East-West PV configuration

Hello,

I am designing a PV system spread over two flat roofs. Due to space limitations I opted for a East-West configuration to fit more modules onto the roof, as I can just place the rows essentially right next to each other.

I have 2 strings with 15 modules in series. I am using a SolarEdge 8K inverter with power optimizers for each module.

I am running into problems modelling this setup though. When I leave the self-shading tool on, it severly reduces the energy output due to shading. However, I assume there should be no shading in this configuration?

For East-West configurations, can I simply turn of the self-shading feature and still get accurate results?

I have attached a layout of the confifuration and the SAM file.

Any help would be much appreciated.

Thanks a lot,

Kevin

Hi Kevin,

Thank you for attaching the SAM files and providing the diagrams.

SAM's inputs are designed either for arrays of modules that are in the same plane, or parallel rows of modules that all have the same orientation (tilt and azimuth angle). There are ways to approximate other layouts like yours.

If you assume there will be no self shading, then I think the approach of modeling all of the east-facing modules in one subarray and the west-facing modules in another subarray without enabling self shading should give you reasonable results. Your diagram shows an odd number of modules in each row, so from the top view, I'm not sure which modules face east and which ones face west.

The way you have the self shading inputs set up in the "eastwest_shading" case , SAM is modeling Subarray 1as 3 rows of east facing modules tilted at 10 degrees and almost touching, so I would expect it to calculate shading losses, especially given the relatively high latitude of almost 50 degrees North. Subarray 2 is 5 rows of west-facing modules.

Best regards,
Paul.

Paul and Kevin,

This information has been helpful. I'm modelling an east-west case as well and I'm trying to figure out how I can take shading into account?

I have subarray 1 set up as an east-facing row, and then subarray 2 is west (sampling just 2 rows but I suppose I could do 4 if needed). Since they are separate subarrays, is there a way I can set it up so SAM knows they are going to be next to each other?

thanks
steph

Hi Steph,

For self shading calculation purposes, SAM models subarrays separately, so it can't model shadows caused by of rows of modules in one subarray on rows in another subarray.

Best regards,
Paul.