Hello everybody. I am a student of chemical engg. I have undertaken "CFD simulation of Gas-liquid-solid (3phase) fluidized bed" as project work. I have created geometry & meshed model using gambit. I am using fluent as solver. The column's diameter is 100 mm, and height is 1.8m. Liq velocity is 0.1 m/s & gas velocity is 0.01 m/s. Solid diameter is 2.5 mm & density 1200 kg/m3. Solid quantity in the column is 1 Kg.

In this system, liquid(water) is primary phase & air & solid (glass beads) are secondary phase.

I am using Realizable K-e model & mixture multiphase model.Simulation is unsteady with 0.01 s timestep.Buoyancy is activated. (9.81 m/s2 in -ve Y direction). Energy equation is not activated. Gidaspow drag model is used for solid - liq & Schiller naumann for liq-air.

There is a perforated plate of 20% opening at the joining of distribution section(bottom cone) & test section(top cylinder), which is used for uniform distribution of fluids as well as entrainment of solids My problem is I am not able to model it correctly for solid phase.

First I have modeled it as porous jump boundary condition. For fluids I have calculated loss coefficient. But for solids whatever value I give, solids pass through the perforated plate.

Later I have modeled the 510 square holes (1.6mm X 1.6mm). But still solids (dia=2.5mm) are passing through the perforated plate and resting on the air sparger located 50 mm below the perforated plate.

Now I am frustrated.

CAN ANYONE PLEASE HELP ME ??

I don't have experience in 3 phases CFBs. I'm modelling the classic CFBCs. I don't have experience in porous media e.t.c. But why do you use mixture model instead of Euler-Euler-Euler? Mixture model assumes that all phases have the same velocities (no slip velocity) or that they don't have the same velocities but they will equilibrate after a sort period of time. It's not strange that your particles pass through the narrow holes if you model them with mixture or Euler models. In Euler model, the only importance of the diameter is in drag coefficient. I would propose that if you can you should set the vollume fraction of solids zero at this boundary.

P.S. if you will enable Energy Equation in the future you should know that in the Mixture model there is only ONE temperature. The temperature of the Mixture.

At the outset, thanks a lot Aris.

I have selected mixture model because porous jump boundary condition is not available with Euler model. Even when I modeled the holes & used Euler model particles passed through the holes.

I am not sure that I can define the volume fraction of solids zero at the perforated plate. Anyway, Thank you very much.

I would be extremely obliged, if you could help me any further.

Regards.

I didn't know about porous media.

About the things I Know: - I'm modelling CFBs so i have a distributer (a plate with small holes). I set it as velocity inlet and I set the velocity of gas. Then i set the volume fraction of solids=0. Thus solids don't pass from this boundary. The problem is that in your case you have to set in the plate an outlet boundary condition. I don't know how you will set for volume fraction of solids Neuman b.c., and in the other hand you will have Dirichlet boundary condition for the other phases. - The simplifications of Mixture model don't mesh up Physics? - Don't forget that in Mixture-Euler the solids are modeled as a continuum media thus it's not strange that particles are passing through your holes. - A thought may be to define this boundary as velocity inlet (with negative velocities for gas and liquid) and zero volume fraction for particles.

Dear Aris, At the outset, I am extremely sorry for not replying early. I think my problem is solved. I was not able to define zero volume fractions of solids at porous plate so I defined the fixed values for porous plate for solids. I defined zero value for all 3 velocity components of solids. This ensured that solids are not passing through the porous plate. But I am not sure whether my approach is correct. I will be extremely grateful for any help I get. Thanks a lot for your support. --Sanyo

[shadowfax]

Hi Sanyo,

I know this post is from a long time ago, but I also have the same problem that my particles are passing through the porous plate. I tried your way and it seems to be working. However, did you manage to find out whether this approach is correct ?