CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (http://www.cfd-online.com/Forums/openfoam-solving/)

 sunnysun February 25, 2009 09:33

Hi, everyone, I have a geom

Hi, everyone,

I have a geometry which is roughly a bifurcation with cylinder tube(flow goes from cylinder to bifurcation, 1 inlet, 2 outlet). How can i set the flow division through boundary condisions? say I would like to have 30% of flow go in to one brance and 70% goes in to another? The inlet flow is pulsatile and I use timeVaringMappedFixedValue which read velocity from data file.

Thanks a lot!!

Vivien

 sunnysun February 25, 2009 11:11

Seems to be a simple question.

Seems to be a simple question...but can somebody give me some ideas?

 sunnysun February 25, 2009 11:11

Seems to be a simple question.

Seems to be a simple question...but can somebody give me some ideas?

Vivien

 kati February 25, 2009 12:28

Hi Vivien, For I know, ther

Hi Vivien,

For I know, there is no outlet BC in OpenFOAM which allows such predefined division of flow to different outlets (some codes do have such BC).

What you could do, is to set a flowRate BC on one of the outlets and define a negative flow rate that is about 30% of the inflow (there is also a timeVarying version of flowRate BC). Then you could use zero pressure on the other outlet for the rest of the outflow.

Regards,
Kati

 sunnysun February 26, 2009 04:59

Hi,Kati, Thanks a lot for y

Hi,Kati,

What is still unclear for me is why should I use negative flow rate at onbe outlet?(say outlet1). Also, what should be the pressure BC for outlet1? Why zero pressure instead of zero gradient pressure BC at outlet2?

Thank you very much!

Vivien

 kati March 2, 2009 13:34

Vivien, Negative flow rate

Vivien,

Negative flow rate because positive would mean inflow.

Normal outlet BC is setting pressure and using zeroGradient for convected fields (actually inletOutlet instead of zeroGradient, more stable).

You could also fix flow rate at both boundaries and then use zeroGradient for pressure, but then you are fixing all flow rates on all boundaries, and I wouldn't do that because it leaves no freedom to the flow solution. It should in theory be okay with incompressible or steady state cases, but in compressible transient it is an error.

You could test different BC combinations and report the results here. I'd be interested to hear whether you had any problems with "fix all flow rates" approach.

Kati

 All times are GMT -4. The time now is 23:57.