Capturing Blunt Trailing Edge

[Avijeet Acharya]

Hello,

I'm trying to model and mesh a NASA-CRM-like wing through OpenVSP (mainly using its Python API). For reference, my model units are in m. The blunt trailing edge has a thickness of around 0.013m (13 mm). I cannot go below 0.015 in the Min Edge Len setting; otherwise, OpenVSP crashes. But with that parameter set to 0.015, I can't get a water-tight mesh. (Lots of Edges>2 Tris).
I tried various techniques to refine the TE within the restrictive Min Edge Len setting, like placing WLine Source, Line Source, etc., at the tip. (Line source was split into two line sources to accommodate the crank in the TE). But they either have little to no effect or end up crashing OpenVSP.

Please kindly suggest a way to solve this issue. Thanks in advance.

Avijeet

P.S.: Is the "Symmetric Plane Splitting" flag not at all exposed to the API?

[Rob McDonald]

You will need to post your model for anyone to be able to help you.

Also, what version of OpenVSP are you using?

Rob

[Avijeet Acharya]

Thanks for replying Rob. My model is attached.
I have am using version 3.48.2 and 3.36.0 on Ubuntu.
I have spent a lot of time trying to tweak the parametrs and get into the Goldilocks zone where I was able to generate the mesh for my current geometry. The thing is this only works on version 3.36. Version 3.48.2 still crashes when I try to generate the mesh.
I am not sure whether even version 3.36 will work if I change my model's cross section or planform.

Please suggest how to work around this.

Avijeet

[Brandon Litherland]

Avijeet,

Part of the problem is likely that you have duplicate trailing edge points in your AF files.
Looking at the first station, for example, you have two upper TE points at 1.0, 0.0, 0.0 and two lower TE points at 1.0, -1.11...e-03, 0.0.

The result is that the wing Bezier surface tries to fit between two coincident points and causes a local loop as shown below. When you see artifacts like this, the AF points are almost always the culprit. You'll see them on the LE as well when there are duplicate points (not in yours) or when the LE is poorly resolved or interpolated from a bad source. 

In short, these artifacts will cause all sorts of problems for the mesh generator because it uses the Bezier surface as the source, not the displayed panels you see on the screen like CompGeom.

You should remove those duplicate points in your airfoil files and try this again.

Section 1 close-up screenshot

[Avijeet Acharya]

Brandon,

Thanks for catching that. It was happening due to a mismatch in the number of points mentioned for the upper surface and the actual number of points provided. The lower surface was okay.
However, I must add that while this helps me generate the mesh properly in version 3.36, version 3.48.2 still keeps crashing.
I'm attaching models with different TE profiles that I tried (All worked in 3.36 but not in 3.48.2).

Avijeet

[Rob McDonald]

It looks like you've found a bug (or a few).  I haven't sorted out a fix yet, but I can give you a partial workaround.

The bug is something with the handling of the outer domain and symmetry plane in CFDMesh.  If you turn those off (and only generate a mesh of the wing), it should work.

I'll try to figure this out and get it fixed.

Rob

[Avijeet Acharya]

I'll look forward to the fix, while using v3.36 in the meantime. Thanks for your replies.

Avijeet