Airfoil simulation solution interfered by mesh

Dvergr · September 23, 2014, 05:26

Hey!

First of all, I am completely new to OpenFOAM (and CFD in general) and it is quite a lot to take in. Secondly, I am not sure whether my problem has to do with the meshing or the solver, so I went with this sub-forum for now.

Problem
I am currently working towards doing some airfoil analysis with pisoFoam and pimpleFoam. Besides the basic OpenFOAM tutorials I have been given the following tutorial:
https://www.hpc.ntnu.no/display/hpc/...l+Calculations

Most of the tutorial went fine, but the results where definitely wrong. Whenever the mesh changes to a more fine mesh layer the velocity decreases to the free stream velocity or very close to it. Screenshots of the velocity field in the refinement box around the airfoil can be seen here:
Without mesh - http://i.imgur.com/mWOMcWF.png
With mesh - http://i.imgur.com/9LYJ2VD.png

Mesh
The mesh was generated from a surfaceFeaturExtract of an .stl file and snappyHexMesh. A detail of the mesh at the top of the airfoil can be seen here: http://i.imgur.com/lT7P4oB.jpg

I have tried changing the mesh in several ways including switching the fine mesh layer from the inside (the refinement box around the airfoil) to the outside (outside the refinement box). BNone of it has helped, and between two mesh layers the velocity always increases slightly on the coarse side (even when far away from the airfoil) and drops to the free stream velocity on the fine side.

Solver
I have tried running the case with simpleFoam, simpleFoam+pimpleFoam (simpleFoam data as initial condition), and lastly pisoFoam (without simpleFoam data). All solvers get approximately the same velocity field with the exact same errors.

The case files can be found here: https://www.dropbox.com/s/icsjdngktw...rfoil.zip?dl=0
The errors can easily be seen after only 10 to 20 cycles with simpleFoam, so it should be quite quick to check out possible improvements.

Does anyone have an idea what might cause this problem? Any help would be greatly appreciated. I am doing this as part of a university internship and I really need to figure this out in order to continue.

Dvergr · September 28, 2014, 03:05

I found the problem and it ended up being a stupid mistake on my part

I hadn't used the extrudeMesh feature since it was never mentioned in the airfoil tutorial. Instead I had just copied the polyMesh folder from my snappyHexMesh folder to my case folder, which definitely does not give the same results.

However, I still do not know exactly what extrudeMesh does compared to just copying the final polyMesh files, so if anyone should stumble upon this post, feel free to post a short explanation of the extrudeMesh feature (and I know it is the wrong subforum, but my original post ended up being here).

September 23, 2014, 05:26	Airfoil simulation solution interfered by mesh	#1
Dvergr New Member Martin Pedersen Join Date: Sep 2014 Location: Xi'an, China Posts: 3 Rep Power: 12	Hey! First of all, I am completely new to OpenFOAM (and CFD in general) and it is quite a lot to take in. Secondly, I am not sure whether my problem has to do with the meshing or the solver, so I went with this sub-forum for now. Problem I am currently working towards doing some airfoil analysis with pisoFoam and pimpleFoam. Besides the basic OpenFOAM tutorials I have been given the following tutorial: https://www.hpc.ntnu.no/display/hpc/...l+Calculations Most of the tutorial went fine, but the results where definitely wrong. Whenever the mesh changes to a more fine mesh layer the velocity decreases to the free stream velocity or very close to it. Screenshots of the velocity field in the refinement box around the airfoil can be seen here: Without mesh - http://i.imgur.com/mWOMcWF.png With mesh - http://i.imgur.com/9LYJ2VD.png Mesh The mesh was generated from a surfaceFeaturExtract of an .stl file and snappyHexMesh. A detail of the mesh at the top of the airfoil can be seen here: http://i.imgur.com/lT7P4oB.jpg I have tried changing the mesh in several ways including switching the fine mesh layer from the inside (the refinement box around the airfoil) to the outside (outside the refinement box). BNone of it has helped, and between two mesh layers the velocity always increases slightly on the coarse side (even when far away from the airfoil) and drops to the free stream velocity on the fine side. Solver I have tried running the case with simpleFoam, simpleFoam+pimpleFoam (simpleFoam data as initial condition), and lastly pisoFoam (without simpleFoam data). All solvers get approximately the same velocity field with the exact same errors. The case files can be found here: https://www.dropbox.com/s/icsjdngktw...rfoil.zip?dl=0 The errors can easily be seen after only 10 to 20 cycles with simpleFoam, so it should be quite quick to check out possible improvements. Does anyone have an idea what might cause this problem? Any help would be greatly appreciated. I am doing this as part of a university internship and I really need to figure this out in order to continue. Last edited by Dvergr; September 26, 2014 at 04:36. Reason: Added more information and link to case files

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
sliding mesh problem in CFX	Saima	CFX	46	September 11, 2021 08:38
how to set periodic boundary conditions	Ganesh	FLUENT	15	November 18, 2020 07:09
Problem with restart solution in shape_optimization.py	robyTKD	SU2 Shape Design	21	May 29, 2013 10:26
[snappyHexMesh] How to Do External Mesh for Airfoil sHM	msuaeronautics	OpenFOAM Meshing & Mesh Conversion	1	September 23, 2012 05:00
Mesh for 3 dim Geometry	Phil	FLUENT	9	July 12, 2000 05:39

September 28, 2014, 03:05		#2
Dvergr New Member Martin Pedersen Join Date: Sep 2014 Location: Xi'an, China Posts: 3 Rep Power: 12	I found the problem and it ended up being a stupid mistake on my part I hadn't used the extrudeMesh feature since it was never mentioned in the airfoil tutorial. Instead I had just copied the polyMesh folder from my snappyHexMesh folder to my case folder, which definitely does not give the same results. However, I still do not know exactly what extrudeMesh does compared to just copying the final polyMesh files, so if anyone should stumble upon this post, feel free to post a short explanation of the extrudeMesh feature (and I know it is the wrong subforum, but my original post ended up being here).