JPS5910846A - Eddy current flaw detector for metallic surface - Google Patents

Abstract

PURPOSE:To detect a surface defect securely by arranging three detection coils so that their center points form a regular triangle, and using differential detection signals between one specific coil and other two coils in combination. CONSTITUTION:The detection coils A, B, and C consisting of exciting coils 1a- 1c and detection coils 2a-2c are arranged at vertexes of a regular triangle to constitute a detector. An oscillator D supplies an AC current to the exciting coils 1a-1c, which are excited to flow the surface eddy current of a material to be detected. Voltages PA-PC induced across the detection coils 2a-2c are inputted to differential amplifiers DA1 and DA2 through amplifiers FA-FC and phase shifters SA-SC to send the difference between the induced voltages PA and PB and the difference between the PB and PC to phase detectors PD1 and PD2, which perform phase detection by a signal from a timing controller PB to obtain the differential detection signals. consequently, a surface defect is detected securely.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an eddy current flaw detection device for gold JFA surfaces, and is capable of detecting metal defects in various directions, regardless of gradual changes in conductivity and magnetic permeability on the surface of a planar metal material, and changes in Surface defects (f
-It is designed to be able to be detected reliably.

Various proposals have been made in the past regarding techniques for automatically detecting various defects on the surface of planar metal materials using the eddy current deep-sea method. -f: Most of the probe-type coils are used, and in order to detect all minute changes in impedance due to surface defects, the standard comparison type (
Detection 1 that has a detection coil +11 and a standard coil (2) that has an absolute value type) coil configuration! Often used for +31. However, in this method, the impedance change (also I''ii induction E change) of the detection coil il) due to defects existing on the surface of the metal material (4) is obtained as an absolute value, and the entirety of defects having various forms can be detected. On the other hand, metal materials (4)
Changes in the impedance of the coil due to surface conductivity, magnetic permeability changes, lift-off (distance between the coil and the material surface) changes, and coil temperature changes. ! , pressure change (
These generally have a larger contribution than those due to defects)
In many cases, the dynamic range of the detector (3) is exceeded and detection becomes impossible.

Therefore, in the case of item (h), two sensing coils f51 (6
A self-comparison type C differential type) coil configuration with 1k is adopted,
A method is used that cancels out changes in conductivity and detects only defects. In particular, in the case of hot flaw detection, the difference in temperature between the two detection coils becomes a problem, so it is common to use a differential configuration.

However, in this case, if the surface defect has various types of roughness, it may become undetectable. For example, as shown in Fig. 3, there is a defect (7) with a main component parallel to the direction in which the detection coil (61fsl) is arranged (hereinafter referred to as the fill direction),
As shown in FIG. 4, this is the case where a defect (7) perpendicular to the fill direction is located between the sensing coils +5i and +6+.

In the case of a wide planar metal material, it is known to scan the probe coils in various directions, or to use multiple probe coils in parallel, but when scanning in the direction of the microcoils, as shown in Figure 4, When there is a defect (7) in the direction perpendicular to the coil direction, it is possible to pass through a point where it can be detected without fail. However, as shown in FIG. 3, when there is a defect (7) in the direction of the coil, only the end portion of the defect (7) is detected, and the entirety of the defect (7) cannot be grasped. In a particularly extreme case, as shown in Figure 5, when the length of the defect (7) is sufficiently long in the coil direction, the depth of the defect (7) suddenly changes with the known differential probe. Only a portion can be detected, and most of the defects are undetectable.

The present invention has solved the above-mentioned problems, and its first feature is that an eddy current flaw detection device that detects surface defects in planar metal materials by the all-prologue eddy current flaw detection method is equipped with six detection coils. They are arranged so that the centers of all the coils are at the vertices of an equilateral triangle, and a differential configuration is established between a specific one of these sensing coils and the other two, and these two types of differential configurations are used. There is a point VC in which defects are detected by a combination of two detection signals.The second feature is that in an eddy current flaw detection device that detects surface defects of planar metal materials using a probe-type eddy current flaw detection method, , a large number of sensing coils are installed across the entire width of the metal material in such a way that the center of each of the three adjacent sensing coils is at the vertex of an equilateral triangle.
They are arranged in rows and have a differential configuration between adjacent detection coils, and defects are detected by a combination of detection signals from these differential configurations.

Hereinafter, the present invention will be explained according to the illustrated embodiments.
Figures 8 to 8 show all embodiments for detecting surface defects on a planar metal material by the mutual induction method, and in Figure 6 and Figure 7, (8) is a detector, and a holder (9) Attach it to the holder (9) and run 3 carp JL/Hobin (1
0A)(l0B)(IOc) and each bobbin Cl0A)(
Detection coils (A) each wound around lOB) (10C)
(B) (C) (!: Consisting of each detection coil (
A) (B) (C) ld outer excitation coil (1B)
(lb) (lc) and inner detection coil (2a) (2b)
(2c) and lffle.

The holder (9) is preferably made of a non-magnetic electrical insulator, but may also be made of a non-magnetic metal. Usually a pot-shaped ferrite core (ferromagnetic material) is used, but a round bar-shaped ferrite core can also be used, and even non-magnetic metal can be used. It can also be used with standard insulators. Each detection coil (A) (
B) (C) are arranged adjacently so that their centers are the vertices of an equilateral triangle (positions shifted from the vertices of an equilateral triangle are also possible), and their diameter and length are the minimum defects to be detected. It is said to be about the same amount or several times that amount.

Figure 8 shows a block diagram of the feeding control system 1, and in the figure,
(++lVi control circuit, the control circuit Ill is an oscillator (
D), amplifier (FA) (FB) (FC), phase shifter (SA
)(SB)(SC), differential amplifier (L)A+ )C[A
2), a phase detector (R) I'I (PD2), a timing controller (0, etc.).The oscillator (D) has a predetermined frequency of AC current detection coils (A) (B) (C
)'s excitation coil (rla) (ib) (lc) is energized by applying VC, causing a purple eddy current (induced current) to flow on the surface of the test material. difference! i)+amplifier (L)A+) is the detection coil (
The induced voltage PA generated in 2a) and the induced voltage PB generated in the detection coil (2b) are inputted via amplifiers (FA) (FB) and phase shifters rsA)/SB),
Find the difference between A and induction chamber 1 (1) B. Differential amplifier (D
A2) is the induction chamber [1fP
B and the induction that occurs in the detection coil 2c)? I! King PC and (I: Amplification Q (FB) (FC) and phase shifter (SB) (
SC) and enter this induction power ■゛13 and induction? l! Freeze King PC Kichi's puki. Timing controller (G
) All output signals of the l-1 oscillator (D) are input 7, and the timing of the phase detector (PI) +)CPD2) is fully controlled. Detection coils (2a) (2b) are detected by phase detector (Fol) and C1 which detects the phase of the output of the differential amplifier (11).
The orthogonal component outputs VA-B and HA-B' on the impedance plane of the impedance difference between the two are obtained. Phase detector (
1'D2) Phase detection of the output of the fi differential amplifier (Ilu) is performed to detect the orthogonal component outputs of the impedance difference between Il (21) and (2c) on the impedance plane T VC-B and He-B 1&: get. Amplifier (FA)
) (FC) and phase shifter (5A) (SC) (7) y
J+' IJ Knee Muke It can be adjusted automatically or manually.

According to the configuration of the above embodiment, each of the detection coils (A), (B), and (C) is initially placed on a healthy surface where there is no harmful defect (7) on or near the surface of the metal material (4). For example, if the volumes of the amplifier (FAMFC) and phase shifter (SA) (SC) are set so that the output of the phase detector (Pr)
When the defect (7) exists only directly under the detection coil (2b) (2c), the induced voltage PB and induced voltage PC of the detection coil (2b) and (2c) do not change, and the induction w1. of the detection coil (2a) does not change. Since only the voltage PA changes, a difference occurs between the induced voltage PA and the induced voltage PB, and as a result of phase detection, the phase detector (Isu1) detects the defect (7) directly under the detection coil (F5''). This phase detection output is approximately proportional to the impedance change time of the detection coil (A) due to the presence of defect (7). As shown in the figure, unlike conventional two-coil differential probes, it is possible to reliably detect cracks (7), which are typical surface defects, in various directions and relative positions with respect to the detection end. .That is, in the case (a) of Fig. 9, Kij VA-B
(Q, VC-B < O, and if (B), then VA-B
> 01VC-B: VA-B for Ol (c)
〉0, VC-BzQ, (gradient) if h VA-B<0.
In the case of Vc-B=0, (e) VA-B:; 0,
If VC-B>0, (to) then K 11 VA-Bku0,
VC-B < 0 and null. Nao detection coil (A) (B)
The probability of the existence of a defect that contributes to either of (c) is extremely rare.

10 to 12 show another embodiment in which surface defects of a planar metal material are detected by a constant current drive method, and as shown in FIGS. 10 and 11, a detector (sl f
i yJ-tv Thailand i91 3 Nohobin (10A)
(10B) (l0C) and detection coil 8) (13) (c)
(!: Constructed. Holder 19Bl) 'Hobinrl
OA) (IORX l0C) K The situation is the same as in FIGS. 6 and 7. Each detection filter (A) (B
) fc) ke each center is a vertex of an equilateral triangle,! : The detection carp should be placed adjacent to each other so that A) (B
) (C) consists only of a detection filter, and its diameter is about the same or several times as long as the smallest defect (7) to be detected. As shown in Figure @12, the control circuit fll+ is the same as in Figure 8, 9<of the oscillator), and an amplifier (FA).
(FB) (IrC), phase shifter ('5A) (SB) (SC
), differential amplifier (DAl) (DA2), phase detector (P
D+) (PD2), a timing controller (G), etc. Input the output of the oscillator (D) and install a constant current drive source (Q) that flows an alternating current with a constant amplitude and phase to each detection filter (A), (B), and (C).
The SEE at the wire end of each detection coil fA) (B) (C) is made to be proportional to the impedance difference of each detection coil (A) (B) (C). Therefore, in this case, the method of detecting defect (7) is the same as in the embodiment described in 1lrJ, and various surface defects +71 k ? lfl can actually be detected.

Next, a method for inspecting surface defects (7) over the main surface of metal materials (4) such as wide steel semi-finished products such as blooms and slabs, and finished products such as thin plates and thick plates will be explained.

In this case, as mentioned above, the detection coil (A) (13) (c
), the sensitivity and resolution of defect detection decreases, so one of the following two methods is applied.

Thousand P! J1 is a method in which a detector (8) is mechanically scanned in the width direction of a wide metal material (4) as shown in FIG. Since the metal material (4) is conveyed at a constant speed on the 1ffl conveyor line in the direction of arrow (a), the scanning speed V
If the following conditions are satisfied, the main surface of the metal material (4) can be inspected without fail.

2.7-8≦” II: 0.7X2.Dl where W
: Test material plate width S: Test material conveyance speed a: Effective detection range D 2" J ile diameter continuous a slur 7'17) - Example rH1W = 1500 + o +
-5=10DO/min Minimum defect length to be detected I
, zDχ3Dsm, so a=42ms, and therefore =1.2 work.

Although the method and apparatus for detecting flaws on the entire surface of a wide metal material by scanning a single detector (8) have been disclosed above, a plurality of detectors (8) may be arranged at regular intervals in the width direction of the material. Therefore, it is of course possible to reduce the scanning width and scanning speed variation.

The second method uses a large number of detection filters (A+XA2)...(AJ1+1)(B T
) (B21/bending (Bn) is placed on the surface of the metal material (4) over the entire width in the width direction of the metal material (4) with a gap in the width direction of the metal material (2), and three adjacent detection coils (A) ( In this method, the centers of each of the sensing coils (A
) (B) indicates the number T n 17) The value is W/I). In the case of continuous cast slag - for example W-1500m, n=
If it is 30 mm, n=50, and approximately 50 each of detection coils A) and (B) are required. As an example of how to create a differential, the following combinations can be considered.

In this case, the control circuit (11) is shown in Fig. 15 or the first
The configuration is as shown in Figure 6. That is, in FIG. 15, one oscillator (D) generates each excitation coil (1a+) (''1a2) of ('2n-1-1) detection coils IA) and (B).
)・=41 an-t-1) (1b+X 1b2)
= ・Dbn)'jc drive, (2 discharges 1) amplifiers (
FA+) (Ph2)・(FA411-+)(FB+)
(FB2)/(FB+-1) and phase shifter (SA1) (S
A2)=-(sAn41M SB+)(SB2)...
(SRn) and 2n fl! Differential amplifier (2) (DA31...(DAhl-11(DB+)
1...l'DBn) and phase detector (Ph3) (PA
s)...(PAnto+)(R3+)...rPBn)
kR is getting better. In the 16th figure, increase Wr m (Ft
vrx;s),! : Phase shifter ('A) (S!'l) and differential intensifier 'flFIL (m) and phase detector (PD) are 1
A switch is provided that receives the output of the timing generator (R) from the detection coils (A) and (B) and sequentially switches the output of the timing generator (R). KL so that the output of (PD) is sampled and held by the sample and hold device (11).

ing. However, in this case, the processing time allowed for one differential pair is as follows. That is, 2nT≦a15
(20-7X2D/S) ζ Here, a - Effective flaw detection range S - Test material conveyance speed change D - Coil diameter changes. In an example of continuous casting slag, W=1500+u+
, D = 30m-n = 50, and S = 1000
m/min, so T < i, 4 D/2 nS;
It becomes 0.025 seconds. And in normal eddy current testing, it is 5KH.
Test frequencies from z to 100 KHz are used, but 5 K
In the case of Hz, one period is 2X10-4 seconds, so D-12511'
This will be the 6th term. Note that the foil circuits shown in Fig. 15 and @16 are configured to detect the defect (7) by the mutual induction method, but instead of this, the circuit is configured to detect the defect (7) by the constant current drive method. m may be harvested.

Next, the experimental results will be described.

Example 0 Test material Hot-rolled steel plate, surface temperature 1000°C, 1200 m/min O detection method Multi-coil (method shown in tK14), control circuit parallel arrangement method (method shown in Fig. 15), coil Several 40 pieces 0 test Wk same wave u 50KHz O coil diameter V2O friend O liftoff 5 rxrx O prologue water-cooled O result flaw detection car 89 duck c Based on the results of several samplings) Example O test Material Continuously cast slag, Surface temperature 850″C1500
mm width C material Speed 1m/min 0 detection force type 6 fixture differential mechanical scanning method C test question failure 100KHz O coil diameter Daro O connection C lift-off 81uR O prologue water-cooled type % type % O result crack detection rate 82 inches (based on the actual response after cooling the slag) The conventional differential probe (when two detection coils are arranged in the width direction of the slab) could only obtain a detection rate of 45 inches; This is thought to be due to the fact that the occurrence rate of transverse cracks was high in the case of cast slabs, and the detection rate was very low.

According to the present invention, the centers of all three sensing coils are arranged at the vertices of an equilateral triangle, and the difference l between a particular one of the sensing coils I and the other two is determined. ! Since defects are detected based on the combination of detection signals provided by the IJ configuration, there is no directivity in detecting defects, and surface defects such as cracks in various directions can be detected with certainty. A large number of sensing coils are arranged in two rows across the entire width of the metal material so that the center of each of the six matching sensing coils is at the vertex of an equilateral triangle, and the distance between each adjacent sensing coil is Defects are detected by a combination of detection signals using a differential configuration, so the detection coil does not have to scan across the width of the metal material, and can reliably detect surface defects in various directions across the entire width of the metal material. It can be detected, and the effect of thousands is significant.

[Brief explanation of the drawing]

Figures 1 and 2 are perspective views of conventional examples, Figures 3 to 5 are plan views of conventional examples, Figure 6 is a side view of an embodiment of the present invention, and Figure 7 is a perspective view of a conventional example. The bottom view, Figure 8 is a block diagram of the control system, Figure 9 is a plan view for explaining the operation, 8g1
Figure 0 is a side view showing another embodiment, Figure 11 is a bottom view of the same,
Figure 12 is a block diagram of the same 1JIf [I] system, Figure 13
The figure is a plan view showing all other embodiments, Figure 14 is a plan view showing another embodiment, Figure 15 is a block diagram of the same control system, and Figure 16 is a plan view showing another embodiment.
This figure is a block diagram of a control system showing a modification of FIG. 15. (4)... Metal material, (7)... Defect, (A) (field (C)... Detection coil, III/VII mechanical circuit. Figure 9 Figure 13 Figure 74 Figure 15 Figure 76

Claims (1)

[Claims] 1. In an eddy current flaw detection device that detects surface defects in planar metal materials using a probe-type eddy current flaw detection method, six detection coils are arranged so that the center of each is the vertex of an equilateral triangle 1. A differential configuration is established between a specific one of these detection coils and the other two, and all defects are detected by a combination of detection signals from these two types of differential configurations. This is an eddy current flaw detection device for metal surfaces. Z In an eddy current flaw detection device f that detects surface defects in planar metal materials using the prologue type eddy current flaw detection method, a large number of detection coils are arranged so that the center of each of the six adjacent detection coils is at the vertex of an equilateral triangle. are arranged in two rows across the entire width in the width direction of the metal material, a differential configuration is adopted between each of the seven adjacent detection coils, and all defects are detected by the combination of detection signals from these differential configurations. This is an eddy current flaw detection device for gold FA surfaces.