ff3d-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [ff3d-users] Code to solve Fixk's law, what's wrong??!!


From: Stephane Del Pino
Subject: Re: [ff3d-users] Code to solve Fixk's law, what's wrong??!!
Date: Sun, 9 May 2004 17:49:35 +0200
User-agent: KMail/1.6.2

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello.

Le dimanche 9 Mai 2004 01:54, Arun Viswanathan a écrit :
> Hi all, I wanted to write a code to solve Fick's law
> (J=-D dC/dx) in Fem3D. I saw the examples, used the
> documentation and came up with this code (below). It
> runs, but:
> 1. I don't know if it is correct, how to check it?
I think you made a few mistakes.
1) You are trying to solve an 'order one' problem. freefem only workes with 
'order two'. Your problem is also 1D ...
2) Since your geometry can be defined using the structured mesh, you do not 
need to define any domain and then use standard FEM using
        solve(u) in M
It is a comon mistake: people usually want to define a POV geometry even if 
they want to use the whole mesh to solve their problem. Fictitious domain 
methods are to be used only when the mesh does not describe the geometry.

By the way you defined a domain without using the 'inside' or 'outside' 
keywords. This should produce an error since this syntaxe is now obsolete.

> 2. I need to enter values for Diffusivity (D) and Flux
> (J) for different chemicals, to solve for
You just have to replace D and J by functions.

> Concentration (C), how do I do that?
> 3. How to use OpenDX to visualize my "p1.dat" file?
This part is not finished yet, but in your case, it should work.
> Structured3DMesh.cpp:33: Not implemented
This only means that ff3d does not know how to save a mesh in the OpenDX 
format, but this is not a problem for you. You just need to save the data. 
Then in DX, you have to use the "Grid or Scattered File" option choosing the 
structured type.

I have to insist that if you want to solve order one problems freefem may not 
be the answer since, up to now, it only deals with elliptic problems. 

Best regards,
Stephane.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.4 (GNU/Linux)

iD8DBQFAnlMVbZyhKhZwdc0RAmlPAKDNtp6N3dzKdZM2e+njGnrVRNdeqgCdFhP6
8rrPjusUWyPt83OmtmnN0p0=
=1VF/
-----END PGP SIGNATURE-----




reply via email to

[Prev in Thread] Current Thread [Next in Thread]