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

net/publication/337819219

Measure of the Shielding Effectiveness in coaxial cables

Chapter · December 2019

CITATIONS READS

0 324

5 authors, including:

Kenedy Marconi Arnaud Colin

Instituto Federal de Educação, Ciência e Tecnologia da Bahia (IFBA) EMCOL Solutions
65 PUBLICATIONS 118 CITATIONS 18 PUBLICATIONS 36 CITATIONS

SEE PROFILE SEE PROFILE

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

Shielding effectiveness View project

EMC simulation View project

All content following this page was uploaded by Kenedy Marconi on 07 March 2022.

The user has requested enhancement of the downloaded file.

 ISEF 2009 - XIV International Symposium on Electromagnetic Fields
in Mechatronics, Electrical and Electronic Engineering
Arras, France, September 10-12, 2009

Measure of the Shielding Effectiveness in coaxial cables

Kenedy Marconi Geraldo dos Santos, Arnaud Colin, Mohamed Ramdani, José Osvaldo
Saldanha Paulino, Leonardo Alvarenga Lopes Santos and Claudio H. Gomes Santos.

Universidade Federal de Minas Gerais

Av. Antonio Carlos 6627 - Cep 31270-901 - Belo Horizonte - BRASIL
École Supérieure d’Électronique de l’Ouest
4, rue Merlet de la Boulaye - BP30926 - 49000 - Angers - FRANCE
[email protected], [email protected], [email protected],
[email protected],
[email protected], [email protected]

Abstract – The coaxial cables’ shielding effectiveness is an important parameter to guarantee the
electromagnetic compatibility of the electro-electronic systems. This concern involves the signal quality in the
cable and the correct functioning of the nearby equipments. It would be interesting to detect such a failure as
soon as possible in order to conceive better systems and avoid unnecessary costs. Within this context, this work
presents a simplified methodology to validate a virtual shielding effectiveness measurement in any coaxial
cables, through a comparison with a physical experiment.

Introduction

The Electromagnetic Compatibility (EMC) is the electro-electronic equipment capacity to operate

properly in a determined electromagnetic environment, without perturbing the adjacent equipments
and without being perturbed by other equipments or by itself [1]. For minimizing EMC problems, it is
possible to act on the source, on the receiver or on the coupling environment. See Fig. 1:

Noise Source Coupling Receiver (Victim)

Fig.1 – Block diagram referent to the interference generation, transmission and victim

Noise source – These sources can be transmitters, motors, computers, rays or any equipment capable
of irradiating electromagnetic energy;
Coupling environment – The coupling can be realized through conductor, inductive or capacitive
material, etc.
Receiver – The victims can be persons, computers, electronic centrals, etc.

In this study, the coupling environment is the element which will be investigated in order to reduce the
irradiated and inducted electromagnetic interferences. One of the techniques used to minimize these
interferences is the coaxial cables’ shielding.

Problem description

A coaxial cable with low shielding effectiveness can present some problems when used in the
neighborhood of sensitive equipments or in an electromagnetic polluted environment.
In a vehicle, if conducting lines are installed close to a high tension inductor or near an electric motor,
they become susceptible to receive the noises emitted from these components. Such disturbs can
generate signal frequency and amplitude variation within the conductor, which might cause errors in
the functioning of electronic centrals connected to it. This paper proposes a method to determine if it is
possible to calculate virtually a coaxial cable’s shielding effectiveness.
 ISEF 2009 - XIV International Symposium on Electromagnetic Fields
in Mechatronics, Electrical and Electronic Engineering
Arras, France, September 10-12, 2009

Shielding Effectiveness

The Shielding Effectiveness (SE) is the attenuation defined as the quotient, in decibels, between the
electric (or magnetic) field with and without shielding; another approach is the quotient, in decibels,
between the inducted tension with and without shielding, as described in Equation 1[2].

V(without shielding) E(electric field without shielding) (1)

SE dB = 20log SE dB = 20log
V(with shielding) E(electric field with shielding)

Methodology

Physical measurement

In a semi-anechoic chamber, a RF signal with constant power is injected into a coaxial cable which
impedance is matched with a 50 Ω load. Then the electric field emissions are measured by a field
probe. This experiment is realized first on a totally shielded cable, then on the same cable with a
shielding opening in its middle – 100mm then 300mm. See Fig. 2 and Fig. 3.

Fig. 2. Measurement setup photo

OBS.: The opening in the cable shielding has been made without disconnecting the cable.

Fig. 3. Measurement setup synoptic [3]

 ISEF 2009 - XIV International Symposium on Electromagnetic Fields
in Mechatronics, Electrical and Electronic Engineering
Arras, France, September 10-12, 2009

The experimental setup is based on the following parameters: L1=50 cm, L2 = 30cm, L3 =30 cm, h=5
cm and load=50 Ω. The coaxial cable under test is a RG-58 (Zc= 50 Ω; dielectric= polyethylene). The
distance between the probe and the cable center has been set to 50cm.
The following equipments have been used:
• Semi-anechoic chamber (efficient between 100kHz to 18GHz)
• Ground plane
• Field monitor HI6100 (ETS)
• Signal generator SML03 (R&S)
• Power meter URV5 (R&S)
• Amplifier 75A250 (AR 10kHz – 250MHz / 75W)
• Amplifier 500W1000A (AR 80 – 1000MHz / 500W)
• Field Probe HI6105 (ETS 100kHz – 6GHz)
• Integration software EMC 32 (R&S)

Fig. 4. Hardware configuration on EMC32 integration software

By means of the EMC32 integration software that controls signals injection and parameters
measurement, a constant 50W power is injected into the cable - considered as an antenna in Fig. 4 -
and the electric field is measured with then without shielding. The Voltage Standing Wave Ratio is
also monitored in the cable behavior. During the experiment the frequency is swept from 30MHz to
1GHz.

Virtual simulation

Fig. 5. Simulation via CST-MWS®

 ISEF 2009 - XIV International Symposium on Electromagnetic Fields
in Mechatronics, Electrical and Electronic Engineering
Arras, France, September 10-12, 2009

In order to determine virtually the Shielding Effectiveness factor of a coaxial cable, a virtual model
has been created based upon the physical setup and the cable, respecting dimensions and materials
characteristics. The software CST-MWS® Microwave Studio has been used - See Fig. 5 - and the
calculated results have been compared with the correspondent physical measurements.
The CST-MWS® simulation is based on the Finite-Difference Time-Domain method to solve the
electromagnetic problem [4].
The power of the signal virtually injected into the cable has been maintained to a 1W fixed value.
The correlation of this model with the physical experiment will permit the significance evaluation of
the Shielding Effectiveness virtual calculation.
If this method is considered pertinent, the model proposed will allow the Shielding Effectiveness
calculation of any coaxial cable, saving time and money.

Results

0,78

0,77

0,76

0,75
Noise floor (V/m)

0,74

0,73

0,72

0,71

0,70

0,69
30 130 230 330 430 530 630 730 830 930
Frequency (MHz)

Fig. 6. Noise floor of the experiment ambient

In order to measure only the coaxial cable behavior, the emissions of the test setup without the cable
under test – only the injection cable matched with a 50Ω impedance - have been evaluated inside the
semi-anechoic chamber. See Fig. 6. This noise would be a problem if superposed to signals which
values are about 1V/m, like in the shielded cable study. See Fig. 9. The injection cable seems to be the
responsible for this noise.
 ISEF 2009 - XIV International Symposium on Electromagnetic Fields
in Mechatronics, Electrical and Electronic Engineering
Arras, France, September 10-12, 2009

VSWR 100

10

1
30 130 230 330 430 530 630 730 830 930
Frequency (MHz)

VSWR with 100mm opening VSWR with 300mm opening

Fig. 7. Voltage Standing Waves Ratio Comparison

The VSWR measurement - see Fig. 7 - shows an important signal reflection between 30MHz and
250MHz that can explain the instability of the electric field behavior in this frequency band.

100

10
E Field (V/m)

0,1

0,01
30 130 230 330 430 530 630 730 830 930
Frequency (MHz)

Simulated E field with 300mm opening Simulated E field with 100mm opening
Measured E field with 300mm opening Measured E field with 100mm opening

Fig. 8. Simulated (1W injected) and measured (50W injected) Electric Field comparison
 ISEF 2009 - XIV International Symposium on Electromagnetic Fields
in Mechatronics, Electrical and Electronic Engineering
Arras, France, September 10-12, 2009

Concerning the electric field, a good correlation of form can be noticed between the simulated values
and the measured one. See Fig. 8. We can observe the resonance frequencies of the 50cm conductor in
the neighborhood of the theoretical values.
3*108 v
For 50 cm = 1,5 * λ , we have λ =33,33cm and f = ≈ = 900MHz
λ 33,33*10−2
3*108 v
For 50 cm = λ , we have f = = 600 MHz ≈
λ 50*10−2
λ v 3*108
For 50 cm = , we have λ =100cm and f = ≈ = 300 MHz
2 λ 100*10−2
λ v 3*108
For 50 cm = , we have λ =200cm and f = ≈ = 150 MHz
4 λ 200*10−2
The difference observed at the λ/4 resonance is due to the ground plane imperfection – monopole
antenna theory - and the high VSWR at between 30MHz and 250MHz that is not taken into account in
the virtual model - see Fig. 10. For the cable with a 300mm shielding opening, the software couldn’t
represent the 1,5 * λ resonance. This behavior of the virtual model needs to be analyzed to know
which phenomenon - skin effect, capacitive coupling, etc. - has been too much idealized.
It can be also noticed an amplitude difference between the simulated and the measured values because
the power injected in the virtual experiment is smaller than in the real one.

10,00000

1,00000

0,10000
E field (V/m)

0,01000

0,00100

0,00010

0,00001
30 130 230 330 430 530 630 730 830 930
Frequency (MHz)

Simulated E field with total shielding Measured E field with total shielding

Fig. 9. Comparison between E field measurement (50W injected) and simulation (1W injected) with total
shielding

The behavior of the shielded cable has been considered as ideal by the software because the model
doesn’t take into account structural imperfections. In the measurement, the setup noise is perceived in
the 30-70MHz, 220-290MHz and 650-780MHz. Moreover, the global shape of the simulated curve is
not similar to the measured one. Actually, the 1W injected power is insufficient to fully cross the
 ISEF 2009 - XIV International Symposium on Electromagnetic Fields
in Mechatronics, Electrical and Electronic Engineering
Arras, France, September 10-12, 2009

shielding. This is an important argument to explain the difference of Shielding Effectiveness between
simulation and measurement – see Fig. 9.

100

90

80

70
Shielding Effectiveness (dB)

60

50

40

30

20

10

0
30 130 230 330 430 530 630 730 830 930
Frequency (MHz)

Simulated SE with 300mm opening Simulated SE with 100mm opening

Measured SE with 300mm opening Measured SE with 100mm opening

Fig. 10. Simulated (1W injected) and measured (50W injected) Shielding Effectiveness Comparison

About the coaxial cable shielding effectiveness, the measured values present the following
characteristics:
- Influence of high VSWR in the beginning of the spectrum.
- Influence of the setup noise between 660MHz and 770MHz.
Comparing it with the simulation results, we can observe a good correspondence in term of resonance
frequencies except for the λ/4 one. For a 300mm opening, the 1,5*λ resonance is not calculated by the
software.

Conclusion

The shielding effectiveness of a coaxial cable has been calculated via measurement and simulation of
the electric field emitted by this cable when a constant power RF signal is injected in it. The cable has
been tested with total shielding, then with a known shielding opening.
The differences observed between the physical measurement and the simulated model’s behavior are
mainly due to a lack of refinement of the model. It could be improved increasing the physical signal-
to-noise ratio and the number of parameters of the model. For example, the model of the connectors or
the ground plane could be refined in order to better approximate the insertion lost, impedance
mismatch and wave reflection. Physically, the signal-to-noise ratio could be increased by improving
the injection cable quality and the measurement equipment precision. It can be considered that the
simulation and the measurement of the coaxial cable shielding effectiveness presented a similar global
behavior - except for the amplitude - between 250MHz and 900MHz but the correlation in high
frequencies is better when the shielding opening is 100mm or lower.
 ISEF 2009 - XIV International Symposium on Electromagnetic Fields
in Mechatronics, Electrical and Electronic Engineering
Arras, France, September 10-12, 2009

Another study is being realized with the software ANSOFT-HFSS® High Frequency Structure
Simulator - based on Finite Elements Method - in order to get another reference of simulation model
behavior.

Acknowledgments

The authors would like to thank Fiat Automóveis plant in Betim - MG , Brazil - for the use of its semi-
anechoic chamber and its RF test equipments.

References
[1] PAUL, C.R, Introduction to Electromagnetic Compatibility, John Wiley & Sons, Inc., New York, New York,
1992.
[2] Donald R.J, Michel Mardiguian, Electromagnetic Shielding, Interference Control Technologies, Inc. Volume
3, Gainesville, Virginia, 1988.
[3] R. De Leo, G. Cerri, V. Mariani Primiani, and R. Botticelli , A Simple But Effective Way for Cable Shielding
Measurement, IEEE Transactions on Electromagnetic Compatibility, Vol. 41, August 1999.
[4] CST MICROWAVE STUDIO® -Overview- , “© 2009 CST Computer Simulation Technology AG.”,
<http://www.cst.com/Content/Products/MWS/Overview.aspx>, 15/06/2009.

View publication stats