Work at SourceForge, help us to make it a better place! We have an immediate need for a Support Technician in our San Francisco or Denver office.

Close

OpenFlower V1.0 released!

New Beta Version 1.0 has just been released !

I decided to call it version 1.0 due to the first implementation of the Navier-Stokes solver.

It includes in particular the following new features :
- A second order convection operator;
- A flexible structure called "Algorithm" which enables one to "easily" build new resolution algorithms and coupling between equations;
- An interface with the LASPack linear algebra library for solving linear systems, in particular the Poisson equation for pressure;
- A Navier-Stokes algorithm based on a finite-volume and pressure-velocity coupling formulation used at Stanford CTR along with a 2nd order Adams-Bashforth time scheme;

This is a Beta release because of the following:
- I did not test it on a wide range of cases (anyway, you can see for yourself on the "examples" directory which type of computations I made)
- The calculation of Pressure gradients is not fully conservative at the moment on highly skewed grids (a least squares minimization method will be implemented soon to take this into account)
- In the input data file, we specify a solver, but it is always a Conjugate gradient method which is used.
- The order of input of the data in the FLW file is not thouroughly checked when preparing the simulation, therefore if you do not follow exactly the same structure I used in the examples, you might get strange errors...
- Only velocity inlet, zero pressure outlet and walls boundary conditions are implemented.
Apart from the above statements, it is (I guess) fully functional and will work on majority of your multi-element mesh cases.

Future developments will focus on:
- Correcting some of the above points...
- Taking into account your comments and reported bugs!
- Post-processing outputs for Paraview and Tecplot
- Segment kind probes in the domain
- Implement periodical and neutral (to perform quasi 2D simulations) boundary conditions
- Release a new version of the Reference Manual (I know it is very important)

Testers are welcome to check the sources and make comments!

See U soon on our web-site.
In the meanwhile, do not hesitate to test the code, send image results (openflower at gmail.com) to post on OpenFlower website and tell us what you think about this Beta release!

VortexFlow.

Posted by Alexandre Chatelain 2005-05-28