Moe-Ron is named after a character in the now-defunct Freakazoid cartoon series who arrives from space to deliver an important message to Earth, but cannot recall the message until it is too late to do anything about it. For those of you who aren't following: He's dumb. My Moe-Ron is so named because he's pretty dumb, too. He has no processor. In fact, there are no semiconductors in his control circuits -- he is controlled by switches, cams and relays.
Robot Sumo is an event in which two robots attempt to locate their opponents and shove them out of the circular Sumo ring. (rules) The two main differences between Sumo and the robot fighting one sees on television are that A) Sumo robots are much smaller -- Moe-Ron is approxmiately 3 pounds 3 ounces B) Sumo robots are real robots -- they fend for themselves during competitions. There is no radio control or human intervention of any kind.
Moe-Ron is driven by two of those Ax-Man gear motors with the brass gears that many of us seem to like so much. The wheels are made of three-quarter inch plywood, and the tires are made from mouse pads. They are connected and driven by ten Resin gears.
There are gears mounted to the shafts of the motors using J-B Weld. I used a dremel to grind a slot in the hole in the gear, and to grind away about 1/3 of the motor shaft to make a flat spot. I then pressed the gear onto the shaft with the gear's slot and the shaft's flat spot lined up, and filled it in with J-B Weld. At first, I just drilled the hole in the gear a bit too small, and then pressed the gear onto the shaft. This appeared at first to work, but as I started applying any amount of force to the system, the gears quickly started to slip. I then cut a slot in the inside of the gear (identical to the final version) and then cut a similar slot in the side of the motor shaft. I also cut a slot across the tip of the shaft. I filled the slot with 90-second epoxy, placing a big glob on the top which was intended to stick to the gear and flow into the slot across the tip of the shaft. This was better, but still ended up giving out. So far, the J-B Weld solution (described first) is holding.
The remaining gears (and the wheels) are mounted on axles made from 1/8 inch piano wire purchased from Scale Model Supplies. To get everything to line up, all four of the panels that run front to back were screwed together and then cut, drilled and sanded simultaneously. There are 1/16" x 1/4" x 1/4" squares of polycarbonate glued over the tips of the axles to keep them confined.
This is a pretty light treatment of this subject. If anyone wants more information, please say so -- but I hate to write something that nobody is going to read.
I made molds and cast the gears using orginal gears purchased from Scale Model Supplies in Saint Paul, and molding and casting materials purchased from Synair.
Both the molds and the casts are made from fairly straightforward two-part mixes.
The molds are made from Synair's Por-A-Mold S-333-Medium. I made the molds by lining a cake pan with aluminum foil, rubber-cementing the orginal gears to the foil, and then pouring the mixed molding material over the gears. This seemed to work fairly well, though it took me two attempts to get the air bubbles down to a point I considered satisfactory -- I didn't care except when the bubbles occurred on the faces of the gear teeth. The two parts of the Por-A-Mold should be mixed thoroughly but gently to keep the bubbles at bay. I also used a toothpick to force the material into the teeth and force the bubbles out. In earlier attempts, I taped the gears to a tongue depressor and lowered them into the mixture from the top. The problem here was that **lots** of air got trapped under the gear as I dipped it, leading to big quasimodo bumps on the finished gears.
The gears themselves were cast from Por-A-Kast Mark 3. I first sprayed a few coats of Synair SynLube 531 release agent into the mold to prevent the cast from sticking. I let it dry between coats. The one thing that isn't made totally clear in the directions is that Por-A-Kast Mark 3 gels *VERY* suddenly. Only a few seconds pass between the time it starts to gel and the time it's too solid to work with any more. I suggest mixing no more than you can finish pouring in about a minute.
I originally tried casting the gears from clear casting material, but (as we found out at the TCRG meeting where I originally talked about my efforts to make gears) they were far too brittle. For those who weren't there, one of the gears got dropped, and about a third of several of the teeth came off. My experience in the workshop was similar -- any attempt to drill, sand, or grind the gears to the exact proper shape resulted in large pieces coming off. Too bad. Clear gears would have been cool. Hmmm, I wonder if it's possible to cast polycarbonate?
Note that all Synair products have an advertized shelf life of only six months. Keep this in mind when buying. If you have any trouble purchasing Synair products, please contact me -- I can help.
The basic idea that drove the entire project was Moe-Ron's sense and attack behavior.
The robot has a bump sensor on each of his four sides. Each bump sensor is a lever switch mounted facing outward in the center of the side of the body, and actuated by a large plastic plate hinged at the top and hanging down covering the entire side.
Moe-Ron has all-wheel drive. The two wheels on each side are locked together, and the robot therefore turns as a tank would -- by driving the two sides of the chassis in opposite directions. Moe-Ron either spins left or right, or drives forward or back. There are no in-between movements.
When any one of the bump sensors is activated, Moe-Ron drives towards the object touching it. (Either by driving forward or backwards, or by spinning in place if hit from the side.)
This is accomplished as follows:
When a corner of the robot falls off the edge of the ring, we want the robot to back away from the edge no matter what else is going on: