[Pw_forum] Help about vc-relax

[Paolo Giannozzi]

On Wednesday 23 November 2005 22:48, You Lin wrote:

> I'm trying to do a variable cell relaxation [...]
> The code stops at the second step [...]
> The error message gives no hint.

well, it gives a hint, if you know what it means (which is far
from obvious)...

>  from checkallsym : error #         2
>  not orthogonal operation

this means that the variable-cell relaxation has broken the
original symmetry of the unit cell. Since the number of k-points
depends on the symmetry of the cell, the run has to be stopped.
Possible solutions:
- start with a nonsymmetric cell
- use a symmetry-conserving algorithm: the Wentzcovitch algorithm
(cell_dynamics='damp-w') shouldn't break the symmetry

Paolo

-- 
Paolo Giannozzi             e-mail:  giannozz at nest.sns.it
Scuola Normale Superiore    Phone:   +39/050-509876, Fax:-563513 
Piazza dei Cavalieri 7      I-56126 Pisa, Italy