EC-24.6 Electromagnetic Interference and Compatibility

What this is and why it exists

A product that interferes with other equipment, or fails when other equipment interferes with it, cannot legally be sold in most places.

The subject looks like a collection of superstitions until it is organised around one structure. There is always a source, a coupling path and a victim, and breaking any one of the three solves the problem.

Most of the effective measures are decisions made during layout, and they cost nothing. Most of the expensive measures are things added afterwards because those decisions were not made.

The vocabulary

• Source — whatever generates the interfering energy.
• Coupling path — how the energy reaches the victim.
• Victim — the circuit or equipment that is disturbed.
• Conducted emission — interference leaving along a cable.
• Radiated emission — interference leaving as a field.
• Common mode — current flowing the same direction on all conductors of a cable.
• Differential mode — current out along one conductor and back along another.
• Ferrite — a lossy magnetic component that attenuates common-mode current.
• Pre-compliance — informal measurement on your own bench before formal testing.

The mental model

Name all three parts before changing anything. Interference problems are frequently attacked by adding filters at the victim, when the cheap fix was at the source or in the path. Working systematically means identifying the source, the path and the victim, then choosing which is cheapest to break.

The frequency tells you which kind of problem you have. Below roughly thirty megahertz, energy mostly escapes along cables and is measured as conducted emission. Above that it mostly radiates. The two are measured differently and fixed differently, so knowing which one a failure is halves the search immediately.

The most useful distinction is common mode against differential mode. A large differential current in a small loop on the board radiates poorly. A small common-mode current on a metre of cable radiates very efficiently, because the cable is an excellent antenna at these frequencies. That is why so many emission failures are cured at the connector rather than in the circuit.

The three remedies do different jobs. A filter attenuates along a conductor. A ferrite adds loss to common-mode current specifically. A shield blocks a radiated path. Using the wrong one spends money without helping, and each is effective only over a particular frequency range.

Immunity is the other half of the requirement. Formal tests inject energy into a product: electrostatic discharge onto its surfaces, bursts onto its cables, surges onto its supply, and fields all around it. Protection has to be at the connector, before the energy reaches anything delicate, because once it is inside no amount of software will help.

Finally, formal testing is expensive to repeat. Finding problems earlier with a near-field probe and a spectrum analyser will not certify anything. It will find most failures while they are still cheap to fix.

What you should now be able to explain or do

• Identify the source, coupling path and victim in an interference problem.
• Say whether a given failure is conducted or radiated, from its frequency.
• Explain why a small common-mode current on a cable radiates so well.
• Choose between a filter, a ferrite and a shield with a reason.
• Say where immunity protection has to be placed, and why.
• Run a pre-compliance measurement and say what it can and cannot tell you.

Check yourself