New Install Solar Edge 3800A with S440

[Chyguy1]

Hey,

I am a solar installer and this is my first time using solar edge with power optimizers. I decided to experiment with a system on my house first before I started putting them on other peoples houses. So I have 14 370 W panels for a total of 5180watts. My inverter is 3800 W with POs S440. During my Peak sun the past few days. I’m getting a Max of 2660 Watts.

I have tested all the panels before I put them on the roof to make sure they were all within spec. I also tested all the power optimizers to make sure they were within range. I have paired them to the inverter. They are only operating at 50% tops.

so some background. I have been on the phone with solar edge technicians for the past few days. I’ve gotten 50 different answers, and none of them solved my issue.. from outdated software to Over watts on string.

So this can inverter can only handle 5100 W on a string. I’m at 5180. so I disconnected which put me at 4810. I Paired the optimizers with the inverter again. It did absolutely nothing. Just produce less. I’m at a loss. I almost feel like the optimizers are capping my voltage and Amperage

I have a 8.1 Kw system with a 6Kw inverter that was producing 6KW at the time during these test. Any Help.??

I’m getting ready to scrap solaredge on the first project. Luckily it’s my house too lol

 

[zanydroid]

I believe SolarEdge sets string to constant 600V while also matching the current along the string (multi parameter optimization).

What are the specs on the panels?

How much shading?

 

[ncsolarelectric]

What is the tilt and azimuth of these panels relative to the sun? If they're lying flat on the ground, you will only get around 50% out of them this time of year. IMO, Solar Edge is crap. I've replaced a bunch of them. I use Hoymiles microinverters. They're a huge advantage over the competition, IMO.

 

[Chyguy1]

No Shading. Brand new panels
under NOCT
Pmax 276wp
Vmp 31.7v
Imp 8.69a
Voc 38.2v
Isc 9.22a

The voltage is 380Vdc at inverter. Looks like the optimizers are keeping at 380. even when i take a panel off.


So They are facing 159E.. on a roof. with 22 degree pitch.

I have other panels that are 4 years old getting 75% same direction

 

[Hedges]

It seems like the panels fit in middle of optimizer specs.
They shouldn't be capping, rather converting as DC/DC.
Output limit is 15A, but at 380V / 14s = 27V, good enough to deliver 407W/panel 5700W/string so not limiting.

https://knowledge-center.solaredge.com/sites/kc/files/se-power-optimizer-s-series-datasheet.pdf

Do you have module-level monitoring? To see if all panels at 50%, or half at 0%.

You're getting 50% of panel STC. 75% of inverter rating (expect it to reach 100%.)
Is there anything that can derate inverter? Heating? Does it do that for high line voltage (rather than going offline)?
Did it get configured for reduced output?

Is that SE 3800H-US?

https://www.solaredge.com/sites/default/files/se-hd-wave-single-phase-inverter-datasheet-na.pdf

Nothing jumps out at me.


Unless some panels aren't producing, must be operating panels far from maximum power point. module level data might show something.

 

[Chyguy1]

So, Inverter is 25 ft from panels. And 6 ft from outdoor Panel..

panels are hitting 180-to 190 at peak sunlight per Solaredge monitoring i

When i tested without optimizers im getting 250-280 a panel at peak sunlight. I should be getting full potential out of this inverter.


It is the 3800A-US Version. According to Solar Edge. Everything is operating normally. there is no way to adjust incoming Voltage.. like i said whether i have 14 panels or 13,12,11,10.. it likes to stay at 380. So frustrating

So my other system. I have 27- 300w panels equal to 8100 watts on a 6000 watt inverter. facing the same way with same set up. I’m hitting the 6kw no problem.

 

[Chyguy1]

Was curious if i split the panels up to 7 in 2 strings to increase amps? i know i would need different optimizers cause i need 8 per string minimum. at this point my just toss this thing

 

[zanydroid]

like i said whether i have 14 panels or 13,12,11,10.. it likes to stay at 380.

This is consistent with the constant string voltage behavior. It’s supposed to be increasing amps to compensate. (380VDC is also coincidentally a good starting voltage for a 240VAC inverter)

Max output current is 15A from S440 which corresponds to the 5kW per string number being tossed around.

Try bypassing one panel at a time. Maybe one of them is messed up and SE is not smart enough to bypass it with the optimizer. You may or may not be able to find it with panel level monitoring due to the fact that there is still the current and total voltage constraint optimization that the system has to do, which may impact the measurements of available PV.

(Could also be two busted panels or optimizers, in that case it would be pretty time consuming to isolate)

 

[zanydroid]

i know i would need different optimizers cause i need 8 per string minimum.

If the inverter insists on operating at 380V then yeah you need 7 optimizers to hit that (and it would have degraded flexibility to achieve 380V while still trying to optimize)

 

[400bird]

Are all your panels in one string?
Have you looked at the volts and amps from each panel? You can see both measurements on the panel and string side for each panel/optimizer combo.

So this can inverter can only handle 5100 W on a string. I’m at 5180. so I disconnected which put me at 4810.

The spec sheet posted above says the 5100 watt limit is for 208v only. For standard 240, it's 5900 watts.

I Paired the optimizers with the inverter again. It did absolutely nothing. Just produce less. I’m at a loss. I almost feel like the optimizers are capping my voltage and Amperage

Possibly, look at panel voltage. If the optimizers are limiting power, panel voltage will be above VMP.

Can you share pics of your set up?

 

[Chyguy1]

Are all your panels in one string?
Have you looked at the volts and amps from each panel? You can see both measurements on the panel and string side for each panel/optimizer combo.


The spec sheet posted above says the 5100 watt limit is for 208v only. For standard 240, it's 5900 watts.

Possibly, look at panel voltage. If the optimizers are limiting power, panel voltage will be above VMP.

Can you share pics of your set up?

All Panels are in 1 String. Looking at voltage and amps from each panel during peak sun through the optimizer monitor shows Panel Voltage 33-35 V, Power optimizer 26-28 Volts, current about 6.5 Amps.

Optimizers are rated up to 60Vdc, 14a short circuit current. Well within spec. Don’t have pictures but i will take them on the next sunny day.

 

[Chyguy1]

This is consistent with the constant string voltage behavior. It’s supposed to be increasing amps to compensate. (380VDC is also coincidentally a good starting voltage for a 240VAC inverter)

Max output current is 15A from S440 which corresponds to the 5kW per string number being tossed around.

Try bypassing one panel at a time. Maybe one of them is messed up and SE is not smart enough to bypass it with the optimizer. You may or may not be able to find it with panel level monitoring due to the fact that there is still the current and total voltage constraint optimization that the system has to do, which may impact the measurements of available PV.

(Could also be two busted panels or optimizers, in that case it would be pretty time consuming to isolate)

I tested each panel and optimizer before i put everything together. All voltages, amps and ohms checked out. Think i should still try this?

 

[zanydroid]

I tested each panel and optimizer before i put everything together. All voltages, amps and ohms checked out. Think i should still try this?

That certainly reduces the chances of a problem.

You need more info to debug this, so maybe?

Are the grid settings correct? IE is there throttling at the inverter?

 

[Hedges]

Looking at voltage and amps from each panel during peak sun through the optimizer monitor shows Panel Voltage 33-35 V, Power optimizer 26-28 Volts, current about 6.5 Amps.

No Shading. Brand new panels
under NOCT
Pmax 276wp
Vmp 31.7v
Imp 8.69a
Voc 38.2v
Isc 9.22a

33V to 35V is well above 31.7V, 35V is getting awfully close to 38.2V

This suggests to me the inverter doesn't want to pull the panels down to where they produce full power.
(unless they are so cold that Vmp rises that far, check the math.)

which is what Zany is also thinking:

Are the grid settings correct? IE is there throttling at the inverter?

Suspicious about the spread 33 to 35V, do they receive different amounts of sun?

 

[zanydroid]

Looking at voltage and amps from each panel during peak sun through the optimizer monitor shows Panel Voltage 33-35 V, Power optimizer 26-28 Volts, current about 6.5 Amps.

This also contradicts my theory testing for a bad panel since this seems normal (symmetric behavior between all panels). Other than going above Vmpp.

 

[Chyguy1]

That certainly reduces the chances of a problem.

You need more info to debug this, so maybe?

Are the grid settings correct? IE is there throttling at the inverter?

so i diasbled power limit and control. Solar edge verified that nothing should be limiting the output. what’s even funnier is that on solar edge designer, When i list the system for 14 370 watt panels and hit generate. Its gives me the 3800watt inverter with s440. So i know that part is good. (according to solar edge)

Could having another inverter on the same service be limiting production? sounds stupid but just trying to run every scenario. Solar edge tech is no help

 

[Hedges]

Above Vmpp means not trying to harvest all power, curtailing.

33V to 35V is a pretty wide spread. What you would see if different orientations in series. But I think too wide a spread if each on its own optimizer.
Vmp doesn't vary much with illumination.

Could having another inverter on the same service be limiting production? sounds stupid but just trying to run every scenario. Solar edge tech is no help

Yes, but only if there is communication with a limited-export system.
Or if it limits due to voltage (check AC voltage). Or even frequency, so check that, but doubt anybody uses it.
Or derates due to heating.

 

[zanydroid]

Could having another inverter on the same service be limiting production? sounds stupid but just trying to run every scenario. Solar edge tech is no help

What is the voltage at the inverter AC output?

This is probably the voltage the inverter self reports.

Standard grid code and not like a funky low grid AC voltage set? Being over nominal grid voltage causes throttling

 

[zanydroid]

I’m very confused by how inverter output throttling would work together with you shortening the string to reduce the output power.

Is there some way in SE to limit the per optimizer output?

 

[Chyguy1]

33V to 35V is well above 31.7V, 35V is getting awfully close to 38.2V

This suggests to me the inverter doesn't want to pull the panels down to where they produce full power.
(unless they are so cold that Vmp rises that far, check the math.)

which is what Zany is also thinking:




Suspicious about the spread 33 to 35V, do they receive different amounts of sun?

They all receive same amount of sun.