CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (https://www.cfd-online.com/Forums/openfoam-solving/)
-   -   Diploma Thesis: Hints highly appreciated. (https://www.cfd-online.com/Forums/openfoam-solving/197957-diploma-thesis-hints-highly-appreciated.html)

Schurl January 22, 2018 11:33

Diploma Thesis: Hints highly appreciated.
 
1 Attachment(s)
Hello Colleagues,

I will be the tutor of a diploma thesis, where we designed and manufactured a jetpump for special exhaust gas sampling.

Please refer to the Attachement, where you can see a sketch of the actual jetpump, where we have also measurement data.

The graduant uses the given CFD Tool Ansys, whereas I will use OpenFOAM in parallel. We both validate the models on the one hand with the limited measurement data and on the other hand by comparing the results of both platforms.

I would like to start to mesh the jetpump according the sketch in blockMesh, where i can create the meshes depending on the parameter
- angel phi of the annular gap
- thickness d of the gap
- and the applied pressure at the inlet of the gap

The Task is to find the maximum flowrate through the orifice at a minimum used compressed air through the gap (see attachment) depending on all 3 parameters by a vast parameter study.

Here, I would highly appreciate any hints from experienced user regarding
- is the approach meshing a quader of the jetpump a good idea?
- from the measurement data, we know that the gas velocity at the gap outlet will be in a range of 50m/s to sonic speed--> which solver do you recommend?
- Are there any known pitfalls regarding BC?

Of course, I am going to share alle the results (comparions to ANSYS and measurement data) here.

Thanks in Advance
George

fisichel January 25, 2018 09:48

Diploma Thesis: Hints highly appreciated.
 
Hi Schurl,

Good luck on your thesis. I hope the following helps you out.

blockMesh is a neat tool, so you can likely implement your plan to vary the angle and thickness of the gap. You'll have to run separate OpenFoam jobs.

If you have suitable symmetry in the model, then a quarter model will be fine. If the symmetry is not so good, then the quarter model will be a decent approximation. It's best to start simple first and get more complicated gradually. So, I think you have a good plan.

The solver can be one of the standard solvers. Just Google "openfoam compressible solvers"

I don't know of any pitfalls regarding boundary conditions. OpenFoam has a good handle on them as far as I can tell. I typically have air going in and then out. They work fine for me.

Best Regards,
Chris



Quote:

Originally Posted by Schurl (Post 678946)
Hello Colleagues,

I will be the tutor of a diploma thesis, where we designed and manufactured a jetpump for special exhaust gas sampling.

Please refer to the Attachement, where you can see a sketch of the actual jetpump, where we have also measurement data.

The graduant uses the given CFD Tool Ansys, whereas I will use OpenFOAM in parallel. We both validate the models on the one hand with the limited measurement data and on the other hand by comparing the results of both platforms.

I would like to start to mesh the jetpump according the sketch in blockMesh, where i can create the meshes depending on the parameter
- angel phi of the annular gap
- thickness d of the gap
- and the applied pressure at the inlet of the gap

The Task is to find the maximum flowrate through the orifice at a minimum used compressed air through the gap (see attachment) depending on all 3 parameters by a vast parameter study.

Here, I would highly appreciate any hints from experienced user regarding
- is the approach meshing a quader of the jetpump a good idea?
- from the measurement data, we know that the gas velocity at the gap outlet will be in a range of 50m/s to sonic speed--> which solver do you recommend?
- Are there any known pitfalls regarding BC?

Of course, I am going to share alle the results (comparions to ANSYS and measurement data) here.

Thanks in Advance
George



All times are GMT -4. The time now is 10:12.