<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=iso-8859-1">
<META content="MSHTML 6.00.2800.1106" name=GENERATOR>
<STYLE></STYLE>
</HEAD>
<BODY bgColor=#ffffff>
<DIV><FONT face=Arial size=2><FONT size=2>
<P>Raymond Michaux (</FONT><A href="mailto:raymond.michaux@freesbee.fr"><FONT
size=2>raymond.michaux@freesbee.fr</FONT></A><FONT size=2>)</P>
<P>Hello gentlemen,</P>
<P>I am a retired engineer and I just recently discovered GetDP on the web. I
think it’s a great idea to propose a unified workshop for the treatment of
partial derivatives and integral equations and to make GetDP available on the
web. Congratulations !</P>
<P>I have two questions :</P>
<P>1. I checked first the magnetostatic exemple without any problem. I am now
trying to make a simple model with a magnet in an airgap for comparison with
analytical solution. I meet some difficulties with FunctionSpace for boundary
conditions. For a 2D Magnetic Potential vector solution, the possible boundary
conditions are</P>
<P> values of potential vector A imposed</P>
<P> magnetic induction B = curl A normal to boundary</P>
<P> and perhaps floating constant A…</P>
<P>For the second condition (B normal) I can’t make the difference between
BF_CurlEdge and BF_CurlPerpendicularEdge and I didn’t find any instruction to
indicate that the tangential component of curl A is 0 (perhaps
CompX[BF_CurlEdge] ????)</P>
<P>2. Later, I would like to make a stepwise analysis of a permanent magnet
motor. Is ref [67] in Christophe Geuzaine thesis (Ferreira da Luz and all.;)
available by internet? The use of periodicity conditions and of a moving band
seem very promising</P>
<P>Thank you for your help</P></FONT></FONT></DIV></BODY></HTML>