BC for turbulent internal flow
 

Hi

I want to compute the turbulent around a cylinder array inside a channel. I've a fine structured grid but I'm not able to find the proper combination of inlet/outlet to solve this : I always end up with diverging time step continuity errors.

I tried many combination :
- fixed inlet and Neumann outlet for U, Neumann Inlet and fixed outlet for P
- massflow inlet and Neumann outlet for U, Neumann Inlet and fixed outlet for P
- fixed inlet and Neumann outlet for U, Neumann Inlet and outlet for P
- fixed inlet and massflow outlet for U, Neumann Inlet and fixed outlet for P

Nothing is working.

If anyone has an idea, I take it.

Many thanks

Yannick

Reynolds Number? Relaxation Factors? Solver used? Transient, steady? Discretization Schemes? Mesh foto?

How far away is your inlet and outlet from the domain of interest?

We need more information about it to solve the problem.

If you are you using a transient problem, are you using pimpleFoam? It will let you control the maxCourant number so that your simulation doesn´t blow because of the Courant, although it may blow for other reasons.

Hi

The configuration and the grids are described in this paper :
http://www.sciencedirect.com/science...89974614000358

I used here SA with pimpleFoam.
Here is the fvscheme :
ddtSchemes
{
default Euler;
}

gradSchemes
{
default Gauss linear;
grad(nuTilda) cellLimited Gauss linear 1;
grad(k) cellLimited Gauss linear 1;
grad(kl) cellLimited Gauss linear 1;
grad(omega) cellLimited Gauss linear 1;
grad(epsilon) cellLimited Gauss linear 1;
grad(q) cellLimited Gauss linear 1;
grad(zeta) cellLimited Gauss linear 1;
grad(v2) cellLimited Gauss linear 1;
grad(f) cellLimited Gauss linear 1;
grad(sqrt(kt)) cellLimited Gauss linear 1;
grad(kt) cellLimited Gauss linear 1;
grad(sqrt(kl)) cellLimited Gauss linear 1;
}

divSchemes
{
default Gauss linear;
div(phi,U) bounded Gauss linearUpwindV grad(U);
div(phi,k) bounded Gauss linearUpwind grad(k);
div(phi,epsilon) bounded Gauss linearUpwind grad(epsilon);
div(phi,zeta) bounded Gauss linearUpwind grad(zeta);
div(phi,q) bounded Gauss linearUpwind grad(q);
div(phi,omega) bounded Gauss linearUpwind grad(omega);
div(phi,nuTilda) bounded Gauss linearUpwind grad(nuTilda);
div(phi,T) bounded Gauss limitedLinear 1;
div(phi,kl) Gauss limitedLinear 1;
div(phi,kt) Gauss limitedLinear 1;
div(phi,R) Gauss upwind;
div(R) Gauss linear;
div((nuEff*dev(grad(U).T()))) Gauss linear;
div(phi,v2) bounded Gauss linearUpwind grad(v2);
div(phi,f) bounded Gauss linearUpwind grad(f);
}

interpolationSchemes
{
default linear;
interpolate(HbyA) linear;
}

laplacianSchemes
{
default Gauss linear limited 0.333;
}

snGradSchemes
{
default limited 0.333;
}

fluxRequired
{
default no;
p ;
}

Yannick,

I guess not everyone have access to sciencedirect papers (indeed, I don't).
However, I did some simple turbulent flow calculation around one cylinder without any problem using the following BC for Inlet and Outlet:
for U:
inlet
{
type fixedValue;
value uniform (toto 0 0);
}
outlet
{
type inletOutlet;
inletValue uniform (toto 0 0);
value $internalField;
}
for p:
inlet
{
type zeroGradient;
}
outlet
{
type zeroGradient;
}
I hope this will help you,

Regards,

Cedric