CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   CFX (https://www.cfd-online.com/Forums/cfx/)
-   -   Homogeneous model (https://www.cfd-online.com/Forums/cfx/103168-homogeneous-model.html)

 talebzarei June 12, 2012 15:40

Homogeneous model

I am simulating a typical distillation tray of Air and water with free surface model and homogeneous model in CFX. System has spray regime so air is continues and water is dispersed on the tray
As you know, in distillation tray, we have a cross contact between gas and liquid. I have some question.
1) Can I use from homogenous model for this system?

I have read in previous posts that “If the phases never mix then use the homogenous model”.
But in my system, gas and liquid mixed together and a turbulent regime of gas and liquid and finally a spray regime was created in tray. The Homogeneous model simulated the spray regime much better than the inhomogeneous.

2) In gas out boundary condition because a same equation was solved for gas and liquid, the separation of gas and liquid in gas out boundary condition is not possible. Is it true?

Because in my simulations, most of water was exited from gas out in homogeneous model, In the inhomogeneous model, this problem was not accrued but the mixing of water and gas is not good.

 ghorrocks June 12, 2012 19:52

Quote:
 I have read in previous posts that “If the phases never mix then use the homogenous model”. But in my system, gas and liquid mixed together
Then that means the homogenous model is not suitable. You need the inhomogeneous model, or possibly a eular-eular model.

2) In the homogeneous model it is usually easy to place a boundary above or below the interface so it only sucks one phase. Your comments suggest the homogenous model is not appropriate so no wonder you had difficulties with this.

 talebzarei June 13, 2012 02:09

I have seen some simulations of distillation trays with VOF in Fluent which have a same flow regime with my system. I have two Question?
1) According to homogeneous definition in CFX documentation that ”In homogeneous multiphase flow, a common flow field is shared by all fluids”. Is Homogeneous and free surface model in CFX as the same as VOF in Fluent? Can CFX simulate VOF model. How? :confused::confused::confused:
2) In CFX documentation have given tow circumstances for homogenous model. The first is a flow under gravity, where the phases have completely stratified, for example a free surface flow. But other is
“If the flow is drag dominated, that is, interphase transfer rate is very large, and there are no body forces, the phase velocities will tend to equalize over very short spatial length scales” I think this case is agree with my simulation. Because gas and liquid effect on each other significantly and drag is very large. This is against your comment” “If the phases never mix then use the homogenous model”.
That’s right?

 ghorrocks June 13, 2012 07:37

Quote:
 Is Homogeneous and free surface model in CFX as the same as VOF in Fluent?
Not the same but they are the equivalent models in both simulation packages.

The two cases you quote are very different. The phases share a velocity field in free surface modelling as the phases never mix. So any point in space or time is either phase 1 or 2, or transitioning between the two at the interface. The phases can also share a velocity field when there is no slip between the phases - imagine tiny nanoparticles in a flow, they will just go where the flow goes with no slip.

 talebzarei June 13, 2012 14:51

But I have been seen some gas-liquid mixing which was simulated with VOF method( for example valve tray). I also have seen structure Packing simulation with VOF in CFX which has a gas-liquid contacting between phases. that's true?
and as CFX documentation mentions that we can use from homogeneous model, If the flow is drag dominated or interphase transfer rate is very large. can you give me some example of this flow.

 ghorrocks June 13, 2012 18:15

I do not understand your first sentence.

As for examples - my previous post quoted two examples.

 sjtusyc June 20, 2012 09:24

You need the inhomogeneous model, or possibly a eular-eular model.

As my understanding,inhomogeneous is also a euler-euler model,am wrong?
The multiphase flow is governed by PDE.

 ghorrocks June 20, 2012 18:50

Both homogeneous and inhomogeneous approaches are eularian models. The alternative is lagrangian models and that means particle tracking.

 sjtusyc June 20, 2012 21:41

But i am still confused about the difference between homogeneous model and the inhomogeneous model.As i now the former model share the velocity field,the latter have each field for the the specific phase.
As you said before"The phases share a velocity field in free surface modelling as the phases never mix. So any point in space or time is either phase 1 or 2, or transitioning between the two at the interface. The phases can also share a velocity field when there is no slip between the phases - imagine tiny nanoparticles in a flow, they will just go where the flow goes with no slip."
But as i know the definition of the multiphase is that phases mix in macroscale, so any point in space or time is either phase 1 or 2 even they mix.And the interface between two phase satisfy the no-slip condition.
sometimes i confused about some concepts,should i turn to the control equations for help?

 ghorrocks June 21, 2012 06:14

I do not understand your question.

 sjtusyc June 21, 2012 08:40

Sorry,i will try to be clear next time.

 All times are GMT -4. The time now is 15:57.