On 12/3/2015 11:32 AM, dpb wrote:
On 12/03/2015 10:45 AM, Don Y wrote:
....

We know it isn't -- because I moved the extension cord to a non-GFCI
circuit and the circuit had no problem holding the load.

But have you done the simple expedient of swapping out extension cords yet?

Different extension cord makes no change in symptoms.

I'll try plugging the extension cord into a kitchen outlet (unloaded
circuit)
as that would test a different GFCI with the existing extension cord
BUT DIFFERENT WIRING (as we've already tried a different GFCI with the
existing extension cord and THE SAME WIRING).

I repeat--

But have you done the simple expedient of swapping out extension cords yet?

(sigh) Sorry, but *I* can make a giant list of all possible
combinations and permutations or circuit breakers, circuit breaker
TYPES, extension cords, number of strings, WHICH strings, internal
wiring, ambient temperature, time between applications of loads,
etc. ...

then, try ALL of those variations to find the one(s) that work and
don't.

That's not troubleshooting. That's what (inept) mechanics/plumbers/PC
technicians/doctors/etc. do day to day:

"Well, let's try replacing the battery to see if that's the reason
your old battery died..."

(weeks later) "Hmmm... I guess it wasn't the battery as that NEW
one has also died! Let's try replacing the alternator!"

(weeks later) "Hmmm... what are the chances that the new/rebuilt
alternator was defective? Maybe the cable harness is bad..."

I'm looking for a *reasoned* approach to a particular cause and effect:
if *this* is the underlying problem, then this experiment will serve
to isolate and identify that as the cause.

[This is easy to do whereas tying the existing wiring to a nonGFCI
breaker is a significant effort]

Just swap a standard breaker into the slot in the box.

Do you understand the difference between "a standard breaker"
and a GFCI?

A standard breaker has two connections: the power from the
distribution bus bar (usually a "snap on") and the *wire*
that feed the branch circuit ("load"). The branch circuit
picks up it's neutral and ground connections from a common
connection point shared among all branch circuits (as well as
the "AC line input")

A GFCI breaker has *four* connections: bus bar, load, NEUTRAL
and NEUTRAL PIGTAIL. I.e., the neutral wire feeding the branch
circuit connects to the breaker, NOT the neutral connection point.

Assuming you remove the GFCI and LEAVE IT DANGLING by it's
neutral pigtail, you still have to route the neutral for that
branch circuit up to the connection point for ALL the
neutral's in the panel. It's not "right next to" the breaker,
alongside the "hot" connection to the breaker!

Would you like me to also replace all of the RED lights in
the strings with BLUE ones? Maybe BOTH extensions are defective?
After all, both have been out in the same environment... maybe
they've both developed the same fault? A fault that HEALS
ITSELF when the lights have had a chance to warm up??

Maybe both GFCI breakers have failed in the same way -- despite
the fact that the other three located within inches haven't?

Sorry, I don't mean to sound ****y but "try this" is not what
I'm looking for. I want to approach the problem logically
not willy-nilly. I'd hoped someone might have *definitive*
information of problems like this instead of a litany of
hit-or-miss attempts. I.e., an explanation that reconciles
ALL of the observations I've posted.

(sigh)