Thoughts on conical flow asymmetry

Abstract

A thorough analysis of experimental results obtained on slender bodies of revolution at high angles of attack has revealed that conical flow asymmetry does indeed exist, but only for very slender cones, where the asymmetric flow separation starts developing well below an angle of attack of 30 deg. When the asymmetry starts at 30deg or higher angles of attack, the asymmetry is not conical even for purely laminar flow conditions, but is very similar to that occurring on cylindrical afterbodies. Furthermore, it is found that an asymmetric vortex pair cannot develop from symmetric crossflow separation unless symmetric crossflow attachment is prevented. Nomenclature c = reference length, d d = maximum diameter L = body length M = Mach number, Mc = M^ sin a N = normal force, coefficient CN = NI($JJ* /2)S n = yawing moment coefficient, Cn = nl(pJ Re = Reynolds number, usually Re = U^d/v^, RLoo = S = reference area, nd2/4 U = horizontal velocity x = axial body-fixed coordinate Y = side force, coefficient CY = Y/(pJI*/2)S a = angle of attack OCAV = angle of attack for incipient asymmetric crossflow separation 9C = cone half-angle v = kinematic viscosity p = air density = body azimuth