[Pw_forum] scf stops after finishing 100 iteration

Gabriele Sclauzero sclauzer at sissa.it
Thu Nov 5 15:34:51 CET 2009



udayagiri sai babu wrote:
> Dear Dimpy Sharma
> you modify your electrons card as below. i.e add the option 
> electron_maxstep=1000 by default it is 100. If convergence is not 
> achieved in 100 steps then it stops. So put the variable 
> electron_maxstep to a high value say 1000 so that the scf calculation 
> runs till it is converged.

Or until it reaches 1000 steps. If two atoms are too close because of wrong input 
coordinates, as Stefano was suggesting, you will probably never reach convergence and lose 
  10 times the cputime you've already lost before... when a calculation does not converge 
in 100 steps, it is always better to ask ourselves why it did not.
Instead of increasing electron_maxstep one could for instance check the structure, check 
cutoffs, smearing... and play a bit with mixing_beta and mixinng_mode.


Cheers

GS


>  
>  &ELECTRONS
>     diagonalization ='david'
>         mixing_mode = 'plain'
>            conv_thr = 1.0d-6
>         mixing_beta = 0.7
>         electron_maxstep=1000
> 
> On Wed, Nov 4, 2009 at 2:35 AM, Stefano de Gironcoli <degironc at sissa.it 
> <mailto:degironc at sissa.it>> wrote:
> 
>     have you verified that your crystal structure is reasonable using a
>     visualizer like xcrysden or similar ?
>     it appears to me that a couple of atoms are very very close to each
>     other.
>     stefano
> 
>     Dimpy Sharma wrote:
>      >
>      > Hi Quantum espresso users,
>      >
>      > I tired to perform scf calculation but it shows me the following
>     error
>      >      'convergence NOT achieved after 100 iterations: stopping'
>      >
>      > However the nscf calculation and projected density of states run
>      > successfully, why is this so?
>      >
>      > My input file is as follows
>      >
>      > &CONTROL
>      >      calculation ='scf'
>      >     restart_mode = 'from_scratch'
>      >           outdir = '/sfiwork/dsharma/PDOSSilane/PDOSsi/si_wfc'
>      >       pseudo_dir = '/sfiwork/dsharma/QE_Pseudos/'
>      >           prefix = 'si'
>      >          tstress = .true.
>      >          tprnfor = .true.
>      >    etot_conv_thr = 1.D-4
>      >    forc_conv_thr = 1.D-3
>      >            nstep = 600
>      >           wf_collect =.true.
>      >  /
>      >  &SYSTEM
>      >         ibrav = 0
>      >     celldm(1) =7.4088
>      >           nat = 6
>      >          ntyp = 1
>      >       ecutwfc = 40
>      >       ecutrho = 160.0
>      >          nbnd = 12
>      >  /
>      >  &ELECTRONS
>      >     diagonalization ='david'
>      >         mixing_mode = 'plain'
>      >            conv_thr = 1.0d-6
>      >         mixing_beta = 0.7
>      > /
>      > &IONS
>      >  /
>      > &CELL
>      >  cell_dynamics='none'
>      >  /
>      > CELL_PARAMETERS cubic
>      >      1.000000000    0.000000000    0.000000000
>      >      0.000000000    2.000000000    0.000000000
>      >      0.000000000    0.000000000    2.000000000
>      > ATOMIC_SPECIES
>      >    Si   28.08600  Si.pz-vbc.UPF
>      > ATOMIC_POSITIONS angstroms
>      >  Si        2.95020       -0.66592       -0.12341
>      >  Si        4.90939        0.61749        0.11358
>      >  Si        0.99475        0.65272       -0.16138
>      >  Si       -0.98472       -0.62740       -0.15123
>      >  Si       -2.94624        0.68229       -0.12124
>      >  Si       -4.89549       -0.61534        0.11905
>      >  K_POINTS automatic
>      >   4 1 1   0 0 0
>      >
>      > Can anybody give me any suggestions?
>      >
>      > Thanks
>      >
>      > Dimpy
>      >
>      > Dimpy Sharma,
>      > PhD
>      > Electronics Theory  Group
>      > UCC,Cork
>      > Ireland
>      >
>      >
>      >
>     ------------------------------------------------------------------------
>      >
>      > _______________________________________________
>      > Pw_forum mailing list
>      > Pw_forum at pwscf.org <mailto:Pw_forum at pwscf.org>
>      > http://www.democritos.it/mailman/listinfo/pw_forum
>      >
> 
>     _______________________________________________
>     Pw_forum mailing list
>     Pw_forum at pwscf.org <mailto:Pw_forum at pwscf.org>
>     http://www.democritos.it/mailman/listinfo/pw_forum
> 
> 
> 
> 
> -- 
> U.Saibabu
> PhD student,
> Deformation mechanisms modeling group,
> Materials engineering department,
> IISc Bangalore,
> India.
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum

-- 


o ------------------------------------------------ o
| Gabriele Sclauzero, PhD Student                  |
| c/o:   SISSA & CNR-INFM Democritos,              |
|        via Beirut 2-4, 34014 Trieste (Italy)     |
| email: sclauzer at sissa.it                         |
| phone: +39 040 3787 511                          |
| skype: gurlonotturno                             |
o ------------------------------------------------ o



More information about the users mailing list