JP3173365B2 - Stress measurement method using magnetostriction effect - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION The present invention relates to a stress applied to a steel structure or a mechanical part by utilizing magnetic anisotropy caused by a magnetostriction effect, and more particularly to a stress applied periodically and repeatedly in a short time. A non-destructive measurement method.

[0002]

2. Description of the Related Art As a method for measuring a stress applied to a ferromagnetic material such as a steel material, there is a stress measuring method utilizing a magnetostrictive effect, that is, a phenomenon in which magnetic properties are changed by the stress. Above all, a stress measurement method utilizing magnetic anisotropy caused by the magnetostriction effect is disclosed in Japanese Patent Laid-Open No. JP-A-121325, JP-A-1-135338, JP-A-7-110
No. 270 or Reference 1 [Sakai et al .: Proceedings of the 50th Annual Scientific Lecture Meeting of the Japan Society of Civil Engineers, P662-663 (1995.
9)].

[0003] This method is based on the following principle. FIG. 7 shows a principle diagram of a stress measurement method using magnetic anisotropy generated by the magnetostriction effect. In FIG. 7, 1 is a magnetostrictive sensor, 11 is a U-shaped yoke wound with an exciting coil, 11a and 11b are open ends of the yoke 11, 12 is a U-shaped yoke wound with a detecting coil, 12a. , 12b are open ends of the yoke 12, 13 is an AC power supply, 14 is a voltmeter,
Reference numeral 20 denotes an object to be measured (magnetic material), and reference numeral 30 denotes a direction in which a magnetic flux flows. Here, the magnetostrictive sensor 1 indicates the whole of the yoke 11, the yoke 12, the AC power supply 13, and the voltmeter 14. The yoke 11 and the yoke 12 are orthogonal to each other at the center of the yoke saddle.

Now, when a tensile stress σ _X acts on the X-axis direction of the DUT 20, X, X,
The magnetic permeability μ _X , μ _Y in the Y-axis direction is related by the following equation (5) due to the magnetostrictive effect, that is, magnetic anisotropy occurs: μ _X > μ _Y (5)

[0005] The magnetostrictive sensor 1 is brought close to the measured object 20 in such a state, and the yoke 1 of the magnetostrictive sensor 1 is moved.
When an object to be measured 20 is excited by passing an alternating current through the exciting coil wound around the coil 1, most of the magnetic flux emitted from the open end 11a of the yoke 11 goes directly to the open end 11b of the yoke 11, but Since a magnetic anisotropy as shown in the equation (5) is generated in 20 by the tensile stress σ _X , a part of the magnetic flux flows to the opening end 11 b of the yoke 11 via the yoke 12. Therefore, an electromotive force V represented by the following equation (6) is induced in the detection coil wound around the yoke 12. V = M ₀ · (μ _X −μ _Y ) · COS [2 · (θ−π / 4)] (6) where V is a rectified value of the AC electromotive force induced in the detection coil, M ₀ is a constant determined by excitation conditions, coil conditions, magnetic characteristics of the device under test 20, etc., and COS [2 · (θ−π /
4)] is a cosine function, and θ is the open ends 12a and 12 of the yoke 12.
The angle between the straight line connecting b and the X axis.

The difference in magnetic permeability (μ _X −μ _Y ) is the difference in stress (σ
_X− σ _Y ), the equation (6) is calculated by the following equation (7).
Can be rewritten as V = M · (σ _X −σ _Y ) · COS [2 · (θ−π / 4)] (7) where M is an excitation condition, a coil condition, and a magnetic characteristic of the DUT 20. It is a constant determined by the above.

From the equation (7), the stress applied to the measured object can be obtained by measuring V.

FIG. 8 shows the relationship between the electromotive force of the magnetostrictive sensor and the applied stress obtained when a stress is repeatedly applied to the same portion of an object to be measured and when the stress is removed.

The load stress and the electromotive force of the magnetostrictive sensor form a so-called hysteresis curve that does not show the same electromotive force for the same stress on the side where the stress is added and on the side where the stress is removed. Therefore, when the stress is repeatedly applied, the stress cannot be uniquely obtained from the measured electromotive force of the magnetostrictive sensor.

[0010] This hysteresis curve is caused by residual magnetization accumulated in the material during repeated measurement. 5-1
No. 0337 proposes a method of measuring the stress after erasing the residual magnetization of the DUT by an AC demagnetization method.

FIG. 9 shows an AC demagnetization treatment method described in the above-mentioned patent publication. 9 denote the same components as those in FIG.

Before measuring the stress of the device under test 20, a yoke 11 around which an exciting coil is wound is placed on the device under test 20, and an alternating current is supplied to the yoke 11 by an AC power supply 13 while gradually attenuating its amplitude. Then, the residual magnetization of the device under test 20 is erased.

FIG. 10 shows the relationship between the load stress after the AC demagnetization treatment and the electromotive force of the magnetostrictive sensor. By performing the AC demagnetization treatment, a hysteresis curve is not shown, and the stress can be uniquely obtained from the measured electromotive force of the magnetostrictive sensor.

[0014]

However, in the method described in Japanese Utility Model Publication No. 5-10337, since one demagnetization process requires several seconds, the cycle of the stress repeatedly applied is equal to the demagnetization time. If the period is shorter than the demagnetization treatment time as in a fatigue test, it cannot be applied as shown below.

FIG. 11 shows that 10 ± 5 kgf / mm ²
The results obtained by applying a load stress that changes at a cycle of 1 Hz to the range described above and obtaining the stress from the electromotive force of the magnetostrictive sensor measured using the calibration curve that combines the demagnetization treatment described above.

When the stress is applied in such a short cycle, the demagnetization treatment is not effective because the stress is significantly different from the actual applied stress measured by the strain gauge.

The present invention has been made in order to solve such a problem, and provides a stress measurement method using the magnetostriction effect that enables accurate stress measurement even when a cycle of stress repeatedly applied is short. The purpose is to:

[0018]

An object of the present invention is to provide a U-shaped yoke around which an exciting coil is wound and a U-shaped yoke around which a detection coil is wound, so that they are orthogonal to each other at the center of the yoke saddle. In this case, the open end side of the U-shaped yoke is brought close to the object to be measured, an alternating current is applied to the exciting coil to excite the object to be measured, and an electromotive force induced in the detecting coil. In the stress measuring method using a magnetostrictive effect using a magnetostrictive sensor capable of measuring the load stress of the object to be measured by measuring the load stress, the magnetostrictive effect is characterized in that a periodic load stress is measured by the following method. Stress measurement method.

(A) The stress on the side of the periodic load stress to which the stress is applied is obtained from a calibration curve of the electromotive force and the stress of the magnetostrictive sensor which has been obtained in advance.

(B) Regarding the stress on the stress-removing side of the periodic load stress, i) Substituting the electromotive force V _x measured by the magnetostrictive sensor into the following equation (4) to standardize V _s Ii) a coefficient when the relationship between V _s and σ _s in which the applied stress σ _x corresponding to V _x is normalized by the following equation (3) is expressed by the following n-th order polynomial (1): the C _k, the maximum load stress sigma _max measured by the magnetostrictive sensor C _k
The following n-order polynomial that represents the relationship between the sigma _max and (2)
By substituting the determined, iii) obtains the V _s in the formula (1) from sigma _s, iv) calculating the sigma _x from the sigma _s and the formula (3).

V _s = Σ (C _k · σ ^k _s ) (1) C _k = Σ (G _k · σ ^k _max ) (2) where k = 0, 1, 2,... ... N _s = σ _x / (σ _max −σ _min ) (3) V _s = V _x / (V _max −V _min ) (4) where σ _min is the minimum load The stress, V _max is the maximum electromotive force measured by the magnetostrictive sensor, V _min is the minimum electromotive force measured by the magnetostrictive sensor, and G _k is the coefficient of equation (2).

FIG. 2 shows the relationship between the applied stress and the electromotive force of the magnetostrictive sensor when a periodic stress is applied by changing the maximum applied stress without performing the demagnetization process.

On the stress application side, the relationship between the applied stress and the electromotive force of the magnetostrictive sensor is represented by a single curve even if the applied stress is repeated or the maximum value of the applied stress is changed. Therefore, if this is used for the calibration curve, the load stress can be uniquely obtained from the measured electromotive force of the magnetostrictive sensor.

On the other hand, since a hysteresis curve different from that on the stress application side is formed on the stress relief side, the load stress cannot be uniquely determined from the measured electromotive force of the magnetostrictive sensor. This hysteresis curve also differs depending on the maximum load stress. Therefore, if a curve on the stress relief side corresponding to the maximum load stress can be determined by any method, the load stress can be uniquely obtained from the electromotive force of the magnetostrictive sensor measured by using the curve as a calibration curve.
The method will be described below.

First, in order to facilitate the mathematical processing, the load stress σ _x on the stress removing side and the rectified value V _x of the AC electromotive force measured by the magnetostrictive sensor are calculated by the above equations (3) and (4). Standardize as follows.

FIG. 3 is a diagram in which FIG. 2 is rewritten using stress and electromotive force normalized according to the equations (3) and (4).

By normalizing, the load stress and the electromotive force of the magnetostrictive sensor are made dimensionless. The curve on the stress relief side in FIG. 3 is represented by the above-described n-th order polynomial (1).

The maximum electromotive force V _max is measured by changing the maximum load stress σ _max using the material constituting the object to be measured in advance, and represents the relationship between the coefficient C _k of equation (1) and the maximum load stress σ _max. The above-mentioned n-th order polynomial (2) is obtained.

Using the calibration curve on the stress application side, the maximum load stress σ _max is measured from the electromotive force of the magnetostrictive sensor.
_By substituting _max into Expression (2) to obtain C _k , Expression (1) is determined.

The electromotive force V _x of the magnetostrictive sensor measured at the time of the stress elimination is standardized by equation (4) to obtain V _s , and this V _s is substituted into equation (1) to standardize the stress σ _s Ask for.

By substituting this σ _s into the equation (3), the electromotive force V _{x of the} magnetostrictive sensor measured when the stress is released is calculated.
It is possible to calculate the applied stress sigma _x corresponding to.

[0032]

BEST MODE FOR CARRYING OUT THE INVENTION The degree n of the polynomials of the above equations (1) and (2) is preferably as large as possible, because the measurement accuracy is improved. It is sufficient for evaluation of the applied stress.

FIG. 4 (C ₃ ) shows the relationship between C _k and the maximum load stress σ _max obtained when n = 3 based on the data of FIG.
These are shown in FIG. 5 (C ₂ ) and FIG. 6 (C ₁ ), respectively. However,
C ₀ is omitted because it is sufficiently small.

Based on this relationship, a third-order polynomial (1) is determined from the maximum load stress σ _max measured by the magnetostrictive sensor, and a curve on the stress relieving side is obtained. Therefore, the load stress can be uniquely calculated from the electromotive force of the magnetostrictive sensor measured also on the stress removing side.

[0035]

Example: 1H in the range of 10 ± 5 kgf / mm ² for steel
A load stress that changes in a cycle of z was applied, and the stress was measured by a strain gauge and the method of the present invention.

FIG. 1 shows the periodic load stress calculated by the method of the present invention. It can be seen that the applied stress calculated according to the method of the present invention substantially matches the actual stress measured by the strain gauge.

[0037]

Since the present invention is configured as described above, it is possible to provide a stress measuring method utilizing the magnetostrictive effect, which enables accurate stress measurement even when the cycle of the stress repeatedly applied is short.

[Brief description of the drawings]

FIG. 1 is a diagram showing periodic load stress calculated by the method of the present invention.

FIG. 2 is a diagram illustrating a relationship between a load stress and an electromotive force of a magnetostrictive sensor when a periodic stress is applied while changing a maximum load stress without performing a demagnetization process.

FIG. 3 is a diagram obtained by rewriting FIG. 2 using normalized stress and electromotive force.

FIG. 4 is a diagram showing a relationship between C ₃ and a maximum load stress σ _max .

FIG. 5 is a diagram showing a relationship between C ₂ and a maximum load stress σ _max .

FIG. 6 is a diagram showing a relationship between C ₁ and a maximum load stress σ _max .

FIG. 7 is a principle diagram of a stress measurement method using magnetic anisotropy generated by a magnetostriction effect.

FIG. 8 is a diagram showing a relationship between an electromotive force and a load stress of a magnetostrictive sensor obtained at the time of adding and removing stress when a repeated stress is applied.

FIG. 9 is an explanatory diagram of an AC demagnetization processing method disclosed in Japanese Utility Model Publication No. 5-10337.

FIG. 10 is a diagram illustrating a relationship between a load stress after an AC demagnetization process and an electromotive force of a magnetostrictive sensor.

FIG. 11 is a diagram showing a result of stress measurement using a calibration curve using a conventional demagnetization process when a short-cycle stress is applied.

[Explanation of symbols]

 Reference Signs List 1 Magnetostrictive sensor 11 U-shaped yoke wound with exciting coil 11a Open end of yoke 11b Open end of yoke 11 12 U-shaped yoke wound with detection coil 12a Open end of yoke 12 12b Yoke 12 Open end 13 AC power supply 14 Voltmeter 20 DUT 30 Direction of magnetic flux

──────────────────────────────────────────────────続 き Continued on front page (58) Field surveyed (Int.Cl. ⁷ , DB name) G01L 1/00 G01L 1/12

Claims (1)

(57) [Claims] 1. A U-shaped yoke around which an exciting coil is wound and a U-shaped yoke around which a detection coil is wound are arranged so as to be orthogonal to each other at the center of a yoke saddle. The yoke's open end side is brought close to the DUT, an alternating current is applied to the excitation coil to excite the DUT, and the electromotive force induced in the detection coil is measured to measure the DUT. A stress measuring method using a magnetostrictive effect using a magnetostrictive sensor capable of measuring a load stress of a sample, wherein a periodic load stress is measured by the following method. (A) The stress on the stress-applied side of the periodic load stress is obtained from a previously obtained electromotive force of the magnetostrictive sensor and a calibration curve of the stress. (B) Stress stress relief side of the cyclic loading stress, i) the electromotive force V _x to be measured by magnetostrictive sensor into Equation (4) below, it obtains the V _s normalized, ii) A coefficient C _k when the relationship between V _s and σ _{s obtained} by standardizing the applied stress σ _x corresponding to the above V _x by the following equation (3) and the following n-th order polynomial (1) is , the maximum load stress sigma _max measured by the magnetostrictive sensor wherein C _k
The following n-order polynomial that represents the relationship between the sigma _max and (2)
By substituting the determined, iii) obtains the V _s in the formula (1) from sigma _s, iv) calculating the sigma _x from the sigma _s and the formula (3). V _s = Σ (C _k · σ ^k _s ) (1) C _k = Σ (G _k · σ ^k _max ) (2) where k = 0, 1, 2,... _{n σ s = σ x / (} σ max -σ min) ··· (3) V s = V x / (V max -V min) ··· (4) where, sigma _min minimum load stress, V _max is the minimum emf, G _k is the maximum electromotive, V _min, as measured by a magnetostrictive sensor that is measured by the magnetostrictive sensor is a coefficient of the formula (2).