[guhard]

So I am working on a project at the UNI.

I was told to simulate the turbulent interaction by modifying sources in K and epsilon equations. I am using Eulerian model with Per phase turbulence model. The source is of following equation
PI_k = 4/3 * C_d * rho_c *alpha / d_b *|U_r|^3

where:

• C_d - drag coefficient,
• rho_c - density of continuous phase - water
• d_b - bubble diameter - of air - I use C_PHASE_DIAMETER for it,
• U_r - relative velocity

Unfortunately I dont have much information about that model for turbulent interactions, hopefully, that won't be so important.
Anyway, I wrote this udf:

DEFINE_SOURCE(PI_E_wat, cell, phase_t, dS, eqn)
{
    Thread *mt;
    Thread **all_phase_t;
    Thread *thread_c, *thread_d;
    const real c1 = 3.0;
    const real c2 = 4.0;
    const real g_force = 9.81;

    real C_drag, Re_b, U_rel_x, U_rel_y, U_rel, U_r;
    real rho_c, rho_d, delta_rho, alpha_d, PI_k, PI_E;
    mt = THREAD_SUPER_THREAD(phase_t);
    all_phase_t = THREAD_SUB_THREADS(mt);

    thread_c = all_phase_t[0];  /*THREAD_SUB_THREAD(mix_thread, WATER_ID);  liquid phase */
    thread_d = all_phase_t[1];  /*THREAD_SUB_THREAD(mix_thread, AIR_ID);  gas phase */

    U_rel_x = C_U(cell, thread_d) - C_U(cell, thread_c);
    U_rel_y = C_V(cell, thread_d) - C_V(cell, thread_c);
    U_r = sqrt(U_rel_x*U_rel_x + U_rel_y*U_rel_y); /* relative velocity from udf */

    real d = C_PHASE_DIAMETER(cell, thread_d);

    alpha_d = C_VOF(cell, thread_d);

    rho_c = C_R(cell, thread_c);  /* water density */
    rho_d = C_R(cell, thread_d); /* air density */
    delta_rho = rho_c - rho_d; /* delta rho */

    Re_b = rho_c * fabs(U_rel) * d / C_MU_L(cell, thread_c);  /* Reynolds number with relative velocity*/
    C_drag = 24 / Re_b*(1 + 0.15*pow(Re_b, 0.687)); /* Drag coefficient */
    PI_k = 3.0 / 4.0*C_drag*rho_c*alpha_d / d*pow(fabs(U_r), 3);
    PI_E = 1.2 * C_D(cell, phase_t) / C_K(cell, phase_t) * PI_k;
    dS[eqn] = 0;

    return PI_E / 2;
}

Since the original model is not k dependent I assumed dS[eqn] = 0. Unluckily, I am afraid this is causing covergence problems.


Without the UDF i was able to run multiphase simulation with all possible 2-phase interaction in fluent, with Coupled pressure-velocity scheme and solution converged for steady state(good job me).
Whenever I include (accordingly written) sources for both phases and turbulent equations, I see this

iter  continuity     u-water       u-air     v-water       v-air     k-water       k-air   eps-water     eps-air         iac      vf-air     time/iter
# Divergence detected in AMG solver: k -> Decreasing coarsening group size!
# Divergence detected in AMG solver: k -> Increasing relaxation sweeps!
# Divergence detected in AMG solver: pressure correction -> Turning off correction scaling!
# Divergence detected in AMG solver: pressure correction -> Increasing relaxation sweeps!

Error: Divergence detected in AMG solver: pressure correction

Error: Divergence detected in AMG solver: pressure correction
Error Object: #f

Error: WorkBench Error: Could not handle event: SolutionStatusUpdate
Error Object: #f

I have no idea what I can do.
From other posts I learned that lower URF, or "coarsening" my mesh might help, so far, nothing works.


Can't wait for your suggestions, pointing out my mistakes or in general pointing the direction I should take.


Regards,
Michal

[guhard]

Solved the problem, thread can be deleted.

[pakk]

Good for you! It could be helpful if you give the solution to your problem in a few words, for if somebody else has a similar problem in the future.

[guhard]

Sure, although it wasn't really "fluent related".
In my formula for Reynolds I used "U_rel", but the relative velocity was assigned "U_r". As a result, Re_b became 0, thus drag coefficient became infinite.

The moral of this story is if you see error as I saw, check your variables names

[daliyou]

Hello guys,



I would like to define a two phase threads and calculate the relative velocity as you post, while it always appear erros.




I tried to fix the problem line by line. I found the problem is the line 'thread_c = all_phase_t[0]'


The code is as following:



#include "udf.h"
#include "sg.h"
#include "sg_mphase.h"
#include "flow.h"
#include "mem.h"
#include "metric.h"


/* ************************** Memory and scalar description **************** *\
UDM-0: Gb
/* ************************************************** ************************* */

DEFINE_SOURCE(Gb, c, phase_t, dS, eqn)
{

real Cb, U_rel_x, U_rel_y, U_rel, U_r, source;
real rho_c, rho_d, delta_rho;
const real g_force = 9.81;
Thread *thread_c,*thread_d,*mt;
Thread **all_phase_t; /*array*/

mt=THREAD_SUPER_THREAD(phase_t);
all_phase_t=THREAD_SUB_THREADS(mt);

thread_c = all_phase_t[0];/*THREAD_SUB_THREAD(mt, 1); THREAD_SUB_THREAD(mix_thread, WATER_ID); liquid phase */
thread_d = all_phase_t[1]; /*THREAD_SUB_THREAD(mix_thread, AIR_ID); gas phase */

U_rel_x = C_U(c, thread_d) - C_U(c, thread_c);
U_rel_y = C_V(c, thread_d) - C_V(c, thread_c);
U_r = sqrt(U_rel_x*U_rel_x + U_rel_y*U_rel_y); /* relative velocity from udf */
/* U_r=sqrt(pow(U_rel_x,2.0)+pow(U_rel_y,2.0))*/
Cb= 0.8;
rho_c = C_R(c, thread_c); /* water density */
rho_d = C_R(c, thread_d); /* air density */
delta_rho = rho_c - rho_d; /* delta rho */
source= Cb*C_MU_L(c,phase_t)/C_MU_EFF(c,phase_t)*delta_rho*g_force*C_VOF(c,phas e_t)*U_r;*/
dS[eqn]= 0.0;

return source;

}


The error message:
999999 (..\src\mpsystem.c@1172): mpt_read: failed: errno = 10054

999999: mpt_read: error: read failed trying to read 4 bytes: Invalid argument
MPI Application rank 0 exited before MPI_Finalize() with status -1073741819
The fl process could not be started.


Thanks for your reading and help!


Best regards
Dali