CFD Online Discussion Forums

CFD Online Discussion Forums (http://www.cfd-online.com/Forums/)
-   Main CFD Forum (http://www.cfd-online.com/Forums/main/)
-   -   FOAM (http://www.cfd-online.com/Forums/main/8112-foam.html)

CFDtoy September 20, 2004 13:12

FOAM
 
Hello All,

Has anyone used FOAM Software for simulating unsteady flows?

Thanks

CFDtoy

luiz September 20, 2004 21:57

Re: FOAM
 
Yes. I have. LESimulation.

baal September 21, 2004 07:41

Re: FOAM
 
is it a free software? how can i get it?

thanks.

matej September 21, 2004 09:40

Re: FOAM
 
you can have a free license for academic use; surf to www.nabla.co.uk, or you can search this forum, I gues the link is here about 10 times already. You may also find google or other search engine to be usefull.

matej

CFDtoy September 21, 2004 10:44

Re: FOAM
 
Hi Luiz,

I would like to know if you were able to access the source code as claimed by FOAM. Were you able to implement any of your own algorithms or additional subroutines ?

CFDtoy

luiz September 21, 2004 14:09

Re: FOAM
 
Yes. The source code is really available, and yes, I have done some interesting things myself. I think the main advantage of FOAM is its OO philosophy, which makes it very easy to implement new algoritms. Since there are many differential operators readily available (both spatial and temporal), you can mount new equations/variables/models adding a couple of lines to the original code, without having to worry about the lower layers... For instance, if you want to solve a completely new equation for a tensor T, you could just add:

fvMatrix T { fvm::ddt(T) - fvm::div(T) == fvc::curl(U) } T.solve()

This little piece of code is almost self explanatory: 1) build a matrix T to be solved; 2) obeying the equation ddt(T) -div(T) = curl(U) (That is just an example meaning nothing, just to show you are not tied to navier stokes eq); here fvm is a class of implicit operators, while fvc is for explicit ones...; 3) then solve the eq.

This way, you can QUICKLY implement new models, or even new algorithms, focusing only on what you need. And I think that is why they are able to offer so many turbulence models, multiphase, etc ready to use with their package. I also expect them to be quicker to release new algorithms, models, etc than other commercial codes.

As you can see, I am enjoying a lot the code, and I am becoming a big fan of C++ and FOAM...

good luck


All times are GMT -4. The time now is 14:57.