CL dnoyeB Gilbert ) said...

It would be a pretty rare occurance for a GFCI to trip off due to
transients flowing strictly due to the inductive and capacitive load
of the power line. I wouldn't say impossible, but think it is fair game
to say I have a better chance of winning a lottery.


Isint this contrary to your statement about refrigerators which are an
inductive load?

No, the "inductive and capacitive load of the power line" that I was
speaking of is strictly that: of the power line.

When you apply AC power to a pair of wires with no load at the other end,
there will be a small current flowing in that wire, mostly due to the
capacitive effect of the wire. The inductive effects generally don't come
into play until you try to disconnect those wires and a tiny bit of arcing
may be seen.

This effect can be noticed, albeit very tiny, when a branch circuit with
nothing on it is disconnected from the power source. I have seen this effect
quite dramatically when a substation was being disconnected from its
source -- after the load is shut off, then the disconnects are opened
at the top of the substation, the power company then had a guy in a cherry
picker disconnect the wires at the pole. With about 50' of wires connected
to nothing at the far end, there is quite a visible and audible spark when
those wires are disconnected from the supply (I don't recall the voltage,
but it was at least 4600 and not likely over 16,000).

Getting back to my point, with only the effects of a non-loaded cable
on a GFCI, I do not believe that it is likely that any transients can
trip the GFCI. With a load on the cable downstream from the GFCI (lights
or a refrigerator for instance), the odds improve significanty. Not that
you will see it occuring on a weekly basis, but going from "practically
never" to, say, 1 in 10000, is a significant increase.


--
Calvin Henry-Cotnam
"Never ascribe to malice what can equally be explained by incompetence."
- Napoleon
-------------------------------------------------------------------------
NOTE: if replying by email, remove "remove." and ".invalid"