Getting started in FEM, pressurized vessel

[NewJoker]

Mine is (from the distro repos) 4.8.4.

The newest one is 4.11.1, it can be downloaded from Gmsh website. There are cases when one of the meshers built into FreeCAD doesn’t work properly. Then it’s usually sufficient to use the other one (Netgen in this case) but adjustment of settings may also help. Especially the element size.

[BDL42]

Mine is (from the distro repos) 4.8.4.

The newest one is 4.11.1, it can be downloaded from Gmsh website. There are cases when one of the meshers built into FreeCAD doesn’t work properly. Then it’s usually sufficient to use the other one (Netgen in this case) but adjustment of settings may also help. Especially the element size.

That's good to know there is a more modern version of gmsh. I will look into that. It is puzzling that it fails at times, but is perfectly fine on similar models.

I was spectacularly unsuccessful getting netgen to work with a FreeCAD build. The more I "fixed things", the more fatal build errors popped up. Got to the point of trashing both the source and build directories, and doing a fresh build to get FreeCAD working at all. Basically I was getting a lot of the same errors that people in 2017 were. (I think that's the time frame of that FreeCAD thread on netgen.) Not anxious to try again, unless there's a good path forward on that. I tried the simple approach, and that just didn't work.

Edit: I got the source file, and the binary. There's no instructions with the binary. There's a bin and share directory. What do you do with them?

[thschrader]

No problems with GMSH 4.11 on my side:

Code: Select all

OS: Windows 10 Version 2009
Word size of FreeCAD: 64-bit
Version: 0.20.2.29177 +426 (Git)
Build type: Release
Branch: (HEAD detached from 0.20.2)
Hash: 930dd9a76203a3260b1e6256c70c1c3cad8c5cb8
Python 3.8.10, Qt 5.15.2, Coin 4.0.1, Vtk 8.2.0, OCC 7.6.3
Locale: German/Germany (de_DE)
Installed mods: 
  * BIM 2021.12.0
  * CfdOF 1.22.5
  * CurvedShapes 1.0.4
  * Curves 0.6.8
  * Defeaturing 1.2.0
  * DesignSPHysics 0.6.3(24-09-2022)
  * dodo 1.0.0
  * fasteners 0.4.54
  * Manipulator 1.5.0
  * Plot 2022.4.17
  * Reinforcement
  * WorkFeature

gmsh_4_11.JPG (60.02 KiB) Viewed 1038 times

[BDL42]

Ok. Had some issues with extraction of the gmsh tar.gz file. Seemed to have exploded the extraction into a single directory, and splattered itself everywhere. Finally cleaned up the mess and extracted it into an ordered are.

FreeCAD is now finally using the new gmsh. And the newer version of gmsh (4.11.1) has no problem with my file with the radiused inside corners now. Meshed in a few seconds with no problems. Now the pressure cylinder has lower stresses, simply by effectively filleting the inside corners - as theory would indicate. I wouldn't use such a cylinder at 3.5X it's nominal operating pressure, but it seems a lot better behaved!

Synopsis: Old gmsh (from package manager) failed filleted inside corners. New gmsh had no issues. Weird errors went away. What a difference! Thanks.

pt_3p5x_workingPressure_7075T6_5mmfillet.png (626.35 KiB) Viewed 969 times

[BDL42]

Is there some way to find the elements (nodes?) that have the peak stresses? My apologies for not knowing the correct nomenclature, I'm new at this. Moving a cursor over the CCX results only indicates the x,y, and z coordinates, but not the result type values. It would be very useful to annotate the view with selected element values. I think other FEA tools can do this. (Solidworks?)

I am also finding that the legend is very inconsistent. Sometimes it is present, other times not, one time there were 3 widely spaced color patches with nothing but the background color in between. If present, it seems to ALWAYS land on top of the navigation cube and is obscured. The font used is a tad too small for me to read. Where are the settings for the legends? If the legend was 80% of the size, it would be ok.

cutaway_plane_vm_5mmradius_1p2mm.jpg (41.89 KiB) Viewed 925 times

One more observation, from run to run, there can be a color shift. This makes it hard to compare different simulations because the colors have shifted.

Overall, I'm pretty pleased with this foray into using FEA on FreeCAD. Had a steep learning curve, but glad I have persisted. I have a lot more to learn yet though. I can see there is still quite an art in meshing. Minor differences in mesh size can have big effects in memory footprint, quality of mesh, and of course computation time. Using a 1mm max mesh resulted in 28GB of RAM used. The difference in the results were sub 1% between a 1.5mm and 1mm mesh, so I believe the 1.5mm mesh was adequate.

[NewJoker]

@BDL42 You may want to export the results to a .vtk file and post-process them in ParaView. There you can do such operations and much more easily. FreeCAD offers only basic postprocessing.

[BDL42]

@BDL42 You may want to export the results to a .vtk file and post-process them in ParaView. There you can do such operations and much more easily. FreeCAD offers only basic postprocessing.

I read about that. It's a possibility, but, was hoping to avoid it for my simple uses. Honestly, I'm not a very sophisticated user at this point. For super post processing, you are right, there are better tools. I've used Mayavi in the past but not ParaView.

Perhaps my expectations are too high, but I'd expect some basic point picking functionality.

Legends be rendered properly, and consistently. At the moment, in this large file, that is not the case. Often there is no legend when viewing CCX_Results. Sometimes there is. It may have to do with the model size. Small models are ok. Larger ones, are not consistent. Right now the model is 209MB. As for the number of colors in the view I posted, I was surprised to note there are only three colors. I would expect to see some documentation about that. I'm positive there are more than 3 different values. Seems rather parsimonious compared to all the rest of the visualization capabilities on FreeCAD.

The documentation at https://wiki.freecad.org/FEM_PostPipelineFromResult seems to have more features than I have, including editing the legend scale. This is not available in 0.21.0 that I am running. I understand the documentation runs behind, but it seems the features have been de-contented, since they were written. (Or no one has been available to work on them...)

[BDL42]

I slightly decreased the mesh size and found that I seem to have better looking answers. And perhaps strangely enough, the Pipeline CCX Results look better. I am now getting a full continuum for the legend, rather than only three discrete values. This is a welcome good change.

I was able to export the vtk file and use Paraview to view some of the file. It will take me a while to be able to even duplicate FreeCAD results... There's a lot going on in ParaView! I do think the rendering is more sophisticated, which is good. Still don't know how to do much of anything there, it will take some time to get any interesting results.

[thschrader]

I slightly decreased the mesh size and found that I seem to have better looking answers. And perhaps strangely enough, the Pipeline CCX Results look better. I am now getting a full continuum for the legend, rather than only three discrete values. This is a welcome good change.

I was able to export the vtk file and use Paraview to view some of the file. It will take me a while to be able to even duplicate FreeCAD results... There's a lot going on in ParaView! I do think the rendering is more sophisticated, which is good. Still don't know how to do much of anything there, it will take some time to get any interesting results.

Here is an example how use paraview for postprocessing:
https://wiki.freecad.org/Post-Processin ... h_Paraview

Done by @HarryvL
Based on FC0.19, but you get the idea.

[BDL42]

Redid my model to more accurately represent a particular air tube. Have to say, the 2nd or 3rd time around, your modeling gets more streamlined. Added the constraints and material. However, I can't seem to get the model to mesh using gmsh, V4.11.1. I have tried auto mesh, and various max cell sizes, with similar results. Attached is the model file. Here are the messages I am getting:

Code: Select all

10:33:57  Delete: Selection not restricted to one sketch and its subelements
10:36:56  Traceback (most recent call last):
  File "<string>", line 1, in <module>
<class 'ValueError'>: Datum 34.960 mm for the constraint with index 11 is invalid
10:36:56  App.getDocument('Unnamed').getObject('Sketch003').setDatum(11,App.Units.Quantity('34.960000 mm'))
10:50:32  Recompute failed! Please check report view.
10:58:42  Recompute failed! Please check report view.
11:02:10  Recompute failed! Please check report view.
11:29:56  Unknown command 'FEM_SolverCalculiX'
11:30:51  
11:30:51  Material card chosen:
    /home/bruce/Apps/freecad-build/share/Mod/Material/StandardMaterial/Steel-S335N.FCMat
11:30:51  ThermalConductivity not found in S335N
11:30:51  SpecificHeat not found in S335N
11:30:51  
11:31:03  Material card chosen:
    /home/bruce/Apps/freecad-build/share/Mod/Material/StandardMaterial/Aluminum-2024T6.FCMat
11:31:03  
11:37:14  Fem::FemMeshObjectPython: Link(s) to object(s) 'Body' go out of the allowed scope 'FEMMeshGmsh'. Instead, the linked object(s) reside within 'Part'.
11:37:46    ElementDimension: 3
11:37:46    Group meshing for analysis is set to true in FEM General Preferences. Are you really sure about this? You could run into trouble!
11:37:46    Empty reference: MaterialSolid
11:37:46  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb97f37380>
11:37:46      ConstraintFixed001
11:37:46      [(<PartDesign::LinearPattern>, ('Face3',))]
11:37:46      LinearPattern
11:37:46  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb97f37380>
11:37:46  Body--> Name of the Feature we where searching in.
11:37:46  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:37:46  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb97cf8010>
11:37:46      ConstraintPressure
11:37:46      [(<PartDesign::LinearPattern>, ('Face39', 'Face38'))]
11:37:46      LinearPattern
11:37:46  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb97cf8010>
11:37:46  Body--> Name of the Feature we where searching in.
11:37:46  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:37:46  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb99b68890>
11:37:46      ConstraintPressure001
11:37:46      [(<PartDesign::LinearPattern>, ('Face37',))]
11:37:46      LinearPattern
11:37:46  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb99b68890>
11:37:46  Body--> Name of the Feature we where searching in.
11:37:46  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:37:46    One material with no reference shapes. No need to make a group for materials.
11:37:46  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintFixed001 could not be found!
11:37:46  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintPressure001 could not be found!
11:37:46    /tmp/fcfem___j0v6dw/Body_Geometry.brep
11:37:46    /tmp/fcfem___j0v6dw/Body_Mesh.unv
11:37:46    /tmp/fcfem___j0v6dw/shape2mesh.geo
11:37:46    /home/bruce/Apps/gmsh/gmsh-4.11.1-Linux64/bin/gmsh
11:37:46  Unexpected error when creating mesh: list index out of range
11:37:46  Gmsh had warnings:
11:37:46  list index out of range
11:38:11    ElementDimension: 3
11:38:11    Group meshing for analysis is set to true in FEM General Preferences. Are you really sure about this? You could run into trouble!
11:38:11    Empty reference: MaterialSolid
11:38:11  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb9906fd70>
11:38:11      ConstraintFixed001
11:38:11      [(<PartDesign::LinearPattern>, ('Face3',))]
11:38:11      LinearPattern
11:38:11  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb9906fd70>
11:38:11  Body--> Name of the Feature we where searching in.
11:38:11  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:38:11  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb972d2bf0>
11:38:11      ConstraintPressure
11:38:11      [(<PartDesign::LinearPattern>, ('Face39', 'Face38'))]
11:38:11      LinearPattern
11:38:11  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb972d2bf0>
11:38:11  Body--> Name of the Feature we where searching in.
11:38:11  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:38:11  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb97bcb340>
11:38:11      ConstraintPressure001
11:38:11      [(<PartDesign::LinearPattern>, ('Face37',))]
11:38:11      LinearPattern
11:38:11  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb97bcb340>
11:38:11  Body--> Name of the Feature we where searching in.
11:38:11  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:38:11    One material with no reference shapes. No need to make a group for materials.
11:38:11  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintFixed001 could not be found!
11:38:11  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintPressure001 could not be found!
11:38:11    /tmp/fcfem_o2e5bqcb/Body_Geometry.brep
11:38:11    /tmp/fcfem_o2e5bqcb/Body_Mesh.unv
11:38:11    /tmp/fcfem_o2e5bqcb/shape2mesh.geo
11:38:11    /home/bruce/Apps/gmsh/gmsh-4.11.1-Linux64/bin/gmsh
11:38:11  Unexpected error when creating mesh: list index out of range
11:38:11  Gmsh had warnings:
11:38:11  list index out of range
11:38:57    ElementDimension: 3
11:38:57    Group meshing for analysis is set to true in FEM General Preferences. Are you really sure about this? You could run into trouble!
11:38:57    Empty reference: MaterialSolid
11:38:57  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb97e74090>
11:38:57      ConstraintFixed001
11:38:57      [(<PartDesign::LinearPattern>, ('Face3',))]
11:38:57      LinearPattern
11:38:57  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb97e74090>
11:38:57  Body--> Name of the Feature we where searching in.
11:38:57  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:38:57  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb97b7bf40>
11:38:57      ConstraintPressure
11:38:57      [(<PartDesign::LinearPattern>, ('Face39', 'Face38'))]
11:38:57      LinearPattern
11:38:57  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb97b7bf40>
11:38:57  Body--> Name of the Feature we where searching in.
11:38:57  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:38:57  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb99eca0a0>
11:38:57      ConstraintPressure001
11:38:57      [(<PartDesign::LinearPattern>, ('Face37',))]
11:38:57      LinearPattern
11:38:57  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb99eca0a0>
11:38:57  Body--> Name of the Feature we where searching in.
11:38:57  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:38:57    One material with no reference shapes. No need to make a group for materials.
11:38:57  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintFixed001 could not be found!
11:38:57  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintPressure001 could not be found!
11:38:57    /tmp/fcfem_ysmzfop_/Body_Geometry.brep
11:38:57    /tmp/fcfem_ysmzfop_/Body_Mesh.unv
11:38:57    /tmp/fcfem_ysmzfop_/shape2mesh.geo
11:38:57    /home/bruce/Apps/gmsh/gmsh-4.11.1-Linux64/bin/gmsh
11:38:57  Unexpected error when creating mesh: list index out of range
11:38:57  Gmsh had warnings:
11:38:57  list index out of range
11:39:12  Fem::FemMeshObjectPython: Link(s) to object(s) 'Body' go out of the allowed scope 'FEMMeshGmsh'. Instead, the linked object(s) reside within 'Part'.
11:41:22  Fem::FemMeshObjectPython: Link(s) to object(s) 'Body' go out of the allowed scope 'FEMMeshGmsh'. Instead, the linked object(s) reside within 'Part'.
11:41:31  Fem::FemMeshObjectPython: Link(s) to object(s) 'Body' go out of the allowed scope 'FEMMeshGmsh'. Instead, the linked object(s) reside within 'Part'.
11:41:48    ElementDimension: 3
11:41:48    Group meshing for analysis is set to true in FEM General Preferences. Are you really sure about this? You could run into trouble!
11:41:48    Empty reference: MaterialSolid
11:41:48  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb98927e10>
11:41:48      ConstraintFixed001
11:41:48      [(<PartDesign::LinearPattern>, ('Face3',))]
11:41:48      LinearPattern
11:41:48  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb98927e10>
11:41:48  Body--> Name of the Feature we where searching in.
11:41:48  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:41:48  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb972226c0>
11:41:48      ConstraintPressure
11:41:48      [(<PartDesign::LinearPattern>, ('Face39', 'Face38'))]
11:41:48      LinearPattern
11:41:48  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb972226c0>
11:41:48  Body--> Name of the Feature we where searching in.
11:41:48  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:41:48  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb97bd14f0>
11:41:48      ConstraintPressure001
11:41:48      [(<PartDesign::LinearPattern>, ('Face37',))]
11:41:48      LinearPattern
11:41:48  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb97bd14f0>
11:41:48  Body--> Name of the Feature we where searching in.
11:41:48  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:41:48    One material with no reference shapes. No need to make a group for materials.
11:41:48  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintFixed001 could not be found!
11:41:48  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintPressure001 could not be found!
11:41:48    /tmp/fcfem_tfbdiac0/Body_Geometry.brep
11:41:48    /tmp/fcfem_tfbdiac0/Body_Mesh.unv
11:41:48    /tmp/fcfem_tfbdiac0/shape2mesh.geo
11:41:48    /home/bruce/Apps/gmsh/gmsh-4.11.1-Linux64/bin/gmsh
11:41:48  Unexpected error when creating mesh: list index out of range
11:41:48  Gmsh had warnings:
11:41:48  list index out of range
11:41:56    /home/bruce/Apps/gmsh/gmsh-4.11.1-Linux64/bin/gmsh
11:41:56  file found: /home/bruce/Apps/gmsh/gmsh-4.11.1-Linux64/bin/gmsh
11:41:56  Gmsh: StdOut:
Version       : 4.11.1
License       : GNU General Public License
Build OS      : Linux64
Build date    : 20221221
Build host    : gmsh.info
Build options : 64Bit ALGLIB[contrib] ANN[contrib] Bamg Blas[petsc] Blossom Cgns DIntegration Dlopen DomHex Eigen[contrib] Fltk Gmm[contrib] Hxt Jpeg Kbipack Lapack[petsc] LinuxJoystick MathEx[contrib] Med Mesh Metis[contrib] Mmg Mpeg Netgen ONELAB ONELABMetamodel OpenCASCADE OpenCASCADE-CAF OpenGL OpenMP OptHom PETSc Parser Plugins Png Post QuadMeshingTools QuadTri Solver TetGen/BR Voro++[contrib] WinslowUntangler Zlib
FLTK version  : 1.4.0
PETSc version : 3.14.4 (real arithmtic)
OCC version   : 7.6.1
MED version   : 4.1.0
Packaged by   : geuzaine
Web site      : https://gmsh.info
Issue tracker : https://gitlab.onelab.info/gmsh/gmsh/issues

11:42:12    ElementDimension: 3
11:42:12    Group meshing for analysis is set to true in FEM General Preferences. Are you really sure about this? You could run into trouble!
11:42:12    Empty reference: MaterialSolid
11:42:12  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb97d43f30>
11:42:12      ConstraintFixed001
11:42:12      [(<PartDesign::LinearPattern>, ('Face3',))]
11:42:12      LinearPattern
11:42:12  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb97d43f30>
11:42:12  Body--> Name of the Feature we where searching in.
11:42:12  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:42:12  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb97c8a4a0>
11:42:12      ConstraintPressure
11:42:12      [(<PartDesign::LinearPattern>, ('Face39', 'Face38'))]
11:42:12      LinearPattern
11:42:12  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb97c8a4a0>
11:42:12  Body--> Name of the Feature we where searching in.
11:42:12  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:42:12  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb96f3b5f0>
11:42:12      ConstraintPressure001
11:42:12      [(<PartDesign::LinearPattern>, ('Face37',))]
11:42:12      LinearPattern
11:42:12  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb96f3b5f0>
11:42:12  Body--> Name of the Feature we where searching in.
11:42:12  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:42:12    One material with no reference shapes. No need to make a group for materials.
11:42:12  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintFixed001 could not be found!
11:42:12  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintPressure001 could not be found!
11:42:12    /tmp/fcfem_9rw8othc/Body_Geometry.brep
11:42:12    /tmp/fcfem_9rw8othc/Body_Mesh.unv
11:42:12    /tmp/fcfem_9rw8othc/shape2mesh.geo
11:42:12    /home/bruce/Apps/gmsh/gmsh-4.11.1-Linux64/bin/gmsh
11:42:12  Unexpected error when creating mesh: list index out of range
11:42:12  Gmsh had warnings:
11:42:12  list index out of range
11:43:24    ElementDimension: 3
11:43:24    Group meshing for analysis is set to true in FEM General Preferences. Are you really sure about this? You could run into trouble!
11:43:24    Empty reference: MaterialSolid
11:43:24  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb988c8d40>
11:43:24      ConstraintFixed001
11:43:24      [(<PartDesign::LinearPattern>, ('Face3',))]
11:43:24      LinearPattern
11:43:24  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb988c8d40>
11:43:24  Body--> Name of the Feature we where searching in.
11:43:24  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:43:24  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb9935f880>
11:43:24      ConstraintPressure
11:43:24      [(<PartDesign::LinearPattern>, ('Face39', 'Face38'))]
11:43:24      LinearPattern
11:43:24  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb9935f880>
11:43:24  Body--> Name of the Feature we where searching in.
11:43:24  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:43:24  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb988815c0>
11:43:24      ConstraintPressure001
11:43:24      [(<PartDesign::LinearPattern>, ('Face37',))]
11:43:24      LinearPattern
11:43:24  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb988815c0>
11:43:24  Body--> Name of the Feature we where searching in.
11:43:24  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:43:24    One material with no reference shapes. No need to make a group for materials.
11:43:24  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintFixed001 could not be found!
11:43:24  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintPressure001 could not be found!
11:43:24    /tmp/fcfem_7mhv3kzs/Body_Geometry.brep
11:43:24    /tmp/fcfem_7mhv3kzs/Body_Mesh.unv
11:43:24    /tmp/fcfem_7mhv3kzs/shape2mesh.geo
11:43:24    /home/bruce/Apps/gmsh/gmsh-4.11.1-Linux64/bin/gmsh
11:43:24  Unexpected error when creating mesh: list index out of range
11:43:24  Gmsh had warnings:
11:43:24  list index out of range
11:43:41    ElementDimension: 3
11:43:41    Group meshing for analysis is set to true in FEM General Preferences. Are you really sure about this? You could run into trouble!
11:43:41    Empty reference: MaterialSolid
11:43:41  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb981fe940>
11:43:41      ConstraintFixed001
11:43:41      [(<PartDesign::LinearPattern>, ('Face3',))]
11:43:41      LinearPattern
11:43:41  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb981fe940>
11:43:41  Body--> Name of the Feature we where searching in.
11:43:41  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:43:41  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb97de5ce0>
11:43:41      ConstraintPressure
11:43:41      [(<PartDesign::LinearPattern>, ('Face39', 'Face38'))]
11:43:41      LinearPattern
11:43:41  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb97de5ce0>
11:43:41  Body--> Name of the Feature we where searching in.
11:43:41  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:43:41  Problem: For the geometry of the following shape was no Shape found: <Face object at 0x55fb971d5a20>
11:43:41      ConstraintPressure001
11:43:41      [(<PartDesign::LinearPattern>, ('Face37',))]
11:43:41      LinearPattern
11:43:41  The reference Shape is not a child nor it is the shape the mesh is made of. : <Face object at 0x55fb971d5a20>
11:43:41  Body--> Name of the Feature we where searching in.
11:43:41  LinearPattern --> Name of the parent Feature of reference Shape (Use the same as in the line before and you will have less trouble :-) !!!!!!).
11:43:41    One material with no reference shapes. No need to make a group for materials.
11:43:41  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintFixed001 could not be found!
11:43:41  Error: The shapes for the mesh group for the reference shapes of analysis member: ConstraintPressure001 could not be found!
11:43:41    /tmp/fcfem_8c_bfbf3/Body_Geometry.brep
11:43:41    /tmp/fcfem_8c_bfbf3/Body_Mesh.unv
11:43:41    /tmp/fcfem_8c_bfbf3/shape2mesh.geo
11:43:41    /home/bruce/Apps/gmsh/gmsh-4.11.1-Linux64/bin/gmsh
11:43:41  Unexpected error when creating mesh: list index out of range
11:43:41  Gmsh had warnings:
11:43:41  list index out of range
11:43:54  Fem::FemMeshObjectPython: Link(s) to object(s) 'Body' go out of the allowed scope 'FEMMeshGmsh'. Instead, the linked object(s) reside within 'Part'.
12:06:53    /home/bruce/Apps/gmsh/gmsh-4.11.1-Linux64/bin/gmsh
12:06:53  file found: /home/bruce/Apps/gmsh/gmsh-4.11.1-Linux64/bin/gmsh
12:06:53  Gmsh: StdOut:
Version       : 4.11.1
License       : GNU General Public License
Build OS      : Linux64
Build date    : 20221221
Build host    : gmsh.info
Build options : 64Bit ALGLIB[contrib] ANN[contrib] Bamg Blas[petsc] Blossom Cgns DIntegration Dlopen DomHex Eigen[contrib] Fltk Gmm[contrib] Hxt Jpeg Kbipack Lapack[petsc] LinuxJoystick MathEx[contrib] Med Mesh Metis[contrib] Mmg Mpeg Netgen ONELAB ONELABMetamodel OpenCASCADE OpenCASCADE-CAF OpenGL OpenMP OptHom PETSc Parser Plugins Png Post QuadMeshingTools QuadTri Solver TetGen/BR Voro++[contrib] WinslowUntangler Zlib
FLTK version  : 1.4.0
PETSc version : 3.14.4 (real arithmtic)
OCC version   : 7.6.1
MED version   : 4.1.0
Packaged by   : geuzaine
Web site      : https://gmsh.info
Issue tracker : https://gitlab.onelab.info/gmsh/gmsh/issues

Still getting the message: "Group meshing for analysis is set to true in FEM General Preferences. Are you really sure about this? You could run into trouble!"
In preferences, there is no way to change the option, as it is set, but greyed out. There is also no way to set the default solver.

mjp_L2.FCStd

(116.08 KiB) Downloaded 27 times

pt_preferences1.png (145.01 KiB) Viewed 654 times

What is frustrating is the apparent randomness of the mesher working. Once it works, for a model, it seems to continue to work. There are messages which come out, which should be clues, but I don't know how to interpret them, nor how to make a model which is first time mesh-able.

Can someone take a brief look to see if I'm doing something grievously incorrect? I would think anyway of constructing the model would be as good as others. I happened to use a lot of grooving to remove portions of the solid, to model the piece, similar to a machining operation.