UltraPlanetary Gearboxes Inconsistent Resistance

Technical Motors
1

This year, we are using REV UltraPlanetary Gearboxes for our drive.
Despite being extra careful with them, our robot still experiences sideways drift when trying to accelerate forward in autonomous.
We have properly tightened the bolts holding the gearboxes on so that they do not have any extra resistance, and have made sure that our axles have slight give so that they do not seize up the gearbox with lateral force. The motors are geared 12:1 (3:1 and 4:1 planetaries), the mecanum wheels are brand new, the robot only weighs 13 pounds, and the weight distribution is near centered.

This isolates the issue to the gearboxes themselves - so we tested the gearboxes by taking them off the motors and spinning their output by hand. When taking the front right motor apart and spinning the 3:1, it feels very low friction. However, the other motors’ 3:1s do not spin nearly as easily. All of the 4:1s also spin easily. However, even the worst gearbox on the robot spins much better than brand new gearboxes, which make an “oily” sound and are hard to spin. Our idea is that with use, the robot “breaks in” to the gearboxes, but we don’t want to rely on this because we have no control over how it happens. Because of this, we can’t even replace our gearboxes with brand new ones.
Is there any explanation for why some gearboxes spin better than others? Does anyone know what might be the underlying issue for why these gearboxes might spin this way? Is there any way to adjust the brand new gearboxes so that they all spin freely?

2

There are an enormous number of factors that can make that happen, from a slightly differently cantilevered shaft to slightly different amounts of friction in bearings or between gears.

Running in your gearboxes is a good idea: put your robot on blocks so it can’t go anywhere and drive forward for five minutes in one direction, then for five minutes in the opposite direction. That should help. (You might want to feel the gearboxes during the run and make sure none of them are getting extremely hot, which would (a) indicate a bigger problem and (b) possibly identify which gearbox is the problem.)

…but what you really want in order to be able to drive in the direction you want to during the autonomous period is PID. I don’t remember what it’s called, but there’s a built-in WIPLib tool you can use to tune the PID on your drive train, and it gets the values pretty darn good–you just need a place big enough to drive, and ideally FRC-ish carpet to drive it on.

(Also, for my money, I don’t know that I’d trust UltraPlanetaries on my drive train, and would use Max Planetaries instead. They didn’t exist the last time we used planetaries on our drive train, but back then we used BaneBots P80s.)

3

they’re part of an FTC team so it makes sense to use ultraplanetaries.

1 Like
4

Oh, whoops, missed that! Please disregard that part of my advice.