Experimental Fluid Mechanics
Pressure and velocity fluctuations in wall-bounded turbulent flows
Angeliki Laskari, Roeland de Kat, Bharath Ganapathisubramani
Figure: Tomographic-PIV setup used in this study.
The purpose of this work is to enhance the understanding of wall-bounded turbulent flows, with respect to both pressure and velocity fluctuations. To this end, time-resolved planar and tomographic PIV experiments were performed in a high-Reynolds-number turbulent boundary layer. Starting at the freestream boundary, the structure and evolution of velocity fluctuations within the boundary layer are analysed using the planar PIV database. Across the turbulent/nonturbulent interface, conditional profiles of velocity, vorticity, and Reynolds stress exhibit clear jumps, within an interface thickness that scales with the local Taylor microscale. The interface is tracked in time and a net positive entrainment rate is estimated, exhibiting an increasing trend for higher wall-normal locations. Below the freestream boundary, the flow is organised into zones of uniform momentum, which are detected instantaneously, while a temporal threshold is then applied to remove short-lived zones. A higher than average number of zones is linked with a large-scale Q2 event and an amplification of the Reynolds stress in the log region. On the other hand, a low number of zones is found to be associated with a large-scale Q4 event and a decrease of the turbulent activity in the log region. Zones belonging to a low-zone-number structuring are shown to reside within the measurement plane on average four times longer that those belonging to a high-zone-number case. To gain further physical insight, a pressure estimation method is developed, based on Taylor’s hypothesis, using both planar and volumetric velocity data without the requirement of time information. The method is validated in the case of a DNS channel flow and is found to be robust to noise and grid resolution, performing as well as a pseudo-Lagrangian approach in the case of volumetric experimental data. A 2D formulation of the proposed method, although performing significantly worse than the 3D one instantaneously, provides reliable results in a statistical sense and allows for pressure estimation also for the planar PIV database. Pressure fluctuations are shown to be correlated almost throughout the boundary layer in the wall-normal direction, while within the log region they convect slightly faster than the velocity fluctuations. A high number of uniform momentum zones is linked with an increase of pressure rms values and a larger coherence of pressure fluctuations both in the streamwise and the vertical direction, while the opposite is observed in a low zone number state.
Part of this project has been published in Experiments in Fluids (see https://doi.org/10.1007/s00348-016-2129-5)