can you please post some examples using openfoam with salome thank you

Umair, see the OSSS Manual 1 (OpenFOAM) in the file section of this group.

I have had a look at your OSSS Manual1, and it is very good, but there are a couple of comments I would like to add:

1) After the fuse stage in Salome (Step 6), it may be helpful to change the view. Menu "View", "Display Mode", "Shading", to get the image you illustrate.

2) In Step 7, creating Groups, I found it was necessary to first explode the fuse into faces.

3) In Step 13, saving the *.unv file, I think it would be a good idea to also save the native *.hdf project as well. Although this is a pretty simple geometry and easy to recreate, when one starts looking at more complicated geometries, one can save a whole lot of time when it is necessary to modify the geometry, to have the original available for modification.

I have upgraded the original OpenFoam installation that came with my CAELinux 2010 distro to OpenFOAM 1.71 and ParaView (openfoam) 3.80 (NOT something I would recommend for most people- stay with the original!), and I have had some issues with Paraview. In fact, one of the reasons I went with the update was because I have never been happy with Paraview (I find Salome's Post-processor much more usable, but haven't figured out yet how to get OpenFOAM results into the *.med format...Salome Post Pro also allows for importing from a tabulated file, but I haven't played with that as yet. Another problem I have with Paraview is the inadequacy of the documentation...

When I transfer the project to another computer with CAELinux 2009, I get results in Paraview similar to what you illustrate (although I still can't figure out how to get the big, glaring glyphs that you show!)

CAELinux 2010 comes with a set of applications called "Discretizer" (http://www.discretizer.org/) that are suggested as graphical front ends to OpenFOAM, but their documentation is even more sparse than the documentation for Paraview. This package has some real potential, I believe, especially since it holds the possibility of simplifying the editing of the "p" and "U" files.

Another application I am playing with (still not happy with the results) is VisIt, which is a *.vtk viewer.(https://wci.llnl.gov/codes/visit/). One limitation I have encountered is that when I use "foamToVTK" I get a number of files and folders, and I haven't quite figured out how the data is distributed between the files...VisIt seems to be fairly well-documented, although I haven't fully explored it. It supposedly runs on most platforms...

What would be really nice would be integration of Salome and OpenFOAM, similar to the Code:Saturne/syrthes integration...

Charlie, thank you for your comments. I think all this should be added to the Manual. I have uploaded the orginial odt-version of the Manual in the file section of this group, so you or anyone else may just add the missing things. If you edit the odt file, don't forget to add yourself as an author on the title page of the Manual.

Regarding the ParaView part of the Manual, I know that some steps are missing, I was hoping that all this would be obvious after some trial and error in the Paraview GUI. I will add this to the Manual as soon as possible (may take some time).

Generally, my experience is that ParaView is a really EXCELLENT software, i.e. you can do ANYTHING imaginable with that software in the field of data post-processing. It's really worth to learn using this software.

Hi, I use Salome& Code_Saturne for my study of CFD and I have problems with Code_Saturne ,I don't found good things to understanding to how using Code_Saturne.

Sherko, you should better use OpenFOAM for CFD, see the example in the OSSS Manuals in the file section of this group.

Thank you very much but I don't have a choice ,because I'm student of doctorate and the my Professor who told me to do my work with Code_Saturne.

Sherku,

Perhaps if you could be a bit more specific about your issues with Code_Saturne, we could be of more assistance. You may also find help at the Code_Saturne forums (https://code-saturne.info/products/code-saturne/forums), which includes a section of Flow Modeling Examples.

I am working on a tutorial for using Code_Saturne, a draft of which can be downloaded here (http://ubuntuone.com/p/oKn/).

Thank you ,Charles,

That is very nice from you the (tutorial), My problem is that I don't know how coupling Salome with Code_Saturne(opening Code_Saturne in the option of Salome)CFD module with the others modules of Salome.

Therefore,I opening Code_Saturne & Salome individual and when I running Code_Saturne it stop and write this (ERROR WHILE RUNNING THE PREPROCESSOR) sh:xtrem:command introuvable.

I use Code_Saturne 1.3.3 and Salome 5.1.5

What operating system are you working with?

Regarding the discussion I referred you to, I should point out that, due to the size of the mesh I was working with and the limits of the memory available to my computer, it was best to run Salome and Code_Saturne independently. This is not the source of your error.

The "sh:xtrem:command introuvable" is French for "xtrem: Command not found", issued by your shell interpreter. This suggests that there may be an issue with your $PATH variable or one of the other environmental variables used by sh to find files. "xtrem" is a command being called by Code_Saturne. I do not find a reference to a binary or shell script of this name in my installation, so I am not sure where the command would come from.

When you run Code_Saturne, are you using a special terminal that has been set up for Code_Saturne, or are you rnning it from a conventional shell? In your Code_Saturne folder (if installed normally, this should be int the "/opt" directoryon a Linux system). In this folder is a shell script, "setenv.sh" that should be called before setting up your study. This is usually done from th user's ".bashrc" file. There are other files called from the setenv shell script that must conform to your system setup as well...

I would suggest that, if you are not all that familiar with how your operating system is configured, you consult with someone local who might be of more assistance. Were you using the CAELinux distro, all of this would be pre-configured for you.

Dear Sir,

I use Linux Mandriva 2010 ,you are right the error perhaps (xterm and not xtrem) and I must talking with my professor about this problem and what you suggested . Thank you very much and

good night.

If it is xterm, this is a system program (type "man xterm") that is "a terminal emulator for the X Window System. It provides DEC VT102/VT220 (VTxxx) and Tektronix 4014 compatible terminals for programs that cannot use the window system directly." The error is most likely at the system level, not the Code_Saturne application level.

In your case directory, in the "RESU" subdirectory, you should have a file "listing.########" where the "#" corresponds to a date & time. This is the output of the process, and perhaps you can determine where in the pre-processor hangs. You should also see in the "RESU" directory a "runcase.########" file and an error log with the same date/time stamp. Examining these files can possibly help you figure out where the hangup is. I suspect "xterm" is calling a Code_Saturne function that is not found in the $PATH.

I haven't encountered this problem directly, so without access to your case files, it is difficult to assist further. If you send me a private message and attach the "runcase", "listing" and "error" files for one of your runs, maybe I can help further...

Hi Charles ,Dear Sir,

As I told you before I am working in coupling Salome &Code_Saturne, and I have problems in running Code_Saturne every time and the programme stop calculation and write this sentence ERREUR WHILE RUNNING THE PREPROCESSOR .

./SaturneGUI:line 41 :Erreur de syntaxe pres du symbole inattendu 'else'.

Last time as you told me I fixed the xterm and know the small white window is appear after I lance Code_Saturne but no results on it.

I am really what your advice to fix my programming.

Thank you,

Best regard

Sherko AL-ARKAWAZI

What version of Code_Saturne are you using?

It would be easier to troubleshoot this if you could send me a copy via private message of your GUI file and XML file and your listing file and any error file you might find. These are located in your Case directory. The GUI file can be found in the DATA subdirectory, and the listing and error files can be found in the RESU directory.

The GUI file is build by Code_Saturne, and it reads the XML file that is built when you define your case parameters. You should have "listing.########" and "listpre.########" files are the outputs of the program as it runs. If the program is hanging on the preprocessor, you may not have the listing file. Make sure when you send me the files that they have the same time stamp (the "#"'s in the file name). There should be an error file and a summary file with the same time stamp. These would also be helpful.

I would, of course, rather look at these in English, but I can translate from the French if needed (it just takes a bit longer!).

Hi Charles,

I use Code_Saturne version 1.3.3 and I think there is a problem in XML files.

I will try to do what you say, thank you very much for your effort with me.

Sherko-

I am not sure why, but it appears that your case.html file may be incorrect. It seems that line one may have the wrong version number. If I am reading your error report correctly, then the solution might be to manually edit the file, which should be in your CASE/DATA folder, and in the first line of this file, change the end of the line to read

version="1.3">

and see if this helps. You should be able to edit this file with any text editor- but I do not recommend using a word processor for editing these files, because they have a tendency to sneak hidden codes into the final document which can wreak havoc on your program

It looks like you are getting through the preprocessor- you should have a file in your RESU folder named listingpre.05231513 (the numbers being month, day and time of the particular run). This is a text file that can be easily read with any editor. There may be some sort of clue in here, if the preprocessor is not completing its tasks correctly. Look especially at the end for something to the effect:

Time and memory summary

-----------------------

User CPU time (sec) : 1.61

System CPU time (sec) : 0.13

Total time (sec) : 1.73

Total CPU time / Total time : 1.00

Memory use summary:

Total memory used: 105.391 Mb

Theoretical instrumented dynamic memory: 72.069 Mb

.-----------------------.

| |

| Preprocessor finish |

| |

`-----------------------'

The important thing is no errors noted. This file should also give you a full list of all your mesh components...

It appears we have made some progress. Your mesh appears to be OK, or, at least, the preprocessor is not finding any faults with it.

You have achieved a core dump. We may still not have the version number set properly, and I am not sure why the original setup did not put the correct version number in the *.html file, but for now, let's move on.

A common cause of core dump (but not necessarily the only cause) is a divide-by-zero error. It is now time to review your material properties and initial conditions. These are all embedded in the *.html file. Open the *.html file, copy it into a plain text editor, and let's have a look at that.

It appears that your *.exe file is compiling without initializing the variables stdtcl_, tridim_, and caltri_. My preliminary investigations have not uncovered just what these variables are, but I am working with a newer version of Code_Saturne. I have an older version installed on another computer, and I will need to check that one to see if I can identify where these variables are initialized. I suspect we are not coupling an appropriate header file, or possibly one of the *.f90 source files that should be included.

Meanwhile, cd to your /tmp/tmp_Saturne/RECTANGULAR.CAS1.05251140/ folder and run from the command line "./cs13.exe --track-origins" and see what sort of output we get from that.

Sherko-

Sorry it has taken me so long to get back to you, but it has taken me a while to track down what is going on. From the real Gurus at Code_Saturne, I learn that stdtcl, tridim and caltri come from Fortran routines that are added at compile time (which happens when you click the "Run Case"). Of special interest in this instance is stdtcl, which is a routine that "is used to compute boundary conditions based on data given in the GUI (such as computing a velocity given a flowrate, or computing default turbulence inlet conditions)." This suggests that the issue you are having is related to the boundary conditions you are setting (or, possibly not setting). Again, the key will show up in the *.html file. What turbulence model are you using? Are you properly specifying the inlet velocity?

You may want to try running the case first with no turbulence model.

If you are not using the GUI to set your boundary conditions, then they are set in the file [CASENAME]/SRC/REFERENCE/base/usclim.f90, which will have to be copied from the "base" subdirectory up one level (to the "SRC" subdirectory) before editing. This is an ASCII file that is pretty well commented and explains the various settings. and how to designate which groups the conditions apply to. However, there is nothing to set for a symmetry wall condition.

It is much easier to use the GUI. If you are using the GUI, the document "Tutorial.pdf" in the /opt/saturne/cs-1.3.3/share/doc/ncs folder. Beginning on Page 43 of this document is a step-by-step procedure for setting up your case from the GUI, which includes instructions on how to import groups from the mesh, how to specify wall conditions, and how to initiate boundary conditions. Again, there is nothing to set for the symmetry wall condition, but this will help you insure your inlet conditions are set correctly.

I believe the *.unv file does not retain the information about groups, which I believe is crucial for identifying the boundaries. I have always used *.med files for Code_Saturne.

In your mesh in Salome, are you identifying groups of faces?

I'm not sure why the preprocessor would reject your *.med file. Can you send it to me via private e-mail? The *.hdf file from Salome would be the appropriate file to send. I will send you my e-mail address via private message, and you can send the file as an attachment to a regular e-mail.

I'm newbie in coupling code field. I want to coupling code among neutronic code , thermalhidraulics code and structural code. Any suggest for me about it. I mean, what code could be coupled which suitable/proper for my purpose?

Hi Kai,

Can you tell me how I can download paraview in Linux (ubunto10) for visu. the reasults of fortran code.

Hi, where I can view group file section ? Thanks,

Hello ,

I'm trying to use Salome and OpenFOAM to simulate my propeller in MRF,

But i had problems,

When I created the boolean cut into my propeller, the process was stopped,

I suggest my model is not solid,

so anybodies have the same problems?

Best regrads,

Aldi