From EMC Introduction Every reader knows regulatory compliance is a big deal. Compliance with EMI/EMC and safety performance requirements and standards is a gating step on the way to market. The types and amounts of equipment subject to EMI/EMC requirements has exploded in the past quarter-century. In response, a whole industry has arisen to serve this need. Today’s EMI/EMC compliance engineers have much more to do, but they also have more tools to help them. This article reviews the trends and changes that affected EMI suppression components over the past 25 years.
Many compliance engineers think of the FCC’s 1979 Part 15 requirements on digital device emissions and authorizations as the “big bang” which got everything started. While the imposition of EMI limits in the world’s largest consumer market heightened the importance of emission control, it wasn’t the first instance of commercial EMI regulations. Germany (then West Germany) had had been using the FTZ/VDE 0875/0871 emissions requirements for some time.
Large multi-national companies such as IBM, Honeywell, DEC, Sperry, and Burroughs who sold products in Germany were well aware of EMC design issues. In addition, there were many companies who had experience with EMI/EMC issues because they were involved with military procurement. However, smaller commercial companies—many of them new ventures—unfamiliar with the German market had to rush to get up to speed with the new FCC requirements.
Fortunately, there was a reservoir of electrical and mechanical design engineers with military EMC experience who moved to the commercial sector, as well as a growing cadre of EMC consultants offering design services and educational seminars. Component vendors also leapt into the breach created by the new commercial demand, offering their products, selection guides, and application support.
The solutions developed by EMI/EMC engineers can be broadly conceived of as belonging to two related areas: design techniques and components. Design techniques include issues such as layout, board stackup, bypassing, clock shaping, component placement, filtering, and shielding. EMI/EMC components supply the means of implementation.
Trends
Back in 1979, the experience gained from military applications shaped some of the approaches to EMI containment. However, often the military and commercial applications differed substantially. In the military, cost was often less of a design issue than performance. Also, construction practices differed. Many military products were fully enclosed with many fasteners. This was partly due to environmental concerns, but also due to the fact that immunity to strong signals over a wide frequency range was a necessity in military applications. At the dawn of the commercial EMI era, emissions control was the only concern. Thus, most equipment was put together with a minimal of fasteners and clocking speeds were low (remember the 4 MHz 8088 used in the first IBM PC?). In addition, many plastic enclosures, rather than ones made from metal, were far more common in the commercial arena.
Hence, a number of EMI shielding solutions available for military use were not cost-effective, or were too high-performance, or not easily applied to commercial products. This was particularly true for conductive EMI gasketing, where continuous contact and moisture exclusion were not commonly required for commercial units, and in the areas of conductive coatings.
Over the years, some of the differences in the two environments have narrowed, although the commercial emphasis on price has never gone away, and never will. Clock and data rates have increased, forcing shielding to perform at higher and higher frequencies. A similar widening of the frequency range for viable EMI/EMC solutions has been caused by the introduction of commercial immunity requirements, particularly those of the European Union.
Let’s look at some of the changes that have occurred.
EMI Gasketing In 1979, EMI gaskets were used extensively in military shielding applications. The readily available EMI gasketing materials included spring finger stock, most commonly made of beryllium copper, knitted wire mesh or mesh-over-foam gaskets, and conductive elastomers (usually silicone rubber loaded with conductive particles of various compositions). However, these were not heavily used in commercial equipment for several reasons: (1) They often were unnecessary, because the lower operating frequencies of commercial equipment allowed for adequate shielding even with relatively large gaps (infrequent enclosure contact points). (2) They were expensive. (3) Some had mechanical problems (closure force, galvanic incompatibility) when used in commercial applications. Further, the commercial market didn’t want to pay for more performance than was necessary to meet the FCC/VDE requirements. For example, conductive elastomers not only provided excellent conductivity and shield closure, but also provided a hermetic seal which was unnecessary in indoor applications.
Recognizing these issues, and the potential of a burgeoning shielding market, EMI gasket manufacturers began to introduce commercially focused products. Conductive elastomers began to appear with carbon loading instead of silver, new standardized profiles and hollow extrusion shapes were developed, and a whole new type of gasket line eventually developed with conductive fabric over foam gaskets. Fingerstock manufacturers varied the profiles and materials of their products. Stainless steel started to be used, along with a variety of galvanically friendly platings such as nickel and tin. Attention was also paid to closing force, so assembly of gasketed units would be easier and not deform large, light gauge metal panels.
Today, conductive fabric over foam gaskets are probably the most prevalent type used in indoor commercial equipment. For outdoor applications, the predominant choices are either a combination of an outer non-conductive gasket for sealing combined with an inner conductive gasket that is not moisture resistant, or a nickel/graphite based conductive elastomer gasket.
Shielding: Conductive Coatings If shielding is required for emissions containment (or for that matter immunity), it has to be provided by conductive surfaces. Where sheet metal can be used, the issue is how to mate the different pieces with compatible coatings, fasteners, gaskets, and where necessary, screened or honeycombed ventilation panels. Where the enclosure must be plastic, conductive coatings can be applied.
Initially, most commercial conductive coating was provided by nickel-loaded paints. Carbon and silver-filled paints were available in 1979 to 1980, but the former was ineffective due to its rather high resistivity, and the latter was prohibitively expensive—this was the time of the great silver bubble. Copper filled paints were in their infancy, as their longevity was in question. Soon, paints featuring stable, non-oxidizing copper filler as their main conductive component came to dominate the market. Copper is several times more conductive than nickel, and hence provides better shielding. Some of today’s copper paints also employ silver plated copper.
In addition to paints, there has been activity in several other types of conductive coating technologies. Flame-sprayed zinc has been around since the beginning of this era, and is still used occasionally. Vacuum metallized plastic has seen occasional bursts of popularity, but is not often seen today as a shielding method. Part of the problem stems from the fact that in the 1980’s many metallized parts used aluminum in film thicknesses that were too thin to provide long term reliable conductive contact. However, one new non-painting surface coating technology has arisen which works very well: electroless plating.
Shielding: Conductive Laminates The use of EMI shielding laminates is something of a rise and fall story. Shielding laminates consist of a copper or aluminum foil laminated to a dielectric material. The dielectric provides structural support for the thin foil, and sometimes imparts additional properties such as insulation, abrasion resistance, or heat tolerance. Although available as rectangular sheets for prototyping and experimentation, for production purposes they are often pre-creased and pre-cut so they can be used as components of a shielding enclosure or as an auxiliary ground plane. One virtue of laminate solutions was that the tooling required to produce them was relatively inexpensive. Labs could prototype them with a pair of scissors and some copper tape. Often, “steel rule” dies were made to stamp them out.
Since the requirements for shielding in the early days of EMI were relatively modest, a number of small regional companies built up businesses supplying them. Eventually this solution fell from favor, because the shielding requirements for commercial equipment increased over the years.
Magnetic Shielding Magnetic field shielding is not an area that has seen dramatic changes over the years. One reason is that regulatory magnetic shielding requirements are modest. The number of standards and products affected by magnetic field requirements is not large. Although the early German VDE requirements imposed magnetic field requirements, as did the military standards of MIL-STD-461-A, the FCC and later the European Union’s EN 55022 did not. It should be mentioned that in the past decade, concern over the possible adverse health effects of low frequency magnetic fields has led to increased shielding of CRT monitors.
A second reason is that most equipment does not require specialized magnetic solutions. At high frequencies, the same shielding techniques that work for electric fields—conductive surfaces—tend to be effective for magnetic fields, because wave impedances are modest and skin depths are small. Problems that are specifically magnetic field oriented in nature and also require special measures for solution usually occur at switching or CRT monitor sweep frequencies and their harmonics, ranging from tens of kilohertz to about one megahertz. Solutions in this range do require specialized high permeability magnetic materials.
A magnetic shielding material industry existed prior to 1979, and continues to exist today. However, this segment of the shielding industry has not been subject to the dramatic pressures and demands caused by increasing computational speed and density, and hence it has not changed as much as the others. The sources of low frequency magnetic fields have not changed in the dramatic way that sources of high frequency RF signals have. Many of the manufacturers of magnetic shielding materials in existence then continue to do business today.
Power Line Filters The FCC rules had a dramatic impact on the demand for power line filters and the fortunes of their manufacturers. The requirement to control power line emissions mandated power line filtering, especially where switching power technologies were employed. Manufacturers who had been oriented towards military applications, and those who had been offering products for the German VDE requirements (which extended to a lower frequency range) re-engineered their products to meet the FCC’s requirements and stepped up their marketing efforts. The increased volume led manufacturers to augment their offerings in several ways:
FCC test sites soon proliferated, and nearly all of them became potential selling points for recommending and verifying the line filter performance. A number of suppliers responded by developing the sample kits we see today, which make component selection at the test site much easier than it used to be.
When the European Union passed its EMC Directive, the EMI requirements of EN 55022 (then CISPR 22) became applicable. This standard set 150 kHz as the lower frequency bound for conducted emissions regulation. Soon, other countries, such as Australia and New Zealand also adopted equivalent requirements, and more recently, the FCC followed suit. Thus, a common frequency range for AC mains line conducted emissions requirements now exists world wide. This helps power-line filter manufacturers further streamline their product lines, and makes component selection easier for the EMC engineer.
Ferrites It didn’t take long for EMC designers to realize that ferrites could provide EMC suppression in two situations. First, they were useful as common mode chokes. Often they would be applied to exiting cables as sleeves to suppress common mode radiation. Second, they could be used as lossy in-line elements to lengthen rise time in signal leads.
The initial offerings of ferrites for EMC use were usually made by manufacturers who made broad lines of ferrites for other purposes, such as switching power transformers and inductors, or RF baluns. Materials which also provided attenuation in the RF range were pressed into service as cable sleeves and for in-line signal attenuation, beads with leads were made available as tape-and-reel components.
At first, the ferrite components offered for EMI use were relatively limited in number and form. Soon, the variety available increased. Special flat ferrites were designed for use with ribbon cables, or with holes so they could be inserted over connector pins. The real change, though, came more recently with surface mount technology. Now, surface-mount beads and common mode chokes are available in a wide variety of single and multiple-lead forms for signal and power lead use.
Another interesting use of ferrites that has developed over the past 25 years is the introduction and wide application of ferrite tiles as radiated emission absorbers in full and semi-anechoic chambers. Prior to 1979, military laboratories and test procedures had widely employed shielded screen rooms. These were inexpensive, but because of internal reflections and resonances, could not be correlated to the open-field radiated emissions tests mandated by the FCC, VDE, and other agencies. The RF anechoic chambers of that time consisted of a screen room lined with conductive foam cones. Unfortunately, for the chambers to work properly (i. e., for the cones to be absorptive) they had to be 6 to 8 feet in length at 30 MHz. This made for a huge, and very expensive structure if the performance of an open-field test site was to be replicated.
It was soon realized that ferrite tiles of the proper composition could be used to line screen rooms and achieve a reasonable level of absorption over the frequency range of 30 MHz to 1 GHz in a structure of much smaller volume. Considerable attention was paid to the problem of optimizing the ferrite tile absorber, in terms of the best material composition and mounting method (the spacing of the tile from the chamber wall, typically on a dielectric backing such as plywood is an important factor). Chambers lined with ferrite, or ferrite in combination with small carbon loaded foam cones to extend the upper frequency range, soon made indoor testing possible for laboratories and companies of modest means. These chambers also proved essential for delivering uniform fields in the RF susceptibility tests later developed by the IEC (IEC/EN 61000-4-3).
Shielded Cables And Connectors In the early 1980’s, engineers discovered that a common obstacle to meeting the radiated emissions requirements was radiation from input/output (I/O) cables. Shielding of the equipment’s cabinet wasn’t enough by itself—frequently the shielding had to be extended to include the exiting communication and data leads. EMI test labs began stocking shielded cables. Good quality shielded cables – in terms of both the shield and the termination – were available, but not universally employed. The specification of shielded cables was acceptable to regulatory authorities such as the FCC since they were not normally supplied to the customer and could be ordered by end users once they were specified as a necessary installation accessory.
The words “good quality” merit some elaboration. At that time, cable shields varied significantly in quality. Braided shields might exhibit good tight weaves, or on the contrary, might be loose and open; mylar/foil shields varied significantly in the tightness of the wrap, the quality of the conductive foil, and the way the drain wire made contact with the shield. Quality of termination was also an issue. It was not uncommon to see cables grounded through long drain wires, or for ground terminations to be brought in through a connector pin. At the frequencies of interest, above 30 MHz, these types of terminations were inadequate. A 360 degree surrounding conductive termination was needed for good shield termination.
Shielded cable and connector manufacturers caught on and over time produced a variety of innovative connector backshells and terminations methods involving platings, clamps, crimping, conductive tape, and conductive gaskets. Cable manufacturers, connector suppliers, and cable assembly houses began to produce both components and improved, fully assembled shielded cables. The first area to show improvement was the round style cable used for RS-232 communications and parallel printer interfaces. Shielded flat cables and modular cables soon followed. Today, the importance of shielded cables is so well understood that there are many suppliers for an item that has, through their efforts, become a commodity.
Shielding Effectiveness Testing While a detailed discussion of the measurement of shielding effectiveness (SE) is outside the scope of this article, a few words should be said about the evaluation of components’ effectiveness. Vendors test their EMC shielding components to standards and present the data to prospective customers for their consideration. The measurement standards all define and require standardized repeatable setups. However, the results obtained in these standardized setups do not always translate into predictable performance when the component is used in specific applications.
Many of the SE test methods used in 1979, and their successors today, are derived from military experience and involve the use of shielded rooms. The most cited reference of this type is MIL-STD-285, first issued in 1956, and since retired and replaced by IEEE-STD-299 in 1991. These SE tests have been used to evaluate EMI gaskets, conductive coatings, shielding laminates, and EMI honeycomb vents, and EMI windows.
Sometimes, as in the case of EMI gaskets, test parameters such as frequency range, the method of reference measurement, and test configurations have been modified, unfortunately in different ways, by different manufacturers. While these modifications may have been made for good reasons, and may have helped individual manufacturers standardize their results, it has led to de-standardization across the industry.
In an attempt to address this issue in the EMI gasket community, the IEEE EMC Society Standards Committee published IEEE-STD-1302. This standard is a recommended guide for EMI gasket characterization which compares and contrasts several existing SE test methods as well as identifying other emerging but not yet widely accepted procedures. The standards addressed are: SAE ARP 1705, SAE ARP, MIL-G-38528, DEF-STD-59-103, and stirred mode test techniques.
In the case of conductive coatings, both radiated techniques based on MIL-STD-285 are used along with the simpler method of measuring DC volume resistivity. The ASTM (American Society for Testing and Materials) standard ASTM D 4935, which uses a table-top, closed loop coaxial test method, has gained acceptance.
Measurement techniques of cable and connector shielding effectiveness continue to embrace many of the same methods and setups used 25 years ago, including several variations of cable substitution, as well as coaxial and triaxial measurement techniques.
We can see that some component tests have been with us for the duration, sometimes unchanged, sometimes improved. It is interesting to note that while many new products have evolved to meet the changing demands of the commercial EMI marketplace, many of the basic techniques used to characterize their performance have not changed nearly as dramatically.
Looking Back In the 25 years since the FCC gave a big push to commercial product EMC, a lot has happened. Technology has advanced tremendously, as have product performance requirements. EMC theory and practice has evolved substantially. We see that progress in the components and materials that we use to design in EMC performance. n
About The Author Joseph Butler currently has dual roles of New Business Development Manager and Conductive Elastomers and Compounds Product Development Manager for the Chomerics Division of the Parker Hannifin Corporation. His current responsibilities include providing marketing and business development support and new product development/R&D leadership for the conductive compounds and conductive elastomer business units at Chomerics. Joe has over 35 years experience as an electromagnetic compatibility engineer, having worked in both the military and commercial industry for Raytheon, GenRad, and RCA in addition to his past 18 years at Chomerics. He is a senior life member and Past-President of the Institute of Electrical and Electronic Engineers (IEEE) Electromagnetic Compatibility Society (EMCS). Joe is also a past member of the IEEE EMCS Standards Committee and has been involved with EMC standards development within the American National Standards Institute (ANSI), the American Society for Testing and Materials (ASTM), the Association for the Advancement for Medical Instrumentation (AAMI) and the SAE (Society of Automotive Engineers). He has also previously been involved with EMC standards work with the Radio Technical Commission for Aeronautics (RTCA), the Electronic Industries Alliance (EIA), and the Scientific Apparatus Makers Association (SAMA). He is a National Association of Radio and Telecommunications Engineers (NARTE) certified EMC engineer and is a past member of the Board of Directors of NARTE. Joe has a BS in Engineering Physics from Merrimack College, an MA in Physics from Williams College, and an MBA from Northeastern University. © Copyright 2007 Conformity |