View Single Post
  #11   Spotlight this post!  
Unread 26-12-2014, 22:58
GeeTwo's Avatar
GeeTwo GeeTwo is offline
Technical Director
AKA: Gus Michel II
FRC #3946 (Tiger Robotics)
Team Role: Mentor
 
Join Date: Jan 2014
Rookie Year: 2013
Location: Slidell, LA
Posts: 3,693
GeeTwo has a reputation beyond reputeGeeTwo has a reputation beyond reputeGeeTwo has a reputation beyond reputeGeeTwo has a reputation beyond reputeGeeTwo has a reputation beyond reputeGeeTwo has a reputation beyond reputeGeeTwo has a reputation beyond reputeGeeTwo has a reputation beyond reputeGeeTwo has a reputation beyond reputeGeeTwo has a reputation beyond reputeGeeTwo has a reputation beyond repute
Re: pic: Proto Type Base Built

The bracket mounting the gearboxes looks like a flat bar. Does it have any extension in the vertical direction? If not, you may have some flex in this member in a head-on or rear-end collision. This could momentarily reduce the chain tension. Combined with the sudden change in the traction force this could lead to losing the chain from the sprockets. I would actually consider that bracket (or some supplementals) to be part of the chassis rather than just a gearbox/wheel mount since it provides the continuous front-rear loading.

Also, why the 1-3-1 wheel count per axle? Does it help traction that much, or do you figure that it'll reduce the fraction drive force that must be delivered by chain (vice direct drive)?

I'm also in agreement with philso as to feeling queasy about not keeping the wheels aligned in a diagonal collision, or more properly with not keeping the sprockets in the same plane and losing the chain. We have never depended on a control board for this; at least since our second year we either used triangular structure or large (8") flat braces on everything but our wooden prototypes. I wasn't in the mix our rookie year (Rebound Rumble), but the control board was vertical, so it wasn't providing stiffness. On the other hand, I don't recall many robot-to-robot collisions in RR, so we may have just skated on that one.

Thinking about the things that can go wrong with this design makes me like the simple pseudo-WCD* design we're using for our prototyping tank chassis - short of bending the main c-channel members, there's no way a bumper collision can mess up the sprocket alignment or spacing.

* Pseudo-WCD in that we're using cantilevered axles, but not direct driving the center shaft. We're also using wheel bearings on fixed-mounted axles so imperfect machining won't cause (as much) wobble. Along the same line, we're using COTS shifters: AndyMark Super Shifters, but the design is adaptable to Sonic Shifters with just two more holes in the chassis and longer chains.
Reply With Quote