705.12(A)

[HoosierSparky]

I am having discussions with applicants on what this section requires. Per NEC 2020 705.12(A), "Each source interconnection of one or more power sources installed in one system shall be made at a dedicated circuit breaker or fusible disconnecting means." The way I read this is you can have several power sources (inverters, ESS, generators, etc.), but they must 1st be combined prior to making a single connection at the main panel. This provides a single disconnect of all supplemental power sources backfeeding at the main panel.

The key wording is SHALL BE MADE AT A DEDICATED CIRCUIT BREAKER OR FUSIBLE DISCONNECTING MEANS. Specifically the connection SHALL be made at A. A being singular. In addition it remains singular when stating CIRCUIT BREAKER. BREAKER is singular also. It doesn't say breakers.

What I think the CMP was and is looking for is a single disconnecting means AT the MSP for all supplemental power, not multiple PV breakers at the MSP. Am I wrong and if so how?

 

[wwhitney]

Yes, you are mistaken. It says "each source interconnection of one or more power sources" shall have a dedicated disconnecting means. By using the word "each" at the beginning that language recognizes you may have multiple source interconnections. There is no language requiring a single source interconnection for all the power sources.

So if you have, say, 6 inverters, you can choose to use 6 source interconnections, where each interconnection is for one power source, i.e. one inverter. That would give you 6 disconnecting means, each one of which is dedicated to one interconnection of one source. Or you could choose to combine all 6 inverters into a single interconnection and use a single disconnecting means for a single interconnection of 6 sources. Or you could combine them into 2 groups of 3, and have two disconnecting means for two interconnections, each interconnection being of 3 sources. Etc.

Cheers, Wayne

 

[jaggedben]

No, you are incorrect. It simply means that sources may not be put on a branch circuit with loads. There is no compelling safety reason for all sources to be combined in a single place (as long as signage is provided) and in some cases it is impossible to do so. It also would conflict, for example, with 690.15(C) which speaks of a PV system with up to six switches as a disconnecting means.

Put another way, I can call anything I want 'one system'. The PV is one (or more) system(s), the ESS is one (or more) system(s), etc. There is no requirement to combine all systems.

To be fair, I think the wording is confusing and the CMP has still not gotten it right. But if you go back and look at the revision information for I think the 2014 and 2017 codes, you will see the intent. At one point there was a proposal to actually add language about not sharing with loads but it was rejected as unnecessary, or something.

Edit: posted a second after Wayne.

 

[ggunn]

Yes, you are mistaken. It says "each source interconnection of one or more power sources" shall have a dedicated disconnecting means. By using the word "each" at the beginning that language recognizes you may have multiple source interconnections. There is no language requiring a single source interconnection for all the power sources.

So if you have, say, 6 inverters, you can choose to use 6 source interconnections, where each interconnection is for one power source, i.e. one inverter. That would give you 6 disconnecting means, each one of which is dedicated to one interconnection of one source. Or you could choose to combine all 6 inverters into a single interconnection and use a single disconnecting means for a single interconnection of 6 sources. Or you could combine them into 2 groups of 3, and have two disconnecting means for two interconnections, each interconnection being of 3 sources. Etc.

Cheers, Wayne

Agree but with one caveat; some AHJs require that there not be multiple points of interconnection on a single service. The most common cause of this I have run into is when they require metering of PV output; they usually do not want multiple PV meters.

 

[wwhitney]

Agree but with one caveat; some AHJs require that there not be multiple points of interconnection on a single service.

I think you mean POCO in the above, rather than AHJ enforcing the NEC. I infer that because you deal with various municipal utilities you sometimes conflate the two. But an AHJ enforcing the NEC would have no justification to impose such a requirement. A POCO of course can add their own rules.

The most common cause of this I have run into is when they require metering of PV output; they usually do not want multiple PV meters.

A single PV meter could still supply multiple 705.12(A) disconnects, so I don't see this alone is not a reason for a POCO to require a single overall disconnect.

Cheers, Wayne

 

[HoosierSparky]

I am STARTING to get it. I am just hung up on the part where it states one or more systems. I get that. It is where it states where they are installed in ONE system. I take that as the MSP. At that point is would be ONE PV BREAKER. To get there they would need to connect ALL systems at a combiner and then to the MSP. Now my head starts to hurt!!!

 

[ggunn]

I think you mean POCO in the above, rather than AHJ enforcing the NEC. I infer that because you deal with various municipal utilities you sometimes conflate the two. But an AHJ enforcing the NEC would have no justification to impose such a requirement. A POCO of course can add their own rules.

In the two cities where I have run into this most often, the AHJ and the POCO are one and the same - municipally owned utilities. Unless there are distinctions of which I am unaware, AHJ is the more general term; I take it literally at face value. To my view, an authority having jurisdiction is any entity that has the legal authority to tell me what I can and cannot do, and there can be more than one. It seems to me that the definition in Article 100 backs this up.

A single PV meter could still supply multiple 705.12(A) disconnects, so I don't see this alone is not a reason for a POCO to require a single overall disconnect.

Cheers, Wayne

Perhaps I am misunderstanding you, but I don't know of a case where multiple inverters have been combined into a single feeder for metering and then split out into multiple points of interconnection; I know that I have never designed one that way and I question whether it would be approved anywhere that I have done work. As I said, I may not be getting what you are saying, but two points of interconnection from a single feeder would be connecting those points to each other; that doesn't seem right.

 

[ggunn]

I am STARTING to get it. I am just hung up on the part where it states one or more systems. I get that. It is where it states where they are installed in ONE system. I take that as the MSP. At that point is would be ONE PV BREAKER. To get there they would need to connect ALL systems at a combiner and then to the MSP. Now my head starts to hurt!!!

Yes, and 705.12 governs the connections in the combiner panel as well as the load side connection to the service.

 

[wwhitney]

To my view, an authority having jurisdiction is any entity that has the legal authority to tell me what I can and cannot do, and there can be more than one.
It seems to me that the definition in Article 100 backs this up.

Usage-wise, in this forum, I take AHJ to be the entity enforcing the NEC, and would use a different term for any entity enforcing another set of rules (as well as specifying what that other set of rules is). Of course, your usage may vary.

Perhaps I am misunderstanding you, but I don't know of a case where multiple inverters have been combined into a single feeder for metering and then split out into multiple points of interconnection

I am proposing the following as compliant with 705.12(A) while also providing a single PV meter. From the utility towards the PV: service point -- common service conductors -- service conductors for PV only -- PV meter -- multiple sets of service entrance conductors -- one PV/service 705.12(A) disconnect for each set of PV service entrance conductors -- one or more inverters connected to each PV/service disconnect.

Cheers, Wayne

 

[wwhitney]

I am STARTING to get it. I am just hung up on the part where it states one or more systems. I get that. It is where it states where they are installed in ONE system.

Where does it state they are all installed in ONE system? 2020 NEC 705.12(A) says:

705.12(A) Dedicated Overcurrent and Disconnect. Each source interconnection of one or more power sources installed in one system shall be made at a dedicated circuit breaker or fusible disconnecting means.

It just says one dedicated circuit breaker or fusible disconnecting means per interconnection per system. You could have multiple interconnections per system, and multiple systems per service.

Cheers, Wayne

 

[ggunn]

Usage-wise, in this forum, I take AHJ to be the entity enforcing the NEC, and would use a different term for any entity enforcing another set of rules (as well as specifying what that other set of rules is). Of course, your usage may vary.

The definition in Article 100 does not say that. Its definition is decidedly more general, i.e., "An organization, office, or individual responsible for enforcing the requirements of a code, or standard, or for approving equipment, materials, an installation, or a procedure". That's my usage.

I am proposing the following as compliant with 705.12(A) while also providing a single PV meter. From the utility towards the PV: service point -- common service conductors -- service conductors for PV only -- PV meter -- multiple sets of service entrance conductors -- one PV/service 705.12(A) disconnect for each set of PV service entrance conductors -- one or more inverters connected to each PV/service disconnect.

Sorry, I am having a little trouble visualizing that, but if I get what you are saying I think that once the inverter outputs are combined to go through a PV meter, if you then split it into multiple feeders to hit multiple points of interconnection, you would be also connecting those points to each other through the point where the PV conductors diverge, and I do not think you can do that.

 

[wwhitney]

The definition in Article 100 does not say that. Its definition is decidedly more general, i.e., "An organization, office, or individual responsible for enforcing the requirements of a code, or standard, or for approving equipment, materials, an installation, or a procedure". That's my usage.

OK, but I guess I'm saying that when we write in this forum "the AHJ requires," if the requirement being enforced is unspecified, it by default means the NEC, and the AHJ in question is the AHJ enforcing the NEC. So I'm suggesting it would be clearer if you specify "the POCO requires" when you are referring to an AHJ enforcing POCO requirements that are not NEC requirements.

Sorry, I am having a little trouble visualizing that, but if I get what you are saying I think that once the inverter outputs are combined to go through a PV meter, if you then split it into multiple feeders to hit multiple points of interconnection, you would be also connecting those points to each other through the point where the PV conductors diverge, and I do not think you can do that.

The PV meter would be on the utility side of the points of interconnection, not on the PV side. If you prefer to think of the direction power flows from the PV towards the utility, it would be Inverters - multiple disconnects/points of interconnection - service conductors combine - single PV meter - service point.

Cheers, Wayne

 

[ggunn]

OK, but I guess I'm saying that when we write in this forum "the AHJ requires," if the requirement being enforced is unspecified, it by default means the NEC, and the AHJ in question is the AHJ enforcing the NEC. So I'm suggesting it would be clearer if you specify "the POCO requires" when you are referring to an AHJ enforcing POCO requirements that are not NEC requirements.

Use the term however you want but please stop telling me that the way I use it is wrong. I am using it per the definition in Article 100 of the NEC. Any authority which has jurisdiction over any aspect of my design of a PV system is, by definition, an AHJ. Sometimes there are more than one and sometimes the issues they are concerned with overlap, but hopefully they will agree.

The PV meter would be on the utility side of the points of interconnection, not on the PV side. If you prefer to think of the direction power flows from the PV towards the utility, it would be Inverters - multiple disconnects/points of interconnection - service conductors combine - single PV meter - service point.

What you have described is a 705.11 interconnection, not a 705.12 connection as we were discussing; there is no way I know of to combine inverter outputs to go through a PV meter and then split the output to connect to multiple points on the load side unless maybe they both connect on the load side of a main service disconnect, but why would anyone want to do that?

 

[Elect117]

I am STARTING to get it. I am just hung up on the part where it states one or more systems. I get that. It is where it states where they are installed in ONE system. I take that as the MSP. At that point is would be ONE PV BREAKER. To get there they would need to connect ALL systems at a combiner and then to the MSP. Now my head starts to hurt!!!

Mine too. lol.

"Each source interconnection of one or more power sources installed in one system shall be made at a dedicated circuit breaker or fusible disconnecting means."

Each source (solar , batteries, wind) of one or more supplies installed in one system (one inverter? or multiple inverters being one solar system?) shall be made at a dedicated breaker or fusible disconnecting means".

Ill expand?

Lets say we got a generator for back up and some peaker batteries that can also be on back up, and some solar. You wouldn't have one disconnect for the batteries. You would have the ATS and the inverter's disconnect ya? So two?

The batteries + generator = one system on the ATS

Batteries + solar = second system on a load side connection

Or is that a bad example because the ATS wouldn't be load side?

Or is one system to indicate that each inverter's output requires a individual breaker or fused D/C?

 

[HoosierSparky]

Some of you have taken a left turn. POCO's and AHJ's never were mentioned. Keep on point.

I've attached a sketch of what I've been talking about and what I think 705.12(A) was talking about.

 

[Elect117]

That looks compliant with the rule.

PV backfeed breaker is present. That would be the 705.12(A).

I would also expect each inverter output to have their own breakers in that combiner panel as well, so that would also be compliant depending on interpretation of system.

 

[ggunn]

Some of you have taken a left turn. POCO's and AHJ's never were mentioned. Keep on point.

I've attached a sketch of what I've been talking about and what I think 705.12(A) was talking about.

That looks OK to me as long as all your wire sizes and OCPD ratings are correct and you observe 705.12(B) in your MDP and combiner panel, although I do not think 705.12(A) applies except for the feeder between the disco and the combiner; what is your question?

 

[ggunn]

Some of you have taken a left turn. POCO's and AHJ's never were mentioned. Keep on point.

I've attached a sketch of what I've been talking about and what I think 705.12(A) was talking about.

You may be in trouble in your main panel, though. You didn't list any inverter sizes, but you show that the MDP has a 200A bus with a 200A main breaker. That limits you to a total of 32A from all your inverters per 705.12(B)(2), aka the 120% rule. You combiner panel is 200A; that implies to me that your total inverter current is more than 32A

 

[jaggedben]

Some of you have taken a left turn. POCO's and AHJ's never were mentioned. Keep on point.

I've attached a sketch of what I've been talking about and what I think 705.12(A) was talking about.

If I get rid of the 200A combiner panel and the PV disconnect and just land all four inverter breakers in the main panel I can just call each inverter a separate system, each with its own dedicated breaker. The code doesn't define a source or a system in any way that prohibits this.

It's also not defined by inverters or anything else so particular. Possibly the systems are not similar enough to be considered one system. For example, I might have a site with an older string inverter system that was installed first, then I come along and add a separate microinverter system later. So I have a new system and an old system and there is no sense in calling them one system. One system is a single source and the other is multiple sources installed in one system; doesn't matter.

I also might have a system on a main building connected directly to the service and another on an outbuilding connected to a subpanel that is connected to the service by a feeder. No problem with that either.

All that is required is that whatever is connected to my breaker or disconnect qualifies as a power source and I don't have unrelated loads connected to the breaker.

 

[ggunn]

If I get rid of the 200A combiner panel and the PV disconnect and just land all four inverter breakers in the main panel I can just call each inverter a separate system, each with its own dedicated breaker.

But he is still limited to 32A of total backfeed. (32A)(1.25) + 200A = 240A, which is 120% of the bus rating.