CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   FLUENT (http://www.cfd-online.com/Forums/fluent/)
-   -   why enthalpy residual is great ??? (http://www.cfd-online.com/Forums/fluent/43279-why-enthalpy-residual-great.html)

Asghari January 9, 2007 11:43

why enthalpy residual is great ???
 
Hi CFD specialists ;

I am working on two phase flow with mass transfer process;I need to check following expression for computing residual of energy equation without energy source term in steady state. ( div.[ rho U (E+p/rho)] -div.[k (grad (T))]=0.0 (k is effective conductivity)

Fluent uses finite volume method for discretization of left hand side of upper equation according to the following form :

S= [(sigma on faces) {rho U.dA (E+p/rho)}-(sigma on faces ) { k grad(T).dA}]/ cell volume

In single phase flow without energy source term , S should be zero. I use from a udf in order to certify this object , Udf computes right hand side of upper equation ( S); The part of udf which calculates S is presented at following:

thread_loop_c (t,domain)

{

begin_c_loop(c,t)

{

int i=1;

C_UDMI(c,t,4)=0.0;

c_face_loop(c,t,i)

{

f=C_FACE(c,t,i);

tf=C_FACE_THREAD(c,t,i);

if(BOUNDARY_FACE_THREAD_P(tf)) {

BOUNDARY_FACE_GEOMETRY(f,tf,A,ds,es,A_by_es,dr0);

convection=(F_UDMI(f,tf,1)+F_UDMI(f,tf,2)/C_R(c,t))*F_FLUX(f,tf) ;

diffusion=C_UDMI(c,t,11)*A_by_es/ds*(F_T(f,tf)-C_T(c,t)); } else{

INTERIOR_FACE_GEOMETRY(f,tf,A,ds,es,a_by_es,dr0,dr 1);

if(F_C0(f,tf)==c) {

F_T_G[0]=0.5*(C_T_G(c,t)[0]+C_T_G(F_C1(f,tf),t)[0]);

F_T_G[1]=0.5*(C_T_G(c,t)[1]+C_T_G(F_C1(f,tf),t)[1]);

diffusion=0.5*(C_UDMI(c,t,11)+C_UDMI(F_C1(f,tf),t, 11))*A_by_es/ds*(C_T(F_C1(f,tf),t)-C_T(c,t));

if(F_FLUX(f,tf)>0) {

convection=C_UDMI(c,t,12)*F_FLUX(f,tf); /*method is first order upwind*/

else {

convection=C_UDMI(F_C1(f,tf),t,12)*F_FLUX(f,tf);

} }

else {

F_T_G[0]=0.5*(C_T_G(c,t)[0]+C_T_G(F_C0(f,tf),t)[0]);

F_T_G[1]=0.5*(C_T_G(c,t)[1]+C_T_G(F_C0(f,tf),t)[1]);

diffusion=0.5*(C_UDMI(c,t,11)+C_UDMI(F_C0(f,tf),t, 11))*A_by_es/ds*(C_T(F_C0(f,tf),t)-C_T(c,t));

if(F_FLUX(f,tf)<0) {

convection=-C_UDMI(c,t,12)*F_FLUX(f,tf);

} else {

convection=-C_UDMI(F_C0(f,tf),t,12)*F_FLUX(f,tf);

} } }

C_UDMI(c,t,4)=C_UDMI(c,t,4)-diffusion+convection ;

}

C_UDMI(c,t,4)=C_UDMI(c,t,4)/C_VOLUME(c,t) ;

}

end_c_loop(c,t) }

C_UDMI(c,t,11) indicates to effective conductivity in last iterration;

C_UDMI(c,t,12) indicates to [e+P/rho] in last iteration;

F_UDMI(f,tf,1)indicates to enthalpy in face;

F_UDMI(f,tf,2)indicates to pressure in face;

C_UDMI(c,t,4) is equal to S in context .

But in any iteration I faced to S in range of -250000 to 350000 ; whereas in any iteration S should be zero . Why it happended ?

Have you a suggestion in order that I could remedy this problem ? Does fluent use another method for calculating of S ?

I am appreciated for everybody who help me in solving this problem.

Thank's.

Asghari.



All times are GMT -4. The time now is 23:58.