code_saturne User's Forum

Hello everyone

Now I'am working on the natural convection (Boussinesq approximation) with a high Ra number 1.5x10⁹ in a square geometry(1mx1m).The mesh is 100x100. With this mesh for a small Ra=5x10⁶ and laminar flow, the result is good. When I increase the Ra number with 1.5x10⁹, the flow is turbulent, so I use the LES model(Smagorinsky), some errors happens as shown in the file attachement. I think it is not the problem of the mesh because I've tried with a mesh 150x150. If I forget to add subroutine when I use the model LES? Besides,Smagorinsky model propose the subgrid viscosity for the Navier-stokes, if I need to add the model of subgrid thermal diffusivity for energy equation?

Thank you very much

Chengan

Hello,

You seem to have forgotten the attachment.

Regards,

Yvan

Sorry, I've added the attachment 'listing.txt', Thanks

Hello,

Are you using the usini1.f90 user subroutine, the GUI, or both ? Did you check for the options mentioned in the error message ? Could you post that data ?

Regards,

Yvan

Hello,

In fact, I couldn't find the options in "cavite.xml" they mentioned in the listing

@ La version actuelle ne le permet pas lorsque l'une des
@ options suivantes a ete activee (c'est le cas ici) :
@ - utilisation d'un estimateur d'erreur (IESCAL)
@ - couplage instationnaire (IPUCOU)
@ - prise en compte specifique de la pression
@ hydrostatique (IPHYDR et ICALHY)
@ - pas de temps variable en temps ou en espace ou
@ algorithme stationnaire (IDTVAR)

By the way, If you have some exemple which use LES model?

Thanks.

Chengan

Hello,

Are you using usini1.f90?
Could you attach your user Fortran files?

Best regards,
JF

Hello,

IPHYDR is the keyword for the "improved" option of "pressure interpolation in stratified flow", so you must use the "standard" option.

Best regards,

Yvan

Hello

I didn't add usini1.f90 in folder "SRC".

Best regards.

chengan

Hello,

finally when I change the scheme as "centered"(before I use upwind for the Velocity). The software can work now. It is strange that we don't need the upwinding for the velocity.
Here I have another question:
When I select the LES model, the momentum equation is automatically changes with turbulent viscosity. But for the scalar equation, like temperature, if I need add some turbulent model (like considering the turbulent thermal diffusivity) to the source term?

Best regards

Chengan