[zhixin]

Hi all,
I've been calculating the cell wall distance of the first layer. I used the C_WALL_DIST(c,t) macro, but the result seems to be untrue, so I used the following udf and got two different result.

t=Lookup_Thread(d,11);//*11 is the wall number
begin_f_loop(f,t)
{
F_CENTROID(x,f,t);
c0=F_C0(f,t);
t0=THREAD_T0(t);
C_CENTROID(y,c0,t0);
dist=ND_SUM(pow(x[0]-y[0],2),pow(x[1]-y[1],2),pow(x[2]-y[2],2));
dists=sqrt(dist);
C_UDMI(c0,t0,1)=dists;//*dists stands for the distance between cell centriod of first layer and the face centroid of boundary face
dt=C_WALL_DIST(c0,t0);//*to compare the result between macro and the expression
C_UDMI(c0,t0,2)=dt;
}
end_f_loop(f,t)

So can anyone tell me whether the udf is right or not? Or how does the C_WALL_DIST(c,t) macro get the result?

Best, Gao

[zhixin]

I have another question:
I use ICEM to draw the hexahedral structure mesh and set the height of first layer as 0.002mm. So can I say the distance between the first layer and boundary is 0.001mm?

[pratikddhoot]

Hi,

I used the same macro, but my solution fails to initialize. How did you run the solution?

did the macro C_WALL_DIST work without initialization?

Thanks,
Pratik

[zhixin]

Hi Pratik,

When I used the C_WALL_DIST(c,t) macro, I based on this,
http://www.cfd-online.com/Forums/flu...-centroid.html
and yes I initialize before calculation, I think it's needed.

Can you post your error? Maybe me or others can find out the reason.

Best.

[pratikddhoot]

Originally Posted by zhixin

Hi Pratik,

When I used the C_WALL_DIST(c,t) macro, I based on this,
http://www.cfd-online.com/Forums/flu...-centroid.html
and yes I initialize before calculation, I think it's needed.

Can you post your error? Maybe me or others can find out the reason.

Best.

It says, "received a fatal signal (segmentation fault)" when I hit initialize.

[zhixin]

Have you defined the "user defined-memory?" if you used the memory to store message.

[pratikddhoot]

Originally Posted by zhixin

Have you defined the "user defined-memory?" if you used the memory to store message.

No. I have not. I am new to UDF and I could interpret and run a few other codes.
Only when I use real "y = C_WALL_DIST(c,t);" is when the solution fails to initialize.

[zhixin]

I don't know why this error happened, y=C_WALL_DIST(c,t) seems right, so would you mind post the key part about this macro.

[pratikddhoot]

Originally Posted by zhixin

I don't know why this error happened, y=C_WALL_DIST(c,t) seems right, so would you mind post the key part about this macro.

#include "udf.h"


DEFINE_PROFILE(wall_d_bc,t,i)
{
Thread *t0;
face_t f;
cell_t c, c0;
double F_x,area,A[ND_ND],yplus,wshear;

begin_f_loop(f,t)
{
c0 = F_C0(f,t);
t0 = THREAD_T0(t);

yplus = C_STORAGE_R(f,t,SV_WALL_YPLUS_UTAU); /* Y+*/

F_x = F_STORAGE_R_N3V(f,t,SV_WALL_SHEAR)[0];
F_AREA(A, f, t);
area = NV_MAG(A);
wshear=-1*F_x/area;

/* wshear = C_MU_L(c0,t0)*C_U(c0,t0)/C_WALL_DIST(c0,t0); wall shear stress */
F_PROFILE(f,t,i) = 6.*wshear/(0.075*C_MU_L(c0,t0)*SQR(yplus));
}
end_f_loop(f,t)
}




Look at this function for velocity profile. Save it as .c and interpret it in FLUENT.

Tell me if you can initialize.

If you can directly reply in email, my id is dhoot.p@husky.neu.edu

[zhixin]

I can initialize it.

[pratikddhoot]

Originally Posted by zhixin

I can initialize it.

Had you set inlet velocity to wall_d_bc for whatever 2D case you're running in FLUENT?

[zhixin]

Sorry I also can't initialize. I don't know why.
I suggest you ask others in new thread.