The drive system for Deadblow 1999 had four wheels, with one wheel in each corner of the robot. The reason for this arrangement is that if a lifting robot (such as Voltarc, a heavyweight robot with a powerful lifting arm) attempted to raised the robot, at least two wheels would be on the ground at any one time, allowing me to escape. I hate to watch two-wheeled robots being helplessly paraded around the arena. I used two 12V DC wheelchair motors, with each motor driving two wheels (one in front and one in back) connected by a drive chain. The advantage of this drive system, as mentioned above, is that you can't be immobilized by lifting. The tradeoff (and there's always a tradeoff) is that in order to turn, all wheels actually have to slip tangentially around the turn circle, which translates into a large amount of power wasted.

click on an image to enlarge it

This is a side view of the drive system. The front and rear wheels are driven together by a long #35 3/8" pitch chain. Note the position of the drive chain tensioner, which engages a large number of teeth on the main drive sprocket. Before adding this new tensioner, the robot would make a machine-gun noise whenever I tried to turn (which requires a large amount of torque).

This is a closeup of the drive tensioner, designed by crew member Jon Foreman. The main drive sprocket is on the left, and the tensioning idler, which spins freely on its own needle bearing, is on the right. The angled bracket on the tensioner goes down and mounts to the base. Tension is maintained by the horizontal jack screw at the far right. It pulls the idler away from the drive sprocket and two screws help secure it in place. Many people just use the two screws to keep tension, but the jack screw is the key to keeping your chains from loosening up over time.


This is a side view of the same drive system with the wheels installed. The black clamp in the middle of the picture secured a drive battery to the frame during the testing phase. The bottom plate is 1/4" 6061-T6 aluminum, which was marginal for this application. In fact, during battle, the frame would flex when I fired the hammer. The wheel hubs are also machined out of 6061 aluminum and the tires are filled with a rigid foam, so I didn't suffer from flats. Note that the tensioner design described above had not yet been implemented, and so during this phase of testing, the chain slipped horrbily because I did not have enough of it wrapped around the drive sprocket.

This is the front view of the drive system. You can see that the sprockets and chains on each side of the robot are in a u-shaped channel formed by the inner and outer bearing blocks. Since I needed 8 pieces, I CNC machined the bearing blocks for this robot. The 1/2" axles proved to be too wimpy, and Knee Breaker bent one of them at the conclusion of the middleweight melee.

Here is the drive system without the motors, chains, and electronics. You can see the 8 bearing blocks and 1/8" side armor mounted to the 1/4" base. With wheels at each of the four corners, the turning axis is at the center of the robot. This means that the wheels have to slip when I turn the robot.

The front and rear edges of the frame and armor have an angle cut at the bottom to assist with escaping from lifting robots. Here, the angles have not been cut, so you can see that if the robot was lifted from the front or rear, it would be immobilized on the bottom corner of the armor.