Hi,
I modified the built-in
Hi,
I modified the built-in kEpsilon model of OpenFoam according to a paper. unfortunately the results are not at all what we expected. I used the standard kEpsilon Model and turbFoam to produce vortex shedding behind a square cylinder (Re = 20000) and then wanted to restart with my modified model. The paper states (results with an proprietary code) that the effect should be almost instantly - considering I practically change C1_ in the code that makes sense to me. But nothing happens the drag and lift coefficients are (periodic) exactly the same as before. I tried a lot and I can rule out a wrong formula - what I can't rule out is the accidental misuse of variables while coding - therefore I hope you can take a look at my modification and check if I do something what should not be done. Here is the important extract of my modified file: RASModel::correct(); volScalarField G = nut_*2*magSqr(symm(fvc::grad(U_))); volVectorField dU = fvc::ddt(U_); volScalarField dUx = dU.component(0); volScalarField dUy = dU.component(1); volScalarField Gstar = 0.5*magSqr(U_)+k_; volScalarField C1star = C1_+C1_*0.38*k_/epsilon_*1.0/(Gstar) * mag(U_.component(vector::X)*dUx + U_.component(vector::Y)*dUy + fvc::ddt(k_)); Info << "Using modified kEpsilon model" << endl; # include "wallFunctionsI.H" // Dissipation equation tmp<fvscalarmatrix> epsEqn ( fvm::ddt(epsilon_) + fvm::div(phi_, epsilon_) - fvm::Sp(fvc::div(phi_), epsilon_) - fvm::laplacian(DepsilonEff(), epsilon_) == C1star*G*epsilon_/k_ - fvm::Sp(C2_*epsilon_/k_, epsilon_) ); epsEqn().relax(); And here is the Link to the complete File: (I am sorry if your browser is showing a blue text between the code - just save the file to disk and it should be fine) http://harly.de.vu/www/kEpsilonMod.C Hope you find something - harly |
Harly, what paper you used as
Harly, what paper you used as your reference?
|
Hey no problem:
http://harl
Hey no problem:
http://harly.de.vu/www/Paper.pdf I am implementing the formulation (15) in the kEpsilonMod.C. I am using the same mesh (described in the paper) and also Re = 20000 please don't distribute the paper - thanks - harly |
Thanks - I asked you because I
Thanks - I asked you because I plan to do a similar thing. I was involved in simulating atmospheric boundary layer with modified two- equation models of turbulence (I'll send you a paper if you're interested) so I wanted to implement such "Atmospheric Boundary Layer k-e" here.
I'll take a look at your modification and see if I can help something. Nikola |
Hi,
I am back from the hol
Hi,
I am back from the holidays and still can't find my mistake. So please if someone has an idea I am open for everything. @Nikola: Did you have time to take a look at my code? - and yes please send me the paper of your modification. - harly |
All times are GMT -4. The time now is 23:46. |