Possible Instabilities with Panel Method

[Ronald Pandolfi]

Our interns at Kashmir World Foundation recently submitted a paper for publication on design of a blended wing body aircraft using OpenVSP, and they received feedback from reviewers to compare panel mode with VLM mode. They had attempted to use panel mode during analysis, but they excluded it from the paper because of instabilities. 

I know there were instabilities with panel mode many years back, but I thought the issues had been corrected.

Attached is an update from our Aerospace Team lead. I would much appreciate guidance from someone who has properly compared panel mode with VLM mode.

V/r,

Ron

Ronald Pandolfi, Ph.D.

Chief Technology Officer

Kashmir World Foundation

 www.kashmirworldfoundation.org

 Watch the trailer: Beyond the Noise – A KwF Project to Listen, Learn, and Protect

"Extinguish your spot fires before the winds of chaos ignite a wildfire in your path."

— from the reflections of Princess Aliyah Pandolfi

[Rob McDonald]

VSPAERO was essentially unchanged from OpenVSP release 3.34.0 to 3.44.X.  It received a major update in 3.45.0.  Any problems you've had with 3.34 would still be there in 3.44.

If you are using VSPAERO for serious work, you should re-set your processes and expectations with 3.45.

When VSPAERO produced erratic load distributions like you are seeing, it is usually just an artifact of post-processing the load distribution -- it is not an indication of any problem with the solution.

That said, even with a fat BWB, the thin-surface VLM mode should be fine (usually better than the thick-surface mode) -- so long as your primary interest is integrated forces and moments.  You really only need to step up to the thick-surface formulation when you are interested in local surface pressures or velocities.

Rob