[alexskerhut]

Hi,
I am interpreting a viscosity UDF which is dependent upon temperature and strain rate. The UDF interprets fine but does not lead to convergence and starts diverging from the very start and does not run even a single iteration. My UDF is as follows:

#include<udf.h>
DEFINE_PROPERTY(cell_viscosity,c,t)
{
double stress = 22.22e+06;
double A = 8.86e+06;
double energy = 145000;
double R = 8.314;
double n = 3.55;
double mu_laminar,temp,strainrate,Z,D,E,F;
temp= C_T(c,t);
strainrate= C_STRAIN_RATE_MAG(c,t);
Z= strainrate*exp(energy/(R*temp));
D= pow((Z/A),(1/n));
E= pow((Z/A),(2/n));
F= pow(E+1,0.5);
mu_laminar= (stress*log(D+F))/(3*strainrate);
return mu_laminar;
}

A similar viscosity model named carreau dependent upon strainrate and temperature converges fine but my udf does not even start iterating. It gives the error message

#Divergence detected in AM solver: x-momentum - Increasing relaxation sweeps!

Kindly help me in this my complete project depends upon the successful implementation of the viscosity UDF.

[pakk]

Print the values to your screen.

Message("Z,D,E,F: %f,%f,%f,%f\n",Z,D,E,F);
Message("Mu: %f\n",mu_laminar);
return mu_laminar;

(If I remember the syntax correctly, otherwise check the manual.)

Then check if the values that are shown agree to what you want. If they don't, find out why not and fix your equations.