[gpaw-users] Problem with anisotropic mesh and symmetries

[Jens Jørgen Mortensen]

On 01/19/2016 04:04 PM, Aldegunde Rodriguez, Manuel wrote:
>
> Dear all,
>
>
> I am simulating an alloy (diamond lattice) and the unit cell of some 
> configurations is quite anisotropic. If I let gpaw generate the grid 
> for me, I get an assertion error in the call tosymmetry.check_grid 
> because this is not commensurate with one of the symmetries.
> To avoid this, I can switch off symmetries or set the number of grid 
> points in every direction to be the same. If I use the latter 
> option, then I get an error in the creation of 
> the GridDescriptor because it says it is too anisotropic. Again, I can 
> avoid this by changing the threshold of "sanity" in GridDescriptor to 
> a higher value.
>
> I find these workarounds a bit unsatisfactory, so I was wondering if 
> there is a better solution to this.
>
>
> A minimal working example to reproduce this error attached.
>

I have added an experimental option to GPAW that will turn off 
symmetrization of the density.  It's in the SVN-trunk version of the 
code and it works like this:

calc = GPAW(mode=PW(400),
             kpts={'size': (4, 4, 4), 'gamma': True},
             symmetry={'do_not_symmetrize_the_density': True})

That should work for your crystal.

GPAW will for historical reasons symmetrize the density with all the 
symmetries of the crystal even though that should not be necessary. Some 
day I'd like to skip this step, but for now you can use the trick above.

Jens Jørgen


>
> Thank you in advance,
>
>
> Manuel
>
>
> ---
>
>
> Manuel A. Aldegunde Rodrigez | Research Fellow
>
> Warwick Centre for Predictive Modelling | School of Engineering (D206)
>
> University of Warwick | Coventry | CV4 7AL
>
> Tel.: +44 (0)24 765 74564| E-mail: m.a.aldegunde-rodriguez at warwick.ac.uk
>
> Skype: manuel.aldegunde
>
>
>
> _______________________________________________
> gpaw-users mailing list
> gpaw-users at listserv.fysik.dtu.dk
> https://listserv.fysik.dtu.dk/mailman/listinfo/gpaw-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.fysik.dtu.dk/pipermail/gpaw-users/attachments/20160120/e8ac0e04/attachment-0001.html>