[ztdep]

thank you very much

[billy]

I know SimpleFoam allows steady state calculation. What do you want to simulate?

[ztdep]

i want to simulate the most simple case "


"laminar lid driven cavity"

thank you very much,

[alberto]

Use simpleFoam, switching off turbulence and choosing "laminar" as turbulence model.

Good work,
Alberto

In icoFoam is already the lid driven cavity implemented as example/tutorial case. And you
don't have any unnecessary turbulence models.
Furthermore take a look into the UserGuide, it describes the cavity case very comprehensive.

-marcus

[ztdep]

but how to simulate a steady problem use the icofoam solver.

i have tried a very large time step with the implicit EULER scheme, but i always divergence.

but the implicit scheme shold convenge at any time step

[henry]

icoFoam uses the PISO pressure-velocity coupling algorithm which is an efficient semi-implicit method for transient simulations but is not stable for very large time steps. This is not a proablem because very large time steps are not appropriate for transient flows.

simpleFoam is for steady-state flows and used the SIMPLE pressure-velocity algorithm which is efficient for this purpose.

If you insist on using a transient solver for steady-state flows you will have to write one using the transient SIMPLE algorithm which effectively means substituting the SIMPLE algorithm from simpleFoam into the time loop of icoFoam and introduce a convergence criterion. Although this will be stable for very large time-steps it is not an efficient way to solve steady-state problems.

[crosby]

What would be the best way to get to a compressible steady solver? sonicTurbFoam needs very small time steps, thus a rather large number of them to get to a steady state solution. How difficult is it likely to be to implement a form of local time stepping?

[henry]

rhoSimpleFoam can be used for sub-sonic compressible steady flows but for super-sonic cases under-relaxation is problematic and local time stepping may be a better option. We are investigating this at the moment.

[crosby]

Does anybody have some example setup files for rhoSimpleFoam?

[vvqf]

Hi Henry,

did you mean rhoSonicFoam by "rhoSimpleFoam" ?

and what did you mean for super-sonic cases?
did you mean for super-sonic compressible steady flows, "rhoSimpleFoam" can also be used by local time stepping as a better option, rather than sonicFoam, which is but a transient solver?

[tak]

Hi everyone!

Could anyone teach me the physical meaning of
- fvm::Sp(fvc::div(phi), U)
in UEqn.H, and
fvc::div(phi/fvc::interpolate(rho)*fvc::interpolate(p))
- p*fvc::div(phi/fvc::interpolate(rho))
in hEqn.H of rhoSimpeFoam.

Thank you.

[mecbe2002]

Hi all,

I am also interested in implementation of local time stepping. Even though its a very very old post, i hope you people will reply.

Can you suggest me some guideline to implement local time stepping for "rhoSonicFoam" OpenFOAM 1.6 solver.

thanks
mecbe