Electromagnetic (EM) and radio-frequency (RF) energy sources surround us in the form of various electronic devices, all of which are capable of causing interference for one another. Fortunately, proper application of EM and RF shielding materials can minimize opportunities for interference. The large number of sources that can generate EM interference (EMI) and RF interference (RFI) often makes it necessary to install multiple levels of EM shielding into a design, from the circuit and package levels through the enclosure level.

Proper use of shielding materials helps a large number of high-frequency circuits co-exist when fairly tightly spaced in the same environment. The right amount of shielding material can prevent leakage of unwanted EM and RF radiation and for most designs the trick is knowing which type of shielding material to use and just how much shielding is enough.

Shielding materials come in many forms, including meshes, foils, cloths, tapes, and heat shrink tubing (Fig. 1). Many of these materials can provide very high suppression of EM and RF radiation levels. Of course, simply adding shielding materials to an electronic design can be expensive; the time-honored tradeoff is to use enough shielding materials to bring EM and RF radiation levels within desired boundaries, but without adding excessive cost to the design.

1. EMI shielding gaskets and other materials can be added at different stages of a design to limit EMI/RFI radiation. (Photo courtesy of Tech-Etch.)

How well an EM/RF shield performs is usually denoted by a parameter known as shielding effectiveness (SE), which measures the amount of power or field strength around an electronic device before and after shielding has been applied:

SE (in dB) = 10log(Pi/Pe)

or

SE (in dB) = 10log(Fi/Fe)

where:

Pi = the incident power density,

Pe = the exit power density,

Fi = the incident field strength, before shielding, and

Fe = the exit field strength, after shielding.

A shield can dissipate or absorb energy, and also reflect or redirect energy—the net result is that the shield has reduced the amount of energy around an electronic device. Various means are used to measure the SE of different materials, with probably the best known approach being the military standard MIL-STD-285. This standard is based on a high-level signal source and two antennas (loop, rod, or dipole antennas): one to transmit and one to receive at the frequencies of interest. More severe shielding measurement standards, such as TEMPEST requirements, exist, but these are applied as needed for different applications, such as for US National Security Agency (NSA) applications.

Different techniques developed over the years for measuring SE include the open-field or free-space method (which essentially detects radiation escaping from a finished product), the shielded box method, the shielded room method, and the coaxial transmission-line method. The shielded-box method is often used to compare the performance levels of different shielding materials. It employs a metal box with tight seam and sample port in one wall fitted with a receiving antenna. A transmitting antenna and signal source are placed outside of the box, and measurements are made with and without the material sample over the test port.

Unfortunately, this approach does not correlate well from one facility to the next and is limited to about 500 MHz in usable frequency range. The limitations of this method are overcome somewhat by using a larger shielded room, in which the dimensions of everything in the test setup (including the test material sample) are larger.

The coaxial transmission-line method, with measurements performed on small donut-shaped samples, has become the preferred SE test method. Measurements can be made at specific frequencies using a modulated signal generator, crystal detector, and tuned amplifier, or in swept mode with a tracking generator and a spectrum analyzer. A variable attenuator is set to maximum and a measurement made without the sample holder in place. Then the sample is added and the attenuation reduced until the reading is the same, which is the SE of the sample.