<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">
<TITLE>2d DC Conduction</TITLE>
<META content="MSHTML 6.00.2800.1106" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN
class=033515123-01032006>Hello:</SPAN></FONT></DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN
class=033515123-01032006></SPAN></FONT> </DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN
class=033515123-01032006></SPAN></FONT> </DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN class=033515123-01032006>I have
found my error. A simple typographical error where the "NameOfConstraint"
did not match the</SPAN></FONT></DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN class=033515123-01032006>name
in the Constraint list. I had checked this, I thought. This was also
the source of the </SPAN></FONT></DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN
class=033515123-01032006></SPAN></FONT> </DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN class=033515123-01032006>"<FONT
color=#000000>Warning: Null right Hand Side in Linear Equation<FONT
face="Times New Roman" size=3> " message</FONT></FONT></SPAN></FONT></DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN
class=033515123-01032006></SPAN></FONT> </DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN class=033515123-01032006>I have
modified my simpleBar.pro file (fixed a couple of other things that I had
meant</SPAN></FONT></DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN class=033515123-01032006>to
reset before sending the file out previously) and have attached it just in case
anyone reads this</SPAN></FONT></DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN class=033515123-01032006>in the
archives and wants to see the resolution.</SPAN></FONT></DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN
class=033515123-01032006></SPAN></FONT> </DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN class=033515123-01032006>Sorry
for the wasted bandwidth</SPAN></FONT></DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN
class=033515123-01032006></SPAN></FONT> </DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN class=033515123-01032006>Laird
Bolt</SPAN></FONT></DIV>
<DIV><FONT face=Arial color=#0000ff size=2><SPAN
class=033515123-01032006></SPAN></FONT> </DIV>
<DIV><FONT face=Tahoma size=2>-----Original Message-----<BR><B>From:</B>
getdp-bounces@geuz.org [mailto:getdp-bounces@geuz.org]<B>On Behalf Of </B>Laird
Bolt<BR><B>Sent:</B> Tuesday, February 28, 2006 5:13 PM<BR><B>To:</B>
getdp@geuz.org<BR><B>Subject:</B> [Getdp] 2d DC Conduction<BR><BR></FONT></DIV>
<BLOCKQUOTE dir=ltr style="MARGIN-RIGHT: 0px"><!-- Converted from text/rtf format -->
<P><FONT face=Arial size=2>I apologise in advance for sending such a
newbie question to the list. In Payment if someone could</FONT>
<BR><FONT face=Arial size=2>set me on my way I will add my simple example(s)
to the main example list.</FONT> </P>
<P><FONT face=Arial size=2>In an effort to walk (more of a crawl) before I run
I am trying to simply</FONT> <BR><FONT face=Arial size=2>calculate the voltage
and current density distribution in a flat bar of copper (no temperature
effects assumed)</FONT> </P>
<P><FONT face=Arial size=2>Fragments for this exist in the mail archives as
well as documentation. However,</FONT> <BR><FONT face=Arial size=2>I
must be doing something wrong as my results are always zero when using
Dirichlet BC at both ends</FONT> <BR><FONT face=Arial size=2>and constant (but
non-zero) when I use dirichlet at one end and Neumann BC at the other.</FONT>
</P>
<P><FONT face=Arial size=2>I assume that the null Neumann BC along the long
sides is implicit.</FONT> </P>
<P><FONT face=Arial size=2>I have attached my .geo, .pro and solver.par files
for the Dirichlet BC (1V at one end 0v at the other)</FONT> </P>
<P><FONT face=Arial size=2>THe Formulation was :</FONT> </P>
<P><FONT face=Arial size=2>Formulation {</FONT> <BR><FONT face=Arial
size=2> { Name Voltage_Steady ; Type FemEquation ;</FONT> <BR><FONT
face=Arial size=2> Quantity { </FONT><BR><FONT face=Arial
size=2> { Name Volt ; Type Local ; NameOfSpace
Vgrad ; }</FONT> <BR><FONT face=Arial size=2> }</FONT>
<BR><FONT face=Arial size=2> Equation {</FONT> <BR><FONT
face=Arial size=2> Galerkin { [ invrho[] *
Dof {d Volt} , {d Volt} ] ; </FONT><BR><FONT face=Arial
size=2>
In Bar ; Jacobian JVol ; Integration GradGrad ; }</FONT> <BR><FONT face=Arial
size=2> }</FONT> <BR><FONT face=Arial size=2> }</FONT>
<BR><FONT face=Arial size=2>}</FONT> </P>
<P><FONT face=Arial size=2>When I run the Dirichlet at both ends I receive a
Warning: Null right Hand Side in Linear Equation</FONT> <BR><FONT face=Arial
size=2>which I assume is correct given the problem formulations. I also
get an indication that NaN is</FONT> <BR><FONT face=Arial size=2>being
generated in the SOlver (GMRES).</FONT> </P>
<P><FONT face=Arial size=2>Not surprisingly the Warning goes away when I use
the Neumann BC. THe Solver does not report any</FONT> <BR><FONT
face=Arial size=2>NaN problems.<FONT face=Arial size=2>
<<simpleBar.geo>> </FONT><FONT face=Arial
size=2><<simpleBar.pre>> </FONT><FONT face=Arial
size=2><<solver.par>> </FONT></FONT></P><BR>
<P><FONT face=Arial size=2>Thanks, the mail archives have been a big
help</FONT> </P>
<P><FONT face=Arial size=2>Laird Bolt</FONT> </P></BLOCKQUOTE></BODY></HTML>