Learning Forum

[AnhTu]

Hi everyone

I am doing a simulation problem in Ansys Static Structural using nonlinear contact.

The model consists of 2 parts as shown below. Right from the beginning, part A was tilted and came into contact with part B at point C.

I use Frictional contact with settings as below and try to run solve the model.

After solving I got the reaction force at the contact position from the model even though no force or displacement boundary conditions were applied.

Can someone explain to me why this is the case?

I have attached the project below.

Thank you.

[Erik Kostson]

Hi

Ansys employees are not allowed to download anything from the website - perhaps some other users/members of this forum (e.g., and ) can help out here.

One likely/possible reason is that you have penetration in the Contact which the solver will resolve by pushing the cylinder back.

All the best

Erik

[Rameez_ul_Haq]

I downloaded your archieve file and took some snaps from it. [I manually created a contact tool since it was not already there in your file].
Below shows the minimum, maximum and results of both supports and contact in your analysis:
As it can be seen that the max and min forces over the time for the Fixed Support and Remote Displacement are both equal and almost zero, this signifies that there is no external force acting on the structure. Same trend can be seen in the contact resultant force, which means that there actually is no force passing through that contact but those extremely small values are a result of the nature of numerical analysis techniques like FEA. In analytical world, you would get all the forces in all directions for above three as exactly equal to zero. But FEA softwares like ANSYS don't proceed in the manner like what we see in the analytical world. They have to numerically approximate the resultant force reactions and therefore, the external forces are not exactly equal to the sum of reaction forces.
Note that the X and Z axis forces under Remote Displacement are exactly zero since there is no constraint on them. But rest of the forces (in any of the support or Contact) cannot be exactly zero since there exists a constraint on them, but can still be approximated as zero since there is no external force acting on it.

[Rameez_ul_Haq]

,and I forgot to mention one more thing. The reason (according to what my mind says) that why couldn't you conduct an analysis without a force for a linear contact and you could for a non-linear contact might be that, as has already mentioned, the solver will try to push the cylinder away from the ring if there exists penetration more than the tolerance (before the analysis starts). This is when the force will automatically be developed from within the analysis without the need for an external force, and you would be seeing a reaction force coming from the contact. But since you already have no initial penetration (as shown in the contact tool figure I shared), therefore this is still not the reason why you are seeing a resultant force reaction at this contact.

[AnhTu]

Thanks for your answer
Can you explain more to me how the software calculates and generates resultant force reactions even though there is no external force acting on the structure?
Is there any way to eliminate or minimize resultant force reactions in such a case.

[Rameez_ul_Haq]

,a good question would be should there be any displacements in your model when there are no external forces acting on it? See the picture below.
So your model has displacement values on the cylinder even when the external forces are zero for this type of non-linear contact. If there are no external forces acting then the global matrix equation should not be able to solve for the displacements at all since [F] vector in the global matrix equation has unknown nodal forces and known external forces only. First unknown nodal forces and then unkowns reactions are calculated, after displacements are first calculated from the known external forces. Now, the displacements should be equal to zero (according to my understanding) since they are calculated from the external forces within the global matrix equation and external forces themselves are exactly zero. I would recommend taking opinion on this one that why do displacements exist in the first place when inputted external forces are exactly zero.
Anyways, if the displacements exist within the model sujected to zero external forces for any reason, it should also give a rise to internal forces and hence to the reactions. This is why you see a values for reaction forces within your model (even when external forces are exactly zero).

[Rameez_ul_Haq]

I have commented on your reply here but I edited something in my reply and it is pending approval for now.

[AnhTu]

Hi everyone
I changed the boundary condition of the problem. I have replaced the original displacement boundary condition with a fixed boundary condition. After solving I obtained very large contact reaction force results. I think it is not the error of the calculation, can you explain it to me.
Thank you!

[Erik Kostson]

As we saw, there is some penetration (0.01378 mm) in the contact as seen below. This is resolved during solution thus giving a reaction (parts are pushed away from each other to resolve the penetration)- try to move the cylinder away from the disc so they do not overlap and there is no penetration and there should not be any reaction (this is a quick test to show that it is the solution of the penetration that gives rise to the reaction force).

Erik

[Rameez_ul_Haq]

,I was thinking since the contact appears gray so there are no penetrations because if there are, then it should appear orange. But then I realized that it is for large gap or penetration only. For small penetrations, I guess it will still show gray. Thank you for the indirect clarification :)