[Muad-iiib]

Hello everyone,
I am currently trying to simulate blood flow through a tubular sudden expansion channel. For this a two-fluid model is applied, where the plasma is treated as a Newtonian fluid and the RBCs are treated as shear-thinning fluid. In this model, RBCs have a non-constant viscosity which needs to be implemented through a UDF.

Here is the C++ code I am using :

#include "udf.h"

real mu_inf;
real mu_0;
real mu;
real L[3][3];
real gamma;

DEFINE_PROPERTY(viscosity, c, t)
{
real phi = C_VOF(c, t);
mu_inf = 27.87 * phi * phi * phi - 21.22 * phi * phi + 14.44 * phi;
mu_0 = 537 * phi * phi * phi + 55.01 * phi * phi - 0.13 * phi;

L[0][0] = C_DUDX(c, t);
L[0][1] = C_DUDY(c, t);
L[0][2] = C_DUDZ(c, t);
L[1][0] = C_DVDX(c, t);
L[1][1] = C_DVDY(c, t);
L[1][2] = C_DVDZ(c, t);
L[2][0] = C_DWDX(c, t);
L[2][1] = C_DWDY(c, t);
L[2][2] = C_DWDZ(c, t);

gamma = sqrt(0.5*(4*L[0][0]*L[0][0]+4*L[1][1]*L[1][1]+4*L[2][2]*L[2][2]+2*(L[0][1]+L[1][0])*(L[0][1]+L[1][0])+2*(L[0][2]+L[2][0])*(L[0][2]+L[2][0])+2*(L[1][2]+L[2][1])*(L[1][2]+L[2][1])));
mu = mu_inf + (mu_0 - mu_inf) * (1 + log(1 + 11 * gamma)) / (1 + 11 * gamma);
return mu;
}

Gamma stands for the shear rate, mu the viscosity, phi the volume fraction and L the velocity gradient matrix.

The code doesn't send any error message when compiled on fluent however when initialising the simulation and after every iteration, I get several lines of this message : "Chip : invalid opcode 0 : pc = 8". The simulation doesn't stop though and continues until the end.

Could anyone explain to me what this mean and how it impacts the result of the simulation ?

Thanks for your help !