<!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html;charset=ISO-8859-1">
<title></title>
</head>
<body text="#000000" bgcolor="#ffffff">
Dear Felix,<br>
<br>
Here are your files extended for what you are interested in.<br>
<br>
What I have added and modified:<br>
<br>
* In EleKin_EleMuscCyl.pro:<br>
- 'EleKin' and 'Electrokinetics' instead of 'EleSta' and
'Electrostatics', which corresponds to the implemented model.<br>
- change of sign of the Neumann term in the formulation (the convention
is usually to consider an outward normal to the studied domain).<br>
- global quantities (currents I and floating potentials V) associated
with electrodes (the new model is explained below).<br>
<br>
* In EleMuscCyl09.pro:<br>
- two ways to solve the problem can be used:<br>
<br>
Model 1. your initial model:<br>
The given current density (j[]) is actually the normal current density
entering the studied domain. Nothing is fixed regarding its tangential
component, which explains the non-perpendicularity of the current
density to the electrodes. -> Model 2 solves this problem.<br>
Moreover, the distribution of the current density is usually not known
in advance and should rather be the solution of the problem. For that,
it is better to fix the total current through each electrode) ->
Model 2 allows that.<br>
<br>
Model 2. A new model with floating potentials:<br>
To force the perpendicularity of the current density on the electrodes,
each electrode has to support a constant potential (of floating
nature). The associated global quantity will then be the current
through each electrode. The current density distribution will vary
along each electrode.<br>
<br>
* In PhysicalValue.dat:<br>
- change of the sign of J1, J2 and J3, coherent with the change of sign
of the Neumann term in the formulation. J1, J2 and J3 are then the
normal current densities leaving the studied domain.<br>
<br>
* In EleMuscCyl09.geo:<br>
- a Physical Point 3000, for fixing a reference potential for your
initial formulation (to fix the potential at one node, for a rigorous
treatment in the absence of any other Dirichlet condition)<br>
<br>
I have written comments in all these files describing these additions
and modifications.<br>
<br>
Some references for a good understanding of the floating potential
model:<br>
<p class="Publications"><span lang="EN-GB">"Une méthode naturelle de
traitement des
potentiels flottants associée à la méthode des éléments finis",<span
style=""> </span><br>
P. Dular, F. Henrotte, B. Meys, A.
Genon, W. Legros,<span style=""> </span><br>
Journal de Physique III France, <u>Vol.7, No.11</u>, (1997), pp.
2201-2209.</span></p>
<p class="Publications"><span lang="EN-GB">"Coupling of local and
global quantities in
various finite element formulations and its application to
electrostatics,
magnetostatics and magnetodynamics",<span style=""> </span><br>
P. Dular, W. Legros, A. Nicolet,<span style=""> </span><br>
IEEE Transactions on Magnetics, <u>Vol.34, No.5</u>, (1998), pp.
3078-3081.</span></p>
<br>
I hope all that will be useful.<br>
<br>
Kind regards,<br>
<br>
Patrick<br>
<br>
<br>
<br>
<br>
Félix Buhlmann wrote:<br>
<blockquote type="cite"
cite="midDJEDKEOIKHPBJHGIIHMAMEIGCIAA.felix.buhlmann@compex.ch">
<pre wrap="">Dear Patrick,
Thank's for the answer.
I want to simulate a cylindrical problem composed of 3 electrodes and a
cylindrical conductive medium (conductivity sig[]). There are 2 coaxial
electrodes (ELEC1,2) on the top and one on the bottom (ELEC3).
I found the weak form as (sig[] Grad v,Grad v') on V and (-j[] ,v) on S1.
S1 are the electrodes.
I want to give the current density on the electrodes J1 = (I13-I21)/S1, J2 =
I21/S2, J3 = -I13/S3 (Neumann boudary).
I have attached the files with the problem description. Is that correct ?
It seems that the current density is not constant under the electrodes, see
for example electrode 2 (case I21=0.1; I13=0 in file PhysicalValue.dat)
I though that I will get a current density perpendicular to the electrodes
and it's not the case. Why ?
How can I be sure that the total current is zero. Must I give a constraint
on the currents ?
I thank you very much for your answer.
Félix
-----Message d'origine-----
De : Patrick Dular [<a class="moz-txt-link-freetext" href="mailto:Patrick.Dular@ulg.ac.be">mailto:Patrick.Dular@ulg.ac.be</a>]
Envoyé : vendredi, 29. avril 2005 09:02
À : Félix Buhlmann
Cc : <a class="moz-txt-link-abbreviated" href="mailto:getdp@geuz.org">getdp@geuz.org</a>
Objet : Re: [Getdp] Introduction de source de courant
Felix- It is an electrokinetic problem you want to solve, isn't it? I
build a model for such a problem with current constraints and send it to
you. Do you need a coupling with electrical circuits as well?
Patrick
Félix Buhlmann wrote:
</pre>
<blockquote type="cite">
<pre wrap="">Cher Christophe,
Je simule actuellement un problème electrostatique comprenant plusieurs
électrodes.
J'aimerais pouvoir spécifier le courant injecté (source externe) entre
</pre>
</blockquote>
<pre wrap=""><!---->paire
</pre>
<blockquote type="cite">
<pre wrap="">d'électrode.
Par exemple,si j'ai 3 électrodes (1,2 et 3), j'aimerais spécifié le courant
entre les électrodes
1,2 avec une source de courant (source I1) et le courant entre les
électrodes
1,3 avec une autre source de courant (source I2).
Comment puis-je faire ?
merci pour ta réponse.
Félix
Félix Buhlmann, Ing. phys. dipl. EPF
R&D Project Manager
Compex Médical S. A.
Zone Industrielle"Larges Pièces A"
Chemin du Dévent
CH-1024 Ecublens
Switzerland
Tél : 41 21 695 23 65
Fax : 41 21 695 23 61
Website: <a class="moz-txt-link-freetext" href="http://www.compex.ch">http://www.compex.ch</a>
_______________________________________________
getdp mailing list
<a class="moz-txt-link-abbreviated" href="mailto:getdp@geuz.org">getdp@geuz.org</a>
<a class="moz-txt-link-freetext" href="http://www.geuz.org/mailman/listinfo/getdp">http://www.geuz.org/mailman/listinfo/getdp</a>
</pre>
</blockquote>
<pre wrap=""><!---->
--
Patrick Dular, Dr. Ir., Research associate, F.N.R.S.
Department of Electrical Engineering and Computer Science
Unit of Applied Electricity
University of Liege - Montefiore Institute - B28 - Parking P32
B-4000 Liege - Belgium - Tel. +32-4 3663710 - Fax +32-4 3662910
E-mail: <a class="moz-txt-link-abbreviated" href="mailto:Patrick.Dular@ulg.ac.be">Patrick.Dular@ulg.ac.be</a>
</pre>
</blockquote>
<br>
<pre class="moz-signature" cols="72">--
Patrick Dular, Dr. Ir., Research associate, F.N.R.S.
Department of Electrical Engineering and Computer Science
Unit of Applied Electricity
University of Liege - Montefiore Institute - B28 - Parking P32
B-4000 Liege - Belgium - Tel. +32-4 3663710 - Fax +32-4 3662910
E-mail: <a class="moz-txt-link-abbreviated" href="mailto:Patrick.Dular@ulg.ac.be">Patrick.Dular@ulg.ac.be</a></pre>
</body>
</html>