Sizing the Neutral

[erickench]

I understand what you are saying about bringing the grounded conductor into the service enclosure and grounding it. It would carry the fault current yes. But is the grounded/neutral conductor large enough to handle the fault current? Do you think that NEC 250.66 would be adequate enough for a grounded system? Maybe NEC 220.61 would be adequate since it would give a higher ampacity. But what if there is no disconnect in the grounded conductor?

 

[don_resqcapt19]

I understand what you are saying about bringing the grounded conductor into the service enclosure and grounding it. It would carry the fault current yes. But is the grounded/neutral conductor large enough to handle the fault current? Do you think that NEC 250.66 would be adequate enough for a grounded system? Maybe NEC 220.61 would be adequate since it would give a higher ampacity. But what if there is no disconnect in the grounded conductor?

It is assumed that the fault current will be of short duration and the higher current for short time will not damage the conductor. There is a formula that says you can run 1 amp for 5 seconds for each 48 circular mils of conductor size without damage to the conductor insulation. (The 48 cm is just from memory and may not be correct, but it is close) From this you can get the amperes squared second value that is used to solve for the actual fault clearing time.

 

[erickench]

Okay, I'm going to settle this right now. NEC 250.24 (C) (1) should be deleted and replaced with a rule that says "the grounded conductor brought to the service should be sized according to NEC 220.61". This way it would be able to handle the maximum unbalanced load current. You're right the overcurrent device would interrupt the circuit in case of a fault. But in an unbalanced condition the current could go as high as the feeder rating. Suppose you had a three-wire single phase circuit. If one of the ungrounded conductors is open(no load) then the maximum current from the other ungrounded conductor would flow along the feeder/service neutral and then back to the source which is the utility. NEC 250.66 would not give you an adequate conductor size for the neutral regardless if the system is grounded or not.

 

[roger]

Okay, I'm going to settle this right now. NEC 250.24 (C) (1) should be deleted and replaced with a rule that says "the grounded conductor brought to the service should be sized according to NEC 220.61". This way it would be able to handle the maximum unbalanced load current. You're right the overcurrent device would interrupt the circuit in case of a fault. But in an unbalanced condition the current could go as high as the feeder rating. Suppose you had a three-wire single phase circuit. If one of the ungrounded conductors is open(no load) then the maximum current from the other ungrounded conductor would flow along the feeder/service neutral and then back to the source which is the utility. NEC 250.66 would not give you an adequate conductor size for the neutral regardless if the system is grounded or not.

I see the problem, you can not visualize a service without it serving loads that need a neutral.

If a neutral is needed to serve L-N loads 220.61 would come into play, if a neutral is not needed, ie all L-L loads, 250.24(C)(1) is all you need to consider.

Roger

 

[erickench]

Yes but in a three wire single phase system the grounded conductor is the centertap of a transformer and not the centerpoint of a wye. You would still have the same problem even if it's not an actual neutral. You could have a corner grounded delta system. But the grounded conductor must still be able to handle the same amount of maximum current coming from one of the other ungrounded corners(conductor). Think of it this way. You pour water into a bucket with a hole in the bottom. The same amount of water poured in must come out of the hole.

 

[erickench]

Roger, here is the only explanation that I could come up with. In a grounded system the current going back to the source(utility) first comes to the service enclosure and then divides into two components. One component flows along the grounded service conductor and the other flows though the grounding electrode into the earth and finds it's way back to the source(utility). The current flowing through the grounded service conductor is not as high as the current flowing through the ungrounded service conductor. This has to be the reason for sizing the grounded service conductor in accordance with 250.66.

 

[roger]

Yes but in a three wire single phase system the grounded conductor is the centertap of a transformer and not the centerpoint of a wye.

It doesn't matter and is irrelevant, the grounded conductor in either system will clear a fault.

You would still have the same problem even if it's not an actual neutral. You could have a corner grounded delta system.

And the next fault of either of the other phases would open the service OCPD.

But the grounded conductor must still be able to handle the same amount of maximum current coming from one of the other ungrounded corners(conductor). Think of it this way. You pour water into a bucket with a hole in the bottom. The same amount of water poured in must come out of the hole.

You are still thinking about a neutral, just think about an EGC.

Roger

 

[erickench]

Okay, then let's look at these conclusions. For grounded systems the grounded service conductor would be sized according to NEC 250.66. For ungrounded systems the grounded service conductor would be sized according to NEC 220.61. One tends to think of the earth as a very high resistance and therefore most of the current would go through the wire. Thus the confusion.

 

[Smart $]

Okay, then let's look at these conclusions. For grounded systems the grounded service conductor would be sized according to NEC 250.66.

Incorrect!

For ungrounded systems the grounded service conductor would be sized according to NEC 220.61.

There is no grounded service conductor for ungrounded systems.

One tends to think of the earth as a very high resistance and therefore most of the current would go through the wire. Thus the confusion.

Your statement is accurate, but I don't see where the confusion stems from.

 

[erickench]

I meant the neutral conductor in an ungrounded system. That is sized in according to 220.61. All the return current would flow over the service neutral in an ungrounded system. So both the feeder and the service neutral would be sized according to 220.61 in an ungrounded system. But in the case of a grounded system the return current would divide at the service. Part of it goes through the earth and the rest flows through the grounded service conductor. Thus the grounded service conductor does not have to be as large as the ungrounded conductor. It could be sized according to NEC 250.66 but the feeder neutral would have to be sized according to 220.61. You would then have different size neutral conductors terminating at the neutral terminal bus inside the service neutral conductor. I told you this subject is confusing.

 

[erickench]

Boy am I getting mixed up with this. I meant different size neutral conductors terminating inside the service enclosure. In a grounded system the grounded feeder conductor would be sized according to 220.61 since it carries all the current. But the grounded service conductor does not carry all the current so it's sized according to 250.66 which would give it a smaller size. Please excuse my previous misstatements.

 

[Smart $]

Boy am I getting mixed up with this. I meant different size neutral conductors terminating inside the service enclosure. In a grounded system the grounded feeder conductor would be sized according to 220.61 since it carries all the current. But the grounded service conductor does not carry all the current so it's sized according to 250.66 which would give it a smaller size. Please excuse my previous misstatements.

Service and Feeder Neutral conductors both are sized according 220.61, i.e. their minimum ampacity. In most cases these Neutral conductors are also the grounded conductor. Therefore the service grounded conductor must also meet the requirement of 250.24(C)(1), which requires the grounded conductor be no smaller than the GEC specified in T250.66 and not larger than the largest ungrounded service entrance conductor. This is because not only must it be sized to handle load current, it must also be large enough to handle ground fault current.

Therefore the grounded service conductor's size must be at least the larger of the two required minimums of 220.61 and 250.24(C)(1). AFAIK there would be no instance where the grounded service conductor (line side) would be smaller than the grounded feeder conductor (load side) at the service enclosure.

 

[erickench]

Then why even have 250.24 (c) (1) if you know that you would have a grounded conductor no smaller than the ungrounded conductors? All it would do is just confuse everybody. I think the reason is because some of the current goes through the grounding electrode conductor and into the earth. In the NYC Transit Authority we never grounded the neutral because the stray earth currents would cause electrolysis and then corrosion of structural metal. Our neutrals were always sized the same as the phase conductors. But in a grounded system the code states in 250.24 (c) (1) that you could have a smaller neutral, otherwise the rule itself would serve no purpose.

 

[Smart $]

Then why even have 250.24 (c) (1) if you know that you would have a grounded conductor no smaller than the ungrounded conductors?

That's not what I said. The size comparison is grounded service conductor to grounded feeder conductor at the service disconnect. The minimum required size of the former is never smaller than the latter.

Personally I do not believe the NEC considers current returning to its source via earth in the determination of conductor size for ground fault currents.

 

[erickench]

Yes I was a bit confused at first because the earth has such a high resistance and that very little current would flow through it. But if you would examine NEC 250.66 you would note that the area(cmils) of the conductor is 1/4 of that of the ungrounded service conductor. Therefore the resistance is increased by a factor of four. This creates heat. Now let as suppose you had a 100A 3-wire single phase system. The circuit breaker is going to allow a maximum of 100A to go through. Assume that one of the ungrounded conductors is open(no load). The maximum current flowing through the neutral would be 100A. You can't just assume a partial load as being the maximum. It would be good engineering practice to assume the maximum permissible amps flowing through the circuit breaker and through the neutral as well. So what is the purpose of 250.24(c)(1) if you know that you are going to have so many amps and that you would have to size the grounded(or neutral) conductor the same as the ungrounded conductors? Why was this written into the code? Service conductors do have overload protection but how is the protection going to work if the code says that you are permitted to have a smaller grounded(or neutral) conductor? That cable is going to fry. Of course in the real world we could deal with this problem very easily just by sizing the grounded conductor(or neutral) to be as large as the ungrounded conductor.

 

[roger]

Where is there a "neutral load" in the illustration below?

Roger

 

[erickench]

Okay the motor does not have a neutral. But what if we had a load that did have a neutral? What if someone came along and changed the system components and subsequently removed the motor? Why does 250.24(c)(1) allow the neutral conductor to be sized smaller than the ungrounded conductors if it's known that approximately the same amount of current(minus earth current) is going to flow through the service neutral? How is this smaller conductor going to handle all this current? I'm not questioning anybody's methods in solving the problem. I'm questioning the code rule itself. You have to assume changing conditions.

 

[iwire]

You have to assume changing conditions.

In my opinion the NEC does not assume anything about the future, only what exists today.

For example check out Exception 3 to 230.90(A). In that section the NEC is counting on the calculated load to protect the service conductors which means in the future any additions to the load would have to be scrutinized against the service conductor ampacity.

90.1(B) is also worth a look.:smile:

 

[erickench]

Yes overload protection is required to protect the ungrounded conductors which are larger. But what protects a grounded neutral that is sized smaller?
Normally you would want a neutral sized the same as the ungrounded conductors. Okay maybe they should add a note that for loads without a neutral, the grounded service conductor can be sized according to NEC 250.66.
Otherwise it's just going to confuse people. I know that the NEC is updated and changes are made at every cycle. As anyone ever put through a proposal to change NEC 250.24(c)(1)?

 

[roger]

Otherwise it's just going to confuse people. I know that the NEC is updated and changes are made at every cycle. As anyone ever put through a proposal to change NEC 250.24(c)(1)?

Eric, don't take this wrong but, since it is so confussing to you, why don't you send in a proposal and maybe when the CMP explains their action it might be more clear to you than the way we are trying to explain it.

Roger