coefficient of friction

An "average" coefficient of friction is meaningless here. This is especially true if the mechanism has some nonlinear relation between the input forces and the output forces such as might be found in a linkage or cam system.

The friction force that presumably would have to be overcome is the summation of the friction forces from the three feet. For each foot it is the coefficient of friction times the normal force if the surfaces are relatively smooth and don't deform. The forces will be those exerted by your mechanism however it is contrived and actuated. Calculate those first. Only you know the design of your mechanism. We can't help there with the information you've given so far.

Your coefficients of friction for each material combination can be approximated from some of the limited available data on the subject in reference books or on the internet. This can get you through the first iteration of your design. But to insure that your device will work as intended you are best advised to do some testing. Such tests are simple and straightforward; but will generally require that you have some reasonably accurate method of directly measuring forces and possibly torques.

Ed Weldon

The coefficient of friction in that case is static, that is hgher than the dynamic that is referred in the literature, and it is dependent on the load force.

Presumably the 3 feet in the chuck are spaced 120° around the inside of the pipe and the movable foot only moves a small distance. The clamp probably has enough mechanical slop so the forces normal to the surface equalize for each foot.

I'd put a small load cell between the movable foot and the tube wall, clamp it up and find the force applied. Or if that's not possible, estimate the force using the systems hydraulic pressure/mechanical leverage etc.

The coefficient of friction between most materials can then be found by Googling.

To find friction, multiply each normal force by the appropriate CoF and sum.

(If the force on each foot is the same it would be sensible to use an average CoF).

Note, I've assumed that the force due to the tube's weight is small compared with the clamping forces, if it's not you'll have to use some vectors to find each force.

Hardly you'll need exact values:So take this technical approach:rarely between static and dinamic values are differents in more than 20%,2)high value like friction material is close to 0,35;steel-steel 0,12 or 0,15 up to 0,18;smaller some plastics..Isn't it enough?

I think that there's one mistake in the general reasoning so far: if we're talking about static friction then the normal force times the coefficient of friction gives you the maximum force before slippage. So the maximum force the device is able to withstand before starting to move is three times the minimum of CoF x NormalForce ... After that the thing starts to move (slip) and becomes much more complicated ...

Static friction is almost universally much higher than kinetic friction.

Why include this 'three times' factor. The static friction of a 6-jaw chuck with the same bearing surface area as his 3-jaw chuck is identical, but you would multiply the COF (which is most likely much lower than the static friction) by 6, thereby giving an erroneous result.