[CP2K-user] [CP2K:19937] help: Full_kinetic not available for semi-empirical methods

[Jürg Hutter]

Hi

I trust that you have checked your input geometry and box sizes and that
this is not the cause of the convergences problems.

Why are you using LSD for a system with even number of electrons? Do you
expect a complicated spin state?

The main source of problems is most likely the electrostatic terms in PBC with the
PM6 Hamiltonian. There are two options 1) a cutoff and long range correction and
2) a multipole Ewald method. You should only choose one, don't mix them.
This code is rather old and is not used often (most people use TB for such systems). The algorithms need a careful selection of rather many parameters.
I would suggest to start testing with a smaller, simpler system and find a set
of parameters that you can trust.

regards
JH

________________________________________
From: cp2k at googlegroups.com <cp2k at googlegroups.com> on behalf of Alex Lyu <stefani.alexandria at gmail.com>
Sent: Tuesday, February 20, 2024 7:26 PM
To: cp2k
Subject: Re: [CP2K:19935] help: Full_kinetic not available for semi-empirical methods

Hi JH,

Thank you, I have just noticed your reply. Since then I have started PM6 semi empirical method but the SCF is not converging... not sure what I can change?
I have tried changing the minimiser, preconditioner, extrapolation, scf_guess etc but nothing is working right now.
On Monday, January 22, 2024 at 1:08:51 PM UTC Jürg Hutter wrote:
Hi

without further information this question cannot be answered.

regards
JH

________________________________________
From: cp... at googlegroups.com <cp... at googlegroups.com> on behalf of Alexandria Setiono <stefani.a... at gmail.com>
Sent: Wednesday, January 17, 2024 11:39 PM
To: cp2k
Subject: Re: [CP2K:19770] help: Full_kinetic not available for semi-empirical methods

Hi JH,
thank you for the reply! I managed to get it working, however my SE calculations and my DFT calculations takes about the same time for the same structure. I was wondering how to fix this, perhaps maybe some sources that would help? Because I was under the impression that SE is order of magnitudes faster.

Thank you so much once again!
Stefani
On Thursday, January 4, 2024 at 9:36:42 AM UTC Jürg Hutter wrote:
Hi

you need to choose another preconditioner for the OT method.
See
FORCE_EVAL/DFT/SCF/OT/PRECONDITIONER
for options. I would suggest to use
FULL_SINGLE_INVERSE

regards
JH

________________________________________
From: cp... at googlegroups.com <cp... at googlegroups.com> on behalf of Alexandria Setiono <stefani.a... at gmail.com>
Sent: Wednesday, January 3, 2024 12:01 AM
To: cp2k
Subject: [CP2K:19693] help: Full_kinetic not available for semi-empirical methods

Dear CP2K community, I am a first year PhD Student new to CP2K. I am trying to use the Semi Empirical methods to get MD trajectories, however, I was returned with the following error:
'Full_kinetic not available for semi-empirical methods'

I am wondering where I can find further informations about semi-empirical methods on cp2k or a way to fix this error. Any help is welcomed! :)
thank you so much in advance!

Many thanks,
Stefani

--
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<mailto:cp2k+uns... at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/0623481b-e2c9-41af-999b-80beb0df898dn%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/0623481b-e2c9-41af-999b-80beb0df898dn%40googlegroups.com?utm_medium=email&utm_source=footer>.

--
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<mailto:cp2k+uns... at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/74241591-0cc1-4e79-a081-7e1ad5c799f5n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/74241591-0cc1-4e79-a081-7e1ad5c799f5n%40googlegroups.com?utm_medium=email&utm_source=footer>.

--
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+unsubscribe at googlegroups.com<mailto:cp2k+unsubscribe at googlegroups.com>.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/3910ad64-8e85-44c3-a35e-1394695656d3n%40googlegroups.com<https://groups.google.com/d/msgid/cp2k/3910ad64-8e85-44c3-a35e-1394695656d3n%40googlegroups.com?utm_medium=email&utm_source=footer>.

-- 
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+unsubscribe at googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/cp2k/ZR0P278MB0759E5D151F4C98BF42D2B5D9F572%40ZR0P278MB0759.CHEP278.PROD.OUTLOOK.COM.