New language in 705.12(A)

[Elexit]

Hello All. This is my first post on this forum.

The rules for supply-side interconnection of a PV system have been moved, from 690.64(A) in the 2008 NEC, to 705.12(A) in the 2011 NEC. The reason for the move is simple, since the same rule ought to apply to wind turbines, gas microturbines, and any other generator that might be paralleled with the utility on the customer side of the meter.

So far, so good. But along with the move, a sentence was added which I feel is too vague and could lead to misinterpretation:

The sum of the ratings of all overcurrent devices connected to power production sources shall not exceed the rating of the service.

705.2 defines Power Production Equipment, so it is quite clear that this definition excludes the utility. No confusion there.

But what does connected to mean? I interpret it to mean that one or more service disconnecting means (that is what a supply-side PV disconnect is, even though the NEC doesn't explicitly call it that) each of which is dedicated to power production equipment.

In other words, if you had two PV inverters, and each one was connected to a bussed service gutter through a 200A safety switch fused at 175A, and one or more service disconnecting means for the building loads, your service conductors and gutter busbars would have to have an ampacity not less than 350A. The minimum ampacity is the larger of the two: calculated load or installed PV OCPD.

But the phrase connected to doesn't say all of that. One could very easily interpret it in light of the old song "Dem Bones": The knee bone connected to the thigh bone, the thigh bone connected to the hip bone, etc, etc. In this interpretation, the service conductors would require sufficient ampacity for the sum of all service OCPD. After all, they are all connected, right?

I feel strongly that the former interpretation is correct, but it has sparked arguments within our office. Sooner or later, we will encounter an AHJ who uses the latter interpretation.

I am interested in other people's opinions on this subject.

 

[marti smith]

For some clarification, read: http://www.iaei.org/magazine/2005/09/making-the-utility-connection/

I'm sure it will help, although there are some folks here on the forum that, once seeing your post, will have some truly insightful additions, if you haven't seen already- and welcome, this is a good place!

 

[jaggedben]

But the phrase connected to doesn't say all of that. One could very easily interpret it in light of the old song "Dem Bones": The knee bone connected to the thigh bone, the thigh bone connected to the hip bone, etc, etc. In this interpretation, the service conductors would require sufficient ampacity for the sum of all service OCPD. After all, they are all connected, right?

No, not right. I'm taking a strong position. That interpretation is simply wrong.

I feel strongly that the former interpretation is correct,

To repeat myself, I agree with you.

but it has sparked arguments within our office.

I find it difficult to understand why. There are rules for how big a service has to be for the OCPDs of the loads connected directly to it. Those rules are elsewhere than 705. 705 deals with power production sources, not loads. Further to that point, if the wrong interpretation that you are devils-advocating were correct, there would be no need for the words "power production sources" to appear in that sentence of 705.12(A). Instead, the sentence could read "The sum of the ratings of all overcurrent devices shall not exceed the rating of the service." That is not what the CMP approved so I don't believe it is what they meant. Maybe it's naive of me...... But I believe that words are put in the code for reason, and that the reason is not merely to add more pages to the book. So when they say 'connected to power production sources', they are calling out a set of OCPDs, and they do not mean to include OCPDs connected to loads.

Finally there is common sense. If the loads cannot draw more power than the service is rated for, and the power production sources cannot produce more than the service is rated for, then the service cannot be overloaded, plain and simple.

 

[ggunn]

No, not right. I'm taking a strong position. That interpretation is simply wrong.



To repeat myself, I agree with you.



I find it difficult to understand why. There are rules for how big a service has to be for the OCPDs of the loads connected directly to it. Those rules are elsewhere than 705. 705 deals with power production sources, not loads. Further to that point, if the wrong interpretation that you are devils-advocating were correct, there would be no need for the words "power production sources" to appear in that sentence of 705.12(A). Instead, the sentence could read "The sum of the ratings of all overcurrent devices shall not exceed the rating of the service." That is not what the CMP approved so I don't believe it is what they meant. Maybe it's naive of me...... But I believe that words are put in the code for reason, and that the reason is not merely to add more pages to the book. So when they say 'connected to power production sources', they are calling out a set of OCPDs, and they do not mean to include OCPDs connected to loads.

Finally there is common sense. If the loads cannot draw more power than the service is rated for, and the power production sources cannot produce more than the service is rated for, then the service cannot be overloaded, plain and simple.

Maybe this is what you are saying, but there is no requirement that the sum of the load breakers be less than the service because it is very unlikely that all the load breakers would be fully loaded simultaneously. Even in that case, the main breaker would trip and the service would still be protected.

 

[jaggedben]

Maybe this is what you are saying, but there is no requirement that the sum of the load breakers be less than the service because it is very unlikely that all the load breakers would be fully loaded simultaneously. Even in that case, the main breaker would trip and the service would still be protected.

Pretty much exactly what I was saying. I would only add that the same reasoning still applies even if there are multiple service disconnecting handles. Otherwise you win on succinctness. :thumbsup: