|
[Sponsors] |
May 12, 2003, 18:51 |
Any CFD commercial packages
|
#1 |
Guest
Posts: n/a
|
are written by C or C++, rather than fortran?
|
|
May 12, 2003, 20:59 |
Re: Any CFD commercial packages
|
#2 |
Guest
Posts: n/a
|
I believe EFD.Lab is. But, I'm curious, why do you ask?
|
|
May 12, 2003, 21:53 |
Re: Any CFD commercial packages
|
#3 |
Guest
Posts: n/a
|
When I attended a Fluent users' meeting, perhaps 8 years ago, the developers announced (to groans from the users, who were mostly Fortran-trained engineers) that they were going to write the next major release in C++.
|
|
May 13, 2003, 07:11 |
Re: Any CFD commercial packages
|
#4 |
Guest
Posts: n/a
|
Hi,
I know that FOAM of Nabla (http://www.nabla.co.uk) is commercial and written in C++,although the academic license is free, which is something your do not want? there is also one free I know about, called mouse http://fire8.vug.uni-duisburg.de/MOUSE/ why do you need commercial? matej |
|
May 13, 2003, 07:31 |
Re: Any CFD commercial packages
|
#5 |
Guest
Posts: n/a
|
Hi,
Fluent6.1 is fully written in C, as pointed out Mouse is in C++, and their is another free product called UG (see Resources section, Books (Computational Gas Dynamics (Laney)/Software), you can get information about both. Apurva |
|
May 14, 2003, 12:42 |
Re: Any CFD commercial packages
|
#6 |
Guest
Posts: n/a
|
Yup. It's called FOAM (Field Operation and Manipulation). It does more than just CFD, although that's where it has started from: people are doing fluids - incompressible, compressible, turbulence (RANS + LES, combustion, spray, free surface, multiphase and more), solids (linear and non-linear stress analysis, contact problems, crack propagation, solid-fluid interaction etc.), electromagnetics, and more.
The idea of the code (there's many ways of doing CFD in C++) is to represent the equations in the code as closely as possible to their mathematical form, so a piece of top-level would look like this and you guess what it does (fvm = finite volume method): fvVectorMatrix UEqn ( fvm::ddt(U) + fvm::div(phi, U) - fvm::laplacian(nu, U) ); solve(UEqn == -fvc::grad(p)); Also, the code is set up to deal with all kinds of mesh complexity you've got in other commercial packages: unstructured meshes of arbitrary polyhedra, moving meshes (automatic mesh motion!), layer addition/removal (for the next release), sliding meshes etc. It runs massively parallel and it's got the same efficiency as STAR or Fluent. With FOAM you also get pre and post-processing. The pre-processor is written using Java-Corba-C++ setup and it basically edits the humanly-readable dictionaries that control the code. The "standard" post-processor is OpenDX (open source) with a FOAM module configured for use; there are also interfaces to Ensight and Fieldview for people who have them. The package is commercial and currently licenced free of charge for academic use. It runs on Linux/Unix and it's actually used by a number of universities and some commercial companies at the moment. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Future CFD Research | Jas | Main CFD Forum | 10 | March 30, 2013 13:26 |
Evaluating CFD Packages for Hydrodynamic Drag in Underwater Vehicles | Dundee | Main CFD Forum | 4 | December 5, 2011 07:31 |
CFD packages for ProE | Alasdair | Main CFD Forum | 6 | January 31, 2008 23:50 |
CFX vs 'cheap cfd packages' | Andrew | CFX | 7 | December 21, 2007 12:00 |
Commercial CFD Codes | Praveen Chandrashekar | Main CFD Forum | 8 | September 30, 1999 13:18 |