Needle bearing allowable axial misalignment and small angle rotary oscillation.

[cybanical]

I've been digging through the catalogs but have had some difficulty finding the allowable axial misalignment of a needle bearing (drawn cup or otherwise). I've got an oscillatory application (~10degrees motion) that has limited radial space, and requires a yoke to reach to either side of the oscillating component. Though it's just two parts, due to the yoke design I'm a bit concerned that there may be a bit of misalignment of the bearing housings due to machining. I thought perhaps I can thin the yoke to allow some flex in that direction (stiff in the loading direction). I'm curious if anyone has a suggestion regarding either allowable misalignment or allowing for misalignment in this type of application?

there had been a tad of conversation about this in this thread(thread404-229350), but more so about oscillatory motions for plain bearings. I was a bit surprised about the igus recommendation for oscillatory motion as I guess I would've thought creep of the polymer bearings might become a problem.

 

[tbuelna]

The first thing to consider with a small diameter needle bearing subject to limited angle oscillatory motion, is that just a few rollers support the radial load, and only pass back/forth over a small sector of the race surfaces. This type of operating condition usually leads to short fatigue life with needle bearings. Most commercial needle bearings have some amount of profile crown on the rollers to reduce edge loading. But the large L/D ratio of needle bearing rollers does not allow for much angular misalignment capability between the inner/outer races.

What you should consider is that with the operating conditions noted for your bearing, a plain bearing/bushing will likely provide better efficiency and service life than a needle bearing.

 

[cybanical]

Thanks.
What about starting torque (stiction) between a plain bearing and needle? You noted higher efficiency expectation for the plain bearing due to shorter radial distance to the applied friction for similarly rated component. But I wonder is the static friction or roller inertia greater force? I am using this in a force measuring application, though my guess is these force components are negligible compared to the large radial load being applied.

 

[tbuelna]

cybinal,

You mention this is a rotational joint for a force measuring device, so it seems that minimum/consistent friction characteristics would be a concern. One thing to remember with your needle bearing subject to oscillatory motion is that at the moment the rotation reverses direction the contacts can experience some sliding, which will affect your friction torque. Here's a link to the very good SKF technical web page covering rolling element bearing friction.

Your idea of using a thin flexure is a good one if you can make it work. This approach is used in Formula 1 to replace the spherical bearings at the inboard attachments of the A-arms. The flexures provide the limited angular movement required in the vertical direction and high stiffness required in the other directions, while eliminating friction of the spherical bearings.

One other approach you might consider if your design permits, is a rocking contact pin joint. The most common application of this design is used to minimize friction in silent (inverted tooth) chain joints. There are also some examples where the concept is used to minimize friction of engine joints with similar motion to your application.

Good luck with your project.

 

[cybanical]

Oh, great recommendations. I'm looking into the plain bearings now.

I had never heard of the rocking contact pin joint, though I have been considering something similar for another application that uses straps to enforce the no-slip rolling condition. Thanks for that recommendation, it's good to see its use in industrial applications.