Measurements of wall pressure and streamwise velocity fluctuations in a turbulent boundary layer on a cylinder in an axial air flow (δ/a = 5.04, Reθ = 2870) have been used to investigate the turbulent flow structures in the cylindrical boundary layer that contribute to the fluctuating pressure at the wall in an effort to deduce the effect of transverse curvature on the structure of boundary layer turbulence. Wall pressure was measured at a single location with a subminiature electret condenser microphone, and the velocity was measured throughout a large volume of the boundary layer with a hotwire probe. Auto- and cross-spectral densities, cross-correlations, and conditional sampling of the pressure and streamwise velocity indicate that two primary groups of flow disturbances contribute to the fluctuating pressure at the wall: (i) low-frequency large-scale structures with dynamical significance across the entire boundary layer that are consistent with a pair of large-scale spanwise-oriented counter-rotating vortices and (ii) higher frequency small-scale disturbances concentrated close to the wall that are associated with the burst-sweep cycle and are responsible for the short-duration large-amplitude wall pressure fluctuations. A bidirectional relationship was found to exist between both positive and negative pressure peaks and the temporal derivative of u near the wall. Because the frequency of the large-scale disturbance observed across the boundary layer is consistent with the bursting frequency deduced from the average time between bursts, the burst-sweep cycle appears to be linked to the outer motion. A stretching of the large-scale structures very near the wall, as suggested by space-time correlation convection velocity results, may provide the coupling mechanism. Since the high-frequency disturbance observed near the wall is consistent with the characteristic frequency deduced from the average duration of bursting events, the bursting process provides the two characteristic time scales responsible for the bimodal distribution of energy near the wall. Because many of the observed structural features of the cylindrical boundary layer are similar to those observed in flat-plate turbulent boundary layers, transverse curvature appears to have little effect on the fundamental turbulent structure of the boundary layer for the moderate transverse curvature ratio used in this investigation. From differences that exist between the turbulence intensity, skewness, and spectra of the streamwise velocity, however, it appears that transverse curvature may enhance (i.e. energize) the large-scale motion owing to the reduced constraint imposed on the flow by the smaller cylindrical wall.

The fluctuating wall pressure on a circular cylinder in cross flow has been investigated experimentally in a water tunnel to examine the effect of yaw angle on the spectral characteristics of the wall pressure field and to provide insight into the periodic and turbulent structures of the flow. Wall pressure was measured at five azimuthal positions around the periphery of the cylinder for seven yaw angles from ␣ϭ90°͑normal flow͒ to ␣ϭ0°͑axial flow͒ at three subcritical Reynolds numbers (Re D ϭ7200, 13 500, 27 600). As the yaw angle is varied from ␣ϭ90°to 0°, large systematic and nonmonotonic variations in both the narrowband ͑periodic vortex shedding͒ and broadband ͑turbulent͒ spectral levels occur. The results provide additional insight into the structural characteristics of the yawed cylinder flow regimes over that which has been identified previously through measurements of mean wall forces and flow field characteristics. Significant Reynolds number effects were also observed over the range of the measurements, particularly at small yaw angles, that may indicate fundamental shifts in the structure of the wake or underlying regime transition mechanisms. Possible effects due to vortex-induced cylinder vibrations are not entirely clear but predominantly confined to the larger yaw angles. Simultaneous wall pressure, cylinder vibration, radiated acoustic pressure, and wake velocity measurements are required to appropriately isolate the various potential contributions to the wall pressure field and thus provide a clearer understanding of the underlying boundary layer separation, transitional wake, and fluid-structural coupling mechanisms.

Experiments were conducted to characterize the interaction between two sprays of electrically charged ethanol droplets. The micrometer-size droplet sprays were generated electrohydrodynamically by applying a high positive voltage to two adjacent parallel needles that were located above a distant, electrically grounded funnel. The resultant droplet axial and lateral velocity components and diameter were measured as a function of needle spacing and applied voltage using a Phase Doppler Particle Analyzer. Data were acquired at two axial positions below the needles' tips, for two needle spacings, four applied voltages and at a single flow rate.The results revealed that an increase in applied voltage yielded an increase in the spray charge density. This produced an increase in both the axial and lateral droplet velocity components and a decrease in the droplet Sauter mean diameter and in its variation across the spray. An increase in needle spacing yielded a decrease in the axial velocity component. The lateral velocity component and the Sauter mean diameter, however, were not noticeably affected by this increase. Photographic data established a relationship between the lateral half-width of the spray and axial distance. This was used to identify a nondimensional similarity between the axial mean velocity component and lateral position. The results collectively support that appropriate variations in the applied voltage and needle spacing can yield more spatially uniform mean velocity component and Sauter mean diameter profiles. These variations bring about increased mixing between the two needles' sprays and, thus, an enhanced development of the combined droplet spray.

No abstract