Transformer OCPD

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Consulting Electrical Engineer - Photovoltaic Systems
There is language in the NEC somewhere that talks about inverters through a transformer that exempts the circuit from OCPD or maybe a disconnect where the source is current limited to less than the transformer rating, but I haven't looked at it a while and I can't find it. Anybody?
 
There is language in the NEC somewhere that talks about inverters through a transformer that exempts the circuit from OCPD or maybe a disconnect where the source is current limited to less than the transformer rating, but I haven't looked at it a while and I can't find it. Anybody?
Never mind; found it.

The exception to 690.9(D)
 
Never mind; found it.

The exception to 690.9(D)

How is this meant to work with 240.21(C), assuming the transformer type would require an OCPD? Also I notice that the 690.9(D) exception only indicates inverter (singular), seemingly implying that this only applies for a single inverter, and not multiple inverters sharing a transformer.

If 240.21(C) requires an OCPD for the secondary conductors in most cases, it seems like a rare situation where you could take credit for this rule. Unless this exception is also an exception from 240.21(C).
 
No indication if this is for a particular version of the NEC so I'll be pulling from the 2023.
How is this meant to work with 240.21(C), assuming the transformer type would require an OCPD? Also I notice that the 690.9(D) exception only indicates inverter (singular), seemingly implying that this only applies for a single inverter, and not multiple inverters sharing a transformer.
The exception to 690.9(D) is interesting. Does it only apply to a single inverter? What the exception is allowing is that if the XFMR is rated for the current that can be provided by the inverter then there is no protection required for the XFMR when considering only the inverter operation. This applies to a single 500kW inverter but why not five 100kW inverters combined and connected to the XFMR? From the perspective of the XFMR and the XFMR protection, they look the same and probably have a very similar current. So I would allow this exception for multiple inverters that are combined.
If 240.21(C) requires an OCPD for the secondary conductors in most cases, it seems like a rare situation where you could take credit for this rule. Unless this exception is also an exception from 240.21(C).
NEC 690.9(D) points us to 705.30(F), and 705.30(F) says:
The following apply to the installation of transformers:
(1) For the purpose of overcurrent protection, the primary side of transformers with sources on each side shall be the side connected to the largest source of available fault current.
(2) Transformer secondary conductors shall be protected in accordance with 240.21(C).
This was added to clear up the whole primary/secondary confusion in systems with multiple sources. The primary side of a PV system XFMR will almost always be the side facing the utility as the largest source. The secondary is then facing the inverters. Then 240.21(C) is applied to the secondary conductor protection on the inverter side.
To sum it up; the XFMR conductors need protection from the utility and the XFMR is rated for the full output of any inverters feeding it.
 
No indication if this is for a particular version of the NEC so I'll be pulling from the 2023.

The exception to 690.9(D) is interesting. Does it only apply to a single inverter? What the exception is allowing is that if the XFMR is rated for the current that can be provided by the inverter then there is no protection required for the XFMR when considering only the inverter operation. This applies to a single 500kW inverter but why not five 100kW inverters combined and connected to the XFMR? From the perspective of the XFMR and the XFMR protection, they look the same and probably have a very similar current. So I would allow this exception for multiple inverters that are combined.

NEC 690.9(D) points us to 705.30(F), and 705.30(F) says:

This was added to clear up the whole primary/secondary confusion in systems with multiple sources. The primary side of a PV system XFMR will almost always be the side facing the utility as the largest source. The secondary is then facing the inverters. Then 240.21(C) is applied to the secondary conductor protection on the inverter side.
To sum it up; the XFMR conductors need protection from the utility and the XFMR is rated for the full output of any inverters feeding it.

That's what I thought. It's a rare situation where the exception in 690.9(D) isn't also governed by 240.21(C) to require protection anyway.

Examples I can think of, where you can take credit for this exception:
1. An inverter connected to a transformer with a qualifying topology in the first part of 240.21(C), that allows the primary OCPD to protect the secondary conductors. I find this to be uncommon.
2. An off-grid system with nothing but inverters forming the grid, where the inverter side would be considered the primary, and the load side would be considered the secondary.
 
This was added to clear up the whole primary/secondary confusion in systems with multiple sources. The primary side of a PV system XFMR will almost always be the side facing the utility as the largest source. The secondary is then facing the inverters.
That was a much needed clarification, since prior to that code change, we'd split hairs on deciding how to name primary and secondary.
 
How is this meant to work with 240.21(C), assuming the transformer type would require an OCPD? Also I notice that the 690.9(D) exception only indicates inverter (singular), seemingly implying that this only applies for a single inverter, and not multiple inverters sharing a transformer.
Why would that be? The point is that there is a limit to the current that can be supplied by the inverter(s), and there is no difference between a single inverter and a bank of them feeding a combiner panel.
 
Why would that be? The point is that there is a limit to the current that can be supplied by the inverter(s), and there is no difference between a single inverter and a bank of them feeding a combiner panel.
The wording doesn't reflect the flexibility to apply it for multiple inverters, even if the physical basis does. I agree there is no difference from the physical basis for this rule, between 5x 100A inverters and 1x 500A inverter.
 
The wording doesn't reflect the flexibility to apply it for multiple inverters, even if the physical basis does. I agree there is no difference from the physical basis for this rule, between 5x 100A inverters and 1x 500A inverter.
I'd argue that a strict interpretation of the exception's language doesn't preclude application to multiple inverters, and actually sort of provides a loophole. In that, if I have multiple inverters but each one has a continuous output less than the transformer rating, then for each inverter I'm compliant. Although it's written as if there would only be one inverter it doesn't actually say that's required.

I'm also baffled that the exception wasn't moved to 705 with the rest.
 
The wording doesn't reflect the flexibility to apply it for multiple inverters, even if the physical basis does. I agree there is no difference from the physical basis for this rule, between 5x 100A inverters and 1x 500A inverter.
For this rule I just assume that there is no difference between a single inverter and multiple inverters feeding an agg panel as long as the total current is low enough. I have never been challenged on it.
 
For this rule I just assume that there is no difference between a single inverter and multiple inverters feeding an agg panel as long as the total current is low enough. I have never been challenged on it.
Kind of like how a bank of transformers, not paralleled, are considered a single source?
 
Kind of like how a bank of transformers, not paralleled, are considered a single source?
I don't know about that, but this rule takes advantage of the fact that a PV inverter is a current limited source, and a set of inverter outputs combined in a panel is just as current limited as is a single inverter.
 
So I’ve been brushing up on transformer protection rules for PV installations. I can’t help but think that exceptions in the nec to codes typically mean that you don’t have to follow the main wording that the exception is being applied to. In this case, how is it not saying that we don’t have to refer to 705.30(F) and therefore 240.21(C)? Otherwise, why would this exception even exist? If we have secondary ocpd per 241.21(C) then we have protection from the inverter.
 
So I’ve been brushing up on transformer protection rules for PV installations. I can’t help but think that exceptions in the nec to codes typically mean that you don’t have to follow the main wording that the exception is being applied to. In this case, how is it not saying that we don’t have to refer to 705.30(F) and therefore 240.21(C)? Otherwise, why would this exception even exist? If we have secondary ocpd per 241.21(C) then we have protection from the inverter.
You don't need protection from the inverter(s) if the system is designed correctly, i.e., if all the wiring, switches, transformers, etc. are designed to be able to withstand the maximum current output of the inverter(s).
 
690.9(D) is practically a copy-editing error at this point. But I don't see how its exception could apply to 240.21(C). 240.21(C) applies by default to all transformer installations; Chapters 5-7 could modify that in this case, but they don't.
 
Top