Modelling circular saw blade

barrti

Hi

Does anyone have any experience modeling a circular saw blade?

I am looking at the stress caused by centrifugal force as well as impacts on the saw teeth not to mention the forced vibration resulting.

I suppose I am needing to look at a 3D non linear dynamic response analysis type although I don't think this will include centrifugal forces.

To model the impacts on the teeth as they hit the work, I was considering applying the impact forces to successive teeth and the correct time using the above analysis type.  This is a kind-of quasi static approach and I am unsure of its validity.

I am also unsure of whether to account for the fact that I am trying to analyze a dynamic impact situation in a somewhat static way.  What I mean here is that if the saw blade has considerable inertia (this one does) then as the impact accurs, part of the force instantly comes from the motor driving torque and part comes from the rotational inertia of the blade.

Any comments are welcome.

Thanks for reading!

Victor

Two main approaches:

A) Use a reference frame rotating with the blade.

Linear dynamic analysis is probably sufficient since there are unlikely to be any large displacements. It depends how much rotational vibration the shaft has because angular velocities in the model are changes in angular velocity of the real blade - if the real blade is rotating at a constant speed, then the model won't rotate.

I don't think there's any problem because of torque transmitted from both the motor and the teeth. If the motor provides constant speed then constrain rotation at the shaft's location. If it provides constant torque or some combination of them, constrain rotation through a torsion spring. You might also need to include the motor's rotational inertia with a lumped rotational inertia or lumped masses. It won't include angular acceleration driven by the motor, such as at start-up. But dynamic forces on the teeth will still be resisted by both the inertia of the blade and the torque at the shaft.

Include a centrifugal force load. Even though the angular velocity will be oscillating, it should be approximately constant so you can use a constant centrifugal force.

Use a non-rotating reference frame.

Conceptually simpler, may be easier to set up but computationally more expensive.

Requires nonlinear dynamic analysis because the mesh goes through large rotations, perhaps several revolutions.

Centrifugal force will be automatically included.

Drive it with a torque load at the shaft. Then startup acceleration will be automatically included. This might need a huge number of time steps to get it up to full speed though.

You might be able to model tooth impacts by running them over the surface of an object with friction or colliding them with some deformable massive obstacle. Again, this will be a huge number of contacts opening and closing and might take too long to solve.

In either case, you might need to check for out-of-plane vibration separately because the periods of those oscillations could be so much longer than the time for a few tooth impacts that you can't capture them in a small enough number of fine time steps.

barrti

Thanks Victor
I think I will try the first approach.
How would I apply a torsional spring at the inside diameter of the saw blade disk? Do you mean that I should use a beam connected to the disk by rigid links and rely on torsion of the beam?

Do you think that my proposed method of applying successive impacts to the teeth is valid?
Thanks again.

Victor

Yes and yes.

barrti

Thank you Victor.