[runyu]

Dear all,
I want to define my shear stressx=a*d2u/dx2,and this is my code

# include "udf.h"
# include "mem.h"
# include "math.h"
# define domain_ID 1

DEFINE_ADJUST(adjust_gradient, domain)
{
  Thread *t;
  cell_t c;
  face_t f;

  domain = Get_Domain(domain_ID);
  /* Fill UDS with the variable. */
  thread_loop_c(t,domain)
       begin_c_loop (c,t)
         {
		   C_UDSI(c,t,0) = C_DUDX(c,t);
         }
       end_c_loop (c,t)
    }

DEFINE_ON_DEMAND(store_gradient)
{
  Domain *domain;
  face_t f;
  cell_t c;
  Thread *t;

  domain=Get_Domain(1);

  /* Fill the UDM with magnitude of gradient. */
  thread_loop_c(t,domain)
       begin_c_loop (c,t)
         {
           C_UDMI(c,t,0) = NV_MAG(C_UDSI_G(c,t,0));
         }
       end_c_loop (c,t)
 }

DEFINE_PROFILE(shear_stressx, t,i)
{
    face_t f;
    cell_t c;
    real xw[ND_ND],xc[ND_ND],taux;
    real a = 0.00012;
    Thread *t0 =THREAD_T0(t);

    begin_f_loop(f,t)
        {
         cell_t c0 = F_C0(f,t);
	 taux = -a*C_UDMI(c0,t0,0);
	 F_PROFILE(f,t,i) =taux;
        }
    end_f_loop(f,t)
}

after hook the ajust,change my shear stress I only get my taux=0,and I found my UDS=0 also.
can anybody help me?

[zobekenobe]

Hi, you can try the code posted below.


I have defined and calculated the UDMI in the define_adjust macro.


In the code you posted domain was already passed by fluent in define_adjust so you were overriding it Get_Domain(Domain_ID)


also DEFINE_ON_DEMAND doesnt calculate before or after every iteration but only when activated that's one reason why the UDMI was empty and hence giving zeros.


Anyway try this and see if there is any difference

DEFINE_ADJUST(adjust_scalar, domain_pointer)
{
    Thread* thread_pointer;
    cell_t cellt;

    thread_loop_c(thread_pointer, domain_pointer)
    {
        begin_c_loop(cellt, thread_pointer)
        {
            C_UDSI(cellt, thread_pointer, 0) = C_DUDX(cellt, thread_pointer);
            C_UDMI(cellt, thread_pointer, 0) = NV_MAG(C_UDSI_G(cellt, thread_pointer, 0));
        }
        end_c_loop(cellt, thread_pointer)
    }
}

DEFINE_PROFILE(shear_stress, thread_pointer, position_index)
{
    face_t facet;
    cell_t c0;

    real a = 0.00012;

    Thread *t0 = thread_pointer->t0;

    begin_f_loop(facet, thread_pointer)
    {
        c0 = F_C0(facet,thread_pointer);
        F_PROFILE(facet, thread_pointer, position_index) = -a * C_UDMI(c0, t0, position_index);
    }
    end_f_loop(facet, thread_pointer)
}

[runyu]

Thank you for your help.And I have tried your code,but I got the segmentation fault when I calculated it.

[zobekenobe]

Did you allocate memory for the UDMI??

[runyu]

Originally Posted by zobekenobe

Did you allocate memory for the UDMI??

I defined UDM=1,UDS=1

[zobekenobe]

hey runyu,


I ran the code for a rough geometry I have and it didn't give a segmentation error.


Did you allocate memory as


Define -> user-defined ->Memory-> Number of user-defined-memory locations = 1

??
Usually a segmentation error occurs when something goes wrong with the memory allocated

[runyu]

Originally Posted by zobekenobe

hey runyu,


I ran the code for a rough geometry I have and it didn't give a segmentation error.


Did you allocate memory as


Define -> user-defined ->Memory-> Number of user-defined-memory locations = 1

??
Usually a segmentation error occurs when something goes wrong with the memory allocated

hi,zobekennobe,
Thank you for your careful help,I tried it again,and I found it can be calculated,maybe I did something wrong yesterday.But my shear tress still=0.Could it be something wrong with my Settings?

[runyu]

Dear friend,
I'm extremely grateful to you,and my udf is working now.

[letsroll]

Dear Runyu
May you share the modified code here?
Thank you for your kindness.