Discussion Forum

Hello! Lately I have been struggling a bit with understanding how these formulation are actually implemented in COMSOL Multiphysics (v 6.1).

I am currently working mainly on modal and frequecy analyses, and since I am dealing with "very long" geometries in the out of plane dimension, but with a constant cross section, I have always used 2D elements combined with the plain stress formulation. Since I am working in the frequency domain, I have always enabled the "Out-of-plane mode extension (time-harmonic)", in order to be able to compute out-of-plane displacements and stresses associated with mode shapes properly.

Now I need to perform prestressed eigenfrequency analyses, in order to take into account the effecs of thermal expansion; hence, I am running a first Stationary step before the eigenfrequency analysis.

Now the real problem: from the results I got and the related documentation, it looks like the Plain Strain formulation with the "Out-of-plane mode extension" is the correct choice for the eigenfrequency analysis; however, when I perform the stationary analysis with thermal expansion, the plain stress condition generates a lot of stress in the out of plane direction (sZ), due to the constrained displacements. This looks to be solved with the generalized plain strain; on the other hand, this last option does not allow me to properly perform the modal analysis since out of plane displacement are set to 0, in-plane.

Hence, I would need to implement the generalized plain stress with the out-of-plain mode extension options. Does anyone have any idea on how to do that? Or am I missing something here?

One possible solution that came to my mind is to generate 2 solid mechanics physics, 1 for the stationary study and 1 for the eigenfrequency study; than solve the 2 steps with different phisics. But it looks like an overkill solution.

Thank you in andvance! Kind regards, Federico Sordo