The ABINIT Users Mailing List ( CLOSED )

[verstraete@pcpm.ucl.ac.be]

No, rather use iscf 3

the default iscf 5 is faster but doesn't work well for complicated cases. 
iscf 3 is more robust. This should be in the FAQ or in the definition of 
iscf. Everyone has the same problem. It's possible that increasing nband 
can work (test it!) but in some cases it won't.

Matthieu

On Wed, 20 Aug 2003, James Raynolds wrote:

> 
> 
>       Increase the value of "nband".
> 
> 
> -----Original Message-----
> From: homegchiche@yahoo.fr [mailto:homegchiche@yahoo.fr]
> Sent: Wednesday, August 20, 2003 8:02 AM
> To: forum@abinit.org
> Subject: [abinit-forum] Nickel GGA simulation
> 
> 
> Dear Abinit users,
> I have some problems in the GGA studies of  Bulk Nickel,  the calculs are
> not converge, I have got this error message :
>  findmin : WARNING -
> (scfcge) The second derivative is negative, equal to -4.3426189766E-05.
> findmin : lambda_predict  -3.0200E+00 etotal_predict  -3.1416489113E+01
> scfcge : ERROR -
> Potential-based CG line minimization not converged after 13 restarts.
> Action : read the eventual warnings about lack of convergence.
> Some might be relevant. Otherwise, raise nband.
> leave_new : decision taken to exit ...
> 
>  my input files is :
> 
> strprecon 0.1
>  ndtset 2
>  getwfk2 1
>  acell 3*6.6496
>  rprim  0.0  0.5  0.5
>         0.5  0.0  0.5
>         0.5  0.5  0.0
>  ntype 1
>  zatnum 28.0
>  natom 1
>  type  1
>  xred 0.0 0.0 0.0
>  nsppol 2
>  spinat 0.0 0.0 0.5
>  optcell 1
>  ionmov  3
>  nstep  60
>  ntime  1000 
>  toldfe 1.0d-6
>  ecut 50.0
>  ixc  11
> occopt 7
>  tsmear 0.01 
>  nband 6    
>  kptopt 1
>  ngkpt  4 4 4
>  nshftk 1
>  shiftk
>  0.5 0.5 0.5
> 
> The input files to generate the pseudopotentiel is : 
> 
> 28.00  5  3  6  1.00 : z  nc  nv iexc rnlc 
>     1  0   2.00      : n  l   f 
>     2  0   2.00 
>     2  1   6.00 
>     3  0   2.00 
>     3  1   6.00 
>     3  2   8.00 
>     4  0   2.00 
>     4  1   0.00 
> 2  t                 : lmax  s_pp_def 
> 0  2.10 0.0 t
> 1  2.91 0.0 t
> 2  2.08 0.0 t
> 
> i wait your help and thanks for your times
> Megchiche
> 

-- 
===================================================================
Matthieu Verstraete               mailto:verstraete@pcpm.ucl.ac.be 
PCPM, Boltzmann, pl. Croix du Sud, 1            tel: 010/ 47 86 81
B-1348 Louvain-la-Neuve  Belgium                fax: 010/ 47 34 52