Radiated Emissions and Experimental Precautions of
Equipment with Cables in GTEM Cells
Zaid Muhi-Eldeen Al-Daher
#1
, Angela Nothofer
#1
, Christos Christopoulos
#1
, Steve Greedy
#1
#1
George Green Institute for Electromagnetic Research
1
zaid.muhi-al-daher@nottingham.ac.uk
Abstract— Any electric or electronic equipment sold within the
European Union has to comply with the EC Directive on
Electromagnetic Compatibility (EMC). To achieve compliance,
the equipment must be tested for radiated/conducted emissions
and immunity. A wide range of national and international testing
methods and standards are in force such as the IEC 61000-40-20
[1]. However, standards in general lack of describing testing
methods for equipment with cables. Since most devices and
systems contain leads and cables; the current standards cannot
be directly employed. In an effort to approach this crucial
matter, we present the outcomes of measurements conducted on
an EUT (metal box) with different cable bundle configurations,
in conjunction with both, the correlation algorithm given in the
standards and the repeatability concerns between different
GTEM cells.
I. INTRODUCTION
There is an apparent lack of standards describing test
methods for equipment with cables. The main standards, IEC
61000-4-20 [1] for example, explicitly exclude equipment
with external cables. Other standards, like the CISPR 16 and
CISPR 22 series [2] [3], include procedures for cable layouts,
but these are restricted to specific applications and only valid
for a limited frequency range (up to 1 GHz). In the absence of
regulations, most of EMC test laboratories would either have
their own in-house scenarios or would consider the worst test
configuration setup and as a result, repeatability in most cases
is unattainable. Therefore, there is a strong demand from
industry to improve on repeatability of emission and immunity
measurement results by engaging in studies of cable and
bundling effects in GTEM cells [1-5].The need to establish
more scientifically sound techniques for testing of equipment
with cables is not only essential but will also aid in comparing
measurement outcomes of different environments to those
obtained in GTEM cells. For this purpose, a number of cable
bundles have been considered and measured in two different
GTEM cells in order to provide an insight into the effects of
cable bundles and their repeatability’s in different cells. The
latter is another important topic that has rarely been addressed
in the literature. Different and even similar size GTEM cells
of the same manufacture can perform differently due to their
alterable physical construction nature and therefore parts of
this paper are devoted to this topic.
This paper is organized as follows; section II depicts on
important GTEM characterisation methods; section III
expounds on cable routing and testing procedures; and section
IV presents some measurements precautions and outcomes of
experiments conducted in two different GTEM cells which are
referred here as to the George Green’s cell (GG) and the
National Physics Laboratory cell (NPL). The frequencies
investigated range from 30MHz to 2GHz.
II. CHARACTERIZATION OF GTEM CELLS
As with any RF testing facility, it’s rather important to
characterize the GTEM cell prior to any measurements to
certify the validity of the recorded data and to ensure
reasonable levels of accuracy. There are three important
parameters of a GTEM cell operation that ought to be
controlled and checked periodically. These are, firstly; the
constant characteristic impedance throughout its length which
is expected to be 50 +/- 2 . This is usually measured using a
Time-Domain-Reflectometry technique which is also used for
locating any impedance mismatches across the cell’s length.
Secondly, the cells reflections at the input port which are
often ignored (also known as the return loss S
11
), ought to be
below -20dB (equivalent to a VSWR value of 1.22) across the
frequency range of interest. The GTEM consists of a two
termination modes; one is for the surface currents terminated
by a 50 resistor network and another is for the RF fields
terminated by a wideband pyramidal absorber. Any defects in
these termination modes can lead to significant effects on the
cell’s response and the measurement system dynamic range.
Additionally, any changes in the GTEM geometrical structure
(and consequently its characteristic impedance) or the
existence of any obstacles within the cell (such as ropes or
supporting objects) could easily disturb the propagating fields
and the cell’s reflections particularly at high frequencies.
Therefore, when using GTEM cells it’s extremely vital to
ensure what is measured is purely due to the EUT’s response
and not due to a contribution from the cell itself. The obtained
S
11
responses of GG and NPL’s GTEM cells are shown in Fig.
1. Strong peaks are observed below 200MHz. These are
mainly due to the crossover region from the current to the
wave termination whereby at these intermediate frequencies,
both terminators are not completely effective [6]. Such
frequencies are referred to as the characteristic frequencies of
the cell. The difference in their values between the two cells is
mainly down to the quality and size of the RF pyramidal
absorbers. Further away from the characteristic frequency,
S11 peaks tend to occur at 2 λ intervals.
Thirdly, the dominance of the primary (vertical) electrical
field component and its uniformity within the working volume
of the GTEM cell; verification methods and processes can be
found in [1].
978-1-4244-7306-9/10/$26.00 ©2010 IEEE
2010 Loughborough Antennas & Propagation Conference 8-9 November 2010, Loughborough, UK
133
978-1-4244-7307-6/10/$26.00 ©2010 IEEE
brought to you by CORE View metadata, citation and similar papers at core.ac.uk
provided by Glamorgan Dspace