[Pw_forum] stress in CP and FFT grids

[Axel Kohlmeyer]

On Tue, 23 May 2006, Konstantin Kudin wrote:

KK>  Hi all,

hi kostya,

KK>  It appears that the stress that comes out of the CP code is rather
KK> sensitive to the FFT grids used.

what kind of cutoffs are you using? 
does your system expand or shrink? and how much?

of course the results will always depend on the grids 
(which are defined implicitely by plane wave and density 
cutoff). due to the implementation, you cannot easily 
change grids during the simulation. so you _always_ have 
to restart or use a higher cutoff to start with, so you 
won't have too much troubles with the effects you are 
seeing (and also to reduce pulay stress).

best regards,
   axel.

KK>  For example, I ran a constant pressure simulation ('vc-cp') started
KK> from a certain unit cell. Naturally, the code kept all the FFT grids
KK> unchanged throughout (generated via the initial cell dimensions), and
KK> optimized the cell vectors to certain values.
KK> 
KK>  Then, when I started a fresh simulation with the optimized cell
KK> dimensions, the stress was quite a bit different, on average about 0.6 
KK> GPa higher than before. Looking at the settings, it appears that the
KK> FFT grids have changed because of the fresh start, and that made the
KK> stress different.
KK> 
KK>  Any suggestions on how to deal gracefully with such situations,
KK> especially when the system is "soft" along certain cell coordinates? It
KK> hardly comforting to have about 5% variations in the cell parameters
KK> due to the grids ...
KK> 
KK>  Thanks!
KK>  Kostya
KK>  
KK>  
KK> 
KK> __________________________________________________
KK> Do You Yahoo!?
KK> Tired of spam?  Yahoo! Mail has the best spam protection around 
KK> http://mail.yahoo.com 
KK> _______________________________________________
KK> Pw_forum mailing list
KK> Pw_forum at pwscf.org
KK> http://www.democritos.it/mailman/listinfo/pw_forum
KK> 

-- 
=======================================================================
Axel Kohlmeyer   akohlmey at cmm.chem.upenn.edu   http://www.cmm.upenn.edu
   Center for Molecular Modeling   --   University of Pennsylvania
Department of Chemistry, 231 S.34th Street, Philadelphia, PA 19104-6323
tel: 1-215-898-1582,  fax: 1-215-573-6233,  office-tel: 1-215-898-5425
=======================================================================
If you make something idiot-proof, the universe creates a better idiot.