Elmer Discussion Forum

Hi "Integral Heat Source" was accidentally documented in Models Manual as "Integral Heater Control". I corrected that. And Eelis is right with the units. It would have a better choice to have the units [P]/[V] rather than [P]/[m]. This could be added by a new keyword e.g. "V...

Hi You might improve on the preconditioner (with the expense of memory) or try some other iterative solver. For example ILU1 or ILU2 with BiCGstabl and "Bicgstabl Polynomial Degree = 4" works often rather well. The structured mesh in the tutorial is rather optimal and you may require a muc...

Hi gos, If you have many holes I would rather recommend gmsh, or some other Delaunay triangulation software. ElmerGrid was not really never planned any wider use. Still it is many times used mainly because it provides a parametrisized way of defining simple structured geometries. So whenever someone...

Hi Nice work! Perhaps you could share some picture as well. Now I haven't done this myself so it's difficult to say where it all breaks up. Things that might affect: - You don't state where the problems occure. In linear system, nonlinear system or coupled system level. The lienar system may become ...

Hi A better method to do this could be the Elmer-plugin, see http://www.csc.fi/english/pages/elmer/interfaces Of course the ElmerGrid interface should work as well but I think it was left for little use as this plugin was implemented, and I don't have a huge desire to revisit it if the plugin works....

Hi

You could try the IsosurfaceSolver that creates a surface and let ResultOutputSolver save it. See an example I just sent:
viewtopic.php?f=18&t=1757

-Peter

Hi Tibi If your steady state temperature is known (131 C) and you set the external tempeature the same then that would require a coefficient -> infinity. Instead if you have a cooled system, say with water of 30 C you should rather fix the coefficient so that you get the known temperature as steady ...

Hi All, Attached is a case that demonstrates that the users need not to constrain themselves when it comes to the number of Solvers. This case has two primary solvers, FlowSolver and HeatSolver, and ten auxiliary solver that are used to compute various derived fields and save them in different forma...

Hi

I don't think you need the "Heat Flux BC" keyword. The "Heat Transfer Coefficient" depends on the fluid, Reynolds number, geometry etc. You may find it listed for some canonical cases, see

http://en.wikipedia.org/wiki/Heat_transfer_coefficient

-Peter

Hi

I made a small test case (fem/tests/fluxsolver3) with FluxSolver using DG. This allows for visualization of discontinuities in ElmerPost and VTU formats. You might give it a try.

-Peter