Polyhedron and OpenFOAM
Hi,
Do you know if polyhedron are supported by OpenFOAM? well I would likt to export a StarCD mesh and use OpenFOAM... Thank you! |
Re: Polyhedron and OpenFOAM
Yes, fully supported.
Enjoy, Hrv |
Re: Polyhedron and OpenFOAM
May I suggest you re-write the readme file that comes with OpenFOAM and make it more accessible to people, like me for instance, who are new to the linux OS and have real difficulties installing the programme. I downloaded the 1.5 files, unpacked them, run the check script, the result is ok, then I have to source a lot of steps. I do not know what sourcing does mean in an Linux OS, I have a 1000 pages tome on opensuse, I have to find what sourcing does mean and how to apply it correctly. If you could rewrite that part and give a step by step instructions on how, that would be a real gift. You see, you have developed the programme and know all the details, but I am just a user who wants to look at the bright side of CFD, and, in my opinion, I do not need to be a guru on Linux Thank you
|
Re: Polyhedron and OpenFOAM
Great, that's good news!
Thanks for your answer |
Re: Polyhedron and OpenFOAM
It works as long as you don't have to do postprocessing with ParaView for real world problems including Polys ;-)
|
Re: Polyhedron and OpenFOAM
Well, I'm a bit better than that. For post-processors that cannot cope with polyhedral cells, OpenFOAM will automatically split the cell on the fly and present it to the post-processor as a set of tets/hexes/pyramids.
Happy? :) Hrv |
Re: Polyhedron and OpenFOAM
Not happy. It's just a workaround. Splitting a 2 million cell poly mesh will degrade the ParaView performance.
The poly implementation in the latest ParaView version is buggy and even slower. Ok you get what you pay for and if you want a nice postprocessing EnSight is still available :) Funny observation in ParaView: Export from StarCD in EnSight Gold format. Pressure file has a name case.P and temperature has case.T . Paraview is unable to read the pressure file but reads the temperature file (1 character variable name). |
Re: Polyhedron and OpenFOAM
yes, you are correct. Paraview is big hurdle/ if the mesh is slighly larger there is big problem doing post processing with paraview.
|
Re: Polyhedron and OpenFOAM
Do we have any alternatives? Paraview is good, visualisation is nice, does not cost a fortune etc. I am sure the VTK and Paraview developers will sort out the polyhedral treatment soon if there is sufficient interest.
Hrv |
Re: Polyhedron and OpenFOAM
I use VisIt (http://www.llnl.gov/visit) regularly. The binaries work for me, it's a major pain to build from source. It has a lot more features than Paraview and generally supports more data formats. The UI is slightly less intuitive in my opinion.
|
Re: Polyhedron and OpenFOAM
Ahmed, I understand your frustration
"source" is actually a unix command. The syntax is usually source SomeInputFileHere Try using the "man" command to get an idea of how source works. ie man source You may find that you will get errors that the file you specified is not found which will seem odd to you becuase it is right there in the directory where you issued the command. This is because "." (or as we say DOT) is not in your path. So then the source command becomes source ./SomeInputFileHere I am sure the SUSE manual is overwhelming. Try and find a simple unix book. Is there a "unix for dummies"? A lot of open source stuff is very linux (unix) centric so having some idea of what a shell is and how it might be configured will definately come in handy. Good luck - Andy R |
Re: Polyhedron and OpenFOAM
`source' is not and cannot be implemented as a program, it must be implemented in the shell because it manipulates the shell's environment. Also `man source' will not give you anything useful on most systems (`man bash' followed by a search for `source filename' will). You will frequently see it spelled `. filename'.
The purpose of this is to manipulate your environment (like PATH) so that the correct programs can be run without giving the whole path. Something like this can't be avoided unless the program is installed to a standard location. |
All times are GMT -4. The time now is 03:00. |