CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Periodic boundary conditions in 3D Eulerian granular flow simulations (http://www.cfd-online.com/Forums/fluent/84581-periodic-boundary-conditions-3d-eulerian-granular-flow-simulations.html)

dsm February 2, 2011 22:23

Periodic boundary conditions in 3D Eulerian granular flow simulations
 
Hi,

I am running a granular flow simulation for water flow inside a horizontally placed rectangular channel. The channel is initially half filled with sand (the secondary phase). I am interested in the steady state solution, and therefore using periodic boundary conditions. Since the simulations are computationally expensive I can not afford to run them in a long channel and wait until I reach a steady state. I want to study the interface between the water and the sand.

In single phase flow simulations, Fluent allows a user to prescribe mass flow rate while implementing periodic BCs and calculates pressure difference as an output. But for granular flow simulation I am not able to prescribe mass flow and have to give pressure difference between the periodic inlet-outlet. My simulation does not seem to reach a steady state. The velocities of water and sand continue to increase all the time. I have tried several grids, time steps, grain diameters, etc.

Does anyone have experience with similar type of Fluent simulations? More specifically, is there a way I can prescribe water+sand mass flow rate instead of the pressure difference as my periodic BCs? Could someone please provide information about that multiphase pressure drop (what are phasic pressures and how are they related to the net pressure) ?

Thanks.

sandisk February 2, 2011 23:52

Hi,
Even I am working on water sand slurry in pipe and am interested in the pressure drop. However, after trying with steady state, I am now running it in unsteady state. I am really not sure about your question, but is your question that how can we specify the volume fraction of sand ? I think we can do that in Boundary conditions option !!

dsm February 3, 2011 20:33

Periodic BCs with granular flow
 
For granular flow simulations Fluent doesn't allow the user to impose mass flow rate, if periodic bcs are used. I have to prescribe pressure difference between the inlet-outlet as the input parameter.
I guess, I have two main question:
1)Does anyone have experience in running granular flow simulation with periodic BCs in Fluent?
2)If yes, is there a way to prescribe mass flow rate as the input parameter for the periodic BCs. Similar as your objective, I am also interested in seeing pressure drop as the output.
**************
@Sandisk: On a different note, is your case 3D? Do you have any suggestion on choosing the ratio of cell size to sand diameter. How long does your typical simulation take to reach a steady state? Did you need to use any advanced models for granular bulk viscosity, frictional viscosity, drag, lift etc to achieve good results? I am new to granular flow simulations and any answers or suggestions will be helpful.
Thanks

sandisk February 3, 2011 21:02

@ dsm,
I am sorry I am also new in Fluent. I am using Unsteady state simulation, as I have been advised by some member that slurry flow needs to be simulated with transient simulations. I am just using the usual models and not some advanced models. Sorry, couldn't help you much.

xiaokang March 2, 2012 19:04

Quote:

Originally Posted by dsm (Post 293389)
Hi,

I am running a granular flow simulation for water flow inside a horizontally placed rectangular channel. The channel is initially half filled with sand (the secondary phase). I am interested in the steady state solution, and therefore using periodic boundary conditions. Since the simulations are computationally expensive I can not afford to run them in a long channel and wait until I reach a steady state. I want to study the interface between the water and the sand.

In single phase flow simulations, Fluent allows a user to prescribe mass flow rate while implementing periodic BCs and calculates pressure difference as an output. But for granular flow simulation I am not able to prescribe mass flow and have to give pressure difference between the periodic inlet-outlet. My simulation does not seem to reach a steady state. The velocities of water and sand continue to increase all the time. I have tried several grids, time steps, grain diameters, etc.

Does anyone have experience with similar type of Fluent simulations? More specifically, is there a way I can prescribe water+sand mass flow rate instead of the pressure difference as my periodic BCs? Could someone please provide information about that multiphase pressure drop (what are phasic pressures and how are they related to the net pressure) ?

Thanks.

Dear DSM,
Did you solve your problem finally? I am doing the same thing as you did. I saw the time is like last year, hope to discuss it.thanks!
xkyan


All times are GMT -4. The time now is 17:21.