[zahraorveh]

i write this udf

#include "udf.h"
#include "sg.h"
#include "sg_mphase.h"
#include "flow.h"
#include "mem.h"
#include "metric.h"
#include "unsteady.h"

#define K 3.65e-4
#define gama 0.146
#define sigma 4.796e-4
#define alpha 1.935e-3
#define beta 5.7848e-7
#define kx 0.27318
#define kw 0.2
DEFINE_ADJUST(adjust_for_force, domain)
{
Thread **pt;
Thread *t;
cell_t c;
Domain *pDomain = DOMAIN_SUB_DOMAIN(domain,1);
real x,D;
real xc[ND_ND];
begin_c_loop(c,t)
C_CENTROID(xc,c,t);
C_UDMI(c,t,0)=500*exp(-1*(kx*(C_UDSI(c,t,0))+kw)*(0.04-xc[1]));
D=alpha*beta*pow(C_UDMI(c,t,0),2)+(alpha+beta)*sig ma*C_UDMI(c,t,0)+sigma*gama;
C_UDMI(c,t,1)=(sigma*gama+C_UDMI(c,t,0)*beta*sigma )/D;
C_UDMI(c,t,2)=alpha*sigma*C_UDMI(c,t,0)/D;
C_UDMI(c,t,3)=(alpha*beta*pow(C_UDMI(c,t,0),2))/D;
C_UDMI(c,t,4)=K*gama*C_UDMI(c,t,2);

end_c_loop(c,t)

}



it has to calculate 5 c_udmi for each cell of cfd computational cell. but it doesn't work

[`e`]

Is this DEFINE_ADJUST macro being called by the solver; have you hooked it?

[zahraorveh]

Originally Posted by `e`

Is this DEFINE_ADJUST macro being called by the solver; have you hooked it?

Yes,i hooked it and i defined memory location!!!

[`e`]

Check which value is being assigned to the UDM with:

Message("udm-0 = %e\n",500*exp(-1*(kx*(C_UDSI(c,t,0))+kw)*(0.04-xc[1])));

[zahraorveh]

Originally Posted by `e`

Check which value is being assigned to the UDM with:

Message("udm-0 = %e\n",500*exp(-1*(kx*(C_UDSI(c,t,0))+kw)*(0.04-xc[1])));

Value below 500 has to be assigned to udmi_0!!!!!
But udmi isnt calculated!!!!

[Bruno Machado]

I think you need to add a " thread_loop_c " to your code.

[pakk]

Originally Posted by Bruno Machado

I think you need to add a " thread_loop_c " to your code.

Good observation.
Zahraorveh, didn't you see any compilation warnings/errors? Similar to "warning: variable t undefined"?

[zahraorveh]

Originally Posted by pakk

good observation.
Zahraorveh, didn't you see any compilation warnings/errors? Similar to "warning: Variable t undefined"?

when i compiled it no error i saw!!!!
Can you send me your email to i send my case and data to you???
Have you free time?

[pakk]

No, please don't send me case and data. That will not help anybody.

When I compile it, I see the following:

test.c: In function adjust_for_force:
test.c:27: error: sig undeclared (first use in this function)
test.c:27: error: (Each undeclared identifier is reported only once
test.c:27: error: for each function it appears in.)
test.c:27: error: expected ; before ma
test.c:22: warning: unused variable x
test.c:21: warning: unused variable pDomain
test.c:18: warning: unused variable pt

So, no warning/error about t. The biggest problem from here is that you wrote "sig ma" in stead of "sigma".

But that does not change that t is never defined, it only means that the compiler did not catch this problem. Bruno Machado already gave you the solution.