JP6970425B2 - Axle ultrasonic flaw detection method and its system - Google Patents

Description

The present invention relates to an ultrasonic flaw detection method for an axle and a system thereof.

In railcars, the inspection of the axle to which the wheel is fitted includes the state of using the vehicle, that is, the alternation inspection that inspects without disassembling, the bogie inspection that inspects by removing other than the bogie, and the inspection by disassembling the vehicle and taking out only the axle. There is a general inspection. In each of the above inspections, the area inside the axle is inspected for scratches near the outer circumference of the axle. Usually, the depth from the outer peripheral surface of the axle to be inspected in the above-mentioned alternation inspection, bogie inspection and general inspection is different.
In each of the above inspections of the axle, an ultrasonic flaw detection method is used as one of the above flaw detection methods.
Ultrasonic flaw detection is performed by scanning a probe that transmits and receives ultrasonic waves, that is, an ultrasonic probe. For example, in a hollow axle, that is, a boring axle, which is used for a Shinkansen to reduce the weight of a vehicle, bevel flaw detection is performed by scanning a probe in the hollow portion of the axle (Patent Document 1).

The ultrasonic flaw detection is performed by setting a flaw detection gate, that is, an echo recording gate, in a part of the range to be flawed by scanning the probe in the ultrasonic flaw detector conventionally used.
The rectangular frame shown in gray is the flaw detection gate, which is described as "gate (gray frame)" in the A scope display shown in FIG. 17 (A). The upper side of the rectangular flaw detection gate is the threshold value in the flaw detection gate. It also defines the range of the flaw detection gate in the range where the front and back sides of the rectangle detect flaws.
In FIG. 17A, the horizontal axis is the inner surface of the axle, that is, the inner peripheral surface of the axle as a 0 point (0 position), and shows the beam path from the 0 point. In FIG. 17A, the vertical axis indicates the echo height. As shown in FIG. 17A, the distance through which ultrasonic waves are transmitted from the probe (ultrasonic probe) is displayed in mm as the beam path. The inner peripheral surface of the axle is set to 0 mm, and the farther it goes to the right in FIG. 17 (A), the farther it is from the inner peripheral surface.

As shown in FIG. 17A, the A scope displays the ultrasonic waveform according to the beam path. In the A scope, all the waveforms of 360 degrees in the circumferential direction of the axle are synthesized, and the maximum value and the minimum value of the received echo are shown. In FIG. 17A, the wavy line shown in white shows the maximum value, and the wavy line shown in black shows the minimum value.
The operator who detects the flaw on the axle records the A-scope waveform and sets the position of the flaw detection gate used for determining the flaw while observing the recorded waveform. What is described as "scratch waveform" in FIG. 17A is a scratch echo waveform.

The operator wants to observe the flaw detection gate each time, that is, sets the flaw detection gate at an appropriate position where the occurrence of scratches is expected, and then performs flaw detection. The flaw detection gate once set has a fixed position within the flaw detection range. Met.
For this reason, it is necessary to set the flaw detection gate in advance at a position where the echo from the defect, that is, the scratch is not leaked, and it takes a lot of time and effort to properly set the flaw detection gate.
That is, using a test piece provided with an artificial defect (artificial scratch), that is, a model axle, the operator has to set a flaw detection gate for each axle actually detected by referring to the result of flaw detection performed on the model axle. I was taking it.

Further, when the flaw detection gate fixed as described above is used, when the shape echo (FIG. 17 (B)) enters the flaw detection gate, the shape echo caused by the step on the outer peripheral surface of the axle has the maximum peak in the flaw detection gate. Was erroneously determined to be a defect.
For example, in FIG. 17B, a scratch echo marked as "scratch" is inside the flaw detection gate marked as "judgment gate", but a shape echo is approaching the right side of the flaw detection gate, and scanning is performed from the flaw detection gate. If the scratch echo shifts and the shape echo enters the flaw detection gate, the above-mentioned erroneous determination occurs.

The axles of railway vehicles are journals that can be received by bearings, slinger that is an oil drain seat / rear lid seat, wheel seats that can be fitted with wheels, bearing seats, gear seats that can be fitted with gears, earth ring seats that are collector seats, and brakes. Each part in the axial direction, such as a brake disc seat into which a disc is fitted, has a different diameter and has the above-mentioned step.
The shape echo is an echo from such a step.

More specifically, as described above, the biggest reason for erroneously determining the shape echo as a scratch echo is that, in the conventional defect detection method, basically all the waveforms appearing in the flaw detection gate exceeding the threshold value of the flaw detection gate appear. This is because it was determined to be a wound echo.
In the conventional determination method, the location where a defect, which is a scratch, is generated is predicted in advance from experience by looking at the waveform, and the scratch detection gate is set based on the prediction.

However, even if the appropriate position range is determined and the flaw detection gate is set each time as described above, the presence or absence of defects is checked by scanning the probe. The echo peak captured in the flaw detection gate (the strongest point of the sound pressure level). The position) also shifts in the axial direction, and moves from the flaw detection gate in the order of the foot of the echo waveform, the peak, and the foot. When the peak passes from inside the flaw detection gate and only the base of the defect echo (scratch echo) remains, and the shape echo enters, the peak of the shape echo larger than the base of the echo is erroneously determined as the maximum echo of the defect.
Of course, the flaw detection gate also moves following the echo as the probe scans, but there is still a high possibility that an erroneous judgment will occur.

In addition to the above, the causes of erroneous judgment and judgment variation include fluctuations in the position of the entire waveform (changes in sound velocity due to changes in temperature) and individual differences in the operator who sets the flaw detection gate.
Further, one of the factors that the shape echo is mistaken as a defect echo is the contact state between the axle surface and the inner peripheral surface of the wheel in the fitting. Specifically, the defect echo that should have a flat head at the peak (top) due to the partial contact between the two generated at the fitting position of the wheel is abbreviated as “circumferential discontinuous echo” in FIG. 18 (A). A peak appears partially, exceeding the threshold value of the flaw detection gate, and the relevant portion is determined to be a defect echo. The "circumferential direction evaluation level" in FIG. 18A is viewed in relation to the angle formed by the upper side (threshold value) of the flaw detection gate with respect to the circumferential direction of the axle, that is, a specific line orthogonal to the center line of the axle. Regarding the erroneous determination shown in FIG. 18A, there was no appropriate workaround in the current determination method.

Further, when the axle is new, even if it is in contact with the wheel at each position of the fitting without a gap, a gap is formed in the fitting with the wheel over time, and it is erroneously determined as a defect echo. If there is a defect, the echo from the defect is detected at a strong sound pressure level.
Specifically, if the metal axle and the metal wheel are fitted tightly in the axial direction, the transmitted ultrasonic waves will escape from the axle to the wheel side and be attenuated, and no strong echo will be returned unless there is a defect.
However, strong reflection of the transmitted ultrasonic wave occurs between the axle (metal) and the air forming the gap, and it is erroneously determined that the echo is a defect echo. Due to the above gap, the sound pressure level of the shape echo was not stable, which was a cause of erroneous determination.

Japanese Patent No. 35558359

The above-mentioned problems related to the prior art are summarized as follows.
Since all echoes that entered the flaw detection gate and exceeded the threshold value were regarded as defect echoes, if the shape echo entered the flaw detection gate at a position where the occurrence of the defect was predicted, the shape echo was erroneously determined as a defect echo. Was there. For this reason, it is necessary to take the trouble of setting the gate strictly using the model axle every time the flaw is detected so that the shape echo does not intervene in the flaw detection gate when the same axle is detected. The detection level of the shape echo differs depending on the axle.
In particular, the gate setting is accompanied by variations among operators.

Further, as described above, even if the flaw detection gate can be set at a position where the occurrence of the scratch is accurately predicted, if the timing of tracking the flaw detection gate is deviated from the scanning of the probe, the shape echo intervenes and the above erroneous determination is made. Sometimes occurred. Further, the speed of sound transmitted in the axle changes due to the change in temperature depending on the season, which induces the above-mentioned erroneous judgment. Further, at the fitting position (seat) of the axle with the wheel, a partial gap is generated over time, and the echo between the air and the metal (the shape echo at the fitting position) induces an erroneous determination.
The present invention provides an ultrasonic flaw detection method for an axle and a system thereof that can avoid the above-mentioned erroneous determinations caused by shape echo in the flaw detection near the outer peripheral surface inside the axle, based on the knowledge of the inventor of the present invention. It is something to do.

In the present invention, when the material to be inspected is a solid axle, the outer peripheral surface of the axle is used, and when the material to be inspected is a mid-sized axle, the outer peripheral surface or the inner peripheral surface of the axle is used. The sound probe is scanned, ultrasonic waves are transmitted, echoes are received, and the vicinity of the outer peripheral surface of the axle inside the axle is detected at an oblique angle. The following configuration is provided for an ultrasonic flaw detection method for an axle that detects a sound pressure level as an echo height.
That is, in this ultrasonic flaw detection method for an axle, the echo height for determining a scratch echo indicating a scratch is set as the first level, and the echo height lower than the first level is set as the second level, and the flaw detection on the axle is performed. With respect to the waveform of the echo having a predetermined height or more detected from the entire range to be performed, when the thickness of the waveform at least at the first level is within the range of the thickness indicating the scratch echo, the thickness of the waveform at the second level. It is characterized in that it is determined whether or not the detected echo is a scratch echo indicating a scratch by examining.
Further, in the present invention, when the material to be inspected is a solid axle, the outer peripheral surface of the axle, and when the material to be inspected is a middle axle, the outer peripheral surface or the inner peripheral surface of the axle, the axial direction and the circumferential direction of the axle. The ultrasonic probe is scanned, ultrasonic waves are transmitted, echoes are received, and the vicinity of the outer peripheral surface of the axle inside the axle is detected at an oblique angle. The following configuration is provided for an ultrasonic flaw detection method for an axle that detects the sound pressure level of the axle as an echo height.
That is, in this ultrasonic flaw detection method for an axle, the echo height for determining a scratch echo indicating a scratch is set as the first level, and the echo height lower than the first level is set as the second level.
The angle with respect to a specific virtual line orthogonal to the center line of the axle is shown on the x-axis, and the echo height of the received echo is shown on the z-axis intersecting the x-axis. For the detected echo waveform of a predetermined height or higher, when the width in the x-axis direction at least at the first level is within the range of the width of the scratch echo indicating the scratch, the width in the x-axis direction at the second level is examined. It is characterized in that it is determined whether or not the detected echo is a scratch echo indicating a scratch.
Further, in the present invention, the echo detected when the width in the x-axis direction at the second level is within the range of the width of the shape echo is determined to be the shape echo of the axle, or the width in the x-axis direction at the second level is determined. The echo detected when the width is not within the width range of the shape echo is determined to be a scratch echo, and the shape echo is the circumferential direction of the axle because the diameter of the axle differs depending on the position in the axial direction of the axle. We were able to provide an ultrasonic flaw detection method for axles, which is characterized by echoes caused by the steps provided in.
Further, in the present invention, when the material to be inspected is a solid axle, on the outer peripheral surface of the axle, and when the material to be inspected is a middle axle, on the outer peripheral surface or the inner peripheral surface of the axle, the axial direction and the circumferential direction of the axle. The ultrasonic probe is scanned, ultrasonic waves are transmitted, echoes are received, and the vicinity of the outer peripheral surface of the axle inside the axle is detected at an oblique angle. The following configuration is provided for an ultrasonic flaw detection method for an axle that detects the sound pressure level of an echo as an echo height.
That is, in this method of ultrasonic flaw detection of an axle, the echo height for determining a scratch echo indicating a scratch is set as the first level, and the echo height lower than the first level is set as the second level, and the axial direction of the axle is set. The position of the echo is shown on the y-axis, and the echo height of the received echo is shown on the z-axis that intersects the y-axis. When the width in the y-axis direction at the first level is within the range of the width of the wound echo indicating a scratch, the detected echo by examining the width in the y-axis direction at the second level is a wound echo indicating a scratch. It is characterized in that it is determined whether or not it is.
Furthermore, in the present invention, the echo detected when the width in the y-axis direction at the second level is within the width range of the shape echo is determined to be the shape echo of the axle, or the y-axis direction at the second level. The echo detected when the width of the shape echo is not within the width range of the shape echo is determined to be a scratch echo. The shape echo is the circumference of the axle because the diameter of the axle differs depending on the position of the axle in the axial direction. We have been able to provide an ultrasonic flaw detection method for axles, which is characterized by echoes caused by steps provided in the direction.
Further, in the present invention, when the material to be inspected is a solid axle, the outer peripheral surface of the axle, and when the material to be inspected is a middle axle, the outer peripheral surface or the inner peripheral surface of the axle, the axial direction and the circumferential direction of the axle. The ultrasonic probe is scanned, ultrasonic waves are transmitted, echoes are received, and the vicinity of the outer peripheral surface of the axle inside the axle is detected at an oblique angle. We were able to provide an ultrasonic flaw detection method for axles that detects the sound pressure level of the axle as the echo height with the following configuration.
That is, in this method of ultrasonic flaw detection of an axle, the echo height for determining a scratch echo indicating a scratch is set as the first level, and the echo height lower than the first level is set as the second level, and the center line of the axle is set. The angle with respect to a specific virtual line orthogonal to the x-axis is shown on the x-axis, the axial position of the axle is shown on the y-axis that intersects the x-axis, and the echo height of the received echo intersects the x-axis and y-axis. In the three-dimensional coordinates shown on the z-axis, at least the area of the region projected onto the xy plane at the first level of the echo waveform of the predetermined height or more detected from the entire range where the axle is detected is scratched. When the area is within the range of the scratch echo shown, it is determined whether or not the detected echo is a scratch echo by examining the area of the region projected onto the xy plane at the second level. It is characterized by that.
Furthermore, in the present invention, the echo detected when the area of the region projected onto the xy plane at the second level is within the area of the shape echo is determined as the shape echo of the axle, or the second level. The echo detected when the area of the region projected onto the xy plane is not within the area of the shape echo is determined to be a scratch echo, and the shape echo is determined by the axial position of the axle. It has been possible to provide an ultrasonic flaw detection method for an axle, which is characterized by an echo caused by a step provided in the circumferential direction of the axle due to the difference in the diameter of the axle.
Further, in the present invention, when the material to be inspected is a solid axle, on the outer peripheral surface of the axle, and when the material to be inspected is a middle axle, on the outer peripheral surface or the inner peripheral surface of the axle, the axial direction and the circumferential direction of the axle. The ultrasonic probe is scanned, ultrasonic waves are transmitted, echoes are received, and the vicinity of the outer peripheral surface of the axle inside the axle is detected at an oblique angle. Provided is an ultrasonic flaw detection system for an axle that detects the sound pressure level of an echo as an echo height and adopts the following configuration.
That is, in this ultrasonic flaw detection system for the axle, the echo height for determining the scratch echo indicating the scratch is set as the first level, and the echo height lower than the first level is set as the second level, and the flaw detection of the axle is performed. For echo waveforms of a predetermined height or higher detected from the entire range to be performed, the thickness of the waveform at the second level is at least when the thickness of the waveform at the first level is within the range of the thickness indicating the scratch echo. By examining the above, it is determined whether or not the detected echo is a wound echo indicating a scratch, and the transmission probe and the reception probe, which are the ultrasonic probes, and the ultrasonic probe A scanning unit for scanning, a position detection unit, an A-scope data acquisition unit, a C-scope expansion unit, a threshold storage unit that holds a predetermined sound pressure level as a threshold, an echo extraction unit, an echo designation unit, and the like. Peak value acquisition unit, 1st level setting unit, 1st level regulation unit, 1st range regulation unit, 2nd level setting unit, 2nd level regulation unit, 2nd range regulation unit, 1st The A-scope data acquisition unit includes a determination unit and a second determination unit, and the A-scope data acquisition unit includes a beam path at the scanning position of the probe detected by the position detection unit, and an echo received by the reception probe for substantially the entire area of the beam path. The A-scope data of the waveform of the echo is sequentially acquired and accommodated from the echo height of the above, and the C-scope expansion unit is orthogonal to the center line of the axle with respect to the waveform data accommodated by the A-scope data acquisition unit. Data showing the angle with respect to a specific virtual line on the x-axis, the axial position of the axle on the y-axis intersecting the x-axis, and the echo height of the received echo color-coded according to the height. Accommodated as C-scope data, the echo extraction unit refers to the threshold value held by the threshold storage unit, extracts the waveforms of all echoes exceeding the threshold value from the data acquired by the C-scope expansion unit, and extracts the extraction. The identification data for identifying each of the generated waveforms is stored, and the echo designation unit automatically or accepts the operation of the operator, and the data of any waveform in the C scope expansion unit extracted by the echo extraction unit is stored. The peak value acquisition unit refers to the C scope expansion unit and acquires the echo height data of the peak of the waveform specified by the echo designation unit, and is the first level. The setting unit holds a calculation standard capable of calculating the echo height data to be the first level from the echo height data, and the first range defining unit is the first level. The upper limit value and the lower limit value of the data regarding the thickness of the waveform used for determining the scratch echo at the level are held, and the second level setting unit obtains the echo height data to be the second level from the echo height data. The second range ruler holds at least the lower limit of the data regarding the thickness of the waveform used for determining the shape echo at the second level, and holds the calculation standard that can be calculated. The defining unit acquires data on the echo height of the waveform specified by the echo designation unit with reference to the first level setting unit, and the first determination unit obtains data on the echo height as the first level. With respect to the waveform designated by the echo designation unit, the specified waveform at the first level calculated by referring to the first level regulation unit, the C scope expansion unit, and the first range regulation unit. It is automatically determined whether or not the data regarding the thickness is within the range specified by the first range specifying unit, and the second level specifying unit describes the waveform specified by the echo designation unit. The data of the echo height to be the second level is acquired by referring to the second level setting unit, and the second determination unit is at least within the range of the thickness to be the scratch echo in the first determination unit. With respect to the waveform determined to be present, the data regarding the thickness of the specified waveform at the second level calculated by referring to the second level defining unit, the C scope expanding unit, and the second range defining unit. However, it is automatically determined whether or not the data is within the range specified by the second range defining unit, and the data regarding the thickness of the waveform includes the width of the waveform in the x-axis direction and the said of the waveform. It has been possible to provide an ultrasonic flaw detection system for an axle characterized by having at least one of a width in the y-axis direction and an area of a region of a waveform projected onto the xy plane.
The above color coding includes not only coloring but also discrimination by luminance (brightness) and shading, and includes not only chromatic colors but also achromatic colors as long as they can be distinguished.
Further, in the present invention, the data regarding the thickness of the waveform includes the width of the waveform in the x-axis direction, the width of the waveform in the y-axis direction, the area of the region projected onto the xy plane of the waveform, and the above. At least one of the total widths of all waveforms appearing in the x-axis direction, and the total width of all waveforms appearing in the x-axis direction is a single waveform having continuous widths in the x-axis direction. It has been possible to provide an ultrasonic flaw detection system characterized by being the sum of all the widths of the waveforms appearing in the x-axis direction regardless of whether the waveforms are discontinuous or a plurality of waveforms.
Further, in the present invention, the transmitting probe is a transmitting / receiving probe that also serves as the receiving probe.
The transmission / reception probe is an array probe in which a plurality of oscillators are arranged in a ring shape along the circumferential direction of the axle, and the transmission / reception probe emits ultrasonic waves at an angle to the center line of the axle. It is a phased array that emits ultrasonic waves at once from a predetermined number of adjacent groups of oscillators among the oscillators arranged in such a manner and constituting the annular probe, and scanning of the transmission / reception probe with respect to the circumferential direction of the axle is performed. This is performed by sequentially shifting the group that emits ultrasonic waves at once in the circumferential direction of the annular array probe, and two transmission / reception probes are arranged before and after the axial direction, and the transmission / reception probe of the transmission / reception probe is arranged in the axial direction. One is directed to the opposite side of the other transmission / reception probe, and both transmission / reception probes are integrally scanned in the axial direction, and are scanned on the axle together with the circumferential scanning to the axle. On the other hand, it has been possible to provide an ultrasonic flaw detection system for an axle, which is characterized in that both transmission / reception probes transmit ultrasonic waves in a spiral shape.
Further, in the present invention, the first range setting unit and the second range setting unit are provided, and the model axle has the same dimensions, shape and material as the actual axle and is predetermined prior to flaw detection of the actual axle, that is, the actual axle. An artificial scratch of the shape and size of the above is provided at a predetermined position, and the flaw is detected.
The A-scope data acquisition unit sequentially acquires and stores the A-scope data of the model axle by flaw detection of the model axle, and the C-scope expansion unit is the model axle acquired by the A-scope data acquisition unit. With respect to the waveform data of the above, the C scope data is acquired and stored, and the echo extraction unit refers to the threshold held by the threshold storage unit and is included in the data of the model axle accommodated by the C scope expansion unit. The waveforms of all echoes exceeding the threshold are extracted from the data, and the identification data for identifying each of the extracted waveforms is stored. The first level setting unit is the waveform to be determined by the first determination unit. The ratio of the echo height as the first level to the echo height of the peak, the ratio of the echo height as the first level to the threshold in the threshold storage unit, and the first to the peak of the waveform to be determined by the first determination unit. Either the value of the difference in the sound pressure level of the level or the value of the difference in the sound pressure level of the first level with respect to the threshold value is held as the calculation reference, and the second level setting unit makes a second determination. The ratio of the echo height as the second level to the echo height of the peak of the waveform to be determined by the unit, the ratio of the echo height as the second level to the threshold of the threshold storage unit, the determination target of the second determination unit Either the value of the difference in the sound pressure level of the second level with respect to the peak of the waveform to be used or the value of the difference in the sound pressure level of the second level with respect to the threshold value is held as the calculation reference, and the operator said. By sequentially designating each of the waveform data in the C scope expansion section extracted by the echo extraction section of the model axle by operating the echo designation section, the peak value acquisition section determines the echo height of the specified waveform. The peak value of the data is acquired, and the first level defining unit refers to the first level setting unit and the peak value acquisition unit to obtain the echo height of the first level for the specified waveform of the model axle. The second level defining unit acquires the value, and the second level defining unit refers to the second level setting unit and the peak value acquisition unit to obtain the echo height of the second level for the specified waveform of the model axle. The first value is obtained by sequentially designating each of the waveform data in the C scope expansion section extracted by the echo extraction section of the model axle by operating the echo designation section of the operator. The range setting unit automatically sets the upper limit value and the lower limit value of the first range defining unit so that all the data related to the thickness of the specified waveform at the first level is within the range. The second range setting unit is to sequentially specify each of the waveform data in the C scope expansion unit extracted by the echo extraction unit of the model axle by operating the echo designation unit of the operator. Is to automatically set the lower limit value of the second range defining portion so that all the data related to the thickness of the specified waveform at the second level is within the range, and the above acquired on the model axle. Axle super-axle characterized in that the first determination unit and the second determination unit refer to each of the upper limit value and the lower limit value in the first range regulation unit and the second range regulation unit for the determination of the actual axle. We were able to provide an ultrasonic flaw detection system.

The present invention is different from the one in which an echo appearing at the flaw detection gate is immediately regarded as a scratch echo by using a conventional flaw detection gate, and the shape echo is erroneously determined as a scratch echo by utilizing the characteristics of the shape echo of the axle. Can be avoided.
In particular, in the present invention, a flaw detection gate is set in a local area where the axle is detected, and instead of observing only the local area, an echo having a height exceeding the threshold value in the axle detection range, that is, the entire beam path. This eliminates the need for the large amount of time and effort required in the past to set the flaw detection gate only at the position where the flaw echo is expected. In addition, there is no need to worry about variations depending on the operator who sets the flaw detection gate.
Specifically, the inventor of the present invention, after diligent research, shows that the shape echo has a thick base and a thin triangular shape on the apex side, as shown in the three-dimensional display of the C scope data shown in FIG. 18 (B). On the other hand, it was found that the wound echo appeared in a columnar shape and the thickness was almost the same between the base and the top side (in FIG. 18B, x is the x-axis, y is the y-axis, and z is the above. The z-axis is shown). Based on this finding, the inventor of the present invention has completed the present invention, eliminating the conventional flaw detection gate, or setting the conventional flaw detection gate to a part of the range in which the flaw detection gate is detected on the axle. What was set is to set the start point and end point of the flaw detection gate for the beam path to the entire beam path, that is, the entire range of the flaw detection of the axle, based on the peak of the waveform whose peak exceeds the threshold value. For echoes that are judged to correspond to the thickness of the scratch echo by determining at least two steps above and below the peak, with the upper row as the first level and the lower row as the second level, at least at the first level. It was decided whether it was a scratch echo or a shape echo from the thickness of the echo at the second level. In this determination, the present invention was able to suppress the erroneous determination of the shape echo as a scratch echo.
The thickness at each of the first and second levels may be at least one of the width in the x-axis direction, the width in the y-axis direction, or the area projected onto the xy plane.
Based on the above findings, in the present invention, for example, the range of the thickness specified in the second level is defined as the range corresponding to the shape echo, that is, the range corresponding to the thickness thicker than the range of the thickness specified in the first level. By determining an echo that falls within the range of the second level as a shape echo, it is possible to suppress an erroneous determination that the shape echo is a scratch echo. Of course, the range of the thickness specified in the second level is set to the range corresponding to the wound echo, that is, the range to be the same as the range specified in the first level, and the echo that falls within the range of the wound echo at the first level is the echo at the second level. It may be determined whether or not it is a wound echo directly by the determination.
In particular, in the present invention, the judgment conditions at the first level and the second level need to be set once in principle for the model axle that is the test piece in the flaw detection of one actual axle, and the same model axle is used for each flaw detection even for the same axle. There is no need to take the conventional trouble of setting the flaw detection gate in advance. Therefore, according to the present invention, the setting time by the model axle can be shortened.
Further, in the present invention, since it is not necessary to strictly set the flaw detection gate, it is difficult for the determination to vary due to individual differences of the operators.

The whole explanatory view of the system which concerns on one Embodiment of this invention. An explanatory diagram focusing on data processing of a system according to an embodiment of the present invention. (A) and (B) are explanatory views of the main parts of FIG. (A) is an explanatory diagram centering on the hardware according to the embodiment of the present invention, and (B) is an explanatory diagram showing a modified example of (A). FIG. 4 (A) and FIG. 4 (B) are enlarged vertical sectional views of a main part. FIG. 5 is a schematic vertical cross-sectional view along the axial direction of the axle showing scanning of the probe of FIG. FIG. 5 is a schematic cross-sectional view along the circumferential direction of the axle showing scanning of the probe of FIG. (A) is an explanatory diagram showing the concept of the C scope of the system according to the present invention, and (B) is an explanatory diagram showing the C scope display of the model axle based on the concept of (A) and the notched front view of the model axle. FIG. 8 is an explanatory diagram of xy (angle-axial direction) coordinates showing scratch echo and shape echo as C scope data of FIGS. 8A and 8B. FIG. 8 is an explanatory diagram of xy (angle-axial direction) coordinates showing shape echoes appearing discontinuously as C-scope data in FIGS. 8A and 8B. The explanatory view which shows the flow of the operation procedure in the actual flaw detection of the flaw detection system which concerns on one Embodiment of this invention. (A) and (B) are explanatory views of the main parts of FIG. The explanatory view which shows the flow in the setting before the actual flaw detection using the flaw detection system which concerns on one Embodiment of this invention. The explanatory view which shows the whole flow of the flaw detection system operation which concerns on one Embodiment of this invention. The explanatory view which shows the example of the operation screen of the flaw detection system which concerns on one Embodiment of this invention. (A) is a schematic vertical sectional view showing another embodiment of the flaw detection head of the present invention, and (B) is a schematic front view showing another embodiment of the scanning unit of the flaw detection device according to the present invention. (A) and (B) are explanatory views by A scope display of the problem by the conventional flaw detection method. (A) is an explanatory view showing a discontinuous echo in the circumferential direction of the axle, and (B) is a perspective view showing a three-dimensional stereoscopic display based on C-scope data of scratch echo and shape echo.

Hereinafter, preferred embodiments of the present invention will be illustrated with reference to the drawings.
(Outline of flaw detection method)
In the present invention, when the material to be inspected is a solid axle, the outer peripheral surface of the axle 100, and when the material to be inspected is the middle axle 100, the outer peripheral surface or the inner peripheral surface of the axle, the axial direction and the circumferential direction of the axle 100. The ultrasonic probe 1 is scanned, ultrasonic waves are transmitted, echoes are received, and the vicinity of the outer peripheral surface of the axle 100 inside the axle 100 is detected at an oblique angle. It relates to the ultrasonic flaw detection method of the axle 100 which detects the echo height with respect to.

In particular, the present invention sets a predetermined sound pressure level, that is, a predetermined echo height as the first level for an echo waveform of a predetermined height or higher detected from the entire range of flaw detection of the axle 100, and is higher than the first level. By examining the thickness of the waveform at the second level when the thickness of the waveform at the first level is within the range of the thickness indicating the scratch echo, with the echo height lower than the predetermined height as the second level. , It is determined whether or not the detected echo is a scratch echo indicating a scratch. In the above, the thickness of the waveform at the second level may be examined only when it is within the range of the thickness of the wound echo at the first level, or the thickness of the waveform at the second level may be examined together with the first level. ..

Then, when the thickness of the detected waveform at the second level is within the range of the width of the shape echo, the detected echo is determined to be the shape echo of the axle, or the thickness of the detected waveform at the second level is determined. The echo detected when it is not within the width range of the shape echo is determined to be a scratch echo.
The above-mentioned shape echo is an echo caused by a step provided in the circumferential direction of the axle because the diameter of the axle differs depending on the position of the wheel in the axial direction.

(Ultrasonic flaw detection system)
Taking the case of detecting a boring axle as an example, an example of the most suitable device for carrying out the present invention is shown.
As shown in FIGS. 1 to 4, the ultrasonic flaw detection system according to the present invention includes a mechanism unit 600 and a flaw detection unit 800.
As shown in FIGS. 1 and 2, the mechanism unit 600 includes an ultrasonic probe 1, a scanning unit 500, and a position detection unit 60. Further, the mechanism unit 600 includes an intermediary supply unit 510 and a sensor unit 520. The mediation supply unit 510 supplies a liquid such as water or oil that mediates ultrasonic waves between the ultrasonic probe 1 and the axle 100. In this example, the mediation of the ultrasonic wave is oil, and the mediation supply unit 510 is a refueling pump. In this example, the sensor unit 520 includes an origin sensor, an interlock sensor, and a temperature sensor.
The origin sensor is a sensor that detects a position that becomes an origin for specifying the position of the ultrasonic probe 1 (probe) with respect to the axle 100. The origin sensor is, in this example, a magnetic sensor that detects magnetism. However, if the position of the origin can be detected, even if an optical sensor that detects the position by receiving the projected light is used, a sensor that detects the position by transmitting and receiving radio waves is used. It may be a thing.
Further, regarding the interlock sensor, a physical sensor for detecting whether or not the axle mounting portion 7 described later is properly mounted on the axle 100, that is, whether or not the contact state of the axle mounting portion 7 with the axle 100 is appropriate. It is a contact type sensor that detects. In the interlock sensor, the scanning unit 500 does not drive the ultrasonic probe 1 (probe holding unit 2) before the ultrasonic probe 1 (probe holding unit 2) is mounted on the axle 100. It is provided as follows. However, the above interlock sensor may be omitted.
The temperature sensor is a sensor that detects the temperature of the ultrasonic mediator. In this example, the temperature sensor is an oil temperature sensor provided in an oil supply tank used as a medium for ultrasonic waves.
In the conventional flaw detection method in which the flaw detection gate is strictly set, it is necessary to adjust the flaw detection gate even when the oil temperature changes. Therefore, the above temperature sensor is an indispensable configuration, but in the flaw detection method of the present invention, the flaw detection gate is set. Since it is unnecessary, it can be carried out without providing the above temperature sensor.

The flaw detector 800 is an electrical component, and the flaw detector 8 connected to the ultrasonic probe 1, the computer 9, and the position detection unit 60, the scanning unit 500, the mediation supply unit 510, and the sensor unit 520 are connected to the computer 9. It is provided with an interface 82 to be connected.
The above-mentioned flaw detector 8, the computer 9, and the software of the introduced flaw detection system mainly construct the flaw detection unit 800 of the flaw detection system according to the present invention (FIG. 2). The details of the flaw detection unit 800 will be described later.

(Device configuration)
First, the device configuration, that is, the hardware of the flaw detection system according to the present invention will be described with reference to FIGS. 4 to 7.
As shown in FIG. 5, the ultrasonic probe 1 is a transmitting probe that transmits ultrasonic waves and a receiving probe that receives ultrasonic waves. In this example, the transmission probe is a transmission / reception probe that also serves as a reception probe, and performs both transmission and reception of ultrasonic waves. However, the transmitting probe and the receiving probe may be formed separately.
In this flaw detection method, ultrasonic waves are transmitted obliquely from the ultrasonic probe 1 to the center line of the axle 100 (virtual line / not shown in FIGS. 4 to 6), that is, the center beam is transmitted as described above. A bevel flaw detection is performed at an angle to the center line.

The ultrasonic probe 1 may be a single probe in which scanning in the circumferential direction of the axle 100 as well as the axial direction of the axle 100 is physically performed, but here, a plurality of oscillators 1p are performed in the circumferential direction of the axle 100. Are arranged in an annular shape, and scanning in the circumferential direction is performed by electronic scanning, that is, an array probe that sequentially vibrates the oscillator 1p arranged in the annular shape is used.

Specifically, as shown in FIG. 5, the ultrasonic probe 1 has the transmission / reception probe as the array probe, and is individually arranged in the hollow portion 103 of the axle in an annular shape along the inner peripheral surface of the axle 100. The vibrator 1p emits ultrasonic waves diagonally with respect to the center line. Μ in FIG. 5 shows the beam path.
In this example, the vibrator 1p constituting the ultrasonic probe 1 shown in FIG. 5 is formed in a conical shape, and each vibrator 1p constituting the array probe is a plane orthogonal to the center line (FIG. 5). The ultrasonic wave is transmitted at an angle θ of 50 degrees with respect to (the broken line), that is, an angle of 130 degrees with respect to the center line.
However, the angle θ can be changed in the range of 35 degrees to 75 degrees. If the angle θ is less than 35 degrees, the transmitted ultrasonic waves (transverse waves) cannot be properly reflected.

In the above electronic scanning, one of the annularly arranged oscillators 1p is one that vibrates at one time, and the one oscillator 1p that vibrates sequentially along the circumferential direction is shifted. However, here, a phased array method is adopted in which ultrasonic waves are emitted from a group of a predetermined number of adjacent oscillators 1p among the oscillators 1p constituting the array probe at one time. In the phased array method, scanning of the transmission / reception probe is performed by sequentially shifting the group that emits ultrasonic waves at one time of the annular array probe in the circumferential direction of the axle 100, that is, in the circumferential direction of the annular shape.

In the present invention, as shown in FIG. 7, as the ultrasonic probe 1, an array probe in which the oscillator 1p is arranged in an annular shape so as to be concentric with the axle 100 is used, but a phased array probe is linearly used. It is the same as the linear array that arranges and shifts the oscillating group as described above. More specifically, after the oscillation of the preceding group g1 oscillator 1p, the next group g2 oscillator 1p displaced by one oscillator 1p oscillates. By sequentially shifting and vibrating the group of oscillators 1p along the circumferential direction of the axle 100 in this way, ultrasonic waves are oscillated from the hollow portion 103 toward the vicinity of the inner surface of the axle 100, and the axle by the phased array is used. Perform 100 flaw detection. In the example shown in FIG. 7, the number of oscillators 1p arranged in an annular shape in one array probe is 64, and a total of 128 oscillators 1p are arranged in two annular array probes in the front and rear as described later.
Further, six oscillators 1p are simultaneously oscillated as the above one group. However, the number of oscillators 1p of each array probe is not limited to 64, and the number of oscillators 1p can be changed according to the processing capacity of the computer 9. Further, the oscillators 1p that simultaneously transmit the vibrations are not limited to the above six, and can be changed depending on the axle to be detected. For example, seven or more oscillators 1p may be oscillated at the same time. A maximum of 32 waveform information in the 64 oscillators 1p can be synthesized by the phased array.

The advantage of adopting the phased array method is that ultrasonic waves synthesized with desired waveforms can be transmitted into the axle 100. For example, in the phased array method, a flat plane wave or a curved surface wave can be easily synthesized, and a wide range in the axle of the material to be inspected can be detected at once.
In the above, it is assumed that the oscillators 1p that vibrate at one time are vibrated at the same time in one group, but the timing of vibration is shifted by applying a delay between the oscillators 1p in one group. It may be a sector scan that changes the frequency.

In both cases where the single probe is adopted for the ultrasonic probe 1 and the array probe including the phased array is adopted, scanning in the axial direction of the axle 100 is performed by the ultrasonic probe 1 and the axle 100. This is done by physically moving one or both of the above. Specific examples of the scanning will be described in the description of the scanning unit 500.

102 in FIG. 6 shows the press-fitted portion of the wheels 101 and 101.
In this example, as shown in FIG. 5, the ultrasonic probe 1, that is, the transmission / reception probe is arranged two in the front-rear direction in the axial direction. That is, it is composed of a probe 1a facing diagonally forward and a probe 1b facing diagonally backward, with the traveling direction of scanning in the axial direction as the front (left side of FIGS. 5 and 6). This point will be described in detail with reference to FIG. When the ultrasonic probe 1 is scanned from the right side to the left side of FIG. 6 by the hollow portion 103 of the axle 100, for example, in the probe 1a in which the flaw detection range is the diagonal line V facing forward, the fit with the wheel 101 on the axle. The flaw can be detected in the left corner B of the portion, that is, the press-fitting portion 102, but the ultrasonic wave cannot be properly directed to the right corner A and the flaw cannot be detected. Therefore, another probe 1b in the opposite direction to the probe 1a is arranged back to back. Since the probe 1b sets the flaw detection range to the region of the diagonal line W facing backward, the probe 1b can detect the corner A.
That is, the ultrasonic wave (diagonal line V) emitted by the probe 1a of the ultrasonic probe 1 has a direction in which the vicinity of the left corner B of the axle 100 can be detected, and conversely, the ultrasonic wave emitted by the probe 1b (diagonal line V). The diagonal line W) has a direction in which the vicinity of the right corner A of the axle 100 can be detected. The same functions and configurations are used for both probes 1a and 1b.
The front and rear probes 1a and 1b may be used properly depending on the left and right corners A and B of the axle 100.

Both the front and rear probes 1a and 1b are integrally scanned in the axial direction of the axle 100. Therefore, the ultrasonic probe 1 emits ultrasonic waves in a spiral shape, that is, in a spiral shape with respect to the axle 100 by physical scanning in the axial direction together with the electronic scanning in the circumferential direction, and emits an echo. I will receive it.
The oscillators 1p of the front and rear probes 1a and 1b having a conical shape are fixed integrally and have a shape like an abacus ball. The outside (the white part in contact with the gray oscillator 1p in FIG. 5) is a resin forming a part of the probes 1a and 1b. That is, the resin covers the outside of the vibrator 1p and forms a part of the cylinder presented by the probe holding portion 2. Oil that mediates ultrasonic waves is supplied between the resin surface and the inner peripheral surface of the axle 100.

Although not shown, the probe holding portion 2 is provided with the mediation supply port before and after the probes 1a and 1b, and the mediation recovery port is provided in the vicinity of the probes 1a and 1b. Packing is fitted in front of and behind the supply port and the recovery port in the axial direction of the probe holding portion 2, and the packing holds the ultrasonic wave medium in the vicinity of the probes 1a and 1b.
4 (A) and 2 of FIG. 5-2 show a cylindrical probe holding portion 2 that holds the probes 1a and 1b. One end (tip) of the flexible tube 4 is attached to the probe holding portion 2 and is electrically connected to the probes 1a and 1b. Further, in the flexible tube 4, a tube for guiding the transmission of the ultrasonic wave from the outside of the probe holding portion 2 to the probe holding portion 2 is provided.

Here, the scanning unit 500 will be described with reference to FIG. 4A.
As shown in FIG. 4A, the probe holding portion 2 is arranged inside the axle 100 from one end of the axle 100 in the state where the wheel 101 is press-fitted. As described above, the probe holding portion 2 is provided at one end (tip) of the flexible tube 4. The flexible tube 4 has a length that allows the probe holding portion 2 to sufficiently follow any position in the axle 100.
First, the axle mounting portion 7 is fitted to the end portion of the axle 100 (the left end in FIG. 4A). The axle mounting portion 7 is a guide member through which the flexible tube 4 is passed.
The other end (base end) side of the flexible tube 4 is connected to a winding means 5 such as a pulley provided outside the axle 100. That is, the base end side of the flexible tube 4 is physically connected to the winding means 5.

40 in FIG. 4A shows an auxiliary pulley for changing the sliding direction of the flexible tube 4. By winding the flexible tube 4 by the winding means 5, the probe holding portion 2 moves to the left side of FIG. 4A, and by unwinding the flexible tube 4, the probe holding portion 2 Moves to the right side of FIG. 4 (A). The flexible tube 4 is arranged so as to be inserted into the axle mounting portion 7. The axle mounting portion 7 is fixed to the end of the axle 100, and includes a position sensor 70 as a position detecting portion 60 that detects the sliding position of the flexible tube 4 and obtains information on the flaw detection position. An encoder can be adopted for the position sensor 70. For example, as the encoder, a rotary encoder that controls the rotary motor of the winding means can be used. The position sensor 70 can measure the rotation of the rotary motor or the pull-out length of the flexible tube 4, and can detect the position of the ultrasonic probe 1 in the axial direction.
If the winding means 5 is provided with a moving means such as casters 50 so that it can be easily moved, the installation position can be quickly moved according to the work environment regardless of the weight of the device, which is convenient. Is.

The flaw detector 8 is physically and electrically connected to the ultrasonic probe 1 via a wiring (not shown) passing through the flexible tube 4. The wiring transmits a signal to apply a spike voltage that vibrates the vibrator 1p of the ultrasonic probe 1 from the flaw detector 8 side toward the ultrasonic probe 1 when transmitting ultrasonic waves toward the axle 100. When the echo reflected from the axle 100 is received, the electrically converted signal of the echo received by the vibration of the vibrator 1p is received.
As described above, the flexible tube 4 itself is internally wired for transmitting information sent from the probes 1a and 1b and the probe holding portion 2, and the winding means 5 provides such information. It is transferred to the flaw detector 8 and the computer 9. Depending on the processing of the data, it is also possible to connect and carry out the printer 90 for outputting the accumulated data or the data obtained in real time.

Further, instead of the flexible tube 4 shown in FIG. 4A, the probe holding portion 2 may be scanned by using the cable bear (registered trademark) 46 shown in FIG. 4B and the scanning chain 47. good.
In the example shown in FIG. 4B, the scanning unit 500 and the flaw detection unit 800 are provided on the electric self-propelled vehicle 49 that can be driven by the operation of the operator.
The cable bear (registered trademark) 46 is a member (cable protection member) for protecting and guiding a movable part, that is, a cable connected to the probe holding part 2, and is generally an industrial machine or a machine tool. It is a well-known one used for such purposes. The cable bear (registered trademark) 46 is made by connecting small parts such as a chain, and can realize a free length setting and flexible movement. As shown in FIG. 4B, the tip of the cable bear (registered trademark) 46 is connected to the probe holding portion 2. A signal line connecting the ultrasonic probe 1, the flaw detector 8, and the computer 9 is passed through the hollow cable bear (registered trademark) 46.
As shown in FIG. 4B, the cable bear (registered trademark) 46 can be inserted into the inside of the axle 100 (hollow portion 103) by extending from the zigzag folded state.
The scanning chain 47 is a chain on which the cable bear (registered trademark) 46 is placed and the cable bear (registered trademark) 46 is moved with respect to the axle 100. 48 in FIG. 4B is a winding device for the scanning chain 47. By rotationally driving the take-up device 48 with a motor, the scanning chain 47 can be pulled out or pulled back. By pulling out the scanning chain 47 from the winding device 48, the cable bear (registered trademark) 46 mounted on the scanning chain 47 can be inserted into the hollow portion 103 of the axle 100 as described above. A well-known plug lock chain (trade name) can be adopted as the scanning chain 47.

The self-propelled vehicle 49 is provided with a lifter (not shown). In the lifter, when the scanning chain 47 is pulled out from the take-up device 48 and the ultrasonic probe 1 (searcher holding portion 2) is attached to the axle 100, the probe holding portion 2 reaches a certain position of the axle 100. It is a well-known lifting device that raises the tip portion of the cable bear (registered trademark) 46 and the scanning chain 47 to which the cable bear (registered trademark) 46 is connected. The lifter lowers the cable bear (registered trademark) 46 and the scanning chain 47 pulled out from the axle 100 after the flaw detection, and accommodates the cable bear (registered trademark) 46 and the scanning chain 47 in the self-propelled vehicle 49.
In the example shown in FIG. 4B, the sensor unit 520 (FIG. 1) includes a lifter upper / lower limit sensor together with the other sensors described above.
The lifter upper / lower limit sensor is a sensor that detects the operating range of the lifter, that is, the upper limit and the lower limit of raising / lowering. As the lifter upper / lower limit sensor, a position sensor similar to the origin sensor may be used, or a physical sensor similar to the interlock sensor may be used.
The matters not particularly mentioned in the example shown in FIG. 4B are the same as those in the embodiments shown in FIGS. 1 to 3, 4 (A), and 5 to 15.

(Data processing configuration)
Next, the configuration of data processing of the flaw detection system according to the present invention will be mainly described with reference to FIGS. 2 and 3. That is, here, the above-mentioned flaw detection unit 800 will be described in detail. The flaw detection unit 800 is constructed by the above-mentioned flaw detector 8, the computer 9 connected to the flaw detector 8, and the software of the flaw detection system installed in the computer 9, and is a configuration of the flaw detector 800. , The control unit 860 shown in FIG. 2, the data processing unit 81, the display unit 840, and the operation unit 850 are constructed.
The control unit 860 of the flaw detection unit 800 controls the transmission of ultrasonic waves and the reception of echoes to the ultrasonic probe 1. Further, the control unit 860 performs scanning control on the scanning unit 500 that scans the ultrasonic probe 1. Further, the control unit 860 controls the data processing in the data processing unit 81. The control unit 860 is mainly composed of the central information processing device of the computer 9 and the above-mentioned software installed in the computer 9 (hereinafter, referred to as flaw detection software if necessary).

As shown in FIG. 2, the data processing unit 81 includes an A-scope data acquisition unit 801, a C-scope expansion unit 802, a threshold value storage unit 803, an echo extraction unit 804, an echo designation unit 851, and a peak value acquisition unit. 805, 1st determination unit 810, 1st level regulation unit 811, 1st level setting unit 812, 1st range regulation unit 813, 1st range setting unit 814, 1st determination result accommodating unit 819. , 2nd determination unit 820, 2nd level regulation unit 821, 2nd level setting unit 822, 2nd range regulation unit 823, 2nd range setting unit 824, 2nd determination result accommodating unit 829, pass / fail. A determination unit 833, a pass / fail determination result accommodating unit 834, and a counting unit 835 are provided.

As shown in FIG. 2, the display unit 840 includes an A scope display unit 841, a C scope display unit 842, a first determination result display unit 843, a second determination result display unit 844, and a pass / fail determination result display unit. It is equipped with 845.
The display unit 842 adopting the above configuration displays the A scope, the C scope, the first determination result, the second determination result, and the axle pass / fail determination result, that is, the flaw detection result, which will be described later, on the monitor of the computer 9. ..
The operation unit 850 can specify an arbitrary echo of the C scope expansion unit 802 through the echo designation unit 851 by accepting the operation of the operator.
The operator uses an input device such as a mouse or keyboard of the computer 9 constituting the operation unit 850 to specify an arbitrary echo from the C scope display displayed on the monitor of the computer 9 by the C scope display unit 842 of the display unit 840. be able to.

FIG. 15 shows the operation screen of the display unit 840, that is, the operation unit 850 displayed on the monitor of the computer 9. To specifically explain the designation of the echo, the cursor pp is moved by the operation of the input device on the C scope display shown in the dark background portion in the lower part of FIG. 15, and the cursor pp is aligned with the desired echo. You can specify an echo. The dark part of the background above FIG. 15 shows the A-scope display. Further, although not displayed on the screen of FIG. 15, the B scope showing the cross section in the direction intersecting the axial direction of the axle 100, that is, the axle 100 being cut into round slices and showing the echo is displayed together with the C scope. May be good.
FIG. 15 will be described in detail together with the later flow of operation of the system according to the present invention.

Each configuration of the data processing unit 81 will be described.
The A-scope data acquisition unit 801 is composed of the above-mentioned flaw detection software and a storage area (internal or external storage device) such as a memory or a hard disk of a computer 9 (A-scope memory). The flaw detector 8 sequentially acquires A-scope data of the waveform of the echo from the beam path at the scanning position of the probe detected by the position detection unit 60 and the echo height of the echo received by the probe for substantially the entire range of the beam path. The control unit 860 accommodates the A-scope data in the A-scope data acquisition unit 801.
The flaw detector 8 includes an A / D converter (analog-decital converter) that converts the waveform received by the ultrasonic probe 1 as an analog signal into digital data. The A-scope data acquisition unit 801 acquires and stores the digital data.

The C-scope expansion unit 802 is composed of the above-mentioned flaw detection software and a storage area such as a memory of a computer 9. The C-scope expansion unit 802 refers to the A-scope data acquisition unit 801 and refers to the waveform data (A-scope data indicated by the orthogonal coordinates of the x-axis and the y-axis) accommodated by the A-scope data acquisition unit 801 of the axle 100. The angle with respect to the specific virtual line xa orthogonal to the center line ya is shown on the x-axis (Fig. 8 (A) left figure), the axial position of the axle 100 is shown on the y-axis intersecting the x-axis, and the received echo. The data showing the echo height of the above in different colors according to the height is accommodated as C scope data (FIGS. 8 (B) and 9). Specifically, in the C scope display for displaying the C scope data, the gradation, that is, the color tone is changed according to the echo height level of the A scope.
However, the color coding is such that the operator who sees the C scope display of the C scope display unit 842 can intuitively grasp the echo height, and the flaw detection system of the present invention has the echo height in the C scope data. It is not necessary to add the color-coded information (data related to the color-coded) in the display to the echo height data as long as the data can be identified. In the system according to the present invention, the information on the eco-height of the C scope is acquired as the value of the z-axis (value of the sound pressure level) that intersects the x-axis and the y-axis, and x on the display unit 840. Only the position on the −y coordinate may be displayed.

Further, the color coding is information that can be visually identified not only by coloring but also by luminance (brightness) and shading. In this example, although it is difficult to distinguish from FIG. 8B, coloring on the monitor of the display unit 840 is performed. And the echo height can be visually recognized by the change in brightness.
The C scope display shown in the upper part of FIG. 8 (B) can be considered to be the development of the left figure of FIG. 8 (A) in the state shown in the right figure of FIG. 8 (A).
In the C scope display, as shown in FIGS. 8A and 8B, the axle is virtually displayed in a developed plane state, and the shape of the NG (defect) determination portion can be visually confirmed as a plane image.
The lower part of FIG. 8B shows the appearance of the axle corresponding to the C scope.
The graph of the C scope in the upper part of FIG. 8B is obtained from the model axle m for adjusting the flaw detection system before the actual flaw detection.
As shown in white on the C scope in the upper part of FIG. 8B, the model axle m is provided with a predetermined artificial defect, that is, an artificial scratch (artificial scratch) at a predetermined position. In FIG. 8B, the positional relationship between the upper C-scope display and each axial part of the lower model axle m (that is, journal, slinger, foil seat, bearing seat, gear seat, earth ring seat, brake disc seat, etc.) is It corresponds.
Note that FIGS. 9 and 10 show C-scope data, and one cell can be considered as one unit (1 bit) of the memory. However, in order to avoid complication of drawing, FIGS. 9 and 10 are drawn with a thinning (low resolution) from the actual memory.

The threshold value storage unit 803 is composed of the flaw detection software and a storage area such as a memory or a hard disk of the computer 9. The threshold storage unit 803 holds a predetermined sound pressure level as a threshold.
The above threshold value is a value set from the beginning and is a constant value that does not change during flaw detection. However, the operator can adjust the threshold value before flaw detection through the operation unit 850 as necessary.
The echo extraction unit 804 is composed of the flaw detection software and a storage area such as a memory or a hard disk of the computer 9.
The echo extraction unit 804 refers to the threshold value held by the threshold value storage unit 803 and extracts data of all echo waveforms exceeding the threshold value from the data acquired by the C scope expansion unit 802. As the data of the waveform, the echo extraction unit 804 contains the identification data for identifying the waveform. For example, the echo extraction unit 804 can accommodate data (x, y) indicating the position of the C scope on the xy coordinates of the waveform to be extracted as the identification data. Further, the identification data may include the echo height (sound pressure level) together with the position data. However, if the echo exceeding the above threshold value can be identified, the echo extraction unit 804 may adopt other means.
The echo designation unit 851 automatically designates (automatically designates) the echo extracted by the echo extraction unit 804 in the actual flaw detection, that is, the actual flaw detection with respect to the actual axle.
The echo designation unit 851 designates an echo that has been designated by the operation of the operator's operation unit 850 in the flaw detection of the model axle m for setting the range regarding the thickness of the waveform before the actual axle flaw detection. However, even in the actual flaw detection, the echo designation of the echo designation unit 851 may be performed by accepting the designated operation from the operation unit 850 of the operator together with the automatic designation or instead of the automatic designation.

The first level setting unit 812 holds a calculation standard (first calculation standard) necessary for calculating the first level.
The first calculation standard is data used for calculating the first level, which differs for each waveform. The first calculation standard is data with a certain value. The first calculation standard is the ratio of the echo height as the first level to the echo height of the peak of the waveform, or the difference in the sound pressure level of the first level with respect to the peak of the waveform to be determined by the first determination unit. Can be.
Specifically, when the echo height of 80% of the echo height of the peak of the waveform is set as the first level, the ratio of 80% is the first calculation standard. Further, when the echo height 3 dB (decibel) lower than the peak of the waveform is set as the first level, the value of -3 dB is the first calculation standard. Therefore, the first calculation standard itself held by the first level setting unit 812 does not change for any of the designated waveforms unless the operator adjusts.
However, instead of the above, the first level setting unit 812 is the value of the ratio of the echo height as the first level to the threshold value of the threshold value storage unit 803, or the value of the difference in the sound pressure level of the first level to the threshold value. Any of them may be used as the first calculation standard.

The second level setting unit 822 holds a calculation standard (second calculation standard) necessary for calculating the second level.
The second calculation standard is the data used to calculate the second level, which differs for each waveform. The second calculation standard is data with a certain value. The second calculation standard is the ratio of the echo height as the second level to the echo height of the peak of the waveform, or the difference in the sound pressure level of the second level with respect to the peak of the waveform to be determined by the second determination unit. Can be.
As the second calculation standard held by the second level ruler 821, the same type of standard as the first calculation standard may be adopted as described above, but the echo height of the second level of the waveform is the echo height of the first level. It is necessary to set the second calculation standard so that it will be lower than that.
Specifically, when the echo height of 60% of the echo height of the peak of the waveform is set as the second level, the ratio of 60% is the second calculation standard. Further, when the echo height 6 dB (decibel) lower than the peak of the waveform is set as the second level, the value of -6 dB is the first calculation standard. Therefore, the second calculation standard itself held by the second level setting unit 822 does not change for any of the designated waveforms unless the operator adjusts.
However, also in the second level setting unit 822, instead of the above, the ratio of the echo height as the second level to the threshold value of the threshold value storage unit 803, or the value of the difference in the sound pressure level of the second level to the threshold value. Any of the above may be used as the second calculation standard.
The calculation standard held by the first and second level setting units 812 and 822 may be any one of the above-mentioned peak of the waveform or the ratio to the threshold value or the difference in the sound pressure level, or the peak of the above-mentioned waveform. Alternatively, each level setting unit 812, 822 may be provided with all the differences in the ratio to the threshold value and the sound pressure level as the calculation standard, and the operator may select from them to set the calculation standard.
The first calculation standard held by the first level setting unit 812 may be the threshold value of the threshold storage unit 803, and in that case, a value 100% or −0 dB lower than the threshold value is used as the first calculation standard. Further, when the threshold value for the first calculation is set as the threshold value of the threshold value storage unit 803, in the flaw detection system according to the present invention, one of the first level setting unit 812 and the threshold value storage unit 803 can be used in combination with the other.

The first level defining unit 811 acquires the echo height data to be the first level by referring to the peak value acquisition unit 805 and the first level setting unit 812 for the waveform designated by the echo designation unit 804.
For example, when the first level setting unit 812 uses 80% of the ratio of the echo height as the first level to the echo height of the peak of the waveform as the first calculation standard, the first level defining unit 811 specifies the echo. With respect to the peak of the waveform specified by the unit 804, the echo height of 80% is acquired as the first level of the waveform. When the peak of the waveform specified above is 10 dB, the first level of the waveform is 10 dB × 80% = 8 dB, and when the peak of the specified waveform is 5 dB, the first level of the waveform is 5 dB × 80% =. It becomes 4 dB.
Further, when the first level setting unit 812 uses the echo height of -3 dB from the echo height of the peak of the waveform as the first calculation reference, if the peak of the waveform specified above is 10 dB, the peak of the waveform is the same. The first level is 10 dB-3 dB = 7 dB, and if the peak of the specified waveform is 8 dB, the first level of the waveform is 8 dB-3 dB = 5 dB. The display on the display unit 840 may be either a ratio (%) or a decibel (dB) of the sound pressure level.

The second level defining unit 821 acquires the echo height data to be the second level by referring to the peak value acquisition unit 805 and the second level setting unit 822 for the waveform designated by the echo designation unit 851. The calculation based on the second calculation standard is the same as that according to the first calculation standard. However, as described above, the second calculation standard is set so that the second level has an echo height lower than that of the first level.
In this example, -3 dB is set as the first level from the peak of the waveform designated by the echo designation unit 851, and -6 dB is set as the second level from the peak. As described above, the difference between the first level and the second level is not limited to 3 dB, but in order to reliably utilize the characteristics of the shape echo shown in FIG. 18 (B) on the axle, the first level is used. It is preferable that the difference between the second levels is 2 dB or more, particularly 3 dB or more. However, if the second level is set too low, the base noise will be picked up. Therefore, it is preferable that the difference between the first level and the second level is 2 dB or more, particularly 3 dB or more, and the base noise is not picked up.
The operator can change the ratio or value of the first calculation standard and the second calculation standard of the first and second level setting units 812 and 822 by operating the operation unit 850.

The first range defining unit 813 holds the upper limit value and the lower limit value of the data regarding the thickness of the waveform to be a scratch echo at the first level. Specifically, as shown in FIG. 3A, the x-axis direction lower limit value accommodating portion 815a accommodating the lower limit value of the width in the x-axis direction as the scratch echo at the first level, and the scratch echo at the first level. The x-axis width upper limit value accommodating portion 815b that accommodates the upper limit value of the axial width, and the y-axis direction lower limit value accommodating portion that accommodates the y-axis direction lower limit value that is the scratch echo at the first level. The 816a, the y-axis width upper limit accommodating portion 816b that accommodates the upper limit of the width in the y-axis direction that is the scratch echo at the first level, and the region projected onto the xy plane that is the scratch echo at the first level. The xy area lower limit accommodating portion 817a accommodating the lower limit value of the area and the xy area upper limit accommodating portion accommodating the upper limit value of the area projected onto the xy plane to be the scratch echo at the first level. 817b, the circumferential total width lower limit value accommodating portion 818a accommodating the lower limit value of the total width of all waveforms appearing in the x-axis direction, and the circumferential direction accommodating the upper limit value of the sum total width of all waveforms appearing in the x-axis direction. A total width lower limit value accommodating portion 818b is provided.
The sum of the widths of all the waveforms appearing in the x-axis direction is the sum of the waveforms appearing in the x-axis direction regardless of whether the widths in the x-axis direction are a continuous single waveform or a plurality of discontinuous waveforms. It is the sum of all widths.
The first range setting unit 814 of FIG. 2 automatically changes the upper limit value and the lower limit value of each accommodating unit of the first range defining unit 813. The details of the first range setting unit 814 will be described later.

The second range defining unit 823 holds the upper limit value and the lower limit value of the data regarding the thickness of the waveform to be a scratch echo in the second level. Specifically, as shown in FIG. 3B, the x-axis direction lower limit value accommodating portion 825a accommodating the lower limit value of the width in the x-axis direction as a scratch echo at the second level, and the scratch at the second level. The x-axis width upper limit accommodating portion 825b accommodating the upper limit of the axial width as an echo, and the y-axis width lower limit accommodating the y-axis width lower limit accommodating the scratch echo in the twelfth bell. A region 826a, a y-axis width upper limit accommodating portion 826b that accommodates the upper limit of the width in the y-axis direction as a scratch echo at the second level, and a region projected onto the xy plane to be a scratch echo at the second level. XY area lower limit value accommodating portion 827a accommodating the lower limit value of the area of A circumference that accommodates the lower limit of the total width of all waveforms appearing in the x-axis direction and the upper limit of the total width of all waveforms appearing in the x-axis direction. It is provided with a directional total width lower limit value accommodating portion 828b.
Also in the second range defining unit 823, the sum of the widths of all the waveforms appearing in the x-axis direction may be a single waveform having a continuous width in the x-axis direction or a plurality of discontinuous waveforms. It is the sum of all the widths of the waveforms appearing in the x-axis direction.
The second range setting unit 824 of FIG. 2 automatically changes the upper limit value and the lower limit value of each accommodating unit of the second range defining unit 823. The details of the second range setting unit 824 will be described later.

The count unit 835 detects the thickness values at the first level and the second level for the waveform designated by the echo designation unit 851.
For example, for the echo k1 shown on the xy plane of FIG. 9, the counting unit 835 counts that the width (maximum width) x1 in the x-axis direction at the first level L1 is 6 (6 squares in the graph). The value counted as one of the values indicating the thickness in the first level L1 is acquired. Further, for the echo k1, the counting unit 835 counts that the width (maximum width) x2 in the x-axis direction at the second level L2 is 8 (eight squares in the graph), and indicates the thickness at the second level L2. Get the value counted as one of the values.
Further, in this example, for the echo k1, the counting unit 835 counts that the width (maximum width) y1 in the y-axis direction at the first level L1 is 4 (for four squares in the graph), and the above thickness at the first level L1. It is acquired as one of the values indicating the value. Further, for the echo k1, the counting unit 835 counts that the width (maximum width) y1 in the y-axis direction at the second level L2 is 6 (eight squares in the graph), and indicates the thickness at the second level L2. Get as one of the values.
Further, in this example, the echo k1 is counted by the counting unit 835 as having an area on the xy plane at the first level L1 of 20 (20 squares in the graph), and the thickness at the first level L1 is shown. Get as one of the values.
Further, for the echo k1, the count unit 835 counts that the area on the xy plane at the second level L2 is 43 (43 squares in the graph), and is 1 of the value indicating the thickness at the second level L2. Get as one.
Further, with respect to the echo k2 shown on the xy plane of FIG. 9, the counting unit 835 counts that the width (maximum width) x3 in the x-axis direction at the first level L1 is 6 (6 squares in the graph). The value counted as one of the values indicating the thickness in the first level L1 is acquired. Further, for the echo k2, the counting unit 835 counts that the width (maximum width) x4 in the x-axis direction at the second level L2 is 12 (12 squares in the graph), and indicates the thickness at the second level L2. Get the value counted as one of the values.
Further, in this example, for the echo k2, the count unit 835 counts that the width (maximum width) y3 in the y-axis direction at the first level L1 is 4 (four squares in the graph), and the thickness at the first level L1 is the above. It is acquired as one of the values indicating the value. Further, regarding the echo k2, the counting unit 835 counts that the width (maximum width) y4 in the y-axis direction at the second level L2 is 8 (eight squares in the graph), and indicates the thickness at the second level L2. Get as one of the values.
Further, in this example, the echo k2 is counted by the counting unit 835 as having an area on the xy plane at the first level L1 of 18 (18 squares in the graph), and the thickness at the first level L1 is shown. Get as one of the values.
Further, for the echo k2, the counting unit 835 counts that the area on the xy plane at the second level L2 is 69 (for 69 squares in the graph), and is 1 of the value indicating the thickness at the second level L2. Get as one.
In the actual flaw detection, the first and second determination units 810 and 820 determine whether or not the value counted by the counting unit 835 falls within the range specified by each range specifying unit 815,823.
In the example of FIG. 9, for example, when the lower limit of the width in the x-axis direction at the first level is 4 and the upper limit is 10, the lower limit of the width in the x-axis direction at the second level is 11 and the upper limit is 20. Since the x-axis width x1 = 6 at the first level of the echo k1 is within the scratch range and the x-axis width x2 = 8 at the second level is not within the shape echo range, the echo k1 is a scratch echo. It is judged. On the other hand, regarding the echo k2, the x-axis direction width x3 = 6 at the first level is in the range of the scratch echo, but the x-axis direction width x4 = 12 at the second level is in the shape echo range, so that the echo k2 is the shape echo. Is determined to be.
The width in the y-axis direction, the area on the xy plane (projected area), and the sum of the widths in the circumferential direction (x-axis direction) are also scratch echoes by comparison with the range set in the same manner as the width in the x-axis direction. It is determined whether it is a shape echo or a shape echo. However, in the actual flaw detection, at the first level, any one of the sum of the width in the x-axis direction, the width in the y-axis direction, the area on the xy plane (projected area), and the width in the circumferential direction (x-axis direction) is used. If it is determined to be a defect echo, that is, a scratch echo, the determination of the remaining items among the above items may be omitted and the determination may proceed to the second level determination. Further, also in the second level, a defect echo, that is, a defect echo in any of the items of the x-axis direction width, the y-axis direction width, the area on the xy plane (projected area), and the circumferential direction (x-axis direction) width. If it is determined to be a scratch echo, the determination of the remaining items among the above items can be omitted.
On the other hand, in the range setting of the range defining unit 814, 823 by the range setting unit 814, 824 using the model axle, the count unit 851 sequentially acquires the count number from the artificial defect, and the range setting unit 814, 824 uses the count unit 851. The lower limit value and the upper limit value of each of the above-mentioned accommodating portions of the first and second range accommodating portions 815 and 823 are updated with reference to the acquired count number.
Specifically, looking at the C-scope display in FIG. 15, the operator sequentially designates artificial defects provided on the model axle m, and updates the lower limit value and the upper limit value of each of the accommodating portions.
In the range setting of the range defining unit 814, 823 by the range setting unit 814, 824 using the model axle m, the count values related to the thickness of the artificial defect newly counted by the counting unit 851 are the first and second range accommodating units. If it is smaller than the lower limit value of each of the above-mentioned accommodating portions of 833 and 823, the lower limit value up to that point is updated by using the count value of the newly counted artificial defect as the lower limit value. Similarly, in the range setting of the range defining units 814 and 823 by the range setting units 814 and 824 using the model axle, the count values related to the thickness of the artificial defect newly counted by the counting unit 851 are the first and second ranges. If it is larger than the upper limit value of each of the above-mentioned accommodating portions 815 and 823, the upper limit value up to that point is updated with the count value of the newly counted artificial defect as the upper limit value. As described above, by aggregating each count value related to the thickness of the artificial defect, the lower limit value and the upper limit value of each of the accommodating portions are determined, and the determined range is used for flaw detection of the actual axle.

In the flaw detection of the actual axle, the first determination unit 810 determines the peak value acquisition unit 805, the first level regulation unit 811, the C scope expansion unit 802, and the first range regulation unit for the above waveform designated by the echo designation unit 851. With reference to 813, the data on the thickness of the specified waveform at the first level, that is, the sum of the width in the x-axis direction, the width in the y-axis direction, the projected area in the xy-plane, and the width of the echo in the x-axis direction, respectively. , It is automatically determined whether or not it is within the range specified by the first range defining unit.
In this example, the first level is calculated using the first calculation standard for the peak of the waveform acquired by the peak value acquisition unit 805. On the other hand, when the first level is calculated for the threshold value using the first calculation standard, the first determination unit 810 stores the threshold value for the above waveform designated by the echo designation unit 851 instead of the peak value acquisition unit 805. Part 803 shall be referred to.
The first determination unit 810 accommodates the determination result data, that is, the first level determination data, in the first determination result accommodating unit 819. The display unit 840 can display the determination data of the first determination result accommodating unit 819 on the first determination result display unit 843.

In the flaw detection of the actual axle, the second determination unit 820 has the peak value acquisition unit 805 and the second level defining unit 821 for the waveform determined by the first determination unit 810 to be within the range of the thickness to be the scratch echo. With reference to the C scope expansion unit 802 and the second range defining unit 823, whether or not the data regarding the thickness of the specified waveform at the second level is within the range specified by the second range defining unit. Judgment is made automatically. That is, the second determination unit 820 is the sum of the width in the x-axis direction, the width in the y-axis direction, the projected area in the xy-plane, and the width of the echo in the x-axis direction as data regarding the thickness of the specified waveform at the second level. Each of the above automatically determines whether or not it is within the range specified by the second range defining unit.
Further, in this example, the second level is calculated using the second calculation standard for the peak of the waveform acquired by the peak value acquisition unit 805. On the other hand, when the second level is calculated for the threshold value using the second calculation standard, the second determination unit 820 stores the threshold value for the above waveform designated by the echo designation unit 851 instead of the peak value acquisition unit 805. Part 803 shall be referred to.
The second determination unit 820 accommodates the determination result data, that is, the second level determination data, in the second determination result accommodating unit 829. The display unit 840 can display the determination data of the second determination result accommodating unit 829 on the second determination result display unit 844.
In this example, the pass / fail determination unit 833 makes a pass / fail determination of the axle 100 with reference to the first determination result accommodating unit 819 and the second determination result accommodating unit 829, and transfers the result of the pass / fail determination to the pass / fail determination result accommodating unit 834. Contain. The pass / fail determination result display unit 845 displays the result of the pass / fail determination housed in the pass / fail determination result accommodating unit 834.
However, the display of the judgments of the first judgment unit 810 and the second judgment unit 820 is used as the pass / fail judgment of the axle, and the pass / fail judgment unit 833, the pass / fail judgment result accommodating unit 834, and the pass / fail judgment result display unit 845 are not separately provided as described above. It may be carried out in.

Each of the above-mentioned regulation units 811, 815, 821, 823 is composed of the above-mentioned flaw detection software and mainly a storage area such as a memory or a hard disk of the computer 9.
Although not shown, the B-scope expansion unit and the B-scope display unit may be provided side by side, and the B-scope may be displayed on the display unit 840 as described above.
The B-scope expansion unit refers to the A-scope data acquisition unit 801 and shows an echo cut into round slices on a plane orthogonal to the axial direction of the axle 100. In the B-scope display, if it is a shape echo, it appears in a circular or arc shape, which contributes to facilitating the visual identification of the shape echo.

The system according to the present invention can automatically perform the presence or absence of a defect in the axle, that is, the pass / fail determination of the axle for the flaw detection of the actual axle. However, the operator can also specify the waveform to be determined whether or not it is a scratch among the displayed waveforms by operating the echo designation unit 851 of the operation unit 850 while looking at the C scope display unit 842 of the display unit 840. ..

(Outline of judgment flow)
As shown in FIG. 11, the flaw detection of the actual axle 100 by this ultrasonic flaw detection method includes the A scope data recording step S100, the C scope expansion step S110, the echo detection step S120 exceeding the threshold value, and the first level determination step S130. , The second level determination step S140 is executed, and the pass / fail determination of the axle 100 is performed.

In step S100, the A-scope data acquired from the ultrasonic probe 1 and the position detection unit 60 is accommodated in the A-scope data acquisition unit 801 as described above under the control of the control unit 860.
In step S110, the data of the A-scope data acquisition unit is expanded and stored as C-scope data in the C-scope expansion unit 802 under the control of the control unit 860.
In step S120, the echo extraction unit 804 extracts echoes exceeding the threshold value from the entire beam path and houses the data of each echo in an identifiable manner.

In step S130, under the control of the control unit 860, the data of the thickness of the waveform at the first level is examined to determine whether or not it is a scratch echo. The method of the determination is as described in the counting unit 851 with reference to FIG. 9 above.
Specifically, in the first level determination step S130, under the control of the control unit 860, the first determination unit 810 executes the determination steps of S131 to S134 shown in FIG. 12 (A).
In the first determination unit 810, the width of the waveform detected in step S131 in the x-axis direction is the range held by the first range defining unit 813, that is, the lower limit value held by the x-axis direction lower limit value accommodating unit 815a and the x-axis. It is determined whether or not the upper limit value held by the direction width upper limit value accommodating portion 815b is within the specified range.
The first determination unit 810 holds the waveform whose width in the y-axis direction is held by the first range defining unit 813 in step S132 for the waveform that is not in the range in step S131, that is, the y-axis direction lower limit value accommodating unit 825a. It is determined whether the lower limit value and the upper limit value held by the y-axis direction width upper limit value accommodating portion 825b are within the specified range.
The first determination unit 810 has a range in which the projected area on the xy plane is held by the first range defining unit 813 in step S133 for the waveform that is not in the range in step S132, that is, the xy area lower limit value accommodating unit 835a. It is determined whether the lower limit value held by the xy area lower limit value and the upper limit value held by the xy area lower limit value accommodating portion 835ab are within the specified range.
The first determination unit 810 determines the range held by the first range defining unit 813 in step S134, that is, the lower limit value held by the circumferential total width lower limit value accommodating unit 818a and the circumferential direction for the waveform determined not to be in the range in step S133. It is determined whether or not the upper limit value held by the total width upper limit value accommodating portion 818b is within the specified range.
For the axle determined by the first determination unit 810 to be within the range in step S134, the pass / fail determination unit 833 determines that the axle has passed, that is, there is no defect.

Control when it is determined in any of steps S131 to S133 that it is within the range of scratches, or when it is determined that it is not within the range of shape echo in step S134, that is, when it is determined that there is a defect in the first level determination S130. Unit 860 shifts the flaw detection system to the process of step S140 of the second level determination.
In step S140, under the control of the control unit 860, the data of the thickness of the waveform at the second level is examined to determine whether or not it is a scratch echo. In step S140, as shown in FIG. 12 (B), the second determination unit S 141 to S144 is executed under the control of the control unit 860, and the second determination unit 820 performs the same as in FIG. 12 (A). Each decision is made at two levels. When the waveform determined to be in the scratch range at the first level is not within the scratch range in any of steps S141 to S143 and is not within the shape echo range in step S144, that is, the second When it is determined in the level determination S140 that there is no defect, the pass / fail determination unit 833 determines that the axle detected for scratches is acceptable, as shown in FIG.
When the waveform determined to be in the scratch range at the first level is determined to be in the scratch range in any of steps S141 to S143, or is not in the shape echo range in step S144, that is, If it is determined in the second level determination S130 that it is not a shape echo, the pass / fail determination unit 833 determines that the flaw-detected axle is rejected.
The order of steps S131 to S133 is not limited to the above, and can be performed in a sequence different from the above. Further, each step of steps S141 to S143 can also be carried out in a different order from the above.

By performing the above flow shown in FIGS. 11 and 12, for example, when echo k1 appears in the C scope data expanded and accommodated in the C scope expansion unit 802, as shown in FIG. 9, the first level The x-direction width x1, the y-direction width y1, and the area of the dark gray portion of L1 are each within the above range held by the first range defining portion 813, or the total sum of the x-direction widths is held by the first range defining portion 813, respectively. If it is determined whether or not it is within the above range and the above determination indicates a defect, the x-direction width x2, the y-direction width y2, and the area of the light gray portion (including the dark gray portion) of the second level L2. It is determined whether any of the sum of the widths in the x-direction falls within the range indicating the defect. In the example of FIG. 9, the echo k2 is a shape echo as described above. The echo k2 is almost the same as the x-direction width x3 = 6, y-direction width y3 = 4, and area 18 at the first level L1 with respect to x1 = 6, y1 = 4, and area 20 of echo k1, which is a scratch echo. , Each numerical value (number of squares) of the second level L2 is thick as x2 = 8, y2 = 6, area 43 of echo k1, and x4 = 12, y4 = 8, area 69 of echo k2, and echoes. It can be seen that k2 is a shape echo in which the tendency of the graph shown in FIG. 18 (B) is remarkable.
By performing each step of FIG. 13 using the model axle m prior to the scratch detection of the actual axle, the scratch echo and the shape of each range held by the first and second range defining portions 815 and 823 as described above. It can be set to an appropriate range where the difference in echo can be discriminated.
In the example shown in FIG. 9, since both echoes k1 and k2 are continuous one waveform in both the first level and the second level, the total sum of the x-direction widths of the echoes is the above-mentioned x-axis at the first level, respectively. It matches the direction widths x1 and x3, and at the second level, it matches the above-mentioned x-axis direction widths x2 and x4, respectively.
On the other hand, as shown in FIG. 10, when a plurality of discontinuous echoes appear, even if the widths x5, x6, x7 of the individual echoes k3, k4, and k5 do not fall within the range of the shape echo, x In the determination of the total sum of the direction width x1, when the total sum x8 (= x5 + x6 + x7) falls within the range of the shape echo, it is determined to be a shape echo. In FIG. 10, only the second level region (light gray) is shown to avoid complication of the drawing.

As described above, steps S10 to S16 shown in FIG. 13 are performed on the model axle m by the operation of the operator and the automatic data processing of the control unit 860 (computer 9), and through each setting unit shown in FIG. Acquire the upper and lower limits of each range specification part shown in FIG. In FIG. 13, "(operator)" indicates a step that requires an operation by an operator, and "(PC)" indicates a process performed by the computer 9. In FIG. 13, only "(PC)" is described in step S13, and the display of "(operator)" is omitted, but the computer moves the cursor pp by operating an input device such as a mouse by the operator.
Specifically, regarding steps S10 to S16 of FIG. 13, on the screen of the monitor shown in FIG. 15, the operator aligns each of about 10 artificial defects of the model axle m with the cursor pp and presses the setting button, and each numerical value is pressed. Is acquired, and the range of each range defining unit is set through each setting unit in the range where all of the above artificial defects can be detected. In FIG. 15, the cursor pp is aligned with one of the artificial defects, and the upper limit value and the lower limit value are adjusted so as to include the artificial defect.
FIG. 15 shows a monitor screen during aggregation of data from an artificial defect of the model axle m.
In FIG. 15, "judgment level 1" is the first level, "judgment level 2" is the second level, and "continuous (long) in the axial length direction" is the y-axis below and to the right of the C scope display. The direction width, "continuous in the circumferential direction (points)" is the width in the x-axis direction, "area (over the level)" is the projected area on the xy plane, and "total number of points in the circumferential direction" is the circumferential direction. The total width is shown respectively.
In FIG. 15, as shown in the lower left of the C scope display, the reference for determining the first level and the second level is the threshold value (scratch determination gate), and the echo height of the threshold value is set to the first level (“determination level 1” 100”. %), And the echo height of 55% of the threshold value is set as the second level (judgment level 2). Therefore, if the threshold is set to 60% of the maximum echo height (100%) that the system can perceive, the first level is the maximum echo height of 60% and the second level is the echo of the above. It will be 33%. However, the first level and the second level may be determined based on the peak (gate super maximum value) of the waveform exceeding the threshold value. Further, the first level is set to an echo height lower than the above standard by a predetermined level (indicated as "-3" dB from the above standard in FIG. 15), and the second level is further set to a predetermined level (in FIG. 15, "-5" from the above standard). (Displayed as dB) It may be set as a low echo height.
The data related to the thickness of the defect pointed to by the cursor pp in FIG. 15 is displayed on the right side of the C scope display.
Judgment level 1: 100.0 [%]
Continuous in the axial length direction: 4 [mm]
Circumferential direction: 6 [points]
Area: 21 [points]
Total number of points in the circumferential direction: 8 [points]
Judgment level 2: 55.0 [%]
Continuous in the axial length direction: 6 [mm]
Circumferential direction: 6 [points]
Area: 29 [points]
Total number of points in the circumferential direction: 6 [points]
Is displayed as.
In the artificial defect of the cursor pp of FIG. 15, the threshold value is the maximum echo height of 100.0%. Therefore, the first level is 100.0% and the second level is 55.0% with respect to the maximum echo height.
Further, FIG. 15 is in the middle of counting due to artificial defects, and only the width in the y-axis direction has a lower limit value of 2 and an upper limit value of 8.
In steps S14 to S16 of FIG. 13, each artificial defect of the model axle m is sequentially determined using the range related to the thickness of the waveform at the first and second levels, and the above aggregation is performed.
After completing each step of FIG. 13, steps S21 to S25 shown in FIG. 14 are executed to perform verification on the model axle and the actual axle.
Specifically, in steps S21 to S22 of FIG. 14, all artificial defects of the model axle m are detected using the range of the thickness indicating the waveform defects at the first level and the second level defined in the flow of FIG. It is verified whether it can be done, the above aggregation is repeated until all can be detected, the above threshold value is adjusted, and the above aggregation is repeated to reset the range indicating the above thickness. Specifically, the threshold is reset to be lower than the echo peaks of all artificial defects.
If all the artificial defects can be detected in steps S21 to S22, in step S23, it is examined whether or not an erroneous determination occurs, that is, whether or not a scratch echo is detected, using an actual axle that has been confirmed to have no scratches by prior scratch detection. If no scratch echo is detected in step S23, the actual axle detected in step S26 is determined to be acceptable. If a scratch echo is detected in step S23, the threshold value and the calculation standard for each level are adjusted in step S24, and the above aggregation is repeated in step 25 to reset the range related to the thickness.
After the end of step S25, the process returns to step S23 and the above process of step S23 is repeated.

(Change example)
In the above-described embodiment, the range of the thickness of the scratch echo at the first level and the range of the thickness of the scratch echo or the shape echo at the second level are determined based on the numerical values aggregated from the artificial defects of the model axis. It is the range that was done. The first-level echo height and the second-level echo height of the echo waveform detected during the flaw detection of the actual axle are often the first-level echo height and the second-level echo height of each artificial defect. Although it is different, the width of the thickness to be used as a scratch echo or shape echo is determined from the total of the thicknesses of multiple artificial defects, and the model axle is used as a model, that is, the actual axle targeted by the model axle is uniformly applicable. The scratch echo and the shape echo are judged by referring to the width.
However, it is not limited to setting the range of the thickness of the second level irrespective of the setting of the range of the thickness in the first level as in the embodiment. When the waveform of the detected echo is displayed three-dimensionally as shown in FIG. 18B, the characteristic of the scratch echo shown as "artificial scratch" in the figure, that is, there is no significant change in the thickness at each height of the echo. If the defect is judged by utilizing the feature that the shape echo becomes thicker from the upper side to the lower side in the figure, the range of the thickness used for the judgment of the first level and the second level. May be determined by another method. For example, the range of the scratch echo or the shape echo of the second echo may be determined based on the range of the thickness of the scratch echo in the first echo. Specifically, the thickness at the first level is compared with the thickness at the second level, and if the thickness at the second level is larger than the thickness at the first level within a predetermined ratio range, it is determined to be a shape echo. May be.

Further, in the examples shown in FIGS. 2 and 3, the specified echo defines the lower limit value and the upper limit value within the range of the wound echo, and the wound echo is positively detected, but it is not a wound echo. It may be possible to specify a range and positively detect that it is not a wound echo.
Regarding the shape echo, the lower limit value and the upper limit value that fall within the range of the shape echo are specified and the shape echo is positively detected. It may be detected in. For example, by detecting a range that does not become a shape echo in the determination at the second level, an echo in which a suspicion of a scratch echo occurs at the first level may be passively determined as a scratch echo.
Further, in addition to the upper and lower two monitoring levels of the first level and the second level, more multi-stage monitoring levels such as the third level or the fourth level or higher may be provided to perform each of the above determinations.

In the embodiment shown in FIGS. 4 to 7, an annular array probe is arranged in the hollow portion 103 of the axle 100, and ultrasonic waves are emitted toward the outside diameter of the axle 100 to detect flaws by the phased array. Although not shown, an annular array probe may be arranged on the radial outer side of the axle 100, and ultrasonic waves may be emitted toward the radial inner side of the axle 100 to detect flaws by the phased array.
Further, the ultrasonic probe 1 may be replaced with the array probe, and a single probe may be used for flaw detection using the axle 100.
An example in which a single probe is adopted as the ultrasonic probe 1 will be described with reference to FIG. 16 (A). Similar to the embodiment shown in FIG. 4A, the rear portion of the probe holding portion 2 is connected to the tip end side of the flexible tube 4, and the front portion thereof is an ultrasonic probe 1 for bevel flaw detection. The rotary head portion 10 provided with the above is connected.
The ultrasonic probe 1 is composed of two probes 1a and 1b. Each of the probes 1a and 1b is a single probe provided with one oscillator for transmission and reception.

The rotary head portion 10 is rotatably fixed to a rotary connector 12 provided inside the probe holding portion 2 via a shaft 13. The rotary connector 12 itself transmits a signal from the rotary head 10. The outside of the shaft 13 is placed in a state of being contained in a tubular body 14 directly fixed to the rotary head portion 10. The tubular body 14 is provided with an oil supply pipe 15 that mediates ultrasonic waves of the probes 1a and 1b inside, and is connected to the oil supply pipe 16 of the probe holding portion 2. This connection is made by incorporating a rotary joint (not shown). A gear 44 is provided on the rear outside of the tubular body 14.

Inside the rear of the probe holding portion 2, a gear 42 connected to the flexible shaft 41 included in the flexible tube 4 is provided. The gear 42 rotates in response to the rotation of the flexible shaft 41 that is rotated by the rotation of the rotary motor 3 described above, and is pivotally stopped by the rotation position detector 6 included in the probe holding portion 2. The rotation of the gear 42 is transmitted to the gear 44 via a shaft 43 provided with gears engaged with the gear 42 and the gear 44 at both ends. As a result, the rotary head portion 10 rotates in response to the rotation of the rotary motor 3.

The rotary head portion 10 is fixed with the ultrasonic probe 1, that is, the two probes 1a and 1b back to back. Reference numeral 11 indicates a signal line for sending the detection signal obtained by the probes 1a and 1b to the flaw detector 8 via the flexible tube 4.
The two probes 1a and 1b are fixed to the rotary head portion 10 so that their orientations face the inner peripheral surface of the boring portion of the axle 100, respectively. A spring 18 is provided between the probes 1a and 1b, and both the probes 1a and 1b are urged to the inner peripheral surface of the boring portion of the axle 100, respectively.
An axial position detecting wire 17 is provided on the metal fitting for fixing the flexible tube 4 to the probe holding portion 2.

A rotary motor is connected to the end 3 (FIG. 4 (A)) of the flexible tube 4 on the winding means 5 side.
In addition, in FIG. 16A, contrary to what is shown in FIGS. 4A and 4B, the right end of the probe holding portion 2 is connected to the flexible tube 4, and the winding means 5 (not shown). The probe holding portion 2 is drawn so as to move to the right side by winding the probe.
For example, when scanning the probe in a spiral direction in the direction indicated by the arrow AZ in FIG. 16A, the rotary head portion 10 is rotated in the direction of the arrow AY by the rotation of the rotary motor, and the flexible tube of the winding means 5 is used. By the unwinding operation of 4, the flexible tube 4 is slid and the rotary head portion 10 is transferred together with the probe holding portion 2 in the AX direction (left direction). By the rotation and transfer, the inside of the boring axle 100 can be spirally detected from the inside. When the flaw is detected in the direction opposite to the arrow AZ, the rotary head portion 10 is transferred together with the probe holding portion 2 in the direction opposite to the AX direction by the winding operation of the winding means 5. At this time, if the rotary motor is rotated in the reverse direction of the above and the rotary head portion 10 is rotated in the direction opposite to the arrow AY, spiral scanning in the direction opposite to the above can be performed.
In the case of normal scanning, in the case of normal scanning, the probe holding unit 2 and the rotating head are in the direction opposite to the AX direction, that is, in the direction in which the winding means 5 is moved by the winding operation. It is preferable to transfer the portion 10 to perform flaw detection.
Also in the embodiment shown in FIG. 16A, the cable bear (registered trademark) 46 and the scanning chain 47 shown in FIG. 4B can be used instead of the flexible tube 4.

In the above-described embodiment, the device configuration suitable for the flaw detection of the boring axle is shown, but the same flaw detection can be performed for the configuration of the device suitable for the flaw detection of the solid axle.
The following is an example of a device suitable for flaw detection of a solid axle. As shown in FIG. 16B, the axle 100 is mounted on a roller 51 that supports and rotates the wheels 101 fitted to the axle 100. In FIG. 16B, only two rollers 51 can be seen, but two rollers 51 are paired with one of the left and right wheels 101, that is, two sets of left and right rollers, that is, If a total of four rollers 51 support the wheels 101 and 101, the wheels 101 can be held more reliably. However, the number of rollers 51 is not limited to the above, and the number of rollers 51 can be changed as long as the axle 100 can be reliably supported.

In FIG. 16B, a gantry frame 52 is installed in the vicinity of the axle 100.
The gantry frame 52 holds the probe 1. More specifically, the portal frame 52 includes a probe 1x that detects the inside of the axle 100 from the outer peripheral surface 107 of the axle 100. In FIG. 16B, 108 shows a gearbox provided on the axle 100.

The probe 1x includes an actuator 20 and is pressed against the surface of the axle 100 at each position. The actuator 20 can press the probe against the surface of the axle 100 by an appropriate fluid pressure such as air or oil pressure or other mechanical means.

The probe 1x detects the inside of the axle 100 from the outer peripheral surface 107 of the axle 100, and can slide the probe 1x along at least the thrust direction e of the axle 100, that is, the axial direction of the axle 100. The sliding means 53 is provided.
Explaining what is illustrated, the sliding means 53 includes a shaft 54 having a threaded peripheral surface, a motor 55 provided at one end of the shaft 54, and a jig 56 screwed to the shaft 54. .. The shaft 54 is arranged parallel to the thrust direction e of the axle 100, and the jig 56 is screwed onto the shaft 54.
An actuator 20 of the probe 1x is fixed to the jig 56. The rotation of the motor 55 causes the shaft 54 to rotate, and the jig 56 screwed to the shaft 54 slides in the longitudinal direction of the shaft 54.
As a result, the probe 1x slides along the thrust direction e of the axle 100.

The sliding means 53 can be implemented in any configuration, and is not limited to the one shown in the figure.
The probe 1x can spirally scan at least the outer peripheral surface 107 of the axle 100 by receiving the rotation from the roller 51 together with the sliding by the sliding means 53.
However, flaw detection may be performed using an array probe in which the vibrators 1p are linearly arranged along the axial direction of the axle 100, and even in the example shown in FIG. 16B, the sliding means 53 is used. It is not limited to physically scanning the probe 1x.

In each of the above-described embodiments, the probe 1 is scanned in a spiral shape on both the boring axle and the solid axle, but the scanning method is not limited to this. It is also possible to scan the probe 1 so as to draw another trajectory with respect to the axle. For example, the probe 1 linearly moves one pitch along the thrust direction of the axle, and after this movement, the axle rotates in the circumferential direction, and such an operation is regarded as one process, and this process is repeated an appropriate number of times. It is also possible.

(summary)
In the present invention, when the material to be inspected is a solid axle, the outer peripheral surface of the axle is used, and when the material to be inspected is a mid-sized axle, the outer peripheral surface or the inner peripheral surface of the axle is used. It scans the sound wave probe, transmits ultrasonic waves, receives echoes, and detects the vicinity of the outer peripheral surface of the axle inside the axle, and detects the reach of the ultrasonic waves in the transmission and reception, that is, the echo height with respect to the beam path. In the ultrasonic flaw detection method for the axle, the echo height that is a predetermined level lower than the predetermined threshold is set as a reference, based on the peak of the echo waveform that exceeds the predetermined threshold or the peak detected from the entire range of the axle flaw detection. The echo height, which is a predetermined level lower than that of the first level, is set as the second level, and the waveforms of the detected echoes are at least those in which the thickness of the waveform at the first level is suspected to be a scratch echo. The thickness of the waveform at the two levels is examined to show the characteristics of the wound echo shown as "artificial flaw" in FIG. 18 (B), or the characteristics of the shape echo shown as "shape" in FIG. 18 (B). It enables automatic judgment as to whether or not it indicates, and eliminates erroneous judgment of scratch echo.

1 Ultrasonic probe 1a probe 1b probe 1p oscillator 1x probe (ultrasonic probe)
2 Detector holding part 3 End part 4 Flexible tube 5 Winding means 6 Rotation position detector 7 Axle mounting part 8 flaw detector 9 Computer 10 Rotating head part 17 Wire 18 Thread 20 Actuator 40 Auxiliary pulley 41 Flexible shaft 46 Cable bear ( Registered trademark)
47 Scanning chain 48 Winding device 49 Self-propelled vehicle 50 Caster 51 Roller 52 Gate frame 54 Shaft 55 Motor 56 Jig 60 Position detection unit 70 Position sensor 81 Data processing unit 82 Interface 90 Printer 100 Axle 101 Wheel 102 Press-fitting unit 103 Hollow Part 107 (Axle 100) Outer surface 500 Scanning part 510 Mediation supply part 520 Sensor part 600 Mechanical part 800 Damage detection part (electrical part)
801 A Scope data acquisition unit 802 C Scope expansion unit 803 Threshold storage unit 804 Echo extraction unit 805 Peak value acquisition unit 810 1st judgment unit 811 1st level regulation unit 812 1st level setting unit 813 1st range regulation unit 814 1st Range setting unit 815a x Axial width lower limit value accommodating unit 815b x Axial width upper limit value accommodating unit 816ay Axial width lower limit value accommodating unit
816by y-axis width upper limit value accommodating part
817a xy area lower limit accommodation unit 817b xy area upper limit accommodation unit 818a Circumferential total width lower limit accommodation unit 818b Circumferential total width upper limit accommodation unit 819 1st judgment result accommodation unit 820 2nd judgment unit 821 2 Level regulation unit 822 2nd level setting unit 823 2nd range regulation unit 824 2nd range setting unit 825a x Axial width lower limit value storage unit 825b x Axial width upper limit value storage unit 826ay y Axial width lower limit value storage unit
826by y-axis width upper limit value accommodating part
827a xy area lower limit accommodation unit 827b xy area upper limit accommodation unit 828a Circumferential total width lower limit accommodation unit 828b Circumferential total width upper limit value accommodation unit 829 2nd judgment result accommodation unit 833 Pass / fail judgment unit 834 Pass / fail judgment Result accommodating unit 835 Counting unit 840 Display unit 841 A Scope display unit 842 C Scope display unit 843 First judgment result display unit 844 Second judgment result display unit 845 Pass / fail judgment result display unit 850 Operation unit 851 Echo designation unit 860 Control unit A Right corner B Left corner e (Axle 100) Thrust direction (axial direction)
pp Cursor V diagonal line (range of probe beam)
W diagonal line (range of probe beam)

Claims (11)

When the material to be inspected is a solid axle, on the outer peripheral surface of the axle, when the material to be inspected is a mid-sized axle, on the outer peripheral surface or inner peripheral surface of the axle, ultrasonic probes for the axial and circumferential directions of the axle. Is scanned, ultrasonic waves are transmitted, echoes are received, and the vicinity of the outer peripheral surface of the axle inside the axle is detected at an oblique angle. In the ultrasonic flaw detection method for axles detected as echo height,
The echo height for determining the scratch echo indicating a scratch is set as the first level, and the echo height lower than the first level is set as the second level.
With respect to the waveform of the echo having a predetermined height or more detected from the entire range where the axle is detected for scratches, when the thickness of the waveform at least at the first level is within the range of the thickness indicating the scratch echo, the echo waveform is at the second level. by examining the thickness of the wave state, and are not for determining the echoes detected whether a flaw echo that shows the wound,
The thickness of the waveform indicates the angle with respect to a specific virtual line orthogonal to the center line of the axle on the x-axis, and the axial position of the axle on the y-axis intersecting the x-axis, and the echo height of the received echo. At the coordinates indicated by the x-axis and the z-axis intersecting the y-axis, the width of the waveform in the x-axis direction, the width of the waveform in the y-axis direction, and the region projected onto the xy plane of the waveform. of the area of the ultrasonic flaw detection method of the axle, wherein at least any one Tsudea Rukoto.
When the material to be inspected is a solid axle, on the outer peripheral surface of the axle, when the material to be inspected is a mid-sized axle, on the outer peripheral surface or inner peripheral surface of the axle, ultrasonic probes for the axial and circumferential directions of the axle. Is scanned, ultrasonic waves are transmitted, echoes are received, and the vicinity of the outer peripheral surface of the axle inside the axle is detected at an oblique angle. In the ultrasonic flaw detection method for axles detected as echo height,
The echo height for determining the scratch echo indicating a scratch is set as the first level, and the echo height lower than the first level is set as the second level.
The angle with respect to a specific virtual line orthogonal to the center line of the axle is shown on the x-axis, and the echo height of the received echo is shown on the z-axis intersecting the x-axis. For the detected echo waveform of a predetermined height or higher, the width in the x-axis direction at the second level is examined when at least the width in the x-axis direction at the first level is within the range of the width of the scratch echo indicating the scratch. A method for ultrasonically detecting an axle, which comprises determining whether or not the detected echo is a scratch echo indicating a scratch.
The echo detected when the width in the x-axis direction at the second level is within the range of the width of the shape echo is determined to be the shape echo of the axle, or the width in the x-axis direction at the second level is the shape echo. The echo detected when it is not within the width range is determined to be a scratch echo.
The ultrasonic wave of the axle according to claim 2, wherein the shape echo is an echo caused by a step provided in the circumferential direction of the axle because the diameter of the axle differs depending on the position of the axle in the axial direction. How to detect a wound.
When the material to be inspected is a solid axle, on the outer peripheral surface of the axle, when the material to be inspected is a mid-sized axle, on the outer peripheral surface or inner peripheral surface of the axle, ultrasonic probes for the axial and circumferential directions of the axle. Is scanned, ultrasonic waves are transmitted, echoes are received, and the vicinity of the outer peripheral surface of the axle inside the axle is detected at an oblique angle. In the ultrasonic flaw detection method for axles detected as echo height,
The echo height for determining the scratch echo indicating a scratch is set as the first level, and the echo height lower than the first level is set as the second level.
The axial position of the axle is indicated on the y-axis, and the echo height of the received echo is shown on the z-axis intersecting the y-axis. With respect to the echo waveform, when the width in the y-axis direction at least at the first level is within the range of the width of the scratch echo indicating scratches, the echo detected by examining the width in the y-axis direction at the second level An ultrasonic flaw detection method for an axle, which comprises determining whether or not it is a scratch echo indicating a scratch.
The echo detected when the width in the y-axis direction at the second level is within the range of the width of the shape echo is determined to be the shape echo of the axle, or the width in the y-axis direction at the second level is the shape echo. The echo detected when it is not within the width range is determined to be a scratch echo.
The ultrasonic wave of the axle according to claim 4 , wherein the shape echo is an echo caused by a step provided in the circumferential direction of the axle because the diameter of the axle differs depending on the position of the axle in the axial direction. How to detect a wound.
When the material to be inspected is a solid axle, on the outer peripheral surface of the axle, when the material to be inspected is a mid-sized axle, on the outer peripheral surface or inner peripheral surface of the axle, ultrasonic probes for the axial and circumferential directions of the axle. Is scanned, ultrasonic waves are transmitted, echoes are received, and the vicinity of the outer peripheral surface of the axle inside the axle is detected at an oblique angle. In the ultrasonic flaw detection method for axles detected as echo height,
The echo height for determining the scratch echo indicating a scratch is set as the first level, and the echo height lower than the first level is set as the second level.
The angle with respect to a specific virtual line orthogonal to the center line of the axle is shown on the x-axis, the axial position of the axle is shown on the y-axis intersecting the x-axis, and the echo height of the received echo is shown on the x-axis. And in the three-dimensional coordinates shown on the z-axis that intersects the y-axis, the region of the echo waveform of the predetermined height or higher detected from the entire range where the axle is detected is projected onto the xy plane at least at the first level. Whether or not the detected echo is a scratch echo by examining the area of the region projected onto the xy plane at the second level when the area of is within the range of the scratch echo area indicating the scratch. An ultrasonic flaw detection method for an axle, which is characterized in that the determination is made.
The echo detected when the area of the region projected onto the xy plane at the second level is within the area of the shape echo is determined to be the shape echo of the axle, or the xy plane at the second level. The echo detected when the area of the area projected onto is not within the area of the shape echo is determined to be a scratch echo.
The ultrasonic wave of the axle according to claim 6 , wherein the shape echo is an echo caused by a step provided in the circumferential direction of the axle because the diameter of the axle differs depending on the position of the axle in the axial direction. How to detect a wound.
When the material to be inspected is a solid axle, on the outer peripheral surface of the axle, when the material to be inspected is a mid-sized axle, on the outer peripheral surface or inner peripheral surface of the axle, ultrasonic probes for the axial and circumferential directions of the axle Is scanned, ultrasonic waves are transmitted, echoes are received, and the vicinity of the outer peripheral surface of the axle inside the axle is detected at an oblique angle. In an axle ultrasonic flaw detection system that detects as echo height,
The echo height for determining the scratch echo indicating a scratch is set as the first level, the echo height lower than the first level is set as the second level, and the height is equal to or higher than the predetermined height detected from the entire range for detecting the axle. Regarding the echo waveform, when the thickness of the waveform at least at the first level is in the range of the thickness indicating the scratch echo, the echo detected by examining the thickness of the waveform at the second level is scratched. It is to judge whether or not it is a wound echo indicating
A transmission probe and a reception probe that are ultrasonic probes, a scanning unit that scans the ultrasonic probe, a position detection unit, an A-scope data acquisition unit, a C-scope expansion unit, and a predetermined sound. A threshold storage unit that holds the pressure level as a threshold value, an echo extraction unit, an echo designation unit, a peak value acquisition unit, a first level setting unit, a first level regulation unit, a first range regulation unit, and a first It is provided with a two-level setting unit, a second level regulation unit, a second range regulation unit, a first determination unit, and a second determination unit.
The A-scope data acquisition unit is an A-scope of the waveform of the echo from the beam path at the scanning position of the probe detected by the position detection unit and the echo height of the echo received by the receiving probe for substantially the entire area of the beam path. Data is sequentially acquired and stored,
The C-scope expansion unit indicates the angle of the waveform data contained in the A-scope data acquisition unit with respect to a specific virtual line orthogonal to the center line of the axle on the x-axis, and the axial position of the axle is indicated by the axis. The data shown on the y-axis intersecting the x-axis and showing the echo height of the received echo in different colors according to the height is stored as C-scope data.
The echo extraction unit refers to the threshold value held by the threshold value storage unit, extracts all echo waveforms exceeding the threshold value from the data acquired by the C scope expansion unit, and identifies each of the extracted waveforms. It contains data and
The echo designation unit automatically or accepts an operator's operation and specifies data of an arbitrary waveform in the C scope expansion unit extracted by the echo extraction unit.
The peak value acquisition unit acquires data on the echo height of the peak of the waveform specified by the echo designation unit with reference to the C scope expansion unit.
The first level setting unit holds a calculation standard capable of calculating the echo height data to be the first level from the echo height data.
The first range defining part holds the upper limit value and the lower limit value of the data regarding the thickness of the waveform used for the determination of the scratch echo at the first level.
The second level setting unit holds a calculation standard capable of calculating the echo height data to be the second level from the echo height data.
The second range defining unit holds at least a lower limit value of data regarding the thickness of the waveform used for determining the shape echo at the second level.
The first level defining unit acquires data on the echo height as the first level with reference to the first level setting unit for the waveform specified by the echo designation unit.
The first determination unit refers to the first level regulation unit, the C scope expansion unit, and the first range regulation unit for the waveform designated by the echo designation unit, and calculates the first level. It is automatically determined whether or not the data regarding the thickness of the specified waveform in the specified range is within the range specified by the first range defining unit.
The second level defining unit acquires data on the echo height as the second level with reference to the second level setting unit for the waveform specified by the echo designation unit.
The second determination unit defines the second level, the C scope expansion unit, and the second range for a waveform determined by the first determination unit to be at least in the range of the thickness to be a scratch echo. With reference to the section, it is automatically determined whether or not the calculated data regarding the thickness of the specified waveform at the second level is within the range specified by the second range defining section. can be,
The data regarding the thickness of the waveform is at least one of the width in the x-axis direction of the waveform, the width in the y-axis direction of the waveform, and the area of the region projected onto the xy plane of the waveform. An axle ultrasonic flaw detection system characterized by being present.
The data regarding the thickness of the waveform appears in the x-axis direction, the width of the waveform in the x-axis direction, the width of the waveform in the y-axis direction, the area of the region projected onto the xy plane of the waveform, and the x-axis direction. At least one of the sum of the widths of all waveforms.
The sum of the widths of all the waveforms appearing in the x-axis direction is the sum of the waveforms appearing in the x-axis direction regardless of whether the widths in the x-axis direction are a continuous single waveform or a plurality of discontinuous waveforms. The ultrasonic flaw detection system for an axle according to claim 8, wherein the total width is the sum of all widths.
The transmitting probe is a transmitting / receiving probe that also serves as the receiving probe.
The transmission / reception probe is an array probe in which a plurality of vibrators are arranged in a ring shape along the circumferential direction of the axle, and the transmission / reception probe emits ultrasonic waves at an angle to the center line of the axle. It is a phased array that emits ultrasonic waves at once from a predetermined number of adjacent groups of oscillators among the oscillators that are arranged in such a manner and that constitute the annular array probe.
Scanning of the transmission / reception probe with respect to the circumferential direction of the axle is performed by sequentially shifting the group that emits ultrasonic waves at one time in the circumferential direction of the annular array probe.
Two transmission / reception probes are arranged in front of and behind the transmission / reception probe, one of the transmission / reception probes is directed to the opposite side of the other transmission / reception probe in the axial direction, and both transmission / reception probes are the same. Scanned together in the axial direction
The ultrasonic wave of an axle according to claim 8 or 9, wherein both transmission / reception probes spirally transmit ultrasonic waves to the axle by scanning in the circumferential direction and scanning in the axial direction. Flaw detection system.
A first range setting unit and a second range setting unit are provided.
Prior to flaw detection of the actual axle, that is, the actual axle, flaw detection is performed on a model axle having the same dimensions, shape, and material as the actual axle, and having an artificial scratch of a predetermined shape and size provided at a predetermined position. It is a thing
The A-scope data acquisition unit sequentially acquires and accommodates the A-scope data of the model axle by flaw detection of the model axle.
The C-scope expansion unit acquires and accommodates the C-scope data for the waveform data of the model axle expropriated by the A-scope data acquisition unit.
The echo extraction unit refers to the threshold value held by the threshold value storage unit, extracts waveforms of all echoes exceeding the threshold value from the data of the model axle accommodated by the C scope expansion unit, and extracts each of the extracted waveforms. It contains identification data that identifies the waveform.
The first level setting unit is the ratio of the echo height as the first level to the echo height of the peak of the waveform to be determined by the first determination unit, and the echo height as the first level with respect to the threshold value of the threshold storage unit. The ratio, the value of the difference in the sound pressure level of the first level with respect to the peak of the waveform to be judged by the first determination unit, or the value of the difference in the sound pressure level of the first level with respect to the threshold value is used for the calculation. It is to be kept as a standard and
The second level setting unit is the ratio of the echo height as the second level to the echo height of the peak of the waveform to be determined by the second determination unit, and the echo height as the second level with respect to the threshold value of the threshold storage unit. The ratio, the value of the difference in the sound pressure level of the second level with respect to the peak of the waveform to be judged by the second determination unit, or the value of the difference in the sound pressure level of the second level with respect to the threshold value is used for the calculation. It is to be kept as a standard and
By operating the echo designation unit of the operator to sequentially specify each of the waveform data in the C scope expansion unit extracted by the echo extraction unit of the model axle, the peak value acquisition unit obtains the specified waveform. Get the peak value of the echo height of
The first level defining unit acquires the value of the echo height of the first level for the specified waveform of the model axle with reference to the first level setting unit and the peak value acquisition unit.
The second level defining unit acquires the value of the echo height of the second level for the specified waveform of the model axle with reference to the second level setting unit and the peak value acquisition unit.
The first range setting unit is designated by sequentially designating each of the waveform data in the C scope expansion unit extracted by the echo extraction unit of the model axle by operating the echo designation unit of the operator. The upper limit value and the lower limit value of the first range defining portion are automatically set so that all the data related to the thickness at the first level of the waveform is within the range.
The second range setting unit is designated by sequentially designating each of the waveform data in the C scope expansion unit extracted by the echo extraction unit of the model axle by operating the echo designation unit of the operator. The lower limit value of the second range defining portion is automatically set so that all the data related to the thickness at the second level of the waveform is within the range.
The first determination unit and the second determination unit refer to each of the upper limit value and the lower limit value in the first range regulation unit and the second range regulation unit acquired on the model axle for the determination of the actual axle. The ultrasonic flaw detection system for an axle according to any one of claims 8 to 9.

Images