Aerodynamics Lab II

AE39001
Experiment 9

Smoke Flow Visualisation

Professor
Dr. Sandeep Saha
Author
Arun Kumar Suthar
14AE30005
Experiment performed at:
Aerodynamics lab
Department of Aerospace Engineering
IIT Kharagpur
25 October 2016

1.1 Aim

To visualize flow past cylinder and airfoil using smoke.

1.2 Apparatus

1. Wind tunnel
2. Models( small cylinder, large cylinder and airfoil)
3. Smoke flow setup
4. Smoke generator
5. Smoke injector

1.3 Introduction

Smoke flow visualization is a qualitative measurement technique used to analyze the flow behavior. The
smoke follows the air currents, allowing the observer to visualize the flow. A smoke generator is coupled
with an air blower and a settling chamber. From this chamber smoke is fed to smoke injector, which has
a shape of an airfoil with a number of small diameter tubes placed at trailing edge. The smoke exit the
injector as a sheet. The smoke sheet moves with the air and deforms, describing the flow pattern. The
ability to see flow pattern on a model often gives insight into a solution to an aerodynamic problem.
1. Smoke generator: Smoke for flow visualization purposes may be generated by burning tobacco,
wood, and wheat straw, or by vaporizing hydrocarbon oils like white mineral oil or propylene glycol.
Regardless of the source of the smoke used, it is essential that the smoke meets the following criteria:
1. It must be able to track the flow field accurately. In other words, the smoke particles must be
sufficiently small that their motion reflects the motion of the flow.
2. It must not significantly affect the flow field under investigation.
3. It must possess high reflective properties.
4. It must be non-toxic.
1.2. Smoke-wire technique: Here, smoke is produced by vaporizing oil from a fine wire heated by an
electric current. Most metals with sufficient strength and electrical resistivity can be used, but the three
most commonly used wires are made of stainless steel, nichrome, or tungsten. The size of the wire is
dictated to some extent by the flow speed. For low-speed applications, it is better to use a smaller
diameter wire because a smaller wire produces smoke which is sharper. At higher speeds (a few meters
per second), a larger diameter wire is recommended because its larger surface area can maintain a
higher smoking rate, and since the wires are typically stretched taut, it will be able to accommodate
the required tension at a higher temperature. There are a variety of oils which can be used to produce
smoke filaments, including paraffin, kerosene, lubricating oil, silicone oil, and model train oil.
Paraffin is perhaps the most effective.
1.3. (Titanium tetrachloride):
The technique makes use of the fact that when TiCl4 is exposed to moist air, it develops dense white
hydrochloric acid fumes and minute particles of titanium dioxide according to the reaction:
4 + 22 = 2 + 4

TiC4 is inexpensive and can be purchased commercially. However, hydrochloric acid fumes are toxic and
can pose a serious health hazard. Experiments should therefore be carried out in a well-ventilated
environment, and if possible the fumes should be exhausted to the outside of the laboratory.
2. Smoke Injector: Smoke can be injected from the surface or dispersed with a hollow wand that can be
moved through the flow field. In the case of our experiment, the smoke was injected with help of a
series of small vertical tubes placed enough far away from the test subject. The generated smoke was
allowed to settle down in a large chamber before the smoke is extracted and used for visualization.

1.4 Procedure:

1. The cylinder with the smaller radius was placed inside the test section of the wind tunnel. It is to be
noted that the injection tubes were ahead of the cylinder but were present in the test section itself,i.e. ,
the tubes were near the start of the test section. This was done in order to have the least amount of
disturbance in the smoke flow before it reaches the test geometry.
2. The smoke generator was started. The fuel used for smoke generation was polypropylene glycol
mixed with distilled water and rose fragrance.
3. The smoke from the generator was allowed to settle down in a cascade of chambers comprising of
two chamber, one large and other small chamber. The smoke generator was connected to the lower end
of the small chamber while the outlet of this was connected to the lower part of the large chamber.
Finally the smoke was extracted from the larger chamber outlet connected to the top of the chamber.
Again it was done to reduce the perturbation as much as possible.
4. The wind tunnel suction fan was started and the smoke was injected into the flow through the
injection pipes. The amount of smoke was determined using a regulator.
5. After visualization over the static cylinder, it was rotated (rotation by hand in both anti-clock wise and
clock wise sense) and the effects were observed.
6. The same steps were performed again with the cylinder with larger radius and airfoil. The cylinder
with larger radius was also rotated with hand and the effects were observed.
7. In case of the airfoil, instead of rotation, the angle of attack was changed from a negative arbitrary
value to a positive arbitrary value and the flow was observed.

Observation
Model dimensions:
1.Cylinder(small): Diameter = 15mm
2.Cylinder(large): diameter = 44.5mm
3.Airfoil: Chord = 160m

Flow observation in the case of smaller cylinder

1. For the flow velocity in the experiment the flow just ahead of the cylinder was laminar and steady as
explained by the potential flow theory.

Fig. 1 The flow just ahead of the cylinder

2. The past the cylinder was unstable behind the cylinder, vortices being shed alternately from the top
and bottom surface in a regular fashion downstream. The pattern is called Karman vortex street.

Fig. 2 The vortices being shed past the cylinder

3. The flow separated from the surface of the cylinder (boundary layer separation) forms a wake behind
the cylinder. The wake has slow oscillations in which the velocity is periodic in time and downstream
distance, with the amplitude of the oscillation increasing downstream. They form the Karman vortex
street.
4. The Eddies are shed continuously from each side of the body, forming rows of vortices in its wake.
The alternation leads to the core of a vortex in one row being opposite the point midway between two
vortex cores in the other row.
5. The eddies grew in size downstream and become weaker as the energy of the vortices is consumed by
viscosity as they move further downstream, and the regular pattern disappears.
6. The frequency of production of eddies increased as the flow velocity was increased.
7. When the cylinder was rotated, the rotation seemed to add/subtract with the vortices, i.e., when the
rotation was clockwise, the vortices formed from the upper surface had larger strength while the same
formed from the lower surface had lower strength and vice versa.
8. In the case of cylinder with larger diameter the frequency of the oscillations decreased while the
amplitude increased. For both the cylinders when the velocity was increased, the frequency of the
oscillations increased. Else all other observations were similar to the case of cylinder with smaller
diameter.
Flow observation in the case of airfoil
1. At 0 degree angle of attack, there was no flow separation.

Fig 3. Flow at zero angle of attack

2.As the angle of attack was increased, the flow was still attached but the stagnation point moved
backward which initially was at the trailing edge itself.

Fig. 4. Flow at some angle of attack

3. After a certain angle of attack, the flow was separated from the surface of the airfoil at a certain
point, i.e., the airfoil was stalled. Beyond this point, as angle of attack was increased, the point of
separation moved ahead towards the leading edge. Behind the point of separation, a wake was formed.
Increasing the angle of attack made the size of the eddies in the wake bigger.

Fig. 5. Flow separation

4.When the angle of attack was decreased below 0 degree (angle of attack was negative), the flow was
attached to the surface but only till a certain angle of attack. The value of this angle was much smaller
than the other one (the case of positive stall angle).
5. In the case of negative angle of attack, the flow separated from the lower surface of the airfoil, as
opposed to the upper surface as observed in the case of positive angle of attack. Again, trailing edge
vortices were formed after flow separation which grew bigger as the angle was decreased to even a
lower value.

Discussions

Q1.Discuss different methods for flow visualisation.

The various flow visualization methods are:
1.Dye visualization : This method is employed in a water tunnel or water channel. The method is to
introduce dye in the flow field by various methods like using a dye probe, using ports or slits as part of
the model or coating the test model with concentrated solution of dye and alcohol.
2. Oil-film method: This method is to apply some viscous fluid mixed with solid powder, whose color is
easily visualized by eye, on the tested model. Having had the model in the test section of a wind tunnel
subject to a desirable speed, the pattern revealed by the oil film provides the information of flow near
the model surface, i. e., limiting streamline pattern which implies the shearstress pattern.
3. Smoke-wire technique : With oil coated on a wire, which is heated by electric current continuously,
the smoke is generated and carried downstream by the flow to reveal the flow motion.
4. Pulsed-wire technique: Same as the smoke-wire set-up, except that the wire is heated by pulsed
electric current.
5.Particle tracer methods: In this particles are added to the flow and illuminated. These particles (seeds)
have same physical properties but different optical properties.eg-Particle Image Velocimetry.
6.Optical methods - Some flows reveal their patterns by way of changes in their optical refractive index.
More directly, dyes can be added to (usually liquid) flows to measure concentrations. It includes the
following :
1. Shadowgraph
2. Schlieren
3.Interferometry
2. What are the advantages and disadvantages of smoke flow visualisation?
2. The following are the advantages and disadvantages of smoke flow visualization:
2.1.Advantages:
2.1.1. Easy to produce-Producing smoke is quite cheap and easy and can even be done without any
sophisticated equipment.
2.1.2. Quick positioning-Which helps to visualize the flow at any point on the surface of the body

2.2 Disadvantages
2.2.1. Extended use fills the tunnel with smoke. The tunnel must be vented to remove the smoke
2.2.2. An oily residue is left on whatever the smoke touches.
2.2.3. Complex 3D flows cannot be visualized using this.
2.2.4. Turbulent flows cannot be properly analyzed using this technique.
2.2.5 Use of materials like 4 in smoke flow can produce toxic gases which is harmful

3.Explain whether the smoke lines observed in the wind tunnel are streamlines, streaklines, or both?
3. The smoke lines observed in the experiment were streak lines. This goes by the definition of streak
lines itself, i.e, the locus of all points originating from a same point in the past. If the flow is to be
considered as a steady flow, then the smoke lines can also be considered to be streamlines as the flow is
already incompressible.
4.Can we apply Bernoulli's principle to this flow? Why or why not?
4. Bernoullis equation is applicable to an irrotational, inviscid flow. In both the cases of cylinder and the
airfoil, there exists vortices which can be expressed as curl of velocity with some manipulation. Thus it is
not valid for these flows. But it can be applied on those stream lines which are not a part of the vortices,
that is, it is applicable in case of airfoil when the flow is still attached to the surface.

5.What kind of vortices were generated for the flow over cylinder and airfoil? Discuss separately.
5. The vortices generated in case of cylinder were alternate periodic vortices from the top and bottom
surface of the cylinder. The formation stability and shedding was explained by Von Karman thus named
as Karman Vortex Street. The alternation leads to the core of a vortex in one row being opposite the
point midway between two vortex cores in the other row. The eddies grew in size downstream and
become weaker as the energy of the vortices is consumed by viscosity as they move further
downstream, and the regular pattern disappears. The frequency of production of eddies increased as
the flow velocity was increased.
In the case of airfoil, the vortices formed are known as the starting vortices
The formation of these vortices can be reasoned with Kelvins circulation theorem. According to this
circulation around a closed curve which encloses the same fluid element moving with the fluid, remain
constant with time. Thus if a vortex element is present in the fluid domain, it cannot end in it. We know
that there exists a bound vortex in an airfoil. Since, when the flow started, there was no circulation
present in the fluid domain, thus to counter the above there exists a vortex filament with opposite sense
of rotation.
The other type of vortices that are formed is when the airfoil has stalled. When this happens, then there
is a continuous formation of vortices from the trailing edge which are of similar behaviour(formation of
vortex in both clock wise and anti-clock wise direction from the upper and the lower surface ) as those
formed in the case flow of cylinder.
References
1. Smits, A. J. and Lim, T. T. (ed.) (2000). Flow Visualization. Techniques and Examples. London: Imperial
College Press.
2. Lienhard, J.H "Synopsis of Lift, Drag and Vortex Frequency Data for Rigid Circular Cylinders," College of
Engineering Bulletin No. 300, Washington State University, Pullman. Washington, 1966.
3. https://en.wikipedia.org/wiki/Flow_visualization.