JPH0350989B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a method for detecting subcutaneous defects using ultrasonic flaw detection, and more specifically, its purpose is to distinguish and detect surface defects and subcutaneous defects of a test material by practical means. In recent years, manufacturers of secondary processing of wire rods and steel bars have been proceeding with process omissions for the purpose of saving labor, energy, and reducing costs, and the processing conditions for wire rods and steel bars are becoming increasingly severe. Therefore, the wire rod
Processing cracks during cold forging and wire breakage during wire drawing caused by minute inclusions existing inside steel bars have become a problem, and out-of-furnace refining is required to remove inclusions and prevent inclusions during the ironmaking and steelmaking stages. From the viewpoint of technology and quality assurance, it has become essential to establish an inspection technique to detect the presence or absence of fine inclusions. As for this inspection technology, for steel bar products, there is an ultrasonic flaw detection device that rotates the bar or rotates a probe around the steel bar while injecting a focused ultrasonic beam into the material, but for wire products, the diameter is small. Due to its thinness, similar flaw detection methods cannot be applied, and it is currently impossible to perform flaw detection on the entire circumference and length of the coil at the final stage. Incidentally, problematic inclusions in wire rods and steel bars are present from the material stage, so detecting inclusions at the stage of rectangular steel pieces guarantees the internal quality of the product, especially in wire rods. In this case, ultrasonic flaw detection inside a square steel billet is used as a substitute for internal flaw detection of the product. By introducing internal flaw detection into this intermediate process, defective pieces of steel can be removed in advance, and compared to flaw detection at the product stage, the length of the tested material is shorter and the flaw detection process is more efficient. However, in order to ensure its reliability, it is required to exhibit high detection accuracy even for minute inclusions. Conventionally, as an ultrasonic flaw detection method for rectangular steel pieces, a vertical flaw detection method using a two-piece probe is known.
However, this method has a major problem in that, although defects existing inside can be detected, defects existing beneath the surface of the square steel slab cannot be detected. That is,
When actually processing square steel pieces into products, inclusions that can cause problems such as processing cracks are often found beneath the surface of the steel, and it is desired to establish a flaw detection method that detects internal defects, including those beneath the surface. Therefore, the applicant of the present application proposed a surface layer flaw detection method applying angle angle flaw detection in Japanese Patent Application No. 57-233944.
In flaw detection that applies this angle flaw detection, defects in the surface layer and inside can be detected from the surface of a square steel piece without a dead zone. For these reasons, the accuracy of estimating the position of the detected defect is not very good. Therefore, it has been difficult to determine whether the detected defect is a superficial defect or a subcutaneous defect. Defects on the surface (surface flaws) can be removed by chipping or grinding in the billet processing process, and do not pose a problem during secondary processing of the finished product. Therefore, it is important to distinguish whether the detected defect is a surface defect or a subcutaneous defect of the square steel piece. Therefore, as a method for distinguishing between surface defects and subcutaneous defects, a method disclosed in, for example, Japanese Patent Application Laid-open No. 132054/1983 has already been proposed. This conventional discrimination method uses a magnetic flaw detector and an angle angle flaw detector, and defects detected only by the magnetic flaw detector are treated as surface defects, while defects detected by both the magnetic flaw detector and the angle angle flaw detector are treated as surface defects. It was classified as a subcutaneous defect (subcutaneous defect). However, since it is not possible to determine whether a defect detected by an angle angle flaw detector is a surface defect or a subcutaneous defect, just because a defect is detected by a magnetic flaw detector does not necessarily mean that it is a subcutaneous defect. I can't say for sure. Further, since there is a difference in accuracy between defect position estimation by the angle flaw detection device and defect position estimation by the magnetic flaw detection device, it is uncertain whether the defects detected by both devices are the same defect. Therefore, the conventional method for detecting subcutaneous defects is as follows:
It was a step up in accuracy. Therefore, the present invention was invented in view of the above-mentioned problems, and it is an object of the present invention to provide a method for distinguishing and detecting surface defects and subcutaneous defects of a test material by practical means. , is characterized by a longitudinal wave angle flaw detection device using an electronically scanned array type probe. Moreover, by setting the probe at a predetermined angle to the material surface and scanning the probe using a combination of electronic sector scanning and electronic linear scanning, the incident point on the material surface is kept approximately constant and the angle of incidence on the material surface is kept approximately constant. By detecting the surface layer inside and on the side surface adjacent to the entrance plane, defects in the surface layer of the material to be inspected are detected, the position of the surface layer defect is estimated, and the position is set as information 1, and surface defect detection is performed. The device detects a surface defect on the material to be inspected, estimates the position of the surface defect, gives a range larger than the position estimation error by the angle angle flaw detection device, and calculates the range based on the position of the surface defect. is defined as a surface defect area, that area is defined as information 2, and by subtracting information 2 from said information 1, only subcutaneous defects are detected. The present invention will be explained in detail below. First, the introduction of the oblique flaw detection method in the present invention will be explained. Conventionally, the ultrasonic vertical flaw detection method has been applied to detect internal defects in square steel pieces, as described above, but in this case, as shown in Fig.
The ultrasonic wave S emitted from the
S ₁ has the disadvantage that a reflected echo B ₁ is generated at the bottom surface, and the area near the surface becomes a dead zone. On the other hand, when using the angle angle flaw detection method, as shown in Fig. 2, the reflected echo _S1 from the incident surface of the ultrasonic wave S remains, but the reflected echo from the side surface of the square steel piece 2 remains. Since the echo S ₂ travels downward, there is no dead zone caused by reflected echoes on the sides. Therefore, by using the angle angle flaw detection method and setting the side surface adjacent to the entrance plane as the flaw detection area, it becomes possible to detect flaws from the surface to the inside of the square steel piece 2 without a dead zone. Furthermore, when testing the surface layer of the surface facing the entrance surface, as shown in FIG. 3, no dead zone is generated due to the reflected echo _S1 from the bottom surface, as in the case of testing the side surface. Therefore, in oblique flaw detection, the surface layer of the bottom surface, which is opposite to the entrance surface, can be used as the flaw detection area. The basic principle of the present invention is to detect flaws in the surface layer of the square steel piece 2, which is the test material, by such an angle flaw detection method. Next, in oblique angle flaw detection, the range of refraction angles of ultrasonic waves that can be used to detect flaws in the surface layer will be explained. As shown in Figures 4 and 5, the refraction angle θ is
When the angle exceeds 30°, the bottom echo _B1 becomes extremely small. That is, there is no dead zone in the surface layer of the bottom surface, and surface layer flaw detection becomes possible. Therefore, in order to detect flaws in the surface layer of the bottom surface, it is desirable that the refraction angle θ is 30° or more. When using the side surface layer as the flaw detection area as shown in Figure 6, based on the results for the bottom surface above, if the ultrasonic wave hits at an angle of 30° or more with respect to the normal line to the side surface, there will be an echo reflected from the side surface. It can be seen that flaw detection on the surface layer is possible without causing any problems. That is, it is desirable that the refraction angle θ is 60° or less. Next, we will explain whether flaw detection is performed using longitudinal waves or transverse waves. FIG. 7 is a graph showing the relationship between the angle of incidence α, the angle of refraction θ, and the round-trip passage rate when ultrasonic waves are incident on iron from water. In the graph, T _S indicates a transverse wave and T _D indicates a longitudinal wave. As is clear from FIG. 7, in the region where the refraction angle is 35° or more, a higher passing rate is obtained for the transverse wave T _S than for the longitudinal wave T _D. Therefore, if a transverse wave T _S with a refraction angle of 35° or more is used, no longitudinal wave T _D is generated, so there is no confusion. In other words, if longitudinal wave angle flaw detection is performed in this range, transverse waves will be generated at the same time, and the shear waves may pick up defect echoes from different directions. is desirable. When using the side surface layer as the flaw detection area, the refraction angle θ is
It is desirable that the angle of refraction θ is 60° or less.
In a small area of 35° or less, if longitudinal waves T _D are used that provide a higher round trip pass rate than transverse waves T _S , the usable angular range of the refraction angle θ becomes wider. Therefore, it is preferable to use longitudinal wave oblique angle flaw detection for flaw detection on the side surface layer. However, an appropriate refraction angle θ of the transverse wave or longitudinal wave may be selected depending on the purpose of flaw detection and the area to be flaw detected. Note that the relationship between the refraction angle θ and the incident angle α is expressed by the following equation. sin α / sin θ = C ₁ / C ₂ α; angle of incidence θ; angle of refraction C ₁ ; speed of sound of the material on the incident side C ₂ ; speed of sound of the material on the refracting side A scanning technique for flaw detection over the entire length will be explained. In order to perform online flaw detection over the entire length of the square steel piece 2, the ultrasonic beam S must be scanned at high speed. This scanning method can be roughly divided into mechanical scanning and electronic scanning, and for each of them, linear scanning in which the ultrasonic beam is moved in parallel, and sector scanning in which the ultrasonic beam is waved can be considered. In other words, this can be summarized as follows.

[Table] As mechanical linear scanning, there is a belt-type linear scanning shown in FIG. In this system, a large number of transducers 4 are arranged on a belt 3, and while the belt is rotated, the ultrasonic beam is scanned by relatively moving the rectangular steel piece 2. Furthermore, as the mechanical linear scanning, it is also possible to employ a vibrating linear scanning as shown in FIG. This scanning means arranges a plurality of transducers 4 in a row, vibrates them in the width direction of the square steel piece 2, and moves them relative to each other to scan the ultrasonic beam. What is shown in FIG. 10 is mechanical quasi-linear scanning, in which a large number of vibrators 4 are arranged on the outer circumference of a disk 5, and while rotating the disk 5, a square steel piece 2 is scanned relative to each other. It moves to scan the ultrasonic beam. As a mechanical sector scan, there is one shown in FIG. 11, in which the vibrator 4 is fixed to a reflection plate 6, and the reflection plate 6 is vibrated to swing and scan the ultrasonic beam S against the square steel piece 2. It is something. Next, electronic scanning will be explained. For electronic scanning, an electronic scanning array type probe 7 as shown in Fig. 12 is used.
is used. This probe 7 is constructed by arranging a large number of transducer elements 8 in a line on the plane of a base 9, and applying a coating 10 to the surface thereof. The ultrasonic beam from the probe is controlled by adjusting the timing of wave transmission and reception of each element 8 using a delay time control circuit. For example, if the delay time is not set as shown in FIG. When set, the beam S can be tilted as shown in Figure 13a, or the beam S can be tilted as shown in Figure 13b.
It is possible to freely form a wavefront that narrows the beam S as shown in FIG. 13, or narrows and tilts the beam S as shown in FIG. The characteristics of electronic linear or sector scanning, in which such a probe is electronically scanned and the ultrasonic beam is translated or swung onto the incident surface, are summarized as follows. (i) High-speed scanning ability High-speed scanning is easier than mechanical scanning. (ii) Sharp directivity Since an electronic scanning probe operates many transducer elements simultaneously, the probe as a whole is the same as a large-diameter transducer, and has sharp directivity. (iii) Electron focusing As mentioned above, electronic scanning probes can be used for flaw detection by narrowing the beam and increasing resolution, similar to concave vibrators and lensed vibrators, by giving a predetermined delay time to the transmitted and received signals. enable. Since this focal length can be set arbitrarily, by converging the beam on the flaw detection area in the material, the accuracy of detecting minute defects can be improved, and at the same time, the accuracy of estimating the defect position can also be improved. For reference, Fig. 14 shows the relationship between the ultrasonic beam diameter during flaw detection of a square steel piece and the S/N for a φ2 horizontal hole. (iv) Since the ultrasonic beam can be scanned with the probe fixed, a wide flaw detection area can be obtained with one probe. Next, a case where electronic linear scanning is performed using the electronic scanning array type probe 7 having such characteristics,
A case of electronic sector scanning will be compared and explained. First, in the case of electronic linear scanning, the probe 7 is set in a predetermined state, and the ultrasonic beam S is moved parallel to the incident surface as shown in FIGS. The flaws are detected on the surface layer side of the adjacent side surface. An example of the scanning circuit required in this case is shown in FIG. That is, for example, by using a linear array probe 7 with a total of 8 elements (64) as one set of 16 elements 8, the probe 7 and the transceiver 11 are connected to the delay circuit 12 described above as well as a reed relay circuit. 13, and the transmitting/receiving elements 8 are sequentially shifted using a changeover switch to scan the ultrasonic beam. On the other hand, in the case of electronic sector scanning, the probe 7 is set in a predetermined state, and the ultrasonic beam S is swung toward the incident surface as shown in FIGS. 17a, b, and c. The flaws are detected on the surface layer side of the lower half of the adjacent side surface. An example of the scanning circuit required in this case is shown in FIG. That is, for example, the total number of divided elements
A delay circuit 12 connects the 32 probes 7 and the transmitter/receiver 11.
By sequentially changing the delay time setting by the delay circuit 12, the inclination angle of the ultrasonic beam is changed and scanning is performed by swinging the beam. Comparing the above-mentioned electronic linear scanning and electronic sector scanning, the latter can be said to be advantageous in that the following disadvantages mentioned above are overcome. (i) In the case of electronic linear scanning, the transmitting and receiving elements are sequentially shifted, which increases the total number of elements and increases the overall transducer diameter. (ii) A large number of relays are required for switching, and the lifespan of these relays is short. (iii) Since the transmitter/receiver elements are sequentially shifted, there is no one-to-one correspondence between the transmitter/receiver (T/RUnit) and the element, making it difficult to adjust sensitivity variations. (iv) Since the amount of movement of the incident point of the ultrasonic beam is large, as shown in FIG. 19, the refraction angle changes due to the influence of the surface unevenness of the square steel piece 2, and the defect position estimation accuracy deteriorates. On the other hand, in electronic sector scanning, there is no need to sequentially shift the transmitting and receiving elements, so the total number of elements is small, and the relay 13 is not required.
However, since one handset 11 corresponds to one element 8, variations in sensitivity can be easily adjusted. Furthermore, in electronic sector scanning, since the amount of movement of the incident point of the ultrasonic beam is small, it is less susceptible to the effects of irregularities on the incident surface, as shown in FIG. Thus, in the case of electronic sector scanning, there is no particular problem in practical use, but in order to further reduce the influence of irregularities on the incident surface, it is desirable to use electronic sector + linear scanning, which will be described below. As shown in FIG. 21I, electronic sector + linear scanning is performed to correct the deviation of the incident point of the ultrasonic beam S due to electronic sector scanning. That is, as shown in FIGS. 21a, b, and c, the transmitting and receiving elements are shifted (linearly scanned) according to each beam inclination angle, thereby minimizing the deviation of the incident point, and performing electronic sector scanning. As described above, the defect position estimation accuracy is improved by making it less susceptible to the influence of the entrance surface. Next, a method of setting a probe for performing electronic scanning having such characteristics will be explained. In an electronically scanned array type probe, the transducer element spacing d required to prevent the occurrence of grating globe is d<λ/1+|, assuming that the maximum inclination angle of the ultrasound beam S is ±io. sin io | λ; expressed by the wavelength in the ultrasonic propagation medium. Therefore, in order to increase the inclination angle of the ultrasonic beam S, the element spacing d must be reduced when λ is constant (Figures 22a and b show the relationship between the element width and the maximum beam inclination angle io). ). That is, the element width needs to be made small. Here, the grating globe refers to the generation of a beam with high intensity in a direction other than the desired direction as shown in FIG. 13 described above. Particularly in the case of water immersion or local water immersion as described in the acoustic coupling method described below,
Since the wavelength λ in water is small, it is important to suppress the generation of grating globe by reducing the element spacing d. Now, we will specifically explain how to set up the probe when performing electronic linear scanning. In this case, there are two methods: (i) One method is to position the probe at a predetermined distance from the surface of the rectangular steel piece in a plane perpendicular to its axial direction, and set the probe parallel to the plane of incidence.
However, in this case, the width of each element of the probe is made smaller and the number of divisions is increased. A feature of this setting method is that since the maximum beam inclination angle is large, it is possible to detect flaws on both sides adjacent to the entrance surface. (ii) The other method is to position the probe at a predetermined distance from the surface of the rectangular steel piece in a plane perpendicular to its axial direction, and tilt the probe at a predetermined inclination angle i ⁰ ₀ with respect to the incident plane. This method is to set the A feature of this setting method is that by setting the above-mentioned inclination angle i ⁰ ₀ to be the same as the incident angle, there is no need for a delay operation of each element to incline the beam. Further, the divided element width of the probe may be large, and when comparing probes with the same element division, a larger angle of incidence can be obtained than when the probe is set horizontally. Now, when the probe 7 is set in the predetermined state as described above with respect to the square steel piece 2, and the situation is conceptually shown when performing linear scanning, it is as shown in Fig. 15 a, b, and c. . FIG. 23 shows an example of how the probes 7 are arranged on the square steel piece 2. In this case, one probe 7 is arranged horizontally on each surface of the square steel piece 2. That is, one probe 7 is used to detect flaws on one side or both sides of the square steel piece 2 adjacent to the entrance plane as shown in the figure, but by arranging four probes 7, flaws can be detected. Thus, the entire surface layer of the square steel piece 2 can be inspected for flaws. In addition, when testing both sides, the entire surface layer will be tested redundantly from two directions, but if redundant testing is not required, one probe will be placed on each of the adjacent entrance surfaces of the square steel piece 2. If arranged, it becomes possible to detect flaws on the entire surface layer with two probes. Then electronic sector scan or electronic sector +
The method of setting the probe for linear scanning will be explained. (i) One method is to position the probe at a predetermined distance from the surface of the square piece of steel in a plane perpendicular to the axial direction of the piece, and set the probe horizontally to the plane of incidence.
However, in this case, the width of each element of the probe is made smaller and the number of divisions is increased. A feature of this setting method is that since the maximum beam inclination angle is large, it is possible to detect flaws on both sides adjacent to the entrance surface. (ii) The other method is to position the probe at a predetermined distance from the surface of the square steel piece in a plane perpendicular to the axis of the piece, and to make the probe the center of incidence for sector scanning with respect to the incidence plane. This is a method of setting it by tilting it by an angle. The feature of this setting method is that the absolute value of the beam inclination angle becomes small, so
The maximum delay time can be small. Furthermore, since the absolute value of the beam inclination angle becomes small, the number of element divisions of the probe may be small. Now, the situation when the probe 7 is set in the above-mentioned predetermined state with respect to the square steel piece 2 and the sector is scanned is conceptually shown as shown in Fig. 17 a, b, and c. . The state of electronic sector + linear scanning is conceptually shown in FIGS. 21a, b, and c. 24 and 25 show an example of how the probe 7 is arranged with respect to the square steel piece 2. First, in the case of the example shown in FIG. 24, one probe 7 is arranged parallel to each surface of the square steel piece 2. That is, as shown in the figure, one probe 7 detects flaws in the lower half of both sides of the square steel piece 2 adjacent to the entrance surface, and four probes 7 test the entire surface layer of the square steel piece 2. It is. On the other hand, in the case of the example shown in FIG. 25, two probes 7 are arranged at a predetermined angle to each surface of the square steel piece 2. That is, in this case, one probe 7
As shown in the figure, the lower half of one side of the square steel piece 2 adjacent to the entrance plane is inspected for flaws, and a total of eight probes 7 are used to detect flaws on the entire surface layer of the square steel piece 2. In either arrangement, high-speed flaw detection can be performed online over the entire surface layer of a square steel piece. Next, the probe 7 of the test material 2 used in the present invention
The acoustic coupling method will be explained. Acoustic coupling methods include a direct contact method and a water immersion method. In the direct contact method, the probe and the ultrasound entrance surface of the specimen are in contact through a thin film of water or other liquid couplant. The present invention can be applied to the case where the array type probe 7 is arranged parallel to the material surface as shown in FIGS. 23 and 24. However, in the case of direct contact, the efficiency of transmitting and receiving ultrasonic waves tends to fluctuate due to the thickness of the couplant and the roughness and irregular shape of the ultrasonic incident surface of the specimen 2. The water immersion method is a method in which ultrasonic waves emitted from a probe are propagated through water (or liquid) over a fairly long distance and then made to enter the specimen.As shown in Figure 26, A water tank penetrating method (Fig. 26a) in which the rectangular steel piece 2, which is the test material, is immersed in water 14 for a certain length and the ultrasonic wave incident surface of the test material 2 of the probe 7 is locally immersed. A local water immersion method (FIGS. 26b and 26c) that satisfies the above conditions can be considered. In the case of this water immersion method, the ultrasonic waves are first emitted from the probe 7 into the water 14.
Sound transmission and reception is more stable than the direct contact method. In the present invention, either may be used. Now, the above explanation is a detailed explanation of a specific means for "oblique angle flaw detection of the surface layer of the test material 2 using the ultrasonic probe 7". By introducing the angle angle flaw detection as described above, surface subcutaneous flaw detection can be performed from the surface of the material to be inspected 2 without a dead zone, but at this time there is a risk that surface defects (flaws on the surface) may also be detected at the same time. There is. These surface defects can be removed by chipping or grinding in the billet processing process, and do not pose a problem during secondary processing of wire rod and steel bar products. Therefore, as already mentioned at the beginning, it is necessary to distinguish and extract detected defects into surface defects and subcutaneous defects. Therefore, next, the surface defect discrimination process, which is the most important point of the present invention, will be explained. When estimating the defect position based on the detection time difference between the surface echo (S ₁ echo) and the defect echo F, the refraction angle θ
The settings of the ultrasonic beam S are known, and no matter which scanning method described above is used, the point of incidence of the ultrasonic beam S can be determined by calculation from the positional relationship between the probe 7 and the specimen 2. can. Now, assuming that the position of the incident point x ₀ is determined by calculation with the upper right corner of the square steel piece 2 shown in FIG. 27 as a reference, the defect position (x, h) is expressed by the following equation. x=x ₀ −ΔT×C/2sinθ h=ΔT×C/2cosθ ΔT; Time difference between S1 echo and defect echo F C; Sound speed in steel In reality, the spread of the ultrasonic beam, the irregular shape of the specimen 2 ( There is an error in the estimated defect position due to surface irregularities, sagging, etc.), tracking error of the probe 7, etc.
It is difficult to distinguish between surface defects and subcutaneous defects using oblique flaw detection information alone. Therefore, in order to distinguish between these, only the surface defects are separately detected using a surface defect detection method, and the surface defect information is subtracted from the above-mentioned oblique inspection information, which includes the surface defect and subcutaneous defect detection information. By doing so, only surface subcutaneous defects are extracted. As an information processing method for surface defect discrimination, basically all you have to do is subtract information 2 from information 1, as shown in Figure 28. However, as mentioned above, angle flaw detection is difficult to estimate defect position accuracy. Information 1 is less reliable than the defect position information (information 2) from surface defect detection because it includes many factors that reduce the defect position, and the defect position estimation error range is large, so it cannot be simply subtracted. That is, as shown in FIG. 29, for an actual surface defect 15, the defect position detected by the above-mentioned ultrasonic angle flaw detection is the position marked with ◎ in the figure, and information 1 is used.
If information 2 is obtained by assuming that the surface defect position detected by another surface defect detection device is the position marked with ○× in the figure, if these information are processed as is,
Each defect is determined to be a separate defect, and information 1 is left as subcutaneous defect information. That is, even though it actually has only surface defects 15, it is treated as a subcutaneously defective material. In order to prevent this, information 2 has a certain area, and information 1 that falls within that area is canceled, thereby preventing erroneous detection of a surface defect as a subcutaneous defect. Here, the area provided for information 2 is the area shown by the dashed line in FIG. 29, and needs to be larger than the defect position estimation error range of information 1 (the area indicated by the dotted line in the figure). Therefore, with respect to the estimated surface defect position (○× mark) of information 2, the width of ±ΔH in the width direction of the square steel piece,
In the depth direction, information 1 within the entire flaw detection area δ (area indicated by a dashed-dotted line) is canceled. In this way, information 2 can be given a certain area and information 1 can be determined to be a surface defect. Here, canceling the angle inspection information for the width of ±ΔH in the width direction of the square steel slab 2 means:
Even though a subcutaneous defect in that area is actually detected, it may be determined that the defect is not a subcutaneous defect. In order to reduce the probability that this contradiction will occur, the width of ±ΔH must be made small. That is, in oblique flaw detection, it is better to introduce a scanning method with high defect position estimation accuracy. In the above-mentioned linear scanning, the influence of irregularities in the shape of the material to be inspected is large, and the defect position estimation accuracy is worse than in the case of sector scanning. Therefore, sector scanning is preferred.
Furthermore, it goes without saying that sector+linear scanning, which minimizes fluctuations in the position of the incident point, is the most advantageous. In sector scanning and sector+linear scanning, the following two methods can be used in combination to further improve defect position estimation accuracy. One method is to make the ultrasonic wave incident from the center of the surface, where the influence of irregularities on the incident surface is considered to be the least. In other words, by doing this, Figure 30 a, b
As shown in the figure, even if the surface of the square steel piece 2 is uneven,
The central part of the surface can be approximated as being substantially flat. Another method is to change the apparent angle of refraction (So→
In order to correct S), the echoes from the corners are detected and the angle of incidence indicating the maximum value thereof is determined. Then, the slope of the incident surface is calculated from that value, and the incident angle is corrected so that the beam enters the desired flaw detection area. Next, a surface defect detection device for detecting the surface defect 15 will be explained. Surface flaw detection methods include magnetic particle flaw detection, eddy current flaw detection, surface wave flaw detection, and optical flaw detection, and equipment suitable for each method is well known. However, the detection ability differs depending on each of the above-mentioned surface defect detection methods and the type of surface defect. For example, the detection performance of each flaw detection method for various surface defects as shown in FIG. 32 is as shown in the following table.
15a in Figure 32 is cracked, 15b is broken, 15c
indicates oyster scratches.

[Table] That is, according to the above table, surface wave flaw detection detects all surface defects detected by oblique flaw detection, but magnetic particle flaw detection hardly detects anything other than the crack 15a. However, in surface wave flaw detection, if there is one surface defect, the propagation of sound waves to other areas is extremely reduced, so if there are many surface defects, it may be overlooked. As described above, each surface defect detection method has advantages and disadvantages, and the discrimination characteristics between the surface defect 15 and the subcutaneous defect differ depending on which method is combined with the angle inspection. Next, as a specific example of the present invention, surface layer defects and surface Information processing for discrimination from defects will be explained. First, a method for processing oblique flaw detection data will be explained. As shown in FIG.
Sector + linear scanning is performed in steps, and if there is an echo higher than the preset echo level (function value) within the flaw detection area for each step, the step number, echo height, and detection time where the maximum echo was obtained are displayed. , and the two-axis direction position of the square steel piece are taken in as flaw detection data, and first, the width direction position h of the square steel piece (see page 23) is calculated from the refraction angle and detection time. Next, in order to evaluate the continuity of defects in the axial direction of the square steel piece, position information (estimated projected position) obtained when the width direction position is divided at a constant resolution is substituted. FIG. 34a shows a cross section of the square steel piece 2, 17 indicates the detected defect position, and 18 indicates the projected position information of the defect. FIG. 34b shows the projected surface matrix of the square steel piece 2, where P1 is the dividing pitch in the width direction;
2 is the flaw detection pitch in the axial direction of the square steel piece. Based on the estimated defect projection position information 18 on this matrix, the continuity of defects in the axial direction of the steel piece is evaluated. This information 1
If 8 does not occur consecutively two or more times in the axial direction of the square steel piece, it is determined to be noise 19 and is not treated as defect information. Therefore, information 18 that occurs two or more times consecutively is treated as defect information. That is, the detected defect 17 is the third
As shown by the thick frame in Figure 4b, the final defect No.
The information includes the defect origin (in the axial direction), defect length, and position in the width direction. In addition, in FIG. 33, the flaw detection gate 16 in the flaw detection area is designed for a long period of time so that flaw detection can be performed completely from the surface, taking into account changes in the path due to deformation of the test material 2 and the like. However, at corners, limit the range so that corner echoes are not picked up. Here, we have given an example in which the flaw detection area is sector + linear scanned in 8 steps using an electronic scanning flaw detection device, but it is clear that increasing the number of steps allows for more detailed flaw detection and improves discrimination performance. Next, a method of processing surface defect detection data will be explained. As shown in FIG. 35, a tire probe 22 is used for surface wave flaw detection. In this case, the flaw detection area is the surface facing the ground side surface of the probe 22, and is divided into three parts and gated on each surface and corner part. For the first gate 23 and second gate 24, up to three echoes exceed the threshold, and from the corner gate 25, for the maximum echo exceeding the threshold,
Each echo height, detection time, and position in the axial direction of the square steel piece were taken in as flaw detection data, and the estimated position of the defect in the axial direction of the square steel piece was calculated from the detection time, similar to the aggregation of oblique flaw detection data, and divided at a constant pitch. The information is converted to the position information in the width direction of the square steel piece, and the continuity in the axial direction of the square steel piece is checked to obtain information on the defect number, defect origin (axial direction), defect length (axial direction), and position in the width direction. The information format for magnetic particle flaw detection data from magnetic particle flaw detection equipment will be the same as the aggregated information for angle angle flaw detection and surface wave flaw detection. Then, when discriminating only the subcutaneous defect information from the information obtained by the above-mentioned angle flaw detection and the information obtained by surface defect flaw detection, as described above, surface defect area information is obtained by adding a certain area to the surface defect information. , this information is subtracted from the angle angle flaw detection aggregate information. The processing block diagram is shown in FIG. The above surface defect area information includes the width of several blocks in the width direction of a square steel strip, defect origin and length information, in addition to the width direction position information of surface defect aggregation information of surface wave flaw detection, magnetic particle flaw detection, or both. The area is extended by 10 mm. FIG. 37 shows the concept of the discrimination processing of the present invention, and shows that the combination of surface wave flaw detection and magnetic particle flaw detection prevents erroneous detection in which surface defects are treated as subcutaneous defects. The probe arrangement for full-surface flaw detection during oblique flaw detection has already been described in FIGS. 23 and 24. Therefore, to explain the arrangement of tire probes for full surface flaw detection, as shown in FIG. It is sufficient to provide two. Of course, the test material is moved in the axial direction. In the above embodiment of the present invention, by combining an electronic scanning flaw detection device, a surface wave flaw detection device, and a magnetic particle flaw detection device, surface defects and subcutaneous defects can be identified and determined. There are fewer oversights than with flaw detection. In particular, magnetic particle testing has low detection ability for defects at corners, so
You can use the flaw detection information from the second gate of tire flaw detection. It should be noted that the present invention is not limited to the above-mentioned embodiments, and the material to be tested may be in the shape of a round bar. It is possible to perform flaw detection on the surface layer and the surface at the same time using this method, and to discriminate the defects. As detailed above, the present invention does not distinguish a defect as a subcutaneous defect when it is detected by both an angle flaw detector and a surface defect detector, as in the conventional subcutaneous defect discrimination method. , subcutaneous defects are discriminated by subtracting the defect information obtained by the surface defect detection device from the defect information obtained by the angle inspection device. Therefore, the probability that the discriminated defect is a subcutaneous defect is higher than that of the conventional example. In addition, since there is a difference between the accuracy of defect position estimation by an angle flaw detector and the same accuracy by a surface flaw detector, it becomes difficult to judge whether or not the defects detected at the same time are the same defect. Since a predetermined area is given to the defect position detected by the surface defect detection device with good quality and subtracted from the defect information obtained by the angle angle flaw detection device, erroneous detection of subcutaneous defects is prevented. Furthermore, according to the present invention, a longitudinal wave angle flaw detection device using an electronically scanned array type probe is used to detect the probe at a predetermined distance from the surface of the steel piece in a plane perpendicular to the axial direction of the steel piece. Moreover, by setting the probe at a predetermined angle to the material surface and scanning the probe using a combination of electronic sector scanning and electronic linear scanning, the incident point on the material surface is kept approximately constant and the angle of incidence on the material surface is kept approximately constant. The surface layer inside and on the side surface adjacent to the entrance plane is inspected to detect defects in the surface layer of the material to be inspected, and the positions of the surface layer defects are estimated. Therefore, the surface layer defect position estimation accuracy is improved, and subcutaneous defects can be detected with high accuracy. That is, the surface of the steel material has irregularities as shown in FIGS. 19, 20, and 30, and it is impossible to specify the position of surface layer defects using the methods shown in FIGS. 19 and 20. Therefore, it is not possible to extract only subcutaneous defects by subtracting the detection results of surface defects from all detected defects. By the way, generally, the shape of the steel material is stable in the longitudinal direction of the central part of the steel material, which is concave or convex. Therefore, in the present invention, the incident position of the ultrasonic wave in the oblique ultrasonic flaw detection method for detecting flaws in the surface layer is configured to be at a certain position on the incident surface (injecting from the central part. The accuracy of defect location estimation has increased, resulting in an increase in flaw detection accuracy.

[Brief explanation of drawings]

Fig. 1 is a diagram showing a dead zone by the conventional vertical flaw detection method, Fig. 2 is a diagram showing a side reflection echo by the oblique flaw detection method according to the present invention, Fig. 3 is a diagram showing the bottom reflection echo, and Fig. 4 is a graph showing the relationship between reflected echo height and refraction angle, Figure 5 is a diagram showing the refraction angle in the case of bottom echo, Figure 6 is a diagram showing the refraction angle in the case of side echo, Figure 7 is the angle of incidence, Graph showing the relationship between refraction angle and reciprocating pass rate, Figure 8 is a diagram of the principle of belt-type linear scanning, Figure 9 is a diagram of the principle of vibration-type linear scanning, Figure 10 is a diagram of the principle of mechanical pseudo-linear scanning,
Figure 11 is a diagram of the principle of mechanical sector scanning;
The figure is a perspective view of an electronically scanned array type probe, Figure 13 is a diagram showing how the ultrasonic beam is controlled by the electronically scanned array type probe, and Figure 14 is a diagram showing the relationship between the ultrasonic beam diameter and S/N. Fig. 15 is a diagram showing the principle of electronic linear scanning, Fig. 16 is a control circuit diagram of electronic linear scanning, Fig. 17 is a diagram showing the principle of electronic sector scanning, and Fig. 18 is a control circuit diagram of electronic sector scanning. , Figure 19 is a diagram showing variations in electronic linear scanning, Figure 20 is a diagram showing variations in electronic sector scanning, Figure 21 is a diagram of the principle of electronic sector + linear scanning, and Figure 22 is the element width and beam inclination angle. 23 is a diagram showing the arrangement of the linear scanning probe, FIGS. 24 and 25 are the arrangement diagram of the sector scanning probe, and FIG. 26 is a diagram showing the acoustic coupling method. Fig. 27 is a diagram showing the defect position, Fig. 28 is a flowchart of the discrimination process, Fig. 29 is a diagram of the principle of the discrimination process, Fig. 30 is a diagram showing the incident point of the ultrasonic beam, and Fig. 31 is a diagram showing the defect position. Figure 32 is a diagram showing the type of surface defects; Figure 33 is a diagram showing electronic sector scanning;
Figure 4 is an explanatory diagram for aggregating detected defects, No. 35
36 is a flowchart of the discrimination process, FIG. 37 is a diagram showing the concept of the discrimination process, and FIG. 38 is a diagram of the arrangement of the tire probes. 2... Square steel piece (test material), 7... Probe, 15
...Surface defect, 17...Detected defect position, 18...
Projection position, 20... Defect aggregate information, 22... Tire probe (surface defect detection device), S... Ultrasonic wave,
α……Incidence angle.

Claims (1)

[Scope of Claims] 1. A longitudinal wave angle flaw detection device using an electronically scanned array type probe is used to detect the probe at a predetermined distance from the surface of the steel piece in a plane perpendicular to the axial direction of the piece. In addition, by setting the probe at a predetermined angle to the material surface and scanning the probe using a combination of electronic sector scanning and electronic linear scanning, the incident point on the material surface is kept approximately constant and the inside of the square steel piece is scanned. and the surface layer on the side surface adjacent to the entrance surface, detects defects in the surface layer of the material to be inspected, estimates the position of the surface layer defect, and uses the position as information 1, and the surface defect detection device Detects a surface defect on the material to be inspected, estimates the position of the surface defect, and calculates the range by giving a range larger than the position estimation error by the angle angle flaw detection device, centering on the surface defect position. A method for detecting subcutaneous defects by ultrasonic flaw detection, characterized in that only subcutaneous defects are detected by defining a surface defect area as information 2, and subtracting information 2 from information 1.