Welcome to Winter (Solstice)

Just a stab in the “dark”: dibs on the two 23.5 degrees off the equator.

 

Felt somewhat of the trendsetter with panels still being quite expensive in the 2000s . Our system was commissioned january 2008. Rated at 8kW AC.



(always thought DC ratings were a marketing gimmick as that's not what your system ultimately needs to output). Charging 2 Cars + home electricity - it paid for itself in just over 6 years (as feds picked up ⅓ of the cost). Our electric bill was usually a couple hundred dollars a month average (So Cal) prior to 2008 - having some of the highest rates in the nation. The utility rates just kept going up & up, & PV kept zeroing out.
Now, we're living up north and are considering bifacial panels - which run 15%-30% more efficient:



Ideally aligned from North to south, they convert sunlight on both sides, & in snowy conditions especially - light capturing can include reflection off snow. Our roof in the NW is horrible for locating PV on any south facing directions - but we have a good amount of land - which will make this weird verticle install more practical. Plus, way way easier to keep Montana snow off - compared to flat / relatively flat roof mounts.

 

Now that others have had some time to chime in, here is the solution I'm leaning towards, at least so far. It uses astronomical equations very similar to what I located and put on my old unix workstation back before they migrated us to WinDoze:

https://oikofuge.com/which-place-gets-the-most-daylight/


Shortest total daylight at the South Pole with 4388 hours, longest on Arctic Circle with 4649 hours, a difference of 261 hours. Go mountain climbing, and get 5052 hours atop Mount Forel in Greenland. Denali and a few other peaks in Alaska and Siberia exceed 5000 hours, but still fall more than a day short of Forel.

Other charts show the work path towards this final chart. He also mentions some additional complications not factored in.

I've long accounted for the solar disc not being a point, and for atmospheric refraction, but for some reason had never considered the north-south difference caused by Earth's elliptical orbit. This non-circular orbit effectively lengthens the North's summer a bit when Earth is farther away from the Sun and moving slower, while shortening the South's summer when closer and moving faster. Perihelion (closest point to Sun) is in early January, aphelion (farthest point) in early July.

= = = =

This has a significant discrepancy between its Arctic Circle and your 80 degrees latitude. The discrepancy deserves some exploration.

 

Geesh, I wasn't even close...

Article Circle equals the points beyond which you'll experience at least one day per year where the sun will never set, and a reciprical day where it'll never get above the horizon.

 

My understanding is that this is an economic thing: when considering where to spend your last incremental system dollars, do you get greater production $$ return by adding more DC capacity (larger or more PV panels), or more AC (larger inverters), or both in balance?

When DC capacity exceeds AC capacity, the peak PV output under the best conditions gets clipped and some is lost. But the system spends very little total time in this peak condition, the great majority of operating time is on the 'shoulders' of the daily output curve, and all the extra PV output can be processed by the inverters. In non-ideal weather, hazy conditions, and non-ideal orientations (both directional and seasonal), the PV simply won't reach its rated output capacity, so there is even less clipping, if any at all.

IOW, extra DC will be contributing most of the time, while extra AC will be used only rarely.

The ideal economic balance ratio between DC and AC capacity is a function of the relative prices of the available PV panels vs inverters, which are continually moving, so there is no fixed answer. And available equipment choices at any given moment greatly constrain the practical solutions.

This orientation also produces a very different daily power/energy schedule than the more common south facing orientation, which doesn't fit any utility's demand or load pattern.

Absent batteries, a mixture of orientations can better fit the load patterns. But no mix of solar orientations can fit the 24 hour demand pattern, so batteries or other sources are needed. Figuring the best fit is a very complex problem.

For those of us on simple flat rate net metering, the goal is to maximize annual production, regardless of timing. For most, facing due south works best for that. When home solar producers get put on Time Of Use/Production metering, everything changes.

 

The picture of our PV above shows 2 gables, each array with 4kWs (ideal conditions) max output. The system had two 4kw inverters (micro inverters hadn't reached their potential/reliability/pricepoint back then). Cold days late spring with light cloudy conditions - the panels could cool down to maybe low 60s°f then the sun would come out from behind the clouds. The Gable facing the sun would over produce as much as 20% for a few brief moments until the panels heated back up. The inverter cooling fans would really rev! Apparently the inverters are underrated - in order to accommodate such conditions because they never let the smoke out.