Experimental study of eddy structures in a turbulent boundary layer using particle image velocimetry

Abstract

Particle image velocimetry experiments have been performed in a turbulent boundary-layer wind tunnel in order to study the coherent structures taking part in the generation and preservation of wall turbulence. The particular wind tunnel used is suitable for high-resolution experiments ($\delta \gt 0.3$ m) at high Reynolds numbers (up to $R_{\theta} = 19\,000$ in the present results). Eddy structures were identified in instantaneous velocity maps in order to determine their mean characteristics and possible relationships between these structures. In the logarithmic region, the results show that the observed eddy structures appear to organize like elongated vortices, tilted downstream, mainly at an angle of about 45° and having a cane shape. The characteristics of these vortices appear here to be universal in wall units for $R_{\theta}\,{\leq}\,19\,000$. They seem to find their origin at a wall distance of about 25 wall units as quasi-streamwise vortices and to migrate away from the wall while tilting to form a head and a leg. Away from the wall, their radius increases and their vorticity decreases very slowly so that their circulation is nearly constant. Near the wall, the picture obtained is in fair agreement with existing models. The analysis of the results indicates a universality of the buffer-layer mechanism, even at low Reynolds number, and a sensitivity of the logarithmic region to low-Reynolds-number effects.

The role of coherent structures in subgrid-scale energy transfer within the log layer of wall turbulence

“…This spatial resolution is adequate enough to resolve the vortical structures germane to this study. 28,30 Standard two-frame cross-correlation methods were used to analyze the PIV images with 50% overlap to satisfy Nyquist's criterion and the time delay between images was chosen to minimize the relative velocity measurement error to less that 1%. Table I summarizes the flow parameters for the experiment.…”

Experimental manipulation of wall turbulence: A systems approach

“…This is a thresholding problem: rotational motion is still captured for both senses of vortex rotation, but the usual thresholding techniques applied to velocity fields lead to a systematic under-accounting of retrograde vortices. McKeon, Sharma, and Jacobi 50 explored this phenomenon in the context of this work with respect to observations of swirl in the streamwise/wallnormal plane of a zero pressure gradient turbulent boundary layer, while Carlier and Stanislas 58 documented earlier the general characteristics of the distributions of pro-and retro-grade hairpin vortices.…”

Self-consistent high-Reynolds-number asymptotics for zero-pressure-gradient turbulent boundary layers

“…However, it has recently become increasingly clear that this proliferation of models is principally due to the scarcity of independent wall shear stress measurements and the limited range of experimental Reynolds numbers, besides, of course, the experimental uncertainties. This realization is the result of recent advances in measurement techniques, equipment, and facilities, which have enabled experiments at high Reynolds numbers; for instance, by Hites, 1 Österlund, 2 DeGraff and Eaton, 3 Metzger and Klewicki, 4 Knobloch and Fernholz, 5 Nagib et al, 6 Carlier and Stanislas, 7 Kunkel and Marusic, 8 and Nickels et al 9 ͑in chronological order͒. An overview of recent developments in the field can be obtained from Ref.…”

“…Therefore, we have made the quality assessment of more than a dozen ZPG TBL data sets known in the literature the subject of a separate publication. 11 Nevertheless, data of Karlsson, 12 Smith, 13 Hites, 1 Bruns,14,15 DeGraff and Eaton, 3 Carlier and Stanislas, 7 and Nickels et al 9 have been included in some figures. ͑For a justification of this data selection, see Ref.…”