[Pw_forum] convergence issue

Ranasinghe, Jayangani jir520 at mail.usask.ca
Sat Aug 26 00:30:19 CEST 2017


part of INPUT in the scf .in is below


&SYSTEM
   nat=162,
  ntyp = 2,
    degauss = 0.02,
    nspin = 2,
    ecutwfc = 100,
    ibrav = 0,
    nbnd=1000,
    occupations = 'smearing',
    smearing = 'mp',
    starting_magnetization(1)=0.5,
    starting_magnetization(2)=0.0,
 /
&ELECTRONS
    conv_thr = 1e-10,
    mixing_beta = 0.3,
    electron_maxstep = 900,
    mixing_mode='local-TF',
    mixing_ndim=10,




Thank you


Jayangani


________________________________
From: pw_forum-bounces at pwscf.org <pw_forum-bounces at pwscf.org> on behalf of Ranasinghe, Jayangani <jir520 at mail.usask.ca>
Sent: Friday, August 25, 2017 4:26 PM
To: pw_forum at pwscf.org
Subject: [Pw_forum] convergence issue


Dear QE developers and experts


I am trying to do third order force constant calculation in a 3x3x3 supercell.  I have 190 input  files and I am continuing the scf runs for each. I have set the convergence criteria 1.0E-10. However SCF run doesn't stop after this is achieved. As copied below ethr went to 10^-11 at 108 th step and decreased 10^-14 by 900th step. But didn't stop. My problem is why this is happening and how to control the convergence criteria to meet the given threshold.?



I played with mixing_mode, mixing_ndim etc... but it didn't help




Thank you


Jayangani

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.quantum-espresso.org/pipermail/users/attachments/20170825/598870a4/attachment.html>


More information about the users mailing list