[shashankmechguy]

Hi All,

I am using udf Execute_At_End to print in console time dependent value of torque.
The problem I have is , the example given in fluent udf manual is for entire domain. I only want to calculate torque for one boundary zone which is a face thread of blades of a rotor.

This is the code I am using

#include"udf.h"
DEFINE_EXECUTE_AT_END(torque_time)
{
Domain *d;
Thread *t;
face_t f;
real NV_VEC(A);
real NV_VEC(force);
real NV_VEC(torque);
real x[ND_ND];
real NV_VEC(radius);
real NV_VEC(torque_final);
real tor;
NV_S(force,= ,0.0);
NV_S(torque_final,= ,0.0);
NV_S(torque,= ,0.0);
d = Get_Domain(9);
begin_f_loop(f,t)
{
F_AREA(A,f,t);
F_CENTROID(x,f,t);
NV_S(force,= , F_AREA(A,f,t), * , F_P(f,t));
NV_D(radius,= ,x[0],x[1],x[2]);
NV_CROSS(torque,radius,force);
NV_VV(torque_final,= ,torque_final,+, torque);
tor= NV_MAG(torque_final);
}
end_f_loop(f,t)
printf("Value of torque at the blades is = %f", tor );
}


I used d= Get_Domain(9) because the integer id of that face thread is 9. But its not printing anything on the console and I know I cannot use Get_Domain(9).

Can anyone advise me how to specify one particular face thread for execute_at_end macro

[GM_XIII]

you need to do the following:

d=Get_Domain(1);
t = Lookup_Thread(d, Zone_ID); /*Here is where you give the id read in BC as zone_id*/

[shashankmechguy]

Thanks David,

but still I am not able to print anything in console window.
I dont know why. There is no error while compiling or interpreting this udf.

[GM_XIII]

Your case is multiphase? Have you tried changing the ZOne_id? Post here the new UDF code

[shashankmechguy]

Originally Posted by GM_XIII

Your case is multiphase? Have you tried changing the ZOne_id? Post here the new UDF code

It started working once I changed the printf to message but now I am getting
received a fatal signal : Segmentation violation.

[shashankmechguy]

Here is the UDF Code:

#include"udf.h"
DEFINE_EXECUTE_AT_END(torque_time)
{
Domain *d;
Thread *t;
face_t f;
real NV_VEC(A);
real NV_VEC(force);
real NV_VEC(torque);
real x[ND_ND];
real NV_VEC(radius);
real NV_VEC(torque_final);
real tor;
NV_S(force,= ,0.0);
NV_S(torque_final,= ,0.0);
NV_S(torque,= ,0.0);
d = Get_Domain(1);
t = Lookup_Thread(d, 12);
begin_f_loop(f,t)
{
F_AREA(A,f,t);
F_CENTROID(x,f,t);
NV_S(force,= , F_AREA(A,f,t), * , F_P(f,t));
NV_D(radius,= ,x[0],x[1],x[2]);
NV_CROSS(torque,radius,force);
NV_VV(torque_final,= ,torque_final,+, torque);
tor= NV_MAG(torque_final);
}
end_f_loop(f,t)
Message("Value of torque at the blades is = %f", tor );
}


I am running Fluent on Linux.

[blackmask]

can your code be compiled at all?
IMO,

NV_S(force,= , F_AREA(A,f,t), * , F_P(f,t));

should be replaced by

 F_AREA(A,f,t);
NV_VS(force,= , A, * , F_P(f,t));

[shashankmechguy]

Originally Posted by blackmask

can your code be compiled at all?
IMO,

NV_S(force,= , F_AREA(A,f,t), * , F_P(f,t));

should be replaced by

 F_AREA(A,f,t);
NV_VS(force,= , A, * , F_P(f,t));

Hi BlackMask,

Thanks for the input. I made that correction.
And now I am getting at least print message with value always =0.000.
I thought I am doing something wrong with cross product so I tried printing force value, still 0.000. Then for the sake of checking, I tried printing sum of area values, that is also 0.000.
I dont know what is wrong in the logic of udf?

[blackmask]

Make sure that your thread id is correct. What is the result of

t = Lookup_Thread(d, 12); 
printf("%d\n", THREAD_ID(t));