Planetary Rotary Mechanism Design

Recently I’ve been designing a mini Swerve Bot for possible off-season event use as well as outreach use. Part of this entails a rotating arm. Due to the small size constraints (16.5"x16.5") I was forced to be a bit creative in how the arm mechanism rotates.

Overview of CAD:
The central gear that is along the main axle is fixed in position. This means that the two falcons(geared to 60:1 each with MaxPlanetary gearboxes) climb along the fixed gear as they rotate. Essentially, the arm is rotating its self around the shaft, not being rotated from an external gearbox.

I would love any critiques as I haven’t seen anything similar to this approach and don’t know its viability as a mechanism



I would caution against using gears as the final stage on any rotating arm that can see shock loads (e.g. extends beyond the edge of the robot). Use a sprocket instead. The issue with gears is that only one tooth is bearing the entire load, versus sprockets/chain which spreads the load much better. A shock load can easily shear off the gear teeth.


Several folks have used something similar, this season and in the past.

Try it with chain, for the reasons Peter suggested.

I’ll echo the thoughts above about using a gear as the final stage. Steel gears would likely fair much better, but aluminum is definitely likely to shear. Sprocket is the conventional and proven way to go here.

The bigger problem I see here is the long hex shaft. We built an alphabot in week 1 this year that looked very similar. The long shaft ended up as a torsion spring and made the arm very difficult to control. Eventually we bent the entire shaft quite significantly.
The better way we found was to minimize shaft length, and maximize cross sectional area. The way we accomplished this was by moving the supports all the way up to the arm, and driving the arm directly from the sprocket such that the force was going through the large bolt pattern, instead of through the small hex shaft.

Alpha bot

Comp bot


Probably not the best application to purpose using gears on the arm like this. We did something similar with our Infinite Recharge robot, using a gear on hex shaft to drive two arms from the base that supported our power cell scoop (we built a low-goal robot for that game.) It had the drive gears oriented the same way and, while we didn’t shear any gears (they were steel) we did manage to shear off the motor mount twice when the scoop was impacted and lots of extra torque was applied from that. The mount was 1/4" welded aluminum plate, by the way. One way or another, any serious impact is likely to break something, even if the gear teeth survive. This year, we drove our arm using the much wiser chain and sprocket system that most teams have chosen. We also had a break due to an over-application of torque (from a failure to stop the arm from driving itself into the ground due to a sensor fault) and we broke…one link of each of the two drive chains. Easily repaired and the arm was right back in action none the worse for it.

We did use a gear system like the one you’re showing this season though, on our wrist mechanism that connected the end effector to the arm. But there, any potential stress was perpendicular to the gear direction and the mechanism itself was relatively small and light. No real danger of shearing gear teeth off or twisting a shaft. It worked beautifully and even won us the Industrial Design award at one of our competitions. So there are roles for the type of gearing you’re looking at, but probably not for the application you’re contemplating.

By the way, here’s a picture of the wrist so you can see what I mean. The large gear is welded to the yoke that attaches the end effector to the end of the arm. The drive motor and it’s gear move around the large gear. The entire end effector spins with the central shaft (a hex shaft) with bearings in the yoke while the shaft runs all the way through the middle of the end effector.

First - this mechanism isn’t a planetary gear system, that is a different thing. Just a terminology thing.

Second, 1/2" hex is a versatile and decently strong option for shafts in many applications, but it is not a “drop in and use anywhere” solution. Long arms subject to large moments are a bad application of hex, especially since this appears to be an arm that could easily be on a dead axle and thus has no need for the hex! You can get an arm shaft that weighs the same, or less, but is far, far stronger in taking bending loads and such, by using a large OD, thin wall round tube. Alternately, the REV product line has MaxSpline which conveys similar benefits while also allowing torque transfer if that’s what you need.

As for how the mechanism is powered - aside from the points already made about gear teeth as the final stage for a long arm, that sort of “motors move around the thing” mechanism is certainly a valid approach. That said, chain is really the right move here - the load is distributed across several teeth and not just one, and chain handles shock loads much better than gear teeth do.

1 Like

Seen and worked with similar mechanisms. Biggest problem working with them was gear teeth breaking, but our application moved extremely fast and I dont know if it will occur with this design, depends on the speed of the arm. On one of our systems we used this same approach but two gears and a planetary was mounted on the sides as to not be directly in the middle and stress that shaft. Either move the triangle supports in, or spread the arm out and use two gears, one on each side.

Otherwise, looks good! Very creative

This topic was automatically closed 365 days after the last reply. New replies are no longer allowed.