DIY Chatroom Home Improvement Forum

DIY Chatroom Home Improvement Forum (http://www.diychatroom.com/)
-   Electrical (http://www.diychatroom.com/f18/)
-   -   Which should open first GFCI or breaker (http://www.diychatroom.com/f18/should-open-first-gfci-breaker-36312/)

jamiedolan 01-19-2009 12:18 AM

Which should open first GFCI or breaker
 
My thought was the GFCI would open first, and thought it might open fast enough not to even open the breaker. I caused a direct short from a pass and seymour GFCI outlet to ground. The breaker opened pretty quickly, but there was an arc flash (more than just a tiny spark), and I sure would not have want that to have been a real situation where I was trusting the GFCI with my life...

I thought the gfci's were suppose to open with 5mv, that couldn't cause an arcflah of any size could it?

When I used a solinoid tester from hot to ground on the GFCI, it triped instantly, and did NOT open the breaker.

Jamie

InPhase277 01-19-2009 02:32 AM

A dead short causes a few thousand instantaneous amps to flow. Likely the GFCI was overwhelmed by the intense electromagnetic field. And if you think about it, the current must come from the breakers first, so some small interval of time must pass before it reaches the GFCI, so the breaker is carrying the current before the GFCI is. But it could be a toss up to manufacturing tolerances. In other words, another GFCI and breaker may do the opposite.

Just Bill 01-19-2009 07:22 AM

GFCI and breaker protect against different things. The breaker is strictly on overcurrent device, the GFCI protects against a ground or neutral fault, usually trips with a difference of <.005A; the breaker, 15-20A or whatever is the rating. If the GFCI trips on a current overload(short), it because it was overwhelmed, as suggested above.

rgsgww 01-19-2009 08:05 AM

I would say that the gfi was overwhelmed by the overcurrent.

Billy_Bob 01-19-2009 09:37 AM

A GFCI is an electronic device which runs off the line power. I suppose if you have a dead short, then this would also remove power from the GFCI and it might not have power to turn itself off?

With a solenoid or relay, there is "switch bounce" which momentarily makes contact, then stops making contact, than makes contact again, etc. So this might allow the GFCI to have power for a long enough time to open.

I suppose this also could be different depending on the design of the GFCI circuitry.

As to this being an issue or not, a person being electrocuted would not be a dead short (there would be a lot of resistance there) and the GFCI circuits would continue to have a bit of line power. So not an issue. The GFCI would protect with an electrocution as intended and the breaker would cover the dead short situations.

Might want to try a solid state relay which should not have any switch bounce.

Here is a picture of switch bounce...

http://www.micahcarrick.com/files/at...3/debounce.png

HouseHelper 01-19-2009 09:43 AM

How did this short occur? If it was on the line side of the GFCI, the GFCI would not be affected and would not respond, only the breaker would.

jamiedolan 01-19-2009 11:30 AM

Quote:

Originally Posted by HouseHelper (Post 215484)
How did this short occur? If it was on the line side of the GFCI, the GFCI would not be affected and would not respond, only the breaker would.

Househelper: I caused a direct short in a test setup from Line side to ground. Why wouldn't the GFCI read see that current was not being returned on it's neutral and trip?

Jamie

rgsgww 01-19-2009 11:32 AM

Quote:

Originally Posted by jamiedolan (Post 215525)
Househelper: I caused a direct short in a test setup from Line side to ground. Why wouldn't the GFCI read see that current was not being returned on it's neutral and trip?

Jamie


What brand breakers do you have and how old are they? How old was the gfci?

jamiedolan 01-19-2009 11:37 AM

Quote:

Originally Posted by rgsgww (Post 215527)
What brand breakers do you have and how old are they? How old was the gfci?

Brand new Cuttler CH, with a brand new P&S GFCI. 12awg, 20A run in EMT.

Jamie

joed 01-19-2009 11:40 AM

Short on line is not going through the GFCI. It would not trip. The line is not protected by the GFCI. Think about it. How could the GFCI disconnect the line. That would be like asking a 15 amp branch circuit breaker to trip if the main line to your house shorted to ground. It can only disconenct the receptacle or the load side.

rgsgww 01-19-2009 11:53 AM

Oh, I didn't see.

Short from line to ground...no current will flow through the gfci.

jamiedolan 01-19-2009 02:01 PM

Quote:

Originally Posted by rgsgww (Post 215546)
Oh, I didn't see.

Short from line to ground...no current will flow through the gfci.

Ok no one try this for safety reasons. Here is the explaination to make it clear what I did, I didn't mention at first, because I didn't want anyone to think it was ok to try this. I took a piece of 12awg thhn, and striped the ends. One end I insterted into the hot side of the GFCI. The other end I touched to a piece of emt that was fully bonded to the panel.

Jamie

Stubbie 01-19-2009 02:33 PM

One of the things we did in school at the UAW was build a gfci testing board to look at the behavior of gfci during overcurrent and leakage events. We simply took a 4x8 sheet of plywood and built the circuit on it. We used a gfci receptacle some single pole switches and a 60 watt light fixture. Off the load side of the gfci we wired in a switch in series with the hot and grounded leg prior to the load and the same for the equipment ground. We energized the circuit then threw the switch connected to the grounded leg causing an overcurrent fault. The breaker always tripped out and the gfci always held. However there would be such an energy release that you would actually see the romex jump when the fault was created. You would also see a large arc inside the GFCI and was obvious if you had the lights in our testing lab off. Power was restored after the switch was returned to the off postion and the breaker reset. We then tried the test button on the gfci. In most cases the gfci would survive the first overcurrent event but rarily survived the second one. The test button would not trip the gfci and when disassembled the gfci's points were welded together or big time damage to the gfci's circuitry.
So gfci's don't like overcurrent passed through them and are likely unreliable after such an event.

rgsgww 01-19-2009 03:22 PM

Quote:

Originally Posted by Stubbie (Post 215639)
So gfci's don't like overcurrent passed through them and are likely unreliable after such an event.

Whats interesting about gfcis is when there is an undervoltage spiking to overvoltages at different frequencies...I've seen them reset themselves!

PaliBob 01-19-2009 04:03 PM

1 Attachment(s)
Here is a good pdf on GFI's
GFI-Hows & Whys

.


All times are GMT -5. The time now is 07:13 PM.