Tuesday, August 22, 2006
this post has been a while in the making. it started when i got my first issue of make magazine about 4 months ago. i really, really wanted to try the kite arial photography bit.
after taking a closer look at the article i realized that the author looked like he knew his way around a wood shop a lot better than me. on top of that, he used a disposable 35mm camera that would be triggered in about 3 minutes based on the vicousity of silly putty. i didn't like the idea of sending the kite up, taking a shoot, reeling the kite in and doing it all over.
so i had an old 2MP camera just lying around and tried to think of some way to get that to take pictures while up in the air. my first thought was to build some type of mechanical arm that would swing around and push the shutter button. i even went as far as to buy a power ranger spinning lollipop to take out the motor. ultimately there wasn't enough torque on the motor to actually push down the button. on top of that, the arm spun around way too quickly.
i started talking to a (brilliant) co-worker about the idea, and he said he could wire up something digitally to do the trick. the first phase used a 555 timer and actually didn't quite work. then he gave up and went the overkill route -- he programmed a $.99 microcontroller to send a signal every 90 seconds or so. i wired this to my camera and everything seemed to work. if you are interested, here is the ridicously over-commented source code for the microcontroller.
i finally got a chance to go out to the coast this weekend to try it out. i purchased a very large $30 kite that was made for pulling long tails. i thought that would be close enough.
the wind was great and the kite finally lifted off the ground with me running (i'd tried this many times at the park down the street from my house). overcome with excitement and without really thinking i was ready to put the camera on the kite.
i wound the camera harness around the kite string and off it went. for some reason, my plain boring non-stunt kite was flying all over the place. so the camera was also swaying all over the place. crash #1: the camera swayed itself over 180 degrees and wrapped around the kite line. estimated flying time...oh roughly 25 seconds.
not discouraged (and certainly not learning anything from the first crash), i got it up in the air again. i had it up there for about 2 minutes or so, but was getting worried about it swinging around. so i thought i'd bring it down gracefully.
crash #2: i didn't bring it down gracefully (and i actually got an action shot of this). i looked at the pictures on the card, and didn't see anything from the air. not about to drive home without an arial shot, and still learning nothing from these 2 crashes i got ready to send it up again. i had the harness in my hand, and then a gust of wind came along. the kite took off. miyoko and todd ran after the kite, but the nylon string holding the camera onto the kite snapped. so that was the end of the camera flying.
after we got home, i noticed the camera was no longer working. it was easy enough to tell that the camera had sand in the lens. you could hear the poor motor trying to extend the zoom, but there was too much sand jammed in there. no problem, i thought, i can clean this out.
well, i did clean it out, but i also broke some rather important wires inside the camera while doing this ("see all those wires in there, homer? that's why your robot didn't work"). talking it over with my (brialliant) co-worker, he seems to think he may be able to fix it.
but all was not lost. while doing all this i noticed that there were more pictures on the memory card than i thought. i stuck the card in my computer and found that i had 1 shot from the air, and one dramatic shot of the camera hitting the ground.
here they are:
so what did i learn?
1) tails give a kite stablity. (who knew?)
2) nylon string breaks. my dad is sending me fishing line he used to use to catch sharks
3) sand is bad
4) i'm not brilliant