Mechanum Gear Ratio Recommendations

[TubaMorg]

We've always just gone direct drive straight from the tough box (12:1) with no problems.

[bakejame]

Thanks everyone for your suggestions. I guess I need to order some larger sprockets for the wheels (we only have 26 tooth or smaller). We really want to use the CIMple Box due to its lower weight. With the 4.8:1 GR out of the box and a 12 tooth driver sprocket, I need to get a 28 to 30 tooth sprocket at the wheels to get close to the 12:1 GR.

By the way, how was it to strafe sideways at the 12:1 GR? Did it still perform pretty well?

[dtengineering]

Here's what 12:1 banebots direct drive to 8" AM mecanums looks like.

http://www.youtube.com/dtengineering#p/u/1/gyn9e0cT3-A

Heck... that thing even picks up tubes. Add a minibot and you're good to go.

Jason

P.S. If we rebuilt that machine all over again, I'd be tempted to consider the 16:1... the 12:1 had lots of speed, but I think the 16:1 would offer better acelleration and more precise control. I'd also use the Jaguar speed controllers to implement closed-loop speed control on the controller so you can take the PID loops and encoder reading out of the programming process. Not that it is hard to do that in labview... but back in the PIC days you actually had to have students who knew how to program in order to get 4 PID loops running! With labview and/or the jags, you just set your coefficients and go back to building the minibot.

[bakejame]

Originally Posted by dtengineering

Here's what 12:1 banebots direct drive to 8" AM mecanums looks like.

http://www.youtube.com/dtengineering#p/u/1/gyn9e0cT3-A

Heck... that thing even picks up tubes. Add a minibot and you're good to go.

Jason

P.S. If we rebuilt that machine all over again, I'd be tempted to consider the 16:1... the 12:1 had lots of speed, but I think the 16:1 would offer better acelleration and more precise control. I'd also use the Jaguar speed controllers to implement closed-loop speed control on the controller so you can take the PID loops and encoder reading out of the programming process. Not that it is hard to do that in labview... but back in the PIC days you actually had to have students who knew how to program in order to get 4 PID loops running! With labview and/or the jags, you just set your coefficients and go back to building the minibot.

We are definitely using the Jaguars, but will be using CIM motors. The larger motors should give us the extra torque needed for acceleration and, as a result, more precise control. By the way, I really liked the arm design you all came up with. Nice compact design.