Semaphore-bot

Last spring in an effort to graduate on time (it worked) I got one of my professors to agree to let me do an independant study which I titled "Physical Manifestation of Digital Information"

When my original project design hit some serious hickups that I would not be able to overcome in the time I had remaining, I quickly designed and built this: youtube.com/watch?v=9wEIciYcvJM

Semaphore-bot was designed to read junk emails (via a php script) and then display them by waving them out with a pair of flags using flag code (semaphore). The viewer however is not told the nature of the message being presented, but is instead provided only with a key for decoding the message. As they desipher it they reveal messages like "The Most Reliable Pharmacy Online Free 4 or 12 ViagraPills with any purchase its 100 percent FREE, no gimmick. Use the free pills to satisfy your woman" and the ever popular "dripping sluts are waiting for you" they are forced to wonder: "Was it worth it?"

Currently the bot is in his beta stages as, inorder to get a full 360 wave out of each arm I had to stack a pair of servos. I'm currently building a gear box to translate 180 to 360 (simple 2:1 gear ratio) so that I can run it from 2 servos instead of 4. The arduino is also being replaced by one WITHOUT auto-reset as that proved to invoke some bugs during long runs.

When I shot that video I had not properly calibrated one of the servos, so the positioning isn't perfect (the right arm is off by 1 position) but he should be spelling "Hello World" because really, that should be the first thing spoken by any machine.

More info here: andrewparnell.com/?/blog&id=1388191952829577084

I like your approach. Robot spelling out questionable material via semaphore flags! I read on this forum that you can modify a servo to rotate a full 360 degrees. I wonder if that would be an easier fix than the gearbox?

Really the gear box to grant a servo 360 range is 2 gears. Aside from that, the modification is not a good approach for my project as it would lose a very important aspect of a servo: precision. One of the steps in that modification is to remove the potentiometer in the servo (and replace it with a pair of resistors) the pot in a servo serves to track what angle the servo is at.

Very nice!

Probably too late by now, but a pulley/belt system might have been sufficient and easier to fabricate than a gear box... at least for prototyping.

Hrmm... you could be right on that front. I hadn't considered that.

Heck, two wheels with rubber tires and different radii pressed against each other can do the job if the torque is low enough. :D

Yeah, but at that point you might as well just use gears.

Would this have worked?http://www.budgetrobotics.com/shop/?shop=1&cat=157

It would indeed, and is exactly what I’m talking about when referring to building a 2:1 gear box.