[CP2K-user] [CP2K:17586] Re: SGCPMD: number of steps per SCF run increase suddenly after initial 400 MD steps
Onofrio Tau
onofriotau at gmail.com
Thu Sep 1 13:44:07 UTC 2022
Thank you Marcella. I have reduced the timestep to 0.25 from 0.5 but the
number of convergence steps per SCF run is still too high. To point out
this issue, I attached below two following plots for sgcpMD runs with 0.25
of timestep:
1) The convergence value of the first OT step vs MD step. For example, the
bold value as follows:
***************************
Step Update method Time Convergence Total energy
Change
------------------------------------------------------------------------------
1 OT DIIS 0.10E-01 4.6 *0.00034360* -7197.3373211744
-7.20E+03
[...]
***************************
2) Number of steps per SCF run vs MD step.
I do not know why the convergence value of the first OT step suddenly
raises at certain MD steps, even with 0.25 of timestep.
Any suggestion?
Thanks a lot for your support.
Regards
Onofrio
--
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/e65f7b9c-29e4-4359-9054-0c8bd7a6ef98n%40googlegroups.com.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20220901/576649b3/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Number of steps per SCF run.png
Type: image/png
Size: 94066 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20220901/576649b3/attachment-0002.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Convergence Value of First OT step.png
Type: image/png
Size: 113526 bytes
Desc: not available
URL: <https://lists.cp2k.org/archives/cp2k-user/attachments/20220901/576649b3/attachment-0003.png>
More information about the CP2K-user
mailing list