K. A. Batchelder, R. J. Moffat
Department of Mechanical Engineering
Stanford University, Stanford CA 94305-3030, USA
Correspondence to: K. A. Batchelder
Experiments in Fluids 25 (1998) 104 —107 Springer-Verlag 1998
Surface flow visualization using the thermal wakes of small heated spots
K. A. Batchelder, R. J. Moffat
This note describes a surface ﬂow visualization
technique which uses the thermal wakes of an array of small
heated spots to infer the local ﬂow direction. The thermal wake
is made visible using wide band thermochromic liquid crystals.
The technique is illustrated using the endwall ﬂow under
a horseshoe vortex at the base of a right circular cylinder in
a turbulent boundary layer. Comparisons to results generated
using the oil of wintergreen technique were in good agreement.
In addition to surface ﬂow direction, the technique has the
potential to be used to measure the heat transfer coefﬁcient at
each spot. Data are presented in terms of photographs of the
actual visualization surface. The techniques is suitable for low
Surface ﬂow visualization is a primary tool for the study of
complex three dimensional ﬂows. There are many established
techniques that work quite well; for an overview see Bradshaw
(1970). Every technique has its advantages and disadvantages.
The disadvantage of many techniques is that they involve
introducing some sort of messy material into the wind tunnel.
Another disadvantage of other techniques is that they involve
tracing streaklines over a short duration test period which
includes an initial transient. This often means that a separate
visualization surface must be prepared for each ﬂow situation
to be investigated. There is a need for a surface ﬂow visualiz-
ation technique that uses a passive scalar to trace the surface
streaklines; one that can easily be introduced non-intrusively
into the ﬂow during steady state operation.
A small heated spot on an otherwise adiabatic wall will
generate a tear drop shaped thermal wake whose tail points
downstream. If this wake or its trace on the wall can be made
visible, then the ﬂow direction can be seen. For such an
indication to be useful it must not alter the ﬂow ﬁeld. Thus, the
local buoyancy must be controlled. Also, if the wall layer ﬂow
direction is to be visualized the wake must not extend far into
the boundary layer. Therefore the spots must be small in
size relative to the boundary layer and operated at a small
temperature rise above the freestream.
There are no real surfaces with zero thermal conductivity or
thickness. Thus, there will always be some lateral conduction in
the substrate. The effect of lateral conduction will be to obscure
the asymmetry in the isotherms caused by convective trans-
port. For this reason it is important to minimize the thermal
conductivity and thickness of the surface substrate.
Either infra-red imaging or wide band thermochromic liquid
crystals can be used to make the thermal wakes visible. Liquid
crystal images can be recorded by conventional video, or
photographed with ordinary color ﬁlm and require no special
window materials for viewing. For these reasons liquid crystals
were chosen for the present demonstration.
Construction of the visualization surface
Figure 1 is a cross sectional view of the visualization surface.
Four heat sinks are used, each heat sink is 2.54 cm square,
made of aluminum and each contains 100 equally spaced 1 mm
square pin ﬁns approximately 6 mm in length. There is a gap
which forms a cross through each heat sink, meant to hold the
mounting bracket which holds the electronic chip. These gaps
appear in Fig. (3) as a void lane in the array of heated spots.
The heat sinks are positioned in a 10 cm square hole cut in
a piece of foam core. The ends of the pin ﬁns, having been
machined ﬂat, are level with the top surface of the foam core.
The entire 10 cm square area is then covered with 0.25 mm
thick polyester tape. The adhesion makes good contact
between the tape and the tops of the pin ﬁns, providing
a smooth and continuous ﬂow surface.
The surface is then coated with a layer of black paint
and a layer of liquid crystals. Each coat is applied using an
artists airbrush and each is approximately one half mil thick.
A complete description of the liquid crystal preparation and
painting technique may be found in Farina (1994). The average
thermal conductivity of the polyester tape and painted layer is
estimated to be on the order of 0.1 W/mK. The crystals used in
this investigation activate at 25 °C and have a range of 5 °C,
going from red to green to blue over that range.
Figure (2) shows a schematic of the test facility. The light
source is a Kodak slide projector with a 300 W quartz halogen
bulb, type ELH. The light source provides a focused beam of
light that is uniform in intensity within 20% over the region of