Looking for 1/2" id flanged bearing .25" thick or less (with flange)

[Bill Gold]

Originally Posted by Travis Covington

Have you? I didn't think so...

Nope... but I did make you a few 3D models for the hell of it

Originally Posted by Travis Covington

I shall reveal my design to no one!

Um... I saw your design... I gues my username should be "No One"...

[Dick Linn]

Travis,

I'n no engineer, so take this with a grain of salt. I'm curious as to what you are working on this year that is so critical with respect to bearing eccentricity and the actual difference between a reasonably high-precision ball bearing and a decent bronze bearing. [Especially considering last year's effort without bearings ] Bronze bearings aren't all that bad, mostly depending on lubrication, of course. As far as friction goes, sealed ball bearings do need to come up to operating temperature, don't they? Do they achieve this in a match?

I would think losses in other stages of the drivetrain/chassis would be more significant than this (alignment, gear losses...). I can see being worried about the bearing stiffness on a 300 MW generator (like this: http://bpb-co.com/bpb/articles/2Q01bently4.htm), but not on a robot that operates all over the spectrum for 2 minutes or so. At any rate, I'd appreciate any practical info with respect to bearing applications and the like.

[Travis Covington]

I am no engineer either. But I can try to explain, at least how I understand it as a mechanical engineering student, the need for roller bearings.

Ignore the comment about eccentricity and frictional losses as that was not the primary reasoning behind roller bearings.

I will start by telling you the solution we came up with last year was not nearly as elegant as I had hoped it would be. More recently the gears have started to wobble on the shafts (ever so slightly). Obviously, that is not good for wear on the gear teeth nor is it good for frictional resistance. The reasoning behind roller bearings vs. bronze bearings is that there is a very small overall width. If it were to be two bearings, with a shaft running in between them at a distance of an inch or two, I would agree that bronze bearings probably would work well enough for our purposes. But in this application (a technokat style shifting tranny) the 3/8" gear face is the total width that the bearing is supporting. The actual gear itself has to be kept perpendicular to the shaft (that is moving at a different rate than the gear) solely through the presence of a single bearing. So, in this particular case the roller bearing is better suited as it does not rely on the materials frictional coefficient or lubrication to run true and smooth. The roller bearing has tighter tolerances and is more likely to keep the gear perpendicular to the shaft with a lesser amount of contacting shaft area. A gearbox that is 3-4" wide with a shaft in the middle can easily use bronze bushings on the 2 side plates because the combination of the two bearings, at a distance of 3-4" easily keeps the shaft perpendicular to the side plates and gears. This case is quite a bit different.

I hope I made sense of this, but you can get a better idea if you download Andy Baker's whitepaper on team 45's 2003 2-speed shift on the fly transmission.

I know alot of other people have worked with similar things and maybe one of them can comment a little more on the reasoning behind this particular example.

[Dick Linn]

Makes sense that you'd want some sort of needle bearing, or a plain bearing and some thrust bearings if your design allows it.