CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM (http://www.cfd-online.com/Forums/openfoam/)
-   -   static pressure BC for interFoam (http://www.cfd-online.com/Forums/openfoam/65052-static-pressure-bc-interfoam.html)

kwardle June 2, 2009 10:34

static pressure BC for interFoam
 
All,
I am in need of a static pressure boundary condition for use in the interFoam solver to specify a fixed liquid height. I found one that works but there are some issues.

As an example--consider a u-tube manometer where the initial liquid height is higher in the left channel than in the right one. Obviously, the two should slosh back and forth until both are the same height. That is indeed what you get.

u-tube manometer

Now, say you wanted to have submerged boundary condition for the right hand side such that you specify a 0 static pressure at a certain height. The left hand side should then do the same thing and oscillates down until the liquid height in the left hand side is equal to the height of the submerged boundary on the right side. I tried many different boundary types for this submerged outlet. In every case (but one) the liquid completely drained out until the right side lost liquid contact and the remaining liquid settled to the bottom with the liquid level about half filling the bottom cross-section of the tube. However, if I use a fixedGammaPressure boundary condition on the right side I can actually get the correct behavior.

partial u-tube manometer

For this one the fixedGammaPressure was not zero but 100--as you can see, the extra liquid height in the left side is about 1cm (for reference the width of the tube is 1 cm) since 100 Pa ~= 10 mm H2O.

The problem is that I don't think this boundary condition is even meant for this type of outlet boundary patch. I believe it was developed in order to specify the static pressure on wall patches used in 6DOF simulations. As such, it has a little problem when air bubbles cross the boundary (continued in next post).

kwardle June 2, 2009 11:24

use of gammaFixedPressure BC on outlet boundary
 
Now, if we change the geometry such that there is a tube injecting air bubbles into the right hand side of the partial u-tube manometer, initially I had big troubles when the air bubbles hit the boundary.

http://sites.google.com/site/kwardle..._05in_025s.png http://sites.google.com/site/kwardle..._05in_030s.png http://sites.google.com/site/kwardle..._05in_035s.png

For this, I used gammaFixedPressure for pd and I believe I tried both pressureInletOutletVelocity and zeroGradient for U on the outlet boundary and got this explosion. However, the problem is solved if I just use a straight inletOutlet BC for U with the inletValue set to (0 0 0).

partial u-tube manometer with bubbles

So, things seem to work OK and so I thought I would post this as it might be useful to someone. As I said though, this is not the intended use for the gammaFixedPressure BC (it doesn't even show up in fvPatchFields--actually I can't recall at the moment where I even found it...). Perhaps there are some things that could be done to make this BC type more generalized for application to outlet pressure boundaries. Thoughts?

medak19 March 4, 2015 05:29

Hi,

Your case is the problem I am trying to solve. But, the pressure in my case is excited with a time-varying sinusoidal function. I have two questions for the case of u-tube manometer.
First, I tried many BCs but the only case that would work, meaning the water surface shows oscillation, is if I specify the pressures both at the inlet and outlet. For velocities, I use pressureInletOutletVelocity and the solver is interFoam. Is this the correct way since I prescribe the pressure at inlet and enforce the pressure at outlet too?

Second, if I do not want to enforce the pressure at outlet but only at inlet, what BC should I use for Pressure and velocity at the outlet? I know I can not used zeroGradient for both Pressure and velocity there.

Any thoughts are appreciated.


All times are GMT -4. The time now is 20:40.