Does C_VOF_G works.....does it calculate the gradient vector of volume fraction of the phases...

shekhar

Hi,

Yes. It does work but you have to set the following in Fluent text prompt.

solve/set/expert/Keep temporary solver memory from being feed? : yes

Default is "no".

Thank you

FJ

Thanks FJ, I already did that but still when i start my iterations, this error comes:

Error: FLUENT received fatal signal (ACCESS_VIOLATION) 1. Note exact events leading to error. 2. Save case/data under new name. 3. Exit program and restart to continue. 4. Report error to your distributor. Error Object: ()

shekhar

This macro seems to be restricted to the VOF model. Are you using the VOF model ? Thomas

no thomas, i am using the eulerian multiphase model... actually i am adding the capillary pressure as a momentum source term in my multiphase flow problem, and there i need to compute the gradient of volume fractions of the phases..........how is that possible with eulerian multiphase model !!!

shekhar

I wouold suggest to write a UDS to store the VOF value and then you can get your VOF gradient using the command C_UDS_G.

Thomas

Hi,

I think all gradient macro doesn't have the values at first iteration. So at first iteration, you don't hook your function and the next iteration , let's hook it.

I think it will be O.K.

Thank you

FJ

Thanks FJ, I already tried that too....but this time there comes a different error: Error: divergence detected in AMG solver: pressure correction Error Object: ()

Shekhar

Hi,

Oh, really?. If your UDF function is not called, is your calculation no problem?.

Thank you

FJ

thanks FJ, Ya there isnt any problem when i do not link the UDF. The problem only comes after linking the UDF. I guess its not calculating the gradient of the volume fractions. Well, I am pasting here the UDF as well. this UDF is for adding a momentum source term to phase (i am modeling the hydrodynamics of gas and liquid flow in a packed bed).

#include "udf.h"

DEFINE_SOURCE(cell_x_source, cell, thread_l, dS, eqn) { real source, thita_s, thita_g, rho_g, rho_l, n1, n2; Thread *mixture_thread = THREAD_SUPER_THREAD(thread_l); Thread **pt; pt = THREAD_SUB_THREADS(mixture_thread); thita_g = C_VOF (cell, pt[0]); thita_s = C_VOF (cell, pt[2]); rho_g = C_R (cell, pt[0]); rho_l = C_R (cell, pt[1]); n1 = C_VOF_G (cell, pt[0]) [0]; n2 = C_VOF_G (cell, pt[2]) [0]; source = 96.015*(pow((1-thita_g),0.5)/pow(thita_s,1.5)*n2 + 1/(pow(thita_s,0.5)*pow((1-thita_g),0.5))*n1); dS[eqn] = 0.0; return source; }

Hi,

Year, I've got the same error as your remarks. I considered your UDF but I don't really understand why this kind of error is happened. So I re-coded with my way for unsteady calculation.It was O.K under my conditions.

#include "udf.h"

DEFINE_EXECUTE_AT_END(calculate_value)

{ Thread *t; Thread **pt; cell_t c; Domain *domain = Get_Domain(1); real thita_s, thita_g, rho_g, rho_l, n1, n2;

mp_thread_loop_c (t,domain,pt)

{

begin_c_loop(c, t)

{

thita_s = C_VOF (c, pt[2]);

thita_g = C_VOF (c, pt[0]);

rho_g = C_R (c, pt[0]);

rho_l = C_R (c, pt[1]);

n1 = C_VOF_G (c, pt[0])[0];

n2 = C_VOF_G (c, pt[2])[0];

C_UDMI(c, t, 0) = thita_s;

C_UDMI(c, t, 1) = thita_g;

C_UDMI(c, t, 2) = rho_g;

C_UDMI(c, t, 3) = rho_l;

C_UDMI(c, t, 4) = n1;

C_UDMI(c, t, 5) = n2;

C_UDMI(c, t, 6) = 96.015*(pow((1-thita_g),0.5)/pow(thita_s,1.5)*n2

+ 1/(pow(thita_s,0.5)*pow((1-thita_g),0.5))*n1);

}end_c_loop(c, t)

} }

DEFINE_SOURCE(cell_x_source, c, t, dS, eqn) { real source;

source = C_UDMI(c, t, 6); dS[eqn] = 0.0;

return source; }

Thank you

FJ

thanks FJ, i ll try it n let u know

shekhar

[Fourati]

Hello, I am having the same error message when trying to hook a source term in liquid phase of gas/liquid flow that is function of C_VOF_G. Did you find what's going on with yours please ? Thanks

[pannsave]

Hi Fourati,
I am having problem with the implementation of volume fraction gradient in E-E simulation for my source term in momentum equation.
I have used the following UDF to store my memory which is used for my source term calculation. When i run my simulation, the user defined memory shows always zero, which results zero in my source.
I would be thankful to you if you have any suggestion for my problem.
#include "udf.h"
#define CON1
DEFINE_ADJUST(myadjust,d){ if(ROOT_DOMAIN_P(d)){cell_t c;Thread *t;
Domain *sd = DOMAIN_SUB_DOMAIN(d,0); Alloc_Storage_Vars(sd,SV_VOF_RG,SV_VOF_G,SV_NULL); Scalar_Reconstruction(sd, SV_VOF,-1,SV_VOF_RG,NULL); Scalar_Derivatives(sd,SV_VOF,-1,SV_VOF_G,SV_VOF_RG,&Vof_Deriv_Accumulate);thread _loop_c(t,d)
if(FLUID_THREAD_P(t))
{Thread *pt = THREAD_SUB_THREAD(t,0);
begin_c_loop(c,t) {C_UDMI(c,t,0) = C_VOF_G(c,pt)[0];
C_UDMI(c,t,1) = C_VOF_G(c,pt)[1];}end_c_loop(c,t)}
Free_Storage_Vars(sd,SV_VOF_RG,SV_VOF_G,SV_NULL);} }