universal joint needle bearing

Yes, it is proportional to the sinus of the shafts angle.

Thanks for the reply. So lets say, the shaft is rotating at 1000 RPM and the angle is 30 degrees, can i say that the needle bearing is rotating at 1000SIN30 = 500 RPM?

I'm probably going to have to sit in a quiet place and think this one through but surely the needle bearing rollers oscillate rather than rotate? The oscillation may well be more than a full rotation of each roller but it would necessarily be the same in both directions for one rotation of the shaft, the speed and the magnitude of the roller oscillation would be proportional to joint angle as well as rpm. So a cardan shaft running straight would produce no movement in the roller bearings in the UJ's no matter what it's rpm.

The total rotational movement and velocity of the needle rollers would then be proportional to the joint angle, cross width (some are uneven, vertical being different to horizontal) and journal diameter as well as the rotational speed of the assembly. Nice little calculation that one.

Take a worn unit apart and you will see that they do not rotate, only oscillate.