John Gilmer wrote:
UL changed the standard, and for "new" GFCIs I believe that if the
neutral is open the receptacle is disconnected from the line.


Nope!

The GFCI needs power to operator it's sensing circuitry. Without a neutral
it doesn't have any power and it will not trip.


Anyone tested either?


Yes, I have.

You threatened to test this in a previous thread. Thanks for posting.

I have a cord set salvaged from an old lamp or some such. I put female
"spade" connectors on the wire end. If you take a little solid wire and
put it the back of the "line" side of the GFCI, you can run all kinds of
experiments in RELATIVE safety. Don't get me wrong, you will have bare HOT
terminals so you might consider powering your experiment from a GFCI
protected circuit. AND BE CAREFUL. (I have been doing stupid stuff like
this since I was about 7 years old butt I certainly would not permit my kid
to do it.)

(I have a cord set with insulated aligator clips at the end and am
fairly paranoid when using it.)


Basically, if you reset the GFCI with power on, turn power off, disconnect
neutral from GFCI, and put power back on you can draw all the power you want
between GFCI load side HOT and GROUND or another neutral. There just will
not be any power going into the sensing circuit of the GFCI.

Was this a "new" UL standard GFCI or an "old" one?

When reverse powered to the "load" teminals (GFCI not initally tripped),
the receptacle will be live for an "old" one (not required, but that is
how they were made) and dead for a "new" one.

My guess is that the "old" GFCIs use a mechanically held, electrically
tripped relay that operates as you describe.

My guess is that a "new" GFCI has normally open relay contacts that
connect "line", "load" and receptacle together only when powered. If
that is true, a "new" GFCI would not connect "line" to the receptacle
with no neutral.

One change in the "new" UL standard is that reverse line-load wiring
will not power the receptacle or "line" terminals. I believe that
requires a relay that does not close unless powered from the "line" side.

Another change in the "new" UL standard is that the GFCI be more likely
to fail safe. "Old" GFCIs had a habit of failing with the "line"
connected to the receptacle and "load", and not tripping anymore. That
is one reason I think "old" GFCIs have mechanically held contacts. And
my guess is that "new" GFCIs require the relay to be powered to energize
either the receptacle or "load" terminals.

------
Has anyone who has investigated a GFCI with a hammer looked at how the
"old" or "new" relays work?

--
bud--