[dkjknobel]

Hi all,
I have problem with my UDF. I am modelling a problem where apart from continuity and momentum, I have to solve energy, species transport and one UDS equation (for the volume fraction of Nano particle). The properties of nano fluid are calculated through UDF which depends on (UDS)Volume fraction of nano particle. The energy, species transport and UDS equations have Source term(in red color in attached Image) having terms like properties which are function of variable and other variable are being solve. Some part of the UDF code is written below it is giving parse error for the line
C_UDSI(c,t,2)= dtc * (C_YI_G(c,t,0)[0]);
C_UDSI(c,t,3)= dtc * (C_YI_G(c,t,0)[1]);

c_source= dct*(C_UDSI_G(c,t,4)[0] + C_UDSI_G(c,t,5)[1]);
while interpreting.I know that something is wrong with udf. (to calculate laplace of C and T). Also I do not have much idea about how to define unsteady term phi (Volume Fraction) instead of (phi * density) or User defined memory allocation. I don't know how exactly do it
I will be grateful for your help.



DEFINE_ADJUST(adjust_UDS,domain)
{
Thread *t; cell_t c; face_t f;
domain = Get_Domain(domain_ID); /*Fill UDS with the variable*/
thread_loop_c (t,domain)
{
begin_c_loop (c,t)
{
C_UDSI(c,t,0) = C_VOF(c,t);
}
end_c_loop (c,t)
}
}

DEFINE_DIFFUSIVITY(UDS_diffusivity,c,t,i)
{
real gamma_UDS,temp,phi,mu_nf,rho_nf,d_b;
temp = C_T(c, t); /*cell value of temperature*/
phi=C_UDSI(c,t,0);
mu_nf=mu_bf/pow((1-phi),2.5);
rho_nf=phi*rho_p+(1-phi)*rho_bf;
d_b=(k_b * temp)/(3*pi*mu_nf*d_p);
gamma_UDS=d_b;
return gamma_UDS;
}

DEFINE_SOURCE(UDS_source,c,t)
{
real x[ND_ND],vof_source,temp,phi,mu_nf,rho_nf,D_T;
temp = C_T(c, t); /*cell value of temperature*/
phi = C_UDSI(c,t,0);
mu_nf=mu_bf/pow((1-phi),2.5);
rho_nf = phi*rho_p + (1-phi)*rho_bf;
D_T = 0.26*(k_bf/(2*k_bf+k_p*rho_nf))*(mu_nf/rho_nf)*phi;
C_UDSI(c,t,1)=D_T/temp;
vof_source = C_UDSI_G(c,t,1)[0] + C_UDSI_G(c,t,1)[1];
return vof_source;
}


DEFINE_SOURCE(temp_source,c,t)
{
real x[ND_ND],t_source,temp,phi,d_b,hcr,term1,term11,term12,
term2,term3,term4,D_T,rho_nf,alpha_nf,C_P_nf,mu_nf ;
temp = C_T(c, t);
phi = C_UDSI(c,t,0);
hcr = (rho_p * C_P_p)/(rho_nf/C_P_nf);
d_b=(k_b * temp)/(3*pi*mu_nf*d_p);
term11=d_b*(C_UDSI_G(c,t,0)[0]*C_T_G(c,t)[0]);
term12=d_b*(C_UDSI_G(c,t,0)[1]*C_T_G(c,t)[1]);
term1=term11+term12;
D_T = 0.26*(k_bf/(2*k_bf + k_p*rho_nf))*(mu_nf/rho_nf)*phi;
term2=hcr*(D_T/temp)*(SQR(C_T_G(c,t)[0])+ SQR(C_T_G(c,t)[1]));
C_UDSI(c,t,2)= dtc * (C_YI_G(c,t,0)[0]);
C_UDSI(c,t,3)= dtc * (C_YI_G(c,t,0)[1]);
term3= (C_UDSI_G(c,t,2)[0]);
term4= (C_UDSI_G(c,t,3)[1]);[/COLOR]
t_source = term1 + term2 + term2 + term4;
return t_source;
}


DEFINE_SOURCE(conc_source,c,t,dS,equ)
{
real x[ND_ND],c_source;
C_UDSI(c,t,4)= C_T_G(c,t)[0];
C_UDSI(c,t,5)= C_T_G(c,t)[1];
c_source= dct*(C_UDSI_G(c,t,4)[0] + C_UDSI_G(c,t,5)[1]);
return c_source;
}

[AlexanderZ]

first of all compile your UDF instead of interpret,

next

DEFINE_ADJUST(adjust_UDS,domain)
{
Thread *t; 
cell_t c; 
thread_loop_c (t,domain) 
{ 
begin_c_loop (c,t)
{ 
C_UDSI(c,t,0) = C_VOF(c,t);
}
end_c_loop (c,t)
}
}

you have error here

C_UDSI(c,t,2)= dtc * (C_YI_G(c,t,0)[0]);
C_UDSI(c,t,3)= dtc * (C_YI_G(c,t,0)[1]);

because dtc is not defined.

probably, code has other problems, compile it and show log

best regards

[dkjknobel]

Thank you Sir,
I have implemented the suggestion given by you. After compiling the code I am getting error message like..
Copied C:\Users\Deepak\Desktop\RaT=7E6_Rrho=6_900x900_del T=0.05Sec_SCNHM_1/UDF_SCNHM.c to libudf\src
Creating user_nt.udf file for 2ddp_host ...
(system "copy "C:\PROGRA~1\ANSYSI~1\v194\fluent"\fluent19.4.0\sr c\udf\makefile_nt.udf "libudf\win64\2ddp_host\makefile" ")
1 file(s) copied.
(chdir "libudf")(chdir "win64\2ddp_host")# Generating ud_io1.h
UDF_SCNHM.c
..\..\src\UDF_SCNHM.c(184) : error C2059: syntax error : '='
..\..\src\UDF_SCNHM.c(185) : error C2059: syntax error : '='
..\..\src\UDF_SCNHM.c(193) : error C2061: syntax error : identifier 'con_source'
..\..\src\UDF_SCNHM.c(193) : error C2059: syntax error : ';'
..\..\src\UDF_SCNHM.c(193) : error C2059: syntax error : 'type'
Creating user_nt.udf file for 2ddp_node ...
(system "copy "C:\PROGRA~1\ANSYSI~1\v194\fluent"\fluent19.4.0\sr c\udf\makefile_nt.udf "libudf\win64\2ddp_node\makefile" ")
1 file(s) copied.
(chdir "libudf")(chdir "win64\2ddp_node")# Generating ud_io1.h
UDF_SCNHM.c
..\..\src\UDF_SCNHM.c(184) : error C2059: syntax error : '='
..\..\src\UDF_SCNHM.c(185) : error C2059: syntax error : '='
..\..\src\UDF_SCNHM.c(193) : error C2061: syntax error : identifier 'con_source'
..\..\src\UDF_SCNHM.c(193) : error C2059: syntax error : ';'
..\..\src\UDF_SCNHM.c(193) : error C2059: syntax error : 'type'

Done.

The full udf that I have compiled is as below..

#include "udf.h"
#define s0 0 /* Initial Salinity of Mixture*/
#define t0 300 /* Initial Temperature of Mixture Unit:K*/
#define beta_t 0.0002 /*Coef of ther exp (base fluid) Unit:/degC*/
#define beta_s 0.0008 /*Coef of Salinity(base fluid) Unit:%(-1)*/
#define beta_t_p 0.0000008 /*Coef of ther exp(N-P)Unit:/degC*/
#define ks 0.0000000014 /*Mass Diffusivity Unit: m^2/s */
#define rho_bf 1000 /* Density of base fluid waterUnit:kg/m^3*/
#define rho_p 3950 /*Density of Nano-Particle Unit:kg/m^3 */
#define k_bf 0.5974 /*Ther Cond of base fluidUnit:w/m-k*/
#define k_p 40 /*Ther Cond of Nano-ParticleUnit:w/m-k*/
#define C_P_bf 4182 /*Sp Heat of base fluid Unit: J/kg-K*/
#define C_P_p 765 /* Sp Heat of Nano-Particle Unit: J/kg-K*/
#define mu_bf 0.000955 /* Dy Visco of base fluid Unit: kg/m-s*/
/* #define phi 0.0002 /* Vol Frac of Nano-Particle*/
#define n 3 /* N-P shapefactor=3 forspherNP*/
#define k_b 1.38064852e-23 /*Boltz Const UNIT:m^2kgs^-2K^-1*/
#define d_p 0.00000008 /* Nano-Particle Diameter*/
#define pi 3.14 /* pi */
#define dtc = 0.000000001 /* Assumed */
#define dct = 0.0000000009 /* Assumed */
#define S 0.5 /* Soret Parameter (Assumed)*/
#define D_f 0.5 /* Dufour parameter (Assumed) */
#define g 9.81
/*#define Th 308*/ /*Hot Fluid or Wall Temp(B.C or patching)*/
/*to calculate dct and dtc UNIT:K*/
/*#define Tc 300*/ /*Cold Fluid or WallTemp(B.C or patching)*/
/*to calculate dct and dtc UNIT: K*/
/* #define Ch 1*/ /*Hot Fluid or Wall Conc(B.C or patching)*/
/*(to calculate dct and dtc)*/
/* #define Cc 0*/ /*Cold Fluid or Wall Conc(B.C or patching)*/
/*to calculate dct and dtc*/
# define domain_ID 1


DEFINE_ADJUST(adjust_UDS,domain)
{
Thread *t;
cell_t c;
thread_loop_c (t,domain)
{
begin_c_loop (c,t)
{
C_UDSI(c,t,0) = C_VOF(c,t);
}
end_c_loop (c,t)
}
}


DEFINE_PROPERTY(nf_density,c,t)
{
real rho_nf,phi;
phi = C_UDSI(c,t,0);
rho_nf = phi*rho_p + (1-phi)*rho_bf;
/*density of Nano-Fluid*//*beta2 UNITS:%(-1)*/
return rho_nf;
}

DEFINE_PROPERTY(nf_thermalexpansioncoeff,c,t)
{
real beta_t_nf,rho_nf,phi;
beta_t_nf=((1-phi)*rho_bf*beta_t+phi*rho_p*beta_t_p)/rho_nf;
return beta_t_nf;
}

DEFINE_PROPERTY(nf_diffusivity,c,t)
{
real gamma,delta_t,delta_s,phi;
gamma=ks; /*MassDiffusivity Unit:m^2/s*/
return gamma;
}

DEFINE_PROPERTY(nf_spheat,c,t)
{
real C_P_nf,rho_nf,delta_t,delta_s,conc_salt,temp,phi;
temp=C_T(c,t);
conc_salt=C_YI(c,t,0);
delta_t=temp-t0;
delta_s=conc_salt-s0;
phi=C_UDSI(c,t,0);
rho_nf=phi*rho_p +(1-phi)*rho_bf;
C_P_nf=((1-phi)*rho_bf*C_P_bf+phi*rho_p*C_P_p)/rho_nf;
return C_P_nf;
}

DEFINE_PROPERTY(nf_th_conductivity,c,t)

{
real k_nf,k_nf_d,k_nf_n,phi;
phi=C_UDSI(c,t,0);
k_nf_d=k_bf*(k_p +(n-1)*k_bf-phi*(n-1)*(k_bf-k_p));
k_nf_n=(k_p+(n-1)*k_bf+phi*(k_bf-k_p));
k_nf=k_nf_d/k_nf_n;
return k_nf;
}

DEFINE_PROPERTY(nf_viscosity,c,t)
{
real mu_nf,phi;
phi=C_UDSI(c,t,0);
mu_nf=mu_bf/pow((1-phi),2.5);
return mu_nf;
}

DEFINE_PROPERTY(nf_th_diffusivity,c,t)
{
real alpha_nf,k_nf,k_nf_d,k_nf_n,rho_nf,C_P_nf;
real delta_t,delta_s,conc_salt,temp,phi;
temp=C_T(c,t);
conc_salt=C_YI(c, t, 0);
delta_t=temp - t0;
delta_s=conc_salt-s0;
phi=C_UDSI(c,t,0);
rho_nf = phi*rho_p + (1-phi)*rho_bf;
k_nf_d=k_bf*(k_p +(n-1)*k_bf-phi*(n-1)*(k_bf-k_p));
k_nf_n=(k_p+(n-1)*k_bf+phi*(k_bf-k_p));
k_nf=k_nf_d/k_nf_n;
alpha_nf= k_nf/(rho_nf*C_P_nf);
return alpha_nf;
}

DEFINE_DIFFUSIVITY(UDS_diffusivity,c,t,i)
{
real gamma_UDS,temp,phi,mu_nf,rho_nf,d_b;
temp = C_T(c, t); /*cell value of temperature*/
phi=C_UDSI(c,t,0);
mu_nf=mu_bf/pow((1-phi),2.5);
rho_nf=phi*rho_p+(1-phi)*rho_bf;
/*densityofNano-Fluid*//*beta2 UNITS:%(-1)*/
d_b=(k_b * temp)/(3*pi*mu_nf*d_p);
/*d_b Brownian Diffusion Coefficient UNIT:m^2/s*/
gamma_UDS=d_b;
return gamma_UDS;
}

DEFINE_SOURCE(ymom_source,c,t,dS,eqn)
{
real source,delta_t,delta_s,conc_salt,temp,phi,rho_nf,b eta_t_nf;
temp = C_T(c, t); /* cell value of temperature */
conc_salt = C_YI(c, t, 0); /* cell value of concentration */
delta_t = temp - t0; /* temperature difference */
delta_s = conc_salt - s0;
phi=C_UDSI(c,t,0);
rho_nf=phi*rho_p +(1-phi)*rho_bf;
/* C_CENTROID(x,c,t);*/
source = -rho_nf * (1 - beta_t_nf * delta_t + beta_s * delta_s)*g;
return source;
}

DEFINE_SOURCE(UDS_source,c,t,dS,eqn)
{
real vof_source,temp,phi,mu_nf,rho_nf,D_T;
temp = C_T(c, t); /*cell value of temperature*/
phi = C_UDSI(c,t,0);
mu_nf=mu_bf/pow((1-phi),2.5);
rho_nf = phi*rho_p + (1-phi)*rho_bf;
D_T = 0.26*(k_bf/(2*k_bf+k_p*rho_nf))*(mu_nf/rho_nf)*phi;
C_UDSI(c,t,1)=D_T/temp;
vof_source = C_UDSI_G(c,t,1)[0] + C_UDSI_G(c,t,1)[1];
/*D_T = 0.26*(k_bf/(2*k_bf+k_p*rho_nf))*(mu_nf/rho_nf)*phi;*/
/*D_T thermophoresis Diffusion Coefficient UNIT:m^2/s*/
/*C_CENTROID(x,c,t);*/
return vof_source;
}

DEFINE_SOURCE(temp_source,c,t,dS,eqn)
{
real tm_source,temp,phi,d_b,hcr,term1,term11,term12,
term2,term3,term4,D_T,rho_nf,alpha_nf,C_P_nf,mu_nf ;
temp = C_T(c, t);
phi = C_UDSI(c,t,0);
/*dtc = D_f*alpha_nf*(Th-Tc)/(Ch-Cc);*/
/*dtc Diffusivity of Dufour TypeUNIT:m^2/s*/
hcr = (rho_p * C_P_p)/(rho_nf/C_P_nf);
d_b= (k_b * temp)/(3*pi*mu_nf*d_p);
term11= d_b*(C_UDSI_G(c,t,0)[0]*C_T_G(c,t)[0]);
term12= d_b*(C_UDSI_G(c,t,0)[1]*C_T_G(c,t)[1]);
term1= term11+term12;
D_T= 0.26*(k_bf/(2*k_bf + k_p*rho_nf))*(mu_nf/rho_nf)*phi;
/*D_T thermophoresis Diffusion Coefficient UNIT:m^2/s*/
/*term2=hcr*(D_T/temp)*((pow(C_T_G(c,t)[0]),2)+(pow(C_T_G(c,t)[1]),2));*/
term2= hcr*(D_T/temp)*(SQR(C_T_G(c,t)[0])+ SQR(C_T_G(c,t)[1]));
C_UDSI(c,t,2) = dtc * (C_YI_G(c,t,0)[0]);
C_UDSI(c,t,3) = dtc * (C_YI_G(c,t,0)[1]);
term3 = (C_UDSI_G(c,t,2)[0]);
term4 = (C_UDSI_G(c,t,3)[1]);
/*C_CENTROID(x,c,t);*/
tm_source = term1 + term2 + term3 + term4;
return tm_source;
}

DEFINE_SOURCE(con_source,c,t,dS,eqn)
{
real con_source;
/*dct = -(S*beta_t_nf)*(ks/beta_s);*/
/*dct Diffusivity of Soret TypeUNIT:m^2/s*/
/*C_CENTROID(x,c,t);*/
C_UDSI(c,t,4)= C_T_G(c,t)[0];
C_UDSI(c,t,5)= C_T_G(c,t)[1];
con_source= dct*(C_UDSI_G(c,t,4)[0] + C_UDSI_G(c,t,5)[1]);
return con_source;
}


Also i need some insight for the part..
C_UDSI(c,t,2) = dtc * (C_YI_G(c,t,0)[0]);
C_UDSI(c,t,3) = dtc * (C_YI_G(c,t,0)[1]);
term3 = (C_UDSI_G(c,t,2)[0]);
term4 = (C_UDSI_G(c,t,3)[1]);

and

C_UDSI(c,t,4)= C_T_G(c,t)[0];
C_UDSI(c,t,5)= C_T_G(c,t)[1];
con_source= dct*(C_UDSI_G(c,t,4)[0] + C_UDSI_G(c,t,5)[1]);
Is this the right way to do the same??

I have read udf manuals and other related threads on this forum, but still unable to get rid of this problem.

[AlexanderZ]

DEFINE_SOURCE(con_source,c,t,dS,eqn)
{
real con_source;

change here name of function or variable.

C_UDSI(c,t,2) = dtc * (C_YI_G(c,t,0)[0]);
C_UDSI(c,t,3) = dtc * (C_YI_G(c,t,0)[1]);

there is some problem here, I'm not sure what is the problem, try to define dtc in other way -> don't use #define, but use real dtc=0.000001

All other parts seems correct and compiler didnt show other errors.
By the way - it is very very bad, when you have space in folder names.

C:\Users\Deepak\Desktop\RaT=7E6_Rrho=6_900x900_del T=0.05Sec_SCNHM_1/UDF_SCNHM.c

IT is better if your source file is located in working folder, the same with fluent case file, which you are using

I don't go deep into your code, don't check if it is correct in terms of equations.
I assume you know, what you are doing. So I'm trying to help you with syntax

best regards

[dkjknobel]

ehat is the meaning of this error?
999999: mpt_accept: error: accept failed: No such file or directory

[AlexanderZ]

line 260

Error: No user-defined functions of type udf-type-specific-heat have been loaded.
Error Object: #f

from ansys fluent customization manual

If you would like to use a UDF to define specific heat properties, you must use the
DEFINE_SPECIFIC_HEAT, as described in DEFINE_SPECIFIC_HEAT

best regards

[dkjknobel]

Dear Sir,
I have attached my udf and error that i am getting. The error comes before the first iteration.There is no problem with Interpretation of udf. There is no problem with initialization (and patching the domain ). I have also set the (rpsetvar 'species/save-gradients? #t). I think the problem is related with parallel processing or might be related with Environment Variable. Kindly guide me to find the problem and to do the same.

[AlexanderZ]

C:/Users/WORKSTATION-2/Desktop/scnhm3_rev/scnhm_rev.c:60: macro `DEFINE_DIFFUSIVITY' used with only 3 args

what is it?

DEFINE_DIFFUSIVITY(nf_diffusivity, c, t)

best regards

[dkjknobel]

hi..
This is the mass diffusivity in Species Transport equation. Should it be defined by Define Property ?

[AlexanderZ]

no

the answer is: this is an error, which was clearly reported in YOUR log file
read your logs

should be

DEFINE_DIFFUSIVITY(nf_diffusivity, c, t, i)

best regards

[ghl09013]

Hi，I have some question with your code of the second derivative of the Temperature:
C_UDSI(c,t,5) = C_T_G(c,t)[0];
C_UDSI(c,t,6) = C_T_G(c,t)[1];
concentration_source = density_nf * d_soret * ((C_UDSI_G(c,t,5)[0]) + (C_UDSI_G(c,t,6)[1]));

Is that right to do so?

[am2612]

Hello everyone...
The modified code has been successfully compiled and initialised. But during iteration, following message shows and iteration does not start:

Node 0: Process 5584: Received signal SIGSEGV

MPI application rank 0 exited before MPI_Finalise () with status 2

The f1 process could not be started.


Please help in rectifying it.

[am2612]

please elaborate the steps followed for complete iteration.