CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   Eulerian model (http://www.cfd-online.com/Forums/fluent/115367-eulerian-model.html)

devesh.baghel March 29, 2013 05:22

Eulerian model
 
Hi,
http://www.cfd-online.com/Forums/C:\...c.html (1).png

I am running a case of mixture (air+sand) with Eulerian model.
I reached mid of the required solution, where solid & air particles reached to outlet boundary. I want to allow only Air to pass through outlet not the solid particle but currently I loosing solid particle from the outlet along with air flow.

Can any body having any idea, how to stop solid particle at outlet and only air should escape through outlet boundary ???

Please help me......It is urgently required ..!!!!! :(

oj.bulmer March 29, 2013 11:20

How about a mass source flux at the outlet? It will gulp down the air and leave the particles alone.

OJ

chaosh March 29, 2013 12:05

In your actual geometry is there a screen or something at the outlet that traps the sand particles?

devesh.baghel March 29, 2013 12:34

Thanks guys for your reply...

OJ: I need air has to pass through outlet boundary. I want to keep amount of solid particles constant into the domain. Air is entering with some defined mass flow rate which can not be lower down. Height of domain will also be constraint throughout the simulation. Below is the geometry
http://www.cfd-online.com/Forums/dat...BJRU5ErkJggg==
http://www.cfd-online.com/Forums/dat...BJRU5ErkJggg==Lets say a vertical rectangular block is of height of 5 meter, divided into 5 equal segments. Inlet is at bottom most face & top most surface is defined at outlet. From bottom most segment, 2nd zone is filled with solid particle (like fluidized bed). Air is entering from inlet and outlet is pressure outlet. I need to keep amount of solid particles constant i.e solid particles should not leave the domain.

Please let me know if I missed any points to explain.
http://www.cfd-online.com/Forums/dat...BJRU5ErkJggg==
Please throw some light for this problem ...!!!!!

chaosh March 29, 2013 12:45

Cannot see pics...

Are solid particles coming in through the inlet? Or are they just 'there' in the system. If you want to see the effect of the solid particles on your air, you could just create a zone which is porous and mimics the effect of your solid particles on the air (treat the solid particles like you would treat a porous bed of filter media).. Then you would only need to run air through the system.

If you have solid particles coming in with the air, will they not settle out? The crucial thing that determines if the particles will settle out or advect with the air is the particle diameter. Do you know your particle diameter? The default in fluent is small - 10 microns I believe. Increasing this will trap your solid particles in your system.

You could also use a DPM and set walls to 'trap' particles.

devesh.baghel March 30, 2013 02:48

Hi

you are true chaos; first case: where solid particles are just kept on. solid particles are not entering with air, Also there is no porous zone defined in the domain.

if I can fix the escaping of solid particles just before or at outlet; can you please help me how to fix it.

Thanks in advance

oj.bulmer March 30, 2013 07:19

Have you explored a mass sink using an udf? There is a good reference of such an udf in manual, which is given for a typical bubble column, where gas is inserted in a tank of liquid, and when gas reaches the the top boundary, the mass sink "eats" only the air - or secondary phase, while acting as a free-slip wall to water so water can't escape. I remember FLUENT 14.5 has now degassing boundary as a standalone feature. I think you can't use it out-of-the-box, since you have dispersed solid which will be gulped down with that boundary.

Below is the part of UDF that eats the secondary phase:

*#include*"udf.h"
*#include*"sg.h"
*#include*"sg_mphase.h"
*#include*"flow.h"
*#include*"mem.h"
*#include*"metric.h"
*
*DEFINE_SOURCE(degassing_source,*cell,*thread,*dS, *eqn)
*{
*real*source;
*Thread**tm*=*THREAD_SUPER_THREAD(thread);
*source*= -C_R(cell,thread)*C_VOF(cell,thread)/CURRENT_TIMESTEP*;
*C_UDMI(cell,tm,0)*=*source;
*dS[eqn]*= -C_R(cell,thread)/CURRENT_TIMESTEP;
*return*source;
*}
*
See if you can implement that mass sink part for your top boundary, but to eat air instead of solids.

OJ

devesh.baghel March 31, 2013 01:43

Thank you so much OJ & Chaos

your ideas are appreciable.

OJ: As you suggested before, to define mass source flux what I did is, I have defined top most cell zone with fixed value of velocity 0 m/s for x,y & z direction. Now onwards solid particles are not loosing from outlet and only air is allowed to escape. Once solid particles reached at top surface, recirculating within the domain or else accumulating on top surface.
I guess it is working fine and requirement fulfilled.

If you find anything wrong, please let me know

Thanks guys for your ideas\suggestions :)

devesh.baghel April 3, 2013 12:40

please find below the problem statement

devesh.baghel April 4, 2013 11:20

Hi OJ

I tried to run through UDF as you suggested above, but I am getting error message below.


Copied E:\Project_Mane\April_02/E:\Project_Mane\April_02\udf_mass_sink.c to libudf2\src
(system "copy "C:\PROGRA~1\ANSYSI~1\v140\fluent"\fluent14.0. 0\sr c\makefile_nt.udf "libudf2\win64\3d\makefile" ")
1 file(s) copied.
(chdir "libudf2")()
(chdir "win64\3d")()
'nmake' is not recognized as an internal or external command,
operable program or batch file.
'nmake' is not recognized as an internal or external command,
operable program or batch file.


After finish-up of above lines, I tried to load the lib file but getting an error below:


Opening library "E:\Project_Mane\April_02\libudf2"...
Error: The UDF library you are trying to load (libudf2) is not compiled for 3d on the curent platform (win64).

The system cannot find the file specified.

E:\Project_work\April_02\libudf2\win64\3d\libudf.d ll


Please throw some light on that, it is urgently required.

thanks

oj.bulmer April 4, 2013 11:27

You can not use the same UDF! You need to devise your own with its reference and do a bit of trial and error to get the UDF working.

You need to be sufficiently comfortable in UDFs to be able to do it :)

OJ


All times are GMT -4. The time now is 08:00.