Hi Emilio,
In which step does that happen? Is it in the first MD step, or the calculation runs for a while before crashing?
(You might want to update to Amber14, and make sure you apply all updates. A recent update to Amber 14 was related exactly to handling new DFTB SLK files.)
Gustavo Seabra
> Em 01/10/2015, à(s) 09:47, Emilio Lence <guitarro.de.meixente.gmail.com> escreveu:
>
> Dear amber users
> Recently we downloaded the latest version of the 3ob set for use used it to
> run calculations at DFTB level with a Mg atom in the QM region, in order to
> compare the previously obtained results using AM1/d. We tried different
> starting geometries and 2 different proteins, but in all cases the result
> is the same, we have a problem in the convergence, related with the
> Eigenvalue solver of the form:
>
> QMMM SCC-DFTB: ***************************************************
> QMMM SCC-DFTB: ERROR ON EWEVGE (Eigenvalue solver).
> QMMM SCC-DFTB: ewevge: ier = 307 inner_scf_count= 1
> QMMM SCC-DFTB: ***************************************************
>
> QMMM SCC-DFTB: !!!! ============= WARNING ============= !!!!
> QMMM SCC-DFTB: Convergence could not be achieved in this step.
> QMMM SCC-DFTB: The calculation will continue, but energies and
> QMMM SCC-DFTB: forces for this step will not be accurate.
>
> After a search in the list, we came with a few options that we discarded.
> Some pointed to a problem in the system, other told to change the
> dftb_telec flag. The later was to no use, the former does not seem to be a
> the problem with the structure. Either way, all the search results, were
> older than the new data set 3ob-3-1 released.
> As well, other processes with the a QM region with only H, C, N and O atoms
> seem to be running OK. When I take of the QM region the Mg and the ATP, the
> calculations run without that problem, but with either of them shows the
> same problem. It seems that something happen when it has to use the slk
> files for the P or the Mg atom.
> Thanks for your kind attention
> Best Wishes
>
> Dr. Emilio Lence Quintana
> Organic Chemistry Department-CIQUS
> Univesity of Santiago de Compostela
> 15782 Santiago de Compostela, Spain
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Oct 01 2015 - 07:00:03 PDT
