Say I feed a multiple gang box with a single 14/2 cable fed by a 15 amp cb, wire-nut seven neutrals, pigtail the hot to 6 switches, then feed six 14/2 cables up the wall and through a single bored hole in the top plate. These six cables continue on to six individual recessed luminaires.
According to code, I have 12 current carrying conductors through this hole. Building code requires me to draft stop this hole, and 334.80 requires me to derate these conductors per 310.15(B)(2)(a).
But, because I feed all six cables from a single 15 amp breaker, am I exempted from the derating requirement? (after all, I am limited to 15 amps shared among all cables)

What you describe can happen in pipe systems as well.

In my opinion it should be exempt.

That said I do not believe there is anything in the NEC that allows us to 'exempt it'.

Bob

I could not agree more Bob. each wire will carry it's own current and generate it's own heat which will be additive and hence a large amount of heat right at the upper plate where they are closest together.

How long is the hole? Even if it is a double top plate and the hole is 3" long, if the total conductor length not in the hole is at least 30" then 310.15(A)(2) exceptiion allows us to use the ampacity of conductors outside of the hole for that portion in the hole.

I hate to be the spoil sport here but the code is wrong.
The whole derating issue assumes all conductors will be on separate circuits so the heat adds. If they are on the same circuit it actually subtracts where they are paralleled.
If the total amps is 15 and it is spread out across 1 pair of conductors "in" and 5 "out" pairs of conductors the heat generated in the insulation is ~60% of what it would be with a single Romex in and out. Resistance of the "outs" is 1/5th of a single pair and total heat is still I2R.
The only way this current could add is if this was a series loop and full current comes in and out 3 times.

eprice- I think you are mistakenly using 310.15(A) because 334.80 specifically says that "B" applies and using your approach it would rarely be the case where there would be derating of NM cables. And I do understand the Exception in "A" of 310.15.

I do not think anyone here is disputing that, I said as much in my post.

The code is not wrong ---

(Assuming that the plate is thick enough that derating is required ...)

The code simply says that if you derate this situation then the AHJ will approve this portion of the work.

If you do not wish to derate you need to do engineering. In this case, engineering will show that derating below 15amp is not required.

quote:
--------------------------------------------------------------------------------
I hate to be the spoil sport here but the code is wrong.
--------------------------------------------------------------------------------

I do not think anyone here is disputing that, I said as much in my post.

The code is never wrong. Maybe poorly written in many instances, but not wrong. These examples are where you talk to the AHJ and get permission. When you parrallel conductors you do not significantly increase the heat when they are bundled or in raceway. If the AHJ doesn't know that he shouldn't be inspecting.

George, the exception in (A) of 310.15 was discussed in the IAEI news a while ago. The assumption is the thermal conductivity of the conductor will tend to average out the hot spot. In that regard I don't understand the new language of 336.80. I guess the CMP has the same issue you do with the exception but they won't address it.
Sometimes I think the guys who write new language do not rationalize it with old language they let stand.

I made a 2008 proposal to delete the 334.80 derating requirement, based on the allowance of 310.15(A)(2) exception. I am interested to see what panel 7 says about it.

George Little,

Well, here is how I think these sections work together. I agree that 334.80 does not reference 310.15(A). It tells us how to calculate the ampacity that would apply to the part of the circuit that is in the hole and it uses 310.15(B) to do that. When we're done with that step, we have two (maybe more) portions of the circuit that have different ampacities. Now comes the step where we determine the ampacity of the circuit as a whole. To do so we have to go to 310.15(A). When we get there, we find that the exception to 310.15(A)(2) nullifies the calculation we just did in the first step, because it allows us to apply the ampacity of the portion of the circuit that is not in the hole, to the portion of the circuit that is in the hole. I agree with Ryan, in that, IMO, the second paragraph of 334.80 will hardly ever (only when we have a very short total circuit length) have any relevance, because of the allowance in the exception to 310.15(A)(2).

Edit to add: I do believe that because of the exception in 310.15(A)(2), the derating of NM cable is not necessary as often as many believe. If they are bundled for say 24", then the length of conductors not bundled would need to be less than 20 feet for derating to be necessary. I don't know that I would call that circumstance rare, but it is often not the case.

[This message has been edited by eprice (edited 03-23-2006).]

Ryan,
I did ask Jeff Sargent of NFPA if he thought 310.15(A)(2)exc. (10 foot or 10% rule) cancelled 334.80's second paragraph, and he answered it would seem that it does.
Plus the fact that up to four cables of small conductors can be bundled without regard for derating anyway. If you do the math, you still will be allowed to connect #14s to a 15 amp cb, #12s to a 20, and #10s to a 30 amp cb.