Author Topic: fatal error in nucex  (Read 6791 times)

Molcasito

  • Newbie
  • *
  • Posts: 7
  • Karma: +0/-0
fatal error in nucex
« on: April 17, 2008, 01:56:26 PM »
Dear TM users
I have a problem with TM running on parallel
after I do the statistics and submitt the job, I found the error

fatal error in nucex.
dscf ended abnormally


                           
It is a SCF/DFT calcualtion on parallel
Any help about this error will thankful

Thanks in advanced
M.

Arnim

  • Developers
  • Sr. Member
  • *
  • Posts: 235
  • Karma: +0/-0
Re: fatal error in nucex
« Reply #1 on: April 18, 2008, 09:43:27 AM »
Hi!

Are you sure that your geometry is set up correctly?

Best,
Arnim

Molcasito

  • Newbie
  • *
  • Posts: 7
  • Karma: +0/-0
Re: fatal error in nucex
« Reply #2 on: April 20, 2008, 02:57:07 PM »
Hello!
It looks like to me to be rel;ated to the $user-defined bonds keyword in the coord file.
Do I really need to include this keyword in the coord and/or control file?
I am only using cartesian coordinates

Many thanks
M.

uwe

  • Global Moderator
  • Sr. Member
  • *****
  • Posts: 486
  • Karma: +0/-0
Re: fatal error in nucex
« Reply #3 on: April 25, 2008, 03:12:47 PM »
Hi,

the user defined bonds have nothing to do with the error message.

Either you have entered Angstrom values as coordinates instead of atomic units, so that all distances are unreasonable, or you have changed the symmetry group by hand without letting define adapt the coordinates to that symmetry.

Please check that your coordinates are in atomic units, and rerun define: enter the geometry section and say 'desy' to let define determine the symmetry.

Regards,

Uwe