No abstract

Wall static-pressure and longitudinal-velocity fluctuations are measured in a pressure-induced turbulent separation bubble generated on a flat test surface by a combination of adverse and favorable pressure gradients. The Reynolds number, based on momentum thickness upstream of separation, is Re θ ≃ 5000 at a free-stream velocity of U ref 25 m∕s. The results indicate that the flow is characterized by two separate time-dependent phenomena: a lowfrequency mode, with a Strouhal number St 1 ≃ 0.01, which is related to a global "breathing" motion (i.e., contraction/ expansion) of the separation bubble, and a higher-frequency mode, with a Strouhal number St 2 ≃ 0.35, which is linked to the roll-up of vortical structures in the shear layer above the recirculating region and their shedding downstream of the bubble. These two phenomena are reminiscent of the "flapping" and "shedding" modes observed in fixed-separation experiments, though their normalized frequencies are different. The breathing mode is also shown to be strikingly similar to the low-frequency unsteadiness observed in shock-induced separated flows at supersonic speeds. Nomenclaturepower spectrum of movable sensor G xx = measured power spectrum of reference (fixed) sensor G yy = measured power spectrum of movable sensor h = maximum height of dividing streamline L = length scale in shock-induced separated flows L b = length of separation bubble, which is equal to 0.42 m p = static pressure p 0 = fluctuating static pressure Re θ = Reynolds number based on momentum thickness St, St 1;2 = Strouhal number, which is equal to fL b ∕U ref St f , St s = Strouhal number of flapping and shedding modes, which is equal to fx R ∕U ∞ St L = Strouhal number in shock-induced separated flows, which is equal to fL∕U ∞ St δ ω = Strouhal number of free shear layers, which is equal to fδ ω ∕ U T C = time constant of constant-voltage anemometer circuit Tu = turbulence level U = longitudinal velocity U = average velocity in the shear layer, which is equal to U max ∕2 U c = convective velocity U s = mean longitudinal velocity of inviscid flow at separation V w = voltage across hot-wire probe x = longitudinal position in test section x R = length scale in fixed-separation flows x det = longitudinal position of transitory detachment on test-section centerline, which is equal to 1.75 m x = nondimensional longitudinal position in test section, which is equal to x − x det ∕L b x D = short-time average of nondimensional instantaneous detachment position x R = short-time average of nondimensional instantaneous reattachment position y = vertical position under test surface, positive going down z = spanwise position in test section δ = 99% boundary-layer thickness δ ω = shear-layer vorticity thickness, which is equal to U max − U min ∕∂U∕∂y max γ = forward-flow fraction γ 0 = short-time average of forward-flow fraction ν = kinematic viscosity θ = momentum thickness ρ = air density Subscripts ref = measurement at wind-tunnel reference location (center of contraction exit area) rms = root mean...

The design, construction, and validation of a new academic wind tunnel is described in detail. The wind tunnel is of a classical, blow-down type and generates a pressure-induced, turbulent separation bubble on a flat test surface by a combination of adverse and favorable pressure gradients. The Reynolds number, based on momentum thickness just upstream of separation, is Re θ  5,000 at a free-stream velocity of U ref = 25ms -1 . The length of the separation bubble is estimated at 0·42 ± 0·02m by three different methods. Results of a numerical simulation demonstrate the absence of flow separation in the wind-tunnel contraction. This results in a turbulence level of about 0·05% in the test section. Oil-film visualisation experiments show that the flow near the wall is strongly three-dimensional in the recirculating region and that the topology of the limiting streamlines is consistent with experiments performed on configurations with fixed separation. Finally, spatial variations of the forward-flow fraction have been documented using a thermal-tuft probe and are shown to compare well with the results of the oil-film visualisation.

The validation and testing of a thermal-tuft probe is described in detail. The thermal tuft consists of three parallel wires where the middle wire is heated and the two lateral wires act as resistance thermometers, thereby sensing the flow direction. The probe's function principle is validated in an acoustic resonator that generates a nearly sinusoidal velocity perturbation with zero mean. It is shown that the variation in electrical resistance of the sensing wires is a measure of the flow direction. The probe's sensitivity to the heater current in the central wire and to the flow angle is also investigated. The electronic circuit is validated by placing the probe on a mechanical shaker. The output voltage is shown to be consistent with the variation in electrical resistance of the sensing wires. The flow direction can thus simply be measured by recording the probe's output voltage with a single digital data-acquisition channel. Finally, the thermal tuft is evaluated in a low-speed, pressure-driven, turbulent, separation-bubble flow. It is shown that the forward-flow fraction and the intermittent frequency can be measured with an uncertainty of about ±1.5%. The positions of separation and reattachment in the test section, measured with the thermal tuft, are consistent with flow-visualization experiments reported elsewhere.

No abstract