Need 20kW single axis tracker - any ideas?

[J.P.M.]

The OP stated "Single Axis Tracker" he did not state Tilt required for 42 Degrees - you made that up.

I don't think Bruce made up the latitude of his location. I believe the point was that single axis (w/ N-S axis alignment) tilted at local latitude will probably produce more annual output than the same array size but with the axis same N-S axis of rotation being horizontal, but would perhaps require a beefier design.

All that however, says nothing about the likely, and at this point in the design, unknown lower cost effectiveness and increased maint. requirements of a tracking array vs. a larger fixed array.

[J.P.M.]

I do a fair amount of solar research that is shared with the local utility.
Aside from the perpetual problem of solar intermittency, the daily demand curve does not align with typical fixed array output. We want to make that better.
Obviously, new storage technologies are where we need to be looking..... and waiting.
Personally, I am woking on PV to Heat Pump to Thermal Storage. Several companies are operating in this area and we can expect some exciting new products in the HVAC area.

In the meantime, we wish to demonstrate the ability to generate as much late afternoon PV power as possible - when summer A/C demand is the highest.
Thus, we want to evaluate a rugged single axis tracker.
I have a South facing slope that is 150' long. Horizon to horizon East-West exposure.

Thanks for any ideas.

What's the slope gradient ? If high (steep) enough, that would allow a shorter row pitch for multiple rows. 2 or, more likely 3 rows of panels, depending on particulars. I'd avoid ballasted schemes, particularly on sloped surfaces.

If you're set on single axis tracking, one of many schemes to consider, at least for summer A/C loads, is an array that has "optimal" fixed azimuth vs. demand with the tracking being on the tilt rather than the azimuth. That might mitigate some of the galloping solar variability.

Supply will never match demand. That's why the Almighty made storage. For R.E., that usually means using the grid as a giant electrical storage facility via some form of net metering, or using electrical batteries if using PV, or some form of thermal storage. Any storage will also allow some additional flexibility in array design and orientation, with some of that flexibility as f(storage size). More storage, greater flexibility and generally a trend toward a fixed orientation that maximizes production at peak demand times.

If it was me, I'd store the PV on the grid if possible, and then, depending more than a little bit, but not entirely on cost, in electrical batteries as those losses will be small compared to thermal storage losses and PITA maint. of thermal storage, both of which will be more than you'll design far and greater than electrical storage losses and maint. costs.

Using PV to generate electricity and then turning the low entropy electricity into high(er) entropy heat, and then boosting the high entropy thermal energy to lower entropy thermal energy via electricity (a heat pump) is an unnecessary round trip that will cost you and gain little. If the goal is heat, avoid PV, use solar thermal, go with minimal storage and live with a lower solar fraction of supplied heat. For higher solar fractions, skip the thermal storage and store the energy in electrons.

Take what you want of the above. Scrap the rest.