<div>After many time, finally I ran some simulations (cluster problems), the 'jumps' in the energy convergence remain, but reducing the k-points and cutoffs, was more fast.</div><div><br></div><div>The more fast convergence was (and without many jumps) : </div>
<div><br></div><div>-increasing the mixing_beta from 0.1 to 0.3</div><div>-NO use local-TF in the mixing_mode</div><div>-NO change the mixing_ndim</div><div>-Reduce the vaccum.</div><div>-Reduce cutoff energy.</div><div>
<br>
</div><div>When I only change the mixing_mode and mixing_beta from this previous data (even reduce from 0.1 to 0.09) the convergence problem remain and increase. I hope that with this, the problem disappear.</div><div><br>
</div><div>Probably always was the vaccum, and the default values (except mixing_beta) was fine.</div><div><br></div><div><blockquote class="gmail_quote" style="margin-top: 0px; margin-right: 0px; margin-bottom: 0px; margin-left: 0.8ex; border-left-width: 1px; border-left-color: rgb(204, 204, 204); border-left-style: solid; padding-left: 1ex; ">
After how many SCF cycles these "fluctuations" appear?</blockquote><div><br></div><div>Usually each 10 steps. Related with 0.1 of mixing_beta??</div><div><br></div><div><div>And thank you very much for all the answers, and any other suggestion is welcome.</div>
</div><div><br></div><div>Best Regards</div><div><br></div><div>Joaquin </div><div><br><div class="gmail_quote">On Fri, Sep 3, 2010 at 4:05 AM, Tone Kokalj <span dir="ltr"><<a href="mailto:tone.kokalj@ijs.si">tone.kokalj@ijs.si</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;"><div class="im">On Thu, 2010-09-02 at 08:38 -0500, joaquin peralta wrote:<br>
> Thanks, i obtain more fast convergence for 1V/A, but ...<br>
><br>
> If I use the previous results from 1V/A in order to calculate an<br>
> increase in the Field i obtain fluctuation in the energy.<br>
><br>
> total energy = -261.46819272 Ry<br>
> total energy = -261.46815798 Ry<br>
> total energy = -261.46816720 Ry<br>
> total energy = -261.46260676 Ry<br>
> total energy = -261.37365516 Ry<br>
> total energy = -261.46820680 Ry<br>
> total energy = -261.46821935 Ry<br>
> total energy = -261.46807352 Ry<br>
> total energy = -261.46825528 Ry<br>
> total energy = -261.46820900 Ry<br>
> total energy = -261.46803348 Ry<br>
><br>
> for my case, is recommendable reduce the precision by default of the<br>
> energy convergence from 1D-4 to 1D-3?<br>
<br>
</div>To what energy threshold are you referring: conv_thr (in &electrons) or<br>
etot_conv_thr (in &control). According to the value of 1d-3/1d-4 I would<br>
assume the latter; note however that the latter has nothing to do with<br>
the SCF convergence, as it is an optimization threshold criteria for<br>
geometry optimizer.<br>
<br>
If you use 1d-4 for the conv_thr, then the value is much too high, it<br>
should be 1d-6 or, even better, lower than that.<br>
<br>
After how many SCF cycles these "fluctuations" appear?<br>
<br>
You may try to reduce mixing_beta to 5%, ...<br>
<br>
The following considerations will speed-up your calculations, hence you<br>
will be able to do more SCF convergence tests: from the previous emails<br>
of yours, I notice that you use over 50 Angs of vacuum, which is HUGE (I<br>
would assume that about 25 Angs should be sufficient; in this case there<br>
will be more than 10 Angs spacing from the slab to the dipole-layer). I<br>
also noticed that ecutwfc/ecutrho of 30/500 ry is a very large number<br>
for Aluminum.<br>
<div><div></div><div class="h5"><br>
Regards, Tone<br>
--<br>
Anton Kokalj<br>
J. Stefan Institute, Jamova 39, 1000 Ljubljana, Slovenia<br>
(tel: +386-1-477-3523 // fax:+386-1-477-3822)<br>
<br>
Please, if possible, avoid sending me Word or PowerPoint attachments.<br>
See: <a href="http://www.gnu.org/philosophy/no-word-attachments.html" target="_blank">http://www.gnu.org/philosophy/no-word-attachments.html</a><br>
<br>
_______________________________________________<br>
Pw_forum mailing list<br>
<a href="mailto:Pw_forum@pwscf.org">Pw_forum@pwscf.org</a><br>
<a href="http://www.democritos.it/mailman/listinfo/pw_forum" target="_blank">http://www.democritos.it/mailman/listinfo/pw_forum</a><br>
</div></div></blockquote></div><br><br clear="all"><br>-- <br>----------------------------------------------------<br>Group of NanoMaterials<br>----------------------------------------------------<br><a href="http://www.gnm.cl" target="_blank">http://www.gnm.cl</a><br>
----------------------------------------------------<br>Joaquín Andrés Peralta Camposano<br>----------------------------------------------------<br><a href="http://www.lpmd.cl/jperalta" target="_blank">http://www.lpmd.cl/jperalta</a><br>
<br>In a world without frontiers, <br>who needs Gates and Win.<br>
</div></div>