Dear Eyvaz<br><br>Sorry bout that....yes I meant K_POINTS, is there any thumb rule for deciding how many K-points are good for convergence. Is it dependent on the nature of the system, or on the number of atoms or something else<br><br>Pardon me for my newbie like questions...<br><br>Vidur<br><br><b><i>Eyvaz Isaev <eyvaz_isaev@yahoo.com></i></b> wrote:<blockquote class="replbq" style="border-left: 2px solid rgb(16, 16, 255); margin-left: 5px; padding-left: 5px;"> <br>--- Vidur Parkash <vparkash@mtu.edu> wrote:<br><br>> Thank you everybody for finding a very critical bug<br>> in my setup, yes The distances are actually double<br>> of what they're supposed to be<br>> <br>> Will try to fix that!...another thing ....how does<br>> changing the number of gamma points help the system<br>> to converge?...and is there any thumb rule for<br>> choosing the number of gamma points? <br><br>Can you please explain what do you mean "gamma<br>points"?<br>If
you mean K_POINTS, then their number will not help<br>if atomic positions are wrong (usually poor<br>convergency).<br><br>Bests,<br>Eyvaz. <br> <br><br>> <br>> Vidur<br>> <br>> <br>> <br>> <br>> Stefano de Gironcoli <degironc@sissa.it> wrote: This<br>> may not be totally relevant but ... have you checked<br>> that the <br>> configuration you are using makes sense ?<br>> By visualizing it with xcrysden it appear that the<br>> carbon-carbon <br>> distances are all larger than 3 angstrom ... so<br>> your system is a <br>> collection of weakly interacting atoms, probably<br>> with very narrow bands. <br>> Technically this may be the origin of the<br>> instability, but ...is this <br>> the system you want to study ?<br>> Hpe it helps<br>> stefano<br>> Stefano de Gironcoli - SISSA and DEMOCRITOS<br>> <br>> <br>> <br>> Vidur Parkash wrote:<br>> > Hi<br>> > <br>> > I'm trying to
perform a SCf calculation on a<br>> carbon nanotube cluster<br>> > (metallic system), and no matter what i do the<br>> results do not<br>> > converge.<br>> > I've tried changing the mixing factor, add a<br>> smearing, tried a<br>> > different type of charge mixing, change number of<br>> k points etc.<br>> ><br>> > The input file I've attached is the closest it<br>> ever reached towards<br>> > convergence with accuracy better than 0.1 Ry...but<br>> ultimately failing<br>> > to converge. <br>> > <br>> > <br>> > Can you please suggest some solutions. The input<br>> file is attached<br>> > this time.<br>> > <br>> > Thanks<br>> > <br>> > Vidur Parkash<br>> ><br>> > Michigan Tech University<br>> > 716 EERC, 1400 Townsend Drive<br>> > Houghton MI-49931, USA<br>> > Tel: 1-906-487-0023<br>> ><br>>
><br>><br>------------------------------------------------------------------------<br>> ><br>> > _______________________________________________<br>> > Pw_forum mailing list<br>> > Pw_forum@pwscf.org<br>> > http://www.democritos.it/mailman/listinfo/pw_forum<br>> > <br>> <br>> <br>> <br>> Vidur Parkash<br>> <br>> Michigan Technological University<br>> 716 EERC, 1400 Townsend Drive<br>> Houghton, MI-49931, USA<br>> Tel: 1-906-487-0023<br>> > _______________________________________________<br>> Pw_forum mailing list<br>> Pw_forum@pwscf.org<br>> http://www.democritos.it/mailman/listinfo/pw_forum<br>> <br><br><br>-------------------------------------------------------------------<br>Prof. Eyvaz Isaev, <br>Theoretical Physics Department, Moscow State Institute of Steel & Alloys, Russia, and <br>Condensed Matter Theory Group, Uppsala University, Sweden Eyvaz.Isaev@fysik.uu.se,
eyvaz_isaev@yahoo.com<br><br><br> <br>____________________________________________________________________________________<br>Need a vacation? Get great deals<br>to amazing places on Yahoo! Travel.<br>http://travel.yahoo.com/<br></degironc@sissa.it></vparkash@mtu.edu></blockquote><br><BR><BR>Vidur Parkash<br><br>Michigan Technological University<br>716 EERC, 1400 Townsend Drive<br>Houghton, MI-49931, USA<br>Tel: 1-906-487-0023<br>