[kflora]

Hello FOAMERs,
I am trying to setup a simple precursor simulation using OpenFOAM 5.0 (I know this is quite old, but I need to use it for other reasons) a rigid lid for a simulation of flow in a flume so that I can develop the flow for inflow characteristics of future simulations. I am using RAS with k-omegaSST with pisoFoam.


I setup the boundaries in my blockMeshDict as:
boundary
(
inlet
{
type cyclic;
neighbourPatch outlet;
faces
(
(0 4 7 3)
);
}

outlet
{
type cyclic;
neighbourPatch inlet;
faces
(
(1 2 6 5)
);
}

top
{
type wall;
faces
(
(4 5 6 7)
);
}

bottom
{
type wall;
faces
(
(0 3 2 1)
);
}

frontAndBack
{
type wall;
faces
(
(0 1 5 4)
(3 7 6 2)
);
}
);

and my 0/U as:
internalField uniform (1 0 0);

boundaryField
{
inlet
{
type cyclic;
neighbourPatch outlet;
value uniform (1 0 0);
}

outlet
{
type cyclic;
neighbourPatch inlet;
}

top
{
type slip;
value uniform (0 0 0);
}

bottom
{
type fixedValue;
value uniform (0 0 0);
}

frontAndBack
{
type slip;
}
}


and 0/p as:


internalField uniform 0;

boundaryField
{
top
{
type zeroGradient;
}
bottom
{
type zeroGradient;
}

inlet
{
type cyclic;
neighborPatch outlet;
}

outlet
{
type cyclic;
neighborPatch inlet;
}

frontAndBack
{
type zeroGradient;
}
turbine
{
type zeroGradient;
}
}

The case does run cyclically;however, while the flow develops a vertical profile, it loses velocity (and mass) as the flow develops. As shown in the attached 2 files showing a Slice in the spanwise direction of the velocity, the flow at 10 seconds is 1 throughout most of the flow (except at the bottom with has a fixed condition), but by 100 seconds, as the flow develops, the velocity magnitude throughout most of the flow is less than 0.5.

My goal is to have an average velocity throughout the flow of 1 and so I would expect velocities to redistribute with velocity magnitudes near the surface to exceed 1.

Clearly, something is still wrong. Any suggestions are greatly appreciated.

[Severus]

Hello,
From what I see, this reason could be due to not applying a source term that will drive the flow.

Check out my answer here: 3D channel flow cyclic BC's pisoFOAM

Thanks

[kflora]

Severus,

Thank you for you advice. This took care of my problem nicely and now have the correct velocity profile.