[alex::cfd]

Hi all,
I'm new to open foam, however not new to fluid simulation. I was working on wave energy converters some years ago. Now, for a spare time project, I'd like to delve into open foam multiphase simulation (submerged structure in flow channel).
I worked through the tutorials to get an overall impression and now am setting up a flow channel mainly based on the dambreak turorial, having a look also into other tutorial use-cases that come with open foam (great work, by the way!).

I have several questions, but will start with my main concern: setting the pressure boundary conditions. I read and tested a lot, but can not achieve sensible results.
Perhaps someone could give me a hint.

I first want to setup the flow channel only:

I have the following velocity boundary conditions (which I think are Ok):

FoamFile
{
    version     2.0;
    format      ascii;
    class       volVectorField;
    location    "0";
    object      U;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions      [0 1 -1 0 0 0 0];

internalField   uniform (0 0 0);


boundaryField
{
    leftInlet
    {
        type            fixedValue;
        value           uniform (8.333 0 0); //8.333=30m/s
    }
    rightOutlet
    {
        type            zeroGradient;
    }
    lowerWall
    {
        type            noSlip;
    }
    atmosphere
    {
        type            pressureInletOutletVelocity;
        value           uniform (0 0 0);
    }

    defaultFaces
    {
        type            empty;
    }

}

I would use these pressure boundary conditions (p, not p_rgh) - wich is no more in use as far as I understood:

FoamFile
{
    version     2.0;
    format      ascii;
    class       volScalarField;
    object      p;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions      [0 2 -2 0 0 0 0];

internalField   uniform 0;

boundaryField
{
    leftInlet
    {
        type            zeroGradient;
    }

    rightOutlet
    {
        type            fixedValue;
        value           uniform 0;
    }

    lowerWall
    {
        type            zeroGradient;
    }

    atmosphere
    {
        type            zeroGradient;
    }

}

I did not manage to setup a sensible p_rgh file (I think only inlet / outled bc are not OK):

FoamFile
{
    version     2.0;
    format      ascii;
    class       volScalarField;
    object      p_rgh;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

dimensions      [1 -1 -2 0 0 0 0];

internalField   uniform 0;

boundaryField
{

    leftInlet
    {
      // type            fixedFluxPressure;
      // value           pressureInletOutletVelocity;
      //type            fixedFluxPressure;
      //value           uniform 0;
      //  type            fixedValue;
      //  value           uniform 0;
      type              zeroGradient;

    }

    rightOutlet
    {
      //type        freestream;
    type            fixedValue;
    value           uniform 0;
      //type            fixedFluxPressure;
      //value           pressureInletOutletVelocity;

    }

    lowerWall
    {
        type            fixedFluxPressure;
        value           uniform 0;
    }

    atmosphere
    {
        type            totalPressure;
        p0              uniform 0;

    }

    defaultFaces
    {
        type            empty;
    }

    
}

// ************************************************************************* //

any hints how they should look like?


A related question:
I use one inlet / outlet patch for air/water in total, as both fractions should have the same inlet velocity. Is this ok?

Thank you a lot
Alexander

[alex::cfd]

Hi all,
I see I should have been more specific, sorry for that.
First of all I use:
openfoam on linux - v1806
solver: interFoam
laminar

For the inlet (left side in the above figure), i have two patches one 'inlet' at the bottom for water, the other 'inletAir' for air. They have the same boundary conditions regarding U and p_rgh:
U

    inlet
    {
        type            fixedValue;
        value           uniform (8.333 0 0); //8.333=30m/s
    }
    inletAir
    {
        type            fixedValue;
        value           uniform (8.333 0 0); //8.333=30m/s
    }

p_rgh:

    inlet
    {
      type              zeroGradient;

    }
    inletAir
    {
      type              zeroGradient;

    }

alpha.water

    inlet
    {
      type                fixedValue;
        value             uniform 1;
    }
    inletAir
    {
      type                fixedValue;
       value             uniform 0;
    }

I initialized the case with two fraction of water and otherwize empty flow channel (just as a test):


and it develops as intended (1.5s later):


I also checked the velocities of the inlet air / water inlet everything looks as expected.

However, I can not manage the outlet such that the water/air flows out. E.g. (after 6s):

8s


There is a velocity of 4.x m/s in x (normal to the outlet), which is about 1/2 of the inlet velocity (8.33m/s) and a y-component (to the top) of 3.x m/s -> looks like a partial permeable wall leading to water rising and partly flowing out via the top atmosphere patch.

I use:
U

    outlet
    {
        type            zeroGradient;
    }

    outletAir
    {
        type            zeroGradient;
    }

p_rgh

    outlet
    {
    type            fixedValue;
    value           uniform 0;
    }
    outletAir
    {
    type            fixedValue;
    value           uniform 0;
    }

alpha.water

    
    outlet
    {
        type            inletOutlet;
        inletValue      uniform 0;
    }

    outletAir
    {
        type            inletOutlet;
        inletValue      uniform 0;
    }

As far as I conclude after some hours reading posts and checking other cases I guess U and p_rgh bc should be ok.
Regarding alpha.water I'm unsure. Any hint about where the meaning of the options inletValue, Value etc. are explained? I can find the listing of bc in the manual, however all the options per bc are not explained.

One option could be to use an outlet with the same fixed velocity as the inlet (in the stream channel, mass conservation holds (in mean what comes in must get out).
However, when having submerged structures, there might be some surface waves which might lead to velocities not exactly equal to 8.3 m/s across the outlet - or not?
And the alpha.water might change across the outlet when waves crossing (e.g. part of the 'air' zone might be occupied by water).
As far as I can see, if setting alpha.water bc on the outlet patch, I fix the value there - not?

Any idea how to configure the outlet to get what I want?

Thanks
Alex

[pfemi]

I also have Problems to run a stable Modell whenn it Comes to 2 Phase open channel flow.

For the Alpha file:
if you have a seperate outlet for water, i think you should put Alpha = 1 there.
Same for the inlet.

I use only one outlet patch with Zero Gradient for Alpha and 2 inlet Patches with Alpha = 1 for water and Alpha = 0 for air.

I hope this could help.

If I dont forgett i will put the files in here later on.

[mariofcordova]

I am not sure why you introduced 8.333 as 30 m/s.