More PV transformer / AHJ issues

How does everyone here handle their PV transformer disputes with AHJ's or are we the only one having issues? It seems at least half the time we have a transformer in a design we get an inspector that wants to have an opinion about it. Usually saying we are conflicting ourselves with our specifications of the primary winding towards the utility. I feel like the NEC doesn't spend enough time clearly depicting how to design a transformer into a PV system and it creates problems like this because those who do not have a more in depth knowledge of transformer anatomy and the reasoning behind design considerations do not understand why engineer's make certain decisions regarding system design.

The latest issue was that the AHJ wanted us to remove our neutral bond on the PV side of the (480Y-208Y) transformer (within the first disconnecting means and move it to to the utility side of the transformer because "...that is the secondary of the transformer..." I told him that made no sense and that we'd then have a floating neutral on one side and a neutral bonded in two places on the other side and explained the problems that created. He responded that hes not an engineer so he doesn't have to understand that level of detail and the code book doesn't require an engineer to be able to follow it. If that were the case then why bother getting an engineers stamp on drawings instead just have an NEC compliant guaranteed stamp that electricians can carry around and say they followed the code? To clarify, that is by no means a knock on electricians, its more a reference that the electrician's license requires a working knowledge of the NEC and secondarily its about the fact that we each have our own role in the equation, they keep us honest when we don't think about real life problems, and even when we make mistakes, and we work out the details that require our background.

That digression aside the solution of the inspector was to continue to keep a second bond on the utility side of the transformer but to bond the neutral at the first disconnect from the inverter. That makes no sense at all and isn't safe but I can't find anywhere in the NEC that tells him that is wrong, just engineering knowledge of the system, which he negates, and other code references such as IEEE, which he also negates because that is an "engineer's code and hes not an engineer" which again, seems quite unjustified; how would you handle this, or is there a reference I am missing with bonding that says something clear enough in this situation (I've already tried the argument that design should consider looking at primary in both directions when choosing OCPD, but he again didn't understand that, says it only applies to OCPD - which isn't wrong but hes missing that its the point of inferring that a transformer isn't straight forward, and he says in that case it shouldn't be bonded on either side then 🤦‍♂️ . The neutrals are not physically connected so we have to call it separately derived whether he likes it or not, the problem is he is saying that the PV is the generation source so it is 'deriving' the neutral on the secondary side of the transformer which is the utility side without considering that the PV is actually a delta generator with a neutral reference (and the capability of reactive power support which can balance power to a neutral now on smart-grid inverters); PV is also not a voltage source but a current source but the voltage source (which distinguishes the potential difference between conductors) is in fact the utility.

This is more than a specific problem I am looking to resolve, more looking to compile an arsenal of info regarding thoughts on PV transformers, and the problems you face with confusion, the solutions you came up with - more of a lessons learned post and see if there are enough repeated issues out there that we should write a letter to the CMP for Art 690 for clarification in 2023.

What other problems do you folks face with PV transformer confusion? How do you handle them?
 

jaggedben

Senior Member
Preliminary to my substative suggestions...
- I have no direct experience negotiating with AHJs on this, nor do I typically deal with transformers in my PV systems.
- But I understand everything you're saying and why he is totally wrong
- Not sure how helpful I can be about a seemingly unreasonable person, but here goes...

To my knowledge, the code does not define primary and secondary. Ask him, since he's so up on the letter of the code being understandable to a non-engineer, which code section he's looking at that tells him to ignore how you're defining it. When he says the utility side is the secondary, just say "Well, I don't think so. Where in the code does it say that?" And then repeat broken-record style that the primary is always the side towards the utility, i.e. the side that energizes it first.

You also might appeal 110.3(B) with regard to the use of the transformer, which the manufacturer might have something to say about. Check with them first though, obviously.
 
It seems like the definition of SDS would help.....

Also, regarding the bond on the utility side, there are code articles that specifically prohibit connecting a grounded conductor to ground after the service equipment (250.142(B) and 250.24(A)(5)

Regarding the bond on the PV side, ask him how a fault will clear with no grounded conductor.
 
"... ask him how a fault will clear with no grounded conductor." My guess is that his response would be I don't know, I'm not an engineer...haha.

Already quoted all pertinent sections of Art. 250 to him. If he was smart I suppose he could at minimum confuse the situation and make his case by saying he is considering it not separately derived (because article 250 itself really described when a permanently connected generator switches a neutral vs. a transformer with conductors that are inductively connected vs. physically connected) which would at least justify his case to bond the neutral at the generator itself (at the moment working on something that uses 2014 so that would be 250.35 (B), not sure if that changed in 2017 or 2020 off the top of my head).

Glad to see you guys confirm that I'm not going crazy and missing something here.

On the note of saying the primary is always towards the utility side I don't know I would want to go that route specifically, I think you (jaggedben) nailed it in your first statement where I was trying to validate I wasn't missing something on. The code doesn't define a primary and secondary side of a transformer so we shouldn't either - there is industry terminology that incorrectly typically designates the high side as the primary and the low side as the secondary - we should just consider the effects of the sections of the code that require us to do certain things when we see a transformer and apply those requirements as the system functionally exists. I think this was what I was getting at in the post is that I wish that the NEC would clarify this and say directly that the primary side and secondary side should be considered from a functional perspective for each design consideration (e.g. OCPD, separately considered from bonding, ect.) unless the manufacturer states otherwise that it must functionally be utilized based on the location of the windings. Another issue that comes to head involved bringing the state in and as mentioned before they came up with some real strange solution that I needed to have both primary and secondary protection on both sides. It was a costly random decision that affected position of meters, added extra protection and made a physical mess out of everything. Sometimes it is nice when instead of expecting the inspector to come up with the code reference you have your own to say ah you see, I'm doing this per 690.X. If my area is the only one though with inspectors with a bone to pick on transformers then I suppose I'll just have to continue to expect the problem when we do not have an alternative to putting in transformers.
 

tortuga

Senior Member
Location
(44.057116, -123.103394)
Occupation
field supervisor
In reply to this;
To my knowledge, the code does not define primary and secondary.
The transformer manufacturer is required to define a primary and secondary by voltage on the nameplate. 450.11(B) puts limits on using the secondary as the supply side. So in a usual round-about way it does.

On the note of saying the primary is always towards the utility side I don't know I would want to go that route specifically, I think you (jaggedben) nailed it in your first statement where I was trying to validate I wasn't missing something on. The code doesn't define a primary and secondary side of a transformer
Ill add the text:
2017 NEC said:
450.11 Marking.
(A) General. Each transformer shall be provided with a name-
plate giving the following information:
(4)Primary and secondary voltage
and
2017 NEC said:
450.11(B) Source Marking. A transformer shall be permitted to be
supplied at the marked secondary voltage, provided that the
installation is in accordance with the manufacturer’s instructions.
and this install seems fine.

Also make note of 705.65:
2017 NEC 705.65 said:
(B) Power Transformers. Overcurrent protection for a transformer with a source(s) on each side
shall be provided in accordance with 450.3 by considering first one side of the transformer,
then the other side of the transformer, as the primary.

Exception: A power transformer with a current rating on the
side connected toward the interactive inverter output that is not
less than the rated continuous output current of the inverter shall
be permitted without overcurrent protection from that source.
Cheers
 

CS2

Member
Location
NY
Occupation
Engineer
There is definitely not a straight forward answer to this. You for sure need a N-G bond on the PV side. Now, on the utility side, where is your interconnection point (line-side or load side)? If load side I would have a N-G bond if line side do you consider PV a service and always establish a N-G in the service disco. Clearing up these questions with the AHJ might help first. Do you have a SLD?

Also the Ho is always the higher voltage and consider the "primary" voltage but I have seen this argued.
 
There is definitely not a straight forward answer to this. You for sure need a N-G bond on the PV side. Now, on the utility side, where is your interconnection point (line-side or load side)? If load side I would have a N-G bond if line side do you consider PV a service and always establish a N-G in the service disco. Clearing up these questions with the AHJ might help first. Do you have a SLD?

Also the Ho is always the higher voltage and consider the "primary" voltage but I have seen this argued.
I disagree, it is very straightforward. I dont see what line vs load side connection and if line side whether the disco is treated like a service disconnect or not has to do with the transformer utility side connections.
 

CS2

Member
Location
NY
Occupation
Engineer
I disagree, it is very straightforward. I dont see what line vs load side connection and if line side whether the disco is treated like a service disconnect or not has to do with the transformer utility side connections.
First I'm making some assumptions. I'm considering this is a LV transformer (something like 480 to 208) and also the primary and secondary windings are Wye-G. If this is a loadside connection, there is already an establish N-G bond at the main service/building disconnect. I can add as many means of grounding (ground rods, ufer ground, etc.) on the utility side of the PV system but I'm not establishing a N-G anywhere on the utility PV side because it would result in a parallel path to ground. Do you agree?
 

jaggedben

Senior Member
First I'm making some assumptions. I'm considering this is a LV transformer (something like 480 to 208) and also the primary and secondary windings are Wye-G. If this is a loadside connection, there is already an establish N-G bond at the main service/building disconnect. I can add as many means of grounding (ground rods, ufer ground, etc.) on the utility side of the PV system but I'm not establishing a N-G anywhere on the utility PV side because it would result in a parallel path to ground. Do you agree?
I think all electrofelon is saying is that you treat it like any other transformer you might install, ignoring the fact that powerflow may reverse direction when PV starts outputting. That's the sense it which it is straightforward.

Regardless of whether the connection is supply side or load side, the service should already be grounded. The controversy surrounding whether or not to add an additional N-G bond for a supply side connection has nothing to do with whether or not there is a transformer downstream. You are going to bring the same wires to the transformer regardless.
 
I think all electrofelon is saying is that you treat it like any other transformer you might install, ignoring the fact that powerflow may reverse direction when PV starts outputting. That's the sense it which it is straightforward.

Regardless of whether the connection is supply side or load side, the service should already be grounded. The controversy surrounding whether or not to add an additional N-G bond for a supply side connection has nothing to do with whether or not there is a transformer downstream. You are going to bring the same wires to the transformer regardless.
Correct. I did miss that the OP was talking about a wye wye, and there is some common confusion about when that is an SDS and whether to run a utility side neutral, but it seems pretty clear to me you can't make a N-G bond if its not and SDS (unless it's MV and an MGN system). There were some recent threads on wye wye, but I don't have the link handy.
 

tortuga

Senior Member
Location
(44.057116, -123.103394)
Occupation
field supervisor
Correct. I did miss that the OP was talking about a wye wye, and there is some common confusion about when that is an SDS and whether to run a utility side neutral, but it seems pretty clear to me you can't make a N-G bond if its not and SDS (unless it's MV and an MGN system). There were some recent threads on wye wye, but I don't have the link handy.
Here are some of the Y-Y threads

I missed that Y-Y part also :
The AHJ wanted us to remove our neutral bond on the PV side of the (480Y-208Y) transformer
. \..snip..\ The neutrals are not physically connected so we have to call it separately derived
 
Top