120% Rule with no Disconnect Breaker

[GoldDigger]

I don't think so. See 705.12(A). If it's on the supply side of the service disconnecting means, it's a supply side connection. Otherwise, it's a load side connection. The presence or absence of loads is inconsequential.

If there are loads (other than incidental PV related loads such as inverter idling) on the same side of the service breaker as the PV backfeed, I would call that a service disconnect. To me the PV disconnect must disconnect only the PV. Otherwise you would have the somewhat strange situation where the main breaker in the service panel has to be considered as the PV disconnect and therefore not required to be service rated. :happysad:

 

[ggunn]

If there are loads (other than incidental PV related loads such as inverter idling) on the same side of the service breaker as the PV backfeed, I would call that a service disconnect. To me the PV disconnect must disconnect only the PV. Otherwise you would have the somewhat strange situation where the main breaker in the service panel has to be considered as the PV disconnect and therefore not required to be service rated. :happysad:

That notwithstanding, the distinction between a load side and a supply side connection is clear. If it's between the service disconnect(s) and the meter it's a supply side connection, and if it's on the other side (the load side) of the disconnecting means it's a load side connection. The placement of loads doesn't have any bearing on that. I suppose that there could be a question if the MDP were MLO with an unfused disco and no other OCPD ahead of it, but I have never seen that (not to say it can't happen).

I have encountered MLO MDP's with no disconnecting means ahead of them; any PV landed in that MDP is of course supply side connected, but the AHJ's would not allow a backfed breaker. I had to have a bladed fused disco on the wall next to the MDP (less than 10 feet away as the conductor flies) with conductors landed on the MDP bus.

 

[jaggedben]

I've always believed that when connections are made in series, each connection is considered in it's own context. Thus, in the OPs diagram where you have two 90A service disconnects, one of which has solar connected to a downstream sub, the connection in the sub is a load side connection and the connection in the main is a supply side connection. Whether there are loads on the same service disconnect or not, it's still a supply side connection. As long as you have less than six service disconnects and are not claiming that the connection(s) with PV is exempt from service disconnect rules, it doesn't matter.

Perhaps it would help if the code actually contained some language about connections in series.

 

[PWDickerson]

The PV system requires a dedicated OCPD, so there can't be any premises loads on the load side of the PV system OCPD.

 

[jaggedben]

The PV system requires a dedicated OCPD, so there can't be any premises loads on the load side of the PV system OCPD.

True, but that's rather besides the point. If the OCPD serves both loads and PV, then that isn't the PV OCPD. But it still has to follow appropriate rules if it serves as part of the inverter output circuit.

 

[ggunn]

I've always believed that when connections are made in series, each connection is considered in it's own context. Thus, in the OPs diagram where you have two 90A service disconnects, one of which has solar connected to a downstream sub, the connection in the sub is a load side connection and the connection in the main is a supply side connection. Whether there are loads on the same service disconnect or not, it's still a supply side connection. As long as you have less than six service disconnects and are not claiming that the connection(s) with PV is exempt from service disconnect rules, it doesn't matter.

Perhaps it would help if the code actually contained some language about connections in series.

I'm not sure what you are driving at. The language "supply side connection" means something specific, and it's a connection on the supply side of the service disconnect(s). If the connection is on the supply side of a subpanel and the load side of the MDP it's still a load side connection.

 

[jaggedben]

I'm not sure what you are driving at. The language "supply side connection" means something specific, and it's a connection on the supply side of the service disconnect(s). If the connection is on the supply side of a subpanel and the load side of the MDP it's still a load side connection.

Let's consider the OP's example...

The Main only had two breakers. Each feeding existing sub panels...

View attachment 16596

...and now let's suppose he wanted to backfeed more than 90A, without upgrading the service. He can change the rating of the new subpanel he's installing to meet the load side rules, but does anything else in the code impose a limit on how much he can backfeed? Answer: yes, the rating of the service. The 90A breaker he'd be replacing has to comply with 705.12(A), in my opinion.

That's what I'm driving at. You have two connections in series. You shouldn't be able to ignore what's happening at the service just because you have a load side connection downstream.

 

[ggunn]

Let's consider the OP's example...



...and now let's suppose he wanted to backfeed more than 90A, without upgrading the service. He can change the rating of the new subpanel he's installing to meet the load side rules, but does anything else in the code impose a limit on how much he can backfeed? Answer: yes, the rating of the service. The 90A breaker he'd be replacing has to comply with 705.12(A), in my opinion.

That's what I'm driving at. You have two connections in series. You shouldn't be able to ignore what's happening at the service just because you have a load side connection downstream.

I agree, but it's a load side connection because it's on the load side of the 90A "sub panel breaker", which is a service disconnect. If he were to backfeed more than 90A with no active loads, that breaker would trip.

With series connections 125% of the maximum rated current of the inverter counts in every one of them. I do not think, however, that 705.12(A) applies because it specifically applies to supply side connections, which this is not. 705.12(D) is the governing article.

 

[ggunn]

I agree, but it's a load side connection because it's on the load side of the 90A "sub panel breaker", which is a service disconnect. If he were to backfeed more than 90A with no active loads, that breaker would trip.

With series connections 125% of the maximum rated current of the inverter counts in every one of them. I do not think, however, that 705.12(A) applies because it specifically applies to supply side connections, which this is not. 705.12(D) is the governing article.

Upon further review I can see how you could look at it either way; the subpanel could be a AC combiner (though it's a bit of a stretch since it doesn't combine anything as drawn), but if you go with that, at least in the jurisdictions I work in, you'd have to have a fused bladed disco instead of a breaker to connect it to the service. No AHJ I work with will allow a backfed PV breaker in an MLO MDP.

 

[jaggedben]

Upon further review I can see how you could look at it either way; the subpanel could be a AC combiner (though it's a bit of a stretch since it doesn't combine anything as drawn), but if you go with that, at least in the jurisdictions I work in, you'd have to have a fused bladed disco instead of a breaker to connect it to the service. No AHJ I work with will allow a backfed PV breaker in an MLO MDP.

We all know you work with these weird AHJs that make up their own rules. :lol: FWIW, I've never run into an AHJ that prohibited that. A couple I had to explain it to a few times, but eventually they caught on.

 

[SolarAlternativesDesign]

214MTB meter main connection

214MTB meter main connection

I do this all the time. As you noted, for a 60A breaker, you'd need a 200A sub. You could go to 50A in your example. But in my experience that's always been good enough. There's not too many 100A service homes that need an 11kW solar system.

I think this 100A panel (B-Line 214MTB) we're going to connect through is a similar case.

We plan to swap out the 60A breaker for a 100A breaker and feed 62.5A of PV through a 200A subpanel that will have a 100A main breaker, so the subpanel busbar is adequately sized. There isn't a busbar in the service panel, so no calculation needed there. There will be the feeder from the 100A breaker in the service panel to the subpanel, so we will size its ampacity to be not less than the subpanel main breaker, that is, 100A (2014 NEC 705.12(D)(2)(1)(b)).

Does this work?

 

[jaggedben]

I think this 100A panel (B-Line 214MTB) we're going to connect through is a similar case.
...

Does this work?

Yes. Although I don't see why you need to upsize the breaker to 100A if there are no loads being added and this is not a dwelling (and the bypass suggests to me it's probably a commercial space?). You could size your feeder and breakers to 80A if you wanted, assuming your 62.5A is before the 125% factor is added in.

 

[SolarAlternativesDesign]

Yes. Although I don't see why you need to upsize the breaker to 100A if there are no loads being added and this is not a dwelling (and the bypass suggests to me it's probably a commercial space?). You could size your feeder and breakers to 80A if you wanted, assuming your 62.5A is before the 125% factor is added in.

Right - The subpanel breaker could be 80A. The subpanel could have a few additional loads in the future, but probably not over 60A. It currently serves a well (which dried up) and lights.

Thanks

 

[jaggedben]

Right - The subpanel breaker could be 80A. The subpanel could have a few additional loads in the future, but probably not over 60A. It currently serves a well (which dried up) and lights.

Thanks

One thing... Just make sure that the service conductors are rated 80A, or at least that the utility considers it rated 80A.