SnappyHexMesh Internal Mesh problem
 

I trying to do a complicated internal mesh and I am running into some issues. The flow pass through contains some solid pieces which is where I think the problem lies.
I've been able to accomplish some internal meshes before and I pretty much used the same method.

I attached a shareable folder that contains the files after I ran snappyhexmesh.

I was able to run snappyhexmesh without any errors but it doesn't recognize the internal solids.



https://drive.google.com/file/d/1RJA...ew?usp=sharing


https://drive.google.com/file/d/1XRn...ew?usp=sharing





Thanks in advance

To trouble shoot I've created a simpler object and refined the tolerances on the stl file. I ran admesh --fill-holes "file" to ensure there are no holes. I still came up with this problem. The point I have chosen is in the object but worst case if it was outside the object it should still create a mesh around it.. So either there is a whole in the stl, which there shouldn't be, or something else is wrong.
Any advice would be appreciated. Thank you

Hello,
could you please grant the access to your file? I'd like to investigate it, since I worked on some similar geometries. What are the rough boundary conditions for you simulation and expected mesh accuracy?
best,
Damian

See if the link is view-able now?
I want to simulate constant volumetric flow rate. As for pressure, I want to fix the outlet pressure to atm and set the inlet to zero-gradient to see what pressure it takes at the inlet to maintain the volumetric flow.

But as of right now I need to start figuring out the mesh problem and I'll go from there..

ok, I was now able to download the file and will investigate your case in the next 2 days.

Hi,
I reviewed the geometry, it might be a challenge:)
Before we do a deep dive into the meshing some remarks in advance:
1.) try to simplify your basis geometry.
- remove not relevant geometry details (groove, see my screenshot)
- merge the springs and the sealing ring with the spool / poppet valve.
2.) use the springs models for extra refinement or use cylinder for refinement
3.) make sure, the STL files are water-tight (use the forum SEARCH function!)
4.) I guess, you will need robust boundary layer to resolve the flow, especially on the poppet valve and downstream of it?
5.) to reduce some cells, the upstream and downstream pipes can be cut for the main meshing process and be extended with the 'extrudeMesh' command and grading factor afterwards. I use a automated script for this
6.) also the lower duct / channel seems to be not needed for the simulation, can be removed to reduce cells?
7.) what is exactly your technical question? which effect should be simulated? simple pressure-drop study or more specific ? from own experience, I guess the springs might not have big impact in this configuration / spool position
8.) can you also share the STEP geometry to create a negative 3D model and check what's possible with Salome?
best,
Damian


https://www.cfd-online.com/Forums/me...9-020-pic1.png

Thanks for all those ideas. They all sound helpful!

I have the pipes upstream and downstream to create developed flow before entering the area where I expect the energy loss to occur.

But yeah a rather simple analysis. I will be running multiple geometries at similar conditions and I would like to the pressure drop for each one.

So to start and trouble shoot (no being able to snappyhexmesh to develop an internal mesh) I created a simpler geometry and still ran into the same problem. I know my internal point is within the geometry but it still doesn't develop a mesh inside the stl. It doesn't develop a mesh around it either so to me it sounds like something is is wrong with how I setup my files... It just reshows the blockmesh in paraFoam.
I went back to one of my other simulations where I created an internal mesh using snappyhexmesh, which it worked great before, and I tried to rerun it and the same problem happened...

I have no idea what is going on, any advice or experience with this?

1.) regarding the inlet/outlet extensions. That's clear, it's just my best practice to create the mesh in 2 stages. It means, to extend it with the 'extrudeMesh' command by 10x diameter, for example
2.) your STL is not of the best quality, and I guess not watertight / waterproof, check this links to see how to create perfect STL
https://www.cfd-online.com/Forums/op...sh-salome.html
https://www.youtube.com/watch?v=P4_nKarYzHA
3.) try to split the meshing domain into simple sectors by using additional dummy walls. This approach will help to detect the problematic area
4.) you could also check other tools:
- cfMesh
- GMSH
- Salome
- TCFD
- HEXPRESS/Hybrid