Elmer Discussion Forum

Hello, I have a strange behaviour I have never experienced so far! Some of the .f90 files build from the fem/src .src files contain errors: at least, each time there is '//', the end of the line is deleted in the f90 file. For example, the line interface operator(//) in iso_varying_string.src writes...

Hi Peter, Making a mesh with 715 element type using ElmerGrid -triangles -increase seems to work now. The point now is that element 715 are not recognized and I got the following message for the Stokes solver: WARNING:: GetElementType: Element type code 715 not found. Ignoring element. WARNING:: Loa...

Hello again,


I have also tried to start from a 706 element types mesh and do
ElmerGrid 2 2 cube -increase

I then get 715 type element in mesh.elements but the boundary condition file mesh.boundary contains 0 type boundary element...

Regards
Olivier

Hello, Is it possible to generate element 715 with ElmerGird ? I did try triangles = True Element Degree = 1 but I get element 820... and a message ElementsToTriangles: not implemented for elementtype 408 If I add Element Innernodes = True then I have the following message CreateKnotsExtruded: not i...

Hi Denis,

The problem is in the Material part of the sif file.

Seems that "temp heat capacity" is not well defined (may be it is just a Real missing before the value ?

regards
Olivier

Hello,

What are the Stabilization possibilities for elements 808, 820, 706, 715 for the Navier-Stokes solver between
- Stabilized
- P2/P1
- Bubbles
- pbubbles (which p, b values ?)

Thanks
Olivier

Hi Juha and other, I did some more tests and the difficulty to converge when using the Newton scheme is certainly due to the non-linearity of the power-law and not the cutting for low strain rate. Indeed, using the Carreau law, which do not have this cutting, does not improve the convergence. Adding...

Hello, Using the last version, I did test the Newton method for the non-linear iteration of the Navier-Stoke system. In my case, I'm solving the Stokes equation (Flow Model = Stokes) and use the power law rheology with an exponent 1/3. I found that when it is converging it is just faster than the Pi...

Hi ElmerUsers, I'm wondering what are the available stabilization for the Navier-Stokes equations: - for linear element (303, 404) both Stabilization and Bubbles are well working - for non-linear elements, the P2P1 stabilization seems to work with triangles (306) but not with quadrangle (409). Is th...

Hi Elmer Users,

Do you know if there is any attend to make aversion of Elmer running using GPU ?

Regards
Olivier