Hi,

I am trying to find the onset of separation along a wall in posdat.f and feed this data back to some other subroutines. My program seems to give me Segmentatin fault AFTER A FEW ITERATIONS if I use 'write(*,*) IB,IREG' but works allright if I use 'write(*,*) IB,IRN(IB)'... I have checked to see if anuy of the IRN(IB) are -ve (just in case) but none are...

I have no idea why it should do so only due to a diferent write command??? Anyone has any idea? Please help... I am getting paranoid that STAR-CD has its own mind and its not on my side...

StarCD

I don't think IREG is defined in POSDAT: have a look in nom.inc. As far as I know, IREG is only defined for subroutines which are called per boundary, so it tells you which region the boundary belongs to. POSDAT is called for the whole domain so there isn't a unique value which could be given to IREG.

It is correct to use IRN in POSDAT, but make sure you only address it with IB between 1 and NWLMX, otherwise you'll get a segmentation fault.

Thanks Richard. May be it is the NWLMX... I was using NBP23...

Will check it out soon