[radin_85]

Hi everyone,
I have a weird problem is writing the results from DEFINE_EXECUTE_AT_END to a file.
As you can see below, I have written a UDF to calculate the kinetic energy during my simulation and write it in a file at each time step:

DEFINE_EXECUTE_AT_END(EK)
{
Domain *d;
real Ek=0.0;


Thread *t;
cell_t c;
d = Get_Domain(3); /* returns phase with ID=3 domain pointer*/


thread_loop_c(t,d)
{
begin_c_loop(c,t)
{
Ek += 0.5 * C_R(c,t) * C_VOLUME(c,t) * (pow(C_U(c,t),2) + pow(C_V(c,t),2) + pow(C_W(c,t),2) );
}
end_c_loop(c,t)
}


{
FILE *fp;
fp = fopen("Ek.dat","a");
fprintf(fp,"%f %f\n",CURRENT_TIME, Ek);
fclose(fp);
}
}

The code works file however in my Ek.dat file for each time step I get to rows one with 0 values and the other one is the actual results:

0.000000 0.000000
0.000000 0.000000
0.000100 0.000000
0.000100 0.000005
0.000220 0.000000
0.000220 0.000022
0.000364 0.000000
0.000364 0.000060
0.000537 0.000000
0.000537 0.000132
0.000744 0.000000
0.000744 0.000255
0.000993 0.000000
0.000993 0.000457
0.001292 0.000000
0.001292 0.000777
0.001650 0.000000
0.001650 0.001273

Does anyone have any suggestions to get rid of those zero values at each time step??
Many thanks

[LuckyTran]

By chance are you running in parallel on 2 cores?

[radin_85]

That was my first guess. but no this is not a parallel simulation.

[LuckyTran]

Then... are you running a dynamic mesh?

[AlexanderZ]

what fluent version are you using?
starting from v20 fluent starts in parallel by default
(even it says 1 core, it uses 1 computational node and host node)

[radin_85]

I see. Well, I am using 2021. So what can I do to avoid this problem?