Yes Jeff, this was new in NEC 2002.
See? I learned something new today 
Actually, reidentifying a white conductor as a phase conductor and being required to color it other than white, gray or green was a change added in the '99 NEC.
ROP 5-25A (P 79) and ROC 5-26, 27 (P 78-79)
Pierre, I thought the vertical line along side the paragraoph meant a change from the previous cycle? And the bullet means something else.
In 2002 NEC, 200.7(B) & © have this line.
My 2005 Handbook has nothing.
My 1999 got soaking wet a few years ago and is long gone. I really should get another copy. I like having the back refrence. I have only back to '87 myself.
The Vertical line in a margin does identify where a change in the code is from a previous edition of the NEC. The change may only be a minor editorial change, or it could be a large change, such as a new paragraph.
The line you see is some change in the paragraph, but not a wholesale change. The '99 is when the change mentioned actually occurred.
I got it in my 1996 200.7:D
So Jeff you are right once again.
A white wire (neutral) is a current carrying conductor, so in essence it is always hot (just get in between the neutral & the load for a shocking discovery), under normal circumstances though there is no potential (voltage) to ground.
I thought I remembered it as early as the 1993 edition, but as I said, I can’t verify this as I seem to have “misplaced” my reference material :roll:
BTW - congrats on your new Cert’ Bruce.
The only place a white wire could be used as an ungrounded conductor without reidentification was as the supply side of a switch loop. That “loop” hole was what got changed in 1999.
I am not sure how long ago you would have to go before it was required to reidentify a white wire in a cable used as a hot in anything but a switch loop. 1975 won’t do it. (my oldest book)
I don’t know if this helps anyone understand why a neutral is not required, but here’s proof… my power is unfortunately for the example 120/208 (and a bit on the high side of that), but the idea is no different for 120/240 or 110/220.
In Canada, kitchen recepticals are wired “split-circuit,” that is each half duplex is wired to a different phase with a common neutral. Here is a picture with my multimeter showing that there is indeed 208v (well… 212v) between the two hots without involving the neutral.
On 120/240v, it would show ~240v instead.
Hope this helps.
With the split circuit in mind, how do you provide GFCI protection for the circuits without providing it in the panel?
Excellent question!
Two ways as of the 1998 code (the latest book I have). First they are only required within a certain distance from the sink; the rest are ok as is. I forget the distance and my code book is packed for my upcoming move, plus it may be different in the latest code.
However, It can be handled either:
a) 2 pole GFCI breaker
b) ammedment to the code that allows 20A circuits instead, thus allowing ordinary 20A GFCI recepticals to be installed where required. This section appears to be compatible with the US methods.
Now the new codes may tighten this more (I am not sure), but this place was built in 1982 and only has a GFCI in the bathroom.
However, without them being wired like that I wouldn’t have been able to show that example 