[whynam]

Hello guys, I'm new with UDF and I'm trying to implement a new viscosity model, but I've got some issues, if someone can help me, I'll be really thankful:

#include "udf.h"
DEFINE_PROPERTY(generalized_power_law,c,t)
{
real mu;
real strain_rate;

real k;
real n;

real eta_inf=0.035;
real dEta=0.25;
real a=50;
real b=3;

real n_inf=1;
real dn=0.45;
real c=50;
real d=4;

strain_rate=C_STRAIN_RATE_MAG(c,t);

k=eta_inf+dEta*exp(-(1+(strain_rate/a))*exp(-b/strain_rate));
n=n_inf-dn*exp(-(1+(strain_rate/c))*exp(-d/strain_rate));

mu=0.1*k*pow(strain_rate,n-1);
return mu;
}

I've got these line of error :

c definition shadows previous definition
Error: D:/***/udf/viscosity/viscosity_GPL.c: line 21: non-integer subscript expression: float.

I have no idea what to do..
Thank you very much.

[annan]

Hello Whynam,
I think your problem might come from these two lines :

real c=50;

strain_rate=C_STRAIN_RATE_MAG(c,t);

For C_STRAIN_RATE_MAG, what you need is a cell and a cell thread => C_STRAIN_RATE_MAG(cell,cell_thread). The problem is that you named the constant and you cell "c", so Fluent cannot find the strain rate value.

What I'd recommand is to name your constant with another name rather than c (real constant = 50; ), and don't forget to change its name in the expression of n otherwise you'll get an error too.

n=n_inf-dn*exp(-(1+(strain_rate/c))*exp(-d/strain_rate));

Hope this will help.
Good luck

[whynam]

Hi,

Thank you very much, it was helpful