[Tobi]

Hi all,

just a question to all who are using interFoam.
I made a lot of simulations with that solver but never with a closed tank or something like that.

My idea was to create a tank with p=101300 Pa in it.

I have one inlet and the rest are walls. Now I want that a fluid (water) gets into the tank with 200000 Pa.

Normally the water get in the tank till the compressed air has exact 200000Pa.

The problem I have is to set correct BC for that. Or is it possible to calculate this problem with cfd?

Due to the fact of momentum equation ... mass is going into the system but not out of the system!

Thanks in advance
Tobi

[nimasam]

well, i guess interFoam is not suitable as it consider both phases incompressible, you may want to use compressibleInterFoam

[Tobi]

Originally Posted by nimasam

well, i guess interFoam is not suitable as it consider both phases incompressible, you may want to use compressibleInterFoam

Oh thanks for that hint
thats sure!



I will test it!

[Tobi]

Hi,

okay the solver is running.

But I am not sure how I should set the BC for the inlet?


As I mentioned in the first post, the water should go inside with 2 bar while the air is with 1 bar. After compression of air to 2bar no water is going into the tank.

[nimasam]

well, i guess i use pressureInletVelocity , then i will fix pressure at inlet
please share the result here

[Tobi]

Hi all,

well my first quess was to set U to pressureInletVelocity and p to fixedValue like you mentioned.

But it is not working. The simulations is total unphysical and I got an Floating Point exeption.

First:

p_inlet = 2 bar
P_internalField = 1 bar


Second:

p_inlet = 1.005 bar
p_internalField = 1 bar


(Nothing changed)

hmmm any hints are appreciate

[nimasam]

at least post the error file , and your setup here

[Tobi]

Hi,

I will do it today evening.

[Tobi]

Here the U - file:

boundaryField
{
    wall
    {
        type            fixedValue;
        value           uniform (0 0 0);
    }

    inlet
    {
        type            pressureInletVelocity;
        value           uniform (0 0 0);
    }
}

and the p_rgh file

boundaryField
{
    wall
    {
        type            fixedFluxPressure;
        value           uniform 1e5;
    }

    inlet
    {
        type            fixedValue;
        value           uniform 1.01e5;
    }
}

[nimasam]

1- whats your alpha1 file?
2- how is your initial setup?
3-how is your mesh?
4-which version of OF do you use?

[Tobi]

Hi,

initialValues:

 
p = 1e5 (boundary = calculated)
p_rgh = 1e5
U = (0 0 0)
alphaWater = 0 (boundary = ZeroGradient for the wall and inlet = fixedValue = 1)

T = 300 (boundary = zeroGradient; inlet 300 fixed)

Mesh is a cube 0,5 x 0,5 x 0,5 m with round about 200.000 cells (not many but okay). The inlet is a extra cube with 0,05 x 0,05 x 0,05 m. Mesh made with salome only Hexaeder cells.


I think alphaWater = 1 = phase1 (water) and alphaWater = 0 = phase2 (air)

I use always the latest OpenFOAM Version (2.2.x).

[nimasam]

what about air and water properties , how did you setup transportproperties?

[Tobi]

Hi,

well I do not have a look at that file since now. I copied the tutorial of the compressibleInterFoam solver. I thought that there is water and air included?

I will have a look at that file today evening.

Question to you - did you ever simulated a closed tank or something like that?

[nimasam]

Originally Posted by Tobi

Hi,

well I do not have a look at that file since now. I copied the tutorial of the compressibleInterFoam solver. I thought that there is water and air included?

well, perfect simulation :P

Originally Posted by Tobi

I will have a look at that file today evening.

Question to you - did you ever simulated a closed tank or something like that?

nope, i never did, but as it compressible, for me those configuration seems okay

[Tobi]

Hi,

as I thought everything is correct!

phase1 = water; phase2 = air.

[Tobi]

Okay its working now.

But one thing is still strange (attachment)...

Does anyone know why I have so strange gradients in alpha1 ?

[nimasam]

well
1- how did you solve your first problem (float exception?)
2- whats wrong with alpha1 gradient? distribution seems OK with me

[Tobi]

Hi,

I thought that I solved the first problem but with a p_rgh of 2 bar at the inlet its not working anymore.

Hmmm ... I am out of mind and it is so a simple geometry... hmmm

[cfdan]

don't mind to hijack this thread, but can anyone assist me here too:

http://www.cfd-online.com/Forums/ope...modelling.html

my problem seems easier to solve compared to the one here.