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

[shortcircuit2]

Maybe I am confused. In the drawing in post #17, when the transfer switch is in the generator (off grid) position, the SI's are connected to the generator. Am I missing something?

Of course, the safe way to do this is to connect the Sunny Boys to the main service panel. They would shut down when the grid is down, but how significant is that? If the grid is up most of the time, the difference in the SB's contribution to the bottom line could very well be negligible and not worth the trouble and risk to the genny.

Can someone post a link to the manual for review about the generator issue...please?

 

[ggunn]

Can someone post a link to the manual for review about the generator issue...please?

http://files.sma.de/dl/15216/SI45_60-eng-TB-TUS120812.pdf

I have set one of these up in a laboratory situation to start and stop a generator. The relay you use for generator start can be set to open (or close) on a user chosen number for battery SOC to start the genny, and set to close (or open) on another SOC number to stop the genny. One wrinkle is that if the batteries have not been fully charged in the past two days, it will ignore the stop point and run the genny until the battery SOC reaches 100%.

 

[shortcircuit2]

http://files.sma.de/dl/15216/SI45_60-eng-TB-TUS120812.pdf

I have set one of these up in a laboratory situation to start and stop a generator. The relay you use for generator start can be set to open (or close) on a user chosen number for battery SOC to start the genny, and set to close (or open) on another SOC number to stop the genny. One wrinkle is that if the batteries have not been fully charged in the past two days, it will ignore the stop point and run the genny until the battery SOC reaches 100%.

236 pages...thats alot of reading.

"In grid parallel operation the Sunny Island automatically detects a power outage and automatically switches the Sunny Boy inverter to "Off‑Grid" mode. To set up this arrangement, the Sunny Island and the Sunny Boy have to be connected with a communication cable and the parameters of the Sunny Boy inverters has to be set according to this documentation. In case the grid returns, the Sunny Island switches back to grid‑tie mode according to "UL 1741".

So in "Off-Grid" mode the system prevents backfeed into the gen-set.

Thanks for the link to the install manual.

OK, lets get back to the OP issue.

 

[sandsnow]

Of what you cited, I don't see it. You'll have to elaborate.

(BTW, 408.16 is regarding switchboards in damp locations)

300.3(B)(1) all conductors of the same circuit. Feeding one panel to another is done by one circuit normally. I don't think the arrangement in the manual is addressed one way or another in the NEC.

310.4 - Parallel conductors - We are splitting the load between multiple conductors. Starts at one point spreads out and comes back together at the load end.

240.15 Circuit breaker shall have handle ties at a minimum to open all conductors of the circuit.

408.36 - Panelboard shall be protected by an oCPD. Code text use of the word "an" indicates one OCPD. Here is where I'm really not sure. Exception 2 allows two OCPD to protect a panelboard. Reading this says two mains. I would see this as as two breakers either internal common trip or handle ties. Even with the exception there are too many OCPD protecting the panelboard.

 

[sandsnow]

Generator Issue - Paper explains everything

http://www.sma-america.com/en_US/pr...pe=281&cHash=ef50ec5afb66795d5d9ae2e080f22e17

 

[shortcircuit2]

300.3(B)(1) all conductors of the same circuit. Feeding one panel to another is done by one circuit normally. I don't think the arrangement in the manual is addressed one way or another in the NEC.

310.4 - Parallel conductors - We are splitting the load between multiple conductors. Starts at one point spreads out and comes back together at the load end.

240.15 Circuit breaker shall have handle ties at a minimum to open all conductors of the circuit.

408.36 - Panelboard shall be protected by an oCPD. Code text use of the word "an" indicates one OCPD. Here is where I'm really not sure. Exception 2 allows two OCPD to protect a panelboard. Reading this says two mains. I would see this as as two breakers either internal common trip or handle ties. Even with the exception there are too many OCPD protecting the panelboard.

Is your concern that there are multiple sources supplying one piece of equipment and there are multiple disconnects that don?t simultaneously kill the power?

Would 705.22(5) require only one disconnect that your looking for?

shortcircuit

 

[shortcircuit2]

Generator Issue - Paper explains everything

http://www.sma-america.com/en_US/pr...pe=281&cHash=ef50ec5afb66795d5d9ae2e080f22e17

So it can be done with a relay. I have seen this done with the Outback equipment.

Thanks for the link!

 

[ggunn]

236 pages...thats alot of reading.

"In grid parallel operation the Sunny Island automatically detects a power outage and automatically switches the Sunny Boy inverter to "Off‑Grid" mode. To set up this arrangement, the Sunny Island and the Sunny Boy have to be connected with a communication cable and the parameters of the Sunny Boy inverters has to be set according to this documentation. In case the grid returns, the Sunny Island switches back to grid‑tie mode according to "UL 1741".

So in "Off-Grid" mode the system prevents backfeed into the gen-set.

Sort of. The SI can be set to start and stop the generator on battery SOC setpoints, and it's when the batteries are full that the danger of backfeeding the generator is there. At that point the generator will be offline. When the SB(s) is/are in off-grid mode, the SI can cut their output so that you do not overcharge the batteries when the demand is low and the grid is down.

 

[sandsnow]

Is your concern that there are multiple sources supplying one piece of equipment and there are multiple disconnects that don?t simultaneously kill the power?

Would 705.22(5) require only one disconnect that your looking for?

shortcircuit

I believe the disconnect at each inverter (sunny boy) satisfy that.

 

[Smart $]

300.3(B)(1) all conductors of the same circuit. Feeding one panel to another is done by one circuit normally. I don't think the arrangement in the manual is addressed one way or another in the NEC.

We have two views here IMO. "Yours" and that each SI is a piece of equipment. As a piece of equipment each Si is supplied with a single circuit. The SI goes well beyond the function of a basic device. It is UL listed, so its up to the AHJ to approve (or not).

310.4 - Parallel conductors - We are splitting the load between multiple conductors. Starts at one point spreads out and comes back together at the load end.

True in the sense of grid power supplying loads on the "off-grid" side when needed. Its not actually parallel conductors as described in 310.10(H), so the the actual section that applies is 240.8. Here again I have to state, it is UL listed, so its up to the AHJ to approve (or not).

240.15 Circuit breaker shall have handle ties at a minimum to open all conductors of the circuit.

Here again I have to state, it is UL listed, so its up to the AHJ to approve (or not).

408.36 - Panelboard shall be protected by an oCPD. Code text use of the word "an" indicates one OCPD. Here is where I'm really not sure. Exception 2 allows two OCPD to protect a panelboard. Reading this says two mains. I would see this as as two breakers either internal common trip or handle ties. Even with the exception there are too many OCPD protecting the panelboard.

Nothing in 690 or 705 deals with this specifically. However... 705.12(D)(7) states regarding inverter output connection, "Unless the panelboard is rated not less than the sum of the ampere ratings of all overcurrent devices supplying it, a connection in a panelboard shall be positioned at the opposite (load) end from the input feeder location or main circuit location." This infers more than one inverter ocpd can supply a panelboard in addition to an ocpd for a primary or alternate source. 705 overides 408.

 

[ggunn]

300.3(B)(1) all conductors of the same circuit. Feeding one panel to another is done by one circuit normally. I don't think the arrangement in the manual is addressed one way or another in the NEC.

310.4 - Parallel conductors - We are splitting the load between multiple conductors. Starts at one point spreads out and comes back together at the load end.

240.15 Circuit breaker shall have handle ties at a minimum to open all conductors of the circuit.

408.36 - Panelboard shall be protected by an oCPD. Code text use of the word "an" indicates one OCPD. Here is where I'm really not sure. Exception 2 allows two OCPD to protect a panelboard. Reading this says two mains. I would see this as as two breakers either internal common trip or handle ties. Even with the exception there are too many OCPD protecting the panelboard.

Doesn't your AHJ approve or reject your design based on your drawings when you apply for the interconnect agreement?

 

[sandsnow]

Doesn't your AHJ approve or reject your design based on your drawings when you apply for the interconnect agreement?

I am not the applicant.
I work for the AHJ as a plan checker and inspector.

The interconnect agreement has nothing to do with us.
The purpose of the OP was to determine Code compliance since we have never seen this before.

 

[sandsnow]

We have two views here IMO. "Yours" and that each SI is a piece of equipment. As a piece of equipment each Si is supplied with a single circuit. The SI goes well beyond the function of a basic device. It is UL listed, so its up to the AHJ to approve (or not).

My "view" needs to be in quotations? We are the AHJ. Yes it is listed and we are in contact with UL on it. The product being listed is only part of the story. Is the product being used in the way it was evaluated?


True in the sense of grid power supplying loads on the "off-grid" side when needed. Its not actually parallel conductors as described in 310.10(H), so the the actual section that applies is 240.8. Here again I have to state, it is UL listed, so its up to the AHJ to approve (or not).

Thanks for the 240.8 section. The breakers in the heavy load panel and protected load panel are definitley in parallel and are not part of a listed assembly. Again we are not concerned with off grid function. It is ON GRId is where we see a possible code violation. No it's not really parallel as far as the ungrounded conductors are concerned. I was off base there. The neutral conductors are connected in parallel, but if UL stands behind the configuration I doubt that will be an issue.

Here again I have to state, it is UL listed, so its up to the AHJ to approve (or not).

See above.


Nothing in 690 or 705 deals with this specifically. However... 705.12(D)(7) states regarding inverter output connection, "Unless the panelboard is rated not less than the sum of the ampere ratings of all overcurrent devices supplying it, a connection in a panelboard shall be positioned at the opposite (load) end from the input feeder location or main circuit location." This infers more than one inverter ocpd can supply a panelboard in addition to an ocpd for a primary or alternate source. 705 overides 408.
The rule from 705 only kicks in when we are in off grid mode and the batteries are supplying power through the SI inverter. Again we are fine with the off grid function.

q

 

[Smart $]

My "view" needs to be in quotations? We are the AHJ.

I put "Yours" in quotations to indicate collectively you as one AHJ and all others having the same view.

Yes it is listed and we are in contact with UL on it. The product being listed is only part of the story. Is the product being used in the way it was evaluated?

That's the proper position for an AHJ to take on the matter. I'm just looking at it from a Code view, as I don't have access to the entire standard by which the product was evaluated. I just know by Code the AHJ approves listed equipment (or not) and it gets installed per manufacturers instructions as part of the listing.

Thanks for the 240.8 section. The breakers in the heavy load panel and protected load panel are definitley in parallel and are not part of a listed assembly. Again we are not concerned with off grid function. It is ON GRId is where we see a possible code violation. No it's not really parallel as far as the ungrounded conductors are concerned. I was off base there. The neutral conductors are connected in parallel, but if UL stands behind the configuration I doubt that will be an issue.

The neutral conductors are no more in parallel than the ungrounded. Each neutral supplied from heavy loaded panel lands on one terminal of its respective SI. Each neutral connected to protect panel lands on one terminal of its respective SI... and that is not the same terminal nor an electrically joined terminal as the grid side. Yes the neutrals on each side are likely "bridged internally", but how else can multiple parallel power sources be connected between two panels??? If rejected as you are implying, the manufacturer will have to combine all four units into one, so we can run just one neutral, before you'll accept it... or do we have to run 1/0AWG copper :huh:

The rule from 705 only kicks in when we are in off grid mode and the batteries are supplying power through the SI inverter. Again we are fine with the off grid function.

I don't see how 705,12(D)(7) only kicks in when in off-grid mode. Can you elaborate on how you see it this way?

 

[sandsnow]

I put "Yours" in quotations to indicate collectively you as one AHJ and all others having the same view.


That's the proper position for an AHJ to take on the matter. I'm just looking at it from a Code view, as I don't have access to the entire standard by which the product was evaluated. I just know by Code the AHJ approves listed equipment (or not) and it gets installed per manufacturers instructions as part of the listing.


The neutral conductors are no more in parallel than the ungrounded. Each neutral supplied from heavy loaded panel lands on one terminal of its respective SI. Each neutral connected to protect panel lands on one terminal of its respective SI... and that is not the same terminal nor an electrically joined terminal as the grid side. Yes the neutrals on each side are likely "bridged internally", but how else can multiple parallel power sources be connected between two panels??? If rejected as you are implying, the manufacturer will have to combine all four units into one, so we can run just one neutral, before you'll accept it... or do we have to run 1/0AWG copper :huh:


I don't see how 705,12(D)(7) only kicks in when in off-grid mode. Can you elaborate on how you see it this way?

Because that's when the inverter is energized. In On grid mode the current is just passed through. With regard to the neutral, how do we keep the right amount of current flowing on the respective neutrals? While maybe a long shot, if the difference in length is long enough then the division of current will be unbalanced enough to overheat one of the neutral conductors. Again, when in on grid mode. With regard to 1/0 copper, the Code allows reduction to #2 when done under engineering supervision. If UL confirms use of product in this manner, I could consider that engineering supervision. For some reason, my enter key is not working and I cannot form paragraphs.

 

[Smart $]

Because that's when the inverter is energized. In On grid mode the current is just passed through.

We are talking more than one supply (two per the exception) to a panelboard, correct? So you have no problem in this regard when in off-grid mode, as 705 kicks in. Your objection is in on-grid mode when current just passes through. Your saying the protected loads panel is being supplied from grid by more than one supply ocpd. Okay, I see your objection hmy:

Part of the Sunny Island listing as a UL category QIKH "Static Inverters, Converters and Accessories for Use in Independent Power Systems" is that it has been evaluated as a utility-interactive inverter. As such, I feel it's association with the panels to either side is as a utility-interactive inverter. What operation mode the SI is in doesn't matter... 705 applies.

http://database.ul.com/cgi-bin/XYV/...n=versionless&parent_id=1073991993&sequence=1
http://database.ul.com/cgi-bin/XYV/...n=versionless&parent_id=1073991993&sequence=1

With regard to the neutral, how do we keep the right amount of current flowing on the respective neutrals? While maybe a long shot, if the difference in length is long enough then the division of current will be unbalanced enough to overheat one of the neutral conductors. Again, when in on grid mode.

We're getting back to parallel conductors on this matter, but not in the sense the NEC uses under 310.10(H). Since this is a split-phase 3-wire system, the parallel neutral currents establish conductors of the circuit (i.e. parallel currents but not NEC parallel conductors), so 300.3(B) does apply, but not 300.3(B)(1). However, consider that 310.10(H) parallel conductors establishes rules to keep the circuit currents balanced when separated. That is impossible in this installation when SI conductor pairs are separated, so the conductors of all SI's will have to be run in the same raceway etc.

As to overheating a single neutral, that is likely to never occur. Say we have an extreme unbalanced condition... two SI's of the same leg at max input/output (112A, theoretically?). That will be on two ungrounded conductors... but will return on four neutrals. The current on the neutrals won't even get close to their ampacity... probably not even half ...and that's the worst case under nominal conditions. I don't believe any realistic amount of distance will change that enough to have current in excess of any one neutral's ampacity.

With regard to 1/0 copper, the Code allows reduction to #2 when done under engineering supervision. If UL confirms use of product in this manner, I could consider that engineering supervision.

As I have been saying, the setup does not establish parallel conductors under the stipulations of 310.10(H), so the minimum 1/0 or the exception for #2 is not applicable.

For some reason, my enter key is not working and I cannot form paragraphs.

It's a problem with Internet Explorer 10 (perhaps limited to IE10 on Windows 8... don't know for certain) and forum software (not just this forum, but any that uses the same base coding). Only happens in WYSIWYG Mode of the reply editor. You can copy and paste a "return" or switch the reply editor to Source Mode (upper left icon).

 

[sandsnow]

If it doesn't matter which mode the SI is in, then UL should be able to address 240.8. UL has been slow to respond.

Yeah I agree the neutrals overheating are a long shot. Still they are connected together at each end, but the intent is not to form a single conductor as the Code says.

I think this Sunny Island is a great product and what it does is great. We still have to stick to what is permitted by Code and what is not.

We're asking the manufacturer for their input.

Based on what we have we cannot accept installation as proposed. The product itself is not being rejected, just how it is wired.

 

[ggunn]

If it doesn't matter which mode the SI is in, then UL should be able to address 240.8. UL has been slow to respond.

UL is notorious for their slow response, but to be fair, they have a lot on their plate. I don't see why the designer wouldn't want to split the SIs' output into a pair of subpanels if it would avoid this issue.

 

[Smart $]

UL is notorious for their slow response, but to be fair, they have a lot on their plate. I don't see why the designer wouldn't want to split the SIs' output into a pair of subpanels if it would avoid this issue.

In all fairness to sandsnow's perspective (to which I've only been taking the position of devil's advocate to stimulate the discussion), when SI's are 3?-connected, more than one per phase must connect to the system through a multicluster "box", and no more than one SI per phase per cluster. It seems if this multicluster interface is needed for more than one SI per phase on 3?, a similar "device" is needed for more than one SI per leg on split phase....

Another question I have is where the diagrams posted by the OP'er came from. I've been all over the website and cannot find any similar diagram depicting two SI's per leg on split phase system.

 

[ggunn]

In all fairness to sandsnow's perspective (to which I've only been taking the position of devil's advocate to stimulate the discussion)...

Which I take to mean, "I'll argue about anything." I know this about you.

Another question I have is where the diagrams posted by the OP'er came from. I've been all over the website and cannot find any similar diagram depicting two SI's per leg on split phase system.

The first drawing showed one SI per phase (to neutral) on a three phase system and three SB's phase to phase on the protected loads panel, which isn't a problem.