40th Fluid Dynamics Conference and Exhibit AIAA 2010-4629

28 June - 1 July 2010, Chicago, Illinois

Aerodynamic Characterization of a NACA 0018 Airfoil at

Low Reynolds Numbers

Ryan Gerakopulos,* Michael S. H. Boutilier,† and Serhiy Yarusevych‡

University of Waterloo, Waterloo, Ontario, N2L 3G1, Canada

The lift and separation bubble characteristics of a NACA 0018 airfoil are investigated
experimentally. Surface pressure measurements are presented for Reynolds numbers from
80x103 to 200x103 and angles of attack from 0° to 18°. These data were used to characterize
the separation bubble and determine lift coefficients. From these results, two distinct regions
in the lift curves can be identified: a region of rapid and linear growth of the lift coefficients
at low angles of attack and a region of more gradual and linear growth at higher pre-stall
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

angles. Furthermore, the slope of the lift curve in each region is found to be linked to the
rates of change in separation, transition, and reattachment locations with the angle of attack.
These findings are substantiated by an analysis of the available experimental data for a
NACA 0012 airfoil.

Nomenclature
c = airfoil chord length
Cl = lift coefficient
Cp = surface pressure coefficient
R = reattachment location
Rec = chord Reynolds number, Uoc/ν
S = separation location
T = transition location
Uo = free-stream velocity
x = streamwise coordinate
α = angle of attack
ν = kinematic viscosity
d( )/dα = rate of change of ( ) with angle of attack

I. Introduction

A wide range of engineering devices employ airfoils operating at relatively low chord Reynolds numbers.
Specifically, airfoil performance in the chord Reynolds number range from 104 to 106 is of interest for such
applications as small-to-medium scale wind turbines and unmanned aerial vehicles.1-4 In this Reynolds number
range, the laminar boundary layer on the upper surface of an airfoil is susceptible to separation, even at low angles
of attack. When laminar separation occurs, the evolution of the separated shear layer has a strong influence on the
entire flow field. For Reynolds numbers below about 30x103, the flow does not reattach,5 leaving a wide wake
behind the airfoil. For higher Reynolds numbers, the separated shear layer undergoes laminar-to-turbulent transition
over the airfoil surface. This may result in flow reattachment, closing the recirculating flow into a separation bubble.
Below a Reynolds numbers of about 300x10 3, the separation bubble may occupy upwards of 15% of the chord.6
Independent of the flow regime, flow separation usually has a detrimental effect on airfoil performance and may
also contribute to undesirable noise generation. Thus, knowledge of the existence and extent of the separated flow
region is vital in low Reynolds number airfoil design and performance assessment.


Graduate student, Mechanical & Mechatronics Engineering, 200 University Ave. W, AIAA student member.
†
Graduate student, Mechanical & Mechatronics Engineering, 200 University Ave. W, AIAA student member.
‡
Assistant Professor, Mechanical & Mechatronics Engineering, 200 University Ave. W, AIAA member.

1
American Institute of Aeronautics and Astronautics

Copyright © 2010 by S.Yarusevych. Published by the American Institute of Aeronautics and Astronautics, Inc., with permission.
 A variety of qualitative and quantitative experimental techniques have been employed to examine airfoil
performance7-9 and airfoil boundary layer characteristics6,10-12 at low Reynolds numbers. Previous studies have
demonstrated that both the angle of attack and the Reynolds number have a significant effect on separated shear
layer characteristics over an airfoil. 6,8,11 When a separation bubble forms on the upper surface of an airfoil,
increasing the Reynolds number reduces the length of the bubble.6,8,10 Furthermore, the laminar separation location
propagates toward the leading edge with increasing angle of attack until stall.6,10 It should be noted that, despite
considerable effort to understand the evolution of laminar separated shear layers, available models for characterizing
the separated flow region on an airfoil still depend on empirically derived criteria and tend to be unreliable.13
Symmetric airfoils have been used extensively as aerodynamic test models. A NACA 0012 profile is arguably the
most studied symmetric airfoil. McCroskey14 reviews experimental aerodynamic performance data from over forty
different wind tunnel experiments on a NACA 0012 profile, some performed at low Reynolds numbers. In addition
to the results available for a NACA 0012 profile, a substantial experimental database is available for a NACA 0015
airfoil.15-19 Previous studies performed on these two profiles have advanced understanding of the influence of vortex
shedding, near wake development, and boundary layer characteristics on the aerodynamic performance of an airfoil
at low Reynolds numbers.20-23
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

Thick airfoil sections are required in applications such as vertical axis wind turbines and water turbines to
improve blade stiffness. 2 Since the blades oscillate through positive and negative angles of attack, relatively thick
symmetric airfoils are used in these applications to maintain uniform blade loading.3,24 However, the extensive
collection of literature available for thin symmetric airfoils at low Reynolds numbers is not sufficient for accurate
design estimates over thicker symmetric airfoils.
An experimental investigation of the effect of airfoil thickness on aerodynamic performance at low Reynolds
numbers was undertaken by Jacobs and Sherman25 for various NACA profiles within the thickness range from 9% to
18%. The results cover a Reynolds number range from approximately 40x10 3 to 3x106 and angles of attack from -4°
to 27°. Because these experiments were performed in a wind tunnel with a relatively high free-stream turbulence
intensity of about 2%, the results are expected to differ from those obtained in a low free-stream turbulence
facility.26 A more recent investigation performed by Sheldahl and Klimas24 presents experimental performance data
for symmetric 4-digit NACA profiles with thicknesses from 9% to 15%, Reynolds numbers from 350x103 to 106,
and angles of attack from -25° to 25°. These experimental results were also used to extrapolate performance data for
profiles of 18% to 25% thickness.
Timmer,27 Raghunathan and Ombaka,3 and Nakano et al.28 have investigated a NACA 0018 airfoil in low
Reynolds number flows. Timmer,27 reports lift coefficient data for several angles of attack from 0° to 30° and
Reynolds numbers ranging from 150x103 to 106. Raghunathan and Ombaka3 report surface pressure coefficient
distributions and lift coefficients for angles of attack from 0° to 16° and Rec = 240x103. Nakano et al.28 measured
surface pressure distributions for α = 0°, 6° and 15° and measured the velocity field at α = 6° for Rec = 160x103.
Yarusevych et al.29-31 have examined the boundary layer and turbulent wake development on a NACA 0025 airfoil at
angles of attack of α = 0°, 5°, and 10° for Reynolds numbers from approximately 50x10 3 to 200x103.
This study is motivated by the need to complement the presently limited body of knowledge for thick symmetric
airfoils at low Reynolds numbers. Thus, the present investigation is focused on examining the effect of the Reynolds
number and the angle of attack on the performance characteristics of a NACA 0018 airfoil and relating the
performance characteristics to the separated shear layer development. With this choice of airfoil profile, the results
will also facilitate establishing trends in thickness effects on airfoil characteristics.

II. Experimental Setup and Procedure

Experiments reported here were conducted in an adaptive-wall open-return suction-type wind tunnel at the
University of Waterloo. The 6.0 m long rectangular test section of the tunnel has a height of 0.89 m and a width of
0.61 m. Flow enters the test section through a honeycomb and four screens positioned upstream of a 9:1 contraction,
with a background turbulence intensity level less than 0.3%. The uncertainty of the free-stream velocity
measurements was estimated to be less than 2.5%.
A NACA 0018 aluminum airfoil with a chord length of 0.2 m and a span of 0.61 m was mounted horizontally in
the test section 2 m downstream of the contraction. The airfoil model spanned the entire width of the tests section.
Angle of attack adjustments were facilitated by a model-support mechanism, with an angular resolution of 0.1°.
Surface pressure measurements were obtained from 65 pressure taps symmetrically distributed on the upper and
lower surfaces along the model midspan. A pressure tap installed in the test section upstream of the model served as
a reference free-stream static pressure. A multichannel digital pressure scanner was employed to acquire pressure
measurements. For data presentation, the origin of the coordinate system is located at the leading edge of the airfoil.

2
American Institute of Aeronautics and Astronautics
 For all of the cases investigated, the uncertainty of the surface pressure measurements is estimated to be less than
about 2% of the dynamic pressure.

III. Experimental Results

Experimental results presented here pertain to a range of Reynolds number from 80x10 3 to 200x103 and a range of
angles of attack from 0° to 18°. In this parameter range, both boundary layer flow regimes are observed: flow
separation with subsequent reattachment and flow separation without subsequent reattachment.

A. Lift Coefficient Curves

Lift coefficient data, computed based on surface pressure distributions, are presented in Fig. 1a. The results show
that, on the average, the stall angle increases with increasing the Reynolds number. An increase in the Reynolds
number from Rec = 80x103 to Rec = 200x103 results in an increase in the stall angle from α = 10° to α = 14° (Fig. 1a)
and an increase in the maximum lift coefficient from 0.89 to 1.03 (Fig. 2). For a given angle of attack in the range
0° ≤ α ≤ 6°, the lift coefficient decreases as the Reynolds number increases. In contrast, at higher angles of attack up
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

to the stall angle, the lift coefficient increases with increasing Reynolds number.

a) b)

Figure 1. Lift coefficient a) present study and b) a comparison with previous investigations.

For a given Reynolds number, a conventional nearly linear growth of the lift coefficient with increasing angle of
attack occurs from α = 0° to α ≈ 6° (Fig. 1a). This is followed by a more gradual growth up to the stall angle. At the
stall angle, a sudden and significant reduction in the lift coefficient is observed. The lift coefficient subsequently
recovers with increasing post-stall angle of attack.
Present results are compared with experimental lift coefficient data from other studies in Fig. 1b. The lift
coefficient data of Timmer27 were acquired using a force balance for Rec = 150x103 and 300x103. Agreeing well
with the present findings, the results of Timmer27 indicate that, for 0° ≤ α ≤ 7°, higher lift coefficients are obtained at
lower Reynolds numbers for a given angle of attack. A reverse trend is observed for α = 7° and up to the stall angle.
Furthermore, these two angle of attack ranges correspond to distinctly different lift curve slopes, similar to the trend
observed in the present study. The present results compare well with the data of Jacobs and Sherman25 and of
Raghunathan and Ombaka,3 for α < 6°. The variation between these data sets at higher angles of attack is attributed
to variation in experimental conditions, which are known to have a significant effect on flow development over an
airfoil at low Reynolds numbers.7,26,32

B. Pressure Coefficient Distributions

To investigate the development of the separated flow region over the upper surface of an airfoil, surface pressure
measurements were acquired and used to estimate locations of flow separation, transition, and reattachment. The
laminar boundary layer separation point can be approximately identified as the start of the region over which the

3
American Institute of Aeronautics and Astronautics
 1.2

0.8

Cl,max
0.6

0.4

0.2 present study

25
Jacobs & Sherman
Raghunathan & Ombaka 3
Timmer 27
0
50000 100000 150000 200000 250000 300000 350000
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

Re
c

Figure 2. Maximum lift coefficient. Note that discontinuous increase in the maximum lift coefficient with the
Reynolds number is attributed to a relatively coarse angle of attack increment of 2° used to acquire data near
the stall angle in the present study.

surface pressure is nearly constant.1 When the flow fails to reattach, this constant-pressure region extends to the
trailing edge.1 Conversely, a sudden surface pressure recovery following the constant-pressure region is an
indication of transition and subsequent flow reattachment. Thus, a constant-pressure region followed by sudden
surface pressure recovery signifies the presence of a separation bubble on the airfoil surface.
Upper surface pressure coefficient distributions for the Reynolds numbers and angles of attack investigated are
presented in Fig. 3. For a given Reynolds number, the separation bubble moves upstream toward the leading edge
and the length of the separation bubble decreases as the angle of attack increases. Increasing the angle of attack up
to the stall angle causes the separation bubble to burst near the leading edge, resulting in a sudden and significant
decrease in the suction peak. For example, for Rec = 80x103, the separation bubble is present for angles of attack
from 0° to 10° (Fig. 3a). Increasing the angle of attack to α ≈ 12° causes the bubble to burst, reducing the suction
peak by approximately 80%.
As the Reynolds number increases from 80x103 to 200x103, the stall angle increases from 10° to 14°, with a
corresponding increase in the maximum suction peak from Cp = -2.6 to Cp = -4.0 (Figs. 3a and 3f). Upper surface
pressure coefficient distributions for Reynolds numbers from 80x103 to 200x103 at α = 8° are presented in Fig. 4.
The results show that, for a given angle of attack, an increase of the Reynolds number causes a reduction in the
length of the separation bubble, evidenced by a decrease in the length of the nearly-constant pressure region
centered at about x/c = 0.25.

C. Separation Bubble Characteristics

In the present study, the following methodology was employed to estimate the locations of boundary layer
separation, transition, and reattachment, denoted by S, T, and R respectively. As illustrated in Fig. 5, the separation
location was estimated as the intersection of (i) a linear fit to the nearly linear surface pressure recovery region
following the suction peak and (ii) a linear fit to the nearly constant surface pressure region within the laminar
portion of the separation bubble. To estimate transition and reattachment locations, a shape-preserving polynomial
fit to the discrete surface pressure distribution in the regions of constant surface pressure and subsequent rapid
surface pressure recovery was used. The transition location, associated with a sudden increase in surface pressure in
the aft portion of the separation bubble,33 was estimated as the local maximum in the second derivative of the
polynomial fit. O’Meara and Mueller10 proposed that the reattachment location can be identified as the location
downstream of the transition point where a rapid decrease in the rate of surface pressure recovery is observed.10 In
accordance with this approach, the reattachment location was estimated as the location of the local minimum in the
second derivative of the polynomial fit. This methodology was verified on published pressure distributions for
which estimates of the S, T, and R locations determined based on velocity measurements were available. The
uncertainty in the S, T, and R locations determined using the described methodology was estimated to be less than
0.04c in the present investigation.

4
American Institute of Aeronautics and Astronautics
 a) Rec = 80x103 b) Rec = 100x103
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

c) Rec = 120x103 d) Rec = 140x103

e) Rec = 160x103 f) Rec = 200x103

Figure 3. Upper surface pressure distributions.

5
American Institute of Aeronautics and Astronautics
 −3
3
Re=80x10
3
Re=100x10
3
−2.5 Re=120x10
3
Re=140x10
3
Re=160x10
3
Re=200x10
−2

−1.5
Cp

−1

−0.5

0.5
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

x/c

Figure 4. Upper surface pressure distributions at α = 8°.

Figure 6 depicts the effect of the angle of attack on the S, T, and R locations. For all the Reynolds numbers
investigated, increasing the angle of attack results in the separation bubble propagating upstream and reducing in
length. For instance, for Rec = 100x103, the separation bubble located between 0.54 ≤ x/c ≤ 0.88 at α = 0° moves to a
region 0.08 ≤ x/c ≤ 0.24 at α = 10°, corresponding to a 50% reduction in the separation bubble length (Fig. 6b). A
similar trend is observed in the results of Nakano et al.28 obtained based on liquid-crystal visualization for
Rec = 160x103 (Fig. 6e). For α < 6°, separation and reattachment locations measured by Nakano et al.28 agree well
with present results obtained for the matching Reynolds number. However, for α ≥ 6°, there is deviation between the
corresponding data sets.
The effect of the Reynolds number on the S, T, and R locations is depicted in Fig. 7. The results suggest that at a
given angle of attack, the separation location does not vary appreciably with the Reynolds number. Furthermore, on
the average, both the transition and reattachment locations move upstream with increasing Reynolds number,
resulting in a reduction of the separation bubble length. For example, as the Reynolds number is increased from
80x103 to 200x103 at α = 6°, the separation location remains at approximately x/c = 0.19, the transition location
advances from x/c = 0.40 to 0.30, and the reattachment location advances from x/c = 0.48 to 0.33. As a result, the
separation bubble length is reduced by about 50%.

−3.5

−3 S T

−2.5

−2 R
Cp

−1.5

−1

−0.5

0
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.5
x/c

Figure 5. Estimates of separation, transition, and reattachment locations for Rec = 160x103 at α = 10°.

6
American Institute of Aeronautics and Astronautics
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

a) Rec = 80x103 b) Rec = 100x103

c) Rec = 120x103 d) Rec = 140x103

e) Rec = 160x103 f) Rec = 200x103

Figure 6. Variation of separation, transition, and reattachment locations with the angle of attack.

7
American Institute of Aeronautics and Astronautics
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

a) α = 0° b) α = 2°

c) α = 4° d) α = 6°

e) α = 8° f) α = 10°

Figure 7. Variation of separation, transition, and reattachment locations with the Reynolds number.

8
American Institute of Aeronautics and Astronautics
 D. Effect of Separation Bubble Size and Position on Airfoil Lift
An analysis of the lift coefficient data (Fig. 1) has revealed two distinct regions in the lift curves, which will be
referred to as Region I and Region II. Region I is associated with a linear growth in the lift coefficient at low angles
of attack, whereas Region II is associated with a more gradual linear growth in the lift coefficient at higher pre-stall
angles. These two regions are depicted in Fig. 8 for the present results and those of Timmer.27 Figure 8a illustrates
that, for 80x103 ≤ Rec ≤ 200x103, Region I extends to slightly higher angles of attack with increasing Reynolds
number. In contrast, the extent of Region II remains approximately constant, but shifts to higher angles of attack as
the Reynolds number increases. It should be noted that, the results of Timmer27 for Rec = 300x103 (Fig. 8b), show a
nearly constant lift coefficient at angles of attack approaching stall (13.5° ≤ α ≤ 17°).
Slopes of the lift curves within Regions I and II are plotted in Fig. 9 for the range of Reynolds numbers
investigated. The results show that the rate of change of the lift coefficient with the angle of attack in Region I is
about four times higher than that in Region II. The slope of the lift curve in Region I, on the average, decreases with
increasing Reynolds number for the cases investigated (Fig. 9), and the same trend can be seen in Timmer’s27 results
(Fig. 8b). Within Region II, the slope remains nearly constant for Rec ≥ 120x103 (Fig. 9), agreeing with Timmer’s27
data (Fig. 8b).
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

a) b)

Figure 8. Two distinct regions of linear growth in lift coefficient curves a) present results and b) data from
Timmer.27

Figure 9. Lift curve slopes for Regions I and II.

9
American Institute of Aeronautics and Astronautics
 The appearance of two distinct regions in the lift curves must be related to boundary layer and separated shear
layer development. A comparative evaluation of the lift coefficient results (Fig. 1) and the estimated separation
bubble parameters (Fig. 6) suggests that, within both Regions I and II, the lift curve slope is linked to the rate of
advancement of the S, T, and R locations. The results presented in Table 1 show that, within each of the two regions,
the rate at which the S, T, or R locations advance towards the leading edge as the angle of attack increases is
constant and proportional to the rate of change of the lift curve slope. This link is depicted in Fig. 10 for Rec =
140x103. A higher slope of the lift curve in Region I, dCl/dα = 0.12, corresponds to a more rapid advancement of S,
T, and R locations towards the leading edge, with dS/dα = 0.06, dT/dα = 0.07, and dR/dα = 0.08, respectively.
Furthermore, in Region II, a lower slope of the lift curve, dCl/dα = 0.02, corresponds to a more gradual change in S,
T, and R locations, with dS/dα = 0.02, dT/dα = 0.02, and dR/dα = 0.03, respectively. It should be noted that, within
either Region I or Region II, variations in dS/dα, dT/dα, and dR/dα with the Reynolds number are within the
experimental uncertainty associated with the estimation of the S, T, and R locations. Thus, the Reynolds number
effect on these quantities and the corresponding variation in dCl/dα cannot be assessed.

Table 1. Magnitudes of lift curve slopes and rates of upstream advancement of the S, T, and R locations.
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

Rec Regime dCl/dα dS/dα dT/dα dR/dα

Region I 0≤α≤6 0.14 0.06 0.07 0.08
80x103
Region II 6 ≤ α ≤ 10 0.03 0.03 0.04 0.04
Region I 0≤α≤6 0.12 0.06 0.07 0.08
100x103
Region II 6 ≤ α ≤ 10 0.04 0.03 0.04 0.04
Region I 0≤α≤6 0.13 0.06 0.08 0.08
120x103
Region II 8 ≤ α ≤ 12 0.03 0.02 0.02 0.02
Region I 0≤α≤6 0.12 0.06 0.07 0.08
140x103
Region II 8 ≤ α ≤ 12 0.02 0.02 0.02 0.03
Region I 0≤α≤8 0.11 0.05 0.07 0.07
160x103
Region II 10 ≤ α ≤ 14 0.02 0.01 0.02 0.02
Region I 0≤α≤8 0.11 0.05 0.06 0.07
200x103
Region II 10 ≤ α ≤ 14 0.03 0.01 0.01 0.02

a) b)

Figure 10: a) lift curve slopes and b) rate of upstream advancement of S, T, and R locations for Rec = 140x103.

10
American Institute of Aeronautics and Astronautics
 A similar relationship between the growth of the lift coefficient with angle of attack and the associated rate of
advancement of the S and T locations have been noted by Lee & Gerontakos23 for a NACA 0012 airfoil at low
angles of attack (i.e., within Region I). A comparison of their experimental results obtained for a single Reynolds
number, Rec = 135x103, and the present data for Rec = 140x103 is shown in Fig. 11. Although Lee & Gerontakos23
note the presence of the linear growth in the lift curve only at low angles of attack (Region I), Region II can be
identified in their lift data at pre-stall angles (Figs. 11a). Within this region, extending from 11° to 13°, more
gradual, linear growth of the lift coefficient corresponds to a more gradual advancement of the S and T locations
towards the leading edge, agreeing with the present findings. Comparing the results for the two airfoils, it can be
seen that, within Region I, the lift curve slope and the rate of advancement of the S and T locations towards the
leading edge are lower for the NACA 0012 airfoil. Decreasing the airfoil thickness extends Region I to higher angles
of attack but reduces the extent of Region II, with the latter terminating at the stall angle for both airfoils. Within
Region II, the rate of growth in lift coefficient is comparable for both NACA 0012 and NACA 0018 airfoils, and so
is the rate of change of the transition location with the angle of attack. Note that the difference in the slopes of the
corresponding separation position plots within Region II is attributable to the inherently higher uncertainty in
determining the separation location for the two experiments. The lift curves in Fig. 11a show that the thinner NACA
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

0012 profile produces lower lift in the range of angles of attack from about 3° until stall, hence, producing lower
maximum lift, but stalls at a higher angle of attack compared to the NACA 0018 airfoil.

1 0.8
S − NACA 0018, present study
0.9 T − NACA 0018, present study
0.7 S − NACA 0012, Lee & Gerontakos
23
23
T − NACA 0012, Lee & Gerontakos
0.8
0.6
0.7
0.5
0.6
Cl x/c
0.5 0.4

0.4
0.3
0.3
0.2
0.2
NACA 0018, present study 0.1
0.1
NACA 0012, Lee & Gerontakos23
0 0
0 5 10 15 20 0 2 4 6 8 10 12 14
α, deg α, deg
a) b)

Figure 11. Comparison of a) lift curves and b) S and T locations for a NACA 0018 airfoil at Rec = 140x103
(present study) and a NACA 0012 airfoil at Rec = 135x103 [Ref. 23].

IV. Conclusions
The lift and separation bubble characteristics of a NACA 0018 airfoil were investigated experimentally for
80x103 ≤ Rec ≤ 200x103 and 0° ≤ α ≤ 18°. Surface pressure distributions were measured and used to investigate the
relationship between airfoil performance and the development of the separated flow region. Furthermore, the
dependence of separation bubble characteristics on the Reynolds number and angle of attack were examined.
Examination of the pressure coefficient distributions reveal that, for a given Reynolds number, increasing the
angle of attack causes an upstream advancement of the separation bubble and a decrease in the separation bubble
length. At the stall angle, the bubble bursts, resulting in a sudden and significant reduction in the suction peak and
the lift coefficient. As the Reynolds number increases, on the average, the length of the separation bubble decreases
and the bubble persists at higher angles of attack. Consequently, the stall angle increases, and a higher maximum lift
coefficient is produced.
The analysis of the lift coefficient results from the present investigation and those from Timmer27 revealed two
distinct regions in the lift curve, referred to as Region I and Region II. Region I is associated with low angles of
attack, and Region II extends over higher pre-stall angles. Within both regions, linear growth of the lift coefficient
11
American Institute of Aeronautics and Astronautics
 with the angle of attack is observed; however, the slope of the lift curve is approximately four times higher in
Region I. For the Reynolds numbers investigated, Region I extends to higher angles of attack as the Reynolds
number is increased; whereas, the extent of Region II remains unchanged, but the region shifts to higher angles of
attack. Furthermore, on the average, the slope of the lift curve in Region I decreases with increasing Reynolds
number, such that the lift coefficient increases as the Reynolds number decreases at a given angle of attack within
0° ≤ α ≤ 6°.
A comparative analysis of the lift curves and pressure distributions showed that the two identified regions are
linked to distinct trends in the rate of advancement of the separation bubble towards the leading edge with increasing
angle of attack. Specifically, within either of the two regions, a constant lift curve slope is associated with constant
rates of advancement in S, T, and R locations towards the leading edge with increasing angle of attack. Moreover,
similar to the lift curve slopes, the rates of advancement of S, T, and R locations are substantially higher within
Region I compared to those within Region II. Analogous trends were observed in NACA 0012 airfoil measurements
obtained by Lee and Gerontakos.23 For a NACA 0012 profile, Region I extends to higher angles of attack. Within
Region II, a NACA 0012 airfoil has a lower rate of advancement in the S location than that observed in the present
study for a NACA 0018 airfoil. However, the lift slopes and associated rate of advancement in the T location within
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

Region II for the two profiles are comparable. Although stall occurs at a lower angle of attack for a NACA 0018
airfoil than for a NACA 0012 airfoil, the thicker section has higher maximum lift coefficient for Rec ≈ 140x103.

Acknowledgments
The authors thankfully acknowledge the Natural Sciences and Engineering Research Council of Canada for
funding of this work.

References
1
Carmichael, B. H., “Low Reynolds number airfoil survey,” NASA Contract Report No. 165803, Vol. I., 1981.
2
Tangler, J. L., and Somers, D. M., “NREL Airfoil Families for HAWTs,” American Wind Energy Association WindPower
’95 Conference, Washington, DC, 1995.
3
Raghunathan, S., and Ombaka, O. O., “A thick symmetrical aerofoil oscillating about zero incidence angle,” International
Journal of Heat & Fluid Flow, Vol. 7, No. 2, 1986, pp. 155-159.
4
Mueller, T. J., and DeLaurier, J. D., “Aerodynamics of Small Vehicles,” Annual Review of Fluid Mechanics, Vol. 35, 2003,
pp. 89-111.
5
Lissaman, P. B. S., “Low-Reynolds-Number Airfoils,” Annual Review of Fluid Mechanics, Vol. 15, 1983, pp. 223-239.
6
Brendel, M., and Mueller, T. J., “Boundary-Layer Measurements on an Airfoil at Low Reynolds Numbers,” Journal of
Aircraft, Vol. 25, No. 7, 1988, pp. 612-617.
7
Marchman, J. F., “Aerodynamic Testing at Low Reynolds Numbers,” Journal of Aircraft, Vol. 24, No. 2, 1987, pp. 107-114.
8
Mueller, T. J., “The Influence of Laminar Separation and Transition on Low Reynolds Number Airfoil Hysteresis,” Journal
of Aircraft, Vol. 22, No. 9, 1985, pp. 763-770.
9
Hoffman, J. A., “Effects of Freestream Turbulence on the Performance Characteristics of an Airfoil,” AIAA Journal, Vol. 29,
No. 9, 1991, pp. 1353-1354.
10
O’Meara, M. M., and Mueller, T.J., “Laminar Separation Bubble Characteristics on an Airfoil at Low Reynolds Numbers,”
AIAA Journal, Vol. 25, No. 8, 1987, pp.1033-1041.
11
Brendel, M., and Mueller, T. J., “Boundary Layer Measurements on an Airfoil at a Low Reynolds Number in an Oscillating
Freestream,” AIAA Journal, Vol. 26, No. 3, 1988, pp. 257-263.
12
Mueller, T. J., and Batill, S. M., “Experimental Studies of Separation on a Two-Dimensional Airfoil at Low Reynolds
Numbers,” AIAA Journal, Vol. 20, No. 4, 1982, pp. 457-463.
13
Weibust, E., Bertelrud, A., and Ridder, S. O., “Experimental Investigation of Laminar Separation Bubbles and Comparison
with Theory,” Journal of Aircraft, Vol. 24, No. 5, 1987, pp. 291-297.
14
McCroskey, W. J., “A Critical Assessment of Wind Tunnel Results for the NACA 0012 Airfoil,” NASA Technical
Memorandum 100019, 1989.
15
McAlister, K. W., and Takahashi, R. K., “NACA 0015 Wing Pressure and Trailing Vortex Measurements,” NASA
Technical Paper 3151, 1991.
16
Greenblatt, D., “Management of Vortices Trailing Flapped Wings via Separation Control,” 43rd AIAA Aerospace Sciences
Meeting and Exhibit, 2005-0061, Reno, NV, 2005, pp. 1-22.
17
Gerontakos, P., and Lee, T., “PIV study of flow around unsteady airfoil with dynamic trailing-edge flap deflection,”
Experiments in Fluids, Vol. 45, No. 6, 2008, pp. 955-972.
18
Gerontakos, P., and Lee, T., “Dynamic Stall Flow Control via a Trailing-Edge Flap,” AIAA Journal, Vol. 44, No. 3, 2006,
pp. 469-480.
19
Gerontakos, P., and Lee, T., “Trailing-Edge Flap Control of Dynamic Pitching Moment,” AIAA Journal, Vol. 45, No. 7,
2007, pp. 1688-1694.

12
American Institute of Aeronautics and Astronautics
 20
Huang, R. F., and Lin, C. L., “Vortex Shedding and Shear-Layer Instability of Wing at Low-Reynolds Numbers,” AIAA
Journal, Vol. 33, No. 8, 1995, pp. 1398-1403.
21
Alam, M. M., Zhou, Y., Yang, H. X., Guo, H., and Mi, J., “The ultra-low Reynolds number airfoil wake,” Experiments in
Fluids, Vol. 48, No. 1, 2009, pp. 81-103.
22
Lee, T., and Basu, S., “Measurement of unsteady boundary layer developed on an oscillating airfoil using multiple hot-film
sensors,” Experiments in Fluids, Vol. 25. No. 2, 1998, pp. 108-117.
23
Lee, T., and Gerontakos, P., “Investigation of flow over an oscillating airfoil,” Journal of Fluid Mechanics, Vol. 512, 2004,
pp. 313-341.
24
Sheldahl, R. E., and Klimas, P. C., “Aerodynamic Characteristics of Seven Symmetrical Airfoil Sections Through 180-
Degree Angle of Attack for use in Aerodynamic Analysis of Vertical Axis Wind Turbines,” Sandia National Laboratories,
Albuquerque, NM, SAND80-2114, March 1981.
25
Jacobs, E. N., and Sherman, A. “Airfoil Section Characteristics as Affected by Variations of the Reynolds Number,” NACA
Technical Report No. 586, 1937.
26
Laitone, E. V., “Wind tunnel tests of wings at Reynolds numbers below 70 000,” Experiments in Fluids, Vol. 23, No. 5,
1997, pp.405-409.
27
Timmer, W. A., “Two-dimensional low-Reynolds number wind tunnel results for airfoil NACA 0018,” Wind Engineering,
Vol. 32, 2008, pp. 525-537.
Downloaded by Indian Institute of Science on May 6, 2025 | http://arc.aiaa.org | DOI: 10.2514/6.2010-4629

28
Nakano, T., Fujisawa, N., Oguma, Y., Takagi, Y., and Lee, S., “Experimental study on flow and noise characteristics of
NACA0018 airfoil,” Journal of Wind Engineering, Vol. 95, 2007, pp. 511-531.
29
Yarusevych, S., Sullivan, P. E., and Kawall, J. G., “Coherent structures in an airfoil boundary layer and wake at low
Reynolds numbers,” Physics of Fluids, Vol. 18, No. 4, 044101, 2006, pp. 1-11.
30
Yarusevych, S., Kawall, J. G., and Sullivan, P. E., “Unsteady Separated Flow Characteristics on Airfoils Using Time-
Resolved Surface Pressure Measurements,” AIAA Journal, Vol. 46, No. 2, 2008, pp. 508-516.
31
Yarusevych, S., Sullivan, P. E., and Kawall, J. G., “On vortex shedding from an airfoil in low-Reynolds-number flows,”
Journal of Fluid Mechanics, Vol. 632, 2009, pp. 245-271.
32
Ol, M. V., McAuliffe, B. R., Hanff, E. S., Scholz, U., and Kähler, C., “Comparison of Laminar Separation Bubble
Measurements on a Low Reynolds Number Airfoil in Three Facilities,” 35th AIAA Fluid Dynamics Conference and Exhibit,
AIAA 2005-5149, Toronto, Ontario, 2005, pp.1-11.
33
Tani, I., “Low Speed Flows Involving Bubble Separations,” Progress in Aerospace Sciences, Vol. 5, 1964, pp. 70-103.

13
American Institute of Aeronautics and Astronautics