[ana86]

Dear all,



I want to write an udf to calculate a temperature that depends on the concentration of my material at the interface.

But when I use the macro C_YI(c,t,i) in my code the simulation crashes.

Can you help me, how to use this macro, or does anybody know, what is the best way to take the concentration (or mass fraction) into account?

I use the model "species transport". And I have a mixture-material consisting of two materials.



Thank you for your help!

[AlexanderZ]

show your code, describe in details, how do you start your simulation

best regards

[ana86]

Hello Alexander,



for the test I used a simple 2D Case.

The interface motion depends on the temperature and the temperature on the other hand, depends on the concentration at the interface. At the interface I have a "specified mass fraction" boundary condition.



My udf:



#include "udf.h"
#include "sg.h"
#include "mem.h"
#include "stdio.h"
#include "stdlib.h"


DEFINE_GRID_MOTION(interface, domain, dt, time, dtime)
{

Thread *tf = DT_THREAD(dt);
face_t f;
cell_t c;
Node *v;
int n, temp;

int i;

real x[ND_ND];
real T = 1000.;

SET_DEFORMING_THREAD_FLAG(THREAD_T0(tf));

begin_f_loop (f, tf)
{
F_CENTROID(x,f,tf);

T = F_T(f,tf);

f_node_loop (f, tf, n)
{
v=F_NODE (f, tf, n);

if (NODE_POS_NEED_UPDATE (v))
{
real TT = T + T * F_YI(f,tf,0);

dispz=(Tm-TT)/1000.0;
NODE_X(v)+=dispz;
NODE_POS_UPDATED (v);
}
}

}
end_f_loop (f, tf)
}


If I use this udf without F_YI(f,t,0) then it works fine.



Thank you und best regards
Ana

[AlexanderZ]

Did you switched on species model?

by the way you had several errors in your code, so I don't know, how it could work

#include "udf.h"
#include "sg.h"
#include "mem.h"
#include "stdio.h"
#include "stdlib.h"


DEFINE_GRID_MOTION(interface, domain, dt, time, dtime)
{

Thread *tf = DT_THREAD(dt);
face_t f;
cell_t c;
Node *v;
int n, temp;

int i;

real x[ND_ND];
real T = 1000.;
real TT = 0.0;
real dispz = 0.0;
real Tm;

SET_DEFORMING_THREAD_FLAG(THREAD_T0(tf));

begin_f_loop (f, tf)
{
F_CENTROID(x,f,tf);

T = F_T(f,tf); 

f_node_loop (f, tf, n)
{
v=F_NODE (f, tf, n);

if (NODE_POS_NEED_UPDATE (v))
{
TT = T + T * F_YI(f,tf,0);

dispz=(Tm-TT)/1000.0;
NODE_X(v)+=dispz;
NODE_POS_UPDATED (v);
}
} 

}
end_f_loop (f, tf)
}

real Tm; is not defined in your code. So dispz=(Tm-TT)/1000.0; can be calculated

best regards

[ana86]

Yes, I switched on the species model, otherwise I couldn't give a mass fraction boundary condition...

The mistakes in the code (e.g. Tm) occurs, because I had to change the code a bit for the forum. I'm not allowed to show the whole code, so it seems that I didn't check the code correctly. But it works fine without F_YI.

I know that I have to set "solve/set/expert" to yes, if I want to use the temperature gradient, maybe there is something equivalent for species transfer?


best regards

[AlexanderZ]

what kind of error do you have?

best regards

[ana86]

The error is:

Received signal SIGSEGV


After that, fluent automatically closes.

[AlexanderZ]

you've mentioned about temperature gradient
but in your code there is no temperature gradient.

all seems to be correct in code above, check parts of code, which you didn't show

best regards

[ana86]

I run the simulation with this simple code, with defined Tm and without any temperature gradient. The simulation works, when I did not use the Macro F_YI (I just changed the line TT = T + T*F_YI(f,t,0) to TT = T + T), but when I use the macro fluent crashes. So it seems to be a problem of this macro... I also tried with C_YI(c,t,0), but the same problem occurs...