|
[Sponsors] |
[foamyMesh] Can foamyHexMesh be used on fully defined point set? |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
March 14, 2017, 09:16 |
Can foamyHexMesh be used on fully defined point set?
|
#1 |
New Member
Philip Fackler
Join Date: Mar 2017
Posts: 2
Rep Power: 0 |
Hello all,
I'm quite new to openFOAM. For my dissertation research, my code distributes points on curves, surfaces and the interior volume of a domain. I need a way to test the validity of the point placement by running a flow solver on the points. I couldn't find an open-source mesh-free solver that wasn't just for shallow water problems. So, looking at openFOAM, foamyHexMesh looked promising as a way to get a mesh from my points (and subsequently solve on them). But as I figured out how to use it, it appears it only takes interior (volume) points as initial points and then it inserts points on the boundaries. So, in case I'm missing something (which is highly likely ), I'm going to ask here. I have a discrete (triangles) geometry definition. And I have points distributed (without any connectivity), some belonging to the interior (volume), some belonging (strictly) to a surface, some belonging to a curve (two surfaces), and some belonging to curve end points (two or more surfaces, two or more curves). Is there a way to run foamyHexMesh on all these points (without inserting any new points) to get a polyMesh conforming to the aforementioned geometry? Thank you. PS: If it's not possible with foamyHexMesh, perhaps someone could direct me to another solution (either a FOSS meshless solver that I could run on the points directly, or another option for meshing the points to a polyMesh or tetmesh). |
|
March 17, 2017, 10:42 |
|
#2 |
Senior Member
Join Date: Nov 2010
Location: USA
Posts: 1,232
Rep Power: 25 |
AFAIK there aren't many "meshless" flow solvers. There are solvers that claim to be but merely produce an easy cartesian grid in the background.
If you only need to test your point validity it wouldn't be a wild jump to just write a simple flow solver. If you don't want to go that far you could just connect your points into tets and run that as a finite-volume mesh. OpenFOAM is finite-volume, which does not lend well for use on point clouds. The whole point is to have volumes which are conservative. |
|
March 17, 2017, 12:18 |
|
#3 |
New Member
Philip Fackler
Join Date: Mar 2017
Posts: 2
Rep Power: 0 |
I think you misunderstood my question. I'm asking for a way to make a mesh from my points so that I can run a finite volume solver. I'm asking specifically about how to use foamyHexMesh. As an alternative, I'm asking generally, what can I use to make a tetmesh or Voronoi mesh from my point set.
|
|
March 17, 2017, 12:23 |
|
#4 |
Senior Member
Join Date: Nov 2010
Location: USA
Posts: 1,232
Rep Power: 25 |
No, I understood your question well. Any of the OpenFOAM meshers (foamy, snappy, blockMesh) are not designed for your inputs, and you rightly concluded they're not useful.
I don't know of any open-source (or even closed-source) meshers which will take what you have as input off-the-shelf. Putting points inside the domain, and even on the boundaries in some cases, is the job of the mesher, not the user. You can look at the netgen libraries for tet mesher code, but you will have to hack it up so it takes your point cloud as input. Honestly if I were you I'd just write your own tet mesher. If all you need is to connect the vertices and verify validity that shouldn't be terribly difficult. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
[snappyHexMesh] How to define to right point for locationInMesh | Mirage12 | OpenFOAM Meshing & Mesh Conversion | 7 | March 13, 2016 15:07 |
OpenFOAM static build on Cray XT5 | asaijo | OpenFOAM Installation | 9 | April 6, 2011 13:21 |
Env variable not set | gruber2 | OpenFOAM Installation | 5 | December 30, 2005 05:27 |
Plot temperature vs time for a defined point | FREDERIK | FLUENT | 0 | August 26, 2005 06:49 |
CFX4.3 -build analysis form | Chie Min | CFX | 5 | July 13, 2001 00:19 |