CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Simulating a manometer in CFX (https://www.cfd-online.com/Forums/cfx/239697-simulating-manometer-cfx.html)

Abdulla060 November 19, 2021 14:19

Simulating a manometer in CFX
 
Hello.

I'm very new to Ansys and CFD in general. I've been practicing CFX and I wanted to create a very simple manometer simulation. however, I can't figure out how I can fill a region of my domain with a different fluid (for example mercury in the bottom of the manometer)

my model

https://i.imgur.com/SJezUOo.png

ghorrocks November 19, 2021 18:38

If you want to have a mercury free surface and the pressure difference pushes the free surface up and down then you will need to do a multiphase simulation with a free surface model.

But you should look at this model and think what are you really trying to do here? A much better approach might be to not model the manometer, but replace it with a blind passage with a pressure measurement there. This model is single phase, so much easier, and will give you a pressure difference across the device. Then you can do a second simulation with the pressure difference applied over a manometer to see the free surface move. This two simulation method will be much easier to do compared to everything in one.

But it all depends on what you are trying to achieve.

Abdulla060 November 20, 2021 07:34

Quote:

Originally Posted by ghorrocks (Post 816916)
If you want to have a mercury free surface and the pressure difference pushes the free surface up and down then you will need to do a multiphase simulation with a free surface model.

But you should look at this model and think what are you really trying to do here? A much better approach might be to not model the manometer, but replace it with a blind passage with a pressure measurement there. This model is single phase, so much easier, and will give you a pressure difference across the device. Then you can do a second simulation with the pressure difference applied over a manometer to see the free surface move. This two simulation method will be much easier to do compared to everything in one.

But it all depends on what you are trying to achieve.

Thank you very much for the reply.

My main goal is to learn multiphase simulation in CFX and I thought that this practice will both achieve this goal whilst being easily verifiable. my biggest problem now is that I don't know how to assign each fluid to a region (similar to Fluent patch) if such a thing is even possible in CFX.

Gert-Jan November 20, 2021 09:43

The numerical solution will tell you where which fluid will be present.
You problem is the initialisation. In principle you can do this using a step function for the initial volume fraction. Alternatively, you can split the geometry (keep the split alive through Spaceclaim, Mesher and Pre), approximately where the thin line is now in your figure. This allows you to specify different initial conditions in Pre.

Remember that since your liquid volume is quite thin, surface tension might play a role, which however is quite difficult to capture using CFX. Better use a larger volume in the bend.

Also, be aware that the start up of such a numerical simulation comes with significant pressure waves. If you don't take any precautions, your liquid might have left the pipe long before you notice.

Summarized, this might be a pretty difficult calculation in the end.

ghorrocks November 20, 2021 20:57

If you are learning CFX then the best place to start is the tutorial examples. They have versions which cover most of the functions in CFX, including multiphase. The tutorial examples are in the ANSYS Customer web page (if you are a commercial customer) or the ANSYS Academic web page (if you are an academic or student).


All times are GMT -4. The time now is 09:02.