The “Slam Dunk” is a device which brings students with virtually no mobility into competitive wheelchair basketball. Chris Marotta at the Henry Viscardi School built the original version which worked but he was interested in updating it to have a more stable hold on the ball and to use more readily available components. Here’s some video of Slam Dunk 2.0 in action:

The original Slam Dunk used a ball pushing mechanism. One of the goals however was to increase the grip on the ball while the user was driving their PWC around. One way to do this was a larger “bowl” that the ball would sit further down in. This however made a pushing mechanism more involved to implement due to how much further the ball would have to go to get all the way up and forward over the lip. I switched to using a dumping mechanism which was pretty simple to get going with a motor and a limit switch. Here’s the first prototype of that method:

I opted for a worm gear motor so that it would hold position without needing power. Only needing to power the motor while dumping the ball in this way increases battery life significantly.

The control electronics consist of a microcontroller waiting for a switch activation, controlling a motor driver, and monitoring a limit switch. The limit switch sets the home position. The dump position is arrived at through a fixed duration of movement away from home. Several alternate methods would also work. A second limit switch could be used to detect arriving at the dump position. Or an absolute encoder on the shaft could be used to locate both home and dump positions. Further, if using limit switches, instead of these being polled by the microcontroller, they could be in series with the motor (parallel to appropriate diodes for reversal) so that the motor would stop at its limits regardless of what signals the microcontroller is producing. In that case the MCU could just send forward and reverse commands to the motor and not worry about manually stopping the motor, which would be handled directly by the limit switches. There are pros and cons to all these methods. Here are some pictures of building the Slam Dunk:

bt

Here are some pictures of the Slam Dunk clamped to a wooden chair for initial testing: