New Clothes

Our new tracks and running gear arrived from China in double-quick time and as soon as they did we were straight on to designing a rough chassis to fit them to and to get testing. This was done in double quick time and the parts were cut by the Tech teacher at Harry’s school. A happy evening with tacky pva glue, clamps, a spanner and Allen keys led to a sleeker version of KEITH than we had previously seen. The wheels from China consist of two stainless steel ‘rims’ with spacers between. The drive cogs bolt to hex hubs which fit the motors. The free-running wheels have bearings on each side and run on bolt axles. I modified the hex hubs to reduce the overall width of the assembly (deft use of the Dremel to chop out unwanted bits of brass!) to ensure that everything fits within the 225mm width limit.

Smart Wheels! Smart!

The tracks, as we feared when we ordered them, are a hard plastic. They are manufactured for one of the radio-control tanks on the market and work extremely well on soft surfaces (carpet, dirt etc) but they slide on hard smooth surfaces. Not what we want for PiWars! That said, they are better than the Vex tracks used last year in that respect. We have a few ideas about how to add grip and will be trying those out in due course.

We installed the Pi, the makeshift motor drivers and RC receiver then plugged in one of our new LiPo batteries. Harry fired up the code and off we went! Wheelspin (trackspin?) on the kitchen floor and drifting round corners was fun but perhaps not as precise as we need! Turning on the spot is very fast and the prototype ran well on carpet too. All very promising. The kittens were very interested and either followed it or just sat in the way. No doubt they will learn to interfere properly as testing continues…

Work on the drawings for the competition chassis is now advancing. Changes required from the initial prototype have been made and the skittles-flicker has been tested. Sensor locations have been sketched in and ideas for the golf challenge attachment have been drafted out and would appear to be workable. The aim here is to fit a single servo into the front of the robot that will do two things for different challenges, but without needing different code or any changes in electrical connections. That was a lesson learned in the first PiWars when KEITH did not do as well as wanted when we relied on attachments. Last year KEITH Evo had everything built in. That proved reliable, but we had all sorts of difficulties in fitting everything in and making it work practically.

Harry has tested his glycerol-based ideas and they look positive. It now falls to me to try to hack various bits of hardware without breaking it :-)

We are also really pleased to have been invited to a workshop on 11th February in Bradford where we will meet a number of local PiWars teams. The aim is to swap notes, hints and tips with KEITH, KEITH Evo and possibly the prototype KEITH 3 (Mostly Harmless) strutting their stuff!


If you haven't already, don't forget to follow us on Twitter @KEITHtheRobot for updates on when new blog posts are live!