It started out as the Viper kit from LynxMotion. I noticed that they changed the Viper kit slightly just after I purchased it. The new kit appears to make it easier to change batteries.

This is the overall view of Phantom. It's pretty much the stock Viper kit, except that I machined up a nice wedge for the front. The 4 screw holes across the front secure the top to the scoop. The screws were not present in the picture. The wheels have a thin coating of RTV Silicon on them for improved traction. Phantom made its debut appearance at the International Robot Sumo Tournament in Seatte in March, 2004. Unfortunately the motors I selected didn't have enough torque - Sigh. Bigger...

Here's the top view. The red bush button is a reset switch. The black pushbutton is the start button. The leftmost toggle is for the motor power, and the one next to it is the CPU power. The third toggle is a run/start switch, and the rightmost toggle is a hint to determine if Phantom should turn left or right when starting.

When the run/start switch is in the start position, then the startup code pauses and waits for the black push button to be pressed before starting the 5 second countdown. If the run/start switch is in the run position, then the startup code doesn't pause. This allows the robot to not pause for 5 seconds if it gets reset accidentally. Bigger...

Here's a closeup of the electronics. I'm using the "Maxi Dual 10A H-Bridge" from the "Robot Store" (which doesn't seem to exist anymore), and the MegaBitty purchased from the "Mark III Robot Store". The MegaBitty includes a couple of H-Bridge chips, but they're only rated for 500 ma, so I didn't install them and just soldered wires from the appropriate pads.

The arrangement of the connectors on the MegaBitty wasn't convenient for all of the sensors that I had, so I created a little breakout area on a Radio Shack perfoard to rearrange the connectors the way I wanted them.

The MegaBitty has a 9v battery dedicated to it, and the motors have 10 NiMH AA cells. Bigger...

Here's a closeup of the MegaBitty, which gives more than enough CPU power for this sumo in a 0.9" square package. The connectors on the MegaBitty board have a 0.05" spacing, where the connectors on my breakout area and the perfboard are 0.1". Those are 0-80 screws in the corners. Bigger...

Here's the rear view, where the serial port (using a standard telephone jack) and the ISP (In System Programming) port are visible. The reason that the hole around the telephone connector is too big is because I originally used the handset connector instead of the normal line connector. Whoops. Bigger...

The MegaBitty only has TTL level RS-232 on it, so some type of level shifting is required. "HVWTech" made a very nice little adapter. Once feature that I really like about the HVWTech adapter is that by soldering some 0.1" headers in the 4 pins along the back and the two pins next to the "HVWTech" on the board, and the two pins in the same spot across the board, you can then plug it into a solderless breadboard. Bigger...

In order to fit on the aluminum scoop, I had to cut 3/4" off the stock scoop as well as cutting the tips off the vertical members. In this picture you can also see the mounting spots for the rare earth magnets. It turns out that with the motors I have and the softness of the wheels, I couldn't use the magnets. Bigger...

Here's a picture of the scoop prototype that I made out of MDF, alongside the finished scoop made from aluminum. The angle of the tip is 35 degrees. Bigger...

Here you can see the Jig I used to cut the rough 35 degree angle using the bandsaw, as well as the finishing cuts on the mill. It's made from two 3/4" pieces of MDF which have been glued together. Bigger...

I used a piece of plywood for my fence to do the rough cut on my metal bandsaw. Bigger...

Here you can see the rough cut being done. By the time I was finished, the aluminum was too hot to touch. Bigger...

Here's the line sensor that I'm using (one on each side - also from the Mark III Store). It was intended to be used as a micro-sumo line sensor, but I used one on each side. I just had to file down the sides a wee bit to get it to fit. The mount was machined from Delrin. I was originally going to put two slots for mounting, but realized that I only needed one in the center. This allows the distance from the sensor to the ring to be adjusted. Bigger...

Here's the sensor mounted on the mount Bigger...

And here we se it mounted on the sumo. Bigger...