Inverter size

Status
Not open for further replies.

inforaj

Member
Location
Chiago
Occupation
Member
Can you help with this sizing calculation?
There are 191 SunPower Modules(SPR-E20-330-com), and each has 330W, So the total power DC KW = 63.03. The module has an open-circuit voltage=64.9V, a short circuit current = 6.52A, and a rated current=6.04A.
Open-circuit power generation with rated current = 64.9 X 6.04 = 74.87123 KW
After the low temperature at rated voltage calculation, the voltage at each module is 62.46V and rated current = 6.04, So the generated power = 72.05635 KW.

The preliminary design has an SMA inverter (STP-Core1 50-US S, 480V) with a maximum capacity of 75KW and efficiency = 97.5%.

As per the above calculation, I'm assuming this inverter (STP-Core1 50-US S, 480V) is undersized. I guess we will go for a larger size of the inverter.

I really appreciate any help you can provide.
 

winnie

Senior Member
Location
Springfield, MA, USA
Occupation
Electric motor research
1) voltage under load is lower than open circuit voltage, so your power calculation is too high.

2) peak power generation only occurs infrequently. It often makes sense to intentionally undersize the inverter relative to the DC array power.

Jon
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
Can you help with this sizing calculation?
There are 191 SunPower Modules(SPR-E20-330-com), and each has 330W, So the total power DC KW = 63.03. The module has an open-circuit voltage=64.9V, a short circuit current = 6.52A, and a rated current=6.04A.
Open-circuit power generation with rated current = 64.9 X 6.04 = 74.87123 KW
After the low temperature at rated voltage calculation, the voltage at each module is 62.46V and rated current = 6.04, So the generated power = 72.05635 KW.

The preliminary design has an SMA inverter (STP-Core1 50-US S, 480V) with a maximum capacity of 75KW and efficiency = 97.5%.

As per the above calculation, I'm assuming this inverter (STP-Core1 50-US S, 480V) is undersized. I guess we will go for a larger size of the inverter.

I really appreciate any help you can provide.
A challenge in pairing that module with an SMA Core 1 inverter is that the E20-330 is a high voltage, low current module; even in areas that don't get very cold the maximum 1000V string length is only 14 modules. The Core 1 inverters have 6 MPPT channels and each has two unfused inputs, so the maximum number of E20-330 modules you can connect to any Core 1 inverter without resorting to triple stringing and string fusing is (12)(14) = 168 modules, or 55440W DC. That's only a 1.11:1 DC:AC ratio for an STP50 and 0.887:1 for an STP62.
 

inforaj

Member
Location
Chiago
Occupation
Member
A challenge in pairing that module with an SMA Core 1 inverter is that the E20-330 is a high voltage, low current module; even in areas that don't get very cold the maximum 1000V string length is only 14 modules. The Core 1 inverters have 6 MPPT channels and each has two unfused inputs, so the maximum number of E20-330 modules you can connect to any Core 1 inverter without resorting to triple stringing and string fusing is (12)(14) = 168 modules, or 55440W DC. That's only a 1.11:1 DC:AC ratio for an STP50 and 0.887:1 for an STP62.
That's right, thank you. Here is the Inverter and string summary. Each inverter has 191 modules.
 

Attachments

  • Inverter & String.png
    Inverter & String.png
    13.1 KB · Views: 10

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
You definitely don't need a 75kW inverter for 63kW nameplate of modules, as a matter of general principle. Unless the inverter is already purchased or there is some other devil in the details, my guess is that it would be more cost effective to provide a few small fused combiners on a smaller inverter.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
You definitely don't need a 75kW inverter for 63kW nameplate of modules, as a matter of general principle. Unless the inverter is already purchased or there is some other devil in the details, my guess is that it would be more cost effective to provide a few small fused combiners on a smaller inverter.
He could use a 50kW inverter if loading 126% is OK for him. The SMA Core 1 is only available in 33.3kW, 50.0kW, and 62.5kW versions. They all have the same DC section, though, so he will be triple stringing in any case.

Sunpower modules and SMA Core 1 inverters have a special relationship with respect to rapid shutdown; maybe that's what he is dealing with. SPR MLSD (SunSpec compliant) modules use the Tigo TS4-R-F device and I haven't been able to find an inverter company other than SMA that will certify communication with the TS4-R-F.
 

inforaj

Member
Location
Chiago
Occupation
Member
A challenge in pairing that module with an SMA Core 1 inverter is that the E20-330 is a high voltage, low current module; even in areas that don't get very cold the maximum 1000V string length is only 14 modules. The Core 1 inverters have 6 MPPT channels and each has two unfused inputs, so the maximum number of E20-330 modules you can connect to any Core 1 inverter without resorting to triple stringing and string fusing is (12)(14) = 168 modules, or 55440W DC. That's only a 1.11:1 DC:AC ratio for an STP50 and 0.887:1 for an STP62.
The SMA claims the DC/AC ratio can be as high as 1.5 according to SMA Core1 specifications sheet. But we stay on the range of 1.0 – 1.37 to avoid clipping losses. The maximum we may go for 1.37 ratio.
 

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
But we stay on the range of 1.0 – 1.37 to avoid clipping losses. The maximum we may go for 1.37 ratio.
A fine choice, but understand that it is primarily economics that should guide that choice. If a system with a 1.5 DC/AC ratio produces the same annual energy as a system with a 1.37 ratio, but costs less, it's a better deal. That assumes that "annual energy production" is the proper metric of utility; with time based electric rates and other considerations, it may not be, so substitute the appropriate metric for your comparison.

In other words, a priori "clipping losses" are no worse than "underutilized inverter losses".

Cheers, Wayne
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
A fine choice, but understand that it is primarily economics that should guide that choice. If a system with a 1.5 DC/AC ratio produces the same annual energy as a system with a 1.37 ratio, but costs less, it's a better deal. That assumes that "annual energy production" is the proper metric of utility; with time based electric rates and other considerations, it may not be, so substitute the appropriate metric for your comparison.

In other words, a priori "clipping losses" are no worse than "underutilized inverter losses".

Cheers, Wayne
And clipping losses in the middle of the day are sometimes made up for, or more, by increased production on the shoulders of the power curve.
 

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
And clipping losses in the middle of the day are sometimes made up for, or more, by increased production on the shoulders of the power curve.
Generally not on a straight kWh basis, to my understanding, but quite possibly on a $ basis, depending on the TOU rates applicable.

Cheers, Wayne
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
Generally not on a straight kWh basis, to my understanding, but quite possibly on a $ basis, depending on the TOU rates applicable.

Cheers, Wayne
Some analyses I have seen from PVsyst have shown otherwise.
 

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
Some analyses I have seen from PVsyst have shown otherwise.
I don't think "generally not" precludes "occasionally."

Anyway, that would be possible if inverter efficiency is highest at 100% nameplate power, with a significant dropoff at lower powers. Is that true of modern inverters? I.e. does the PVsyst analysis have the correct efficiency curve?

Cheers, Wayne
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
Generally not on a straight kWh basis, to my understanding, but quite possibly on a $ basis, depending on the TOU rates applicable.

Cheers, Wayne
The 'shoulders' will always produce more energy on a straight Kwh basis for some range of DC/AC values between 1 and the point of diminishing returns (assuming the insolation on each module is roughly uniform).
 

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
The 'shoulders' will always produce more energy on a straight Kwh basis for some range of DC/AC values between 1 and the point of diminishing returns (assuming the insolation on each module is roughly uniform).
Certainly if you fix the inverter rating and increase DC/AC ratio by increasing the PV DC rating. But if you fix the PV DC rating and decrease the inverter rating, I don't think that's particularly true. It may be slightly true to the extent that the inverter efficiency is not constant with changing DC input power levels.

Cheers, Wayne
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
Certainly if you fix the inverter rating and increase DC/AC ratio by increasing the PV DC rating. But if you fix the PV DC rating and decrease the inverter rating, I don't think that's particularly true. It may be slightly true to the extent that the inverter efficiency is not constant with changing DC input power levels.

Cheers, Wayne
Naturally it matters what you start with and what you change. If you fix the inverter size and increase the DC:AC ratio past the point where the output in the middle of the day on the most productive days of the year starts to clip, the production will nearly always continue to increase. If you start with a fixed array size and decrease inverter size until clipping occurs, the output will of course decrease.
 

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
And clipping losses in the middle of the day are sometimes made up for, or more, by increased production on the shoulders of the power curve.
If you start with a fixed array size and decrease inverter size until clipping occurs, the output will of course decrease.
Those two statements seem to me in contradiction.

If the increased production on the shoulders more than makes up for the clipping losses at the peak, then the total output would have to be increasing with decreasing inverter size. At least for some inverter sizes between "size at which the inverter first starts clipping" and the size at which the first statement above is true.

Cheers, Wayne
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Electrical Engineer - Photovoltaic Systems
I don't see how I could state it any clearer than I did in post #16, so I will stand pat.

Edit: Maybe I can. For a given system with no changes, if insolation increases to the point where the output clips, does that clipping represent "losses". No, not necessarily; the output is also increasing during the rest of the day, on the shoulders of the curve. That's what I meant.

Would a larger inverter have captured the DC input when the smaller one clipped and therefore produce more, net? Yes, of course, but you'd pay more for the larger inverter, so it might not be a net gain.
 
Last edited:

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
Edit: Maybe I can. For a given system with no changes, if insolation increases to the point where the output clips, does that clipping represent "losses". No, not necessarily; the output is also increasing during the rest of the day, on the shoulders of the curve. That's what I meant.
Ah, to me "clipping losses" always means the difference between what your system is producing, given that the inverter is maxed out during part of the day, and what your system would be producing if the inverter size were increased just enough so that it was never maxed out.

So my responses have all been based on that definition.

Cheers, Wayne
 

jaggedben

Senior Member
Location
Northern California
Occupation
Solar and Energy Storage Installer
Certainly if you fix the inverter rating and increase DC/AC ratio by increasing the PV DC rating. But if you fix the PV DC rating and decrease the inverter rating, I don't think that's particularly true. It may be slightly true to the extent that the inverter efficiency is not constant with changing DC input power levels.

Cheers, Wayne
If you start with a given DC size and a DC/AC ratio of 1, and lower the inverter size, you will lose increasing losses to clipping. There will be a point at which the inverter produces less energy from the last module in the array than it clips. That's the equivalent point of diminishing returns going in that direction. Remember the question is whether "clipping losses ... are ... made up for ... by increased production". In practical terms, that's on a per panel basis.

The ranges will change depending on whether you choose to look at clipping on an annual basis or for some other defined chunk of the year.
 
Status
Not open for further replies.
Top