Turbulent Flow Simulations with Elmer - A Request

Numerical methods and mathematical models of Elmer
kishpishar
Posts: 54
Joined: 17 Jun 2015, 10:04
Antispam: Yes

Re: Turbulent Flow Simulations with Elmer - A Request

Post by kishpishar »

And here the case files and the solver source file.
Attachments
KEpsSolver.F90
(40.08 KiB) Downloaded 228 times
case_CoupledMode.sif
(3.44 KiB) Downloaded 232 times
case_SegregatedMode.sif
(4.41 KiB) Downloaded 231 times
julien givernaud
Posts: 86
Joined: 18 Nov 2014, 18:18
Antispam: Yes

Re: Turbulent Flow Simulations with Elmer - A Request

Post by julien givernaud »

Hello,

Thank you for your efforts!

I would try to use your modified solver KEpsSolver.F90.
I work on Windows and I found only KEpsSolver.dll file. How can I integer your file in my installed Elmer directory?
Sorry for the basic question.

Julien
kishpishar
Posts: 54
Joined: 17 Jun 2015, 10:04
Antispam: Yes

Re: Turbulent Flow Simulations with Elmer - A Request

Post by kishpishar »

Hi Julien,

Just compile the solver file in the case directory, you'll find KEpsSolver.dll in the same directory. Please add the statement

Procedure = "KEpsSolver" "KEpsSolver" in the solver section of your case file. In my example, the respective case files are given above and you'll find the mesh files in a previous post above (2DJetMixingTank.zip)

Thanks & regds
Kumar
kishpishar
Posts: 54
Joined: 17 Jun 2015, 10:04
Antispam: Yes

Re: Turbulent Flow Simulations with Elmer - A Request

Post by kishpishar »

Hi Peter,

I have also tried using the FCT version of the solver. When FCT corrections are applied, the solver runs just fine in the beginning but eventually crashes. To activate the FCT, I have turned both the stabilization and Bubbles off (standard SUPG stabilization anyway is not implemented for the KE solver) and activated the keyword

Linear System FCT = Logical True

Another interesting (and intriguing!) feature is that the solver in FCT mode converges when the FCT corrections are turned off! This is effected by the statement

FCT Correction Skip = Logical True

I'm finding it really difficult to understand this result. As far as I can see, FCT is implemented in the subroutine FCT_Correction ( Solver ) in the SolverUtils.F90 file in the last section through the statement

u = u + ucorr

However, applying the corrections seem to result in instability. I am stuck at this stage :-)

I think I am making a mistake somewhere; Could you please take a look at the case file, Peter? Thanks,

Kumar
Attachments
case_FCT.sif
(4.59 KiB) Downloaded 210 times
Post Reply