The objective of this effort is to study the interaction
between streamwise vorticity and Tollmien-Schlichting
(TS) waves in an effort to gain new
insights into the physical processes responsible for
bypass transition. Transition by means of amplification
and nonlinear breakdown of initially small-amplitude
TS waves is only observed for low
free-stream turbulence (FST) levels. "Bypass" of TS
instability mechanisms has been hypothesized for
higher FST levels. However, recent observations
demonstrate that TS waves still play an active role at
moderate FST levels. Elevated FST levels also appear
to be associated with streamwise vorticity within the
layer that originates at the leading edge. Therefore,
the influence of streamwise vortices on TS waves
may provide new insights into bypass transition,
which has remained a mystery since the 1930s.
Free stream nonuniformity (FSN) is deliberately
introduced into an otherwise highly uniform free
stream. The FSN is in the form of a laminar wake
from a fine wire (d = 0.002 inch, Rd = 16) located
7,250d upstream of the leading edge of a flat plate.
Interaction of the wake with the leading edge results
in a pair of weak counterrotating vortices embedded
within the Blasius boundary layer. The characteristics
of two-dimensional TS waves generated by a vibrating
ribbon have been determined with extensive hot-wire
measurements, both with and without the
presence of the vortices.
The ribbon is located just upstream of Branch I of
the neutral stability diagram (F = 60 x 10-6, R = 485)
and it is active over the full span of the test section
thereby allowing the wave behavior to be studied
over large streamwise distances. The wave amplitude
grows by almost two orders-of-magnitude between
Branch I and Branch II for this operating point. The
development of peak root-mean-square (rms) wave
amplitude with streamwise distance conforms with
predictions from linear stability theory. However,
large variations in the rms wave amplitude emerge in
the spanwise direction despite the relatively small
wave amplitude (µ/U1 ≈ 0.5%). Spanwise profiles (not
shown) of the rms wave amplitude have the same
form of peak-valley splitting initially observed in
1962. The phenomenon is now known as K-type
secondary instability and it has been subject to
extensive theoretical study.
The vortices introduce considerable phase
distortion of the TS waves as shown in Figure 1.
Initially, the rms wave amplitude is reduced in the
vicinity of the vortices for a substantial streamwise
distance, as shown in Figure 2(a). A remarkable
feature has been captured in the contours for
R = 914, shown in Figure 2(b), that is, the appearance
of two small regions at y ≈ 1.5 mm with exceptionally
low rms amplitude. This point marks a critical
change in the wave behavior since a small increase
in Reynolds number leads to a completely different
distribution in which the maximum wave amplitude
now occurs between the vortices, as shown in
figure 2(c). A further increase in Reynolds number
results in rapid growth in amplitude (e.g., µ/U1 ≈
10% for R = 984) and in the onset of random behavior,
which is a characteristic of the final approach to
breakdown to turbulence.
A different type of secondary instability mechanism
appears to be associated with the vortices,
which leads to transition at a lower Reynolds number.
The results help explain the adverse effects of
wind tunnel flow quality on tests concerning bodies
with substantial regions of laminar flow.
Point of Contact: J. Watmuff
(650) 604-4150
jwatmuff@mail.arc.nasa.gov
Back To Top
Previous Paper
Return to Revolutionary Technology
Next Paper 
