The following pictures show the result of a simple line to ground fault on a 277 volt lighting circuit.

You can see the plastic switch housing melted from the heat.

Luckily the cover plate was in place when this flashed, all that happened to the person that turned the switch on was a good scare.








[This message has been edited by iwire (edited 04-06-2004).]

Wow! Since I primarily do residential work, I've never encountered a 277/480 volt circuit.

However, this forum and pictures like these have made me well aware of the added danger.

Bob, was this caused by a wiring error (skinned wire, etc) or a fault with the motion sensor itself?

This was a wiring problem.

I do not know if the wire was damaged by the roto zip of the dry wall guy or if the wire was pinched when stuffing the MS into the box.

It was the wire on the load side of the switch, the MS was installed and the circuit turned on, then when someone pushed the button on the MS it "lit up".

Remember this is just a 277 volt fault not a 480 fault and the 20 amp breaker never tripped, the wire was blown apart clearing the fault.

Just something to think about before sticking your fingers into a live 277 volt (or any voltage) box.

On the same job a sprinkler contractor thought he would help me out and move a live 277 volt 2' x 4' troffer, well the wiring plate came off when it was jammed against duct work and the 12 AWG was cut into by the troffer.

Again the breaker did not trip, the wire just blew apart, this guy did not even know how lucky he was not to have gotten blasted.

My guess is the wooded ladder kept him from being a conductor, as the light fixture enclosure was 'hot' for a short period of time.

I use 4x21/8 box on all dimmers GFCI and other biger switches that takes more
space than a general snap switch. That is why i do it big guy

Big Guy That describes every box on this job.

All the boxes I purchase (the Company purchases ) are 4" Sq deeps and this has a 3/4" raised ring on it.

Only 3 - 12/2s in this box, out of about 100 of these MS, only one blew up.

ok it looked like a 4x1.5 anyway thing like this happen's Hay maybe no anti-short
LOL just joking

Bob, just curious, do you also use the color coded MC cable? like the brown/gray for 277 etc. Not a code requirement, I realize, but makes the job easier.

Also, when you say the wire "blew apart," I assume you mean the copper vaporized and splattered everywhere?
Peter

[This message has been edited by CTwireman (edited 04-06-2004).]

CT,
Technically it is a code requirement, though rarely followed. Not the colors necessarily but differentiating systems. Take a look at 210.4(D)

"The Watt Stops Here"...well, almost
Where did you find cable with all black conductors?

Seriously, this could very easily happen to a finger or hand, and they can't be replaced like boxes and wire.

can anyone get me some info. i still don't look through these things, like breaker reset arc-flash ect.
never heard of things like these here.
If you short 230V to N/ground here, you will trip the breaker, and thats it.
and breakers usually trip before you fire up a milimeter of wire.

I think what wireman ment was that there is know code that say you must use certain
colors for 120 to 480
the only thing on colors is for high legs
grounded and grounding conductors.
I could use brown for 120 and red for 480volt if i what two

Peter.


We have, but I do not agree it makes things easer.

Try having all the right color combinations on hand.

Brown, Orange, Yellow.

Brown Gray

Orange Gray

Yellow Gray.

Etc.

Nick


210.4(D) only applies to the conductors of a multi-wire circuit, not to the building as a whole.

Straight three phase circuits, two wire circuits or feeders are not required to be identified at all.

It happens when we pipe and pull we do color code everything, but in general there is no requirement to.

Bob

Bob,
True, but a vary large portion of a building fits the multi-wire branch circuit definition. I don't think I have ever seen it followed. And I agree doing this in MC cable is an inventory nightmare. I have had to do it per spec. Sorry to get off topic.

Nick go off topic all you want, it is all electrical as far as I am concerned.

A lot of our work is multiwire branch circuits, and if done in MC the requirement gets ignored just as you say.

Bob

And by the way, nice PVC work.

No offense meant, but this shouldn't happen. I'm a bit worried by the opinions in this thread: "It's 277V" is no excuse. It should be no different compared to a 120V system. If things like this happen from time to time, something is seriously wrong.

C-H No offense taken at all.

But I do not understand your statement.

Are you saying mistakes should not happen?

Or that the breaker should open quicker?

What would be your solution to either of these?

Bob

Bob,

I should have said "easier to identify." It just seems like a great idea (color coded MC cable) but apparently not easy to implement. Oh well...

Peter

Hi Peter, it does sound great at first but it is tough to have the right cable always on hand.

We have some strict company rules, one of these is the color coding and marking of all junction boxes.

As an example all 277 lighting J box covers will be yellow, 120 lighting green.

A J box cover with 277 emergency lighting inside will be 1/2 orange 1/2 yellow.

We have a list of about 7 colors we use to identify deferent voltages and systems.

We also use a sharpie to write the panel and circuit on the cover.

Bob

iwire you go around painting boxes?
that is a good way to do things for you and your company. Most of the time the specs
will say how thay want it done .Like on my job i have to paint all fire alarm boxes
red and also the pipe.What i would like to know is do you give the owner's a copy of what the colers stand for?

As an example all 277 lighting J box covers will be yellow, 120 lighting green


Green is for ground can you paint the box that color? I seen someware that if you
used a green wire nut on hot conductors
that it would not be to code. So would that be saying it is all grounds inside?

Generally we spray paint a ton of box covers ahead of time.

We always have spray paint, the same colors are used to paint the circuit numbers on the floor where cables need to be pulled to.

When te cables are pulled and coiled at the ceiling waiting for walls the coils are tagged and yes hit with the color spray paint.

By the way the tags on the cables with circuit numbers are made of colored electric tape matching the color code.

No inspector I am aware of has a problem with green box covers and I do not know of any NEC rule that prohibits this.

No I do not think the color code gets turned over but with the circuit numbers on the boxes 'breaking the code' is not rocket science.

Bob

Bob:

As an AHJ, it would be a pleasure to inspect any of your work!!

As to the "green" box covers, NO PROBLEM, as long as the paint does not interfere with the metal-to-metal surface contact for grd/bond purposes.

That goes for the red, yellow, black, blue, rainbow, multi-color, faux, marbled, or whatever.

John

I have to side with :Andy: (Germany) and C-H (Sweden) here and question why the breaker didn’t trip before this level of damage occurred. Having worked with 240V phase to ground in both Britain and South Africa, a short circuit like described would have tripped the breaker almost instantaneously – I have personal experience unfortunately.

Was this an unusual high impedance, relatively long duration fault that allowed this amount of energy to be dispersed without overloading the breaker - or are there different fault-time curves for European versus American breakers?

Thanks Hotline, I am proud to say I do not remember failing an inspection on a job I ran.

Part of that is refusing to waste the inspectors time no matter how much pressure the GC puts on, I call when the job is ready, not before.

Bob

I too have to wonder what caused this without tripping the breaker. Like Hutch and Andy, I've seen plenty of dead-short faults on our 240V to ground systems, with nothing more than a tripped breaker.

It must have been either an intermittent short, each period of contact too short to trip the breaker, or a high-resistance connection somewhere on the circuit which didn't allow enough current to flow when the short occurred.

I wonder whether there's also a possibility that the source impedance of American supplies is generally quite a bit lower than over here, resulting in a greater amount of "let-through" energy even when a breaker does trip out.

Bob (iwire),

Sorry, I meant that the breaker should open. Everyone makes mistakes and the protective devices are there to save you. I was implying that the breakers are no good.

There's a good possibility that this "smoke job" was caused by an internal fault in the occupancy sensor.
There's internal circuitry that uses the ground. (as a matter of fact, some of the newer ones come with 2 grounds. 1 to be used for 120V, the other for use @ 277V).
This could account for the massive amount of smoke before/without tripping the c/b.
Good chance that the internal circuitry had to burn before the circuit went directly to ground.

We use the "different" colors of M/C cable where we have 4 wire circuits, but like Bob & Nick, have only used the "brn/gry, yell/gry, etc. a few times due to the logistic nightmare of stocking all the different colors.
Bob, I like the idea of painting the plates, we have just labeled them in the past...I'll give it a try.

[This message has been edited by electure (edited 04-08-2004).]

Very few electrical failures are of the "bolted fault" type. With any other type of fault the arc itself will limit the fault current. This impedance will increase the trip time of the OCPD. This is especially true of 480/277 Y systems. The tendency for the arc to limit the current is exactly why the code requires GFP for OCPDs rated 1000 amps or more on these types of systems. The arc limits the current to the point that the breaker "thinks" that it is just seeing a "normal load" current, but in reality the switchgear in "melting down".
In the case that started this thread, assuming a typical 20 amp circuit breaker is used as the OCPD, a total impedance in the fault circuit of 1.6 ohms could result in a breaker clearing time of 5 seconds.
Don

[This message has been edited by resqcapt19 (edited 04-08-2004).]