See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/305430906

Coil Design of the Semicirucular ∞ Coils for Practical use

Article · January 2016

DOI: 10.3233/978-1-61499-639-2-293

CITATIONS READS

0 257

6 authors, including:

Iliana Marinova Yoshifuru Saito

Technical University of Sofia Hosei University
91 PUBLICATIONS   219 CITATIONS    472 PUBLICATIONS   5,072 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Electromechanical Systems with Magnetic Modulation View project

Inverse Electromagnetic Problems View project

All content following this page was uploaded by Yoshifuru Saito on 21 August 2018.

The user has requested enhancement of the downloaded file.

 Coil Design of The Semicirucular ∞ Coils
for Practical use
Takahiro HYUGAa,1, Iliana MARINOVAb, Yoshifuru SAITOa,
Manabu OHUCHc, Hideo MOGIc and Yoshiro OIKAWAc
a
Graduate School of Electrical and Electronics Engineering, Hosei University, 3-7-2
Kajino Koganei, 184-8584, Tokyo, Japan
b
Technical University of Sofia, 1756, Sofia, Bulgaria
c
EMIC (Denshijiki Industry) Co, Ltd, 115-0051, Tokyo, Japan

Abstract. Previously we have succeeded in exploiting a new high sensibility eddy

current testing (ECT) sensor called the infinite (∞) coil. Operating principle of this
∞ coil is that two adjacent coils constructing the north and south poles
alternatively set a zero magnetic potential region and keep the same situation when
eddy currents are flowing along the paths in parallel to the exciting coils. If the
eddy currents could not flow the paths in parallel to the exciting coil on the test
specimen, then a zero magnetic potential region between the two exciting coils
moves toward the other positions. This means that a sensing coil located at a zero
magnetic potential region between the two exciting coils is possible to detect the
disturbed magnetic fields, i.e., the defect in the test specimen could be detected
from sensing coil signal. In this paper, we improve the ∞ coil in an optimum
manner which leads to a practical factory product. Employment of the two planar
semicircular exciting coils has made it possible to enlarge the zero magnetic
potential region. Further a planar rectangular shape exciting coils may be
considered as a deterministic design from the viewpoints of practical production.

Keywords. nondestructive testing, high sensibility ECT, practical design

Introduction

Most of the industrial products and infrastructures are composed of the mechanical
frames, and they support our modern civilianized human life. Each of the machines is
composed of the mechanical parts to support its mechanical strength as well as their
shapes. Major frame materials are the pure metallic or their compound, and are always
stressed to maintain the highly civilized modern human life. The stress is classified into
mechanical, thermal and so on, which lead to cause of fatigue and defect. To keep the
highest safety and reliability, various maintenance works to the machines as well as
infrastructures are continuously required as one of the most important tasks. To
evaluate the metal fatigue, one of the most fundamental and important methodologies is
the nondestructive testing to the metallic materials [1].
The nondestructive testing makes it possible to evaluate the defect size, shape and
situations without any destroying and decomposing the products. Further, we have
___________________________________
1
Corresponding author, E-mail: takahiro.hyuga.8k@stu.hosei.ac.jp
c/o Y.Saito Lab. of The Graduate School of Electrical Engineering, Hosei University, 3-7-2 Kajino-cho,
Koganei, Tokyo 184-8584, Japan
several nondestructive methods such as the eddy current testing (ECT), electrical
potential method, ultrasonic imaging, x-ray tomography and so on [2]. The ECT is
composed of the relatively simple parts and needs not the direct contact to the tested
specimen. Furthermore it is possible to work in a relatively high speed, even though
sometimes it is difficult to detect the defects in the target metallic object depending on
the eddy current flowing direction. However, the ECT has versatile capabilities as the
nondestructive testing method.
Previously, we had exploited a new highly sensitive sensing coil type sensor called
the ∞ coil. Principal purpose of this paper is to design the ∞ coil in the high
predictability and factory productivity by means of the intensive three dimensional
finite element (3D FEM) simulations along with the practical experiments.
In our numerical simulations, we have assumed a line defect searching problem.
Namely, our design target of the sensor is how to enhance the catching ability of the
magnetic flux caused by the detouring eddy currents of this line defect.
A ∞ coil having the planar semicircular exciting coils is a good coil design by
comparing the results of 3D FEM simulations along with the practical experiments.
Examine the results obtained by 3D FEM simulations along with the practical
experiments with more carefully suggests to a ∞ coil having the planar rectangular
exciting coils as the best ∞ coil for the high predictability and factory productivity.

1. Fundamentals of the ∞ coil

Figure 1 shows a conventional planar ∞ coil which is composed of two planar spiral
exciting coils and one sensing coil wound around ferrite bar. According to the shape of
the two exacting coils, we had named this sensor as “the flat/film ∞ coil” or simply “∞
coil” [3,4].

Figure 1. Schematic diagram of the ∞ coil, where circular coils on both side are the planar exciting coils. The
red rectangular box is the sensing coil wound around a ferrite bar. The light blue arrows show the magnetic
fluxes caused by the exciting currents when no target metallic materials.

1.1. Operating principle of the ∞ coil

When an alternating exciting current flows through the spiral exciting coils in an
opposite direction, both two exciting coil construct the north and south magnetic poles,
alternatively. Thereby, as shown in Figure 1, a magnetic flux density flux flows
throughout both of the two spiral exciting coils.
Observe the intensity of exciting magnetic flux distributions shown in Figure 2
reveals that there is the zero magnetic potential region between the two adjacent
exciting coils. When we put on the ferromagnetic materials into this zero magnetic
potential region, i.e., a space between the two adjacent exciting coils, no magnetic field
disturbance is occurred because of the zero magnetic potential. Thereby, it is possible
to set the sensing coil, which is a receiving coil catching the disturbed magnetic fields
caused by the defect detouring eddy currents, wound around a ferrite bar in the zero
magnetic potential region between the two adjacent exciting coils shown in Figure 2.

Figure 2. The magnetic flux density distribution caused by the ∞ coil (simulation).

Even if the exciting magnetic flux passes through this zero magnetic potential
region, the surface of sensing coil is directed in parallel to this magnetic flux. Thereby,
any voltage is not induced in the sensing coil. If we put the ∞ coil on a surface of a
target metal having no defect, then the eddy currents flow as the image currents
flowing in an opposite direction to the exciting currents. When there is no defect in a
target object, the magnetic fields caused by the eddy currents never induce the voltage
in the sensing coil. The reason why no induced voltage in the sensing coil is that any of
the magnetic fields are directed in a parallel direction to the sensing coil surface.
However, if there is a defect in the metallic target object, the detour eddy currents
essentially flow along with the defect. Thus, the magnetic flux caused by detour eddy
currents induces a voltage in the sensing coil, and this induced voltage reveals the
defect in the target object [4].

1.2. Simulations of the ∞ coil

To show the operation principle of the ∞ coil, we carried out the three dimensional
finite elements (3D FEM) simulations. Table1 lists various constants used for the
simulations.
Specification of the simulation model is as follows: A test ∞ coil is located in
parallel to the copper test plate whose dimensions are 100×100×1mm3. A line defect is
emulated by a 2×60mm2 slit, and this slit shape defect is set to 0, 45, 90 degrees to the
sensing coil axis, where 0.2mm lift-off distance is assumed. In addition to the Table.1,
the 300(rms)mA 256 kHz sinusoidal exciting current is flowing thorough the two spiral
exciting coils.
 Table. 1 Various constants of the simulation model.
Exciting coil Sensing coil
Outer diameter 24.0 [mm] Outer diameter 1.4 × 2.4 [mm]
Inner diameter 8.0 [mm] Inner diameter 1.4 × 2.0 [mm]
Length 0.4 [mm] Length 6 [mm]
Number of turns 20 turn Number of turns 100 turn
Input voltage(peak) 1 [V] Axis core Mn-Zn/ferrite
Frequency 256 [kHz] (Permeability: 3000)
Figures 3 and 4 show the eddy current distributions on the target copper plate and
the magnetic flux density distributions in the ferrite core, respectively.
Figure 3(a) shows the eddy current density distributions when no defect in the
target copper plate. Figure 4(a) shows a flux density distribution in the sensing ferrite
core. The flux densities caused by the eddy currents are flowing along with the sensor
coil surface in parallel. This means no induced voltage in the sensor coil when no
defect in the target copper plate. Figure 3(b) shows the eddy current distribution when
the line defect is located in parallel to the sensor coil axis. The eddy currents flow
along with the defect so that all of the magnetic flux caused by the eddy currents flow
in a parallel direction to the sensor coil surface. Figure 4(b) shows the flux density
distributions in the ferrite core of the sensor coil. Figure 3(c) shows the eddy current
distributions when the line defect is located 45 degree to the sensor coil axis. The eddy
currents flow along the line defect so that these eddy currents yield the perpendicular
magnetic flux component to the sensor coil surface. Figure 4(c) shows the magnetic
flux densities in the ferrite core directed 45 degree to the ferrite core axis. Thereby, the
perpendicular magnetic flux component to the sensor coil surface induces the output
voltage in the sensor coil. Figure 3(d) shows the eddy currents flowing the detour paths
to the defect when the line defect is located in a perpendicular to the sensor coil axis.
Because of the flowing direction of the eddy currents, the magnetic flux flows in a
perpendicular direction to the sensor coil surface. Although all of the magnetic flux
caused by eddy currents are directed to the perpendicular to the sensor coil surface,
they are cancelled each other. This means that it is difficult to detect the line defect
under this condition.

(b) The line defect is located in parallel to

(a) No defect the sensor coil axis.

(c) The line defect is located 45 degree to (d) The line defect is located in 90 degree
the sensor coil axis. to the sensor coil axis.
Figure 3. Eddy current distributions in various defect conditions.
 (a) No defect (b) 0 degree (c) 45 degree (d) 90 degree
Figure 4. The magnetic flux density distributions in the ferrite core.

Figure 5 shows the induced voltage waveforms when the line defect is directed 0,
45, 90 degrees to the sensor coil axis. From the above results, it is found that the sensor
coil could be induced the voltage when the detoured eddy current causes the
perpendicular component of the magnetic flux to the sensor coil surface. The induced
voltage in the sensor coil depends on the perpendicular magnetic flux component to the
sensor coil surface so that the ∞ coil is one of the directional sensors.

Figure 5. The induced voltages whose magnitude depends on

the angle between the sensor coil axis and the magnetic flux
flowing directions

2. Improved planar ∞ coils

The operating principle is summarized into the two key ideas. The first is that there is
zero magnetic potential region between the north and south poles when both of the
north and south poles are constructed by the exciting coils. When this zero magnetic
potential region becomes a non zero magnetic potential, the defect is caught by the ∞
coil. This is a fundamental qualitative function of the ∞ coil, i.e. this is the same to the
on/off function of a switch. The second is that the magnetic flux is passing through the
zero magnetic potential region from the north to south poles. This magnetic flux is
always directed in parallel to the sensor coil surface so that never induce the voltage in
the sensor coil. If the direction of this magnetic flux flow to the ferrite core axis is
changed within 90 degree, then the changed angle is mostly proportional to the induced
voltage in the sensor coil. In the other words, the directional change of the magnetic
flux is the quantitative function of the ∞ coil, i.e. the directional change within 90
degree is mostly proportional to the defect magnitude.
In the present paper, we propose the exciting coils having the semicircular and
rectangular shape to enlarge the zero magnetic potential region between the north and
 south poles. Both of the semicircular and rectangular shapes exciting coils construct the
north and south poles alternatively by flowing the alternative exciting currents. A
region between the two adjacent straight line of the two shapes of the planar exciting
coils construct a line shape zero magnetic potential region. The planar semicircular and
rectangular ∞ coils are composed of a pair of semicircular and rectangular exciting
coils, respectively. Both of the semicircular and rectangular ∞ coils are also composed
of the sensing coil wound around a ferrite core as shown in Figure 6.

(a) Semicircular type. (b) Rectangular type.

Figure 6. Proposed new planar type ∞ coils.

2.1. Simulation of the planar semicircular ∞ coil

To verify the operation of the planar semicircular/rectangular ∞ coil, we carried out the
3D FEM simulations. Table.2 lists various constants of the exciting and the sensing
coils. Specification of the simulation model is as follows: The planar
semicircular/rectangular ∞ coil is located in parallel to the copper test plate whose
dimension is 100×100×1mm3. A line defect is emulated by a 2×60mm2 slit, and this slit
shape defect is set to 0, 45, 90 degrees to the sensing coil axis, where 0.2mm lift-off
distance is assumed.

Table. 2 Various constants of the planar semicircular/rectangular ∞ coil.

Semicircular exciting coils Rectangular exciting coils Sensing coil
Outer diameter 36.0 [mm] Outer rectangle 36.0×18.0 [mm] Outer diameter 1.4 × 2.4 [mm]
Inner diameter 12.0 [mm] Inner rectangle 12.0×6.0 [mm] Inner diameter 1.4 × 2.0 [mm]
Coil length 0.3 [mm] Coil length 0.3 [mm] Length 6 [mm]
Number of turns 20 turn Number of turns 20 turn Number of turns 100 turn
Input voltage(peak) 1 [V] Input voltage(peak) 1 [V] Axis core Mn-Zn/ferrite
Frequency 256 [kHz] Frequency 256 [kHz] (Permeability: 3000)

Induced Voltage V Induced Voltage V

3 3

2 2

1 1
Time s 6 6 6 Time s
2. 10 6
4. 10 6
6. 10 6
8. 10 6 2. 10 4. 10 6. 10 8. 10 6
1 1
2 2
3 3
(a) The output signals of the planar semicircular ∞ coil. (b) The output signals of the planar rectangular ∞ coil.
Figure 7. The output signals whose magnitude depend on the angle between the sensor coil axis and the magnetic flux
flow direction in the sensing coil.
 Figure 7 shows an induced voltage waveform when the line defect is directed 45 degree
to the sensor coil axis. The planar semicircular/rectangular ∞ coil yields a higher sensor
induced voltage compared with those of conventional one. Also, the new
semicircular/rectangular ∞ coil yields a lower sensor induced voltage compared with
those of conventional one when no detecting the defect. Consequently, comparison of
the signal to noise ratio between the planar semicircular/rectangular ∞ and
conventional ∞ coils suggests that the planar semicircular ∞ coil proposed here has a
higher sensibility than those of conventional one.

2.2. Comparison of the simulated and experimented results

The tested copper plate has a 1×1mm2 line defect. Table.3 lists various constants of the
tested ∞ coils. The peak excitation voltage was 1V and 256kHz in every tested sensors.
Figure 8 shows the prototype of ∞ coils.
Table. 3 Various constants of the tested ∞ coils
(a) Circular exciting (b) Semicircular exciting
(c) Rectangular exciting coil. (d) sensing coil
coil. coil.
Coil outer Outer coil Outer coil 36.0×18.0 Outer 1.4×2.4
9.0 mm 18 mm
diameter diameter diameter mm diameter mm
Coil Inner Inner coil Inner coil 12.0× 6.0 Inner 1.0×2.0
0.5 mm 1 mm
diameter diameter diameter mm diameter mm
Coil length 0.3 mm Coil length 0.3 mm Coil length 0.3 mm Coil length 0.1 mm
Number of Number of Number of Number of
20 20 20 100
turns turns turns turns
Input voltage Input voltage Input voltage Mn-Zn/
1V 1V 1V Axis core
(peak) (peak) (peak) ferrite
Frequency 256kHz Frequency 256kHz Frequency 256kHz (permeability:3000)

(a) Planar circular exciting coils. (b) Planar semicircular exciting coils.

(c) Planar rectangular exciting coils. (d) Sensing coil.

Figure 8. The pictures of tested ∞ coils, the exciting coils and sensing coil wound around a ferrite bar, respectively.
 Figure 9 shows the experimentally obtained induced signal voltage waveforms of
the conventional, planar semicircular, and rectangular ∞ coils. Table 4 summarizes the
simulated and experimented results of the conventional, semicircular, and rectangular
∞ coils.
Induced Voltage mV
150
100
50
Time s
2. 10 6 4. 10 6 6. 10 6 8. 10 6
50
100
150
(a) Conventional ∞ coil (b) Semicircular ∞ coil. (c) Rectangular ∞ coil.
Figure 9. Comparison among the experimented output signal voltages of the conventional, planar semicircular and
rectangular ∞ coils.

Table.4 The experimented S/N ratios and the peak induced signal voltages.

Conventional ∞ coil Semicircular ∞ coil Rectangular ∞ coil

Simulation Experiment Simulation Experiment Simulation Experiment

No-defect 0.042 0.022 0.028 0.022 0.014 0.018
45° 2.387 0.100 2.896 0.156 2.765 0.160
S/N ratio 56.65 4.55 121.7 7.09 197.5 8.89
Even though the S/N ratio of rectangular ∞ coil takes the best value "197.5" in the
simulations, it has been revealed that the semicircular ∞ coil takes a higher sensor
sensibility compared with that of the rectangular and of the conventional ∞ coils.
Because of the poor skill to work out the tested coils in the laboratory, the large
differences between the simulated and experimented results in the conventional,
semicircular and rectangular ∞ coils could be found in Table.4. The differences
between the conventional and its experimented values may be smaller than that of
others when using the professional works of the ∞ coils.
Thus, if we can construct the practical planar rectangular ∞ coil with professional
skill, then the planar rectangular ∞ coil may take the much better practical S/N ratio.

3. Conclusion

As shown above, we have proposed the new semicircular and rectangular ∞ coil in the
present paper. Employment of the two planar semicircular and rectangular exciting
coils has made it possible to enlarge the zero magnetic potential region. Thereby, it has
been possible to realize the new high sensitive ECT sensor. Further, the results of the
3D FEM simulations and of the practical experiments have verified that the new
developed semicircular ∞ coil yields the higher S/N ratio compared with that of
rectangular ∞ coil and of conventional ∞ coil.
Thus, the ∞ coil having the planar semicircular exciting coils has been the best coil
design by comparing the results obtained by 3D FEM simulation along with the
practical experiments. Examine the results obtained by 3D FEM simulations along with
 the practical experiments with the highest carefully has suggested to the ∞ coil having
the planar rectangular exciting coils as the best ∞ coil for practical factory product.

References

[1] Narishige,S., Nishimizu,A., Koike,M. Abe, M., Narumi, Y, Tsuge, T, (2010), Improved Detection
Performance using a Multi-frequency Method of Eddy Current Testing for Outside Circumferential
Cracks near an Expansion of Heat Exchanger Tubes, Non-Destructive Testing and Condition
Monitoring, Vol. 52(6), June 2010, pp. 298-301.
[2] Hashizume, H., Yamada, Y., Miya, K., Toda, S., Morimoto, K., Araki, Y., Satake, K. and Shimizu, N.,
(1992), Numerical and Experimental Analysis of Eddy Current Testing for a Tube with Cracks, 1992,
Vol.28, pp.1469 - 1472.
[3] Kikuchihara, H. and Saito,Y., (2013), Enhance the Sensibility of the ECT Sensor, Journal of the Japan
Society of Applied Electromagnetics and Mechanics, Vol.21(3), 2013.
[4] Maruyama,K., Marinova,I., and Saito,Y., (2015), Developments of Flat ∞ Coil for Defect Searching in
the Curved Surfaces, E-Journal of Advanced Maintenance, (EJAM), Vol.7-1, 2015, pp. 59-65.

View publication stats