Calhoun: The NPS Institutional Archive

Theses and Dissertations Thesis and Dissertation Collection

2008-12

Aerodynamic predictions, comparisons, and

validations using Missilelab and Missile Datcom (97)

Teo, Hoon Hong.

Monterey, California. Naval Postgraduate School

http://hdl.handle.net/10945/3708
 NAVAL
POSTGRADUATE
SCHOOL
MONTEREY, CALIFORNIA

THESIS

AERODYNAMIC PREDICTIONS, COMPARISONS, AND

VALIDATIONS USING MISSILELAB AND MISSILE
DATCOM (97)

by

Teo, Hoon Hong

December 2008

Thesis Advisor: M. S. Chandrasekhara

Approved for public release; distribution is unlimited

THIS PAGE INTENTIONALLY LEFT BLANK
 REPORT DOCUMENTATION PAGE Form Approved OMB No. 0704-0188
Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instruction,
searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send
comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden, to
Washington headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA
22202-4302, and to the Office of Management and Budget, Paperwork Reduction Project (0704-0188) Washington DC 20503.
1. AGENCY USE ONLY (Leave blank) 2. REPORT DATE 3. REPORT TYPE AND DATES COVERED
December 2008 Masters’ Thesis
4. TITLE AND SUBTITLE Aerodynamic Predictions, Comparisons, and 5. FUNDING NUMBERS
Validations Using MissileLab and Missile Datcom (97)
6. AUTHOR(S) Teo, Hoon Hong

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION

Naval Postgraduate School REPORT NUMBER
Monterey, CA 93943-5000
9. SPONSORING /MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORING/MONITORING
N/A AGENCY REPORT NUMBER

11. SUPPLEMENTARY NOTES The views expressed in this thesis are those of the author and do not reflect the official policy
or position of the Department of Defense or the U.S. Government.
12a. DISTRIBUTION / AVAILABILITY STATEMENT 12b. DISTRIBUTION CODE
Approved for public release; distribution is unlimited
13. ABSTRACT (maximum 200 words)
Aerodynamic prediction software is often used in the early stages of missile systems designed to quickly and
accurately estimate the aerodynamics of a wide variety of missile configuration designs operating over many
different flight regimes. It is also possible to use these empirical packages to validate flight data collected from
wind tunnel tests and other open sources. Analysis of such data provides users with insights to the performance of
a particular missile system and if necessary, enables the development of an appropriate defense system.

Wind tunnel test data on an SA-2 class missile modified by suitable modeling was provided by MSIC. For this
Thesis, this data set became the bench-mark for validating the Missile Datcom (97) empirical code that was used to
compute the performance of the missile. The missile geometry was modeled using the interface MissileLab. A
series of simulations for different flight operating conditions was carried out. The primary quantities compared
were the axial force coefficient, CA and the skin friction coefficient, Cf. This Thesis describes the results obtained
along with the geometry changes that became necessary to obtain reasonable agreement.

14. SUBJECT TERMS 15. NUMBER OF

PAGES
Missile Aerodynamics, Missile Datcom (97), MissileLab 51
16. PRICE CODE
17. SECURITY 18. SECURITY 19. SECURITY 20. LIMITATION OF
CLASSIFICATION OF CLASSIFICATION OF THIS CLASSIFICATION OF ABSTRACT
REPORT PAGE ABSTRACT
Unclassified Unclassified Unclassified UU
NSN 7540-01-280-5500 Standard Form 298 (Rev. 2-89)
Prescribed by ANSI Std. 239-18

i
THIS PAGE INTENTIONALLY LEFT BLANK

ii
 Approved for public release; distribution is unlimited

AERODYNAMIC PREDICTIONS, COMPARISONS, AND VALIDATIONS

USING MISSILELAB AND MISSILE DATCOM (97)

TEO, Hoon Hong

Major, Republic of Singapore Air Force
B.Eng., University of Manchester, Institute of Science and Technology, 2000

Submitted in partial fulfillment of the

requirements for the degree of

MASTER OF SCIENCE IN MECHANICAL ENGINEERING

from the

NAVAL POSTGRADUATE SCHOOL

December 2008

Author: TEO, Hoon Hong

Approved by: Prof. M. S. Chandrasekhara

Prof. Knox T. Millsaps

Chairman, Department of Mechanical and Astronautical
Engineering

iii
THIS PAGE INTENTIONALLY LEFT BLANK

iv
 ABSTRACT

Aerodynamic prediction software is often used in the early stages of missile

systems designed to quickly and accurately estimate the aerodynamics of a wide variety
of missile configuration designs operating over many different flight regimes. It is also
possible to use these empirical packages to validate flight data collected from wind tunnel
tests and other open sources. Analysis of such data provides users with insights to the
performance of a particular missile system and if necessary, enables the development of
an appropriate defense system.

Wind tunnel test data on an SA-2 class missile modified by suitable modeling was
provided by MSIC. For this Thesis, this data set became the bench-mark for validating
the Missile Datcom (97) empirical code that was used to compute the performance of the
missile. The missile geometry was modeled using the interface MissileLab. A series of
simulations for different flight operating conditions was carried out. The primary
quantities compared were the axial force coefficient, CA and the skin friction coefficient,
Cf. This Thesis describes the results obtained along with the geometry changes that
became necessary to obtain reasonable agreement.

v
THIS PAGE INTENTIONALLY LEFT BLANK

vi
 TABLE OF CONTENTS

I. INTRODUCTION........................................................................................................1
A. DESCRIPTION OF WORK ...........................................................................1
B. MISSILE PREDICTION EMPIRICAL SOFTWARE CODE ...................2
C. OUTPUT ANALYSIS......................................................................................3
D. MISSILE PERFORMANCE ..........................................................................3
E. IMPORTANCE OF CA AND CF ANALYSIS ...............................................4
1. CA Analysis ...........................................................................................4
2. Cf Analysis ............................................................................................6
3. Other Force Components ....................................................................7
II. METHOD OF APPROACH .......................................................................................9
A. DATA GIVEN ..................................................................................................9
1. MSIC Data Source .............................................................................10
a. Wind Tunnel Testing ..............................................................10
b. Simulink ..................................................................................12
B. RESEARCH METHODOLOGY .................................................................13
1. Geometry Coding ...............................................................................13
a. Nose Geometry ........................................................................13
b. Fin Geometry...........................................................................14
c. Body Geometry ........................................................................14
d. Surface Roughness .................................................................14
2. Operating Conditions ........................................................................15
a. Boundary Layer Conditions ...................................................15
b. Fins Deflections ......................................................................15
c. Angle of Attack and Altitude ..................................................16
d. Base Drag ................................................................................16
III. RESULTS ...................................................................................................................17
A. SIMULATION CONDITIONS.....................................................................17
B. COMPONENTS INVOLVED IN AXIAL FORCE COEFFICIENT .......19
C. COEFFICIENT OF SKIN FRICTION .......................................................20
1. Effects of Power on Coefficient of Skin Friction.............................20
2. Effects of Surface Roughness on Coefficient of Skin Friction .......21
3. Effects of Mach Number on Coefficient of Skin Friction...............23
4. Effects of α on Cf ................................................................................23
D. AXIAL FORCE COEFFICIENT.................................................................23
1. Effects of Power on Axial Force Coefficient....................................23
2. Effects of Surface Roughness on Axial Force Coefficient ..............24
3. Effects of Mach Number on Axial Force Coefficient......................25
4. Effects of Angle of Attack on Axial Force Coefficient....................25
5. Effects of Altitude on Axial Force Coefficient.................................26
E. DATA COMPARISON WITH MSIC DATA .............................................27
1. Skin Friction Coefficient Comparison with MSIC Data ................27
vii
 2. Axial Force Coefficient Comparison with MSIC Data...................28
IV. CONCLUDING REMARKS ....................................................................................31
A. LIMITATIONS ..............................................................................................31
B. FUTURE WORK ...........................................................................................32
LIST OF REFERENCES ......................................................................................................33
INITIAL DISTRIBUTION LIST .........................................................................................35

viii
 LIST OF FIGURES

Figure 1 Kill Envelope of Various Generations of the S-75 System [From 6] ................3
Figure 2 Component Build-up Model [From 1]...............................................................5
Figure 3 Variation due to Mach Number of Drag Components [From 1] .......................6
Figure 4 Cf vs. Reynolds Number [From 1].....................................................................7
Figure 5 Missile Dimension Provided by MSIC ............................................................10
Figure 6 Cf vs. Reynolds Number [From 2]...................................................................12
Figure 7 Coded 3-D Sketch of Missile with Booster Attached from MissileLab ..........13
Figure 8 Coded 3-D Sketch of Missile After Booster Separation..................................14
Figure 9 CA Components vs. Mach Number (Power Off, Turbulent, Roughness
0.001016, α = 0 deg, Alt. = 0 m)......................................................................19
Figure 10 CA Components vs. Mach Number (Power On, Turbulent, Roughness
0.001016, α = 0 deg, Alt. = 0 m).....................................................................20
Figure 11 Cf vs. Mach Number (Power Off, Turbulent, Roughness 0.001016, 0 deg
α) ......................................................................................................................21
Figure 12 Cf vs. Mach Number (Power On, Turbulent, Roughness 0.001016, 0 deg
α) ......................................................................................................................21
Figure 13 Cf vs. Altitude (Power Off, Turbulent, Roughness 0.001016, 0 deg α)...........22
Figure 14 Cf vs. Altitude (Power Off, Turbulent, Roughness 0, 0 deg α)........................22
Figure 15 Comparison of Altitude and α Using Cf vs. Mach Number (Power Off,
Turbulent, Roughness 0.001016).....................................................................23
Figure 16 Comparison of Altitude and α Using CA vs. Mach Number (Power Off,
Turbulent, Roughness 0.001016).....................................................................24
Figure 17 Comparison of Altitude and α Using CA vs. Mach Number (Power On,
Turbulent, Roughness 0.001016).....................................................................24
Figure 18 Comparison of Altitude and α Using CA vs. Mach (Power Off, Turbulent,
Roughness 0)....................................................................................................25
Figure 19 Comparison of α Using CA vs. Mach Number (Power Off, Turbulent,
Roughness 0, 0m).............................................................................................26
Figure 20 Comparison of Altitude Using CA vs. Mach Number (Power Off,
Turbulent, Roughness 0)..................................................................................26
Figure 21 Cf vs. Mach Number (Power Off, Turbulent, Roughness 0.001016, 0 deg
α) ......................................................................................................................27
Figure 22 CA vs. Mach Number (Power On, Turbulent, Roughness 0.001016, 0m, 0
deg α) ...............................................................................................................28
Figure 23 CA vs. Mach Number (Power Off, Turbulent, Roughness 0.001016, 0m, 0
deg α) ...............................................................................................................29

ix
THIS PAGE INTENTIONALLY LEFT BLANK

x
 LIST OF TABLES

Table 1. Flow Conditions for Cf Analysis......................................................................18

Table 2. Flow Conditions for CA Analysis ....................................................................18

xi
THIS PAGE INTENTIONALLY LEFT BLANK

xii
 ACKNOWLEDGMENTS

I would like to thank Professor M.S. Chandrasekhara for his continuous guidance.
He has been a source of inspiration. Also, I would like to give special thanks to LCDR.
Robert DeWitt, MSIC, for sponsoring and supporting this project, providing the wind
tunnel and simulated performance data, and for following up steadily during the course of
this work.

I would also like to give special thanks to Dr. William Blake of USAF and Mr.
Lamar Auman, U.S. Army AMRDEC for permitting the use of Missile DATCOM (97)
code and the U.S. Army MissileLab code. This made my task significantly easier.

I would like to express my gratitude to my wife, Clarice Chew and daughter

Carlin Teo, who have supported me throughout my time in NPS. Their unwavering
support and encouragement have made my Thesis process enjoyable.

xiii
 NOMENCLATURE

CA = Axial force coefficient

CAb = CA due to base pressure

CAf = CA due to skin friction

CALE = Leading edge CA

CA0 = CA at zero normal force

CAp = CA due to pressure

CATE = Trailing edge CA

CAw = CA due to wave drag

CAwn = CA due to nose

CAwa = CAw due to after body

CD = Coefficient of drag

CD0 = CD at zero lift

Cf = Coefficient of skin friction

CL = Coefficient of lift

CN = Coefficient of normal force

Cp = Coefficient of pressure

Re = Reynolds number

α = Angle Alpha; Angle of Attack

xiv
 I. INTRODUCTION

A. DESCRIPTION OF WORK

Aerodynamic predictions, using various prediction tools, have been practiced for
many years. Comparisons and validations are usually made between computations,
simulations, and wind tunnel test data in an attempt to predict actual flight performance.
In this Thesis, a combination of experimental wind tunnel test data and simulations data
became available for use to compare with that predicted by codes. This was an attempt to
investigate the model geometry and flight conditions that will provide the aerodynamic
properties matching the given data. When fully implemented and validated, the technique
becomes another tool in the hands of an analyst to assess the performance of missile.

The missile system of particular interest to the sponsor was the SA-2 like
configuration. Its flight characteristics were experimentally determined and subsequently
also modeled. This combined data set for the missile system was provided by Missile and
Space Intelligence Center (MSIC). Such data, however, tends to be incomplete due to
limitations of testing some of which include altitude simulation, scale effects and free
stream turbulence effects. Thus, it is imperative that the performance prediction code
used be capable of allowing these factors to be included in the modeling to generate a
resultant data base that can be used for a comprehensive range of flight conditions
Therefore, to validate and analyze the missile system, a methodology was developed to
build up the missile model accurately using the US Army AMRDEC package, described
as MissileLab [5]. It automatically generates the input files for the USAF missile
performance evaluation code Missile Datcom (97) [4]. The design was put through
several simulations. This involved studies at many different flow conditions, such as:
• Mach numbers
• Altitudes
• Angles of Attack
• Missile surface geometry conditions
• Boundary layer conditions
1
 Results from such an extensive research analysis were analyzed and refined
through changes in missile configurations or test conditions. These results were then
compared to the data provided and comments and recommendations were made.

B. MISSILE PREDICTION EMPIRICAL SOFTWARE CODE

Although many prediction codes are available in the market, it is the sponsor's
requirement to produce as many aerodynamic performance coefficients as possible that
led to the use of Missile Datcom (97) [3]. Studies have been conducted to validate the
results produced by Missile Datcom (97) and empirical data [3]. Under certain condition,
results have shown [3] that prediction of axial force by Missile Datcom (97) falls within
11.69% accuracy. A similar and comparable performance prediction code, namely AP05,
is available but it predicts fewer performance coefficients. It is more suitable for 2D
modeling; whereas Missile Datcom (97) can be used for 3D aerodynamic prediction.

Missile Datcom (97) has many key features [3] which make it an attractive
aerodynamic prediction tool. It possesses a trimmed flight function and is able to use
experimental data to model a particular configuration. Missile Datcom (97) has the
capability of modeling standard airfoil design and user defined airfoils. It also allows air
breathing propulsion systems to be analyzed. However, it lacks plotting functions as well
as geometry sketch functions, requiring the user to have a strong understanding on
FORTRAN programming and the need for a post-processing tool. Although the output
files for TECPLOT can be generated, the unavailability of this graphics package at NPS
implied use of other methods. In this Thesis, post processing is done using Excel to
present the results.

Interface with Missile Datcom (97) is through a Windows-based software known

as MissileLab. MissileLab allows users with minimum knowledge of FORTRAN
programming to utilize the aerodynamic prediction capability of Missile Datcom (97).
One of the main features of MissileLab is the capability to produce 3D sketches of any
parts of a missile configuration as it is being built, thereby eliminating user input errors.

2
C. OUTPUT ANALYSIS

From the data provided by MSIC and the time frame of the research available,
two parameters of significant importance were analyzed in this Thesis. These were: axial
force coefficient CA for the body and fins and, also, CA due to skin friction for the body
only. Both these parameters provide insights into the performance of the missile as
discussed below.

D. MISSILE PERFORMANCE

The SA-2 class missile in this Thesis operates in a wide range of conditions. The
missile has a 2-stage propulsion system, with the booster detaching itself 3 to 5 seconds
after launch. The main engine then provides the thrust for another 25 seconds before
burning out [6]. By then, the missile would have reached Mach 3.

With its first employment in 1957 and its first successful publicized engagement
of the U2 Reconnaissance aircraft in 1960 piloted by Francis Gary Powers [6], it has
since been widely used in many air defense systems. The operating altitude of the U2 is
27,000m [6]. The missile is therefore required to operate over a wide regime of altitudes,
accelerating through subsonic to supersonic and high Angle of Attacks (α) due to its
maneuvering. The kill envelope is shown in Figure 7.

Figure 1 Kill Envelope of Various Generations of the S-75 System [From 6]

3
E. IMPORTANCE OF CA AND CF ANALYSIS

The standard coefficients such as various forces (axial, normal and side) and of
the various moments are the quantities of interest. However, the wind tunnel test data
placed notable emphasis on the axial force skin friction coefficient, with and without
engine power. Thus, the comparisons to be made will focus on these two quantities. It is
noted here that other quantities were also computed and would be discussed as well. The
axial force is a critical quantity in the design of a missile and primarily determines the
range and the maneuverability of the missile. In addition, the drag of a missile is a strong
function of the flight Mach number. Different flow related mechanisms come into play at
different speed regimes, as has been described in detail in [1]. As such, the aerodynamic
prediction code must be capable of computing it by switching to the appropriate method
of the computation based on the speed regime without user interference. Likewise, the
skin friction also has to be computed based on the flow conditions at the wall.

Even though all these details are considered, it will still not be possible to
properly quantify these for the actual vehicle owing to the scale effects even in modeling.
This is particularly true for small protuberances and appendage that a full-scale missile
carries and the inclusion of their effects on the potential flow pressure distribution at the
various speed ranges and on each other.

1. CA Analysis

It is well known that more than 50% of drag is body induced and thus, drag will
be the primary focus in this Thesis through CA analysis in the predictions of the
performance of the given missile system. The study of CA will allow predictions in the
range of the missile, its speed, size of propulsion system, and its carriage effect on
airborne systems. The overall CA of a missile body is further broken down into CAf (CA
due to skin friction) and CAp (CA due to pressure). Thus,

C Ao = C Af + C Ap

Predictions from CA and CN will then allow us to generate CL and CD of the

missile using the following equations:
4
 C D = C A cos α + C N sin α
C L = C N cos α − C A sin α

The general components making up the missile drag are shown in Figure 2. Drag
prediction is done using a component buildup method by adding the individual drag of
each component [1]. Missile Datcom (97) allows each component's drag coefficient to be
analyzed.

As the missile operates through a wide range of Mach numbers, the drag
components will vary significantly, as shown in Figure 3. At subsonic speeds, skin
friction dominates while at higher Mach numbers, wave drag will dominate [1].

Figure 2 Component Build-up Model [From 1]

5
 Figure 3 Variation due to Mach Number of Drag Components [From 1]

2. Cf Analysis

A boundary layer exists between the body of the missile and the freestream. The
shear force due to fluid viscosity at the wall is the source of the skin friction drag. The
amount of skin friction drag depends not only on whether the layer is laminar or turbulent
and can vary as much as 10 times [1], but also on whether a laminar separation bubble
forms at different altitudes and flow Reynolds numbers. In fact, multiple regions of local
separation are also possible in missile flow due to many vane sets and differently
deflected flaps. A turbulent flow condition generally results in higher Cf values;
compared to a laminar flow condition. It is also well known that Cf strongly depends on
Reynolds number as shown below [2]:

ρ ∞V∞ d
Re =
μ∞

As Re increases, Cf decreases. And since the missile in question has an operating

altitude from 0 to 30000m, it is crucial that thorough understanding be made of its
operating conditions because its unit Reynolds number can change by three orders of
magnitude from 105 to 108.

Typical Reynolds number for the three different flow conditions: Laminar,
Transitional and Turbulent is shown in Figure 4.
6
 Figure 4 Cf vs. Reynolds Number [From 1]

3. Other Force Components

In addition, the general flow over a missile is very complicated. The strong
vortical flow at angles of attack which develops over a body of revolution that comprises
the missile also affects the local flow and its separation. This can be especially significant
at large pitch and roll angles. The asymmetric formation of these structures and their
interactions could lead to very large and unexpected side forces that are difficult to
predict. Furthermore, these lead to unsteady forces that affect the maneuver
characteristics of the missile. Thus, even though much emphasis is placed on analyzing
the axial force and skin friction behavior, the other aspects are also briefly considered.

7
THIS PAGE INTENTIONALLY LEFT BLANK

8
 II. METHOD OF APPROACH

A. DATA GIVEN

A limited aerodynamic data set was provided by MSIC for comparison and use in
this study. As already stated, this is a combination of experiments and simulation. Thus,
the data set represents a sparse matrix of flow conditions and to a part of the much larger
aerodynamic flight regime of the missile. Furthermore, since the SA-2 type missile is an
incoming missile, it is difficult to establish the full performance details easily. In the data
set provided, the skin friction values are given only for 0 deg α.

MSIC also provided the geometry of the missile and is at best estimated from a
scaled diagram with minimum dimensions. Most parts of the geometry are estimated
through scaling and knowledge based on high speed missile systems. These estimations
include airfoil geometry (cross section), nose geometry, center of gravity, and pivoting
points for the wing control surfaces.

9
 Figure 5 Missile Dimension Provided by MSIC

1. MSIC Data Source

a. Wind Tunnel Testing

The primary data set for validations was obtained from wind tunnel tests.
Subsequently, Simulink was used to complement this set and generate additional data. As
this study compares the predicted results for a full-scale missile with the wind tunnel
data, the following limitations must be kept in mind.

It is generally not possible to adequately simulate the free flight, full-scale

Reynolds number in experiments on scaled models. The maximum unit Reynolds number
10
that can be achieved in wind tunnel testing is about 105 – 106. On the other hand, the
missile experiences at least an order of magnitude higher unit Reynolds number over the
different altitudes that it flies in. Thus, unless the boundary layer is tripped artificially to
reproduce the effects of the higher flight Reynolds number, the laboratory data includes
the effect of natural transition in addition to the presence of natural laminar flow over
some of the body length. In order to induce transitional and turbulent flow over the
object, trips were employed. However, utilization of trips to create the required flow
condition is indeed an art and to accomplish it consistently to obtain different Reynolds
numbers is difficult.

Another well known limitation is the scale effect created due to the
necessarily smaller scale of the laboratory scale model. A full scale wind tunnel testing of
the model will not be practical. With a scaled model, the forces acting on the test object
will be reduced proportionately while the coefficients remain the same. This reduction
will increase measurement difficulties and reduce the accuracy of the measurements. For
incompressible laminar flow over a flat plate the integrated skin friction coefficient [2] is:

1.328
Cf =
Re1c/ 2

And for incompressible turbulent flow [2]:

0.074
Cf =
Re1c/ 5

Thus, the laminar and turbulent skin friction coefficients vary as Re-1/2 and
Re-1/5 respectively. As Reynolds number depends on the overall length 'c' over which the
fluid is flowing, testing on smaller scale models tends to reduce the overall skin friction
values when compared to that are encountered by the full scale missile. The relationship
between Cf values and Reynolds number is shown in Figure 6.

11
 Figure 6 Cf vs. Reynolds Number [From 2]

Determination of these quantities over the full range of flight Mach

numbers from subsonic through to hypersonic in the laboratory is a significant challenge.
There is no one wind tunnel that can cover this wide range of flow conditions. The wind
tunnel size tends to decrease as the Mach number increases. The test times also decrease.
To minimize wind tunnel blockage effects, the models become even smaller at the higher
Mach numbers. Thus, the uncertainty associated with the test data on some of the
quantities of interest here increases. This is one of the reasons for extrapolating the wind
tunnel test data to generate Simulink data for the higher flight conditions.

b. Simulink

A portion of the data given is supplemented by results generated from

MATLAB/ Simulink modeling. This model is, however, developed from mathematical
flight mechanics models, measured airframe properties and generated aerodynamic
performance quantities. Thus, it lacks the fidelity and prediction capabilities of the 3-
dimensional 6 DOF aerodynamic prediction software. A prominent limitation, for
example, was the state of boundary layer on the missile -- whether it was natural,
transitional, or fully turbulent.

12
B. RESEARCH METHODOLOGY

1. Geometry Coding

Coding of the geometry is done through MissileLab. The Windows User Interface
in MissileLab enables users to check the geometry at every input stage. This includes 3-
dimensional sketches of every component. This eliminates the need for tedious
FORTRAN programming. To reduce design errors, MissileLab includes a self-check
process to ensure that all geometries are consistent. A 3-dimensional sketch output from
MissileLab is shown in Figure 7. This view is available throughout the geometry input
stage allowing the user to make changes as needed.

Figure 7 Coded 3-D Sketch of Missile with Booster Attached from MissileLab

a. Nose Geometry

A tangential ogive nose has been selected based on the scaled drawing
provided by MSIC. This ensures that there is no discontinuity in the curvature at the
nose-body junction.

13
 b. Fin Geometry

Fin geometries are measured by hand to the best of the author's ability.
The thickness of the fins and geometry of the airfoil have been selected based on
knowledge of supersonic missiles. The geometry of the airfoil is an elliptical shape.

Due to the lack of detailed technical drawings, the pivot point for
controllable fins is assumed to be in the middle of the cord length.

c. Body Geometry

The data given for the body was with the booster attached. MSIC’s
interest, however, is in the missile body aerodynamics after the booster separation phase.
This occurs 3 to 5 seconds after launch. The MissileLab output for the missile after
booster separation is shown in Figure 8

Figure 8 Coded 3-D Sketch of Missile After Booster Separation

d. Surface Roughness

As surface roughness was not recorded in the data given by MSIC, an

iterative process was adopted in an attempt to map the data provided. For CA mapping,
even for a smooth surface (roughness = 0,) computed CA values from MissileLab yielded
14
readings well above those from MSIC tests. In the Cf mapping, using a surface roughness
of 0.001016 produced results that were very close to the data provided. Therefore, this
surface roughness has been selected. It is equivalent to a relatively smooth surface on
which the paint is carefully applied. More discussions on the comparison will be carried
out in later sections.

2. Operating Conditions

In general, the SA-2 type missile can operate over a wide range of flow conditions
from M = 0.8 to M = 2.5. However, the modern missiles may be expected to fly at even a
higher speed range and also perform maneuvers that are quite dramatic and uncommon.
Due to lack of specific data on the operating conditions of the missile, several
permutations of conditions were arbitrarily selected for predicting the missile
performance.

a. Boundary Layer Conditions

The state of the boundary layers is a significant parameter in the

aerodynamic coefficients of interest. The large change in Reynolds number with the
significant change in the altitude experienced by the missile means that both natural and
fully turbulent boundary conditions have to be modeled. In addition, a combination of the
two as also the possible formation of a laminar separation bubble will need to be
considered at very high altitudes due to the large kinematic viscosity of air that can drop
the Reynolds number to sufficiently low values where the effects of the bubble can begin
to alter the skin friction values and also, CA significant. This was not pursued, however.

b. Fins Deflections

A trimmed flight condition has been modeled with no fin deflections.

15
 c. Angle of Attack and Altitude

Twelve angles of attack, α values were used in the simulations starting

from 0 deg α to 24 deg α in increments of 2 deg. Altitudes selected were 0m, 5000m,
10000m, 20000m, and 30000m.

d. Base Drag

With base drag forming as much as 50% of the total missile, data was
obtained with Base Drag (power off) and without Base Drag (power on) for comparison.
Moreover, after booster separation, the missile will continue to receive thrust from its
second stage propulsion for another 20 seconds before cruising without power in the
terminal stage. This necessitates the need to obtain data for both power on and power off.
Based on MSIC recommendations, this Thesis’ work will focus on the power off
condition. This is because of the missile flies under these conditions during the later stage
of the cruise profile and, most importantly, in the terminal phase.

16
 III. RESULTS

A. SIMULATION CONDITIONS

Part of the effort was aimed at generating the missile aerodynamic performance
characteristics at conditions that envelope the flow conditions for which MSIC has
supplied either the test data or the simulation data. It is also hoped that after establishing
confidence in the computations by validating against the measured data, the
characteristics for intermediate flow conditions can also be provided through this effort.
The conditions chosen are shown in Table 1.

Boundary Fixed
Surface Varying
Power Layer Plots (α:
Roughness (Altitude: m)
Conditions deg)
Off Natural Smooth Cf vs. Altitude Mach Number (1.2, 3, 4.5) 0
Cf vs. Mach Number Altitude (5000, 30000) 0
Cf vs. Altitude Mach Number (1.2, 3, 4.5) 12
Cf vs. Mach Number Altitude (5000, 30000) 12
Rough Cf vs. Altitude Mach Number (1.2, 3, 4.5) 0
Cf vs. Mach Number Altitude (5000, 30000) 0
Cf vs. Altitude Mach Number (1.2, 3, 4.5) 12
Cf vs. Mach Number Altitude (5000, 30000) 12
On Natural Smooth Cf vs. Altitude Mach Number (1.2, 3, 4.5) 0
Cf vs. Mach Number Altitude (5000, 30000) 0
Cf vs. Altitude Mach Number (1.2, 3, 4.5) 12
Cf vs. Mach Number Altitude (5000, 30000) 12
Rough Cf vs. Altitude Mach Number (1.2, 3, 4.5) 0
Cf vs. Mach Number Altitude (5000, 30000) 0
Cf vs. Altitude Mach Number (1.2, 3, 4.5) 12
Cf vs. Mach Number Altitude (5000, 30000) 12
Off Turbulent Smooth Cf vs. Altitude Mach Number (1.2, 3, 4.5) 0
Cf vs. Mach Number Altitude (5000, 30000) 0
Cf vs. Altitude Mach Number (1.2, 3, 4.5) 12
Cf vs. Mach Number Altitude (5000, 30000) 12
Rough Cf vs. Altitude Mach Number (1.2, 3, 4.5) 0
Cf vs. Mach Number Altitude (5000, 30000) 0
Cf vs. Altitude Mach Number (1.2, 3, 4.5) 12
Cf vs. Mach Number Altitude (5000, 30000) 12
On Turbulent Smooth Cf vs. Altitude Mach Number (1.2, 3, 4.5) 0
Cf vs. Mach Number Altitude (5000, 30000) 0
Cf vs. Altitude Mach Number (1.2, 3, 4.5) 12
17
 Cf vs. Mach Number Altitude (5000, 30000) 12
Rough Cf vs. Altitude Mach Number (1.2, 3, 4.5) 0
Cf vs. Mach Number Altitude (5000, 30000) 0
Cf vs. Altitude Mach Number (1.2, 3, 4.5) 12
Cf vs. Mach Number Altitude (5000, 30000) 12

Table 1. Flow Conditions for Cf Analysis

Boundary
Surface Varying Fixed
Power Layer Plots
Roughness (α: deg) (m)
Conditions
Off Natural Smooth CA vs. α Mach Number (1.2, 3, 4.5) 0
CA vs. Mach Number α (0, 12) 0
CA vs. α Mach Number (1.2, 3, 4.5) 20000
CA vs. Mach Number α (0, 12) 20000
Rough CA vs. α Mach Number (1.2, 3, 4.5) 0
CA vs. Mach Number α (0, 12) 0
CA vs. α Mach Number (1.2, 3, 4.5) 20000
CA vs. Mach Number α (0, 12) 20000
On Natural Smooth CA vs. α Mach Number (1.2, 3, 4.5) 0
CA vs. Mach Number α (0, 12) 0
CA vs. α Mach Number (1.2, 3, 4.5) 20000
CA vs. Mach Number α (0, 12) 20000
Rough CA vs. α Mach Number (1.2, 3, 4.5) 0
CA vs. Mach Number α (0, 12) 0
CA vs. α Mach Number (1.2, 3, 4.5) 20000
CA vs. Mach Number α (0, 12) 20000
Off Turbulent Smooth CA vs. α Mach Number (1.2, 3, 4.5) 0
CA vs. Mach Number α (0, 12) 0
CA vs. α Mach Number (1.2, 3, 4.5) 20000
CA vs. Mach Number α (0, 12) 20000
Rough CA vs. α Mach Number (1.2, 3, 4.5) 0
CA vs. Mach Number α (0, 12) 0
CA vs. α Mach Number (1.2, 3, 4.5) 20000
CA vs. Mach Number α (0, 12) 20000
On Turbulent Smooth CA vs. α Mach Number (1.2, 3, 4.5) 0
CA vs. Mach Number α (0, 12) 0
CA vs. α Mach Number (1.2, 3, 4.5) 20000
CA vs. Mach Number α (0, 12) 20000
Rough CA vs. α Mach Number (1.2, 3, 4.5) 0
CA vs. Mach Number α (0, 12) 0
CA vs. α Mach Number (1.2, 3, 4.5) 20000
CA vs. Mach Number α (0, 12) 20000

Table 2. Flow Conditions for CA Analysis

18
B. COMPONENTS INVOLVED IN AXIAL FORCE COEFFICIENT

The total axial force component of a missile includes several components. Missile
Datcom (97) calculates the axial force coefficient by including the various components of
friction, pressure/wave, and base drag for the missile body as well as all the fins. In
addition, the values arising from their interference are also calculated and output in a
table. The contributions of each of these factors are shown in Figure 9 and Figure 10. It
can be seen that base drag is the largest component and it contributes almost 50% to the
overall missile axial force coefficient CA. There is also a significant change in CA with
Mach number, which will be discussed in later sections.

CA Components vs Mach Number

(Power Off, Turbulent, Roughness 0.001016, 0 deg α, 0m)

0.5

0.45

0.4

0.35
CA Pressure/Wave
0.3 CA Base Drag
CA Friction
CA

0.25 CA Fin Set 1

CA Fin Set 2
0.2 CA Fin Set 3
0.15

0.1

0.05

0
0 1 2 3 4 5
Mach Numer

Figure 9 CA Components vs. Mach Number (Power Off, Turbulent, Roughness

0.001016, α = 0 deg, Alt. = 0 m)

19
 CA Components vs Mach Number
(Power On, Turbulent, Roughness 0.001016, 0 deg α, 0m)

0.3

0.25
CA Pressure/ Wave
CA Base Drag
0.2 CA Friction
CA Fin Set 1
CA Fin Set 2
CA
0.15
CA Fin Set 3

0.1

0.05

0
0 1 2 3 4 5
Mach Number

Figure 10 CA Components vs. Mach Number (Power On, Turbulent, Roughness

0.001016, α = 0 deg, Alt. = 0 m)

C. COEFFICIENT OF SKIN FRICTION

1. Effects of Power on Coefficient of Skin Friction

The effects of power on a missile showed no change to the skin friction

coefficient as seen in both Figure 11 and Figure 12 where Cf is plotted for altitudes of
5000 m and 30,000 m. for the power off conditions. The skin friction values decrease
monotonically with Mach number, perhaps due to increasing Reynolds number over the
body length. The nearly parallel curves lend support to this inference. The value at the
higher altitude is higher possibly due to the increased viscosity (by nearly two orders) at
the very large altitude of 30,000m for which this computation was carried out. However,
turning the power 'On' or 'Off' did not significantly affect the base drag.

20
 Cf vs Mach Number
(Power On, Turbulent, Roughness 0.001016, 0 deg α)

0.25
0.2

Cf 0.15 Datcom 5Km

0.1 Datcom 30Km

0.05
0
0 2 4
Mach Number

Figure 11 Cf vs. Mach Number (Power Off, Turbulent, Roughness 0.001016, 0 deg α)

Cf vs Mach Number
(Power Off, Turbulent, Roughness 0.001016, 0 deg α)

0.25

0.2

0.15 Datcom 5Km

Cf

0.1 Datcom 30K

0.05

0
0 1 2 3 4 5
Mach Number

Figure 12 Cf vs. Mach Number (Power On, Turbulent, Roughness 0.001016, 0 deg α)

2. Effects of Surface Roughness on Coefficient of Skin Friction

Skin friction is largely affected by the state of the boundary layer forming on the
missile surface. Surface roughness plays an important role in altering the boundary layer
state and thus, in determining skin friction drag. Figure 13 and Figure 14 show plots with
a surface roughness of zero compared to a surface roughness of 0.001016m. On a real
missile, this roughness value represents a reasonably well painted surface. The plots from
Missile Datcom showed significantly lower Cf values at lower altitude (20000m) for a
smooth surface. Above 20000m, Cf was observed to be insensitive to surface roughness.
21
An analysis of the variation of the properties of air with altitude shows that up to20000
m, the kinematic viscosity increase by an order, however between 20000 m and 30000 m,
it raises by an order. Thus, at an altitude of 20000m and above, the Reynolds number can
be as low as 3.0E+05. This resulted in a thick laminar layer in which the roughness
height was only a small fraction of the total boundary layer height, which caused the Cf to
become insensitive to surface roughness. Another possibility is the formation of a long
laminar bubble at low Reynolds number and thus having surface roughness insensitive Cf
values due to local flow separation inside the bubble.

Cf vs Altitude
(Power Off, Turbulent, Roughness 0.001016, 0 deg a)
0.25

0.2

0.15
Cf

Datcom Mach 1.2

0.1 Datcom Mach 3
Datcom Mach 4.5
0.05

0
0 10000 20000 30000
Altitude (m)

Figure 13 Cf vs. Altitude (Power Off, Turbulent, Roughness 0.001016, 0 deg α)

Cf vs Altitude
(Power Off, Turbulent, Roughness 0, 0 deg a)
0.25

0.2

0.15
Datcom Mach 1.2
Cf

Datcom Mach 3
0.1
Datcom Mach 4.5

0.05

0
0 5000 10000 15000 20000 25000 30000
Altitude (m)

Figure 14 Cf vs. Altitude (Power Off, Turbulent, Roughness 0, 0 deg α)

22
 3. Effects of Mach Number on Coefficient of Skin Friction

Cf values generally decrease with increasing Mach number as seen in Figure 13

and Figure 14. The lowest Cf values are observed for the highest Mach number of 4.5 for
the missile. The insensitivity of Cf at M 4.5 is still being analyzed.

4. Effects of α on Cf

Cf values for different angles of attack are plotted in Figure 15. Increasing α
reduces the Cf values, although not as drastically as increasing the Mach numbers. This
can be seen in Figure 15. As the missile increases in α, separation of air begins which
reduces the viscous effect and results in lower Cf values.

Comparison of Altitude and a using Cf vs Mach Number

(Power Off, Turbulent, Roughness 0.001016)

0.25

0.2
5000m, 0 alpha
0.15 5000m, 12 alpha
Cf

30000m, 0 alpha
0.1 30000m, 12 alpha
0.05

0
0 1 2 3 4 5
Mach Number

Figure 15 Comparison of Altitude and α Using Cf vs. Mach Number (Power Off,
Turbulent, Roughness 0.001016)

D. AXIAL FORCE COEFFICIENT

1. Effects of Power on Axial Force Coefficient

Without power, the base drag of a missile can be as high as 50% of the total drag
[1]. For the power off condition in Figure 16 the average CA is about 1.0 and, in Figure
17 the average CA is about 0.55. That is a significant increase of 45% in CA with power
off.

23
 Comparison of Altitude and α Using CA vs Mach Number
(Power Off, Turbulent, Roughness 0.001016)

1.2

0.8 0m, 0 alpha

0m, 12 alpha
0m, 24 alpha
CA
0.6
20000m, 0 alpha
20000m, 12 alpha
0.4 20000m, 24 alpha

0.2

0
0 1 2 3 4 5
Mach Number

Figure 16 Comparison of Altitude and α Using CA vs. Mach Number (Power Off,
Turbulent, Roughness 0.001016)

Comparison of Altitude and α Using CA vs Mach Number

(Power On, Turbulent, Roughness 0.001016)

0.6

0.5

0.4
0m, 0 alpha
0m, 12 alpha
0.3 0m, 24 alpha
20000m, 0 alpha
CA

20000m, 12 alpha
0.2 20000m, 24 alpha

0.1

0
0 1 2 3 4 5
-0.1
Mach Number

Figure 17 Comparison of Altitude and α Using CA vs. Mach Number (Power On,
Turbulent, Roughness 0.001016)

2. Effects of Surface Roughness on Axial Force Coefficient

At 0m, CA reduces slightly with smooth surfaces, as shown in Figure 16 and

Figure 18 This is an expected phenomenon as a smooth surface will have a reduced Cf
resulting in overall CA. However, in both plots, at an altitude of 20000m, CA becomes
relatively insensitive to surface roughness. The lower Reynolds number at high altitude
(3.0E+05) causes Cf to vary little as discussed in Figure 6. Thus, the resulting CA remains
unaffected.
24
 Comparison of Altitude and α Using CA vs Mach Number
(Power Off, Turbulent, Roughness 0)

1.2

1
0m, 0 Alpha
0.8 0m, 12 Alpha
0m, 24 Alpha
20000m, 0 Alpha
CA 0.6 20000m, 12 Alpha
20000m, 24 Alpha
0.4

0.2

0
0 1 2 3 4 5
Mach Number

Figure 18 Comparison of Altitude and α Using CA vs. Mach (Power Off, Turbulent,
Roughness 0)

3. Effects of Mach Number on Axial Force Coefficient

Variations of Mach number have a very large change on the axial force, as can be
seen in Figure 18. Most changes are observed through the transonic Mach number
regime. At transonic speeds, shocks begin to form on the missile body. The shocks that
form interact with the local boundary layer, shock oscillations also can be present and an
abrupt increase in pressure drag occurs and thus also, in the CA values. As the Mach
number continues to increase, these shocks will move towards the nose and the tail and
become detached from the missile body. The air on the missile body will now see a
velocity slower than the free-stream velocity and, thus, result in lower CA values.

4. Effects of Angle of Attack on Axial Force Coefficient

With Mach number less than 2.5, increasing the angle of attack results in lower
CA values as shown in Figure 19. This is due to flow separation from the surface of the
missile body. The kink seen for the case of α = 24 deg seems to indicate that this is a
likely scenario. Above Mach 2.5, a crossover in CA can be seen. This could be due to the
possibility of asymmetric vortices being generated, which is typical of missile body at
high angles of attack. These vortices will also induce side forces on the missile body,
further affecting the resulting overall CA.
25
 Comparison of Altitude and a Using CA vs Mach Number
(Power Off, Turbulent, Roughness 0.001016)

1.2

0m, 0 Alpha
0.8

0m, 12 Alpha
CA

0.6
0m, 24 Alpha
0.4

0.2

0
0 1 2 3 4 5
Mach Number

Figure 19 Comparison of α Using CA vs. Mach Number (Power Off, Turbulent,

Roughness 0, 0m)

5. Effects of Altitude on Axial Force Coefficient

As altitude increases, CA increases as shown in Figure 20. As explained in Figure

12 and Figure 13 earlier, Cf value can increase with altitude as the Reynolds number
becomes smaller. It should be noted here that Cf constitutes about 20-30% of the total
axial force.

Comparison of Altitude and a Using CA vs Mach Number

(Power Off, Turbulent, Roughness 0)

1.2

0.8 0m, 0 Alpha

5000m, 0 Alpha
10000m, 0 Alpha
CA

0.6
20000m, 0 Alpha
30000m, 0 Alpha
0.4

0.2

0
0 1 2 3 4 5
Mach Number

Figure 20 Comparison of Altitude Using CA vs. Mach Number (Power Off, Turbulent,
Roughness 0)

26
E. DATA COMPARISON WITH MSIC DATA

1. Skin Friction Coefficient Comparison with MSIC Data

The values for CA due to friction are close to the data provided by MSIC by using
a surface roughness of 0.001016 at an altitude of 5000m as shown in Figure 21. With
similar surface roughness, however, the data does not match at an altitude of 30000m. At
30000m, the same roughness may correspond to a different relative roughness since the
boundary layer is much thicker and the Reynolds number is much lower. Thus, obtaining
a better match with the data will require experimenting with different surface roughness
values. It is not clear whether the Simulink model included these effects. As mentioned
earlier, wind tunnel testing typically gives lower Reynolds numbers compared to values
obtained from actual airplane flight [2]. Studies have shown that low Reynolds number
shows high-drag coefficients. MSIC data shows a trend of over-predicting the Cf values
in Figure 21. This could be an indication of the low Reynolds number generally realized
in wind tunnel testing. Disparity from Missile Datcom (97) result is more prominent at
30000m. At very high altitudes of 30000m, where the medium possess large deviations in
properties from sea level, actual flight characteristics of the missile will prove difficult to
replicate in wind tunnels.

Cf vs Mach Number
(Power Off, Turbulent, Roughness 0.001016, 0 deg a)

0.3

0.25

0.2 Datcom 5km

Data 5km
Cf

0.15 Datcom 30km

Data 30km
0.1

0.05

0
0 1 2 3 4 5
Mach Number

Figure 21 Cf vs. Mach Number (Power Off, Turbulent, Roughness 0.001016, 0 deg α)

27
 2. Axial Force Coefficient Comparison with MSIC Data

Through clarification with MSIC, the CA data provided consists of CA due to

pressure drag and CA due to base drag. For power ‘on’ condition, a relatively close match
from MissileLab can be achieved. Both data can be matched relatively closely at lower
Mach numbers up to Mach 2.0 as shown in Figure 22. This could indicate the possibility
of inaccuracies in the creation of flow conditions for high Mach numbers in wind tunnel
environment.

CA vs Mach Number
(Power On, Turbulent, Roughness 0.001016, 0 deg a, 0m)

0.6

Datcom Total CA
0.5
Data CA
0.4
Datcom Body Only
CA

0.3 Datcom Base Drag

Datcom Press/Wave +
0.2 Base
Datcom Body Only + Base
0.1

0
0 1 2 3 4 5
Mach Number

Figure 22 CA vs. Mach Number (Power On, Turbulent, Roughness 0.001016, 0m, 0 deg
α)

For the power 'off' condition, a relatively close match can be achieved by
mapping the base drag from Missile Datcom output. The CA values from MSIC, however,
are notably lower and appear to capture only the base drag component.

With the assumption that MISC data captures only the base drag, results from
Missile Datcom (97) show a relatively close match. Parameters affecting base drag
include Reynolds number, Mach number, α, body fineness ratio, fin proximity, and the
presence of a boat tail or flare [1]. But the contribution of Reynolds number, α, fineness
ratio and fin proximity to base drag is small, and hence, the satisfactory match.

28
 CA vs Mach Number
(Power Off, Turbulent, Roughness 0.001016, 0m, 0 deg α)

1.2

1
Datcom 0 Alpha
0.8
Data 0 Alpha
CA

0.6
Datcom Body Only 0 Alpha
0.4 Datcom Base Drag 0 Alpha

0.2

0
0 1 2 3 4 5
Mach Number

Figure 23 CA vs. Mach Number (Power Off, Turbulent, Roughness 0.001016, 0m, 0 deg
α)

In both cases, as CA is a function of Mach number and Reynolds number,

reproductions of the true operating conditions are crucial, and this is a challenge in wind
tunnel testing and software modeling.

29
THIS PAGE INTENTIONALLY LEFT BLANK

30
 IV. CONCLUDING REMARKS

A systematic study using DoD supplied empirical software packages was carried
out to generate the aerodynamic performance coefficients of an SA-2 class missile. The
results were compared against a data set provided by MSIC. The model geometry was
generated using MissileLab and the computations were conducted using Missile Datcom
(97). The study enveloped a large matrix of flow conditions. A reasonable match was
obtained for Cf using a roughness of 0.000106m at sea-level conditions. At higher
altitudes, a mismatch remained and this difference was explained as due to the effects of
relative roughness that change steadily with altitude owing to the state of the boundary
layer and its height changes concomitant with the Reynolds number changes with
altitude. For CA, a good match was obtained for the power ‘on’ condition up to about
Mach 2.0. Additional work needs to be carried out for a higher degree of agreement.

A. LIMITATIONS

This study has some limitations, ranging from the use of a geometry that is
derived from visual observations and also, from generating it from scaled figures. Data
from a limited experimental test matrix was the basis for comparison of the computed and
test results. The well known, inherent limitations of test data limit the range of
comparisons that can be made. Factors such as free-stream turbulence, scaling effects and
Reynolds number simulation affect the wind tunnel testing results which were not
included in the studies.

Despite its capabilities, Missile Datcom also has some limitations which are
reported in the literature. Future studies can overcome these by modeling the appropriate
flow regimes for estimation of CA suitably. Improvements have also been made in later
versions of Missile Datcom through the investigation of fin normal force and center of
pressure prediction [8].

With increasing computing powers, use of CFD to solve complex aerodynamic

problems can provide more comparable results.

31
B. FUTURE WORK

One of the objectives for this Thesis was to set up an in-house capability for
missile performance computation. Due to the limited amount of time available for this
Thesis, operating conditions and analysis of the missile were restricted to a few values.
Further iterations on the conditions can be made with added simulation runs performed
on missile geometry changes, such as trimmed flights or varying the wings deflection.

32
 LIST OF REFERENCES

[1] Michael R. Mendenhall, Tactical Missile Aerodynamics: Prediction Methodology,

Progress in Astronautics and Aeronautics, Volume 142, 1992, pp. 1-60.

[2] John D. Anderson, Jr., Fundamentals of Aerodynamics, 2001, pp. 60-71.

[3] Thomas J. Sooy and Rebecca Z. Schmidt, Aerodynamics Predictions,

Comparisons and Validations Using Missile Datcom (97) and Aeroprediction 98
(AP98), AIAA 2004, 1246, 2004.

[4] William B. Blake, Missile Datcom User's Manual - 1997 FORTRAN 90 Revision,
AFRL-VA-WP-TR-1998-3009, 1998.

[5] Lamar M. Auman, MissileLab User's Guide, AMR-SS-06-12, Feb 2006.

[6] Steven Zaloga, Red Sam: The SA-2 Guideline Anti-Aircraft Missile (New
Vanguard), 2007.

[7] Amanda N. Horton and Melissa A. McDaniel, Identification and Correction of

Axial Force Prediction Discrepancies due to Angle of Attack Effects in Missile
Datcom, AIAA-2005-4833, 2005.

[8] Christopher Rosema, Mark Underwood and Lamar Auman, Recent Fin Related
Improvements for Missile Datcom, AIAA-2007-3937, 2007.

33
THIS PAGE INTENTIONALLY LEFT BLANK

34
 INITIAL DISTRIBUTION LIST

1. Defense Technical Information Center

Ft. Belvoir, VA

2. Dudley Knox Library

Naval Postgraduate School
Monterey, CA

3. Professor M. S. Chandrasekhara
Department of Mechanical and Astronautical Engineering
NASA Ames Research Center, M.S. 215-1
Moffett Field, CA

4. Robert L. Dewitt
Defense Intelligence Agency
Missile & Space Intelligence Center, MSD-2/ Dewitt
Redstone Arsenal, AL

5. Jay A. Witt
Defense Intelligence Agency
Missile & Space Intelligence Center
Redstone Arsenal, AL

6. Lamar M. Auman
Chief – Aerodynamic Technology
U.S. Army AMRDEC
AMSRD-AMR-SS-AT
Redstone Arsenal, AL

7. Professor Yeo Tat Soon, Director

Temasek Defence Systems Institute
National University of Singapore
Singapore

8. Tan Lai Poh (Ms), Assistant Manager

Temasek Defence Systems Institute
National University of Singapore
Singapore

35