Fluorescent lights and wall wart voltage spike

metron9

As I was ready to go to bed tonight I turned off the fluorescent light above the panel project I was working on, I use a 12V supply and a 6V wall wart to power the panel. Well it turned off when I turned the light off. Hmm I said and did it again, so I hooked up the scope to trigger on a falling edge connected it to the wall wart (6V wart putting out 9V its a cheap one for sure) here is the electrical glitch that happens when I turn off the light, this 24V spike is somehow generated just by turning the light off when both the light and wall wart are connected to the same AC plug outlet.
The fluorescent light is one of those round bulb types with the magnifying glass in it. It has its own on off switch and that is what triggers this electrical glitch.

It made me wonder about problems people have with intermittant problems and microprocessors and AC power conditions. Adding a cap to the output of the wart lowers the glitch but it is still a pretty big glitch at about 1000uS and 24 volts and the glitch has to do with the AC power line feeding the wart so it is just another wierd and strange problem to deal with.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Think outside the BOX!

Paul Baker

Fluorescent lights incorporate ballasts to regulate thier current flow. The cheapest way of doing this is using a "magnetic" ballast, which is nothing more than an inductor. When turned off inductors will do a voltage kickback, and this is what you are seeing. If you can hear a humming from the light, a magnetic ballast is being used. "Electronic" ballasts convert the AC mains into a higher frequency while regulating current via transistors and other semiconductors. These are more expensive, but dont cause the major inductive kickbacks into the power mains that the magnetic ballasts do when switched off.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Paul Baker
Propeller Applications Engineer

Parallax, Inc.