Discussion Forum

Hi,

I have a 3D model with materials of different CTEs stacked together and I'm analyzing the thermal expansion of some metal structures inside them over a range of temperatures.

From the blog below - I see that the Rigid motion suppression domain constraint is a better one to use - than a 'fixed' constraint' : https://www.comsol.com/blogs/how-to-provide-structural-stability-in-thermal-expansion-simulations/ . My model structure is similar to this - though in 3D and having many stacked layers.

To choose the domains on which to set the 'Rigid motion', I ran a quick eigenfrequency analysis [with a coarse mesh and the bottom-most surface 'fixed'] as suggested - but that didn't yield any eigenfrequencies close to 0. In fact they all were in the order of E7.

In the end, I had to manually pick a couple of the domains to run this through - but the run times were pretty huge [6.5hrs] - compared to a run [with the exact same model] that had a 'fixed' constraint on the bottom surface [1hr] .

I'm still new to COMSOL - and my model only involves a Solid-mechanical + electrostatics multiphysics, so any pointers on what to check for would be good - particularly the eigenfrequency analysis. I looked through the forum for anything similar posted on eigenfrequency - but couldn't find any.

Thanks Anand