CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   Simple question (https://www.cfd-online.com/Forums/main/6351-simple-question.html)

reader July 14, 2003 09:59

Simple question
 
<html> <meta name="GENERATOR" content="TtH 3.37"> <style type="text/css"> div.p { margin-top: 7pt;}</style> <style type="text/css"> </style>

Consider the equation in a cube <br clear="all" /><table border="0" width="100%"><tr><td> <table align="center" cellspacing="0" cellpadding="0"><tr><td nowrap="nowrap" align="center"> u<sub>t</sub>=u<sub>xx</sub>+u<sub>yy</sub>+u<sub>zz</sub></td></tr></table> </td></tr></table> I replace derivatives with differences, e.g., <br clear="all" /><table border="0" width="100%"><tr><td> <table align="center" cellspacing="0" cellpadding="0"><tr><td nowrap="nowrap" align="center"> u<sub>xx</sub>=</td><td nowrap="nowrap" align="center"> *1 <div class="hrcomp"><hr noshade="noshade" size="1"/></div>h<sup>2</sup>
</td><td nowrap="nowrap" align="center"> (u<sub>i<font face="symbol">-</font> 1,jk</sub><font face="symbol">-</font> 2u<sub>ijk</sub>+u<sub>i+1jk</sub>)</td></tr></table> </td></tr></table> and then use explicit Runge-Kutta method with automatic choice of timestep. On a grid 100 x 100 x 100 all is OK. But if I take a grid 500 x 500 x 500 the timestep becomes too small since the stiffness raises. How to overcome this? Thank.

</html>

cfd dude July 14, 2003 13:29

Simple answer
 
Use an implicit method.


All times are GMT -4. The time now is 06:45.