CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Fluent UDF and Scheme Programming (http://www.cfd-online.com/Forums/fluent-udf/)
-   -   udf for wall slip (http://www.cfd-online.com/Forums/fluent-udf/111520-udf-wall-slip.html)

ammm January 9, 2013 06:39

udf for wall slip
 
i need a udf for slip wall
i want to use udf for setting this boundary condition:
us=B*(du/dy)
what should i do?

msaeedsadeghi January 10, 2013 02:49

I have already written it. Email me please.

ammm January 10, 2013 06:02

i emailed you thank you in advance

fahmadi January 12, 2013 13:58

Quote:

Originally Posted by msaeedsadeghi (Post 401099)
I have already written it. Email me please.

Dear Mr Sadeghi
I also have written a UDF for slip velocity which works correctly, but when I want to add the term due to thermal creep, I get the below error:
"FLUENT received fatal signal (ACCESS_VIOLATION)."

I recognized that this error is due to using the expression "C_T_G(cell,t)" in my UDF.

Could you help me to solve this problem?

Thank you so much!

msaeedsadeghi January 13, 2013 00:26

Quote:

Originally Posted by fahmadi (Post 401513)
Dear Mr Sadeghi
I also have written a UDF for slip velocity which works correctly, but when I want to add the term due to thermal creep, I get the below error:
"FLUENT received fatal signal (ACCESS_VIOLATION)."

I recognized that this error is due to using the expression "C_T_G(cell,t)" in my UDF.

Could you help me to solve this problem?

Thank you so much!

In most Micro-Channel problems there is no need to add the thermal creep section. You can neglect it.

ammm January 15, 2013 10:19

i write this udf for slip. is it correct???

#include "udf.h"
DEFINE_PROFILE(shear_stressx, thread, n)
{
face_t f ;
cell_t c0 ;
Thread *t0 =THREAD_T0(thread) ;
float u0,uw,vderivate, sstressx ;
begin_f_loop(f, thread)
{
c0 = F_C0(f ,thread) ;
vderivate= C_DUDY(c0,t0) ;
u0 = C_U(c0, t0);
uw=u0+1*pow(10,-6)*vderivate ;
F_U(f, thread)=uw;
sstressx = -.001003*vderivate ;
F_PROFILE(f,thread,n) = sstressx ;
}
end_f_loop(f,thread)

asm January 16, 2013 05:49

How to write a slip condition in wall
 
dear Mr Sadeghi and ammm:

i need a udf(fluent or cfx is ok) for slip wall.As ammm metioned,the slip wall

which can be attached with thermal creep is the best.But if the thermal creep is

excluded,that is ok. could you send me a udf you have writtern.please send

the email:asm23374634@gmail.com

thank you very much

ammm January 17, 2013 06:03

my udf works good with low slip length but for larger slip length not converged. Can anyone guide me for this convergence problem?
i use fluent 6.2 and when i dont use udf it works corrctly
please help me:(

Shayan February 22, 2013 17:10

Quote:

Originally Posted by fahmadi (Post 401513)
Dear Mr Sadeghi
I also have written a UDF for slip velocity which works correctly, but when I want to add the term due to thermal creep, I get the below error:
"FLUENT received fatal signal (ACCESS_VIOLATION)."

I recognized that this error is due to using the expression "C_T_G(cell,t)" in my UDF.

Could you help me to solve this problem?

Thank you so much!

I had same problem, to avoid this problem,you should not hook all your UDFs at the same time; Step by step! first apply the udf of slip velocity, then let the iterations go, after your solution is converged`,unhook it and apply Thermal creep udf, let it go for numerous iterations, then re-apply slip-velocity udf again,
All of these operations is due to suitable initializations which help Fluent to start a solution from a rather logic guess.
each UDF produces a sudden jump in the resodual graph, so it's better avoiding 3 jumps at the same time, instead, imply them gradually.

flotus1 February 25, 2013 05:32

Quote:

Originally Posted by ammm (Post 402407)
my udf works good with low slip length but for larger slip length not converged. Can anyone guide me for this convergence problem?
i use fluent 6.2 and when i dont use udf it works corrctly
please help me:(


Apply some under-relaxation.

Additionally, I can only emphasize once more that fluent comes with a BC of the type us=B*(du/dy).
It is called "Slip Boundary Formulation for Low-Pressure Gas System" and even includes the term for thermal creep.

msaeedsadeghi February 26, 2013 00:44

I have done that before without thermal creep, but today have found how to consider thermal creep also.
It needs step by step enabling and disabling equations.

Katerina March 4, 2013 06:34

Dear all,
I'm doing a simulation with slip and VOF. I would like to define slip - shear stress for liquid only. But I'm doing it wrong.
My shear stress=mu*dudx*vof
What should I change to get C_VOF for correct thread ?

#include "udf.h"

DEFINE_PROFILE(wall_shear_y,thread,position)
{
face_t f;
Thread *t0, *t2, *t3;
cell_t c0;
double TAUxy, mu;
double dudx;
double vof;

begin_f_loop(f,thread)
{
t0=THREAD_T0(thread); /*adjacent cell index*/
c0=F_C0(f,thread); /*adjacent cell thread*/

t2 = THREAD_SUB_THREAD(t0, 0); // take ID of phase no.1
t3 = THREAD_SUB_THREAD(t0, 1); // take ID of phase no.2

dudx=C_DUDX(c0,t0);
mu=C_MU_L(c0,t0); /*viscosity*/

begin_c_loop(c0,t3)
{
vof=C_VOF(c0,t3); /*volume of fluid value in adjacent cell for phase 2*/
}
end_c_loop(c0,t3)

TAUxy=mu*dudx*vof;
F_PROFILE(f,thread,position)=TAUxy;
}
end_f_loop(f,thread)
}

Regards,
Katerina

elaf_2003 March 26, 2014 05:41

slip wall shear
 
I was wondering if any one can help me?
I am using flow around rotating cylinder, the shear stresses at the wall need to be specified to constant value. In this option, there are 3 components for the shear stresses, i.e. x direction components and so on.

Does anyone know for example how can set these values for cylinder surface
Regard.

m zahid November 13, 2014 05:25

hi
 
hi, i need a UDF for slip boundary condition at the bottom of the domain or at the ground (wall), 3D case is just like a flow over a building. if anybody have sample UDF please share this, z is a vertical axis of my domain. thanks

mziqureshi@hotmail.com

regards,


All times are GMT -4. The time now is 11:08.