[MNMK]

Hi,

I am simulating a Water tank which is filled with air at the beginning and gradually water flows through a tube in it. I would like to know whether a free surface model is a right model? I simulate it transient.

regards,

[ghorrocks]

Probably yes, but it depends on the details of what you are trying to do and the physics at work.

Is the water/air interface bubbles, foam, froth? Or is it just a simple, roughly flat water/air interface?

What fluid velocities does this use?

Most importantly - why are you doing this? What are you trying to learn from this simulation? This tells us what factors are important and you can develop the model to achieve your goal.

[evcelica]

Originally Posted by ghorrocks

Most importantly - why are you doing this? What are you trying to learn from this simulation? This tells us what factors are important and you can develop the model to achieve your goal.

Exactly! Result of interest is the most important information.

[MNMK]

Originally Posted by ghorrocks

Probably yes, but it depends on the details of what you are trying to do and the physics at work.

Is the water/air interface bubbles, foam, froth? Or is it just a simple, roughly flat water/air interface?

What fluid velocities does this use?

Most importantly - why are you doing this? What are you trying to learn from this simulation? This tells us what factors are important and you can develop the model to achieve your goal.

Dear Ghorrocks,

My model is a tank which will be filled with water. The water inlet temperature increases as time passes.
what should be considered is that on the water surface, a small area of bubbles can be seen in test.

One of my goals is to determine the water height in my tank when the pressure difference between 2 measure points( e.g. Up and Down in the Tank)reaches 0.2 bar. I measure the water height by the expression : lengthInt(Water.Volume Fraction )@Line 3

The point which is not clear to me is the value of the water height. Because of the range of water volume fraction at the end of the simulation ( from 0 to 1 Water Volume Fraction), I am not able to see a clear interface. I think the kind of interface model has an influence on it. doesn´t?

I have another question. When I make a temperature contour, I can just choose the water. Temperature or Air. Temperature as a variable which look exactly the same. Why do the contours look the same? Even the air and water temperatures do not have the same temperatures, especially long from the beginning.

best regards,

[ghorrocks]

One of my goals is to determine the water height in my tank when the pressure difference between 2 measure points( e.g. Up and Down in the Tank)reaches 0.2 bar.

Can't you work this out from simple hydrostatic pressure? Why is CFD required for this?

I think the kind of interface model has an influence on it. doesn´t?

Yes, of course it does. Please post your output file so we can understand what you have done. Also, what is the water/air interface? Is it bubbly, foamy, or a clear interface?

Why do the contours look the same?

They look the same because the air and water phases have the same temperature field. What do you expect here? Are you expecting the air and water to have different temperature fields? Why?

You have not answered the most important question:

Most importantly - why are you doing this? What are you trying to learn from this simulation? This tells us what factors are important and you can develop the model to achieve your goal.

If I knew what you are trying to do then we might be able to help you.

[MNMK]

Hi again,

I would like to design a water Tank, so I have to check my model that works correctly like in Test. The model which I am simulating is more complex than in the photo. I am not able to show the exact geometry.
1-About the Water/Air interface, as I said, I just saw in Test the bubble which is in Water some times( imgain you pure Water in Glass which you see some bubbles).
2- About the countor, because the air is located at top of the tank and takes longer to be warm in comparison with water. Firstly, I had expected I could make a contour with both air and water temperature variable. Secondly, every contour for each variable would be different but do no.
3. As I said, I would like to validate my simulation model with test results to modify the geometry, the position of water duct in the tank etc.

If there are other questions or unclear points, I will be glad to know. I need your support on this topic.

wishes,

[ghorrocks]

OK, so you are reproducing a test result and you want to look at design aspects like the position of the inlet, but what are the parameters important to the design? Do you want the pressure field? Flow loss? Maximum velocity? Mixing ratio? How much foam is formed? I have no idea what is important in your analysis.

Please show an image which shows the unexpectly warm air on the top of your domain.

[MNMK]

I appreciate your rapid reply.

Let me describe an example. As can be seen in the attached photo. I have a tank which is filled with water and air. The initial water height is L and the water and air temperatures are 25°C. Then warm water flows into the tank and cold water flows outside.We imagine the mass flow at the inlet and outlet are the same. Surely we have bubbles on the water surface(see photo).

What I would like to see is the temperature contour as time passes. Also, what is the air and water volume fractions in the tank. When is possible, where are the air particles in water?

wishes,

[ghorrocks]

It appears accurate modelling of the bubble and foam formation due to the inlet flow will be important. This is a challenging topic, especially as it is combined with a free surface flow.

The approach I would try first is an inhomogeneous free surface model. This will allow you to capture both the free surface effects of the inlet jet and the bulk water, but also capture a blurred water/air interface for the bubbly/foamy region. But I suspect the model of generating bubbles in this approach may not be accurate - but it is worth a try as it is the easiest approach and if it works that is great.

If that approach does not work and you want to model the bubble formation then things are going to get complex. The obvious approach would be to do a super-high resolution homogeneous free surface model, but the mesh required for this will be huge (I hope you have a supercomputer to run it). There may be other more sophisticated models which deal with the bubble formation better, I would have to look into that. But they definitely will not be for CFD beginners, so should only be considered if you have extensive CFD experience.

[MNMK]

as I described in the last example,
1.May you please describe to me why I can not make a different air and water temperature contour. (they look the same)
2.Also, when I want to monitor the water volume fraction, the scale from 0 to 1 is close to the test or 0.8 to 1 ( I mean which one is closer to reality) because I do not see in the test results a variety of volume fractions in the tank. I can guess a limited scale, but I want to be sure.

[ghorrocks]

May you please describe to me why I can not make a different air and water temperature contour. (they look the same)

Because you have modelled it incorrectly. If you want us to help you understand what you have modelled incorrectly you will need to upload your output file, and image of the results you are getting and an image of the results you expect.

Also, when I want to monitor the water volume fraction, the scale from 0 to 1 is close to the test or 0.8 to 1 ( I mean which one is closer to reality) because I do not see in the test results a variety of volume fractions in the tank. I can guess a limited scale, but I want to be sure.

I do not understand your question. Are you asking why is there not a clear interface where it goes from a volume fraction of 0 to 1? Or are you asking why is it modelling a large region of volume fraction of around 0.8?

Again, an image of the results you are getting and what you expect would be really helpful here.