1511 has been working on assembling a prototype drivetrain that utilizes Vex Pro Ballshifters Gearboxes and a pneumatically controlled drop down “belly pan.” Four pneumatic pistons can push the belly pan against the floor and raise our wheels off the ground. The bottom of the belly pan has wheel tread to increase the friction between the robot; the intent of the belly pan is to make the robot “unpushable” once the pan is dropped down.
Why have such a huge reduction off the ball shifter if your just going to multiply it in the last stage? Couldn’t you just remove the external gear reduction ?
Just wondering, how much stress testing (at last years max weight) have you done with that belly brake system? Is it robust enough that you will start sliding before damage is done to your belly pan? Also, do you have a safety in case you lose pressure during a match?
IIRC, 1114 has used a similar chain setup in the past as far as tensioners and #35 chain. They use the biggest sprocket possible and nylon tube on a dead axle to act as tensioners that push on top of the chain in their sheet metal drive from 2013. Take a look if you have a chance. I did a lot of research on that bot and found a lot of the stuff they did to be quite simple and elegant.
As long as it isn’t soft, or overly grippy plastic this type of tensioning system works well. As with any chain system, it will wear down over time and need to be replaced.
The material is akin to delrin or HDPE, and eventually wear out, but not before multiple hours of run time have passed in a much harsher environment than FRC
We haven’t done much actual stress testing yet, though we have had multiple people try to push it on carpet with the pan down to no avail (other than the carpet moving). This design has no safety factor in case of pressure loss, but for our build season design (if the game is conducive to this kind of base) we will more likely go with four dropping “feet” and a static pan. It was a cool idea, but I’m sure it scares our electrical mentors considerably.
I also plan to experimentally test how much of a factor surface area is in terms of tread traction. Conventional physics says surface area has no role in frictional force, but that only really applies for flat surfaces, not ones with grooves like the blue nitrile tread we used on the bottom.
Were I to design a drive like this, I’d personally forget about the omni wheels on the corners and use normal traction wheels. At high speeds with a square or slightly long frame, a 6 wheel dropped center drive turns a little on the fast side, even without omnis.
If you don’t want to drop the center and want to use 4 traction + 2 omni’s, that’s cool too.
For an example of a drive which uses omnis to make it more manuverable than the average tank drive, check out 1625’s 2014 drive. I heard that weight distribution is important for something like that , though.
I’d also eliminate the dropdown belly pan brake. It’s a cool idea for a game where you really don’t want to go anywhere, but, as I said in the other thread, I’d rather make sure I have high traction on my wheels and a way of locking them (encoders and smart code?) For a game where alignment (like, say, to shoot) is key, odds are a robot’s momentum when it hits you will knock you far enough out of alignment that you have to switch back to wheels to realign.
Plus, “Our brakes are down and we can’t move” is a bad way to end a match.
The drop down belly pan is a good idea. We had a “friction plate” on 340 last year.
The key is spring return cylinders in case of loss of air and another thing is don’t lift your wheels off the ground. Have the plate push against the ground and as long as your wheels don’t come up and the stroke on your cylinders is small enough >1" they will be fine under side load. Looks like a great drive train.