CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   CFX (http://www.cfd-online.com/Forums/cfx/)
-   -   Multiple source files (initial condition) (http://www.cfd-online.com/Forums/cfx/94247-multiple-source-files-initial-condition.html)

TWA November 10, 2011 06:04

Multiple source files (initial condition)
 
Hi,

I want to use multiple source files (.res files) as an initial condition for a transient simulation. In my .def file I have a fluid and a solid domain. I have one .res file with an identical mesh as in the .def file (for both fluid and solid), and I have one .res file with only the fluid mesh. In my transient simulation I want to use the initial values for the fluid domain from the "fluid-only" .res file, and initial values for the solid domain from the other .res file. I specify both files in the setup, but my problem is that the solver uses values for the fluid domain from the wrong .res file. It does not help to change the order of how I specify the .res files (Initial values 1 and Initial values 2 in Define run). Can I control which .res file which has the highest priority, or something like that?

ghorrocks November 10, 2011 17:19

You cannot specify multiple initial conditions files. You are going to have to combine both res files into one file to use it - have a look at the cfx5interp.exe command and the options it has for combining res files. You may have to delete the solid domain from the fluid run, then interpolate the fluid run as one step, the solid run as the next step or something like that.

TWA November 11, 2011 02:41

Aha, thank you. I managed to solve my problem with an alternative approach, but I'll look into this for future reference.

ghorrocks November 11, 2011 05:04

Can you describe the alternate approach?

TWA November 11, 2011 07:54

It was just something that was possible in this specific problem. In the fluid+solid simulation, I was able to retrieve an almost identical result by removing the fluid domain and instead apply a constant temperature on the walls where the fluid had been. (I'm really just interested in the temperature distribution in the solids in this steady state simulation.) So I ended up with one res file with only solid domains and one res file with only the fluid domain. I specified both in Define run and the solver picked the correct initial values for each domain from the two res files.


All times are GMT -4. The time now is 13:27.