Using two single phase inverters instead of a split phase inverter

[harrisonpatm]

This query is relating to typical North American split-phase house wiring.

I understand how a split phase inverter connects to both legs and the neutral in a house's panel. I understand that a simple single-phase inverter can't be used for both legs. If you try to wire the single hot line into both legs of the panel, then you risk current doubling up on the neutral and quickly exceeding your house wire ampacity. So that's not what I'm suggesting.

Instead, I'm thinking what would happen if you use two different single phase inverters, one on each leg. Then you would be able to energize each leg via a separate power source, and each inverter uses a different neutral. The disadvantage of course, is that this would not work for any of your split-phase 220-240v circuits in your house, since the two inverters have no way to communicate and offset their phases by the correct degree to energize 220-240v appliances.

Why bother? Because a decent split phase inverter in the 6-8000w range that I would need for my whole house needs can run $1200-2000 USD. While I could get 3000w/4500peak single phase inverters for $150-200 and stack them in parallel. Less than half the cost. Plus you could get a third for redundancy and backup, still way under the budget of a split-phase inverter. Also, I only have a single 220v appliance in my house, the clothes dryer, which is far from an essential appliance; that's what a clothesline is for.

Anyway, is there something electrically wrong with this plan? Safety wise. I'm mostly concerned as to whether the neutral lines on either leg will still overlap current. It's not a concern at all, for now, that this doesn't work with 220-240 circuits.

 

[OffGridInTheCity]

What I've seen (in youtube hacks) is that if you only have 120v going to a distribution panel, you can run 1 wire to the neutral and the other to both sides of the panel. I gather it's not recommended to do this but it will energize all the breaker positions at 120v so you can place more breakers / distribute more circuits. But of course you can't do 240v circuits.

In other words - "1 x 120v source -> 1 x distribution panel" seems to be the rule. If you have more than 1 x 120v sources, then you'd need more than 1 x distribution panels.

 

[harrisonpatm]

What I've seen is that if you only have 120v going to a distribution panel, you can run 1 wire to the neutral and the other to both sides of the panel.

I was told by an electrician that if I did this, it could double the current on the neutral if I load both legs at the same time. Are you suggesting that it doesn't matter? Not an attack, I'm actually asking, because if you're right, then yes, this would absolutely be better that 2 inverters in parallel, I would love to do it, if its safe.

I'm pretty sure a scheme to have 2 x 120v sources -> 1 x distribution panel is not only against code but a really bad idea

What would happen if I did?

 

[OffGridInTheCity]

You caught me mid-editing - e.g. "... but a really bad idea...". This IS what I think but I don't mean my posts to come across judgmental and especially since I'm not an electrician at all.... which is why I removed that phrase. I'm guessing the proper advice is 1 dist panel per 120v source - and perhaps only use 1/2 the panel slots for breakers/circuits.

As long as the total current is within spec for the panel - I don't think panel overload would be an issue weather you do 1/2 panel or both sides.

I'll be interested to follow the discussion

 

[harrisonpatm]

As long as the total current is within spec for the panel - I don't think panel overload would be an issue weather you do 1/2 panel or both sides.

Hm. So single phase inverter, powering both legs on the same phase, plus the neutral. At that point, I don't think the issue would be panel overload, but rather inverter overwork. Now in that case I'd need 1 inverter capable of 6000w or so. Still might be worth in, economically, as those are still several hundred less than a split-phase inverter.

 

[OffGridInTheCity]

I'd agree that 6000w @ 48v is the upper end for mainstream inverters at 120v.

Low to Middle examples
- I'd peg "Reliable" inverters at the lower end $ (high frequency) - https://www.ebay.com/itm/203041064898 - price point. I've had 3 of these and they all worked OK.

- MPP Solar (https://watts247.com/product/hybrid-lv6048-split-phase-120v-240v/), EG4 (https://signaturesolar.com/eg4-6k-off-grid-inverter-6000ex-48hv/) , GroWatt etc in the middle price point ~ $1300

But at $1300 you can also get 240v/120v as well....
Such as SGP - https://www.amazon.com/Inverter-Charger-Frequency-Output-SUNGOLDPOWER/dp/B0BZCKWMY6/ref=sr_1_1_sspa

Are you thinking all-in-one or just inverter?

 

[harrisonpatm]

Probably just inverter. I've used 1 Reliable as well.

But back to issue, would you still say that the electrician misled me? Single phase inverter hot split to both legs, one neutral, and I should be able to safely power the whole panel, minus any 240v circuits? Because I can work with that, I just want to make sure its safe.

 

[OffGridInTheCity]

Probably just inverter. I've used 1 Reliable as well.

But back to issue, would you still say that the electrician misled me?

I don't know what you're electrician said/ment - so I can't comment on that.

Single phase inverter hot split to both legs, one neutral, and I should be able to safely power the whole panel, minus any 240v circuits? Because I can work with that, I just want to make sure its safe.

I have my shed sub-panel temporarily wired with 120v -> both sides of the panel. Let me post this pic showing the wiring to help clarify how I view this....

In this pic, the Black and Red and White would be split-phase 240v/120v with the White as neutral. The Red is not energized right now... so I used the Black (energized 120v leg) and White neutral for my 120v and jumpered the left bus of the panel to the right bus with that small Red jumper.

This powers all the breaker slots with 120v. One item that I don't think is to code is that I have the Black+Red-jumper wire mashed in the same screw down connector. Note the grounds are twist tie connected (instead of a terminal strip) which is the correct spirit of safety but I'm not sure if a ground terminal block is required - it may be. In either case, I know for sure that one doesn't bond neutral and ground at a sub-panel when they are bonded at the main panel.

As far as max power - the wire is 6AWG so 50a @ 120v is the max wire wise and the panel is 125a rated - https://www.amazon.com/gp/product/B00002N7KK/ref=ppx_yo_dt_b_search_asin_title - which I understand means you can do up to 80% of that rating = 100a total.

Hope this clarifies....

 

[harrisonpatm]

This clarifies a lot, thank you. It's what I have pictured in my head, so its nice to see its also what you're doing.

 

[OffGridInTheCity]

This clarifies a lot, thank you. It's what I have pictured in my head, so its nice to see its also what you're doing.

Yea well - remember that I just blew up a brand new inverter thru a dumb move. So not sure how much assurance one should take from my shed wiring

I found this youtube on 2 x 120v sources -> 1 distribution panel. One interesting. tidbit.... If you measure the voltage between the 2 sources 'hot legs' you can see 0v if the phasing is in sync vs voltage if the phasing is out of sync and clarifies why you don't want to try to power any 240v circuits/loads.

View: https://youtu.be/WZS7-0KPYwU

 

[harrisonpatm]

Ah,thanks for finding that video! I was googling electronics forums, hadn't made my way to youtube yet. So he did exactly what I had imagined. It was really interesting to see the phases moving between the two hots. Also it seems like my plan of 2ea 3000W inverters will work, if that's the way I want to go.

To code? I'm sure it won't be! Anyway, I had planned on having an electrician do the housework with the generator inlet and subpanel, so at least that part will be to code. The cord to connect the inverters to the inlet, on the other hand, was certainly going to be homemade!

 

[cak]

I have my shed sub-panel temporarily wired with 120v -> both sides of the panel. Let me post this pic showing the wiring to help clarify how I view this....

This is how my farmhouse has been functioning for over 20 years and it is perfectly safe with correctly sized breakers.

The breaker back on your main panel needs to be sized for that wire and so in any overload situation that breaker would trigger. With this method you are basically merging the two buss bars into one 120v bus bar. Loads on each bus bar will add onto each other but equally on the hot and neutral so breaker will protect as long as there is only that one power source for the breaker box. 240 appliances would not be happy as others have warned.

Now for your situation of trying to parallel two independent 120v inverters and systems there are MANY potential risks and concerns of integrating them into the same breaker panel. Ideally if the inverters are not communicating with each other and particularly with the cheap inverters that you would likely be using you would want to make sure that you also separate the neutrals into separate buss bars to match with their respective inverter or you will likely get undesired results and possible sparks and damage. I have noticed many cheap inverters like the reliable ones have a readable voltage on the neutral so issues would be highly likely if the two inverters neutrals or hots were crossed. But if you perfectly keep each inverter isolated but do so in the same electrical box everything should work fine. However I doubt it is worth the risks for minimal cost savings and the lower reliability of the cheap inverters.

 

[Korishan]

Not sure if it was mentioned or not (I didn't fully read each post), but there's something that needs to be addressed when connection any AC generating device with another. And that's accurate synchronization. If the phases are not synced, then you will get oscillating voltage.
There are 'hacks' that can be done, but not guaranteed to work.

So basically, if the inverters are not designed to be connected this way, then changes are it won't work, or won't work as expected. Worse case is that one of both inverters are destroyed, or Neutral ends up carrying way too much current (in a 240V situation, Neutral doesn't carry any current unless there is some circuitry that only uses one of the 120V legs)

 

[harrisonpatm]

Can someone explain why this did indeed work in the video linked by OGITC? The youtuber just wired two different , isolated battery backups (which have internal inverters for their ac output) into his home generator inlet. Each power source carried one leg, plus the neutral, meaning the neutral was shared by both inverters.

I'm aware that one success doesn't make for a bill of recommendation. Just trying to understand.

In the house, and it's breaker panel, the neutral is always shared by both legs anyway, right?

 

[Korishan]

Can someone explain why this did indeed work in the video linked by OGITC?

Yup, he says it right towards the beginning:

0:39 mark: View: https://youtu.be/WZS7-0KPYwU?t=39

"The one caveat with this is that we are not going to be able to power our 240V circuit and the most we're going to be able to power without using a transformer that actually steps the voltage up is going to be 120V

and then later on:
1:48 mark: View: https://youtu.be/WZS7-0KPYwU?t=108

"This power bank here is essentially going to power every other leg in our 240V panel, and this one will serve every other leg"

And at this mark he shows exactly what I was referring to with oscillating voltage
4:08 mark: View: https://youtu.be/WZS7-0KPYwU?t=248

As you can see when he connects across the two hot legs, the voltage starts at ~195VAC, then is slowly goes up to ~240VAC, then it starts to drop all the way down to almost zero VAC, and then starts to climb again

He refers to it as "Weird voltage". But what is happening is that the two sine waves are not in sync with each other. And as time passes, the waves will become 180* of each other, which cancels out and becomes zero volts. Then as they swing back to 0* of each other, 240VAC will show up again.
The reason this is happening is that if he were to show the frequency of each of the inverters, you'd see that they are
1) not identical. One of them could be 58Hz, and the other 60Hz, or even 59Hz and 62Hz, or something of that sort
2) not in sync and drifting back and forth, because of #1

This is the same phenomenon that occurs with sound canceling headphones. The mic picks up the audio on one side, and then outputs it on the speaker side 180* off so that the sound is canceled out

 

[harrisonpatm]

Oh, that's fine then. I don't need 240, as I mentioned in the first post. But two inverters should have no problem with just powering each leg at 120 only.

Anyway, is there something electrically wrong with this plan? Safety wise. I'm mostly concerned as to whether the neutral lines on either leg will still overlap current. It's not a concern at all, for now, that this doesn't work with 220-240 circuits.

So I don't care if it doesn't work with 240 circuits.

 

[Korishan]

Yeah, it 'can' work. Neutral won't cross carry, especially if Neutral and Ground are bonded in the panel. But you just have to make absolutely sure that there are zero circuits that would require 240V. Otherwise there could be damaged devices.

Unless you already have the equipment, it would be more feasible, safe, and less hassle to just get an inverter that can do 240V output and use a transformer to get 2 hot legs and neutral. The transformers are relatively inexpensive. The only ones that go up in price are the smart transformers that auto shift central point to maintain balance. Or just make your own with a MOT with a 1:1 coil with a center tap on the output. But this is getting into the unsafe territory again

 

[harrisonpatm]

It's all up in the air, there are lots of pros and cons both ways. One of the biggest variables is that in my area, in my house, I will never actually be able to go off grid electric; the power company won't allow that. Another huge variable is that my house electric usage is tiny compared to most, way under the national average. It's never going to make economic sense to power the house from solar and a powerwall.

My current plan, a year or so down the road, is to have an electrician safely install a subpanel/MTS feeding every circuit in my house except for my one 240v circuit, which is the clothes dryer. This way, I should be able to safely power a majority of my household off the powerwall that I'm still building, while keeping the dryer on grid use. I'm simply exploring options for how to get 120v output from my powerwall. This is just one idea, it's certainly not the only option.

Or just make your own with a MOT with a 1:1 coil with a center tap on the output. But this is getting into the unsafe territory again

Eh, I like a lot of DIY, but even that sounds a little sketchy to me!

Another benefit to two small inverters that i considered is less idle draw. I do have one of them already, measuring less than 20w idle current. That's quite a bit less than one large inverter.

 

[OffGridInTheCity]

except for my one 240v circuit, which is the clothes dryer.

There may be a 120v dryer solution - https://www.appliancesconnection.com/blog/best-120-volt-compact-electric-dryers - so you could be 100% at 120v.

 

[harrisonpatm]

Certainly, thought my proposed system would actually still have the dryer 240v on grid.

Are the 120v clothes dryers any good? I can't imagine they would be. Using heat to remove moisture from clothes is a matter of energy, one way or another. If a 5000w dryer can do it in an hour, a 120v 15a 1800w dryer will take 3x longer to do the same amount. It's not to say that there's potentially a super efficient and effective 120v dryer out there, but if you really need it done, there's a clothesline.

Or, if I really wanted the 240v dryer powered from the powerwall, I could get a dedicated 48v-240v inverter run to the dryer directly. But at that point even I would admit I'm overcomplicating things.