Actual Solar Generated Near Me

[bhbaker220]

I’ve only had a 4.1 kw (10 panels) grid tied system running since August. I had very comparable results in power generated vs PVWatts in August but have been running significantly lower since early September. Some of the deterioration into the fall is a clipping due to the sun not rising as high and shading from trees. However, I think a good portion is due to more cloudy days than what PVWatts is predicting.
Is there any data available that would give me a perspective of actual solar generated in the fall days and months this year vs prior years in the Houston area with a south facing roof mount system? Seems like there would be site where users can share their system design and actual solar generation.

 

[hwy17]

pvwatts uses simulated weather so real weather will definitely vary from that, but I'm not aware of the kind of resource you're looking for. ERCOT might have solar production data you can compare to last year but that would be for regions or the whole state probably.

 

[Shimmy]

The best information you will get to better understand your situation is a meter that gives you sub-minute production data from your own system. It really is transformative in understanding what things impact production.

Shading kills solar production, especially if you are not using microinverters or optimizers on each panel.

 

[bhbaker220]

The best information you will get to better understand your situation is a meter that gives you sub-minute production data from your own system. It really is transformative in understanding what things impact production.

Shading kills solar production, especially if you are not using microinverters or optimizers on each panel.

I have Emporia Vue installed and therefore have the instantaneous measurement of solar and know I have shading problems. What I don’t have is a year over year comparison that would shed some light (no pun intended) on the number of cloudy days from one year to another. Seems like I’ve had 2-3 weeks of pretty cloudy weather where as the data from PVWatts must average historical data over several years. We should have full sun for two days this week, so I’ll get a better perspective of what a good day in November looks like.
I’d be willing to pay a small one time fee to access the customer data (anonymously) that is in the cloud by companies like Growatt if they had search criteria to screen for data near me. Perhaps I’ll have to wait till AI matures.

 

[LydMekk]

Usually you get around 75% of the rating of your panels.

 

[Shimmy]

Unfortunately you can only really baseline to what an average month is, and for that you need history or to trust PVWatts.

If you want to play with the math, you can map your noontime peak production and fit a curve to it with start/end of production to give you a best-case for the solar day. Integrate for energy.

Personally, I find that cloud impact is highly variable day-to-day and week-to-week. Trending things, my system in Honolulu generally goes from ~40kWh/day November 1 to ~30kWh/day November 30th. In August it peaks around 55-60kWh on a perfect day. But, bad days can easily be less than half that. Then some years I can magically get 20% above those numbers on a given day or week. It was frustrating to me at first, but now I get that the numbers only really work in aggregate.

There are paid services though that offer better statistics and forecasts though; they are intended for large solar farms and day-ahead price bidding.

 

[2TrevorJ]

I'm about 3 hrs NW of you:

Aug was 3.1% over PVWatts' estimate
Sept was 3.0% under
Oct was 8.2% under

Pretty sure it's just this year's weather/clouds coming in worse than PVWatts' historical data.

 

[bhbaker220]

I'm about 3 hrs NW of you:

Aug was 3.1% over PVWatts' estimate
Sept was 3.0% under
Oct was 8.2% under

Pretty sure it's just this year's weather/clouds coming in worse than PVWatts' historical data.

Thanks. That's exactly the feedback/data I was looking for.