pic: 4587 Swerve 1.0

The issue is that 6800’s tend to be larger than silverthins, espiecally the thickness, and that makes it weaker once you stick it into a 1/4" plate. I would like silverthin-esque bearings for very large IDs actually for an offseason project I’m working on (either a turret or some large swervey thing) because 68xx bearings can only get so large, around 6" max.

I found 3.5”ID x 4.0”OD x 0.25” bearings for about 25$ a piece not including shipping. Search for KA035CP0 on alibaba and you should find it. Let me know if you guys have any more questions. I’ve emailed other suppliers from alibaba as well to shop around for prices more results will come in the next few days.

You way want to look into the X contact versions of bearings. They are rated for thrust loads and momentum loads as well as radial loads.

I’ll spend an extra $80 for a robot’s worth of parts if it means I get them in 2 days instead of 2 weeks, especially during a build season.

Sure, you can order things ahead of time, but I like knowing if I need more of something during the season I can get it quick.

How does a thicker bearing result in it being weaker? Just counterbore the 1/4" thick plate you’re recessing the bearing into and then just clamp the outer race of the bearing to the plate? I really don’t see an issue here.

If you want a larger overall diameter then I can understand going to a different series of bearing, but while we’re talking about swerve modules, the 68xx series is just fine.

If I apply a sideways torque to a bearing that’s sticking way out of a plate compared to one that’s completely captured by the plate, it seems like the force would be more likely to crush the plate where it sticks out, but maybe my intuition is mistaken.

We ran like this for our entire 2018 season and the offseason before with no issues. No wear at all on the bearing or the plate. We had 3 small blocks equidistant around the bearing that clamped the bearing to the plate.

Do you have pics of this?


A true “X contact” bearing has a “gothic arch” path geometry, compared to a simple radius profile used in a traditional radial contact bearing. The gothic arch path geometry allows each ball to contact both the inner and outer races in 2 places. If you draw lines thru these contact points and the center of the ball, the lines form an “X”, which is where the bearing gets its name. They are also referred to as “4-point contact” or “X-type” bearings.

A 4-pt contact bearing behaves like a simplified version of a duplex pair of angular contact bearings. Compared to a comparably sized radial contact bearing, they will have considerably higher capacity and much less axial and moment/tilt “play”, making them much more appropriate for the azimuth position on a swerve drive. This would also be beneficial in cases where you have mating gears that mesh via the bearing. Tight internal clearances are also important in this regard. I don’t know what you get with a “low cost” bearing, but you can hide a lot of imperfections in a bearing by making it looser.

(full disclosure) I am a Product Design Engineer for Kaydon Bearings/SKF. Kaydon invented thin section bearings about 60 years ago, and I chuckle and cringe when I see people refer to a “genuine Silverthin” bearing as compared to a Chinese copy (they are all copies, IMHO). Kaydon may not be the “low price leader”, but our quality standards and engineering support allow us to participate in some extremely cool projects.

What sort of loading do you all expect the azimuth (steering) bearing on an FRC swerve module to experience? “Normal” loading seems rather mild, but the shock loads could be incredibly high for some scenarios (like dropping the robot or getting hit from the side when your wheel was up against something, etc.). I would like to do a load analysis to see what sort of stresses are generated.

I’ll try to grab some within the next week.