Bearing Preload to Torque Equation 1

[Mandrill22]

Is there an equation to convert bearing preload to torque required to rotate the shaft from a static position? I'm looking at a precision angular contact bearing with an OD of 40mm (ID is negotiable) or less and I want it to provide 2 N-m of rotational resistance. Essentially, I want the shaft to be static unless a torque of more than 2 N-m is applied. If the OD would need to be larger, please let me know roughly how large. Not sure if a tapered roller bearing would be better

 

[geesamand]

You're poking at one of the trickier questions in rolling bearing tribology.

There are equations for rolling bearing friction vs axial preload. In my experience, the breakaway torque isn't much higher than rolling but predicting that in advance is a very rough estimate.

 

[3DDave]

Easy. Tell the assembler to tighten down on the bearing retainer until they get that resistance as measured with a torque indicator. And then never let the temperature change.

A spring-loaded clutch plate would be better as that will have a uniform performance over time and temperature.

 

[Tmoose]

With no information (pictures) provided as to your application I am left on my own as to what you wish to accomplish.

3DDave's recommendations are nice practical solutions.

You consistently refer to your "bearing" as singular.
I suspect you will need to use two bearings separated axially to support your device's shaft.
You will need two opposed bearings, with one bearing's outer ring capable of sliding axially to generate adjustable preload.
If the shaft is oriented anything but vertical whatever is afixed to the shaft may be eccentric ( CG off center) and thus tend to rotate the shaft by itself.
Your first choice of (pricey) precision angular contact bearing(s) are generally of lower contact angle (15° and 25°) than "normal" angular contact bearings (40 something °).
Does contact angle play into your undisclosed requirements?