Wave speed test for different Poisson's ratios

[johnh]

Dear FEAP Admin,

The geometry of the model is like a thin pizza box with few number of element through the thickness direction.
The dofs of out of plane is not constrained (mimicking the plane stress condition).

I wrote two input files:

1) E= 1671430000, nu=0.3
2) E= 131471000, nu=0.49

Theoretically, those have the same P-wave speed V_p = 1500m/s, but the second case with nu=0.49 gives about one third of the expected velocity we obtained by nu=0.30.

It is somewhat surprising.

Could you suggest any remedy to resolve this issue?

Thank you.

Best wishes,
John Hong

[FEAP_Admin]

Can you try this also with the 2D element and let us know what happens.

[Prof. R.L. Taylor]

You are seeing the plane stress wave speed.  The wave reflects off the faces and changes the response.  If you restrain your mesh on the y-faces you will see the plane strain response which is what you want.  It looks good for both cases.  Also, your input forces are not consistent the central node (30452) should have double the force so that the input is a uniform line load.

If you want to see the P-wave in a 3-d problem you need dimensions in all directions so that wave reflections do not cancel the front.

[johnh]

Dear Professor Taylor,

Thank you very much for your help.
We will run 2D cases and others you recommend, and will post the result.

Best wishes,
John Hong

[johnh]

Dear Professor,

As summarized in the attached ms-word files, the reduction of the wave speed is observed in both 2D and 3D in plane stress condition, compared to the wave speed in the plane strain condition.

Thank you for your help.

Best wishes,
John Hong

Input files : http://aem.kaist.ac.kr/temp/Input.zip
Results (2D) : http://aem.kaist.ac.kr/temp/2D_results.docx
Results (3D) : http://aem.kaist.ac.kr/temp/3D_results.docx