New DegenGeom Format

[Jason Fugate]

I have a question about the DegenGeom csv file being used in OpenVSP 3.13.3 and VSPAero 4.2. I noticed that there is a new section within the DegenGeom file, which specifies the following parameters:

"lex,ley,lez,tex,tey,tez,cgShellx,cgShelly,cgShellz,cgSolidx,cgSolidy,cgSolidz,toc,tLoc,chord,Ishell11,Ishell22,Ishell12,Isolid11,Isolid22,Isolid12,sectArea,sectNormalx,sectNormaly,sectNormalz,perimTop,perimBot,u,t00,t01,t02,t03,t10,t11,t12,t13,t20,t21,t22,t23,t30,t31,t32,t33,it00,it01,it02,it03,it10,it11,it12,it13,it20,it21,it22,it23,it30,it31,it32,it33,toc2,tLoc2,anglele,anglete,radleTop,radleBot,"

Without this section, VSPAero produces the following error: "GeomIDFlags are NOT defined!" As I have mentioned previously, I am developing Matlab functions capable of converting an *.hrm wireframe geometry directly to a DegenGeom file and running VSPAero's VLM solver manually. The code that I had developed for previous versions no longer works with the new release because of this error, but I'm wondering if it's necessary to make all of these calculations or not. Inserting dummy values for the entire section produces a segmentation fault, so some of the parameters are obviously utilized in the solver. I was hoping that you might be able to clarify which values are critical to the analysis and which could be omitted/replaced with a dummy value.

Thank you!

Jason

[Jason Fugate]

Apologies, I now realize that only the following parameters were added to the previously existing section:

"toc2,tLoc2,anglele,anglete,radleTop,radleBot,"

Could you instead clarify the definitions of these parameters so that I can make the necessary calculations?

Thank you!

[Rob McDonald]

Sure,

toc2 is a second method to calculate the airfoil t/c. tLoc2 is the
corresponding location of the maximum thickness as calculated by the
new method.

The toc and tLoc variables that are already there estimate t/c and its
location by only considering the discrete points in the wireframe.
More resolution will lead to a more exact estimate.

The toc2 and tLoc2 variables are calculated more exactly by
considering the thickness of the Bezier curves -- not just the
wireframe points. I don't think VSPAERO uses these quantities, but
it might. In practice, toc2 and toc are very similar.

anglele and anglete are the leading edge angle and trailing edge
angle. For most rounded leading edges, the anglele will be 180.0.
For a typical airfoil, anglete will be something like 10 degrees.

These are calculated based on the underlying Bezier curve -- you could
certainly approximate them from a wireframe. VSPAERO does not use
these numbers.

radleTop and radleBot are the leading edge radius (over chord)
calculated just above and below the LE point. Imagine a biconvex
airfoil with a sharp LE. The radius is undefined at the LE, but it
does exist just above/below the LE point. You can check for sharpness
by looking at anglele. The radius numbers are calculated based on the
underlying Bezier curve -- not just from the wireframe. Curvature is
related to second derivatives -- this will be tough to estimate from a
course wireframe.

The latest version of VSPAERO added a leading edge suction model based
on Carlson NASA TP 3557. This model needs an estimate of the leading
edge radius. This was the motivation for adding all these quantities.

Rob

> --
> You received this message because you are subscribed to the Google Groups
> "OpenVSP" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to openvsp+u...@googlegroups.com.
> For more options, visit https://groups.google.com/d/optout.