4kw system outputting 2.5kw. Why?

[Hedges]

Your earlier meter photos showed near full Imp current. But you had well below Vmp voltage. That is highly suspicious.

Even if you thought sky was clear, you can't know without instruments. But 10.47A reading vs. 10.96 Imp says it was a cool, clear sunny day.

So possibly, this is a behavior from the inverter I am not a subject matter expert at this, so I don’t know if my conclusion is accurate

Not unless it is a bug. The inverter would draw voltage down low to search for higher wattage peaks, but only briefly. It would dwell in vicinity of highest peak.

My WAG is that previous connection to 120V caused inverter to set about 2500W maximum (due to max current through transistors driving AC). But I would have expected it to curtail wattage with higher voltage, reduced current. Not lower voltage.
I don't suppose you have any way to read and possibly set parameters? (Which you just did, later post shows normal wattage) SB can probably talk by Ethernet, possibly WiFi unless built during the chip shortage. If we do find something screwy, given internet connection SMA support might be able to reach in and fix it.

It seems clear (or most likely assuming my Wild @$$ Guess is incorrect) that either there is voltage drop (and power dissipation) somewhere along the circuit path, or else some panels (or RSD) are letting current bypass them without delivering any themselves. I think a part shaded panel would then deliver its (shaded) Isc, probably in parallel with diode bypassing balance of current.

Them newfangled RSD boxes would be my preferred guess.
If you had access to IR camera, you might spot the outlier panel.
If those were optimizers you could probably read parameters per panel.

You could bypass optimizers one at a time, or do a binary search (half at a time.) Disconnecting half the optimizers would be more work than one at a time. Jumper to bypass half the array would just be one wire. I think MPPT will operate down to 100V so half probably works, but can't repeat binary search with half of a half. At least the middle row of panels isn't convenient to reach and rewire. Do shut off by opening AC breaker, then DC disconnect, before messing with DC wiring!

Inverter parameters:

"[US] Country Set UL1741/2016/120"

I thought "120" was odd, but that's just L-N voltage. manual says,

"240 V split-phase system L1, L2 and N • UL1741/2016/120 L-N-L"

 

[Solarisium]

Your earlier meter photos showed near full Imp current. But you had well below Vmp voltage. That is highly suspicious.

Even if you thought sky was clear, you can't know without instruments. But 10.47A reading vs. 10.96 Imp says it was a cool, clear sunny day.



Not unless it is a bug. The inverter would draw voltage down low to search for higher wattage peaks, but only briefly. It would dwell in vicinity of highest peak.

My WAG is that previous connection to 120V caused inverter to set about 2500W maximum (due to max current through transistors driving AC). But I would have expected it to curtail wattage with higher voltage, reduced current. Not lower voltage.
I don't suppose you have any way to read and possibly set parameters? (Which you just did, later post shows normal wattage) SB can probably talk by Ethernet, possibly WiFi unless built during the chip shortage. If we do find something screwy, given internet connection SMA support might be able to reach in and fix it.

It seems clear (or most likely assuming my Wild @$$ Guess is incorrect) that either there is voltage drop (and power dissipation) somewhere along the circuit path, or else some panels (or RSD) are letting current bypass them without delivering any themselves. I think a part shaded panel would then deliver its (shaded) Isc, probably in parallel with diode bypassing balance of current.

Them newfangled RSD boxes would be my preferred guess.
If you had access to IR camera, you might spot the outlier panel.
If those were optimizers you could probably read parameters per panel.

You could bypass optimizers one at a time, or do a binary search (half at a time.) Disconnecting half the optimizers would be more work than one at a time. Jumper to bypass half the array would just be one wire. I think MPPT will operate down to 100V so half probably works, but can't repeat binary search with half of a half. At least the middle row of panels isn't convenient to reach and rewire. Do shut off by opening AC breaker, then DC disconnect, before messing with DC wiring!



"[US] Country Set UL1741/2016/120"

I thought "120" was odd, but that's just L-N voltage. manual says,

"240 V split-phase system L1, L2 and N • UL1741/2016/120 L-N-L"

Logged in to the Sunnyportal and I see that at peak it is at 2.7kw of a nominal 4.09kw. Can’t see voltage remotely. Need to knock on the inverter to see that!!!! Weird German engineering!



One thing caught my attention is this:
String derated!



Second is the phase not being configured: L1, L2, L3. Is that required or having an impact?

 

[400bird]

Have you found the reason for the MPPT derate?
I wonder if it is because you are over the 10 amp limit on the MPPT?

Can you explain how the AC was hooked up incorrectly initially? I've seen you post what appears to be correct wiring twice.

Doing some math, I've got one idea that it could be some swapped wires.

If you had 7 panels in series, that would be 7 X 41.3v (VMP) = 289
Any chance you've got some wires swapped on the roof and only have 7 panels in series,

 

[Solarisium]

Have you found the reason for the MPPT derate?
I wonder if it is because you are over the 10 amp limit on the MPPT?

Can you explain how the AC was hooked up incorrectly initially? I've seen you post what appears to be correct wiring twice.

Doing some math, I've got one idea that it could be some swapped wires.

If you had 7 panels in series, that would be 7 X 41.3v (VMP) = 289
Any chance you've got some wires swapped on the roof and only have 7 panels in series,

Apparently SMA says the inverter get hot so it derates to protect sensitive electronics. I have no idea how much it derates! It is not clipping for sure.

Regarding how I connected the inverter to the main service panel.
Since the main service panel supplies both 120 and 240v, I had incorrectly connected the inverter to a 120 v source. Look at my drawing (this is my best at drawing)


I ran the inverter at 120v for a couple of hours and configured it too which is incorrect. Future readers, don't do this.

Later I connected it correctly to 240v. Refer to my drawing:



It had nothing to do with the panels and how they are connected in series.
But your math has a good point. It seems like I have 7 panels voltage down here at the inverter.

During the RSD installation I had two positive end at the end due to odd orientation of one single panel. Once I fixed it, I did use a ammeter and checked the voltage of each panel connected to the RSD device individually. Each panel delivered ~0.7 DC as expected from APSMART RSD.
Since the mc4 connections are idiot proof, it isn't possible to incorrectly connect positive instead of negative ends.


But it is possible that two single panels are totally defective and only bypass the current to the next panel.

I made a mistake and did not test each individual panels prior to lifting them up to the roof. Only tested 4 of them on the ground to test drive the system, and now that I am typing this, I remember the voltage was not what I expected. It was read by the inverter around 120 to 130 vdc from 4 panels. I associated the low voltage to the fact that it was late evening, there was shade on on panel, and they were laying flat on the floor. So did not question it furthermore.

Thanks for asking that question! With this, the RSDs are ruled out as the culprit.
Looks like it is an unintended consequence of connecting the inverter to 120vac! How can I reset or reconfigure it? Is it possible?

Is it needed? The system was up and running all day today and generated 21kw as expected from a 4kw system per PV NREL website. So it is generating as expected but running at a lower voltage, and as Hedges stated, the SMA us drawing more current instead of a running at a higher voltage.

Does this make sense to anyone? I’d rather run at a higher voltage than higher current.
Looking into how to reset and reconfigure an SMA SB…

 

[Solarisium]

Have you found the reason for the MPPT derate?
I wonder if it is because you are over the 10 amp limit on the MPPT?,

Also is the mppt supposed to run below 10 amp?
I wonder if the low voltage/ high current is causing the inverter to send that message of derating mode?
The inverter is installed in shade, and the housing temperature is not higher than 80 F.

 

[Hedges]

One thing caught my attention is this:
String derated!

Typically means too hot.
Are fans working? Is sun beating on the inverter?

Also is the mppt supposed to run below 10 amp?
I wonder if the low voltage/ high current is causing the inverter to send that message of derating mode?
The inverter is installed in shade, and the housing temperature is not higher than 80 F.

OK, not in the sun.

10A is max operating current (per MPPT), a bit under 10.96A Imp of your panel. You should be able to connect MPPT inputs A and B in parallel to handle the full current.

Maybe derating due to temperature rise of the MPPT?

Derating ought to mean higher PV voltage, not lower. Low PV voltage is still something to figure out.

Second is the phase not being configured: L1, L2, L3. Is that required or having an impact?

I don't know what phase refers to, unless three SB were installed on 3 phases of a 3-phase system.

Apparently SMA says the inverter get hot so it derates to protect sensitive electronics. I have no idea how much it derates! It is not clipping for sure.

Later I connected it correctly to 240v. Refer to my drawing:

Your drawing shows green ground going to center terminal, which is N (neutral).
Connecting neutral is optional, white wire should be used. Your breaker panel may have same bar for G and N, or may be separate.
There is a separate screw in SB for ground.

Thanks for asking that question! With this, the RSDs are ruled out as the culprit.
Looks like it is an unintended consequence of connecting the inverter to 120vac! How can I reset or reconfigure it? Is it possible?

I don't follow how RSD has been ruled out.

Simply feeding 120V should cause it to keep waiting until it sees the correct voltage.
Did it do anything with 120V connected?

Some settings can be changed more easily, others require "Grid Guard" code, settings locked after 10 hours.

Does this make sense to anyone? I’d rather run at a higher voltage than higher current.

Check the instructions on paralleling two MPPT and try that.
"Derating" is a clue, and trying to get all power through one MPPT may be a problem, even though voltage within limits and we'd expect it to just draw 10A max.

 

[Solarisium]

I don't suppose you have any way to read and possibly set parameters? (Which you just did, later post shows normal wattage) SB can probably talk by Ethernet, possibly WiFi unless built during the chip shortage. If we do find something screwy, given internet connection SMA support might be able to reach in and fix it.

Connected to the web interface and took screenshots:

 

[Solarisium]

' Typically means too hot.
Are fans working? Is sun beating on the inverter?

I wasn’t home at the time of "derating" message. Not sure if fan was working. But generally I haven't heard any fan noise so far from the inverter.

OK, not in the sun.

10A is max operating current (per MPPT), a bit under 10.96A Imp of your panel. You should be able to connect MPPT inputs A and B in parallel to handle the full current.

Maybe derating due to temperature rise of the MPPT?

Sounds reasonable. If the MPPT is drawing more current at lower voltage it should get hotter (I think).

Derating ought to mean higher PV voltage, not lower. Low PV voltage is still something to figure out.
I don't know what phase refers to, unless three SB were installed on 3 phases of a 3-phase system.

There are three radio buttons that I can select only one of them. I didn't make any deflection since I want sure.

Your drawing shows green ground going to center terminal, which is N (neutral).
Connecting neutral is optional, white wire should be used. Your breaker panel may have same bar for G and N, or may be separate.
There is a separate screw in SB for ground.

My service panel has one bar for ground and Neutral. So I have connected the N terminal to that bar.
Also one more finding: I initially had connected the inverter to 120 volts AC. One hour later I connected two poles of the breaker (which was reading 240vac) to L1 and L2 terminals. No N or Ground was connected. In less than 5 minutes I got an error message. Looked it up : High voltage grid was the error. When connected the breaker panel Neutral to N, the error message disappeared.

I don't follow how RSD has been ruled out.

Apt a month ago when I first connected the inverter to 120vac, I tested it with 4 panels without RSDs. Each panel at 49voc I was expecting a voltage a bit lower than 200vdc like 160vdc. Instead I got 120vdc to 130vdc from 4 panels without RSDs. The inverter without the RSDs when connected to 120vac, was showing lower than expected voltage I.e. ~120vac instead of ~160vac. So RSDs don't play a role I think.

Simply feeding 120V should cause it to keep waiting until it sees the correct voltage.
Did it do anything with 120V connected?

Yes, it output ~120VAC. I can try this tomorrow . And send a picture.it does not wait instead it starts generating at 120vac.

Some settings can be changed more easily, others require "Grid Guard" code, settings locked after 10 hours.

the inverter runtime after running it all day today is now 24 hours so certainly locked. I tried logging in the inverter interface using my installer account.

Check the instructions on paralleling two MPPT and try that.
"Derating" is a clue, and trying to get all power through one MPPT may be a problem, even though voltage within limits and we'd expect it to just draw 10A max.

Meaning splitting the single array into two Arrays? I certainly can do that too,

 

[Solarisium]

As you have mentioned right in the beginning, the inverter is locked at 120v and needs grid guard code to change.

Created an installer account at https://my.sma-service.com, pending account activation.
The registration screen/bot did not like the fact that as an installer I have a Gmail account for my email!

I logged in to my inverter interface as installer and saw these screens that confirmed the system works at 120VAC!

 

[Hedges]

I wasn’t home at the time of "derating" message. Not sure if fan was working. But generally I haven't heard any fan noise so far from the inverter.

Sounds reasonable. If the MPPT is drawing more current at lower voltage it should get hotter (I think).

Yes, more I^2R loss in certain parts.
Also, efficiency curves show lower efficiency at lower voltage. For a switcher, greater voltage conversion causes greater losses. 240Vrms is 340Vpeak, when highest current is delivered.

There are three radio buttons that I can select only one of them. I didn't make any deflection since I want sure.

My service panel has one bar for ground and Neutral. So I have connected the N terminal to that bar.
Also one more finding: I initially had connected the inverter to 120 volts AC. One hour later I connected two poles of the breaker (which was reading 240vac) to L1 and L2 terminals. No N or Ground was connected. In less than 5 minutes I got an error message. Looked it up : High voltage grid was the error. When connected the breaker panel Neutral to N, the error message disappeared.

Some SB distinguish 208/240/277V by voltages between L1, N, L2, and phase, when jumpers set for autoselect. Three other jumper positions would explicitly set voltage without needing N.

It seems yours sensed and set undocumented 120V, so gave error for 240V. Then with N connected, reconfigured its setting.
I'm a bit surprised because manual says 120 L-N-L for 120/240V split phase and 208Y. 208 L-L for 208 delta.
Oh, you configure either UL1741/2016/120 for 240V and 208V using N, or UL1741/2016/208 not using N.
Because you selected UL1741/2016/120 but N was floating (may have appeared same as wired to L1), that would have been connection used for 277V of other models but appears to be undocumented 120V.

Do you also have a ground wire? SB case should be grounded with a green wire. Neutral should be connected with a white wire.

Apt a month ago when I first connected the inverter to 120vac, I tested it with 4 panels without RSDs. Each panel at 49voc I was expecting a voltage a bit lower than 200vdc like 160vdc. Instead I got 120vdc to 130vdc from 4 panels without RSDs. The inverter without the RSDs when connected to 120vac, was showing lower than expected voltage I.e. ~120vac instead of ~160vac. So RSDs don't play a role I think.

OK, same voltage in that test with or without RSD. Also consistent with your low voltage seen now.
Question is why so low.

Yes, it output ~120VAC. I can try this tomorrow . And send a picture.it does not wait instead it starts generating at 120vac.

Does not wait the standard 5 minutes? If AC is present when PV applied, that may be what it does.

Producing 120V appears to be an undocumented feature. They used to have 700W model up to 1700W which were 120V, but dropped those.
If 120V is available from all the -41 models, that could be great news for people who want to do a small backup or offgrid system with single Sunny Island. Wattage of SB would be limited to about same current as 208V spec.

the inverter runtime after running it all day today is now 24 hours so certainly locked. I tried logging in the inverter interface using my installer account.

Installer lets you change some things. Grid Guard code is given to people who declare (convince?) that they are qualified to mess with stuff.
Not clear to me that anything is wrong with your configuration. UL1741/2016/120 is the setting for 120/240V split-phase and for 208Y. Apparently also for 120V single phase.

The other setting you might want or need is CA Rule 21. Utility probably requires it but doesn't confirm. If you add battery backup you will want it.
I've seen some indication SMA think on-grid settings even Rule 21 don't work great for battery backup, to be determined.

Meaning splitting the single array into two Arrays? I certainly can do that too,

While that would split the wattage between two MPPT, it would drop the voltage low, within MPPT operating range but below spec range. Lower efficiency and may hit tighter limits. It still can exceed 10A limit.

What I mean is connecting input A and input B in parallel, so they share the current of a single string.

Page 20 of manual:

"Parallel Operation of the DC Inputs A and B
The DC inputs A and B of the inverter can be operated in parallel and up to three strings can be
connected to it in parallel. As a result, as opposed to normal operation, up to three strings can be
connected directly to inverters with two DC inputs and up to four strings to inverters with three DC
inputs. The inverter automatically detects the parallel operation of the DC inputs A and B."

That would have 20A max operation, which your 10.5A fits easily.

I think that may fix your issue because each carrying half the current means 1/4 the power dissipation of resistive components (times 2 paths). Each circuit path runs much cooler.


You could later parallel another string of different orientation, because they wouldn't hit peak current at same time.

 

[Hedges]

As you have mentioned right in the beginning, the inverter is locked at 120v and needs grid guard code to change.

Created an installer account at https://my.sma-service.com, pending account activation.
The registration screen/bot did not like the fact that as an installer I have a Gmail account for my email!

I logged in to my inverter interface as installer and saw these screens that confirmed the system works at 120VAC!

Not sure, but I suspect "120V" is correct. Maybe somebody else with a -41 inverter can check and confirm.

My 480V delta TriPower says "277V"

I think I've heard of others with Gmail having an issue. I have Yahoo and did not have a problem, but I was communicating with a human not a bot, and I provide credentials.

 

[Hedges]

Connected to the web interface and took screenshots:

Seems reasonable. Whole lot of settings, not all of which apply - Battery, CT.
My 30kW TriPower had some similar things about minimum power, one referring to battery.

When I first operated it would sweep DC voltage and turn off again. I had small 2500W array and inverter was set for 1500W minimum. I reduced that to 500W and it would then begin operating.

 

[Solarisium]

Yes, more I^2R loss in certain parts.
Also, efficiency curves show lower efficiency at lower voltage. For a switcher, greater voltage conversion causes greater losses. 240Vrms is 340Vpeak, when highest current is delivered.



Some SB distinguish 208/240/277V by voltages between L1, N, L2, and phase, when jumpers set for autoselect. Three other jumper positions would explicitly set voltage without needing N.

It seems yours sensed and set undocumented 120V, so gave error for 240V. Then with N connected, reconfigured its setting.
I'm a bit surprised because manual says 120 L-N-L for 120/240V split phase and 208Y. 208 L-L for 208 delta.
Oh, you configure either UL1741/2016/120 for 240V and 208V using N, or UL1741/2016/208 not using N.
Because you selected UL1741/2016/120 but N was floating (may have appeared same as wired to L1), that would have been connection used for 277V of other models but appears to be undocumented 120V.

Do you also have a ground wire? SB case should be grounded with a green wire. Neutral should be connected with a white wire.



OK, same voltage in that test with or without RSD. Also consistent with your low voltage seen now.
Question is why so low.



Does not wait the standard 5 minutes? If AC is present when PV applied, that may be what it does.

Producing 120V appears to be an undocumented feature. They used to have 700W model up to 1700W which were 120V, but dropped those.
If 120V is available from all the -41 models, that could be great news for people who want to do a small backup or offgrid system with single Sunny Island. Wattage of SB would be limited to about same current as 208V spec.



Installer lets you change some things. Grid Guard code is given to people who declare (convince?) that they are qualified to mess with stuff.
Not clear to me that anything is wrong with your configuration. UL1741/2016/120 is the setting for 120/240V split-phase and for 208Y. Apparently also for 120V single phase.

The other setting you might want or need is CA Rule 21. Utility probably requires it but doesn't confirm. If you add battery backup you will want it.
I've seen some indication SMA think on-grid settings even Rule 21 don't work great for battery backup, to be determined.



While that would split the wattage between two MPPT, it would drop the voltage low, within MPPT operating range but below spec range. Lower efficiency and may hit tighter limits. It still can exceed 10A limit.

What I mean is connecting input A and input B in parallel, so they share the current of a single string.

Page 20 of manual:

"Parallel Operation of the DC Inputs A and B
The DC inputs A and B of the inverter can be operated in parallel and up to three strings can be
connected to it in parallel. As a result, as opposed to normal operation, up to three strings can be
connected directly to inverters with two DC inputs and up to four strings to inverters with three DC
inputs. The inverter automatically detects the parallel operation of the DC inputs A and B."

That would have 20A max operation, which your 10.5A fits easily.

I think that may fix your issue because each carrying half the current means 1/4 the power dissipation of resistive components (times 2 paths). Each circuit path runs much cooler.


You could later parallel another string of different orientation, because they wouldn't hit peak current at same time.

I see what you meant by parallel connection to input A and B. I will try that this evening or this afternoon. Perhaps I’ll try to shoot a video to show more details.

 

[Solarisium]

Connected the inputs in parallel to A and B.
this should solve the derating issue per Hedges advice.

Look at this video and see how I connected the DC and AC to the inverter.

 

[Hedges]

Looking forward to hearing if it does. Then I'll start to be expert on the models I haven't even touched.

Input C is zero volts (DC) as expected.

People with some hybrids have reported getting shocked by PV inputs, with no PV connected.

For informational purposes, check AC and DC voltages of input positive to negative, and positive to ground. Negative to ground as well, more as a check because that could be calculated (or maybe not for AC, which adds as vectors.)
(A and B are paralleled, same voltage, so no need to check both.)

Operating at reasonable load, I've seen AC Vpp between positive and negative around 3Vpp or about 1% of VDC. That's the ripple of capacitors supplying sine wave AC current from DC source.

Positive and negative may have DC voltages symmetric around ground. It is a transformerless design. But the (buck-boost?) converter likely is single ended, drives one phase relative to the other, so while its output will be symmetric the DC input may not be. (The inverter only monitors neutral, doesn't drive the two phases relative to it.)

If PV DC is asymmetric, it may carry AC voltage relative to ground. I don't think I saw that for my transformerless 10000TL-US. Those appear to be buck only, not buck boost, because PV DC input voltage had to be higher than Vpp.

 

[Solarisium]

@Hedges, what should I select as the country standard when/ reconfigured?

How should I connected the terminals?how the L1, N, L2, Ground should be connected to the main breakers panel?

Some SB distinguish 208/240/277V by voltages between L1, N, L2, and phase, when jumpers set for autoselect. Three other jumper positions would explicitly set voltage without needing N.

It seems yours sensed and set undocumented 120V, so gave error for 240V. Then with N connected, reconfigured its setting

I'm a bit surprised because manual says 120 L-N-L for 120/240V split phase and 208Y. 208 L-L for 208 delta.
Oh, you configure either UL1741/2016/120 for 240V and 208V using N, or UL1741/2016/208 not using N.
Because you selected UL1741/2016/120 but N was floating (may have appeared same as wired to L1), that would have been connection used for 277V of other models but appears to be undocumented 120V.

Do you also have a ground wire? SB case should be grounded with a green wire. Neutral should be connected with a white wire.

 

[Solarisium]

Looking forward to hearing if it does. Then I'll start to be expert on the models I haven't even touched.

Input C is zero volts (DC) as expected.

People with some hybrids have reported getting shocked by PV inputs, with no PV connected.

For informational purposes, check AC and DC voltages of input positive to negative, and positive to ground. Negative to ground as well, more as a check because that could be calculated (or maybe not for AC, which adds as vectors.)
(A and B are paralleled, same voltage, so no need to check both.)

Operating at reasonable load, I've seen AC Vpp between positive and negative around 3Vpp or about 1% of VDC. That's the ripple of capacitors supplying sine wave AC current from DC source.

Positive and negative may have DC voltages symmetric around ground. It is a transformerless design. But the (buck-boost?) converter likely is single ended, drives one phase relative to the other, so while its output will be symmetric the DC input may not be. (The inverter only monitors neutral, doesn't drive the two phases relative to it.)

If PV DC is asymmetric, it may carry AC voltage relative to ground. I don't think I saw that for my transformerless 10000TL-US. Those appear to be buck only, not buck boost, because PV DC input voltage had to be higher than Vpp.

Didn't capture in video, but when measured voltage between DC terminals and ground, I read some voltages. Didn't take notes but will do it again and report.

Also there was DC voltage between A and L. Will check and send later.

 

[Hedges]

@Hedges, what should I select as the country standard when/ reconfigured?

I think you have a suitable standard now. I think it will disconnect immediately above 60.5 Hz, and not ramp down watts. That was they way we used to do things, and your utility might want Rule 21 frequency-watts and other functions. I don't know if system has already been approved etc., but if so no need to change.

Rule 21 should be fine too. That's is the newer preferred setting according to utilities. More important for you, it works with AC coupled battery backup (Sunny Island, etc.) according to SMA's published documentation. So switching to Rule 21 should be better, at least no harm.

SMA America's video said when using SB behind Sunny Island, to set SB to offgrid. That goes against their written documentation, and I have concerns. For them to give that advice suggests Rule 21 does not work properly with their battery inverter. I have some guesses/theories.

Since you're found your SB will operate at 120V, you might be able to do backup with a single Sunny Island and not use a transformer. Probably requires manually switching off and back on so it recognizes the changed voltages. Could be better to use a transformer. With 3.8kW, current is within what SI relay can handle. However, this would need a 40A breaker in your panel (or another transformer). Is your main panel 100A or 200A?

How should I connected the terminals?how the L1, N, L2, Ground should be connected to the main breakers panel?

If your breaker panel has a single bar for neutral and ground, bonded to the enclosure, then neutral and ground are almost the same thing. They should be kept as separate branching circuits going to outlets, sub panels, etc. so "objectionable" neutral current doesn't flow in ground wires, only connecting together at that one bar.

You have a white wire from the neutral/ground bar to "N" terminal of SB, which is correct (unlike your schematic with green line.) There should be a green or bare wire from the bar to enclosure of SB (manual shows a terminal.) Per code, that should run together with line and neutral wires. I saw bare copper wires running outside of conduit. If that goes to SB it work, but isn't properly in the conduit. Maybe conduit itself meets code as grounding conductor, but all joints would have to be securely electrically connected. I run a wire inside the conduit.

Didn't capture in video, but when measured voltage between DC terminals and ground, I read some voltages. Didn't take notes but will do it again and report.

Also there was DC voltage between A and L. Will check and send later.

A table of all AC and DC voltages will be interesting. It will give clues to inverter design.
SMA has specs on PV panel capacitance. Some technologies have high capacitance and aren't compatible. AC voltage riding on PV would cause ground currents through the capacitance. I think it is only (or mostly) a problem for 208V and 277V, not symmetric 120/240V, but that could depend on inverter architecture.
At power-up, (some?) Sunny Boys swing PV offset relative to ground to check for shorts; that could also be affected by the capacitance.

 

[400bird]

Didn't capture in video, but when measured voltage between DC terminals and ground, I read some voltages. Didn't take notes but will do it again and report.

Is this how you measured the 250-290 volts DC from your solar? If so, that's not how you measure the solar out. You'd measure from PV + to PV - with one meter lead on each.

Also there was DC voltage between A and L. Will check and send later.

What is A?

 

[Hedges]

PV voltage reported by inverter

"A" was the one MPPT input initially connected.

Not sure if "L" is AC line voltage. I would have checked it to ground.
Some interesting things may be happening between PV and AC terminals, since these are transformerless.