Neutral problems

[rockhardhoosier]

I have 4 sunny island 6048 inverters with 4 Fla 540 amp hr batteries wired in parallel. I have a manual transfer switch . Everything had been working fine then I had to switch from solar back to the grid for a few hours. When I switched back to the solar my GFI outlets in the house tripped and anything with a motor, like my fireplace fan motor ninja airfryer/grill , my microwave water heater,all was not working correctly. But when you turn back to the grid it is fine. Well then had happened before but it was a loose neutral. Tightened it up and everything worked fine. This time there are no loose neutral grounds or anything. I even had an electrician come and double check he found nothing. He checked the hertz he checked from the neutral and the grounds to each line and we had 119-121v and when checked across both lines we had arpund 240-243v so that was good. So what could this be? The electrician is at a stand still.

 

[OffGridForGood]

I have 4 sunny island 6048 inverters

24kW system? @Hedges is the resident SI SMA guru...he will be along shortly I expect...

I have a manual transfer switch

Pic? - does your MTS switch the N or do you use a common N?

When I switched back to the solar my GFI outlets in the house tripped

How is the N created and where is it bonded to the Earth-Ground?
Since everything works on Grid, but doesn't on Inverter, seems to point to N-G issue with the Inverter set up. But I am not familar with how the SMA's handle the whole N-G set up, Hedges will know.

 

[rockhardhoosier]

I might also add everything has worked fine for almost a year,besides occasionally having to tighten up a wire. Yes right at 24kw system. I've used the transfer switch many times without an issue. The electrician installed all the neutral and grounds and bonded them. When he showed up he was thinking it was also a loose wire. But wasnt.

 

[Hedges]

I'm not sure that loose/lost neutral would trip GFCI. Maybe one that checks voltages between L/N/G.

If there is a second connection between L and G, maybe resistive or through a capacitor, then quality of N connection could divert more return current into G. But GFCI only compare L current to N current going through them, so would only expect the GFCI outlet which had that "objectionable current" downstream to trip.

A "residual current" clamp ammeter could check for imbalance. EMI filters in electronics tend to put a fraction of a mA from L into G, and large quantity can cause trip.

"Wired in parallel", do you mean inverters are series parallel 2s2p, or actually parallel 2p2s, for 120/240V split-phase? Actually parallel would be for 120V only in US, or 220V only in Europe.

Is transfer switch for L1/L2 only, or does it also switch N? The way mine is set up I only switch L1 and L2, and the N current still gets to travel in same conduit. Wired differently I could see switching N so its current doesn't split off and take a different path.

But switching N from one path to another ought not cause GFCI to trip, at least not in either position. The act of switching N, if it opens first, could make things unhappy - 240V across two 120V loads in series, not dividing voltage equally. But L and N would still have balanced currents, both for GFCI outlets and GFCI breakers.

As off-grid says about considering what is switched, and where bonded.

Do you have Sunny Boys, are they switched, or always on Sunny Island? Any backfeeding inverter must have 5 seconds off after disconnecting from one source before connecting to other.


In each Sunny Island, a relay opens connection between AC1 load and AC2 grid. Current through AC2 is monitored.
N and G are hard connected between AC1 and AC2, no measurement. In fact, use a single wire for each of N and G, maybe the AC1 lug but it doesn't matter.

 

[rockhardhoosier]

I will show this to my electrician as I understand very little of you reply. As I am not an electrician. Also I meant the batteries are wired in parallel. I have the system setup as 1 master and 3 slaves. With the master and slave 2 is on line 1 and slave1 and 3 are on line 2. No sunny Boys. I use midnite charge controllers. But not going back to the GFI outlets tripping. Why would the fans not work properly on the fireplace, convection microwave
and the airfryer. The water heater instantly throws and error code. Might be stupid question ,

 

[12VoltInstalls]

When I switched back to the solar my GFI outlets in the house tripped and anything with a motor, like my fireplace fan motor ninja airfryer/grill , my microwave water heater,all was not working correctly

Describe “not working correctly,” please.

That a) all the GFCI’s tripped, b) “anything with a motor” not working right, and c) you just switched from grid to inverter I would suspect you have a leg imbalance going on - even if that is due to a ‘surge’ condition from one of those motors or the water heater.

I’m not saying that’s it but exploring that would eliminate it from the flowchart of troubleshooting.

Fwiw anecdote:
When I use my larger tablesaw I need to run the generator. Too much surge for a 2000W inverter.
If I am also plugged to my aio for battery charging, when I turn on the tablesaw the GFCI on the generator and the one in the house on the generator feed both trip.

That might be circumstantial in your situation.

 

[rockhardhoosier]

Describe “not working correctly,” please.

That a) all the GFCI’s tripped, b) “anything with a motor” not working right, and c) you just switched from grid to inverter I would suspect you have a leg imbalance going on - even if that is due to a ‘surge’ condition from one of those motors or the water heater.

I’m not saying that’s it but exploring that would eliminate it from the flowchart of troubleshooting.

Fwiw anecdote:
When I use my larger tablesaw I need to run the generator. Too much surge for a 2000W inverter.
If I am also plugged to my aio for battery charging, when I turn on the tablesaw the GFCI on the generator and the one in the house on the generator feed both trip.

That might be circumstantial in your situation.

Ok so the airfryer has a fan on it. When switching to solar the fan doesn't work but the air fryer does heat up. So I took the Fryer around to all the outlets through my whole house and does the same at each one of them. The GFI doesn't kick all the time. Same on microwave which is a convection oven also. The fan on the fireplace works but gets noisy. We did ck the hertz from the solar system and it's good. I know sounds stupid. But the air Fryer is actually the reason I knew something was going on. I actually thought the airfryer went bad so I bought a new one ,and guess what ,wasn't the airfryer, lol. Now not sure if this might have something to do with it but I did have a anti-islanding code on inverter a couple weeks ago while the generator was running and kicked the breaker on the generator. But has never done it again. Thanks

 

[Hedges]

Air fryer is 120V? Then fan and heater both get same power.
I could see an appliance having 240V circuits work but not 120V.

Air fryer 2-prong or 3-prong grounded plug?
That seems like the simplest place to diagnose. Voltage between various pins, with load and no-load.

Grid and generator - how neutral and N/G bond is handled needs to be considered.

If neutral is switched along with L1/L2, there could be voltage glitches.

Wild frequency (and voltage) if loads were driven one leg by grid, one leg by output of another asynchronous source (generator or inverter)

Maybe test loads hardwired directly to grid, to inverter, to generator. Could be the wires and switching between them is messed up some how.
With the resistive load of fryer, use clamp ammeter (on individual wires where accessible) to see where current flows.

 

[rockhardhoosier]

Air fryer is 120V? Then fan and heater both get same power.
I could see an appliance having 240V circuits work but not 120V.

Air fryer 2-prong or 3-prong grounded plug?
That seems like the simplest place to diagnose. Voltage between various pins, with load and no-load.

Grid and generator - how neutral and N/G bond is handled needs to be considered.

If neutral is switched along with L1/L2, there could be voltage glitches.

Wild frequency (and voltage) if loads were driven one leg by grid, one leg by output of another asynchronous source (generator or inverter)

Maybe test loads hardwired directly to grid, to inverter, to generator. Could be the wires and switching between them is messed up some how.
With the resistive load of fryer, use clamp ammeter (on individual wires where accessible) to see where current flows.

Airfryer is 120 with 3 prong plug..Neutral is not switched along with L1/L2 in transfer switch .

 

[robbob2112]

Just a dumb thought here, fans and motor's not working correctly usually means not a pure sine wave or very bad frequency variance. Not sure what effect this may have on GFCI butI am betting not good.

In general there are two methods of converting DC to AC. Switching fets that turn on/off for various durations that make a stepped square wave. That paired with caps and other ciruits to smooth it and make it pure sine. The other uses a large induction in a tuned tank setup that spits out a pure sine in the end. The second is how victron works and is why it can be much better for inductive loads. The large inductors/torid coils can be costly which is why higher end uses them and cheaper end uses the first method.

My theory would be one of your inverters has flipped out and isn't providing pure sine anymore. One that supplies your kitchen outlets. Have you tried your airfryer on a circuit that is not on the same inverter pair with the kitchen outlets. Another question is if all the fireplace fans are on the same inverter pair. Because of how modern difgital multimeters work they would probably still read correct voltage and frequency even if the waveform is messed up. (High sample rate with averaging)

Easy way to prove/disprove the theory would be to take a small fan and plug into each room of the house and take note if it works in some rooms not others. This can be carting around the air fryer or any other small appliance with a fan built in.

A better way would be to go out to the inverters and check each output with an O'scope to see that the pure sine is present and the frequency is correct. Of course you need a working O'scope to check that out.

 

[Hedges]

Airfryer is 120 with 3 prong plug..Neutral is not switched along with L1/L2 in transfer switch .

Just a dumb thought here, fans and motor's not working correctly usually means not a pure sine wave or very bad frequency variance. Not sure what effect this may have on GFCI butI am betting not good.

...

A better way would be to go out to the inverters and check each output with an O'scope to see that the pure sine is present and the frequency is correct. Of course you need a working O'scope to check that out.

Something has to be wrong with voltage or frequency, as seen at the outlet. Under load, measure it while fryer is plugged into other socket.

 

[rockhardhoosier]

Just a dumb thought here, fans and motor's not working correctly usually means not a pure sine wave or very bad frequency variance. Not sure what effect this may have on GFCI butI am betting not good.

In general there are two methods of converting DC to AC. Switching fets that turn on/off for various durations that make a stepped square wave. That paired with caps and other ciruits to smooth it and make it purse sine. The other uses a large induction in a tuned tank setup that spits out a pure sine in the end. The second is how victron works and is why it can be much better for inductive loads. The large inductors/torid coils can be costly which is why higher end uses them and cheaper end uses the first method.

My theory would be one of your inverters has flipped out and isn't providing pure sine anymore. One that supplies your kitchen outlets. Have you tried your airfryer on a circuit that is not on the same inverter pair with the kitchen outlets. Another question is if all the fireplace fans are on the same inverter pair. Because of how modern difgital multimeters work they would probably still read correct voltage and frequency even if the waveform is messed up. (High sample rate with averaging)

Easy way to prove/disprove the theory would be to take a small fan and plug into each room of the house and take note if it works in some rooms not others. This can be carting around the air fryer or any other small appliance with a fan built in.

A better way would be to go out to the inverters and check each output with an O'scope to see that the pure sine is present and the frequency is correct. Of course you need a working O'scope to check that out.

I did just try it on an outlet that wasn't on the same line and still done same thing.

 

[robbob2112]

I did just try it on an outlet that wasn't on the same line and still done same thing.

Right, but I assume your load center is typical and divides the bus into half which should be basically balanced. On another circuit is not nessassarily the same thing as on the other half. I will make the grand assumption that one pair of your inverters drive half the load and the other pair drives the other half the load. Between them you have 240vac and a neutral/ground between them (bonded only in the main panel).

So look in the main breaker panel and see which is/are the kitchen and which breaker is next to it. My folks had two circuits run to the kitchen outlets, half on a 20 amp breaker and the other half on another 20 amp breaker. I guess pop had trouble with turning on the toaster and coffee pot at the same time and popping a breaker one place he lived.

Are the inverters the type that switch the neutral/ground bond on and off automatically? If one of them was incorrectly providing a bond internally via relay and if it were on it could cause issues. looked at the manual - never mind.

 

[rockhardhoosier]

Right, but I assume your load center is typical and divides the bus into half which should be basically balanced. On another circuit is not nessassarily the same thing as on the other half. I will make the grand assumption that one pair of your inverters drive half the load and the other pair drives the other half the load. Between them you have 240vac and a neutral/ground between them (bonded only in the main panel).

So look in the main breaker panel and see which is/are the kitchen and which breaker is next to it. My folks had two circuits run to the kitchen outlets, half on a 20 amp breaker and the other half on another 20 amp breaker. I guess pop had trouble with turning on the toaster and coffee pot at the same time and popping a breaker one place he lived.

Are the inverters the type that switch the neutral/ground bond on and off automatically? If one of them was incorrectly providing a bond internally via relay and if it were on it could cause issues. looked at the manual - never mind.

Ok ,yes you are correct on how the inverters are paired and there are to for line one and two for line 2. So what I did (could have been right or wrong) my 4 inverters run to what I guess you call a combiner box. I just used a 200amp home main panel box for this. So 2 inverters run to one half of the box(L1) and two to the other half(L2). The power then runs out of this box to the transfer switch.
Each inverter has its on breaker 2 breakers on L1 ,2 breakers L2. So I shut down the 2 breakers that fed the outlets that the airfryer was originally hooked to. Found an outlet that still had power to it and tried the airfryer there. Still same thing. Maybe I'm mistaken but shouldn't that been correct on saying it's doing it on both lines or both halves on the breaker box? Sorry if I sound like and idiot, but I thought that was correct. Let me also add in that combiner box I have a breaker that feeds a smaller 50amp box for my pavilion and it also does the same thing when plugging in the airfryer. I'm kinda on the same page as thinking one of the inverters has flipped out. But with that being said the other inverter providing power to that line would be sufficient in supplying power for the airfryer correct? I don't think the GFI kicking had anything to do with this,I think they kicked because of me shutting power down then turning it back on ,because once I reset the GFI it didn't do it again. Once again ignore my stupidity on this if I'm incorrect.

 

[robbob2112]

First - from here on out I am just spitballing and making semi-uneducated guesses. So think hard before you do any of it.

Ideally you would turn everything in the house off and only run 1 inverter and see if the test air fryer works correctly. By this I mean turn off 3 breakers to the other inverters. If the single inverter is misbehaving repeat the test with another one. If a second one misbehaves I would think that means that the data cables between the inverters that keep everything in sync are suspect.

Have you checked all the data cables to verify they are good and are plugged in properly? If they aren't I think it would result in the combined waveform being out of sync with each other.... assuming they are not in sync I am not sure what the result would be.

Also, have you tried a complete go dark power everything off and do a startup from there? I assume your installer left instructions on the proper order to power things on in?

 

[rockhardhoosier]

Something has to be wrong with voltage or frequency, as seen at the outlet. Under load, measure it while fryer is plugged into other socket.

Hey so sorry been gone for awhile, got back to this now. When I'm turning on the inverters without any load it is going up to 69.9 hertz. 119v. I measured the voltage on the other plug when turning airfryer on and it dropped a little to like 115, but does same on the grid. Seems this probably a hertz problem. What could cause this?

 

[robbob2112]

Have you checked your interconnection data wires? If one of those is loose or damaged the inverters can't sync the frequency correctly.

 

[rockhardhoosier]

Have you checked your interconnection data wires? If one of those is loose or damaged the inverters can't sync the frequency correctly.

Yes none of them look damaged and and the connection feels tight

 

[robbob2112]

Might be worth powering down the system and testing continuity on all the data cables. A cheap ethernet cable tester would assist with this, while you are at it verify pinout.

This would also do a cold boot on the whole system, in case they used windows as the os.

 

[rockhardhoosier]

Might be worth powering down the system and testing continuity on all the data cables. A cheap ethernet cable tester would assist with this, while you are at it verify pinout.

This would also do a cold boot on the whole system, in case they used windows as the os.

I got a tester and all the cables tested out good, still running almost 70 hertz