[idefix]

Hello together,

I thought it would be easy to calculate but it´s not working.
I have a vertical cylinder with the lenght x = 20 mm. At x = 10 mm there is a channel entering the cylinder. The channel has an angle of 40° to a horizontal line. I am just calculating a 2d plane of the cylinder and y-direction is in the vertical direction (parallel to the gravity).
I want to calculate the following case:
The liquid is flowing through the channel and enters the cylinder. The cylinder is filled with air at 1 bar. I expected that the flow goes vertical down. It only goes vertically down when I use setFields and initialise the liquid flow.
But one strange thing happens. At the exit of the cylinder a "liquid-mountain" is created. It seems like the liquid can´t leave the domain. I tried a lot of different boundary conditions but I can´t solve the problem.
I use
Here is what I use:

alpha1:
INLETchannel
{
type fixedValue;
value uniform 1;
}
OUTLET
{
type inletOutlet;
inletValue uniform 0;
value uniform 0;
}
INLET-AIR-Cylinder
{
type fixedValue;
value uniform 0;
}
wall
{
type zeroGradient;
}
defaultFaces
{
type empty;
}

For U:
INLETchannel
{
type fixedValue;
value uniform (1 2 0);
}
OUTLET
{
type inletOutlet;
inletValue uniform (0 0 0);
value uniform (0 0 0);
}
INLET-AIR-Cylinder
{
type pressureInletOutletVelocity;
value uniform (0 0 0);
}
wall
{
type fixedValue;
value uniform (0 0 0);
}
defaultFaces
{
type empty;
}

For p_rgh:
INLETchannel
{
type zeroGradient;
}
OUTLET
{
type fixedValue;
value uniform 100000;
}
INLET-AIR-Cylinder
{
type totalPressure;
p0 $internalField;
U U;
phi phi;
rho rho;
psi none;
gamma 1;
value uniform 100000;
}
wall
{
type fixedFluxPressure;
adjoint no;
}
defaultFaces
{
type empty;
}

Thanks a lot for your help

[adambarfi]

hi

I think the conditions you have specified for outlet have some problems and it resists against the fluid to come out, try the following BCs:
-for velocity at outlet: zeroGradient
-for p_rgh at outlet: fixedFluxPressure or buoyantPressure

[idefix]

Hi,

thanks for your fast reply.
I am trying it at the moment. I will inform you if everything works.
But could you just explain me, why this boundary conditions are better than the others? I don´t understand in which way the first boundary conditions prevent the flow from going out.

Thanks a lot

[adambarfi]

when you are using inletOutlet condition with inlet value of zero it means that on that boundary, the velocity gradient and magnitude are zero, so the fluid will stay there as near a wall.
for pressure condition the buoyantPressure sets fixedGradient pressure based on the atmospheric pressure gradient, it means that the pressure at outlet don't apply any force to the fluid, AFAIK.
fixedFluxPressure is for any boundary at which the flux is fixed, e.g. wall and inlets but not usually outlets, I'm not sure .

[idefix]

Hello,

the simulation is working fine now. Thanks
But still I am not very familiar with the boundary conditions.
I used buoyantPressure for the outlet as boundary condition for the pressure.
Am I right that this boundary condition calculated the gradient between the pressure in the last cell before the patch and the atmospheric pressure?
If yes, which atmospheric pressure is used? Is it 1.013 bar or 1 bar?

My second question is concerning the pressureInletOutletVelocity, which I understood in the following way:
I have for an outflow zeroGradient and for an inflow the velocity is calculated from the pressure at the patch (the last cell before the patch)?

Thanks a lot for your help