Any thoughts on this problem? I just can't win

[Markyrocks69]

Ok so I've had my system up and running for a few days. So I just been crossing some t's and dotting some i's. Getting the inverters connected to the internet and getting monitoring going ect.

I'll cut to the chase my one inverter seemed to be performing like 25% of the other one. So yes I have 2x sma sb 7.7 us 40 inverters. Each inverter has same number of strings same # of panels. All panels azimuth is the same angle everything identical.

I had hoped a software update would fix it, no dice. What is happening is when the sun is in the peak position.. the one inverter is basically pumping out the maximum. 7.7 kw the other errors out and sits idle saying it's an issue with the grid, grid fault, overvoltage ect.

So today I came home early from work. Turned the inverter that seems to be performing correctly off to see what the other one did. Amazingly it started pushing out the max 7.7kw almost immediately. That being said it kinda rules out the inverter being bad or any issues associated with it being connected correctly ect.

That narrows it down to having something to do with my grid. Obviously the electrical grid (as it relates to the poco) is fine if the system has no issues keeping 1 inverter happy. The only thing at this point that I can really say maybe the issue is that I have both inverter connected to a subpanel and theres no other loads in the panel (yet). The first thing I'm going to try is to move one of the inverter breakers to the main panel and see if the fact that theres constant loads in there will lower the voltage some and allow it to operate as intended. If for some reason that doesn't work then I'm pretty much at a loss. I know I've seen other setups with more than one inverter and the manual even shows different (mounting) scenarios with more than one so I know this is supported. Anyone have any experience dealing with this type of problem ? Any input would be appreciated.

[Ampster]

Can your inverters be stacked electronically? Are L1 and L2 connected the same phase on each inverter? If it was me I would do the same thing, ie connect to the main panel.Are you getting any clues from the line voltages measured by the inverters?

[ButchDeal]

When the one shuts down what is the AC voltage at the inverter?

[Mike90250]

Sounds like your main AC feed to the pole has too much resistance, and the inverters raise the voltage enough to cause an issue. You need to check grid voltage with good meters, with & w/o inverters on in the daylight. Other neighbors feeding into the grid may be raising the lines some too.

[bcroe]

That happened to me. The pair of 7.5KW inverters took turns knocking the other one
off line, because with both on the line terminals got up to 264VAC. Fix that by getting
the PoCo to lower their voltage, reduce the V rise with bigger wire at your end, or set
the voltage monitors (including line to neutral) to a higher value. Bruce Roe

[Markyrocks69]

Can your inverters be stacked electronically? Are L1 and L2 connected the same phase on each inverter? If it was me I would do the same thing, ie connect to the main panel.Are you getting any clues from the line voltages measured by the inverters?

I'm not exactly sure what you mean by "stacked electronically ". The l1 and l2 should be on the same phase.

[Markyrocks69]

When the one shuts down what is the AC voltage at the inverter?

I had been messing around with it earlier and I was having connection issues and it was reading like 255v or less which I found kinda odd bc supposedly the max is 264v and it still wasn't able to connect.

[Ampster]

I'm not exactly sure what you mean by "stacked electronically ". The l1 and l2 should be on the same phase.

Probably not an issue because these are Grid Tie. Off Grid inverters can be stacked to run in parallel but Grid Tie inverters dont need that because they sync to the grid.

[bcroe]

I had been messing around with it earlier and I was having connection issues and it was reading like 255v or
less which I found kinda odd bc supposedly the max is 264v and it still wasn't able to connect.

At the inverter terminals you will see a voltage when they are off, and a higher voltage
with one running. That increment might double when both run, though you may not see
it after one stops. Bruce Roe

[foo1bar]

I had been messing around with it earlier and I was having connection issues and it was reading like 255v or less which I found kinda odd bc supposedly the max is 264v and it still wasn't able to connect.

7kw at 240V is ~30A.

So, I think the question is - what happens when you try to shove another 30A down that line to the POCO?
Essentially - what's the resulting voltage from: 255V + 30A*(resistance_to_POCO_transformer) ?
If you're outputing 30A and the resistance is higher than about 0.3 ohms, you'll be above the 264V that you've said that's the max for the inverter.
(depending on the inverter you may be able to increase that number)

If this is a DIY install, I'd check what the voltage is at the first accessible spot after the meter. If it's too high at the meter you can talk to your POCO about it - they have a spec that they're supposed to keep it within. It may be a corroded connection between you and their transformer that's causing a higher resistance and higher voltage when you're pushing current into the grid (and lower voltage when consuming). Or it could be they have it set to a higher voltage than it should be for you. (my parent's house was near the substation and the voltage was always on the high end)
If it's not a DIY install, I'd have the installer investigate. They should be able to determine if it's too much resistance from the meter to the inveter. Or if it is too high of voltage at the meter. (or both)

[Mike90250]

Can your inverters be stacked electronically?.....

NO, there is no stacking or sync cable for GT inverters.
Grid Tie inverters nearly always connect to L1 & L2 of the grid. There is no stacking, just wired to the grid

[Markyrocks69]

7kw at 240V is ~30A.

So, I think the question is - what happens when you try to shove another 30A down that line to the POCO?
Essentially - what's the resulting voltage from: 255V + 30A*(resistance_to_POCO_transformer) ?
If you're outputing 30A and the resistance is higher than about 0.3 ohms, you'll be above the 264V that you've said that's the max for the inverter.
(depending on the inverter you may be able to increase that number)

I'm failing to see the point here. The voltage at my house is a hair on the high side. At the main panel its like 250-252 or something normally. I don't remember exactly but I did notice that b4. I realize adding another 30amp to the service entrance conductors will add resistance and force the inverter to increase voltage but it's not even like getting to that point. Maybe I'm looking at this with a simplistic pov, but in my peanut sized brain the order of operations would ge through is 1. Look at the grid voltage ect, 2 connect to the grid sync ect, 3 start producing power. It's not even connecting. It connects initially then once a certain load is reached it kicks off, then attempts to reconnect and obviously doesn't like what it sees and just keeps erroring out with the reconnection fault grid error.

The graph I'm posting isn't entirely representative of a normal day bc around 220pm I shut the other inverter off and it started working correctly but here's some more info

[foo1bar]

I'm failing to see the point here. The voltage at my house is a hair on the high side. At the main panel its like 250-252 or something normally. I don't remember exactly but I did notice that b4. I realize adding another 30amp to the service entrance conductors will add resistance and force the inverter to increase voltage but it's not even like getting to that point.

The data you have indicates that it is getting to that point.

You can basically model things as if the resistance from the inverter to the POCO's transormer is a fixed amount. And that the voltage at the transformer is fixed at a specific value.
Let's say you're running a couple of motors, and an electric oven and an electric drier. The voltage at your house is going to be lower than the voltage at the transformer. This is called "voltage drop".
When you're producing power, the opposite happens - the voltage at the inverter is higher because of current * resistance.
The inverter can try to put 30A out toward the grid. It can see the voltage rise, and determine in less than 1/60th of a second that the voltage is way higher than what it should be when it starts pushing that power out toward the grid. When it sees that happen it'll stop trying to send out power, report an error like what you're seeing and then try again later. And if it sees the same condition it will report that error again, and wait a bit and try again a little later. And so on.

the order of operations would ge through is 1. Look at the grid voltage ect, 2 connect to the grid sync ect, 3 start producing power. It's not even connecting. It connects initially then once a certain load is reached it kicks off, then attempts to reconnect and obviously doesn't like what it sees and just keeps erroring out with the reconnection fault grid error.

Part of that "connecto to the grid" is pushing power out to the grid. If it finds it can't push power out for even 1/120th of a second without exceeding it's parameters for max voltage, that is a grid error condition.

Is there a DC shutoff separate from the AC connection?
ie - can you easily disconnect the panels while still using the inverter to monitor the AC voltage?
If so, you can look at that, and see what the AC voltage is at different times of the day.

You can see how much it changes due to just fluctuations in the grid (usually will vary by a few volts over the day - maybe you have a lot of people running AC in the afternoon in your city and the voltage goes down 1% due to that - or maybe you have a lot of dairy farmers nearby running motors from ~6AM-8AM for morning milking along with all the people cooking breakfast and that causes enough demand that the voltage drops by 2%. Does the voltage go up by a few percent at around noon because all your neighbors are running solar?)

And you can see how much it changes due to fluctuations in your consumption. Like does it drop by 2% when you are running enough things to consume 30A*240V more than you were 3 minutes prior when you measured?

If 7kW of consumption causes a 10V decrease, most likely 7kW of production will cause a 10V increase. Probably you'd have even more of an increase because you'd have an additional I*R for the wires between your inverter and your main panel.

[Salts]

Turn on your air conditioner, all the lights in the house, television, vacuum cleaner and anything else you can think of.. See if the other inverter trips out when everything is running.

If it does, its probably not a line issue.. If it doesn't, either your pipe isn't big enough, or its clogged. I'd start by checking the contacts in your own breaker panel and make sure everything is tight and secure and not corroded. Then check the contacts at the meter. You might want to call your utility and let them know you need to cut that seal and get their approval so they can replace it.

Beyond your meter, you have to have things serviced by the utility company.

60 amps is a lot of power going backwards. I read the thread and I didn't see where you said what size feed line is going to these inverters and how far it runs back to your main breaker box. For that kind of amperage, you'd need a 1ga or 2ga feed line, and that's if the run is short. If you're running 100 feet or more, you almost need to jump up to something like 1/0 copper.

I only push 30 amps and I installed 4ga for a 130ft run. A bit over-kill, but I wanted to leave room for upgrades.

What is the distance to your two arrays? Is there a significant difference in how far they each are from your main panel? If there is, is it the closest one that seems to be dominating?

[Markyrocks69]

I've been all through the panels everything is good there. I'm not touching the meter socket. I only have one big array but one inverter is like 30 ft further away and that is the one that kicks out . The inverters are probably like 150 and 180 ft away from the panel, there is no combining of the ac out by the inverters. Theres 2 separate circuits, the wire is way overkill for the 32 amps. I used 2 gauge aluminum but there is very little loss. In the panel its usually 252v at the inverters it was around 250v. I'm not sure where you're getting your information about wire sizing ect but for 40 amps 8 gauge is usually fine for short runs. I used line loss calculators and all kinds of research b4 I did this and 2 gauge seemed to be the point of diminishing returns for the balance between price, ease of installation and potential power gains. The wire is pretty good, I forgot exactly what the specs said but its actually the same size as 2 gauge copper bc of whatever technology. I just checked it out so the wire is compact stranded meaning that its about as close to solid wire as it can be without actually being solid. So I'm assuming it would have the least amount of resistance. But the 2 gauge aluminum is 90 degree use-2 I believe 100 amps but again each inverter is serviced by it's own set of wires. The wire that goes between the main panel and the meter is 4/0 alluminum use-2 90 degrees.

I'm going to move one of the solar breakers and do a test today . I attached a pic of how its connected now vs how I'm going to test it. I honestly think the inverters are fighting in the panel bc the breakers are right next to each other and theres no other loads in there. I'm going to split them up hopefully it fixes it. Just kinda annoying.