Fluent udf give zero instead of coordinate location
 

I am working on DEFINE_TURBULENT_PROFILE. However, x[ND_ND] doesn't seems to work for me as it gives me zero. When I try monitoring its output (e.g. Message("x=%g and y=%g \n", x[0], x[1]);) on Fluent's console, I saw this instead - x=0 and y=0.

Basically it just keeps showing me zero. Even when I try "Message("x=%g and y=%g \n", x[0], 2);" but it shows 0 instead of 2! When I try outputting other output like S, the message doesn't show 0.

Is x[ND_ND] ONLY for the "DEFINE_PROFILE" udf? Does it works for others like mine case?

Typecast
 

That's because you are expecting the compiler to do the typecasting for you. 2 is an integer and you are trying to print it as float or double. Use 2. or 2.0. Furthermore, %g is not a good format specifier for such numbers, use %f or %lf.

Thanks for the reply. My issue is I am trying to define eddy viscosity as a function of position. I use xx=x[0] and y=x[0] but the position keep giving me 0. Because of this, my eddy viscosity is no longer a function of position as the position is constantly 0!

E.g. I try
if (y>0.1) {S=2}
else {S=1}.

My solver keeps giving y=0 and thus the above sample if-else loop is meaningless as S is constantly 2.

The Code
 

Most likely the reason is as I mentioned earlier, use of int values in place of real. If it is something other than that, I'd suggest you to copy whole code here for faster debugging.

The code works fine other than the if-else which does not switch as x & y are constantly equals to 0.

DEFINE_TURBULENT_VISCOSITY(test_mu_t,c,t)
{
real x[ND_ND];
real xx,y;
real mu_t;
real rho = C_R(c,t);
real k = C_K(c,t);
real d = C_D(c,t);
real Omega, alpha;
real dudx = C_DUDX(c,t);
real dudy = C_DUDY(c,t);
real dvdx = C_DVDX(c,t);
real dvdy = C_DVDY(c,t);

xx = x[0];
y = x[1];
alpha = 1+y;
if (xx >= -0.1 && xx <= 0){
Omega = sqrt(abs(2*((0.5*(dudy-dvdx)-(1.76656/(alpha)))*(0.5*(dvdx-dudy)+(1.76656/(alpha)))*2))); //+ (d/k)*Cw*epsilon_xy*(31.76656/(alpha*R));
}
else {
Omega = sqrt(abs(0.5*((dudy-dvdx)*(dvdx-dudy) + (dvdx-dudy)*(dudy-dvdx))));
Message("Check to see if loop reaches here. x=%g.\n",xx);
}

mu_t = Cmu*rho*k*5/(SQR(Omega));

return mu_t;
}

Centroid
 

xx and y are 0 because x itself is 0. You have to use C_CENTROID(x, c, t) before setting xx = x[0] and y = x[1].

May i know how to use C_CENTROID(x, c, t)?
Do I just include C_CENTROID(x, c, t)? According to this site (https://www.afs.enea.it/project/nept...ec-define-init),


DEFINE_TURBULENT_VISCOSITY(test_mu_t,c,t)
{
real x[ND_ND];
real xx,y;
real mu_t;
real rho = C_R(c,t);
real k = C_K(c,t);
real d = C_D(c,t);
real Omega, alpha;
real dudx = C_DUDX(c,t);
real dudy = C_DUDY(c,t);
real dvdx = C_DVDX(c,t);
real dvdy = C_DVDY(c,t);
C_CENTROID(x, c, t)

xx = x[0];
y = x[1];
alpha = 1+y;
if (xx >= -0.1 && xx <= 0){
Omega = sqrt(abs(2*((0.5*(dudy-dvdx)-(1.76656/(alpha)))*(0.5*(dvdx-dudy)+(1.76656/(alpha)))*2))); //+ (d/k)*Cw*epsilon_xy*(31.76656/(alpha*R));
}
else {
Omega = sqrt(abs(0.5*((dudy-dvdx)*(dvdx-dudy) + (dvdx-dudy)*(dudy-dvdx))));
Message("Check to see if it reaches here. x=%g.\n",xx);
}

mu_t = Cmu*rho*k*5/(SQR(Omega));

return mu_t;
}

Thanks so much! It works!

Good
 

Nice to know.