<div style="line-height:1.7;color:#000000;font-size:14px;font-family:arial">Dear Abolore Musari:<br>In my opinion, lots of paper ran DFPT at 4x4x4 q-mesh. Sometimes (not often), 3x3x3 q-mesh may also provide reasonable results.<br>The easiest way may be similar as k-mesh test, you can gradually increase the q-mesh from 2x2x2 to NxNxN until the frequencies converged (the threshold can be defined by yourself).<br><div>--<br>GAO Zhe<br>CMC Lab, Materials Science &amp; Engineering Department,<br>Seoul National University, South Korea<br>        
&nbsp;</div><div id="divNeteaseMailCard"></div><br><pre><br>At&nbsp;2012-08-12&nbsp;12:47:12,"Abolore&nbsp;Musari"&nbsp;&lt;abmus007@gmail.com&gt;&nbsp;wrote:
&gt;Dear&nbsp;sir.
&gt;Sir&nbsp;l&nbsp;just&nbsp;want&nbsp;to&nbsp;ask&nbsp;how&nbsp;do&nbsp;l&nbsp;know&nbsp;the&nbsp;proper&nbsp;nq1&nbsp;nq2&nbsp;nq3&nbsp;to&nbsp;use&nbsp;in
&gt;my&nbsp;phonon&nbsp;calculation&nbsp;after&nbsp;my&nbsp;scf&nbsp;calculation.
&gt;Am&nbsp;sorry&nbsp;l&nbsp;ask&nbsp;this&nbsp;type&nbsp;of&nbsp;question&nbsp;l&nbsp;will&nbsp;be&nbsp;so&nbsp;happy&nbsp;if&nbsp;u&nbsp;can&nbsp;kindly&nbsp;explain
&gt;thanks
&gt;Musari.&nbsp;A.A
&gt;UNAAB&nbsp;Nigeria
&gt;_______________________________________________
&gt;Pw_forum&nbsp;mailing&nbsp;list
&gt;Pw_forum@pwscf.org
&gt;http://www.democritos.it/mailman/listinfo/pw_forum
</pre></div><br><br><span title="neteasefooter"><span id="netease_mail_footer"></span></span>