Egc 250-122b

[steve66]

Steve, I think that I agree with you.

So in your opinion you would start with the smallest permitted conductors and any larger deviation from those conductors would make 250.122(B) applicable? So in the case of this scenario the EGC could be the same size?

Rob:

The starting point does seem a little vague here. But I think one could reasonably claim that either #8 or #6 is the "starting size conductor" for 50A. So I wouldn't have any problem with someone running a #10 ground with either.

Another issue that I think is somewhat similar: For a 400 amp feeder, I think one could claim that either 500KCM or 600 KCM is the starting size conductor.

In both cases, I would tend to give the "benefit of doubt", and say that the larger size is what you use as your starting point.

This may sound like I'm contridicting my previous point of "use the smallest size allowed", but for example, #8 wouldn't be allowed for a 50A circuit if the terminations are only rated 60 deg.

Steve

 

[Dennis Alwon]

Steve I think you are just showing what Charlie and I have stated. It isn't clear where the starting point is. If you are using 60C conductors then I believe I would start where the 60C conductor is. If I am using 90C then I would start there. I don't think you can use the 90C as a starter if you are using 60C insulation.

Now-- what difference would the insulation make? That is one of the problems with this article.

 

[steve66]

Steve I think you are just showing what Charlie and I have stated. It isn't clear where the starting point is. If you are using 60C conductors then I believe I would start where the 60C conductor is. If I am using 90C then I would start there. I don't think you can use the 90C as a starter if you are using 60C insulation.

Now-- what difference would the insulation make? That is one of the problems with this article.

Dennis:

I agree the starting point may be a little vauge in some cases, but I think that's pretty much limited to one or two wire sizes : #8 vs. #6 or 500KCM vs. 600KCM for the two examples that have been mentioned.

I don't think the same argument can be extended to claim that 3 sets of 750's are needed for an 800 amp breaker. To me it seems obvious that here the wire size has been increased due to mutual heating in a duct bank, or for some other reason.

Steve

 

[Dvanosdall]

Steve,
You are 100% correct. The sizing is due to being in a concrete encased duct bank, heat, with 4 of these 800 amp 480 volt circuits from one Switchboard.

 

[charlie b]

No, he is only 50% correct. It is true that the heat from nearby conduits has an influence on the ampacity of a conductor. It is not true that the ampacity values of table 310.16 COULD have been used, so that use of a larger conductor constitutes "increasing in size." That table flat out does not apply to underground ductbanks in which more than three conductors are in close proximity. That table is irrelevant to this installation. It is not the "starting point" for determining whether the ungrounded conductors are bigger than they have to be. If an engineering calculation determines the minimum size, the size that has just barely enough ampacity for the calculated load, then you don't get to look at table 310.16 to see if a smaller size could have been used. That table simply does not take into account all the factors that determine the "current, amperes, that a conductor can carry continuously" under these particular conditions of use, "without exceeding its temperature limit (adapted from the article 100 definition of ampacity).

 

[charlie b]

Someone is likely to call me on this point, so I will preempt that question.

That table (310.16) flat out does not apply to underground ductbanks in which more than three conductors are in close proximity.

And yet I have used it in exactly that context. How was I able to justify that use, and how am I not contradicting myself here? It is because most of my designs involve new buildings, or renovations, in which the load is calculated per article 220.

Situation #1:
It is commonly enough recognized that 220 is conservative, that if you calculate a load of 800 KVA you are likely never to see a real load higher than 400 KVA. Indeed, if you tell the utility that your calculated load is 800 KVA, they will give you a 300 KVA transformer. Thus, if you supply this building via an underground ductbank that has three sets of 500 MCM in three conduits that share an encasement, each conductor will not really be able to carry up to 380 amps without exceeding its temperature limit. But they won’t be called upon to carry that much current, since the real load will be far less than the calculated 800 KVA. I have a written statement from the senior inspector for the State of Washington, confirming that for new installations the state inspectors will allow the use of Table 310.16 for determining ampacity in underground ductbanks.

Situation #2:
The following situation is much different, however. Consider an existing Building B that is sub-fed from existing Building A, via an overhead carrier between the buildings. The owner plans to renovate Building B, and to replace the feeder with an underground ductbank. The load served via the existing overhead feeder is measured for 30 days. We add 25%, then add any planned increase in loads within Building B, and call that the “calculated load.” I would absolutely not accept the table 310.16 ampacity values for designing the new underground feeder. I would perform a Neher McGrath calculation, and I am certain the ampacity values of the conductors would be significantly smaller than the values given in 310.16.

What is different between the two situations? In Situation #1, I can count on the real load being well below the calculated load, and therefore the smaller wires that 310.16 would let me use will perform satisfactorily. In Situation #2, I expect the real load to be close to the calculated load, and therefore I cannot justify using the smaller wires that 310.16 would let me use.

 

[jumper]

Charlie,
I have no problem following what an EE tells me to do - it is usually overkill , so what do I care - , but as to the OP's problem. What if the calculation is wrong. Who is responsible if damage occurs? The engineer,who designed it, or the AHJ, who approved it?

 

[kingpb]

When an engineer puts his stamp on the document, he is the responsible party.

I have run into this situation myself, and to avoid conflict with the AHJ, I simply increase the EGC. I have found that it's ok to be right, but not at the expense of the project, or to have AHJ looking for things to nail you on.

So, in the end who wins; the engineer's right, the project is delayed for two days, the Owner's pissed off and won't hire you again, was it worth being "right"? My time is more valuable than that.

Just from this forum, you can see everyone has an opinion, and in each of our minds I'm sure we are convinced we have the best answer, especially when it comes to the ambiguities of the NEC. But, keep in mind the NEC is NOT a design document, it is minimum standards to ensure safety, and there are multiple ways to meet those minimum standards.

Jumping off soap box know.

 

[iwire]

Well put Kingpb.