<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
 <channel>
  <atom:link href="http://www.ch.comsol.com/community/forums/general/rss/thread/3886.rss" rel="self" type="application/rss+xml"/>
  <title>COMSOL Forums: what kind of mesh method do i need to choose?</title>
  <link>http://www.ch.comsol.com/community/forums/general/thread/3886/</link>
  <description>Most recent forum messages</description>
  <pubDate>Wed, 17 Mar 2010 12:39:05 +0000</pubDate>
  <image>
   <title>COMSOL Forums: what kind of mesh method do i need to choose?</title>
   <url>http://www.ch.comsol.com/shared/images/logos/comsol_logo.gif</url>
   <link>http://www.ch.comsol.com/community/forums/general/thread/3886/</link>
  </image>
  <item>
   <title>Re: what kind of mesh method do i need to choose?</title>
   <link>http://www.ch.comsol.com/community/forums/general/thread/3886/#p10234</link>
   <description>So cool. Thanks very much. I will try that!</description>
   <pubDate>Wed, 17 Mar 2010 12:39:05 +0000</pubDate>
   <guid isPermaLink="false">3886.1268829545.10234</guid>
  </item>
  <item>
   <title>Re: what kind of mesh method do i need to choose?</title>
   <link>http://www.ch.comsol.com/community/forums/general/thread/3886/#p10162</link>
   <description>Hi&lt;br /&gt;&#13;
&lt;br /&gt;&#13;
There are no easy reply, in a general sens, but ONE important feature of the mesh is that it MUST represent correctly the gradients, or deformations of your model. Therefore one often check the mesh density by &amp;quot;refining&amp;quot; once and resolving, just to ensure hat you have less then 5-10% changes in results. To much changes indicte a too coarse mesh, too small changes might indicate a very fine mesh. Now there is no danger in making the mesh too fine, its only that you might hit a RAM limit, or simply it wuld solve very slowly, just as frustrating.&lt;br /&gt;&#13;
&lt;br /&gt;&#13;
For your case of a hollow cylinder (with or without a closed end) I would use the symmetry, and sketch it in a 2D-axi geoemtry, an then revolve mesh it into a 3D shape (with tri and/or quads) this gives you a regular mesh. You can then compare the results with a 3D free-teth mesh&lt;br /&gt;&#13;
&lt;br /&gt;&#13;
Good luck&lt;br /&gt;&#13;
Ivar</description>
   <pubDate>Wed, 17 Mar 2010 07:13:09 +0000</pubDate>
   <guid isPermaLink="false">3886.1268809989.10162</guid>
  </item>
  <item>
   <title>what kind of mesh method do i need to choose?</title>
   <link>http://www.ch.comsol.com/community/forums/general/thread/3886/#p10120</link>
   <description>Hi,&lt;br /&gt;&#13;
i'm modeling a hollow cylinder with one end closed and want to see the defomation under eignfrequencies. During the process of mesh, i dont know how to mesh the cylinder, i mean, is there any standard or requirment for hollow cylinder under eignfrequency analysis? &lt;br /&gt;&#13;
&lt;br /&gt;&#13;
I choose initialized mesh method and another mesh method desided by myself, the eignfrequencies from the two methods are different. Which way is the proper way to mesh my model?&lt;br /&gt;&#13;
&lt;br /&gt;&#13;
Thanks!</description>
   <pubDate>Tue, 16 Mar 2010 16:11:12 +0000</pubDate>
   <guid isPermaLink="false">3886.1268755872.10120</guid>
  </item>
 </channel>
</rss>

