Choosing the Right Chamber Depends on the - ETS

advertisement
Anechoic Chambers
© Photographer: Madartists | Agency: Dreamstime.com
on
Choosing the
Right Chamber
Depends on the Application
by Martin Wiles
Conformity FEBRUARY 2008
Editor’s Note: Since the original publication of this article
in February 2007, new methods of chamber validation have
become well established, and work has begun on product
measurements above 1 GHz, where the debate continues
around the issue of height scanning. In the meantime, other
standards, such as CISPR 22, have started to align themselves
with CISPR 16 by referring to the chamber tests from their
respective documents. This will carry CISPR 22 forward until
CISPR 32 replaces CISPR 22 and 13. We will cover the likely
impact of these changes in a future issue of Conformity.
A
nechoic chambers are used today for performing
EMC measurements according to a variety of
different published EMC standards. There are many
different fields of application including consumer electronics,
automotive, aerospace, military, medical, telecommunications
and others. Standards are developed and published worldwide
by different organizations resulting in different requirements
and consequently different chamber types. Written for
engineers new to EMC anechoic chambers, this article will
discuss the type of chamber that is required for different
applications and their standards, in particular the changes to
CISPR 16, and how these chambers are calibrated against the
requirements.
RF Absorber Technology
The RF absorber materials that line the surfaces of the
shielded room play a major part in the chamber design. There
are 3 basic types involved:
• Microwave pyramidal absorber ‑ electric losses
°
This is the preferred technology for high frequencies
°
Usable frequency range 100MHz‑18GHz
°
°
The losses are provided by carbon loading of a
pyramid structure
Low frequencies limited by size: 2.4m absorber
required at 30MHz
• Ferrite tile ‑ magnetic losses
°
This is the preferred technology for low frequencies
°
Magnetic losses provided by 5 to 6mm tile
°
Usable frequency range 30‑1000MHz
°
Space efficient but heavy.
°
Disadvantage is that it cannot be used for high
frequencies and must be used in combination with
hybrid pyramid foam
• Hybrid absorber ‑ electric and magnetic losses
Today’s Technology
°
This is the preferred technology for broadband EMC
EMC anechoic chambers are primarily used for testing
radiated emissions (RE) and immunity (RI) in the frequency
range from 30 to 1000MHz, with extensions to 18GHz, or
even 40GHz becoming more frequent. Different methods
and criteria for validating chambers and performing EMC
measurements for testing immunity or emissions are
standardized, including test distances, field levels, emission
limits, pass criteria, equipment set up and so on.
°
Usable frequency range 30MHz‑18GHz
As a consequence and depending on the need to fully
comply with these standards, EMC anechoic chambers
can be typically divided into two groups: pre‑compliance
(testing is research and development or the purchase budget
is low) or full‑compliance (testing is for type approval).
Whilst a manufacturer who can fill the test schedule with
its own internal needs will focus the chamber requirements
on its own product standards, an independent test lab must
broaden its scope to be able to meet the requirements of as
many standards as possible. Through put is often important
in both cases and quick change from one test to another is an
important part in chamber ergonomics.
The EMC anechoic chamber will be an RF shielded room
with RF absorber materials installed on the four walls and
ceiling and possibly on the floor. The design of EMC anechoic
chambers is dictated by the standards and the available RF
shield, absorber and RF test equipment technology. Let us
look at what is typically used today.
°
EMC Anechoic Chambers
In general, EMC anechoic chambers are different and can be
divided into different groups as follows:
• Partially lined: Surfaces are not fully covered with
absorbers
• Semi anechoic SAR: The walls and ceiling are covered
with absorbers whilst the floor is a metal reflecting ground
plane
• Fully anechoic FAR: All surfaces are covered with
absorbers
In addition and more specifically in the last decade, EMC
chamber technology has been able to establish the following
two categories of chamber:
• Pre‑compliance compact chambers
RI compliance/RE pre‑compliance
°
°
°
Conformity FEBRUARY 2008
The hybrid pyramid foam must have special formula
for good matching with ferrite tile at the bottom. At
high frequencies its performance is not as good as
microwave pyramid of equal size.
Compliance to IEC 61000.4.3 radiated immunity
Pre‑compliance to the radiated emission standards
such as CISPR 22
Typical size: 7.2 x 3.0 x 3.0 m L x W x H
• Full compliance chambers
RI compliance/RE compliance
°
°
°
Compliance to IEC 61000.4.3 radiated immunity
Compliance to the radiated emission standards such as
CISPR 22
Typical sizes:
•
3m: 8.5 x 6.0 x 6.0 m; 2m diameter quiet zone
(Figure 1)
•
5m: 11.5 x 7.0 x 6.0 m; 2m diameter quiet zone
(Figure 2)
•
10m: 18.8 x 11.6 x 8.5 m; 3m diameter quiet
zone (Figure 3)
The most common type of chamber will be a compact or full
3m type. The compact chamber offers the advantage of being
able to fit into the majority of buildings due to their limited
height of 3m. The full 3m and larger chambers will be part of
a dedicated building purposely built in many cases to house
the chamber. In parallel with the transient nature of some
markets like telecoms most of these chambers today offer
the flexibility of being removable or modified to a different
size if the requirements of the testing change or the company
moves to a different location. The life cycle of the products
has increased together with the quality of some of the key
maintenance items like shielded doors.
CISPR 16‑1‑4 will prefer the higher accuracy of the biconicals
and log periodics. This is due to the smaller electrical size
of the antenna and subsequent better definition of the phase
center.
In addition, a combination antenna is about 1.5m long and
has its low frequency section at the back. Assuming the phase
center is defined as the mechanical center of the antenna,
the two low frequency sections will be separated at a 3m
distance by 3m plus an extra 1.5m (i.e., 50% increase in
the separation). For the most difficult chamber calibration
described by the NSA method, this has a not insignificant
effect, and means that biconicals and log periodics are
preferred at least for the chamber calibration. The FAR (fully
anechoic rooms) testing methods, described later, have refined
this method using a small biconical but this has encountered
other problems and these methods are likely to return to
standard size biconicals.
Antennas
In general, the antennas used for radiated field EMC
testing are of key importance as compared to the rest of the
instrumentation. Over the past ten years antenna technology
has clearly progressed and adapted to the different tests
required by the standards. The older biconical and log periodic
designs have seen the arrival of hybrid or combination
antennas that can cover the whole frequency band 26MHz to
3GHz or 80MHz to 6GHz in one sweep by combining and
matching a biconical section to a log periodic section. Such
combination antennas are often greater than a meter in width
and length.
Figure 1: Full 3m Chamber
Radiated immunity testing will require low VSWR and high
gain so that the smallest possible amplifier can be used to
generate the required field. Here, combination antennas are
the preferred technology
Radiated emissions testing requires high precision calibrated
antennas. Antennas have technically changed very little
with biconicals and log periodics remaining the preferred
technology, although there are some combination antennas
which offer sufficient precision for compliance.
For fast testing and minimizing setup changes, combination
antennas can be used for emission and immunity testing at
the same time, so many EMC product measurements will be
performed with combination antennas. However, chamber
calibration according to radiated emission standards such as
Conformity FEBRUARY 2008
Figure 2: Full 5m Chamber
For radiated immunity testing, the
antenna is generally calibrated at 1 or
3m distance in order to give an idea
to system designers how much power
can be developed with a given signal
input.
Consumer Electronics
Typical chamber type: Compact,
Full 3m/5m/10m
Typical frequency range:
30MHz‑18GHz
Radiated emissions requirements
are covered under the basic standard
CISPR 16, whilst specific products
groups are covered by CISPR 11,
14 and 22. In the U.S., FCC Part
15 requirements are similar. CISPR
16 has undergone some significant
changes in recent years and now
includes chamber validation methods
for both Semi Anechoic Chambers
Figure 3: Full 10m Chamber
(SACs) and Fully Anechoic Rooms
(FARs) in CISPR 16‑1‑4 and most
Applications
recently for chamber validation above 1GHz. These changes
are slowly being adopted by the product standards of which
The main EMC test applications are consumer electronics,
telecom, medical, automotive and military/aerospace. There is CISPR 22 (ITE) and CISPR 13 (Broadcast devices) will
merge in the future into CISPR 32, the so‑called multimedia
no single organization regrouping all the standards covering
standard.
the above groups. The European Union (EU) has played a
For radiated emission testing,
antennas are calibrated as pairs on a
reference OATS according to ANSI
C63.5, and CISPR 16‑1‑4 defines
the FAR method for calibrating the
free space antenna factor. Note that
a new draft standard CISPR 16‑1‑5
is currently being developed by IEC/
CISPR but this will not be published
for several years.
key role in recent years in initiating a major overhaul in the
standards available and has produced a number of directives
such as the EMC Directive (2004/104/EC), the Automotive
Directive (2004/104/EC), the Low Voltage Directive (73/23/
EEC), the R&TTE Directive (99/5/EC), and the Medical
Devices Directive (2001/104/EC).
These directives require the testing of all products to be
sold within the EU according to standards which have been
either developed (mostly by CENELEC and ETSI), or
which have been derived
from existing standards.
Standards organizations
like the IEC/CISPR, who
publish standards on a global
level, are extremely active
and their newly published
documents need to be
monitored on a regular basis
in order to remain up to date
with current requirements.
We will now discuss these
applications and describe
the typical chambers used
to carry out EMC testing
related to them. A quick-look
reference guide can be found
in
Table 1 which summarizes
the applications with the type
of chamber required.
The CISPR 16‑1‑4 chamber validation method <1GHz
for SACs remains the well established Normalised Site
Attenuation method and this document will most likely replace
the EN 50147‑2 in Europe under new proposals as the CISPR
document is adopted as an EN. Using a pair of antennas
previously calibrated on a suitable OATS, the chamber floor
will be a ground plane and validation will be carried out from
30‑1000MHz. A transmit antenna will be placed at different
positions and heights on the turntable and a receive antenna
will be scanned from 1‑4m at
3m (CISPR 22 only) or 10m
distance. When normalized
to theoretical values the
site’s NSA must be within
+/‑4 dB. The size of the
EUT/quiet zone will vary
and will dictate the size of
the room and the distance at
which the measurements are
made. For example, EUTs
larger than 2m are typically
not tested at 3m since their
front face would be in the
near field of the antenna,
producing unreliable results
and therefore a 10m distance
would be required.
EMC Anechoic Chamber ‑ Automotive testing at GM
Conformity FEBRUARY 2008
The antenna calibration plays
a key role in the uncertainty
budget of this measurement
and is typically carried
out per the ANSI C63.5
Specifications
Commercial
Electronics
& Medical
ANSI C63.4‑2003
General
Description
Frequency Range
Min Chamber type
Methods of measurement of radio‑noise
emissionsfrom low voltage electrical and
electronic equipment in the range 9KHz to
40GHz
RE
9KHz to 40GHz
Full SAC 3m,5m,10m
CISPR 11 (2004)
Industrial, scientific and medical (ISM)
radio‑frequency equipment ‑ Electromagnetic
disturbance characteristics ‑ Limits and
methods of measurement
RE
30‑1000MHz
Full SAC 10m
CISPR 14‑1 (2005)
Electromagnetic compatibility ‑ Requirements
for household appliances, electric tools and
similar apparatus ‑ Part 1: Emission
RE
30‑1000MHz
Full SAC 3m,5m,10m
CISPR 14‑2 (2001)
Electromagnetic compatibility ‑ Requirements
for household appliances, electric tools and
similar apparatus ‑ Part 2: Immunity ‑ Product
family standard
RI
30‑1000MHz
Full SAC 3m,5m,10m
CISPR 16‑1‑4 (2004)
Specification for radio disturbance and
immunity measuring apparatus and methods
‑ Part 1: Radio disturbance and immunity
measuring apparatus
RE
30MHz‑18GHz
Full SAC/FAR 3m,5m,10m
CISPR 22 (2005)
Information technology equipment ‑ Radio
disturbance characteristics ‑ Limits and
methods of measurement
RE
30MHz‑6GHz
Full SAC/FAR 3m,5m,10m
CISPR 24 (1997)
“Information technology equipment ‑
Immunity characteristics ‑ Limits and methods
of measurement “
RI
30‑1000MHz
Full SAC 3m,5m,10m
Electromagnetic compatibility (EMC) ‑ Part
4‑3: Testing and measurement techniques
‑ Radiated, radio‑frequency, electromagnetic
field immunity test
RI
80 MHz to 6 GHz
COMPACT
CISPR 22 (2005)
Information technology equipment ‑ Radio
disturbance characteristics ‑ Limits and
methods of measurement
RE
30MHz‑6GHz
Full SAC 3m,5m,10m
CISPR 24 (1997)
“Information technology equipment ‑
Immunity characteristics ‑ Limits and methods
of measurement “
RI
30‑1000MHz
Full SAC 3m,5m,10m
TR 102‑273
Electromagnetic compatibility and Radio spectrum Matters (ERM); Improvement on Radiated Methods of Measurement (using test site) and
evaluation of the corresponding measurement
uncertainties; Part 3: Anechoic chamber with a
ground plane
RE
30MHz‑40GHz
Full FAR 3m,5m
ETS 300‑328
Radio Equipment and Systems (RES);
Wideband transmission systems; Technical
characteristics and test conditions for data
transmission equipment operating in the 2,4
GHz ISM band and using spread spectrum
modulation techniques
RE
30MHz‑4GHz
Full SAC 3m,5m,10m
Aircraft
RTCA DO 160
Environmental conditions and test procedures
for airborne equipment
RE/RI
100MHz‑6GHz
Mil‑ Std chamber
Military
Electronics
MIL‑STD‑461E requirement RE 103
Department of Defense: Requirements for
the control of electromagnetic interference
characteristics of subsystems and equipment Radiated Emissions, Electric Field, 10KHz to
18GHz
RE
80 MHz to 18 GHz
(Absorber only)
Mil‑ Std chamber
MIL‑STD‑461E requirement RS 103
Department of Defense: Requirements for
the control of electromagnetic interference
characteristics of subsystems and equipment
Radiated Susceptibility, Electric Field, 2MHz
to 18GHz
RI
80 MHz to 18 GHz
(Absorber only)
Mil‑ Std chamber
IEC 61000‑4‑3 (2006)
Telecom
Title of Standard
Table 1: Standards And Recommended Chamber Types
Conformity FEBRUARY 2008
Specifications
Title of Standard
General
Description
Frequency Range
Min Chamber type
Automotive
CISPR‑12 (2005)
Vehicles, boats, and internal combustion
engine driven devices ‑ radio disturbance
characteristics ‑ limits and methods of
measurement
RE
150KHz to
30MHz limits
have not been set
30‑1000MHz and
1000MHz‑18GHz under study
Standard 10m chambers
with large QZ diameters
can meet this standard.
CISPR‑25 (2005)
Limits and methods of measurement of radio
disturbance characteristics for the protection
of receivers used on board vehicles
RI
150KHz‑ 1000MHz
CISPR 25 type
ISO 11451
Road vehicles ‑ vehicle test methods for
electrical disturbances by narrowband radiated
electromagnetic energy
RE/RI
check equivalent
SAE standard
check equivalent SAE
standard
ISO 11452
Components‑ test methods for electrical
disturbances by narrowband radiated
electromagnetic energy
RE/RI
check equivalent
SAE standard
check equivalent SAE
standard
SAE J551
vehicle testing
‑2
Test limits and methods of measurement of
radio disturbance characteristics of vehicles,
Motorboats, and spark‑ignited Engine Driven
Devices
RE
30MHz ‑ 1000MHz
Full 10m
‑4
Test limits and methods of measurement of
radio disturbance characteristics of vehicles
and devices, broadband and narrowband,
150KHz to 1000MHz
RE
150KHz‑1GHz
CISPR 25 type
‑11
Vehicle Electromagnetic immunity‑Off vehicle
source
RI
100KHz‑18GHz
Full 10m
‑12
Vehicle Electromagnetic Immunity‑On board
transmitter simulation
RI
1.8MHz‑1.3GHz
Full 10m
‑13
Vehicle Electromagnetic Immunity‑Bulk
Current injection
RI
1MHz‑ 400MHz
Full 10m
SAE J1113
component testing
‑21
Electromagnetic compatibility measurement
procedure for vehicle components‑ immunity
to electromagnetic fields 10KHz‑18GHz
absorber lined chamber
RI
10Khz‑18GHz
CISPR 25 type
‑25
Electromagnetic Compatibility Measurement
Procedure for Vehicl3e Components‑Immunity
to radiated electromagnetic fields, 10KHz to
1000MHz
RI
10KHz‑1000MHz
CISPR 25 type
‑41
Limits and methods of measurement of radio
disturbance characteristics of components and
modules for the protection of receivers used
on board vehicles
RI
150KHz‑1000MHz
CISPR 25 type
2004/104/EC European
Automotive Directive
Vehicle and Component testing
ANNEX IV ( CISPR 12 )
Method of measurement of radiated
narrowband emissions from vehicles
RE
30‑1000MHz
Full 10m
ANNEX V (CISPR 12
and 25 )
Method of measurement of radiated
narrowband emissions from vehicles
RE
30‑1000MHz
Full 10m
ANNEX VI (ISO 11451‑2)
Method of testing for immunity of vehicles to
electromagnetic radiation
RI
20MHz‑2GHz
Full 10m
ANNEX VII (CISPR 25 )
Method of measurement of radiated
broadband electromagnetic emissions from
electrical/electronic sub‑assemblies
RE
30‑1000MHz
CISPR 25/ SAR 3m
ANNEX VIII ( CISPR 25 )
Method of Measurement of radiated
narrowband electromagnetic emissions from
electrical/electronic subassemblies
RE
30‑1000MHz
CISPR 25/ SAR 3m
ANNEX IX (ISO 11452 )
Method’s) of testing for Immunity of electrical/
electronic sub‑assemblies to electromagnetic
radiation
RI
20‑1000MHz
CISPR 25 type
Table 1: Standards And Recommended Chamber Types continued
Conformity FEBRUARY 2008
document. There is much discussion currently at CISPR 16
level concerning the antenna calibration under the draft CISPR
16‑1‑5 document and also the NSA method itself and the so
called RSM (Reference Site Method ) that differs in some
details to the NSA method fundamentally.
The CISPR 16‑1‑4 chamber validation method <1GHz
for FARs has incorporated much of the well known work
previously developed by CENELEC and is known as the Free
Space NSA or FSNSA method. This is again a volumetric
test with a pair of antennas but differs from NSA in that it
specifies the use of a small biconical transmit antenna for
the whole range 30‑1000MHz and the receive antenna must
be the same as used subsequently for the product tests. The
transmit antenna is placed at 3 different heights and 5 different
positions and the FSNSA measured at 3 or 5m separation. In
addition, both antennas are allowed to be tilted towards each
other thus allowing the use of one individual antenna factor –
much more simple than the multiple geometric factors
required for NSA tests. The pass criteria is again +/‑4dB and
so far most chambers are showing similar behavior as a FAR
(with floor absorber) or SAC with (Ground plane) in that a
compliant 3m SAC will also be a compliant 3m FAR with the
same true for the pre‑compliant smaller chambers
An additional development for CISPR 16‑1‑4 was the recent
publication of a chamber validation method for sites above
1GHz called the Site VSWR or SVSWR test. This method has
implications for existing chambers because it will test them
more severely than the Free Space Transmission Loss (FSTL)
method that has been commonly used until now. One of the
key differences is the SVSWR’s use of an isotropic source
antenna in conjunction with a broadband horn receive antenna
compared to FSTL’s use of typically two horn antennas so
that SVSWR exposes more of the chamber to the antenna
beam. Several different positions are measured by moving
the antenna along the line of sight to specified test points and
then calculating the SVSWR of the data group at that position
with the criteria now being SVSWR < 6dB to pass. Another
point to note with this test is that the floor absorbers cannot
be higher than 30cm. Experience with this measurement is
currently somewhat limited but the implications so far are that
compliant FARs will pass easily as long as the floor anechoics
mirror the ceiling anechoics. SACs will pass as long as any
partial lining designs with hybrid absorbers increase the
coverage of hybrid and use 30cm microwave absorbers on the
floor.
This test ensures that EUTs are subjected to known field levels
during the radiated immunity tests. Absorbers are partially
required on the floor for this test and should be removed when
carrying out the radiated emissions tests. Hybrid absorbers
on rolling carts are typically used and moved off to the side
walls when not in use, thus making the change from RE to
RI testing practical and fast. When RE testing is not required
some chambers are left as FAR with full floor coverage.
Both the compact and full compliance chambers will typically
pass these criteria without problem using ferrite or hybrid
absorbers. It should be noted however that the relatively new
standard for medical devices, IEC 60601‑1‑2 (a derivative
of IEC 61000.4.3) now requires testing to 2.5GHz which, in
effect, becomes 3GHz (also for the combination antennas). So
now most compact chambers will need to use hybrid materials
and not just the ferrite‑only design whose performance drops
off from 1GHz onwards.
Further, as a result of its extension to 6GHz, IEC 61000.4.3
has now developed procedures for validating sites above
1GHz involving the use of smaller test planes–windowing–or
multiple transmit positions to take into account the directive
nature of the antennas at higher frequencies.
IEC – CISPR Joint Task Force on Fully Anechoic Rooms
(DRAFT IEC 61000.4.22) At this point it is important to
note that the IEC and CISPR have created a group from IEC
TC77B and CISPR A to create a draft document describing
methods of measurement in Fully Anechoic Rooms. The idea
is to simplify the logistics of the measurement such that there
is one single chamber validation test instead of two different
tests, one for emissions and one for immunity. The proposal
is quite radical and will evolve over time until its proposed
publication date sometime in 2010.
Radiated immunity requirements are covered under the basic
standard IEC 61000.4.3, with various other standards (i.e.,
CISPR 24 ITE immunity, IEC 60601‑1‑2 2001‑medical)
referring back to this standard. The measurement of field
uniformity is carried out over a specified test area (typically
1.5 x 1.5m), made up of 16 points at 80cm above the floor.
The test criteria is for 75% of the 16 points to be within 0‑6
dB of each other.
Compact Chamber
Conformity FEBRUARY 2008
Telecommunications
Typical chamber type: SAC/FAR, Compact, Full 3m
Typical frequency range: 30MHz‑18GHz
This market has been one of the most active in the last five
years in line with the growth in the mobile phone market, with
a significant number of chambers being built and used either
for product development or type approval testing. In Europe,
EMC testing comes under the R&TTE directive, while in the
U.S. testing falls under FCC Part 15 and Bellcore regulations.
Although there are many variations on basic tests due to
specific product functionality tests, most of the type approval
requirements remain under the methods described in CISPR
22 and CISPR 24 and related IEC 61000 series standards.
The chamber requirement for compliance will be a full
3m SAC or FAR designed for 30MHz‑18GHz for radiated
emissions and 80MHz‑18GHz for radiated immunity with
removable absorber on the floor (SAC)
Product development and R&D activity will generally choose
a fully anechoic compact chamber using the chamber for
pre‑compliant radiated emissions work and compliant radiated
immunity.
Specific product standards developed by ETSI define many
other test methods in addition to those mentioned above and
the only one that changes the chamber requirements is the
transmitter spurious emissions that can be found in a vast
majority of the large number of different standards ETSI has
produced. The requirements of the anechoic chamber define a
fully lined chamber for 3 and 5m spurious emission tests using
1m absorber specified against the following table of frequency
vs. performance.
‑10dB ‑20dB ‑30dB 30‑100MHz
100‑300MHz
300MHz‑10GHz
The strict application of this requirement does not therefore
always take place due to ETSI’s poor definition of the
absorber and chamber requirements, since there is no
definition of how to measure these values. If this is applied
to the letter, and the above values are interpreted as absorber
reflectivity values, it will mean hybrid absorber of 1000mm in
length. This then becomes a large full 3m chamber, but may
require rethinking the treatment of the floor which would be
difficult to remove for the basic ground plane tests required by
CISPR 22. If however it is allowed to interpret these values
in a more flexible way by measuring them using the very old
industry method “Termination VSWR” then these values
can be achieved with a 1m long pyramid and the chamber is
therefore a lot cheaper. The “Termination VSWR” method
is basically a slotted line technique that was used in the dark
ages of EMC chamber testing some 20 years ago and can
produce some very attractive looking data. Unfortunately
Conformity FEBRUARY 2008
for many who choose this more attractive route they soon
discover that product testing below 100MHz can be limited
since 1m pyramidal absorbers do not work well enough,
something we also knew 20 years ago.
Automotive
Typical chamber type: Small partial lined to full 10m
Typical frequency range: 80MHz‑40GHz
Automotive requirements tend to make chamber design
variable due to the large number of standards available and the
need to test at component or vehicle level. As one solution,
General Motors recently decided to build to the hardest
common denominator for testing, which is the 10m emission
test, and also cater for the next 20 years of standards changes.
Now they have 4 identical 10m chambers. Although this is
probably an extreme case, it is an example of the dilemma that
manufacturers face. At the same time, every manufacturer will
still have its own unique requirements which are often very
difficult to meet.
Of recent note is the Automotive EMC Directive, requiring
car accessories to meet the CE mark requirements. This has
been a source of discussion between automotive and telecom
manufacturers for some time at ETSI in order to avoid overlap
of testing under the R&TTE and the Automotive EMC
Directives. Such issues will increase in this market as the
products become more sophisticated.
The automotive standards these manufacturers apply are
basically quite simple, with the most common coming from
CISPR, SAE, and the ISO. These standards are usually copies
of each other, with small differences.
• CISPR 12 Vehicles, boats, and internal combustion engine
driven devices ‑ radio disturbance characteristics ‑ limits
and methods of measurement
Typical chamber type: Standard 10m chambers with large
QZ diameters meet this standard.
Typical frequency range: (150KHz) ‑30MHz‑1000MHz
The 10m emission testing locates the antenna 10m from
the outer shell of the vehicle. The antenna is not scanned
but located at 3m height (for 3m testing, the antenna is
located at 1.8meters). Both sides of the vehicle and both
polarizations are tested and the antenna is to be in line with
the middle point of the engine compartment.
A two antenna position chamber makes the test much
easier. A monopole is used for the range 150KHz to
30MHz, and only vertical polarization measurements
are made. For 30MHz to 200MHz, a biconical antenna
is used, and the log periodic is used for the range
200MHz‑1000MHz. Alternatively, tuned dipoles can be
used for the entire range.
An RF Chamber for Evaluating Automotive Components and Systems
Though some people fondly remember the days when
automobiles had relatively few electronic components,
and a disabled vehicle could often be fixed with nothing
more than a wrench, the fact is that today’s cars and
trucks are safer, more fuel efficient, and significantly
more reliable than their purely mechanical ancestors.
Electronic systems enhance the performance of the engine,
transmission, steering and brakes, while helping to keep
the driver alert and informed. New vehicles rely on dozens
of microprocessors and miles of wiring to meet stringent
government standards for safety and fuel economy, while
also satisfying various consumer requirements for comfort,
convenience and style.
Integrating dozens of electronic systems in the confined
space of an automobile along with a growing number of RF
transmitters and receivers for communications, navigation
and entertainment presents a considerable electromagnetic
compatibility challenge. So when Clemson University
was setting up their International Center for Automotive
Research (CU-ICAR) in Greenville, South Carolina, the
ability to make EMC-related measurements was a high
priority. However, the design of this facility presented some
unique challenges.
For example, university research often requires highly
accurate measurements of well defined sources under
tightly controlled conditions. Commercial automotive
EMC tests, on the other hand, require a wide variety
of test equipment and environments. Furthermore, the
devices under test in automotive applications range from
small electronic modules a few centimeters long and
weighing a few ounces to vehicles that are several meters
long weighing over a ton. The operating frequencies
of automotive systems range from DC to tens of GHz.
Clemson needed to design a test facility capable of
performing the widest possible range of automotive EMC
tests with a limited budget and a fixed amount of space.
The original plan was to devote more than two thirds of
the 50’ x 30’ electronics lab floor space to an RF chamber
large enough for full vehicle testing. However, it soon
became apparent that it was desirable to perform many
measurements outside the chamber and that the extra
space inside the chamber would not be well utilized.
Clemson then worked with ETS-Lindgren to design a 22’ x
30’ chamber that required less than half of the available lab
space, but was still large enough to hold a mid-sized car.
The resulting chamber is a modified version of a standard
3-m EMC test chamber with a reinforced floor and a large
air-driven sliding door. The chamber is sunk into the floor
Conformity FEBRUARY 2008
of the lab so that the raised metal floor of the chamber is
level with the concrete floor in the lab.
Clemson also worked with a variety of test equipment
vendors to come up with a list of antennas, analyzers,
amplifiers, cables, probes and test fixtures to support the
testing of greatest importance for automotive research.
Specific equipment required for tests relevant only to
certain automotive OEMs was not purchased. Instead, more
flexible test equipment capable of simulating a wider variety
of test conditions was chosen. Although some of this test
equipment is more expensive than the application specific
equipment it replaces, the overall equipment costs were
significantly below what it would have cost to support all
possible automotive EMC test procedures.
As a result, the Clemson facility (which is scheduled to
be completed in the spring of 2008) will be capable of
simulating the conditions of most commercial automotive
EMC test procedures while also supporting FCC and CISPR
EMC tests, antenna measurements, model validation and
research projects at frequencies from DC to several GHz
and higher. All of this at a cost of just under $1.5 million
(USD).
Electronics and effective electronic systems integration
will be a key factor affecting the success or failure of
automotive companies in the coming years. An ability to
make accurate and meaningful EMC measurements can
make a tremendous difference in the cost and reliability of
automotive systems. Anyone who is currently planning
to build or upgrade an automotive electronics laboratory
is welcome to contact Professor Todd Hubing at Clemson
(Hubing@clemson.edu), who will be happy to share
his experiences and discuss the trade-offs involved in
designing and equipping an automotive EMC test facility on
a budget.
• For 10m testing, the antenna is located 3m over ground and
it is not scanned. The antenna is 10meters from outer skin
of vehicle and in line with engine mid point. Both sides
of vehicle are tested. For 3m testing, the antenna is placed
at 1.8m, and both horizontal and vertical polarizations are
measured.
• CISPR 25 Limits and methods of measurement of radio
disturbance characteristics for the protection of receivers
used on board vehicles
Typical chamber type: 7.1x6.85x4.3m (36” pyramidal
absorber only)
Typical frequency range: 70MHz‑1000MHz
Testing to this standard requires an absorber lined chamber
where the absorption of the material has to be better than
6dB for the range 70MHz 1000MHz. For the chamber
testing procedure, a monopole is used for the range
150KHz to 30MHz, a biconical antenna is used for the
range 30MHz to 200MHz, and the log periodic is used for
the range 200MHz‑1000MHz. Whole vehicle testing used
to see how the radio or radios in the car are affected by
the different systems in the vehicle (for example, how the
radio is affected by windshield wipers).
A CISPR‑25 chamber can be used for EU, SAE and ISO
automotive standards. The chamber is meant to be used
for automotive component testing; with proper floor
reinforcement it can be used to test whole vehicles as
indicated in the standard document
• SAE J551 Vehicle testing and ISO 11451
Typical chamber type: Full 10m
Typical frequency range: 10KHz‑18GHz
For 10m testing, the antenna is located 3m over ground
and it is not scanned. The antenna is 10meters from outer
skin of vehicle and in line with engine mid point. Both
sides of vehicle are tested. For 3m testing antenna is placed
at 1.8m and both horizontal and vertical polarizations are
measured.
There is no absorbent material between antenna and EUT.
The antenna is placed at least 2m from the vehicle engine’s
center point, the uniformity plane is horizontal, and it is a
1.5 diameter circle where the field for frequencies above
200MHz is between +‑3dB for 80% of the frequencies.
• SAE J1113 Component testing and ISO 11452
Typical chamber type: Similar to that of CISPR25
Typical frequency range: 10KHz‑18GHz
An absorber lined chamber is required. Antennas and field
generator to cover the range are required. No need to scan
10 Conformity FEBRUARY 2008
antenna, a test bench is required, DUT is placed on the
bench with the wiring extended to a LISN, the antenna is
placed at 1m distance.
• 2004/104/EC
Typical chamber type:
°
°
Annex 1‑6: Standard 10m chambers with large QZ
diameters can meet this standard.
Annex 7,8,9: CISPR 25 chamber is enough
Typical frequency range: 10KHz‑18GHz
The absorber lined shielded enclosure must be able
to correlate to OATS, and NSA measurement of the
chamber meeting the +4dB should be acceptable. For 10m
testing, the antenna is located 3m over ground and it is
not scanned, the antenna is 10meters from outer skin of
vehicle and in line with engine mid point. Both sides of
vehicle are tested. For 3m testing, the antenna is placed
at 1.8m and both horizontal and vertical polarizations are
measured.
The antenna is placed no less than 1.5m above the
ground and no less than 2 meters from the center of the
engine. The field uniformity requirement is that points
on a horizontal line of 1m in length, perpendicular to the
antenna line of sight, must be within 6dB
Military
Typical chamber type: Small with partial lining of
microwave absorbers
Typical frequency range: 80MHz‑40GHz
Military standards will vary from country to country and
will have little to do with current civilian standards, and are
published by respective departments of defense. However the
chamber requirements are typically quite simple
• MIL‑STD‑461E: Requirements for the Control of
Electromagnetic Interference Emissions and Susceptibility
• There is no chamber specification or validation procedure.
Absorber specification is 6 dB @ 80 MHz, rising to 10
dB 250 MHz to 40GHz. A 60cm/ 24 inch microwave
absorber used partially will fulfill the requirements of this
specification.
• The basic chamber size will be approx. 6.1x6.1x3.7m
Aerospace
Typical chamber type: Small, partial lining of microwave
absorbers; same as Mil Std
Typical frequency range: 80MHz‑40GHz
Radiated field EMC testing on full scale EUTs, such as
aircraft, is rare and often carried out in the open at the edge
of military airfields. Most testing is carried out at component
level according to the RTCA DO 160D.
• Main configuration is similar to the MIL STD 461E with
a ground plane table in a relatively small chamber. RF
Absorbers are installed to minimize the reflections from
the shielded room.
• RS testing is carried out from 100MHz‑18GHz at different
field strength levels although mode‑stirred / reverberation
chambers are becoming more common due to the high
levels of Electric field ( 200V/m) required for some
equipment.
• RE testing is carried out from 1MHz 6GHz. Typical
antennas used are Rod antenna 10KHz 30MHz, biconical
30‑200MHz and log spiral 200MHz‑1GHz and Horn >
1GHz.
• There is no chamber specification or validation procedure
only a minimum RF absorption for the RF absorbers that
partially line the shielded room.
°
100 to 250MHz ‑6dB
°
Above 250MHz –10dB
Conclusion
This overview of EMC anechoic chambers cannot be
exhaustive, in particular for the automotive standards,
but should give a general perspective to EMC engineers
unfamiliar with this technology on the range of different
facilities that exist today.
There are many current developments within the standards
with the most significant coming out of CISPR 16 leading
to new procedures above 1GHz and alternative test methods
using FARs and these should be monitored closely.
From this article the reader should understand that EMC
chamber design is basically quite well established and as a
consequence most manufacturers should be able to give a
potential EMC anechoic chamber customer a very accurate
idea of the chamber required for a given set of
specifications.
Martin Wiles is product manager for the Anechoic and
EMCO product lines at ETS‑Lindgren, in Stevenage, England.
He is a UK delegate for CISPR A and the IEC‑CISPR Joint
Task Force on Fully Anechoic Rooms. He can be reached by
e‑mail at martin.wiles@ets‑lindgren.com.
11 Conformity FEBRUARY 2008
Download