This project is a bit of a random one…as if any of my projects aren’t, I suppose. But a couple of weeks ago, I noticed that the spoon that has been living in my sugar container has quite a bit more shine to it than any of its siblings.

That got me wondering if table sugar works as a polishing media. So naturally, the next step was to design and build myself a tumble polisher.

I figured all I would need is some sort of drum to put the part(s) and sugar in, and then a mechanism to continuously rotate that drum for hours at a time. Up above is an image of how that came out, and here it is in its Fusion form.

For a ‘drum’ I just went with a coffee can. It was available and seemed to serve my needs. My only real concern with it was/is the lid. It holds pretty snug, but the idea of it popping off and dumping sugar all over my table isn’t ideal. So I decided to just use some duct tape for now to give me some extra peace of mind.

And sitting under that coffee can is the fun part.

The can sits nested between two sets of rollers, one powered and one passive.

The passive side, shown on the right in the above image, has two 608-2RS roller bearings (aka roller skate bearings).

The below cross section shows the assembly a bit more clearly. The mount has M4 heat sets inserted into each end for attaching the wheels.

The ID of the roller bearings is only 8mm, and the OD on the heat sets I have is 6.3mm. So I don’t like to try and have this side of the interface sit inside of the bearing race. Instead, I went with a separate “Retainer” that has a snug fit on the ID of the bearing and sandwiches the bearings inner race between this retainer and the Mount.

On the outer race of the bearing I added some printed wheels. The primary purpose for these was to constrain my coffee container from moving axially. The wheels are shaped to register against the grooves that run around my coffee can. As long as the flanges on the wheels (they only are constrained on the outer race on one side) are on opposite sides for the two wheels, the can won’t be allowed to creep forward or back and fall off of my tumbler.

The one other feature I wanted to comment on with the passive roller side is the notch flexure.

This notch will act like a pivot point (see below simulation), allowing the axis between the two wheels to tilt. I want the drive drum to be in charge of defining my axis of rotation, so having this degree of freedom for the passive rollers allows them to conform to the axis defined by the drive roller.