Adding LSORBIT Causes Errors in WAVECAR

Problems running VASP: crashes, internal errors, "wrong" results.

Moderators: Global Moderator, Moderator

Post Reply
Posts: 1
Joined: Thu Feb 13, 2020 8:48 am

Adding LSORBIT Causes Errors in WAVECAR

#1 Post by josh_reding » Mon Mar 30, 2020 10:08 am

I'm doing DFT calculations on 2D perovskite materials. I can run them without issue without using LSORBIT, but when I add it to my INCAR, I get a slew of errors in my slurm file related to the WAVECAR. INCAR and errors are listed below. Any help would be appreciated.

!Parallel Options
LPLANE = .TRUE. (Real space distribution)
!NPAR = 1 (Max is no. nodes)
ALGO = Normal

!Electronic Relaxation
PREC = Low (Precision level for FFT grid)
LREAL = .TRUE. (Projection operators: automatic)
!ENMAX = 600.00 eV (Plane-wave cutoff)
NELM = 100 (Max number of SCF steps)
NELMIN = 5 (Min number of SCF steps)
EDIFF = 3E-09 (SCF convergence criteria)

!Spin-orbit Coupling (This section new for SOC calculations)
LSORBIT = .TRUE. (Spin-orbit coupling)

!Output files
LORBIT = 11 (Prints PDOS)
!EMIN = 20 (EMIN for DOS)
!EMAX = 30 (EMAX for DOS)

Error Messages:
LDA part: xc-table for Pade appr. of Perdew
found WAVECAR, reading the header
number of bands has changed, file: 280 present: 720
trying to continue reading WAVECAR, but it might fail
number of k-points has changed, file: 4 present: 5
trying to continue reading WAVECAR, but it might fail
WAVECAR: different cutoff or change in lattice found
POSCAR, INCAR and KPOINTS ok, starting setup
FFT: planning ...
reading WAVECAR
ERROR: while reading WAVECAR, plane wave coefficients changed 44842

Posts: 122
Joined: Mon Sep 24, 2018 9:39 am

Re: Adding LSORBIT Causes Errors in WAVECAR

#2 Post by merzuk.kaltak » Mon Mar 30, 2020 1:02 pm

You haven't uploaded a proper error report, so it is quite hard for us to answer your question and we are forced to guess what the problem with your setup.
The most common mistake with spin-orbit coupling switched on is that one uses the std version of VASP.
The wiki entry about LSORBIT, however, explains that the non-collinear version of VASP should be used for these kind of calculations.

Post Reply