Learning Forum

[javat33489]

Hi all.

I'm a little confused.

I'm doing a calculation of two cylinders, where one moves inside the other (it has protrusions, with which it touches the outer cylinder). I apply force to the inner cylinder, with this force it will stop on the sides.

 

I set the force on the end of the cylinder, but the problem is not solved, no matter what contact there is (any settings) and even if I set 100N, the calculation still ends in error and time goes backwards. I also tried grinding the mesh, but that didn't help.

 

But if, under the same conditions, I use displacement at the same end of the cylinder, the calculation works perfectly. Why?

What important factors need to be taken into account in order for the calculation with force to end successfully?

 

I also tried using light springs, this helps for a while, but then the error occurs again, the load of 22 tons is very high and the springs fail:

 

I can’t apply symmetry, because then I will need to bend the entire structure radially from the center by 1 degree and also move the inner cylinder by force.

[Kaushal Vadnere]

 

Hi,

Applying displacement as a BC instead of Force aids in convergence as specifying displacement means the solution is defined at each time point, making it easier to converge. On the other hand, when a force is applied, the program may struggle to solve the case if the connection slips, as there is no stiffness to carry the load.
See if this post on similar issue helps: Inputting displacement to get reaction force results work but not the other way (ansys.com)

Thanks,
Kaushal

 

[mjmiddle]

Are you using frictional contacts? They will need a normal pressure to develop any tangential frictional force. I don't see anything in your model to do that. If the outer cylinder radius is slightly smaller than the inner cyinder radius, then you have a press-fit type of model that will have a normal force on the contacts.

Or if you fix one end of the inner cylinder, then the force will cause the cylinder to expand and there will be a normal force on the contacts. Of course, the inner cylinder is already constrained from motion, then, which may not be the intent in your model.