Hybrid System , PV, Battery, Generator, ATS Confused Designer? Parallel feeders &ocpd

[TrickleCharge]

Hybrid System , PV, Battery, Generator, ATS Confused Designer? Parallel feeders &ocpd

Ok so I have a designer/contractor who is proposing this system with the ats, generator, pv, batteries, etc. Now the one thing that I have a question on is the "parallel feeds" that come from one panel to feed the sub panel. Basically these feeds go thru the (4) sunny island inverters via (4) 70A s.p. breakers to a sub panel which land on (4) 70A s.p breakers. Do you think that section 300.3 applies for parallel conductors which takes you to 310.4 for the requirements such as min 1/0 conductors etc? The colored diagram is a vague plan on how this will be done. The second hand written draft is showing the principal of parallel feeders. The other input that I gave the designer is compliance with 240.15 in regards to having the (4) single pole breakers feeding the sub panel. Thanks for the input. View attachment Wayne Dr 2.pdf View attachment Wayne Dr 1.pdf


View attachment Wayne Dr 1.pdf
View attachment Wayne Dr 2.pdf

 

[jaggedben]

There are some others on this forum who are more expert on Sunny Islands than I am, so hopefully they will chime in here. With that said, here's my feedback:

1) The drawings don't match and are unclear as to their meaning. In the hand drawing, do the "3/c" and "2/c" notations include the neutral conductor or not? If they do, then you've got a double-split-phase system, which makes sense for 4 inverters. If they do not, and it's a 3 phase system....then the whole thing is not right. I hope that the adapted drawing from SMA, which shows a 3 phase system, is just being used for convenience.

2) Assuming it's a double-split phase system: Interesting question about the parallel feeds, but in my opinion 310.4 (or 310.10 in 2011) does not apply. The individual circuits feeding each Sunny Island can be viewed as individual branch circuits, especially in off grid mode. Each separate feed is sufficiently protected by its own breaker, so it is safe. Moreover the maximum size that the Sunny Island terminals will accept is 4wg, so there's no question of going up to 1/0. With that said, it would be advisable to adhere to 310.10(H)(2) as closely as possible. (I'm using 2011 NEC here, 'cause that's what I have in front of me.) That is, same length (approximately), size, and type. Put the Sunny Islands that are on the same leg closer to each other to facilitate this.

If it's a 3 phase system with 3 inverters, there are no parallel feeds.

3) With regard to 240.15 I'm not sure what issue you are concerned about. Note that this can NOT be a multi-wire branch circuit setup. You need to send a separate neutral to each Sunny Island. Thus separate circuit breakers are appropriate.

4) Not sure whether the slave Sunny Islands will break the AC1-AC2 connection if the master loses grid/generator. Maybe someone else here can say.

 

[jaggedben]

3) With regard to 240.15 I'm not sure what issue you are concerned about. Note that this can NOT be a multi-wire branch circuit setup. You need to send a separate neutral to each Sunny Island. Thus separate circuit breakers are appropriate.

Ammending my previous comment after re-reading your comments in the picture ... I see that you are concerned about 240.15(B), and you are treating it as one circuit.

A couple points:
1) See point (4) above. If the Sunny Islands can be programmed to drop the AC1-AC2 connection when their counterparts lose grid/generator voltage, then this would fit under "automatically" in 240.15(B).
2) Another option might be a 4-pole breaker, if such a thing exists for the panel they are installing on.

 

[ggunn]

Is your house supplied with 208/120 3 phase? It looks like there are three Sunny Boys wired 208V phase to phase and three Sunny Islands wired 120V phase to neutral. That should work, though the programming of the SI master so that you don't backfeed the generator when the SB's are producing more than your loads and your batteries are full, and starting the genny when the SB's are not producing and the batteries are low, etc. can be a bit tricky. Is your prospective installer a Sunny Island expert?

There aren't any parallel conductors or parallel feeds. Each SI feeds one of the three phases of the 208 and gets return current on the shared neutral. The three SB's are in a normal 208 configuration.

 

[sandsnow]

Im working on this with the op. He should have posted the sma single phase drawing with 4 SI. Then the crude drawing (mine) would hopefully make more sense.
Our concern is when the system is in off grid mode. Not sure what happens in the si, but doesnt the ac just feed right through? Like transfer switch in normal mode.
If so then we have a parallel feed from one panel another. We definitely would not supply a subpanel as shown in the crude drawing, right? The crude drawing would be a violation of 310.4 i believe.
What does the addition of the si change?

Were just trying to get an understanding of these hybrid systems
Thanks

 

[TrickleCharge]

Oops I uploaded the 3phase diagram. The system is a typical residential 1p 3w system. To reiterate on another point is the fact of supplying power to a sub panel from a series of s.p. breakers. If I shut off one of the breakers my sub panel is still energized by the other 3 sp breakers. How is this considered branch circuits when ultimately we are providing power to a 200a sub panel. Take away the SI and you have 4 separate branch ckts to suplly pwr to a sub panel.....is that practical? Thanks again for your input.

 

[jaggedben]

Im working on this with the op. He should have posted the sma single phase drawing with 4 SI. Then the crude drawing (mine) would hopefully make more sense.
Our concern is when the system is in off grid mode. Not sure what happens in the si, but doesnt the ac just feed right through? Like transfer switch in normal mode.

Maybe you just mis-typed, but I believe you have it backwards. It's in the on-grid mode that the ac just feeds through. In this case it would also be true if the generator is operating.

We definitely would not supply a subpanel as shown in the crude drawing, right? The crude drawing would be a violation of 310.4 i believe.

Would that last sentence be true if the conductor was 1/0? Unusual, yes, but a code violation?

If I shut off one of the breakers my sub panel is still energized by the other 3 sp breakers.

Again, worth finding out if the SIs can 'automatically' prevent this. I'm not sure of the answer myself, but it might be possible. And again, maybe it can be done with a 4-pole breaker. Or perhaps some other contactor setup can do it.

Take away the SI and you have 4 separate branch ckts to suplly pwr to a sub panel.....is that practical?

If by 'practical' you mean 'will it work?', then yes, it is practical.
More to the point, I would take issue with your phrase "Take away the SI". The inclusion of the SIs is a reasonable and valid justification for doing it this way. Brings up some sticky code issues maybe, but still reasonable and valid. The question is, is it safe? Besides, the question of de-energizing the sub with one handle, which can probably be dealt with, I don't see why it isn't per se. He'll need to watch out for 300.20, but that's doable.

Some further food for thought:

-Does the designer actually need a single critical loads subpanel? Or would he be willing to split it into two subpanels and avoid the whole controversy?
-Supposing he can't/won't do that, do you guys ever grant variances? Seems to me this would be the place for one; 310.4 clearly wasn't written with this in mind. He can satisfy almost all the conditions for the exception and with his conductors each protected by two breakers, it seems unlikely there'd be a problem. That is, if he can satisfy you on 240.15(B) and 300.20.

 

[sandsnow]

Maybe you just mis-typed, but I believe you have it backwards. It's in the on-grid mode that the ac just feeds through. In this case it would also be true if the generator is operating.

Yes I mis-typed - ON GRID Mode is correct


Would that last sentence be true if the conductor was 1/0? Unusual, yes, but a code violation?

I am really not completely sure. If there were 240V loads in the protected panel, then definitely. We could not separate the ungrounded conductors.
Since the SB inverter(not SI) is 240V output, then those phases have to be kept together. 300.3(B).

There are two exceptions to 408.36. The general rule is a panelboard has to be supplied by AN ocpd.
Exc 1 seems to apply only where 230.71 is used.
Exc 2 only allows two main circuit breakers to supply the panel.

Both exceptions utilize the term main circuit breaker or sets of fuses which "traditionally" imply a 2 pole or 3 pole C/B or single fusible switch.




Again, worth finding out if the SIs can 'automatically' prevent this. I'm not sure of the answer myself, but it might be possible. And again, maybe it can be done with a 4-pole breaker. Or perhaps some other contactor setup can do it.



If by 'practical' you mean 'will it work?', then yes, it is practical.
More to the point, I would take issue with your phrase "Take away the SI". The inclusion of the SIs is a reasonable and valid justification for doing it this way. Brings up some sticky code issues maybe, but still reasonable and valid. The question is, is it safe? Besides, the question of de-energizing the sub with one handle, which can probably be dealt with, I don't see why it isn't per se. He'll need to watch out for 300.20, but that's doable.

I don't see "yet' how the SI makes everything ok. Yes it has electric interlock for shutdown etc, but we still have the issue of the 240V output being split. I don't see how that is mitigated unless the SI is an SDS between the two AC connections. In the end it appears to me the SI is just a fancy transfer switch between batteries and the grid.

Some further food for thought:

-Does the designer actually need a single critical loads subpanel? Or would he be willing to split it into two subpanels and avoid the whole controversy?
-Supposing he can't/won't do that, do you guys ever grant variances? Seems to me this would be the place for one; 310.4 clearly wasn't written with this in mind. He can satisfy almost all the conditions for the exception and with his conductors each protected by two breakers, it seems unlikely there'd be a problem. That is, if he can satisfy you on 240.15(B) and 300.20.

Yes two subpanels could be suggested. We do grant variances and allow for new equipment or technologies where the Code is not caught up yet. Basic Code rules still have to followed which is our concern here. We try to be open minded, but we won't accept everything just because the applicant wants to do it.

I'll come back to this thread on Monday. Thanks for the replies.

 

[ggunn]

-Does the designer actually need a single critical loads subpanel? Or would he be willing to split it into two subpanels and avoid the whole controversy?
-Supposing he can't/won't do that, do you guys ever grant variances? Seems to me this would be the place for one; 310.4 clearly wasn't written with this in mind. He can satisfy almost all the conditions for the exception and with his conductors each protected by two breakers, it seems unlikely there'd be a problem. That is, if he can satisfy you on 240.15(B) and 300.20.

I don't see what the controversy is. Each SI just passes one phase of the service onto the three phase 208 protected loads sub panel, connected phase to neutral. The SB's are phase to phase 208 feeding the other end of the sub panel busbars and the SI's are transparent to power (up to ~60A IIRC connected through a 3P backfed breaker to the main panel) flowing either direction through them. Nothing is in parallel. When the grid goes down the transfer switches in the SI's cut the protected loads from the main and each one of them energizes one of the busbars in the sub from batteries. The SB's keep pushing power onto the sub panel phase to phase as long as the SI's let them.

When the batteries get depleted the master SI starts the generator and closes the three SI transfer switches to energize the protected loads panel from the genny. The SI senses when the master transfer switch has the genny (rather than the grid) selected so it can keep the SB's from backfeeding the genny (assuming that the power from the genny is clean enough for the SB's to synch with it, otherwise they just shut down). The SI programming to make all that happen is a bit complex (and you have to get into the SBs' firmware to make them think they are off grid all the time) , but it's all in the manual. I have set up a split phase system with one SI, one SB, a generator, and an autotransformer that ran this way.

Am I missing something? If the service is not 208/120 three phase, then I have no clue.

 

[jaggedben]

Am I missing something? If the service is not 208/120 three phase, then I have no clue.

Yeah, you missed where they said that the system is not 3-phase. We're talking about 4 SIs in a dual-split-phase system.

 

[jaggedben]

Basic Code rules still have to followed which is our concern here. We try to be open minded, but we won't accept everything just because the applicant wants to do it.

How bout this...
Would you feel better if the applicant used a pre-wired system such as this one?

http://www.midnitesolar.com/product...productCatName=E-Panel - SMA&productCat_ID=38
http://www.midnitesolar.com/pdfs/SMA_QUAD_BATT__COMBINER_AC-DC_wiring.pdf
http://www.midnitesolar.com/product...e-Wired Systems - AC Coupled&productCat_ID=31

Basically all the code issues you've brought up would be eliminated outside of a listed assembly. You'd have a single circuit from the grid, to the assembly, and on to the subpanel. I'd bet that Midnite Solar would be happy to talk through the various code implications with you. They've probably thought about them. They also have their own forum on their website.

I was going to make a point for point argument about how it is probably possible to install everything with a method that addresses the various code concerns, and then it occurred to me that such a method would probably look like a Midnite E-panel.

Admittedly, from what you've told us, it looks like the applicant has some homework to do as well. The more I reflect on the single pole breakers the less I want to defend that. :roll:

 

[ggunn]

Yeah, you missed where they said that the system is not 3-phase. We're talking about 4 SIs in a dual-split-phase system.

You're right, I did miss that.

The drawing in the OP showed three SI's and three SB's, with different colored (red/blue/green, blue/black/green, and red/black/green) conductors coming from each SB, which looks clearly phase to phase three phase to me with neutral as green. The SI's conductors were red/green, blue/green, and black/green, which again shows a phase to neutral three phase connection.

 

[TrickleCharge]

Ok here is the updated "simplified" version of the proposed installation. Thanks for the input!

 

[Smart $]

Ok here is the updated "simplified" version of the proposed installation. Thanks for the input!

I'm just not seeing any basis to your objections. Each SI is a separate source, separate load, and wired as such using individual neutrals. Suggest some labeling which is more obvious than a panel directory [see 705.12(D)(7)].

What I do see as a problem, is compliance with 705.12(D)(2), the so-called 120% rule. Your Heavy Loaded Panel has 200A, 70A, and 70A supplies (340A summed) supplying a 200A panel (currently assuming it has 200A bus and feeder if MLO).

 

[jaggedben]

I'm just not seeing any basis to your objections. Each SI is a separate source, separate load, and wired as such using individual neutrals.

The concern is when the SIs are not operating and simply pass through the connection from one panel to the other. The OP is correct that this is an electrically parallel connection, whether or not you think certain code provisions apply.

What I do see as a problem, is compliance with 705.12(D)(2), the so-called 120% rule. Your Heavy Loaded Panel has 200A, 70A, and 70A supplies (340A summed) supplying a 200A panel (currently assuming it has 200A bus and feeder if MLO).

The SIs will never supply the Heavy-Loaded Panel. That's not how this system works. They will supply the "Protected AC loads panel", where they will supply up to 140A per leg.

The 120% rule does need to be applied, separately, to the Sunny Boys plus the grid/generator, and to the Sunny Boys plus the SIs. But it hasn't been stated how many Sunny Boys there are, if any, and what their output is.

 

[jaggedben]

The 120% rule does need to be applied, separately, to the Sunny Boys plus the grid/generator...

Scratch the word 'generator'. (There's only one breaker anyway.)

 

[Smart $]

The concern is when the SIs are not operating and simply pass through the connection from one panel to the other. The OP is correct that this is an electrically parallel connection, whether or not you think certain code provisions apply.

Still, each SI is a separate source, separate load. Doesn't matter when passing current through that the power of each SI has the same origin.

Now if the SI's weren't in the picture, the violation would be 240.8.

The SIs will never supply the Heavy-Loaded Panel. That's not how this system works. They will supply the "Protected AC loads panel", where they will supply up to 140A per leg.

Okay. When I googled them, the snippets indicated they were inverters... but I didn't get into their operation. Had get info off mfg website to understand what they did and why they were called inverters.

 

[sandsnow]

How bout this...
Would you feel better if the applicant used a pre-wired system such as this one?

http://www.midnitesolar.com/product...productCatName=E-Panel - SMA&productCat_ID=38
http://www.midnitesolar.com/pdfs/SMA_QUAD_BATT__COMBINER_AC-DC_wiring.pdf
http://www.midnitesolar.com/product...e-Wired Systems - AC Coupled&productCat_ID=31

Basically all the code issues you've brought up would be eliminated outside of a listed assembly. You'd have a single circuit from the grid, to the assembly, and on to the subpanel. I'd bet that Midnite Solar would be happy to talk through the various code implications with you. They've probably thought about them. They also have their own forum on their website.

I was going to make a point for point argument about how it is probably possible to install everything with a method that addresses the various code concerns, and then it occurred to me that such a method would probably look like a Midnite E-panel.

Admittedly, from what you've told us, it looks like the applicant has some homework to do as well. The more I reflect on the single pole breakers the less I want to defend that. :roll:

Using that equipment and being a listed system - great! Install per instructions and go home.

I am trying to get an answer from UL about the SI. If UL has investigated them for the quad installation, then great. So far no answer.

 

[sandsnow]

Still, each SI is a separate source, separate load. Doesn't matter when passing current through that the power of each SI has the same origin.

Now if the SI's weren't in the picture, the violation would be 240.8.



Okay. When I googled them, the snippets indicated they were inverters... but I didn't get into their operation. Had get info off mfg website to understand what they did and why they were called inverters.

We have no issue with the SI being used as a battery inverter.

When the SI are in grid mode we thought the current will be passing straight through as if the SI weren't there.

 

[BillK-AZ]

In reading an install manual (Version 2, downloaded in 2012) for the SI5048U it seems that the following statement by jaggedben:

The SIs will never supply the Heavy-Loaded Panel. That's not how this system works. They will supply the "Protected AC loads panel", where they will supply up to 140A per leg.

May not be true, the SI can backfeed the grid. The manual states

AC feed-in generators on the stand-alone grid side (Sunny Boys) can feed their energy into the grid through the internal transfer relay of the Sunny Island 5048U, for limitations, see section 14.2.8 "Limits and Power Adjustment" (Page 116).
Following grafic (sic) shows the energy flow direction for net metering and energy consumption from public grid:

If feeding the grid is a possibility, then 705.12(D)(2), the so-called 120% rule, applies and the Heavy Loaded Panel is undersized. The maximum AC out to the grid is not clear from the manual I have, but can be 56A per SI and there are two on each phase (the SI is 120V), a 70A CB would be used for each SI. Smart $ seems correct.