FAQ: When is a neutral a current carrying conductor?

[winnie]

I'd like to propose an addition to the FAQ.

<B> When does a neutral count as a current carrying conductors for purposes of 310.15(B)(4)?</B>

Code requires that the ampacity of conductors be adjusted if several conductors are bundled together, for example in a single raceway. The reason is _heat_. The rules of 310.15(B)(2) are there to prevent excessive temperatures caused by the heat production of several conductors bundled together.

In general, <B> all </B> of the conductors in a circuit will carry some current, and this includes the neutral conductor. Thus, in general <B> all </B> conductors count when applying 310.15(B)(2).

There are a few specific circumstances where the neutral doesn't 'count' for 310.15(B)(2); these circumstances are outlined in 310.15(B)(4).

These rule can be better understood if the focus upon the neutral is ignored. Rather than asking 'does the neutral carry current', ask 'Given this set of 3 (or 4) conductors, how much heat is produced?'

In general, the neutral will carry some current, even when 310.15(B)(4) tells us that we don't have to count it.

There are _two_ specific situations where this applies.

1) The neutral conductor of a 3 wire circuit derived from a _single_ phase system. In this case, even though there are _three_ conductors which potentially carry current, the _net_ heat produced by these three conductors is always less than or equal to that of _two_ fully loaded conductors.

2) The neutral conductor of a 4 wire circuit derived from a three phase wye connected system where there the major portion of the load is linear. In this case, even though there are _four_ conductors which potentially carry current, the _net_ heat produced by these four conductors is always less than or equal to that of _three_ fully loaded conductors.

In the following cases, the neutral must count for 310.15(B)(2)

a) Single phase circuits with _dedicated_ neutrals. In this case, there are _two_ conductors which potentially carry current, and the net heat produced will be that of two conductors.

b) three wire circuits consisting of two phase conductors and a neutral, derived from a three phase wye connected system. In this case, there are _three_ conductors which potentially carry current, and the net heat produced will be as much as that of _three_ fully loaded conductors.

c) four wire circuits consisting of three phase conductors and a neutral, where the major portion of the load is non-linear. In this case, there are _four_ conductors which potentially carry current, and the net heat produced can exceed that of _three_ fully loaded conductors.

Basically the implications of 310.15(B)(4) are that we only have to count a total number of current carrying conductors equal to the worst case heating expected for a given circuit.

-Jon

 

[charlie b]

Sorry, but I cannot go along with your approach.

The reason is heat. The rules of 310.15(B)(2) are there to prevent excessive temperatures caused by the heat production of several conductors bundled together.

That is true. It is the physics behind the rule. But it is not the wording of the rule.

Thus, in general all conductors count when applying 310.15(B)(2).

I wouldn?t start the ?answer? to a FAQ there, because the code does not start the rule there. It starts with the notion of unbalanced current.

Rather than asking, ?does the neutral carry current,? ask, ?Given this set of 3 (or 4) conductors, how much heat is produced??

But that is not the way the rule is worded. It is no easy task to determine which combination of currents in various conductors will produce the most heat. So I would not call that an ?easy? way to judge any particular situation.

(2) The neutral conductor of a 4 wire circuit derived from a three phase wye connected system where there the major portion of the load is linear. In this case, even though there are four conductors which potentially carry current, the net heat produced by these four conductors is always less than or equal to that of three fully loaded conductors.

That is also true. I once took the time to prove it mathematically. But I would not expect all electricians to have loved mathematics enough to remember how to do some of the things taught in High School Algebra (e.g., manipulation of inequalities and factoring quadratic expressions). So I wouldn?t use this as the means of helping an electrician understand how to apply the rules.

I appreciate your attempt to help us here. But I tend to prefer to keep things as simple as possible. The simplest approach I can find on this topic is simply to read the rules, as written. They tell you whether to count the neutral in a number of scenarios. I think that is enough information for my purposes.

 

[winnie]

Given the number of times people ask about when to count the neutral as a CCC, I think that this does need to be addressed with something more than read the code

However I see your point about starting with the code itself, giving and giving the FAQ answer based upon the code. I'll mull on this and write something up later that gives the simple answer, possibly putting the physics in _after_ the answer.

-Jon

 

[tallgirl]

Winnie,

Something else to look at using in your explanation is how the current is balanced, because that's the language the code uses -- a neutral does not count as a current carrying conductor if all it is carrying is the unbalanced current.

A MWBC derived from a single phase system balances the current between the ungrounded phase conductors (the same thing happens with 3 phase, but I digress). As I understand the code, the issue is if the circuits are fully loaded, do the ungrounded phase conductors "balance" the load between themselves such that no current is carried on the neutral.

That's where I'd put the focus -- what does "balance" mean and how can you know if that's what's going on. THEN I'd explain why that's important using the physics.

 

[charlie b]

As I understand the code, the issue is if the circuits are fully loaded, do the ungrounded phase conductors "balance" the load between themselves such that no current is carried on the neutral.

Julie, I believe that you understand it correctly, but I am not sure you said it correctly. Perhaps what follows is essentially saying the same thing; perhaps not. You judge for yourself.

It is not necessary that the circuits are fully loaded, nor that at any point in time the neutral carries no current. But if, at a given moment, the loads on the ungrounded conductors are balanced, and if at that moment the neutral carries no current, and if the neutral would only carry current when you change the balance of loads (e.g., turn something on or off on one of the phases), then in that particular configuration the neutral is not counted as a “current-carrying conductor.” I think this is essentially what you were trying to say.

For example, in a MWBC consisting of two ungrounded conductors and one grounded conductor, all derived from a 120/208V three phase system, you could perfectly balance the loads on the two sides, and the neutral will still carry current. It doesn’t just get the unbalanced current, it gets it all.

Let me go back to Winnie’s point. Consider a MWBC consisting of two ungrounded conductors and one grounded conductor, all derived from a 120/240V single phase system.

CASE 1: Load Phases A and B fully and equally. Both conductors carry their maximum current, and both generate heat. The neutral carries no current and generates no heat.

CASE 2: Now turn off half of the loads on Phase A only. Phase B has the same current and generates the same heat. Phase A has a lower current and generates less heat. The neutral now has current, so it generates some heat.

RESULTS: The total heat in the second case must necessarily be equal to, or lower than, that in the first case. It can be proven with a bit of messy math. That is why it is OK to not count the neutral in this configuration. Nothing you can do with neutral current will give you more heat than you had in the balanced case.

 

[allenwayne]

To answer the OP`S question. On a 120 volt circuit the neutral is always counted a CCC.

But when you get into bundling the same count happens. A 3 wire is counted as 3 CCC`s a 2 wire is counted as 2 CCC`s.You just have to derate what is there depending on the install.Every install has their own particulars.JMHO
.

 

[George Stolz]

I'll mull on this and write something up later that gives the simple answer, possibly putting the physics in _after_ the answer.

How about writing the simple answer, and then we can edit in links to related threads (ie this one) at the bottom, to add in the rest of the story? For an example, check out this one. I wrote a brief explanation of the main point of each thread.

I agree with including this in the FAQ, it's a good topic - and I appreciate you going through with it, Jon. I look forward to seeing your next draft.

 

[George Stolz]

Cough cough, I look forward to seeing your next draft.

Man, I can be a nuisance...

 

[dnem]

There's 2 ways to go.
The long explanation and the short list.

You're working on the long explanation, I'd rather make the short list.

Don't count any of the following:

1] Non-current carrying conductors
. a] Equipment bonding/grounding conductor
. b] Travelers between 3ways or 4ways are never both current carrying conductors at the same time so don't count both. . Only count the pair as one.

2] Unbalanced current only conductors
. When all grounded hot conductors of the system are sharing a grounded conductor in a multiwire feeder or branch circuit and present in the wireway in question.

David

 

[George Stolz]

Pushy, Pushy

Is the neutral a current carrying conductor? How do I apply 310.15(B)(4)?

First the silly answer: the neutral always carries current.

310.15(B)(4) tells us that we don't need to account for neutral conductors that only carry 'the unbalanced current from other conductors of the same circuit'. These conductors still carry some current, but we are permitted to ignore them in our accounting when applying the adjustment factors of table 310.15(B)(2)(a).

This is found in two common situations.

1) A single phase feeder or multi-wire branch circuit consisting of two 'hots' (ungrounded conductors) and a single 'neutral' (grounded conductor). In this case, the neutral carries only the unbalanced current of the two hot conductors, we would count a total of two current carrying conductors.

2) A three phase feeder or multi-wire branch circuit consisting of three 'hots' (ungrounded conductors) and a single 'neutral' (grounded conductor) where the major portion of the load is linear. In this case, the neutral carries only the unbalanced current of the three hot conductors, we would count a total of three current carrying conductors.

The rest of the time we have to count the neutral.

To understand this, remember that 310.15(B)(4) is all about accounting, not about reality. This is about figuring out the correct number from 310.15(B)(2)(a) to apply, not about the minute details of how many conductors actually have a bit of current flowing on them.

The main requirement is that the conductor carry only the unbalanced current of the other conductors in the same circuit. This tells us that there have to be other conductors, and that these other conductors have to be able to carry the _balanced_ current of the circuit. If the other conductors can possibly carry all of the balanced current, then only the unbalanced current is left for the neutral. This is the case when 310.15(B)(4) kicks in. The neutral may carry some current, but it doesn't count for 310.15(B)(2)(a).

But if the 'hot' conductors are not balanced, then some of the balanced current must flow on the neutral. This this case with any single 'line-neutral' circuit, and is also the case when you have two phases of a three phase wye system sharing a neutral. The neutral carries both unbalanced and balanced current, and must be counted for 310.15(B)(2)(a).

Harmonics confuse this a bit: the neutral has to carry the _balanced_ triplen harmonic current of the load. We are permitted some leeway, but when the major portion of the load is non-linear, the neutral is no longer considered to carry _only_ unbalanced current from the other conductors, and thus must be counted.

-Jon

Sold.

 

[SurfSide EC]

. When all grounded hot conductors of the system

David

Is this a typo or am I missing something?

 

[winnie]

That is a typo. David meant to type 'ungrounded hot conductors'.

I have to admit the value of the simple list that he presents; my personal style tends toward explaining the physics of what is going on rather than just giving the answers. George pushed me on that exact point.

I personally believe that the answer is not good enough without some of the background; this way new situations can be handled correctly. But too much background is not a good thing; if you want to actually get work done than it is important to actually _know_ the answer rather than have to figure it out each time

Pop quiz:

A 120/208 4 wire feeder from a wye source feeds a panel. The panel is fully loaded with a large heater connected from the A leg to the B leg, with 120V loads fed between the C leg and the neutral. Is the neutral a 'current carrying conductor'.?

-Jon

 

[tallgirl]

Pop quiz:

A 120/208 4 wire feeder from a wye source feeds a panel. The panel is fully loaded with a large heater connected from the A leg to the B leg, with 120V loads fed between the C leg and the neutral. Is the neutral a 'current carrying conductor'.?

-Jon

Unless I'm missing something that's not a particular hard question and the answer is "Yes".

Which means I'm missing something and the answer is "No" ...

 

[charlie b]

I see something tricky in the wording of the question. If that trickiness is intentional, it may be enough to turn the answer into a "no."

The question is poorly worded, in two ways.

First, it talks of the 120 loads in the plural, but talks of the neutral in the singular. So when it asks whether the neutral (singular) is (again, singular) a current-carrying conductor (and yet again, singular), it should ask whether each neutral is a CCC. That answer would be ?yes.?

Secondly, it uses the word ?neutral? in two different contexts. In its final sentence, the only sentence that is a question, it uses the word ?neutral? in the context of a conductor. But in the preceding sentence, it uses the word ?neutral? in the context of the ?neutral bar,? the center point of the WYE configuration. So if I were to read the question literally, it is asking whether the neutral bar needs to be counted as a current-carrying conductor, as though it were a participant in the derating process. That answer is certainly ?no.?

So fess up, Winnie, what's your answer? What's your game?

 

[winnie]

My bad. I am not trying to be tricky in the wording, but trying to describe a tricky situation.

In the scenario I described, I want to know if the neutral of the feeder to the panel is a current carrying conductor. Clearly the neutral(s) to the single phase 120V loads are current carrying conductors.

As I see it, we have a 'full boat' going to this panel. Three hot conductors and a neutral. However the panel has been set up to be quite imbalanced in a tricky way. The load on the phase conductors is the same, but some of the load doesn't return to neutral, and some does.

Seeing that full boat, and not having any non-linear loads specified, most of us would say that the neutral doesn't count as a CCC.

By the physics of this rather hypothetical situation, the neutral most definitely _is_ a CCC.

The question, and one that I don't know the answer to: Given how the code is written, are we required to count this neutral as a CCC?

-Jon

 

[don_resqcapt19]

Jon,

A 120/208 4 wire feeder from a wye source feeds a panel. The panel is fully loaded with a large heater connected from the A leg to the B leg, with 120V loads fed between the C leg and the neutral. Is the neutral a 'current carrying conductor'.?

I am assuming that we are talking about the feeder grounded conductor and if so, they for the purposes of derating, it is a current carrying conductor in this case. If you would put some 120 volt loads on the other two phases, then the feeder neutral would no longer be a current carrying conductor.
Don

 

[charlie b]

By the physics of this rather hypothetical situation, the neutral most definitely is a CCC.

Not true. Definitely not true. (EDIT: After I posted this, I noticed that Don posted first, and that his answer disagrees with mine. We'll just have to see how this works its way out.)

What is true is that the neutral (meaning the one that is part of the panel feeder) will carry current. But the fact that it is carrying current does not make it a “Current-Carrying-Conductor.”

It may seem like I am being facetious, but I am not. The phrase, “Current-Carrying-Conductor,” although itself not included in the Article 100 definitions, does have a clearly intended meaning. It is used in Table 310.15(B)(2)(a) as a criterion for derating the ampacity of conductors. Not all conductive materials that happen to be carrying current are intended to be covered by that table.

The neutral is addressed in 310.15(B)(4). If that article says that a particular neutral must be counted as a CCC, then it must be counted.

The question, and one that I don't know the answer to: Given how the code is written, are we required to count this neutral as a CCC?

We are not. There is (perhaps, sometimes) an imbalance in loads being served by the three phase conductors. When that happens, the neutral will carry current. But the total heat generated by four wires (A, B, C, N) in this imbalanced situation is never higher than would be generated by three wires (A, B, C) in a balanced loading condition. The code authors knew that to be true. That is why we need not count the neutral as a CCC, when the only current it carries is the imbalanced current from the three phases.

 

[winnie]

As I said, the tricky part of the situation is the fact that the panel is imbalanced in a non-obvious way. As set up, all 4 conductors could carry full current. IMHO the neutral in this case _should_ count as a CCC.

Consider the parallel situation. A 20A 208V load and a 20A 120V load share the same conduit. 4 wires, 4 current carrying conductors. The code is clear here, because these are two _separate_ circuits. The neutral doesn't carry the imbalance from the other three conductors; it carries the current of the 120V circuit.

But a feeder is generally considered a single circuit by the code.

With the imbalanced panel that I suggested, current on leg A balances leg B, and current on the neutral balances leg C. The worst case scenario is that all four conductors could be fully loaded.

I could argue that since legs A and B are never connected to the neutral, this feeder really is two separate circuits. I could also argue that the neutral does not carry only imbalance current, since if all legs are loaded the neutral is still fully loaded. But I don't know if either of these arguments actually hold water in a code sense.

-Jon

 

[don_resqcapt19]

Jon,

As set up, all 4 conductors could carry full current. IMHO the neutral in this case _should_ count as a CCC.

Based on this set up, I think that the code requires this grounded conductor to be counted as a current carrying conductor because it is not carrying the unbalanced load other conductors in the circuit. It is my opinion that 310.15(B)(4)(a) supports this conclusion.
Don

 

[dnem]

. When all grounded hot conductors of the system

David

Is this a typo or am I missing something?

Oops !
Those 2 little letters sure change alot
Let me try this again

The short list.

Don't count any of the following:

1] Non-current carrying conductors
. a] Equipment bonding/grounding conductor
. b] Travelers between 3ways or 4ways are never both current carrying conductors at the same time so don't count both. . Only count the pair as one.

2] Unbalanced current only conductors
. When all ungrounded hot conductors of the system are sharing a grounded conductor in a multiwire feeder or branch circuit and present in the wireway in question.

Edited to add: . If harmonics or solid state devices are present that screw up the sinewave and load current onto the grounded conductor beyond the expected unbalanced current, then the grounded conductor is not a neutral and must be counted

David