proper string inverter sizing

[sensij]

I should add... inverter efficiency measurements are not likely to be the right way to tell how the inverter will operate outside of the MPPT range. Efficiency generally refers to the conversion of the DC to AC. The MPPT function is what helps the panels produce to the maximum ability. So if the panels operate at some point other than the maximum, it doesn't really matter that the DC to AC conversion is still 95% or better... the losses are upstream of that. The penalty for operating at something other than the maximum might be small, or might be significant. It just depends on what the inverter is doing in place of MPPT.

[charles2]

If the SMA tech so goes for it, who am I to say otherwise? The attached CSI data doesn't show below 270 V, but I'll take your word for it that the jpg shows data from this inverter.

Well, I'd say you have a keen eye for details and that official pronouncements are not always correct. I've certainly learned not to trust the claims of our incompetent and corrupt local government officials.

[charles2]

sensij, what do you make of the attached simulations? Do they clarify the issue or further obfuscate?

[solarix]

If you look at the "specific energy yield" for those two SMA designs, the single inverter produces 1449kW/kWh and the dual produces 1450kW/kWh. so all the extra expense of doing a dual install is 1 point better. KISS.

[sensij]

Solarix, glad you jumped in. The question here is how the inverter handles operating within the dc range, but outside the mppt range. I'm not sure I would trust the simulation results without knowing the answer to that question. Do you know?

[charles2]

If you look at the "specific energy yield" for those two SMA designs, the single inverter produces 1449kW/kWh and the dual produces 1450kW/kWh. so all the extra expense of doing a dual install is 1 point better. KISS.

By that metric the dual install is actually worse if you consider that the single inverter system used 1 less module (26) than the dual (27). What accounts for this?

[charles2]

Speaking very generally, by how much does MPPT increase inverter output?

[charles2]

sensij, here is a quote from an NREL engineer:

"You're right to question whether the efficiency is any good for DC voltages below 270V. If you look at the plot of efficiency that is shown in the datasheet for the 5000TL, when you go to a low DC input voltage of 175V, the efficiency has dropped from 97.5% to 96%. In order to maintain a weighted average CEC efficiency of 96%, SMA has changed the 'Rated' MPPT voltage range minimum from 125V to 270V for the 7700TL. The only reason to have two different MPPT input ranges is to game the CEC efficiency rating.

"As far as how low the actual efficiency is going to be, I would expect that anything much lower than 93%-94% would start to lead to problems in the heat dissipation and lifetime of components in the inverter. The fact that they even allow the 7700TL to connect at these low voltages indicates that the efficiency can't be too catastrophic. It's impossible to know without actually testing the inverter at these low input voltages, but I would bank on operation around 96% efficient (max) at 270V input, and possibly down to 92-93% efficient around the minimum of 125V. If you really need to operate at these low input voltages, you might be better off with two SB4000TL inverters."

I went with the 3.8kw inverters that you recommended. Thanks for your advice!

[inetdog]

Speaking very generally, by how much does MPPT increase inverter output?

For a typical insolation profile corresponding to clear sky and with the panels aimed for on solar noon, my SWAG would be that:
1. If your inverter did not have MPPT it would have to be customized specifically for whatever your personal peak panel power was.
2. Given that, you would lose about 1/3 of the total available power over the course of the full day.
3. For a constant impedance design, anything that cut the maximum panel output by a factor of two would drop the inverter output by a factor of four. Etc.

[sensij]

sensij, here is a quote from an NREL engineer:

"You're right to question whether the efficiency is any good for DC voltages below 270V. If you look at the plot of efficiency that is shown in the datasheet for the 5000TL, when you go to a low DC input voltage of 175V, the efficiency has dropped from 97.5% to 96%. In order to maintain a weighted average CEC efficiency of 96%, SMA has changed the 'Rated' MPPT voltage range minimum from 125V to 270V for the 7700TL. The only reason to have two different MPPT input ranges is to game the CEC efficiency rating.

"As far as how low the actual efficiency is going to be, I would expect that anything much lower than 93%-94% would start to lead to problems in the heat dissipation and lifetime of components in the inverter. The fact that they even allow the 7700TL to connect at these low voltages indicates that the efficiency can't be too catastrophic. It's impossible to know without actually testing the inverter at these low input voltages, but I would bank on operation around 96% efficient (max) at 270V input, and possibly down to 92-93% efficient around the minimum of 125V. If you really need to operate at these low input voltages, you might be better off with two SB4000TL inverters."

I went with the 3.8kw inverters that you recommended. Thanks for your advice!

Thank you very much for following up on this thread. What the NREL engineer is saying makes a lot of sense to me. I think you'll be happy with the design flexibility of the 3.8's, and can be confident you'll be getting the performance you are looking for.