Our driveway is on the northwest side of our house, and the sidewalk beside it stays shaded and icy even when the front walk has melted dry.
Put a 2′ diameter mirror on the west edge of the roof to reflect sunlight down onto the sidewalk, with servos to aim. Photosensors could detect whether the sky was clear enough to be worth tracking the sun, or it could just do it anyway. Sweep back and forth along the length of the sidewalk on a one-hour period, to melt and dry the whole thing evenly.
It could use solar declination and time of day to calculate aim totally open-loop, or use photovore behavior to track the sun and then calculate the appropriate angle to aim at the sidewalk, or put sensors beside the walk to run as a remote photovore, or put a camera by the mirror and watch where the bright spot is. Or recognize that I only need it to work in the winter, assume a known solar declination, and write a fixed program for solar tracking and sidewalk sweeping.
Two addenda:
I’m talking about a flat mirror, not a solar death ray. One sunlight is enough to clear the front walk; that’s all I want to add to the driveway walk.
I’m not a mechanical engineer and I can’t quite picture it, but it seems to me that it might be possible to make one motorized photovore to track the sun, one motorized sweeper to move a pointer back and forth aimed at the sidewalk, and attach the mirror to a geared angle-splitter that bisects the angle between the two.
You can bisect the angle using a linkage of four bars in a diamond. One joint is fixed, two are attached to the two trackers, and the fourth points where the mirror should. (It also moves in and out, so you need a slotted fifth bar which the mirror is mounted to the fixed end of.)
If you’d really rather use gears, then you can do it with three bars, the outer two of which have meshing gears at their pivot points.
If these descriptions aren’t sufficient, let me know and I’ll build examples from Lego. Or draw diagrams.
Interesting… I’m playing with the same idea for solar cooking and heating. I’m using an Arduino board with a couple of LDRs as a voltage divider then strapping that assembly onto the top of a servo. The servo is then swung to balance the LDR at a value of 512 (centre). I then know where the sun-servo is as an angle so do some simple math to drive another servo carrying a mirror. With a pot to determine where I want the reflection I …. is this making any sense?
Without some math beyond my ability I’m only bouncing the reflection off within the plane of the track of the sun – not tracking a sidewalk in 3D!
If I get somewhere with it – and get some pics online, I shout.
[m]
I built a small solr concentrator out of a small satellite dish.
I wanted to motorize it but never got around to it.
Check out this great link for a bunch of solar trackers though!
http://www.redrok.com/electron.htm#led3x
One of the problems that you’ll encounter is that snow is quite reflective. Thus, most of the solar energy you end up directing onto the sidewalk will be reflected back into space.
I’m wondering if you might not have better luck with a modified version of a system like this:
http://news.yahoo.com/s/ap/20071231/ap_on_hi_te/solar_roads;_ylt=AuEFouXxz16nP8MRlInTJMms0NUE
In the extreme case, you could chip up the existing side walk and pour a new one with heat pipe built into it, with the collector powering them located in a sunny spot, but even a retrofit version which simply collected heat from a sunny location and blew it (via heated air?) onto the sidewalk might work reasonably well.
Dave
Hi, Dave!
I shovel my walks, and the sun does a nice job of drying the front walk and melting any residual slush. I’d be delighted to achieve the same level of meltage on the walk beside the driveway, which in theory should be possible.
Saw this on hackaday, sounds very similar to your sidewalk melter idea:
http://hackedgadgets.com/2008/12/06/sun-track-reflects-light-into-your-home/
Jin, very cool! Thanks a bunch for remembering my project idea and coming back to let me know about this!