[DFTB-Plus-User] Failure in diagonalisation routine zhegvd

[Bálint Aradi]

Dear Juita,

On 12/05/2013 12:41 AM, . Juita wrote:
> Dear Users,
> 
> I am the new user of DFTB+.  I tried the molecular dynamics simulations
> with the package, however, they stopped with error message "Failure in
> diagonalisation routine zhegvd, non-positive definite overlap!
> Minor      1 responsible".  The attached is the input file for the run.

If the error happens during the MD between two SCC iterations (so not
right at the beginning of a geometry step), then it is a problem with
the LAPACK-diagonalization routine we use per default. You can then try
the relatively robust instead by setting

Eigensolver = RelativelyRobust {}

in the Hamiltonian block.

If the error happens at the very beginning of a geometry step (1st SCC
cycle) then probably two atoms are too close to each other.

  Best regards,

    Bálint


-- 
Dr. Bálint Aradi
Bremen Center for Computational Materials Science, University of Bremen
http://www.bccms.uni-bremen.de/cms/people/b_aradi/


-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 263 bytes
Desc: OpenPGP digital signature
URL: <http://www.dftb-plus.info/pipermail/dftb-plus-user/attachments/20131205/8f209a3a/attachment.sig>