GFCI problem

[John Gold]

My project: Wiring/installing a light in my bathroom.

My procedure: I am running the power from a GFCI protected outlet.

The hot lead on the Load side of theoutlet connects to one terminal on a switch.

Another hot lead takes power from the switch to the light overhead. the neural wire from the switch line and outlet line are tied together.

Both the switch and outlet are grounded.

My problem: The GFCI trips immediately when I activate the circuit (at the circuit breaker).

Other details:

-- The GFCI outlet is fine -- If I wire it without the switch/light there are no problems. I even replaced it with a different one, just to check.

-- The switch/light circuit appears to work OK if I bypass the GFCI

I suspect there is a problem with the wiring leading from the outlet to the light. I've checked everything I can see, and it all looks OK.

My questions:

-- What obvious problems should I be looking for?

-- Is there a way to test the wires for a problem (like maybe a sheetrock screw embedded in one)?

Thanks,

John

[gabriel]

Use a continuity tester, disconnect all fixture wires at the switch box and take out the light bulb on the fixture. There should not be continuity:

- between the hot wire and anything else

- between the neutral wire and anything else

[Anthony Straight]

It isn't entirely clear just what you mean, here, but if you have routed the "hot" lead through the GFI, and bypassed the GFI with the "neutral" lead, that will trip the GFI. Take both the hot lead and the neutral lead from the "line" terminals of the GFI.

I presume this wiring is one with cable, and the cable runs from the GFI to the switch, and a second cable runs from the switch to the light fixture. The switch should connect to the hot (black) wire from the GFI and to the light. The two white wires should be connected together at the switch.

Sometimes, on a very long run, having a switch in one side of the line will cause too much unbalance in the normal leakage current because when the switch is open, the entire neutral wire from the GFI to the light fixture, through the light, and back to the switch will be sensed by the GFI, while only the portion of the hot wire from the GFI to the switch will be sensed to balance it. Since the two lengths are unequal, the difference in leakage may trip the GFI. Try it with the switch on. If it works, then trips the GFI when the switch is turned off, this may be the problem.

Finding a nail or screw penetrating to the neutral wire only is a difficult task unless you have a special test instrument that measures very small leakage currents at a few hundred volts, and it requires proper preparation to avoid damaging electrical equipment during the test. If you suspect this, you should probably have an electrician perform this test or otherwise troubleshoot the problem.

It is not required that the bathroom light be protected by a GFI. You can supply the light from the line side of the GFI.

There may actually be a ground fault in the light fixture or elsewhere. Disconnect the fixture (both conductors) and see what happens. If that doesn't remove the fault, disconnect the line feeding the switch (both conductors) at the switch and see what happens. This should help you narrow down the location of the problem.

See also:

Good Luck

[Anthony Straight]

OK, that was wrong. Take both leads from the "load" terminals of the GFI. Connecting to the "line" side will remove GFI protection from the light circuit, which is allowable.

[gabriel]

Oh, if that is the case, then your advice is correct. Both hot an neutral have to pass through GFCI. Now what is happening it makes sense.

To the OP: Don't bother with the continuity tests I suggested in my other post, do what Anthony said.