TL:DR: I saw a set of cheap panels with weird specs at Home Depot. I bought some to experiment on and to use as portables to augment my mounted solar.
Home Depot is selling 200w of panel for $114. That’s $0.57/Watt. Not amazing compared to used panels (typically $0.33/Watt) but HD is all over the place and has free ship-to-store.
It also comes with mounting brackets and one of those single-stage PWM controllers. I’m not bagging on that kind of controller, but it’s not a great fit for this particular set of panels.
#THE CATCH
The panels are a weird design, apparently 24 cells in series. Normal “12v nominal” panels have 36 in series for ~18Vmp. These have a Vmp of 12.0v, so I think we would call them “8v nominal”.
This makes them practically unusable in parallel for charging lead or LiFePO4.
You could run the panels in series on the PWM controller since it has a 50v input max and the series Voc would be 30v. But, due to the way PWM works the panels would be running at in the 14v range at the most. This is way, way off the 24.0Vmp of the series array. I’d expect a max harvest of ~120w with that kind of setup. If these were normal panels in parallel and on PWM I’d expect a max of ~160w. We can go into the math on that if anyone wants.
The best case scenario IMO would be to run the panels in series with an MPPT controller. This would get us closer to ~170w max harvest.
some other thoughts:
- The panels might work well enough in parallel for 3S Li-NMC because of that chemistry’s lower voltage
- HD has a 10% discount program for veterans if you provide them with a bit of documentation.
It’s cos(your latitude-23.44°), or just 1 in the tropics, unless I’m misunderstanding the question. I assume this is for a panel mounted level? Make sure your calculator isn’t expecting radians!
If you mean ideal at a specific time and date, you can use the formula from the Wikipedia page. Substituting in 23.44°*sin(years since last March equinox*360°) for the (rough) solar declination, and longitude+UTC for hour angle, you get:
sin(latitude north) * sin(23.44°*sin(year fraction*360°)) + cos(latitude north) * cos(23.44°*sin(year fraction*360°)) * cos(longitude+UTC time).
If you’re just looking to avoid writing that out every time, I’ll happily implement it in the common programming language of your choice. If you really care about the deviation caused by the elliptical orbit of the Earth you’ll need more complicated math.
Edit: That should be signed latitude on the second equation, and March equinox, not solstice. Fixed.