JP2012127812A - Method and device for quality evaluation of billet - Google Patents

Abstract

PROBLEM TO BE SOLVED: To perform flaw detection of a wide range at one time by reducing a surface dead zone without heightening frequency and by widening a flaw detection range in a depth direction.SOLUTION: When flaw detection of a continuously casted billet is performed by ultrasonic waves and quality of the billet is evaluated based on the flaw detection result, a first piezoelectric vibrator 20 which can transmit a line focus ultrasonic beam for performing flaw detection of a surface layer part of a material 8 to be inspected through water, and a second piezoelectric vibrator 30 having a line focus field, are arranged facing each other across a sound isolation plate 40 so that focuses of an ultrasonic beam 21 transmitted from the first piezoelectric vibrator 20 and an ultrasonic beam range (a reception signal field 31) to be received by the second piezoelectric vibrator 30 intersect at a predetermined depth F in the billet, to perform the ultrasonic flaw detection.

Description

  The present invention relates to a steel piece quality evaluation method and apparatus for ultrasonically testing a continuously cast steel piece and evaluating the quality of the steel piece based on the flaw detection result.

  A thin steel plate is manufactured through various processes such as continuous casting, hot rolling, pickling, cold rolling, and galvanization. Defects that are problematic in the final product are surface defects that are problematic in appearance and defects that are directly under the surface layer that are manifested by pressing. Defects may be generated in each of the above-described steps. For example, examples of castability defects include alumina inclusions and powder inclusions. It is also said that bubbles under the surface of the slab appear on the surface when the scale is turned off in a hot-rolling heating furnace, and the scale is clogged in the hole. Usually, these inclusions and bubbles are directly below the surface of the continuously cast steel slab, so defects can be found directly below the surface of the continuously cast steel slab as much as possible by electromagnetic stirring and braking techniques in the mold during continuous casting. Efforts are made to prevent the occurrence of

  As one of the methods for evaluating the quality of continuously cast steel slabs to determine whether the casting conditions for continuous casting are appropriate, the specimen to be inspected is extracted from the continuously cast steel slab and the surface of this sample is ground for several millimeters. There is a method of evaluating the quality by smoothing the surface, flaw-detecting it with ultrasonic waves, and statistically examining the distribution of bubbles and inclusions.

  The ultrasonic flaw detection method is used for flaw detection in steel products (bars, plates, pipes, etc.). A generally well-known ultrasonic flaw detection method is the vertical pulse echo method, in which the ultrasonic probe and the material to be inspected are acoustically coupled with oil or water, and the ultrasonic probe is perpendicular to the material to be inspected. In this method, an ultrasonic wave is transmitted to an echo and an echo reflected by a defect existing inside is received by an ultrasonic probe. This method can sufficiently detect the central part of the thickness of the material to be inspected, but in the extreme surface layer (several millimeters below the surface), the transmission pulse or surface echo becomes a dead zone (an area that cannot be detected). Can not be flawed.

  Therefore, as a method for detecting defects on the surface layer portion of the material to be inspected, a method of increasing the frequency of ultrasonic waves (for example, 50 MHz) is generally performed.

(1) A method of increasing the frequency of ultrasonic waves In a method of detecting flaws by water using a vertical pulse echo method, as shown in FIG. The frequency of the ultrasonic beam 11 to be transmitted is increased, the vibration diameter is increased, and the ultrasonic beam 11 is focused by an acoustic lens or the like, so that a strong reflected wave (F echo) is generated from the defect 8F near the focal point. 1), and as shown in FIG. 1B, the time width of the reflected wave (S echo) received from the surface 8S to be inspected by high frequency is shortened to reduce the dead zone as much as possible. To do. Although the dead band due to surface echo is not completely eliminated, for example, when the frequency is 5 MHz, the dead band of about 4 to 6 mm can be reduced to about 1 to 2 mm when the frequency is increased to 10 times 50 MHz. .

  However, if the ultrasonic frequency is increased and the ultrasonic beam is focused, the flaw detection range in the depth direction is also very limited. Therefore, after measuring the flaw detection range, A method has been implemented in which flaw detection is repeated while cutting in the depth direction. In this method, for example, even if only a depth range of 5 mm is measured, flaw detection and processing of the object to be inspected are repeated. As the frequency is increased in order to reduce the dead zone, the depth range in which flaw detection can be performed becomes narrower. Therefore, it is necessary to increase the number of inspections and flaw detections, and time is required.

(2) Method described in Patent Document 1 Therefore, in Patent Document 1, in the technique in which the frequency of the ultrasonic wave is increased, a multiple reflection echo generated from a defect in the extreme surface layer portion is detected, so that it has conventionally been in the dead zone. It is possible to detect defects that are hidden and cannot be detected.

  However, due to the influence of the defect shape (such as a slanted or triangular pointed object), multiple reflection echoes are less likely to occur, and undetected defects are likely to occur. Furthermore, since multiple reflection echoes are used, the depth of the defect cannot be accurately known, and the exact depth is not known. Therefore, the defect size is determined from the signal amplitude based on the DGS characteristic (distance-signal amplitude characteristic). There is a problem that cannot be estimated more accurately. Furthermore, this method also has the problem described in (1) because it transmits and receives high-frequency ultrasonic waves.

JP-A-1-237449

  The conventional method of ultrasonically testing continuously cast steel slabs and evaluating the quality of the steel slabs based on the results of the flaw detection results in a depth direction flaw detection because the depth direction flaw detection range is narrow. The larger the area, the greater the number of flaw detections and sample processings required to evaluate the quality of the defect, and there was a problem that it took longer to obtain the result.

  The present invention has been made to solve the above-described problem, and an object of the present invention is to enable quick evaluation of continuously cast steel pieces using ultrasonic waves.

  The present invention is a line focus type in which a continuous cast steel slab is flaw-detected with ultrasonic waves and the surface layer portion of a material to be inspected is detected through water when the quality of the steel slab is evaluated based on the flaw detection result. An ultrasonic beam transmitted from the first piezoelectric transducer, and a first piezoelectric transducer capable of transmitting the ultrasonic beam and a second piezoelectric transducer having a line-focused field of view. Ultrasonic flaw detection is performed by placing the acoustic separators facing each other so that the focal points of the ultrasonic beam range for reception by the second piezoelectric transducer intersect at a predetermined depth in the steel. This solves the problem.

  Here, the ultrasonic waves can be focused in the long side direction of the first and second piezoelectric vibrators, and the ultrasonic waves can be diffused in the short side direction.

  Further, the propagation path of the ultrasonic wave is calculated from the depth of the defect and the arrangement of the first piezoelectric vibrator and the second piezoelectric vibrator, and the depth position of the defect is calculated based on the calculated propagation path. It can be corrected.

  Further, a plurality of the predetermined steel intermediate depths can be provided, and the first and second piezoelectric vibrators can be arranged for each of the steel intermediate depths.

  The present invention also provides a steel piece quality evaluation apparatus for ultrasonically testing a continuously cast steel piece with ultrasonic waves and evaluating the quality of the steel piece based on the flaw detection result. A first piezoelectric vibrator capable of transmitting a line-focused ultrasonic beam for flaw detection on a surface layer portion and a second piezoelectric vibrator having a line-focus-like field of view are connected to the first piezoelectric vibrator. The ultrasonic beam transmitted from the child and the focal point of the ultrasonic beam range for reception by the second piezoelectric transducer are opposed to each other with the acoustic separator interposed therebetween so as to intersect at a predetermined depth in the steel. By providing the ultrasonic flaw detection means arranged, the above-mentioned problem is solved in the same manner.

  Here, the first and second piezoelectric vibrators can focus ultrasonic waves in the long side direction and diffuse ultrasonic waves in the short side direction.

  Further, the propagation path of the ultrasonic wave is calculated from the depth of the defect and the arrangement of the first piezoelectric vibrator and the second piezoelectric vibrator, and the depth position of the defect is calculated based on the calculated propagation path. Means for correcting can be provided.

  Further, a plurality of the predetermined steel intermediate depths can be provided, and the first and second piezoelectric vibrators can be arranged for each of the steel intermediate depths.

  The inspection object extracted from the continuously cast steel pieces has less dead zone than the conventional technology, and the inspection range in the depth direction can be increased to detect a wide range at a time. The time required for body quality evaluation is greatly shortened, and quick feedback to the production conditions for continuously cast steel slabs becomes possible, making it possible to produce steel slabs of higher quality.

Cross-sectional view and waveform image diagram for explaining conventional problems 1A is a cross-sectional view and FIG. 1B is a top view illustrating a basic configuration of an embodiment of the present invention. A perspective view showing an example of a piezoelectric vibrator usable in the present invention Sectional drawing which shows arrangement | positioning in the said embodiment The figure which similarly shows the example of the relationship between the distance (incident angle) between incident position P1-P2 for every depth in steel of a defect, and S / N Sectional view illustrating the principle of the present invention The block diagram which shows the whole structure of 1st Example of this invention. A perspective view showing the flaw detection situation Sectional view of the sensor head Flow chart showing the processing procedure (A) Cross-sectional view and (B) Correlation diagram showing the same calculation method The figure which shows the correlation by (A) conventional method and (B) this invention method The figure which compares and shows the flaw detection range by a conventional method and 1st Example Sectional drawing which shows the principal part structure of 2nd Example of this invention. The figure which compares and shows the flaw detection range by a conventional method and 2nd Example

  Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 2A is a cross-sectional view showing the basic configuration of the present embodiment, and FIG. 2B is a top view of the same. In the figure, 20 is a first piezoelectric transducer, 21 is an ultrasonic beam transmitted from the first piezoelectric transducer 20, 30 is a second piezoelectric transducer, and 31 is the second piezoelectric transducer. Received signal field of the child 30, 40 is an acoustic separator, φ1 is an angle with respect to the normal of the surface 8S of the first piezoelectric vibrator 20, and φ2 is a surface of the second piezoelectric vibrator 30 being inspected An angle with respect to the normal line of 8S, Dc is a measurement depth range, and F is a focal depth in steel.

  In this embodiment, the acoustic coupling method is a water immersion method, and transmission and reception of ultrasonic waves are performed by dividing the first piezoelectric transducer 20 and the second piezoelectric transducer 30. As shown in the cross-sectional view shown in FIG. 2A and the top view shown in FIG. 2B, the first piezoelectric vibrator 20 and the second piezoelectric vibrator 30 are line-focused ultrasonic waves. The ultrasonic wave is unfocused and diffuses in the direction perpendicular to the surface where the first piezoelectric transducer 20 and the second piezoelectric transducer 30 face each other, and the parallel plane is The shape of the vibrator is determined so that the ultrasonic wave is focused. For example, a rectangular piezoelectric vibrator is prepared, and a curvature is given to converge on the long side as shown in FIG. 3, and the flat side is kept flat so as not to converge on the short side. Alternatively, an acoustic lens may be attached so that a line-focused beam can be obtained on a flat piezoelectric transducer. The reason for focusing on the long side here is to prevent the ultrasonic waves from diffusing too much and detecting small defects.

  The arrangement of the first piezoelectric vibrator 20, the second piezoelectric vibrator 30, and the acoustic separator 40 will be described in detail with reference to FIG. In FIG. 4, P <b> 1 is a crossing position between the sound pressure central axis of the first piezoelectric vibrator 20 and the surface 8 </ b> S to be inspected, and P <b> 2 is the sound pressure central axis of the second piezoelectric vibrator 30 and the surface of the material to be inspected. Crossing position with 8S, Wd is the distance between positions P1 and P2, Pc is the focal position in the specimen (steel) 8, d1 is the distance between positions Pc and P1, and d2 is the distance between positions Pc and P2. is there.

  The distances d1 and d2 and the angles φ1 and φ2 may be different values. However, assuming that ideal ultrasonic reflection occurs at the focal position Pc, the distances d1 and d2 and the angles φ1 and φ2 are the same value. Thus, the first piezoelectric vibrator 20 and the second piezoelectric vibrator 30 are preferably arranged in line symmetry. When the first piezoelectric vibrator 20 and the second piezoelectric vibrator 30 are arranged so as to be line symmetric, the acoustic separator 40 is arranged on the center axis of line symmetry.

  If the depth F in steel to be flawed is set in advance before the flaw detection work, a path through which the ultrasonic wave propagates from the positions P1 and P2 to the position Pc in the shortest is obtained, and the angles θ1 and θ2 can be calculated. From the angles θ1 and θ2, the angle φ1 and the angle φ2 are calculated for each position P1 and P2 using Snell's law expressed by the following equation.

Where φ: incident angle θ: refraction angle Vw: underwater sound velocity Vs: steel sound velocity

  It is preferable to arrange the position P1 and the position P2 where the S / N can be taken most with respect to a certain steel depth F in a chart. For the creation of the chart, a horizontal hole artificial barb and a flat bottom hole artificial bark are processed for each depth F, and the optimum conditions are investigated based on this, for example, as shown in FIG. Based on the arranged results, the flaw detection is performed by selecting the optimum condition for each depth F.

  In addition, although it is suitable for making arrangement | positioning easy that position P1, P2, and Pc are the same straight line form, it is not limited to this.

  With the arrangement shown in FIG. 4, the first piezoelectric vibrator 20 and the second piezoelectric vibrator 30 face in a direction different from the regular reflection direction. Therefore, as shown in FIG. 6, the signal transmitted from the first piezoelectric transducer 20 and reflected by the surface 8S of the material 8 to be inspected becomes difficult to be received by the second piezoelectric transducer 30, and the surface The amplitude of the reflected wave (S echo) is reduced. In addition, an acoustic separator 40 is provided to reflect the reflected wave on the surface 8S to be inspected and to block the scattered wave leaking into the second piezoelectric transducer 30, thereby further reducing the amplitude of the surface reflected wave and the surface dead zone. Can be reduced as much as possible.

  Also, by arranging the line-focused beam diffusing side to intersect the depth direction, the reflected wave from the defect in the intersecting range, that is, the measured depth range Dc in FIG. 2 is received. Can do.

  In the conventional technique, in order to narrow the dead zone of the S echo, the ultrasonic waves are focused in order to increase the frequency and receive the reflected wave from the defect with good S / N. As the dead zone is shortened, there is a conflicting problem that the frequency range capable of flaw detection becomes narrower while higher frequency and focusing of ultrasonic waves are required.

  According to the present invention, it is possible to solve the problems in the reciprocal relationship, to reduce the surface dead zone and to detect a wide range in the depth direction without increasing the frequency.

[First embodiment]
A first embodiment of the present invention will be described with reference to FIGS.

  In FIG. 7, 50 is a sensor head, 52 is an ultrasonic transmission unit, 54 is an ultrasonic reception unit, 56 is an A / D conversion unit, 60 is a signal processing unit 64, an ultrasonic transmission / reception control unit 66, and a scanner control unit. A computer including ultrasonic flaw detection control software 62 including 68, 70 a scanner mechanism unit, and 72 an output unit.

  Inspected material 8 (width Sc / 2 [m] × length Sl [m] × thickness) cut out from a continuously cast steel piece (width Sc [m] × length Sl [m] × thickness St [mm]) As shown in FIG. 8, St [mm]) is installed in a large water tank 80, and the sensor head 50 is scanned in the XY directions by the scanner mechanism 70 shown in FIG.

  The sensor head 50 will be described with reference to FIGS. In FIG. 9, 52 is a shoe that contacts the material 8 to be inspected. As shown in FIG. 9, the sensor head 50 has a structure in which water is supplied from the outside and can be acoustically coupled by a local water immersion method. Since the material to be inspected is large, the all-immersion method sinks the material to be inspected in a tall and large water tank. Therefore, it is necessary to change the water in the water tank every time the inspection material is installed and taken out with a lifting magnet (lifting magnet) or a crane, and the working efficiency is deteriorated. In this embodiment, the sensor head 50 capable of local water immersion shown in FIG. 9 is used. As shown in FIG. 8, the water used for acoustic coupling is received by the water tank 80 and circulated by the pump P. I was able to detect flaws. The time required for water replacement by the all-immersion method is almost zero, and the efficiency of the flaw detection work is improved compared to the all-immersion method. In addition, it is also possible to make it spill without circulating by the pump P.

  In the sensor head 50, the first piezoelectric vibrator 20 and the second piezoelectric vibrator 30 are installed so as to be focused at a depth F [mm] as shown in FIG. The incident angles φ1 and φ2 and the positions P1 and P2 of the incident points of the first piezoelectric transducer 20 and the second piezoelectric transducer 30 are a horizontal hole artificial with a diameter of 2 mm that is previously drilled at a depth F [mm]. The S / N was investigated using scissors, and the optimum value (the condition with the best S / N) obtained from the survey results was adopted.

  The ultrasonic transmission unit 52 has a frequency f [MHz], a wave number K [wave number], and a transmission voltage Vt [V], which are arbitrarily set in advance by the ultrasonic transmission / reception control unit 66, in the sensor head 50. The piezoelectric vibrator 20 is driven to transmit ultrasonic waves. The reflected wave from the material to be inspected 8 is received by the second piezoelectric transducer 30 in the sensor head 50, filtered and signal amplified by the ultrasonic receiver 54, and A / D converter 56 performs A / D converted and taken into the computer 60.

  In this embodiment, the frequency is 5 [MHz], the wave number is 1 [wave number], the transmission voltage is 100 [V], and the ultrasonic receiving unit 54 uses an analog bandpass filter of 3 MHz to 5 MHz.

  The scanner control unit 68 controls the scanner mechanism unit 70 to perform XY scanning. At this time, a timing signal is output for each set flaw detection density based on an encoder installed in the scanner mechanism unit 70 and a driving pulse signal. Using this timing signal as a synchronization signal, the ultrasonic transmitter 52 drives the first piezoelectric transducer 20, and the ultrasonic receiver 54 transmits the reflected wave from the material 8 to be inspected by the second piezoelectric transducer 30. Then, the ultrasonic reception unit 54 performs filtering and amplification processing, and the A / D conversion unit 56 performs A / D conversion.

  The signal processing unit 64 performs the processing shown in FIG. 10 on the reflected wave A / D converted by the A / D conversion unit 56 and taken into the computer 60.

  First, in scanning flaw detection (step 100), a waveform is stored in a memory (not shown) for each coordinate (x, y), and a synchronous addition signal process (aperture synthesis calculation) and a digital filter process are performed on the waveform at each coordinate. Then, full-wave rectification is performed, the maximum value Pxy in an arbitrarily set gate is calculated, the luminance of the maximum value Pxy is converted, and two-dimensional mapping is performed to create a C scope image (step 110). A primary determination process (step 120) is performed on the C scope image, and the maximum amplitude Pxy and the propagation time Txy of the soot echo are calculated at locations where the amplitude is larger than the arbitrarily set threshold value θ1. In this embodiment, the position of the -9 dB level of the maximum amplitude position of the soot echo is read. The method for reading the propagation time is not limited to the method in this embodiment.

  Based on the maximum amplitude Pxy and the propagation time Pt at the coordinates (x, y) determined in the primary processing, the estimated depth and the estimated diameter are calculated (step 130).

When calculating the depth Dxy [mm], it is simply assumed that the ultrasonic wave travels straight without being refracted, and the velocity of sound in steel is Vs [m / sec]. .
Dxy = Txy × Vs / 2 (2)

  However, in this embodiment, in consideration of refraction, a relational expression with an actual propagation path (propagation time) is prepared in advance so as to satisfy Snell's law for each depth as shown in FIG. Based on the propagation time, the depth is calculated.

  FIG. 12A is a comparison between the estimated depth and the actual depth calculated by the equation (2) without considering the actual propagation path of the ultrasonic wave, and FIG. 12B is the actual propagation path of the ultrasonic wave. This is a comparison between the estimated depth and the actual depth according to this embodiment in consideration of the above. In FIG. 12A, it is estimated that the defect at the shallow position is located at a position deeper than the actual depth. However, in the present embodiment, as shown in FIG. 12B, it can be seen that the depth can be estimated with higher accuracy than the case where the calculation is performed by the equation (2).

  After calculating the depth Dxy, the amplitude Pxy is corrected based on the depth Dxy. In the case of the present embodiment, since the flaw detection sensitivity decreases except in the vicinity of the focal depth F, an amplitude distance propagation characteristic is calculated in advance using an artificial scissors, and the amplitude Pxy is corrected based on this characteristic. And the estimated defect diameter Fxy is calculated using the conversion formula of amplitude and defect diameter investigated in advance.

  In the secondary determination process (step 140) of the ultrasonic instruction, a threshold S [mm] or more that is arbitrarily set in advance is extracted as a defect instruction, and the number of defects and the coordinates (X, Y) of each defect instruction. The estimated defect diameter Fxy and the estimated depth Dxy are output (step 150).

  FIG. 13 shows the effect of the present invention when a defect having an estimated diameter of φ0.5 mm or more is detected. In FIG. 13, the example of the flaw detection range shows an example of a depth range in which flaw detection is possible when flaw detection is performed at 50 MHz. By using the present embodiment, it can be seen that flaw detection can be performed from a shallower side than the conventional method to a wider side.

[Second Example]
A second embodiment will be described with reference to FIG. In the first embodiment, a single sensor head is used. However, it is possible to further expand the flaw detection range by combining sensor heads optimized for each flaw detection range.

  In the second embodiment, as shown in FIG. 14, the first piezoelectric vibrator 20 and the second piezoelectric vibrator 30 are arranged so that the focal point intersects at the focal depth F1 [mm], and the focal depth is obtained. The third piezoelectric transducer 22 and the fourth piezoelectric transducer 32 were arranged so that the focal points intersect at F2 [mm] (F1 <F2). The first piezoelectric vibrator 20 and the second piezoelectric vibrator 30 arranged with respect to the focal depth F1 [mm] are used as the first sensor head, and the first piezoelectric vibrator 20 arranged with respect to the focal depth F2 [mm]. The three piezoelectric transducers 22 and the fourth piezoelectric transducer 32 are used as second sensor heads, and flaw detection is performed by simultaneously scanning the inspection object 8 with the two sensor heads.

  FIG. 15 shows the results of the flaw detection range with an estimated diameter of 0.5 mm or more according to the second embodiment. FIG. 15 shows that a wider range of flaw detection is possible with this embodiment than with the first embodiment.

  It is possible to provide three or more sensor heads.

DESCRIPTION OF SYMBOLS 8 ... Test object 8F ... Defect 20, 22, 30, 32 ... Piezoelectric vibrator 21 ... Ultrasonic beam 31 ... Received signal visual field 40 ... Acoustic separator 50 ... Sensor head F, F1, F2 ... Depth of focus

Claims (8)

When a continuously cast steel slab is flaw-detected with ultrasonic waves and the quality of the steel slab is evaluated based on the flaw detection result,
A first piezoelectric transducer capable of transmitting a line-focused ultrasonic beam for flaw detection on the surface layer of the material to be inspected through water, and a second piezoelectric transducer having a line-focused visual field The acoustic beam is transmitted so that the ultrasonic beam transmitted from the first piezoelectric transducer and the focal point of the ultrasonic beam range for reception by the second piezoelectric transducer intersect at a predetermined depth in steel. A method for evaluating the quality of a steel slab, characterized in that ultrasonic flaw detection is performed by placing the separators facing each other across the separator.   The method for evaluating the quality of a steel slab according to claim 1, wherein the ultrasonic wave is focused in the long side direction of the first and second piezoelectric vibrators and the ultrasonic wave is diffused in the short side direction.   3. The ultrasonic propagation path according to claim 1 or 2, wherein an ultrasonic propagation path is calculated from the depth of the defect and the arrangement of the first piezoelectric vibrator and the second piezoelectric vibrator, and the defect is based on the calculated propagation path. A method for evaluating the quality of a steel slab characterized by correcting the depth position of the steel.   4. The method according to claim 1, wherein a plurality of the predetermined steel intermediate depths are provided, and the first and second piezoelectric vibrators are disposed for each of the steel intermediate depths. A method for evaluating the quality of steel slabs. In the steel piece quality evaluation apparatus for detecting the continuously cast steel piece with ultrasonic waves and evaluating the quality of the steel piece based on the flaw detection result,
A first piezoelectric transducer capable of transmitting a line-focused ultrasonic beam for flaw detection on the surface layer of the material to be inspected through water, and a second piezoelectric transducer having a line-focused visual field The acoustic beam is transmitted so that the ultrasonic beam transmitted from the first piezoelectric transducer and the focal point of the ultrasonic beam range for reception by the second piezoelectric transducer intersect at a predetermined depth in steel. An apparatus for evaluating the quality of a steel slab, comprising ultrasonic flaw detection means arranged opposite to each other with a separator interposed therebetween.   6. The first and second piezoelectric vibrators according to claim 5, wherein ultrasonic waves are focused in the long side direction and ultrasonic waves are diffused in the short side direction. Billet quality evaluation equipment.   7. The ultrasonic propagation path according to claim 5 or 6, wherein an ultrasonic propagation path is calculated from a defect depth and the arrangement of the first piezoelectric vibrator and the second piezoelectric vibrator, and the defect is determined based on the calculated propagation path. An apparatus for evaluating the quality of a steel slab comprising means for correcting the depth position of the steel slab.   In any one of Claims 5 thru | or 7, The said predetermined steel intermediate depth was made into multiple, and said 1st and 2nd piezoelectric vibrator was each arrange | positioned with respect to each steel intermediate depth, Steel bill quality evaluation device.