L. Jourdan et al., THE EFFECT OF DRAG-REDUCING POLYMER ADDITIVES ON WALL-PRESSURE FLUCTUATIONS IN TURBULENT CHANNEL FLOWS, European journal of mechanics. B, Fluids, 17(1), 1998, pp. 105-136
The present paper is concerned with the experimental study of the infl
uence of Toms' phenomenon on the convective part of the hydrodynamic t
urbulent wall-pressure field. The experiments were carried out in a wa
ter tunnel, at flow velocities of 2.3, 3.3 and 4.3 m/s, where amounts
of drag reduction up to 40% were obtained by injecting dilute polymer
solutions. An investigation of the spectral feature of the wall-pressu
re frequency spectrum and the wall-pressure frequency cross-spectrum w
as achieved from data taken using pinhole transducers. The scaling law
s for the mid-and high-frequency regions of the spectrum were examined
both for the water how and the drag-reducing flow. The use of a noise
cancellation procedure allowed observation of the maximum of the spec
trum in the mid-frequency range at an approximately reduced frequency
w delta/u(tau), of 80 for the water flow. This maximum is shifted towa
rds a higher value of the reduced frequency in the presence of Toms' e
ffect. From cross-spectral data obtained over separation distances ran
ging from 0.46 to 2.86, the effect of drag reduction was estimated on
the coherence functions as well as the convection velocity. These data
show a correlated shift of the threshold which delineates the low wav
e number group identified as the large-scale turbulence contributor in
the outer region of the flow and the high wave number group associate
d with turbulent sources in the log-region. The effect of drag reducti
on on the convection velocity data supports the fact that the contribu
tions to the different wave number groups should be attributable to ac
tivity in the outer and inner regions, respectively. In addition, a co
mputation of the root mean square pressure was achieved from the wall-
pressure frequency spectrum. When expressed as functions of the percen
tage of drag reduction, the rms pressures related to the wall shear st
ress are found to increase monotonicaly. The equivalent ratio obtained
by scaling the rms pressures using the dynamic pressure is found to d
ecrease in the drag-reducing flow. Finally, an estimation of the proba
bility density distribution showed that the skin friction reduction af
fects predominantly the occurrence of higher pressure amplitudes as we
ll as very low pressure amplitudes. This confirms the important contri
bution of the large pressure peak events to the overall features of th
e wall-pressure. (C) Elsevier, Paris.