AIAA Journal, Vol.34, No.9, 1996, p.1958 - 1960.

J. P. Hubner and N. M. Komerath
School of Aerospace Engineering
Georgia Institute of Technology
Atlanta, GA  30332-0150

 This Note describes the experimental finding of counter-rotating structures, with axes oriented approximately spanwise, located between the surface and the vortex core on a 60-degree delta wing at high incidence.  The finding is explained using laser sheet flow visualization, and spectral analysis of hot-film data on and above the surface. It is confirmed by quantitative analysis of velocity data, synchronized in phase to the signal from a hot-film.  The finding is related to previous work on the origin and effects of such fluctuations, and its implications explored.  A mechanism based on centrifugal instability is proposed.

 Previous work at this and other laboratories1-5 has shown that the vortex flowfield over a swept  wing at 25 to 35 degres angle of attack  develops organized velocity fluctuations under steady freestream conditions. The fluctuations are concentrated within a narrow frequency band, but can only be described as quasi-periodic,  with the phase not repeating exactly from cycle to cycle. These have been observed on isolated delta wings as well as on wing-bodies and full models of fighter configurations. Ref. 2 showed that the fluctuations maintain a constant value of reduced frequency (or Strouhal number) at a given angle of attack, over a large range of Reynolds number. Data at 20 degrees angle of attack near the vertical tail of an F-15 model, extrapolated from 1/32 and 1/7-scale model tests using this constant Strouhal number, were shown to match the measured fin vibration frequency on a full-scale F-15.  Flowfield studies on moderately-swept (<60 deg..) wings at angle of attack above 25 degrees shows that vortex breakdown occurs essentially at the apex. Ref. 5 showed that the fluctuations originate close to the wing surface on a 60-deg. cropped delta wing, at or upstream of the 30% root chord station.  They then amplify, and focus into a narrow frequency band. The peak  frequency decreases as the measurement location is moved downstream.  In a cross-flow plane at the wing trailing edge,  the frequency content is uniform except in the  post-burst core region, where other phenomena appear to dominate.  This Note focuses on the phenomena occurring near the surface. Redionitis et al. (Ref. 6) attributed the fluctuations to vortex shedding; however,  the velocity field over a 60-degree wing at   a < 30 deg. is steady in the mean, and vortex shedding is not  a plausible explanation.  Gursul7 proposed a "helical mode"  oscillation  in the post-burst flowfield characteristic of moderately-swept wings. The correlation with experimental evidence for sweep angles > 60 degrees  was  encouraging, but the correlation for sweep < 60 deg. was  ambiguous. While the geometry of the flow, and the two-point surface pressure correlations of Gursul7   do allow a "helical mode" description, this does not complete the physical explanation for the origin, amplification and focusing of the phenomenon. Here we report a  detailed investigation using multiple planar, surface and single-point measurement techniques. Spectral content of the fluctuations is measured using both hot-films and laser velocimetry (LV).  The velocity components are resolved using  LV, thereby avoiding the ambiguity inherent in hot-film measurements.
 Fig 1 shows the focusing and amplifying of the fluctuation energy for the 59.3° delta wing model4.  The flow unsteadiness is visualized near the surface and under the core using laser sheets illuminated by smoke introduced through various surface ports, aligned parallel and perpendicular to the surface.  Surface hot-film sensors and sensors in the flow above the wing are used to generate the spectra of the flow fluctuations and to determine the frequency of the spectral peak.  The repetition rate of patterns in the laser sheet images is counted from video frames and checked against the hot-film spectra peak, both by extrapolation to higher velocities and by direct measurement.  The structures are then captured quantitatively by laser velocimetry--phase-synchronized with a trigger generated from a surface hot-film sensor at the upstream end of the measurement grid.