Experiment No.

Low Speed Flow Over an Airfoil

AE611: Aerodynamics Lab Report
17th October 2023

Instructor:
Prof. Vineeth Nair
Department of Aerospace Engineering
Teaching Assistants:
Ashutosh
Samad
Department of Aerospace Engineering

Submitted by
Sanjay R
Roll No: 23M0002
Department of Aerospace Engineering
Contents

1 Introduction 1

2 Objectives 1

3 Theory 1
3.1 Geometric characteristics of an airfoil . . . . . . . . . . . . . . . . . . . . . 2
3.2 Quantifying Aerodynamic forces . . . . . . . . . . . . . . . . . . . . . . . . 3
3.3 Results from thin airfoil theory . . . . . . . . . . . . . . . . . . . . . . . . 4
3.4 Estimates from XFLR5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

4 Experimental setup 4

5 Procedure 5

6 Observations and Results 6

6.1 Cp distribution around the airfoil . . . . . . . . . . . . . . . . . . . . . . . 6
6.2 Velocity profile and estimation of drag . . . . . . . . . . . . . . . . . . . . 11
6.3 Performance curves of the airfoil . . . . . . . . . . . . . . . . . . . . . . . . 12

7 Comparison with literature estimates 20

8 Sources of errors 22

9 Conclusion 22

10 References 24
 Low Speed Flow Over an Airfoil

1 | Introduction
Airfoils are cross-sectional shapes of lift-producing devices such as wings in aircraft.
Understanding the behavior of fluids around airfoils is of great importance in fields such
as aerodynamics and fluid mechanics. It allows us to optimize the design of aircraft and
control the lift and drag forces, thereby enhancing their efficiency and performance. In this
experiment, a symmetric airfoil NACA 0012 was kept in the test section of a closed wind
tunnel, and the surface pressure and the wake velocity profile were studied to find the lift
and drag forces of the airfoil. The measurements obtained in this experiment have been
compared to the theoretical predictions. The shape of an airfoil is shown in Figure 1.1

Figure 1.1: Major Airfoil profile [4]

2 | Objectives
■ To study the upper and lower surface pressure distribution of the airfoil (cpu and cpl )
at various angles of attack from the freestream.

■ To measure the sectional lift coefficient (cl ) and sectional moment coefficient at the
leading edge (cmLE ) and at quarter chord point (cmc/4 ) at these respective angles of
attack.

■ To find the pressure drag (cdp ) at these corresponding angles of attack.

■ To find the total drag coefficient of the airfoil (cdt ) in the flow with the wake velocity
profile and the skin friction drag coefficient (cdf ) of the airfoil by subtracting the
pressure drag coefficient.

■ To locate the center of pressure and aerodynamic center with the obtained measure-
ments and compare it with their locations predicted by thin airfoil theory.

■ To compare the results obtained with the theoretical predictions.

3 | Theory

Page 1
 Low Speed Flow Over an Airfoil

3.1 | Geometric characteristics of an airfoil

Figure 3.1: Airfoil nomenclature [2]

Figure 3.1 represents the major properties of the geometry of the airfoil. The average of
the upper and lower surfaces of an airfoil represents the camber line of the airfoil and the
distance between the camber line and any of the surfaces represents the thickness of the
airfoil.
There are many configurations of the geometric properties leading to various airfoil shapes.
One such typical configuration comes from the NACA 4-digit series of airfoils which is
defined in the x − y plane with Equation 3.3 and Equation 3.4.
m
yc = p2
(2px − x2 ) , x ∈ [0, p]
m 2 (3.1)
yc = (1−p)2
(1 − 2p + 2px − x ) , x ∈ [p, 1]

 √ 
yt = 5t 0.2969 x − 0.1260x − 0.3516x2 + 0.2843x3 − 0.1015x4 (3.2)

yu = yc + yt (3.3)

yl = yc − yt (3.4)

The NACA 4-digit series (eg. NACA XXXX) airfoil is configured as

■ 1st digit: Maximum camber in the percentage of chord = m

■ 2nd digit: Location of maximum camber point in ten times the percentage of chord
=p

■ Last 2 digits: Maximum thickness in the percentage of chord = t

Page 2
 Low Speed Flow Over an Airfoil

3.2 | Quantifying Aerodynamic forces

The aerodynamic forces are generated during the fluid flow on the surface, contributed by
the pressure and shear stresses on the surface. These forces act on the center of pressure
of the airfoil.

Figure 3.2: Aerodynamic forces experienced by an airfoil[3]

Here F ′ is the resultant aerodynamic force, N ′ is the Normal component, A′ is the axial
component, L′ is the lift force generated by the airfoil and D′ is the drag force of the
airfoil. The non-dimensional parameters are calculated with the formula given below.
" ! #
1 Zc Z c
dyu dyl
Cn = (Cpl − Cpu ) dx + C fu − C fl dx (3.5)
c 0 0 dx dx
" ! #
1 Zc dyu dyl Z c
Ca = Cp u − Cp l dx + (Cfu + Cfl ) dx (3.6)
c 0 dx dx 0

 R R   
c
1 (Cpu − Cpl ) xdx − 0c Cfu dy u
− Cfl dyl
xdx
CmLE = 2  0R c  dyu

c
 dx
dyl
dx   (3.7)
c
R
+ 0 Cpu dx + Cfu yu dx 0 Cpl dx + Cfl yl dx

Since the contribution from viscous forces is very small compared to the pressure forces,
the Cfu and Cfl products can be neglected. With that, the above equations become
1
Z c 
Cn = (Cpl − Cpu ) dx (3.8)
c 0

" ! #
1 Zc dyu dyl
Ca = Cp u − Cp l dx (3.9)
c 0 dx dx
" ! #
1 Zc Z c
dyu dyl
CmLE = 2 (Cpu − Cpl ) xdx + y u Cp u − yl Cpl dx (3.10)
c 0 0 dx dx
With the normal and axial force coefficients, the lift and pressure drag coefficients are
calculated as
Cl = Cn cos α − Ca sin α (3.11)

Page 3
 Low Speed Flow Over an Airfoil

Cdp = Cn sin α + Ca cos α (3.12)

The location of the center of pressure at a small angle of attack is assumed to act at
′
MLE
xcp ≈ − (3.13)
L′

3.2.1 | Wake measurement for total drag estimation

Due to the momentum conservation, the drag force can be estimated by the momentum
loss created by the airfoil. The coefficient of drag can be estimated from the wake velocity
profile using the formula
" ! #
2 Z b u(y) u(y)
Cdt = 1− dy (3.14)
S a V∞ V∞
where S is the surface area of the airfoil and a and b are the bottom and top of the
velocity profile locations respectively. The skin friction coefficient can be estimated from
the difference between the total and pressure drag coefficients

Cdf = Cdt − Cdp (3.15)

3.3 | Results from thin airfoil theory

From thin airfoil theory, the predictions of the sectional lift and moment coefficients can
be estimated with the following equations.

cl = 2πα (3.16)

cl
cmLE = − (3.17)
4

cmc/4 = cmcp = 0 (3.18)

Also, the location of the center of pressure is at 0.25 times the chord.

3.4 | Estimates from XFLR5

An open-source software XFLR5 was used to get the estimates of the aerodynamic forces
using the panel method for the sectional coefficients. It gives the pressure distribution
around the airfoil and the drag, lift, and moment coefficient measurements. These are
used for comparison with the experimental measurements.

4 | Experimental setup
An open loop, suction-type wind tunnel is used to generate the airflow, and the NACA
0012 airfoil is set up in the test section of the wind tunnel. This means that the airfoil
has no camber and a thickness of 12%. The dynamic pressure of the flow and the gage
pressure measured using pitot probes in the test section are matched to have a velocity of
V∞ = 17m/s in the stream-wise axis. The chord of the airfoil (c) is 0.15m.

Page 4
 Low Speed Flow Over an Airfoil

Figure 4.1: Depiction of the shape of NACA0012 airfoil

The apparatus required is as follows

■ Wind tunnel
■ NACA 0012 airfoil
■ Rotating setup to change the angle of attack
■ Pitot Probe placed in the airfoil for surface measurements
■ 2 channel selector boxes
■ Manometer
■ Mechanical Traverse Assembly
The x-axis is set to be along the streamwise direction of the flow, the y-axis is the vertical
axis and the z-axis is the span axis for the calculations in the experiment. The flow is
assumed to be uniform and the airfoil is assumed to have negligible roughness.

5 | Procedure
■ The ambient pressure and temperature of the lab during the experiment is noted
and the density is calculated using the equation of state. With that, the dynamic
pressure for U∞ = 17m/s is calculated and matched with the manometer reading in
the test section to start the measurements.
■ After the flow is settled with the airfoil setup, the angle of attack of the airfoil is
fixed and the surface pressure from each port in the airfoil is measured. All of the
ports are connected to the channel selector box and the output of the box gives the
particular pressure at the port using the manometer.

Page 5
 Low Speed Flow Over an Airfoil

■ The maximum and minimum of the fluctuations are recorded and averaged to get
the optimal pressure at a point.

■ This process is repeated for angle of attacks from -8 to 8 with a step increase of 2.
The cp recorded from each location is then used to calculate cl , cdp and cm using the
Equation 3.11, Equation 3.12 and Equation 3.10 respectively.

■ The total drag of the airfoil is measured by the wake velocity profile. The pitot-static
probe is traversed along the y-axis behind the airfoil with a small step increase to
note the pressure readings until it reaches the dynamic pressure of the flow. The
velocity is calculated with Equation (5.1).
s
2(P0 − P )
v= (5.1)
ρ

■ With the total drag and the pressure drag coefficient obtained, the skin-friction drag
coefficient is calculated using Equation 3.15.

■ The obtained measurements are compared with the theoretical estimates from the
thin airfoil theory and the results from XFLR5.

6 | Observations and Results

The ambient conditions during the experiment are mentioned in Table ??. With this, the
Re of the airfoil has been calculated to be 162315.

Properties Values
Ambient Pressure (Pa) 100240
Ambient Temperature (K) 300.65
3
Fluid Density (kg/m ) 1.1615
Dynamic Pressure (Pa ) 167.843
Dynamic Viscosity (N s/m−2 ) 1.849 × 10−5

Table 6.1: Ambient conditions

6.1 | Cp distribution around the airfoil

The pressure distribution around the airfoil is non-dimensionally represented as Cp and
used for estimating the aerodynamic performance parameters and center of pressure of
the airfoil. Figure 6.1 to Figure 6.9 shows the results of the pressure distribution around
the airfoil of the experiment compared with the results from XFLR5.

Page 6
 Low Speed Flow Over an Airfoil

Figure 6.1: Cp vs x/c for α = −8◦

Figure 6.2: Cp vs x/c for α = −6◦

Page 7
 Low Speed Flow Over an Airfoil

Figure 6.3: Cp vs x/c for α = −4◦

Figure 6.4: Cp vs x/c for α = −2◦

Page 8
 Low Speed Flow Over an Airfoil

Figure 6.5: Cp vs x/c for α = 0◦

Figure 6.6: Cp vs x/c for α = 2◦

Page 9
 Low Speed Flow Over an Airfoil

Figure 6.7: Cp vs x/c for α = 4◦

Figure 6.8: Cp vs x/c for α = 6◦

Page 10
 Low Speed Flow Over an Airfoil

Figure 6.9: Cp vs x/c for α = 8◦

It can be seen that the trend with the experimental results correlates well with the panel
method estimates from XFLR5. The minimum pressure point at α = −8◦ and 8◦ shows
low values due to errors in the corresponding probe setup. Tbe area enclosed by the upper
and lower surfaces gives the lift coefficient of the airfoil from Equation 3.11.

6.2 | Velocity profile and estimation of drag

Figure 6.10 shows the velocity profile in the wake of the airfoil at α = −2◦ , 0◦ , 2◦ .

Page 11
 Low Speed Flow Over an Airfoil

Figure 6.10: Wake velocity profile measurements for airfoil at α = −2◦ , 0◦ , 2◦

It can be seen that the velocity profiles at α = −2◦ and 2◦ has higher peak than at α = 0◦
due to higher resistance to the momentum of the flow. Therefore, the drag coefficient
would be higher.

6.3 | Performance curves of the airfoil

The cl , cd , and cm are highly important to the performance of the airfoil at their corre-
sponding α. Figure 6.11 shows the variation of cl obtained in the experiment with the
theoretical predictions with the thin airfoil theory from Equation 3.11 and the results from
XFLR5.

Page 12
 Low Speed Flow Over an Airfoil

Figure 6.11: cl vs α compared with thin airfoil theory and XFLR5 results

The thin airfoil theory results in a straight line as it has a linear relationship from Equa-
tion 3.16. The XFLR5 results vary to it as the numerical solution takes viscosity and
accounts for the formation of Laminar Separation Bubble. The experimental results are
closer to these results and the linear fit of the experimental data aligns with the theoretical
solutions as shown in Figure 6.11.

The total drag measured from the wake of the airfoil is shown in Figure 6.12. This is also
compared with the numerical results of XFLR5. The drag from the thin airfoil theory
cannot be predicted as it is 0 for all cases.

Page 13
 Low Speed Flow Over an Airfoil

Figure 6.12: cdt vs α compared with the XFLR5 results

The drag coefficent calculated from the measurements is higher than the results from
XFLR5. This is associated with the errors with the experimental setup.

The results from XFLR5 provide the data for both total and pressure drag coefficients
and therefore the pressure drag can also be compared with the experimental data.

Page 14
 Low Speed Flow Over an Airfoil

Figure 6.13: cdp vs α compared with the XFLR5 results

The values of the pressure drag should follow a parabolic profile. The measurements from
the experiment show a higher pressure drag due to the errors of the experimental setup.

The skin friction coefficient is calculated from the Equation 3.15 for both the experimental
and the XFLR5 data and is plotted as shown in Figure 6.14

Page 15
 Low Speed Flow Over an Airfoil

Figure 6.14: cdf vs α compared with thin airfoil theory and XFLR5 results

The skin-friction coefficient calculated from the experiment also is slightly higher than the
numerical results.

The moment coefficient from the thin airfoil theory at the quarter chord is zero from
Equation 3.18, as the aerodynamic center, center of pressure lies at the quarter chord
point. But the solutions from XFLR5 and the experimental data contradicts it as it varies
non-linearly as shown in Figure 6.15.

Page 16
 Low Speed Flow Over an Airfoil

Figure 6.15: cmc/4 vs α compared with thin airfoil theory and XFLR5 results

The moment coefficient at the Leading edge varies linearly as shown in Figure 6.16. A
good correlation is obtained between the thin-airfoil theory, XFLR5 and the experimental
results with the cmLE .

Page 17
 Low Speed Flow Over an Airfoil

Figure 6.16: cmLE vs α compared with thin airfoil theory and XFLR5 results

The center of pressure location must ideally be at the quarter chord according to thin
airfoil theory. The result from XFLR5 and the experimental measurements varies closely
to the quarter chord line as shown in Figure 6.17. Also, the linear fit of the experimental
data aligns slightly higher to the quarter chord line. Note that the XFLR5 prediction of
center of pressure is irrelevant at α = 0◦ since there is no lift force on the airfoil.

Page 18
 Low Speed Flow Over an Airfoil

Figure 6.17: xCP vs α compared with thin airfoil theory and XFLR5 results

The data obtained from the experimental measurements are shown in Table 6.2 and
Table 6.3.

α cl cmc/4 cmLE xCP

-8 -0.80704 0.07185 0.27361 0.33903
-6 -0.76283 0.00954 0.20025 0.26251
-4 -0.47537 0.01104 0.12988 0.27322
-2 -0.31663 0.00611 0.08527 0.26930
0 -0.05957 0.00137 0.01627 0.27303
2 0.38671 -0.00909 -0.10576 0.27350
4 0.43241 -0.00807 -0.11617 0.26867
6 0.62966 -0.01269 -0.17010 0.27015
8 0.73881 -0.05600 -0.24070 0.32579

Table 6.2: Tabulation of the experimental data from surface pressure measurements

Page 19
 Low Speed Flow Over an Airfoil

α cdtotal cdpressure cdf riction

-2 0.10604 0.08475 0.02129
0 0.07838 0.07255 0.00583
2 0.13905 0.07119 0.06786

Table 6.3: Tabulation of the experimental data using wake velocity profile

6.3.1 | Estimation of aerodynamic center

The aerodynamic center is a point where the pitching moment remains constant with the
angle of attack. Thin airfoil theory predicts that the aerodynamic center is at the quarter
chord of the airfoil, which may not be true for the experimental results. The aerodynamic
center can be determined with the following equations.
dcl
= a0 (6.1)
dα

dcmc/4
= m0 (6.2)
dα

xAC −m0
= + 0.25 (6.3)
c a0
The values of the aerodynamic center predicted by the experimental and the theoretical
results are

xAC experiment = 0.2514 (6.4)

xAC thin airf oil theory = 0.25 (6.5)

7 | Comparison with literature estimates

The results of the experiment are compared with the classic literature of Abbott and
Von-Doenhoff, Theory of Wing sections [1], with a Re = 3 × 106 . Figure 7.1 shows the
comparison of the experimental results of cl VS α from the literature, where a good
correlation between the data can be seen.

Page 20
 Low Speed Flow Over an Airfoil

Figure 7.1: Comparison of cl vs α plots between the experimental results and Abbott
and Von-DoenHoff [1], Theory of Wing Sections

The results from Abbott and Von-DoenHoff is more linear than the present data, since
the Re of the experiment is higher. A linear fit of the present data aligns with the data of
Abbott and Von-DoenHoff.

Page 21
 Low Speed Flow Over an Airfoil

Figure 7.2: Comparison of cmc/4 vs α plots between the experimental results and Abbott
and Von-DoenHoff [1], Theory of Wing Sections

The variations seen in Figure 7.2 for Abbott and Von-DoenHoff data is much smaller than
the present data, as it may be more closer to the aerodynamic center than the present
data and the Re being different.

8 | Sources of errors
■ The traversing mechanism was found to be loose and had errors, and the pitot probe
was not completely aligned with the flow due to bending.
■ There was a slight leakage of the flow due to the inserted instruments, which may
affect the total drag measured in the wake of the airfoil.
■ The airfoil may have slight roughness which may increase the drag coefficient mea-
sured.
■ There may be human errors in recording the measurements.

9 | Conclusion
1. The low-speed flow over the airfoil resulted in a Re of 162315, and with the surface
pressure and wake velocity profile measurements, the performance results of the

Page 22
 Low Speed Flow Over an Airfoil

airfoil cl , cdt , cdp , cdf , cmLE , cmc/4 is obtained with the locations of center of pressure
and aerodynamic center.

2. The coefficient of pressure recorded around the airfoil resulted in a similar trend to
the panel method estimates from XFLR5.

3. The velocity profile measured at the wake shows large deficits at higher angles of
attack as the flow resistance increases.

4. The results of cl match well with the XFLR5 results and the thin airfoil solutions.
Still, there are slight variations associated with the discrepancies in the experimental
setup.

5. The estimated location of the aerodynamic center from the experimental measure-
ments is closer to the thin airfoil theory prediction.

6. The literature data of Abbott and Von-DoenHoff [1] correlate well with the current
experimental measurements.

Page 23
 Low Speed Flow Over an Airfoil

10 | References
[1] Ira H Abbott and Albert E Von Doenhoff. Theory of wing sections: including a
summary of airfoil data. Courier Corporation, 2012.

[2] Geometric CAE. Airfoil terminology, 2018. https://geometriccae.medium.com/

airfoil-terminology-423c79d5ed8.

[3] Prof. Aniruddha Sinha. Introduction to aerodynamics. AE-707: Aerodynamics of

Aerospace Vehicles, Department of Aerospace Engineering, IIT Bombay, 2023.

[4] Cadence Design Systems. Cambered airfoil explained: The effects of

shape on lift, 2022. https://resources.system-analysis.cadence.com/blog/
msa2022-cambered-airfoil-explained-the-effects-of-shape-on-lift.

Page 24