<table cellspacing="0" cellpadding="0" border="0" ><tr><td valign="top" style="font: inherit;"><div class="plainMail"><a href="http://www.democritos.it/mailman/listinfo/pw_forum" target="_blank"></a>Hi Paolo<br><span style="font-weight: bold;">by far the most common reason for bad scf convergence is </span><br style="font-weight: bold;"><span style="font-weight: bold;">incorrect input data. This is the first thing to verify.</span><br><span style="font-weight: bold;"><br></span>Of course, I use Xcrysden or Vista to check everything, bond lenght, bond angle, atomic coordinates before doing anything further. I think there are two reasons for the slow convergence in my case related to two following problems:<br>1. I have not done structural optimization first.<br>2. My generated pseudopotentials by ld1.x is not good. I have to spent more time on this one. Like you mentioned before, generate a pseupotentials is easy but to obtain a good one, it is not
simple. I am still learning.<br>Anyways, thanks for reply.<br>Sau Nguyen<br>U of Houston<span style="font-weight: bold;"><span style="font-weight: bold;"><br></span></span></div></td></tr></table><br>