Plan Check + Neutral Discussion

[atlsolar]

I posted this question elsewhere but trying here since I've gotten no response yet. Attached is a drawing of my overall proposed layout.

Summary of My Setup and Question​

I have a home with two separate 200A services that are spliced in a wire trough to feed three main panels. I’m repurposing one Siemens 200A panel as a sort of “splitter”, removing its branch breakers and installing only three breakers (100A, 100A, 125A) in their place to feed two inverters and a new subpanel. The new subpanel will hold the original branch circuits.

The other two panels each have their own main breakers (one is 200A ITE, the other is a 150A breaker feeding another 200a ITE panel). Each panel will be backfed from the inverters via a new output panel into a backfeed breaker (with interlock kit) or a manual transfer switch (MTS).

My inverters (EG4 12000XP’s) maintain a continuous neutral between AC input and output but do not create a neutral-ground bond, so they must be set to floating mode since my main panel already has a neutral-ground bond. All subpanels have isolated neutrals.

The question is about whether I must run a dedicated neutral from the inverter output panel to each panel’s backfeed breaker and manual transfer switch—especially given that all panels share the same house neutral bus anyway. If I do run new neutrals, will this possibly create parallel neutral paths?

---

Avoiding the Parallel Neutral Issue​

I found a discussion here (link here) where someone warns against running two neutrals—one from the main panel to the inverter input and another from the inverter output panel to the backfed panel. The concern is that this would create parallel neutral paths, which violates NEC 310.10(H) and can cause unexpected current flow, but I have separately red that 300.3(B) & 310.10(H) require all conductors of a circuit be in the same cable/conduit to avoid inductive heating.

Additionally, a video discussing this same issue (YouTube link) explains why running two neutrals is unnecessary and can lead to grounding and troubleshooting issues. The person in the video emphasizes that the backfed panel already has a neutral from the main service, so running another neutral from the inverter output panel is redundant and potentially problematic.

---

Final Questions​

Given this setup, do I need to run a neutral from the inverter output panel to each backfeed breaker and the manual transfer switch, or is the existing shared neutral sufficient? If I do run new neutrals, would this cause parallel neutral paths, and is there any scenario where running an extra neutral is actually needed? When I am moving the branch breakers to the subpanel, can I tie the original and new panels' neutral bus bars together with a single thicker gauge wire and only move the hot L1 and L2 lines?

Thanks.

 

[Ted1000]

I noticed in the video from Ray Loveless that you linked that the L1 and L2 grid connection wires are in a separate conduit without a neutral. Isn't that wrong? If you share a neutral shouldn't they be in the same conduit?

 

[atlsolar]

That’s essentially what I’m trying to figure out. Ray’s video has just the two which seems wrong, but Jesse Muller’s video has all three and gets called out for having two neutrals.

But without the neutral, you aren’t putting the full circuit through the same conduit.

It’s like a circle, and I can’t figure out which is technically correct and/or practically correct in my situation. Plus, not feeding the neutral back to all three saves me about $1000 in 1awg neutral…

 

[CMDRZOD]

Canadian here. We have similar codes but not identical.

I wouldn't run neutrals from EG4 outputs to the backfed breakers. Reasoning:

1. Panel is already supplied a common neutral. Parallel paths is not code. This would overrule needing all circuit conductors IMO.

2. It can be argued that the neutrals are not technically circuit conductors. They are neutrals and conduct unbalanced current only. There are cases where they carry zero current.

3. You already have achieved a solid connection back to neutral.

However your inspection authority might not agree. I would talk with them. I would run bonding conductors with each run with omitted neutrals.

 

[atlsolar]

Thanks @CMDRZOD. That's what I'm thinking also, but everywhere I look, I get conflicting responses and explanations.

To clarify what I'm thinking in this moment:

• existing main (turned "splitter) to new subpanel via 125a breaker & bus bars - include 1awg L1, L2, N, and 6awg G
  • Thoughts on being able to just pull 1awg N from old bus to new bus, and then not have to move neutrals on each branch? Pretty similar question, and in my head, I shouldn't need to pull individual neutrals and can just pull the 2 hots.
• 100a "splitter" to each inverter's IN/bus - 3 awg L1, L2, N and 6awg G
• Inverter load/bus to combiner output panel - 3awg L1, L2, N, and 6awg G
• Combiner output panel to each landing spot (backfeed breaker or manual transfer switch) - 1awg L1, L2 (NO N or G since already present in all landing spots)

Then, assuming I run with the above plan, I think either will "work" functionally, i.e. lights will light up either way. So is there a way I can then test for potential issues, crossing current, or anything else that might become an issue that will go unnoticed? At least that way, I can test, verify it's good, or identify an issue and shut it off and run the extra N.

 

[timselectric]

Every situation is different. I can't comment on yours because I don't know enough about it.
But for rays (video link above ) setup. I would have ran a single conduit with L1 and L2 input, L1 and L2 output, a single neutral, and ground.

 

[atlsolar]

@timselectric - I guess one large difference between Ray's and mine is that the inverter is being fed from Panel 1, and then is backfeeding Panel 2,3, and 4 (new subpanel). Panel 1, 2, and 3 all are directly fed from the same 2x 200a service lines that are spliced into 3 lines before entering any panel.

Not sure, but maybe a couple photos will help???

Trough pre-panels where 2x 200a become 3.


Current panels 1,2,3 + illustration of where MTS will go on it's way to the third 200a panel in my shop. 4th/new subpanel will go to the left of panel 1.


So turning panel 1 into splitter panel, move ALL existing breakers in panel 1 into new subpanel (panel 4). Panel 1 feeds each inverter via 100a breakers. Feeds new subpanel via 125a breaker.

then inverter combiner output panel feeding panel 2 (backfeed breaker/interlock kit), 3 (MTS), and 4 (backfeed breaker/interlock kit)

 

[timselectric]

@timselectric - I guess one large difference between Ray's and mine is that the inverter is being fed from Panel 1, and then is backfeeding Panel 2,3, and 4 (new subpanel). Panel 1, 2, and 3 all are directly fed from the same 2x 200a service lines that are spliced into 3 lines before entering any panel.

Not sure, but maybe a couple photos will help???

Trough pre-panels where 2x 200a become 3.
View attachment 276913

Current panels 1,2,3 + illustration of where MTS will go on it's way to the third 200a panel in my shop. 4th/new subpanel will go to the left of panel 1.
View attachment 276914

So turning panel 1 into splitter panel, move ALL existing breakers in panel 1 into new subpanel (panel 4). Panel 1 feeds each inverter via 100a breakers. Feeds new subpanel via 125a breaker.

then inverter combiner output panel feeding panel 2 (backfeed breaker/interlock kit), 3 (MTS), and 4 (backfeed breaker/interlock kit)

Pictures help a lot.
The trough makes everything easier.
I would run a single 200a (3/0 copper) neutral from panel #1 into the trough. Then run a single (appropriately sized) neutral for each inverter and sub panel.
You can use a terminal block or polaris type connector in the trough to connect them together.

 

[atlsolar]

@timselectric - Thanks for the reply. I think I may not 100% understand plus have a couple problems that might be in the way.

There is NO service disconnect at my house prior to the three panels being directly fed. Before realizing this, I had planned to use polaris splitters pre-panel and split off to an inverter feeder panel. Now that I know, I'm removing all the breakers, turning Panel 1 into the "splitter", and moving the breakers to a new subpanel. Knowing this, I don't have an easy way to tap the existing 4/0 service neutral to run the 3/0 neutral into the trough, do I? I don't think I can simply turn off the three main breakers fed from the same spliced service and pop out the neutral to work on it, can I?

I'm also not understanding how what you are saying is different than what exists today. I'm sure it is, I just am missing something. Today, there is the main service neutral that is spliced and goes to the existing 3 panels. You mentioned running a single neutral for each inverter and SUB panel, so you are not suggesting to touch the existing spliced neutral to the two other existing panels fed directly from the service lines, or are you? If so, how is that different from today's setup + my suggested neutrals run from Panel 1 to the inverters and new sub panel?

If I did what you're saying - maybe just use a 5-port polaris inside of panel 1 - 1 port coming from spliced service, 1 to panel 1, 1 to inverter 1, 1 to inverter 2, 1 to new sub panel. Then, I'd still run the neutral from the (BOTH?) inverter bus to the combiner panel, but NOT run the neutral back to ANY of the 3 panels (2,3,and sub-4)? Is that correct? I also would not need to move individual neutrals to the sub-panel when I move the branch circuits/hots?

Thanks for the help. I've just been spinning in circles on this issue for a week now. I think it's the last big hurdle I have left to figure out in what's definitely not your standard house setup.

 

[timselectric]

@timselectric - Thanks for the reply. I think I may not 100% understand plus have a couple problems that might be in the way.

There is NO service disconnect at my house prior to the three panels being directly fed. Before realizing this, I had planned to use polaris splitters pre-panel and split off to an inverter feeder panel. Now that I know, I'm removing all the breakers, turning Panel 1 into the "splitter", and moving the breakers to a new subpanel. Knowing this, I don't have an easy way to tap the existing 4/0 service neutral to run the 3/0 neutral into the trough, do I? I don't think I can simply turn off the three main breakers fed from the same spliced service and pop out the neutral to work on it, can I?

I'm also not understanding how what you are saying is different than what exists today. I'm sure it is, I just am missing something. Today, there is the main service neutral that is spliced and goes to the existing 3 panels. You mentioned running a single neutral for each inverter and SUB panel, so you are not suggesting to touch the existing spliced neutral to the two other existing panels fed directly from the service lines, or are you? If so, how is that different from today's setup + my suggested neutrals run from Panel 1 to the inverters and new sub panel?

If I did what you're saying - maybe just use a 5-port polaris inside of panel 1 - 1 port coming from spliced service, 1 to panel 1, 1 to inverter 1, 1 to inverter 2, 1 to new sub panel. Then, I'd still run the neutral from the (BOTH?) inverter bus to the combiner panel, but NOT run the neutral back to ANY of the 3 panels (2,3,and sub-4)? Is that correct? I also would not need to move individual neutrals to the sub-panel when I move the branch circuits/hots?

Thanks for the help. I've just been spinning in circles on this issue for a week now. I think it's the last big hurdle I have left to figure out in what's definitely not your standard house setup.

OK, I understand a little more about your existing setup.
Turning panel #1 into a "splitter" for the inputs of the inverters is a good idea. Since there's no other disconnect before the trough.
But you can still tap the neutral in the trough, with a single neutral to each inverter. Just like the existing panels are done.
If you turn off all loads (main breaker in each panel), the neutral is safe to work with. The neutral is only dangerous when it's carrying current and you try to disconnect it.

 

[timselectric]

I also would not need to move individual neutrals to the sub-panel when I move the branch circuits/hots?

Technically no, but I still would.
You don't want to turn off the main breaker in a panel and think it's safe.
Only to find out that there are neutrals in there that are still carrying current through that panel.

 

[atlsolar]

@timselectric, Thanks again for the replies. Can you help me understand why tapping the neutral pre-panel is better or any different than just using the panel's neutral bus bar to lead to each inverter and the new subpanel? I understand what you're suggesting, and it makes sense on HOW, just not understanding why it's functionally different, and the bus bar version is much easier to accomplish.

Then, assuming I have that in place, I should NOT run a neutral from the inverter combiner panel back to panels 2, 3, and 4, right, as the same neutral already exists in each, right?

Similar to the neutral question - If I'm following the neutral plan with my ground too, running one line to the new subpanel and one to each inverter, then inverter to combiner output panel, do I need a ground from the output panel back to the places I'm backfeeding (panel 2,3, and new sub panel)? All already share a ground, right, but I'm thinking more is better with the ground - essentially, whichever is the shortest/easiest/fastest path, I should provide. So with that, I SHOULD include a ground wire from the inverter combiner output panel back to the panels 2,3,and 4 - even if I'm NOT adding a neutral.

All of that sound correct?

 

[timselectric]

@timselectric, Thanks again for the replies. Can you help me understand why tapping the neutral pre-panel is better or any different than just using the panel's neutral bus bar to lead to each inverter and the new subpanel? I understand what you're suggesting, and it makes sense on HOW, just not understanding why it's functionally different, and the bus bar version is much easier to accomplish.

Then, assuming I have that in place, I should NOT run a neutral from the inverter combiner panel back to panels 2, 3, and 4, right, as the same neutral already exists in each, right?

Similar to the neutral question - If I'm following the neutral plan with my ground too, running one line to the new subpanel and one to each inverter, then inverter to combiner output panel, do I need a ground from the output panel back to the places I'm backfeeding (panel 2,3, and new sub panel)? All already share a ground, right, but I'm thinking more is better with the ground - essentially, whichever is the shortest/easiest/fastest path, I should provide. So with that, I SHOULD include a ground wire from the inverter combiner output panel back to the panels 2,3,and 4 - even if I'm NOT adding a neutral.

All of that sound correct?

Single neutral and single ground to each panel or inverter.
The main objective is to not run the neutral for one item through another item. Unless you are also running the line conductors through it. And to avoid loops, or parallel paths.
For the same reason as moving the load circuit neutrals. You don't want to turn off the main breaker in panel #1, and assume that it's safe. Only to get a big surprise because the neutral currents from the inverters are still flowing through that panel.

 

[atlsolar]

@timselectric - I'm getting ready to work on everything this weekend, and I want to run a couple details by you.

Like you suggested, I'll be splitting the neutral pre-panel, using a 5-port polaris connector. Service neutral in, then out to the splitter panel, 2x inverters, and new sub panel. Likely will do this in the trough below the panel.

For the ground, can I run this from the existing panel/future-"splitter panel's" ground bus bar to the new panel, or do I also need to tap the ground pre-panel like the neutral? tapping from the existing panel's bus will be much easier and more accessible, just using #6 from one panel to the next.

Knowing I only want one N-G bonding in the main panel never in another panel down the line, my scenario is no different, regardless of how the neutral is being tapped, regardless of the fact that the new panel won't have it's neutral tied to the main panel's neutral bus where the Neutral-ground bond is?

I think I know the answers on all of this, but just want to triple check I'm not assuming something incorrectly.

Thanks.

 

[timselectric]

For the ground, can I run this from the existing panel/future-"splitter panel's" ground bus bar to the new panel, or do I also need to tap the ground pre-panel like the neutral? tapping from the existing panel's bus will be much easier and more accessible, just using #6 from one panel to the next.

You can do that. But I would prefer to tap the ground in the trough. For the same reasons as anything else. If something goes wrong in one panel, I wouldn't want to lose my grounding system for everything else.

Knowing I only want one N-G bonding in the main panel never in another panel down the line, my scenario is no different, regardless of how the neutral is being tapped, regardless of the fact that the new panel won't have it's neutral tied to the main panel's neutral bus where the Neutral-ground bond is?

I'm not sure if I follow you.
All neutrals are connected together.