<?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.comsol.fr/community/forums/general/rss/thread/3748.rss" rel="self" type="application/rss+xml"/>
  <title>COMSOL Forums: Basic solving (3d)</title>
  <link>http://www.comsol.fr/community/forums/general/thread/3748/</link>
  <description>Most recent forum messages</description>
  <pubDate>Fri, 12 Mar 2010 06:23:43 +0000</pubDate>
  <image>
   <title>COMSOL Forums: Basic solving (3d)</title>
   <url>http://www.comsol.fr/shared/images/logos/comsol_logo.gif</url>
   <link>http://www.comsol.fr/community/forums/general/thread/3748/</link>
  </image>
  <item>
   <title>Re: Basic solving (3d)</title>
   <link>http://www.comsol.fr/community/forums/general/thread/3748/#p9781</link>
   <description>Hi &lt;br /&gt;&#13;
I'm not sure if I fully understand, you import a meshing from NASTRAN neutral file of a &amp;quot;hollow&amp;quot; tube&amp;quot; = the metal part I understand, and then you want to simualte the flow of some liquide going through it ?&lt;br /&gt;&#13;
&lt;br /&gt;&#13;
In which case yes you nees the volume of the liquide that is most probably the interiour ofthe tube, hence xou must define the entrance boundary(ies) and the exit boudary(ies) and ensure that COMSOl manages to create a closed domain for the NS sollving. On thes BC you can define your inlet and outlet conditions.&lt;br /&gt;&#13;
&lt;br /&gt;&#13;
A priori, if the nasrtan calualtion of the tube worke, COMSOl should not require any different mesh&lt;br /&gt;&#13;
&lt;br /&gt;&#13;
Now I'm not sure I fullky understood the question, in anycase hope it helps&lt;br /&gt;&#13;
Goodluck&lt;br /&gt;&#13;
Ivar</description>
   <pubDate>Fri, 12 Mar 2010 06:23:43 +0000</pubDate>
   <guid isPermaLink="false">3748.1268375023.9781</guid>
  </item>
  <item>
   <title>Basic solving (3d)</title>
   <link>http://www.comsol.fr/community/forums/general/thread/3748/#p9727</link>
   <description>I am inporting a Nas mesh file and that is a tube. There is an open inlet and outlet. I would like to just perform a simple navier stokes solve on this tube. However, I keep getting the error that their is no convergence.&lt;br /&gt;&#13;
&lt;br /&gt;&#13;
My questions are:&lt;br /&gt;&#13;
&lt;br /&gt;&#13;
1) Should my inlet and outlet be closed? or can they be left open? If they are open, then there is no face created, that can be defined as the inlet...there is online a solid wall thickness around the tube that can be defined as the inlet/outlet.&lt;br /&gt;&#13;
&lt;br /&gt;&#13;
2) Could my mesh size also be the problem?&lt;br /&gt;&#13;
&lt;br /&gt;&#13;
Thank you.&lt;br /&gt;&#13;
</description>
   <pubDate>Thu, 11 Mar 2010 18:17:44 +0000</pubDate>
   <guid isPermaLink="false">3748.1268331464.9727</guid>
  </item>
 </channel>
</rss>

