FEA Simulation Isolated Surface Distortion

[268318]

I am currently designing a telescope mirror cell.  I need to have the surface of the mirror deflect less than 5 nanometers in order to preserve optical quality.

In order to determine whether my design is sufficient I am utilizing inventor's built-in FEA simulator.  My problem is that, when I go to see the displacement, the majority of the displacement is due to the bulk rotation of the mirror as seen below.

Does anyone know if there is a way I can view the distortion of the mirror face rather than its overall displacement?  Thanks in advance!

 

[blair]

I would try the Vector Specific direction. I can't tell which axis would be perpendicular to the mirror face in your picture. Somewhere on the web, is a piece of code written for this to determine optimal mirror cell design. It was web-based and you just plugged in the data.

[268318]

Thanks for the advice, unfortunately it does not seem to have worked in this case.

Looking at the Z-Axis displacement I get the following:

This does not seem to me like a very useable result.  The displacement seems to match the parabolic contour of the mirror surface rather than warping with the applied stresses.  I would expect something more like this:

Image from previous simulation with fixed contact constraints:

Let me know if you have any ideas on how to get better results.  Thanks for the help!

[admaiora]

Have you tried witha  shell analysis?

[268318]

No I have not yet tried a shell analysis.  I have not used shell analysis before so I will have t o go learn how to use it.  Ill let you know how it works out.

Will it let me view the small deformation rather than the overall motion of the part?  As far as I know shell only helps with simulation speed but as I said I have never used it...

[blair]

If you want, you can Pack-and-Go the files to me and I'll run it through Simulation Mechanical. blair dot stunder at arcticmfg dot com

when I'm not at work, you can reach me at: www.pgrasc.org

[raviburla]

Hi,

By : " .... I can view the distortion of the mirror face rather than its overall displacement"

Do you mean that you want to eliminate the rigid displacements and just look at the deformations? This would make sense if the model is relatively thin ( so that it is similar to a beam or shell). For those cases, even though strains are small there could be large rigid (stress-free) deformations. However, there is no way to isolate the elastic deformations from rigid deformations in Inventor Simulation. 

But, have you tried looking at "Actual Deformation Scale". This is probably what you are looking for. By default Inventor Stress Analysis shows deformations scaled so that the max displacement is 10% of the model bounding box.

Please let me know if you have more questions/comments.

Thanks,

Ravi Burla (Autodesk)

[268318]

It is my understanding that the "actual deformatin" simply scales back the visual distortion effect.  What I am looking for is not that.  My problem I was having was that the mirror was moving as a whole relitive to the constrained components.  The overall motion is not of concern to me only the deformation of the glass itself.  What I would like is the ability to see the distortion relitive to the mirror geometry itself while using the interacting componenets only to determine the forces applied to the mirror.