
[Sponsors] 
September 26, 2003, 09:32 
Re: scalar transport equation problem

#1 
Guest
Posts: n/a

Hi all,
I set a scalar transport equation in fluent, but it surprised me that when the Iterating began, the solution of the scalar converged at once, that is, Residual was equal to perfect zero. When I finished the iterating and plot the result, I found that the value on the zone were all equal to zero. I think it means that such calculation of the scalar is invalid. Could somebody tell me how this happens? Thank you! Benny 

September 26, 2003, 12:19 
Re: scalar transport equation problem

#2 
Guest
Posts: n/a

Hi
I suppose that your scalar is not entering your system, but it is generated in the domain, isnt'it? You can try to answer to these questions: did you define the right source term for your UDS? did you enabled the source term in the boundary condition panel for 'fluid' (you have to click in the box of the 'source terms', and then to choose in the list the right source for each UDS')?  if your source term is partially evaluated in a DEFINE_ADJUST function, have you selected it in the Function Hooks panel (define>user defined> function hooks)?  if the source term should be zero(your UDS simply transported in your system), check your boundary conditions. Hope this helps. Regards, Ale 

September 26, 2003, 21:51 
Re: scalar transport equation problem

#3 
Guest
Posts: n/a

Hi Ale,
Thank you so much! I defined the right source term for my UDS, right sources have been selected after enabling the 'source term' in the Fluid Zone, and DEFINE_ADJUST has also been seclected in the Function Hooks panel. So I think that myabe I defined wrong Boundary Conditions of the UDS. I'd like to know whether we should set the same value for the same scalar but in different zone. For instance, in the Wall and Inlet zone, should the value of a scalar be equal to each other; or we should set different value for it according to the real conditions? Thank you! Best regards! Benny 

September 29, 2003, 05:20 
Re: scalar transport equation problem

#4 
Guest
Posts: n/a

You should set the correct b.c. for each zone; e.g. at walls you should set flux=0 if your scalar is a chemical species, the real flux if it is heat (or zero if your system is adiabatic), at inlets you should set the real value you have in the inlet stream and so on...
If you have nonzero values at inlets you must have nonzero values in the domain. Check the value of diffusivity, especially if you are not computing the convective flux (to compute the convective flux you have to select 'mass_flow_rate' as flux function for your UDS), and the value of your source terms (by UDMs). Let me know if you find the problem... Best luck and best regards, Ale 

May 9, 2010, 22:12 
Scalar transport: NEED HELP!

#5  
New Member
Gloria
Join Date: Feb 2010
Posts: 16
Rep Power: 8 
Hi,
I also meet the same problem. I use scalar transport equation to simulate the particle transport in a porous zone in 2D. The source term at each cell is U*A*CE*Cin/Vcontrol (The UDS equation has no density term), U is the cell velocity A is face area, which is also the face normal vector of the face (dx for V, and dz for U) CE is the collection efficiency of porous zone at each cell, Cin is the UDS which is the number concentration of the particle Vcontrol is the control volume of the cell which is dx*dz in 2D. The following is the code I wrote. Please help me to find the error, I am a starter and nobody near me can help me :(. Oh, BTW, as for the boundary of the porous zone (rectangular W*H), I set "interface" for three sides of the zone where it stand in the air, and the bottom is set to wall. Quote:
#include "udf.h" #include "stdio.h" #define dp 0.875e6 /*particle diameter (um) */ #define Dcyl 0.006 /*cylinder diameter of tree stem&twig (m) */ #define Dleaf 0.001 /*leaf diameter of tree stem&twig (m) */ #define K 1.38e23 /*Boltzmann constant (Nm/K)*/ #define T 293 /*Temperature at 20C (K)*/ #define mu 1.789e5 /*viscosity (N s/m^2) */ #define rhoP 1050 /*density of particle (Kg/m^3)*/ #define rhoA 1.225 /*density of air (Kg/m^3)*/ #define pi 3.1415926 /*constant Pi */ #define lambda 0.066 /*mean free path (um) */ #define g 9.81 /*gravity acceleration m/s^2*/ #define Cpol 1.32 /*Polhausen coefficient*/ #define dx 0.16 /*dx=W/n, W=1.6m, n=10 (m)*/ #define dz 0.1375 /*dz=H/n H=2.2m, n=16, m */ /*Calculated constant by giving a dp*/ #define Cc 1.177 /*Slip corrction factor*/ #define dm 3.22492e11 /*laminar diffusion coefficient (m^2/s) */ #define Trel 2.49589e6 /*relaxation time of the particle (s)*/ #define Usettle 0.00002882 /*settling speed of particle (m/s) */ #define B 8.55378e7 #define D 68458.70124 /************************************************** ************************************************** ***************/ DEFINE_SOURCE(uds_source0,c,t,dS,eqn) { real source; face_t f; real x[ND_ND], z, Ucell, volume,NV_VEC(A), NV_VEC(Uface); z=x[1]; Ucell=C_U(c,t); volume=C_VOLUME(c,t); F_AREA(A,f,t); int i; /**********************For collection efficiency of leaves and cylinder of the tree (CEcell) ************************/ /*CE is the total collection efficiency =CEcyl+CEleaf (<=1)at each cell */ real CEcell,CEcell_c,CEcyl,CEleaf, SADcell; real Ec,Ec_c, El, El_c;/*collection efficiency of each cell due to all deposition mechanisms*/ real Edc, Edl; /*collection efficiency of cylinder&leaf due to DIFFUSION*/ real Eiic, Eiil; /*collection efficiency of cylinder&leaf due to INTERCEPTION+IMPACTION*/ real Eic, Eil; /*collection efficiency of cylinder&leaf due to IMPACTION*/ real Rec, Rel,Stkc, Stkl, Rc, Rl; Rec=D*Dcyl*Ucell; /*Reynolds number for flow around a cylinder*/ Rel=D*Dleaf*Ucell; /*Reynolds number for flow around a leaf*/ Stkc=Trel*Ucell/Dcyl;/*Stk number for flow around a cylinder*/ Stkl=Trel*Ucell/Dleaf; /*Stk number for flow around a leaf*/ Rc=dp/Dcyl; /*interception parameter for flow around a cylinder*/ Rl=dp/Dleaf; /*interception parameter for flow around a leaf*/ /************************************************** ************************************************** ********************/ begin_f_loop(f,t) { C_CENTROID(x,f,t); /*for cylinder collection efficiency at each cell due to all deposition mechanisms*/ Edc= B/pow((Dcyl*Ucell),0.5); Eiic = (Stkc/(Stkc+0.8)(2.56log10(Rec)3.2*Rc)/(10*sqrt(Stkc)))*(1+Rc); Eic = Eiic; if (Eic >(1+Rc)){ Eiic = pow(Stkc/(Stkc+0.8),2); } else if (Eic< 0){ Eiic= pow(Stkc/(Stkc+0.8),2); } else{ Eiic= Eic; } /*cylinder*/Ec = (Edc + Eiic); Ec_c=Ec; if (Ec_c>=1){ Ec=1; } else if (z>((3/8)*x[0]+2.5)&&z>((3/8)*x[0]+1.9)&&z<(2.1*x[0]1.46)&&z<(2.1*x[0]+1.9)){ Ec=0; } else { Ec=Ec_c; } /*Surface area density (SAD)of foliage */ if (z <= 2.07){ SADcell = 3.99043*pow(z,3)+12.8994*pow(z,2)7.58257*z+1.89096; } else { SADcell = 224.925*pow(z,2)+912.217*z918.31; } CEcyl = 1exp(1SADcell*Ec*dx); /*total collection efficiency at each cell */ CEcell=2*CEcyl; CEcell_c=CEcell; if (CEcell_c>=1){ CEcell=1; } else { CEcell=CEcell_c; } source = C_UDSI(f,t,0)*CEcell*Ucell*F_AREA(A,f,t)/volume; dS[eqn] = CEcell*Ucell*F_AREA(A,f,t)/volume; } end_f_loop(f,t) return source; } 

Thread Tools  
Display Modes  


Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
UDF Transport equation for transition momentum thickness Reynolds number  Wantami  FLUENT  0  April 9, 2011 02:50 
Scalar transport in Lid driven cavity  CFDtoy  Main CFD Forum  2  February 19, 2011 10:43 
User defined scalar transport problem (UDS)  jpcfd  FLUENT  0  June 8, 2010 04:50 
Is it possible to influence the velocity of an scalar transport equation?  jannnesss  CFX  1  January 11, 2010 13:02 
Energy equation convergence problem  Reza  Main CFD Forum  0  August 27, 2003 13:09 