Anechoic Chambers 4/11/05 11:08 AM Page Feature Article Brian F Lawrence Anechoic Chambers, Past And Present his article will review the historical development of absorber materials and anechoic chambers, which play an important role in the work of today’s EMC test engineer We will discuss early attempts to achieve correlation between Anechoic Chambers and Open Area Test Sites and trace the improvements made through the years that resulted in current industry practice T Different industries and regulatory authorities place different priorities on emissions in comparison with susceptibility If a home computer malfunctions it can be inconvenient, but if an automobile, or worse, an aircraft were to malfunction it could be disastrous On the other hand, if a mass-produced item must be removed from store shelves as a result of regulatory spot checks, this could be a different kind of disaster — an economic one — for the manufacturer and retailer Responsible companies and independent test laboratories, therefore, responded to the emerging EMC Regulations by developing their own EMC testing capabilities, including the construction of OATS facilities and RF Shielded chambers The ideal OATS, as defined in the standards, is practically impossible to create, although with the right location and careful design, there are now a number of near perfect OATS facilities in operation Typical problems associated with the use of an OATS could include; ambient RF interference, poor grounding conditions, inclement weather, remote locations and testing time limited to the daylight hours If weather protection is provided, dielectric reflection from wood or plastic walls, as well as reflection from wiring and lighting, are also of concern While theoretically an RF shielded chamber could solve some of these problems, its imperfections will result in internal surface reflections, cavity resonances, yielding poor site attenuation (in the case of emissions tests) and non-uniform field conditions (in the case of susceptibility testing) Lining the internal surfaces of RF Shielded chambers with an ideal absorbing material would have the effect of simulating OATS conditions within a convenient, indoor, weather protected test chamber An RF anechoic chamber could become an ideal EMC test site, useful for both emissions and susceptibility tests, if the absorber materials can adequately eliminate internal surface reflections over the test frequency range Producing such absorber material was the challenge presented to the anechoic chamber industry during the 1970’s Absorber materials were not new They had been used in anechoic chambers for many years to create test facilities for radar and microwave antenna evaluation Absorbers were typically manufactured by impregnating conductive carbon into a foamed plastic medium, such as polyurethane or polystyrene These carbon-impregnated materials were fashioned into tapering wedge and pyramid shapes to provide a suitable impedance match between free space and the resistive absorber medium Balancing the carbon content with the shape of the tapering material provided efficient and predictable FEBRUARY 2005 CONFORMITY® www.conformity.com As international regulatory agencies introduced RF emission and susceptibility requirements and standards in the 1970’s and 1980’s, the need to make accurate EMC tests gained increasing importance Regulatory standards define not only the permitted characteristics of the equipment under test (EUT), but also the test procedures and the calibration of the test equipment and test facility Only by addressing all of these points can the standards foster correlation between measurements made at different locations and times by different engineers using different instrumentation In general, standards on the measurement of radiated electromagnetic emissions prescribe the use of open area test site (OATS), while those concerned with RF susceptibility define an RF Shielded environment in which a uniform field can be established Surrounding the susceptibility test area with an RF shield is necessary to prevent the test, which deliberately creates strong radiated signals, from interfering with communications outside of the test area Anechoic Chambers 4/11/05 11:08 AM Page Feature Article absorption of RF energy from Microwave frequencies to below 500 MHz, where the tapered length of the absorber would be greater than one wavelength The lower frequency of good absorption performance was strongly related to the length of the absorber (and still is, for conductive foam pyramidal units) Extrapolating this established technology down to 30 MHz and below was the basis of many early EMC Anechoic Chambers of the 1980’s Pyramidal absorbers six feet, eight feet, and even up to twelve feet long were produced and installed in large RF shielded chambers with mixed success Not only were there physical problems in manufacturing, handling and installing these materials, but new methods of factory quality testing also had to be developed in order to make meaningful and reliable production tests on such large pyramidal shapes and at frequencies down to 30MHz www.conformity.com Multi-national corporations such as IBM and Hewlett Packard were particularly interested in taking advantage of anechoic chambers for their EMC Test Programs Available OATS facilities, often remote from their manufacturing plants, could not keep up with their demanding test schedules The RF chamber industry responded and in1982 the first full size, 3-meter range, EMC Anechoic Chamber was built for IBM in Boca Raton, Florida This chamber’s site attenuation was tested according to the site attenuation methods developed for ANSI C63.4 and accepted by the FCC as modeling open area test site performance and suitable for testing to the FCC Part 15 Rules The chamber was designed and built by Ray Proof at a cost of almost $2M (two million dollars) and required Ray Proof’s Absorber Division to install a 50 foot long, walk-in waveguide to test the foot long pyramids of foam that lined the walls and ceiling of the IBM Chamber More 3m range anechoic chamber installations followed the success at IBM, but this chamber performance and Absorber technology did not conveniently scale up to a 10m range length, desirable for testing Class A computing devices Something different was needed The next step in chamber development came as the result of a partnership between customers, industry, and academia Funding provided to the University of Colorado at Boulder by IBM and Ray Proof resulted in the development of a numerical model for absorber materials The model used a homogenization principle to simulate a distribution of pyramidal shapes as a series of layers having different impedances This absorber model gave industry the capability to design and build Absorber Materials with much improved performance in the VHF band A chamber simulation program was also developed that imported the material performance files from the absorber models to predict the field conditions that would exist in the final chamber construction This new tool allowed design engineers to optimize chamber shaping and absorber layout to provide the desired OATS equivalence In parallel, the chamber industry had to design and install more sophisticated and accurate test equipment able to verify the actual performance of the optimized absorber designs A huge ft square coaxial line, having a ft square center conductor was installed at Ray Proof, able to measurement the low frequency reflectivity of absorbers in groups of units This original Ray Proof test system, together with an array of more modern systems instrumented with network analyzers is installed at the ETS-Lindgren absorber plant in Durant, Oklahoma Anechoic Chambers that could meet both the 3m and the 10m range OATS characteristics of site attenuation according to such standards as ANCIC63.4 and CISPR16 became available by 1990 However, they were monsters, large and expensive, and outside the economic range of the majority of potential customers In 1969 the University of Tokyo patented the use of ferrite tiles in EMC Anechoic Chambers Sintered ferrite tiles, only a few millimeters thick, can exhibit excellent absorption properties at frequencies below 100 MHz By the late 1980’s many ferrite tile lined chambers were being used in Japan as EMC Test Sites The great advantage of the ferrite tile technology was that chambers could be dramatically reduced in size The surrounding shield did not have to be sized to CONFORMITY ® FEBRUARY 2005 Anechoic Chambers 4/11/05 11:08 AM Page Feature Article accommodate a large volume of thick absorber lining in addition to the active test volume However, ferrite was still a very expensive material to produce, and was even more expensive when the tiles were packed and shipped to sites outside Japan When the original ferrite tile patent expired in the mid 1980’s, competitive pressures reduced the cost of a ferrite lined chamber Again, it was IBM who in 1986 became the first company in the U.S.A to install a high performance 10m range chamber using ferrite tile technology at their facility in Austin, Texas By the 1990’s ferrite tile suitable for EMC test chambers were being produced by several companies in Asia, the U.S.A and Europe The original absorber numerical modeling programs and their later derivatives had been modified to include ferrite parameters together with dielectric matching layers As a result, a new generation of optimized, hybrid absorbers combining the best features of ferrite and conductive foam could be designed and applied to EMC chambers Having reached this point of development with the EMC Test Chamber, the focus on improving site attenuation correlation to the Normalized Site Attenuation, NSA, of an ideal OATS has moved on from the absorber and chamber to the calibration As the EMC practice has evolved, so too has chamber test site design For less than a $100K investment, companies can now own and operate a compact 7m x 3m x 3m precompliance EMC Chamber Facilities like this demonstrate +/- 6dB correlation to NSA at low frequencies and within +/- 4dB at frequencies from 100 MHz to millimeter wave frequencies Slightly larger chambers can be designed to demonstrate +/-4dB correlation to NSA for smaller EUTs and with reduced scanning height of the antenna For Engineers who are evaluating product susceptibility and who self certify product emissions, such pre-compliance chambers provide ideal indoor test site convenience The next step up from the precompliance chamber is the fullcompliance, 3m-range facility Such a chamber which cost IBM $2M in 1982 is available today in a much reduced 9m x 6m x 6m size, or even smaller, offering exceptional performance, for around $300K At the higher end, base models of 10m range anechoic chambers start below $1M Emerging test requirements have lead to chamber designs that offer specialized or combination test capabilities These include, for example, EMC and wireless testing in a 3m to 5m range length according to ETSI standards and special chambers for automotive test applications according to CISPR-25 For the ultimate susceptibility testing of complex and large EUTs there are also fully qualified, non-anechoic reverberation chambers Today there is a test chamber for almost every EMC test requirement, making the emulation of a wide variety of test conditions possible EMC test engineers can now take chamber anechoic performance for granted and concentrate on selecting other chamber features and accessories The optimal choice of antenna frequency ranges, antenna patterns, equipment handling ramps, hoists, towers, turntables, automated sliding doors, and other accessories will improve the ease of use, lower the cost of ownership, and allow the chamber user to take full advantage of the performance that the modern chamber can deliver About The Author Brian Lawrence is the Director of Sales & Marketing for ETS-Lindgren, Europe Prior to the sale of Lindgren RF Enclosures, Inc to ESCO Technologies Corporation in March of 2000, Brian Lawrence was responsible for Lindgren’s EMC Test Chamber business worldwide Brian Lawrence has over 40 years experience in Anechoic Chamber and Absorber Material development and has worked for Ray Proof USA and Ray Proof UK during his career FEBRUARY 2005 CONFORMITY ® www.conformity.com Chamber simulation programs, incorporating hybrid absorber models have been responsible for the modern generation of EMC Anechoic chambers These chambers cannot only meet OATS standards but will beat almost every actual OATS site in terms of correlation to the theoretical ideal site model across the entire test frequency range Typical regulatory standards require an acceptable OATS test site to demonstrate site attenuation correlation to the ideal model within +/- 4dB Modern 10m and 3m range chambers available from ETS-Lindgren are guaranteed to correlate to within +/3 dB of the ideal standard, using optimized hybrid absorber technology and design of the Antennas that will be used in the chambers Anechoic Chambers 4/11/05 11:08 AM Page ... eight feet, and even up to twelve feet long were produced and installed in large RF shielded chambers with mixed success Not only were there physical problems in manufacturing, handling and installing... plant in Durant, Oklahoma Anechoic Chambers that could meet both the 3m and the 10m range OATS characteristics of site attenuation according to such standards as ANCIC63.4 and CISPR16 became available... standards and special chambers for automotive test applications according to CISPR-25 For the ultimate susceptibility testing of complex and large EUTs there are also fully qualified, non-anechoic