Neutrals as current carrying conductors

[crawdad]

I have recently been involved in the following scenario.

A single conduit 3/4" EMT contains 5 single phase 120V circuits fed off a 208/Y/120V panel (3 phase, 4 wire). This conduit contained 7 #12 AWG conductors and one #12 AWG ground conductor.

I had immediate concerns and so separate neutrals were indicated to bring the number of conductors in the raceway to the count of 10 plus a ground.

I still have the concern of de-rating and installing a ground for each circuit for a total of 5 #12 AWG ground wires.

The point of conflict has arisen over neutrals in this scenario and whether they are current carrying. I can explain this via vector analysis and Ohm's law as yes, but I am unable to prove (argue) this point from a Code perspective.

HELP

Thanks in advance

 

[charlie b]

Re: Neutrals as current carrying conductors

Every once in a while we get what appears to be a simple issue, and it turns out to be a very interesting (if not downright entertaining) question. I think this is one. Thanks for bringing this issue up.

First of all, your answer lies in 310.15(B)(4). Every neutral in this conduit will carry the current associated with its corresponding hot wire. That is not the same as carrying ?only the unbalanced current from other conductors of the same circuit.? Therefore, they count as being current-carrying conductors, for the purposes of derating per Table 310.15(B)(2)(a). You are correct.

Now, why do I find this question interesting? Because if you kept just the 7 original conductors in the conduit, using a shared-neutral configuration, then the derating factor would have been 70%. But by adding 3 conductors, so that there are no longer any shared neutrals, you bring the total number of current-carrying conductors to 10, and therefore must use a 50% derating factor. In other words, your modification didn?t reduce the total neutral current, and in fact it REDUCED (!!!) the total I^2 R losses within the conduit, yet the code makes you derate even further. Go figure!

 

[noxx]

Re: Neutrals as current carrying conductors

I don't have my book in front of me to quote from so I'll phrase this as a question. Isn't there an exception in the derating table for diverse loads, which would bring this conduit back to 70%, given of course that the loads are in fact diverse, and not continous lighting etc?

-Noxx

 

[crawdad]

Re: Neutrals as current carrying conductors

Ooops!

Thanks for reminding me "charlie b"

One crucial bit of information. ALL FIVE CIRCUITS are on PHASE A at the panelboard. Hence this is why the mod to five neutrals.

 

[Ed MacLaren]

Re: Neutrals as current carrying conductors

I still have the concern of de-rating and installing a ground for each circuit for a total of 5 #12 AWG ground wires.

The word "ground" actually has no meaning in a technical discussion. Did you mean grounded (neutral) conductor, or grounding (equipment grounding) conductor?

Ed

[ April 30, 2003, 12:15 PM: Message edited by: Ed MacLaren ]

 

[crawdad]

Re: Neutrals as current carrying conductors

Thanks Ed

I did leave that information out and I meant "grounding conductor" as in a safety ground.
My concern is the ability to clear a fault when all circuits are faulted.
For this scenario I figured I could make a case for the additional grounds.
Once again this was a design I was inspecting and not designing so I did not do calcs for this area. But I would not use just a single #12 AWG if I were designing the circuit.

 

[charlie b]

Re: Neutrals as current carrying conductors

Originally posted by noxx: Isn't there an exception in the derating table for diverse loads, which would bring this conduit back to 70%, given of course that the loads are in fact diverse, and not continuous lighting etc?

For not having the book handy, I?d say you have a good memory. But it?s not an exception, it?s a Fine Print Note. It refers you to the non-enforceable Annex B, Table B310.11. The difficulty with using this Table is the notion of ?load diversity.? That term is not related to ?continuous versus non-continuous,? and taking advantage of the Table may require ?engineering supervision.? The term ?diversity factor? relates to the sum of the loads on all five circuits, as compared to the total load observed at any one time. In other words, if you never run all five loads at the same time (e.g., you turn off one fan before you turn on another fan, just to give them equal wear and tear), then there is a possibility of using Table B310.11.

 

[charlie]

Re: Neutrals as current carrying conductors

In other words, your modification didn?t reduce the total neutral current, and in fact it REDUCED (!!!) the total I^2 R losses within the conduit, yet the code makes you derate even further. Go figure!

Actually, the reason for not counting the neutral is that it carries the unbalanced current. Looking at a single three phase circuit, you will have only three wires carrying current OR the neutral will carry the current the phase conductors are not carrying. The net result is the amount of current that three wires would carry. By adding the other conductors, you have increased the current by a factor of two more wires since each grounded conductor will be carrying the same current as the phase conductor.

This is all predicated on no harmonics to throw a monkey wrench into the calculations. HMMM . . . do we use monkey wrenches? LOL I guess we do, at times.

 

[charlie b]

Re: Neutrals as current carrying conductors

This specific question relates to a three phase panel, but all five loads powered through this one conduit are on Phase A. So it?s not a matter of the neutrals carrying unbalanced currents from a collection of a Phase A hot, a Phase B hot, and a Phase C hot. In the initial configuration, the total current in the five hot wires will equal the total current in the two neutrals. In the revised configuration, the total current in the five hot wires (which did not change) will equal the total current in the five neutrals. Thus, the total neutral current will be the same in both cases, but the amount of current in each neutral wire will have dropped.

My point of interest comes from the fact that the total heat losses will be less in the revised configuration, but that we must derate the revised configuration more than the original. It?s true, the code requires it, but I find it curious nonetheless.

 

[charlie]

Re: Neutrals as current carrying conductors

all five loads powered through this one conduit are on Phase A

OOPS Sometimes ya gotta pay attention!

 

[dana1028]

Re: Neutrals as current carrying conductors

crawdad - RE:

I did leave that information out and I meant "grounding conductor" as in a safety ground.
My concern is the ability to clear a fault when all circuits are faulted.
For this scenario I figured I could make a case for the additional grounds.
Once again this was a design I was inspecting and not designing so I did not do calcs for this area. But I would not use just a single #12 AWG if I were designing the circuit.

In the NEC Handbook (1999 & 2002) - Art. 250-122(c)commentary - "A single EGC is required to be sized for the largest OCPD and IS NOT required to be sized for the composite of all the circuits in the raceway. It is not anticipated that all circuits will develop faults at the same time." ...in this same vein, it is not necessary to have a separate EGC for every circuit in a raceway as 'it is not anticipated that all circuits will develop faults at the same time.'

Another piece of information (as a FYI) regarding the ability of a #12 to clear ALL five circuits should they all fault simultaneously - though the NEC does not permit a #12 to be used as an EGC for a 100A circuit (5 ckts x 20A), if you take a look at Soares 'Book on Grounding' - Chap. 11 (8th Ed.) - they cover the subject of 'Clearing Ground Faults' . They state: "It can be shown...using values from the Insulated Cable Engineers Assoc. Pub. P 32-382 that an insulated copper conductor...may safely carry one ampere for every 42.25 circular mils for 5 seconds without destroying its validity." .... 5 seconds...that's about 75 times longer than it takes to clear a fault. Using these values, a #12 could carry 155A for 5 seconds without damage.