Resolving the structure of Turbulence in Rough Wall Channel Flows Using 3D Time Resolved Multiscale Velocity Measurements
Abstract
Project Summary This proposal is submitted both to NSF and ONR, and the budget is divided among the two programs. The similarity hypothesis for rough wall boundary layers states that outside of the roughness sublayer, the dynamics of turbulent motions are independent of roughness geometry. Yet, our measurements in high Reynolds number turbulent channel flows have shown that while Reynolds stress profiles agree with the similarity hypothesis, the signature of roughness scale turbulence clearly persists in the outer layer, and significantly impacts the energy flux across turbulence scales. Furthermore, interactions of outer layer sweeps with the roughness elements generate U-shaped vortices with quasi-streamwise legs. They develop as spanwise vorticity rolls up in a region of low velocity, upstream of roughness peaks, and their legs stretch downstream in regions of elevated streamwise velocity (flow channeling) between roughness elements. Flow induced by adjacent legs of neighboring U-shaped structures cause ejections, which lift the vortex legs away from the wall. These processes appear to be major contributors to subgrid scale (SGS) energy flux. These observations indicate that to develop of physical models for roughness induced momentum and energy fluxes, it is essential to understand how roughness scale turbulence is generated and transported away from the wall, and how it interact with outer layer eddies. We propose to investigate how the roughness height, wavelength, spatial arrangement and uniformity, affect the generation of U shaped vortices, their evolution in time, their interactions with outer layer structures, and impact of these interactions on turbulence statistics. To elucidate the multi-scale interactions, one has to simultaneously resolve the small-scale turbulence near the wall along with the large structures hovering above. For this purpose, we have integrated two time-resolved 3D velocity measurements techniques with different resolution, namely tomographic particle image velocimentry (TPIV) and digital holographic microscopy (DHM). TPIV performs ‘coarse’ measurements (~0.5 mm vector spacing) over a volume that covers more than half channel height. DHM simultaneously provides data at a much high resolutions (15–60 ?m) for a sample located in the roughness sublayer and embedded within the TPIV volume. The combined TPIV-DHM instrument resolves 3.5 orders of magnitude of length scales. To perform measurements within the Roughness sublayer, including the space between roughness elements, we match the optical refractive index of the transparent (acrylic) rough wall with that of the liquid - a concentrated aqueous NaI solution. Time resolved (0.7 – 2.25 kHz) data will be uses to examine several turbulent channel flow cases: (i) flow around isolated and pairs of roughness elements with different size and spacing, (ii) rough walls with densely and sparsely packed elements, and (iii) a rough wall with embedded elements of different size. Statistical analysis involving conditional sampling and comparisons to instantaneous data will examine the impact of roughness geometry on strength, orientation, and eruption rates of roughness scale eddies, as they interact among themselves and with outer layer structures. Resulting effects on the outer layer turbulence, SGS fluxes, Reynolds stresses, turbulence production and dissipation rates will be quantified.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Aug 12, 2016
- Source ID
- N000141512404
Entities
People
- Joseph Katz
Organizations
- Johns Hopkins University
- Office of Naval Research
- United States Navy