About the Performance of Non-Multiplication Magnetization Method in a Magnetic Particle Testing

Transcription

1 19 th World Conference on Non-Destructive Testing 016 About the Performance of Non-Multiplication Magnetization Method in a Magnetic Particle Testing Michitaka HORI 1, Arihito KASAHARA 1 1 Nihon Denji Sokki Co., LTD, Tachikawa-city, Tokyo, Japan Contact hori@j-ndk.co.jp Abstract. In the magnetic particle testing, it has been advanced non-contact of the test object, improvement of detection performance of surface defects is required. We propose a magnetization method using the non-multiplication magnetization of high frequency. It was confirmed that to improve the defect detection performance by increasing the magnetization frequency. The use of non-multiplication magnetization scheme multidirectional coil, it was confirmed that all directions of the defect detection can be facilitated in a non-contact. We confirmed the characteristics of the new proposed method by using the simulation and experiment, and report these results. Introduction Recently, in the magnetic particle testing, it has been advanced non-contact of the test object, improvement of detection performance of surface defects is required. To perform the magnetic particle test in a non-contact, the coil magnetization method is widely used. In the conventional phase control using thyristors, it could not be easily changed frequency. Power devices used for magnetizing power supply will become high performance, and became a low cost. In addition, the inverter technology has advanced. By using the inverter, it is now possible to easily change the frequency of the magnetizing current. We propose the magnetization method using the non-multiplication magnetization of high frequency. It was confirmed that to improve the defect detection performance by increasing the magnetization frequency. The use of non-multiplication magnetization scheme multidirectional coil, it was confirmed that all directions of the defect detection can be facilitated in a non-contact. We confirmed the characteristics of the new proposed method by using the simulation and experiment, and report these results. 1. Magnetization power supply structure Fig.1 shows the structure of the basic circuit in a magnetization power supply. It has individual magnetization power supply and magnetization coils for each direction (x,y,z). The magnetization power supply are capable of individually setting magnetization current frequencies and strengths. The magnetizing currents can be passed for each direction without current phase synchronization. In the effective magnetic field range, complex magnetic fields of various frequencies is generated by coils. License: 1 More info about this article:

2 3Phase Power Sup. Inverter X Axis Mag. coil Current Sensor Frequency & Current Controller Inverter Y Axis Mag. coil Current Sensor Frequency & Current Controller Inverter Z Axis Mag. coil Current Sensor Frequency & Current Controller Fig. 1. Configuration of magnetization power supply. Conventional coil magnetization method.1 Shape of test specimen There is a possibility that the defect detection performance may deteriorate due to the influence of the demagnetizing field in a conventional coil method. When the ratio of length L to diameter of a specimen is small, it is necessary to increase the ratio of L / D such that have a yoke.. Multidirectional magnetization field It is necessary to magnetize to 90 degrees relative to the defect in the magnetic particle testing. When the orientation of the defect is not known, the orientation of the magnetization must be changed. In multiple alternating magnetization using conventional commercial threephase power supply the composite magnetic field of the coil installed in each axis is not uniform in all directions. It is difficult to equalize the multi-directional flaw detection performance in the conventional method.

3 Hy = A sin(ωt) Hy = A sin(ωt) Hx = A sin(ωt + π/3) Hx = A sin(ωt+π/) H Hx Hy A sin ( t) sin ( t / 3) H Hx Hy A Fig.. Magnetic field locus vector diagram Fig.3. Magnetic field locus vector diagram In the conventional method using a commercial three-phase power supply, Fig. shows a composite vector diagram of a magnetic field generated in the orthogonal coil. Conventional magnetization power supply using the phase control method of thyristors. The phase difference of the magnetizing current of each axis varies coil load state. The phase difference may not become 10 degrees. Phase difference as shown in Fig.3 becomes uniform magnetic field in all directions and at 90 degrees, uniformly would be able to detect all directions of the defect. For the whole direction of defect detection, change of power phase, various changed in adjusting the magnitude of the current conditions is necessary. 3. Characteristics of high magnetization frequencies When using the coil magnetization method, defect detection performance suffers for small test specimens with strong demagnetizing fields. The frequency of the generated magnetic field can be raised to concentrate the magnetic flux on the surface of the specimen, improving the surface defect detection performance. Generally, the depth of penetration of the magnetic flux is determined using equation (1). 0 S f (1) δ: Penetration depth [m] μ=μ 0 μ s: Magnetic permeability[h/m] ρ: Resistivity [Ωm] f: Frequency [Hz] Fig.4 shows the penetration depth of the magnetic flux for soft iron material between the frequencies of 5Hz and 300Hz, according to equation (1). Extremely small surface defects have a depth of roughly several hundred μm from the surface of the specimen, so a magnetic flux penetration of 1mm below the surface is sufficient, and therefore we set the maximum magnetization power supply frequency to 300Hz. We compared the surface defect detection performance for magnetization currents with different frequencies. Magnetic particle flaw detection test was carried out under the condition that the magnetic flux density in the coil center is 50Hz and 300Hz same. Table 1 shows the results of the magnetic flux density at each frequency was measured using a Hall element, and Table shows the test conditions. 3

4 Penetration depth [mm] Frequency [Hz] B [mt] 1 1 at Current is 600[A] Table 1. Measurement results of magnetic flux Density Frequency [Hz] Test specimen 60mm cube Material soft iron Test sample JIS A1 Type 15/100 Magnetic particle liquid 1.5 g / liter Fig.4. Penetration depth by equation (1) Table. Test conditions Fig.5-1 shows the result of defect testing when the magnetization current for the test specimen (JIS A1 type standard test piece) was 50[Hz]/600[A], and Fig.5- shows the result when the magnetization current was 300[Hz]/600[A]. Fig.5-1. Testing Result at Frequency 50[Hz] Fig. 5-. Testing Result at Frequency 300[Hz] When the ratio of length L to diameter D of the test specimen is small, the influence of the demagnetizing field makes it difficult to detect defects. We confirmed that improves the performance of the defect detection by increasing the magnetization frequency. It is thought to be due to focus the magnetic flux in the inspected object surface in the skin effect. 4. Defect characteristics by non-multiplication magnetization method 4-1. Basic characteristics Multiple alternating magnetization using the conventional coil method uses commercial power supply, so the composite vectors for each axis are not uniform. This makes it difficult to uniformly detect all orientations of defects. We propose the magnetization method using non-multiplicative frequencies. It is possible to generate a uniform magnetic field in all directions by the proposed method. It will be described in detail below the proposed method. Fig.7 shows a vector locus generated in two coils centered orthogonal with equations () and (3). In Fig.7-A the X axis frequency is 300[Hz] and the Y axis frequency is 30[Hz]. Fig.7- B shows calculation results for doubled frequencies, and Fig.7-C shows calculation results 4

5 for tripled frequencies. Doubling or tripling the frequencies of the axes produces dead zones, making it impossible to perform defect detection for all directions. However, Nonmultiplication magnetization method is the maximum area of the magnetic field is a square, and can eliminate the problem of the magnetic field becoming small for a specific direction. The proposed method is possible to make uniform the magnetic field without relation to the load conditions. Hy = A sin(ωt+δ 1) Hx = A sin(ωt + δ ) H tan 1 Hx Hy A sin ( 1t 1) sin ( t ) sin( 1t 1) sin( t ) () (3) Fig. 6. Structure of X axis, Y axis coils Case A. Non-multiplication Case B. Doubling Case C. Tripling Fig. 7. Coil center magnetic field locus simulation results 4-. Bidirectional magnetic particle testing and results We confirmed the effectiveness of the non-multiplication magnetization by the experiment shown below. Two directions of the coils are constructed to be orthogonal. The strength of the magnetic field in each direction was adjusted to be the same. We were using the 100mm cube of soft iron material as a test object, and performed magnetic particle testing using the following conditions. Case 1. Phase control power source using a commercial frequency Case. Identical frequencies X axis: 50[Hz]/500[A], Y axis: 50[Hz]/600[A] Case 3. Non-multiplication X axis: 300[Hz]/500[A], Y axis: 30[Hz]/600[A] In the experiment using the JIS A1 type 15/100 standard test piece, magnetic particle content is 1.5 [g / l]. Fig. 8-A, 8-B, and 8-C show the results of measurement of the magnetic fields at the centers of the coils under test conditions 1,, and 3, respectively. Measurement was performed using Hall elements, and X-Y waveforms were recorded. When a magnetic field was generated using non-multiplicative frequencies, the coil center magnetic field matched the calculation results shown in Fig.7 Case A. 5

6 Magnetic flux density: 0.01[T] / div A. (Case 1. Phase control) B. (Case. Identical frequency) C. (Case3.Non-multiplication) Fig.8. Magnetic field measurement results Fig.9 A, B, and C show photographs of the results of magnetic particle testing in each of the conditions. A. (Case 1. Phase control) B. (Case. Identical frequency) C. (Case3.Non-multiplication) Fig.9. Magnetic particle testing Results The results of the above test showed that non-multiplicative frequencies could be used to detect the defect in all directions, clearly identifying the circular defect on the standard test piece Tridirectionally magnetic particle testing and results Next, in order to confirm the effectiveness of the non-multiplication magnetization was performed three directions of magnetic particle testing. A JIS A1 type 30/100 standard test pieces were placed at the center of the three plane of the test specimen (100x100x100 cube) as shown in Fig.10. The coil center was aligned with the center of the test specimen. Three directions of the coils are orthogonal with each 90 degrees. Plane 1 Plane 3 Plane Fig.10. Coil magnetic field orientations and test specimen 6

7 Using the technique of the conventional method (identical 50Hz frequencies) and the high non-multiplicative frequencies (130Hz to 300Hz), it was compared by experiments the performance of the magnetic particle testing. These results are shown in Fig.11 and 1. Plane 1 Plane Plane 3 Fig. 11. Magnetic particle testing results (Condition: Conventional method) Plane 1 Plane Plane 3 Fig. 1. Magnetic particle testing results (Condition: High non-multiplicative frequency) The results of the above testing found that the defect detection capabilities of the conventional method suffered from dead zones, and there were defects on all faces of the test specimen which could not be detected. Testers therefore had to use trial and error to adjust the sizes of the magnetization currents and current phases to detect the defects in all directions. However, We proposed the method for magnetizing the specimen by a coil of each axis using high nonmultiplicative frequencies. The proposed method can be uniformly magnetized the entire surface of the specimen, and has the uniform defect detection capability. It is possible to eliminate the dead zone. We confirmed these advantages. Conclusion We have proposed a new method as a magnetization method for performing magnetic particle testing. We confirmed the following characteristics using by experiments and simulations. In the case of small specimens, flaw detection is difficult due to the influence of the demagnetization field. We confirmed to be able to improve the detection performance by increasing the magnetization frequency to small specimens. It is necessary to detect the multidirectional defect in magnetic particle testing. We used the magnetizing current of the nonmultiplicative frequency to the coils in multiple directions. We confirmed that the defects in 7

8 JIS A1 type standard test pieces were clearly detected, and that defects were detected in all orientations. Reference [1]M.Hori The New Magnetization Method by Non-Multiply Magnetization, Journal of JSNDI Vol.58, No.3,