Glad to see trackers that are working.
I'd looked into a few designs. One I liked was the one with two tanks of refrigerant gas, a couple shade-reflectors, and a motorcycle shock absorber, that work by the sun pumping the refrigerant between tanks, unbalancing the panel so it rotated to face the sun and balance the weight. I thought that would work fine with propane for the volatile liquid (when used outdoors and far from the house - like near the propane tank for the heating gas) to make it cheap (though the similar system to close venetian blinds at night and slant them for maximum solar gain in the day would require a nontoxic, nonflammable, working fluid.)
I've looked at trackers for my hypothetical future solar RE component at my Nevada place. The site is almost always cloudless (at 5015 feet and in a rain shadow, just downwind of a mountain range with the nearest pass at 8,314'). It has no vegetation that shadows the house - because the firebreak requires 50+ feet of bare dirt, then everything after that at least six times as far away as it's height. (Also: It's just sage and the like. No trees for a couple miles on the sunny side.) But it has strong, gusty, winds for part of most days, making mouint strength an issue. It is in a valley, with mountain shadows cutting off about the first half hour and last hour of daylight. It has over five acres, so there's no shortage of area limiting the number of panels that can be installed.
For solar at this site, trackers have never made sense to me. They only get about an extra 27% out of the panels in ideal conditions - and the mountain shadows near sunrise and sunset eat into that terribly. With panels running well under $1/watt, it makes far more sense to put the money into more panels and maybe more storage than into a tracker. It also lets me avoid several added points of failure, wear, and/or maintenance (such as the bearings, strong movable frame, wires that get flexed daily, whatever drives it, ...). It makes more sense to me to put in more panels, accept the cosine error, and just adjust the tilt seasonally, if at all.
IMHO Trackers sometimes made sense when panels were costly, and might still make sense for a site where there is a limited "sweet spot" for the panels but sky visibility from there to both the eastern and western horizon. But for my situation (lots of area, no obstructions to the south, mountains to the east and west) they look more like a reduction in performance/price ratio and reliability, and an increase in labor.