3 Phase 3 Wire Minimum Size Neutral and EGC

[mbrooke]

For a 4000amp 3 phase 3 wire service, what is the minimum size service neutral needed? And for each parallel set? What about the EGC past the main disconnect to the main lug main distribution gear?

 

[LarryFine]

For a 4000amp 3 phase 3 wire service, what is the minimum size service neutral needed?

Same as GEC.

And for each parallel set?

Each EGC needs to be sized as if it was the only one.

What about the EGC past the main disconnect to the main lug main distribution gear?

Ditto.

 

[Carultch]

For a 4000amp 3 phase 3 wire service, what is the minimum size service neutral needed? And for each parallel set? What about the EGC past the main disconnect to the main lug main distribution gear?

Seems like a trick question. You may have not intended it that way, but a 3 phase/3-wire service by definition would not have a neutral. This would be a service like an ungrounded delta or a corner-grounded delta. While it is a grounded conductor, the grounded phase of a corner-grounded delta grid is not technically a neutral. In those cases, all three wires would need to be the full size, and the neutral sizing algorithm wouldn't apply.

If it were a 3-phase/4-wire system, I agree with Post #2.

 

[augie47]

See 250.24(C) for service and 250..122 for load side.

 

[mbrooke]

Seems like a trick question. You may have not intended it that way, but a 3 phase/3-wire service by definition would not have a neutral. This would be a service like an ungrounded delta or a corner-grounded delta. While it is a grounded conductor, the grounded phase of a corner-grounded delta grid is not technically a neutral. In those cases, all three wires would need to be the full size, and the neutral sizing algorithm wouldn't apply.

If it were a 3-phase/4-wire system, I agree with Post #2.

Right, but then I would have folks asking me about line to neutral loads after the disconnect.

 

[mbrooke]

See 250.24(C) for service and 250..122 for load side.

Am I correct to assume that the grounded service conductors would be smaller then the load side EGCs?

 

[tom baker]

If there are no line to neutral loads then you size it per 250.122
For the service conductors, the neutral is a white wire with an imaginary green stripe. You size it to the larger of unbalanced current - white or 250.66 green wire. Like Augie said.
Sometimes the neutral can be smaller than the EGC just depend on the OCPD sizing

 

[texie]

The grounded conductor minimum size would be based on 250.102, not 250.66. In larger services like this it will make a difference.

 

[don_resqcapt19]

Am I correct to assume that the grounded service conductors would be smaller then the load side EGCs?

They will be, because the size of the grounded conductor is based on the size of the largest ungrounded conductor in each raceway, while the EGC has to be full sized in each raceway based on the upstream OCPD.

 

[kwired]

Three phase three wire has no neutral. this has to be either corner grounded or ungrounded delta.

Most the time with either of those you need all three same size per the load, but with corner grounded delta I guess if you actually have less load on the grounded conductor it would be possible for it to be smaller than the ungrounded conductors. still would not be allowed to be smaller then the required grounding electrode conductor though.

Load size of the service regardless of supply type, it can't be any smaller than the required EGC of the feeder circuit it is associated with. There is no minimum grounded conductor size on branch circuits as a general rule other than it must be able to carry the load. Seem to recall there is a minimum size for ranges or maybe even some other specific applications though.

 

[texie]

They will be, because the size of the grounded conductor is based on the size of the largest ungrounded conductor in each raceway, while the EGC has to be full sized in each raceway based on the upstream OCPD.

Yep, one of the quirks of the code. This is what many point out in the discussions about revamping the code in regards to sizing EGCs, supply side stuff, etc.

 

[mbrooke]

They will be, because the size of the grounded conductor is based on the size of the largest ungrounded conductor in each raceway, while the EGC has to be full sized in each raceway based on the upstream OCPD.

That doesn't make any sense. If the service conductors can survive the fault duration so can the EGCs.

 

[mbrooke]

Three phase three wire has no neutral. this has to be either corner grounded or ungrounded delta.

Most the time with either of those you need all three same size per the load, but with corner grounded delta I guess if you actually have less load on the grounded conductor it would be possible for it to be smaller than the ungrounded conductors. still would not be allowed to be smaller then the required grounding electrode conductor though.

Load size of the service regardless of supply type, it can't be any smaller than the required EGC of the feeder circuit it is associated with. There is no minimum grounded conductor size on branch circuits as a general rule other than it must be able to carry the load. Seem to recall there is a minimum size for ranges or maybe even some other specific applications though.

Right, but in 3 phase 3 wire you still need to bring a grounded conductor to the service when a center point is grounded down.

 

[don_resqcapt19]

That doesn't make any sense. If the service conductors can survive the fault duration so can the EGCs.

That has been told to CMP 5 a number of times in various proposals. While it just makes sense that the supply side bonding jumper is the worst case and permits a smaller conductor than on the load side of the OCPD, no one has submitted a technical substantiation to the code panel to make the change. I had some discussions with one of the panel members, and he agrees that it is logical, but logic is not considered a technical substantiation, and without a technical substantiation, CMP 5 will not make a change.

 

[mbrooke]

That has been told to CMP 5 a number of times in various proposals. While it just makes sense that the supply side bonding jumper is the worst case and permits a smaller conductor than on the load side of the OCPD, no one has submitted a technical substantiation to the code panel to make the change. I had some discussions with one of the panel members, and he agrees that it is logical, but logic is not considered a technical substantiation, and without a technical substantiation, CMP 5 will not make a change.

Understood, and well said.

However ohms law is enough logic... fault current flows in a loop. So if part one of the loop can survive it so can part two of the loop assuming wire temps, types and insulation is the same.

 

[kwired]

Right, but in 3 phase 3 wire you still need to bring a grounded conductor to the service when a center point is grounded down.

If you have a grounded center point it is three phase four wire not three wire.

 

[kwired]

Understood, and well said.

However ohms law is enough logic... fault current flows in a loop. So if part one of the loop can survive it so can part two of the loop assuming wire temps, types and insulation is the same.

EGC never has to be larger than the ungrounded conductors. sometimes can be smaller though.

 

[mbrooke]

If you have a grounded center point it is three phase four wire not three wire.

4 wire up to the main, after that its technically 3 wire.

 

[don_resqcapt19]

Understood, and well said.

However ohms law is enough logic... fault current flows in a loop. So if part one of the loop can survive it so can part two of the loop assuming wire temps, types and insulation is the same.

You will have an opportunity to try and convince CMP 5 if you make a Public Comment on one of the following Public Inputs that were submitted on this subject. Those PIs are 564, 1591, and 2773. These all addressed the issue of the size of the EGC in a parallel raceway being larger than the ungrounded conductors in that raceway. These PIs were all "resolved" (rejected) at the panel meeting a couple of weeks ago. The panel comments will be published as part of the First Draft Report on 7/2. At that point anyone can comment on the panel action in an attempt to change the minds of the panel members.

 

[mbrooke]

You will have an opportunity to try and convince CMP 5 if you make a Public Comment on one of the following Public Inputs that were submitted on this subject. Those PIs are 564, 1591, and 2773. These all addressed the issue of the size of the EGC in a parallel raceway being larger than the ungrounded conductors in that raceway. These PIs were all "resolved" (rejected) at the panel meeting a couple of weeks ago. The panel comments will be published as part of the First Draft Report on 7/2. At that point anyone can comment on the panel action in an attempt to change the minds of the panel members.

PIs are here?

NFPA 70®: National Electrical Code®

<p>90.2(C) Installations Covered.<br /><br />This Code covers the installation and removal of electrical conductors, equipment, and raceways; signaling and communications conductors, equipment, and raceways; and optical fiber cables for the following:<br /><br /> (1) Public and private...

www.nfpa.org