CUTOFF and REL_CUTOFF

234 views
Skip to first unread message

Pierre-André Cazade

unread,
Sep 12, 2019, 9:24:43 AM9/12/19
to cp2k
Dear CP2K users,

Following the tutorial to tune the CUTOFF and the REL_CUTOFF for two of my systems, I find that quite large value of the CUTOFF  (700 Ry for beta glycine and 900Ry for quercetin) are necessary to achieve a decent level of precision (about 1.e-6). I am a bit surprised considering my systems are made from C, N, O, H. On top of that, for beta glycine, the last grid is not used that much, unlike what is recommended in the tutorial. Therefore, I have a few questions:

1/ Are these large cutoffs normal, even for small atoms? Is the cutoff more dependent on the crystal structure than the atom type?
2/ Is it normal to have still a large number of gaussians distributed on the first grid? Should I always aim for having a larger number of gaussians on the coarsest grid than on the thinnest?
3/ How do one decide on the number of grids? Is 4 kind of a magic number or should one aim to tune this as well? How does it affect accuracy/performance?

I attached the input file I used to generate the series of CUTOFF and REL_CUTOFF for both systems as well as the results I obtained.

Thank you for your help.

Best regards,
Pierre  
quercetin_cutoff.inp
quercetin_cutoff.dat
bglycine_cutoff.dat
bglycine_cutoff.inp

hut...@chem.uzh.ch

unread,
Sep 13, 2019, 8:39:22 AM9/13/19
to cp...@googlegroups.com
Hi

1) the cutoff depends on the pseudopotential, F will need much
higher cutoff than Se.
2) Defaults are okay (logarithmic)
3) For high cutoffs, I usually use 5 or 6 grids

regards

Juerg Hutter
--------------------------------------------------------------
Juerg Hutter Phone : ++41 44 635 4491
Institut für Chemie C FAX : ++41 44 635 6838
Universität Zürich E-mail: hut...@chem.uzh.ch
Winterthurerstrasse 190
CH-8057 Zürich, Switzerland
---------------------------------------------------------------

-----cp...@googlegroups.com wrote: -----
To: "cp2k" <cp...@googlegroups.com>
From: "Pierre-André Cazade"
Sent by: cp...@googlegroups.com
Date: 09/12/2019 03:24PM
Subject: [CP2K:12218] CUTOFF and REL_CUTOFF
--
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...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/0f9bfb64-3184-4098-a87a-c5cdfbed512f%40googlegroups.com.


[attachment "quercetin_cutoff.inp" removed by Jürg Hutter/at/UZH]
[attachment "quercetin_cutoff.dat" removed by Jürg Hutter/at/UZH]
[attachment "bglycine_cutoff.dat" removed by Jürg Hutter/at/UZH]
[attachment "bglycine_cutoff.inp" removed by Jürg Hutter/at/UZH]

Pierre Cazade

unread,
Sep 13, 2019, 9:16:48 AM9/13/19
to cp...@googlegroups.com
Hi Juerg,

Many thanks.

Regards,
Pierre
--
Dr Pierre Cazade, PhD
AD3-023, Bernal Institute,
University of Limerick,
Plassey Park Road,
Castletroy, co. Limerick,
Ireland

Reply all
Reply to author
Forward
0 new messages