[MissCFD]

I run a multiphase simulation with water and air as a continuous phases. I set the air density as the reference density. I ran also the same simulation but without the buoyancy activated to see the effect. And I would like to know if it’s normal to have a different value of the Pressure variable at the same point in the two simulations ? But I find the same water flowrate as a result … I thought the activation of the buoyancy just affect the absolute pressure not the pressure variable, am I wrong ?

Does the outlet boundary condition (opening with 0 relative pressure) have a different meaning in the two simulations ?

Is there a way to see the same Pressure in the two simulations with the creation of an expression ?

Does the relative pressure at the outlet boundary condition need to contain the water hydrostatic pressure ?

Thank you for your help !

[Opaque]

In general, for natural convective flows, the pressure fields would be different.

If gravity is present definitely, the absolute pressure must contain the hydrostatic component, but adding buoyancy to the system is more than just changing the definition of absolute pressure.

Let see it with an example, a container filled with water and air on top. We put a sphere in the system and expect it to float. How will the system account for the buoyancy in such a case? We activate the gravity, and solve the problem.

The solution is expected to be a rise in water height followed by an increase in pressure at the bottom, i.e. the pressure field is different everywhere.

Adding gravity is not an offset in pressure ONLY, but much else.

[MissCFD]

Thank you very much for your answer and your example. What do you mean by natural convective flows ?

In my case, it’s just an unfilled pipe with a reservoir in the middle at a height lower than the pipe. I modelise it as a multiphase flow with water and air which are continuous phases. So for me, with and without the buoyancy activated should give me the same value for the Pressure variable, doesn’t it ?

Also, an other question that I have in my mind, I set the air density as the reference density. So, I was wondering if the value of the relative pressure set at the outlet needs necessary to contain the water hydrostatic pressure (because there is a water level at the outlet) ?

Originally Posted by Opaque

In general, for natural convective flows, the pressure fields would be different.

If gravity is present definitely, the absolute pressure must contain the hydrostatic component, but adding buoyancy to the system is more than just changing the definition of absolute pressure.

Let see it with an example, a container filled with water and air on top. We put a sphere in the system and expect it to float. How will the system account for the buoyancy in such a case? We activate the gravity, and solve the problem.

The solution is expected to be a rise in water height followed by an increase in pressure at the bottom, i.e. the pressure field is different everywhere.

Adding gravity is not an offset in pressure ONLY, but much else.

[Opaque]

I assume there is no flow at all in your system; therefore, the solution for the Pressure variable should be the same assuming absolutely everything else is the same.

I also assume you have set up your initial conditions to the expected solution; otherwise, how is supposed the system to converge with air on top and water at the bottom? What will drive it to the static condition?

Natural convective flows are "Buoyancy driven flows"

[MissCFD]

I initialized my simulation with the automatic values, just air and no water at the beginning. There is a mass flow condition at the inlet of my domain that will drive the water flow. I attached a picture of the kind of flow I want to simulate if it can help you to visualize

Originally Posted by Opaque

I assume there is no flow at all in your system; therefore, the solution for the Pressure variable should be the same assuming absolutely everything else is the same.

I also assume you have set up your initial conditions to the expected solution; otherwise, how is supposed the system to converge with air on top and water at the bottom? What will drive it to the static condition?

Natural convective flows are "Buoyancy driven flows"

[ghorrocks]

Also be aware that the definition of the variable "pressure" changes when you turn gravity on. With gravity "pressure" includes a hydrostatic component. Look in the documentation for details. So this is another way how pressure between your two runs could be different.

[Opaque]

Ok. Let us try another approach in the thinking process.

The introduction of reference values creates confusion all over. The final solution should be irrelevant to the reference value selected, agree?

Let say the air in your model is compressible (ideal gas). Which pressure in the solution do you expect to match to the experiment/real problem?

If we agree on the above, I will assume you answered Absolute Pressure to the last question; therefore, the variable Pressure would be a function of which reference value you selected since it must be the Absolute Pressure minus reference value contributions, correct?

That is, if you select Air, the Pressure field would be different (hopefully by a constant) to the Pressure field if you selected Water

[MissCFD]

Yes the absolute pressure includes the hydrostatic pressure with the density reference we set. So if I put the air density as reference what happens for the relative water pressure ? Is the water relative pressure included automatically the value of the water hydrostatic pressure in ?

Originally Posted by ghorrocks

Also be aware that the definition of the variable "pressure" changes when you turn gravity on. With gravity "pressure" includes a hydrostatic component. Look in the documentation for details. So this is another way how pressure between your two runs could be different.