JP4882703B2 - Hardening pattern inspection method and inspection apparatus - Google Patents

Description

  The present invention relates to a quenching pattern inspection method and an inspection apparatus for inspecting a quenching pattern of a steel part that has been subjected to quenching such as induction hardening, such as an automobile part.

Conventionally, steel parts (quenched parts) that have been subjected to quenching such as induction quenching are often used for apparatus parts such as automobile parts. For such a hardened part, in order to assure its quality, a hardened pattern (hardened hardened layer) formed on the surface portion of the hardened part is inspected.
As for the inspection of the quenching pattern, for example, as disclosed in Patent Document 1, there is a method using an eddy current. The method disclosed in Patent Document 1 measures the depth (quenching depth) of a hardened hardened layer of a steel material in a nondestructive manner. The quenching depth of the steel material and the output voltage of the detection coil Correlation data is obtained in advance, the target steel material is magnetized by the excitation coil, the induction magnetic field induced by the eddy current generated thereby is detected by the detection coil, and the output voltage of this detection coil is the correlation data that becomes known Is to calculate the quenching depth.

  In addition, for the inspection of the quenching pattern, the quality of the quenched part is assured by detecting the quenching pattern breakage (a portion where the quenching hardened layer is interrupted or locally extremely shallow). The quenching pattern of the quenching parts is currently guaranteed by a torsion test by sampling in an assembly line such as an automobile. In other words, an assembly in which the quenched parts are used and assembled to the assembly state (assembled state) is periodically extracted on the assembly line, and the assembling pattern is confirmed from the guarantee of mechanical strength by a torsion test on the assembly. Inspection is performed.

Specifically, as shown in FIG. 12, as an assembly configured by using a hardened part, for example, there is a CVJ assembly 101 configured by incorporating a constant velocity joint (CVJ: Constant Velocity Joint). . The constant velocity joint is a joint portion to which rotation is transmitted at a constant velocity so that no difference in velocity occurs between the connected shafts at any joint angle.
In the CVJ assembly 101 according to this example, joint portions 103 and 104 that are constant velocity joints are formed on both ends of the drive shaft 102, and are components that constitute the joint portions 103 and 104. The shaft members 105 and 106 connected to both end sides exist as hardened parts.

Then, when inspecting the quenching pattern of the shaft members 105 and 106 that are the quenching parts, a torsion is caused by applying a predetermined torque (see arrows T1 and T2) to the CVJ assembly 101 having a shaft shape as a whole. A test is conducted. In other words, when the quenching pattern is cut off in the shaft members 105, 106, etc. constituting the CVJ assembly 101, the portion where the quenching pattern is cut becomes a portion having a lower strength than the other portions, so that torque is applied. Will break. The occurrence of the quenching pattern is detected by the occurrence of such a broken portion.
As described above, detection of quenching pattern breakage in induction hardening or the like is a very important quality problem, but in the past, it could only be guaranteed by a torsion test by extracting an assembly including a quenched part. Is the current situation.

However, when a quenching pattern is inspected by a torsion test by extracting an assembly including a quenched part, there are the following problems.
That is, first, the torsion test by sampling as described above applies a predetermined torque to the object to be inspected and detects the occurrence of a fractured portion thereof, and therefore it takes a considerable time until the result is obtained. .
In addition, since the torsion test is a destructive test, the loss cost of the parts to be inspected is higher than that of the non-destructive test.
Further, since the inspection target is an assembly including a hardened part, man-hours to the assembly state are required, and the assembling cost is wasted. Similarly, since the inspection target is an assembly, the entire assembly to be inspected, including parts other than the part having the quenching pattern cut, is wasted. That is, since it is a torsion test after assembling to the assembly state, the loss cost is high with respect to the assembling cost and the cost of parts other than the hardened parts.
Furthermore, since the inspection is only by sampling and not 100% inspection, it cannot be guaranteed in the assembly line or the like.

On the other hand, the method disclosed in Patent Document 1 described above is a nondestructive method for measuring the quenching depth of a steel material. However, the quenching evaluation is performed in addition to the quenching depth. Since the hardness (hardness of the hardened part) is also an object, efficient inspection suitable for the actual quenching state can be performed by adding the quenching surface hardness to the evaluation object.
Moreover, about the method currently disclosed by patent document 1, when it measures by an actual assembly line etc. and it is going to use the measurement result for the quality determination of hardened parts, it is the temperature about the components which are the measurement object. There is room to consider changes and shape errors.
JP 2002-14081 A

  The present invention has been made in view of the above-described problems of the prior art, and the problem to be solved is a quenching pattern of a steel part subjected to induction hardening or the like. Quenching pattern inspection method and inspection that enable in-line inspection by non-destructive inspection of cuts, improve quenching quality, and reduce loss costs and time during quenching pattern inspection To provide an apparatus.

  The problem to be solved by the present invention is as described above. Next, means for solving the problem will be described.

That is, in claim 1, an excitation coil for applying a predetermined AC excitation signal to a component to be inspected, and a detection signal for detecting an eddy current from the component to be inspected to which the AC excitation signal is applied. A quenching pattern inspection method for determining the quality of a quenching pattern for a component to be inspected using an eddy current sensor having a detection coil, and showing a value resulting from a phase difference of the detection signal with respect to the AC excitation signal A number of non-defective products measured using the eddy current sensor using a coordinate plane defined by a first coordinate axis and a second coordinate axis perpendicular to the first coordinate axis and indicating the magnitude of the detection signal Based on the distribution of points on the coordinate plane determined from the value resulting from the phase difference and the magnitude of the detection signal, an allowable error region in the coordinate plane is preliminarily determined. Constant, and the value of the distance along the shape of the boundary line with respect to the boundary line that divides the allowable error area on the coordinate plane is defined as separation values were measured using the eddy current sensor, having a quenching pattern breakage The phase difference for the inspection site is set so that the separation value of the point on the coordinate plane determined from the value due to the phase difference for the defective product and the magnitude of the detection signal is relatively large. Set the frequency of the AC excitation signal used for measurement of the resulting value and the magnitude of the detection signal, use the eddy current sensor, use the value resulting from the phase difference for the inspection site of the inspection target component, and the Whether the magnitude of the detection signal is measured and a point on the coordinate plane determined from the measured value of the phase difference and the magnitude of the detection signal is within the allowable error area. More, it is to determine the quality of the quenching pattern for the component being tested.
This makes it possible to inspect all steel parts that have been hardened by induction hardening, etc., in-line by non-destructive inspection of the quenching pattern cut, improving the quenching quality and improving the quenching quality. Loss cost and time can be reduced when inspecting the incoming pattern.
In addition, the accuracy of the determination of the quality of the quenching pattern can be improved, and for the eddy current measurement in the quenching pattern inspection, the optimum condition for the frequency of the AC excitation signal, which is one of the measurement conditions, Based on the index of separation value, it can be derived regardless of the knowledge and skill level of the measurer's eddy current measurement.

According to a second aspect of the present invention, a point on the coordinate plane determined from the value resulting from the phase difference and the magnitude of the detection signal for a defective product having a quenching pattern breakage measured using the eddy current sensor is used. The position of the eddy current sensor with respect to the examination site is set so that the separation value is relatively large .
This makes it possible to inspect all steel parts that have been hardened by induction hardening, etc., in-line by non-destructive inspection of the quenching pattern cut, improving the quenching quality and improving the quenching quality. Loss cost and time can be reduced when inspecting the incoming pattern.
In addition, it is possible to improve the accuracy of the determination of the quality of the quenching pattern, and for the eddy current measurement in the quenching pattern inspection, the optimum condition for the coil position, which is one of the measurement conditions, is called a separation value. Based on the index, it can be derived regardless of the knowledge or skill level of the measurer's eddy current measurement.

According to a third aspect of the present invention, the shape of the permissible error region is determined so that the direction of the first principal component, which is the component having the highest contribution ratio to the distribution, is the major axis direction, and the second principal component is orthogonal to the first principal component. The direction is the minor axis direction, and the intersection of the major axis and the minor axis is determined from a value indicating the center of the distribution, and the distribution is averaged at the intersection point with the major axis direction being the direction of spread. An ellipse having a predetermined spread based on the standard deviation in the case of a normal distribution is used.
As a result, when determining the quality of the quenching pattern, the variation in eddy current measurement values (measurement error) due to the effects of temperature conditions, shape errors, etc. of the parts to be inspected is allowed within a range corresponding to the tendency of each variation. Therefore, it is possible to perform accurate determination (equal evaluation) according to the properties of the inspection target component.
Therefore, it is possible to improve the quenching quality, and it is possible to reduce over-detection (determining what is not actually a defective product as a defective product) when determining the quality of the quenching pattern, thereby improving productivity. can do.

The detection coil for detecting a detection signal due to an eddy current from an excitation coil for applying a predetermined AC excitation signal to the inspection target component and the inspection target component to which the AC excitation signal is applied. An eddy current sensor, a value caused by a phase difference of the detection signal with respect to the AC excitation signal, a measuring means for measuring a value of the magnitude of the detection signal, and a first value indicating the value caused by the phase difference Using the coordinate plane defined by the coordinate axis and the second coordinate axis that is orthogonal to the first coordinate axis and that indicates the value of the detection signal, the position of a number of non-defective products measured by the measuring means is measured. Based on the distribution of the points on the coordinate plane determined from the value resulting from the phase difference and the magnitude of the detection signal, an allowable error area in the coordinate plane is set in advance, and the measurement means Whether the measured point on the coordinate plane determined from the value resulting from the phase difference and the value of the magnitude of the detection signal for the inspection site of the inspection target component is within the allowable error region, A quenching pattern inspection apparatus comprising: a determination unit that determines whether or not the quenching pattern of the inspection target part is good, and conforms to a shape of the boundary line with respect to a boundary line that defines the allowable error region on the coordinate plane The coordinate plane determined from the value resulting from the phase difference and the magnitude of the detection signal for a defective product having a quenching pattern breakage, which is defined as a separation value as a separation value and measured using the eddy current sensor The AC excitation signal used for measuring the value resulting from the phase difference and the magnitude of the detection signal for the examination site so that the separation value of the upper point is relatively large. It is used to set the frequency of.
This makes it possible to inspect all steel parts that have been hardened by induction hardening, etc., in-line by non-destructive inspection of the quenching pattern cut, improving the quenching quality and improving the quenching quality. Loss cost and time can be reduced when inspecting the incoming pattern.
In addition, the accuracy of the determination of the quality of the quenching pattern can be improved, and for the eddy current measurement in the quenching pattern inspection, the optimum condition for the frequency of the AC excitation signal, which is one of the measurement conditions, Based on the index of separation value, it can be derived regardless of the knowledge and skill level of the measurer's eddy current measurement.

In claim 5, the point on the coordinate plane determined from the value resulting from the phase difference and the magnitude of the detection signal for a defective product having a quenching pattern breakage measured using the eddy current sensor. The position of the eddy current sensor with respect to the examination site is set so that the separation value is relatively large .
This makes it possible to inspect all steel parts that have been hardened by induction hardening, etc., in-line by non-destructive inspection of the quenching pattern cut, improving the quenching quality and improving the quenching quality. Loss cost and time can be reduced when inspecting the incoming pattern.
In addition, it is possible to improve the accuracy of the determination of the quality of the quenching pattern, and for the eddy current measurement in the quenching pattern inspection, the optimum condition for the coil position, which is one of the measurement conditions, is called a separation value. Based on the index, it can be derived regardless of the knowledge or skill level of the measurer's eddy current measurement.

7. The determination unit according to claim 6, wherein the shape of the tolerance region is orthogonal to the direction of the first principal component, which is the component having the highest contribution to the distribution, and to the first principal component. The direction of the second principal component is the direction of the minor axis, and the intersection of the major axis and the minor axis is determined from a value indicating the center of the distribution, and the distribution is determined at the intersection with the direction of the major axis as the direction of spread. An ellipse having a predetermined spread based on the standard deviation in the case of a normal distribution with an average value.
As a result, when determining the quality of the quenching pattern, the variation in eddy current measurement values (measurement error) due to the effects of temperature conditions, shape errors, etc. of the parts to be inspected is allowed within a range corresponding to the tendency of each variation. Therefore, it is possible to perform accurate determination (equal evaluation) according to the properties of the inspection target component.
Therefore, it is possible to improve the quenching quality, and it is possible to reduce over-detection (determining what is not actually a defective product as a defective product) when determining the quality of the quenching pattern, thereby improving productivity. can do.

As effects of the present invention, the following effects can be obtained.
That is, according to the present invention, it is possible to inspect 100 parts in-line by non-destructive inspection of the quenching pattern cut for steel parts subjected to induction hardening or the like, and improve the quenching quality. In addition, the loss cost and time can be reduced when inspecting the quenching pattern.

The present invention, for example, inspects the quenching pattern (quenched hardened layer) of the steel material (quenched part) that has been subjected to quenching such as induction hardening, such as automobile parts, using eddy current. In addition, for hardened parts that are inspection target parts, good quality products that do not have a hardened pattern cut (parts where the hardened hardened layer breaks or becomes extremely shallow locally) and hardened parts that have a hardened pattern cut. This is for discriminating defective products including NG patterns (NG products) and guaranteeing the quality of hardened parts.
First, prior to the description of the embodiment of the invention, the principle of eddy current measurement related to the present invention will be described.

FIG. 1 shows the relationship between the layer state, hardness, and magnetic permeability in the depth (distance from the surface) direction of a quenched member that is a hardened steel material (such as S45C).
As shown in FIG. 1, in the quenched member, as a schematic structure of the hardened member, from the surface side, a hardened layer 1 that is a portion subjected to quenching and a mother layer 2 that is a portion of the base material are , Formed through the boundary layer 3. Referring to the hardness change curve 4, the cured layer 1 and the mother layer 2 have different hardnesses, and the hardness of the cured layer 1 is larger than that of the mother layer 2. In the boundary layer 3, the hardness gradually decreases from the hardened layer 1 to the mother layer 2. As a specific example of the hardness, Vickers hardness (Hv) is shown, and the hardened layer 1 has a hardness of about Hv = 600 to 700, and the mother layer 2 has a hardness of about Hv = 300.

On the other hand, referring to the magnetic permeability change curve 5, the change in the magnetic permeability with respect to the distance from the surface of the quenched member is approximately inversely proportional to the change in the hardness with respect to the distance from the surface of the quenched member. That is, the magnetic permeability of the hardened layer 1 is larger than that of the mother layer 2 and the boundary layer 3 gradually increases from the hardened layer 1 to the mother layer 2 with respect to the magnetic permeability.
In the eddy current measurement related to the present invention, the relationship between the hardness and the permeability with respect to the distance from the surface in such a quenched member is utilized.

FIG. 2 shows a schematic diagram of an apparatus configuration for performing eddy current measurement related to the present invention.
As shown in FIG. 2, in eddy current measurement, an excitation coil 7 and a detection coil 8 that are adjacently arranged with a common central axis with respect to a measurement part 6 a of a workpiece (magnetic body) 6 that is a measurement target are predetermined. Set to position.
In such a configuration, when a current is supplied to the exciting coil 7, a magnetic field is generated around the exciting coil 7. Then, an eddy current is generated in the vicinity of the surface of the measurement site 6a of the workpiece 6 that is a magnetic body by electromagnetic induction (see arrow C1). As the eddy current is generated on the surface of the measurement site 6a, the magnetic flux penetrates the detection coil 8. And the induced voltage accompanying the eddy current generation in the surface of the measurement site | part 6a is measured by the detection coil 8. FIG.

Both ends (both terminals) of the exciting coil 7 are connected to the AC power source 9. The AC power supply 9 applies a predetermined AC excitation signal (excitation AC voltage signal) V <b> 1 to the excitation coil 7.
Both ends (both terminals) of the detection coil 8 are connected to the measuring device 10. The measuring device 10 has a magnitude of a detection signal (voltage signal for the induced voltage) V2 obtained from the detection coil 8 when the AC excitation signal V1 from the AC power supply 9 is applied to the excitation coil 7, and the detection signal V2. A phase difference (phase delay) Φ (see FIG. 3) with respect to the AC excitation signal V1 is detected. Here, the AC excitation signal V1 (waveform) is given to the measuring apparatus 10 in order to detect the phase difference Φ. FIG. 3 shows the relationship between the AC excitation signal V1 and the detection signal V2 in eddy current measurement.

The detection signal V2 detected by the detection coil 8 reflects the magnetic permeability of the measurement site 6a (workpiece 6). That is, when the magnetic permeability of the measurement region 6a increases, the magnetic flux accompanying the generation of eddy current as described above increases and the detection signal V2 increases, and conversely, when the magnetic permeability of the measurement region 6a decreases, the magnetic flux accompanying the generation of eddy current. Decreases and the detection signal V2 decreases.
In order to quantify (numerize) the detection signal V2 based on this eddy current, as shown in FIG. 3, the amplitude value Y, which is the magnitude of the detection signal V2, and the position of the detection signal V2 relative to the AC excitation signal V1. Focusing on the value X (= Y cos Φ), which is a value resulting from the phase difference Φ, the following knowledge has been obtained.

First, the amplitude value Y of the detection signal V2 may have a correlation with the hardened surface hardness (hardness of the hardened portion).
That is, as can be seen from a comparison between the hardness change curve 4 and the permeability change curve 5 in FIG. 1, there is a relationship that the permeability is high when the quenching surface hardness is low. When the magnetic permeability is high, the magnetic flux generated when the AC excitation signal V1 is applied to the excitation coil 7 increases, and the eddy current induced on the surface of the measurement site 6a also increases. Along with this, the amplitude value Y of the detection signal V2 detected by the detection coil 8 also increases. Therefore, conversely, from the amplitude value Y of the detection signal V2 detected by the detection coil 8, the magnetic flux penetrating the measurement site 6a where the eddy current is generated, that is, the magnetic permeability, is derived. Thereby, the quenching surface hardness can be understood from the relationship between the hardness change curve 4 and the permeability change curve 5 shown in FIG.

Next, the value X resulting from the phase difference Φ of the detection signal V2 with respect to the AC excitation signal V1 may have a correlation with the quenching depth (depth of the quench-hardened layer).
That is, when the quenching depth becomes deep, that is, the hardened layer 1 hardened in the quenching member increases, the low permeability range increases in the depth direction, and the AC excitation signal V1 is increased. The phase delay of the detection signal V2 will increase. Thereby, the depth of the quenching depth can be determined from the magnitude of the value caused by the phase difference Φ.

The embodiment of the present invention will be described based on the above principle of eddy current measurement.
As shown in FIG. 4, the quenching pattern inspection method according to the present embodiment includes an excitation coil 11 for applying a predetermined AC excitation signal V <b> 1 to a workpiece 50 as an inspection target component, and an AC excitation signal V <b> 1. Using the eddy current penetrating coil 13 which is an eddy current sensor having a detection coil 12 for detecting the detection signal V2 due to the eddy current from the applied work 50, the quality of the quenching pattern for the work 50 is judged. .

  In the present embodiment, the workpiece 50 is a steel material (S45C or the like) that has been subjected to induction hardening. As shown in FIG. 4, the workpiece 50 according to the present embodiment is a shaft-like member as a whole, and is a shaft portion 50b that is a shaft-shaped portion and a portion that is expanded in diameter relative to the shaft portion 50b. And a joint portion 50c to which other members are connected, and induction hardening is applied to a portion including substantially the entire shaft portion 50b.

The eddy current penetrating coil 13 has an amplitude value Y (hereinafter also simply referred to as “Y value”) of the detection signal V2 correlated with the hardened surface hardness and a phase difference Φ correlated with the hardened depth (FIG. 3). (Referred to below) is a sensor for measuring a value X (hereinafter also simply referred to as “X value”), and the excitation coil 11 and the detection coil 12 included in the sensor are combined in a state where the central axes are arranged in common. It is accommodated in a case 14 made of resin or the like.
In the present embodiment, the excitation coil 11 is disposed outside the detection coil 12. The case 14 has a rectangular thick plate-shaped outer shape and has a through hole 14a for allowing the work 50 to be inserted therethrough. That is, the excitation coil 11 and the detection coil 12 are accommodated in the case 14 with the coil hollow portions aligned with the through holes 14a.

  The eddy current penetrating coil 13 having such a configuration is movable in the axial direction of the work 50 (left and right direction in FIG. 4) in a state where the work 50 that is an axial member is inserted into the through hole 14a of the case 14. Provided. Then, the eddy current penetrating coil 13 is formed on the surface of the workpiece 50 in a state where the excitation coil 11 and the detection coil 12 are set so as to be in a predetermined position with respect to the inspection site 50a desired in the workpiece 50. The hardened pattern is inspected.

That is, when a predetermined AC excitation signal V1 is applied to the excitation coil 11, an eddy current is generated at the inspection site 50a of the workpiece 50, and a voltage signal regarding the induced voltage is detected by the detection coil 12 as the detection signal V2. Detected.
Then, the Y value (amplitude value Y) correlating with the quenching surface hardness for the inspection site 50a and the X value (value X = YcosΦ) correlating with the quenching depth for the inspection site 50a are measured.

In order to inspect the quenching pattern for such a workpiece 50, as shown in FIG. 5, the quenching pattern inspection apparatus according to this embodiment includes the eddy current penetrating coil 13 and the eddy current penetrating coil 13 that connects the cable 15. And an eddy current flaw detector 20 connected to each other.
The eddy current flaw detector 20 applies the AC excitation signal V1 to the excitation coil 11, measures the X and Y values based on the detection signal V2 detected by the detection coil 12, and measures the measured values of the X and Y values. The quality of the quenching pattern for the used workpiece 50 is determined.

That is, the eddy current flaw detector 20 includes an AC power source unit 21 (corresponding to the AC power source 9) for applying a predetermined AC excitation signal V1 to the excitation coil 11, and a measurement unit 22 for measuring the X value and the Y value (described above). Equivalent to the measuring device 10). That is, both terminals of the excitation coil 11 are connected to the AC power supply unit 21 via the cable 15, and both terminals of the detection coil 12 are also connected to the measurement unit 22 via the cable 15. The measurement unit 22 is given an AC excitation signal V1 (waveform) in order to detect the phase difference Φ.
Further, the eddy current flaw detector 20 includes a determination unit 23 that determines the quality of the quenching pattern for the workpiece 50 using the measurement values of the X value and the Y value by the measurement unit 22.
Thus, in this embodiment, the eddy current flaw detector 20 functions as a measurement unit that measures the X value and the Y value, and a determination unit that determines the quality of the quenching pattern for the workpiece 50.
Further, the eddy current flaw detector 20 is provided with a display unit 24 for displaying a measurement result by the measurement unit 22, a determination result by the determination unit 23, and the like.

Further, in the quenching pattern inspection apparatus according to the present embodiment, a measuring jig 30 for holding and fixing the workpiece 50 in a predetermined posture is provided.
The measuring jig 30 includes a base 31, and the mounting table 32 that forms a mounting surface 32 a on which the workpiece 50 is mounted on the base 31. The workpiece 50 is held and fixed in a state where it is placed in a predetermined posture on the placement surface 32a. In the present embodiment, the workpiece 50 that is a shaft-shaped member has an upright posture such that the axial direction is the vertical direction (vertical direction in FIG. 5) as the predetermined posture, and the workpiece 50 is lifted from above. It is held and fixed on the mounting table 32 by a fixing member 33 or the like that is fixed by pressing.

Further, the measurement jig 30 is provided with a movement support member 34 for supporting the eddy current penetrating coil 13 so as to be movable in a posture corresponding to the workpiece 50. The movable support member 34 is a base 31.
It is supported so as to be movable by a column 35 erected on the top.
In the present embodiment, the movement support member 34 is configured by a plate-like member having a through hole 34a, and the work 50 that is held and fixed in a predetermined posture on the mounting table 32 is inserted into the through hole 34a. The eddy current penetrating coil 13 is fixedly supported to the moving support member 34 through the case 14 in a posture corresponding to the workpiece 50, that is, with the workpiece 50 inserted through the through hole 14a. That is, in a state where the eddy current penetrating coil 13 is fixedly supported by the moving support member 34, the through holes 34a and 14a are substantially coaxial, and are held in a predetermined posture on the mounting table 32 by the through holes 34a and 14a. The work 50 to be fixed is penetrated.

The movement support member 34 is supported by the column 35 so as to be movable in the axial direction (vertical direction in FIG. 5) of the workpiece 50 held and fixed on the mounting table 32 in a predetermined posture. As a result, the vortex penetrating coil 13 is provided so as to be movable in the axial direction of the workpiece 50 in a state where the workpiece 50 is inserted through the through hole 14 a of the case 14.
In the measuring jig 30 having such a configuration, when measuring the eddy current of the workpiece 50, the workpiece 50 is held and fixed in a predetermined posture (upright posture) and the eddy current penetrating coil 13 is moved relative to the workpiece 50 (in the axial direction thereof). For the direction, the required predetermined accuracy is ensured.

Specific contents of the quenching pattern inspection method according to the present embodiment by the quenching pattern inspection apparatus having the above-described configuration will be described.
In the quenching pattern inspection method described below, as shown in FIG. 6, the X axis 40x which is the first coordinate axis indicating the value (X value) resulting from the phase difference Φ of the detection signal V2 with respect to the AC excitation signal V1. And an XY plane 40 that is a coordinate plane defined by a second coordinate axis (Y axis 40y) that is orthogonal to the X axis 40x and that indicates an amplitude value (Y value) that is a magnitude value of the detection signal V2. Use.
That is, the coordinate plane defined by the X axis 40x indicating the output value (output value X) for the X value and the Y axis 40y indicating the output value (output value Y) for the Y value is the XY plane 40. Therefore, the output value X and the output value Y are values output from the measurement unit 22 of the eddy current flaw detector 20, and the values are voltage values.

Then, on the XY plane 40 determined from the X axis 40x and the Y axis 40y, a tolerance zone 41, which is an allowable error region for determining the quality of the quenching pattern for the workpiece 50, is set in advance.
The tolerance zone 41 is based on the distribution of points on the XY plane 40 determined from the X value (output value X) and Y value (output value Y) of a number of non-defective products, measured using the eddy current penetrating coil 13. Set.

Here, the non-defective product and the defective product (hardened pattern cut product) of the workpiece 50 according to the present embodiment will be described with reference to FIG.
FIG. 7 shows a cross-sectional view in which the longitudinal direction (axial direction) of the workpiece 50, which is a shaft-shaped member subjected to induction hardening, is a shear direction, and FIG. FIG. 4B shows a cross-sectional view of the quenched pattern.
As shown in FIGS. 7A and 7B, the workpiece 50 according to the present embodiment subjected to induction hardening is induction-hardened at a portion including substantially the entire shaft portion 50b as described above. The In other words, in the work 50, a hardened layer 51 (see the light-colored portion) is formed on the base layer 52, which is a base material portion, on the entire surface of the shaft portion 50b. . Although not shown in the drawing, a boundary layer is formed between the hardened layer 51 and the mother layer 52.

In the non-defective workpiece 50 shown in FIG. 7 (a), the hardened layer 51 has a portion where the quenching depth gradually decreases at both ends in the axial direction. A quenching pattern with depth is formed.
On the other hand, in the workpiece 50 of the quenching pattern cut product shown in FIG. 7 (b), the hardened layer 51 has a quenching pattern cut 53 which is a portion that is interrupted or locally extremely shallow. .
That is, as shown in FIG. 7A, the work 50 in which the quenching pattern (cured layer 51) is uniformly formed in a desired portion without the quenching pattern cut 53 as described above is a non-defective product. A workpiece 50 other than a non-defective product, such as a quenching pattern cut product having a quenching pattern cut 53 as shown in FIG. Becomes defective.

When setting the tolerance zone 41 used for the determination of the non-defective product / defective product for the workpiece 50, a large number of eddy current measurement values (X value and Y value) for many good products are plotted on the XY plane 40. This point is used.
The setting of the tolerance zone 41 will be described below using an example of eddy current measurement results for the X value and the Y value shown in FIG.

FIG. 6 shows an example of eddy current measurement results of the output value X (mV) and the output value Y (mV) when the frequency of the AC excitation signal V1 is 120 Hz for the non-defective and defective workpieces 50.
In the eddy current measurement relating to the setting of the tolerance zone 41, the presence or absence of a quenching pattern breakage or the like in a cross-sectional view as shown in FIG. 7 for a workpiece to be measured (different from the workpiece 50 to be inspected for a quenching pattern) A non-defective product or a defective product is determined in advance by a cutting inspection that visually inspects the workpiece, and a workpiece that is known to be a non-defective product is measured.
That is, a plurality of (at least two) workpieces subjected to induction hardening under the same conditions are produced, and among these workpieces, a non-defective product / defective product is determined in advance by cutting inspection as described above. The X value and the Y value are measured for the remaining part of the workpiece, and plotted on the XY plane 40. Thereby, the point (data) on the XY plane 40 about many good products is acquired.

FIG. 6 shows a portion of the output value X in the range of −165 to −125 (mV) and the output value Y in the range of −3960 to −3820 (mV) on the XY plane 40. In the XY plane 40 in such a range, data (output value X and output value Y) for a large number (for example, about several hundreds) of non-defective products are acquired.
As shown in FIG. 6, the non-defective product data 42 that are the measurement points of the non-defective workpieces indicated by white circles are intensively distributed in a predetermined area on the XY plane 40.
Specifically, in the eddy current measurement according to this example, the non-defective product data 42 is in the range of about −145 to −130 (mV) for the output value X and in the range of about −3940 to −3880 (mV) for the output value Y. The result was distributed.

On the other hand, the following data was obtained for defective products.
That is, as shown in FIG. 6, among the defective products, the pattern cut product data 44 that is the measurement point for the workpiece of the quenching pattern cut product indicated by the black circle is distributed in the non-defective product data 42 on the XY plane 40. Located in a region that is deviated from the region. Here, as shown in FIG. 7B, the quenching pattern-cut product refers to a poor product having a clear quenching pattern cut and having a remarkable degree of failure with respect to the quenching pattern.
In addition, for defective products, a workpiece having a lower degree of defect with respect to the quenching pattern than that of the quenched pattern was used as a standard product, and standard product data 45, which is data about this, was obtained. In other words, the standard product is a work having a hardened pattern that is not uniform compared to a non-defective product, and has a portion in which the hardened layer 51 is locally shallow, etc., and the degree of failure of the hardened pattern is medium. The standard product data 45 is located in a region deviating from the region in which the non-defective product data 42 is distributed on the XY plane 40, but the degree of divergence is lower than the pattern cut product data 44.

As described above, in the XY plane 40, the degree of deviation from the region in which the non-defective product data 42 is distributed increases as the degree of defect in the quenching pattern increases with respect to the measurement points of the X value and the Y value for the workpiece. It turns out that it becomes high. That is, on the XY plane 40, the distinction between the non-defective product and the defective product is clearly shown by the position of the measurement point, and the non-defective product data 42 has a concentrated distribution with respect to a predetermined area. .
Therefore, such a predetermined area in which the non-defective product data 42 is distributed on the XY plane 40 is set in advance as the tolerance zone 41 and is used to determine whether the quenching pattern for the workpiece 50 is good or bad.

The tolerance zone 41 is set in consideration of variation in the distribution of the non-defective product data 42 on the XY plane 40. The variation in the distribution of the non-defective product data 42 has a tendency because properties such as conductivity, magnetic permeability, and shape error of the inspection portion 50a of the work 50 contribute. For this reason, the tolerance zone 41 is set in consideration of the tendency of variation in the distribution of the non-defective product data 42 so that almost all non-defective products in a range in which the degree of failure of the quenching pattern is acceptable are included.
That is, the tolerance zone 41 is a region that allows variations in measurement values (measurement errors) due to the conductivity and the like of the inspection site 50a of the workpiece 50.

In the present embodiment, as shown in FIG. 6, the tolerance zone 41 has an oval shape in consideration of the tendency of variation in the distribution of the non-defective product data 42. That is, the boundary line 43 between the tolerance zone 41 and other portions on the XY plane 40 has an elliptical shape.
As shown in FIG. 5, in the quenching pattern inspection apparatus according to the present embodiment, the XY plane 40 representing the tolerance zone 41 and the eddy current measurement values (output values X and X) on the XY plane 40. The points determined (plotted) from the output value Y) are displayed using the computer 25 connected to the eddy current flaw detector 20. In other words, the computer 25 can visually grasp the points determined from the eddy current measurement values of the tolerance zone 41 and the workpiece in the XY plane 40.

In this way, the tolerance zone 41 is set in advance on the XY plane 40, and if the point determined from the eddy current measurement value for the workpiece 50 related to the determination of the quality of the quenching pattern is within the tolerance zone 41, the tolerance zone 41 is regarded as a good product. If it is outside the zone 41, it is determined as a defective product.
That is, in the quenching pattern break inspection method according to the present embodiment, after setting the tolerance zone 41 in advance as described above, the eddy current penetrating coil 13 is used to determine the X value and the Y value for the inspection portion 50a of the workpiece 50. The workpiece 50 is quenched depending on whether or not a point on the XY plane 40 (hereinafter referred to as “measurement point”) determined from the measured X and Y values is within the tolerance zone 41. Judge the quality of the pattern.

In such determination using the tolerance zone 41, that is, whether or not the measurement point for the workpiece 50 is within the tolerance zone 41, the separation value is used as an index indicating the distance to the boundary line 43 that defines the tolerance zone 41. Is used.
The separation value is a value of a distance along the shape of the boundary line 43 with respect to the boundary line 43 that defines the tolerance zone 41 on the XY plane 40. Therefore, for the tolerance zone 41 having an elliptical shape as in the present embodiment, the measurement point having the same separation value is the distance in the major axis direction with respect to the boundary line 43 on the XY plane 40. On the other hand, the distance in the direction of the minor axis with respect to the boundary line 43 becomes shorter.
As for the separation value, the separation value = 1 is set when the measurement point of the workpiece 50 is located on the boundary line 43 of the tolerance zone 41.

The quenching pattern inspection method according to the present embodiment will be described with reference to the flowchart of the determination algorithm shown in FIG.
First, the tolerance zone 41 is set from the variation in the distribution of the non-defective product data 42 on the XY plane 40 (S100). That is, by measuring eddy currents, a large number of measurement points (good product data 42) are obtained for workpieces that are known to be good in advance, and the tolerance zone 41 is set in consideration of variations in the distribution of the good product data 42. To do.

  Next, a separation value on the XY plane 40 is calculated from the eddy current measurement values (X value and Y value) of the workpiece 50 that is the inspection target of the quenching pattern (S110). That is, eddy current measurement is performed on the workpiece 50 using the eddy current penetrating coil 13, measurement points are plotted on the XY plane 40 from the output value X and the output value Y, and the separation value of the measurement point with respect to the tolerance zone 41 is measured. Is calculated.

If the calculated separation value is 1 or less, an OK determination is made that the workpiece 50 is non-defective (S120). That is, when the separation value is 1 or less, the measurement point for the workpiece 50 exists in the tolerance zone 41 including the boundary line 43, so that the workpiece 50 is determined to be non-defective. To do.
On the other hand, when the calculated separation value is larger than 1, an NG determination is made that the workpiece 50 is defective (not good) (S130). That is, if the value of the separation value is larger than 1, the measurement point for the workpiece 50 exists outside the tolerance zone 41, and therefore the workpiece 50 is determined as a defective product.

Such a determination algorithm is executed in the determination unit 23 provided in the eddy current flaw detector 20.
That is, the eddy current flaw detector 20 uses the XY plane 40 defined by the X axis 40x indicating the X value and the Y axis 40y indicating the Y value, and distributes a large number of non-defective product data 42 measured by the measuring unit 22. The tolerance zone 41 in the XY plane 40 is set in advance, and the workpiece 50 depends on whether or not the measurement point for the inspection site 50a of the workpiece 50 measured by the measurement unit 22 is in the tolerance zone 41. It functions as a judging means for judging the quality of the quenching pattern.

  As described above, on the XY plane 40 representing the eddy current measurement value, the tolerance zone 41 is preset from a large number of non-defective product data 42, and the quality of the quenching pattern for the workpiece 50 is determined using the tolerance zone 41. As a result, it is possible to inspect all parts in steel by induction hardening and other in-line inspections by non-destructive inspection for quenching pattern breakage, improving the quenching quality and improving the quenching quality. Loss cost and time can be reduced when inspecting the incoming pattern.

That is, by using a preset tolerance zone 41 on the XY plane 40 representing the eddy current measurement value, the quenching pattern of the workpiece 50 from the measurement value (X value and Y value) by the eddy current measurement which is nondestructive measurement. Therefore, it is possible to perform an in-line 100% inspection (guaranteed for 100%) of the quenching pattern.
As a result, the torsion test by sampling, which is a destructive test that has been performed conventionally, can be abolished, and loss costs and man-hours associated with the abolition of the torsion test can be reduced. As a result, the quenching quality can be improved, and loss costs and time can be reduced when inspecting the quenching pattern.

Further, when continuously inspecting the quenching pattern for a plurality of workpieces 50, by constantly monitoring the eddy current measurement result, for example, cooling water clogging or power in an apparatus for quenching such as induction quenching. When quenching abnormality such as sudden fluctuation occurs, it can be detected in real time, and early countermeasures can be taken for quenching abnormality.
Also, traceability management can be performed by acquiring eddy current measurement values of all parts in an actual assembly line or the like. In relation to this, it is possible to investigate the correlation between the occurrence of quenching abnormality as described above and the occurrence of quenching pattern breakage in the workpiece 50, and to know the manufacturing requirements and management requirements for non-defective products. Become.

Next, a preferable setting method for the tolerance zone 41 will be described.
In the quenching pattern inspection method according to the present embodiment, the shape of the tolerance zone 41 is set such that the direction of the first principal component, which is the component having the highest contribution ratio to the distribution of the non-defective product data 42, is the major axis direction. The direction of the second principal component orthogonal to the principal component is the minor axis direction, the intersection of the major axis and the minor axis is determined from the value indicating the center of the distribution, and the distribution is expanded in the major axis direction. It is preferable that the direction is an ellipse having a predetermined spread based on a standard deviation in the case of a normal distribution having an average of the intersections.

FIG. 9 shows an image diagram of the tolerance zone 41 and the separation value.
As shown in FIG. 9, when setting the tolerance zone 41, first, the direction of the first principal component (see the arrow 49x), which is the component having the highest contribution rate to the distribution of a large number of non-defective product data 42, is obtained.
As described above, the distribution of the non-defective product data 42 on the XY plane 40 has a tendency depending on the properties of the workpiece 50. That is, the distribution of the non-defective product data 42 on the XY plane 40 varies not with random variation but with a tendency due to the properties of the workpiece 50.

In the XY plane 40 according to the present embodiment, the distribution variation due to the influence of the electrical conductivity (electrical conductivity) of the workpiece 50 corresponds to the component having the highest contribution rate to the distribution of many good product data 42. In other words, the change in the conductivity of the work 50 to be inspected appears most prominently as the movement distance of the measurement point on the XY plane 40. The direction of variation in the distribution of the non-defective product data 42 due to the influence of the conductivity is the direction of the first main component indicated by the arrow 49x in FIG.
The electrical conductivity is the reciprocal of the resistivity (electrical resistivity), and the resistivity changes under the influence of the temperature. Therefore, the first main component has a distribution variation due to the temperature of the workpiece 50. It can also be said.

When the direction of the first principal component is obtained, the direction of the second principal component (see arrow 49y) is determined as the direction orthogonal to the direction.
The second main component is a variation in distribution due to the influence (mainly the influence of the magnetic permeability) such as the magnetic permeability and shape error (variation between lots) of the workpiece 50. That is, a change in magnetic permeability and a shape error of the workpiece 50 to be inspected appear as a moving distance in the direction of the second principal component of the measurement point on the XY plane 40.

  The directions of the first principal component and the second principal component obtained in this way are taken as the major axis and minor axis directions of an ellipse (hereinafter referred to as “set ellipse”) that partitions the tolerance zone 41, respectively. Then, the intersection point O1 between the major axis and the minor axis is determined from a value indicating the center of the distribution of a large number of non-defective product data. As a value indicating the center of the distribution of the large number of non-defective data 42, for example, the following values are used. The “value” here refers to the eddy current measurement value (X value and Y value).

That is, with respect to the distribution of a large number of non-defective products data 42 on the XY plane 40, the least square method is used, and the first principal component direction (major axis direction) and the second principal component direction (minor axis direction) respectively. It is conceivable that a straight line (linear function) on the XY plane 40 is approximated and the value at the intersection of these straight lines is used as a value indicating the center of the distribution of a large number of non-defective product data 42.
In addition, regarding the distribution of a large number of non-defective products data 42 on the XY plane 40, a straight line in the direction of the first principal component (the direction of the major axis) passing through the average value in the direction of the second principal component (the direction of the minor axis). It is conceivable to use the value at the intersection of the first principal component direction and the straight line of the second principal component direction passing through the average value as the value indicating the center of the distribution of the non-defective product data 42.

In this way, after obtaining the major axis / minor axis direction and the intersection O1 thereof, the distribution of a large number of non-defective data 42 is a normal distribution in which the values at the intersection O1 are averaged with the major axis direction as the spreading direction. If the ellipse has a predetermined spread based on the standard deviation, the focal point of the ellipse and the distance from the focal point are determined, and the set ellipse is determined.
That is, as shown in FIG. 9, when the distribution of a large number of non-defective data 42 is the direction of the long axis indicated by the arrow 49x as the spreading direction, and the value at the intersection O1 of the long axis with the short axis is the average μ, In other words, the standard deviation σ (dispersion: σ <SUP> 2 </ SUP>) when the origin is the intersection O1 on the coordinate plane defined by the major and minor axes intersecting at the intersection O1 (when μ = 0) The distribution is N (μ, σ <SUP> 2 </ SUP>) (see the normal distribution curve 47).

In the present embodiment, when determining the set ellipse, a range of μ ± 3σ is used as the predetermined spread (limit value) in the normal distribution. That is, the set ellipse has a spread of ± 3σ around the intersection O1 in the major axis direction.
In this way, the length of the major axis of the set ellipse is determined by using the range of μ ± 3σ when the distribution of the non-defective product data 42 is a normal distribution as the spread in the major axis direction. Then, the focal point of the ellipse is obtained based on the length of the major axis and the variation in the direction of the minor axis of the distribution of the large number of non-defective product data 42. Thereby, the setting ellipse is determined (see the boundary line 43).
Further, when determining the set ellipse, a large number of non-defective product data 42 is almost entirely included based on the average value of the eddy current measurement values (X value and Y value) based on the distribution variation of the large non-defective product data 42. After determining the focal point to be a point on the long axis in advance, the set ellipse may be determined by determining the distance from the focal point using the above-described μ ± 3σ spread in the normal distribution.
As described above, when the tolerance zone 41 is set, an ellipse having a spread of μ ± 3σ in the normal distribution in the direction of the long axis is used, so that the non-defective product data 42 is theoretically about 99.7% in the direction of the long axis. Will be included.
The tolerance zone 41 (the shape of the boundary line 43) on the XY plane 40 is set by determining the set ellipse as described above.

Here, the relationship between the separation value and the spread of the ellipse when the set ellipse is determined with the distribution of a large number of non-defective products 42 as a normal distribution in the direction of the major axis will be described.
As described above, the measurement value on the boundary line 43 that is the set ellipse has the separation value = 1. Therefore, as shown in FIG. 9, on the boundary line 43 of the tolerance zone 41 which is an ellipse (3σ ellipse) having a spread of μ ± 3σ in the normal distribution described above, the separation value = 1.
On the other hand, the separation value is a value of the distance along the shape (elliptical shape) of the boundary line 43 with respect to the boundary line 43 that defines the tolerance zone 41 on the XY plane 40. On an ellipse (6σ ellipse) 46 that is similar to a certain 3σ ellipse and has a spread of μ ± 6σ in a normal distribution similar to that used when determining the set ellipse with the intersection O1 as the origin, The separation value = 2. Similarly, the separation value = 3 on an ellipse (9σ ellipse) 49 having a spread of μ ± 9σ in the normal distribution.
Therefore, in the XY plane 40 shown in FIG. 9, the pattern cut product data 44 is distributed at positions where the separation value is about 3.

As described above, FIG. 10 shows a quenching pattern inspection method in the case where the tolerance zone 41 is set by determining a set ellipse by using a normal distribution or the like based on variations in the distribution of a large number of non-defective product data 42. A description will be given according to the flowchart of the determination flow shown.
First, from the variation of the non-defective product data 42 on the XY plane 40, the first principal component having the highest contribution rate to the distribution of the large number of non-defective product data 42 is obtained (S200). Here, the variation in distribution due to the influence of the conductivity of the workpiece 50 is the first main component (see FIG. 9, arrow 49x).

  Next, a second principal component orthogonal to the first principal component is obtained (S210). The second main component is a variation in distribution mainly due to the influence of the magnetic permeability of the workpiece 50 (see FIG. 9, arrow 49y).

  Subsequently, using the dispersion of the non-defective product data 42 and the spread of μ ± 3σ in the normal distribution having the direction of the first principal component (the direction of the major axis) as the direction of the spread, An ellipse having the direction and the direction of the second principal component as the minor axis direction is set as the tolerance zone 41 (S220). That is, as described above, the set ellipse (the shape of the boundary line 43) is determined by using the first principal component, the second principal component, the non-defective data variation, and the μ ± 3σ spread in the normal distribution, and the tolerance zone. 41 is set.

  Then, as in the case of the flowchart shown in FIG. 8, the separation value on the XY plane 40 is calculated from the eddy current measurement values (X value and Y value) of the workpiece 50 that is the inspection target of the quenching pattern ( S230), if the calculated separation value is 1 or less, OK determination is made that the workpiece 50 is a non-defective product (S240). If the calculated separation value is greater than 1, the workpiece 50 is defective. An NG determination is made as being (not good) (S250).

Such a determination flow is executed in the determination unit 23 provided in the eddy current flaw detector 20.
That is, the eddy current flaw detector 20 makes the shape of the tolerance zone 41 orthogonal to the direction of the first principal component, which is the component having the highest contribution ratio to the distribution of a large number of non-defective data 42, and to the first principal component. The direction of the second principal component is the minor axis direction, and the intersection point O1 between the major axis and the minor axis is determined from the value indicating the center of the distribution, and the distribution is determined at the intersection point O1 with the major axis direction as the spreading direction. An ellipse having a predetermined spread μ ± 3σ based on the standard deviation σ when the value is a normal distribution with the value μ as an average.

By setting the tolerance zone 41 as described above, when determining whether the quenching pattern is good or bad, the variation (measurement error) in the eddy current measurement value due to the influence of the temperature condition, shape error, etc. of the workpiece 50 can be changed. It is possible to allow within a range according to the tendency, and it is possible to perform an accurate determination (equivalent evaluation) according to the properties of the workpiece 50.
As a result, the quenching quality can be improved, and over-detection (determining what is not actually a defective product as a defective product) can be reduced when determining the quality of the quenching pattern. Can be improved.

Next, when performing eddy current measurement in the quenching pattern inspection method as described above, a preferable setting method for the frequency (hereinafter referred to as “excitation frequency”) of the AC excitation signal V1 in the excitation coil 11 of the eddy current through coil 13 is described. explain.
The XY plane 40 used in the quenching pattern inspection method described above includes the Y value that is the magnitude (amplitude value) of the detection signal V2 from the detection coil 12 of the eddy current through coil 13 and the AC excitation of the detection signal V2. This is a coordinate plane determined from the X value that is a value (Ycos Φ) resulting from the phase difference Φ with respect to the signal V1, and corresponds to an impedance plane that indicates the relationship of impedance to conductivity and permeability.

For this reason, when the excitation frequency changes, the portion (numerical range) corresponding to the impedance plane of the XY plane 40 also changes with the change. When the portion corresponding to the impedance plane of the XY plane 40 changes, the tendency of the distribution of the non-defective product data 42 and the like on the XY plane 40 also changes, and the shape of the tolerance zone 41 (the shape of the set ellipse) also changes. Change. Accordingly, the separation value of the measurement point on the XY plane 40 also changes.
That is, when the excitation frequency changes, even if the non-defective product data 42 and the non-patterned product data 44 are acquired using the same non-defective product and defective product (hardened pattern cut product) for the workpiece 50, the impedance plane of the XY plane 40 is obtained. And the values of the X value and the Y value are different, and the shape of the tolerance zone 41 set based on the non-defective product data 42 and the separation values of the measurement points are different.

  Therefore, using the separation value of the pattern piece data 44 as an index, the excitation frequency is set so that the separation value of the pattern piece data 44 is large. That is, as the separation value of the pattern cut product data 44 is larger, the distinction between the non-defective product data 42 and the pattern cut product data 44 on the XY plane 40 becomes clearer. By using a large excitation frequency, the accuracy of the determination of the quality of the quenching pattern for the workpiece 50 is improved.

  That is, in the quenching pattern inspection method according to the present embodiment, the value of the distance along the shape of the boundary line 43 with respect to the boundary line 43 defining the tolerance zone 41 on the XY plane 40 is defined as a separation value. The separation value of the point (pattern breakage data 44) on the XY plane 40 determined from the X value and the Y value of the quenching pattern breakage measured using the eddy current penetrating coil 13 is relatively large. It is preferable to set the excitation frequency used for the measurement of the X value and the Y value for the inspection site 50a of the workpiece 50.

An example of setting the excitation frequency will be described with reference to FIG.
FIG. 11 shows an example of measurement results for the change in the separation value of the tolerance zone 41 and the pattern piece data 44 with the change in the excitation frequency. 11A to 11H show the cases where the excitation frequency is 4.8 kHz, 6.0 kHz, 7.5 kHz, 10.0 kHz, 12.0 kHz, 15.0 kHz, 20.0 kHz, and 24.0 kHz, respectively. The distribution of a large number of non-defective product data 42 on the XY plane 40, the tolerance zone 41, and the distribution of the pattern cut product data 44 are shown. Note that the set ellipse (the shape of the boundary line 43) that is the shape of the tolerance zone 41 is determined using the variation in the distribution of the non-defective product data 42 and the spread of μ ± 3σ in the normal distribution as described above.

As can be seen from the present measurement result example shown in FIG. 11, when the excitation frequency changes, the shape of the tolerance zone 41 on the XY plane 40, and the major axis (first principal component) and minor axis of the ellipse as the shape ( It can be seen that the direction of the second main component) and the separation value of the pattern cut product data 44 change.
Therefore, when focusing on the separation value of the pattern cut product data 44, as shown in FIG. 11A, when the excitation frequency is 4.8 kHz, the separation value is in the range of 3.26 to 3.45. In the case of 6.0 kHz, the separation value is in the range of 4.13 to 4.31 (FIG. 5B), and in the case of 7.5 kHz, the separation value is in the range of 3.56 to 3.73 (the same figure). (C)) In the case of 10.0 kHz, the separation value is in the range of 3.06 to 3.34 ((d) in the figure). In the case of 12.0 kHz, the separation value is 3.32 to 3.45. In the case of the range (Fig. (E)), 15.0 kHz, the separation value is in the range of 2.52 to 2.72 (Fig. (F)), and in the case of 20.0 kHz, the separation value is 1.52.
In the case of ˜1.88 (FIG. (G)) and 24.0 kHz, the separation value was in the range of 2.34 to 2.38 (FIG. (H)).

When the above result is obtained with respect to the relationship between the excitation frequency and the separation value of the pattern piece product data 44, the excitation frequency having the largest separation value of the pattern piece product data 44 is set as the optimum frequency. This is used as an excitation frequency when eddy current measurement is performed on the workpiece 50 to be inspected.
That is, in this measurement result example, 6.0 kHz (separation value 4.13 to 4.31), which is the excitation frequency when the separation value of the pattern breakage product data 44 becomes the largest, is selected, and the selected excitation frequency. Is set as the optimum frequency.

  Thus, by setting the excitation frequency so that the separation value of the pattern cut product data 44 is relatively large on the XY plane 40, the accuracy of the determination of the quality of the quenching pattern can be improved. In addition, for eddy current measurement in quenching pattern inspection, the optimum condition for excitation frequency, which is one of the measurement conditions, is determined based on the measurer's knowledge and proficiency in eddy current measurement based on an index called separation value. I can guide you.

In the eddy current measurement in the quenching pattern inspection according to the present embodiment, the position of the eddy current penetrating coil 13 with respect to the inspection part 50a of the workpiece 50 is preferably set using the separation value of the pattern piece data 44 as an index. .
That is, since the position of the exciting coil 11 and the detection coil 12 in the axial direction of the workpiece 50 with respect to the inspection site 50a changes due to the change of the position of the vortex penetration coil 13 with respect to the inspection region 50a, The separation value of the measurement point at is also changed.
Therefore, using the separation value of the pattern breakage product data 44 as an index, the position of the eddy current penetrating coil 13 relative to the inspection site 50a (the position in the axial direction of the workpiece 50, below) so that the separation value of the pattern breakage product data 44 becomes large. It is preferable to simply set “coil position”.

Specifically, as shown in FIG. 5, the workpiece 50 (inspection site 50 a) of the eddy current penetrating coil 13 is moved by moving a moving support member 34 that movably holds the vortex penetrating coil 13 in the measurement jig 30. While changing the position (position in the axial direction) with respect to, for example, a 1.0 mm pitch or a 0.5 mm pitch, the separation value of the pattern piece data 44 at each position is measured.
Then, in the predetermined range in which the coil position is moved, the coil position where the separation value of the pattern breakage product data 44 becomes the largest is set as the optimum coil position, and the optimum coil position is set as the coil position for the workpiece 50 to be inspected. Used as

  Thus, by setting the coil position on the XY plane 40 so that the separation value of the pattern cut product data 44 is relatively large, the accuracy of the determination of the quality of the quenching pattern can be improved. In addition, for eddy current measurement in quenching pattern inspection, the optimum condition for coil position, which is one of the measurement conditions, is determined based on the measure and knowledge of the eddy current measurement of the measurer based on an index called separation value. I can guide you.

  As described above, by using the separation value as an index, the optimum excitation frequency and coil position can be easily set, and conditions suitable for eddy current measurement in the quenching pattern inspection can be easily realized. Become.

The figure which shows the relationship of the layer state of the depth direction of a hardening member, hardness, and magnetic permeability. The schematic diagram which shows the apparatus structure for performing the eddy current measurement relevant to this invention. The figure which shows the relationship between the alternating current excitation signal and detection signal in eddy current measurement. The partial cross section figure which shows the structure of a workpiece | work and a eddy current penetration coil. The figure which shows the whole structure of the hardening pattern inspection apparatus which concerns on one Embodiment of this invention. The figure which shows the example of an eddy current measurement result about X value and Y value. Sectional drawing which shows the quality goods about a workpiece | work and a quenching pattern piece. The flowchart which shows an example of the determination algorithm used for the quality determination of a quenching pattern. The image figure of a tolerance zone and a separation value. The flowchart which shows an example of the flow of the determination used for the quality determination of a quenching pattern. The figure which shows the example of a measurement result about the tolerance zone and the change of a separated value with the change of an excitation frequency. The figure which shows the assembly (CVJ assembly) which is the object of the conventional hardening pattern test | inspection.

11 Excitation coil 12 Detection coil 13 Eddy current penetrating coil (eddy current sensor)
20 Eddy current flaw detector (measuring means, judging means)
22 measurement unit 23 determination unit 40 XY plane (coordinate plane)
40x X axis (first coordinate axis)
40y Y-axis (second coordinate axis)
41 Tolerance Zone (Allowable Error Area)
42 Good product data 43 Boundary line 44 Pattern cut product data 50 Work piece 50a Inspection part

Claims (6)

An eddy current sensor having an excitation coil for applying a predetermined AC excitation signal to a component to be inspected, and a detection coil for detecting a detection signal due to an eddy current from the component to be inspected to which the AC excitation signal is applied Is a quenching pattern inspection method for judging the quality of a quenching pattern for a part to be inspected,
It is determined from a first coordinate axis indicating a value resulting from a phase difference of the detection signal with respect to the AC excitation signal, and a second coordinate axis orthogonal to the first coordinate axis and indicating a value of the detection signal. Using a coordinate plane,
Based on the distribution of points on the coordinate plane determined from the values resulting from the phase differences and the magnitudes of the detection signals for a number of non-defective products measured using the eddy current sensor, an allowable error region on the coordinate plane Set in advance,
A distance value along the shape of the boundary line with respect to a boundary line defining the allowable error region on the coordinate plane is defined as a separation value;
The separation value of the point on the coordinate plane determined from the value caused by the phase difference and the value of the detection signal for a defective product having a quenching pattern breakage measured using the eddy current sensor is compared. So that
Used to measure the value resulting from the phase difference and the magnitude of the detection signal for the examination site, and setting the frequency of the AC excitation signal,
Using the eddy current sensor, measure the value resulting from the phase difference and the value of the detection signal for the inspection site of the inspection target component,
Depending on whether or not a point on the coordinate plane determined from the measured value of the phase difference and the magnitude of the detection signal is within the allowable error region, the quenching pattern of the inspection target part is determined. A quenching pattern inspection method characterized by determining pass / fail. The separation value of the point on the coordinate plane determined from the value caused by the phase difference and the value of the detection signal for a defective product having a quenching pattern breakage measured using the eddy current sensor is compared. So that
The quenching pattern inspection method according to claim 1, wherein the position of the eddy current sensor with respect to the inspection site is set. The shape of the tolerance region is
The direction of the first principal component, which is the component having the highest contribution to the distribution, is the major axis direction, the direction of the second principal component orthogonal to the first principal component is the minor axis direction, and the major axis and the minor axis. Is determined from a value indicating the center of the distribution, and a predetermined spread based on a standard deviation when the distribution is a normal distribution in which the direction of the major axis is a spreading direction and the values at the intersections are averaged is defined. The quenching pattern inspection method according to claim 1, wherein the quenching pattern inspection method is an ellipse. An eddy current sensor having an excitation coil for applying a predetermined AC excitation signal to the inspection target component, and a detection coil for detecting a detection signal due to eddy current from the inspection target component to which the AC excitation signal is applied;
Measuring means for measuring a value resulting from a phase difference of the detection signal with respect to the AC excitation signal, and a value of the magnitude of the detection signal;
Using the coordinate plane defined by a first coordinate axis indicating a value resulting from the phase difference and a second coordinate axis perpendicular to the first coordinate axis and indicating the magnitude of the detection signal, Based on the distribution of points on the coordinate plane determined from the value resulting from the phase difference and the magnitude of the detection signal for a number of non-defective products measured by the above, an allowable error region in the coordinate plane is set in advance. The point on the coordinate plane determined from the value resulting from the phase difference and the magnitude of the detection signal for the inspection region of the inspection target part measured by the measuring means is within the allowable error region. depending on whether or not, a quenching pattern inspection apparatus and a determining means for determining quality of quenching pattern for the component being tested,
A distance value along the shape of the boundary line with respect to a boundary line defining the allowable error region on the coordinate plane is defined as a separation value;
The separation value of the point on the coordinate plane determined from the value caused by the phase difference and the value of the detection signal for a defective product having a quenching pattern breakage measured using the eddy current sensor is compared. Hardening pattern inspection characterized in that the frequency of the AC excitation signal used for measurement of the value resulting from the phase difference and the magnitude of the detection signal for the inspection region is set so as to be larger apparatus. The separation value of the point on the coordinate plane determined from the value caused by the phase difference and the value of the detection signal for a defective product having a quenching pattern breakage measured using the eddy current sensor is compared. So that
The quenching pattern inspection apparatus according to claim 4, wherein a position of the eddy current sensor with respect to the inspection site is set. The determination means includes
The shape of the tolerance region is
The direction of the first principal component, which is the component having the highest contribution to the distribution, is the major axis direction, the direction of the second principal component orthogonal to the first principal component is the minor axis direction, and the major axis and the minor axis. Is determined from a value indicating the center of the distribution, and a predetermined spread based on a standard deviation when the distribution is a normal distribution in which the direction of the major axis is a spreading direction and the values at the intersections are averaged is defined. The quenching pattern inspection apparatus according to claim 4 or 5, wherein the quenching pattern inspection apparatus has an ellipse.

Images