Airfoil Meshing & CFD
***11/10/23, 14:00 Edit***
Meshing issue appears to be resolved. Every subtract or combine operation had the parasolid option selected. I also extended the airfoil 10 units passed the surfaces just to be sure. Paying careful attention to the procedure, no problems arose or suggested themselves.
*** Start of Initial Post***
Hey all,
INTRO) I'm having issues, in 2 different forms, trying to run CFD on an airfoil. I'm following ACU-T: 2100 (turb. flow, airfoil) to the letter (However, I am aware there is a moderate probability I'm making an error while following regardless. I'm very conscious of the details). I'm importing a STEP AP214 file (generated by SolidWorks) of the airfoil (1mm chord, 50mm length). I generate the circular solids and alter them accordingly in HyperworksCFD. WHEN COMBINING, I have taken 2 different approaches; keeping parasolid geometry and not. Below are my approaches and their problems;
1) Keeping Parasolid Geometry
-I am able to achieve a perfect mesh (see image).
-I ran acusolve on 4 cores (I have a 2018 Dell Precision, i7, 16GB of RAM) and it took about an hour and a half to solve (maybe a little less). While running, with the Acusolve run status bar indicating ~66% complete, a HW Mesh process appeared out of nowhere and ran.
-Upon viewing the results I found that something caused the airfoil geometry to get essentially reset and what resulted was a slanted rectangle (see image).
2) Combining Without Keeping Parasolid Geometry
-I am unable to achieve a mesh. See image (Mesh Error: Volume Mesher Failed. The log output is attached)
2b) The Issue Is Clearly Airfoil Related (see image)
-When establishing Boundaries, I'm not confident the Default Wall selects what it needs to. (I later on changed the order in which I selected boundaries. This is because the default wall kept on selecting the smaller circle face when selected and removing it from the +Z Slip boundary. This had no impact and the mesh still failed)
-When establishing an Edge Layer I am able to select the airfoil. However, I select the parent surf, it highlights, I press the play button, BUT AFTER pressing the parent surf is no longer recognized and the (1) disappears.
MY QUESTION[S])
1)Does anyone have any ideas on what might be going wrong? I'm not deviating from ACU-T: 2100 to the best of my knowledge (I'm very aware of how specific & detailed the software is).
2) Is there any best practice out there that I can apply? (I'm sure ACU-T: 2100 covers 90+% of it) For example, best approach to pre-processing/meshing. Or is there an known procedure that differs from or expands upon ACU-T: 2100's approach?
3) Should I be generating the airfoil in Hyperworks or Hypermesh? If so, can you provide a detailed procedure or point to a similar procedure? I have under a year in Hyperworks/Mesh. I've spent 3+ years in MotionView, HyperView and have used hypermesh/works on rare occasions.
I've taken the utmost care in the CAD to ensure the airfoil is perfect. I shouldn't have any issues with geometry. It's just a spline, mirrored (so 2 splines) and extruded (1 volume).
***
(NOTE: The pictures shown below are also attached)
Answers
-
I would also recommend reviewing this validation:
https://help.altair.com/hwcfdsolvers/acusolve/topics/acusolve/turbulent_flow_over_a_naca_airfoil_3.htmYou may want to add some zone mesh controls to keep the mesh a bit finer outside the boundary layers, and in the wake region.
Also - you'll probably want to zoom into the leading age region to make sure the geometry is smooth rather than faceted. Sometimes it's difficult to tell what is the faceted geometry representation line and what is the resulting mesh line.
0
Altair Employee