Getting started in FEM, pressurized vessel

About the development of the FEM module/workbench.

Moderator: bernd

BDL42
Posts: 218
Joined: Mon Jun 06, 2022 5:55 pm

Getting started in FEM, pressurized vessel

Post by BDL42 »

Have created a toy example, which I would like to analyze using FEM. This example is similar to a real object which I saw online, and it is known to have issues. I've never done FEM before. I did read the tutorial on the workbench, and watched a video, but that is the extent of my knowledge. I'm familiar with the basic concepts, but am struggling with actually implementing the forces. I'd like to analyze the stresses due to the sharp corner at the end of the tube. This cylinder is pressurized to 30 MPa. 6061 is totally inadequate for this use, and perhaps 7075 as well, especially with the sharp corner. I have been unable to find the material card equivalent to 7075-T651, but it's possible create a custom material with judicious editing.

I added a fixed constraint at the end of the tube. That was easy, although not intuitive. I had to click on the face twice. Not sure why that is. My next step was to add a pressure constraint, rather than force. Only the bottom of the cylinder and the 3mm walls see the 30 MPa. I was able to add just the bottom face of the pressure constraint. If I added the sidewalls, it appears that the direction of force is backwards. The pressure is always outwards. I solved this problem by having only one face under pressure and making a second pressure constraint.

Double clicking on MaterialSolid lead to a traceback:

Code: Select all

Traceback (most recent call last):
  File "/home/bruce/Apps/freecad-build/Mod/Fem/femviewprovider/view_material_common.py", line 58, in setEdit
    view_base_femconstraint.VPBaseFemConstraint.setEdit(
  File "/home/bruce/Apps/freecad-build/Mod/Fem/femviewprovider/view_base_femobject.py", line 93, in setEdit
    task = TaskPanel(vobj.Object)
  File "/home/bruce/Apps/freecad-build/Mod/Fem/femtaskpanels/task_material_common.py", line 201, in __init__
    self.selectionWidget.has_equal_references_shape_types()
  File "/home/bruce/Apps/freecad-build/Mod/Fem/femguiutils/selection_widgets.py", line 576, in has_equal_references_shape_types
    r = geomtools.get_element(ref[0], ref[1])
  File "/home/bruce/Apps/freecad-build/Mod/Fem/femtools/geomtools.py", line 189, in get_element
    return part.Shape.getElement(element)  # Face, Edge, Vertex
<class 'Part.OCCError'>: Index out of bound
Can someone explain this and how to fix this?

Code: Select all

OS: Pop!_OS 22.04 LTS (pop:GNOME/pop)
Word size of FreeCAD: 64-bit
Version: 0.21.0.32959 (Git)
Build type: Unknown
Branch: master
Hash: bfb3efbc72a2b562a5ed7a0696e4a72f563bfd2b
Python 3.10.6, Qt 5.15.3, Coin 4.0.0, Vtk 7.1.1, OCC 7.5.1
Locale: English/United States (en_US)
Installed mods: 
  * fasteners 0.4.56
  * kicadStepUpMod.backup1663010049.9382765
  * Render 2023.2.3
  * Help 1.0.3
  * A2plus 0.4.60k
  * fcgear 1.0.0
  * lattice2 1.0.0
  * kicadStepUpMod 10.18.8
I'm also having issues with meshing, but let's solve things in a methodical fashion. Appreciate any and all help.
pressuretube.FCStd
(767.54 KiB) Downloaded 33 times
User avatar
NewJoker
Veteran
Posts: 3018
Joined: Sun Oct 11, 2020 7:49 pm

Re: Getting started in FEM, pressurized vessel

Post by NewJoker »

Selections in the FEM workbench are not very intuitive at the moment - you have to click Add and then select faces.

It's hard to say what causes the issue with the MaterialSolid object but I would delete it, add a new one and only specify the values. It will be assigned to the whole part automatically.

Meshing should work in this case if you specify the proper element size (not too large). It would be best to remove the thread for the analysis unless it's important for you to take it into account but it shouldn't have a significant impact on the results. Then you could also utilize symmetry by modeling only a quarter of the part and applying symmetry boundary conditions (fixing displacements in directions normal to faces belonging to the cut planes).
BDL42
Posts: 218
Joined: Mon Jun 06, 2022 5:55 pm

Re: Getting started in FEM, pressurized vessel

Post by BDL42 »

Thanks for the response. Yes, I can remove the threads, if it is messing up the meshing.

Where are the FCMat files stored? I thought I had saved a file (with properties like 7075) but it seemed to have vanished. Maybe there's a problem with the model that prevented the save? I did a search on my whole SSD and couldn't find a single FCMat file!

When I select the body then Gmesh, I get the message:

Code: Select all

Fem::FemMeshObjectPython: Link(s) to object(s) 'Body' go out of the allowed scope 'FEMMeshGmsh'. Instead, the linked object(s) reside within 'Part'.
I don't even know what this is telling me. It sounds like something is incorrect.
Using the default settings 0 = auto, and Apply, I get the following:

Code: Select all

14:17:10    ElementDimension: 3
14:17:10    Group meshing for analysis is set to true in FEM General Preferences. Are you really sure about this? You could run into trouble!
14:17:10  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x5619b502e3c0>
14:17:10      ConstraintFixed
14:17:10      [(<PartDesign::SubtractiveHelix>, ('Face4',))]
14:17:10      SubtractiveHelix
14:17:10  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x5619b502e3c0>
14:17:10  Body--> Name of the Feature we where searching in.
14:17:10  SubtractiveHelix --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
14:17:10  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x5619b3ce2ab0>
14:17:10      ConstraintPressure001
14:17:10      [(<PartDesign::SubtractiveHelix>, ('Face43',))]
14:17:10      SubtractiveHelix
14:17:10  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x5619b3ce2ab0>
14:17:10  Body--> Name of the Feature we where searching in.
14:17:10  SubtractiveHelix --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
14:17:10    Empty reference: MaterialSolid001
14:17:10    One material with no reference shapes. No need to make a group for materials.
14:17:10  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintFixed could not be found!
14:17:10  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintPressure001 could not be found!
14:17:10    /tmp/fcfem_ks5fim0t/Body_Geometry.brep
14:17:10    /tmp/fcfem_ks5fim0t/Body_Mesh.unv
14:17:10    /tmp/fcfem_ks5fim0t/shape2mesh.geo
14:17:10    /usr/bin/gmsh
14:17:10  Unexpected error when creating mesh: list index out of range
14:17:10  Gmsh had warnings:
14:17:10  list index out of range
14:17:23  Fem::FemMeshObjectPython: Link(s) to object(s) 'Body' go out of the allowed scope 'FEMMeshGmsh'. Instead, the linked object(s) reside within 'Part'.
Some strange things going on. Not obvious for this beginner. Besides computation time, is there any compelling reason to use symmetry? Seems rather old school these days with lots of memory and decent computing platforms.
pressuretube04302023.FCStd
(769.35 KiB) Downloaded 30 times
BDL42
Posts: 218
Joined: Mon Jun 06, 2022 5:55 pm

Re: Getting started in FEM, pressurized vessel

Post by BDL42 »

I see this message: Group meshing for analysis is set to true in FEM General Preferences. Are you really sure about this? You could run into trouble!
Where is FEM General Preferences? I see this, but there is no group meshing under general, nor any of the other tabs.
pt_preferences.png
pt_preferences.png (151.18 KiB) Viewed 1472 times
Why is a highly experimental option both chosen, and unable to be unselected? I also seem to have no default solver.
BDL42
Posts: 218
Joined: Mon Jun 06, 2022 5:55 pm

Re: Getting started in FEM, pressurized vessel

Post by BDL42 »

Finally got it to mesh. This is not intuitive at all. I have a mesh, but have two warnings.

Going to mechanical analysis --> Write .inp file states "No mesh object defined in the analysis." I am using SolverCcxTools.
How does on select the mesh object? If I click on FEMMeshGmsh and then Solve>Solver CalculiX Standard it does not bring me to the SolverCcxTools menu. If I double click on SolverCcxTools, then the mesh is unselected!

This is because the object was created not under the Analysis object! I didn't put it in the wrong place, FreeCAD did. Once the mesh was inside the Analysis object, I could proceed further.

A little closer. Unfortunately I get the non-positive Jacobian error. Gee whiz, this is fighting me all the way. Well, if anyone could help, I'd appreciate it a lot.
pt_prelim_mesh_ccxfail.png
pt_prelim_mesh_ccxfail.png (268.12 KiB) Viewed 1450 times
pt_test04302023.FCStd
(454.59 KiB) Downloaded 28 times
CharlesM
Posts: 15
Joined: Thu Feb 02, 2023 4:14 pm

Re: Getting started in FEM, pressurized vessel

Post by CharlesM »

Dear BDL42,

I won't be able to help you with your specific file as I have never used FreeCAD for FEA (I am more of an Abaqus, Code_Aster guy). If you're new to FEA, directly jumping into a problem such as the study of stress concentration might be a big step. Your mesh seems too coarse, and you need to ensure a certain number of elements in the thickness of your vessel. For this type of simulation, a 2D axisymmetric analysis would be typical. I don't know if it is feasible through FreeCAD.

Start with a simpler model that you can compare with an analytical solution, such as a simple cylinder with internal pressure (https://en.wikipedia.org/wiki/Cylinder_stress). Check the influence of the mesh and the boundary conditions on the results of your simulation. And look at tutorials on youtube (study of pressure vessels is quite common in fea) even if they are for other softwares.

Once you get that running, you can move on to the real case. :)
BDL42
Posts: 218
Joined: Mon Jun 06, 2022 5:55 pm

Re: Getting started in FEM, pressurized vessel

Post by BDL42 »

CharlesM wrote: Mon May 01, 2023 10:08 am Dear BDL42,

I won't be able to help you with your specific file as I have never used FreeCAD for FEA (I am more of an Abaqus, Code_Aster guy). If you're new to FEA, directly jumping into a problem such as the study of stress concentration might be a big step. Your mesh seems too coarse, and you need to ensure a certain number of elements in the thickness of your vessel. For this type of simulation, a 2D axisymmetric analysis would be typical. I don't know if it is feasible through FreeCAD.

Start with a simpler model that you can compare with an analytical solution, such as a simple cylinder with internal pressure (https://en.wikipedia.org/wiki/Cylinder_stress). Check the influence of the mesh and the boundary conditions on the results of your simulation. And look at tutorials on youtube (study of pressure vessels is quite common in fea) even if they are for other softwares.

Once you get that running, you can move on to the real case. :)
Thanks for your comments. I'd agree the mesh seems coarse in sections of the model. Some of running comments in the code however say they are writing a "ginormous file" for the input file. I used the auto mesh for min and max cell sizes, so it would seem there are some issues. I will attempt at least to restrict the maximum size and try again. There are also occasionally some problems at sharp edges. I will try to smooth them to see if the meshing goes easier. I don't see a way to control the number of elements in the thickness in FreeCAD.

Good idea on tutorials on pressure vessels. I did do a search on FEA pressure vessel FreeCAD but or something like that and didn't find much. I will relax the requirement for FreeCAD. But in the end, I do need to accomplish this on FreeCAD. It seems the FEM documentation is lagging the software, especially in the tutorial area. I can't follow the FreeCAD FEM tutorial, because even the file names used in the tutorial have changed! And I don't know where they are stored either. Talk about making it hard to start.
thschrader
Veteran
Posts: 3129
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: Getting started in FEM, pressurized vessel

Post by thschrader »

My test-file:
pressuretube_ts.FCStd
(768.49 KiB) Downloaded 39 times
Mesher Netgen, runs pretty fast, even when including the threads.
Only a part of the pressure arrows (blue) are shown here.
tube001.JPG
tube001.JPG (40.47 KiB) Viewed 1221 times
BDL42
Posts: 218
Joined: Mon Jun 06, 2022 5:55 pm

Re: Getting started in FEM, pressurized vessel

Post by BDL42 »

thschrader wrote: Tue May 02, 2023 4:13 pm My test-file:
pressuretube_ts.FCStd

Mesher Netgen, runs pretty fast, even when including the threads.
Only a part of the pressure arrows (blue) are shown here.

tube001.JPG
Interesting.
What version FreeCAD are you using? My version:

Code: Select all

OS: Pop!_OS 22.04 LTS (pop:GNOME/pop)
Word size of FreeCAD: 64-bit
Version: 0.21.0.33041 (Git)
Build type: Unknown
Branch: master
Hash: b5a548e3fd931745667dd482aa39305f4db49ecf
Python 3.10.6, Qt 5.15.3, Coin 4.0.0, Vtk 7.1.1, OCC 7.5.1
Locale: English/United States (en_US)
Installed mods: 
  * fasteners 0.4.56
  * kicadStepUpMod.backup1663010049.9382765
  * Render 2023.2.3
  * Help 1.0.3
  * A2plus 0.4.60k
  * fcgear 1.0.0
  * lattice2 1.0.0
  * kicadStepUpMod 10.18.8
My version is built without NETGEN support. I cannot mesh this using NETGEN.
Did a build, but it broke my version of FreeCAD. Got an error

Code: Select all

make[2]: *** No rule to make target '/usr/lib/x86_64-linux-gnu/libmetis.so', needed by 'lib/libSMDS.so'.  Stop.
Hmm, need to recover from this..
Output of cmake was:

Code: Select all

$ cmake .
-- Compiler: GNU, version: 11.3.0
-- prefix: /usr/local
-- bindir: bin
-- datadir: share
-- docdir: share/doc/FreeCAD
-- includedir: include
-- libdir: lib
-- cmake: 3.22.1
-- Compiling with Qt 5
-- Checking for connection to GitHub...
-- GitHub connection established for FetchContent
-- Module support is disabled.
-- Version: 9.1.0
-- Build type: 
-- CXX_STANDARD: 17
-- Required features: cxx_variadic_templates
-- fmt was downloaded using FetchContent into /home/bruce/Apps/freecad-build/_deps/fmt-src
-- PyCXX found:
--   Headers:  /home/bruce/Apps/freecad-source/src
--   Sources:  /home/bruce/Apps/freecad-source/src/CXX
--   Version:  7.1.7
-- -- Found OCE/OpenCASCADE version: 7.5.1
-- -- OCE/OpenCASCADE include directory: /usr/include/opencascade
-- -- OCE/OpenCASCADE shared libraries directory: /usr/lib/x86_64-linux-gnu
-- The imported target "vtkParseOGLExt" references the file
   "/usr/bin/vtkParseOGLExt-7.1"
but this file does not exist.  Possible reasons include:
* The file was deleted, renamed, or moved to another location.
* An install or uninstall procedure did not complete successfully.
* The installation package was faulty and contained
   "/usr/lib/cmake/vtk-7.1/VTKTargets.cmake"
but not all the files it references.

-- The imported target "vtkRenderingPythonTkWidgets" references the file
   "/usr/lib/x86_64-linux-gnu/libvtkRenderingPythonTkWidgets.so"
but this file does not exist.  Possible reasons include:
* The file was deleted, renamed, or moved to another location.
* An install or uninstall procedure did not complete successfully.
* The installation package was faulty and contained
   "/usr/lib/cmake/vtk-7.1/VTKTargets.cmake"
but not all the files it references.

-- The imported target "vtk" references the file
   "/usr/bin/vtk"
but this file does not exist.  Possible reasons include:
* The file was deleted, renamed, or moved to another location.
* An install or uninstall procedure did not complete successfully.
* The installation package was faulty and contained
   "/usr/lib/cmake/vtk-7.1/VTKTargets.cmake"
but not all the files it references.

-- The imported target "pvtk" references the file
   "/usr/bin/pvtk"
but this file does not exist.  Possible reasons include:
* The file was deleted, renamed, or moved to another location.
* An install or uninstall procedure did not complete successfully.
* The installation package was faulty and contained
   "/usr/lib/cmake/vtk-7.1/VTKTargets.cmake"
but not all the files it references.

-- VTK components: vtkCommonCore;vtkCommonDataModel;vtkFiltersVerdict;vtkIOXML;vtkFiltersCore;vtkFiltersGeneral;vtkIOLegacy;vtkFiltersExtraction;vtkFiltersSources;vtkFiltersGeometry;vtkIOMPIParallel;vtkParallelMPI;vtkhdf5;vtkFiltersParallelDIY2;vtkRenderingCore;vtkInteractionStyle;vtkRenderingFreeType;vtkRenderingOpenGL2
-- Check for medfile (libmed and libmedc) ...
-- We guess that libmed was built using hdf5-openmpi version
-- Found Boost: /usr/lib/x86_64-linux-gnu/cmake/Boost-1.74.0/BoostConfig.cmake (found suitable version "1.74.0", minimum required is "1.65") found components: filesystem program_options regex system thread date_time 
-- Found Netgen: /usr/lib/cmake/netgen
-- Found NETGEN version 6.2, calculated: 393728
-- Set up to compile with Qt 5.15.3
-- Shiboken2Config: Using default python: .cpython-310-x86_64-linux-gnu
-- SHIBOKEN_PYTHON_INCLUDE_DIRS computed to value: '/usr/include/python3.10'
-- SHIBOKEN_PYTHON_LIBRARIES computed to value: ''
-- libshiboken built for RelWithDebInfo
-- PYTHON_CONFIG_SUFFIX: .cpython-310-x86_64-linux-gnu
-- libshiboken built for RelWithDebInfo
-- PySide 5.15.2 Python module found at /usr/lib/python3/dist-packages/PySide2.

-- Found Matplotlib: /usr/lib/python3/dist-packages/matplotlib (found version "3.5.1")
-- Platform is 64-bit, set -D_OCC64
-- Performing Test _flag_found
-- Performing Test _flag_found - Success
-- Performing Test _flag_found
-- Performing Test _flag_found - Success
-- Performing Test _flag_found
-- Performing Test _flag_found - Success
-- Performing Test _flag_found
-- Performing Test _flag_found - Success
-- [E57] Revison ID: E57Format-2.2.1-x86_64-linux-gcc11
-- [E57] Building static library
-- setting gcc options: -Wall -Werror -Wno-deprecated -pedantic-errors
-- Found Boost: /usr/lib/x86_64-linux-gnu/cmake/Boost-1.74.0/BoostConfig.cmake (found version "1.74.0") found components: python310 
-- found Boost: 1_74
-- boost-incude dirs are: /usr/include
-- boost-python lib is: 
-- boost_LIBRARY_DIRS is: /usr/lib/x86_64-linux-gnu
-- Boost_LIBRARIES is: Boost::python
-- area module (for Path Workbench) will be installed to: lib
-- Performing Test _flag_found
-- Performing Test _flag_found - Success
-- Performing Test _flag_found
-- Performing Test _flag_found - Success
<string>:1: DeprecationWarning: The distutils package is deprecated and slated for removal in Python 3.12. Use setuptools or check PEP 632 for potential alternatives
<string>:1: DeprecationWarning: The distutils.sysconfig module is deprecated, use sysconfig instead
-- Coin3D doc is not installed

   ==============
     System
   ==============

-- bindir:                      bin
-- BLAS:                        -undefined-
-- CMAKE_VERSION:               3.22.1
-- Compiler:                    /usr/bin/c++ (11.3.0)
-- datadir:                     share
-- docdir:                      share/doc/FreeCAD
-- includedir:                  include
-- libdir:                      lib
-- prefix:                      /usr/local
-- Python:                      3.10.6 [/usr/bin/python3.10] Suffix: [.cpython-310-x86_64-linux-gnu]

   ==============
     Config
   ==============

-- BUILD_FEM:                   ON
-- BUILD_GUI:                   ON
-- BUILD_SMESH:                 ON
-- BUILD_START:                 ON
-- BUILD_TECHDRAW:              ON
-- BUILD_TEST:                  ON
-- BUILD_VR:                    OFF 
-- BUILD_WEB:                   ON
-- CMAKE_BUILD_TYPE:            OFF 
-- CMAKE_CXX_FLAGS:             -Wall -Wextra -Wno-write-strings 
-- CMAKE_CXX_STANDARD:          17
-- CMAKE_INSTALL_PREFIX:        /usr/local
-- CMAKE_PREFIX_PATH:           -undefined-
-- FREECAD_CREATE_MAC_APP:      -undefined-
-- FREECAD_LIBPACK_USE:         OFF 
-- FREECAD_QT_VERSION:          Auto
-- FREECAD_USE_EXTERNAL_KDL:    OFF 
-- FREECAD_USE_EXTERNAL_SMESH:  OFF 
-- FREECAD_USE_FREETYPE:        ON
-- FREECAD_USE_PYBIND11:        OFF 
-- PYTHON_EXECUTABLE:           /usr/bin/python3.10
-- PYTHON_LIBRARY:              /usr/lib/x86_64-linux-gnu/libpython3.10.so
-- USE_CUDA:                    -undefined-
-- USE_OPENCV:                  -undefined-

   ==============
     Libraries
   ==============

-- Boost:                       1.74.0
-- Coin3D:                      4.0.0 [/usr/lib/x86_64-linux-gnu/libCoin.so] [/usr/include]
-- Coin3D_DOC:                  not found
-- DesignerPlugin:              not built (BUILD_DESIGNER_PLUGIN is OFF)
-- Doxygen:                     1.9.1 Language: English
-- Eigen3:                      3.4.0
-- fmt:                         Sources downloaded to /home/bruce/Apps/freecad-build/_deps/fmt-src
-- Freetype:                    2.11.1
-- HDF5:                        1.10.7
-- Matplotlib:                  3.5.1 PathDirs: /usr/lib/python3/dist-packages/matplotlib
-- MEDFile:                     4.1.0 [/usr/lib/x86_64-linux-gnu/libmedC.so] [/usr/include]
-- NETGEN:                      6.2.0 (6.2-dev) [-DNO_PARALLEL_THREADS;-DOCCGEOMETRY;-DNETGEN_VERSION=393728] [] [/usr/include/netgen/include;/usr/include/netgen] [nglib] [/usr/include/netgen/include;/usr/include/netgen]
-- OCC:                         7.5.1 [/usr/lib/x86_64-linux-gnu] [/usr/include/opencascade]
-- OCC_Libs:                    [TKFillet;TKMesh;TKernel;TKG2d;TKG3d;TKMath;TKIGES;TKSTL;TKShHealing;TKXSBase;TKBool;TKBO;TKBRep;TKTopAlgo;TKGeomAlgo;TKGeomBase;TKOffset;TKPrim;TKSTEPBase;TKSTEPAttr;TKSTEP209;TKSTEP;TKHLR;TKFeat]
-- OpenGL_Lib:                  [/usr/lib/x86_64-linux-gnu/libGL.so]
-- OpenGLU_Incl:                [/usr/include]
-- OpenGLU_Lib:                 [/usr/lib/x86_64-linux-gnu/libGLU.so]
-- pivy:                        0.6.5
-- pybind11:                    not enabled
-- PYCXX:                       7.1.7 Incl: /home/bruce/Apps/freecad-source/src Src:/home/bruce/Apps/freecad-source/src/CXX
-- PySide:                      5.15.2 [/usr/lib/../include/PySide2]
-- PySideTools:                 v: 2  uic: [/usr/lib/qt5/bin/uic]  rcc: [/usr/lib/qt5/bin/rcc]
-- QtConcurrent:                5.15.3
-- QtCore:                      5.15.3
-- QtNetwork:                   5.15.3
-- QtOpenGL:                    5.15.3
-- QtPrintSupport:              5.15.3
-- QtSvg:                       5.15.3
-- QtUiTools:                   5.15.3
-- QtWebEngineWidgets:          5.15.9
-- QtWidgets:                   5.15.3
-- QtXml:                       5.15.3
-- Rift:                        not enabled (BUILD_VR)
-- Shiboken:                    5.15.2 [/usr/lib/../include/shiboken2;/usr/include/python3.10]
-- SMESH:                       7.7.1.0 build internal
-- SPNAV:                       [/usr/lib/libspnav.so] [/usr/include]
-- SWIG:                        4.0.2
-- VTK:                         7.1.1
-- XercesC:                     3.2.3 [/usr/lib/x86_64-linux-gnu/libxerces-c.so] [/usr/include]
-- ZLIB:                        1.2.11

=================================================
Now run 'cmake --build /home/bruce/Apps/freecad-build' to build FreeCAD
=================================================

-- Configuring done
-- Generating done
-- Build files have been written to: /home/bruce/Apps/freecad-build
Edit: It seems there are issues with netgen, found a thread in 2019, that never concluded about issues. And my version of netgen (today) is older than what they were talking about!
thschrader
Veteran
Posts: 3129
Joined: Sat May 20, 2017 12:06 pm
Location: Germany

Re: Getting started in FEM, pressurized vessel

Post by thschrader »

BDL42 wrote: Tue May 02, 2023 6:22 pm ...
Edit: It seems there are issues with netgen, found a thread in 2019, that never concluded about issues. And my version of netgen (today) is older than what they were talking about!
My FC is version 0.20 portable.
I am writing from another machine.
Try GMSH:
Max element size 1,5 mm, no mesh refinements.
As I did with Netgen.

The inner face of the end-cap is approx 7,3 cm^2,
Times 30 MPa gives around 21 kN resulting force in y-direction,
see dat file after running sim. This is you checksum.

I will try gmsh tomorrow...

Regards Thomas
Post Reply