Question regarding 705.12(2)(3)(c) for SOLAR

[shocknocka]

Hello all,

2014 deleted 705.12(D)(7), which covered tying into sub panels(solar). The replacement is the all encompassing 705.12(D)(2)(3), and specifically, the 3 options(a-c).

My question.
In this new code section, do any of these options overrule another? For example, I have a project that complies under option 2 but will not under option 3 due to the sum of the branch circuits.
I don't understand why the sum of the Circuit breaker ratings would be considered as apposed to the the actual accumulative load.
I have a 225/200 MSP with a 125/100 sub. I intend to install 50 amps of backfeed breaker in the sub. It falls under the 120% rule (now 705.12(D)(2)(3)(b)...) in both locations. However, b/c the sum of the branch circuit breaker ratings exceeds the bus's rating, I have no room for a backfeed....at least according to 705.12(D)(2)(3)(c).

thoughts?

 

[Carultch]

Hello all,

2014 deleted 705.12(D)(7), which covered tying into sub panels(solar). The replacement is the all encompassing 705.12(D)(2)(3), and specifically, the 3 options(a-c).

My question.
In this new code section, do any of these options overrule another? For example, I have a project that complies under option 2 but will not under option 3 due to the sum of the branch circuits.
I don't understand why the sum of the Circuit breaker ratings would be considered as apposed to the the actual accumulative load.
I have a 225/200 MSP with a 125/100 sub. I intend to install 50 amps of backfeed breaker in the sub. It falls under the 120% rule (now 705.12(D)(2)(3)(b)...) in both locations. However, b/c the sum of the branch circuit breaker ratings exceeds the bus's rating, I have no room for a backfeed....at least according to 705.12(D)(2)(3)(c).

thoughts?

705.12(D)'s subsections are options for how to comply. Not concurrent requirements. You are permitted to comply with any one of them, but you do not necessarily have to comply with all of them. If you comply with the 120% rule, you don't need to think about summing up all the unrelated branch breakers in the panelboard. Provided that your 120% accounts for all sources.

NEC2014 now allows you to not let rounding errors be a show stopper. Rather than 1.25*system current rounded up to the breaker you are using, you simply use 1.25*system current, while using any breaker that is this value or larger. Then you add this with the main supply breaker, and divide by busbar rating.

 

[shocknocka]

That's the way I interpreted it when I first read it. The lead engineer for the designer we use at my company has a different opinion. I won't say which company, but they service a large number contracts on the east coast, thousands for us a year alone. I generally do not question their opinion.
I would like to get someone's experienced opinion on the why behind the what on (c).

 

[Carultch]

That's the way I interpreted it when I first read it. The lead engineer for the designer we use at my company has a different opinion. I won't say which company, but they service a large number contracts on the east coast, thousands for us a year alone. I generally do not question their opinion.
I would like to get someone's experienced opinion on the why behind the what on (c).

The idea behind part C of this rule, is that all branch circuits operating simultaneously, regardless of if they are sources or loads, will never load the busbar more than the sum of all the branch breakers combined. One of the branch breakers would trip, before the busbar overloads.

The reason this rule is added, is it allows for mixed use panelboards that do not comply with the 120% rule. Likely because they are primarily used for inverter interconnection, with a few auxiliary loads as well. The sources could be 80% of the busbar, while the loads might be 20% of the busbar. The main supply would at least be 80% of the busbar, if not 100%. Difficult to make this work with the "120% rule", because sources dominate the branch circuits instead of loads.

Bill Brooks has a good explanation at 1:06:50 in this video:
https://youtu.be/7vD0r_bVjuo?list=PLRNS1x1jcKbG_gZuhD8Wemyllq0WUldGD&t=4010

 

[shocknocka]

The idea behind part C of this rule, is that all branch circuits operating simultaneously, regardless of if they are sources or loads, will never load the busbar more than the sum of all the branch breakers combined. One of the branch breakers would trip, before the busbar overloads.

The reason this rule is added, is it allows for mixed use panelboards that do not comply with the 120% rule. Likely because they are primarily used for inverter interconnection, with a few auxiliary loads as well. The sources could be 80% of the busbar, while the loads might be 20% of the busbar. The main supply would at least be 80% of the busbar, if not 100%. Difficult to make this work with the "120% rule", because sources dominate the branch circuits instead of loads.

Bill Brooks has a good explanation at 1:06:50 in this video:
https://youtu.be/7vD0r_bVjuo?list=PLRNS1x1jcKbG_gZuhD8Wemyllq0WUldGD&t=4010

Carultch,
Thank you very much, this was helpful. I'm going to assume this is the general consensus since there are almost 150 views at this point and no one has chimed in with a differing opinion. I personally stand behind the logic of this interpretation. Lets hope the inspector will too.

 

[ggunn]

Carultch,
Thank you very much, this was helpful. I'm going to assume this is the general consensus since there are almost 150 views at this point and no one has chimed in with a differing opinion. I personally stand behind the logic of this interpretation. Lets hope the inspector will too.

I had this discussion with an inspector, too. He was wanting a system to comply with a, b, and c instead of a, b, or c. I don't win them all but I won that one.

Among other things, option c legitimizes the AC combiner panels we have been using for years.

 

[jaggedben]

The code couldn't be more clear.

705.12(2)(3)...
Busbars. One of the methods that follows shall be used to determine the ratings of busbars in panelboards.

Emphasis added.