 and 22.

The rear drive unit 40 is provided with a measurement head scanning unit 43 for scanning the measurement head 60 (FIG. 1), and is movable in the axial direction with the hollow cylindrical central shaft 42 as a guide surface. Here, the central shaft 42 is a linear guide with a semi-circular groove formed therein, and linear bearings 44 and 44 'are attached to the measurement head scanning unit 43, so it is possible to move in the axial direction even with a very small driving force. Is. A ball screw bearing 45 and a ball screw 46 are provided at one end of the measurement head scanning unit 43. The ball screw 46 passes through a bearing provided on the rear block 41 ′ of the rear drive unit 40,
A worm receiving gear 47 is attached to one end thereof. A worm original gear 48 is attached to the shaft tip of the reduction gear of the DC servo motor 8 for driving the measurement head axial direction attached to the rear block 41 'so as to mesh with the worm receiving gear 47.

The rotation of the measurement head scanning unit 43 in the circumferential direction is suppressed by the engagement between the central shaft 42 and the linear bearings 44, 44 ', and the ball screws 45, 46 for driving the axial direction. A linear bearing 50 is provided at one end of the measurement head scanning unit 43, and a shaft 49 is provided at the rear block 41 'to prevent backlash.

The measurement head scanning unit 43 has a circumferential rotating unit having a gear 52.
51 is provided via a bearing, and can be driven in the circumferential direction by the measurement head circumferential direction driving DC servomotor 12 attached to the measurement head scanning unit 43. The circumferential rotating part 51 is provided with a measuring head driving air cylinder 16, and a measuring head 60 made of an electrically non-conductive material is attached to the shaft end thereof.

It should be noted that in order to detect the axial position inside the rear drive 40 of the measuring head 60, the encoder 10 also
Although not shown in the figure, a limit switch 11 (FIG. 3) for preventing runaway of the measurement head is provided in the measurement head scanning unit 43.
An L-shaped lever for driving the limit switch is attached to the rear block 41 '. Further, the limit switch 15 (not shown, which will be described later in FIG. 4) for confirming the origin position is also used for detecting the circumferential position of the measuring head 60.
And an encoder 14 for detecting the rotation angle (which will be described later with reference to FIG. 4). In FIG. 2, provided on both sides of the air cylinder 22 are guides for centering the main body of the pipe defect inspection apparatus. The inside of the member 75 has a hollow cylindrical shape, and its side wall serves as a guide surface for the cylinder.
There are 22 axially movable members 76 inserted. Member 76
A ball plunger 77 having a rotating front end is attached to the front end thereof, and a coil spring for pushing out the member 76 is inserted into the rear end thereof. If the spring constant of the coil spring is set appropriately, the center of the inspection device main body can be made substantially coincident with the center of the pipe P. In this case, the tip of the ball plunger 77 is pressed against the inner surface of the pipe P and is released, but since the ball rotates, it can move smoothly without frictional resistance.

The centering guide as shown in FIG. 2 requires less driving force because the frictional resistance is less when the inner surface of the pipe is relatively smooth. For example, in the case of a poorly shaped weld, the diameter of the ball is small and the weld metal is small. It may not be possible to exceed the (bead). FIG. 3 shows an embodiment in which a warp-shaped leg 78 is used in place of the ball plunger so that even if such unevenness is somewhat large, it can be overcome. Since both ends of the leg 78 are shaped like sleds, the protrusions that are lower than the rising height of the sled can be overcome. At this time, since the legs must not rotate, a vertical groove may be dug in the side surface of the member 76, and a pin may be inserted into the groove from the side surface of the member 75 to lock the rotation.

The axial position of the pipe defect inspection device can be detected from the amount of movement, but there is a risk that the position will not be known if slippage occurs between the pipe and the leg, or if the device goes down during movement. Therefore, as shown in FIG. 2, an air cylinder is formed on the rear surface of the rear block 41 'of the rear drive unit 40.
24 was attached, and an AE oscillator 26 was provided at the end of the shaft. To detect the position, the acoustic oscillation element 26 is pressed against the inner surface of the pipe to oscillate sound (AE), and the two AE oscillation elements attached to the member at a certain interval are pressed against the outer surface of the pipe to determine the time difference between the vibration The axial position of the pipe defect inspection device is detected from the difference in vibration receiving time.

Fig. 4 shows a system diagram of the control, drive, and measurement system of the pipe inspection device. Reference numeral 1 is a computer incorporating an external storage device such as a hard disk for storing data and programs. Computer 1 is Interface 4 or GP
-Various drive devices and solenoid valves via the IB interface 5,
It controls measuring instruments, captures measured values, processes them, and outputs the results. As shown in Fig. 1, the piping inspection device is DC
It is driven by a servomotor or an air cylinder. The entire pipe inspection device is axially driven by a front drive cylinder 18, an axial drive cylinder 20, and a rear drive cylinder 22. These air cylinders are solenoid valves 19 and 2, respectively.
Connected to the compressed air source 27 through 1,23, solenoid valve 19,21,23
Are controlled by the computer 1 via the interface 4. Similarly, the cylinder for the measurement head drive 16 and the cylinder for the AE sensor for position detection of the pipe inspection device are solenoid valves 17,2.
The solenoid valves 17 and 25 are connected to the compressed air source 27 through 5 and controlled by the computer 1 via the interface 4. DC for driving the measurement head 60 in the axial direction
Servo motor 8, DC servo motor for measuring head circumferential drive
Power is supplied from DC motor drive devices 9 and 13, respectively, and a signal for driving 12 is output from the computer 1 via the interface 4.

The direct current from the direct current power source 2 provided in the same number as the supply electrode property is controlled by the computer.
Thus, the polarity is switched at regular time intervals and is supplied to each power supply terminal 65. The potential difference between a large number of measurement terminals 67 (described later with reference to FIG. 5) is measured by connecting to the minute potentiometer 6 by switching the measurement terminals to be measured by the multiplexer 7. The measured potential difference is transferred to the computer 1 via the GP-IB interface 5. The computer 1 determines the size of the crack from the potential difference distribution in the axial and circumferential directions of the pipe by the method described later. Here, the multiplexer 7 and the minute potentiometer 6 for switching the potential difference measurement terminals are controlled by the computer 1 via the interface 4 or the GP-IB interface 5.

5 to 9 show an example of the structure and terminal shape in the vicinity of the measuring head 60 shown in FIG. As described above, the electrical head limited head 60 is attached to the shaft end of the measurement head driving air cylinder 16 provided in the circumferential rotation portion 51 of the measurement head scanning portion 43. However, in this state, the limited head freely rotates about the axis of the air cylinder 16.
Therefore, as shown in FIG. 5, two cylindrical members 61 are provided on both sides in the circumferential direction of the air cylinder in the circumferential rotation unit 51, and the linear guides 62 are inserted inside the cylindrical members 61, whereby the radial direction is freely set. The structure is such that it can move around, but cannot rotate around the axis of the air cylinder 16.

The internal structure of the measuring head 60 is shown in FIG. Basically, two power supply terminals 65, 65 'for DC current supply are arranged at both ends, and measuring terminals 67, 67' for measuring the potential difference are located near the center of the terminals.
Are arranged in a line. A step is provided in the middle of the cylindrical terminals 65, 65 ', 67, 67' to insert a washer-shaped spring seat, and the measurement head
A coil spring 70 is compression-inserted between 60 and the spring seat to push the terminal out of the measuring head. When the pressing force of the terminals is weak in measuring the potential difference, contact resistance occurs between the measuring terminals 67 and 67 'and the material to be measured, and the potential difference of the material to be measured itself cannot be accurately measured. For this reason, it is desirable to use a coil spring having a spring constant of approximately 100 g / mm or more in order to secure a sufficient pressing force.

FIG. 7 shows the shape of the terminal. Usually, the shape of the terminal is cylindrical, the tip is conical, and the rear end is threaded to fix the lead wire. However, as described above, a step is provided to receive the spring seat for supporting the coil spring that applies the pressing force. The terminals having the shape shown in FIG. 7 may be arranged on a straight line as shown in FIG. 6, but the distance between the measuring terminals is inevitably long. Therefore, as shown in FIG. 8 (front view) and FIG. 9 (side view), the conical tip of the measuring terminal is bent, and a part of the columnar portion is shaved to provide a flat surface. (See FIG. 6) The measuring head 60 is provided with a guide surface 68 for guiding the cylindrical portion and a guide surface 69 for guiding the flat surface, so that the measuring terminal 67 can freely move in the vertical direction in FIG. Move, but not rotate. With this structure, the spacing between the cylindrical parts of the measuring terminals is 10
Even if it is about mm, the distance between the tips of the measuring terminals can be about 2 mm. The reason why the distance between the measuring terminals is shortened is to improve the detection accuracy of the surface crack shape described later.

FIG. 10 shows an example of the terminal arrangement in the measurement head 60. Plural sets of power supply terminals 65 and 65 'are arranged at both ends of the measurement head, and plural sets of measurement terminals 67 and 67' are arranged at the center thereof. First
In the example of FIG. 0, the number of power supply terminal pairs is 5 and the number of measurement terminal pairs is 4, but any number may be used. Power supply terminal 65, 65 '
The reason for arranging a plurality of sets is to make the electric field uniform.

Therefore, it is necessary to arrange the measuring terminals 67, 67 'at a uniform location of the electric field (specifically, facing the portion that produces a uniform electric field when the test object is not defective).
It must be located at least inside the pair of power supply terminals. In FIG. 6, the four pairs of measuring terminals 67 and 67 'are arranged at the center of the five pairs of opposing feeding terminals 65 and 65'.

There are many circumferential cracks on the inner surface of the pipe. Therefore, as a method of attaching the measuring head 60 shown in FIG. 10 to the measuring head scanning unit 43, as shown in FIG. 1, the power supply terminals 65 and 65 'and the measuring terminals 67 and 67' are to be inspected. The pipes may be attached so as to be aligned in the axial direction of the pipe. However, rarely cracks also occur in the axial direction. When trying to detect cracks in both the circumferential and axial directions, for example, after first detecting the circumferential cracks, take out the inspection device once from the pipe and change the direction of the measurement head by 90 °, Cracks must be detected. This method increases the inspection time. When it is desired to detect cracks in the circumferential direction and the axial direction at the same time, it is preferable to use a measuring head having a structure as shown in FIG. Feeding terminals 65, 65 'and measuring terminal 6 for detecting circumferential cracks
The arrangement with 7,67 ′ is the same as that in FIG. 10, but in addition to the above, for the axial crack detection, a pair of feeding terminals 72,72 ′, and the feeding terminals 65,65 ′ for the circumferential crack detection are provided. The measurement head 60 is provided so as to face the circumferential direction of the measurement head 60 so as to be orthogonal to. In this case, the measurement terminal also serves as the measurement terminal for detecting circumferential cracks, and the circumferential potential difference distribution is measured using terminals 67-2 and 37-3, and 67'-2 and 67'-3. I shall.

FIG. 12 shows a measurement head of an embodiment different from the above in FIG. At the four corners of the measuring head 60, axial feeding terminals 65, 65 'and circumferential feeding terminals 72, 72' are provided as in FIG. Inside the pair of power supply terminals, the measuring terminals 67 and 67 'for measuring the axial potential difference distribution and the measuring terminals 74 and 74' for measuring the circumferential potential difference distribution are arranged so that the axial distance and the circumferential distance are equal. To place. In measuring the potential difference, there are two methods, one is to measure the circumferential potential difference distribution in the entire symmetric range and then again to measure the whole axial potential difference distribution, and the other is to measure the circumferential potential difference distribution and the axial potential difference distribution simultaneously. There is a method.

The reason for providing the axial power feeding terminals 65, 65 'and the circumferential power feeding terminals 72, 72' and flowing the current in two directions of the axial direction and the circumferential direction is as follows. Now, if the crack is in the axial direction of the pipe, the electric field is not disturbed by the crack because the electric field is in the axial direction even if a current is passed in the axial direction, so the measured potential difference distribution is It is exactly the same as when there is no, and it is judged that there is no crack. However, when an electric current is applied to the axial crack of such a pipe in the circumferential direction, the circumferential electric field is greatly disturbed by the crack, resulting in an irregular potential difference distribution. The size of the crack can be determined. Similarly, when an electric current is applied axially to a crack in the circumferential direction of the pipe, the axial electric field is greatly disturbed by the crack, resulting in an irregular potential difference distribution. The size can be determined. If a crack is generated with an inclination in both the axial direction and the circumferential direction of the pipe, determine the shape of the crack from the potential difference distribution measured by applying current from each direction and including the inclination. Is possible.

FIG. 13 shows another embodiment of the measuring head. Measuring head
In the four corners of 60, as in Fig. 11, axial feed terminals 65, 6
5'and the circumferential direction power supply terminals 72, 72 'are provided. Inside the pair of power supply terminals, the measurement terminals 67 are arranged in a matrix so that the intervals in the axial direction and the intervals in the circumferential direction are equal.
In the example of FIG. 13, four measuring terminals are arranged in the axial direction and four measuring terminals in the circumferential direction. Therefore, when the measuring head 60 is pressed against the inner surface of the pipe once by the measuring head driving air cylinder 16, the measurement of the axial potential difference distribution is performed. In this case, the potential difference distribution can be measured at a total of 12 locations, 3 locations in the axial direction and 4 locations in the circumferential direction, and the measurement time can be greatly shortened.

Next, a pipe inspection method will be described. Figure 14 shows the overall flow chart for detecting the defect shape of a pipe. When the inspection is started, the measurement range is set first. Assuming that the coordinate in the axial direction is z and the coordinate in the circumferential direction is θ, the measurement start point in the axial direction is z ₁ , the measurement end point z ₂ , the axial measurement points are θ ₁ and θ ₂ , and the measurement pitches are Δz, Set as Δθ. The pipe inspection device first moves to the measurement starting point (z ₁ , θ ₁ ) and then starts measuring the potential difference distribution. First, the potential difference distribution over the entire measurement range is measured with a coarse pitch. From the measured potential difference distribution, the potential difference near the measurement start point with no crack is set as the reference potential difference V ₀ , and the distribution of the potential difference ratio V / V ₀ is obtained. FIG. 15 shows a schematic diagram of the distribution of the potential difference ratio V / V ₀ obtained by measuring the potential difference in the axial direction by passing a current in the axial direction. In this figure, the horizontal direction of the paper is the axial direction of the pipe, and the vertical direction is the circumferential direction. Since the electric field is disturbed around the crack, the potential difference ratio becomes large. Crack difference ratio V / V ₀
The crack is a circumferential crack because the area where is large extends long in the circumferential direction. Therefore, the place where the potential difference ratio is the largest is detected, and it is determined that there is a crack where the potential difference ratio V / V ₀ is larger than 1.02, for example. Next, the potential difference distribution is measured with fine pitches in both the axial and circumferential directions only around the crack. For example, in Figure 15, Since the potential difference ratio V · V ₀ is larger than 1.02 in all ranges,
Measure to include this area. However, since the reference potential difference V ₀ is necessary, Measure a region that is somewhat wider than the range. Figure 16 shows a schematic diagram of the potential difference ratio distribution around the crack. In the case of circumferential cracks, it is determined that there is a crack at the point where the maximum potential difference in the axial distribution of the measured potential difference distribution,
At the same time, the distribution of the maximum potential difference in the circumferential direction is determined as the potential difference distribution along the crack. The potential difference distribution is used to determine the crack shape by a simple surface crack shape determination method described later, and the crack shape is displayed on the CRT screen (not shown) of the crack shape computer 1 (FIG. 4). The output of the measurement results is a hard copy of the crack shape or a printer output of the crack dimensions (coordinates, potential difference, potential difference ratio, crack depth, etc.).

The details of the flow chart in FIG. 14 will be described below.
First, the flow chart of Fig. 17 shows how to move the pipe inspection device. As shown in FIG. 4, the front drive cylinder, the axial drive cylinder, and the rear drive cylinder are all controlled by the computer 1 via solenoid valves. Here, the state in which the air cylinder is extended is represented as ON, and the state in which it is contracted is represented as OFF. In the initial state, the front and rear cylinders are ON, and the axial drive cylinders are OFF. When the axial movement range is set, for example, in the case of forward movement, the front cylinder is turned off, the axial drive cylinder is turned on, the front cylinder is turned on, and the front drive unit 30 is moved relative to the rear drive unit 40. It is possible to move forward by the stroke. Next, when the rear cylinder is turned off, the axial drive cylinder is turned off, and the rear cylinder is turned on, the rear drive unit 40 returns to the original position with respect to the front drive unit 30. As a result, the entire pipe inspection device has advanced by the stroke of the axial drive cylinder. Repeat this to move to the set position. If you want to move backward, you can reverse the procedure, so you can move backward as shown on the left side of FIG.

FIG. 18 shows a flow chart for measuring the potential difference distribution when a measuring head for detecting circumferential cracks as shown in FIG. 10 is used. After setting the measurement range, move to the measurement start point (θ ₁ , z ₂ ) and turn on the measurement head drive cylinder 16 to turn on the power supply terminal 6
5, Press the measuring terminal 67 against the inner surface of the pipe. The terminal for measuring the potential difference of four channels by the multiplexer 7 (FIG. 4) is switched and measured by the micro potentiometer 6. The measured potential difference is recorded in the computer 1 via the GP-IB interface 5, and at this time, the coordinates of the measurement position are assigned to the measured value and recorded. Next, the cylinder 16 is turned off, the measurement pitch Δθ is rotated in the circumferential direction, and the measurement is performed again. Repeat this to measure the potential difference distribution in the circumferential direction,
Since the measurement is performed using four pairs of measuring terminals, if the value obtained by multiplying the circumferential pitch Δθ by a value obtained by adding 1 to the number of times of measurement n exceeds the circumferential distance Θ between the terminal pairs, ΔΘ = Θ × 4
Just rotate. When the measurement up to the measurement range θ _{2 in the} circumferential direction is completed, the measurement head scanning unit 43 is moved by Δz in the axial direction by the axial direction driving DC servo motor 8. Next, the circumferential direction is measured in the reverse direction, and when the measurement is completed up to the circumferential measurement range θ ₁ , the measurement head scanning unit 43 is moved by Δz in the axial direction, and the measurement is performed again in the original direction. As described above, the potential difference distribution on the entire θ-z plane is measured by so-called rectangular scanning. Of course, when the measurement range exceeds the axial movement range of the measurement head scanning unit, the entire pipe inspection device is moved in the axial direction to measure the potential difference distribution.

In the above-mentioned potential difference measurement, if there is some temperature difference distribution in the sample to be measured (pipe), a thermoelectromotive force is generated between the measuring terminal and the sample to be measured, which is the average potential difference among the measured potential differences. Will be included as Therefore, in order to measure the potential difference of the measured sample itself, the thermoelectromotive force must be removed by some method. One method is to subtract the potential difference measured by cutting off the current from the potential difference measured by passing a current. Another method is to intermittently switch the polarity of the direct current and measure the amplitude of the potential difference. Since the latter has a larger absolute value of the measured potential difference, the measurement accuracy is improved accordingly. Further, the method of cutting off the current has a drawback that it takes time until the current stabilizes after the current has flowed, but the method of switching the polarity of the current has an advantage that the current stabilizes instantly. The device for switching the polarity of this current is the current polarity conversion device 3
(Fig. 4).

The method for determining the crack shape from the potential difference distribution along the crack is shown below. The flow chart of the surface crack shape determination method
Shown in Figure 19. In advance, a general-purpose large-scale computer was used to analyze the electric field for cracks with various aspect ratios, for example, a / c = 1.0,0.5,0.25,0.1, and the potential difference distribution on the surface in the direction perpendicular to the crack surface was stored in the internal storage device of the computer 1. Alternatively, it is stored in an external storage device. As an example of the stored potential difference distribution, Fig. 20 shows the potential difference distribution for each crack depth with an aspect ratio a / c = 0.5. This figure was obtained by analyzing the electric field by FEM for the case where there is a crack in the center of a flat plate with plate thickness t = 20 mm. The crack depth a / t normalized by the plate thickness t is 0,0.125,0.25,0.375,0.5,0.652 at the deepest point in the center of the crack and
It is 0,75. When there is no crack (a / t = 0), the potential difference is proportional to the distance z from the feeding terminal. On the other hand, when there is a crack, the potential difference increases near the crack. These potential difference distributions are approximated to the nth order by the computer 1 (4th
(Fig.). In determining the crack shape, the surface crack length 2c was determined from the potential difference distribution around the crack measured first.
* And maximum potential difference ratio V / V ₀ max are calculated. As an example, Fig. 21 shows the potential difference distribution around the crack introduced into the inner surface of the stainless steel 12B pipe by fatigue. The potential difference is almost constant where there is no crack, and when the average is obtained, V ₀ = 37,25 μV is obtained as the reference potential difference. The potential difference increases at the cracked portion, and the potential difference distribution in this portion is approximated to the nth order. FIG. 21 shows a curve obtained as a result of the fourth-order approximation. When the crack length 2c on the surface is obtained from the intersection of this fourth-order approximation curve and the reference potential difference V ₀ , 2c = 22.5
mm is obtained. From the approximate curve, the maximum potential difference ratio V / V ₀ max corresponding to the deepest point of the crack is determined. Vm in the case of FIG. 21
Since ax = 38.0, V / V ₀ max = 38.0 / 0 / 24.75 = 1.535 was obtained. Next, from the potential difference distribution shown in FIG. 20, the potential difference ratio V / V ₀ and the crack depth a for cracks of various aspect ratios a / c are shown.
In order to create the relationship with / t, the relationship between the potential difference ratio V / V ₀ and the aspect ratio a / c is created. In this case, since the electric field analysis by FEM analyzes a flat plate with a plate thickness t = 20 mm, the relationship between the potential difference ratio V / V ₀ and the aspect ratio a / c at the measurement position d * corresponding to the distance d between the measurement terminals is analyzed. Must be created. Therefore, d corrected by the plate thickness t * of the ratio measuring member
The potential difference is calculated for each crack depth at the position of * = d × 20 / t *, and the relationship of the aspect ratio a / c of the potential difference ratio V / V ₀ is created as shown in FIG. The relationship between the potential difference ratio V / V ₀ and the aspect ratio a / c is calculated by n-th order approximation for each crack depth a / t.
It is stored in the storage device (FIG. 4). Next, the potential difference ratio V /
Using the relationship between V ₀ and the aspect ratio a / c, the aspect ratio a / c
A master curve of the relationship between the potential difference ratio V / V ₀ for 0.5 and the crack depth a / t is created as shown in Fig. 23. Also in this case, the relationship between the potential difference V / V ₀ and the crack depth a / t is approximated to the nth order, for example, the fifth order. The crack depth a * is obtained by substituting the maximum potential difference ratio V / V ₀ max obtained by fourth-order approximation of the potential difference distribution into this master curve. Then, the aspect ratio a * of the crack is calculated by the surface crack length 2c * (= 2c × 20 / t) with the thickness corrected.
Find / c * and compare with the aspect ratio a / c of the master curve. If they do not match, the potential difference ratio V / V ₀
Aspect ratio a / c =
A master curve of the relationship between the potential difference ratio V / V ₀ for a * / c * and the crack depth a / t is created, and the maximum potential difference ratio V / V ₀ max is substituted to obtain the crack depth a *. This operation is repeated until they agree with each other, for example, until the difference between a / c and a * / c * becomes 0.01 or less. The shape of the entire crack is determined by substituting the potential difference ratio at each measurement position into the master curve of the relationship between the potential difference ratio V / V ₀ to the aspect ratio and the crack depth a / t when the two match. . In this case, as the potential difference ratio, the potential difference ratio at each measurement position may be substituted, or the potential difference distribution approximated to the nth order may be substituted.

The concrete calculation results of the potential difference distribution around the fatigue crack shown in Fig. 21 are shown below. Since the plate thickness of the stainless steel tube is t * = 15.8mm and the distance between the measuring terminals is d = 5mm, each aspect ratio at the position of d * = d × 20 / t * = 5 × 20 / 15.8 = 6.3mm The electric potential difference for each crack depth is calculated. However, since the potential difference is measured so that the crack is located at the center of the measuring terminal, the potential difference at the position of z = d * / 2 = 3.15 mm is obtained, and the potential difference ratio V / V ₀ as shown in Fig. 22 is obtained. Create a relationship with the aspect ratio a / c. Using these relationships, a master curve of the relationship between the potential difference ratio V / V ₀ for the aspect ratio a / c = 0.5 and the crack depth a / t is created as shown in FIG. Substituting the maximum potential difference ratio V / V ₀ max = 1.535 into this curve, a * / t = 0.266 and a * = 5.31 mm is obtained. Surface crack length 2c * ＝ 22.5mm when plate thickness is corrected to 2c ＝
22.5 × 20 / 15.8 = 28.48 mm, and the crack aspect ratio is a * / c * = 5.31 / 14.28 = 0.37. So next
When a master curve for a / c = 0.37 is created and the crack depth is obtained, a * = 4.97 mm is obtained, and a * / c * = 0.348
Becomes Once again, when a master curve for a / c = 0.34 is created and the crack depth is obtained, a * = 4.92 mm is obtained.
* / C * = 0.344, and the aspect ratios were almost the same.
This is the result of manual calculation, but when calculated by computer 1, the master curve for a / c = 3.348 was created and a * = 4.94mm, a / c = 0.345 were obtained, and the aspect ratios were almost the same. . The correspondence between the surface crack shape thus obtained and the beach mark on the fracture surface after fracture is shown in FIG. As shown in FIG. 21, the cracks are slightly shallow near the crack tip on the surface due to the coarse potential difference measurement interval, but they are very well in agreement except for that. Therefore, if the potential difference distribution can be measured with fine pitches, it will be more accurate.

However, the above-mentioned method can be applied to the case where the crack is perpendicular or parallel to the axial direction, and cannot be applied to the inclined crack as it is. FIG. 25 shows a method of determining the crack position when the crack is inclined with respect to the axial direction and the circumferential direction. FIG. 25 shows the potential difference distribution in the axial direction. The shaded area in the figure shows the maximum potential difference in the axial direction. Since the position of the crack is located at the center of each measurement block, it is determined that it becomes like a solid line connecting black dots. In such a case,
As shown in FIG. 25, the black point coordinate points representing the crack position are linearly approximated by the method of least squares to determine the angle with respect to the axial direction, and the crack length 2c * is determined from the coordinates at both ends. At this time, if the angle between the crack direction and the electric field direction is Θ,
The potential difference ratio V / V ₀ ′ becomes smaller than the potential difference ratio V / V ₀ when the crack is at right angles to the electric field.
V ₀ ′ = V / V ₀ · cos Θ. Therefore, when the crack shape is obtained by the above method, it is necessary to correct the measured potential difference ratio V / V ₀ ′ with Θ and evaluate it with V / V ₀ = V / V ₀ ′ / cos Θ. . However, if Θ exceeds 45 °, the accuracy will deteriorate, so it is better to use the measured value for the electric field where Θ is smaller than 45 °.

〔The invention's effect〕

As described above, according to the pipe defect inspection apparatus of the present invention, not only horizontal pipes but also vertical pipes and bent pipes can be moved, and the axial feed terminals of the measurement head by the DC potentiometric method provided in the device can be moved. Position of the crack by measuring the potential difference distribution in both the axial direction and the circumferential direction of the pipe with the measurement terminals arranged in a matrix or in a cross shape at equal intervals in both the axial direction and the circumferential direction and at the axial direction and the circumferential direction. Since the shape and shape of the pipe can be detected, it is possible to accurately inspect the soundness of the pipe.

[Brief description of drawings]

FIG. 1 is a side view showing an embodiment of a pipe defect inspection apparatus according to the present invention, with a part cut away. FIG. 2 is a rear front view of the above embodiment. FIG. 3 is a partial perspective view of an embodiment different from the above. FIG. 4 is a system diagram of the embodiment shown in FIG. 5 to 13 are explanatory views showing the structure of the measurement head,
Figure 14 is a flow chart of the entire defect inspection, Figures 15 and 16 are schematic diagrams of the measured potential difference ratio distribution, Figure 17 is an axial drive flow chart of the pipe defect inspection device, and Figure 18 is the potential difference distribution. Flow chart of measurement, FIG. 19 is a flow chart of the method for determining crack shape, FIG. 20 is a diagram showing an example of the potential difference distribution on the surface of the plate material obtained by FEM, and FIG. 21 is around the crack of the stainless pipe. The measured potential difference distribution chart, FIG. 22 is a chart showing the relationship between the potential difference ratio and the aspect ratio, FIG. 23 is a chart showing the relationship between the potential difference ratio and the crack depth,
FIG. 24 is an explanatory diagram showing a comparison between the actual crack shape and the determined crack shape, and FIG. 25 is an explanatory view showing a crack position determining method. 1 ... Computer, 2 ... DC power supply, 3 ... Current polarity converter, 4,5 interface, 6 ... Small potentiometer,
7 ... Multiplexer, 8 ... Measurement head axis direction, DC servo motor for driving, 12 ... Measurement head circumferential direction DC
Servo motor, 16 ... Cylinder for driving measurement head, 18 ...
… Front drive cylinder, 20 …… Axial drive cylinder,
22 …… rear drive cylinder, 26 …… position detection AE sensor, 27 …… compressed air source, 30 …… front drive, 33 …… universal joint, 40 …… rear drive, 43 …… measurement head Scanning unit, 46
...... Axial drive ball screw, 51 …… Circular rotation part, 60
…… Measuring head, 65 …… Axial power supply terminal, 67,74 …… Measurement terminal, 72 …… Circular power supply terminal, 78 …… Centering leg.

Claims (4)

(57) [Claims] 1. A pipe defect inspection device having a self-propelled function, which is inserted into a pipe for use, and has a structure in which a front mechanism and a rear mechanism are connected by a universal joint and a telescopic mechanism.
Further, the front mechanism and the rear mechanism each have at least three legs that are controlled to expand and contract in the radial direction of the pipe, and (a) a measuring head used facing the inner surface of the pipe, (B) On the surface of the measuring head facing the inner surface of the pipe,
A plurality of pairs of power supply terminals arranged facing each other in the circumferential direction of the pipe and spaced apart from each other; (c) a pair of measurement terminals facing and adjacent to each other in the circumferential direction of the pipe in the vicinity of the central portion of the pair of power supply terminals; ) A means for pressing the measuring head toward the inner surface of the pipe, and (e) a surface on which the measuring head faces the inner surface of the pipe,
A plurality of pairs of power supply terminals opposed to each other in the axial direction of the pipe, spaced apart from each other; (f) opposed to each other in the axial direction of the pipe in the vicinity of the center of the pair of power supply terminals opposed to each other in the axial direction; Adjoining measuring terminal pairs, (g) means for scanning the measuring head in the axial direction of the pipe, (h) means for scanning the measuring head in the circumferential direction of the pipe, (i) the power supply Means for electrically connecting a DC power supply to the terminals, the means having the same number of DC power supplies as a plurality of power supply terminal pairs, (j) a means for electrically connecting the measurement terminals to a micro potentiometer, and (k) the micro potentiometer And a means for inputting the output signal of the above into a computer. 2. The measuring terminal pair according to claim 1, wherein the pair of measuring terminals opposed to each other in the circumferential direction are plural pairs, and the pair of feeding terminals opposed to each other in the circumferential direction are arranged in the circumferential direction. 1. A pipe defect inspection apparatus characterized in that the number of pairs of measuring terminals facing each other is increased by one pair, and the opposing line of the measuring terminal is located in the middle of the opposing line of the power feeding terminal. 3. The power supply terminal as set forth in claim 1 or 2, wherein the pair of measurement terminals opposed to each other in the axial direction is a plurality of pairs, and the measurement terminals are opposed to each other in the axial direction. The number of pairs is one more than the number of pairs of measuring terminals opposed to each other in the axial direction, and the opposing line of the measuring terminal is located in the middle of the opposing line of the feeding terminal. Inspection device. 4. The oblique cone according to any one of claims 1 to 3, wherein the pair of measuring terminals has a bottom surface of the same size as the cross section of the cylinder at the tip of a cylindrical main body. And a guide member that is slidably contacted with the guide surface, wherein at least a part of the cylindrical main body is provided with a guide surface that is cut in a plane parallel to the center line. A pipe defect inspection device characterized in that it is slidable in the axial direction by means of which the rotation around the axial center is locked.