FOAM FATAL IO ERROR: keyword features is undefined in dictionary??
1 Attachment(s)
Hello all!!
Can someone tell me why i am getting this error? I followed some tutorials and i still not knowing.. why?? I am really new and alone with this.. i have also attached the SnappyHexMeshDict (Maybe is not good enough but at the beginning i only want that Snappy works with my sistema.STL) In my .STL i named "solid objeto" and also at the end "endsolid objeto" Thanks a lot!!! --> FOAM FATAL IO ERROR: keyword features is undefined in dictionary "/home/amorenelchino/openfoam/amorenelchino-2.2.0/run/sistema/system/snappyHexMeshDict.castellatedMeshControls" file: /home/amorenelchino/openfoam/amorenelchino-2.2.0/run/sistema/system/snappyHexMeshDict.castellatedMeshControls from line 37 to line 60. From function dictionary::lookupEntry(const word&, bool, bool) const in file db/dictionary/dictionary.C at line 402. FOAM exiting |
Hi
if you read the error message carefully you will recognize it says Code:
keyword features is undefined in dictionary This means you haven't defined the keyword "features" in your snappyHexMeshDict file To fix this include something like: Code:
castellatedMeshControls surfaceFeatureExtract to resolve knuckle lines you have to fill in that part as well e.g. Code:
{ I hope I could contribute regards |
Yuhuuuu! :D so fast!!
Thanks Colin!
sometimes i have the things in front of me..:-/ I was trying a mesh, but i would need to use these eMesh, because my geometry has sharp corners, i tried to get a tutorial, but the link in OpenFOAMWiki is not working...:-( Do you know one?? Really thanks!! |
Hi
the command you are looking for is: surfaceFeatureExtract How to use it correctly type: surfaceFeatureExtract -help For me it usually works with: surfaceFeatureExtract -includedAngle 150 constant/triSurface/your.stl "patchname" patchname is without the inverted commas and consist usually of the stl file name without .stl and the name of the solid. for reference see here http://www.cfd-online.com/Forums/ope...ture-edge.html post #4 regards |
If i understand good...with only typing this command in terminal, works these "Surface Feature", but it is also correct write the "SurfaceFeatureExtractDict" and type the command? and another question...the ".eMesh" is the result? :o
Sorry, because i am trying to use openfoam since two weeks!and in my freetime! i am getting crazy! :eek: Thank you very very much!! Cheers |
Hi
yes it is enough to simply type the command with the required options. I never used a surfaceFeatureExtractDict file, however I have never used of220 maybe in the latest version it is required. the *.emesh file is indeed the result of that command and will be stored in the triSurface folder. It also will be automatically read out by sHM when specified in the features section. regards |
Thanks you are being a great help!!
Ok! i was using "surfaceFeatureExtractDic", it worked... now i want to use "decomposePar" i did it and was ok, if i want to make SHM with this decompose'Par...what should I type in terminal?? only the command "SnappyHexMesh"? And the last...i have done "SHM", everything was running ok, but in "Surface refinement iteration 4"----> Bus error (core dumped) any idea? Always appears something...i am not going to see my mesh Thanks in advance! :) |
puh it is a little bit difficult to figure out what is wrong with so little
information: common mistakes when running sHM (in parallel) are - using wrong decompostion method (simple is the one you need) -overseeing that snappyHexMesh is writing the mesh in different time steps if the option -overwrite is not used - since you are asking what to type for sHM in parallel the question is: did you use a command like this: Code:
mpirun = multiprocessor run np gives the number of processors you use so 2 has to be replaced with your number if you use more than 2 snappyHexMesh -parallel should be clear -overwrite see above |tee snappy_Log will write the output on the screen in a log file called "snappy_Log" -> this is handy if you want to ask questions in the forum you can simply copy and paste the error message in your post and it helps us to analyse your problem. If I haven't posted the solution up there yet you might want to provide us your snappyHexMeshDict decomposeParDict and the log of your meshing process regards |
3 Attachment(s)
Really thanks for your time...
Ok, i was trying..and trying..and i think my .stl in ASCII was not good, because paraView gave me an error and closed suddenly when i tried to open it...but the geometry was ok, so i tried with .STL but binary and it was better...Paraview has opened my .stl without problems.. So i made it again, but i still have problems.. I attach the documents that you told me... if I get this i have to at least thank you in my work! i want to write my thesis.. I had to delete i bit of snappy_log, it was too heavy... |
Hi Laura,
sry for my late reply, but i had some urgent projects to work on and a long weekend so I haven't been at the computer too much the recent time. OT: The lines of your snappy_log kind of confuse me, especially the ones with the numbers in front. The only explanation which comes to my mind is that you forgot the -parallel flag when using the mpirun command. If you don't use it several calculations will be started but not in parallel. Concerning ASCII and binary I made the experience that the ASCII works usually very good, however when you use spaces in the name of the solids this can cause trouble for paraview (paraview even complains when a not OF related file in the same folder with improper naming is present). Binary files i think can not be treated by sHM. When you look at the meshing time of your snappy log you see that the time is very short which is actually unusual for sHM and that means he probably skipped some parts. Combined with the huge file size I would say that the meshing went wrong ( the huge file size can come from the error messages which are sometimes printed for each cell which leads to big files) Therefore my hint, try it again with a ASCII file and if it goes wrong post the error message which would give us another hint. regards |
All times are GMT -4. The time now is 23:23. |