The numerical study of the interaction of a potential vortex with a stationary surface recently published by Kidd and Farris [1] is extended through a transformation of the boundary-value problem to Volterra integral equations. The new calculations verified the results by Kidd and Farris and improved the bounds of the critical Reynolds number Nc, beyond which no self-similar vortex flows exist, to 5.5 < Nc < 5.6 The breakdown of the self-similar motions develops through an instability in the lower boundary layer, which is indicated by two inflection points in the tangential velocity profile. At the critical Reynolds number the lower inflection point reaches the surface and indicates the beginning of boundary-layer separation in the wake-type flow. If the Stokes linearization is applied, one arrives at a new Stokes paradox. However, this “paradox” can be resolved by correcting the free-stream pressure distortion of the Stokes approximation. The new slow-motion approximation is nonlinear and yields an integral which is also free of the Whitehead paradox. The properties of the new exact solution confirm the novel flow features previously detected in almost self-similar motions, which were constructed by adjustable local boundary-layer approximations.

This content is only available via PDF.
You do not currently have access to this content.