Another 120% busbar question

[Jeff S]

I've seen considerable discussion about the 120% rule on this and other forums. 705.12(D) (2) from NEC 2011 is frequently noted. The 120% is stated there for load side of service. It would seem as though 705.12 (A) Supply Side connection isn't included with the 120% rule but in much of the discussion, there's been no differentiation between the two points of connection. Am I missing something here because it seems as though many opinions are using the 120% rule for line side connections. I believe John Wiles wrote in an article that he considers a new service (Line Side connection) to be a load side connection if you add an inverter or solar combiner panel to it. Others seem to disagree.
Any insight into how this is interpreted would be appreciated.


Thanks,
Jeff S

 

[don_resqcapt19]

My understanding that the amount of solar connected to the line side of the service can equal the rating of the service. Everything is designed to carry the rated amps of the service and the addition of the solar system would not change this.

 

[ron]

Line side connections are a pain. You need to get the utility involved and the AHJ doesn't have as many prescriptive requirements, so they tend to make stuff up (rules) on the line side.

 

[Sierrasparky]

Most POCO will not allow a line side , however if you have a meter/main (200a)then the 200afeeder that carries the entire load to a main breaker panel (200a) then if you place the PV tie inbetween I think you might be able to still consider this a line side tap. My reasoning is that the feeder to the sub will never see greater amps than the main breaker it has. The meter main will never see greater than 200A also.

 

[Jeff S]

Haven't had problems.

Haven't had problems.

The local utilities haven't had any issues with line side connections, it's just a pain to schedule a time to shut down the system. We are considering using some Ilsco insulated taps that would allow us to tap the conductors in the CT or main enclosures. They are made to tap live and would eliminate the scheduling issues. Obviously you want to wear protective gear if you attempt this.
For sizing, I just make sure that my fused disconnect fuses and wire are large enough for the combined solar combiner box inverter loads X 1.25. I agree with others that it makes no sense to size this for what someone might add in the future (like tap the conductors). The wrong person could overload any system.
Thanks for your input.

 

[jaggedben]

To answer the original question, yes, there is a difference in the rules for line side and load side connections. The 120% rule should not apply if the solar disconnect is also part of the service disconnecting means (one of up to six handles in a location, per Article 230).

Regarding the Wiles opinion: If you add a solar combiner with a main breaker, then any connection beyond that main breaker is a load side connection. Unless a solar combiner is configured without a main breaker - that is, with each solar breaker as part of the service disconnecting means, limit six handles - then you are unequivocally doing a load side connection. So if the AHJ insists on enforcing the 120% rule you pretty much have to listen.

 

[jaggedben]

Most POCO will not allow a line side , however if you have a meter/main (200a)then the 200afeeder that carries the entire load to a main breaker panel (200a) then if you place the PV tie inbetween I think you might be able to still consider this a line side tap.

It is not line side because there is a circuit breaker in the meter/main. That breaker is the service disconnecting means, and anything downstream of it (feeding from the utility) is load side. Tapping the 200A feeder in your example would follow the tap rules in 240.21(B). The tapped conductor is also subject to the 120% rule, and the solar backfeed couldn't could exceed 40amps. (This last bit, in my opinion, is an overreach in the code. But it is what the code currently requires.)

Your POCO really should be allowing line side taps; the solar is required to have disconnecting means just like any other part of the service. But if they allow your example, that's not an exception to their policy.

 

[jaggedben]

The local utilities haven't had any issues with line side connections, it's just a pain to schedule a time to shut down the system. We are considering using some Ilsco insulated taps that would allow us to tap the conductors in the CT or main enclosures. They are made to tap live and would eliminate the scheduling issues. Obviously you want to wear protective gear if you attempt this.
For sizing, I just make sure that my fused disconnect fuses and wire are large enough for the combined solar combiner box inverter loads X 1.25. I agree with others that it makes no sense to size this for what someone might add in the future (like tap the conductors). The wrong person could overload any system.
Thanks for your input.

Safety issues aside, I think the problem with this is that if you make a mistake you will fault back to the transformer and blow out power to the whole neighborhood you are working in. How many shutdowns are worth avoiding before that happens? I have heard enough horror stories that I would never try a live tap with those Ilsco taps. The only type of live supply-side connection I'd attempt, with proper PPE of course, would be adding a breaker in a MLO panel.

 

[Marvin_Hamon]

The supply side interconnection option is surprisingly unqualified, the NEC says we can do it but does not provide any guidance about how to do it or what NEC requirements should be applied. That leaves it up to interpretation, and we all love the grey area now don't we?

What I have seen, and what I apply myself, is that the circuit from the supply side interconnection to the first OCPD / disconnect is installed according to 230 for service entrances. From the first OCPD / disconnect to the inverter the circuit is installed according to the load side interconnection requirements. There is nothing in the NEC that explicitly requires this but it has become a best practice and I have yet to find an AHJ who does not accept this design.

So does the 120% rule apply to the supply side connection? It depends on how you and the AHJ choose to go about the supply side interconnection. But if you choose not to go by the load side interconnection standards what do you use? There are really no other code sections that seem to provide any guidance.

 

[Jeff S]

You have a good point about the safety. We've only discussed this and haven't ever done it. I suspect that we probably would have the utility shut down the power if we tap the conductors. Up to this point we've always been able to add additional lugs to make a line side connection but sometimes a meter base isn't well suited to that. Also the uitility and the inspector need to approve the different lugs although so far it's never been a problem.

After reading a number of forum comments regarding the 120% load side discussions about AC combiner panels I have these comments:
Apparently it's fairly common for installers to view an AC combiner panel (with perhaps microinverters) in the same light as a DC combiner even though it doesn't meet the rule as written. By that I mean that the main OCPD is sized large enough for the combined loads (or sources) and just that OCPD (not added to the individual OCPDs) is used for 120% sizing of conductors and busbars. If there were readily available AC combiners that didn't accept additional ordinary breakers we probably wouldn't be having all of these discussions. It seems to me that the code panel's rejection of the "Solar PV circuits only- no other circuits to be added" is primarily based on their concern that it's too easy to add to the standard load centers that are used for combiners. Where it gets extreme is where they've rejected this even when there are no more available slots to expand because of concerns that the lines could be tapped.
No inspector has demanded that we strictly follow the rule because as many have noted, it's not likely to be dangerous if you size the busbar and conductors for the largest OCPD that feeds from either end and they're probably not aware of the rule.
As noted, the safe thing would be to get permission from the AHJ but we deal with numerous jurisdictions and would need to do that for every project. It's also costly to upsize everything just in case an inspector wants to strictly adhere to the rule.
Bottom line, I wish that the code panel would come up with a suitable exception to the rule that would be safe and cost effective.

 

[Jeff S]

The supply side interconnection option is surprisingly unqualified, the NEC says we can do it but does not provide any guidance about how to do it or what NEC requirements should be applied. That leaves it up to interpretation, and we all love the grey area now don't we?

What I have seen, and what I apply myself, is that the circuit from the supply side interconnection to the first OCPD / disconnect is installed according to 230 for service entrances. From the first OCPD / disconnect to the inverter the circuit is installed according to the load side interconnection requirements. There is nothing in the NEC that explicitly requires this but it has become a best practice and I have yet to find an AHJ who does not accept this design.

So does the 120% rule apply to the supply side connection? It depends on how you and the AHJ choose to go about the supply side interconnection. But if you choose not to go by the load side interconnection standards what do you use? There are really no other code sections that seem to provide any guidance.

Thanks for your comments. It may be easier for me to explain this with an example instead of just a description. For a Line-Side connection for (4) 15A circuits in an AC solar combiner panel stating no other circuits and with (41) Enphase inverters: For this example, the four 15A circuit loads have a total of (41) M210 240VAC inverters so 41 X 0.88A X 1.25=45.1A so the fuses in my service disconnect would be 50A. I typically would feed this with the min. #6 conductor for services. The sub-panel typically will be 100A or 125A depending on brand which covers the 120% for the busbars. The real issue is whether the conductors from the fused disconnect to the MLO combiner panel can be #6 (for the actual load and based on only the first OCPD) or need to be sized based on the breakers: (4 X 15A)+50A fuses=110A 110A/1.2=91.7A. If I understand the 120% rule, the line side conductors would now need to be at least #3 (75deg. terminal rating) depending on C.O.U. Is that how you see it?
Thanks,
Jeff S.

 

[jaggedben]

It seems to me that the code panel's rejection of the "Solar PV circuits only- no other circuits to be added" is primarily based on their concern that it's too easy to add to the standard load centers that are used for combiners. Where it gets extreme is where they've rejected this even when there are no more available slots to expand because of concerns that the lines could be tapped.

I think some of the concern has to do with the fact that load centers are not really tested for these types of applications (i.e. backfeeding) by the listing agencies. There is concern that breakers would not operate properly if the real-life thermal loading did not resemble the testing scenario.

Now where I agree it gets extreme is that the same rules are applied to conductors.

If there were readily available AC combiners that didn't accept additional ordinary breakers we probably wouldn't be having all of these discussions.

Now that is an interesting notion. Another interesting notion is that load centers should be tested with a backfeeding scenario during the listing process.

If I understand the 120% rule, the line side conductors would now need to be at least #3 (75deg. terminal rating) depending on C.O.U. Is that how you see it?

That is how I see it, because that's what the code says. Everything after the 50A disconnect is load side. But I believe (in part because Marvin said it in another forum) that the rules for conductors should be different. It depends whether they are tapped for the backfeed (i.e. middle fed) or not, but if they are end fed (as in your case) I believe they should be sized for the largest individual OCPD that supplies them.

 

[Jeff S]

jaggedben, I appreciate your response. I'm now shifting my discussion from line side to load side connections.

I'm not trying to beat this subject to death (I know there are other threads about it also) but especially with the AC modules just coming to market it's important to come up with a solution for "AC combiner panels". Unfortunately this can't wait until the 2014 NEC or later. In our area the utilities all require an external AC disconnect. Since no one wants to put up multiple disconnects (6 max.), we can't take each inverter to a breaker in an existing load center. If we could, the rules would be straightforward. Therefore anytime we have more than one inverter, we need to use an AC combiner panel. 705.12(D) says in part, "Where distribution equipment including switchboards and panelboards is fed simultaneously by a primary source (s) of electricity and one or more utility-interactive inverters, and where this distribution equipment is capable of supplying multiple branch circuits or feeders or both, the interconnection provisions for the utility-interactive inverter(s) shall comply with (D)(1) through (D)(7)." This is clearly written about using an existing general use loadcenter or panelboard with other branch circuits. I understand that it is by default that the discussions on the subject are applying this to AC combiner panels because there isn't anything specifically written about using a dedicated panel. I also understand that the code panel's rejection of labeling and other means of identifying a dedicated panel would cause a person to use the general rules even though the conductor part is overkill. Whether the solution comes from a manufacturer creating a UL listed AC dedicated combiner or the code panel comes up with a separate section for a dedicated solar panel, we need a better solution. If there were to be a section describing the requirements for a dedicated solar AC combiner panel, I would add an exception that would allow a small amount (10% or ? of the ampacity of the panel) of solar PV related loads along with the multiple power sources for monitoring or weather station sensors. Especially with Enphase, the monitoring receptacle needs to be wired close to the termination of the multiple inverter circuits.

I was fat, dumb and happy before I started researching this subject further. As with other subjects, the more you learn, the more you realize what you don't know.
From all of the opinions that I've seen, it kind of leaves me on the fence. To summarize, my take is that the most conservative approach (with inspectors and the NEC) is to use the 120% rule on the conductors (and of course the busbars) feeding the AC combiner panel even though this will increase the conductor and conduit sizes by quite a bit. On the other hand a lot of installers seem to use the 120% rule on the busbars and then size the conductors for the largest OCPD feeding the conductors. Is that the general consensus?
Thanks to all for your patience with my questions. There aren't many places where you can have a serious discussion about details like this.
Jeff S.

 

[jaggedben]

To summarize, my take is that the most conservative approach (with inspectors and the NEC) is to use the 120% rule on the conductors (and of course the busbars) feeding the AC combiner panel even though this will increase the conductor and conduit sizes by quite a bit.

As has been discussed here in other threads, if your AHJ reads the code this way, it's hard to make an argument that the 120% rule doesn't apply to AC combiners. Now some AHJs have apparently been willing to approve variances that the code making panel wouldn't accept. I tend to think a full load center plus permanent labeling is good insurance. But its really a conversation to have with your AHJ, whether they insist on the most conservative interpretation of the code or are willing to be flexible.

With systems under 10kW or so it really doesn't make of a difference in terms of load center cost. Maybe more so for the conductors on long runs from combiner to interconnect.

On the other hand a lot of installers seem to use the 120% rule on the busbars and then size the conductors for the largest OCPD feeding the conductors. Is that the general consensus?

My impression is there isn't a general consensus of any kind on this subject. :happyno: I think a lot of people, both installers and inspectors, are still struggling to remember the language in the code.

 

[Marvin_Hamon]

The key phrase in 705.12(D) is "capable of supplying". Any conductor can be tapped, and any panel can have a load added. Even a dedicated PV panel could have one of the inverters disconnected and a load added so it is "capable of supplying" a load. A lot of people confuse "capable" with "currently", as in there are no loads in the panel today so it should not have to meet the 120% rule. All this has been brought up in the CMP during every code cycle since I have been involved in PV. There is a proposed change to the code for 2014 that will allow dedicated AC combiners and conductors, but I don't have great hope that it will do any better than past attempts.

Any conductor that has an OCPD and sources at both ends gets to meet the 120% rule too, which strangely enough often means that the conductor from the inverter to an OCPD does not need to be oversized. Basically we have to oversize the conductor assuming that the conductor will be tapped in the future, instead of the electrician in the future having to make sure the conductor is sized correctly before it is tapped. It starts to look a little crazy when you end up with 7 parallel conductors per phase to meet the rule. Who is going to tap that to add a load?

A lot of people seem to be ignoring the conductor requirement overall and the panel requirement when doing a dedicated AC combiner and don't get caught during plan check. I just submit a generic Alternative Methods and Materials request with every project where I want to use a dedicated combiner or not oversize a conductor so I am covered. Some AHJs are surprised that one is needed and some are not but I have not had many rejected.

 

[Jeff S]

Marvin and jagged Ben,
Thanks for your insightful comments.

 

[Jeff S]

Utility Disconnect Sizing

Utility Disconnect Sizing

Base on the above thread, it would seem that even though the code doesn't say it, you would need to use the 120% rule on sizing a non-fused utility required disconnect placed between the solar combiner panel and the breaker in the main service panel. For example:
Solar combiner panel with two 2P 15A breakers fed from a 30A breaker in the main service would then need a disconnect based on [2 x 15)+30]/1.2=50A. This would be the same as the busbar and conductors. It just doesn't seem to make sense that a 30A breaker would need a 60A disconnect in series with it. Any thoughts on this?

I've often been accused of thinking too deep on a subject. Guilty as charged.
Thanks,
Jeff

 

[jaggedben]

I would say you have it correct. And while some of this may not make much sense in the big picture, I think it does make sense that a disconnect have an equal or greater rating than the conductors connected to it.

 

[Marvin_Hamon]

705.12(D) does not apply to disconnects because they can't supply multiple circuits. While it's true that double lugged terminals can land multiple circuits at the terminals of the disconnect those are not internal. There is no bus in the disconnect that can have two sources and a load attached.

I have found that when conductors are upsized for the 120% then the disconnect ends up being upsized also because the conductors are now too big to land at the disconnect. Collateral damage from the 120% rule. It can be a real issue when you have to go from a 600A disconnect to a 1200A disconnect.

 

[Jeff S]

Marvin,
Are you saying then that in my example, if the wire fits, you would use a 30A disconnect?

Thanks,
Jeff