Hi,

I am using Eularian granular two phase model (water and sand as two phases). I'm trying to model water flow over the sand. The water&sand interface is predicted similarly to the lab data by FLUENT. However, I would also expect some portion of the sand will be stirred up to the upper water phase close to the free surface. As I plot out phase-2 volume fraction, there is almost nothing above the interface (defined as vf=0.5). That is, sand vf is almost 0 above the interface. It is not correct. There should be some sand concentration in the upper water phase. Do you know the probable reason for this? Thanks.

Basically my model is a rectangular flume with logarithmic velocity inlet and outflow boundary conditions. Sediment concentration profiles were measured at various locations in the test section where a sand bed was present. Upstream of the test section a rigid, flat bed was present. Lab data showed sediment concentrations existing in the upper flow phase.

I am using Eularian multiphase model with syamlal-obrien model for granular viscosity and solids pressure, lun-et-al for granular bulk viscosity and radial distribution, schasffer for frictional viscosity, algebriac for granular temperature, derived for elasticity modulus, and syamlal-obrien for phase interaction. For viscous, I am using k-e turbulence model.

Thanks.

[thanhndb]

Hi, I am doing on it as well. I used Eularian model for unsteady simulation as suggested from published works. I saw the simulation worked well.

I dont know in detail of your settings, however you should create a new material for sand and use gravitational acceleration (9.81 m/s2) for the operating condition.

Regards,

Thanh

[coalgas]

Hello,
I am working with Euler Granular model as well. I am using Coal as a granular phase. Yes, in your problem i guess, you need to define pseudo species named sand. And need to patch its volume fraction in the initialization panel. But you need to work out appropriate drag and lift coefficients. Usually Gidspaw or Syamlal O'brian will work perfectly, well it depends upon physics of your problem.
One thing, stability of these models are very much depends upon initial values.
Looking forward for to share your experiences with Euler-Granular multiphase modelling.

regards
Chetan

[thanhndb]

Hi Chetan,

You are absolute right. However one thing makes me confusing is that, when I defined new material named sand I can not find any reference proposed the Viscosity of sand particles. Could you tell me where I can find an appropriate value for that? Thank you very much in advance.

Actually i am doing on gasification in dual fluidized bed gasifiers.

Regards,

Thanh

[md7alizadeh]

Hi
the initial flow field is very important. it's better to run the model with single phase (fluid) and after achieving fully developed flow field, then swich to the two phase eulerian.
good luck

[sri99iitm]

dude can u pls help with writing udf .
Im running simulation with a cylinder ,sphere inside it for Non Newtonian fluid.
I recently started working with fluent ,hope u could lead ur helping hand
Dimensions of cylinder
L=20cms
D=4cms
d(sphere)=2cms
Parabolic Velocity equation be V=Vmax(1-r/R)^(n+1/n)

my code:
#include "udf.h"

DEFINE_PROFILE(inlet_x_velocity, thread, position)

{

real x[ND_ND];

real y,r,R;

face_t f;

begin_f_loop(f, thread)

{

F_CENTROID(x,f,thread);

r = pow(x,2)+pow(y,2);

// Vmax=4.1,R=o.o4//

F_PROFILE(f, thread, position) = Vmax *pow(1-r/R,(n+1)/n);

}

end_f_loop(f, thread)

}

Please tell the changes i have to make ,thanks in advance

[coalgas]

Hello Thanh,

I dont really remember where i read this, but it is not really possible to assign a single number to its viscosity. As, The viscosity of sand is variable and depends upon the composition of the sand, its particle size distribution, the amount of water in the sand and many other variables.

Well, i have found out paper by Biswas et al in Materials and Design 23 (2002) in which they studied app viscosity sand-water slurry on basis of sand weight fraction. For weight fraction of sand ranging from 10% to 30 % app viscosity is around 0.011 poise. you can refer to that paper.

Its good to know about your work in the area of fluidized bed. Are you studying reactions involved in it?? I mean coal-gasification reactions?? if yes, are you using 'DEFINE_HET_RXN_RATE' Macro?? it would be great to share a experience with working with proscribed macro.

regards
chetan

[thanhndb]

Hello Chetan,

Thank you so much for sharing. I have the same thought as yours about the particulate viscosity. I also took a look at the paper you mentioned and the one, Jiradilok et al. Chem Eng Sci (2008). However you mean we have to write an UDF for the particulate viscosity while defining the new pseudo material?

On the gasification reactions on the FB gasifier I dont touch at all. Because we have just start the project in a short time. In our previous project we simulated for the entrained flow gasifier just using DPM model. In that case we considered the reactions.

I will touch the reaction stuff affer finishing the hydrodynamic test.

Thank you for sharing experience Chetan,

Regards,

Thanh

[thanhndb]

Hi Alizadeh,
Thank you very much for your suggestion.
I hope we can share experience on CFD stuff.

Regards,

Thanh

[coalgas]

Yes i reckon, we need to write a new UDF for particulate viscosity while defining new 'Sand' -pseudo specie.

regards
chetan

[phire]

Originally Posted by md7alizadeh

Hi
the initial flow field is very important. it's better to run the model with single phase (fluid) and after achieving fully developed flow field, then swich to the two phase eulerian.
good luck

@md7alizadeh
Pls I am working on scour under pipeline using the eulerian multiphase model. I have two phases water and sand. However, I have not been able to get the right scour profile. I read your comment above and I have tried it to implement it but i keep getting error message and divergent message ( guess i didnt implement it well). Pls can u try to shed more light on how to solve for the single phase field first to get the fully developed flow before switching to the two phase model.
thanks

[linto]

Originally Posted by Kelly
;153765

Hi,

I am using Eularian granular two phase model (water and sand as two phases). I'm trying to model water flow over the sand. The water&sand interface is predicted similarly to the lab data by FLUENT. However, I would also expect some portion of the sand will be stirred up to the upper water phase close to the free surface. As I plot out phase-2 volume fraction, there is almost nothing above the interface (defined as vf=0.5). That is, sand vf is almost 0 above the interface. It is not correct. There should be some sand concentration in the upper water phase. Do you know the probable reason for this? Thanks.

do it for some more number iterations....do it for 10000 iterationslinto p anto

[linto]

hi i need to simulate particle size 5cm in diameter..whether5 i can use eularian granular model to simulate the flow of particle in water?

[apurv]

Hi all,

I am also working on eulerian multiphase to solve flow in fluidized beds. I have an issue of convergence when i give density as a function of pressure in a UDF. Earlier I had tried ideal gas for density but the flow is not compressible so I am now varying only density with pressure in the bed. The continuity diverges rapidly. I am not understanding what further changes to make. I have already tried adjusting URELs and solution as implicit and explicit but there was no improvement in convergence.

Please help me find out what is wrong in this case? Also, does anyone know how speed of sound is related to density for air? In UDF for density, we have to specify speed of sound and I have kept it constant.

Thanks,

Apurv

[thanhndb]

Originally Posted by coalgas

Hello,
I am working with Euler Granular model as well. I am using Coal as a granular phase. Yes, in your problem i guess, you need to define pseudo species named sand. And need to patch its volume fraction in the initialization panel. But you need to work out appropriate drag and lift coefficients. Usually Gidspaw or Syamlal O'brian will work perfectly, well it depends upon physics of your problem.
One thing, stability of these models are very much depends upon initial values.
Looking forward for to share your experiences with Euler-Granular multiphase modelling.

regards
Chetan

Hi Chetan,

I know you have experiences in using Fluent. I am now trying to simulate of fluidized bed with non-unifrom distribution of solid particles. My problem is that I dont know well how write the udf code for the rosin-rammler distribution. So coud you please give me some suggestions or give me a sample of the code?

Thanks so much in advance!

Thanh

[vagaikwa]

Hi,
Can someone please help me with defining the viscosity of the new material that I create in Granular Temperature Model. I dont know the viscosity of the material, I just have the density. What value should I input in that case in the viscosity? Thanks.

[tangfl]

If i hope to study erosion on the wall due to sand impact, do i need to assume amount of sand particles impact the wall?

[goksusoydan]

Hi @phire

We are working on scours around pipelines using eulerian multiphase model. I am creating a new material name as sand in the phases part and change its density. We couldnt get the correct scour around the cylinder. I define a sand region and water region. Patch the sand region with vf 0.5. The water takes the sand away from the channel. We couldn't fix it out. Could you please help us if you have any idea about it.

[ani4377]

Dear,
I am also trying to simulate sand bed in fluent. I am unable to add sand in the phases (That are further patched). I am using water with changed properties(density 2640kg/m3 and viscosity is very high 10^8). Will it help or is there other way.

Another question is how to add sand in fluent materials.
Please help