CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   Which software can model more than two phase fluid (http://www.cfd-online.com/Forums/main/5081-software-can-model-more-than-two-phase-fluid.html)

J.Yang August 27, 2002 11:55

Which software can model more than two phase fluid
 
Hi, everyone,

I want to model the slurry with at least three size of particles distributing in it. I want to know the separation process of the slurry under gravity sedimentation. Anyone know which software has such capability?

Thank you!

J.Yang

alain August 27, 2002 14:17

Re: Which software can model more than two phase f
 
Hi,

CFX can do this kind of simulations.

Best regards

Jan Rusås August 27, 2002 14:25

Re: Which software can model more than two phase f
 
I think most software with an euler euler solver for multiphase flow should be able to do that. If I understand it correctly then it is just a matter of solving an additional similar equation for each size group. I standard answer to such a question in this forum is try to contact the different software vendors and ask them your question - look under the resources pages. Good luck


J.Yang August 27, 2002 22:02

But CFX only can model two phase fluid
 
Hi, thank Alian and Jan for your helping. But CFX only can model two phase fluid. My problem is that I want to model the slurry with various sizes of dispersed solid particles distributing in continuous water. The solid particle size ranges from 0.2 micron to 2 mm. So I need model every particle group using one phase.

Thank you very much!

J.Yang

jan rusås August 28, 2002 03:21

Re: But CFX only can model two phase fluid
 
You still only have two phases - slurry and the particles! The number of size groups has nothing to do with a phase - they are still "a particle phase" You can then solve more equations for your dispersed phase representing each size group. Jan

Free_and_SIMPLE August 28, 2002 03:24

Re: But CFX only can model two phase fluid
 
IMO You should look for a program which solve the transport equations for particle/droplet groups (probably in a Lagrangian way). The particle size range is subdivided into appropriate classes and a transport equation for each particle group is solved.

I think using a multiphase approach is not appropriate for your problem and computational not feasible. If you have a lot of different particle size classes they should not be modelled as a continuum.

jan rusås August 28, 2002 04:58

Re: But CFX only can model two phase fluid
 
Dear Free_and SIMPLE I could be wrong on this, but then hopefully will be corrected.

The selection of how to model a system with a fluid and particles depends mainly on the load, how dense or diluted the dispersed phase is. For very high mass loading is the euler/euler prefered. The main problem with euler/euler and multiple size groups is that a transport equation has to be solved for each size group - so it depends on your hardware. But 3 or 4 size groups should not be a problem, that is at least what I have solved on my small PC. Maybe there also exist euler/euler codes that can handle a size distribution. Jan

Free_and_SIMPLE August 29, 2002 02:28

Re: But CFX only can model two phase fluid
 
Dear Jan,

of course you are correct. The model of choice depends on the load. The problem I see with Euler/Euler continuum treatments is the physical sound formulation. If you treat the size groups as scalar quantities with global interactions it will work, but if you are looking for a "real" multiphase model, where each group has its own velocity field you might run into troubles. An alternative can be a pdf approach which is very common in combustion and spray simulation.

Best Regards.

alain September 3, 2002 06:37

Re: But CFX only can model two phase fluid
 
Hi,

As a matter of fact CFX has the capability of solving EULER/EULER multiphase with more than 2 phase :

In CFX 5 (unstructured) you can use pure euler/euler with up to 5 phase but there is no granular euler which you may need.

In CFX 4 (structured) an arbitrary number of phases may be specified. There is also a specific model MUSIG (Multiple-Size Group) which is available to represent the size distribution of a disperse phase.

There is certainly other software that can do this kind of simulation i think.

From theorical point of view I don't think that euler lagrange approach is suitable for sedimentation problem as this model suppose a finite residence time for particules.

Best regards

Alain


All times are GMT -4. The time now is 07:59.