I reckon it wouldn't hurt to mention some history of this thing on my behalf.
Once I discovered the spaghetti-bowl-in-a-handy-box fiasco, I went to shut off the breaker to the clothes washer. I then checked the voltage with my trusty digital VOM- 15 volts... HMMMM... my thoughts were a possible failing breaker or (perhaps) a minor carbon track somewhere within it, or something. Upon further investigation I found a second outlet not too far away and it appeared to be wired to this same circuit. Checked the voltage- 120. It was only then that I realized the circuit was Edisoned. Found the other breaker three-up from the washer circuit, on the opposite leg, shut it off, checked the voltage- zero volts; a little bit safer to work on.
My thinking about this was that, if this "back EMF" from the opposite leg was somehow "seen" by the GFCI outlet it would cause nuisance tripping but for some reason it would work fine at the end of the circuit run (perhaps) because nothing was downstream of it, somehow upsetting the circuit balance (in retrospect, it appears that my logic has some holes in it. Even so, theory and practice are oftentimes different worlds).
Even if the GFCI outlets do work fine at the beginning of the circuit and using the feedthroughs were no problem, this (in my opinion) should never be done as the integrity of the neutral grounded conductor should be totally device-independent on multiwire circuits. Pigtails prevails... Just my opinion...
[This message has been edited by Sir Arcsalot (edited 03-15-2004).]
[This message has been edited by Sir Arcsalot (edited 03-15-2004).]
