COMPARISON OF PROPERTIES OF NASA LS(1)-0013

AIRFOIL ESTIMATED USING EXPERIMENTAL DATA,

THIN AIRFOIL THEORY AND JAVAFOIL

Balaji Kartikeyan Chandrasekaran1

Wichita State University,Wichita, Kansas,67260

The properties, Lift co-efficient, drag co-efficient and Moment co-efficient, of

NASA/Langley LS(1)-0013 Airfoil was estimated using the thin-airfoil theory and also using
JavaFoil. The data hence obtained was compared with the experimental data presented in
NASA Technical Report1 and the observations has been discussed and it was found that
experimental set of data is more correct as compared to other sources of data.

Nomenclature
c = chord of Airfoil
Cl = Lift coefficient
Cd = drag coefficient
Cm0.25c = Moment coefficient about quarter chord
Cmle = Moment coefficient about leading edge
α = Angle of Attack
Re = Reynolds Number
m.c.l = Mean Camber Line
NACA 0000 = JavaFoil Name for NASA/Langley LS(1)-0013

I. Introduction
As part of the curriculum of AE-414 course of Aerodynamics-I, the work was carried out to compare the properties
of the LS(1)-0013 airfoil as estimated using various tools that was introduced as part of the course. The airfoil
NASA/LANGLEY LS(1)-0013 is designated in the form NASA LS(1)-xxxx, where LS(1) indicates low speed (1st
series) and the first 2 digits indicate the design lift-coefficient in tenths and the last two digits indicate the thickness
in percent chord.
The study was carried out using the following concepts:
1. Thin-Airfoil Theory : It assumes that if an airfoil is thin then it can approximated into a single line
reperesented by its m.c.l. Then, it is assumed that the flow around the airfoil is uniform, inviscid,steady,
incompressible and irrotational. The airfoil is replaced by a vortex sheet with circulation chosen in such a
way that the m.c.l is locus of all the stagnation points in the flow.

2. JavaFoil: It is a simple java based program developed by Dr Martin Hepperle used for analyzing various
airfoils. It enables us to create airfoils of user’s choice and estimate its various properties under different
flow conditions.

The results that were obtained from the above methods were used to compare them with the experimental data
obtained from the NASA Technical Report1

1
Graduate Student, Aerospace Engineering, 1845 Fairmount St, Wichita, Kansas 67260.
1
 II. PROCEDURE

A. Thin Airfoil Theory

The thin airfoil theory when applied for a symmetric airfoil such as NASA/Langley LS(1)-0013 provides the
values of various co-efficients as follows:
Cl = 2πα
Cmle = - Cl /4
Cm0.25c = 0
The co-efficeint of drag is also zero since the model assumes an inviscid flow but in reality there exists a drag. The
values of these co-efficients were calculated and tabulated in Table 1.
Table 1: Thin Airfoil Theory Data

α (Deg) α (Rad) Cl Cmle Cd Cm0.25c

-8 -0.13968254 -0.878 0.219501 0 0
0 0 0 0 0 0
8 0.13968254 0.878005 -0.2195 0 0

B. EXPERIMENTAL DATA

The experimental data was used from the NASA technical Report1 wherein a low speed wind-tunnel testing was
carried out for the chosen airfoil at Langley Low-Turbulence pressure tunnel.
The report consists of data of this particular airfoil collected for variety of flows with different Re varying from
0.6 x 106 to 12.0 x 106 and also the data has been collected for different surface finishes and the performance has
been compared with that of NACA 0012 airfoil and LS(1)-0013 has been found to be similar to that of NACA 0012
with stall angle of one degree higher than that of NACA 0012.
The experimental data can be considered to be highly accurate since it has been carried out by the original
developer of the airfoil and the report is available in free internet.

C. JAVA-FOIL

The Java-foil was utilized to study the characteristics of the airfoil under study for a chosen test condition from
the experimental data. The condition chosen was having the following characteristics:
a. Re = 4.0 x 106
b. Stall Model = eppler
c. Transition Model = Eppler Standard
d. Surface Finish = Smooth
The co-ordnates of the upper and lower surface has been taken from te theoretical value available in UIUC
Airfoil Coordinates database2. This was chosen over the values provided in the NASA Technical Report1 since those
values were measured on an airfoil that was built in their facility and that was not available for measurement when
this report was written.

2
 III. RESULTS
The plots of the three co-efficients estimated from all the three sources has been plooted below.

Figure 1: Cl - Thin Airfoil Theory

Figure 2: Cd- Thin Airfoil theory

Figure 3: Cm0.25c - Thin Airfoil theory

3
Figure 4: Experimental Data

4
Figure 5: JavaFoil

5
 IV. DISCUSSION
A. Co-Efficient of Lift (Cl)
The lift slope of all the three sources, calculated from Figures 1,4 and 5 are given below:
1. Thin-Air foil theory : 0.109751 per Deg
2. Experimental Data : 0.1125 Per Deg
3. JavaFoil : 0.1195 Per Deg
From the above data it can be safely concluded that all the three methods were able to provide the same slope
(approximately) hence are able to predict the variation of lift co-efficient with angle of attack almost accurately.
On looking further in the region of stall, the variation can be observed in the three methods. The thin-airfoil theory
predicts a linear variation even in stall region. While the javafoil predicts stall at almost correct angle of attack in
comparison to experimental data but the value of Cl is much lower. The main reson for this variation might be due
to the chosen Stall model.

B. Co-efficient of Drag (Cd)

The drag estimate of thin-airfoil theory is zero hence it does not do a great job in predicting the drag co-efficient
(D’Alemberts Paradox).
The drag co-efficient minima occurs at Cl=0 (estimated from figures 4 and 5) for both experimental data and
Javafoil but the values differ slightly due to the model difference in JavaFoil (uses theoretical co-ordinates) and
experimental data (uses measured values of co-ordinates of upper and lower surfaces). The difference can also
be attributed to the modelling of the whole problem in JavaFoil.

C. Moment Co-Efficient about 0.25c (Cm0.25c)

The value of moment co-efficient at 0.25c for a symmetric airfoil is zero (theoretically). The value has been
predicted almost almost perfectly by all the three methods with the thin-airfoil theory predicting the best. The
variation of the value of Moment co-efficient beyond stall point of the airfoil in Java foil is slightly different
from that of experimental data mainly due to the model chosen.

V. CONCLUSION

From the above discussion it can be concluded that the experimental data gives the more correct values of the co-
efficients in comparison to JavaFoil and Thin Airfoil Theory. The main reasons for failure of thin airfoil theory in
predicting the correct values of drag and moment co-efficients, and the values of Cl in stall region can be attributed
to its assumptions. The main weakness of the javaFoil is the selection of Stall Model and Trasition Model which
might have a great impact on calculating the values of co-efficients in stall region.

VI. SUPPORTIVE WORK

One of the weakness of the Java Foil was identified as the stall model chosen for the analysis. The model chosen
as part of the experiment was the Eppler model. In this brief work, an attempt has been made to change the stall
model to calcfoil and also removing stall model to study how it affects the JavaFoil aility to calculate the various co-
efficients. The parameters of the airfoil has been kept constant and only the model has been changed with figure 6
providing the results of Calcfoil model. Even though there is an option of no stall model, it basically gives the same
result as that of Thin-Airfoil theory for the lift co-efficient. The results obtained are presented as figure 6.
The results clearly show that the co-efficeints that as been calculated by the javaFoil is very close to the ones
estimated by the NASA wind tunnel. The lift Co-efficient value has dropped in comparison to the previous chosen
model even though the angle of attack at which the Cl(Max) is achieved remains the same. This clearly proves that
the variations in values calculated in Javafoil is not only is a function of the chosen stall model but it is also the
function of the inbuilt algorithm for solving the model. It can also be observed that value of Cm0.25c is ow more close
to theoretical value and the experimental value.
As per the JavaFoil Developer (Dr Martin Hepperle3) JavaFoil uses potential flow analysis (linearly varying
vorticity distribution) which is somewhat similar to the Thin-Airfoil theory. This is an valid explanation for its
values being very close to that of Thin-Airfoil Theory. It is also stated by the developer that JavaFoil does not model
flow separation but uses empirical corrections which is a possible cause for the failure of JavaFoil to detect exact
angle of Attack at which separation occurs in comparison to the Experimental Data.
6
Figure 6: JavaFoil with Calcfoil Stall Model

7
 References

1James C. Ferries, Robert J. McGhee and Richard W. Barnwell, “Low-Speed Wind-Tunnel Results for Symmetrical NASA

LS(1)-0013 Airfoil”, August 1987

2
http://m-selig.ae.illinois.edu/ads/coord/ls013.dat
3
Dr Martin Hepperle http://www.mh-aerotools.de/airfoils/javafoil.htm