June 2024 - Nektar-users - imperial.ac.uk

Forcing / momentum source for incompressible Navier-Stokes
by Kumi Yasuda 28 Jun '24

28 Jun '24

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Hi all, When using the incompressible Navier-Stokes solver, is it possible to add a force that is a function of spatial location, for example a localised "ideal fan" momentum source in a specific region? Could this force also be a function of time and/or of the momentum at a (different) given point or region? Kind Regards, Kumi

3 6

Merging junction example for PulseWaveSolver
by Riemer Kai 27 Jun '24

27 Jun '24

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Hello Nektar users, I would like to simulate a merging junction using the PulseWaveSolver but struggle to do so. As a guide I used the Merging.xml file in PulseWaveSolver > Examples > Merging.xml, however the example is not a merging junction. I confirmed this through the following print command in PulseWaveSystem.cpp which returns 0 (and 1 for bifurcations) cout << "Number of merging junctions: " << m_mergingJcts.size() << std::endl; So, I decided to start from scratch. I can construct any arbitrary network geometry with segments and bifurcations but not if I add a merging junction. The reason my merging junction cannot converge is because P_Au becomes -nan I have tried many different configurations but did not succeed hence I finally decided to reach out. Do you have any suggestion for me or in the best case a minimal working example of a merging junction model.xml file? Best wishes, Kai

1 0

Using FieldConvert checkpoints with "extract" module
by Nande, Ishan A. 27 Jun '24

27 Jun '24

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Hello, I am trying to use the "extract" module to sample some data on a boundary ID. In standalone mode, the module works as intended but I am encountering segmentation fault while using it in FieldCovert checkpoints. I am supplying the settings as a separate file and this is my setup: <NEKTAR> <FILTERS> <FILTER TYPE="FieldConvert"> <PARAM NAME="OutputFile">session.fld</PARAM> <PARAM NAME="OutputFrequency">2</PARAM> <PARAM NAME="Modules">extract:bnd=1</PARAM> </FILTER> </FILTERS> </NEKTAR> Can someone please point out what 's wrong in this setup ? Regards, Ishan PGR

4 4

linear mesh requirements for NekMesh CAD projection
by Kumi Yasuda 21 Jun '24

21 Jun '24

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Hello, Section "4.5.18 Mesh projection" of the user guide suggests that CAD projection can be used to refine a linear mesh using STEP geometry. When I tried to do this, NekMesh gave the warning: "This module is designed for use with Star-CCM+ meshes only; it also requires that the star mesh was created in a certain way." Does anyone know what this "certain way" involves? Are there some criteria that, if satisfied, would allow linear meshes generated without using Star-CCM+ to be refined using CAD projection? Regards, Kumi

1 0

NekMesh: Mesh generation not working
by Alexander Schukmann 18 Jun '24

18 Jun '24

******************* This email originates from outside Imperial. Do not click on links and attachments unless you recognise the sender. If you trust the sender, add them to your safe senders list https://spam.ic.ac.uk/SpamConsole/Senders.aspx to disable email stamping for this address. ******************* Hello everyone, since NekMesh doesn't seem to like my 3D mesh generated with ICEM and Gmsh (see thread "NekMesh: Inconsistent triangular face orientation"), I tried creating it directly with NekMesh by using a mesh configuration file or mcf ("mesh_configuration_file.txt"). However, this doesn't properly work either when I activate the boundary layer refinement or blr feature. I attached all the necessary files converted to txt. I provided the geometry as *.step ("cylinder_geom_step_file1.txt") to the mcf and it works without blr. As soon as I switch on blr it throws the following warning, aborting the volume meshing process: Fatal : Level 0 assertion violation not setup for curve bl refinement [InputMCF] WARNING: Volume meshing has failed with message: [InputMCF] WARNING: Level 0 assertion violation not setup for curve bl refinement [InputMCF] WARNING: The linear surface mesh be written as a manifold mesh If I include all the domain boundaries except for the two side faces along the cylinder axis, i.e. writing <P PARAM="BndLayerSurfaces" VALUE="1,2,3,4,6,7,8,9" /> with 5 and 10 being the two side faces, it works. But obviously, I only need blr on the cylinder surface and not anywhere else. In a second attempt I loaded the geometry into Gmsh first, defined all the physical groups ("geo_file.txt") and again exported as *.step ("cylinder_geom_step_file2.txt"). This doesn't work at all, neither with nor without blr activated in the mcf and produces the following warning before crashing entirely: Fatal : Level 0 assertion violation surface mesh connectivity error [InputMCF] WARNING: Surface meshing has failed with message: [InputMCF] WARNING: Level 0 assertion violation surface mesh connectivity error [InputMCF] WARNING: Any surfaces which were successfully meshed will be written as a manifold mesh. [InputMCF] - Elapsed time: 4.78223s. [OutputNekpp] Writing Nektar++ file 'CylReD3900_3D_G1.xml' Fatal : Level 0 assertion violation Connected edges do not share a vertex. Edges 2347, 2345 terminate called after throwing an instance of 'Nektar::ErrorUtil::NekError' what(): Level 0 assertion violation Connected edges do not share a vertex. Edges 2347, 2345 Any help is very mush appreciated. All the best Alex Sicher versendet mit [Proton Mail](https://proton.me/).

2 1