Boundary Conditions in .vtu File

Numerical methods and mathematical models of Elmer
Post Reply
mabr
Posts: 10
Joined: 19 Apr 2018, 15:53
Antispam: Yes

Boundary Conditions in .vtu File

Post by mabr »

Hello all!

This might be a trivial question, but I haven't found anything in other threads. I am dealing with a heat transfer model. I am trying to map heat flux from an external surface to the surface of my Elmer model and then use this surface as a boundary condition. I did some sort of mapping with my UDF routine. The final result of temperature is close to what I have expected, but I am still not satisfied. I have the feeling that the mapping of heat flux to Elmer surface fails, so i would like to see the boundary condition imposed on my Elmer model in .vtu output file. Is there a way to write boundary condition values to .vtu output file via sif file and then see the distribution of boundary condition on the surface?

Thanks.
mzenker
Posts: 1999
Joined: 07 Dec 2009, 11:49
Location: Germany

Re: Boundary Conditions in .vtu File

Post by mzenker »

Hi,

in fact, you want to see the initial state, before the first timestep?
You can use the resultoutput solver to write the vtu file and set exec to "before timestep". You then need to do one extra timestep at the end. Or maybe you can try to fiddle with the timesteps so that the first one has zero length.
If you do a steady state simulation, you could switch your heat source off and see what happens.

HTH,
Matthias
mabr
Posts: 10
Joined: 19 Apr 2018, 15:53
Antispam: Yes

Re: Boundary Conditions in .vtu File

Post by mabr »

Thank you for the answer, Matthias. Yes, I am looking for initial state of my model. I am dealing with a steady-state model. What do you mean by switching off the heat source? I have no heat source present in my model. My idea was to define the "flux compute" procedure and set the "exec solver = after all". But I assume this calculates fluxes from temperatures and are thus not a BC values. I tried with the "resultoutputsolver", but I guess I am missing something. Below is the code of my resultoutput solver, that doesn't work (obviously :D ). Is there an example on how to write heat fluxes of initial state to the resultoutput solver? What would be the keyword for boundary condition heat flux? I tried "Temperature Flux". I was looking through the test cases, but couldn't find anything useful.

Code: Select all

Solver 2
Exec Solver = before timestep
Equation = "ResultOutput"
Procedure = "ResultOutputSolve" "ResultOutputSolver"
Output File Name = case
Vtu format = Logical True
Vector Field 1 = Heat Flux
End
Thanks in advance.
mzenker
Posts: 1999
Joined: 07 Dec 2009, 11:49
Location: Germany

Re: Boundary Conditions in .vtu File

Post by mzenker »

Hmmm...
I am afraid I don't get what you intend to do. If you could post the whole sif file this would certainly help.
The resultoutput solver is invoked implicitly if you give the "Post File" the ending .vtu. So if you use it explicitly you should leave that field blank. Of course it doesn't make much sense to say "before timestep" if you do a steady state simulation. Perhaps you could do a transient test run just to get your initial state?

HTH,
Matthias
raback
Site Admin
Posts: 4812
Joined: 22 Aug 2009, 11:57
Antispam: Yes
Location: Espoo, Finland
Contact:

Re: Boundary Conditions in .vtu File

Post by raback »

Hi mabr

The saving of VTU files just goes through the existing fields and saves them (if they are active). Keywords are not fields so if you want to visualize them you should make them to fields. You can use the SaveBoundaryValues module (Ch. 49 in current models manual) to do this. Just call the module prior to saving the results.

-Peter
Post Reply