EMC Design Practices: Shielding Playing Hide and Seek
RF energy—now you see it, now you don't. It's much like playing hide and seek when we were younger. As you may recall, it's very difficult to hide when you are out in the open.
La blindage electromagnetique describes it well. The French term for shielding does a great job conveying the idea that we are hiding electromagnetic energy. It's still being generated, we just can't see it—or measure it. Consequently, if we are having difficulty hardening our circuits so that they meet emissions and susceptibility (immunity) requirements, we can always hide them within a shielded enclosure.
Since shielding also provides an isolated, low-inductance ground reference, it effectively reduces internal crosstalk and circuit-path coupling as well. In many cases, shielding can eliminate the need for filters. Even if filters are required, a shield provides a superior RF sink and isolates the filter's input and output leads.
Shielding works, and it doesn't affect signal integrity, an important factor with today's ultra-high-speed systems. It's the only suppression technique with all these advantages, which accounts for shielding being one of the most widely used ways of meeting EMC requirements.
Shielding can be used at the PCB level to protect sensitive circuits from internal and external RF sources as well as at the systems level. Really large systems can be placed in shielded rooms or even shielded buildings. But regardless of the size of the enclosure, shielding works the same way.
Because of the differences in the radiated field and shield impedance, some of the energy incident on the enclosure is reflected (R) from the surface. Some of the energy is absorbed (A) as it travels through the shield material. The remainder exits the other side.
The ratio of the incident field intensity (Fi) entering the shield to the field intensity (Fo) exiting on the other side is the shielding effectiveness. Normally, this is expressed in dB = 20 log (Fi/Fo) and represents the sum of the absorbed and reflected energy (A dB + R dB). This is illustrated in Figure 1.
For most commercial equipment being built for the European marketplace, 50-dB to 60-dB shielding effectiveness is all that is required. If good PCB EMC design practices have not been followed, then 80-dB to 120-dB shielding effectiveness may be required. Be assured, it's much easier to build a 50-dB enclosure than a 120-dB one.
At the higher frequencies, shielding effectiveness is dominated by absorption as expressed by
A dB = Ka t √(ms F)where: A dB = absorption loss in dB
Ka = 3.338 (t in mils) or 131.4 (t in mm)
t = thickness (mils or mm)
m = relative permeabilitys = relative conductivityF = frequency in MHz
Absorption is significant and continues to increase with frequency. The attenuation becomes so large that it's easy to understand that the shield material does not normally limit high-frequency shielding effectiveness. A few examples will make this much clearer.
A typical PCB shield made from alloy 174 BeCu (m = 1, s = 0.50) would have an approximate thickness of 16 mils. At a frequency of 100 MHz, the absorption loss is 23.6 dB per mil of thickness, and the PCB shield attenuation would be 378 dB. A small electronic enclosure made from aluminum (m = 1, s = 0.64) would have an approximate thickness of 40 mils. At a frequency of 100 MHz, the absorption loss is 26.7 dB per mil of thickness, and the enclosure attenuation would be 1,068 dB. At 1,000 MHz, the attenuation would be 3,378 dB.Of course, using typical lab equipment we aren't able to measure attenuation values that large. Measurements are limited to about 140 dB. But in many cases, in spite of the material, the shielding isn't that good.
The calculated value for shielding effectiveness of the material represents a best-case value because it does not take into consideration the effects of holes and other discontinuities. Most shields look like electronic Swiss cheese. They are full of holes needed for cooling, cables, displays, switches, and controls. Or they have seams.
The worst-case attenuation of a single hole or slot aperture that is not penetrated by a conductor in a thin conducting sheet is given by
S dB = 20 log (λ/ 2 L)
where: S dB = slot attenuation
λ = incident wavelength
L = slot length (which is ≤ λ/ 2)
This equation can be solved in terms of L to determine the aperture size that results in a given attenuation.
In general, however, apertures should be smaller than λ/50 and never bigger than λ/20. To achieve acceptable attenuation values at 100 MHz, for example, apertures should be smaller than 10 mm.
In general, seams tend to be the largest and worst-case aperture. As a result, the smaller the enclosure, the better it is. Plus, smaller enclosures have fewer penetrations, and these penetrations are smaller.
Considerable improvement in aperture shielding effectiveness can be obtained (as λ/2 approaches the value L) by increasing the thickness of the material to t ≥ L. At this point, the opening begins to act as a waveguide being operated below its cutoff frequency.
For frequencies that are less than Fco/3, waveguide attenuation is independent of frequency and given by
A dB ≈ 30 t/L
where: A dB = attenuation
t = thickness
L = length or diameter of the opening
This equation is for one waveguide. The constant, 30, is an approximation resulting from a slight variation because of hole shape and ranges from about 28 to 32. For a t/L ratio of 4, typical of high-performance honeycomb, the calculated attenuation of one cell is 120 dB.
In the frequency range between Fco/3 and Fco, the attenuation decreases, reaching 0 dB at Fco. Even so, the waveguide has significantly greater attenuation in this frequency range than the same size hole in a thin sheet. This characteristic is fundamental to the superiority of honeycomb, which is constructed of small waveguides assembled in parallel. Figure 2 shows the shielding relationship between material properties, hole size, and thickness.
Regardless of a panel's attenuation, the shielding effectiveness of the enclosure is determined by the panel's installation. Any ventilation or access panel whose perimeter is not adequately bonded into the enclosure behaves as a lossy antenna structure sitting in a hole.
Grounding the panel at one point will reduce the antenna efficiency and may even solve radiation problems at low frequencies, but it will not eliminate leakage from the rest of the seam. For example, if a 100-dB honeycomb is improperly installed over a 6-in. muffin fan, the perimeter leakage can limit the enclosure attenuation at 100 MHz to about 20 dB.
The best installation methods are welding, brazing, soldering, or riveting—in that order. But these methods prevent easy access for maintenance and repair. Installing closely spaced threaded fasteners or clamps permits, but does not facilitate, field removal. As a result, enclosure manufacturers generally use RF gasket materials to maintain contact across the seams. Depending on the seam design, gaskets will eliminate the need for fasteners or at least permit them to be spaced further apart.
There are three basic designs: the isolated seam (not a seam at all, just a butt joint), the compression seam, and the shear seam. These are illustrated in Figure 3.
The isolated seam is popular in weight-sensitive applications because there are no overlapping surfaces that increase material content. Also, the shielding material does not need to handle gasket compression forces and can be very thin, such as foil. Any RF sealing approach must bridge the gap between the shielding materials.
For one-time applications, conductive tape often is used to seal the seams. For applications where the seam may be opened, the gap typically is bridged by spring finger configurations that can apply forces to both sides or by elastomers that are configured to fit into the gap. If elastomers are used, the shield-material thickness must be adequate to handle the edge loading without buckling. Material thickness, overlap, and tape conductivity determine shielding effectiveness.
The compression seam is the most widely used, typically because it was not part of the initial plans for EMC protection. At the last minute when everything is failing, a compression seam can be reconfigured into an existing enclosure. It might not be the best for the application, but it's easy.
Any gasket material can be used in the compression seam, but since the compression forces are normal to the mating surfaces, this configuration needs closely spaced screws or clamps to preserve the shielding effectiveness. Screw spacing is determined by the stiffness of the cover and the compression force of the gasket materials.
If the gasket material has large variations in conductivity vs pressure, such as carbon- or metal-filled elastomers, then screws must be installed using a torque wrench, or the design must have positive mechanical stops to ensure uniform compression to minimize shielding-effectiveness variation. Using metal spring fingers or high-performance conductive elastomers, this configuration can easily provide more than 100-dB shielding effectiveness.
The shear seam, also called a flange-and-pan or channel-and-pan (knife edge) design, can eliminate the need for fasteners to preserve the shielding effectiveness of the enclosure. This design must be planned—it can't just happen.
To facilitate sliding across the gasket surface, this design requires metal finger gaskets. Although the compression forces are normal to the mating surfaces, this arrangement rotates the forces so that they are parallel with the panel surfaces, eliminating the need for multiple, evenly spaced fasteners.
The elimination of the fasteners and their associated assembly costs can pay for the installed gasket materials. Moreover, shear seams offer superior reliability because compression forces are determined by the size of the panel relative to the size of the opening, and these opening sizes are constant. This configuration provides the greatest shielding effectiveness (more than 140 dB) over the widest frequency range.
About the Author
Ron Brewer is vice president of EMC technical services at Instrument Specialties. He is a NARTE-certified EMC/ESD engineer with more than 25 years in EMC/ESD/Tempest engineering. Mr. Brewer serves on three technical committees and, as an internationally recognized EMC authority, has made more than 185 technical presentations in North America, Europe, Asia, and the Pacific. He also has been named a Distinguished Lecturer by the IEEE EMC Society. Instrument Specialties, P.O. Box 650, Delaware Water Gap, PA 18327, (570) 424-8510. Copyright 1999 Nelson Publishing Inc.