This paper describes experimental results measured in a low-shear turb
ulent boundary layer. The low-shear condition is exerted after the bou
ndary layer reaches Re(theta)similar or equal to 2000 and has the effe
ct of removing the inner layer; thus, these are the first results to s
how the behavior of an outer-layer-only turbulent boundary layer. The
removal of the inner layer causes the gradual decay of the turbulent s
tresses over eleven boundary-layer thicknesses (roughly 20 large-eddy
length scales) of streamwise distance, with the decay beginning at the
wall and propagating into the outer flow with increasing downstream d
istance. However, the structure of the outer layer is little affected
by the perturbation, as demonstrated by stress (anisotropy) ratios, qu
adrant analysis, and spectral measurements. Although the lack of near-
wall production implies this flow must eventually decay into isotropic
turbulence, this decay occurs relatively slowly because the dissipati
on is also greatly reduced with the decrease in near-wall shear. In ad
dition, the outer-layer production is significant in maintaining the t
urbulence level. These results show that, once formed, the outer-layer
characteristics are not explicitly dependent on the presence of the i
nner layer. These results are compared with similar studies of isotrop
ic turbulence near a shear-free wall. Very close to the wall the two f
lows both show that the normal-stress components respond differently t
o the presence of the wall. However, away from the wall the isotropic
results underpredict the distance to which the tangential stresses are
damped by the impermeability condition at the wall. Finally, the resu
lts show general similarities to those in a boundary layer just downst
ream of reattachment, after a similar low-shear condition over the sep
aration bubble. This raises the possibility that many of the important
features of the reattaching flow can be captured by the present, simp
ler experiment. (C) 1996 American Institute of Physics.