DEWBOT IV Drive Train
DEWBOT IV possessed Team 1640's second (and likely last) 2-wheel drive (2wd) drive-train (the rear wheels are driven). It has a 2-speed transmissions powered by 2 CIM motors each to provide both speed for laps and fine control for catching and placing Trackballs. Driven rear wheels allow us to recover by reversing should we become hung up on the Overpass.
The team's original plan was to use (6) wheels with only the middle (2) driven. Front & back wheels were 8" Omni wheels (AndyMark part am-0048). Middle wheels were AndyMark 8" Performance Wheels (am-0105). Traction for middle drive wheels would have been assured by the slightly greater diameteer of these wheels (with all axles on the same plane). This 6-wheel configuration was never fully completed.
The Team's 2-Feb weight crisis forced some changes, including the change from a 6-wheel to a 4-wheel robot. 2wd was retained, with the now rear Performance wheels driven.
Arcade steering (single joystick on the Operator Interface) was employed.
We used AndyMark Shiftboxes with (2) CIMs per side. Taking into account 12-tooth drive sprockets and 42-tooth driven, overall gear reduction ratios were 14.6:1 and 37.3:1. Maximum robot speeds (@ max efficiency) was 11.4 and 4.4 ft/s (in high & low gears, respectively). Low gear was needed for placing Trackballs on the Overpass at the end of the match. High gear was needed for knocking Trackballs off the Overpass at the beginning. Both gears were needed & used.
In retrospect, the 4-wheel 2wd drive-train was a poor decision. Inconsistent weighting of the rear drive wheels made DEWBOT IV extraordinarily difficult to drive. This was made even more problematic when we purposefully shifted weight forward to avoid wheelies (and flipping over on our back) when the lift was raised. Due to some mentor changes, the 2008 drive-train ended up being more of an afterthought than a purposefully designed system. Up to this point, the team had not spent a lot of effort to clearly understand drive-train mechanics.
This clear drive-train performance failure inspired the team initiate drive-train prototype testing during the summer of 2008.
We used the KitBot chassis, shortened to allow us to fit the starting envelope with the Forks in Park. Issue was forced by mentor loss. Wheelbase was therefore short: 21" (long) x 20" (wide).
- 2wd was simple
- 2wd with differentials and angular front-wheel steering makes a lot of sense for cars. It is very energy efficient and also "rubber"-efficient
- some weight savings
- It seemed okay for Overdrive
- In retrospect, it shamed us into really learning about drive-train mechanics in a rigorous & disciplined manner
- What we learned in our drive-train education: never have undriven wheels!. Undriven wheels which support significant weight reduce traction.
- 3 points define a plane - but we've got 4 wheels! 1 wheel was often unloaded in practice. Too often, this was the Left-Rear wheel. This compromised our ability to turn left. A big problem in Overdrive.
- Our efforts to move weight forward to avoid obvious wheelie problems exacerbated the steering problem by shifting weight to the (undriven) front wheels.
- Reliability issue - original chain master link clip was put on outside of chains. This would catch from time-to-time on the rim of the performance wheel and we'd lose the chain. With one chain the robot (at best) drove straight only. Discovered and fixed at Philadelphia.
- Maintenance issue - drive wheel treads wore out quickly and were the very devil to replace
- Maintenance issue - there was no provision made to raise or lower individual wheels to put all (4) in-plane
- Maintenance issue - No thought given to bumper removal & installation. Mentors could get old waiting for this to happen. Students could get older trying to accomplish it.
- 2wd is generally a poor drivetrain choice
- The original 6-wheel 2wd would have probably been superior to the 4-wheel 2wd built in terms of steering and agility. Only negative - if we hung up on the overpass, there would have been no recovery with middle wheel drive.
- Due to uncontrollable changes in the team structure, the drive-train and chassis design ended up being the "poor cousin" next to the Fork-Lift-Catapult. This was an unfortunate situation which we must endeavor never to repeat.