[hanness]

Hi there,
I wanted to run a case with the buoyantSimpleFoam solver but got some strange behaviour of the flow - at an inclined wall the hot air moves in the direction of gravity. For comparison I ran the same case with a vertical wall and got the expected result. Moreover, running the case with the inclined wall with the buoyantBoussinesqSimpleFoam solver also gives the expected results. - see attached images
inclinedWall_Uz2.jpg

straightWall_Uz.jpg

inclinedWallBoussinesq_Uz2.jpg

The boundary conditions were the following:
p_rgh
internalField uniform 101325;
boundaryField
{
hotWall
{
type buoyantPressure;
value $internalField;
}
fixedWalls
{
type buoyantPressure;
value $internalField;
}
}

U
internalField uniform (0 0 0);
boundaryField
{
hotWall
{
type fixedValue;
value uniform (0 0 0);
}
fixedWalls
{
type fixedValue;
value uniform (0 0 0);
}
}

k
internalField uniform 0.1;
boundaryField
{
hotWall
{
type compressible::kqRWallFunction;
value uniform 0.1;
}
fixedWalls
{
type compressible::kqRWallFunction;
value uniform 0.1;
}
}

epsilon
internalField uniform 0.01;
boundaryField
{
hotWall
{
type compressible::epsilonWallFunction;
Cmu 0.09;
kappa 0.41;
E 9.8;
value uniform 0.01;
}
fixedWalls
{
type compressible::epsilonWallFunction;
Cmu 0.09;
kappa 0.41;
E 9.8;
value uniform 0.01;
}
}

mut
internalField uniform 0.01;
boundaryField
{
hotWall
{
type compressible::epsilonWallFunction;
Cmu 0.09;
kappa 0.41;
E 9.8;
value uniform 0.01;
}
fixedWalls
{
type compressible::epsilonWallFunction;
Cmu 0.09;
kappa 0.41;
E 9.8;
value uniform 0.01;
}
}

Can anybody tell me whether this is a bug or my boundary conditions are not suitable?

Thanks a lot,
Hannes

[RMF]

Hi Hannes,

this looks really strange.
Which schemes did you use?

The fluid is an ideal gas?

Bye RMF

[hanness]

Dear RMF,

yes, the fluid is calculated as ideal gas. The used schemes are as follows:

ddtSchemes
{
default steadyState;
}
gradSchemes
{
default Gauss linear;
}
divSchemes
{
default none;
div(phi,U) Gauss upwind;
div(phi,h) Gauss upwind;
div(phi,K) Gauss upwind;
div(phi,k) Gauss upwind;
div(phi,epsilon) Gauss upwind;
div(phi,R) Gauss upwind;
div(R) Gauss linear;
div((muEff*dev2(T(grad(U))))) Gauss linear;
}
laplacianSchemes
{
default none;
laplacian(muEff,U) Gauss linear uncorrected;
laplacian((rho*(1|A(U))),p_rgh) Gauss linear uncorrected;
laplacian(alphaEff,h) Gauss linear uncorrected;
laplacian(DkEff,k) Gauss linear uncorrected;
laplacian(DepsilonEff,epsilon) Gauss linear uncorrected;
laplacian(DREff,R) Gauss linear uncorrected;
}
interpolationSchemes
{
default linear;
}
snGradSchemes
{
default corrected;
}
fluxRequired
{
default no;
p_rgh;
}

anybody any idea?
Thanks a lot
Hannes

[hanness]

thanks to Henry I got the solution:
"Inclining the wall makes the mesh non-orthogonal but you choose an uncorrected laplacian but a corrected snGrad. If you correct for non-orthogonality the flow goes in the expected direction."

[immortality]

hi Hanness.do we use unCorrected in orthogonal meshes only and for non normal faces(non-orthogonal) use corrected?should both laplacian and snGradSchemes be the same?am i correct?when we use limit?