pic: Differential Freewheel Clutch

This is a little project of mine that I’ve been working on.
In short, it’s a passive, non-shifting PTO intended for use on a climber. This is capable of passing the 6 CIMs of the entire drive train to another mechanism without shifting, and still allowing the robot to drive as normal.

The differential takes in both sides of the drive train and pushes them to a single shaft, while still allowing them to drive each side independently. This is transferred to a freewheel clutch. In a bicycle, this is a freehub (the thing that lets you backspin/freespin the wheel without pedalling). The freewheel clutch only allows forward motion through (i.e. the mechanism will only engage when both sides of the drive train are going forward).

The freewheel clutch is then passed into a pair of sprockets. One sprocket is used for ratcheting (can be ignored if not using on a climber), and the other is the output for the mechanism.

The CAD is missing a few things, like springs and bolts, but otherwise it’s mostly complete.

We did something similar for our climb this year…shiftless PTO. We skipped the diff and just used two sprag bearings (freewheels), one on each side.

If I’m not mistaken, this is not quite correct. Rather, the clutch transfers motion iff the net rotation of the differential is forward - i.e., if the average of the rotational velocities of the two sides is forward.

I’d love to see the CAD of this so I can figure out how it all works. We may want this mechanism up our sleeves for future years.

Very cool! I’ve never thought of using a differential to do this before. I like the use of Vex bevel gears in this over custom ones.
I’m a little worried about the clutch sprocket thing you have. Sprag bearings like what ajlapp suggested could make this a bit easier/ less complicated.

You might be able to swap the bevel gear diff to a staight gear diff to make things a little less complicated.

This is correct, but I couldn’t find a proper way to word it. For it to transfer power needed for climbing, however (assuming your ratio is such to warrant it), you would need both sides going forward at a considerable throttle.

I wouldn’t worry about sprockets as ratchets. As sourcing a ratchet gear from here is tricky, to say the least, we’ve historically used steel pawls (the little rectangle bits close to the tangent) and 50t gears without any issues. Switching to a sprocket makes the gear less likely to eat the pawl and will hold just fine. If you’re really worried, I would gear it for speed before the differential, and gear it back down on the output.

The goal of this is to make something that I could create without needing to source special parts. The enclosure for the diff can be welded (or made with brackets), the hub and reducer can be made completely using a drill and (optionally) a lathe. While I doubt I would use this on a real robot, I might 3D print a smaller 1/2 scale prototype as a toy or something. The model you see above can completely fit in a 5in box.

Could you post the CAD for this please? It looks amazing and I would like to take a closer look to understand it better :smiley:

I’ll post a step file when I get back to my computer tonight

Thanks! If I do anything with it I’ll be sure to show you :slight_smile:

CAD available here. GrabCAD does a surprisingly good job with it’s inbuilt 3D viewer.


Thanks! Can O pm you if I have any questions?

Of course