CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Structural Mechanics (http://www.cfd-online.com/Forums/structural-mechanics/)
-   -   Sphere Deformation Simulation (http://www.cfd-online.com/Forums/structural-mechanics/127838-sphere-deformation-simulation.html)

amingh December 23, 2013 06:43

Sphere Deformation Simulation
 
2 Attachment(s)
Hello,
I am trying to simulate the deformation of a sphere which is placed between 2 cylindrical plates and is in contact with them. The plate in the bottom is fixed by FIXED SUPPORT and the plate on the top is under a SURFACE FORCE. I want to simulate the deformation of the sphere. I have couple of questions:

1. How can I give different mechanical data such as young's modules for each of the 3 bodies SEPARATELY? because I don't want the whole geometry to have the same mechanical properties.

2. Since everything is symmetric in my simulation, I expect to have symmetric results, but as you can see in results I get some deformation in the top plate which is not symmetric, I wonder why it is like that????

Thanks for any help:)

flotus1 December 23, 2013 08:10

Your physical model may be symmetric, but your computational model is not.
With the Boundary conditions you described so far, your model is under-constrained and Ansys will add weak springs to prevent the solution from diverging.
You have to add boundary conditions to your model to make fully-constrained.
A simple solution would be to actually make use of the symmetry and model only a segment of the full model. With symmetry boundary conditions at the new boundaries you can already constrain some degrees of freedom which should be enough to make the model fully constrained.

amingh December 23, 2013 08:19

Thanks for your answer, I just wanted to fix the displacement of the plate to prevent it from deformation, but it didn't work.

amingh December 24, 2013 09:16

Quote:

Originally Posted by flotus1 (Post 467413)
Your physical model may be symmetric, but your computational model is not.
With the Boundary conditions you described so far, your model is under-constrained and Ansys will add weak springs to prevent the solution from diverging.
You have to add boundary conditions to your model to make fully-constrained.
A simple solution would be to actually make use of the symmetry and model only a segment of the full model. With symmetry boundary conditions at the new boundaries you can already constrain some degrees of freedom which should be enough to make the model fully constrained.

I have fixed the problem with symmetric deformation, but still I do not know how to give different mechanical properties for each part separately, for example mechanical properties of gold for the sphere and diamond for the plates. (I am using static structural toolbox in ANSYS.)

Florin February 28, 2014 04:52

Hello amingh,
You just have to create the materials in Engineering Data (you must know the properties) and when in Mechanical, you can assign them to the desired geometry.


All times are GMT -4. The time now is 11:24.