Actuators, Structural Design, and Overall Mechanics

General Comments

The mechanics of our robot were our failing. Our robot was too top heavy, causing it to be unbalanced. When driving straight, our robot would kind of waddle. If we had a mechanical engineer on our team, he would have probably recognized this flaw in our design quickly. Alas, we didn't realize the flaw until too late (or too lazy to rebuild anything).

Geartrain

We were so proud of our geartrain after we had built it. We stopped with our 8th iteration. It was compact and fast. We used a 27:1 gear ratio. Each geartrain had three motors for maximum power. In addition we used the smaller 2.5 inch black tires. Too bad our robot lost all speed after we loaded the robot with more parts, the Handyboard, and batteries. In retrospect, we should have used a higher gear ratio and larger tires. However, the bracing around our gearbox was unbelieveably strong.

Shaft Encoders

These things were both good and bad. We had them mounted on the axle just under the motors, which meant they were spinning at 1/3 the speed of the motors. This placement gave us good resolution, but the high resolution was quite unnecessary. By the time our motors reached the desired number of ticks, it still took another 10-50 ticks to stop the robot depending on how fast our robot was initially moving. We felt the shaft encoders were good for getting the robot into the general vicinity of going straight, turning, and going where you wanted to go. However we would not recommend using them for anything too accurate. Use wall following or aligning with the wall for much more accurate movements. One last thing we did was shield the shaft encoders in a cardboard box. The lighting in 26-100 was very strong and could easily throw off the readings if they were exposed.

Ramp to lift up ball

Another part of our robot was a ramp device that could lift up a ball. When the ramp was down, we could jam a ball between the ramp and a wall. Then we would raise the ramp causing the ball to roll into the body of our robot. To power the ramp, we used a servo. We screwd a 24-tooth gear directly onto the servo. Then we used that gear to power another 24-tooth gear that was connected to the ramp. Not having the servo directly connected to the ramp gave use a few problems. Mainly there was a great deal of play between the two gears and if there was too much stress on the ramp, the gears would just give. We would hear these painful clicking noises as the teeth just slide past each other.

Overall, though, the ramp worked fine. Under most conditions it lifted the ball. Occasionally it lifted the ball too well and would flip the ball over the wall. We placed a small bump sensor on the ramp to detect the presence of a ball. In our strategy we could use this sensor to tell us whether or not the robot should make another attempt at picking up the ball. Oh yah, the ramp could be used as a faux switch, which made our robot bounce like the ghetto pimp ride it is.

Arm collection system

We built a very cool arm. It was sturdy as heck. During practice runs we would have our robot bang its arms down on the table. It made a loud 'thud' noise. In addition to being sturdy, the arms had a gate in the front. The gate allowed balls to come into the arms, but not out. We initially planned to use this design for plowing over balls and containing them. By contest time, this feature was completely ignored and we concentrated on using the sturdiness of our arms to bash things.

The arm was connected to a servo on one side and a free spinning grey connection thing on the other. Having the arm powered on one side made it a little unbalanced, but it worked. The arms lifted and lowered very reliably.

 

Back to the Main Page