Exploration Electromagnetic Noise Reduction for IC ... · Exploration Electromagnetic Noise Reduction for IC Electric Field Measurement and ... Plot 6, Bayan Lepas Technoplex, Medan

International Journal of Engineering & Technology IJET-IJENS Vol:14 No:02 32

143502-2727-IJET-IJENS © April 2014 IJENS I J E N S

Exploration Electromagnetic Noise Reduction

for IC Electric Field Measurement and

Prediction in GTEM Cell

King Lee Chua1, Mohd Zarar Mohd Jenu

1, Man On Wong

2, See Hour Ying

2

1Center for Electromagnetic Compatibility, Universiti Tun Hussein Onn Malaysia (UTHM),

Parit Raja, Batu Pahat, 86400 Johor, Malaysia.

[email protected], [email protected] 2Altera Corporation (M) Sdn. Bhd.

Plot 6, Bayan Lepas Technoplex, Medan Bayan Lepas, 11900, Penang, Malaysia.

[email protected], [email protected]

Abstract-- In IEC 61967 standards, there is often interest to

evaluate electromagnetic emission of integrated circuits (ICs) by

mounting IC test board on TEM/GTEM cell wall. The method is

able to separate the desired emission measurement to be

influenced by unintentional interference. However, this method

has limited test device rotation in horizontal position as well as

neglecting vertical polarization which is also a significant source

of emission. Basically, the electromagnetic emission of a device is

contributed by both the horizontal and vertical polarizations.

The limitation can be overcome with three dimensional views as

termed to be three orthogonal rotations in Gigahertz Transverse

Electromagnetic Mode (GTEM). When conduct emission

measurement by placing the whole test device in the GTEM cell,

it is important to ensure that unwanted disruption from

supporting components on the test board and its interface cable

will not interfere with the measurement. Therefore, in this paper,

we present numerous experimental works to tackle these crucial

matters in order to emphasize the technique to quantify IC

electromagnetic emission performed inside a GTEM cell. This is

achieved with application of basic electromagnetic compatibility

(EMC) measurement approaches such as shielding, grounding

and suppression using ferromagnetic material. Following that,

dipole moments model is established to achieve far electric field

approximation and verification with semi-anechoic chamber

(SAC) measurement. The results show strong evidence that the

effectiveness of the new proposed technique for IC emission

measurement in GTEM cell. The obtained results are then

processed further using conventional algorithm for correlation

with field measurement in semi anechoic chamber.

Index Term-- Electromagnetic noise, integrated circuit,

electromagnetic radiation, GTEM cell, SAC

1. INTRODUCTION

The electronic world nowadays is moving towards the fashion

for producing smaller devices with faster clock speed and

higher integration densities. This is generally realized by

embedding electronic components such as transistors, diodes,

capacitors and resistors in a miniature size integrated circuit

(IC). Due to advance semiconductor process technologies,

smaller ICs can be built with very complex structure.

Consequently, ICs have become significant noise source that

cause electromagnetic emission. The fact has led to growing

demands on electromagnetic emission characterization at chip

level. The exploration is importance to provide the

corresponding information for component selection in early

printed circuit board (PCB) design. Therefore, shorter time is

required to develop any electronic system in compliance of

electromagnetic compatibility (EMC) test.

According to the studies in past, two techniques namely

TEM/GTEM cell measurement method [1] and near field

scanning method [2] and [3] are introduced for ICs

electromagnetic emission characterization. Near field

technique has advantage to characterize IC electromagnetic

emission in close vicinity, with assumption the measured

emission is contributing by IC itself if an infinite perfect

ground plane is established around the IC test board. This

ensures IC is the only radiator and reliability of the

measurement is attained. Conversely, ambient noise may be

considered as the main aspect that disturbing measurement

accuracy in the near field method. To prevent disruption of the

ambient noise, the near field system generally is setup inside

shielding room which is costly and not affordable by most of

EMC test laboratories.

Basically, TEM/GTEM cell is an enclosed metallic

structure which provides well isolation between inner and

outer environments of the cell. As the cell is properly closed,

its inner side would neither contribute to nor suffer from any

external interference. The cell cost is also relatively cheaper

than building a shielding room. Another issue of the near field

method is field probe sensitivity must compromise with spatial

resolution and dynamic range. A narrow band probe merely

suitable for field measurement in a specific frequency range,

therefore various set of probes are needed for emission test in

different frequency ranges. In particular, TEM/GTEM cell

septum behaves as receiving antenna which has frequency

response over a wide band of frequency, so no receiving

antenna change is required. The GTEM cell can operate over

wider range of frequencies than TEM cell. In addition, GTEM

cell offers three dimensional field views which provide overall

information for radiated emission testing for the IC.

In this paper, the focus is given on the technique to

characterize radiated electromagnetic emission of a FPGA

chip using GTEM cell. The measurement is conducted inside

mailto:[email protected]






the cell instead of clamping it on the cell body. In order to

ensure that the FPGA chip is continuously active in the

measurement, the chip was configured with toggle flip-flop

(TFF) pattern and exercised at a 100MHz clock signal. Several

techniques such as shielding the FPGA board inside a metallic

enclosure, grounding the enclosure with low impedance

ground strap and suppressing common mode current emission

of interface cables using ferromagnetic material are employed

to avoid interference due to unwanted electromagnetic noise.

Then, the GTEM data is process to extract equivalent dipole

moments for far field prediction and verification with SAC

measurement. The results presented in the study provide

useful alternative to IC emission test methods based on IEC

61967.

2. CHARACTERIZATION OF GTEM CELL

Prior to perform any electromagnetic emission measurement

of ICs, the performance of GTEM cell is important to be

checked and validated. This can ensure the measured data

achieve a reasonable level of accuracy.

GTEM cell can be considered as a rectangular

transmission line which operates in TEM mode. The cell

characteristic impedance Z0 along septum length is expected to

be 502. This parameter must be checked because

impedance mismatching can lead to multiple reflections and

eventually affect reliability and accuracy of collected raw

data. When a network analyzer is connected to cell port, it is

possible to evaluate the overall cell input impedance against

frequency range of interest. Fig. 1(a) shows the input

impedance over a frequency range from 30MHz up to 1GHz.

It is seen that the input impedance varies about 50

frequency of interest.

The reflection properties at GTEM cell is characterized

by measuring return loss of the cell when it is empty. As can

be seen in Fig. 1 (b), the return loss is well below 20dB over

the frequency spectrum. The 50 load and the RF absorber

have minimized reflections and resonances in the cell. The

return loss that exceeds -20dB still considered acceptable

because according to the standard IEC 61967-2, the return loss

ought to be below 14dB and it is equivalent to a VSWR value

of 1.5.

(a)

(b)

Fig. 1. Parameters of GTEM cell, (a) cell impedance vs frequency, and (b)

cell return loss

Basically, a GTEM cell comprises a tapered section with

a single port, 50 ohm characteristic impedance and a RF

absorber. Fig. 2(a) shows the setup of a spectrum analyzer to

measure output voltage due to electromagnetic emission of

any device under test (DUT) in the cell. In accordance to

transmission line theory, the overall structure can be

simplified into an equivalent circuit in Fig. 2(b). Here, it is

assumed that cell septum as inner conductor, cell wall as outer

conductor, the DUT represents a source and the spectrum

analyzer is the load.

(a)

50Ω Z0

l

Transmission line

VS

VLZL

ZS

Vi

z = -lz = 0

z

(b)

Fig. 2. (a) GTEM emission measurement setup,

(b) Equivalent circuit

Voltage reflection coefficient Γ at a load is defined as the

ratio of the amplitudes of the reflected and incident voltage

waves [4]. Assume the GTEM cell as lossless transmission

line, the voltage reflection coefficient can be obtained as

(1)

where is incident wave travelling towards the load,

represents reflected wave travelling towards the source, ZL is

100 200 300 400 500 600 700 800 900 100030

35

40

45

50

55

60

65

70

Frequency (MHz)

Imp

ed

an

ce

(O

hm

)

100 200 300 400 500 600 700 800 900 1000-70

-60

-50

-40

-30

-20

-10

Frequency (MHz)

S1

1 (

dB

)

Septum

RF absorber

50

load



load impedance, Z0 is characteristic impedance, and zL is

normalized load impedance ⁄ . The magnitude of | ( )| at the load is obtained

| ( )| | ( )|

| |[ | | | | ( )]

⁄ (2)

with

phase constant, c is speed of light in free space,

| | is reflection coefficient, and is phase angle.

The voltage at source is determined with

| ( )| ( ) (3)

3. TEST DEVICE PREPARATION

The DUT chosen for emission test is a field programmable

gate array (FPGA) chip which is mounted on a printed circuit

board (PCB) and sitting on top of ground plane. All necessary

components, other than the chip, are soldered on opposite side

of the PCB. Fig. 3 shows the FPGA test board.

(a) (b)

Fig. 3. FPGA test board, (a) Top side, (b) Bottom side

It has been challenging to characterize electromagnetic

emission of IC inside the GTEM cell. This is because all

supporting components and interconnection cables on the PCB

may generate unwanted energy that would also couple to cell

septum. As a result, the measured emission is a combination

field contributed by the IC and the disturbance. However, it is

believe that the interference can be minimized if precautions

are taken into account in the measurement setup.

Shielding is a simple and effective technique that has

widely been used to isolate an environment from

electromagnetic interference. So the PCB is firstly housed

inside a metallic cavity and exposes the IC via a window.

Since conductor is a perfect shield, it can isolate unwanted

energy inside the cavity. This can ensure the FPGA chip is the

only source to contribute electromagnetic emission in

measurement.

In general, the performance of a shield depends on shield

thickness. According to [5], if shield thickness is greater than

skin depth of the material at the frequency of incident field,

there is less likely the multiple reflections and the material is

dominated by absorption loss, so it can behave as a good

solved by

√ ⁄ meter (4)

with f is frequency of incident field, is shield conductivity

and is shield permittivity. Aluminium is a cheap and

versatile conductor with adequate absorption for shield. Its

conductivity, is ⁄ , relative permittivity, is

and permeability, is ⁄ . When

aluminium is selected as shield material to build the cavity,

the skin depth for the lowest frequency of interest, 30MHz is

approximated 0.026mm. Since the shield thickness is 0.2mm,

so it can be assumed that the cavity is working as a perfect

barrier to isolate generated electromagnetic wave to travel

across once the cavity joints are properly shielded.

Although the conductor can behave as a good shield,

however, at the same time, it might be a perfect radiator if the

cavity resonant frequency fr agree with fundamental frequency

and harmonics of electromagnetic energy inside the cavity. In

this particular case, resonance may amplify the energy inside

the cavity and consequently propagate it out of the cavity. The

resonance frequency of the cavity can be determined by

√ √(

)

(

)

(

)

(5)

where

√ is phase velocity of uniform plane wave in the loss

less dielectric medium (=0, , ) filling the cavity, m, n and p

correspond to the number of half-wave variations of the field

in respective x-, y- and z-direction. The lowest resonant

frequency of the cavity is approximated 1.7GHz and it is

greater than the highest frequency of interest. Thus, as long as

the operating frequency and its harmonics stay below this

resonance frequency, the cavity will sustain a free oscillation.

The setup of the FPGA board in the metallic cavity is shown

in Fig. 4.

Fig. 4. Setup of the FPGA board inside metallic cavity

4. ELIMINATION ELECTROMAGNETIC NOISE

4.1. Ambient Noise

It is important to evaluate ambient noise inside the GTEM cell

to ensure that there are no external electromagnetic leakages

due to imperfect shielding of the cell. Therefore, action can be

taken if any voltage peak appears on the desired spectrum

frequency. This is to make sure what is measured is

contributed by the IC and a clean noise floor shall be obtained

when it is not powered.

a. Grounding

As mentioned previously, the cavity may be a good radiator if

not properly grounded. When the IC is energized, the

electromagnetic fields which emitted from package lead frame

and bonding wires of the IC will coupled with the cell septum.



However, part of the fields will also be coupled with the

cavity, resulting in induction current which will produce

secondary emission.

In this case, proper grounding is extremely important for

diverting the current away from the cavity. Thus, the inner

side of the cavity is grounded to PCB using gasket whereas

the outer side is tied to GTEM cell body using ground strap.

The high quality ground strap has to be chosen to provide low

impedance and good bonding between connecting points to

avoid unwanted potential difference developed along the path.

The basic concept is lower impedance path will divert more

current away from the victim.

Fig. 5(a) shows the wire, ribbon cable, and copper tape are

selected for grounding purposes. Fig. 5(b) is a setup inside the

GTEM cell with attachment of copper tape between the cavity

and GTEM cell ground plane clearly shown.

(a) (b)

Fig. 5 Minimizing enclosure effect with different types of ground

straps, (a) selected ground strap, (b) setup of the measurement

Fig. 6 presents ambient n.ises after grounding the cavity

using different ground straps where the FPGA chip is no

powered. No obvious difference observed even after power on

the FPGA chip. It can be observed that leakages from external

sources manage to enter the GTEM cell via the power and

signal cables. These leakages must be removed to ensure

correct measurement of electromagnetic emission due to

FPGA chip. In particular, it is found in the experimental result

that usage of copper tape is effective in removing some of the

external electromagnetic noise.

Fig. 6. Ambient noise while grounding cavity with different

ground straps

b. Location of Grounding

The location of the ground strap plays another important

factor to optimize the effectiveness of grounding. In this

particular case, the ground strap shall be placed near to all

interconnection points so that the electromagnetic noise from

the interconnection points will instantly be diverted to ground.

Fig. 7 shows placement of copper tape for location P1 and P2.

The corresponding result in Fig. 8 indicates further

improvement on the cleanliness of the noise floor while

placing copper tape at both locations P1 and P2.

Fig. 7. Placement of copper ground

Fig. 8. Ambient noise for different ground locations

4.2. Suppression with Ferromagnetic Material

The result indicates grounding effective to partially

remove the external electromagnetic leakage. Therefore,

ferrimagnetic material is employed to suppress the common-

mode current which unintentionally formed at outer layer of

the interconnection cables, with the assumption that common-

mode cable current is the basis for electromagnetic

interference (EMI) and ferrite provides an excellent solution to

suppress EMI [6].

Fig. 9 presents the new noise floor after reducing external

leakages using ferromagnetic material and bundling all the

cables with Ni/Cu woven.

100 200 300 400 500 600 700 800 900 100010

20

30

40

50

60

Frequency (MHz)

Am

plit

ude (

dB

uV

)

No ground wire

Single wire

Ribbon cable

Copper tape

100 200 300 400 500 600 700 800 900 100010

20

30

40

50

60

Frequency (MHz)

Am

plit

ude (

dB

uV

)

No ground

P1

P2

Both

100 200 300 400 500 600 700 800 900 100010

20

30

40

50

60

70

80

Frequency (MHz)

Am

plit

ud

e (

dB

uV

)

Without ferrite & Ni/Cu woven

With ferrite

With ferrite & Ni/Cu woven

Copper

tape

P1 P2



Fig. 9. Radiated emission before and after wrapping cables with ferrite

sheet

4.3. Double Shielded Cables

The previous result still shows the existence of

electromagnetic leakage at about 900MHz even after efforts

had been taken to eliminate it with shielding, grounding and

suppression using ferrite. Since these peaks appear across

GSM mobile phone frequencies, it is suspected that these

peaks originate from nearby base station. The cable

connecting the GTEM cell to spectrum analyzer is able to pick

up GSM signal. The condition can be improved using double

shielded cable as shown in Fig. 10.

Fig. 10. Use double shielded cable for better performance

5. RADIATED FIELD MEASUREMENT

After obtaining a clean noise floor, the FPGA chip was

mounted on the manipulator as shown in Fig. 11. The device

is rotated at three orthogonal orientations for optimizing

emission measurement. As the radiation pattern of the FPGA

chip is unknown. At each orientation, the device is further

rotated across its vertical axis for 0, 45 and 315 degrees. The

voltage of the emission is measured at the GTEM cell port

using spectrum analyzer.

To operate the FPGA chip, a toggle flip-flop (TFF) logic

circuit pattern was created for the chip configuration. The

circuit will occupy 20% of total the FPGA chip space so that it

will generate sufficient energy for the emission measurement.

The TFF circuit has two I/O terminals where the input

terminal is exercised with an external sinusoidal clock signal.

The clock frequency is 100 MHz and the output signal of the

TFF pattern is 50 Mhz.

Fig. 11. Measurement setup in GTEM cell

Fig. 12 shows the measured radiated emission that

captured in x, y and z positions. It is observed that the

frequency spectrum comprises fundamental clock frequency

with integration fundamental and harmonics of the output

signal. The data can now be used for further applications

because the effects of unwanted ambient noise have been

eliminated using the techniques described earlier in this paper.

It is assumed that the measured voltages are contributed by the

FPGA chip.

(a)

(b)

(c)

Fig. 12. Emission of FPGA chip at three orthogonal orientations, (a) x

axis, (b) y axis, and (c) z axis

6. Equivalent Emission Model

As rotating the FPGA chip in the GTEM cell, the

electromagnetic field emitted from the FPGA chip will couple

to differential dipole components and the TEM mode e0y of the

GTEM cell resulting potential at the cell port. However, the

measured GTEM voltages only contain electrical information

relative to the FPGA chip emission in the GTEM cell. We also

have known that the voltage is actually related to the current

source, i.e., electric and magnetic dipole moments as given by

( )

( ) (6)

Assuming the FPGA chip as a general source model, the

dipole modeling process can be carried out to approximate far

field for correlation between GTEM cell and semi anechoic

chamber. In particular, the FPGA chip is represented as a set

of dipole which comprises three electric moments , ,

and three magnetic moments , , as to the

100 200 300 400 500 600 700 800 900 100010

20

30

40

50

60

Frequency (MHz)

Am

plit

ud

e (

dB

uV

)

Single shielded cable

Double shielded cable

100 200 300 400 500 600 700 800 900 100015

20

25

30

35

40

45

50

55

60

Frequency (MHz)

Am

plit

ude (

dB

uV

/m)

0 degree

45 degree

-45 degree

100 200 300 400 500 600 700 800 900 100010

20

30

40

50

60

70

Frequency (MHz)

Am

plit

ude (

dB

uV

/m)

0 degree

45 degree

-45 degree

100 200 300 400 500 600 700 800 900 100010

20

30

40

50

60

70

80

Frequency (MHz)

Am

plit

ude (

dB

uV

/m)

0 degree

45 degree

-45 degree



actual source in the GTEM cell. A coordinate system ( , ,

) is assigned to the FPGA chip.

In conventional algorithm [7], the moments are obtained

with assumption the relative phase differences between the

dipole moments are negligible. The algorithm requires nine

orientations of GTEM voltages in Fig. 12 to obtain the

complete set of dipole moments. Each of the GTEM voltage in

decibels per microvolt (dB V) is normalized by the TEM

mode in order to relate to the magnitude of bij with Eqn. (7).

( ( ) )

(7)

where Zc = 50 is the GTEM cell characteristic impedance.

Subscript i is the three orthogonal orientations and subscript j

denotes rotation at vertical axis. The e0y is numerically

approximated by

( )

⁄ ∑ [ (

)

(

)]

( ) ( ) ( ) (8)

where a, h, g, y are denoted the cell width, the septum height,

gap width, and the test object height, respectively at the

location of measurement.

The following are nine equations correspond to nine

GTEM measurements. The second subscript 1-3 denotes three

rotation angles with = 0, = /4, and = -/4 of vertical

axis at each orientation.

(9)

where is the wave number in free space.

By considering predominant electric (P koM) or

magnetic (P << koM) source of the model is unknown, the

electric dipole moments can directly be solved via three basic

positions i.e. , , and .

(10)

Alternatively, the electric dipole moments also are able to be

determined by summing up the remaining positions.

( )

( )

( )

(11)

The magnetic dipole moments are obtained by

;

;

(12)

where

| |, | |, | | (13)

The extracted electric and magnetic dipole moments are

available to approximate three radiated electric field i.e. ,

, and in a far field of free space. The amplitude of

vertical field | | is | | and the horizontal electric field, | |

is obtained as √| | | |

.

7. SAC VALIDATION

The predicted far radiated electric field based on GTEM cell

measurement is verified with SAC measurement. As shown in

Fig. 11 and Fig. 13, the coordinate system of SAC is defined

differently from that of GTEM cell. In SAC, the horizontal

and vertical electric field strengths are automatically

computed by the program controlling the SAC equipment. In

particular, the GTEM quantities have to be manually adjusted

align with the SAC quantities for comparison.

Fig. 13 shows the experimental setup of the test device in

horizontal position on the turn table. The test device is placed

at a height of 0.8m above a ground plane of the chamber. The

radiated field was measured at 3m distance away from the

centre of the test device using a bilog antenna. The frequency

band of the antenna is 30 ~ 100MHz. The antenna height is

located at 1m to hold the maximum values coupled to the

antenna. The turn table is remotely controlled in the

measurement.

Fig. 13. Radiated emission measurement in SAC

A typical SAC emission measurement (Fig. 14(a)) will

evaluate the horizontal and vertical components by rotating

the antenna at its horizontal axis. In fact, we aware the



horizontal component is contributed by two axis alignments

where the adjacent horizontal source is always neglected in

the typical measurement.

3m

Measuring antenna DUT

Y

X

Z

(a)

3m

Measuring antenna DUT

Y

X

Z

(b) Fig. 14. Changing antenna position to account neglected horizontal

component; (a) typical antenna setup, (b) additional antenna setup for

SAC measurement

To account the missing adjacent horizontal component, we

perform additional measurement by changing the antenna

position in 90 degree across its vertical axis. The horizontal

component is obtained as √| | | |

as shown in Fig.

14(b).

Fig. 15 presents the comparison between GTEM data and

SAC data. The noise floor between the two environments is

similar to achieve meaningful comparison. The electric field

peaks frequency in the spectrum is matching well. However, it

generally observed that the GTEM cell able to capture

stronger emission as the shorter distance between the test

device and the antenna. Therefore, further work will be carried

out to establish correct factor for improving the agreement of

the peak value.

(a)

(b)

Fig. 15. Estimated radiated emission from FPGA chip in GTEM cell as compared to SAC measurement; (a) horizontal electric field intensity and

(b) vertical electric field intensity

CONCLUSION

In this paper, we have presented techniques to eliminate

electromagnetic noise from interfering the radiated emission

due to ICs in a GTEM cell. A proper shielding the IC test

board using metallic cavity combined with the usage of

grounding near interconnection points using low impedance

ground strap and cable shielding using ferrimagnetic

successfully removed the unwanted noise in the emission

measurement. The approximate horizontal and vertical electric

field via dipole moment technique has shown optimistic

insight correlation between GTEM cell and SAC. This work

provides a preliminary input in an effort to perform IC

emission measurement inside a GTEM cell as opposed to

mounting it on the GTEM wall based on IEC 61967.

ACKNOWLEDGEMENT

The authors wish to thank Multimedia Development

Corporation (MDeC) for the financial support and Altera

Corporation (M) Sdn. Bhd. for providing FPGA test board for

the research. The support by UTHM Research, Innovation,

Commercialization and Consultancy Office (ORICC) is also

acknowledged in providing the permission and fee to publish

this paper.

REFERENCES [1]. Standard EMC 61967-2. Integrated Circuits -- Measurement of

Electromagnetic Emissions, 150kHz to 1GHz -- Part 2: Measurement

of Radiated Emissions - TEM Cell and Wideband TEM Cell Method.

Int. Electrotech. Commiss. (2005). [2]. B. Deutschmann, H.P., and G. Langer. Near field measurements to

predict the electromagnetic emission of integrated circuits. Proc. 5th

Int. Workshop Electromgan. Compat. Integr. Cuicuits (2005). [3]. Haixiao, W., D.G. Beetner, and R.E. DuBroff. Prediction of Radiated

Emissions Using Near-Field Measurements. IEEE Transactions on

Electromagnetic Compatibility, (2011). 53(4): p. 891-899.

[4]. F. T. Ulaby, E.M., U. Ravaioli. Fundamentals of Applied

Electromagnetics. 6th Edition: Pearson (2012). p.80-83

[5]. Paul, C.R., Introduction to Electromagnetic Compatibility. 2nd Edition (2006): John Wiley & Sons. p.718-721

[6]. Hubing, T.H. Bundled cable parameters and their impact on EMI

measurement repeatability. IEEE International Symposium on Electromagnetic Compatibility. (1990). p. 576-580

[7]. Wilson, P., On correlating TEM cell and OATS emission

measurements. IEEE Transactions on Electromagnetic Compatibility. (1995). 37(1): p. 1-16.

100 200 300 400 500 600 700 800 900 1000-10

0

10

20

30

40

50

60

70

80

Frequency (MHz)

Ra

dia

ted

Em

issio

n (

dB

uV

/m)

SAC

GTEM

100 200 300 400 500 600 700 800 900 1000-10

0

10

20

30

40

50

60

70

80

Frequency (MHz)

Ra

dia

ted

Em

issio

n (

dB

uV

/m)

SAC

GTEM

Documents

Exploration Electromagnetic Noise Reduction for IC ... · Exploration Electromagnetic Noise Reduction for IC Electric Field Measurement and ... Plot 6, Bayan Lepas Technoplex, Medan