Earlier this week, John and I had a design evening. A few things were accomplished:
I brought in a THB6064 stepper driver (from my CNC machine) and we plugged that in to the stepper. John had written an interrupt based arduino stepper driving program (complete with acceleration control!) and we tried that out. We proved to ourselves that the motors we’ve chosen can indeed get to the maximum speed we’re after, but we’re still not sure about what acceleration they’ll give us. We learned that smoothly accelerating steppers is important and that once they start to slip, you have to drop the step rate right down to re-capture the rotor. More to do here we think…
We had a discussion about how we wanted the robot to look (which has a bearing on how the chassis should be constructed). Likely answer (nothing is set in stone) is that we’ll have flat side panels with a nice profile cut out of the top instead of the flat plate we had last year. This should mean we can make the robot a little prettier and possibly lighter too. The downside is that its harder to make accurately, but if we use my CNC to carve out the side panels, we should be OK.
Material choice was also discussed – wood is nice and cheap, but heavy and needs painting. Metal is probably overkill, but we’ll certainly use it for anywhere we need to reinforce. After a bit of googling, I discovered PVC foamboard – this is a plastic foam sheet with machining properties similar to wood, but less dense. It can also be heat formed, so curves are possible. We’ve got a sample on order from ebay 🙂
Finally, we played skittles with the set we bought which was linked from the event page. This gave us a clue how hard the ball will need to hit the skittles – and now I’d be very surprised if simply pushing the ball to the skittles was enough to get a strike. Looks like we’ll need to design a mechanism to accelerate the ball… Looks like Leo White has been doing the same thing 🙂