JP2009222425A - Nondestructive inspecting device and nondestructive inspecting method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a nondestructive inspecting method capable of more easily measuring a stress intensity factor at a stress concentration part at higher precision without detecting cracks at the stress concentration part of a member to evaluate the degree of fatigue at the stress concentration part even if the crack has actually not occurred at the stress concentration part. <P>SOLUTION: At first the density distribution of magnetic fluxes at the surface of the stress concentration part of a sample is measured. Then, it is detected that the propagation of cracks from the stress concentration part is shown as a curved line protruding toward downward of the magnetic flux density distribution along the estimated direction. When the magnetic flux density distribution shows the curved shape, the distance from the stress concentration part to the top of the curved shape is measured. Based on the linear relationship between the distance and the stress intensity factor at the stress concentration part, the stress intensity factor is determined from the distance. When the magnetic flux density distribution does not show the curved shape, it is determined that the fatigue does not generate at the stress concentration part yet. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to non-destructive inspection of members, and more particularly, to use of non-destructive inspection for strength evaluation of members based on fracture mechanics.

  The strength of the member is evaluated based on the stress distribution in the member. Here, when the member includes a crack or a defect, the stress distribution is specific in the vicinity of the crack or the defect. In addition to cracks and defects, the stress distribution is substantially specific in a portion where stress is very likely to concentrate, such as a corner of a member or a joint between members, that is, a stress concentration portion. It is known that such a specific stress distribution, that is, a stress singular field, is characterized by a parameter called a stress intensity factor. Therefore, a stress intensity factor is generally used for strength evaluation in the vicinity of cracks, defects, and stress concentration portions. For example, if the stress intensity factor of a defect of a certain member is equal to or greater than the fracture toughness value of the member, it can be determined that brittle fracture starting from the defect occurs in the member. In addition, the growth rate of fatigue cracks due to repeated loads is called the width of the fluctuation of stress intensity factors associated with the repeated loads, that is, the stress intensity factor width (also referred to as the stress intensity factor range) and the Paris law. Satisfy the relationship. By utilizing this relationship, it is possible to evaluate the progress stage of the crack from the stress intensity factor width, and further to predict the life until the crack reaches the ductility or brittle fracture of the member.

As a conventional method for obtaining a stress intensity factor from an actual member by nondestructive inspection, for example, a method described in Patent Document 1 is known. This method utilizes the fact that the spontaneous magnetization in the vicinity of the tip of the crack in the metal has a linear relationship with the stress intensity factor width of the crack. Specifically, the spontaneous magnetization is measured near the tip of the crack using a Hall element microscope, and the stress intensity factor width of the crack is calculated from the obtained spontaneous magnetization based on the above linear relationship. In addition, a method for obtaining a stress intensity factor from a half-value width distribution of X-rays transmitted through a member is also known.
Japanese Unexamined Patent Publication No. 2007-071657 JP 2002-277442 A JP 2006-242701 A

  In the conventional method described in Patent Document 1, spontaneous magnetization in the vicinity of the crack tip must be measured with sufficiently high accuracy in order to obtain the stress intensity factor with sufficiently high accuracy. However, since spontaneous magnetization is generally susceptible to environmental temperature, external magnetic field, or ferromagnetic impurities, it is not easy to measure spontaneous magnetization with high accuracy. Furthermore, the above method maintains sufficiently high measurement accuracy of spontaneous magnetization by utilizing the fact that paramagnetic metal changes to ferromagnetism in the plastic region in the vicinity of the crack. Therefore, when the whole member exhibits ferromagnetism, it is difficult to maintain the measurement accuracy of spontaneous magnetization sufficiently high. The above method also requires a step of detecting a crack in advance, and cannot be applied to a member having no crack.

  In the method of obtaining the stress intensity factor from the X-ray half width distribution, it is not easy to sufficiently improve the X-ray resolution so that the stress intensity factor can be obtained with sufficiently high accuracy. Furthermore, it is difficult to further shorten the measurement time.

  The object of the present invention is to measure the stress intensity factor of the stress concentration portion with higher accuracy and more easily without detecting a crack from the stress concentration portion of the member. It is an object of the present invention to provide a non-destructive inspection method that can evaluate the degree of fatigue of the stress concentration portion even if no occurrence occurs.

Unlike the conventional method, the nondestructive inspection method according to the present invention uses the following facts discovered by the inventors of the present application.
If the sample is formed from a material exhibiting ferromagnetism at least in the stress-concentrated region, the surface magnetic flux density distribution has an irregular shape while fatigue is not progressing much at and near the stress-concentrated portion of the sample. Indicates. Here, the stress concentration portion is a portion where stress is particularly concentrated and the stress field becomes specific, and is preferably determined by the structure of the member such as the shape and composition of the member. The irregular shape is different for each sample, but hardly changes until fatigue of the stress concentration portion progresses to some extent. When the fatigue of the stress concentrated portion reaches a certain level, the shape of the magnetic flux density distribution on the surface at the stress concentrated portion and its vicinity changes rapidly from the irregular shape before that. This abrupt change is earlier than the time point when the stress concentration portion is cracked. If a crack occurs in the stress concentration portion, the magnetic flux density distribution on the surface is convex downward or convex upward from the stress concentration portion along the direction of crack propagation. This tendency does not depend on the sample. The downward convex curve shape appears when the magnetic flux density distribution is plotted with the north pole facing upward, and the upward convex curve shape is the opposite. The distance from the stress concentration portion to the vertex of the curved shape, that is, the point where the inclination is equal to 0 is larger than the length of the crack. The distance is determined by the stress intensity factor of the stress concentration portion, and in particular, the distance is proportional to the stress intensity factor width of the stress concentration portion.

  The nondestructive inspection method according to one aspect of the present invention preferably includes the following steps in order. First, the magnetic flux density distribution on the surface of the sample is measured in a predetermined region including the stress concentration portion of the sample. Further, the predetermined region includes a direction in which a crack is expected to develop when a crack occurs in the stress concentration portion. Next, it is detected that the measured magnetic flux density distribution shows a curved shape convex downward or convex upward along a certain direction from the stress concentration portion. When the measured magnetic flux density distribution shows any one of these curve shapes, the distance from the stress concentration portion to the vertex of the curve shape is obtained. By utilizing a predetermined relationship between the distance and the stress intensity factor width of the stress concentration portion, in particular, a linear relationship, the stress intensity factor of the stress concentration portion is determined based on the distance.

  Preferably, data indicating a linear relationship between the distance from the stress concentration portion to the vertex of the curve shape and the stress intensity factor of the stress concentration portion is stored in a database in advance. In that case, in the step of determining the stress intensity factor of the stress concentration portion, data indicating the linear relationship is read from the database and used.

  The nondestructive inspection method according to another aspect of the present invention preferably includes the following steps in order. First, when the stress concentration portion of the sample is in an initial state where fatigue has not yet occurred, the magnetic flux density at at least one point on the surface of the stress concentration portion, or a predetermined direction from the stress concentration portion along the surface of the sample Measure the magnetic flux density distribution in the area extending to, and store the data indicating the measurement results in a database. Next, the magnetic flux density at the same point or the magnetic flux density distribution in the same region is newly measured. Further, the amount of change between the newly measured magnetic flux density at the one point or the magnetic flux density distribution in the region and the magnetic flux density or magnetic flux density distribution stored in the database is compared with a predetermined threshold value. Here, as the amount of change between the magnetic flux density distributions, the maximum value of the amount of change between the magnetic flux densities at each measurement point included in the region is preferably used. If the amount of change is less than the threshold, it is determined that the fatigue of the stress concentration portion has not reached the predetermined level, and if the amount of change exceeds the threshold, the fatigue of the stress concentration portion has progressed to the predetermined level. Judge.

  Unlike the conventional method, the nondestructive inspection method according to the present invention can calculate the stress intensity factor without detecting a crack from the stress concentration portion. Therefore, a crack detection operation is not necessary. In particular, since the apex of the curved shape that is convex downward or upward indicated by the magnetic flux density distribution on the surface develops before the tip of the crack, even if the crack does not progress to the detectable length, The stress intensity factor at the tip can be determined. Furthermore, unlike the absolute value of the magnetic flux density at a single point on the sample surface, the overall shape of the magnetic flux density distribution is not easily affected by disturbances. It is easier to accurately measure the absolute value of the magnetic flux density at one point. Even if the entire sample is made of a ferromagnetic metal, the above distance can be measured with sufficiently high accuracy. Thus, unlike the conventional method, the above-described method according to the present invention measures the stress intensity factor of the stress concentration portion more easily and more easily without actually detecting a crack from the stress concentration portion. it can.

  The nondestructive inspection method according to the present invention further detects an abrupt change in the magnetic flux density distribution on the surface at and near the stress concentration portion during a period in which no crack occurs in the stress concentration portion. Thereby, unlike the conventional method, fatigue of the stress concentration portion can be detected before a crack actually occurs in the stress concentration portion.

Hereinafter, preferred embodiments of the present invention will be described with reference to the drawings.
FIG. 1 shows a block diagram of a nondestructive inspection apparatus according to an embodiment of the present invention. This inspection apparatus preferably includes a magnetic flux density measurement unit 1 and a fatigue evaluation unit 2. As shown in FIG. 1, the magnetic flux density measurement unit 1 preferably includes a probe 11, a gauss meter 12, stages 13A and 13B, position sensors 14A and 14B, and a position control unit 15. The fatigue evaluation unit 2 preferably includes a stress intensity factor calculation unit 21 and a database 22.

  The probe 11 is preferably in the form of an elongated bar, and includes a magnetic sensor 11A at the tip thereof. The magnetic sensor 11A is preferably a Hall element. In addition, a magnetoresistive element may be used. The gauss meter 12 drives the magnetic sensor 11A and receives a signal output from the magnetic sensor 11A accordingly. The gauss meter 12 further determines the magnetic flux density penetrating the magnetic sensor 11A in a predetermined direction from the signal. For example, when the magnetic sensor 11A is a Hall element, the gauss meter 12 applies a predetermined amount of current to the Hall element, and accordingly determines the magnetic flux density penetrating the Hall element from the Hall voltage generated in the Hall element. The Hall element is preferably a GaAs element, and its magnetic sensitive area is preferably 10 μm × 10 μm. Furthermore, the resolution of the Gauss meter 12 in that case is preferably 0.01 mT.

  The stage preferably includes an XY stage 13A and a Z stage 13B. The XY stage 13A is preferably a platform having a horizontal upper surface and can be automatically displaced in a horizontal plane. A sample 4 to be inspected is fixed on the upper surface of the XY stage 13A. The Z stage 13B is preferably a table for fixing the end of the probe 11, and is supported by a support column 13C extending in the vertical direction. Preferably, the Z stage 13B supports the probe 11 so that the level of the output signal of the magnetic sensor 11A corresponds to the strength of the magnetic flux penetrating the magnetic sensor 11A in the vertical direction. The Z stage 13B can automatically slide along the column 13C. The position of the tip of the probe 11 with respect to the upper surface of the XY stage 13A changes due to the displacement of the XY stage 13A, and the distance from the upper surface of the XY stage 13A to the tip of the probe 11 changes due to the displacement of the Z stage 13B.

  The position sensor preferably includes a horizontal laser displacement meter 14A and a vertical laser displacement meter 14B. The horizontal laser displacement meter 14A irradiates the side surface of the XY stage 13A with the laser beam LA and measures the distance from the reflected light to the side surface. The vertical laser displacement meter 14B preferably irradiates a plurality of laser beams LB toward the upper surface of the XY stage 13A, and measures the distance from the light reflected by the upper surfaces of the sample 4 and the tip of the probe 11 to each upper surface.

  The position control unit 15 is preferably a functional unit realized by the personal computer 3 executing a predetermined program. The position controller 15 preferably obtains the position of the XY stage 13A in the horizontal plane based on the distance measured by the horizontal laser displacement meter 14A and the displacement history of the XY stage 13A, and further monitors the position while monitoring the position of the XY. The stage 13A is displaced. On the other hand, the position control unit 15 preferably calculates the height of the magnetic sensor 11A from the upper surface of the sample 4 based on the distance measured by the vertical laser displacement meter 14B and the position of the magnetic sensor 11A within the tip of the probe 11. Then, the Z stage 13B is displaced while monitoring its height.

  The fatigue evaluation unit 2 is preferably a functional unit realized by the personal computer 3 executing a predetermined program. In particular, the database 22 is preferably a storage means such as a semiconductor memory or an HDD built in the personal computer 3. In addition, external storage means externally connected to the personal computer 3 or connected through a network may be used as the database 22. The fatigue evaluation unit 2, preferably the stress intensity factor calculation unit 21, causes the magnetic flux density measurement unit 1 to measure the distribution of the magnetic flux density emitted from the surface of the sample 4 in the vertical direction. That is, the gauss meter 12 is made to repeatedly measure the magnetic flux density at a predetermined interval, preferably continuously, while the position control unit 15 displaces the stages 13A and 13B in a predetermined pattern. Thereby, the magnetic flux density distribution is measured particularly around the stress concentration portion of the sample 4.

  Here, the sample 4 is made of a material exhibiting ferromagnetism, preferably steel, at least in a region where stress is concentrated. In particular, the sample 4 may be formed of a material exhibiting ferromagnetism as a whole, such as a high carbon chromium bearing steel SUJ2. The stress concentration portion of the sample 4 is a portion where stress is concentrated due to the structure of the sample 4 and the stress field becomes specific. For example, when the sample 4 is a rectangular flat plate as shown in FIG. 2 and includes a slit 4A near the center of its long side, the tip 4B of the slit 4A and the vicinity thereof are stress concentration portions. In addition, a corner of a member, a joint portion between different members, or the like can be a stress concentration portion.

  The stress intensity factor calculation unit 21 preferably causes the position control unit 15 to translate the XY stage 13A in a certain direction, thereby causing the magnetic flux density measurement unit 1 to transfer the magnetic flux density from the stress concentration part of the sample 4 in the direction opposite to the translation direction. Let the distribution be measured. Here, the reverse direction is preferably a direction in which a crack generated in the stress concentration portion propagates. When the sample 4 includes, for example, the slit 4A shown in FIG. 2, if a crack occurs at the tip 4B of the slit 4A that is a stress concentration portion, the crack is generally indicated by the sample shown in FIG. It develops in a direction perpendicular to the long side of 4. Therefore, the stress intensity factor calculation unit 21 causes the magnetic flux density measurement unit 1 to measure the magnetic flux density distribution in the region 4C that linearly extends from the tip 4B of the slit 4A to a predetermined distance in that direction.

  The stress intensity factor calculation unit 21 preferably causes the magnetic flux density measurement unit 1 to continuously measure the magnetic flux density during the translation period of the XY stage 13A. At that time, the gauss meter 12 continuously outputs a series of measurement values to the stress intensity factor calculation unit 21 as an analog signal. The stress intensity factor calculation unit 21 stores the analog signal together with the elapsed time from the translation start time of the XY stage 13A. From the elapsed time and the moving speed of the XY stage 13A, the stress intensity factor calculating unit 21 calculates the magnetic flux density distribution from the analog signal by calculating the distance that the XY stage 13A has moved during the elapsed time.

  The stress intensity factor calculation unit 21 preferably removes, as noise, a component having a spatial fluctuation frequency equal to or higher than a predetermined threshold value from the magnetic flux density distribution obtained as described above. In particular, when a Hall element is used as the magnetic sensor 11A, the Gauss meter 12 periodically inverts the polarity of the current applied to the Hall element. At that time, in the analog signal output from the gauss meter 12, noise having the same frequency as the polarity inversion of the current applied to the Hall element, that is, switching noise is generated. Therefore, the stress intensity factor calculation unit 21 preferably includes a low-pass filter or a digital multimeter, and removes components above the switching noise frequency from the analog signal using them.

  Furthermore, the stress intensity factor calculation unit 21 detects that the magnetic flux density distribution after removing the noise is plotted to show a downward or upward convex curve shape. Here, if the fatigue at the stress concentration portion has progressed to some extent, the above-mentioned magnetic flux density distribution generally shows a downward convex curve shape when plotted with the N pole facing upward, and the plot with the S pole facing upward. When it does, it shows a convex curve shape. FIG. 4 shows an example of the magnetic flux density distribution along the linear region 4C shown in FIG. The vertical axis in FIG. 4 indicates the magnitude of the magnetic flux density with the N pole facing upward, and the horizontal axis indicates the distance from the base end O of the slit 4A to the measurement point of the magnetic flux density. In FIG. 4, fatigue progresses around the tip 4B of the slit 4A in the order of the solid line, the broken line, and the alternate long and short dash line. In the initial state where the periphery of the tip portion 4B is not so fatigued, the magnetic flux density distribution is irregular as shown by the solid line in FIG. The irregular shape varies from sample 4 to sample 4. However, if the fatigue reaches a certain level around the tip portion 4B, the above magnetic flux density distribution changes to a downwardly convex curve shape as shown by the broken line and the alternate long and short dash line in FIG. Further, as the fatigue progresses and the crack progresses, the curve shape deforms in the order of the broken line and the alternate long and short dash line in FIG. This tendency does not depend on sample 4. Preferably, data of magnetic flux density distribution measured when the same stress concentration portion of the sample 4 is in an initial state where fatigue has not yet occurred is stored in the database 22 in advance. The stress intensity factor calculation unit 21 uses the data to confirm that the newly measured curve shape of the magnetic flux density distribution has changed significantly from an irregular shape in the initial state to a curved shape convex downward or upward. Determine.

  In addition to the above, the stress intensity factor calculation unit 21 causes the magnetic flux density measurement unit 1 to measure the magnetic flux density distribution in a wide range around the stress concentration part, and protrudes downward or upward in a certain direction from the stress concentration part. A portion showing a curved shape may be searched from the obtained magnetic flux density distribution.

  When the magnetic flux density distribution shows a downward or upward convex curve shape from the stress concentration part along the crack propagation direction, the stress intensity factor calculation unit 21 calculates the distance from the stress concentration part of sample 4 to the vertex of the curve shape. The stress intensity factor width of the stress concentration portion is determined based on the calculated distance. Here, the distance is generally linearly related to the stress intensity factor width of the stress concentration portion. FIG. 5 shows an example of the linear relationship. FIG. 5 is obtained particularly for the tip 4B of the slit 4A shown in FIG. 3, and the vertical axis indicates the distance from the base end O of the slit 4A to the apex of the curved shape, and the horizontal axis. Indicates the stress intensity factor width of the tip 4B of the slit 4A. The database 22 preferably stores in advance data indicating a linear relationship as shown in FIG. The data preferably represents a list of combinations of the above distances and the corresponding stress intensity factor widths. In addition, the slope and intercept of a linear expression that should be satisfied by the distance and the corresponding stress intensity factor width may be represented. The stress intensity factor calculation unit 21 uses the data stored in the database 22 to determine the stress intensity factor width of the stress concentration part corresponding to the distance. The fatigue evaluation section 2 preferably evaluates the crack growth rate generated in the stress concentration section based on the Paris law based on the stress intensity factor width determined as described above, and more preferably predicts the life of the sample 4.

  Preferably, the stress intensity factor calculation unit 21 uses the following fact to determine whether or not the newly measured magnetic flux density distribution has changed from an irregular curve shape in the initial state to a curve shape convex downward or upward. Is determined.

  When the fatigue of the stress concentration portion reaches a certain level, the shape of the magnetic flux density distribution on the surface accompanying the stress concentration rapidly changes from the irregular shape in the initial state at and near the stress concentration portion before the crack occurs. For example, in the case where the sample 4 includes the slit 4A shown in FIG. 3, the rapid change seen in the magnetic flux density distribution in the region 4C linearly extending from the tip portion 4B along the crack propagation direction. As shown in FIG. 6 (a) and 6 (b), the stress intensity factor width in the initial state of the tip 4B of the slit 4A is different. Until the fatigue reaches a certain level, the magnetic flux density distribution in the linear region 4C maintains an irregular curve shape in the initial state shown by the solid line in FIG. In particular, at each point of the curve, the magnetic flux density changes from the initial value to a level that cannot be distinguished by the gauss meter 12 even if it is large. On the other hand, when the fatigue reaches a certain level, the magnetic flux density distribution in the linear region 4C rapidly changes from the initial curve shape shown by the solid line in FIG. 6 to the curve shape shown by the broken line. . In particular, at a point where the distance from the base end portion O of the slit 4A is the same, the magnetic flux density generally changes to such an extent that it can be sufficiently identified by the gauss meter 12. At this stage, no crack is generated in the tip 4B of the slit 4A. In addition, when the curved line shape shown by the broken line in FIG. 6 is regarded as a downwardly curved line shape, the distance between the apex and the base end O of the slit 4A is the tip of the slit 4A at that time. The stress intensity factor width of 4B and the linear relationship shown in FIG. 5 are satisfied.

  Preferably, the stress intensity factor calculation unit 21 first removes a bias component caused by a disturbance from the newly measured magnetic flux density distribution. Here, since the bias component is generally common except for the stress concentration portion, the stress intensity factor calculation portion 21 preferably measures the magnetic flux density even at a point outside the stress concentration portion, and the measured value and the value in the initial state. The bias component is determined from the difference between and. The initial value is preferably stored in the database 22 in advance. Next, the stress intensity factor calculation unit 21 compares the curve shape of the magnetic flux density distribution from which the bias component has been removed with the curve shape of the initial state stored in the database 22, and at least one measurement point located in the stress concentration portion, Preferably, the amount of change in magnetic flux density at a singular point of the stress field or the amount of change in magnetic flux density distribution in the same region including the stress concentration portion is obtained. Here, as the change amount of the magnetic flux density distribution, preferably, the maximum value of the change amount of the magnetic flux density at each measurement point included in the above-described region is used. The stress intensity factor calculation unit 21 further compares the obtained change amount with a predetermined threshold, preferably the resolution of the gauss meter 12. If the amount of change is less than the threshold value, the stress intensity factor calculation unit 21 determines that the fatigue of the stress concentration part has not yet reached the level where the stress intensity factor can be calculated from the curve shape of the magnetic flux density distribution. If the amount of change exceeds the threshold value, the stress intensity factor calculation unit 21 has progressed to the above degree of fatigue at the stress concentration portion, so that the measured magnetic flux density distribution has a downward convex curve shape. Judge.

  Utilizing the above configuration, the above-described nondestructive inspection apparatus inspects fatigue of the stress concentration portion of the sample 4 preferably by the following procedure. FIG. 7 shows a flowchart of the inspection method.

Step S1: The sample 4 is fixed at a predetermined position on the upper surface of the XY stage 13A, and the tip of the probe 11 is disposed above the stress concentration portion of the sample 4. Preferably, the magnetic sensor 11A is disposed immediately above the stress concentration portion. Further, the Z stage 13B is displaced by the position control unit 15, and the magnetic sensor 11A is moved close to a predetermined height, preferably about 1 mm, from the stress concentration portion. For example, when the sample 4 is a rectangular flat plate shown in FIG. 2 and includes the slit 4A shown in FIG. 3, the magnetic sensor 11A is preferably arranged immediately above the tip 4B of the slit 4A.

  Hereinafter, for convenience of explanation, the slit 4A is assumed to be perpendicular to the long side of the sample 4, and the xyz orthogonal coordinate system is fixed to the XY stage 13A. The origin O of the coordinate system is the base end of the slit 4A, the x-axis is the long side direction of the sample 4, the y-axis is the direction from the base end O toward the front end 4B along the longitudinal direction of the slit 4A, and the z-axis Is a vertical direction, that is, a direction perpendicular to the upper surface of the XY stage 13A.

Step S2: The stress intensity factor calculation unit 21 causes the position control unit 15 to translate the XY stage 13A in a constant direction at a constant speed. That is, on the XY plane, the magnetic sensor 11A is moved from the stress concentration portion at a constant speed in the direction opposite to the above direction. For example, for the sample 4 shown in FIG. 2, the magnetic sensor 11A is moved from the tip 4B of the slit 4A in the longitudinal direction of the linear region 4C shown in FIG. 3, that is, in the y-axis direction. While translating the XY stage 13A, the stress intensity factor calculation unit 21 causes the magnetic flux density measurement unit 1 to continuously measure the magnetic flux density. At that time, the gauss meter 12 continuously outputs a series of measurement values to the stress intensity factor calculation unit 21 as an analog signal. The stress intensity factor calculation unit 21 stores the analog signal together with the elapsed time from the time when the XY stage 13A starts moving. The stress intensity factor calculation unit 21 preferably passes a low-pass filter or a digital multimeter before storing the analog signal, and removes a frequency component equal to or higher than a predetermined threshold as noise from the analog signal.

Step S3: When the XY stage 13A is translated by a predetermined distance or time, that is, when the magnetic sensor 11A is moved from the stress concentration portion to a predetermined distance in the XY plane, the stress intensity factor calculation unit 21 The control unit 15 stops the XY stage 13A, and the magnetic flux density measurement unit 1 ends the measurement. In the example shown in FIG. 3, the timing corresponds to the time when the magnetic sensor 11A has finished moving the entire linear region 4C from the tip 4B of the slit 4A.

Step S4: The stress intensity factor calculation unit 21 measures the magnetic flux density at a predetermined point outside the stress concentration part, and based on the difference between the obtained measured value and the magnetic flux density at the same point in the initial state, The bias component is removed from the magnetic flux density distribution stored in steps S2 to S3. Further, the stress intensity factor calculation unit 21 compares the waveform of the analog signal from which the bias component has been removed with the waveform in the initial state stored in the database 22, and calculates the amount of change at the same point or the level within the same region. Find the maximum amount of change.
Step S5: The stress intensity factor calculation unit 21 compares the amount of change obtained in step S4 with a predetermined threshold, preferably the resolution of the gauss meter 12, more preferably 0.01 mT. If the amount of change is less than the threshold value, the process proceeds to step S6; otherwise, the process branches to step S7.
Step S6: The stress intensity factor calculation unit 21 determines that the fatigue of the stress concentration part has not reached a level at which the stress intensity factor can be calculated.

Step S7: The stress intensity factor calculation unit 21 determines that the fatigue of the stress concentration part has reached a level at which the stress intensity factor can be calculated. Further, the stress intensity factor calculation unit 21 regards the waveform of the analog signal from which the bias component has been removed in step S4 as a downwardly convex curve shape, and the level of the stored analog signal decreases from the start of movement of the XY stage 13A. Elapsed time until reaching the peak of the convex waveform is obtained. Further, from the elapsed time and the moving speed of the XY stage 13A, the distance that the magnetic sensor 11A has moved in the horizontal direction during the elapsed time is obtained. In the example shown in FIG. 3, the distance is the horizontal distance between the apex of the downward convex curve shape indicated by the magnetic flux density distribution along the y-axis direction from the tip 4B of the slit 4A and the tip 4B. Equal to the distance.

Step S8: The stress intensity factor calculation unit 21 searches the database 22 for the stress intensity factor width corresponding to the distance obtained in step S7, or calculates it according to the linear expression stored in the database 22. The stress intensity factor width thus obtained is determined as the stress intensity factor width of the stress concentration portion of the sample 4. The fatigue evaluation unit 2 preferably evaluates the growth rate of cracks generated in the stress concentration portion of the sample 4 based on the obtained stress intensity factor width, and further predicts the life of the sample 4.

  As is clear from the above embodiment, the following three facts shown in FIGS. 4, 5, and 6 are important in the present invention. (1) If fatigue progresses around the stress concentration part of the sample, the magnetic flux density distribution on the surface protrudes downward from the stress concentration part toward the direction of crack growth regardless of the type and initial state of the sample. Showing. (2) The distance from the stress concentration portion to the apex of the curved shape has a linear relationship with the stress intensity factor width of the stress concentration portion. (3) Until the fatigue proceeds to some extent from the initial state in the stress concentration part of the sample, the magnetic flux density distribution on the surface from the stress concentration part along the crack propagation direction hardly changes from the irregular shape in the initial state. When the fatigue reaches a certain level, the magnetic flux density distribution suddenly changes from the initial shape to the convex curve. These three facts have been verified by the following experiments.

First, a sample 4 shown in FIG. 2 was used as a test piece and made of high carbon chromium bearing steel SUJ2. The SUJ2 exhibits ferromagnetism and the Vickers hardness is 156 ± 5 kgf / mm ² . The test piece 4 is a rectangular flat plate, the width W is 25 mm, the length L is 125 mm, and the thickness is 8 mm. Further, a slit 4A was created as a pre-crack at the center of one long side of the test piece 4 by wire processing. The length a of the slit 4A is 3.00 mm, and the curvature radius r of the inner surface of the tip portion 4B is 90 μm.

  Next, a four-point bending fatigue test was performed on the test piece 4. That is, two points 4D, 4E, 4F, and 4G that are symmetrical with respect to the center of each long side of the test piece 4 shown in FIG. 2 are used as fulcrums, and the test piece 4 is repeatedly repeated in the short side direction with respect to each fulcrum. A load P was applied. Here, the distance d between the two fulcrums 4D and 4E is 50 mm on one long side, and the distance S between the two fulcrums 4F and 4G is 100 mm on the other long side. The repeated load P was periodically changed in a sinusoidal shape at a stress ratio of 0.1 and a frequency of 20 Hz. However, the maximum value of the load P is constant.

Each time the number of repetitions N of the load P reached a predetermined number, the tip 4B of the slit 4A was observed with a microscope, and the length of the crack generated there was measured. FIG. 8 shows a photograph of the slit 4A observed at that time. (a) shows the slit 4A when the repetition number N is 4.47 × 10 ⁵ , and (b) shows the slit 4A when the repetition number N is 7.08 × 10 ⁵ . As shown in FIG. 8, as the number of repetitions N increases, the crack 4H propagates from the tip 4B of the slit 4A in the extending direction of the slit 4A.

  Further, the test piece 4 is installed on the XY stage 13A of the magnetic flux density measuring unit 1, and the surface magnetic flux density distribution is shown in the region 4C linearly extending in the y-axis direction from the tip 4B shown in FIG. Was measured in the same manner. Here, the tip of the linear region 4C was a point 12.5 mm away from the base end 0 of the slit 4A in the y-axis direction. In addition, the bias component included in the measured magnetic flux density distribution is the magnetic flux density measured at a point (x, y) = (− 9.0 mm, 12.5 mm) sufficiently away from the linear region 4C and the initial state. It was calculated from the difference between the values. FIG. 4 shows a result obtained by removing the bias component and the high-frequency component of spatial variation from the magnetic flux density distribution measured in the linear region 4C. In FIG. 4, in addition to the repetition number N when the magnetic flux density distributions indicated by the solid line, the broken line, and the alternate long and short dash line are obtained, the total extension a of the slit 4A and the crack 4H is shown.

The stress intensity factor K _I of the distal end portion 4B of the slits 4A for the test piece 4, the width W of the test piece 4, the thickness t, the total extension a of the slits 4A and its previous crack 4H, fulcrum on each long side It can be expressed by the following equation (1) as a function of the distances d and S and the load P.

Here, the coefficient F _IP is a constant determined by the experimental method, and is about 1.02 in this experiment. Further, equation (1) is an approximate equation when the ratio d / W of the fulcrum interval d to the width W of the test piece 4 is sufficiently large.

The magnetic flux density distribution shown in FIG. 4 indicates that the stress intensity factor width ΔK _{N = 0} of the tip 4B of the slit 4A in the initial state is 17.9 MPa when the number of repetitions N of the four-point bending fatigue test is zero. Measured when √m. In addition, when the stress intensity factor width ΔK _{N = 0} in the initial state was set to 16.0 MPa√m, the same magnetic flux density distribution as that in FIG. 4 was obtained. In either case, the magnetic flux density distribution showed a downward convex curve before the crack 4H progressed 0.5 mm from the tip 4B of the slit 4A.

After the above magnetic flux density distribution which is as shown a convex curved shape on the lower, the distance Y _b between the tip portion 4B of the vertex and the slits 4A of the curve shape, shown in broken lines and one-dot chain line in FIG. 4 As shown, it increased with the number of repetitions N of the load P. Figure 9 shows the total length a between the distances Y _b and the slits 4A and cracks 4H to increase with the repetition number N of the load P. FIG. 9 also shows the size R in the y-axis direction of the plastic region generated around the tip 4B of the slit 4A. Here, the size R of the plastic region in the y-axis direction is defined by the distance from the base end O of the slit 4A to the boundary of the plastic region farthest in the y-axis direction, and is obtained analytically. From Figure 9, it was found that the above distance Y _b is always greater than any size R in the y-axis direction of the total length a and the plastic region of the slits 4A and crack 4H.

Calculated stress intensity factor range ΔK using equation (1) from the total length a of the slit 4A and crack 4H shown in Figure 9, the plot in association with the distance Y _b shown them in Figure 9 To do. The graph thus obtained is shown in FIG. As shown in FIG. 5, even if the stress intensity factor width ΔK _{N = 0} in the initial state is different, substantially the same linear relationship is recognized between the distance Y _b and the stress intensity factor width ΔK. It was. Accordingly, the stress intensity factor width ΔK can be determined from the distance Y _b as in the above embodiment of the present invention.

The magnetic flux density distribution indicated by the solid line in FIG. 6 is measured from the linear region 4C when the number of repetitions N is 0, that is, when the initial state is maintained. On the other hand, the magnetic flux density distribution indicated by a broken line is measured from a linear region 4C when the repetition number N reaches 1.58 × 10 ⁴ . In FIG. 6A, the stress intensity factor width ΔK _{N = 0} of the tip 4B of the slit 4A in the initial state is 16.0 MPa√m, and in FIG. 6B, it is 17.9 MPa√m. In either case, when the number of repetitions N is about 1.58 × 10 ⁴ , the crack 4H has not yet occurred, so the total extension a of the slit 4A and the crack 4H is equal to the length of the slit 4A of 3.00 mm. Until the repetition number N reached about 1.58 × 10 ⁴ , the magnetic flux density distribution changed from the curve shape in the initial state to a resolution of 0.01 mT or less of the Gauss meter 12. However, when the number of repetitions N reached 1.58 × 10 ⁴ , the magnetic flux density distribution changed greatly from the initial curve shape. From the above, according to the amount of change between the measured curve shape of the magnetic flux density distribution and the initial curve shape as in the above embodiment, the stress concentration is reduced before the stress concentration portion is cracked. It can be detected that the fatigue of the part has reached the extent that the stress intensity factor can be calculated.

  In the above embodiment, the magnetic flux density on the surface is measured while changing the position of the magnetic sensor 11A relative to the surface of the sample 4. In addition, a plurality of magnetic sensors may be arranged at a predetermined interval at a predetermined distance from the surface of the sample or on the surface. For example, for the sample 4 shown in FIG. 2, the probe 41 is preferably placed at a predetermined distance in the y-axis direction from the tip 4B of the slit 4A, as shown in FIG. In the probe 41, a plurality of magnetic sensors 41A are preferably arranged at equal intervals along the y-axis direction. Each magnetic sensor 41A is preferably a Hall element similar to the magnetic sensor 11A shown in FIG. The gauss meter 12 individually drives each magnetic sensor 41A and individually receives an output signal from each. The gauss meter 12 further determines the magnetic flux density penetrating each magnetic sensor 41A in the vertical direction from each output signal. The stress intensity factor calculation unit 21 detects from a plurality of magnetic flux densities obtained by the gauss meter 12 that the magnetic flux density distribution along the y-axis direction has changed beyond the predetermined threshold from the curve shape in the initial state. To do. Thereby, fatigue at the tip 4B of the slit 4A can be detected before the crack 4H actually occurs. The stress intensity factor calculation unit 21 further detects that the magnetic flux density distribution along the y-axis direction has changed significantly from the curve shape in the initial state to show a downwardly convex curve shape. Subsequently, the stress intensity factor calculating unit 21 determines the stress intensity factor width from the distance between the apex of the curved shape and the tip 4B of the slit 4A. Here, as shown in FIG. 9, the apex of the curved shape always precedes the front end of the crack 4H that extends in the y-axis direction from the front end portion 4B of the slit 4A. Therefore, in the method using the probe 41 shown in FIG. 10, the strain at the surface of the sample 4 accompanying the progress of the crack 4H is detected through a strain gauge attached to the surface of the sample 4 at an earlier stage. The progress of crack 4H can be predicted.

The block diagram of the nondestructive inspection apparatus by embodiment of this invention The top view which shows an example of the sample which is a measuring object of the nondestructive inspection apparatus shown by FIG. 2 is an enlarged plan view of the vicinity of the slit of the sample shown in FIG. The graph which shows an example of magnetic flux density distribution along the linear area | region 4C shown by FIG. When the magnetic flux density distribution shown in FIG. 4 shows a downward convex curve shape, the linear relationship between the distance from the stress concentration portion to the apex of the curve shape and the stress intensity factor width of the stress concentration portion is Graph showing A graph showing the magnetic flux density distribution along the linear region 4C before the tip 4B of the slit 4A shown in FIG. 3 is cracked. Flowchart of the measuring method by the nondestructive inspection apparatus shown in FIG. Photomicrograph of the vicinity of slit 4A, taken when a four-point bending fatigue test was performed on the sample shown in FIG. The distance Y _b from the tip 4B of the slit 4A shown in FIG. 3 to the apex of the downwardly convex curved shape shown in FIG. 4, and the y-axis direction of the plastic region generated at the tip 4B of the slit 4A A graph in which the size R and the total length a of the slit 4A and the crack 4H are plotted against the number of repeated loads in a four-point bending fatigue test. Schematic plan view of a probe according to another embodiment of the present invention

Explanation of symbols

1 Magnetic flux density measurement unit
11 Probe
11A magnetic sensor
12 Gauss meter
13A XY stage
13B Z stage
13C prop
14A horizontal laser displacement meter
14B Vertical laser displacement meter
15 Position controller
2 Fatigue evaluation section
21 Stress intensity factor calculator
22 Database
3 PC
4 samples
LA, LB laser light

Claims (8)

A magnetic flux density measuring unit for measuring the magnetic flux density of the surface of a sample made of a material exhibiting ferromagnetism at least in a region where stress is concentrated, and
The magnetic flux density distribution on the surface is measured by the magnetic flux density measurement unit in a predetermined region including the stress concentration part of the sample, and the magnetic flux density distribution is convex downward or upward along a certain direction from the stress concentration part. A stress intensity factor calculation unit that detects a convex curve shape and determines a stress intensity factor of the stress concentration part based on a distance from the stress concentration part to a vertex of the curve shape,
Non-destructive inspection device. The nondestructive inspection apparatus further includes a database holding data indicating a linear relationship between the distance and a stress intensity factor of the stress concentration portion,
The stress intensity factor calculation unit determines the stress intensity factor of the stress concentration part from the distance using the data,
The nondestructive inspection apparatus according to claim 1. A magnetic flux density measuring unit for measuring the magnetic flux density of the surface of a sample made of a material exhibiting ferromagnetism at least in a region where stress is concentrated, and
When the stress concentration portion of the sample is in an initial state where fatigue has not yet occurred, the magnetic flux density at at least one point on the surface of the stress concentration portion measured by the magnetic flux density measurement portion, or the stress concentration portion from the stress concentration portion Including a database holding data indicating a magnetic flux density distribution in a region extending in a predetermined direction along the surface of the sample, and the magnetic flux density at the one point newly measured by the magnetic flux density measurement unit or the region The amount of change between the magnetic flux density distribution and the magnetic flux density stored in the database or the magnetic flux density distribution is compared with a predetermined threshold value. If the amount of change is less than the threshold value, fatigue of the stress concentration portion is predetermined. A fatigue evaluation unit that determines that the degree has not reached the degree, and determines that fatigue of the stress concentration portion has progressed to the degree if the amount of change exceeds the threshold;
Non-destructive inspection device. The magnetic flux density measurement unit
A probe including a magnetic sensor,
A stage for displacably supporting the sample and the probe;
A gauss meter that drives the magnetic sensor and converts the output signal of the magnetic sensor into a measured value of magnetic flux density; and
A position controller that controls the position of the magnetic sensor relative to the sample by displacing the stage;
The nondestructive inspection apparatus of Claim 1 or Claim 3 containing this. The magnetic flux density measurement unit
A plurality of magnetic sensors arranged on the surface of the sample or at a predetermined height from the surface and arranged in a predetermined direction along the surface from the stress concentration portion; and
A gauss meter that drives the plurality of magnetic sensors and converts each output signal of the plurality of magnetic sensors into a measured value of magnetic flux density;
The nondestructive inspection apparatus of Claim 1 or Claim 3 containing this. Measuring a magnetic flux density distribution on a surface in a predetermined region including the stress concentration portion of a sample made of a material exhibiting ferromagnetism in a region where stress is concentrated at least;
Detecting that the magnetic flux density distribution has a downward convex or upward convex curve shape along a certain direction from the stress concentration portion;
When the magnetic flux density distribution indicates the curved shape, obtaining a distance from the stress concentration portion to the apex of the curved shape; and
Determining a stress intensity factor of the stress concentration portion based on the distance;
Non-destructive inspection method.   Data indicating a linear relationship between the distance and the stress intensity factor of the stress concentration portion is stored in a database in advance, and the data is read from the database and used in the step of determining the stress intensity factor of the stress concentration portion. The nondestructive inspection method according to claim 6. For a sample made of a material exhibiting ferromagnetism at least in a region where stress is concentrated, when the stress concentration portion is in an initial state where fatigue has not yet occurred, the magnetic flux density at at least one point on the surface of the stress concentration portion, or Measuring a magnetic flux density distribution in a region extending in a predetermined direction along the surface of the sample from the stress concentration portion, and storing the measurement result in a database;
Newly measuring the magnetic flux density at the one point or the magnetic flux density distribution in the region; and
The amount of change between the newly measured magnetic flux density at the one point or the magnetic flux density distribution in the region and the magnetic flux density or magnetic flux density distribution stored in the database is compared with a predetermined threshold, and the amount of change is If it is less than the threshold value, it is determined that the fatigue of the stress concentration portion has not reached a predetermined level, and if the amount of change exceeds the threshold value, the fatigue of the stress concentration portion has progressed to the above level. The step of judging,
Non-destructive inspection method.