convergence of chemical shift calculus

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


Moderators: Global Moderator, Moderator

Post Reply
Message
Author
Francois Vivet
Newbie
Newbie
Posts: 1
Joined: Thu Oct 09, 2014 5:53 pm

convergence of chemical shift calculus

#1 Post by Francois Vivet » Wed Nov 12, 2014 7:38 pm

I am a new user of VASP (5.3.3). I have been trying to perform calculus of chemical shift for 3 weeks now and I always get the same result (output of the bach) :
generate k-points for: 1 1 1
DFATOM: ERROR, unable to reach convergence in 100 cycles.

and the calculus stops whatever the value of ENCUT I choose or the way I build the K-points. Is someone abble to tell me what these 100 cycles are and which tag(s) I have to change to get a convergence of my calculus?
In addition, here is my INCAR file:
SYSTEM = (Sr Al2 O4)6 (P6_3) ~ (Pearson #1210460)_1 (VASP)

PREC = N
LCHIMAG =.TRUE.
ENCUT = 800.000
EDIFF = 1.0e-8
ISMEAR = 0; SIGMA = 0.1
DQ = 0.001
ICHIBARE = 1
LNMR_SYM_RED = .TRUE.
NSLPLINE = .TRUE.
LREAL = A
NBANDS = 142
NELMIN = 4
ISTART = 0

Best regards.

François Vivet

admin
Administrator
Administrator
Posts: 2921
Joined: Tue Aug 03, 2004 8:18 am
License Nr.: 458

Re: convergence of chemical shift calculus

#2 Post by admin » Wed Nov 26, 2014 2:45 pm

Stopping criterion 1.0e-8 is very severe. Depending on the symetry of the system it can
be difficult to reach the convergence using 1.0e-8 withing a short time.
Try to preconverge the structure in several steps using milder criterion 1.0e-6, 1.0e-7.

Post Reply