Halloween Spider Drop

{ [ ] Projects >>> [Hardware] - [Software] - [Class] - [Other] - [Failed] -|!|!|- [Resources] - [Biography] - [Contact] - [ ] }

Halloween Spider Drop

Note: the text below is in response to a Halloween contest that the project was going to be in a submission in. This project was made in October 2008, and the original thread can be found here.

Well, I tried. Unfortunately, it wasn't strong enough.

My idea for OBCs contest was to have a human controlled spider thread with a spider hanging on the end. I live in a dormatory, and we have a central area four stories tall (~40 feet) with stairs on either side and a central light shaft in the middle. The project design was to have a person wave their hands in front of some IR devices, and that would control the vertical movement of the spider, and how much it swings. This way, the spider has a full 2d plane of movement. Also, a special scenaro (when a hand got really close to the vertical sensor) would drop the spider suddenly.

You can see the center cross here, where the swing servo is mounted. This is just before I took it down.

I'd just like to take a moment to say that my code works perfectly. Rather, the problem that I ran into was my mechanical. The servos that I used (Parallax standard and continuous) don't have enough torque (47 oz in), and so it created problems. My swing servo controlled the swing by running the thread that the spider hangs from through a tube, which was then attached to the servo. This servo was out in the middle of the space, and mounted on some 1/2" angle iron for stability. Anyway, too short of a tube, and I don't get swing, too long and it can't handle it. Problem 1.

The second problem was the vertical movement servo. It's quite slow under load, and it was a dull experience to raise the spider. To suddenly drop the spider, I mounted the vertical servo on another standard servo, which turned to drop so that the string would come off the side of the drum. Imagine a roll of string. Stand it up, and it all falls down. Anyway, apparently the friction is just a little bit too much, and it didn't go out like I wanted. I could have filed it down, I guess, but there were enough other problems that I didn't feel the need to.

The electronics consisted of a PPDB in side an old shoe box, with two IR LEDs and two IR recievers. Also, I put in two 2.1mm Barrel jacks so that I could provide seperate power for my swing servo and my vertical servo. I had tested earlier, and found the regulator on the PPDB getting hot.

As for safety, I took several precautions. Truth be told, I didn't really get any sort of approval for this project, and it looked a bit dubious with the liberal use of zipties. So, I took safety serious. All components out in the air are very light, with gently rounded corners so that they are fairly safe if it should fall. Next, I put lots of sting out to the center. There are seven completely independent sets of tie downs: two along the x axis, two along the z axis, one tied to the cross point, one tied to the servo counter balance, and the spider thread. Each of this were securely tied with a half hitch and a square knot to various railings.

My control box with the three power supplies and the vertical servo and drop servo mounted on a BOE-BOT Chassis.

So, that was to be my submission. Next time (Christmas?) I'll try it with stepper motors or maybe DC motors. Best of luck to everyone else!

Attachments
(Right click and "Save Link As...")

TestIrDutyDistanceDetector.spin