PV watts is giving me higher morning production in summer the higher I tilt a 270 degree array.

[hwy17]

I'm just surprised by this.

It's a bifacial array facing 270 degrees due west.

Comparing tilts of 25 and 35 degrees, the hourly data is showing higher production for the 35 degree tilt in 8-10am for the same days.

Do you think it's like a bifacial gain thing?

I was expecting higher tilts to move production later, at the cost of overall production.

 

[sunshine_eggo]

Is that consistent year round?

 

[DougfromdaUP]

It’s been a while since I’ve used PVWatts, but don’t they express angles as tilt from horizontal? So wouldn’t you expect the 35 degree (closer to vertical) panel to produce more when the sun is at a low angle?

 

[1201]

Pvwatts is not infallible. I've gotten some results that don't make sense

 

[Zwy]

Try JRC https://re.jrc.ec.europa.eu/pvg_tools/en/#PVP

 

[hwy17]

Is that consistent year round?

It does not happen every day year round, but it does occur frequently year round. Most days I check have it. The columns I captured here are DC watts production.

Pvwatts is not infallible. I've gotten some results that don't make sense

Yeah it seems bugged. Maybe especially in the bifacial function.

 

[hwy17]

Do you guys think pvwatts accounts for horizon shading due to mountains? It might be calculating a direct back side illumination that were actually not going to get. We have a maybe 20 degree mountain shadow to the east.

Or maybe it's actually accurately catching that, and this morning bump on the 35 is the direct backside illumination it will catch.

 

[MTM98290]

You can use suncalc.org to set an object and it's height. This will show the shadow cast by the object. It isn't perfect but it gives you an idea of what to expect. This only works on the website version, not the app version as far as I can tell.

 

[sunshine_eggo]

Do you guys think pvwatts accounts for horizon shading due to mountains?

No. It factors in average weather over the last 30 years. It looks at each day, computes the average conditions from all data and then picks a day that's closest to the average and uses it.

It might be calculating a direct back side illumination that were actually not going to get. We have a maybe 20 degree mountain shadow to the east.

Or maybe it's actually accurately catching that, and this morning bump on the 35 is the direct backside illumination it will catch.

It's worth looking at the irradiance columns. Clouds can notably influence the diffuse irradiance value. Do you see a pattern?

 

[hwy17]

It's worth looking at the irradiance columns. Clouds can notably influence the diffuse irradiance value. Do you see a pattern?

I don't immediately see the how a higher tilt would result in better production early on a low irradiance morning. I guess it does look like a backside direct illumination effect if anything, typically disappearing 2ish hours after sunrise.