For perhaps nine years the IBC / IFC has mandated that there be lighting on the 'path of egress.' That is, on the area outside the exit doors. This lighting is required to be on an emergency circuit, and is very often an HID fixture.
What struck me about this report was that the victim appeared to be a very conscientious and careful person. He took the 'as built' wiring diagram home to determine the correct circuit breaker, turned the light switch off and taped it, etc. He was trying to lock and tag out the circuit, he just wasn't trained correctly to test the circuit before working on it. He didn't use a pen tester or meter to check for voltage or use insulated tools, both common mistakes from untrained persons. This should be required reading for new maintenance workers who are frequently required to do similar tasks (change broken light bulbs).
Reno: Egress lighting has to be provided to 'the public way' which is an oft debated line in the sand thing. Defining 'public way' IMHO, is best left to be determined on a case by case basis.
My point was/is...HID fixtures on an E circuit may be a bad choice considering warm-up & restrike time. A 24-7 fixture on an E circuit would have to cool down & restrike before illumination. Right or wrong??
Based on the pic in the link, this does not look to be a fixture with a restrike lamp option.
I think there's a bit of uncertainty as to the type of fixture and the circuit powering it.
Hotline, I think you are imagining one of those 'bug eyes' that only trun 'on' when the power fails. I agree that HID would be a poor choice for those.
I was thinking of a separate circuit that, in addition to the usual power, also had back-up power available (either within the fixture, or serving the entire circuit). In the later instance the light would operate as any other light- on at dusk, etc. The difference being that the light would remain 'on' if there were a power failure.
I have also been shocked by this type of metal halide fixture and survived. It was a lesson in proper testing for live circuits for me. These fixtures have large ballasts and are often in hard ceiling areas of buildings. So changing the ballast is difficult even when the power is off. An option for this fixture type is to have a current sensing relay that turns on a 120 volt E-power fed quartz lamp if the metal halide lamp fails. The relay is often contained in a separate enclosure which has only 3/8" flex grounding it. When I checked for power, I did not notice that there was still emergency power in the relay enclosure. OOPS! Big shock.. Because of the double sources of power and the confined work area these fixtures are very dangerous.
It now makes sense to me to replace these death traps with a compact fluorescent fixture--cheaper than a metal halide ballast and much safer to work on!!!
#197069 - 11/07/1003:47 PMRe: Changing a broken light bulb: a cautionary tale
"I was thinking of a separate circuit that, in addition to the usual power, also had back-up power available (either within the fixture, or serving the entire circuit). In the later instance the light would operate as any other light- on at dusk, etc. The difference being that the light would remain 'on' if there were a power failure."
No matter what controls the circuit you describe above, it would not be an acceptable means of REQUIRED egress lighting.
Circuit 'on' with utility power, all is OK Utility power drops out, HID fixtures extinguish.
Generator (if available) starts & transfer occurs to designated circuit in <10 seconds IF gen is designed and installed as Emergency, HID fixture has voltage. Cool down & restrike require 3.5 to 10 min dependent on bulb, & ballast.
If utility power restored quickly, same conditions exist.
As I said, I did not see a restrike component within the pic in the link, nor is there enough specific info provided to make a firm conclusion.
I just questioned the use oh a metal halide fixture on something referred to as an emergency circuit.
I agree with John, emergency lighting should be virtually instant on (maybe just the CFL lag). As for the original incident; I suppose an answer would be for ballast/lamp holder manufacturers to put a neon indicator on the ballast to indicate that power was present, for safety and as a trouble shooting aid but that would cost an extra few cents a unit and they are unlikely to do it. I am not proposing a code change.
I suppose one can never really account for every possible variation ...
The fact remains that EVERY such 'egress' light I've seen has been some manner of HID. Maybe we are differing in what is meant by an 'emergency' circuit.
For the egress lighting, it is only required that they provide light for 90 minutes after power failure. So, many fixtures have batteries within the housing.
Another variation is to have a circuit of ordinary fixtures that is also supplied by a battery / UPS bank. No flicker / restrike issue there. Why have such a system? I believe it arose because many buildings have multiple tenants with separate metering - while the outside lighting is controlled by a 'house' panel. When the remodel of ONE unit results in the addition of "egress" lighting, sometimes the easiest way is to simply add a battery back-up to the existing outside lighting circuit.
Look, we're not here to debate various code and engineering nuances - at least not in this thread It's just that this well-meaning Handy-Andy got zapped. The idea is to remind the rest of us to make damn sure the circuit is dead, then work like it was hot anyways. Even with a dead circuit, those fixtures have capacitors that can zap you off your ladder; only the coroner will really care what it was, exactly, that killed you - the shock or the fall.
Other times those fixtures are wired to 2-pole circuits, and wires get mixed up. As a result, one of those feed wires might still be hot. Oops. So we check - then check again.