## Universal Joints & Torque

Hello,

I need to select an appropriate electric motor for a drive mechanism which includes a U-Joint. It's understood that rotational variances occur in U Joints as the angle of the input shaft increases relative to the output shaft. Can someone refresh my memory as this relates to torque? In other words, is there a percentage increase in torque to drive the input shaft as its angle increases to say 15, 30, 45, 90 degrees relative to the output shaft that stays at 0 degrees? If so, what's the percentage increase at each increment - or at least a general rule of thumb?

Many thanks.

Bret
 Recognitions: Homework Help It's not so much the change in torque that is the problem with shafts coupled by U-joints. The life of the joint is greatly reduced once the angle increases beyond a very small amount, unless the torque applied to the joint is greatly reduced. For instance, if a given shaft and joint is rated to transmit 300 ft-lb of torque at an angle of 3 degrees, the same shaft will only be capable of transmitting 100 ft-lb of torque if the angle increases to 10 degrees. For larger angles, U-jointed shafts are not practical.
 Thanks SteamKing. Good thing that I asked. I seem to remember now about this limitation but didn't remember it to be as dramatic. I may need to reconsider the electric motor. However, don't you think that at least the wear issue could be reduced by going with lower RPMs , say under 100 RPM or even a manual crank?

Recognitions:
Homework Help

## Universal Joints & Torque

It depends on what kind of motor you want to use. If you have a motor turning at high speed (say 1800 rpm) and you wish to reduce the output speed to 100 rpm, then the speed reduction will also result in a torque magnification. For instance:

1 ft-lb at 1800 rpm = 18 ft-lb at 100 rpm

It's hard to suggest anything without knowing more about how your drive mechanism is situated between the motor input and the output of the drive mechanism.
 What I'll do then is jig a physical test with a torque wrench, create some resistance constant, and see what the readings are with each change of the angles. I'll post my findings.