JPS60161556A - Method of eddy current flaw detection for continuously-cast piece - Google Patents

Abstract

PURPOSE:To enable the detection of a flaw without being affected by a noise signal caused by an oscillation mark signal, by a method wherein probe coils electrified with currents of two frequencies are moved for scanning in the prescribed direction and detection signal components are rotated on a vector plane to discriminate a flaw signal from the noise signal. CONSTITUTION:Probe coils 105a and 105b electrified with currents of frequencies f1 and f2 are moved in reciprocation for scanning in the direction oblique to the direction of the movement of a continuously-cast piece 1. Detection outputs of the frequencies f1 and f2 by detectors 110 and 111 are turned into necessary and unnecessary signals A1' and B1', and A2' and B2', by phasers 112 and 113, and these signals are rotated in phase by a phaser 114 and turned thereby into necessary and unnecessary signals A and B which are perpendicular to each other on another vector plane. Then, the signal A corresponding to a presence/absence component of a flaw is disciminated excellently from the signal B corresponding to a noise component, and the flaw of the continuously-cast piece is detected by an eddy current test from the ratio between the respective differences between the maximum and minimum values of the flaw component and the noise component without being affected by the noise caused by an oscillation mark signal.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an eddy current flaw detection method for detecting defects on the surface of a continuously cast slab.

[Prior art]

Most of the slab slabs that are used as raw materials for thick steel plates, hot rolled steel plates, etc. are cast using a continuous casting process. In recent years, as part of energy-saving measures, the so-called hot charge process has been introduced in the steel sheet manufacturing process, in which after casting these slab slabs, they are charged into the heating furnace for the next process as hot slabs to save fuel in the heating furnace. It is adopted as one. However, there are limits to the application of this hot charging process. In other words, when the hot charging process is carried out, even if there are flaws in the cast slab slab, the flaws are not removed and are loaded into the heating furnace and then rolled, so the flaws do not appear on the steel plate after rolling. Surface flaws caused by this method may remain, making it difficult to apply to cast slabs where many flaws are expected to occur. To make this possible, it is necessary to heat the slab slabs to prevent them from being directly charged into the heating furnace.
It is necessary to accurately determine the presence or absence of flaws at the warm stage, and hot flaw detection methods are being actively developed for this purpose.

For example, optical flaw detection methods have been put into practical use for large flaws such as vertical cracks, but on the other hand, cracks and
It is difficult to detect flaws such as corner cracks using optical flaw detection because the flaw length is short. For this reason, eddy current flaw detection is used to detect even minute flaws.

One of them is, as shown in FIG. 1, two probe coils 105a, , 10 are arranged in parallel in the width direction of the slab slab 1 that is being transferred in the longitudinal direction (in the direction of the white arrow).
5b is scanned back and forth in the width direction, and the eddy current flaw detection is performed using the output difference of each coil.

This method is suitable for detecting vertical cracks due to the relationship between the scanning direction and the flaw direction, but cannot sufficiently detect horizontal cracks, corner cracks, etc. in the width direction of the slab.

Further, as shown in FIG. 2, two probe coils 105a are arranged in parallel in the direction of the slab slab 1 being transferred in the longitudinal direction (in the direction of the white arrow).
, 105b are arranged over the entire width of the slab slab 1 for flaw detection. In the case of this method, due to the arrangement of the probe coils 105a and 105b and the relationship with the direction of the flaw, it is suitable for defects in the width direction of the slab such as horizontal cracks and corner cracks, but is invulnerable to vertical cracks. The disadvantage is that a coil is required and the signal processing system becomes multi-channel and complex.

Flaws such as cracks and corner cracks often occur along oncision marks, and when such flaws are detected, noise is generated by the concave portions of the oncillation marks. Other causes of noise include:
This may be due to so-called lift-off fluctuations in which the slab slab vibrates up and down during flaw detection, causing the distance between the slab slab and the probe coil to fluctuate.

When such noise occurs, it becomes difficult to detect a flaw signal. A phase rotation method is known as a method for distinguishing between noise signals and flaw signals.

As shown in Fig. 3, this method uses, for example, a flaw signal A of a signal trajectory (represented on an impedance complex plane-vector plane with the real part on the X-axis and the imaginary part on the y-axis) obtained by flaw detection. The phase angle is different from that of the noise signal B, and the phases of the flaw signal A and the noise signal B are rotated so that the noise signal B is located on the y-axis (Fig. 4), and the noise signal BO is
By decreasing the X component and increasing the X component of the flaw signal A, the presence or absence, size, etc. of a flaw is determined based on the magnitude of the X component.

However, the phase angle difference between the oncillation mark signal and the flaw signal is actually small, and it is difficult to distinguish between the two using this method.

〔the purpose〕

The present invention has been made in view of the above circumstances, and its purpose is to detect defects such as vertical cracks and horizontal cracks regardless of the extending direction of the defects, and to detect defects using a two-frequency method. To provide an eddy current flaw detection method for continuously cast slabs capable of detecting flaws without being affected by noise signals caused by on-sillation marks even in the case of flaws along sillage marks.

〔composition〕

The eddy current flaw detection method for continuously cast slabs according to the present invention is a method for eddy current flaw detection of continuously cast slabs using a probe coil, in which electric current of two frequencies is applied to the probe coil, and the probe coil is moved in the direction of movement of the slab. The noise signal component of one frequency component of the detection signal and the flaw signal component of the other frequency component are aligned with one orthogonal axis in the vector plane of each frequency component. The phase of one or both of the frequency components of the detection signal is rotated in order to eliminate the noise signal, and a component orthogonal to the noise signal of the one frequency component and a component orthogonal to the defect signal component of the other frequency component are extracted. It is characterized in that the components are orthogonalized in a new vector plane, and a flaw signal and a noise signal are discriminated based on the ratio of the difference between the maximum value and the minimum value of each orthogonal component of the signal locus.

〔Example〕

The present invention will be specifically explained below based on the drawings.

FIG. 5 is a schematic front sectional view of the apparatus used for carrying out the present invention, as seen from the scanning direction, which is inclined at an angle α in the horizontal direction with respect to the conveying direction of the continuously cast slab; Figure 6 is a cross-sectional view of the flaw detection box taken along the line Vl-Vl in Figure 5.
Fig. 7 is a front sectional view showing the flaw detection head and its support tube vicinity, Fig. 8 is a side view showing the frame and the vicinity of the moving body, Fig. 9 is an enlarged elevational sectional view taken along line IX-IX in Fig. 8, FIG. 10 is a schematic diagram showing the lower surface of the flaw detection head.

In the figure, reference numeral 1 denotes a high-temperature continuous casting slab that has just been straightened by pinch rolls of a continuous casting facility (not shown), and is being transported horizontally in its longitudinal direction.

A gate-shaped support frame 2 is installed across the transfer area of the slab 1 at right angles to the transfer direction (only the upper frame material is shown in the figure), and the upper frame material of the support frame 2 is A rectangular tube is turned sideways, and its side surface (example: center portion) is cut out in the horizontal direction. Inside the wheel 32, an axle is fixed near the head of a vertically long hanging member 31, the head of which is semicircular.
32 are removably fitted, and the hanging member 31.3
1 is attached to the upper surface of the box frame 30 of the lifting device 3. A motor (not shown) is attached to one of the wheels 32 via a gear, and the wheel 32 is driven by the motor.
is rotated, and the lifting device 3 is movable in the width direction of the slab 1. An air cylinder 33 is installed inside the box frame 30 and protrudes from the lower surface of a cylinder rod 36, and a rectangular connecting plate 39 is fixed horizontally to the lower end of the cylinder rod 36.

A rectangular connecting member 39a is fixed to the lower surface of the connecting plate 39, and the head of the connecting member 5a having a T-shaped cross section is positioned within the connecting member 39a. The bottom of the holder prevents it from falling and is fixed in place.

The supply/discharge pipes 34 and 35 of the air cylinder 33 are connected to an air supply source (not shown), and the air cylinder 33 is operated by the air supplied from the air supply source.
6 is moved in and out to move the flaw detection box 5 up and down in the vertical direction.

Two anti-rotation ronds 37 and 38 are installed vertically on the connecting plate 39 with the cylinder rond 36 in between, and the upper ends of the ronds 37 and 38 are connected to mooring holes 41 and 42 formed on the lower surface of the box frame 30. The length of the inserted portion is determined so that it will not come out even if the cylinder rond 36 is advanced to its maximum extent. Connecting plate 39 and rotation prevention iron 37.38
prevents the flaw detection box 5 from rotating.

The flaw detection box 5 has the shape of a rectangular parallelepiped box with a central portion of the bottom thereof cut out, and a bottom plate 51 is constructed and fixed around the cutout portion 5b. A rectangular scanning plate 53 is mounted on the bottom plate 51, and a strip-shaped opening 53a of a predetermined size is formed in the center of the scanning plate 53 so that its longitudinal direction coincides with the longitudinal direction of the scanning plate 53. .

As shown in FIG. 6, the bottom plate 51 is provided with a circular opening 51a having a diameter slightly longer than the length of the long side of the opening 53a, with its center aligned with the center of the opening 53a. 53 can rotate around the center of the opening 53a and the opening 51a.

Bearing legs 58 are attached near the four corners of the lower surface of the bottom plate 51, and four rotatable axles 60, 60, . A shaft portion 59 is inserted so that the lower part of the wheel 60 can be seen below the flaw detection box 5.

The flaw detection box 5 is placed on the slab 1 via wheels 60 by adjusting air supply and discharge to the air cylinder 33. The wheels 60 roll as the slab 1 is transferred.

On the scanning plate 53, as shown in FIG.
It extends in the longitudinal direction of the opening 53a across the a,
The exposed rails 55b, a slide member 55 containing a slide bearing slidably fitted to the upper part of the rail 55b.
a, Flaw detection head mounting plate 54 with 55a attached to the bottom surface
is provided. The flaw detection head mounting plate 54 has a rectangular shape whose long side is longer than the interval between the rails 55b, 55b, and its longitudinal direction is perpendicular to the laying direction of the rails 55b, and corresponds to the opening 53a of the scanning plate 53. A circular opening 54a is provided at the position where the opening 54a is located.

Around the opening 54a, as shown in FIG. 7, there is a cylindrical spring receiver 5 whose upper end is closed except for the central opening 56a.
6 is installed.

The openings 54a, 56a and 51a have a flange 5 in the middle.
A support tube 57 for attaching a flaw detection head having a diameter of 7a is attached to the flange 57.
The spring receiver a is inserted so as to be positioned inside the tube 56, and a spring 57b is interposed between the upper rear surface of the spring receiver tube 56 and the flange 57a to surround the support tube 57. , urges the collar portion 57a downward.

The lower end of the support tube 57 is located below the bottom plate 51,
A flaw detection head 100 is connected. The length of the support tube 57 and the position of the flange 57a are adjusted so that the distance between the flaw detection head 100 and the slab 1 is a predetermined length.

A belt-shaped frame 75 whose long sides are longer than the rails 55b is installed on the scanning board 53 via vertical supports 76, 76, with the elongated direction of the frame 75 coinciding with the installation direction of the rails 55b.

On the upper surface of one end of the long side of the flaw detection head mounting plate 54, a moving body 8o having a C-shaped cross section and a thick tip end is connected to its lower surface and a connecting member 72 having an I-shaped cross section. Grooves 80a and 80a are cut in opposing surfaces of the distal end of the movable body 8, respectively.

The upper and lower edges of a frame 75 located closer to the center of the flaw detection head mounting plate 54 than the movable body 8o are slidably fitted into the grooves 80a, 80a, so that the flaw detection head is integrated with the frame 75. Board 54. Guides the movement of the moving body 8o.

The frame 75 has an elongated opening in its longitudinal direction, and a driving sprocket 77 and a driven sprocket 78 are provided at both longitudinal ends of the surface of the frame 75 facing the movable body 8o. A chain 83 is stretched across. Drive sprocket 77
A rotating shaft of a drive motor 71 for one-way rotation, which is placed on a stand 74 on the scanning plate 53, is fixed to the shaft, and the chain 83 is rotated by its drive.

The moving body 80 is provided with a reverse gear 84 meshing with a chain 83, and a driving sprocket 77 and a driven sprocket 7 on the surface of the frame 75 facing the moving body 8°.
Stoppers 81 and 82 for reversing the reverse gear 84 are protrudingly provided at positions close to each of the reverse gears 84. As a result, when the reverse gear 84 is engaged with the upper side of the chain 83, the movable body 8o moves in the direction of the white arrow, and when it comes into contact with the stopper 82, the reverse gear 84 is reversed and moves below the chain 83. They mesh, and the movable body 80 moves in the direction opposite to the direction of the white arrow. When the reverse gear 84 comes into contact with the stopper 81, the reverse gear 84 is brought into mesh with the upper side of the chain 83, and this process is repeated. Therefore, the flaw detection head 00 connected to the moving body 80 scans back and forth.

This direction of reciprocating scanning or the sliding member 55a.

The direction of the rail 55b and frame 75 is the length of the flaw to be detected, the direction of extension of the flaw, and the probe coils 105a, 10.
5b in the width direction of the slab 1, the scanning stroke of the flaw detection heel l"200, etc. It is preferable that the angle α formed is about 10°.

A pair of probe coils 105a and 105b are installed in the flaw detection hem 100 one after the other in the direction of conveyance of the slab 1, and as shown in FIG. Agricultural fields are placed at predetermined intervals.

Each probe coil 105a, l05b is water-cooled, and both ends thereof are connected to the bridge 105 of the signal processing section. As shown in FIG. 11, which is a block diagram of the signal processing section, the probe coils l05a and 105b are two of the four sides constituting the bridge 105, and are connected to the oscillator 1.
The outputs of frequencies fI+r2 of 02 and 103 are sent to the bridge 10 via a frequency mixer 104 that mixes them.
It is given to 5.

The impedance of the probe coils 105a and 105b changes depending on the surface condition of the slab 1, and the resulting unbalanced current in the bridge 105 is sent to filters 106 and 107, where the component signal fI' based on the frequency f1 is transmitted.
and a component signal f2' based on the frequency f2. The separated signals fl, Z, and f2' are sent to amplifiers 10, respectively.
8, 109 and is amplified, and the amplified signal is sent to the detector 1.
10, sent to 111. Detectors 11O2111 are given synchronizing signals from oscillators 102 and 103, respectively, and detectors 110 and 111 are provided with amplifiers 108 and 109.
The signals from the transfer rotators 112 and 113 are synchronously detected, and the detected signals are input to the transfer rotators 112 and 113, respectively. The signal of frequency f1 is the first
As shown in Figure 2(a), the orthogonal coordinate system (X,.

y+) on the vector plane of the necessary signal A1. The unnecessary signal B1 appears, and the signal of frequency f2 is shown in Fig. 12 (b
), the necessary signal A2+ and the unnecessary signal B2 appear on the vector plane of the orthogonal coordinate system (Xl, yz).

B1 81 is located on the yl' axis. The signals after the phase rotation, that is, the signals A+', B+' on the (X+', y1 plane,
' component is input to the phase rotator 114. One phase rotator 113 converts the signal A 2 + B 2 so that A2 is yz'
The phase is rotated so that it is located on the axis, and the signal after the phase rotation is
In other words, the signals A2', B on the (Xl', yz') plane
The x2' component signal of 2' (FIG. 13(b)) is input to the phase rotator 114.

The phase rotator 114 further rotates the phase of the input signal, that is, the xl' and

The signal shown in FIG. 13(C) obtained in this way is
A signal A is obtained by orthogonally expanding the phase angle difference between the necessary signal A1' and the unnecessary signal B2'.

It becomes B. By arranging the signals at right angles in this way, it becomes easy to separate and identify both signals, and by reading the magnitude of the necessary signal A corresponding to the flaw signal, it becomes possible to know the presence or absence and size of a flaw.

A reference value is set in advance in the comparators 115 and 116, and when a signal equal to or higher than this reference value is input, the input signal, in this case the X and Y component signals, is sent to the X component minimum value detection circuit 117 and the maximum Value detection circuit 118, Y
It is applied to the component minimum value detection circuit 119 and maximum value detection circuit 120.

The X component minimum value detection circuit 117, the Y component minimum value detection circuit 119, the X component maximum value detection circuit 118, and the Y component maximum value detection circuit 120 detect and hold the minimum value and maximum value of the input signal.

The reset signal 125 sends a reset signal to each of the detection circuits 117, 118, 119, and 120 at intervals determined based on the frequency output from the oscillators 102 and 103 or the relative movement speed between the probe coils 105a, 105b and the slab 1. The maximum value and minimum value of the X component since the previous reset signal are sent to the differential amplifier 121, and the maximum value and minimum value of the Y component are sent to the differential amplifier 121.
The minimum value is output to the differential amplifier 122, and the differential amplifier 1
21 and 122 calculate the maximum value-minimum value (ΔX and ΔY) of each of the X component and Y component, and send the difference signals ΔX and ΔY to the divider 123, respectively. The divider 123 calculates ΔX/ΔY, and inputs the quotient signals ΔX, /ΔY to the determination circuit 124.

The determination circuit 124 compares the comparison ratio set therein with the input value, and if the input value exceeds the set value, determines that a harmful defect exists and issues a predetermined output.

Next, the principle of discrimination between a noise signal and a flaw signal and the phase adjustment of the phase rotators 112, 113, and 114 of the present invention will be explained. The signal fI' input by the detector 110 is phase rotated by the phase rotator 112 and obtained (xl 'y1
') Flaw signals A, ' with a phase angle difference of θ1 on the plane and oncillation mark signal B1 ' [Figure 14 (a)]
When expressed as vectors, the respective vectors r'.

“Hello? ,/\.

A+ ' = A1 ' sinθ+ X+ +A+ '
cos θ1△ "ko'-B+ ')'1' ... (2)/\7 △ However, Xl + y1' becomes the unit vector of the X1' + 3'l ' axis. (
Xl,)'2) Flaw signal A2 with a phase angle difference of θ2 on the plane
and oncillation mark signal B2 (Fig. 14(b)
] When expressed in bepect, shoulder −A・・i・(α+θ・
) G + △ A2C05 (α + θ2) yz... (3) Storage = B2 s
inαC+132 COsα* ・・−(41However, 0
．． The angle is the unit vector of the X2+yl axis, and α is the angle between the yl axis and B2. (Xl,)'2) Place the plane in the phase rotator 113.
When the coordinates are rotated by an angle θ, the rotated (xl ′
−→7−→.

A2 and B2 in the yz') plane are as in (3) above.
.

From formula (4), A2' = A2 sin (α+φ+θ2) C'
＋／＼.

1-゛(J, l B2 cos (α+φ)yz...
・(6) becomes.

1 in the (xlZX2') plane composed of the signal XIZ x2' input to the phase rotator 114 is the above (11
, +21 and formula (51, f6), go to →.

A = A1' sin θ, X, +.

2/\, Azsin(α+φ+θ2) X2' =・(71
B=82 5in(α+φ)x';' (8)-Ude. In order to make A and B orthogonal in the phase rotator 114, A - B = 0, but the condition is φ
=-α or φ=-(α+θ2).

As understood from FIG. 14(b), the phase rotator 113
If you adjust -1 → and set B2 to be on the y2 axis, φ-
-α is satisfied, and if A2 is set on the y2 axis, φ-1 (α+02) is satisfied. However, the signals A and B are actually not straight vectors but figure-8 trajectories. Considering these actual waveforms, φ−−α, φ−1(α+θ
Signals obtained by simulating the waveform after the rotational scanning in 2) are shown in FIGS. 15 and 13, respectively. In the case of φ=-α, as the phase angle of the signal A1B1' [Fig. 15(a)] approaches that of the signals A2', B2 [Fig. 15(b)], the calculation result signals A and B (FIG. 15(C)) The phase angle difference becomes small and discrimination becomes difficult, but in the case of φ-1 (α+02), the signal A1' + 81' and the signal A2'.

Even when the phase angle difference of B2' is small, the 13th
The signals A and B shown in FIG.

Therefore, in the present invention, the latter condition is used for the rotation operation of the phase rotator 113. Therefore, in the present invention, when the eaves are not present in the valley of the oncillation mark, the output of the phase rotator 114 is Using this method, the eaves can be detected well. That is, if a flaw exists, it can be detected by the presence or absence and size of the signal A in the X direction.

Now, if there is a flaw in the valley of the oncillation mark, the detection trajectory is the detection trajectory of only the oncillation mark [Fig. 16(a)] and the detection trajectory of only the flaw signal, as shown in Fig. 16 tc+. [Fig. 16(b)] is overlapped, and there is almost no difference from the trajectory in which only the oncillation mark was detected [Fig. 16(a)], but the size of the flaw, that is, the flaw signal The phase angle differs depending on the size, and the X component increases and the Y component decreases. Therefore, the second invention of the present application is to convert the maximum value of the X component of the figure-8 trajectory to the minimum value ΔX [Fig. 16 (C)] to the maximum value of the Y component -
The presence or absence of a flaw is determined based on the ΔX/ΔY value divided by the minimum value ΔY (FIG. 16 (C)).

Note that the oncillation mark signal is a signal based on the distance variation between the probe coil and the slab due to surface irregularities, and this distance variation is caused by swinging of the probe coil or vertical movement of the slab 1 during conveyance. Since they are essentially the same as the resulting lift-off fluctuations, it goes without saying that the present invention can also be applied to the case where a signal related to lift-off fluctuations is discriminated from a surface flaw signal.

〔effect〕

Next, in order to detect corner cracks in the hot slab slab that is being continuously cast at a drawing speed of 1.44 m/min, a flaw detection head with three sets of probe coils arranged at a distance of 60+ nm is placed on both sides of the slab. The angle α is 10°.
, the scanning speed of the flaw detection head was 24 m/min.

Eddy current flaw detection was performed according to the present invention with a scanning stroke of 400 mm. In this case, the flaw detection width of the flaw detection head in the width direction of the slab is about 250 mm, and since corner cracks often exist within about 200 mm from the corners on both sides of the slab, the above flaw detection is performed to detect corner cracks. Cover enough range.

FIG. 17 is a diagram showing a substantial scanning locus of the probe coil with respect to the slab 1 at that time, and the arrow mark indicates the scanning direction. In the example above, the pitch of this zigzag trajectory is 8
, approximately 5mm, covering the entire range to be detected.
Note that the angle α, scanning speed, and scanning stroke of the scanning plate 53 may be appropriately determined depending on the drawing speed and the shape and dimension of the flaw.

Figure 18 is a graph showing the flaw detection results, with the horizontal axis representing the flaw depth (mm) and the vertical axis representing the eddy current output value (V). An artificial flaw with a length of 7 mm created with a chisel in the width direction, ◇ is an artificial flaw with a length of 3 mm created in the width direction of the slab slab with a chisel, and e marks 1 mouth marks 1 Δ marks are natural defects that occurred during casting. Horizontal scratches, push-in scratches.

They are based on flaw detection output values for vertical cracks, and the broken line represents the noise output level of 0, which also includes oncillation marks at a depth of about l+++m during this flaw detection.

As can be understood from this figure, the flaw signal output value is 1 or more, which is large compared to the noise output level of about 0.5V, and the flaw signal output value is greater than 1. Can determine defects in

In addition, in the present invention, when detecting flaws in the phase direction of a slab such as vertical cracks, it is of course possible to further rotate the scanning plate so that the scanning direction is perpendicular to the phase direction (α-90°). In this case, the accuracy of detecting vertical cracks etc. can be improved.

In addition, in the present invention, a reverse gear and a slide bearing are used to scan the gutter for flaw detection, but it is of course possible to use other gears other than these. When a bearing is used, there is less vibration when the flaw detection head stops, reverses, and starts, allowing stable flaw detection.

As described in detail above, the present invention uses a probe coil that carries current of two frequencies to tilt in the longitudinal direction of the slab and scans it back and forth, so that the entire area to be inspected can be inspected. Since the noise signal and the flaw signal are orthogonalized using a detector, and the flaws are detected by the ratio of the X component value and the Y component value of the detection trajectory, it is possible to detect flaws regardless of whether they are parallel or perpendicular to the oncision mark. In addition, even if a flaw exists in a place where noise occurs, such as the valley of an oncision mark, it is possible to accurately detect the flaw.Furthermore, since this flaw detection is possible during continuous casting, it is possible to detect the real charge process. Even in this case, the equivalent effect of improving the product yield after rolling can be achieved by separately sending defective slabs to the care process.

[Brief explanation of the drawing]

Figures 1, 2, 3, and 4 are explanatory diagrams of conventional methods;
Fig. 5 is a schematic front cross-sectional view of the apparatus used to carry out the present invention, Fig. 6 is a cross-sectional view of the flaw detection box taken along the v+-vit line in Fig. 5, and Fig. 7 shows the flaw detection head and its support tube vicinity. 8 is a side view showing the frame and the vicinity of the moving body, FIG. 9 is an enlarged elevational sectional view taken along line IX-IX in FIG. 8, and FIG. 10 is a schematic diagram showing the lower surface of the gutter for flaw detection. 11 is a block diagram of the signal processing section of the present invention, FIG.
13, 14, 15, and 16 are diagrams used to explain the principle of the present invention, and FIG. 17 is a diagram showing the scanning locus.
FIG. 18 is a diagram showing the effects of the present invention. 1...Continuous casting slab 53...Scanning plate 71...
Drive motor 80...Moving body 100...Flaw detection head 102, 103...Emission device 105a, 1
05b... blow coil 112゜113, 1
14... Phase rotator 117... X component minimum value detection circuit 118... X component maximum value detection circuit 119...
・Y component minimum value detection circuit 120...Y component maximum value detection circuit 121, 122...Difference amplifier 123・
...Divider patent Applicant: Sumitomo Metal Industries, Ltd. 1
Famous agent Patent attorney Noboru Kono No. +1211 ¥, Figure a 3 [X] Heat + I Figure S Figure 9 (a) (b) (a) (b) Figure 14 (C) 1st horoscope (b) Figure V16 vI'1 country 5 yen (mm) No. 1g7 Continuation of page 1 0 Inventor Tatsuo Hiroshima Inside Nishinagasu Honmichio Technical Research Institute, Amagasaki City

Claims (1)

[Claims] 1. In a method of eddy current flaw detection of continuously cast slabs using a probe coil, the probe coil is energized with electric current at two frequencies, and the probe coil is reciprocated in a direction forming a predetermined angle with respect to the direction of movement of the slab. scan,
One or both of the frequency components of the detection signal are adjusted so that the noise signal component of one frequency component of the detection signal and the defect signal component of the other frequency component are aligned with one orthogonal axis in the vector plane of each frequency component. The phase is rotated, a component orthogonal to the noise signal of the one frequency component and a component orthogonal to the defect signal component of the other frequency component are extracted, and the two extracted components are orthogonalized in a new vector plane, and the signal is An eddy current flaw detection method for continuously cast slabs, characterized in that a flaw signal and a noise signal are discriminated based on the ratio of the difference between the maximum value and the minimum value of each orthogonal component of the trajectory.