Potential bug with VSPAERO stability derivatives for supersonic solutions

[Carl Johnson]

I was testing the default wing in supersonic flow in VSPAERO and the values in the *.flt and *.stab files didn't make sense. For example, it was reporting the neutral point of the wing as being hundreds of length units in front of the wing. The text files also had the total lift coefficient as being somewhere around zero, or even negative. Using the viewer the results looked reasonable; it was only the text files that had the issue.

Digging in it looks like the values for "clo" are the issue. It seems like the derivatives are calculated based on the "clo" instead of total "cl" when calculating things like CL,alpha and Cm,alpha and reporting in the text files. The values for "cli" and "cl" in the results manager seem fine. It's only the values in the *.flt and *.stab for supersonic cases that seem to have the issue; all subsonic results don't have the same problem.

[Rob McDonald]

Interesting, I'll pass on that information.

The clo value comes from a very simple model that is only appropriate for subsonic flow.  I suspect it is not getting turned off -- and it somehow blows up in supersonic flow.  I'll pass this on to Dave and see if we can get it sorted.

Were you trying the adjoint stability mode?  While it may have the same problem, I think it might report the derivatives for cli separately.

Rob

[Carl Johnson]

I ran the "steady" stability option. I did just try the "steady adjoint" and the results seemed strange from that one with the force derivatives with respect to alpha having very large magnitudes.

[Rob McDonald]

I passed your report onto the VSPAERO developer.

This is a bug.  He realized the problem a while ago, but did not remember to resolve it.  Hopefully he'll get it sorted soon.

Rob