[LEE CHENG]

Hello, everyone.
The UDF is used to calculate the slip velocity. In order to make the case easier to converge, the under relaxation factor is added to the UDF code.
The code before adding the under relaxation factor is:
slip = Coeff1[1] * Ls*slip;
The code after adding the relaxation factor is:
slip = (1-underrelaxfactor) *u[1] + underrelaxfactor * Coeff1[1] * Ls*slip;
underrelaxfactor=0.01.
Before I changed the code, changing the value of Ls gave me a different result, but after I changed the code, changing the value of Ls seemed to have no effect on the result of the calculation.
Do I need to add any other Settings in FLUENT or code?I hope the experts can give me some advice. Thank you.

[LEE CHENG]

Originally Posted by LEE CHENG

Hello, everyone.
The UDF is used to calculate the slip velocity. In order to make the case easier to converge, the under relaxation factor is added to the UDF code.
The code before adding the under relaxation factor is:
slip = Coeff1[1] * Ls*slip;
The code after adding the relaxation factor is:
slip = (1-underrelaxfactor) *u[1] + underrelaxfactor * Coeff1[1] * Ls*slip;
underrelaxfactor=0.01.
Before I changed the code, changing the value of Ls gave me a different result, but after I changed the code, changing the value of Ls seemed to have no effect on the result of the calculation.
Do I need to add any other Settings in FLUENT or code?I hope the experts can give me some advice. Thank you.

Thank you in advance for your help

[pakk]

I can not really follow what you are doing, or trying to ask. You also only put one line of code, where is the rest of the UDF? How do you attach it? Why don't you just use the built-in slip? Too many questions.

[LEE CHENG]

I want to calculate the wall slip velocity according to the wall velocity gradient and return it to FLUENT.But if I change the value of Ls it doesn't seem to change.Here's my code:
#include "udf.h"
#include "mem.h"
#include "sg.h"
#include "math.h"
void coord_coeff(Thread *f_thread)
{
face_t face;
cell_t cell;
Thread *c_thread;
real y[ND_ND];
real u[ND_ND], a;
real B[ND_ND], b;
real A[ND_ND];
real dr0[ND_ND], es[ND_ND], ds, A_by_es;
begin_f_loop(face, f_thread)
{
F_CENTROID(y, face, f_thread);
cell = F_C0(face, f_thread);
c_thread = THREAD_T0(f_thread);

BOUNDARY_FACE_GEOMETRY(face, f_thread, A, ds, es, A_by_es, dr0);
ND_SET(u[0], u[1], u[2], C_U(cell, c_thread), C_V(cell, c_thread), C_W(cell, c_thread));
a = NV_MAG(u);
NV_CROSS(B, u, A);
b = NV_MAG(B);
ND_SET(F_UDMI(face, f_thread, 0), F_UDMI(face, f_thread, 1), F_UDMI(face, f_thread, 2), NVD_DOT(u, 1, 0, 0) / a, NVD_DOT(u, 0, 1, 0) / a, NVD_DOT(u, 0, 0, 1) / a);
ND_SET(F_UDMI(face, f_thread, 3), F_UDMI(face, f_thread, 4), F_UDMI(face, f_thread, 5), NVD_DOT(A, 1, 0, 0) / A_by_es, NVD_DOT(A, 0, 1, 0) / A_by_es, NVD_DOT(A, 0, 0, 1) / A_by_es);
ND_SET(F_UDMI(face, f_thread, 6), F_UDMI(face, f_thread, 7), F_UDMI(face, f_thread, 8), NVD_DOT(B, 1, 0, 0) / b, NVD_DOT(B, 0, 1, 0) / b, NVD_DOT(B, 0, 0, 1) / b);
}
end_f_loop(face, f_thread)
}
DEFINE_PROFILE(slip_velocity_x, f_thread, index)
{
face_t face;
cell_t cell;
Thread *c_thread;
real normal_slip, tangential_slip,k_slip;
real slip;
real Ls=0.00005;
real underrelaxfactor=0.01;
real Coeff1[ND_ND], Coeff2[ND_ND], Coeff3[ND_ND];
real u[ND_ND];

coord_coeff(f_thread);

if (Data_Valid_P())
{
begin_f_loop(face, f_thread)
{
cell = F_C0(face, f_thread);
c_thread = THREAD_T0(f_thread);
ND_SET(u[0], u[1], u[2], F_U(face, c_thread), F_V(face, c_thread), F_W(face, c_thread));
ND_SET(Coeff1[0], Coeff1[1], Coeff1[2], F_UDMI(face, f_thread, 0), F_UDMI(face, f_thread, 1), F_UDMI(face, f_thread, 2));
ND_SET(Coeff2[0], Coeff2[1], Coeff2[2], F_UDMI(face, f_thread, 3), F_UDMI(face, f_thread, 4), F_UDMI(face, f_thread, 5));
ND_SET(Coeff3[0], Coeff3[1], Coeff3[2], F_UDMI(face, f_thread, 6), F_UDMI(face, f_thread, 7), F_UDMI(face, f_thread, 8));
normal_slip = -1*NVD_DOT(Coeff1, NV_DOT(Coeff2, C_U_G(cell, c_thread)), NV_DOT(Coeff2, C_V_G(cell, c_thread)), NV_DOT(Coeff2, C_W_G(cell, c_thread)));
tangential_slip= NVD_DOT(Coeff1, NV_DOT(Coeff1, C_U_G(cell, c_thread)), NV_DOT(Coeff1, C_V_G(cell, c_thread)), NV_DOT(Coeff1, C_W_G(cell, c_thread)));
k_slip= NVD_DOT(Coeff1, NV_DOT(Coeff3, C_U_G(cell, c_thread)), NV_DOT(Coeff3, C_V_G(cell, c_thread)), NV_DOT(Coeff3, C_W_G(cell, c_thread)));
slip = Ls*normal_slip;
slip = (1-underrelaxfactor) *u[0] + underrelaxfactor * Coeff1[0] * slip;

F_PROFILE(face, f_thread, index) = slip;
}
end_f_loop(face, f_thread)
}
Before I add this line of code(slip = (1-underrelaxfactor) *u[0] + underrelaxfactor * Coeff1[0] * slip, it's going to change with Ls.
Please take a look. Is there any problem in my code that I have ignored?Thank you very much!

[pakk]

In your initial question, you wrote

slip = (1-underrelaxfactor) *u[1] + underrelaxfactor * Coeff1[1] * Ls*slip;

As far as I see, this should just work.
But in the code, you wrote:

slip = (1-underrelaxfactor) *u[0] + underrelaxfactor * Coeff1[0] * slip;

This is not the same. There is no "Ls" anymore in this line, so that is why your results don't depend on it anymore.

[pakk]

Still, I hope you are aware that Fluent already has built-in slip. If you just want to add to slip to your model, you don't need UDFs, just select the option.

If you are already aware, I put it here for other people who just want to use slip: don't use any UDF, just select the low-pressure boundary slip!

[LEE CHENG]

Thank you.Pakk.
I want the wall slip velocity to vary with the velocity gradient.

[pakk]

Then I repeat again: Fluent has a built-in functionality for slip. Search for "low-pressure boundary slip". It might just do exactly what you want, no UDF required.

[zhangdongjie]

Originally Posted by pakk

Then I repeat again: Fluent has a built-in functionality for slip. Search for "low-pressure boundary slip". It might just do exactly what you want, no UDF required.

thanks for your replay。but Low pressure boundary slip only works in 3D. How can I open this button in 2D?

[zhangdongjie]

Originally Posted by LEE CHENG

Hello, everyone.
The UDF is used to calculate the slip velocity. In order to make the case easier to converge, the under relaxation factor is added to the UDF code.
The code before adding the under relaxation factor is:
slip = Coeff1[1] * Ls*slip;
The code after adding the relaxation factor is:
slip = (1-underrelaxfactor) *u[1] + underrelaxfactor * Coeff1[1] * Ls*slip;
underrelaxfactor=0.01.
Before I changed the code, changing the value of Ls gave me a different result, but after I changed the code, changing the value of Ls seemed to have no effect on the result of the calculation.
Do I need to add any other Settings in FLUENT or code?I hope the experts can give me some advice. Thank you.

I'm simulating a nozzle with rarefied gas. Its kn number is very large. First order slip and temperature jump must be added. I use Maxwell slip equation to write UDF, but I don't know where UDF should be loaded. I look up the data. The sliding speed can be loaded into the shear stress, so where is the temperature jump loaded? Is it loaded directly into the wall temperature?

[LEE CHENG]

I think our UDFs should be similar, and maybe we can learn from each other and share our UDF.

[zhangdongjie]

Originally Posted by LEE CHENG

I think our UDFs should be similar, and maybe we can learn from each other and share our UDF.

haode,xiongdi,wodeQQ1248385182

[zhangdongjie]

Originally Posted by LEE CHENG

I think our UDFs should be similar, and maybe we can learn from each other and share our UDF.

Hello, can you talk to each other? 1248385182@qq.com