[ljpGBL]

I was simulating the separate flow over the sandwave using SST k-w modeling.
As you can see, I wrote an udf in order to calculate the shear-stress and shear-velocity inside the fluid, the compiling process and loading were ok, calculating is also fine, the problem occurs when I display the User Memory 0 and 1. They are separately shear-stress and shear-velocity, but the contour of the shear-velocity apparently has some problem.
I had tried change the options in view menu but it doesn't work.
78c6595105cc19b81c12cd2a2437e6c.jpg
Clip_20181219_225447.png
The code I wrote is as follows. If you had met a similar problem as I do please let me know how you solve it, or any suggestion is also welcome! Thank you!

#include "udf.h"
#include "mem.h"
#include "sg_udms.h"

#define rho 998.2
#define g 9.81
#define wl 0.046

/* #define NUM_UDM 2 */

// static int offset = UDM_UNRESERVED;

DEFINE_PROFILE(vof_profile, t, i)
{
#if !RP_HOST

real x[ND_ND];
face_t f;

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

if (x[1] <= wl)
{
F_PROFILE(f, t, i) = 1;
}
else
{
F_PROFILE(f, t, i) = 0;
}
}
end_f_loop(f, t)

#endif
}

DEFINE_PROFILE(outlet_pressure, t, i)
{
#if !RP_HOST

real x[ND_ND];
face_t f;
begin_f_loop(f, t)
{
F_CENTROID(x, f, t);
if (x[1] >= wl)
{
F_PROFILE(f, t, i) = 0.0;
}
else
{
F_PROFILE(f, t, i) = rho*g*(wl-x[1]);
}
end_f_loop(f, t)
}

#endif
}

DEFINE_PROFILE(inlet_velocity, t, i)
{
#if !RP_HOST

real x[ND_ND];
face_t f;
begin_f_loop(f, t)
{
F_CENTROID(x, f, t);
if (x[1] > wl)
{
F_PROFILE(f, t, i) = 0.0;
}
else
{
F_PROFILE(f, t, i) = 0.21;
}
end_f_loop(f, t)
}
#endif
}

/*DEFINE_EXECUTE_ON_LOADING(on_loading, libname)
{
offset = Reserve_User_Memory_Vars(NUM_UDM);
Set_User_Memory_Name(offset, "shear-stress");
Set_User_Memory_Name(offset+1, "shear-velocity");
} */

DEFINE_EXECUTE_AT_END(calculate_shear_scales)
{
#if !RP_HOST

Domain *d;
Thread *t;
cell_t c;

d = Get_Domain(3); /* The id of the water domain in multiphase case */

thread_loop_c(t, d)
{
begin_c_loop(c, t)
{
C_UDMI(c, t, 0) = C_MU_EFF(c, t)*(C_DUDY(c, t)+C_DVDX(c, t));
C_UDMI(c, t, 1) = sqrt(C_UDMI(c, t, 0)/C_R(c, t));
}
end_c_loop(c, t)
}
#endif
}

[AlexanderZ]

what kind of problem do you have?
what's wrong with velocities?

best regards

[ljpGBL]

Originally Posted by AlexanderZ

what kind of problem do you have?
what's wrong with velocities?

best regards

hi, good to see your asking.
I mean isn't there should be a full mapping with the contour display?
But my first picture shows that there is some "empty data" or display mistake happen, how could I fix this or it is the right displaying? But it makes no sense right?