We changed a number of T12 tubes yesterday- one of which inexplicably tripped breakers on 3 different circuits, some of which are on completely different panels. The only thing these circuits have in common are that all these panels are that they're fed from the same 100kVA transformer (208Y/120V).

What could possibly have caused this? Any of you seen something like this before? Only thing I can think of is harmonics from the ballast causing just enough harmonic surge in the transformer secondary to trip already overloaded circuits. Either that, or maybe the grounds are all messed up.

Was this a one time event or can this be repeated?

Strange...

We stopped replacing bulbs after it tripped... I'll let you all know what happens when we replace the others. IF we replace the others.

Are any of the breakers that tripped GFCI's? (suspecting luminare grounding/bonding issues)

Larry

Nope, no GFCIs, just normal breakers. One of the electricians here had a new theory that sounds pretty likely- he thinks the bulb grounded out; that it was getting high voltage on one side through the ballast, and the other prong shorted out to ground against the side of the fixture. This would have put a fairly high voltage and current on the ground.

If the pins were on the same end of the lamp, under the conditions you described, I've seen many filaments burn out when a pin hits the housing. I've never heard of one tripping a breaker, much less three, in separate panels

Steve,

How would current on the EGC trip other breakers?

This is a 3-phase 208Y/120 setup in a 12-wide trailer, derived from two transformers, each serving a portion of the building. I suspect (but haven't verified) that the neutral bar is bonded to ground in each panel, and all panels grounded to the structural steel, which would allow the fault current to pass through the other circuits in nearby panels as a voltage spike. Several of these circuits are already overloaded (12 computers on a 20A circuit+ peripherals) and wouldn't take much to trip, regardless of how the surve/spike/etc actually got to them.

If the neutrals are bonded properly at a single point, a ground fault couldn't have impacted them in this way.

I hate to say it Steve, but that sounds like fuzzy electrical theory to me. If the ground and neutral are bonded together (as they should be) why would it react any differently to a short to ground than current applied to the neutral under normal operating conditions?

Is it possible there was a loose neutral splice in the fixture that was jostled as the bulb was changed?

Or is it possible you're coming from some sort of a corner grounded system, whether intentionally or inadvertently?

Steve, we have had similar situation in a store. The store was t12 magnetic ballasts and various lights thru out the store were t5 electronic, undercab lights. These lights were causing other circuits to heat up and trip. We had to use a frequency meter to find them. Apparantly the harmonic goes back down the neutral and causes issues with the other circuits. We have had up 100+ cycles and this causes breaker to burn out and all sort of strange things to happen.

Joe Ogle