[CP2K:8184] Separating DFT-D3

[Ondrej Marsalek]

Hi Juerg,

thank you for the suggestion. Making all kinds GHOST was one of the
ways I tried to disable the electronic part of the calculation
completely (not realizing it would disable dispersion as well,
although that makes sense). Unfortunately, doing it for all kinds does
not work. Encouragingly, after 1 SCF iteration (0 does not work), all
the electronic energies are zero, but then I get the first error
below. If I further set the basis sets to NONE, I get the second error
below (even before the SCF). I think most of the code is not ready to
not have anything to do at all. I will try to have a look at bypassing
the electronic calculation entirely (perhaps in qs_forces?), but I am
not sure how large an edit that would be, do you think that's
reasonable?

For reference, I am doing this with a REFTRAJ run.

Ondrej


 *****************************************************************************
 *** 10:17:39 ERRORL2 in                                                   ***
 *** qs_integrate_potential_single:integrate_v_core_rspace processor 0  :: ***
 *** err=-300 condition FAILED at line 849                                 ***
 *****************************************************************************

 ===== Routine Calling Stack =====

            8 integrate_v_core_rspace
            7 qs_ks_build_kohn_sham_matrix
            6 rebuild_ks_matrix
            5 qs_ks_update_qs_env_forces
            4 qs_forces
            3 velocity_verlet
            2 qs_mol_dyn_low
            1 CP2K
 CP2K| condition FAILED at line 849
 CP2K| Abnormal program termination, stopped by process number 0


 ************************************************************************
 *** ERROR in get_neighbor_list_set_p (MODULE qs_neighbor_list_types) ***
 ************************************************************************

 *** No neighbor list set is associated ***

 *** Program stopped at line number 1039 of MODULE qs_neighbor_list_types ***

 ===== Routine Calling Stack =====

            8 build_overlap_matrix
            7 build_core_hamiltonian_matrix
            6 qs_energies_init_hamiltonians
            5 qs_energies_scf
            4 qs_forces
            3 velocity_verlet
            2 qs_mol_dyn_low
            1 CP2K
 CP2K| Abnormal program termination, stopped by process number 0


On Thu, Sep 22, 2016 at 6:43 AM,  <hut... at chem.uzh.ch> wrote:
> Hi
>
> this is a solution with a small code hack.
>
> Add "GHOST" to all your &KIND sections and change the file
> qs_dispersion_pairpot.F
>
> at about line 970
>          dodisp(ikind) = disp_a%defined
>          ghost(ikind) = .FALSE.           !disable ghosts
>          floating(ikind) = floating_a
>
> should work now.
>
> regards
>
> Juerg
>
>
> --------------------------------------------------------------
> Juerg Hutter                         Phone : ++41 44 635 4491
> Institut für Chemie C                FAX   : ++41 44 635 6838
> Universität Zürich                   E-mail: hut... at chem.uzh.ch
> Winterthurerstrasse 190
> CH-8057 Zürich, Switzerland
> ---------------------------------------------------------------
>
> -----cp... at googlegroups.com wrote: -----To: cp... at googlegroups.com
> From: Ondrej Marsalek
> Sent by: cp... at googlegroups.com
> Date: 09/22/2016 01:30AM
> Subject: [CP2K:8179] Separating DFT-D3
>
> Dear all,
>
> I would like to get the forces, stress, and energy from the DFT-D3
> dispersion correction separately, in the normal output files of CP2K.
> I know this is already possible with the verbose output from the DFT-D
> code, but that is not useful for me in this particular case. Is that
> possible? My attempts to disable the electronic part of the
> calculation in Quickstep have so far not been successful.
>
> Thank you,
> Ondrej
>
> --
> You received this message because you are subscribed to the Google Groups "cp2k" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns... at googlegroups.com.
> To post to this group, send email to cp... at googlegroups.com.
> Visit this group at https://groups.google.com/group/cp2k.
> For more options, visit https://groups.google.com/d/optout.
>
> --
> You received this message because you are subscribed to the Google Groups "cp2k" group.
> To unsubscribe from this group and stop receiving emails from it, send an email to cp2k+uns... at googlegroups.com.
> To post to this group, send email to cp... at googlegroups.com.
> Visit this group at https://groups.google.com/group/cp2k.
> For more options, visit https://groups.google.com/d/optout.