[JM27]

Hi there,

I am running some bubble dynamics simulations in compressibleInterFoam. I notice that pMin has a significant effect on the solution.
The initial condition is atmospheric pressure. I set the pressure p to oscillate around atmospheric condition using uniformFixedValue BC. The issue is that when the pressure on the boundary goes to 0 or below 0, the solution diverges. So if p_oscillatory >= p_atm it causes problems with negative pressure/temperature or floating point exceptions!
I understand that this is because of the EoS implementation in this solver, i.e. rho = p * psi which basically means that pressure can never be 0 otherwise density is also 0. So, I tried to bypass this by setting very small values of pMin and the simulation runs completely, however it seems that the values drastically affect the solution leading to unrealistic bubble shapes and in some cases even bubble breakup.
I have tried the similar cases but using a simple fixedValue BC for pressure set at atmospheric and have good and stable results.

I had also tried to implement fvOptions functionality but it seems it is not supported with compressibleInterFoam.

I have been stuck on this issue for a while and I really need to use the oscillatory boundary, please help !

Thanks in advance,
JM

[JM27]

Hi, I am still having this problem, using OF v5.


Can anyone answer my question, please?

[peterhess]

Hello!
I dont use myself the solver compressibleInterFoam...

Anyway, cause the name of the solver has comperssible in it, that means the pressure is principly by definition absolute not static!!!
p_absolute = p_static + 1.013e5

In comperssible solver you dont define static but absolute pressure.

i.e. It is not acceptable to have < 0 pressure!
0 absoulte pressure means absolutly vacuum!
0 static pressure means the pressure you define is 1.013e5!

Any value between 0 and 1.013e5 means vacuum.
And higher than 1.013e5 means higher than atmpspheric...

That means the pressure in the p dictionary should be p + 1.013e5!

The pMin (or pMax) do nothing else than to limit the pressure to the defined value by pMin value, if the calculated pressure went less than the pMin.

i.e. if the calculated pressure went less than pMin, then the calculated value will be replaced by pMin!

Please dont forget to replace also the internalField with: value + 1.013e5!

Else the solver will suffer much to do the first calculation steps and divergence could happend...

Regards

Peter

[JM27]

Dear Peter,

Thank you for your reply! I am aware of all of this and why pMin needs to be specified in the dictionary, however in compressibleInterFoam one also specifies an Equation of state for the different phases. So an equation of state will be defined for air and water in this case. I do define the initial conditions based on absolute pressure, however, it seems that based on the value of pMin that I choose, a region of incorrectly estimated pressure.
As mentioned in my previous post, I am using the solver for bubble dynamics problems. The bubble shape is highly influenced by the pressure field surrounding the bubble so this will result in an incorrect shape representation or artificial breakup.

Looking at the pEqn for compressibleInterFoam, pMin seems to be used for the last non orthogonal correction. My mesh is cartesian so there should be no need for non orthogonal correctors so I set this to 0 in PIMPLE, however pMin still seems to be affecting the solution.

I am therefore very confused by the outcome of my simulations and not sure what else I can try to solve this problem. Please help!

Regards,

JM

[peterhess]

Show me please the pressure dictionary?

Which values are inside?

I still belive that you define the wrong pressures in the dictionary, If I understood your discribtion of the case.

Why you have pMin at all?

I belive you can delete it.

The solver will not find it and will ignore it (I belive).

I viewd the tutorials for the solver u r using.

There is no pMin in any one!

One more time I dont use this solver, but as a generall role it is used if it has been defined.

Regards

Peter

[JM27]

Originally Posted by peterhess

Show me please the pressure dictionary?

Which values are inside?

I still belive that you define the wrong pressures in the dictionary, If I understood your discribtion of the case.

Why you have pMin at all?

I belive you can delete it.

The solver will not find it and will ignore it (I belive).

I viewd the tutorials for the solver u r using.

There is no pMin in any one!

One more time I dont use this solver, but as a generall role it is used if it has been defined.

Regards

Peter

Hi Peter,


Thank you for your reply. Not sure which tutorials you have looked at for compressibleInterFoam (using v5 here) but if you go on tutorials/multiphase/compressibleInterFoam/laminar/depthCharge2D and look in constant/thermophysicalProperties you will see that pMin needs to be defined.

FoamFile
{
    version     2.0;
    format      ascii;
    class       dictionary;
    location    "constant";
    object      thermophysicalProperties;
}
// * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * //

phases (water air);

pMin        10000;

sigma
{
    type    liquidProperties;
    phase   water;
}

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

In my initial conditions and boundary conditions I use the absolute atmospheric pressure which I take to be 10^5Pa.


0.orig/p_rgh reads:

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

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

internalField   uniform 1e5;

boundaryField
{

    farField 
    {
        type                   uniformFixedValue;  // oscillating pressure boundary
        uniformValue           sine;
        uniformValueCoeffs
        {

                amplitude       95000; 
                frequency       20000;           
                scale           1;
                level           100000; // absolute pressure
                t0              0;
        }
    }

    defaultFaces
    {
        type            empty;
    }
}

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

If I do not specify a pMin value in constant/thermophysicalProperties, OpenFOAM will give an error, so I guess it does need to be specified.
Hope this clarifies!

[JM27]

Hi Peter,


Any further advice with regards to my problem?


Do you think my pressure specification is OK?



Thanks,


JM

[peterhess]

Hello and sorry for the late answer.

You are right pMin is needed.

Usually pMin is to be found in fvSolution. That why I did not find it in the tutorial you mentioned.

May I ask you to post the case please.

Regards

Peter