On Jan 9, 12:18*pm, DerbyDad03 wrote:
On Jan 8, 10:24*pm, JIMMIE wrote:





On Jan 8, 4:45*pm, "hr(bob) "
wrote:

On Jan 8, 2:30*pm, JIMMIE wrote:

On Jan 8, 2:14*pm, croy wrote:

Two years ago, I revamped a 20a circuit in my (USA) house. I
added some outlets, and put a GFCI breaker for that circuit
in the service panel.

All was well for a year and a half, then the GFCI breaker
started tripping for no apparent reason. *I assumed that the
breaker had gone bad, so I replaced it. *The new one still
trips, albeit after a longer period of time (old, about
three minutes; new, three to 20 hours).

I've pulled all the outlets and tightened the screws.

Still the breaker trips after some hours, whether the
circuit gets loaded or not.

One thought I've had: *the original romex for the longest
run of this circuit is from that period of time (1960s?)
when the ground wire is smaller gage than the main
conductors. *Could that be causing the problem?

Other ideas?

--
Thanks,
croy

The GFCI breaker works by detecting leakage current on the ground
wire. It only needs to detect about 5ma to cause a trip. Witth no
obvious problems they willl still trip just because the wire is too
long. This is caused by capacitive coupling between the hot wire and
the ground. I dis overed this the hard way when I ran an extension
courd from my back porch to my storage shed about 60 ft away. Forgot
and rediscoverd it again when I ran 100ft of extension cord at an RV
park.

Jimmie- Hide quoted text -

- Show quoted text -

It isn't leakage current on the ground wire. *It is unbalance between
the two leads leaving the GFCI toward the load, it doesn't matter what
they are connected to, it is the unbalance between the current flow in
the two leads coming out of the GFCI that trips it..- Hide quoted text -

- Show quoted text -

* *If there is an imbalance that means current IS flowing in the
ground circuit.
Sticking yor finger in a socket and grabbing the water
faucet will trip the associated GFCI because the current flowed back
to ground creating a current imbalance in the hot and nuetral legs. It
may or may not have went back via any particular ground wire or other
path.. My mentioning *wire was an assumption that had to do with the
OPs problem and my earlier suggestion that the wire may be excessively
long. In that case ground wire is appropriate. Just one possible
solutuion out of many suggest because no one else did. It is equally
correct to expalin the basic operation of a GFCI as detecting ground
current as it is detecting *and imbalance in current in the hot and
return.


That is not true. The second explanation is completely correct and
all that is needed. Here's one to think
about. I connect together two netural wires on a GFI
circuit. What happens? It trips because there is
an imbalance between the hot and neutral currents.
There is no path to ground involved. Or I swap the
neutrals on two differenct circuits. Again the GFI
trips and no ground current is involved.



"If there is an imbalance that means current IS flowing in the
ground circuit."

That may be true, but that is not what casues a GFCI to trip.

Yes, that true, but per above, it's not the only way to
trip one. Originally Jimmie used the term ground wire.
Now he's using ground circuit. I would say that's an
improvement, but still very misleading at best. I
agree with you that few people would hear the term
ground circuit and think that circuit includes them
standing on a wet cement floor. The more natural
association would be to the ground of the circuit
involved.





Your original statement was "The GFCI breaker works by detecting
leakage current on the ground wire" and that is the only point I am
arguing.

Your statement implies that the device is monitoring the current on
the ground wire (how else would it detect the current?) and when it
detects the leakage current it trips.

I agree.


You may call it semantics, but I call it science. The fact remains
that your statement "The GFCI breaker works by detecting leakage
current on the ground wire" is just plain wrong, regardless of how you
try to justify it. I'll even give you some latitude and let you leave
off the word "wire". The fact remains that the GFCI doesn't detect to
leakage current on any type of ground circuit. Period.

It works by detecting the imbalance between the hot and neutral wires.
Regardless of where the leakage current is going to, that leakage
current itself is not "detected" by the GFCI. It is the *imbalance*
between the hot and neutral that is detected.