CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   OpenFOAM Running, Solving & CFD (http://www.cfd-online.com/Forums/openfoam-solving/)
-   -   Some clarifications about twoPhaseEulerFoam (http://www.cfd-online.com/Forums/openfoam-solving/59202-some-clarifications-about-twophaseeulerfoam.html)

mighelone January 22, 2008 14:04

Hello to everybody, I need
 
Hello to everybody,

I need same clarifications about the granular kinetic theory used in twoPhaseEulerFoam.

1) what is the meaning of the properties in the file ppProperties?

2) To avoid unphysichal packaging (alpha to high) do I need to define packingLimiter as Yes?

3) What is the difference between alphaMax and alphaMinFraction in the file kineticTheoryProperties? And the difference between alphaMax in ppProperties and kineticProperties?

4) What is the meaning of HrenyaSinclairCoeffs?

5) Is the keyword equilibrium the switch to solve the PDE equation for granular temperature or to calculate it as function of alpha?

5) What are the guidelines (i.e. Reynolds number) to switch on turbulence models in granular flows (I'm a newby about this kind of flows)?

Thank you for your attention!

Michele

alberto January 22, 2008 19:29

Hello Michele, 1. The ppPro
 
Hello Michele,

1. The ppProperties contains the values of the parameters used when you switch the particle-force "model" on, setting g0 to a positive value. This means that you have an additional term in the alpha equation and in the momentum equation to try to prevent overpacking.

2. Theoretically no. The packing limiter algorithm is a trick developed to avoid the packing limit in some particular models (Gibilaro's particle-bed model and its extensions), where there's no normal stress in the particulate phase to avoid overpacking.
If you use the kinetic theory or the particle force model, you should not need the limiter.

The current implementation of the two-fluid granular model in OF is not always stable if you switch the kinetic theory on however, because of the explicit management of the granular pressure.

3. alphaMax is the maximum value allowed for the dispersed phase. alphaMinFriction is the minimum value of the dispersed phase at which the particulate phase frictional stresses starts to act.

alphaMax in kineticTheoryProperties is used if the kinetic theory is on, the other if the particle force model is on.

4. Check the paper from C. Hrenya and J. Sinclair. They're parameters used in their formulation of the two fluid model.

5. If equilibrium is set to true, an algebraic equation (local equilibrium) is solved instead a PDE for the granular temperature.

6. Sorry to answer with a question, but what kind of system are you trying to simulate?

Usually turbulence models are not adopted in dense simulations (bubbling fluidised beds), while are used when the gas phase is faster and the system more dilute (risers).

I hope this helps.

With kind regards,
Alberto

mighelone January 23, 2008 06:36

Hello Alberto, thank you fo
 
Hello Alberto,

thank you for your detailed answer!

I'm a newbie about multiphase flow!
Can you give same references about the particle-force model, since I don't have understand very well its functions? Is it included in Mfix?

"Effects of particle-phase turbulence in gas-solid flows" is the paper of C. Hernia and J. Sinclair that you are referring?

About the object of my simulations: at the moment I'm training with multiphase flows performing same test cases, in particular fluid and spouted beds. In future the object of my attention is fixed bed gasification.

Is the granular theory still valid for fixed bed, and OpenFoam support the packed options, like fluent and mfix?

My last question: is the implementation of species transport and homogeneous and heterogeneous reaction in openfoam a complex task?

best regards

Michele

miles_davis September 29, 2011 06:56

Quote:

Originally Posted by alberto (Post 181240)
Hello Michele,

1. The ppProperties contains the values of the parameters used when you switch the particle-force "model" on, setting g0 to a positive value. This means that you have an additional term in the alpha equation and in the momentum equation to try to prevent overpacking.

Hi

So this mean that if go is set to zero, the PPproperties file is not used and no particle-particle interaction are taking into account?

regards,

alberto September 30, 2011 10:45

No, if the kinetic theory is used, g0 must be set to zero in ppProperties, and the particle pressure is accounted for elsewhere in the kinetic theory model.

Best,
Alberto

miles_davis October 7, 2011 11:57

Sorry Alberto but I still don't get it.

I am simulating bubbles in water..
Due to the low vol. fraction, I don't want to account for bubble/bubble interaction
How can I do so?
How can I cancel any particle pressure calculation/kinetic theory model or granular temperature?

I have tried to look in the solver but my c++ is very bad...

Regards,

alberto October 8, 2011 01:07

Quote:

Originally Posted by miles_davis (Post 327076)
Sorry Alberto but I still don't get it.

I am simulating bubbles in water..
Due to the low vol. fraction, I don't want to account for bubble/bubble interaction
How can I do so?
How can I cancel any particle pressure calculation/kinetic theory model or granular temperature?

I have tried to look in the solver but my c++ is very bad...

Regards,

Set

- g0 to zero in ppProperties
- set kineticTheory to off in kineticTheoryProperties

You can refer to the bubbleColumn tutorial for a case without the particle pressure and without kinetic theory models. Alternatively, you can use the bubbleFoam solver.

Best,

miles_davis October 11, 2011 11:01

Thanks for the help dans the information you always provide Alberto.

I have used the settings you have juste mentioned.

Miles

P.S.: I can't use bubblefoam. I test the tutorial on my case and it did not work. That's why I decided to use twopahseeulerfoam. I am starting to get results. I do not have time to set up another solver.

alberto October 11, 2011 11:17

Quote:

Originally Posted by miles_davis (Post 327492)
Thanks for the help dans the information you always provide Alberto.

I have used the settings you have juste mentioned.

Good

Quote:

P.S.: I can't use bubblefoam. I test the tutorial on my case and it did not work. That's why I decided to use twopahseeulerfoam. I am starting to get results. I do not have time to set up another solver.
Interesting. The two solvers (bubbleFoam, twoPhaseEuler) share most of the algorithms. The major difference should be the hard-coded drag model.

Best,

Alberto

theja October 26, 2011 08:49

dear alberto

i am simulating dust particles in room with air.
in the twophaseEulerFoam how to simulate the dust particles.. and how to change the dust particle sizes

thanking you in advance

alberto October 26, 2011 11:14

Quote:

Originally Posted by theja (Post 329533)
dear alberto

i am simulating dust particles in room with air.
in the twophaseEulerFoam how to simulate the dust particles.. and how to change the dust particle sizes

thanking you in advance

I do not think an Euler-Euler hydrodynamic model is the best choice for this kind of application (What is the average particle volume fraction?).

If what you want to simulate is simply the normal environmental dust in the environment, its concentration is way too low for the equations in the hydrodynamic limit to be valid. You either can do rough approximations like considering the particles as a transported scalar, or you need more sophisticated models (High-order moment methods or Euler-Lagrange).

Best,

k.kshitij November 7, 2012 01:11

About Using only Frictional viscosity
 
Dear All,

I'm simulating gas-solid flows with high sold phase fraction. I don't want to use the kinetic theory and the particle-particle interaction force or specify a constant viscosity to the solids.
I want the solid viscosity to be calculated from the frictional model, such as Schaeffer, How can this be implemented?
I have set the the g0 term in ppProperties to 0 and also kept the kinetic theory off. Is this the right way.

Thanking in aniticipation

rvmedina20 January 16, 2014 20:14

hey guys,

does anybody have good references for the twophaseeuler model? I am hoping to implement this solver, and would like to understand what the solver is doing and how the parameters are being used. Any help would be highly appreciated. (sorry if i am posting in wrong thread, just thought it seemed more appropriate than starting a new thread). Thanks

-Rick

sharonyue March 25, 2014 22:54

Quote:

Originally Posted by rvmedina20 (Post 470426)
hey guys,

does anybody have good references for the twophaseeuler model? I am hoping to implement this solver, and would like to understand what the solver is doing and how the parameters are being used. Any help would be highly appreciated. (sorry if i am posting in wrong thread, just thought it seemed more appropriate than starting a new thread). Thanks

-Rick

Check this out:

http://powerlab.fsb.hr/ped/kturbo/Op...chePhD2002.pdf

A skeleton of this model you can find in OpenFOAMwiki, wrote by Alberto:

http://openfoamwiki.net/index.php/BubbleFoam


All times are GMT -4. The time now is 17:12.