Content uploaded by Tze Chuen YAP
Author content
All content in this area was uploaded by Tze Chuen YAP on Nov 22, 2014
Content may be subject to copyright.
4th International Conference on Mechanical Engineering, December 26-28, 2001, Dhaka, Bangladesh/pp. IV 31-36
Section IV: Fluid Mechanics 31
THE EFFECT OF TURBULENCE INTENSITY ON THE AERODYNAMIC
PERFORMANCE OF AIRFOILS
T.C.Yap*, M.Z.Abdullah,
Z.Husain, Z. Mohd Ripin, R. Ahmad
School of Mechanical Engineering, Universiti Sains Malaysia,
Engineering Campus, 14300 Nibong Tebal,
Pulau Pinang, MALAYSIA.
Abstract The experiments have been carried out in low speed, open circuit wind tunnel at the
School of Mechanical Engineering, USM to study the effect of turbulence intensity on the airfoil’s
aerodynamic performance. Two types of airfoil i.e. NACA 0015 and Eagle 150 wing airfoils, are
tested at three different Reynolds number. Three different density of wire-mesh are placed before
the wind tunnel test section in order to generate turbulence in the range of 2.4% to 5.4%. The mean
velocity and the turbulence intensity of the free stream flow are measured using a two-component
Laser Doppler Anemometer (LDA). The results show that the increase in turbulence intensity
delayed the stall angle but increased the lift and drag coefficients. The results obtained from the
NACA 0015 and Eagle airfoil show almost similar trend. The results also show the stall is delayed
with the increase of Reynolds number.
Keywords: Turbulence Intensity, Lift coefficient, Drag coefficient, Laser Doppler Anemometer
INTRODUCTION
Most of the aircraft and turbo-machine work in
turbulent environment, the level of the turbulence will
affect the flow's boundary layer separation. The
aerodynamics characteristic of an airfoil is mainly
depended on the flow characteristic (separation and
reverse flow). As a result, the level of turbulent also
affects the lift and drag coefficients of the airfoil. This
information could help the designers and engineers to
improve the performance of the aircraft or turbo
machine.
Many investigators have studied the influence of
turbulence to the separation bubbles in turbines blade
aerodynamics and aerofoil performance. Hiller and
Cherry (1981) have studied the effects of the stream
turbulence on two-dimensional, separated and re-
attached flows. They found that the mean flow-field
responds strongly to the turbulence intensity but with
little effect on integral scale and fluctuating pressures
depend strongly upon both intensity and scale.
However, the mechanism of turbulence interaction with
the shear layer is unclear.
Butler et al. (2001) have studied the effect of
turbulence intensity and length scale on low-pressure
turbine blade aerodynamics. They found that for low
Reynolds numbers (4.5×104-8×104), the boundary layer
on the suction surface of the turbine blade always
separated at lower turbulence intensity (0.4%-0.8%),
increased the turbulence to a higher level (10%) could
prevent the separation and the boundary layer transition
to turbulent.
Mueller and Pohlen (1983) have studied the influence
of turbulence intensity on the Lissaman 7769 airfoil.
They have increased the nominal turbulent intensity
from 0.08% to 0.30%, and tested at the Reynolds
numbers below 3.0×105. They concluded that the
increase in turbulent intensity could eliminate the
hysteresis region, which occurs at the lift, and drag
coefficients results. The increase in free stream
turbulence and acoustic excitation also caused the
laminar shear layer transformed into the transition
region much earlier, thus allowing the flow to reattach.
Hoffmann (1991) has studied on the NACA 0015
airfoil at Reynolds number of 2.5x105. The results show
that the increasing in turbulent intensity from 0.25% to
9% has resulted 30% increased in maximum lift
coefficient. At a higher turbulent intensity (9%), the
maximum lift coefficient reached the saturation. The
results also show that the increase in turbulent intensity
increased the drag coefficient, however, the rate of
change is negligible.
Huang and Lee (1999) had different results, they used
NACA 0012 in their investigation and the Reynolds
number ranged from 5x104 to 1.4x105. Huang and Lee
only investigated turbulent intensity in the limited range
of 0.2% to 0.65%. They found that the variation of lift
and drag are closely related to the behavior of surface
flow. The surface flow and L/D at low free stream
*Email: yaptc99@yahoo.com
ICME 2001, Dhaka, December 26-28
Section IV: Fluid Mechanics 32
turbulence are different from a higher free stream
turbulence (>0.45%). The lift coefficient increased with
the increase in turbulence intensity up to 0.45%.
However, for the turbulence intensity higher than
0.45%, the lift coefficient decreased with the turbulence
intensity. They concluded that the drag coefficient
increases and the ratio of lift and drag coefficient
decreases with the increase in turbulence intensity. At
the lower turbulence intensity (less than 0.45%), the
increasing of turbulence intensity has delayed the stall
angle, however, at higher than 0.45% its influence is
negligible.
EXPERIMENTAL SETUP AND PROCEDURE
The experiments are carried out in the low speed,
open-circuit wind tunnel at the School of Mechanical
Engineering, Universiti Sains Malaysia. The wind
tunnel has a 300 x 300 x 600 mm Plexiglas’s test
section with three components electronic balance for the
measurement of lift, drag and turning moment. The
maximum velocity in the wind tunnel is 38m/s. The
flow’s mean velocity and the fluctuation values are
measured by a DANTEC two component Laser Doppler
Anemometer (LDA), and using a Spectra-Physics
Model 177-G0232 with an air-cooled 300mW Argon
ion laser as the light source. The two-component system
used each the blue and the green laser light for both
components. Signal analysis is obtained by a 58N40
Flow Velocity Analyzer enhanced processor and present
in the FVA software. To measure the desired point more
effectively, the laser probe is mounted on a traversing
mechanism that can be controlled by the FVA software
on the computer.
A smoke generator using the Shell ondina oil 15 is
used as the seeding in this experiment. The smoke
ejector is placed in front of the wind tunnel inlet to
allow the smoke flow into the test section (Fig. 1). The
purpose of this seeding is to allow the laser beam detect
the flow velocity. The airfoils are made from fiberglass
and both ends joined with plates, resulting in a
rectangular box-shaped (bi-wing) assembly. One of the
sides of the model is attached to a rod and connected to
the wind tunnel’s electronic balancing unit.
In order to generate different turbulence intensities in
the test section, the mesh screen with different mesh
density and wire diameter are put after the intake just
before the test section. The mesh density, wire diameter
and the turbulence intensity generated in the experiment
at different Reynolds number are listed in Table 1.
In these experiments, the lift and drag forces of
NACA 0015 and the Eagle’s airfoils are investigated at
three different Reynolds numbers i.e. Re=6.4x104,
1.27x105 and 1.91x105 corresponding to three free
stream velocity of 10m/s, 20m/s and 30m/s respectively.
Four different turbulence intensities are generated in the
experiments and tested at various angles of attack from
0 to 15o.
RESULTS AND DISCUSSION
NACA 0015 Airfoil
Fig. 2 shows the variation of lift coefficient with respect
to angle of attacks of NACA 0015 airfoil at a Reynolds
number of 6.4x104. At the lowest turbulence intensity of
2.45%, the lift coefficient is increased (increment rate ≈
COMPUTER DANTEC LDA SYSTEM
HONEYCOMB MESH SCREEN
LDA PROBE ON TRANVERSE SYSTEM
SMOKE GENERATOR
FAN
AIRFOIL
WIND
TUNNEL
Fig. 1: Experimental Setup
ICME 2001, Dhaka, December 26-28
Section IV: Fluid Mechanics 33
1.67π/rad) with the increase of the angle of attack up to
the stall angle (9o). After the stall angle, the lift
coefficient dropped rapidly. Fig. 2 also shows that the
increasing in the turbulent intensity causes the stall
angle occurs at the higher angle of attack, and also
increases the maximum lift coefficient. This is probably
due to the increase in turbulent kinetic energy produced
at the boundary layer with the higher energy on the
airfoil which delayed the flow separation. When the
stall angle occurs at a higher angle of attack, the lift
coefficient reaches a higher value of Clmax. The effect of
the turbulence intensity on the drag coefficient for
NACA 0015 airfoil is shown in Fig. 3. The result show
that at Re= 6.4x104, the increase in turbulence intensity
caused small increase in the drag coefficient. The result
also shows that the drag coefficient increases slowly
with the increase in the angle of attack (increment rate ≈
0.12π/rad) until it reaches the stall angle, and at the stall
point, the drag coefficient increased suddenly with a
higher slope (≈1.27π/rad).
In Fig. 4 and 5 show the lift and drag coefficients
against angle attack of NACA 0015 at different
turbulent intensity for Re=1.27x105. The results show
similar trends as obtained for Re=6.4x104, the lift and
drag coefficients increase with the increase of the
turbulence intensity.
Furthermore, Fig. 4 and 5 also show the stall angle at
Re=1.27x105 is higher than the stall angle at
Re=6.4x104 (in Fig. 2 and 3), it illustrates that the
increasing of Reynolds number delayed the stall angle.
The Fig. 6 and 7 show variation of lift and drag
coefficients versus angle of attack at higher Reynolds
number (Re=1.91x105), the stall angle is delayed by the
increase in the Reynolds number and the turbulence
intensity. The variation of maximum lift coefficient
with the Reynolds number is shown in Fig. 8. The
maximum lift coefficient is increased with the
turbulence intensity; however, the rate of increment is
not linear with the increment of turbulent intensity. The
result shows that the increase in turbulent intensity
increased the maximum lift coefficient, however, when
the Re increases, the Clmax does not increase as
expected. The Clmax decreases when the Re increased
from 6.4x104 to 1.27x105, however, after Re=1.27x105,
the Clmax increases.
Table 1: Turbulence intensities at different mesh and Reynolds number
Mesh
screen
number
Mesh density
(mesh/cm)
Wire
diameter
(mm)
Turbulence intensity
at Re=6.4x10 4
(%)
Turbulence intensity
at Re=1.27x105
(%)
Turbulence
intensity at
Re=1.91x105
(%)
No mesh - - 2.45 2.39 1.81
M1 1.081 0.9 3.03 3.14 2.80
M2 3.077 0.7 3.36 3.41 3.07
M3 0.769 4 5.39 5.27 -
-2 0 2 4 6 8 10 12 14 16 18
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4 2.45%
3.03%
3.36%
5.39%
Lift c oefficient, C l
Angle of att ack,a
aa
a
Fig 2: Lift coefficient versus angle of attack for
NACA 0015 at Re = 6.4x104.
-20246810121416
0.00
0.05
0.10
0.15
0.20
0.25
0.30 2.45%
3.03%
3.36%
5.39%
Drag coeffi cient,C d
Angle of att ack
Fig 3: Drag coefficient versus angle of attack for
NACA 0015 at Re=6.4x104
ICME 2001, Dhaka, December 26-28
Section IV: Fluid Mechanics 34
-2 0 2 4 6 8 10 12 14 16 18
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2 2.39%
3.08%
3.41%
5.27%
Lift coefficient, Cl
Angle of attack
Fig. 4: Lift coefficient versus angle of attack Re=
1.27x105 (NACA 0015)
-2024681012141618
0.00
0.05
0.10
0.15
0.20
0.25
0.30
2.39%
3.14%
3.41%
5.27%
Drag coefficient, Cd
Angle of attack
Fig. 5: Drag coefficient versus angle of attack at
Re=1.27x105 (NACA 0015)
-2 0 2 4 6 8 10 12 14 16
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.81%
2.8%
3.8%
Lift coef ficient, Cl
Angle of attack
Fig. 6: Lift coefficient versus angle of attack at Re=
1.91x105 (NACA 0015)
-20246810121416
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35 1.81%
2.8%
3.07%
Drag coefficient, Cd
Angle of at tack
Fig. 7: Drag coefficient versus angle of attack at
Re=1.91x105 (NACA 0015)
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
0.95
1.00
1.05
1.10
1.15
1.20
1.25
1.30
1.35 Reynolds numbers
6.4E4
1.27E5
1.91E5
Cl max
T(%)
Fig. 8: Variation of maximum lift coefficient with
Reynold’s number
-2 0 2 4 6 81012141618
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
2.36%
3.06%
3.42%
5.21%
Lif t coeffi cient
Angle of att ack
Fig. 9: Lift coefficient versus angle of attack of Eagle
airfoil for Re=6.4x104
ICME 2001, Dhaka, December 26-28
Section IV: Fluid Mechanics 35
-2024681012141618
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0.35
0.40
0.45
2.36%
3.08%
3.42%
5.21%
Drag coefficient
Angle of attack
Fig. 10: Drag coefficient versus angle of attack of
Eagle airfoil at Re=6.4x104
-2024681012141618
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
2.54%
3.14%
3.46%
Lift coefficient
Angle of attack
Fig. 11: Lift coefficient versus angle of attack of
eagle airfoil at Re=1.27x105
-2 0 2 4 6 8 10 12 14 16 18
0.00
0.05
0.10
0.15
0.20
0.25
0.30
2.54%
3.14%
3.46%
Drag coefficient
Angle of attack
Fig. 12: Drag coefficient versus angle of attack of
eagle airfoil at Re=1.27x105
-2024681012141618
-0.2
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.79%
2.78%
3.1%
Lift coefficient
Angle of attack
Fig. 13: Lift coefficient versus angle of attack of
eagle airfoil at Re=1.91x105
-2024681012141618
0.00
0.05
0.10
0.15
0.20
0.25
0.30
1.79%
2.78%
3.1%
Drag coefficient
Angle of attack
Fig. 14: Drag coefficient versus angle of attack of
eagle airfoil at Re=1.91x105
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5
1.1
1.2
1.3
1.4
1.5
Reynolds number
6.4E4
1.27E5
1.91E5
Cl max
T(%)
Fig. 15: Variation of maximum lift coefficient with
turbulent intensity
.
ICME 2001, Dhaka, December 26-28
Section IV: Fluid Mechanics 36
Eagle 150 wing airfoil
The results obtained from the Eagle airfoils are almost
similar with the NACA 0015’s airfoil. The Fig. 9 shows
the relation between the lift coefficient and the
turbulence intensity at various angles of attack.
Generally, it shows that lift coefficient increases as the
turbulent intensity increases.
The lift coefficient increases up to the stall angle and
after the stall angle, the lift coefficient begin to
decrease, however, the decrement rate is much slower
than the slope of the NACA 0015 at the same situation.
Fig. 10 shows the relation between the drag coefficient
and the turbulent intensity at Re= 6.4x104. The drag
coefficient also increases when the angle of attack
increases. The results show the increment rate is small
at the beginning, however, after the stall, the increment
rate becomes steeper. This is mainly caused by suddenly
increase in the pressure drag force due to flow
separation.
The Fig. 11 shows the results of the lift coefficient at
various angle of attack on Eagle 150 airfoil with three
different turbulence intensity, Ti and at Reynolds
number, Re=1.27x105. The stall angles are 9, 11 and 13°
for the turbulence intensities of 2.54, 3.14 and 3.46%
respectively. The results show that the increase in
turbulence intensity resulted in delaying of the stall
angle. Fig. 12 shows the drag coefficient of Eagle airfoil
at different turbulence intensity versus the angle of
attack. The results show that increase in the angle of
attack resulted in a slight increase in the drag
coefficient, and the drag coefficient increased suddenly
at the stall angle. Fig. 13 and 14 show the results of the
lift coefficient at the higher Reynolds number, Re=
1.91x105. At Re=1.91x105, the airfoil showed a similar
trends as with the previous investigations. The increased
of the turbulent intensity causes delay of the stall angle,
and provided higher Clmax. Fig. 14 shows that the
increase of the turbulence intensity could increase the
drag coefficient, however, the influence of the higher
turbulence intensity to the drag coefficient is negligibly
small. The slope of the drag coefficient perform almost
constant until it reach the stall angle and than increased
rapidly after the stall angle.
The variation of maximum lift coefficient, Clmax with
the turbulent intensity is shown in Fig. 15. At
Re=6.4x104, the rate of increasing of Clmax is almost
linear (≈0.339) for turbulence intensity of 3.5%.
However, the turbulence intensity above 3.5%, the
increment rate becomes lower (≈0.011). At higher
Reynolds number, Re= 1.27x105, the increment rate is
about 0.345 and at Re=1.91x105 increment rate is 0.095.
The results also illustrate that in order to increase the
maximum lift coefficient, two methods could be used,
either increase the turbulent intensity or increase the
Reynolds number.
Comparison
In the investigations, two types of airfoil behave
differently at different Reynolds number and influence
of turbulence intensity has profound. The experiments
show the results of lift and drag coefficients for both
airfoils have similar trends at different turbulent
intensity. In general for particular values of Reynolds
number and turbulence intensity the Eagle airfoil has
higher Clmax compared to NACA 0015. The maximum
lift curve of Eagle 150 airfoil is more stable compared
to NACA 0015’s.
CONCLUSION
The effect of turbulence intensity on the aerodynamics
performance of the NACA 0015 and Eagle 150 airfoils
is profound. The increase in air free stream turbulence
intensity causes delay of the stall angle and the
maximum lift coefficient. However, it causes the
increase in drag coefficient.
ACKNOWLEDGEMENT
This study is supported by the IRPA long-term
research grant provided by the Ministry of Science,
Technology and Environment Malaysia. The authors
also like to thank the National Science Fellowship
(NSF) department for the financial support.
REFERENCES
Butler, R.J., Byerley, A.R., VanTreuren, K., and
Baughn J.W., “The Effect of Turbulence Intensity
and Length scale on Low-pressure Turbine Blade
Aerodynamics”, International Journal of Heat and
Fluid Flow, 22, pp. 123-133 (2001).
Hillier, R. and Cherry, N.J. “The Effects of Stream
Turbulence on Separation Bubbles”, Journal of
Wing Engineering and Industrial Aerodynamics, 8,
pp. 49-58 (1981).
Hoffmann, J.A. “Effects of Free stream Turbulence on
the Performance Characteristics of an Airfoil”,
AIAA Journal, 29(9), pp. 1353-1354 (1991).
Huang, R.F., and Lee, H.W. “Effects of Free stream
Turbulence on Wing-Surface Flow and
Aerodynamic Performance”, Journal of Aircraft,
36(6), pp. 965-972 (1999).
Mueller, T.J., Pohlen, L.J., 1983, “The Influence of
Free-Stream Disturbances on Low Reynolds
Number Airfoil Experiments”, Experiments in
Fluids, 1, pp. 3-14 (1983).
