values stored in UDM are different from their original values
 

Hello,
Now I come across a strange phenomenon that the values or variables distribution stored in UDM are different from their original.

I am debugging a multiphase reaction using Euleran model (gas-solid granular flow, 2 phases). It's always wrong this time. So the reaction source is set to 0 and several UDMs are placed inside the DEFINE_HET_RXN_RATE. The results show that some values stored in UDMs are different from their original value.

Have you guys ever encountered this phenomenon.
How does it happen? And is it reliable to call values in this way during Euleran multiphase simulation.

Thank you very much.

The udf and comparisions are shown below.

DEFINE_HET_RXN_RATE(debug,c,t,hr,mw,yi,rr,rr_t)
{

Thread *tp,*ts ;

ts= THREAD_SUB_THREAD(t,1);
tp= THREAD_SUB_THREAD(t,0);


C_UDMI(c,t,31)= C_T(c,tp);// gas temperature
C_UDMI(c,t,33)= C_T(c,ts); //solid temperature

C_UDMI(c,t,28)= C_R(c,tp); // gas density
C_UDMI(c,t,26)=C_YI(c,ts,2);

*rr = 0.0;

}

The problem has been solved.
The reason maybe from switching between different versions. The original cas and data is established by Version 2020. When it is reopened by version 15, there is some unpredictable chaos. Be carefully open the files (generated from high version) with a lower version.

But it must note that there is still some tiny differences between the UDM and it's twin value, when rebuild by Fluent 15. The differences may have minor impact on the calculation.