[frereubu]

I went on a solar power course at the Centre for Alternative Technology in Wales (~10 years ago, so perhaps this into may be somewhat out of date) and they'd calculated that the extra energy generated by moving a panel to the optimal direction during the day was less than the energy it took to move the panel, let along the extra costs of the mechanics. Bear in mind that in Wales there's plenty of cloudy weather during the year, so light is often diffuse, and it's quite a long way north (the gulf stream tricks people into thinking the UK is further south than it is, but the CfAT is 600 miles further north than Toronto). If you're somewhere with lots of direct sunlight that calculation might be different, but as others have pointed out, that doesn't take into account the cost of the mechanics and controllers.

[tpxl]

Can you make a purely mechanical solution to the tracking problem, at least along a single axis? If you put a water container on the left and the right side of the solar panels, the one in the sun will get hotter, expand, and that can move the solar panel. It's moving parts and extra cost, but no energy consumption.

I'm no engineer, so I can't determine whether it would work, but on the surface it looks like it should?

[dwild]

I remember a guy in a community that I frequent that build a small trailer that had fixed solar panels, he had build one for a reason I don't remember and afterward some companies were interested and that post was showing his second one he had just sold. It was a bunch of panels on an A frame so essentially one side had awful sun visibility but not the other. People wondered why he did it in such ways, as that meant that half the time they would be useless, why not a moving platform on a central pivot that you can adjust? Well turns out it was cheaper to add panels on the other sides than building a more complex frame that allowed pivot. I'm sure it could have been cheaper to build a system sure (maybe not in his case considering it was on a trailer), but I have an hard time believing the maintenance cost and failure potential is worth it. You can just add more panels... they are starting to become that cheap.

[hetspookjee]

I can imagine that this kind of research is done once 15 years ago, as I remember that too, but perhaps with the recent advance of technology might actually not be true anymore. I don’t have sources but in general some types of things get reiterated so long that time overtook it

[wongarsu]

As GP argued, the movement can be entirely precomputed, and even if it isn't "move the panel to face the brightest spot" is trivial to solve with simple components. The only real advances I can imagine to make this profitable would be better motors that are cheaper and use less energy, or new types of solar panels that are more directional (e.g. a solar concentrator setup where you move mirrors to hit a smaller collector).

The latter exist but don't seem more profitable (in the case of PV panels because of lifetime problems due to more heat), and while we have gotten a bit better with motors I don't think there even is a lot of headroom to gain much efficiency.

[thechao]

Y'all have plenty of water in Wales: it feels like a simple water clock could do a trivial repositioning of the panels to do precomputed solar tracking.