Cheaper solar power collector array
Solar panels are just too expensive. This(edit: I put the wrong link in somehow, corrected now) site says the cheapest are $1.55 a watt. In Ottawa here, for a very modest 3.6 kwhr per day ( a normal house uses 24-36), a 1kw array gives you about 2.4 kwhr in winter (search “solar atlas”, also “solar NREL data” gives some actual data, I have to look into it, but hourly data would be really nice), or $2325. And that’s the very cheapest panels, who knows if you could actually get them.
A while back I looked into whether or not using a mirror or tracking system would reduce costs, and by how much, and how to do it inexpensively.
The increased output varies by quite a bit, but it’s usually about 40% more for a sun tracker, and obviously for mirrors it depends on the configuration of mirrors. What I meant, though, is something fairly cheap and easy to build, and which can make use of normal solar panels, since I’m assuming high temperature solar collectors would end up pretty expensive and hard to obtain. Is it economical?
Well, for mounting a tracking system and mirrors on top of an ordinary roof, a whole lot of factors come into play, mostly damage during storms, installation costs, and that most people wouldn’t consider them “attractive”. Unfortunately cosmetic appearance is critical in a tinyhouse, even more than elsewhere, with the uphill battle against bad zoning (still meaning to write a post about that).
But if you could accommodate that somehow, or if it’s a relatively remote area, especially if you can just put them on the ground, or you plan ahead when designing the house, so it is easier to attach the system… The numbers are pretty clear. A combination of tracker and mirrors should be easily less than half the cost, even less in highish latitude like here.
For a tracking system, one of the problems with commercial systems is that they use these fancy actuators, and they use another one for each 600W or so of panels. What if you connect them all together with cables (like fishing line you can get some pretty strong stuff that is very hard to stretch, like dyneema line, which is about as stretchy as steel!) so they move in unison with a single actuator?
Plus, once you have a tracking system, you can use mirrors effectively. Again, wish I could draw this in 3D. Drawing it in 2d in paint probably wouldn’t help much.
Anyway, consider the figure one on this page. Now consider just the left half of the figure. You have the mirror and the PV panel – why not put another mirror, doing the same sort of reflecting, on the other side of the panel? And 2 more above and below the panel? Ignoring reflector efficiency, that gives you 4 times more sunlight on your panel. I don’t know why they don’t do this.
In january, in Ottawa here, we get about 350 watts peak per square meter of sunlight, (perpendicular to the sun, i.e. the optimal configuration to intercept sun). You multiply that by 4, and it isn’t going to damage the panel or anything. The standard test conditions are 1kW/sq meter. The ambient temperature is obviously very low, so it can dissipate heat fine. If the temperature gets too high in summer, if you don’t want to remove some/one of the mirrors, the sun tracking system can detect that and tilt the whole thing so that the panel is shaded a bit more.
Yes, it increases the temperature of the panel a bit, so the efficiency goes down, but you’re still going to get way more power out of the panels. I wonder about the possibility of using infrared reflective film material over the panels to reduce heating. Or maybe IR transmissive mirrors. Or just a fan directed at the back of the solar panel.
But what about a really cheap way to do the sun tracking system? Oh boy, sun tracking systems are really expensive, there’s no way that could ever be economical, some people will say. I don’t believe a word of it.
Suppose you take a mounting system a bit like the one shown in that page’s Fig. 2. It has to be adjusted, ideally, manually with the seasons, but who cares. A little exercise twice a year isn’t gonna kill you. Use some door hinges (well greased) and some wood. Then you need the actuator and the tracking electronics. To keep the cost down, use only one actuator, with the cable tethering thing. The tinyhouse’s programmable logic controller (search this blog for a post on that) can do the sun tracking. Not only does it have a clock in it, but it can measure the the output from the panel to make sure it is optimal. The tracking does not have to be very accurate. The actuator? Well, seriously, all you really need is a gear motor and a spindle. Get some rope, and have it wind up the rope using the spindle (like a winch), ideally with it winding out the rope at the same time, if you see what I mean. If you wrap the rope in the opposite direction on the same spindle, you can get it to both release rope, and to take up the other end of the up rope at the same time. This allows you to exert some tension on the rope without exerting a net force on the gearmotor (the force being exerted to pull the rope off the spindle is canceled out by the force to pull the other end of the rope off the spindle, resulting in no net torque) . You need some tension to prevent any wobbling, maybe you could avoid the need for a spring tensioner by using relatively stretchy rope.
I have not seen, with a fair bit of searching, any reason anywhere why this would not work. You could surely get 2 or 3 times as much energy harvested each day, when you combine the 40% increase from tracking, and multiply with the hypothetical 400% from reflectors.
A tracking system also reduces the need for batteries, because sunlight is available sooner in the morning, and later in the evening. Especially for a relatively large collector array, like for a tinyhouse village, it could save some serious money over an unassisted array.
Edited to add on 2010/11/11: I remember seeing a document with an insolation vs. power out put graph for a solar panel. Looked again, couldn’t find a graph like that, which would easily show the fallacy of assuming that reflectors couldn’t work.
Also found a fairly good article about the greenhouse gas emissions associated with solar panel production, which obviously using reflectors helps with: http://www.lowtechmagazine.com/2008/03/the-ugly-side-o.html