6

Flow Visualization in Wind Tunnels

Muzafferuddin Mahmood
King Fahd University of Petroleum & Minerals
Kingdom of Saudi Arabia
1. Introduction
Flow visualization is an experimental means of examining the flow patterns around a
body or over its surface. The flow is visualized by introducing dye, smoke or pigment
into the flow in the area under investigation. The primary advantage of such a method is
its ability to provide a description of the flow over a model without a complicated data
reduction and analysis. Smoke- flow visualization (described by Bradshaw, 1970 and Rae
and Pope 1984) involves the injection of streams of vapor into the flow. The vapor follows
filament lines (lines made up of all the fluid particles passing through the injection points)
in steady flow the filament lines are identical to stream lines. (Lines everywhere tangent
to the velocity vector). Smoke- flow visualization thus reveals the entire flow pattern
around a body.
It is difficult to exaggerate the value of flow visualization. The ability to see flow pattern on
a model often gives insight into a solution to an aerodynamic problem. Flow visualization can
be divided into two broad categories the first is surface flow visualization when the
visualization media is applied to the surface such as tufts and oil flow etc. The second type
is off surface such as smoke and streams.
There are basically four methods of recording the flow visualization test. The first and the
best but the least permanent method is for the scientist and the engineer to observe with his
eyes. Because of the depth perception one can see a three dimensional picture. The other
three common methods of recording the result of flow visualization are by film, either still
or movie or television camera or video and magnetic tapes. It must be realized that all three
of these methods are using a two dimensional medium to often record a three dimensional
phenomena. This is especially fine when using a smoke or helium bubbles to trace flow
stream lines pass the model. All three of these methods can be used either black and white
or color. The photography methods while recording more time for developing and printing
stills, when compared to video, yield higher resolution.

2. Flow visualization techniques

The present chapter will deal with the visualization of flow past different types of models
like flat plates, delta wings, elliptical cones and rectangular bluff bodies using

smoke wire technique

laser light illumination technique

surface flow visualization techniques

The results obtained are being discussed considering the importance of different
techniques.

www.intechopen.com

100

Wind Tunnels and Experimental Fluid Dynamics Research

2.1 Flow visualization using smoke wire technique

Flow visualization is considered an important tool to understand the nature of flow field. Its
proper utilization will provide reasonable information that will help in influencing the flow
characteristics. This will eventually lead to better design modification leading to
development of favorable pressures.
The visualization of the flow past the building models was made possible by using the
above technique originally developed by Raspet and Moore in early 1950s and
subsequently improved and used by Batill and Mueller [1 ], and was further improved by
Stahl and Mahmood [2 ]. The procedure involves brushing a thin wire, which then forms a
large number of small droplets. Heating the wire evaporates the oil with each droplet
providing a fine streak-line in the flow. Under proper illumination the smoke streaks appear
bright which can be observed and photographed This technique basically requires a fine
wire of about 0.1mm in diameter of Nickel Chromium steel and suitable oil, which can
vaporize quickly, and a DC current to heat the wire.. This technique was utilized, as it is
very useful in studying separation especially in smooth and nominal boundary turbulent
flows. The smoke-wire was placed at the upstream side of the building model to see the
separating bubble from frontal edge and ground vortices at normal incidence, for both sharp
and round edge model. For oblique incidences of 25 and 45 the smoke-wire was placed at
a suitable position to note the separation of shear layer in the immediate vicinity of the
leading edge corner where high suction pressures are developing. This technique provided
a clear influence of rounding of roof edges on the separating shear layers that formed
conical vortices. The influence due to change in the angle of incidence on the separation
process for sharp and round edge models was found to be of interest. The smoke-wire could
also be moved around the model to get useful information about the flow field.
2.1.1 Flow visualization in smooth flows, =90
The results from smoke wire technique for an angle of incidence of 90 for smooth flows are
shown in Figure 1. These indicate that for a sharp edge model the flow incident on the front
surface of the model divides into two parts. Part of the flow goes up and separates from the
front edge with a large bubble and a re-circulation closer to leading edge, and the rest of the
flow descends down and forms a vortex near the ground. In the model with a rounded
edge, the flow does not have a sharp separation but a long bubble with a height smaller than
that noted for the sharp edge model emerges. The balance flow goes down forming a vortex
close to the ground in a way similar to that noted in the case of the sharp edge model. The
difference between smooth flow and turbulent flow is visible in the separation of flow from
the frontal edge. The height of the separation bubble is comparatively smaller in turbulent
flow than the smooth flow.
2.1.2 Flow visualization in smooth flow at = 45, 25
As the region near the leading edge corner is considered critical since maximum mean and
peak suction pressures were recorded at hole # 50205(Texas tech University test building
nomenclature), smoke wire technique was utilized to observe the flow past this region. Batill
and Mueller [1] studied the transition in the flow on an airfoil using this technique. This
technique was also successfully applied in studying the separation of flow past flat plates at
various angles by Mahmood [2]. At incidence of = 45 the smoke streaks incident on the
side wall at a distance of about 15 to 20% of the length of the wall goes up and separates to

www.intechopen.com

101

Flow Visualization in Wind Tunnels

from large separation bubble (bump) on the top surface of a sharp edge model as evident in
Figure 1c. The remaining flow moved down resulting in re-circulation of flow, which
persisted all along the length of wall. But as the rounding is increased, the size of the bump
became smaller indicating the influence of rounding at a specific location. A bump smaller
in size was noted for a round model with radius of 5mm. For a model with a radius of 10
mm, the smoke streaks remain almost attached to the roof and just slide over curved roof
surface with a minimum possible separation, as shown in Figure 1d. Similar changes were
noted for the flow at an of angle incidence of 25. For this incidence angle the sharp edge
model shows more separation and the height of separation of the vortex sheets decreased
with increased rounding as evident in Figure 2.
For a sharp edge model a clear separation was visible at station 1 i.e., around 15 to 20% of
the distance from leading edge corner, as shown in Figure 2a. But when round edge models
were placed the height of separation reduced in proportion to the magnitude of rounding,
especially for R=10 mm model, where the flow remained attached while moving from
sidewall to the top surface, as shown in Figure 2c. This is in line with the observations made
in smooth flow condition and is expected to influence the pressures that are developed on
top surface of the model.
Since the influence of rounding was visible on both flows at oblique and normal incidence, its
influence on the pressure characteristics was investigated and compared in both the cases.

(a)

(b)

(c)

(d)

Fig. 1. Flow visualization using smoke-wire technique (smooth flow condition). View from front.
a. Sharp edge model = 90
b. Round edge model = 10 mm, = 90
c. Sharp edge model = 45
d. Round edge model R = 10 mm, = 45

www.intechopen.com

102

Wind Tunnels and Experimental Fluid Dynamics Research

(a) Sharp edge model.

(b) Round edge model, R = 5.0 mm.

(c) Round edge model, R = 10.0 mm.

Fig. 2. Flow visualization using smoke-wire technique, = 25 (smooth flow condition).
View from front.

www.intechopen.com

Flow Visualization in Wind Tunnels

103

3. Flow visualization using laser light illumination technique

A smoke-probe was utilized to generate heavy smoke upstream side, which then passes
over the model. A section of the flow was then made visible by using a laser-light sheet. This
plane is generated by using a laser light source in conjunction with a lens system. This
technique helped to study the flow on the top of the model. The near roof edge regions are
considered important from pressure point of view, as corner vortices are said to exist there,
which is a cause for severe suctions. The presence of corner vortices and the changes that
take place in them when edges are rounded could also be studied. Still pictures at different
stations were obtained by moving the laser light plane. This technique was used for all the
flow conditions. A video recording of the flow was done and when played at slow rate
provided useful details of the flow over the roof and could provide further information
about the flow phenomenon.
3.1 Flow visualization past building model in smooth flow at =45
The flow was visualized at two stations on the model top surface. Figures 3a and 3b show
the section of the flow near the leading edge corner at station 1 (15-20% of the distance of
the leading edge), whereas Figures 3c and 3d show the section of flow at station 2, which
was at about 50% of the distance. Figures 3a and 3c show that there exist concentrated
vortices springing from both the leading edges at station 1 and 2, respectively in the case
of sharp edge models. A similar observation was also reported by Bienkiewicz and Sun [3]
on a sharp edge cubical model when surface flow visualization was carried out in a
turbulent boundary layer flow. Figure 3b shows some separation and rolling at station 1
for a round edge model with edge radius R=5 mm with a small vortex formation, where
as at station 2 a big size vortex formed as, shown in Figure 3d. This had a diffused or
burst nature, and could be observed during flow visualization. For model with a further
rounded edge of R=10 mm, no separation was noted at station 1 where as proceeding
towards the model end to station 2, formation of a vortex was noted which can be seen in
Figure 3d. This had a loose core and a burst nature. A similar vortex with a loose core is
usually observed for slender sharp edge delta wings at high incidence [4].
3.2 Flow visualization past elliptical cones and delta wings
As it was reported in the literature that the symmetry of vortex formation and the degree of
asymmetry on elliptical cones is influenced by the thickness of the ellipses [5] and the
theoretical predictions about the degree of asymmetry was also made. In order to study this
phenomenon different elliptical cones whose base height to width ratio varied between 20%
to 80% were taken. The model was placed in the wind tunnel using a suitable fixture. Smoke
was generated upstream side of the model using a smoke probe. The laser light sheet
illumination technique was then used to find flow characteristics. Fig 4 shows the vortices
formation for 65% thick elliptical cone. The degree of asymmetry could be found when the
plain of observation is bit normal to the axis of the cone. Similar results were reported on
circular cones. Where it was possible to observe an asymmetric flow changing to almost
symmetric one when a fin was fixed on the cone [6].
Experiments were conducted to study the asymmetry of vortex formation on a sharp edged
delta wing of Aspect ratio A=0.56 in Ref [7]. It was observed that up to certain angle of
attack symmetric vortex formation takes place as can be seen from Fig.5. These vortices

www.intechopen.com

104

Wind Tunnels and Experimental Fluid Dynamics Research

looked quite concentrated sitting on the wing. When the incidence angle was increased one
vortex got burst and further increase in incidence angle resulted in bursting of both the
vortices. The bursting of one of the vortices can be seen in Fig.6 and no vortex asymmetry
was observed for this type of configuration. A study by Terry,N et al [8] shows extensive use
of laser light illumination technique in visualizing the flow past an 80Delta wing

Wind direction
Laser light sheet,
Station 1
Model top
surface

Sectional view from down-stream side

(a)

(b)

(c)

(d)

Fig. 3. Flow visualization (top surface) using laser light sheet illumination technique = 45.
(a) Sharp edge model, (b) round edge model, R = 10.0 mm station 1 (15% of length), (c) sharp
edge model, (d) round edge model, R = 10.0 mm station 2 (50% of length).

www.intechopen.com

Flow Visualization in Wind Tunnels

105

Fig. 4. Vortex formation behind elliptical cone thickness ratio=0.65, Reynolds number =132,000

Fig. 5. Vortex formation behind Delta wing Aspect ratio=0.56, Reynolds number =132,000, =30

www.intechopen.com

106

Wind Tunnels and Experimental Fluid Dynamics Research

Fig. 6. Vortex formation behind Delta wing Aspect ratio=0.56, Reynolds number =132,000 =35

4. Surface flow visualization technique

The study of the flow near the surface of the square plate model was carried out by using
the chalk powder suspended in Kerosene oil and sprayed on the model. Figure7 shows the
setup for surface flow visualization. When the flow passes the model the chalk and
Kerosene mixture settles as per flow lines. The pattern of flow obtained revealed direction of
surface streamlines and features like the separation of flow from the surface. The flow
visualization experiments carried out by using different methods like smoke wire technique,
smoke probe, surface flow etc., can be divided into two groups as one with 28 and
another with 32as the incidence angle of around 30 is considered critical where a
sudden drop in normal force occurs.
4.1 28
Here the investigations are limited to angles of = 26.5 and = 28. The flow is essentially
divided into two parts, namely the wing center flow which is accompanied by separation
and the wing tip flow Fig 8. In the latter case the flow is separating from side edges which
form vortex sheets which curl up over the wing into the side edge vortices. The results from
various flow visualization studies are shown in Figures 12-13,Here, for the above mentioned
angles 28, the three dimensional separation bubble covers the central part of the plate
and the flow is rotating inside the bubble, moving on the plate surface forward and
outward, and frequently being drawn towards wing side edges. As a result of this induced
transverse flows, considerable masses of air are involved in vortex motion along the side
edges on the suction side as evident from Figs 8-9. Since the bubble is closing at a distance of
about 0.1 C from trailing edge, some of the flow is observed as moving forward at this
junction from movie film taken using a movie camera.

www.intechopen.com

Flow Visualization in Wind Tunnels

Fig. 7. Surface flow visualization setup

Fig. 8. Surface flow visualization on wake side of flat plate at = 28

www.intechopen.com

107

108

Wind Tunnels and Experimental Fluid Dynamics Research

Fig. 9. Flow visualization inside of wake using smoke wire technique = 28

Fig. 10. Flow visualization inside of wake using smoke wire technique = 32

www.intechopen.com

Flow Visualization in Wind Tunnels

Fig. 11. Flow visualization using smoke wire technique = 40

Fig. 12. Conjectured flow field on wake side of flat plate at sub critical incidence

www.intechopen.com

109

110

Wind Tunnels and Experimental Fluid Dynamics Research

Fig. 13. Conjectured flow field on wake side of flat plate at super critical incidence
On the portion of the wing near the leading edge, on the suction side, heavy cross flows are
induced which move to the side edges and bend into the stream direction. The axes of these
side-edged vortices when traced reach the leading edge. The angle it makes is about half the
angle of attack which the plate makes with the stream direction. The core of the three
dimensional bubble is near to the leading edge at a distance of about 0.25 C. There was no
trailing edge vortex shedding observed for these incidences. The side-edge vortices extend
downstream at least six to seven times chord. The diameter of these side edges vortices
increases considerably with a little change in angle from = 26.5 to = 28. The side-edge
vortex core axis moves straight up to trailing edge when moving upstream, where it bends a
little and again moves straight up to leading edge. From the outer wake flow pictures it is
observed that the flow at the center seems to be just as big bubble trying to close about 0.1C
to 0.2C from the trailing edge which is quite different from the angle = 32 where the
bubble closes well away from the trailing edge. Also the wake boundary is wavy for = 28
where as for = 32 it is quite sharp as seen in the Figs. 9 and 10

www.intechopen.com

Flow Visualization in Wind Tunnels

Fig. 14. Flow visualization on wake side using tuft probes.

www.intechopen.com

111

112

Wind Tunnels and Experimental Fluid Dynamics Research

Fig. 15. Conjectured flow field on wake side of flat plate at sub critical incidence

www.intechopen.com

Flow Visualization in Wind Tunnels

113

The result of the variation flow visualization techniques have been used to construct the
flow field as is shown in Figs12-13 and Fig.15 (Flow visualization details). This is what has
been observed it is not an interpretation how the actual flow structure is. A first attempt has
been made to get an idea of the flow field in the following way.
4.2 32
For these incidences and higher up to = 50, it was observed that the wing centre flow is
dominating the total span and the tip vortices are not present on the wing as shown in
Fig.10, but appear further down the stream. The three dimensional bubble covers almost
whole of the span with the appearance of trailing edge vortex shedding, and the flow is
rotating inside this bubble. The flow near the plate surface is forward and spreading
outward which is also noted from surface flow visualization pictures.
Trailing-edge vortex shedding and re-circulating flow was also observed by Calvert [9]. In the
region near the leading edge on the suction side, the flow was observed as moving towards the
plate centre from sides, and the flow which is moving forward is joining the main stream flow
reaching from sides to form rotary motion like side edge vortices, the concentration of which is
observed little downstream of the trailing edge. This distance increases gradually as the angle
of incidence is increased. Also, it was found in the surface flow pictures that the two eyes
had disappeared and with them the additional two vortices must have vanished there. For all
these angles 32 trailing-edge vortex shedding was observed, and the frequency of vortex
shedding was observed to increase as the angle of attack increases.

5. Conclusions
Flow visualization is considered an important tool to understand the nature of the flow
field. Its proper utilization will provide reasonable information that will help in influencing
flow characteristics. The above methods mentioned are not the only ways of understanding
the flow but the results obtained using above methods highlight the usefulness of these flow
visualization techniques.

6. Acknowledgements
The author is grateful for the facilities provided by King Fahd University of Petroleum and
Minerals (KFUPM) and University of Sydney, Australia. The author also acknowledges the
support from Dr. K Srinivas from the School of Aerospace, Mechanical & Mechatronics
Engineering, University of Sydney Australia.

7. References
Batill, S.M.,& Mueller,T.M. (1981).Visualization of Transition in the Flow over an
Aerofoilusing Smoke wire technique , AIAA journal,Vol.19, pp.340-345
Bienkiewicz, B.,& Sun, Y .(1992). Local Wind loading on the Roof of a Low-rise Building,
JWEIA,Vol 45, pp.11-24
Calvert,J.R.( 1967). Experiments on the flow past an inclined disk, Journal of Fluid
Mechanics,vol.29, Pt.4, pp.691-703

www.intechopen.com

114

Wind Tunnels and Experimental Fluid Dynamics Research

Fiddes, S.P.,& Williams,A.L.(1989).Recent developments in the study of Separated flows

past Slender bodies at incidence, Conference proceedings Roy.Aeron.Soc.,London,
pp 31.1-31.17.
Mahmood, M.,Low-Speed Experiments on a Flat Square at High Angles of Attack,
M.S.Thesis Dissertation, KFUPM, Dhahran, Saudi Arabia.
Stahl,W.H, Mahmood,M and Asghar,A.,Experimental Investigation of the Vortex flow on
Delta wings at High incidencesAIAA journal, 30(4), April 1992, pp 1027-1032
Stahl,W.H., Mahmood,M.,Asghar,A.,Experimental Investigation of the Vortex flow on a
very slender Sharp-edged Delta wings at High incidence, AIA Ajourn. vol 30,No.
4, pp1027-1032, April 1992.
Stahl,W.H., Asghar,A.,Mahmood,M., Supression of Vortex Assymetry and side force on
acircular cone,DLR-IB222-A12, Cologne, Germany, April 1993.
Terry, N.,Gangulee, D., Secondary flow control on slender Sharp-edged
ConfigurationsAIAA paper93-3470,CP,1993.

www.intechopen.com

Wind Tunnels and Experimental Fluid Dynamics Research

Edited by Prof. Jorge Colman Lerner

ISBN 978-953-307-623-2
Hard cover, 709 pages
Publisher InTech

Published online 27, July, 2011

Published in print edition July, 2011

The book Wind Tunnels and Experimental Fluid Dynamics Research is comprised of 33 chapters
divided in five sections. The first 12 chapters discuss wind tunnel facilities and experiments in incompressible
flow, while the next seven chapters deal with building dynamics, flow control and fluid mechanics. Third section
of the book is dedicated to chapters discussing aerodynamic field measurements and real full scale analysis
(chapters 20-22). Chapters in the last two sections deal with turbulent structure analysis (chapters 23-25) and
wind tunnels in compressible flow (chapters 26-33). Contributions from a large number of international experts
make this publication a highly valuable resource in wind tunnels and fluid dynamics field of research.

How to reference

In order to correctly reference this scholarly work, feel free to copy and paste the following:
Muzafferuddin Mahmood (2011). Flow Visualization in Wind Tunnels, Wind Tunnels and Experimental Fluid
Dynamics Research, Prof. Jorge Colman Lerner (Ed.), ISBN: 978-953-307-623-2, InTech, Available from:
http://www.intechopen.com/books/wind-tunnels-and-experimental-fluid-dynamics-research/flow-visualizationin-wind-tunnels

InTech Europe

University Campus STeP Ri

Slavka Krautzeka 83/A
51000 Rijeka, Croatia
Phone: +385 (51) 770 447
Fax: +385 (51) 686 166
www.intechopen.com

InTech China

Unit 405, Office Block, Hotel Equatorial Shanghai

No.65, Yan An Road (West), Shanghai, 200040, China
Phone: +86-21-62489820
Fax: +86-21-62489821