5% - eTheses Repository - University of Birmingham

The burst-sweep dynamics study of Utami & Ueno (1987) , used flow
visualisations taken by two still cameras for picture processing
of horizontal cross-sections taken just above each other at 0.2s
intervals. They proposed a detailed mechanism for the flow
structure as shown in figure 1. From their visualisations they
concluded that a perturbation in the high shear and vorticity
layer very close to the wall (ie the viscous sublayer) produced
vortex filaments in this sublayer which were lifted and elongated
by the mean shear flow. Such vortices interacted with each other
forming a horseshoe vortex as in figure la.
Utami and Ueno then proposed that the horseshoe vortices rolled
around each other and were stretched further to give the large
scale spiral and slightly inclined vortex shown in figures lb,c.
The fluid between the legs of the main spiral would thus be
driven in the upstream-upward direction consistent with an
ejection and the outer fluid would be driven in the downstreamdownward
direction, consistent with a sweep.
Whether this work presents an accurate picture is still open to
debate, but it is given some credence by the work of Acarlar &
Smith (1987a,b), who studied hairpin vortices shed firstly from
a hemispherical protuberance and secondly by fluid injection in
a developing laminar boundary layer. Their visualisations (shown
in figure 2) and hot-film-anemometry studies have shown that such
boundary layers develop remarkably similar velocity profiles to
those found in a turbulent boundary layer, and that the hairpins
interacted in the exact same way as described by Utami and Ueno.
Al-5