Species model

you should use Define properties function in UDF. it is a simple code

Hello Shayan,

As Mojtaba suggested, the DEFINE_PROPERTY function should work for your case. An example on how to use this function is provided in the ANSYS Fluent UDF manual guide in the section "Temperature-dependent Viscosity Property" (available for download at http://users.ugent.be/~mvbelleg/fludf-12-0.pdf). You can use the C_CENTROID macro to obtain the centroid coordinates of each control volume inside your flow domain and calculate the viscosity with the 2nd element of the space vector (y-direction). It should look something like this:

#include "udf.h"

DEFINE_PROPERTY(fluid_viscosity,c,t)

{

real A = 1.0; /*proportional constant*/

real visc; /*viscosity*/

real x[ND_ND]; /*centroid coordinates*/

C_CENTROID(x,c,t); /*get the centroid coordinates*/

visc = A*x[1]; /*here is the viscosity definition as a function of the y-direction*/

return visc;

}

Hope this helps. Good luck!

Best regards,

Stephan

Thank you all for your comments. 

Dear Stephan, 

You can find below my UDF; actually the problem arises at the interface--pls look at my comment at the end of  the code...

#include "udf.h"

DEFINE_PROPERTY(cell_viscosity, c, t)

{

real mu;

real x[ND_ND];

real Y_coord;

C_CENTROID(x,c,t);

Y_coord=x[1];

if(Y_coord

The code looks good so far. The issue with the contour plot could be a matter of node-values vs. cell-values. Did you show contour plots with the node value option, or did you plot the cell values? As far as I know, the latter uses the centroid values of each control volume, which are obtained by interpolation from the values of all nodes belonging to the control volume. That might explain the deviation 5.7 vs. 6 Pa*s.

Furthermore, you should consider that the else-if case will only be used if the Y-coordinate of the centroid is exactly(!) 7e-6, any diaviation from this will result in a viscosity of 0.001.

Last but not least, I'm not sure about your profile function: it doesn't seem to be physically correct. From your profile, the viscosity increases with increasing distance for y-coordinates smaller 7*e-6, but then there is a dramatic drop to water-like viscosity. If you use the following function the profile would be continuous (see also graph attached): mu=(-857000*(Y_coord-7e-6))+0.001. This makes much more sense from a physical point of view...

Hope this helps.

Best regards,

Stephan

Thanks Stephan. Very helpful...