Session H20: Turbulent Boundary Layers V: Experiments

Skin-friction and Reynolds number scaling of turbulent channel flow

An experimental study was conducted on smooth-wall, fully-developed, turbulent channel flow. The Reynolds number (\textit{Re}$_{m})$ based on the channel height and the bulk mean velocity ranged from 10,000 -- 300,000. Measurements of the flow rate and the streamwise pressure gradient allowed the skin-friction coefficient ($C_{f})$ to be determined, and its variation with Reynolds number will be discussed and compared with previous investigations. Two-component LDV measurements were also made at friction Reynolds numbers \textit{Re}$_{\tau }$ = 1,000 -- 6,000. The scaling of both the mean flow and the Reynolds stresses will be examined. In particular, the variation in these quantities with Reynolds number will be discussed.   

Measurements of the wall-normal velocity component in very high Reynolds number pipe flow

Nano-Scale Thermal Anemometry Probes (NSTAPs) have recently been developed and used to study the scaling of the streamwise component of turbulence in pipe flow over a very large range of Reynolds numbers. This probe has an order of magnitude higher spatial and temporal resolution than regular hot wires, allowing it to resolve small scale motions at very high Reynolds numbers. Here use a single inclined NSTAP probe to study the scaling of the wall normal component of velocity fluctuations in the same flow. These new probes are calibrated using a method that is based on the use of the linear stress region of a fully developed pipe flow. Results on the behavior of the wall-normal component of velocity for Reynolds numbers up to 2 million are reported.   

Experimental study of the boundary layer properties in ultimate Taylor-Couette flow

We report high-resolution measurements of the properties of the velocity boundary layer in turbulent Taylor-Couette flow using time-resolved particle image velocimetry (PIV). The experiments are performed in the Twente Turbulent Taylor-Couette facility (T3C). The Taylor number is varied from 10$^8$ to 10$^{13}$, which covers the ultimate turbulence regime and the transition regime. We also change the rotation ratio of the inner and outer cylinders. The boundary layer profile, thickness, and scaling behavior are experimentally examined. In addition, the measured results are closely compared to the boundary layer properties in turbulent Rayleigh-B\'enard flow.   

New insight on flow development and two-dimensionality of turbulent channel flows

The experimental conditions required for a turbulent channel flow to be considered fully-developed and nominally two-dimensional remain a challenging objective. Oil film interferometry (OFI) and static pressure measurements were carried out over the range $200 < Re_{\tau} < 800$ in an adjustable-geometry channel flow facility. Three-dimensional effects were studied by considering different aspect ratio (AR) configurations, and also by fixing the AR and modifying the hydraulic diameter $D_{H}$ of the section. The conditions at the centerplane of the channel were characterized through the local skin friction from the OFI and the centerline velocity at three different streamwise locations, as well as the wall shear based on the streamwise global pressure gradient. The skin friction obtained from the pressure gradient overestimated the local shear measurements obtained from the OFI, and did not reproduce the same AR-dependence observed with OFI. Differences between the local and global techniques were also reflected in the flow development. Development length of high-aspect-ratio channels scales with the hydraulic diameter of the section, and is around 200 channel full-heights $H$, much larger than the values of around $100-150~H$ previously reported in the literature.   

Multi-component measurements in high Reynolds number turbulent boundary layers

Measurements, with highly resolved spectra are obtained in high Reynolds number ($Re_\tau$) turbulent boundary layers using sub-miniature cross-wires. The probe consists of $2.5 \mu$m diameter platinum wires welded across the sharpened stainless steel prong tips, contained in a volume of $0.4\times0.4\times0.2$mm$^3$. Velocity profiles are measured at various stream-wise positions with nominally matched unit Reynolds number ($U_\infty/\nu$). In this manner the same probe geometry affords approximately matched viscous-scaled sensor length ($l^+$) and sensor spacing ($\Delta s^+$) across the entire range of $Re_{\tau}$, such that Reynolds number trends can be observed free of spatial resolution effects. The probes have matched measurement volumes of approximately $14 \times 14 \times 7$ ($\pm 10\%$) viscous length scales across all $Re_\tau$. The prong is inclined at $10^\circ$ to the horizontal, permitting measurements close to the wall while also minimising blockage effects. The prongs are fabricated to account for this inclination, ensuring that the sensing elements remain parallel to the wall at the desired prong orientation. The resulting highly resolved multi-component velocity statistics up to $Re_\tau = 10,000$ and their associated trends against $Re_\tau$ will be presented.   

Turbulent convection velocities in a turbulent boundary layer

Turbulent convection velocities in a turbulent boundary layer are of crucial importance for bringing together theory and experiment. In this study, we determine probability density functions ($pdf$s) of turbulent convection velocities per wavenumber $k_x$ --using a phase-spectral approach-- from time-resolved PIV measurements in a stream-wise wall-normal plane of a turbulent boundary layer at $Re_\tau\approx2700$. A field-of-view covering approximately $2\times0.5\delta$ with high spatial, $l^+=20$, and temporal resolution, $\Delta t^+= 0.7$, allows us to determine convection velocities for a range of different wall-normal locations ($y/\delta:0.02-0.47$; $y^+:60-1260$). The results indicate that the mode of the $pdf$s coincides with the local mean velocity and that there is a considerable spread around this mode. In the talk, a detailed description of turbulent convection velocities with wall-normal distance will be presented.   

Correction of Pressure Data Close to the Wall in Turbulent Boundary Layer

We have developed a small pressure probe and measured both static pressure and wall pressure simultaneously in turbulent boundary layers up to Reynolds numbers based on the momentum thickness 44000. Experimental data were obtained by the same person using the same techniques at the three large facilities. We find that the measured pressure data are contaminated by the artificial background noise induced by test section and are also affected by the flow boundary conditions. By analyzing data from different wind tunnels acquired at the same Reynolds number, we evaluate the effect of background noises and boundary conditions on the pressure statistics. We also compare the experimental results with results of direct numerical simulations and discuss differences in boundary conditions between real and simulated wind tunnels. The results have already been reported in TSFP7. In the present paper, we are interested in the interaction with pressure probe and the wall. It is very difficult to measure the pressure fluctuation close to the wall. We discuss the effect of solid wall on the pressure data and suggest the method how to correct the data. The measured data are compared with those of DNS.   

Turbulent boundary layer investigation at large Re with micron resolution

The reliable measurement of statistical quantities in turbulent boundary layer flows down to the wall is a challenging problem for many decades. However, due to the progress in laser based experimental techniques in the last years, it is now non-intrusively possible to measure statistical quantities, such as the mean velocity profile, wall-shear stress, Reynolds stresses or the probability density functions of the turbulent fluctuations, with micron resolution (K\"{a}hler et al. Exp. Fluids, 2012). The high spatial resolution allows for accurate measurements as typical bias errors, caused by spatial averaging effects of the probe size, can be avoided. Using advanced optical techniques, we have investigated a turbulent boundary layer flow along a 22 m long flat plate, installed in a wind-tunnel with a 2m by 2m cross-section, at different Reynolds numbers. The statistical results of the investigation will be discussed in the contribution.   

Canonical boundary layer properties at high Reynolds number as measured in the UNH Flow Physics Facility

This presentation describes the characteristics of the flow within the Flow Physics Facility (FPF) at the University of New Hampshire. Having a test section length of 72m, the FPF employs the ``big and slow'' solution to obtaining well-resolved turbulent boundary layer measurements at high Reynolds number. We report on experiments that investigate the wind speed and Reynolds number capability, spanwise uniformity, streamwise pressure gradient, and free-stream turbulence intensity in the FPF. Single element hot-wire measurements of the boundary layer statistical profiles (up to fourth central moment) are presented. These experiments used standard 1mm sensors to generate spatially and temporally well-resolved measurements over the Karman number range 2000 - 20000. Integral parameters and spectra, at a variety of stream-wise locations and Reynolds numbers, are presented, and compared to existing data. For a wide range of test conditions, the FPF is shown to provide high-resolution access to the turbulence of the canonical boundary layer at high Reynolds number.