[Pw_forum] linmin error#2

[Stefano de Gironcoli]

this is an error that appears more often than we would like...
this happens when during the line minimization step the cubic
polynomial used to fit energy and forces of corrent and previous steps
does not find a proper minimum...
A useful--- even if not very automatic ---strategy is to restart the minimization
from the best configuration found so far...

However looking to your forces I notice that some of them are gigantic...
are you sure you are not putting  atoms 3,4,7,8 in clearly wrong positions ?
Your energy went  up  of 20Ry since the previous step (Eold-Etot) ...
I guess you started from an unrealistic configuration and the first relaxation
step kicked the sistem in an even less realistic one and then crashed ...

check your starting configuration...

regards, stefano

mutombo wrote:

> Hello,
> Can someone tell me what is wrong? I am trying to relax a Si(111) slab .The
> code stopped suddenly when searching for  a next position. Below is a part of
> the output file:
> ...
> convergence has been achieved
>
>      Forces acting on atoms (Ry/au):
>
>      atom   1 type  1   force =    -0.00001129    0.20701935   -0.25959104
>      atom   2 type  1   force =    -0.00030341    0.11931099   -0.27967793
>      atom   3 type  1   force =    -0.00424617    0.04599481  -25.20392513
>      atom   4 type  1   force =     0.00114704   -0.19525276   24.40961533
>      atom   5 type  1   force =     0.00444265   -0.06459323   -0.00001344
>      atom   6 type  1   force =    -0.00526230    0.05333743    0.04068551
>      atom   7 type  1   force =     0.00569425    0.02267289  -24.56861329
>      atom   8 type  1   force =     0.00297920   -0.00070163   25.18973657
>      atom   9 type  1   force =    -0.00072625   -0.16743179    0.64279784
>      atom  10 type  2   force =    -0.00371372   -0.02035606    0.02898557
>
>      Total force =   101.548839     Total SCF correction =     0.000638
>      searching for next position (pslinmin)...
>      Eold  =    -60.65963916
>      Etot  =    -40.45408960
>      DEold =     -0.66033281
>      DEtot =    -49.69400583
>      linmin: 3rd order interpolation
>
>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
>      from linmin : error #         2
>      unexpected error
>  %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
>
>      stopping ...
>
> Regards,
>                                              Pingo
>
> _______________________________________________
> Pw_forum mailing list
> Pw_forum at pwscf.org
> http://www.democritos.it/mailman/listinfo/pw_forum