PROCEEDINGS OF SPIE

[Link]/conference-proceedings-of-spie

Design and simulation analysis of a

magnetic shielding box for ring laser
gyroscope

Rong, Chui-yu, Hu, Shao-min, Han, Xiang

Chui-yu Rong, Shao-min Hu, Xiang Han, "Design and simulation analysis of a
magnetic shielding box for ring laser gyroscope," Proc. SPIE 8905,
International Symposium on Photoelectronic Detection and Imaging 2013:
Laser Sensing and Imaging and Applications, 890514 (19 September 2013);
doi: 10.1117/12.2033520

Event: ISPDI 2013 - Fifth International Symposium on Photoelectronic

Detection and Imaging, 2013, Beijing, China

Downloaded From: [Link] on 23 Jan 2022 Terms of Use: [Link]

 Design and simulation analysis of a magnetic shielding box for ring
laser gyroscope

Chui-yu Rong, Shao-min Hu, Xiang Han

College of Opto-electric Science and Engineering, National University of Defense Technology,

Changsha, 410073, China

ABSTRACT
Magnet outside is one of the main reasons that induce the bias shift of the Ring Laser Gyroscope (RLG), and the
most effective way to handle it so far, is to employ a magnetic shielding device. In this paper, we design a new magnetic
shielding box for RLG in Fe-Ni soft magnetic alloy. By 3-D modeling and simulation calculation in finite element
method, we plot the distribution of magnetic flux density of the box in uniform magnet; Besides, distribution of magnetic
shielding effectiveness (SE) on the gaining plane of RLG where is most sensitive to the change of magnet outside, is
given out, too. We also discuss the factors that influence the distribution of SE such as permeability of shielding material,
the thickness of the box and punching on the surfaces of the box. And by the SE test of the box samples, the result of the
simulation analysis was finally proved true. The result shows that, the structure of the box decides the distribution of SE
in it and the permeability of the material plays the predominant role in evaluating the average value of SE. Punching on
the surfaces of the box makes the distribution of SE severely inhomogeneous, and with increase of the permeability, the
influence will be more significant. At last, we put some practical steps forward to enlarge the average value of SE, and
these measures are also applicable in other similar magnetic shielding devices.

Keywords: Laser optics, Ring Laser Gyroscope, Magnetic shielding box, Finite element analysis, Shielding

effectiveness

1. INTRODUCTION
As an ideal component of strap down inertial navigation system (INS), RLG has been widely used in navigation,
positioning and many other fields. The precision of RLG is not only influenced by the manufacturing technique, its inner
structure and some other elements by itself, but also by the environment factors outside. Among these, magnetic field is
an important one. It is acknowledged that, in a gyro with an ideal parallel-plane resonator, laser runs within the plane.
However, the machining errors and cavity-adjusting errors make the polarization of the light a little elliptic. When
magnetic field outside is applied on the gyro, the Longitudinal Zeeman Effect which causes large magnetic bias of gyro
takes place. 1 The result of experiment shows that, for a single two-frequency gyro with linearly polarized light in it,
when we apply the magnetic field along the transvers direction, the magnetic bias of the gyro is about 1(°/h)/Gs; and for
a gyro with rotatory polarized light, the value will be much larger. 2 So, in practical application, a gyro must be equipped

International Symposium on Photoelectronic Detection and Imaging 2013: Laser Sensing and Imaging and Applications,
edited by Farzin Amzajerdian, Astrid Aksnes, Weibiao Chen, Chunqing Gao, Yongchao Zheng, Cheng Wang,
Proc. of SPIE Vol. 8905, 890514 · © 2013 SPIE · CCC code: 0277-786X/13/$18 · doi: 10.1117/12.2033520

Proc. of SPIE Vol. 8905 890514-1

Downloaded From: [Link] on 23 Jan 2022
Terms of Use: [Link]
 with a magnetic shielding device.
In this article, we design a magnetic shielding box for RLG which is based on Fe-Ni soft alloy. By 3-D modeling
and simulation calculation in finite element method, we discuss that, to what extent the common factors during the
designing such as the permeability, the thickness of the material and punching on the surfaces of the box, affect the SE in
a box. Base on the SE tests of the box samples, we compare the results of the actual measurements with the simulated
ones, and some useful conclusions have been drawn.
2. BASIC DESIGNING SCHEME
The magnetic bias of RLG is originated from the coactions of high-frequency AC electromagnetic field (EMF) and
low-frequency DC one. Because the errors generated by the high-frequency AC EMF is comparatively small and the net
result almost equals to zero, the magnetic bias is mainly attributed to the low-frequency DC EMF. Generally speaking,
magnetic shielding of RLG requires weak magnetic material which is with high magnetic sensitivity, low magnetic
resistance and mainly applied in low-frequency EMF. So, we choose Fe-Ni soft magnetic alloy with high initial
permeability as the material of our shielding box.
By comprehensive analysis of the actual situation and the requirement above, we propose our design scheme as is
shown in Figure 1.

Figure 1. Design of A Magnetic Shielding Box for RLG

In our design, the magnetic shielding box of RLG is a cuboid which can be divided into two parts: the body and the
lid. On the front surface of the body, there is an opening for easily connecting to peripheral circuit; on the bottom of the
box, three square holes are designed to fix the gyro; on the front and back surfaces of the lid, four round threaded holes
are punched in order to hold the lid and body together.
3. 3-D MODELING IN ANSYS
The model which needs to be built in ANSYS chiefly comprises two parts: the box and the Helmholtz coil. The
model-building of the box is comparatively easy, and it is just done by drawing the model in SolidWorks and importing it
to ANSYS.
As far as the Helmholtz coil concerned, it consists of two single coils with equal diameter, and the space between
the coils is the same as the diameter. The outer diameter of the Helmholtz coil we used is 36cm; the inner one is 30cm;
the space between the coils is 36cm and the number of turns is 194. When we electrify the Helmholtz coil, 20Gs uniform
magnet can be generated in its center. Figure 2 shows the model of the Helmholtz coil in ANSYS.

Proc. of SPIE Vol. 8905 890514-2

Downloaded From: [Link] on 23 Jan 2022
Terms of Use: [Link]
 Figure 2. Build the Model of the Helmholtz Coil in ANSYS
We combine the two models together, and build the cuboid model of air outside. Thus, the whole simulation model
is finished. Figure 3 shows the complete model for simulation.

Figure 3. The Complete Model for Simulation

4. ANALYSES OF THE SIMULATION

There are two methods to analyze EMF problems in ANSYS: the node and the edge. To a 3-D model which
contains permeability magnetic material, the edge method is our preferred. In solving the 3-D low-frequency static and
dynamic EMF problem, the edge method is more accurate. 3
4.1 Distribution of Magnetic Flux Density in the Shielding Box
According to the Finite Element Method (FEM), we mesh the whole model and apply excitation of current density
to the Helmholtz coil. Then we solve it and plot the distribution of magnetic flux density in the shielding box. The result
is shown in Figure 4.

Proc. of SPIE Vol. 8905 890514-3

Downloaded From: [Link] on 23 Jan 2022
Terms of Use: [Link]
 -- TAN 15 2013
16:11:01

211915 .62202 .935745 [Link]

.155950 .467672
Flle: .... \\magnet shielding box(new).x t

Figure 4. Distribution of Magnetic Flux Density in the Magnetic Shielding Box

Figure 4 shows that, on the edge of the front surface of the box, magnetic flux density B varies dramatically; and
around the three square holes on the bottom surface, the flux leakage is also observed easily. Like these places where the
magnetic line of force isn’t continuous, the reluctance is larger than other places and it will cause the decrease of SE.
4.2 Influence on SE of the Box in Different Magnetic Shielding Materials
On the grounds of the preceding analysis we have made, we choose four common Fe-Ni soft magnetic alloys—1J50,
1J79, 1J80, 1J85 as the magnetic shielding materials of the box. The specific parameters of the four alloys are listed in
Table 1.
Table 1 DC Magnetic Performance of Four Common Fe-Ni Soft Magnetic Alloys 4
Thickness Resistivity Permeability Saturation Flux
Grade
(mm) (μΩ·m) (mH/m) Density (T)
1J50 0.35~0.50 0.45 5.0 1.5
1J79 0.35~1.00 0.55 31 0.75
1J80 0.35~0.50 0.62 44 0.65
1J85 0.35~1.00 0.56 62.5 0.70
Actually, the capillary of RLG is nearly on the middle plane of the shielding box in X direction. Here, X is the
normal direction of the top and bottom surfaces of the box. For convenience’s sake, we take this middle plane as our
object of study. On that plane, firstly we choose 17 paths, PATH1~PATH17, which are symmetrical and evenly spaced.
After that, mapping the sum of flux vectors BSUM to every path, we gain the distribution of magnetic flux density
BSUM (x, y) which is on the middle plane of the box in X direction.
Similarly, when unshielded, we take PATH1~PATH17 in the same position, and we can get the unscreened
distribution of magnetic flux density BSUM’(x, y). Therefore, the SE of the box is considered approximately as

BSUM '( x, y )
SE = (1)
BSUM ( x, y )
According to Eq.(1), we plot the distribution of SE for the box in the four different materials. The results are shown
in Figure 5.

Proc. of SPIE Vol. 8905 890514-4

Downloaded From: [Link] on 23 Jan 2022
Terms of Use: [Link]
 (a)

I:::2872

I 30.08
3310

(c) (d)

278.6

I:::2820

Figure 5. Distribution of SE for the Box in Different Materials. (a) 1J50; (b) 1J79; (c) 1J80; (d) 1J85
From Figure 5, the value of SE reaches maximum at the four corners in the box firstly; then, it decreases rapidly
towards the center of the box and finally forms a flat plane. This plane stands for the average standard of SE in the box.
With the increasing of permeability of shielding materials, the average SE in the box has improved significantly. Before
the permeability approaches maximum in its DC state, the SE of the material increases with the permeability. 5 So, we
can see that, the permeability plays a crucial role in deciding the average SE of the box.
4.3 Influence on SE of the Box with Different Thickness of the Shielding Materials
Taking 1J79 for example, we choose 0.3mm and 0.5mm as the thickness of the materials. Next, solve the model we
made in ANSYS and plot the distribution of SE. Figure 6 shows the results.

(a) (b)

85 20

Figure 6. Distribution of SE in the Boxes with Different Thickness of the Shielding Materials. (a) 0.3mm; (b) 0.5mm

Proc. of SPIE Vol. 8905 890514-5

Downloaded From: [Link] on 23 Jan 2022
Terms of Use: [Link]
 From Figure 6, the thickness of the materials increases from 0.3mm to 0.5mm, SE increases about 46%. Generally,
when magneto motive force of a magnetic circuit is fixed, the reluctance decreases with the enlargement of the thickness
of the material. Thus, more magnetic fields can be bypassed and SE can be improved. However, increasing the thickness
of the shielding material blindly will cause unnecessary waste. So, during the design of the magnetic shielding box, the
total mass and cost of the INS should be taken into consideration.
4.4 Influence on SE of the Box with Holes on Its Surfaces
There are four holes on surfaces of the shielding box, one on the front surface for cable connection and the other
three on the bottom surface for RLG positioning. The size of the cable connection hole is 38.2mm×4.2mm and that of
the positioning holes is 17.3mm×13.1mm. We still take 1J79 for example, plot the distribution of SE with unpunched
situation, punching on the front surface only and punching on the bottom surface only. The results are shown in Figure 7.

=0.000 (b)

S'A

(c)

roo ,zozo

Figure 7. Distribution of SE in the Box with Holes on Its Surfaces. (a) With No Hole; (b) Only with A Hole on the Front Surface; (c)
Only with Three Holes on the Bottom Surface
In Figure 7, from a general view, the hole on the front surface causes a decrease of SE of the box by 2.43%; the
other three holes on the bottom surface cause a decrease of SE by 1.94%. Compared with Figure 6(b), four holes cause a
total decrease of SE by 3.40%. But if we take a view from the depression degree of the SE surface, at the deepest point,
the SE of the box decreases by 60%. This indicates that, although punching on the surfaces of the box has little influence
on the average SE, it leads to a severe uneven distribution of the SE in space.
5. EXPERIMENT
In an effort to testify the magnetic shielding performance of the box which we have designed before, we select five
samples of this box at random for SE tests. The size of the samples is 146mm×124mm×44mm; the thickness of the

Proc. of SPIE Vol. 8905 890514-6

Downloaded From: [Link] on 23 Jan 2022
Terms of Use: [Link]
 shielding material is 0.5mm and the magnetic field generated by the Helmholtz coil is 20Gs. The results of the SE tests
are listed in Table 2.
Table 2 SE Tests Results of the Box Samples
NO. of Type of X-Component Y-Component Z-Component Total Average
Samples Material of SE of SE of SE SE
#1 1J50 23.0 208.2 74.7 93.7
#2 1J50 21.0 169.2 76.9 86.0
#3 1J50 23.7 144.9 73.9 79.3
#4 1J50 23.9 211.6 85.0 102.7
#5 1J50 24.5 163.3 68.4 82.3
From Table 2, the total average SE of the samples is about 80~100 and its Y component is about 140~210. We do
care about the Y component of SE because the magnetic field in Y direction is parallel to the capillary of the gyro which
affects the magnetic bias of the gyro greatly.
After that, we measure the magnetic flux density of 18 points which are symmetrical and evenly spaced in one of
the sample box. By calculating their SE in the magnetic field of 20Gs, we plot the contour maps of SE based on the data
of simulation and the actual measurements. Figure 8 shows the two maps by comparison.

(a) (b)
60

80

29 00

60]I 1 \ I / I 00 90
E É
E 50
.52.40
N 01 N
-20 ar

i,-
20

60
0 -20 o 20 20 40 60 80 100

Figure 8. Distribution of SE in A Box of 1J50. (a) Based On the Data of Simulation; (b) Based On the Data of Actual Measurements
Comparing with Figure 8(a) and Figure 8(b), we can see that the distributions of SE in the two charts are almost the
same and the influence caused by the holes on the bottom surface can easily be identified. During the measurements, the
permeability of the shielding material increases in 20Gs according to its B-H curve, so the SE based on the actual data is
larger as a whole.
Generally, an INS is equipped with multilayer magnetic shielding materials. Each of them is with different initial
permeability and designed for magnetic shielding of different magnetic field intensity and working frequency. In terms of
the magnetic shielding box, good agreement between the measured and simulated results has been achieved, which
demonstrates that the box we designed satisfies the demand of magnetic shielding of RLG.
6. CONCLUSION
Synthesize the analyses above, in design of magnetic shielding box of RLG, our conclusion has three.
(1) For a magnetic shielding box of RLG, with the same structure and magnetic shield intensity, the permeability of
the shielding material plays a decisive role in distribution of SE in the box.

Proc. of SPIE Vol. 8905 890514-7

Downloaded From: [Link] on 23 Jan 2022
Terms of Use: [Link]
 (2) After the selection of the magnetic shielding material is done, we should make a trade off among the
performance indicators, the mass and the total cost of the box in deciding the thickness of the shielding
material.
(3) Punching on the box leads to an uneven distribution of SE in the box, and it’s going to be more distinct with the
increasing of permeability of the shielding material. Therefore, when we are designing a magnetic shielding
box of RLG, we should reduce the size of the holes as much as possible; the holes on the bottom surface should
be near to the side of the box and be far away from the gaining area of RLG. Make sure that the body of RLG
locates in the flat plane that is shown in Figure 5, which will reduce the unexpected errors caused by the uneven
distribution of SE and diminish the magnetic bias of RLG.
ACKNOWLEDGEMENTS
We gratefully acknowledge the technical and financial support from the Opto-Electronic Engineering Department.
REFERENCES
[1] Dong Hongchen, Wang Ke. “Research on the Magnetic Characteristic of Ring Laser Gryscope”, Navigation and
Control, 8(4), 50-54 (2009)
[2] Chen Linfeng, Han Zonghu, Chen Yong, et al. “Study on Characteristic of Ring Laser Gyroscope Sensitive to
Magnetic Fields”, ACTA PHOTONICA SINICA, 35(7), 974-976 (2006)
[3] Sun Mingli, Hu Renxi, Cui Hairong, et al. [Guidance with Examples of Finite Element Analysis in ANSYS 10.0
on Electromagnetics], China Machine Press, Beijing, 2-5 (2007)
[4] China National Standard, GBn 198-88, Technical Condition of Fe-Ni Soft Magnetic Alloy, 1988
[5] Zhao Qibo, Yao Lahong, Wang Xiangxin. “Influence on Magnetic Shielding Effectiveness of Permeability and
Thickness”, Wuhan Vocational College of Engineering and Technology, 19(2), 14-16 (2008)

Proc. of SPIE Vol. 8905 890514-8

Downloaded From: [Link] on 23 Jan 2022
Terms of Use: [Link]