Effects of Surface Irregularity on Turbulent Boundary Layer Wall Pressure Fluctuations

Abstract

Measurements were made of the mean velocity profiles and wall pressure field upstream and downstream of the flow over both a backward-facing and forward-facing step. For each configuration the velocity profiles show that the effects of the separation-reattachment process persist more than 24 step heights downstream of the step. Extremely high values of the RMS wall pressure are measured near reattachment. These values are 5 and 10 times larger than on a smooth flat plate for the backward-facing step and the forward-facing step, respectively. The spectral density of the wall pressure fluctuations in the recirculation region is dominated by low frequency components. Downstream of reattachment there is a reduction in the low frequency content of the wall pressures and an increase in the high frequency components. At the farthest measured position downstream, the spectral density is still higher than that found on a smooth flat plate. These results show that the complex turbulent flow generated by a surface irregularity can significantly increase the localized wall pressure field and these increases persist far downstream of the irregularity. Consequently, a surface irregularity can be a major source of turbulence-induced vibrations and flow noise, as well as a cause of the inception of cavitation in marine applications.

Influence of Separation on Sound Generated by Vortex-Step Interaction

“…If the vortex is assumed to be a crude model of a discrete eddy in a turbulent boundary layer, a larger value of d/h would be representative of turbulent flow over a small skin step, when the boundary-layer thickness tends to be large compared to h. The case in which d/h"4 is illustrated in Figure 5. Here the deflection of the incident vortex is relatively small (about three step heights), and the circular path of the vortex pair terminates just to the right of the step (near ºt/h+3), the trajectory being reminiscent of reattachment profiles observed in experiments (Farabee & Casarella 1984, 1986, 1988: the ''eddy'' formed by the vortex pair is torn apart, proceeding to the right along the wall and away from the step, while is captured within a ''separation bubble''. Actually, d /dt changes sign at ºt/h+5)2, at which time it is ''released'' from the edge S with a final circulation "!1)27…”

“…More generally, reattachment occurs downstream of a stagnation point, and is a region of high-fluctuating wall pressure, where free shear layer eddies bifurcate into upstream and downstream travelling components (Bradshaw & Wong 1972). According to Farabee & Casarella (1984, 1986, low Mach number turbulent wall pressures at reattachment for backward and forward facing steps (with step heights comparable to the boundary-layer thickness) are, respectively, about 5 and 10 times larger than the smooth wall boundary-layer pressures. The wall pressure must ultimately revert to that of a smooth wall boundary layer, but differences have been observed at distances as large as 72 step heights downstream of the step.…”

“…Irregularities in the form of forward or backward facing steps have been studied extensively (Bradshaw & Wong 1972;Moss & Baker 1980;Chandrsuda & Bradshaw 1981;Eaton & Johnston 1981;Farabee & Casarella 1984, 1986, 1988 because their free shear layers possess well-defined separation and reattachment zones. More generally, reattachment occurs downstream of a stagnation point, and is a region of high-fluctuating wall pressure, where free shear layer eddies bifurcate into upstream and downstream travelling components (Bradshaw & Wong 1972).…”

Wall Pressure Fluctuations Induced by Separated/Reattached Channel Flow

“…This is also the case for the streamwise turbulence velocity component and the Reynolds stress [13,21]. separated #ow from Farabee [13] and Farabee and Casarella [34] are shown for comparison. In general, the trends among the various sets of data are similar; however, noteworthy deviations exist in the magnitude of the peak #uctuation within the reattachment location.…”

“…The maximum velocity decreases substantially downstream due to the channel area expansion as well as the interaction between the plate and tunnel wall shear layers. Other experiments [13,34] utilize the free-stream velocity for r.m.s. wall pressure #uctuation normalization, a velocity that is una!ected by adjacent walls.…”