[jackgeo]

I am having some issues with an unsteady vortex lattice method (UVLM) program i've written for aircraft propellers and rotors. The problem being that for accurate propeller geometry such as the Masterscrew GF 10x8 (from UIUC http://m-selig.ae.illinois.edu/props/propDB.html) the trailing edge wake becomes unstable after ~5 time steps.

By unstable i refer to a massive spike in the panel vorticity on the blade and wake which results in a large displacement when a rollup routine is performed. The majority of my program is based on the work from Katz & Plotkin, however the vortex line equations are based off vortex core models (Rankine for blade and Scully for wake) and the core size is usually set to ~0.5xdA where dA is the smallest panel area on the blade.

I was able to deduce that reducing the overall pitch of the blades (aka -10deg across entire blade) produced satisfactory wake shapes but the performance calculations are therefore only valid for that propeller at a -10 pitch setting. The other way to reduce the influence was to increase the vortex core size dramatically (~5-10x) however i can't see any physical way of justifying those values.

The Code was tested with wing and full aircraft configurations and worked perfectly, as well as simple rectangular blades and tapered blades with camber and small twist but not for actual blade geometry.

Im not quite sure what the problem is, as the velocity inflow angle should reduce the effective angle of attack to something in the linear lift profile for an airfoil (where VLM appears to be valid). I feel like i may have neglected some sort of geometric conditions that VLM is sensitive to but am not aware.

Any and all help would be greatly appreciated, i am able to post images if anyone needs a better description of what i am referring to,

Thanks again to anyone who stumbles across this and feels like helping out.

[jackgeo]

I really need help if anyone can offer...