Has anybody managed to export a CGNS file from engrid?
I tried to modify the engrid/src/scripts/engrid_installer_updater.cfg and changed the line
Unfortunately engrid still says that CGNS support has not been compiled.
Last time Oliver had a quick look at it, he decided it was preferable to implement the ability for exporting to SU2: http://engits.eu/vanilla/index.php?p...-engrid-1.3/p1 :)
Anyway, his initial answer back then was this:
I managed to get this working on Ubuntu 12.04. Steps:
thanks for your suggestion. I implemented your changes as best I could, but when compiling I get an error compiling cgnswriter.cpp. I am on Fedora anyway.
This is where I have to give up. It's a pity because otherwise engrid in combination with gmsh is really great. I am a bit astonished that there is no broad acceptance and support of cgns. The NASA tools are not bad either, but do not offer much regarding data import/export.
Maybe foamToCgns could be a solution, but it is not straight out of the box either.
Hi Sebastian & Bruno,
personally I find that CGNS is not much of a standard, because there is a lot of different ways you could organise your CGNS file and it it is no way guaranteed that a given CGNS file works with a given solver.
Which error message did the complier produces when trying to compile the CGNS writer?
What exactly do you want to do? Maybe it would indeed be easier to simply write an export function for the format you need -- other than CGNS ;-)
Out of interest: Do you use Gmsh for surface meshing? Did you ever try enGrid's own surface mesher?
Hello Oliver and Bruno,
I want to transfer data between an inhouse 1D solver and 3D CFD. CGNS appeared to me a good choice since I did not want to reinvent yet another format. With CGNS I also have some experience using the NASA tools to explore and access the data. From what I have learned so far, CGNS is not that widely accepted as a standard format for data transfer between CFD codes. This is probably the right moment to drop the idea of using CGNS. The only requirement for a data exchange standard is that it be widely understood and accepted. Maybe SU2 is an option.
The reason for using gmsh (surface) + engrid is that I have worked out how to obtain an Openfoam mesh this way. If there is an easier way using only engrid, all the better.
|All times are GMT -4. The time now is 20:10.|