Velocity not zero during flow reversal
 

Hi,

I tried to simulate oscillatory channel flow using twoPhaseEulerFoam, the turbulence closure is k-Epsilon model. I use oscillatory pressure gradient to drive the flow, however, I found that the velocity of free stream at the flow reversal is not zero( very small), and it would grow when the case runs longer, could any one give some hint? What causes this problem? How to avoid this? The following is the momentum equation:

fvm::ddt(Ub)
+ fvm::div(phib, Ub, "div(phib,Ub)")
- fvm::SuSp(fvc::div(phib), Ub)
==
pressure gradient+ other terms

Perhaps, it's more informative that I point out the fvSchemes that I use:

ddtSchemes
{
default backward;
}
divSchemes
{
default none;
div(phia,Ua) Gauss limitedLinearV 1;
div(phib,Ub) Gauss limitedLinearV 1;
div(phib,k) Gauss limitedLinear 1;
div(phib,epsilon) Gauss limitedLinear 1;
...
}

One more question, just out of curious, why doesn't the program complain about missing definition of div(phib) in fvScheme?

Thanks!!

Hi Charlie,
I know its quite long time since you have posted this question,,, but I am facing somewhat the same type of problem when having an oscillatory pressure gradient. The mean flow begins to rise and its not zero anymore at the reversal...
Have you managed to find a solution for this?

Sina