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/)
-   -   Particle interaction in twoPhaseEulerFoam (http://www.cfd-online.com/Forums/openfoam-solving/94081-particle-interaction-twophaseeulerfoam.html)

caltantz November 4, 2011 15:38

Particle interaction in twoPhaseEulerFoam
 
Hello everyone,

Although I have been following the forum for some time, this is my first post.

I have very recently switched my interest from combustion to multiphase flows. I intend to study dense particle-gas flows (fluidized bed gasifiers) and I am interested in open source software to do it.

I came across the twoPhaseEulerFoam solver but unfortunately my C++ knowledge is still very limited and I have some questions concerning it.

From what I have understood by following the code itself and the forum, the solver uses two approaches to describe the particle interactions, the powder modulus approach and the kinetic theory of granular flows approach.

I) Concerning the powder modulus, if I am not wrong, the code uses the ppProperties to calculate the solids pressure. How are the solids shear stresses taken into account, if at all?

II) When using the ppProperties, the alphaMax criterion is not fulfilled even if the packing limiter is employed (parameters taken from Bouilard et al., AIChe 1989). Has anybody experienced this problem before?

III) The same problem occurs when I use the KTGF for the more detailed description of the particle collisions. The solution to this problem would be to use an extremely small time step? (The simulations conducted up to now employ dt=10^-6). Are there any other possible solutions? Am I doing something wrong?

I am using the version 2.0.1 and I try to simulate toy problems without employing turbulence modeling.

Thank you very much

alberto November 5, 2011 14:25

Quote:

Originally Posted by caltantz (Post 330768)
I) Concerning the powder modulus, if I am not wrong, the code uses the ppProperties to calculate the solids pressure. How are the solids shear stresses taken into account, if at all?

You can specify a constant viscosity. There is no other effect accounted for in that model (See the literature. For example the first works of Gidaspow).

Quote:

II) When using the ppProperties, the alphaMax criterion is not fulfilled even if the packing limiter is employed (parameters taken from Bouilard et al., AIChe 1989). Has anybody experienced this problem before?
Yes, that is correct. The model does not specify a strict maximum packing, so the phase might pass the packing.

Quote:

III) The same problem occurs when I use the KTGF for the more detailed description of the particle collisions. The solution to this problem would be to use an extremely small time step? (The simulations conducted up to now employ dt=10^-6). Are there any other possible solutions? Am I doing something wrong?
This is a problem in the code and how the particle pressure is managed (explicitly) in the implementation of the kinetic theory. I did some work on the topic (Warning: maketing my own stuff! :-)) http://www.sciencedirect.com/science...32591011003652

Best,

caltantz November 5, 2011 15:04

Hi Alberto,

Thank you very much for your fast and to the point response.

I will study your paper and the code in detail in order to understand the problem thoroughly. I will try to implement your ideas myself.

I am currently trying to find an appropriate tool for simulating fluidized bed gasifiers and I started from open source software since I am much more favorable to them. I will try as much as possible not to work with commercial codes.

Can I ask you another question. I saw in your paper that you have some experience with MFIX. What is your opinion about this tool?

All the Best,

Christos

alberto November 5, 2011 17:07

Quote:

Originally Posted by caltantz (Post 330865)
Can I ask you another question. I saw in your paper that you have some experience with MFIX. What is your opinion about this tool?

MFIX has many models for gas-particle flows, some very recent, since it is used as research code. The features of the code are documented, and it is quite easy to use.

In my opinion, the major limitation is the capability of dealing with complex geometries. However, in recent releases the cut-cell technique was implemented to try to overcome this problem.

Best,

caltantz November 5, 2011 17:37

Thanks for the information Alberto.

All the Best,

Christos

OvGU August 3, 2014 05:25

Quote:

Originally Posted by caltantz (Post 330768)
Hello everyone,

Although I have been following the forum for some time, this is my first post.

I have very recently switched my interest from combustion to multiphase flows. I intend to study dense particle-gas flows (fluidized bed gasifiers) and I am interested in open source software to do it.

I came across the twoPhaseEulerFoam solver but unfortunately my C++ knowledge is still very limited and I have some questions concerning it.

From what I have understood by following the code itself and the forum, the solver uses two approaches to describe the particle interactions, the powder modulus approach and the kinetic theory of granular flows approach.

I) Concerning the powder modulus, if I am not wrong, the code uses the ppProperties to calculate the solids pressure. How are the solids shear stresses taken into account, if at all?

II) When using the ppProperties, the alphaMax criterion is not fulfilled even if the packing limiter is employed (parameters taken from Bouilard et al., AIChe 1989). Has anybody experienced this problem before?

III) The same problem occurs when I use the KTGF for the more detailed description of the particle collisions. The solution to this problem would be to use an extremely small time step? (The simulations conducted up to now employ dt=10^-6). Are there any other possible solutions? Am I doing something wrong?

I am using the version 2.0.1 and I try to simulate toy problems without employing turbulence modeling.

Thank you very much

Hi caltantz
I think I can get some idea that Im struggling last few days. Actually I am trying to simulate euler model for a gas-solid particles. At the 1st time, only passing through Air but Secondly with combustion reactions.

for the first case I can use twoPhaseEuler but for the second case twoPhaseEuler doesn't work because of reactions. can you give some idea which solver or how I can fix this problem?

OvGU August 3, 2014 05:28

Quote:

Originally Posted by alberto (Post 330876)
MFIX has many models for gas-particle flows, some very recent, since it is used as research code. The features of the code are documented, and it is quite easy to use.

In my opinion, the major limitation is the capability of dealing with complex geometries. However, in recent releases the cut-cell technique was implemented to try to overcome this problem.

Best,

Hi Alberto
I think I can get some idea that Im struggling last few days. Actually I am trying to simulate euler model for a gas-solid particles. At the 1st time, only passing through Air but Secondly with combustion reactions.

for the first case I can use twoPhaseEuler but for the second case twoPhaseEuler doesn't work because of reactions. can you give some idea which solver or how I can fix this problem? I have solution in fluent but can't understand how to forward in OpenFoam with this 2 conditions.


All times are GMT -4. The time now is 04:35.