Hi Ajay!

Glad to hear that you are diving into FiPy.

You will see that FiPy includes an explicit diffusion term to illustrate that one can violate the stability criterion. Of course, this is why most of the time you would pick the implicit diffusion term. There are a couple of paragraphs in Numerical Recipes that discuss stability and amplification factors. (I apologize if you already know these things.) If this is the first time you are hearing about this, then read on.

Regarding stability issues, a good thing to do is to have both the FiPy users manual and a copy of a book like Numerical Recipes in front of you at the same time. That should provide some additional background information needed to really understand
the FiPy users manual.

Does this help? Have I missed the point of your question? Maybe you could ask the
question a bit differently?

Good luck!

Best,
Dan


Dan Lewis
Assistant Professor
Materials Research Center, Room 110
Rensselaer Polytechnic Institute
110 8th Street
Troy, NY 12180
[email protected]
518-276-2297

http://www.rpi.edu/~lewisd2






On Oct 23, 2009, at 7:22 AM, Ajay Rawat wrote:

Another Question........
Is there any stability related issue in solving phase field problem with respect to mesh size and time steps.

Thanking you all in advance

I'm really liking Fipy.
Really thankful to Dr. Daniel wheeler , Jonathan Guyer  and his team.

On Fri, Oct 23, 2009 at 12:11 PM, Ajay Rawat <[email protected]> wrote:
Is there any methods to solve stiff differential equation in fipy.
seacuse in my model in need to couple my PDE with stiff differential eqn.
or i had to goto scipy to solve stiff eqn.

--
Ajay Rawat
Kalpakkam, IGCAR

-------------------------------------------------------------------------
Save Himalayas....
-------------------------------------------------------------------------



--
Ajay Rawat
Kalpakkam, IGCAR

-------------------------------------------------------------------------
Save Himalayas....
-------------------------------------------------------------------------


Reply via email to