UAV-tutorial / compressible-solver

[oliveroxtoby]

Hi guys,
I do not know if this fits in here, but I did not want to start a new thread.
I have all tutorials through and am quite familiar with the cfdOF -wb now. Only at the 4th (the UAV tutorial) I can not get on. I did everything as described but I always get an error message and the solver does not expect anything useful (see screenshot).

What I have done:
PhysicsModel: Stady, turbulent case.
FluidProperties: Air
InitialiseFields: Potential flow
openFOAM: parallel true (4), conv.criteria 1e-3, 2000 iterations
Inlet: y = 5m/s and z = -20m/s

Instead of the slice operator I have used XOR (you can see it on the tutorial screenshot as well even if there is written "slice") and it only works with XOR otherwise the model disappears! -> The mesh was well generated (looks like the tutorial)

I don't know why it does not work.
The file is too large to attach.

Maybe double-check your inlet condition. It looks like the solver is probably not seeing any inlet velocity. Otherwise if you can find a way to share the fcstd file, that would probably be the quickest way to get an answer.

[shIxx]

Otherwise if you can find a way to share the fcstd file, that would probably be the quickest way to get an answer.

I have uploaded it to google drive:
https://drive.google.com/drive/folders/ ... 5_mvcKfOn7

[oliveroxtoby]

Otherwise if you can find a way to share the fcstd file, that would probably be the quickest way to get an answer.

I have uploaded it to google drive:
https://drive.google.com/drive/folders/ ... 5_mvcKfOn7

Strangely when I open the file, write the mesh and run the solver, it runs with no problem. Can you paste your system info (Help | About) and attach the complete output of your report view please?

[shIxx]

Of course, thank you for your helpI.

OS: Windows 10
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.18.14825 (Git)
Build type: Release
Branch: master
Hash: a0ff629747d5d73567075821852a8cdaadcf51c5
Python version: 2.7.14
Qt version: 4.8.7
Coin version: 4.0.0a
OCC version: 7.2.0
Locale: German/Germany (de_DE)

I hope you mean the output window "reportview" in the German translation. At the very beginning when I start FC already comes an error and the cfd-wb is not displayed to me although it is installed. that's been around for about a year, since the download on the addon manger works and blueCFD-core is for download in the settings. I think that has nothing to do with this problem but still it is strange.

[oliveroxtoby]

Of course, thank you for your help

OS: Windows 10
Word size of OS: 64-bit
Word size of FreeCAD: 64-bit
Version: 0.18.14825 (Git)
Build type: Release
Branch: master
Hash: a0ff629747d5d73567075821852a8cdaadcf51c5
Python version: 2.7.14
Qt version: 4.8.7
Coin version: 4.0.0a
OCC version: 7.2.0
Locale: German/Germany (de_DE)

I hope you mean the output window "reportview" in the German translation. At the very beginning when I start FC already comes an error and the cfd-wb is not displayed to me although it is installed. that's been around for about a year, since the download on the addon manger works and blueCFD-core is for download in the settings. I think that has nothing to do with this problem but still I find it strange.

Thanks for the output. It looks as if you installed cfMesh a while ago before we made some modifications. As a result, although the mesh looks good, the process is terminating before it is fully complete. If you re-do the cfMesh installation from the Preferences window, I think this should solve your problem.

[oliveroxtoby]

I hope you mean the output window "reportview" in the German translation. At the very beginning when I start FC already comes an error and the cfd-wb is not displayed to me although it is installed. that's been around for about a year, since the download on the addon manger works and blueCFD-core is for download in the settings. I think that has nothing to do with this problem but still it is strange.

The error is generated by the CFD workbench. This is a separate development from the CfdOF workbench so I can't really comment.

[shIxx]

Thanks for the output. It looks as if you installed cfMesh a while ago before we made some modifications. As a result, although the mesh looks good, the process is terminating before it is fully complete. If you re-do the cfMesh installation from the Preferences window, I think this should solve your problem.

Thank you it works.
Straight forward. You look at the stystem info and the output file and you know whats the problem .
You are really a professional. Thumbs up!

I'm surprised that so far all other simulations worked well with this broken installation. Anyway, I reinstalled everything (blueCFD, cfMesh and HiSA)
As I installed cfmesh just showed:

Code: Select all

downloading cfmesh.
extract ...
building .... please wait
and then Log: C:/ ...........................

I thought it was done and it just shows me where to find the logfile, so I hit "Apply". Then the mesher did not worked anymore and broke off with an error. So I installed cfMesher again and found out that something is happening in the output window. I waited until finally "insatllation complete" was showen.
That took almost 15 minutes ... is this normal for a 15mb file?
Same for HiSA and this had only 0.25mb downloaded.

[oliveroxtoby]

So I installed cfMesher again and found out that something is happening in the output window. I waited until finally "insatllation complete" was showen.
That took almost 15 minutes ... is this normal for a 15mb file?
Same for HiSA and this had only 0.25mb downloaded.

The code has to be compiled in the OpenFOAM environment - this is what takes the time.

[thschrader]

Wind on building, comparison of Hisa/simpleFoam solvers.
House 8x8x6,5 m (simulation at half body for saving computing time)
Inlet: v=42 m/s (150 km/h)
Fluid air, temperature 290 K.
Mesher: snappyhexmesh with refinement zones.

Hisa: resulting force on building 72 kN, drag-coefficient 1,27
simplefoam: 65 kN force, drag coefficient 1,14.
According to my DIN-manual, the correct drag-coefficient is between
1,2-1,3 (depending on roof-tilt)

Hisa runs longer, but gives a more accurate p-residual, which is needed for
the force calculation.

In my opinion Hisa can be used for such calculations too, not "only" for
aircraft-design.

house.JPG (78.73 KiB) Viewed 1671 times

hisa.JPG (62.5 KiB) Viewed 1671 times

simplefoam.JPG (62.13 KiB) Viewed 1671 times

[oliveroxtoby]

Wind on building, comparison of Hisa/simpleFoam solvers.
House 8x8x6,5 m (simulation at half body for saving computing time)
Inlet: v=42 m/s (150 km/h)
Fluid air, temperature 290 K.
Mesher: snappyhexmesh with refinement zones.

Hisa: resulting force on building 72 kN, drag-coefficient 1,27
simplefoam: 65 kN force, drag coefficient 1,14.
According to my DIN-manual, the correct drag-coefficient is between
1,2-1,3 (depending on roof-tilt)

Hisa runs longer, but gives a more accurate p-residual, which is needed for
the force calculation.

In my opinion Hisa can be used for such calculations too, not "only" for
aircraft-design.
house.JPG
hisa.JPG
simplefoam.JPG

Interesting observation, thanks. The flux-splitting scheme used does have a low Mach correction and seems to perform quite well under those conditions, but we haven't studied it comprehensively so I am wary about making any promises.