Failure when switching on spin-orbit coupling

Message

wellsdm · #1 Post by **wellsdm** » Mon Nov 05, 2007 7:58 pm

I have made serveral runs on an orthorhombic structure with no problem but then I try to turn on LSORBIT = .true. it fails. I am getting the following error.

Mon Nov 5 14:37:52 EST 2007
/spncpl/zdir
Host is:malachite
Try lamboot
621 /home/users/don/execute/nt_vasp3 running on n0 (o)
622 /home/users/don/execute/nt_vasp3 running on n0 (o)
623 /home/users/don/execute/nt_vasp3 running on n0 (o)
624 /home/users/don/execute/nt_vasp3 running on n0 (o)
625 /home/users/don/execute/nt_vasp3 running on n0 (o)
626 /home/users/don/execute/nt_vasp3 running on n0 (o)
627 /home/users/don/execute/nt_vasp3 running on n0 (o)
628 /home/users/don/execute/nt_vasp3 running on n0 (o)
running on 8 nodes
distr: one band on 8 nodes, 1 groups
vasp.4.6.28 25Jul05 complex
POSCAR found : 4 types and 24 ions
LDA part: xc-table for Ceperly-Alder, standard interpolation
found WAVECAR, reading the header
number of k-points has changed, file: 36 present: 64
trying to continue reading WAVECAR, but it might fail
WARNING: stress and forces are not correct
POSCAR, INCAR and KPOINTS ok, starting setup
WARNING: wrap around errors must be expected
FFT: planning ... 1
reading WAVECAR
reading wavefunctions of collinear run, up
reading wavefunctions of collinear run, down
the WAVECAR file was read sucessfully
charge-density read from file: CsCuYS3
magnetization density read from file 1
LAPACK: Routine ZPOTRF failed! 1
LAPACK: Routine ZPOTRF failed! 1
LAPACK: Routine ZPOTRF failed! 1
LAPACK: Routine ZPOTRF failed! 1
LAPACK: Routine ZPOTRF failed! 1
LAPACK: Routine ZPOTRF failed! 1
LAPACK: Routine ZPOTRF failed! 1
LAPACK: Routine ZPOTRF failed! 1

I made previous self-consistent runs, then keeping everything constant made ICHARG = 11 and the other suggestions made in the manual. Any suggestions would be greatly appreciated.

#2 Post by **admin** » Tue Nov 06, 2007 11:58 am

if the k-point mesh has changed, the charge obtained from the read WAVECAR file may be completely wrong because the ordering of the k-points may differ (i.e. the read 36 kpoints to not correspond to the first 36 k-points of the 64-mesh, and the weights are different,....)
Please check if the errors persist if you either keep the same k-mesh or do not start from the WAVECAR.

tsemi · #3 Post by **tsemi** » Fri Jan 15, 2010 6:51 pm

Hello,
I am having this exact problem. What do you mean by 'keep the same k-mesh'? How do I ensure I am doing this?

Thank you,
tsemi

tsemi · #4 Post by **tsemi** » Mon Jan 18, 2010 4:16 pm

Could someone please explain how to make sure the k-mesh is the same for SOC runs?
Thank you.

#5 Post by **admin** » Thu Mar 18, 2010 12:15 pm

as SOC can only be run without any symmetry imposed on the cell, the k-mesh of the full BZ has to be used. You can eg start a run with the input parameters you use for the SOC run and use the IBZKPT file (written at the beginning of a vasp run) as KPOINTS file