JP3863802B2 - Ferromagnetic material diagnosis method and Barkhausen noise voltage pulse width measurement system - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a non-destructive diagnostic method using Barkhausen noise and a voltage pulse width measurement system thereof, and is particularly used for diagnosing the material of an object to be measured which is a ferromagnetic material, such as a steel material. Application to a ferromagnetic material is preferred.
[0002]
[Prior art]
Conventionally, using the magnetic properties of ferromagnetic materials to change depending on the material, etc., it has been attempted to nondestructively diagnose the material of the object to be measured, which is a ferromagnetic material, using a magnetic method. It has been. Recently, attention has been paid to a method using Barkhausen noise (hereinafter referred to as BHN) resulting from the discontinuity of magnetization, and fatigue strength of mild steel using the method (LP Kajalainen et al., IEEE Trans. Mag. MAG). 16, 514 (1980)) and methods for estimating the toughness of tool steel (Nakai et al., Iron and steel, 75, 833, (1989)) have been proposed.
[0003]
When the object to be measured, which is a ferromagnetic material, is AC-excited, the magnetization changes inside the object to be measured as the exciting magnetic field changes. Usually, the change in magnetization is performed by domain wall motion, and the domain wall motion is discontinuous due to resistance at positions such as precipitates, grain boundaries, and dislocations. For this reason, the magnetization of the object to be measured changes discontinuously, and a high-frequency pulsed voltage waveform corresponding to the discontinuous change is induced in the detection coil. This discontinuous magnetization change is called Barkhausen jump, and a pulsed voltage waveform corresponding to this jump is called Barkhausen noise. Since the state of discontinuous magnetization changes depending on the state of precipitates, grain boundaries, dislocations, etc., BHN has a correlation with the above-mentioned material of the object to be measured, so BHN is useful for material diagnosis of ferromagnetic materials. It is considered a parameter.
[0004]
So far, the effectiveness of the effective voltage of the voltage waveform has been shown as a BHN parameter that correlates with the crystal grain size and precipitate grain size of steel materials (H. Sakamoto, M. Okada, M. Homma: IEEE Trans. Magn., 23-5 (1989), 2236). Specifically, the experimental result shows that the effective voltage is proportional to 1 / √d when the crystal grain size of α-Fe is d, and the effective voltage is proportional to D ² when the precipitate particle size is D. Indicated.
[0005]
Among them, Sakamoto et al. Assumed that the shape of the voltage pulse generated by a single discontinuous movement of the domain wall is a Gaussian pulse, and based on the idea that the grain size and precipitate interval of steel materials are proportional to the Gaussian pulse duration. The duration was defined as the standard deviation of the Gaussian pulse, and the RMS voltage was calculated theoretically. As a result, the calculation result and the above-mentioned experimental result agreed well. This report by Sakamoto et al. Adopts the concept based on the conventional knowledge that one Barkhausen jump has a correlation with structural parameters such as crystal grain size and precipitate spacing of steel. However, the Barkhausen jump that is actually detected does not result in the detection of individual Barkhausen jumps because multiple jumps occur in the same time zone and pulses overlap.
[0006]
Prior to this, it was attempted by PJCoyne et al. To measure individual Barkhausen jumps with silicon steel sheets by reducing the time rate of change of the applied magnetic field to a level of 10 ⁻⁵ to 10 ⁻³ Oe / ms. . As a result, the obtained BHN voltage waveform was not necessarily a pulse waveform corresponding to one Barkhausen jump, but a “pulse cluster” in which pulses overlapped was observed (PJ Coyne and J. Kramer: AIP Conf. Proc. , 24 (1975), 726). As described above, even if the time rate of change of the magnetic field is reduced, it is difficult to separate a plurality of jumps occurring in the same time zone, and it is impossible to detect a BHN voltage pulse corresponding to each jump. there were.
[0007]
[Problems to be solved by the invention]
As described above, in the actually detected voltage waveform of BHN, a plurality of domain walls move in the same time zone, so that individual voltage pulses overlap each other. For this reason, it is impossible to separate individual voltage pulses and measure their time width, and it is difficult to diagnose the material of the ferromagnetic material from the time width.
[0008]
Therefore, the present invention determines the voltage pulse width having a correlation with the material from the BHN voltage waveform even if the BHN voltage pulses overlap, and accurately diagnoses the material such as the structure of the object to be measured which is a ferromagnetic material without destruction. It is an object of the present invention to provide a method and a measurement system that make it possible.
[0009]
[Means for Solving the Problems]
To achieve the above object, the present invention has the following aspects.
[0010]
(1) A method for diagnosing a ferromagnetic material using BHN, the step of exciting an object to be measured, which is a ferromagnetic material, by an exciting unit, and a voltage induced in the detector by a magnetization change of the object to be measured. A step of detecting as a waveform, a step of extracting a voltage waveform of BHN from the detected voltage waveform by frequency filtering, and a zero-crossing time T _i at which the voltage becomes zero after the voltage waveform of BHN is shown in the voltage-time domain. Determining a time difference T _{i + 1} −T _i between adjacent times T _i and T _{i + 1} and defining the voltage pulse width of the BHN, and determining the material of the object to be measured from the voltage pulse width. And diagnosing the ferromagnetic material.
[0011]
(2) An average value of at least two time differences T _{i + 1} −T _i is obtained, and the material of the object to be measured is diagnosed using the average value as a representative value of the voltage pulse width of the BHN ( 1. A method for diagnosing a ferromagnetic material according to 1).
[0012]
(3) A system for measuring a voltage pulse width of a BHN used for diagnosis of a ferromagnetic material, an excitation unit for exciting a measurement object that is a ferromagnetic material, and a magnetization change of the measurement object. a detection unit for detecting a voltage waveform, and a frequency filtering unit for extracting a voltage waveform of BHN from the detected voltage waveform, obtains a zero cross time T _i which the voltage becomes zero in the voltage waveform of the BHN, and adjacent time T _i measuring system of the voltage pulse width of BHN, characterized in that it comprises a calculation unit which obtains a time difference T _{i + 1} -T _i and T i _{+ 1} and the voltage pulse width of the BHN.
[0013]
(4) The calculation unit obtains an average value of at least two T _{i + 1} −T _{i and} sets the average value as a representative value of the voltage pulse width of the BHN. BHN voltage pulse width measurement system.
[0014]
DETAILED DESCRIPTION OF THE INVENTION
The specific ferromagnetic material diagnosis method, BHN voltage pulse width measurement method and measurement system of the present invention will be described below.
As a method of changing the magnetization of the object to be measured, a method of changing a magnetic field by applying a magnetic field from the outside to a local region to be measured is generally used. As a specific means, an excitation part such as an excitation head in which an excitation coil is wound around a core of an excitation coil or a soft magnetic material is used, and a ferromagnetic material is arranged in the excitation coil, or the excitation head is made of a ferromagnetic material. The magnetic field to be applied is changed by changing the value of the current passed through the exciting coil. In the present invention, since the voltage pulse width of BHN may change depending on the time change rate of the magnetic field, a constant value of the change rate of the magnetic field is desirable when measuring BHN. The time change rate of the magnetic field is about 10 ⁻⁸ to several tens of Oe / ms.
[0015]
Magnetization changes that change with changes in the magnetic field are detected as voltage waveforms induced in the detection unit. As the detection unit, a detection head in which a detection coil is wound around a detection coil or a soft magnetic core is used, and the object to be measured is placed in the detection coil, or the detection head is applied to the object to be measured. In particular, when a detection coil is applied to a site to be diagnosed, BHN only from that region can be obtained, so that highly accurate measurement can be efficiently performed.
[0016]
The voltage waveform induced in the detection coil is a state in which a high-frequency BHN voltage waveform is superimposed on a voltage waveform having substantially the same frequency as the frequency of the exciting magnetic field. Therefore, the BHN voltage waveform is extracted from the frequency filtering unit combining the high-pass filter and the low-pass filter. The filter conditions vary depending on the rate of change of the excitation magnetic field with respect to time and the material type of the object to be measured. Usually, the frequency of the high-pass filter is several tens of Hz to several tens of kHz, and the frequency of the low-pass filter is several kHz to several MHz. is there. Here, since the voltage value of the BHN voltage waveform is as small as several μV, it is desirable to amplify the voltage with a low noise type voltage amplifier in advance. Since the influence of noise in the filtering unit can be reduced, it is desirable to perform voltage amplification before filtering.
[0017]
The analysis of the BHN voltage waveform is performed after showing in the voltage-time domain. FIG. 1 shows an example of the BHN voltage waveform obtained. The BHN voltage waveform is composed of a plurality of chevron-shaped waveforms having positive and negative voltage polarities, and a small and narrow pulse waveform may be superimposed on the chevron-shaped waveform. This is a result of detection of overlapping of individual voltage pulses because a plurality of domain walls move in the same time zone. As a result of numerous experiments, the present inventors have detected a voltage zero-crossing time T _{i at} which the voltage becomes 0 as a pulse width of a BHN waveform in which a plurality of voltage pulses overlap, and adjacent times T _i and T _{i + 1} . When the difference T _{i + 1} −T _i is employed, the voltage pulse width has a strong correlation with the material of the object to be measured even when there is an overlap.
[0018]
For example, the third pulse waveform in FIG. 1 is composed of two pulses. Focusing on zero-crossing times T ₃ and T ₄ across the time axis where the waveform voltage is 0, the third waveform The pulse width is T ₄ −T ₃ which is the difference. In contrast, the conventional method of dividing the total generation time by the number of peaks to obtain the pulse width, for example, taking the third pulse as an example, the value obtained by dividing the difference T ₄ -T ₃ by the number of peaks of the waveform 2 In the conventional method of setting the pulse width, the correlation between the pulse width and the material obtained by the influence of overlapping pulses becomes extremely low. As described above, even when a plurality of pulses overlap each other, the inventors have determined that the voltage pulse width is a value obtained from the difference between the voltage zero crossing times, so that the pulse width has a strong correlation with the material. The inventors have found that the material can be diagnosed from the pulse width by using the correlation, and have completed the present invention.
[0019]
Next, the handling when the BHN voltage waveform is composed of discretely generated pulses as shown in FIG. 2 will be described. When the voltage between time T _i and T _{i + 1} is 0, it is considered that there is no BHN pulse. Therefore, the first voltage pulse exists between time T ₁ and T ₂ , the voltage pulse width is T ₂ −T _1, and the voltage is 0 between time T ₂ and T ₃ , so the pulse is It shall not exist. Since the second pulse exists between the times T ₃ and T ₄ , the second voltage pulse width is T ₄ −T ₃ .
[0020]
The analysis of the BHN voltage waveform described above is performed by an arithmetic unit programmed with an algorithm. For example, the BHN voltage waveform can be converted into digital data by AD conversion or the like, and analysis can be efficiently performed by using a computer as the calculation unit. In this case, the voltage zero crossing time when the BHN waveform becomes voltage 0 may be obtained by complementing between discrete data points as shown in FIG.
[0021]
Ferromagnetic materials that can be diagnosed by the voltage pulse width of BHN determined by the method of the present invention are precipitate spacing, crystal grain size, ratio of two-component phase, dislocation spacing, dislocation cell size, etc. It can be applied to places that can be a domain wall pinning site.
[0022]
For example, FIG. 5 schematically shows the result of measuring the pulse number distribution with respect to the BHN voltage pulse width for ferromagnetic materials A and B having different precipitate spacing distributions as shown in FIG. The distribution shape of the number of pulses with respect to the voltage pulse width was similar to that of the precipitates in the ferromagnetic materials A and B. From these, it is possible to create a calibration curve of the voltage pulse width and the precipitate interval, and even for materials whose deposit interval distribution is unknown, if the distribution of the number of pulses with respect to the voltage pulse width is measured, The distribution of precipitate spacing can be diagnosed.
[0023]
In addition, the average value obtained for a plurality of voltage pulse widths is highly reproducible for each measurement, and is optimal for accurate diagnosis. For example, if two or more voltage pulse widths are obtained from three or more times T _i , a plurality of average values can be obtained. Further, if the average value of the voltage pulse width and the calibration curve of the material are obtained in advance, the material can be diagnosed with high accuracy from the average value of the voltage pulse width for an unknown material.
[0024]
Here, it is also possible to obtain a calibration curve for the BHN pulse width and the material and use the calibration curve for other material diagnosis. For example, a calibration curve of the distance between cementite dispersed in ferrite and the BHN pulse width is obtained, and the interval between other precipitates precipitated in the ferrite using this calibration curve and the measured BHN pulse width, It is also possible to measure the distance between dislocations formed therein and the dislocation cell size.
[0025]
An example of the voltage pulse width measurement system of the BHN of the present invention is shown in FIG.
The magnetic head 1 includes an excitation head 11 as an excitation unit and a detection coil 12 as a detection unit. The excitation head 11 is obtained by winding an excitation coil around a silicon steel U-shaped core, and the detection coil 12 is an air-core coil. The detection coil 12 is fixedly disposed between both legs of the U-shaped core so that a change in magnetization in the ferromagnetic body due to excitation can be captured as a voltage waveform. The magnetic head 1 is placed on the surface of the object to be measured, which is a ferromagnetic material, and a change in magnetization at the position of the detection coil 12 is detected.
[0026]
Excitation is performed by controlling the current flowing through the excitation coil by the waveform generator 2 and the power amplifier 3 so that the rate of change of the magnetic field with respect to time is constant. The voltage waveform induced in the detection coil 12 is boosted by the low noise type voltage amplifying device 4 and then subjected to low-pass and high-pass filtering processing by the frequency filter 5 which is a frequency filtering unit. The BHN voltage waveform obtained by filtering is converted into digital data by the digital oscilloscope 6. Then, the data is sent to the computer 7 which is a calculation unit by GP-IB or the like, and analysis for obtaining the voltage pulse width is performed by a dedicated program. The analysis may be performed in real time at the same time as the measurement, or may be performed collectively after the BHN waveform data is stored in the memory.
[0027]
【Example】
Hereinafter, the present invention will be described in detail with some examples.
[0028]
Example 1
Several types of steel materials with different intervals of spheroidized cementite precipitated in α-Fe were prepared, and the cementite interval was diagnosed using BHN. The diagnostic system is as shown in FIG. 6, the excitation condition is a change rate of the magnetic field and takes a constant value of 2 × 10 ⁻⁴ Oe / ms, and the frequency filtering condition is a low-pass frequency of 100 kHz and a high-pass frequency of 500 Hz. .
[0029]
Investigate the correlation between the average value of the voltage pulse width and the average value of the cementite interval, and simultaneously obtain the number of positive and negative peaks from the same BHN voltage waveform. The correlation with the pulse width obtained by dividing was examined.
[0030]
FIG. 7 shows the result. According to the present invention, it has been found that the average value of the voltage pulse width is linearly related to the average value of the cementite interval, and if this calibration curve is used, the cementite interval can be diagnosed for an unknown steel material. On the other hand, with the conventional pulse width obtained using the number of peaks shown as a comparative example, the rate of change of the parameter value with respect to the cementite interval decreases as the cementite interval increases, making it impossible to obtain measurement accuracy in practice. It was found that it can not be used.
[0031]
As described above, it has been found that the use of the diagnostic method and the measurement system of the present invention makes it possible to accurately diagnose the distance between precipitates in a ferromagnetic material in a nondestructive manner.
[0032]
(Example 2)
Permalloys with different crystal grain sizes were prepared, and the crystal grain size was diagnosed using BHN. The diagnostic system is as shown in FIG. 6, and the excitation condition is a constant rate of 0.1 Oe / ms as the rate of change of the magnetic field, and the frequency filtering condition is a low-pass frequency of 100 kHz and a high-pass frequency of 2 kHz.
[0033]
The time of the analyzed BHN waveform was 128 ms, and the individual voltage pulse widths were obtained from the zero crossing time during which the waveform was voltage 0, and the average value of the voltage pulse widths was obtained therefrom. FIG. 8 shows the correlation between the average value of the voltage pulse width and the average crystal grain size obtained from the structure observation. It has been found that the average value of the voltage pulse width is linearly related to the average value of the crystal grain size, and that the crystal grain size can be diagnosed for an unknown steel material by using this calibration curve.
[0034]
As described above, it has been found that the use of the diagnostic method and measurement system of the present invention makes it possible to accurately diagnose the crystal grain size of a ferromagnetic material in a non-destructive manner.
[0035]
Example 3
Using FIG. 7 which is a calibration curve of the BHN pulse width and cementite interval obtained in Example 1, measurement of the dislocation interval and dislocation cell size formed in the steel material from the BHN pulse width was performed. The steel material used is a general structural steel having a C content of 0.16 wt%, and has a ferrite-pearlite structure. Prepare a fatigue test piece with a square cross-section and a constricted R50mm from 10mm to 8mm square in the center, and apply repeated compressive tensile load to this test piece to change the internal dislocation structure. It was. Test pieces A to D having four different dislocation intervals and dislocation cell sizes were prepared by changing the load and the number of repetitions.
[0036]
The BHN measurement was performed by the system shown in FIG. 6, and the BHN was measured by pressing the magnetic head 1 against the central portion of the test piece where fatigue damage was most advanced. The excitation rate is constant at 2 × 10 ^-4 Oe / ms at the rate of change of the magnetic field, and the excitation magnetic field is applied in the longitudinal direction of the test piece so that the magnetization of the object to be measured is reversed in the reverse direction from the saturated state. The excitation magnetic field was changed. The frequency filtering conditions are a low-pass frequency of 100 kHz and a high-pass frequency of 500 Hz. After analyzing the BHN waveform for 64 ms out of all the BHN waveforms and obtaining the zero crossing time when the voltage becomes 0, individual BHN voltage pulse widths were obtained, and the average value of these was used as the representative value of the voltage pulse width.
[0037]
In order to confirm the accuracy of the diagnosis result of the dislocation structure obtained from the BHN pulse width, the dislocation structure at the center of the test piece where the BHN pulse width was measured was observed with an electron microscope, and the measurement results of both were compared. Here, dislocation structures of different sizes were observed in the same test piece, but the average value of all sizes was taken as the measurement value.
[0038]
The dislocation interval and dislocation network size estimated from the BHN pulse width and the calibration curve of FIG. 7 were 0.23 μm for A, 1.1 μm for B, 1.8 μm for C, and 2.5 μm for D. In contrast, the size measured with an electron microscope is 0.30 μm for A, 1.4 μm for B, 2.1 μm for C, and 2.8 μm for D, which is almost the same as the size estimated from the BHN pulse width. I found out.
[0039]
From these, it was found that the BHN voltage pulse width of the present invention can diagnose not only fatigue but also the interval between dislocations formed inside due to various damages and the dislocation cell size. In addition, it is also clear that a material diagnosis can be performed using a calibration curve made of another material, that is, a material diagnosis can be performed even if the material of the measured object and the calibration curve are different. It was revealed.
[0040]
【The invention's effect】
The present invention is a method for measuring the voltage pulse width of BHN of a ferromagnetic material typified by steel materials such as carbon steel and stainless steel which are practically important. By using the voltage pulse width according to the present invention, disturbance noise can be reduced. It is possible to reduce the influence, and even if the BHN voltage pulse is weak, the effect of accurately diagnosing a material such as tissue can be obtained.
[Brief description of the drawings]
FIG. 1 is a diagram showing an example of a BHN voltage pulse width measured by the method of the present invention.
FIG. 2 is a diagram showing an example of a BHN voltage pulse width measured by the method of the present invention.
FIG. 3 is a diagram illustrating an example of a method of determining a zero crossing time T _i when BHN waveform data is discrete.
FIG. 4 is a diagram showing an example of a distribution of occurrence frequency with respect to a precipitate interval in ferromagnetic bodies A and B.
FIG. 5 is a diagram illustrating an example of a pulse number distribution with respect to a BHN voltage pulse width in ferromagnetic bodies A and B;
FIG. 6 is a schematic diagram showing an example of a measurement system for carrying out the method of the present invention.
FIG. 7 is a diagram showing a relationship between a BHN voltage pulse width and a cementite interval measured by the method of the present invention.
FIG. 8 is a graph showing the relationship between the BHN voltage pulse width measured by the method of the present invention and the crystal grain size of permalloy.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Magnetic head 2 Waveform generator 3 Power amplifier 4 Voltage amplifier 5 Frequency filter 6 Digital oscilloscope 7 Computer 11 Excitation head 12 Detection coil

Claims (4)

A method of diagnosing a ferromagnetic material using Barkhausen noise,
Exciting the object to be measured, which is a ferromagnetic material, at the excitation unit;
Detecting a change in magnetization of the object to be measured as a voltage waveform induced in a detection unit;
Extracting a voltage waveform of Barkhausen noise from the detected voltage waveform by frequency filtering;
Obtaining a zero-crossing time T _i at which the voltage becomes zero after the voltage waveform of the Barkhausen noise is shown in the voltage-time domain;
Determining a voltage pulse width of the Barkhausen noise by obtaining a time difference T _{i + 1} −T _i between adjacent times T _i and T _{i + 1} ;
And diagnosing the material of the object to be measured from the voltage pulse width. The average value of at least two time differences T _{i + 1} -T _i is obtained, and the material of the object to be measured is diagnosed using the average value as a representative value of the voltage pulse width of the Barkhausen noise. 2. The method for diagnosing a ferromagnetic material according to 1. A system for measuring the voltage pulse width of Barkhausen noise used for diagnosis of ferromagnetic materials,
An excitation unit for exciting a measured object that is a ferromagnetic material;
A detection unit for detecting a change in magnetization of the object to be measured as a voltage waveform;
A frequency filtering unit for extracting a voltage waveform of Barkhausen noise from the detected voltage waveform;
A zero crossing time T _{i at} which the voltage becomes 0 in the voltage waveform of the Barkhausen noise is obtained, and a time difference T _{i + 1} −T _i between adjacent times T _i and T _{i + 1} is obtained to obtain a voltage pulse of the Barkhausen noise. And a voltage pulse width measurement system for Barkhausen noise, comprising: an arithmetic unit for width. 4. The calculation unit according to claim 3, wherein the arithmetic unit obtains an average value of at least two time differences T _{i + 1} −T _i and uses the average value as a representative value of the voltage pulse width of the Barkhausen noise. Barkhausen noise voltage pulse width measurement system.

Images