ECN Forum Forums General Topics of the Trade General Discussion Area nongrounded circuit extension, GFI won't trip

250.130(C) does not require us to run a EGC back from a two wire branch circuit extension.

250.130(C) permits us to do that if we choose.


Say you wanted to extend from a two wire circuit which is not prohibited, (Think about Article 394) but being the guy you are you want to add a EGC to the circuit.

Without 250.130(C) and it's "shall be permitted" it would be a violation of 300.3(B) to add the EGC to the circuit.

Bob

Wow, now I'm really confused.

If we extend a 2 wire branch circuit ungrounded, we can't use a a 2-wire receptacle, because they are for replacement use only.

I thought the GFCI protection rules were for existing circuits only. If we extend it from an ungrounded circuit to feed receptacles, aren't we required to bring the new part up to code?

Not that Mike Holt is the ultimate code authority, but I have an Mike Holt "Code Basics" article in EC&M (October 2004) that said BC extensions for receptacles are not permitted to be ungrounded.



Peter



[This message has been edited by CTwireman (edited 03-23-2005).]

My take on it now after listening to all of you is that replacments of ungrounded shall be GFIed and extentions shall be grounded.

250.130 is saying it shall be grounded but you have a choice in how you do it per 250.130 1-5

Two questions please: 1. Receptacle does not have equipment ground because it is a 2 wire type?
2. Is the line neutral grounded?
If power line neutral is grounded than GFCI protection againts leakage to ground is maintained and the GFCI will trip when 6ma or more leakage occurs. With 6ma of leakage the trip time could be upto 5.6 seconds.
Normally residential wiring always have the common wire grounded, thus the NEUTRAL is created.
If the line power do not have any grounding than the person touching one of the wires will become grounded, leakage current will flow through the person do to capacitive leakage currents to ground from the other conductor. The amount of current will flow depends on; length and size of the conductor. Leakage could be upto a couple of milliamps.

Les Vass

Edit to remove sales link


[This message has been edited by Roger (edited 11-15-2005).]

Article 406.3 D (a), (b) and (c) allow the replacement of a 2 wire nongrounding receptacle with either another 2 wire nongrounding receptacle or a GFI. Once replaced with a GFI the receptacle is to be marked GFI protected as well as "No Equipment Ground". All receptacles on the load side are to be marked as well. Note: there should be no equipment grounding conductor installed on the load side or between receptacles.

406.3 A says you must install only receptacles of the grounding type of the voltage class and current for which they are rated.

Question does this mean that they have to be installed with a grounding conductor or is this soley to allow the use of three prong cords?

Sorry got called away for a few and didn't finish my thought.

406.3 B says that have grounding contacts shall have them effectively grounded. Exception NO.2 Replacement receptacles permitted by 406.3 D

I would like to know why the high impedance of the neutral in question? Do you have different voltages at different points in this circuit?

Code issues asside, the questions was why didn't the GFI trip.

It didn't trip because there wasn't a ground fault to trip it. The tester you used relies on the GFIs ground prong to provide the ground. You don't have one since there's no ground wire.

If you had tested it by plugging in a toaster and sticking a butterknife into the slots while leaning on the grounded sink, you would have provided a ground fault and it would have tripped. Not the reccomended procedure thought.

I hadn't given it much thought before but... If I wanted to really test a GFCI in the lab, I would probably plot it at several different line to neutral currents. I wonder if the response is linear with increasing currents or if the sensitivity might drop off due to saturation? Maybe I'll run a test after Thanksgiving and get back to you.
Joe

If you take your wiggy and test from the hot side of the GFI to a ground it should trip.
If you are outdoors try sticking one lead to the dirt, it should trip unless the earth is bone dry. If you dont want to stick lead in dirt use screwdriver and touch screwdriver with lead.

I can see how Wiggy tests would be a decent go, no-go test of GFCI function. I always use meters and the datasheet I just d/l'd for the Wiggy, doesn't list input resistance. Unless it's substantially higher than 20 kohm, it is faulting too much current to verify that the GFCI is operating within spec. I'll be using a pot in series with a current limiting resitor to fault current to ground through my Fluke. It will be on AC mA & min/max record. I'll start at 110K and ramp down till it trips and read max off of the meter. I'll start with the fault current being the only load and step in increasing l-n loads to see if fault sensitivity changes.

I'm also going to borrow a Wiggy from one of our electricians for a few minutes and see how much fault current it will flow @ 120V.
Joe

[This message has been edited by JoeTestingEngr (edited 11-19-2005).]