Mecanum drive

[RobotRaider]

Just want to clear some things up and get some input.
For those of you out there who have built a mecanum drive i am wondering a few things:

1. Drive base weight
2. Max front to back fps
3. Max side to side fps
4. Did you use a super shifter on each wheel?

Any other details and advice is greatly welcomed and appreciated. Thanks!

[EricH]

Feet per second is dependent entirely on gearing. Side to side will be slower than front to back, though. Base weight would be a bit heavier than a 4WD 2 tranny system (2 extra trannies, and the wheels are heavier).

[scottydoh]

The use of a super shifter on each wheel seems a little bit overboard too.

[Ether]

The reason that the side-to-side motion is slower than front-to-back (for equal applied voltage) is that the rollers must spin for side-to-side motion, whereas they do not spin for pure front-to-back motion. The friction in the roller's bearings cause the speed to slow down.

High quality mecanum wheels with properly design bearings for the rollers will minimize this effect, but they would be quite expensive.

[RobotRaider]

Thanks for the input.
And I didn't say what i meant entirely. Obviously you wouldn't put a super shifter on each wheel instead with a AM tough box nano on each wheel.

As for speed i know that strafe motion is about half of your forward ability so my question becomes what is the max forward/back fps with a direct CIM into a tough box nano on a mecanum drive.

Sorry about the confusion

[Ether]

Originally Posted by RobotRaider

As for speed i know that strafe motion is about half of your forward ability

Half?? Wow, you must have been watching some bots with pretty beat-up wheels.

Quality mecanum wheels will have nearly equal forward and strafe speed, but they are very expensive. If they get hit and damaged so that the rollers don't spin freely under load, you'll start to see degradation in sideways speed.

[EricH]

Originally Posted by Ether

Half?? Wow, you must have been watching some bots with pretty beat-up wheels.

Quality mecanum wheels will have nearly equal forward and strafe speed, but they are very expensive. If they get hit and damaged so that the rollers don't spin freely under load, you'll start to see degradation in sideways speed.

Or if the rollers too tight, or are rubbing hard against the sideplates. Either of those will cause it. This is another of those cases where theory says one thing, and reality says to theory, "You need to come over to MY place tonight!" Just about every mecanum robot I've seen in FIRST is slower sideways than front-back. 357, 330's testbed, and many others. It's faster than 1/2 speed, but it's nowhere near full speed. Maybe about 3/4 speed.

That's just what I've seen at competitions; the root causes could be any of the ones we've mentioned, or something we haven't thought of.

[Ether]

Originally Posted by EricH

Or if the rollers too tight, or are rubbing hard against the sideplates.

I agree with the above, but will add this: a high-quality mecanum wheel's rollers will not be "too tight" or "rub hard against the sideplates". The rollers will have properly-designed bearings that can withstand the high axial thrust loads and still spin freely with no binding or rubbing. That's what makes them expensive.

This is another of those cases where theory says one thing, and reality says to theory, "You need to come over to MY place tonight!"

There's no reason that theory cannot include a consideration of the effects of friction. In many complex systems, a proper theoretical analysis of frictional effects is absolutely mandatory for developing a proper design.

[EricH]

Originally Posted by Ether

I agree with the above, but will add this: a high-quality mecanum wheel's rollers will not be "too tight" or "rub hard against the sideplates". The rollers will have properly-designed bearings that can withstand the high axial thrust loads and still spin freely with no binding or rubbing. That's what makes them expensive.

Which is probably 50% of the reason they aren't found on FRC robots.

The other 50% is that very few teams, if any, know they exist.