Line Side Tap Neutral count as GEC to main panel?

Status
Not open for further replies.

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
250.24(C) and 250.92(A)(1)

The problem with Smart$ argument is that neither of these sections unambiguously applies to a supply side PV disconnect if such a disco is not already considered a 'Service Disconnecting means' or 'at the service' (respectively).

Again, if we simply change the definition of the service so that applies to any conductors and equipment connecting a utility to a premise, regardless of the direction of energy flow, then most of the ambiguity goes away.

The only remaining problem sections would be those that refer to the 'supply-side of the service disconnecting means' in a way that could be taken to imply, grammatically, that any such thing on the 'supply side' is not the service disconnecting means. This interpretation is contradicted by some other sections, such as 230.82(5) and 230.71(A), which imply that certain (not all?) of such connections are service disconnecting means or service equipment. Perhaps regular use of the word 'normal', as in 'supply side of the normal service disconnecting means' is all that is necessary to nix this interpretation. Currently the word 'normal' appears in only one such spot that I'm aware of (230.40 Exception 5), but at least that means there's a precedent for it.

In all, these proposals would substantially fix the problem, in my opinion, without adding more than one word to any section of the code.:D I think shortcircuit's proposal in this thread is also in the same spirit.
Why couldn't the NEC just say how a PV disco to a line side tap should be configured rather than depend on a chain of logic that is subject to interpretation? Or would that just make too much sense? :D
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
Technically, if you don't require a neutral for the PV System, you do not have to bring one to its disconnect. But then you'd be required to bond through other means as noted in my first sentence.
Some AHJ's require a full sized neutral through the disco and PV meter irrespective of whether or not the inverter uses it.
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
Ummm... For the disconnect enclosure, 250.92(A)(2), likely utilizing (B)(1), but not the only means [i.e. could be bonded using raceway(s) and/or jumper(s)].

Not 250.24(C). That's for Service Disconnecting Means. :angel:

Technically, if you don't require a neutral for the PV System, you do not have to bring one to its disconnect. But then you'd be required to bond through other means as noted in my first sentence.

You're contradicting yourself. In post #26 you said "the disconnect must have a neutral to enclosure bond." ggunn asked for where the code says that, and now your saying that it doesn't say it anywhere. So, if I don't have to bring a neutral and bond it, then why do I have to bond it if I do bring it? And if an inspector says to bond it or not bond it, what section of the code says he's right or wrong?

This is why the code needs to be revised. :roll:
 

Smart $

Esteemed Member
Location
Ohio
You're contradicting yourself. In post #26 you said "the disconnect must have a neutral to enclosure bond." ggunn asked for where the code says that, and now your saying that it doesn't say it anywhere. So, if I don't have to bring a neutral and bond it, then why do I have to bond it if I do bring it? And if an inspector says to bond it or not bond it, what section of the code says he's right or wrong?

This is why the code needs to be revised. :roll:
Sorry to seem contradictory. Not my intention. It is a result of the options available for a code-compliant installation of this disconnect. I started from the perspective of the neutral having been run to the disconnect. The easiest way to establish "grounding" if not already done through a raceway wiring method is to bond to the neutral.

Ultimately, it is a simple matter. Is the enclosure bonded in a compliant manner? If yes, nothing else needs done.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
Another result of misunderstanding.
Not according to my AHJ. They say that their metering must have the neutral to function correctly whether or not the inverter is using it.

But whether it is the AHJ or you who is misunderstanding the NEC is not the issue. As it stands now, in order to build a compliant system one must navigate several overlapping sections of Code which may or may not be consistent and unambiguous, and it is obviously confusing. The pertinent sections were written at different times, and it seems to me that the ways that they can interrelate were not fully vetted. I have to say that bonding the neutral to the enclosure of a switch and not connecting the EGC/GEC to the enclosure makes no sense to me and I have never seen that done. I think that all enclosures should be directly connected to the grounding system.

A simple, concise statement in the disconnecting means section(s) of 690 and/or 705 directing us to bond the neutral to ground (or not) in the disconnect of a utility interactive inverter interconnected with a service via a line side tap would clear the air.
 

Smart $

Esteemed Member
Location
Ohio
Not according to my AHJ. They say that their metering must have the neutral to function correctly whether or not the inverter is using it.
What I've stated has nothing to do with metering (BTW, since when does the AHJ do metering). And I certainly didn't say don't run a neutral when a neutral is required.

But whether it is the AHJ or you who is misunderstanding the NEC is not the issue. As it stands now, in order to build a compliant system one must navigate several overlapping sections of Code which may or may not be consistent and unambiguous, and it is obviously confusing. The pertinent sections were written at different times, and it seems to me that the ways that they can interrelate were not fully vetted. I have to say that bonding the neutral to the enclosure of a switch and not connecting the EGC/GEC to the enclosure makes no sense to me and I have never seen that done. I think that all enclosures should be directly connected to the grounding system.
I do not see where you get that I said a PV System combo EGC/GEC does not get bonded to the enclosure. Any bonding (or not) that I've mentioned is in relation to the service side of the disconnect.

Code already requires all conductive enclosures to be bonded to the GES.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
What I've stated has nothing to do with metering (BTW, since when does the AHJ do metering). And I certainly didn't say don't run a neutral when a neutral is required.


I do not see where you get that I said a PV System combo EGC/GEC does not get bonded to the enclosure. Any bonding (or not) that I've mentioned is in relation to the service side of the disconnect.

Code already requires all conductive enclosures to be bonded to the GES.

From post #34:
========================================================
You said:
As such, there is no need to bond the GES via a GEC or GEC tap to the grounded conductor (neutral). The GES is already bonded to the service neutral at another point or points, and the GEC(s) are sized based on the ungrounded conductors size(s).

And then you said:
...the disconnect must have neutral-to-enclosure bond (albeit a jumper, which is not to be confused as a Main or System Bonding Jumper) to ground the enclosure and establish an EGC for the PV side.
========================================================

From that I gathered that you don't think it is necessary to bond the neutral to the GES (Grounding Electrode System) in the switch, but that it is necessary to bond the neutral to the enclosure. If the neutral is not tied to ground in the switch but is tied to the enclosure, the only way to do that is if the enclosure is not tied to ground. That's my interpretation of what you said, anyway.
 

Smart $

Esteemed Member
Location
Ohio
... From that I gathered that you don't think it is necessary to bond the neutral to the GES (Grounding Electrode System) in the switch, but that it is necessary to bond the neutral to the enclosure.
The GES GEC(s) can be bonded to the Service grounded conductor (aka neutral) anywhere from Service point to Service Disconnecting Means... period. But FWIW, that includes a neutral in the Service - PV System disconnect if one so desires to bond it there.

If you have multiple Service Disconnecting Means enclosures, the rule for multiple GEC(s) or Common GEC with GEC taps kicks in. So is the Service - PV System a Service Disconnecting Means? I say no. As such, there is no need to run GEC(s) or common GEC tap thereto typically. The one exception I can think of is a solar 'farm' where there is no premises load and thus no Service Disconnecting Means enclosures. And keep in mind, I am talking about where this Service - PV System disconnect is in a separate enclosure from any other Service Equipment.

If the neutral is not tied to ground in the switch but is tied to the enclosure, the only way to do that is if the enclosure is not tied to ground. That's my interpretation of what you said, anyway.
The neutral is tied to the GES in the Service Disconnecting Means enclosure(s). That is, the neutral is already tied to ground elsewhere so bonding it to the Service - PV System enclosure also grounds the enclosure.
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
... The one exception I can think of is a solar 'farm' where there is no premises load and thus no Service Disconnecting Means enclosures. ...

So you really think it is really possible to have a (code compliant) service with no 'service disconnecting means'?

Do you also think that the disconnecting means don't have to comply with any of the rules for disconnecting means in article 230?

In my opinion, when the language of the code leads to interpretations that are baldly ridiculous, that's a sign that the language needs to be changed.
 

Smart $

Esteemed Member
Location
Ohio
So you really think it is really possible to have a (code compliant) service with no 'service disconnecting means'?

Do you also think that the disconnecting means don't have to comply with any of the rules for disconnecting means in article 230?

In my opinion, when the language of the code leads to interpretations that are baldly ridiculous, that's a sign that the language needs to be changed.
For a solar farm, do you really think it is a Service per NEC definition?
 

Smart $

Esteemed Member
Location
Ohio
Smart$

If the definition of a service changed to what I suggested, would you revise all your interpretations?
I have not considered the implications of what you suggest. My interpretations are based what is currently in the Code. If the Code changes, my interpretations reflect the change(s)... appropriately I hope. ;)
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
For a solar farm, do you really think it is a Service per NEC definition?

See 230.2(A)(5).

As for the definition, it needs to be changed. If I call up my utility and ask for a connection for my solar farm, they are going to make me fill out a 'Request for Service.'
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
In my opinion, when the language of the code leads to interpretations that are baldly ridiculous, that's a sign that the language needs to be changed.
That is one statement with which I agree wholeheartedly.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
The GES GEC(s) can be bonded to the Service grounded conductor (aka neutral) anywhere from Service point to Service Disconnecting Means... period. But FWIW, that includes a neutral in the Service - PV System disconnect if one so desires to bond it there.

If you have multiple Service Disconnecting Means enclosures, the rule for multiple GEC(s) or Common GEC with GEC taps kicks in. So is the Service - PV System a Service Disconnecting Means? I say no. As such, there is no need to run GEC(s) or common GEC tap thereto typically. The one exception I can think of is a solar 'farm' where there is no premises load and thus no Service Disconnecting Means enclosures. And keep in mind, I am talking about where this Service - PV System disconnect is in a separate enclosure from any other Service Equipment.


The neutral is tied to the GES in the Service Disconnecting Means enclosure(s). That is, the neutral is already tied to ground elsewhere so bonding it to the Service - PV System enclosure also grounds the enclosure.

I'm sorry but I'm still not getting it; let me try another tack. Let's play "what's wrong with this picture?"

Let's say I have to design a PV system too large to interconnect via a backfed breaker, so I build it with a line side tap. I place a fused disconnect less than 10' from the tap. The system inverter uses the neutral. It is not a solar farm; it is attached to a facility with local loads.

In the disco the CCC's go through the switched contacts and fuses back to the service. The neutral and the GEC/ECG pass through the switch back to the service, and the switch enclosure is bonded to the GEC/ECG. Inside the switch enclosure the neutral is either bonded to the GEC/ECG or not, depending on how the AHJ interprets the Code.

It still seems to me that you are saying that in the above scenario if the neutral were not bonded to the GEC/ECG in the disco, it would be the neutral, not the GEC/EGC, that would be bonded to the switch enclosure. Is that what you are saying?
 

Smart $

Esteemed Member
Location
Ohio
I'm sorry but I'm still not getting it; let me try another tack. Let's play "what's wrong with this picture?"

Let's say I have to design a PV system too large to interconnect via a backfed breaker, so I build it with a line side tap. I place a fused disconnect less than 10' from the tap. The system inverter uses the neutral. It is not a solar farm; it is attached to a facility with local loads.

In the disco the CCC's go through the switched contacts and fuses back to the service. The neutral and the GEC/ECG pass through the switch back to the service, and the switch enclosure is bonded to the GEC/ECG. Inside the switch enclosure the neutral is either bonded to the GEC/ECG or not, depending on how the AHJ interprets the Code.

It still seems to me that you are saying that in the above scenario if the neutral were not bonded to the GEC/ECG in the disco, it would be the neutral, not the GEC/EGC, that would be bonded to the switch enclosure. Is that what you are saying?
No. I'm saying your EGC/GEC combo stops at the disco and is bonded to the enclosure (disco's EGC terminal bus).
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
No. I'm saying your EGC/GEC combo stops at the disco and is bonded to the enclosure (disco's EGC terminal bus).
I see. That's another issue. Does everyone agree with that? I'll hang up and listen...
 

Smart $

Esteemed Member
Location
Ohio
I see. That's another issue. Does everyone agree with that? I'll hang up and listen...
No, it's the same issue... just different than you were thinking. :D

Can't say whether everyone agrees, but I certainly believe NEC requirements do.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
No, it's the same issue... just different than you were thinking. :D

Can't say whether everyone agrees, but I certainly believe NEC requirements do.
They are related, certainly. If the grounding conductor does not run from the switch back to the service, then of course the neutral needs to be bonded to it at the switch, but if the grounding conductor does run back to the service, then whether or not the neutral is bonded in the switch is an issue unto itself.
 
Status
Not open for further replies.
Top