multiphaseEulerFoam loosing fluid with zeroGradient for p on walls???
 

Dear fellow foamers,

I've run into another strange problem with multiphaseEulerFoam (mpEF). In order to perform scalability tests with mpEF, I've prepared a simple 3d dam break case (only 2 phases - air and water). I've used standard boundary conditions and solver settings taken from tutorial files (see below). When I use the zeroGradient BC for pressure on solid walls all the water somehow leaks through the walls (with no effect on Courant number!). fixedFluxPressure works just well... Any idea where this behaviour comes from?

--
0.org/alphaair

boundaryField
{
walls
{
type zeroGradient;
}
atmosphere
{
type inletOutlet;
phi phiair;
inletValue uniform 1;
value uniform 1;
}
}

--
alphawater is almost the same, only on atmosphere the inletValue and value are uniform 0
--
0.org/p

boundaryField
{
walls
{
type zeroGradient;
}
atmosphere
{
type totalPressure;
p0 uniform 0;
U Uair;
phi phiair;
rho rho;
psi none;
gamma 1;
value uniform 0;
}

}

--
0.0rg/Uair

boundaryField
{
walls
{
type fixedValue;
value uniform (0 0 0);
}
atmosphere
{
type fluxCorrectedVelocity;
value uniform (0 0 0);
}
}
--
... again, Uwater is the same
--
system/fvSolution

solvers
{
p
{
solver GAMG;
tolerance 1e-7;
relTol 0.05;
smoother DIC;//GaussSeidel;
nPreSweeps 0;
nPostSweeps 2;
nFinestSweeps 2;
cacheAgglomeration on;
nCellsInCoarsestLevel 10;
agglomerator faceAreaPair;
mergeLevels 1;
}

pFinal
{
solver PCG;
preconditioner
{
preconditioner GAMG;
tolerance 1e-7;
relTol 0;
nVcycles 2;
smoother DIC;//GaussSeidel;
nPreSweeps 0;
nPostSweeps 2;
nFinestSweeps 2;
cacheAgglomeration on;
nCellsInCoarsestLevel 10;
agglomerator faceAreaPair;
mergeLevels 1;
}
tolerance 1e-7;
relTol 0;
maxIter 20;
}

pcorr
{
$pFinal;
tolerance 1e-5;
relTol 0;
}

U
{
solver smoothSolver;
smoother DIC;//GaussSeidel;
tolerance 1e-8;
relTol 0.1;
nSweeps 1;
}

UFinal
{
$U;
tolerance 1e-7;
relTol 0;
}
}

PIMPLE
{
// nOuterCorrectors 1;
nCorrectors 3;
nNonOrthogonalCorrectors 0;
nAlphaSubCycles 3;
}

relaxationFactors
{

"U.*" 1;
p 1;

}
--
As for fvSchemes, I use vanLeer and limitedLinear as in damBreak tutorial.

Thanks, Petr.