If you have ever tried to listen to your favorite radio morning man while you shaved or put on your makeup or both and if your bathroom mirror is fitted with fluorescent lights, you have a general idea what the arcing and sparking in a fluorescent light can do for the electromagnetic environment. Of course, fluorescent lights are one source of RF noise that actually does operate from DC to daylight. Well, perhaps not DC, but 60Hz is close enough, isn’t it. Give or take a few RF or chattering relay starters the light works something like this: during the first few milliseconds of the cycle nothing happens. Then the rising potential puts the Hertz on the electronic noise. When the ionization potential of the gas in the tube is exceeded, the plasma ignites. The electrons jump out of the gas atoms and accelerate toward the positive electrode, but they bump into another gas atom in a few nanometers – like turning on a small current and then turning it off. This ensemble of switching transients creates RF, MM waves, heat, light and UV. The UV excites the phosphor on the inside surface of the tube and, voila!, on comes the soft white or harsh white uniform light that we all know and love.

The lights have a current waveform that looks something like the waveform in the figure. It’s sort of a sine wave out of which someone took a bite. This waveform produces a frequency spectrum that looks like the other picture in the figure. This spectrum decreases linearly with frequency, if the interference is capacitively coupled, and it usually is (we are, after all, in the near field at 60Hz even if the light is in Bangor, Maine, and the victim in Miami, Florida). In the victim the result is a noise spectrum which is constant with frequency from 60Hz to a few hundred kHz where the rise time of the waveform kicks in and causes the harmonics to decrease as the square of the frequency.

In order to see what he is doing, your friendly local sawbones uses these lights. They have the advantage of being cool (so you save on air conditioning) and being an efficient light source so that you can save a few pennies a month on your power bill. But, if like some of the physicians, you are using them in a room with sensitive diagnostic equipment, brain wave measuring equipment used in neurosurgery for example, then you get your choice – the light or the information but not both. Now, it is nearly a no-brainer to figure out what to do. You just use inefficient, hot, incandescent lights. It never fails to surprise me how otherwise reasonable people will argue that they can’t accept the easy, cheap answer and need to install power line filters or shielding or both in order to use their beloved fluorescent lights.

Light dimmers have a very similar waveform and spectrum, although the switching is done by silicon-controlled rectifiers. They, however, switch more current and are therefore louder. They too are unnecessary. They had light dimmers in the1920s. The things used variable transformers or rheostats. Both still work, are still available and make no noise. The medicine men, in fact, use blanket and blood warmers which use solid-state switching devices to switch twice each cycle. They make even more noise since they control even more current. They, too, are unnecessarily noisy since a bi-metal thermostat works just as well, switches only once or twice an hour and its clicks are more easily suppressed. Even though these devices are used in close proximity to medical diagnostic equipment, no one as far as I know makes quiet light dimmers or warming ovens and, worse, the hospital folks don’t want to hear about how to avoid the problems.