Two inverters powering independent service panels sharing a battery

[drachjc]

Hello,

I have been reading this site, watching YouTube videos and thinking and researching about my plan to add back-up grid down power to my existing setup for a few months. The amount of information I have learned from everyone on here has been extremely helpful!

I currently have an 8.82kW array going through a SolarEdge inverter (installed 4 years ago by a well known national solar installer). My house has 2 separate (side by side) 200A service panels. I have read many threads about dealing with two main panels and long story short I think the way I want to go is to power each panel totally independently with different inverters using interlocks on each panel. This is simple and straight forward and avoids the whole multiple ground-neutral bonds situation.

Because of my existing solar, I was looking toward Sol-Ark for the AC coupling capability to easily integrate things. I reached out to Sol-Ark asking if 2 Sol-Arks powering separate panels could share the same battery, and they provided a single sentence response saying unless both inverters are connected they cannot use the same battery. There was no elaboration if this meant they just needed to be connected via the communication cables, or actually to the same panel. Unless I missed it I didn't see anything in their documentation mentioning this type of setup.

I wouldn't have minded a more technical explanation, especially as I was considering buying two of them, but since then I decided that a more cost effective option might be to purchase a single Sol-Ark, use that for my AC coupling/battery charging, and then a second much cheaper inverter (such as a Sungold 6000W LFP) that would power the other panel. The only thing shared between them would be the bank of 6 EG4 48V 100Ah batteries. I wouldn't even bother hooking up AC input to the other inverter and just let the Sol-Ark handle the battery charge. Furthermore since my intention is to only ever use this in a grid outage situation I could just leave the second inverter powered off until it is needed.

I haven't found anything to lead me to believe that this type of setup would not be possible, however most people with multiple inverters use them in parallel to power the same panel so this is somewhat different. I have already purchased the batteries, and though I could set them up as 2 batteries I would prefer to use them as one battery.

If anyone has any thoughts as to whether or not this would work or have a better suggestion I would love to hear it. Thanks!

 

[sunshine_eggo]

Hello,

I have been reading this site, watching YouTube videos and thinking and researching about my plan to add back-up grid down power to my existing setup for a few months. The amount of information I have learned from everyone on here has been extremely helpful!

I currently have an 8.82kW array going through a SolarEdge inverter (installed 4 years ago by a well known national solar installer). My house has 2 separate (side by side) 200A service panels. I have read many threads about dealing with two main panels and long story short I think the way I want to go is to power each panel totally independently with different inverters using interlocks on each panel. This is simple and straight forward and avoids the whole multiple ground-neutral bonds situation.

Are you sure it does? Will you drop a second ground rod for the second panel? Will you then bond your ground rods together as required by NEC, or is one panel ground bonded to the other?

Concerning grounding, you want everything above ground tied together and everything below ground tied together with only ONE connection between above and below.

As an overall concept, I would argue there's little difference in what you propose compared to two parallel 120V inverters each supplying one leg of 120/240VAC split phase power. Sure, they're coordinating with each other, but it's very possible that they will have substantially different load requirements/utilization like your two completely separate panels. My Victron's work this way acceptably by providing a N-G bond in only one inverter while the other has the N-G bond relay open.

 

[drachjc]

Are you sure it does? Will you drop a second ground rod for the second panel? Will you then bond your ground rods together as required by NEC, or is one panel ground bonded to the other?

Thanks for your reply!

Am I sure about the questions you asked? I am not. However, maybe I am being unclear or left out too many details in my attempt to be concise. My house was built (in 2018) with 2 separate 200A panels being fed from two separate cables coming from opposite sides of the same meter (located outside on the other side of the wall). There exists no connection on the inside of the house between either panel that I can tell. I have not examined in detail inside the panels, as I am right now at a more theoretical/planning phase. My assumption based on other people's threads with similar situation is that these panels are both considered "main" panels and thus both have the neutral ground bond. I will take a look and verify. I am not an electrician, but my understanding is that if both panels shared the same input (e.g. both panels were wired to the same generator plug, via interlocked breakers of course), this would be an issue, like having a sub panel with the ground screw installed. I assume (possibly incorrectly, despite the myriad of inspection stickers on my panels) that the panels were installed properly as respect to the grounding rods or whatever is required for this type of set up. Everyone in my neighborhood is like this so while it may not be common, it's not uncommon either.

It seems that most people's recommended solution would be an external transfer switch between the meter and the panels. It might be "the best" but requires a much greater expense, power company involvement, etc.

Does that explain what I am asking better? Maybe a better analogy would be to consider each panel as a separate "building" as both get their power upstream directly from the power company. If I want to shut off the entire power to my house I have to flip the 200A breaker at the top of each panel. There is no external 400A breaker or disconnect beyond there. They each are powered from the meter independently.

My idea about having 2 inverters would be like having 2 separate generator inputs, one for each panel (with no link between them not even the same conduit) and powering them from 2 separate generators (i.e. no Y cable).

As an overall concept, I would argue there's little difference in what you propose compared to two parallel 120V inverters each supplying one leg of 120/240VAC split phase power. Sure, they're coordinating with each other, but it's very possible that they will have substantially different load requirements/utilization like your two completely separate panels.

This is basically why I think it would work fine. The inverters also don't need to be synchronized since they would be powering completely separate panels. So if they were different brand inverters I don't think that would matter either.

Thanks again for taking the time to reply.

 

[drachjc]

Because I like analogies... running two different inverters from the same battery would be like two different generators sharing the same fuel tank. That's how I'm thinking about it. But I didn't know if there was a technical reason or safety issue with running two independent inverters with totally independent power outputs from the same battery.

 

[sunshine_eggo]

I think you've explained it, but I think details are omitted/unknown.

Both panels must be earthed. given that you never want multiple paths to earth, and you don't want multiple separate earths, the two panels should be at least G bonded either from panel to panel, or if they have separate ground rods, the ground rods should be bonded, so the panels are not entirely separate. Even if you use two separate inverters, you may have multiple N-G bonds.

The scenario I presented showed that there is a N-G bond in only one inverter by design.

I'm not convinced that your scenario would NOT have multiple N-G bonds due to bonding between the panels.

 

[drachjc]

Thanks again for replying. I will investigate the panels closer and see if I can figure out those details.

 

[drachjc]

I pulled the covers off the panels. I confirmed as you said would be the case that the grounds are connected between the panels. And I also confirmed both panels N-G bond.

So, as long as there is no N-G at either inverter there should be no problem with the setup I described? The Sungold inverter I mentioned has some information in the manual for disabling the N-G bond and mentions doing that if you are using it with a main panel. The Sol-Ark documentation merely states "ground MUST be bonded to Neutral ONLY ONCE in the circuit" so I am not sure if that means I can use it or not for this.

 

[sunshine_eggo]

I pulled the covers off the panels. I confirmed as you said would be the case that the grounds are connected between the panels. And I also confirmed both panels N-G bond.

So, as long as there is no N-G at either inverter there should be no problem with the setup I described? The Sungold inverter I mentioned has some information in the manual for disabling the N-G bond and mentions doing that if you are using it with a main panel. The Sol-Ark documentation merely states "ground MUST be bonded to Neutral ONLY ONCE in the circuit" so I am not sure if that means I can use it or not for this.

I believe that is the case. The only other concern that I would have is if these are split phase inverters, is there any issue with having them effectively share a neutral due to the N-G bond.

 

[iamrich]

I know this was a while back, but what did you end up doing? I am getting close to hooking up my inverters (x2 LVX6048) and I think our setups are similar. My inverters each power a sub panel and are complete independent of each other. I was planning on a similar battery setup (~30kwh) and would like both inverters to share a single battery setup. I am not opposed to running two battery banks, but it seems like having one bigger bank would be more versatile.

 

[Robbert]

Looking into that setup as well.
2 mpp solar inverters, doing their own part of the house.
Don't want parallel, because i need a somewhat redundant setup. Now both have their own battery bank, but it would be much better if they could share one large bank.

Curious if in such a setup there will be a 'fight' between the two different charge controllers, even if they have identical settings and same cable lengths.

I think it should be no issue, but don't dare to hook it up like this before i am sure it good.

Did you already attached your inverters to the same battery?

 

[RCinFLA]

Looking into that setup as well.
2 mpp solar inverters, doing their own part of the house.
Don't want parallel, because i need a somewhat redundant setup. Now both have their own battery bank, but it would be much better if they could share one large bank.

Curious if in such a setup there will be a 'fight' between the two different charge controllers, even if they have identical settings and same cable lengths.

I think it should be no issue, but don't dare to hook it up like this before i am sure it good.

Did you already attached your inverters to the same battery?

There is always some calibration difference on charging voltages and current between inverters.

I have fours inverters sharing a common battery. Particularly on low frequency inverters, an inverter may force an exact, as they see it, float voltage setting causing a little 'trading spit' between inverters charge current with slight float voltage calibration differences.

The calibration error usually doesn't cause too much of an issue if it is not too great. I had a problem with my old Trace/Xantrex inverters with their lead-acid charge voltage temp compensation. The inverters do temp comp adjustment at discrete temp steps and one inverter would jump to next temp step increment before another. I ended up removing remote battery temp sensors to avoid too much float charge current exchange between inverters.

 

[sunshine_eggo]

There is always some calibration difference on charging voltages and current between inverters.

I have fours inverters sharing a common battery. Particularly on low frequency inverters, an inverter may force an exact, as they see it, float voltage setting causing a little 'trading spit' between inverters charge current with slight float voltage calibration differences.

The calibration error usually doesn't cause too much of an issue if it is not too great. I had a problem with my old Trace/Xantrex inverters with their lead-acid charge voltage temp compensation. The inverters do temp comp adjustment at discrete temp steps and one inverter would jump to next temp step increment before another. I ended up removing remote battery temp sensors to avoid too much float charge current exchange between inverters.

My neighbor sees this on his FM-80 MPPT at float. Fortunately, they can be calibrated within 0.1V increments, but even that results in a bit of "trading spit".

 

[Robbert]

I contacted MPP-Solar and according to them it is fine to share one battery with two inverters.
Just ordered new cells to build a second battery that will be paralleled with the one I just finished building. If that is ready, I will parallel the two inverters to this 'big' battery.

 

[iamrich]

Did you already attached your inverters to the same battery?

Only one so far. I am play testing it to see if I can tell I am on an inverter instead of grid. So far, I can't tell. Will probably hook up the other inverter soon. I don't have my panels installed yet, but having a 30kwh battery backup is nice