|
[Sponsors] |
Is my ABL simulation actually using my terrain file? |
|
LinkBack | Thread Tools | Search this Thread | Display Modes |
September 11, 2013, 19:11 |
Is my ABL simulation actually using my terrain file?
|
#1 |
New Member
Stephen Bosch
Join Date: Feb 2013
Location: Germany
Posts: 22
Rep Power: 13 |
I am doing a simulation with complex terrain based on the turbineSiting tutorial. After clearing numerous hurdles the simulation runs, but the result is not quite what I expect to see.
It looks as though the terrain wasn't meshed, though I can't be sure. First, here is the mesh, visualized in Paraview 3.98. This was done with blockMesh and snappyHexMesh. Somehow I expected it to be busier, but maybe this is just a result of the visualization. This is the mesh with the terrain file (STL): I took a y-normal slice. Here it is with terrain: And here is the same slice with the terrain removed: Here's a z-normal with the terrain: Now the same z-normal slice without the terrain: Now, this is a large area and the terrain isn't particularly dynamic, but I can't shake the feeling that the flow isn't reacting with the terrain. Especially the last two images -- wouldn't we expect to see changes in velocity around the hills? I've attached the output of checkMesh as well as the logs. There are some ptscotch errors, but I get those with the tutorial as well, and they don't seem to be relevant. What am I missing here? The case files can be found here. Last edited by pajot; September 11, 2013 at 19:18. Reason: Problems with attachments |
|
September 12, 2013, 08:28 |
turbineSiting tutorial is not meshing the terrain
|
#2 |
New Member
Stephen Bosch
Join Date: Feb 2013
Location: Germany
Posts: 22
Rep Power: 13 |
Okay, now I know something is wrong.
I set up two turbineSiting cases, one I left in its original state, and one from which I explicitly deleted the terrain.stl file so that it would cause a non-fatal error when running snappyHexMesh. Then I ran the Allrun script. I did a y-normal slice of U in Paraview 3.98 for each and put them side-by-side for comparison. I am unable to detect any difference: (Any apparent differences are due to optical illusion caused by the asymmetry in the image.) So this means that in the turbineSiting tutorial, the terrain.stl file is not being included in the mesh. This tutorial has been around for a while, so I'm a little surprised that nobody has noticed this yet. If you don't look critically at the result, you could be duped into thinking the velocity gradient is being caused by the terrain, but can't be if deleting the terrain file has no impact on the result. |
|
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
wmake compiling new solver | mksca | OpenFOAM Programming & Development | 14 | June 22, 2018 07:29 |
what is swap4foam ?? | AB08 | OpenFOAM | 28 | February 2, 2016 02:22 |
[swak4Foam] build problem swak4Foam OF 2.2.0 | mcathela | OpenFOAM Community Contributions | 14 | April 23, 2013 14:59 |
[swak4Foam] problem on installing swak4Foam | navid2 | OpenFOAM Community Contributions | 2 | May 30, 2012 05:32 |
[blockMesh] Axisymmetrical mesh | Rasmus Gjesing (Gjesing) | OpenFOAM Meshing & Mesh Conversion | 10 | April 2, 2007 15:00 |