CFD Online Discussion Forums

CFD Online Discussion Forums (
-   OpenFOAM Running, Solving & CFD (
-   -   Solver for compressible Navier-Stokes equation (

treima May 30, 2012 03:11

Solver for compressible Navier-Stokes equation

I have a problem, which I would like to solve with a solver based on the Navier-Stokes equations.

The solver shoud be based on the compressible, steady state Navier-Stokes with conservation of mass, momentum and energy.
Im calculating with air, so inviscid, mach number about 0.7 up to almost 1. Turbulence dosent matter.

Is there a solver in OpenFOAM, which have these properties? Almost all solvers based on rhoSimple are calculating with RANS-models. Have anyone some experience with the full Navier-Stokes equations in OpenFOAM?

Best regards

romant May 30, 2012 03:45

what do you really want to calculate if turbulence does not matter. The full navier stokes equations can be solved by direct numerical simulation, but in this case you might want to have access to something like a cluster in the top 500 super computer list all to yourself for a year.

OpenFOAM can solve the navier stokes equation and if you set turbulence to laminar it will not model turbulence. however, at the mach numbers that you indicate this won't give you any useful result, except if you solve it DNS like.

for more advanced simulation of the full navier stokes equations try to look into LES (large eddy simulation), which will solve navier stokes on almost all of the grid except when the cell size is larger than the turbulent structures.

treima May 30, 2012 04:17

1 Attachment(s)
Thanks for your answer!

Perhaps i should say something to my background:
For my master thesis, I`m having a look at some equations and the aim is to calculate the adjoint system and implement this in OpenFOAM. One of these equations is the Navier-Stokes equation. I attached the primal equation, which Im considering (sry for writing in German, but I hope, the equations are clear).

For this purpose Im searching for a solver, which i can modify. So is there one in OpenFoam?

How long this solver takes for some easy problems is another question. The first idea was a turbine with an in- and outlet for the air.

sylvester May 30, 2012 05:25

Solvers you could possibly use are sonicFoam ("Transient solver for trans-sonic/supersonic, laminar or turbulent flow of a compressible gas") and rhoCentralFoam ("Density-based compressible flow solver based on central-upwind schemes of Kurganov and Tadmor").

I would reconsider solving the primal equations without turbulence, though. Chances of non-physical results and instabilities are high. For your adjoint equations you can always use the frozen turbulence assumptions.


All times are GMT -4. The time now is 21:04.