Learning Forum

[Danmei]

Hi all,nI'm running a simple simulation using mixture model + standard k-e turbulence model. I need to achieve the extraction of the secondary phase at the bottom wall by using udf. I used DEFINE_ADJUST to calculate the amount of mass I need to take out and DEFINE_SOURCE to add the source term in the cell zone(2nd phase). I am receiving this error message whenever I tried to calculate: Node 0: Process xxxx: Received signal SIGSEGV. nnI have several check points:nThe udf itself has no problem compiling and loading. nThe simulation runs well without udf. nI set the UDM location to 1 before I build and load the librarynI tried initializing the model without udf first and then hook up the udf, it didn't worknI tried to create the parallel version of my code, still it doesn't have any problems with compiling and loading but cannot start the computationnI tried interpret the udf instead, it doesn't do anythingnI checked the zone id, they are all correctnI tried to hook up the DEFINE_ADJUST only, it gave me the same error, it seems that it is the DEFINE_ADJUST part that causes this problemnCan anyone help me?n#include udf.hnnDEFINE_ADJUST(my_adjust1, d)n{ntThread *t, **pt;ntface_t f;ntcell_t c;nntt = Lookup_Thread(d, 9);nttbegin_c_loop(c, t)nt{nttC_UDMI(c,t,0)=0.0;nttpt=THREAD_SUB_THREADS(t);nttif (C_V(c, pt[1]) < 0.0)ntttC_UDMI(c,t,0) = -1.368*C_R(c, pt[1])*C_VOF(c, pt[1])*C_V(c, pt[1]);nt}ntend_c_loop(c, t)nn}nnDEFINE_SOURCE(bot_source1, c, t, dS, eqn)n{ntThread *super_t = THREAD_SUPER_THREAD(t);ntreturn (-1.0*C_UDMI(c,super_t,0));n}n

[Keyur Kanade]

Looks duplicate of n

Hi all,I'm running a simple multiphase simulation of a tank. I use mixture model and standard k-e. nothing else. In the model, I need to "take" the second phase out if it hit the bottom floor with a certain velocity, similar to degassing boundary condition, by using udf. I used DEFINE_ADJUST to calculate the mass source and used DEFINE_SOURCE to add the source term. I am receiving this error message everytime I hit the calculation button: Node 0: Process 25932: Received signal SIGSEGVI have several check points here:The udf compiled and loaded properly I set the UDM location to 1 before I build and load the udfI also tried initializing the system by calculating it without udf first, as suggested on the forum, it didn't helpThe model runs well without udfI tried to hook up the my_adjust only, it gave me the same error, it seems like it has something to do with the DEFINE_ADJUST part onlyI tried both steady-state and transient, the results are the sameI also tried to interpret the udf instead of compiling, it didn't helpI made sure that the zone ID are correctCan someone please help me?#include "udf.h"DEFINE_ADJUST(my_adjust1, d){tThread *t, **pt;tface_t f;tcell_t c;tt = Lookup_Thread(d, 9);ttbegin_c_loop(c, t)t{ttC_UDMI(c,t,0)=0.0;ttpt=THREAD_SUB_THREADS(t);ttif (C_V(c, pt[1]) < 0.0)tttC_UDMI(c,t,0) = -1.368*C_R(c, pt[1])*C_VOF(c, pt[1])*C_V(c, pt[1]);t}tend_c_loop(c, t)}DEFINE_SOURCE(bot_source1, c, t, dS, eqn){tThread *super_t = THREAD_SUPER_THREAD(t);treturn (-1.0*C_UDMI(c,super_t,0));}/forum/discussion/20719/node-0-process-25932-received-signal-sigsegv-fluent-udf-c-udmi

n