MASTER'S THESIS

Installation Effects in Supersonic Aircrafts

Mathilda Johansson
2016

Master of Science in Engineering Technology

Mechanical Engineering

Luleå University of Technology

Department of Engineering Sciences and Mathematics
I NSTALLATION EFFECTS IN SUPERSONIC AIRCRA FTS

A M A S T E R ’ S T H E S I S S TU D Y

BY

M A T HI L D A J O H A N S S O N
A CKNOWLEDGMENT
I would like to direct my sincere gratitude to the persons who been involved in this
project and help me out during struggling times. First and foremost, I want to thank
Professor Pär Jonsén for making it possible for me to accomplish this project and for
being an academic support, both as a supervisor and examiner. Without his help I would
not have finished my master’s thesis at this early time of the semester.

I would also like to extend a big thank you to Richard Avellán at GKN Aerospace
Sweden AB, for his endless knowledge within this field and excellent help during the
project. Another person I would like to thank is Henrik Alvebrink, for excellent guidance
and support as a supervisor during my time at the company.

I would further like to thank Thomas Johansson and Kim Stenholm for their generous
help and support with advanced gas turbine theory and guidance in GasTurb12. You
helped me out many times.

Finally, I would like to thank my colleagues at the Propulsion Engineering department at

GKN Aerospace Sweden AB for the hospitality I have been receiving. A special thanks to
Anna Lind, who have encouraged my ability to continue working within the engine
performance field.
A BSTRACT
This report presents a Master’s thesis study done at GKN Aerospace Sweden AB (GKN)
in Trollhättan for Luleå University of Technology. The objective of this master’s thesis
was to study important installation effects at the air intake and outlet for a supersonic
aircraft with an integrated engine. This was accomplished by developing and applying
equations, formulas and tables for the investigated installation effects from a method
defined by The Boeing Company and implement these into the engine performance
simulation tool, GasTurb12. For validation of the simulation results the equations were
compared to experimental performance data from a McDonnell Douglas F-4J Phantom II
aircraft with installed J79-GE-8 engines.

The validation was divided into two steps, as the existing engine model in GasTurb12,
provided by GKN, was not fully identical to the Boeing material over the whole operating
envelope. The equations from GasTurb12 were therefore implemented into Matlab, where
the code was validated against the experimental data provided in The Boeing Company
material. The GasTurb12 data was then verified against the Mablab code. Consistent
results in the two steps resulted in a correct implementation of the equations.

Simulating a defined aircraft engine model in Gasturb12 using the implemented method,
the magnitudes of the external drags were generated. This enables further studies about
how to reduce the investigated installation effects and minimize the external drag for a
defined aircraft-engine-system. The future work will be to investigate the impact from the
different inlet- and afterbody geometries to find the best possible configuration for
minimal drag. Also some improvement of the implemented method should be made.
Considering the air pre-compression generated from the nose and forebody of the aircraft
or implementing data from various inlets than the three defined in this study are some
examples for further examination.
T ABLE OF C ONTENTS
1. Introduction ..................................................................................................................................... 1
1.1. Background ............................................................................................................................. 1
1.2. Problem statement ................................................................................................................... 2
2. Theory ............................................................................................................................................. 5
2.1. Aircraft engine design ............................................................................................................. 5
2.2. Speed of sound and Mach number .......................................................................................... 6
2.3. Shock waves ............................................................................................................................ 8
2.4. Isentropic flow in ideal gases .................................................................................................. 9
2.5. Incompressible and compressible flows ................................................................................ 10
2.6. Aircraft inlets......................................................................................................................... 11
2.7. Aircraft nozzles ..................................................................................................................... 12
2.8. Drag ....................................................................................................................................... 13
2.9. GasTurb12 ............................................................................................................................. 22
3. Method .......................................................................................................................................... 23
3.1. The development phase ......................................................................................................... 23
3.2. The implementation phase..................................................................................................... 28
3.3. The validation and verification phase.................................................................................... 33
3.4. Influence on the total net thrust ............................................................................................. 36
4. Results and Discussion .................................................................................................................. 37
4.1. The validation step ................................................................................................................ 37
4.2. The verification step .............................................................................................................. 44
4.3. Drag coefficient plots ............................................................................................................ 52
5. Conclusions ................................................................................................................................... 59
5.1. Future work ........................................................................................................................... 59
References ............................................................................................................................................. 60
Appendix 1 - Project plan
Appendix 2 - Nozzle boattail pressure drag coefficient as f(β)
Appendix 3 - Boattail drag correction for nozzle pressure ratios
Appendix 4 - Annular supersonic base pressure correlations
Appendix 5 - Reference base pressure ratio
Appendix 6 - Nozzle interference drag coefficient
Appendix 7 – GasTurb12 code for spillage drag
Appendix 8 - Spillage drag coefficients
Appendix 9 - Bleed mass flow ratio
Appendix 10 - GasTurb12 code for boattail drag
Appendix 11 - GasTurb12 code for base drag
Appendix 12 - Matlab code for spillage drag
Appendix 13 - Matlab code for boattail drag
Appendix 14 - Matlab code for base drag
Appendix 15 - GasTurb12 and Matlab code for interference drag
Appendix 16 - Names of the implemented tables in GasTurb12
Appendix 17 - Exisitng Matlab code for the nozzle boattail pressure drag coefficient as f(β)
Appendix 18 - Matlab code for theooretical maximum base drag
Appendix 19 - Exsisting Matlab code for the nozzle interference drag coefficient
N OMENCLATURE
Area,
The annular base area,
Inlet capture area,
Inlet throat area,
Local stream tube area ahead of the inlet,
Free-stream area of the air entering the inlet,
Area of bleed air,
Speed of sound,
Drag coefficient, dimensionless
Zero flow base pressure, dimensionless
Zero flow boattail drag, dimensionless
Specific heat capacity at constant pressure,
Specific heat capacity at constant volume,
Diameter,
Base diameter,
Jet diameter,
Maximum diameter of the afterbody,
Drag, N
Thrust, N
Ideal gross thrust, N
Specific internal enthalpy,
Specific stagnation enthalpy,
Velocity decay coefficient, dimensionless
Boattail length,
Mach number, dimensionless
Mach number corresponding to nozzle pressure ratio,
dimensionless
Nozzle design exit Mach number, dimensionless
Mass flow rate,
Number of engines
Base perimeter,
Perimeter of circle of area equal to the annular base area,
Jet perimeter,
Pressure, Pa
Base pressure, Pa
Nozzle exit static pressure (just inside the nozzle), Pa
Static pressure, Pa
Total pressure, Pa
Dynamic pressure, Pa
Dynamic pressure for high velocities, Pa
Specific gas constant,
Temperature, K
Stagnation (total) temperature, K
Velocity,
Ramp angle, degrees
Boattail angle, degrees
Specific heat ratio, dimensionless
Normalization parameter, dimensionless
 Deflection angle, degrees or Normalization parameter,
dimensionless
Mach angle, degree
Density,
Wave angle, degree
Thickness of the annular base area
Δ Additional drag correlation, dimensionless

S UBSCRIPTS
Add Additive drag
amb Ambient
Bmax Maximum base drag
BT Boattail
BS Base
eng engines
corr Corrected flow
IN Interference
inlet The entry of the aircraft inlet
j jet
large Large dimensions
m Maximum
ref Reference
s Static
small Small dimensions
spill Spillage
std Standard day conditions
sub Subsonic
sup Supersonic
t Total
∞ Free stream conditions
β Boattail drag
0 Total
1. I NTRODUCTION
Studies of future aircraft engines are regularly performed at GKN to continuously
improve and develop new solutions within the aerospace industry. The requirements of
future engines are specified from the aircraft manufacturers and compiled in a
requirement specification. Of major importance is the performance of the aircraft and
engine, e.g. fuel consumption, thrust, and engine size. The performance requirements are
defined at different flight altitudes and flight speeds (Flight Mach numbers). These are
used to create an engine performance model to specify the engine in important parameters
as pressure, temperature, fuel flow and thrust. The thrust of the engine provides the
aircraft with power to reach a certain speed, but the aircraft is also exposed to different
drag forces [1] [2]. For static equilibrium flight, the engine thrust must be equal to the
drag force.

Drag is defined as air resistance, or forces acting in an opposite direction of a moving

object due to its surrounding fluid, and depends on the area, speed and shape of the
moving object. The drag equation describes the relations between these parameters and is
defined as [3] [4]

(1)
Where is the area, is the dynamic pressure and the drag coefficient, , is dependent
of the object’s shape, Mach number and Reynolds number, .

Aircraft engines have been developed for a long time and are well proven in service. To
further improve the aircraft technology it is important to optimize the engine and aircraft
as a system e.g. in terms of installed performance. There are different kinds of installation
effects, which are mainly occurring due to air intake losses and losses around the
afterbody of the engine. To understand and minimize the installation losses in future
aircraft designs, new modeling methods are needed to improve the prediction of
propulsion characteristics in an aircraft system [2][5].

1.1. B ACKGROUND
Earlier work regarding installed engine performance usually includes only the inlet- and
nozzle pressure losses, as the other types of installation effects has been assigned the
aircraft analysis [6]. In 1972, The Boeing Company presented a method for calculating
propulsion system installation losses, used for preliminary studies of advanced aircraft
configurations [7][8][9][10]. The result was a Fortran computer program and is presented
as a report divided into four volumes. The different volumes include engineering
descriptions of the calculation procedure, programmers manual, sample calculation and
input data, bookkeeping definitions and data correlations. The Boeing material was used
as a template for this master’s thesis.

Up to date GKN have a first order method for analyzing the installed engine performance
for general supersonic engines, without using advanced and time consuming CFD
simulations or data from wind tunnel experiments. The method is explained in [5] and
investigates the drag magnitudes based on calculations in MATLAB. The method is also
based on the Boeing material in [7][8][9][10]. Using the same engine performance

1
simulation tool as for the uninstalled performance, GasTurb12, the method in this
master’s thesis will provide a first order estimate of result for the installation effects and
their impacts on the engine performance.

Due to the sensitive nature of installed aircraft and engine performance, little or no
literature was available in the public domain. Older references as [11], [12] and [13]
explain the phenomena in a physical way, which are still applied today. Newer sources as
[14], [15] and [16] also mention some of the installation effects, but are of limited value
regarding basic understanding of the aerodynamics.

1.2. P ROBLEM STATEMENT

The total installation effects are defined and decomposed into different categories,
depending on where and how the drag is created. The importance of the installation
effects is determined by the flight Mach number and altitude, aircraft type, geometry of
the air inlet and afterbody, number of engines, etc. and is varying from case to case. An
example of how the installation effects may be decomposed is presented in Figure 1.

The objective of this master’s thesis was to study important installation effects at the air
intake and outlet for a supersonic aircraft with an integrated engine, e.g. Lockheed F22
Raptor, Lockheed F35 Lightning II and Eurofighter. The study will provide GKN the
opportunity to simulate and predict installation effects in an early stage and obtain
performance data for the total system of an aircraft and engine. This was accomplished by
developing and applying equations, formulas and tables for the investigated installation
effects from assigned literature and implements these into the engine performance
simulation tool, GasTurb 12. For validation of the simulation results, the equations were
also implemented into MATLAB and compared with experimental performance data
from a McDonnell Douglas F-4J Phantom II aircraft with installed J79-GE-8 engines [9].

Inlet Installed Performance Nozzle/Afterbody

Nozzle
Internal Cfg
P2 Total Pressure Inlet Nozzle CD
Performance
P Recovery Pressure Internal
Recovery Performance

Boattail Drag DBT

Dbleed B.L. Bleed Drag

Engine FN
SFC Interference
Performance DIF
Drag

Dbypass Bypass Drag

Base Drag DBase
Nozzle/Afterb
Inlet Drag ody Drag

Dspill Spillage drag (Scrubbing

Dscrub
Dinlet DAB drag)

Figure 1. Installation effects that contributes to total installed engine performance, where the
green fields are the earlier modeled installation effects and the red fields are the ones that is of
interest in this project [5] [7].

The aim of this master’s thesis project was to develop and apply skills in the choice of
theory and method when solving unstructured problems as a profession of Master of

2
Science Mechanical Engineering within the Engineering Mechanics field. This was
achieved by implementation of the project through different phases to facilitate the
comprehension of the work and provide a simple step-by-step concept. The different
phases and their meaning are presented in Appendix 1 together with a Gantt-chart.

1.2.1. L I MI T AT I O N S
Only the contribution of external drag was analyzed for the installed engine performance,
since data for Inlet Pressure Recovery and Nozzle Internal Performance already exist in
GasTurb12. Inlet pressure recovery includes the amount of free stream pressure recovered
at the first compressor inlet section from the free stream. The nozzle internal performance
refers to the losses from the turbine exit to the nozzle outlet [14], for more information
see Figure 2 and Chapter 2. Scrubbing drag only exists when having parts in the
afterbody that hits the jet beam and is not relevant for the types of aircraft in this analysis
[5].

The installation effects were divided into two categories; prioritized- and secondary
effects, where the prioritized effects were initially investigated and validated. The
secondary effects were to be analyzed in a second stage if there was time left at the end of
the project. The installation effects were divided as follow:

Prioritized Installation Effects Secondary Installation Effects

 Spillage Drag  Interference Drag
 Boattail Drag  Boundary Layer Bleed Drag
 Base Drag  Bypass Drag

A more detailed introduction of the different drags will be presented in Chapter 2.

The effects of pre-compression of the free stream due to shocks from the aircraft nose
cone and the wing leading edge area will not be included in this study. The inlets are
considered non-choked at all time.

3
4
2. T HEORY
This chapter contains an overview of the theory used in the project and equations applied
to obtain the results. The basic aerodynamics, jet engine theory and configurations will be
explained together with important physical phenomenon like shock waves, isentropic
flow and drag.

2.1. A IRCRAFT ENGINE DESIG N

In 1937 two men named Hans von Ohain and Frank Whittle were the first persons
independently inventing a working jet engine, both based on the Brayton cycle concept
[17]. Today’s jet engines basically uses the same architecture, but with more improved
performance and design [1]. The most commonly used engine architecture for supersonic
aircraft in service today is the two spools mixed flow augmented turbofan. The different
components and processes in this engine may be divided into a number of thermodynamic
stations, see Figure 2 [1][18]. The free stream air is set to be station zero, but is also
called the infinity which is used in this report. Station 1 is the aircraft-engine interface
where the aircraft inlet ends and the upstream air is entering the engine. The inlets are
designed to provide the required airflow to the engine at a specified velocity and with
minimum pressure losses. Station 2 is the first compressor inlet, from where the air is
slowed down, compressed and directed to the high-pressure compressor inlet at station
25. At the last compressor exit, Station 3, the air is fully compressed. Next is the
combustion chamber with its burner exit at Station 4, where the temperature has been
increased by combustion of fuel. The combustion gases expand through the first turbine
stator exit at Station 41(the high-pressure turbine inlet). Station 45 is the low-pressure
turbine inlet, following with the low-pressure turbine exit, at Station 5. These two
turbines provide power to drive the compressors by two concentric shafts. The hot gas
passes Station 6, the reheat entry, and Station 7, the reheat exit, where the afterburner is
placed in between. The afterburner is a device for introducing additional heat to obtain
increased thrust when needed, where the extra fuel is added at the reheat entry. The nozzle
throat is placed at Station 8, where the walls are converging to increase the gas velocity
and it becomes sonic. For supersonic velocities, the gas velocity increases with increasing
cross section area,[1] which results in diverging walls at Station 9 to accelerate the air and
decrease the static pressure. Station 9 is the nozzle exit and the last station of the jet
engine, where the jet of hot gas is released and creates thrust which enables the airplane
to move forward [1][18][19].

Figure 2. Schematic diagram of a jet engine with an afterburner [7].

5
2.1.1. E N GI N E CAT E GO R I E S
There are different kinds of aircraft engines used for different missions. The most basic
version of a jet engine is the Turbojet, which only have one compressor and one turbine.
Another type is the Turbofan that is similar to the Turbojet, but has a larger fan that
provides air bypassing the compressor and the turbine and mixes with the combustion
gases before entering station 8, creating more efficient thrust. Turbofans with low bypass
ratios, like the one in Figure 2, are best suitable for supersonic aircrafts, while turbofans
with larger bypass areas are better for subsonic aircrafts. A third kind of jet engine is the
Ramjet engine that only consists of three sections; an diffuser inlet that compress the
incoming air and decreases the velocity, a combustion chamber where the air is
combusted and a convergent nozzle that ejects the heated air and produce thrust. A
Ramjet operates optimally at supersonic speeds and is mostly used for robots and rockets
[1][19].

2.1.2. E N GI N E P ER FO R MAN C E
The performance of an engine is primarily determined by three qualities; the specific fuel
consumption, air flow and thrust, and how they change with altitude and flight speed
(Flight Mach number).

Thrust is defined as the force that moves the aircraft forwards. The general thrust
equation is derived from Newton’s 2nd law and is given for a turbojet with
as [1] [20][6]

(2)
where equals the air mass flow through the engine, is the air velocity, is the exit
area and is the pressure at the specified station. Infinity represents the free stream
properties while Station 9 represents the engine nozzle exit and states the static
pressure.

Afterburners are often employed by military supersonic aircrafts to produce additional

thrust required at certain operation conditions. Due to the high fuel consumption,
afterburners are only used for short durations [6]. Engines including an afterburner must
take the added fuel flow into account, which gives the reheat thrust equation as [1][6]

(3)
where Station 1 represents the aircraft-engine interface, which is the entrance of the air
into the engine.

The thrust requirements for the engine aircraft are given from the required performance of
the aircraft, as the engine thrust must overcome the drag forces exposed at the fuselage
[1] [2]. Different engine types use different principles to achieve high values of the first
term in Equation (3). The Turbojets use a small mass flow, but keeps high velocity of the
air, while Turbofans have a large mass flow with low velocity [17].

2.2. S PEED OF SOUND AN D M ACH NUMBER

The air around us consists of a large number of molecules moving around randomly with
instantaneous velocity and energy changing in time. All these molecules define a mean

6
value of the molecular velocity and energy that only depends on the surrounding
temperature (for a perfect gas). When for example a balloon is burst, energy is released
and absorbed by the closest surrounding molecules in the air, which increases their
molecular velocity and collide with the neighboring molecules. In the collision, energy is
transferred between the molecules and so it proceeds out in space, creating a wave of
energy. The increased energy in the wave also creates a pressure change, which is picked
up by the eardrum and signalizes the pressure variation as sound. The wave is therefore
called sound wave and its penetrating velocity is defined as the speed of sound, see
[16].

The wave approaches the flow in front of the wave with velocity , pressure , density
and temperature , while the flow behind it recedes with velocity , pressure
, density and temperature , see Figure 3. Using the continuity
equation for the two regions gives the velocity

(4)

as the product between two infinitesimal quantities ( ) is almost zero and may be
neglected [16][15].

Figure 3. Flow characteristics before and after passing a sound wave.

Further derivation applying the momentum equation, adopting the propagation of a sonic
wave to be isentropic and using thermodynamic property relations the speed of sound, ,
in an ideal gas is defined as

(5)

where γ is the specific heat ratio of the fluid, is the specific gas constant and the
temperature in Kelvin [16][15].

The Mach number is defined as

(6)

7
where is the actual speed of a fluid or a moving object and is used to define velocities
for fluid flow regimes. As the speed of sound is a function of temperature, the Mach
number for an aircraft is changing depending of the weather conditions for the day.
Following classification of different flow regimes is applied [16][15]:

< 1, subsonic flow

= 1, sonic flow
> 1, supersonic flow.

2.3. S HOCK WAVES

The propagation of a sonic wave may be compared with the rings of the water surface
with the source of sound in the middle. When having a moving source at a subsonic
velocity the waves will look like in Figure 4a and for sonic and supersonic velocities the
source will be in front of the waves. When traveling at sonic velocity ( =1) the waves
are concentrated right at the beginning of the object, as showed in Figure 4b, and together
creating a more intense wave, which is called a shock wave. The flow in front of the
moving object, also called upstream flow, will have no idea of the approaching object, but
then change the direction and decrease the velocity of the flow streamlines as soon as it
has passed the shock. The decrease in velocity results in increasing static pressure,
density and temperature of the flow. For supersonic velocities the waves are too slow to
follow the object and propagate behind the object, creating an angled Mach wave as
shown in Figure 4c and can be observed as a infinitely weak oblique shock [1][3][16].

Figure 4. The propagation of sonic waves from a moving source at a) subso nic, b) sonic and c)
supersonic velocity.

The two kinds of shocks are called Normal shock wave and Oblique shock wave. The
normal shock waves are perpendicular to the free flow. As mention before, the flow
passing a normal shock will have significant velocity losses, but is dependent of the speed
in the “upstream” flow [15][16]. The relationship between the upstream flow Mach
number, and the downstream flow Mach number, , is derived and defined
depending only of the specific heat ratio in [16][15] as

(7)

The equation yields that an upstream supersonic Mach number closer to 1 will give a
higher downstream subsonic Mach number and the velocity losses will be decreased.

8
When moving in high speed the sonic waves are propagating behind the moving object,
which makes the flow in upstream flow completely unaware of the moving object until it
hits the front tip. Oblique shock waves are created when the front of the moving object is
angled and got their name as they obtain a tilted angle from the free flow, see Figure 5.
The angle is called the Mach angle, , and is the narrowest angle and minimum velocity
loss that can be obtained for a straight oblique shock. These types of waves are called
Mach waves and are defined as weak shocks. Stronger oblique shocks, must have a Wave
angle, , greater than µ and less than 90°; otherwise the shock will be detached to the
object and become a normal shock. The wave angle is dependent of the upstream Mach
number and the shape of the moving object [15][16]. An illustration of the oblique shock
geometry is shown in Figure 5.

Figure 5. The oblique shock and its important angels.

As for the normal shock waves, the parallel free flow streamlines passing a straight
oblique shock wave changes direction and proceeds parallel to the object wedge. The
angle of the changed direction is called the Deflection angle, , and is defined by the -
-relation [15][16]

(8)

Equation (8) gives two values of for one solution of θ; the narrower weak shock and
the wide strong shock. Most times it is the weak shock that is desirable when designing
aircraft inlets.

2.4. I SENTROPIC FLOW IN ID EAL GASES

Flow through a tube with gradually decreasing area and then gradually increasing area
(also called a Laval tube [21]) is a reversible process and the conditions of the flow will
return to its origin values. The second law of thermodynamics states that reversible,
adiabatic flows have constant entropy and are defined as isentropic flows [3][15] [14].

For high speed flows the stagnation enthalpy, , represents the total energy of a flowing
fluid stream and is defined as

(9)

9
where the first term, h, is the internal enthalpy and the second term represent the kinetic
energy.

For ideal gases with constant specific heats the stagnation enthalpy is given from the
entropy equations as

(10)

where is the stagnation (total) temperature, is the static temperature and is the
specific heat capacity at constant pressure.

Using the equations of state and the speed of sound equations, the following relations are
derived

(11)

(12)

(13)

where and represent the total quantities and are defined as the isentropic flow
relations [3][15][14]. and are static parameters, is the Mach number and is the
specific heat ratio.

2.5. I NCOMPRESSIBLE AN D CO MPRESSIBLE FLOWS

The viscosity and compressibility characteristics of air are dependent of Mach number
and will changes with speed. At low speeds (< 200 mph) the compressibility effects are
negligible and the flow is considered incompressible. For incompressible flows the
dynamic pressure, is given with as [3][15][14]

(14)

and the mass flow, as

(15)
with as the velocity, as the density and as the cross section area at a given station.

For higher velocities (>200 mph) the air is considered compressible, which defines the
mass flow as [15] [14]

(16)

10
where is the cross section area, is the total pressure and is the total temperature
for each engine station. R is the specific gas constant.

The dynamic pressure for high velocities is rewritten in terms of the Mach number
(equivalent to Equation (14)) and is given as [15][14]

(17)

where is defined as the ambient pressure.

2.6. A IRCRAFT IN LETS

The air inlet is of crucial importance for the engine performance since it delivers the air
flow required by the engine in the entire flight envelope with minimum pressure loss and
distortion. Most compressors have an inlet design Mach number around 0.4 and 0.5,
which is why the intake has to diffuse the air before it reaches the compressor inlet [22].

There are several kinds of aircraft inlets, which can be divided into two main categories;
Normal shock- and Oblique shock inlets. The Normal shock inlets have the simplest
configuration, where no affects from shock waves are considered. The most common
kind is the Pitot inlet, Figure 6, which is designed so that the entering air obtains a Mach
number below 1. The air is slowed down by the normal shock wave in front of the inlet,
followed by the diffuser to reach desired Mach value at the first compressor inlet. Here it
has to be a trade between loss, complexity, weight and costs for a more advanced intake
[21][22][23].

Figure 6. Normal shock inlet; Pitot inlet.

The oblique shock inlets are used for supersonic flights and are designed to decelerate the
flow from supersonic to subsonic, with low pressure losses. A longer cowl lip with gashes
intercepts the oblique shocks and decreases the Mach number for successfully, to finish
with a normal shock at the opening, or inside the intake, before the air enters the
compressor. There are three kinds of oblique shock supersonic inlets; using external- or
mixed compression to decelerate the air, see Figure 7. External compression inlets are
design to place the normal shock just at the inlet entry, where it is transformed into a
diffuser. Mixed compression inlets include oblique shocks compression within a portion
of the inlet and finish with a normal shock with a subsequent diffuser. Supersonic flight
may also use pitot-like inlets, where the air is decelerated by oblique shocks inside the

11
inlet throat [21][22][23]. This type of inlet is called internal compression inlets and is also
shown in Figure 7.

Figure 7. Three different types of compression for oblique shock aircraft inlets [24].

2.7. A IRCRAFT NOZZLES

The continuity equation states that an flow inside a channel has the same mass flow at all
cross section areas. Applying Bernoulli’s equation one can see that increasing the cross
section area decreases the velocity, which result in increased static pressure. At
supersonic velocities these findings apply vice versa [6][15].

At Station 9 of a jet engine, maximum velocity and minimum static pressure difference to
the outer flow of the jet plume is desired. For subsonic flights a convergent nozzle will
satisfy these demands and the air will reach velocities close to Mach 1. For supersonic
flights the outlet gas velocity must reach higher Mach numbers than 1. Using a nozzle
with a convergent part, that increase the air velocity up to Mach 1, and then transforms
into a divergent part will accelerate the exhaust gases up to Mach numbers above one.
Variable nozzle configuration is available for both designs and is used for engine flow
matching and expansion ratio adaption purposes [6][20]. The adjustable geometries are
particularly the throat area at Station 8 and the exit area at Station 9. Figure 8 shows the
principle of the two different nozzle types.

Figure 8. Working principle of a convergent and convergent-divergent nozzle.

12
2.8. D RAG
Drag is defined as the forces acting in an opposite direction of a moving object due to its
surrounding fluid, where the two most dominate types are the skin friction drag and the
pressure drag. Skin friction drag is created due to the viscous forces created at the surface
of the object, in the region called boundary layers. Boundary layers are a thin film region
where the air molecules affect each other and create a non-slip condition where the fluid
is decelerated. Inside the boundary layers the frictional shear stress generates the friction
drag, where its magnitude depends on the Reynolds number, the pressure distribution and
the surface finish. The Reynolds number, , determines if the surrounding air flow is
laminar or turbulent, where a low number, <2300, indicates laminar flow and a high
number, Re>4000 indicates turbulent flow. Values of 2300< <4000 indicated
transitional flow and is a mix of both laminar and turbulent flow [3][15][25]. The
Reynolds numbers is dimensionless and defined as the ratio of internal forces to viscous
forces as

(18)

where is the density, is the mean velocity and µ is the dynamic viscosity of the fluid.
is a characteristic traveling length of the air over the moving object.

If the flow is laminar or turbulent also affects the pressure drag, which is dependent of the
pressure gradient. Increased free stream velocity and deceased boundary layer thickness
decreases the pressure and creates a negative pressure gradient in the boundary layers.
Positive pressure gradient obtains vice versa, see Figure 9. Eventually the pressure
gradient obtains a value that causes flow reversal between the free stream flow and the
surface of the object, which results in flow separation and wake. The more flow
separation created, the more drag is obtained. In summary laminar flow causes less skin
friction drag, turbulent flow is more resistant to flow separation and results in less
pressure drag [3][15][25].

Figure 9. Pressure gradients in accelerating flow (Region 1) and decelerating flow (Region 2).

The total drag of an object is therefore a mix between skin friction- and pressure drag,
which can be reduced by two kinds of measures; streamlining and decelerating. If
pressure drag is the dominant drag, the shape of the object could be formed streamlined to
reduce the risk of flow separation. Skin friction drag is reduced by minimizing the wetted
area or by affecting the surrounding flow. Seeking for laminar flow across the whole
surface is one solution, but is often hard to achieve due to high flight Mach numbers and
Reynolds number. Instead, the transition from laminar to turbulent flow could be delayed

13
by removing slow moving air from the boundary layers using porous surfaces with small
holes. Eventually, when the flow has reached turbulence, the surface has to be as fine and
polished as possible to decrease the shear stress friction or increasing the Reynolds
number. All these parameters are represented by one coefficient; the drag coefficient, CD
[3][15][25]

The size, or area, of an object is also affecting the magnitude of drag and is often used as
the reference when calculating the drag coefficient from the drag force, as shown in
Equation 1. The selected reference area is often the region where the evaluated drag is
acting, but could be any other area preferred. If two drag coefficients are to be compared,
the same reference area has to be used. It is therefore of importance to specify which
reference area used when presenting a drag coefficient [14]. In this report the capture area
and the maximum area will be used for inlet- respective afterbody drag and will be
explained further in this chapter.

2.8.1. S P I L L AG E D R AG
Spillage flow is defined as the stream tube airflow that is captured by the inlet but is not
required by the engine and therefore “spills” on the outside of the inlet and creates
spillage drag, see Figure 10 for illustration. The air inlet is designed for the maximum
engine airflow, which is not always required by the engine, and the amount of spilled air
is then changing with throttle settings [14]. The spillage drag is affected by two things;
the free stream tube momentum loss and the behavior of the air when passing the cowl
lip.

Figure 10. Spillage flow at a Pitot inlet (left) and a n Oblique shock inlet (right) seen from the
side, where blue is the incoming air and red is the air spilled at the side of the inlets.

The first phenomenon is called additive drag or pre-entry drag and is created due to the
airflow losses from the free stream tube to the inlet opening. The additive drag is
dependent of the inlet configuration, free-stream Mach number and shock geometry.
Additive drag is defined as [7][13]

(19)

for an normal shock inlet, as a Pitot inlet, where is the cross-sectional area at the inlet
lip, is the velocity and is the static pressure at 1) the inlet lip and ) the free
stream, see Figure 11. This equation is defined for an external compression Pitot inlet at
all Mach numbers. For oblique shock inlets the additive drag is defined as [7][13]

14
 (20)

for all Mach numbers with the normal shock inside or at the inlet lip, where
is defined as ramp drag and obtained from plotted data in [7]. α is the angle of the ramp
illustrated in Figure 12.

Figure 11. Additive drag of a Pitot inlet [7].

Figure 12. Additive drag of an oblique shock inlet [7].

The second phenomenon is called the lip suction effect and is created when the spillage
air is accelerating over the cowl lip, which decreases the pressure and produces a negative
drag that partly cancels out the additive drag. The lip suction effect depends on the shape
of the cowl lip, bluntness and sideplate cutback and is represented as a correlation
factor , which is multiplied with the additive drag (Equation (19) and (20)) to receive
a total spillage drag [14][26]

(21)
This correlation factor is obtained from experimental data and may be found in graphs
and tables from [10] for different inlet configurations.

The spillage drag coefficient is computed using Equation (1) and the capture area, Ac as
the reference area, which in this case gives

(22)

15
where Q is the dynamic pressure for compressible flows. The capture area for a Pitot inlet
equals the area surrounded by the cowl lip, while for oblique shock inlets the capture area
is illustrated in Figure 13 .

Figure 13. The capture area (A c ), the local stream tube area ahead of the inlet (A 0 ), the free-
stream tube area of air entering the inlet (A 0I ) and the aircraft throat inlet area (A inlet ) for a
oblique shock inlet. The different between A 0I and A 0 is due to bleed air and defined as A 0BLC.

There are two other areas of importance when working with aircraft inlets. These two are
called the local stream tube area ahead of the inlet, and the free-stream tube area of air
entering the inlet, and contribute to different mass flows. The ratios between the mass
flows may be represented by the ratios between the areas, as the density and the velocity
of the entering air are constant, see Equation (15). The local stream tube area ahead of the
inlet and the free-stream tube area of air entering the inlet are shown in Figure 13.

The inlet capture mass flow ratio is defined as the ratio between the free-stream tube area
of air entering the inlet and the capture area, and is given in [7][8] as

(23)

where is the bleed mass flow ratio. The bleed flow is used to remove the boundary
layers close to the inlet throat and is obtained from plotted data in [9][10] and shown in
Figure 13. The local stream tube area ahead of the inlet, is defined from the
compressible mass flow rate equation, Equation (16), as [14]

(24)

16
where the required mass flow from the engine, correlates the size of the area. and
are the total temperature and total pressure in the free stream. is the flight Mach
number and is the specific heat ratio of the surrounding air.

2.8.2. B O AT T AI L DRAG
Boattail drag is defined as the drag that is created at the engine nozzle due to external
airflow, depending on the size of the exit and boattail area. The nozzle opening is
adjustable for different boattail chord angles, β, which is defined as the angle of the
straight line between the largest outer diameter of the nozzle, and the nozzle exit,
, Figure 14. Decreasing the area results in increasing boattail angle, causing separation
of the external flow and creates pressure drag [5][7][15].

Figure 14. The nozzle configuration showing the boattail angle [9].

The boattail drag coefficient for circular arc boattails is defined as

(25)

for supersonic Mach numbers 1.0 ≥ < 3.0 [7][8]. For subsonic Mach numbers 0.4 ≤
≤ 0.95 the impact from the propulsive jet must be accounted for certain nozzle
geometries using the method in reference [12]. Larger dimensions of the maximum
diameter, , generate greater impact from the jet. For small dimensions of , i.e.

(26)

the impact is neglected and the boattail drag coefficient at nozzle pressure ratio ,
is given from tabled data [7][10], plotted in Appendix 2.

Nozzle pressure ratios other than 2.5 at subsonic speed with small dimensions of
must apply an additional drag correction. Because of an increasing pressure ratio, the
pressure difference between the two stages increases, which creates a negative drag. The
additional drag correction, is found in [7], Appendix 3, and the total boattail
drag coefficient for 2.5 ≤ ≤ 8 and M < 1 is then given as

.
(27)

17
When flying at supersonic velocities the negative drag from the plume will not be added.
The reason is that the flow field around the afterbody is complex due to the expansion
waves created around the afterbody and the shock pattern created in the plume [7][27].

For larger dimensions of ,

(28)

the impact of the propulsive jet gives the boattail drag coefficient for pressure ratios
between 2 and 7 as

(29)

where is the zero flow boattail drag, obtained from tables in [12] (Fig. 4) and

is the rate of change of boattail drag with base pressure drag also obtained in [12], (Fig.
7). is the base pressure increment;

(30)

where both the zero jet diameter function, , and the gradient

are also obtained from [12], (Fig.9 and Fig. 10).

The boattail drag is computed using Equation (1) and the maximal area as the reference
area, which in this case gives

(31)
where is the dynamic pressure for compressible flows and is the maximum area
of the afterbody.

2.8.3. B A S E DRAG
The base area, , of a nozzle is defined as the small annular area that is created
between the diameter at the end of the outer shell around the nozzle and the diameter of
the jet, Figure 15. Between the jet and the base there is a small region of low pressure
created. The fast moving air from the jet generates a pumping effect, where the jet wants
to pump the air away, which results in reduced pressure over the base surface. The
boundary layers at the nozzle exit influence the basic drag and works as an insulator, as it
reduces the effective dynamic pressure of the outer flow and decreasing the jet pump
effect [7][11][27].

18
 Figure 15. Illustrations of the base drag parameters [7][10][12].

The base drag is dependent of the pressure ratio between the jet and the outer flow, the
boattail angle, Mach number, nozzle geometries. For subsonic Mach numbers the impact
from the propulsive jet must again be accounted for certain nozzle geometries using the
method in reference [12]. Larger base diameter, , generates greater impact from the jet
plume. For small dimensions of , see Equation (28), the impact is neglected and the
base drag coefficient is given as

(32)

where is the annular base area, is the larger outer area of the nozzle and is
the base pressure ratio plotted in Figure 16 [8].

Figure 16. The base pressure and free ambient pressure ratio plotted against the jet total pressure
and free ambient pressure ratio.

For larger dimensions of , see Equation (28), the impact from the propulsive jet gives
the base drag coefficient

(33)

19
where is the zero flow base pressure and is the base pressure increment, both
obtained from tables and equations in [12] (Fig. 5, Fig. 9 and Fig. 10), and Equation (30).

For supersonic Mach numbers 1.0 ≤ M ≤ 2.2 a method using correlations parameters is
defined. The procedure of determine the base drag is given in six steps [7] [8] [10]1;

1. Compute the correlation parameter C ;

(34)

2. Find the corresponding correlation parameter from [7][10], see Appendix 4.

3. Compute the velocity decay coefficient, for ;

(35)
otherwise .

4. Find from [7][10], see Appendix 5.

5. Compute using B and by

(36)

6. Compute the basic drag coefficient,

(37)

The base drag for all Mach numbers is computed using Equation (1) and the maximal
area as the reference area, which in this case gives

(38)
where is the dynamic pressure for compressible flows and is the maximum area
of the nozzle.

1
Equation (36) was given as two different configurations with the base pressure ratio at K
multiplied in [7] [10] and divided in [8]. A quick plausibility check showed that the
equation in [8] is the one to use, as the other one would results in unreasonable high
negative drag.

20
The base drag upper limit is defined as the maximum pressure difference between the
base pressure at the annulus and the ambient pressure, and will only occur when the base
drag pressure corresponds to vacuum. In [11] and [28] the maximum base drag
coefficient is normalized by the dynamic pressure and defined as

(39)

Using the equation of dynamic pressure, the Mach definition and the ideal gas law, the
coefficient will only depend on the specific heat ratio and the Mach number,

(40)

and gives the maximum base drag as

(41)
where is the dynamic pressure for compressible flows and is the annular base
area.

2.8.4. I N T E R F E R EN C E D R A G
Having two aircraft engines operating close to each other will create a combined drag
which is greater than the two individual free flow drags added together. The difference in
drag is called interference drag. Interference drag is a result of having two bodies within a
common external airstream, where each flow field is modified by the other body [3] [11]
[29].

The space between the two engines, the nozzle exit diameter and the Mach number are
parameters determine the magnitude of the interference drag and are plotted against the
interference drag coefficient, , in Appendix 6 [7][10] which is defined for
numbers of engines as

(42)

The ideal gross thrust, Fg,i, is given as [29] [26];

(43)

where is the nozzle pressure ratio and the function is defined for a convergent-
divergent nozzle as [29] [30]

(44)

The interference drag for one single engine, , is calculated by the number of spaces
between the engines using

21
 (45)

The distance between the engines that affects the magnitude of the drag is called nozzle
spacing, , and is defined as the distance between the centers of the exit nozzles. An
illustration of the nozzle spacing is showing in Figure 17.

Figure 17. The afterbody-looking-forward view of a two-engine aircraft showing the space in
between.

2.9. G AS T URB 12
GasTurb12 is a useful gas turbine cycle program for evaluating the design and off-design
performances of most common types of gas turbines for both steady-state and transient
simulations. Depending on which type of simulation to be performed, there exist different
types of program settings; Basic thermodynamics and Engine design. Both these settings
offer design and off-design simulations.

The engine model geometry is generated in the design phase, where for instance mass
flow and areas are determined. In the off-design phase the engine model is simulated in
the flight envelope. Using component design data the overall engine performance will be
obtained using the conservation of energy, mass and momentum equations. The engine
models are divided into different modules, calculating the thermodynamic quantities for
each module and presented in a summary result sheet [31].

The program provides the user to compose other equations than the ones disposed in the
program as composed values. The results from these equations are presented as a list in
the summary result sheet. The program also enables implementation of tables in a 30x30
matrix, where intermediate values are interpolated by GasTurb12 [31].

Obtaining table values for implementation in GasTurb12 the utility program GasTurb
Details 5 provides the Digitizing a Picture tool, where plotted data are transferred to a txt-
file [32].

22
3. M ETHOD
The project was implemented through different phases to facilitate the comprehension of
the work and provide a simple step-by-step concept. There are mainly three phases that
include the investigation of the installation effects, which are applied for both the
prioritized and secondary installation effects. The different phases and their signification
are presented in Appendix 1, together with a time schedule presented as a Gantt-chart.

The first phase contained construction of equations, based on the different methods for
each installation effect defined in the material from The Boeing Company [7][8][9][10].
In the second phase the equations was implemented into the gas turbine cycle program
used at GKN; GasTurb12. The engine model used for experiments in [9] was not identical
with the engine model provided in GasTurb12, which is why the third phase was divided
into two steps; the validation- and the verification step. The validation step contained
validation between experimental data obtained from tests in The Boeing Company
material [9] and a Matlab code constructed of the implemented equations in GasTurb12.
The second step was the verification of the implemented equations in GasTurb12 with the
same Matlab code used in the validation step.

3.1. T HE DEVELOPMENT PHASE

This phase involves configuration of equations, formulas and tables for the chosen
effects. The material from The Boeing Company [7][8][9][10] consists of analytical and
empirical equations together with Fortran codes, which was used to gather all equations
needed for the simulation in GasTurb12.

3.1.1. S P I L L AG E D R AG
The additive drag calculations are based on a method used in [9] for calculating the
theoretical additive drag, which is based on an existing computer program from [13]. The
method consists of four steps and results in plots of the total spillage drag of an aircraft
inlet. The first step is to establish a suitable baseline mass flow ratio, for both

subsonic- and supersonic Mach numbers to determine an overall reference for the inlet
drag loss calculations. Using the additive drag program in [13], the theoretical additive
drags for chosen inlet geometry is calculated in step two, based on Equation (19) and
(20). In the third step the additive drags are adjusted to the baseline mass flow ratio and
results in plots of the additive drag coefficient, versus the inlet capture mass flow
ratio, . The last step implies multiplication with the correlation factor , obtained
from plotted data in [9] and [10]. The final step results in plots of the spillage drag
coefficient, for inlet capture mass flow ratios up to the baseline mass flow ratio.

The excess airflow from the baseline mass flow ratio up to creates a reference
spillage drag, but is normally associated to the aircraft drag and therefore not treated in
this report [9] [33].

The Boeing Material [9][10] presents plots of the spillage drag coefficient for different
aircraft inlet geometries and using Equation (22) the spillage drag is given as

(46)

23
where is the dynamic pressure for compressible flows. In this report, the dynamic
pressure will be defined with γ=1.4 for all installation effects as

(47)
The local stream tube area ahead of the inlet, is normalized with the corrected airflow,
as

(48)

where

(49)

and and are the normalization parameters defined as

(50)

and

(51)

with and as the standard values, =288.15 K and =101.325 Pa.

In [9] the installation effects of a McDonnell Douglas F-4J Phantom II fighter inlet is
investigated and will be used in this study for validation of the implemented equations.
An illustration of the F-4J inlet is shown in Figure 18. The other two inlets of interest for
implementation in GasTurb12 are a normal shock inlet and a Mach 2.5 external
compression inlet and are illustrated in Figure 19 and Figure 20.

Figure 18. F-4J Phantom inlet implemented in GasTurb12.

24
 Figure 19. Normal shock inlet implemented in GasTurb12.

Figure 20. 2.5 Mach external compression inlet implemented GasTurb12.

3.1.2. B O AT T AI L DR A G
The boattail drag equations were configured for aircraft engines nozzles having no base
area, i.e. the base diameter is equal to the exit diameter;

(52)
for the given Mach numbers 0.4-0.8, 0.85 and any M ≥ 0.9. The jet diameter was defined
as the nozzle exit diameter;

(53)
which gives the nozzle geometry ratio for jet propulsive impact simplified as

(54)

The boattail angle in radians for this study is shown in Figure 21 and by trigonometry
defined as

(55)

25
 Figure 21. Nozzle configuration with no base area.

The boattail drag coefficient value for mach numbers lower than 0.9 and nozzle
geometries is obtained from tabulated data together with the additional drag
correction for different boattail angles in Equation (27) (see Appendix 2 and 3). Mach
numbers lower than 0.9 and nozzle geometries give a boattail drag coefficient
when putting Equation (30) into (29). The resulting equation is given as

(56)

Supersonic Mach numbers, , implies that the boattail drag coefficient is computed
by using Equation (25) [7][10] and for the drag coefficient is interpolated
between the two coefficients for Mach numbers 0.9 and 1.0 [8].

The total boattail drag is calculated for all Mach numbers by Equation (31) [8].

3.1.3. B A S E DRAG
The base drag equations is configured for a nozzle geometry as shown in Figure 14 in two
different scenarios; subsonic and supersonic . For subsonic
flight, the base drag coefficient for small dimensions of is defined by Equation (32),
see Equation (26). For large dimensions of , see Equation (28), the base drag
coefficient is obtained when putting Equation (30) into Equation (33);

(57)

and is defined for pressure ratios of 2 ≤ ≤ 7.

For supersonic flight the six-step-method described in chapter 2.5.3 is used. The areas in
Equation (34) and (35) are rewritten to diameters by knowing that

(58)

(59)

26
and

(60)
Set Equation (25) = Equation (24) gives

(61)

with geometry definitions according to Figure 15.

The nozzle is expected to be ideally expanded, conducting that the Mach number
corresponding to nozzle pressure ratio is defined using Equation (13) as

(62)

since the static pressure at the nozzle exit will be equal the static pressure of the free
stream [1][6] [15].

The resulting correlation parameter and the velocity decay coefficient (for )
expressed in diameters and defined pressure ratios are given as

(63)

and

(64)

where is the nozzle exit static pressure just inside the nozzle, is the nozzle
design exit mach number and is the mach number corresponding to the nozzle
pressure ratio.

The total base drag is calculated for all Mach numbers by Equation (38) [8].

3.1.4. I N T E R F E R EN C E D R A G
The interference drag equations were configured for aircraft engines with two
convergent-divergent nozzles as in Figure 21. The drag is defined for numbers of
engines at Mach number for space ratios by putting Equation
(43) into Equation (42);

27
 (65)

Normalization with the dynamic pressure for compressible flows, Q and the maximum
area gives the interference drag coefficient for a single engine;

(66)

3.2. T HE IMPLEMENTATION PH ASE

This phase included implementation of the constructed equations in Chapter 3.1 into
GasTurb12 as composed values. The tables from [7][9][10][12] where used in GasTurb
Details 5 to obtain table values and implemented by the Table Editor tool in GasTurb12.
The installation effects where created as separate files to facilitate the validation. Each
file starts at line 20, providing the user to insert other equations ahead. In this study, all
files include equations from the engine model created in GasTurb12 for validation. These
will not be commentated.

3.2.1. S P I L L AG E DR AG
Equation (23) and (46) to (51) was defined in the composed values section, where the
spillage drag coefficient was constructed by if-statements for the three different types in
inlet geometries. The names of the inlets are specified in Table 1 and simulated in
GasTurb12 by putting the selected inlet equal to one, while the others are set to zero. The
geometric variables are defined by data from [9] and a list with the names of the
composed variables in GasTurb12 is specified in Table 1. The spillage drag code is
presented in Appendix 7.

28
 Table 1. Composed spillage drag variables in GasTurb12 with units.

Variable in Variable in Unit in

report GasTurb12 Gasturb12
F-4J
Phantom F4J_inlet
inlet
Normal Normal_shock_
shock inlet inlet
2.5 Mach
Supersonic_
external
external_
compression
comp_inlet
inlet
π pi
M XM
R_air [J kg−1 K−1]
γ GM
Ac [ft2]
Pstd [Pa]
Tstd [K]
m2corr [kg/s]
RF RF
Q Q_dyn [Pa]
Ainf [m2]
ABAc

A0IAc
Cdspill
Fdspill [N]

The values of the spillage drag coefficient tables in [9] and [10] for the three selected
inlets, see Appendix 8, were obtained using the tool Digitizing a Picture in GasTurb
Details 5 [32]. The values were plotted and implemented into GasTurb12 by the Table
Editor tool. The bleed mass flow ratio for the F-4J inlet and the Mach 2.5 external
compression inlet where implemented in the same way, while the bleed mass flow ratio
for the normal shock inlet was set to zero.

The plots in Appendix 8 and 9 had a limited range of data for high Mach numbers. As the
Table Editor tool does not allow for undefined cells in the tables, the values of the
spillage drag coefficient and the bleed mass flow where extrapolated in these regions. The
extrapolations for each table are presented as follow;

o Appendix 9, the bleed mass flow for the F-4J inlet: Extrapolation of the three
curves with Mach number 0.9, 1.2 and 1.6 and inlet capture mass flow ratio
. The curve with Mach number 2.0 is extrapolated for inlet capture

mass flow ratio .

29
 o Appendix 9, the bleed mass flow for the external compression inlet:
Extrapolation of the three curves with Mach number 1.7, 2.2 and 2.5 for bleed
mass flow ratios
o Appendix 8, the spillage drag coefficient for the F-4J inlet: Extrapolation of the
two curves with Mach number 1.6 and 2.0 and coefficient values
and respectively.
o Appendix 8, the spillage drag coefficient for the normal shock inlet:
Extrapolation of the two curves with Mach number 1.8 and 2.0 and coefficient
values and respectively.
o Appendix 8, the spillage drag coefficient for the external compression inlet:
Extrapolation of the three curves with Mach number 1.7, 2.2 and 2.5 for
coefficient values

Avoiding inaccurate results for the spillage drag coefficient an if-statement is created,
which interrupts the calculations when an extrapolated value is used. The given value of
the spillage drag coefficient will then be , which indicates error. The
limits of extrapolation are alterable and can be changed by the user.

Appendix 16 presents the tables’ names, numbers and limits for each table created and
implemented in GasTurb12.

3.2.2. B O AT T AI L DR A G
Equations (25) to (31), (56) and (66) were implemented into GasTurb12 in the composed
values section. The boattail drag coefficient is constructed by if-statements, enabling all
the different scenarios due to Mach number and geometries. The geometric variables are
obtained from [9] and a list with the names of the composed variables in GasTurb12 is
presented in Table 2. The boattail drag code is presented in Appendix 10.

30
 Table 2. Composed boattail drag variables in GasTurb12 with units.

Variable in Variable in Unit in

report GasTurb12 GasTurb12
Dmax [m]
Lbeta [m]
S Space [m]
Δ Thick_base [m]
St9_P [kPa]
St9_T [K]
Pamb [kPa]
A8 [m2]
A9 [m2]
Amax [m2]
Dexit [m]
β beta [radians]
β beta_deg [degrees]
Db [m]
Cdbet1
BTCdbeta25
CorrBT
Over025

BTgrad

Cpbetaz
Grad

Zerojet
DeltaCpb
Under025
Cdbet0
Cdbeta
FDbeta [N]

The values of the tables from [12] (Fig. 4, Fig. 7, Fig. 9, Fig. 10), [7] and [10], see
Appendix 2 and 3, were obtained using the tool Digitizing a Picture in GasTurb Details 5.
The values were plotted and implemented into GasTurb12 by the Table Editor tool.
Limited range of data in [12] (Fig. 10) resulted in two extrapolated jet temperature
curves, providing more preferable data for temperatures over 922° K. Following equation
was used for the extrapolation

(67)

31
The plot in Appendix 2 had a limited range of data for the highest Mach number (0.95)
and as the Table Editor tool does not allowed undefined cells in the tables, the values of
the boattail drag coefficient for boattail angles larger than 16, β>16°, and small
dimensions of Dmax (Equation (26)) was set to . Appendix 16 presents the
tables’ names and numbers created in GasTurb12.

3.2.3. B A S E DRAG
Equation (30), (32), (33), (36) to (38), (57), (64) and (66) were defined as composed
values in GasTurb12. In order to facilitate the reading, extensive equations were
fragmented. The base drag coefficient is based on if-statements, enabling all the different
scenarios due to Mach number and geometries. The names of the composed variables and
equation fragments are listed in Table 3 together with the data from [9]. The base drag
code is presented in Appendix 11.
Table 3. Composed base drag variables in gasturb12 with units.

Variable in Variable in Unit in

report GasTurb12 Gasturb12
e e_bas
γ GM
Mdj
St9_Ps [kPa]
Dj [m]
PjePf [kPa]
Mj
C CC
B BB
K KK
PbPfK

PbPfM
Cpbz
Less025
More025
Cdb1
Cdb0
Cdbase
FDbase [N]
FN [N]

The values of the tables from [12] (Fig. 5, Fig. 9, Fig. 10), [7] and [10], see Appendix 4
and 5, were obtained using the tool Digitizing a Picture in GasTurb Details 5. The values
were plotted and implemented into GasTurb12 by the Table Editor tool. Appendix 16
presents the tables’ names and numbers created in GasTurb12.

The two configurations of Equation (36) were investigated to identify which one to apply.
Both versions where composed together with the other equations in GasTurb12 and used

32
to run a simple simulation with a engine model similar to a J79-GE-8 engine for Mach 1.2
at altitude 13716 meter (45 000 feet).

3.2.4. I N T E R F E R EN C E DRAG
Equation (45), (65) and (66) were implemented into GasTurb12 in the composed values
section. The geometric variables are defined by data from [9] and a list with the names of
the composed variables in GasTurb12 is presented in Table 4.

In the Boeing material [8] the function was set to the value of

(68)

which was also done in this report to obtain proper validation in future steps.
Table 4. Composed interference drag variables in gasturb12 with units.

Variable in Variable in Unit in

report GasTurb12 Gasturb12
Neng noe
intdc
FDNIAP [N]
FDin [N]
Cdin

The values of the table in [7] and [10], see Appendix 6, were obtained using the tool
Digitizing a Picture in GasTurb Details 5. The values were plotted and implemented into
GasTurb12 by the Table Editor tool. Appendix 16 presents the table’s name and number
created in GasTurb12.

3.3. T HE VALIDATION AN D VERIFICATION PHASE

The engine model used for simulation in GasTurb12 had approximately the same
characteristics as a J79-GE-8 engine, with some differences in the input data at different
Mach numbers. This resulted in complications with the validation of the experimental
performance data received in [9] and the results from the equations implemented in
GasTurb12, as the results would not be identical. Enabling the results to be validated, a
Matlab code for each installation effect where created. These codes are copies of the
composed values created in GasTurb12 and presented in Appendix 12 to 15. The
validation of the implemented equations in GasTurb12 was therefore performed in two
steps for each installation effect;

 Validation of the Matlab code with the experimental performance data in The
Boeing Company material
 Verification of the implemented equations in GasTurb12 against the Matlab code
used in the validation step.

The validation step implied that the same input parameters used in the Boeing material
were used in the Matlab code. The input data were defined for six different operating
points, with different Mach numbers and flying altitudes. The output drag coefficients for

33
the Boeing material and the Matlab code were plotted in a scatter plot to obtain a holistic
view of the validation results.

The verification step used the same Matlab code, but the input parameters from the
engine model in GasTurb12. The simulation was done for the matching six operating
points and the output drag coefficients were plotted in a scatter plot. The implemented
equations in GasTurb12 will be correct if the output data from the two sources are
consistent.

The Matlab codes where constructed so that the values obtained from plots in [7][9][10]
and [12] were typed in by the user manually. The six simulated operating points are
presented in Table 5.
Table 5. The six operating points presented in altitude and Mach number.

Operating point 1 2 3 4 5 6
Mach no. 0.4 0.6 0.8 1.2 1.6 2.0
Altitude [m] 4572 7620 10668 13716 10668 10668

3.3.1. S P I L L AG E DR AG
In the validation step the two first operating points, were not validated, as no
plotted data of the spillage drag coefficient for are available from the plots. The
third operating point, , was adopted the Mach number when obtaining
values from the bleed mass flow ratio plot. This because no plotted data of the bleed mass
flow ratio for are of knowledge. The values from the plots were obtained by
manual reading of the tables for the F-4J inlet in Appendix 8 and 9, using the Digitizing a
Picture -tool in GasTurb Details 5.

The verification step was performed in the same way as the first, except using the input
data from the engine model in GasTurb12. The two first operating points were not
verified because of the limited range of data. For the Matlab code the third operating
point was adopted the Mach number for the values from the bleed mass flow
plot.

The output data from GasTurb12 was obtained by running the code as composed values.
The program interpolates the values from the implemented tables, which are used to
calculate the drag and the mass flow ratios in the code. The inlet mass flow ratio, and

the inlet capture mass flow ratio, for the validation- and the verification step were
plotted in two scatter plots for comparison.

3.3.2. B O AT T AI L DR A G
For the validation step the values of the drag coefficient at pressure ratio 2.5 were
obtained using an already existing Matlab code from [5], shown in Appendix 17, and
manually typed into the boattail drag Matlab code. The boattail drag correction is
calculated inside the code by a six degree equation.

The verification of the Matlab code data in the second step against the engine model data
from GasTurb12 was implemented the same way as in step one, with the input parameters

34
from GasTurb12. The jet total pressure was assumed to be equal the total nozzle exit
pressure.

All operating points had a ratio of which implied no validation of the

method from [12]. A second validation case of larger dimensions at was
constructed, with a maximum diameter and a boattail length of 100 inches and a base
thickness of 0.1 meter. The other input parameters where set as the ones from GasTurb12.

3.3.3. B A S E DRAG
Validation of the base drag coefficient was not performed, as any experimental
performance data from the Boeing material existed. The base drag code was therefore
verified only by the second step against the GasTurb12 engine model data for the six
operating points. The F-4J nozzle configuration was used for simulation, with an added
estimated base thickness of 0.025 m. The design Mach number was set to 1.3 and the
specific heat ratio at the nozzle was specified in the Boeing material as =1.3. The jet
total pressure was assumed to be equal the total nozzle exit pressure. The equations for
obtaining the correlation parameter B and the base pressure ratio at K in Matlab where
obtained from the code in [5].

Determine the reliability of the base drag coefficient, the results was verified against
another Matlab code, calculating the maximum theoretical base drag coefficient for each
operating point. If the base drag coefficient does not exceed the maximum base drag
coefficient, the result is reliable. The maximum base drag Matlab code is presented in
Appendix 18.

All operating points had a ratio of which implied no validation of

verification of the method from [12]. The same second validation case as for the boattail
drag coefficient was performed, with a maximum diameter and a boattail length of 100
inches and a base thickness of 0.1 meter.

3.3.4. I N T E R F E R EN C E DRAG
Validation of the Matlab code for interference drag for , was not performed, as
values of the interference drag coefficient for number of engines, for
are not available in the plot in Appendix 6. For an already existing
Matlab code from [5], shown in Appendix 19, was used to obtain the values of .
These values were manually implemented into the Matlab code for interference drag,
receiving the normalized interference drag coefficient.

In the verification step the same operating points were verified as in step one and the
input data from the engine model in GasTurb12 were applied. The values of the
interference drag coefficient for number of engines were obtained using the same
Matlab code as in step one.

35
3.4. I NFLUENCE ON THE TOTA L NET THRUS T
In order to obtain a comprehension of the magnitudes of the external drag caused by the
installation effects, a simulation of a common operating point for a supersonic aircraft
was performed in GasTurb12. The engine model of J79-GE-8 engine character with an F-
4J Phantom II inlet, used in the validation and verification phase, was applied at the same
altitude for different Mach number. The drag results were plotted in an area diagram,
showing the different drag contributions from each installation effect and their influence
on the total net thrust.

The simulation was performed for Mach numbers 0.4 to 2.0 at an altitude of 11000 meter.

36
4. R ESULTS AND D ISCUSSION
The results from the implementation phase for the four investigated installation effects
are presented as flow charts of the GasTurb12 implementation in Appendix 20 to 23,
providing a holistic view of the processes.

The validation of the Matlab code with the experimental data obtained from the Boeing
material is presented as tables containing both the input and output data generated. The
verification of the GasTurb12 code and the Matlab code is presented in the same way,
with input data obtained from GasTurb12. The output drag coefficients from both cases
are presented as scatter plots and the results are discussed. The total drag from the
installation effects and their influence on the total net thrust at a certain altitude is
presented in an area plot.

The choice of using the method from The Boeing Company was based on the availability
of open source material. These types of studies are classified within the issuing
companies, but due to prescription this material was released for the public. The copies
are not of high quality, as the papers are written by typewriter and some words or
numbers have been erased during the years. This could result in inaccurate validation and
the figures ans words may have to be “sanity” checked. Though, the results provide a
good estimation of predicting the influence of the total net thrust. It is important to
remember that this way of bookkeeping the drags are just one of several others.

4.1. T HE VALIDATION STEP

The tables of the input- and output data for the four installation effects from the validation
step are presented in Table 6 to Table 11 for each operating point. The margin of error
between the experimental data in the Boeing material and the output Matlab data are
presented in percent, rounded to the closest integer, with the Boeing data as the reference.
The cause of the errors is discussed after the six tables.

37
Table 6. Input- and output data for boattail-, interference- and spillage drag at operating
point 1.

Input Output Experimental Margin of

Variable
Matlab Matlab data (Boeing) error [%]
π 3.14159265
M 0.4
Alt [m] 4572
[kPa] 57.1166
F4JI 1
NSI 0
SECI 0
[J /(kg K)] 287.058
[ft2] 6.82
GM (inlet) 1.4
[K] 288.15
[kPa] 101.325
m2corr [kg/s] 82.369
RF 0.977
Dmax [in] 38.6
S [in] 53.8
noe 1
[in] 23.4
Thick_base [m] 0
GM (nozzle 1.3
Mdj 1.3
A8 [sq,in] 693.72
A9 [sq,in] 719.614
Q [Pa] 6397
Pt8P0 2.414
β [deg] 10.093 10.084 0
0.0299 - -
0.030 0.030 0
[N] 145.7 144.9 0
NaN - -
NaN 0.031 0
[N] NaN 148.8 0
NaN 0 -
[N] NaN 0 -
/ 0.837 0 -
/ 0.837 0 -

38
Table 7. Input- and output data for boattail-, interference- and spillage drag at operating
point 2.

Input Output Experimental Margin of

Variable
Matlab Matlab data (Boeing) error [%]
π 3.14159265
M 0.6
Alt [m] 7620
[kPa] 35.1324
F4JI 1
NSI 0
SECI 0
[J /(kg K)] 287.058
[ft2] 6.82
GM (inlet) 1.4
[K] 288.15
[kPa] 101.325
m2corr [kg/s] 85.704
RF 0.98
Dmax [in] 38.6
S [in] 53.8
noe 1
[in] 23.4
Thick_base [m] 0
GM (nozzle 1.3
Mdj 1.3
A8 [sq,in] 727.199
A9 [sq,in] 777.797
Q [Pa] 8977.8
Pt8P0 2.677
β [deg] 8.663 8.654 1
0.025 - -
0.026 0.025 4
[N] 175.93 172.66 2
0.005 - -
0.025 0.026 -4
[N] 168.11 176.23 5
NaN 0 -
[N] NaN 0 -
/ NaN 0 -
/ 0.6529 0 -

39
Table 8. Input- and output data for boattail-, interference- and spillage drag at operating
point 3.

Input Output Experimental Margin of

Variable
Matlab Matlab data (Boeing) error [%]
π 3.14159265
M 0.8
Alt [m] 10668
[kPa] 23.7734
F4JI 1
NSI 0
SECI 0
[J /(kg K)] 287.058
2
[ft ] 6.82
GM (inlet) 1.4
[K] 288.15
[kPa] 101.325
m2corr [kg/s] 81.132
RF 0.984
Dmax [in] 38.6
S [in] 53.8
noe 1
[in] 23.40
Thick_base [m] 0.0
GM (nozzle 1.3
Mdj 1.3
A8 [sq,in] 771.312
A9 [sq,in] 865.624
Q [Pa] 10650.5
Pt8P0 3.057
β [deg] 6.584 6.573 0
0.0141 -
0.014 0.013 7
[N] 111.860 106.270 5
0.013 - -
0.043 0.043 0
[N] 341.430 346.560 1
0.040 0.044 9
[N] 269.930 296.920 9
/ 0.551 0.547 1
/ 0.542 0.542 0

40
Table 9. Input- and output data for boattail-, interference- and spillage drag at operating
point 4.

Input Output Experimental Margin of

Variable
Matlab Matlab data (Boeing) error [%]
π 3.14159265
M 1.2
Alt [m] 13716
[kPa] 14.7998
F4JI 1
NSI 0
SECI 0
[J /(kg K)] 287.058
2
[ft ] 6.82
GM (inlet) 1.4
[K] 288.15
[kPa] 101.325
m2corr [kg/s] 80.992
RF 0.984
Dmax [in] 38.6
S [in] 53.8
noe 1
[in] 23.4
Thick_base [m] 0
GM (nozzle 1.3
Mdj 1.3
A8 [sq,in] 867.924
A9 [sq,in] 1109.512
Q [Pa] 14918.2
Pt8P0 4.09
β [deg] 1.24 1.23
- -
0.001 0.001 0
[N] 13.413 13.52 1
0.019 - -
0.03 0.03 0
[N] 341.68 337.9 1
0.098 0.098 0
[N] 923.48 926.31 0
/ 0.5511 0.55 0
/ 0.537 0.537 0

41
Table 10. Input- and output data for boattail-, interference- and spillage drag at operating
point 5.

Input Output Experimental Margin of

Variable
Matlab Matlab data (Boeing) error [%]
π 3.14159265
M 1.6
Alt [m] 10668
[kPa] 23.7734
F4JI 1
NSI 0
SECI 0
[J /(kg K)] 287.058
2
[ft ] 6.82
GM (inlet) 1.4
[K] 288.15
[kPa] 101.325
m2corr [kg/s] 78.213
RF 0.935
Dmax [in] 38.6
S [in] 53.8
noe 1
[in] 23.4
Thick_base [m] 0
GM (nozzle 1.3
Mdj 1.3
A8 [sq,in] 812.564
A9 [sq,in] 1170.214
Q [Pa] 42602
Pt8P0 6.466
β [deg] -0.00004 -0.012 0
- -
0 0 0
[N] 0 0 0
0.0063 - -
0.0053 0.0056 5
[N] 170.4 180.3 5
0.1065 0.107 0
[N] 2874.4 2888.2 0
/ 0.619 0.619 0
/ 0.598 0.598 0

42
Table 11. Input- and output data for boattail-, interference- and spillage drag at operating
point 6.

Input Output Experimental Margin of

Variable
Matlab Matlab data (Boeing) error [%]
π 3.14159265
M 2.0
Alt [m] 10668
[kPa] 23.2946
F4JI 1
NSI 0
SECI 0
[J /(kg K)] 287.058
2
[ft ] 6.82
GM (inlet) 1.4
[K] 288.15
[kPa] 101.325
m2corr [kg/s] 63.407
RF 0.904
Dmax [in] 38.6
S [in] 53.8
noe 1
[in] 23.4
Thick_base [m] 0
GM (nozzle 1.3
Mdj 1.3
A8 [sq,in] 835.164
A9 [sq,in] 1170.214
Q [Pa] 66565.6
Pt8P0 8.241
β [deg] -0.00005 -0.012 0
0 - -
0 0 0
[N] 0 0 0
0.0046 - -
0.0025 0.0027 -7
[N] 125.23 138.98 -10
0.071 0.073 -3
[N] 2934.2 3078.8 -5
/ 0.6638 0.663 0
/ 0.6328 0.632 0

43
The results generated from the Matlab codes in the first operating point (Table 6)
correspond to the experimental data from the Boeing material accurately with margins of
error under 1%. In the second operating point (Table 7, green cells) the boattail drag
coefficient differs with 4%. The reason is the six degree equations for the boattail drag
correction in the Matlab code, which follows the curve in Appendix 3 more detailed than
the curve implemented in the Boeing material. The Boeing material defines fewer values
on the x-axis, which gives a more estimated average value between the points. Therefore
should the value in Matlab correspond better to the actual value. This conclusion also
applies for the third operating point (Table 8) as the curves for the nozzle pressure ratios
between 2 and 4.5 in Appendix 3 need more careful plotting.

The interference drag coefficient differs with about 5% in operating point 2 and 5 (Table
7 and Table 10, blue cells) and is explained due to the interpolation of the curves in
Appendix 6. The experimental data in the Boeing material is not exactly the same as the
output from the existing Matlab code for the interference drag coefficient for Neng number
of engines, CDINNeng. A small margin of error is understandable, as the values in the
Matlab code are obtained using a plotting program as Gasturb Details 5. In operating
point 6 (Table 11, blue cells), the margin of error is larger, which may be explained as
inaccurate plotting in the Boeing material. Using the Boeing material output values, the
interference drag coefficient for Neng number of engines is equals the values for the Mach
0.55-curve and not the 1.8-2.4-curve in Appendix 6.

The largest margin of error for the spillage drag coefficient was at the third operating
point (Table 8, red cells), where a higher value of the Mach number was applied in the
bleed mass flow plot from Appendix 9, resulting in a higher value of the inlet capture
mass flow ratio. This generates a higher value of the spillage drag coefficient in Appendix
8. At operating point 6 (Table 11, red cells), the margin of error is due to the difficulty to
manually obtain the precise value of the spillage drag coefficient table.

All margins of error are small relative to the total net thrust and will not be of high
impact. It is notable that Boeing has done some rounding of variables, which results in
margin errors as well. The most obvious evidence for rounding is shown in the percentage
difference between the drag coefficients and the percentage difference between the drags,
e.g. Table 11 (blue cells). The margins of error should be equal for both the drag
coefficients and the drags, as the drag coefficients are multiplied with input data from [9].
The difference is therefore explained due to rounding of the drag coefficients, the input
data or a mixture of both.

4.2. T HE VERIFICATION STEP

The input- and output data for the four installation effects from the verification step are
presented in Table 12 to Table 17 for each operating point. The margin of error between
the output GasTurb12 data and the output Matlab data are presented in percent with the
GasTurb12 data as the reference. The cause of the errors is discussed after the six tables.

44
Table 12. Input- and output data for boattail-, interference- and spillage drag at operating
point 1.

Input Output Output Margin of

Variable
GasTurb12 GasTurb12 Matlab error %
π 3.14159265
M 0.4
Alt [m] 4572
[kPa] 57.1819
F4JI 1
NSI 0
SECI 0
[J /(kg K)] 287.058
2
[ft ] 6.82
GM (inlet) 1.4
[K] 288.15
[kPa] 101.325
m2 [kg/s] 53.679
RF 1
Dmax [in] 38.6
S [in] 53.8
noe 1
[in] 23.4
Thick_base [m] 0
GM (nozzle) 1.3
Mdj 1.3
A8 [m^2] 0.407431
A9 [m^2] 0.429852
Q [Pa] 6404.38
Pt8P0=Pt9P0 2.809
T2 [K] 266.717
P2 [Pa] 63.8492
FN [kN] 49.87 - -
β [deg] 11.444 11.444 0
0.033 0.033 0
0.034 0.034 0
[N] 163.13 162.07 1
0.002 NaN -
0.02 NaN -
[N] 96.6798 NaN -
NaN NaN -
[N] NaN NaN -
/ 0.8535 NaN -
/ 0.8526 0.853 0

45
Table 13. Input- and output data for boattail-, interference- and spillage drag at operating
point 2.

Input Output Output Margin of

Variable
GasTurb12 GasTurb12 Matlab error %
π 3.14159265
M 0.6
Alt [m] 7620
[kPa] 37.6
F4JI 1
NSI 0
SECI 0
[J /(kg K)] 287.058
2
[ft ] 6.82
GM (inlet) 1.4
[K] 288.15
[kPa] 101.325
m2 [kg/s] 41.5735
RF 1
Dmax [in] 38.6
S [in] 53.8
noe 1
[in] 23.4
Thick_base [m] 0
GM (nozzle) 1.3
Mdj 1.3
A8 [m^2] 0.400486
A9 [m^2] 0.453303
Q [Pa] 9475.42
Pt8P0=Pt9P0 3.3691
T2 [K] 255.835
P2 [Pa] 47.9663
FN [kN] 39.55
β [deg] 10.519 10.519 0
0.0316 0.0316 0
0.03 0.03 0
[N] 213.48 213.93 0
0.0036 0.0035 -3
0.0166 0.016 -3
[N] 118.545 114.368 -3
NaN NaN -
[N] NaN NaN -
/ 0.644 NaN -
/ 0.643 0.643 0

46
Table 14. Input- and output data for boattail-, interference- and spillage drag at operating
point 3.

Input Output Output Margin of

Variable
GasTurb12 GasTurb12 Matlab error %
π 3.142
M 0.8
Alt [m] 10668
[kPa] 23.842
F4JI 1
NSI 0
SECI 0
[J /(kg K)] 287.058
[ft2] 6.82
GM (inlet) 1.4
[K] 288.15
[kPa] 101.325
m2 [kg/s] 31.989
RF 1.0
Dmax [in] 38.6
S [in] 53.8
noe 1.0
[in] 23.4
Thick_base [m] 0.0
GM (nozzle) 1.3
Mdj 1.3
A8 [m^2] 0.398364
A9 [m^2] 0.493370
Q [Pa] 10681.3
Pt8P0=Pt9P0 4.11387
T2 [K] 246.876
P2 [Pa] 36.354
FN [kN] 31.29
β [deg] 8.98 8.98 0
0.0267 0.0268 0
0.019 0.019 0
[N] 153.747 155.18 1
0.0115 0.0115 0
0.029 0.029 0
[N] 236.873 237.02 0
0.03 0.03 0
[N] 202.59 200.25 -1
/ 0.567 0.568 0
/ 0.56 0.56 0

47
Table 15. Input- and output data for boattail-, interference- and spillage drag at operating
point 4.

Input Output Output Margin of

Variable
GasTurb12 GasTurb12 Matlab error %
π 3.14159265
M 1.2
Alt [m] 13716
[kPa] 14.7477
F4JI 1
NSI 0
SECI 0
[J /(kg K)] 287.058
2
[ft ] 6.82
GM (inlet) 1.4
[K] 288.15
[kPa] 101.325
m2 [kg/s] 27.8261
RF 1
Dmax [in] 38.6
S [in] 53.8
noe 1
[in] 23.4
Thick_base [m] 0
GM (nozzle) 1.3
Mdj 1.3
A8 [m^2] 0.414289
A9 [m^2] 0.599183
Q [Pa] 14865.6
Pt8P0=Pt9P0 5.5459
T2 [K] 279.182
P2 [Pa] 35.7804
FN [kN] 31.29
β [deg] 5.143 5.143 0
0.114 - -
0.0197 0.0197 0
[N] 220.8 220.8 0
0.019 0.019 0
0.0224 0.0224 0
[N] 252.32 251.9 0
0.1079 0.1084 0
[N] 1016.63 1021.01 0
/ 0.538 0.538 0
/ 0.523 0.523 0

48
Table 16. Input- and output data for boattail-, interference- and spillage drag at operating
point 5.

Input Output Output Margin of

Variable
GasTurb12 GasTurb12 Matlab error %
π 3.14159265
M 1.6
Alt [m] 10668
[kPa] 23.8423
F4JI 1
NSI 0
SECI 0
[J /(kg K)] 287.058
[ft2] 6.82
GM (inlet) 1.4
[K] 288.15
[kPa] 101.325
m2 [kg/s] 64.7861
RF 1
Dmax [in] 38.6
S [in] 53.8
noe 1
[in] 23.4
Thick_base [m] 0
GM (nozzle) 1.3
Mdj 1.3
A8 [m^2] 0.444455
A9 [m^2] 0.755
Q [Pa] 42725.3
Pt8P0=Pt9P0 7.39056
T2 [K] 330.92
P2 [Pa] 101.391
FN [kN] 59.87
β [deg] -0.00082 -0.00082 0
0 0 0
0 0 0
[N] 0 0 0
0.0063 0.0063 0
0.0045 0.0045 0
[N] 144.773 144.87 0
0.1515 0.1514 0
[N] 4101.36 4098.51 0
/ 0.5922 0.5923 0
/ 0.5675 0.5675 0

49
Table 17. Input- and output data for boattail-, interference- and spillage drag at operating
point 6.

Input Output Output Margin of

Variable
GasTurb12 GasTurb12 Matlab error %
π 3,14159265
M 2
Alt [m] 10668
[kPa] 23,8423
F4JI 1
NSI 0
SECI 0
[J /(kg K)] 287,058
[ft2] 6,82
GM (inlet) 1,4
[K] 288,15
[kPa] 101,325
m2 [kg/s] 91,7841
RF 1
Dmax [in] 38,6
S [in] 53,8
noe 1
[in] 23,4
Thick_base [m] 0
GM (nozzle) 1,3
Mdj 1,3
A8 [m^2] 0,47917
A9 [m^2] 0,755
Q [Pa] 66758,3
Pt8P0=Pt9P0 9,6425
T2 [K] 393,744
P2 [Pa] 186,683
FN [kN] 78,99
β [deg] -0,00083 -0,00082 0
0 0 0
0 0 0
[N] 0 0 0
0,0046 0,0046 0
0,0023 0,0023 0
[N] 115,254 114,04 -1
0,0484 0,0481 -1
[N] 2047,29 2033,7 -1
/ 0,6726 0,673 0
/ 0,6429 0,6429 0

50
The boattail drag results from the Matlab code generated corresponding data with
GasTurb12, except two small margin errors due to rounding in operating point 1 and 3
(Table 12 and Table 14, green cells). In the second operation point, the interference drag
Matlab code had a margin of error of 3% (Table 13, blue cells). The explanation is the
interpolation of data in the existing Matlab code received from earlier work and the
interpolation in GasTurb12 for the table shown in Appendix 6. The data in the two tables
are not entirely identical and may vary by one thousand. Operation point 6 (Table 17) has
a small rounding error for the interference drag. The only margin of error for the spillage
drag coefficient was at the last operating point (Table 17, red cells), which is due to the
manually reading of the spillage drag coefficient table.

The verification of the results from the base drag equations implemented into GasTurb12
is presented in Table 18.
Table 18. Input- and output data from GasTurb12 for base drag together with the maximum
theoretical base drag at all six operating points.

Variable GasTurb12 data

Operating
1 2 3 4 5 6
point
e 2.718282 2.718282 2.718282 2.718282 2.718282 2.718282
M 0.4 0.6 0.8 1.2 1.6 2
Alt [m] 4572 7620 10668 13716 10668 10668
Dmax [in] 38.6 38.6 38.6 38.6 38.6 38.6
S [in] 53.8 53.8 53.8 53.8 53.8 53.8
noe 1 1 1 1 1 1
[in] 23.4 23.4 23.4 23.4 23.4 23.4
Thick_base
0.025 0.025 0.025 0.025 0.025 0.025
[m]
GM (nozzle) 1.3 1.3 1.3 1.3 1.3 1.3
Mdj 1.3 1.3 1.3 1.3 1.3 1.3
[kPa] 57.1819 37.6 23.842 14.7477 23.8423 23.8423
Q [Pa] 6404.38 9475.42 10681.3 14865.6 42725.3 66758.3
A9 [m^2]] 0.429852 0.453303 0.493370 0.599183 0.755 0.755
Pt9 [kPa] 160.628 126.68 98.08 81.7887 176.21 229.899
Tt9 [K] 1938.96 1939.09 1938.74 1939.11 1938.55 1938.23
Pt9s [kPa] - - - 14.7471 23.9844 36.014
B - - - 1.2863 1.3768 1.9426
- - - 0.1016 0.0759 0.0563

- - - 0.9058 0.7111 0.7056

0.189 0.1045 0.0684 0.00941 0.01816 0.0118
[N] 913.844 747.512 551.592 105.591 585.78 597.025
FN [kN] 49.87 39.55 31.29 26.79 59.87 78.99
0.018 0.019 0.018 0.004 0.01 0.008
Max. 9.6154 4.2735 2.404 1.0684 0.601 0.3846

51
The result shows that none of the base drag coefficients for the six operating points
exceed the theoretical maximum values with the base pressure set to vacuum. The drag
have a small impact on the total net thrust, which is a reliable result as the dimensions of
base are small. The implemented equations in GasTurb12 are therefore adequate for
estimation of base drag.

For cases with the base- and boattail drag where , the dimensions of the
maximum- and base diameters have to be large relative the exit diameter. Investigating a
case with defined boattail angles, the boattail length, , have to be extended to a higher
value as well. This type of aircraft engines are not of high interest for GKN and will
therefore not be treated more than just implementation of the equations. A quick
simulation in GasTurb12 with large dimensions showed that the boattail drag effects are
small and have almost no impact on the total net thrust. The resulting base drag was
negative and should be invested more if this type of engines is used.

4.3. D RAG COEFFICIENT PLOT S

The drag coefficients from the validation- and verification step for each installation effect
plotted as scatter plots at all six operating points are presented as follow.

4.3.1. S P I L L AG E DR AG
The inlet mass flow ratio and the inlet capture mass flow ratio are both plotted in the
same scatter plot, showing the conformity in the validation- and verification step. For
accurate conformity, the blue and red markers should be plotted equal and the green
marker should equal the purple.

Inlet mass flow- and inlet capture mass flow ratio

Boeing_A0Ac Matlab_A0Ac Matlab_A0iAc Boeing_A0iAc
1
0,9
0,8
0,7
0,6
Ratio

0,5
0,4
0,3
0,2
0,1
0
0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 1,8 2
Mach

Figure 22. The inlet mass flow ratio (A0Ac) plotted together with the inlet capture mass flow
ratio (A0iAc), both output from Matlab and Boeing in the validation step.

52
 Inlet mass flow- and inlet capture mass flow ratio
Matlab_A0Ac Gasturb_A0Ac Gasturb_A0iAc Matlab_A0iAc
0,9
0,8
0,7
0,6
0,5
Ratio

0,4
0,3
0,2
0,1
0
0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 1,8 2
Mach

Figure 23. The inlet mass flow ratio (A0Ac) plotted together with the inlet capture mass flow
ratio (A0iAc), both output from Matlab and GasTurb12 in the verification step.

All output data for the inlet mass flow ratio in the Matlab code are consistent with the
experimental data from [9] and gives the same values at all operating points. The same
applies for the verification step with the output data from Matlab and Gasturb12. The
inlet mass flow ratios also have consistent data for both the validation- and the
verification step. The result shows that the implemented equations for the two ratios in
GasTurb12 correspond to the experimental data in the Boeing material.

A summary from chapter 4.1 and 4.2 of the resulting spillage drag coefficients from the
validation- and verification step are plotted in Figure 24 and Figure 25 as scatter plots.

53
 Spillage drag
Boeing Matlab
0,12

0,1

0,08
CDspill

0,06

0,04

0,02

0
0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 1,8 2 2,2
Mach

Figure 24. The spillage drag coefficient at the four simulated operation points for the first
validation step.

Spillage drag
Matlab Gasturb
0,16

0,14

0,12

0,1
CDspill

0,08

0,06

0,04

0,02

0
0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 1,8 2 2,2
Mach

Figure 25. The spillage drag coefficient at the four simulated operation points for the second
validation step.

4.3.2. B O AT T AI L DR A G
A summary from chapter 4.1 and 4.2 of the resulting boattail drag coefficients from the
first and second step of validation are plotted in Figure 26 and Figure 27 as scatter plots.

54
 Boattail drag
Boeing Matlab
0,035

0,03

0,025

0,02
CDβ

0,015

0,01

0,005

0
0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 1,8 2 2,2
Mach

Figure 26. The boattail drag coefficient at the six simulated operation points for the first
validation step.

Boattail drag
Matlab Gasturb
0,04

0,035

0,03

0,025
CDβ

0,02

0,015

0,01

0,005

0
0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 1,8 2 2,2
Mach

Figure 27. The boattail drag coefficient at the six simulated operation points for the second
validation step.

55
4.3.3. I N T E R F E R EN C E DRAG
A summary from chapter 4.1 and 4.2 of the resulting interference drag coefficients from
the first and second step of validation are plotted in Figure 28 and Figure 29 as scatter
plots.

Interference drag
Boeing Matlab
0,05
0,045
0,04
0,035
0,03
CDIN

0,025
0,02
0,015
0,01
0,005
0
0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 1,8 2 2,2
Mach

Figure 28. The interference drag coefficient at the five simulated operation points for the first
validation step.

Interference drag
Matlab Gasturb
0,035

0,03

0,025

0,02
CDIN

0,015

0,01

0,005

0
0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 1,8 2 2,2
Mach

Figure 29. The interference drag coefficient at the five simulated operation points for the second
validation step.

56
4.3.4. B A S E DRAG
A summary from chapter 4.2 of the resulting base drag coefficients from the verification
between GasTurb12 and the theoretical maximum base drag coefficient are plotted in
Figure 30 as a scatter plot.

Base drag
GasTurb Maximum base coeff.
10
9
8
7
6
CDB

5
4
3
2
1
0
0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 1,8 2
Mach

Figure 30. The base drag coefficient calculated in GasTurb12 (blue) and the theoretical
maximum drag coefficient calculated from the Matlab code (red).

4.3.5. I N F LU EN C E O N T H E T O T A L N ET T HR US T
The results from the simulation of the engine with approximately the same characteristics
as a J79-GE-8 engine and an F-4J Phantom II inlet in GasTurb12 are plotted in Figure 31.

Alt = 11000m
Spillage drag Boattail Base drag_0.01m Interference drag
10%

8%
∑FD/FN

6%

4%

2%

0%
0,4 0,5 0,6 0,7 0,8 0,9 1 1,1 1,2 1,3 1,4 1,5 1,6 1,7 1,8 1,9 2
Mach

Figure 31. The total installation drag influence caused by spillage- boattail- base- and
interference drag of a J79-GE-8 engine at altitude 11000 meters.

57
The maximum total influence of the four installation effects is about 8% for this type of
aircraft flying at an altitude of 11000 meters. The most influent installation effect is the
spillage drag and is most critical at Mach 1.5 to 1.7. After Mach 1.7 the mass flow
required for the engine is starting to approach its design value, generating lower drag
coefficient. At the designed Mach number there will still be some spillage drag, as this
decreases the risk of having the shock entering the inlet. The boattail drag in Figure 31 is
zero for Mach higher than 1.5 as this engine is designed for having maximum open
nozzle exit at . Regarding base drag, the influence of the total net thrust is
almost constant, with a smaller decrease at supersonic flight. What is important to take
into account is the designed Mach number that in this case is estimated and not defined in
the Boeing material. This is because J79-GE-8 engines normally do not have any base
area and the design Mach number is therefore not a required input variable.

The cause of the peak at for the boattail drag and the interference drag is due to
the different flow and shocks created in the transonic area. The calculations within this
range need more profound knowledge and additional parameters need to be considered.

58
5. C ONCLUSIONS
The method that has been developed in this study is one version of bookkeeping the
installation effects in supersonic aircrafts and is applicable for engines with similar inlet-
and afterbody configuration as the GE-J79-8. The method also works for aircrafts with
normal shock inlets or oblique shock inlets, flying at Mach numbers up to 2.5.

The implementation of the equations into the gas turbine cycle program GasTurb12 will
provide GKN with the ability of forecasting the impact of installation effects on engine
performance. The outcome from the verification step show that the Matlab code results
correspond accurately to the outputs in GasTurb12. Together with the results from the
validation step, the adoption of a correct implementation into GasTurb12 could be stated.
Simulating a defined aircraft engine model in Gasturb12 using the implemented method,
the results will represent the magnitudes of the external drags as expected. This enables
further studies about how these drags can be minimized and thereby increasing the total
net thrust.

The limitations of simulating a supersonic aircraft using this method are primarily when
at low subsonic speeds. The spillage drag for the F-4J inlet is not defined for operation
points at Mach , as the inlet is defied not to spill below this value. This makes it
hard to validate the spillage drag at these lower Mach numbers. Also, the base drag
calculations were not validated against any empirical data, but gave reasonable values of
the drag compared to the calculated theoretical maximum base drag together with results
for the entire afterbody drag. The installation effect is therefore included in the system.

In conclusion, the method works well and can be used as a tool for predicting installation
effects for supersonic aircrafts as a first order estimation.

5.1. F UTURE WORK

Further improvement of the implemented method is to consider the air pre-compression
generated from the nose and forebody of the aircraft. The entering air of the implemented
inlets is in this study adopted to be free stream air, which have not been affected by any
further parts of the aircraft. Another improvement is to implement data from other inlets
than the three defined, obtaining a wider range of applications.

Ensuring that the equations are valid for similar engines should be performed by
simulating other types of aircrafts with existing experimental data. The implemented
method supports further studies of how to reduce the investigated installation effects and
minimize the external drag for a defined aircraft-engine-system. The impact from the
different inlet- and afterbody geometries should be examined to find the best possible
configuration for minimal drag.

59
R EFERENCES

[1] D. Jonsson and P. L. Hansen, Jetmotorlära: Flygvapnet, Stockholm: Allmänna förlaget,

1973.

[2] H. Alvebrink, "Examensarbete installationsmotstånd för överljudsplan, VOLS: 10216303,"

GKN Aerospace, Trollhättan, 2015.

[3] B. R. Munson, D. F. Young, T. H. Okiishi and W. W. Huebsch, Fundamentals of Fluid

Mechanics, 7 ed., New York, US: John Wiley & Sons Inc, 2012.

[4] E. E. Covert, Thrust and Drag: Its Prediction and Verification, New York: American
Institute of Aeronautics and Austronauties, Inc., 1985.

[5] R. Avellán, "Metod för installationsmotstånd för supersoniska flygplan, VOLS:

10154732," Volvo Aero, Trollhättan, 2012.

[6] P. P. Walsh and P. Fletcher, Gas Turbine Performance 2nd ed., Oxford, UK: Blackwell
Science Ltd, 2004.

[7] W. H. Ball, "Propulsion System Installation Corrections, Vol. I -Engineers Manual," The
Boeing Company, Dayton, USA, 1972.

[8] W. H. Ball, "Propulsion System Installation Corrections, Vol. II -Programmers Manual,"

The Boeing Company, Dayton, USA, 1972.

[9] W. H. Ball, "Propulsion System Installation Corrections, Vol. III -Sample Cases," The
Boeing Company, Dayton, USA, 1972.

[10] W. H. Ball, "Propulsion System Installation Corrections, Vol. IV -Bookkeeping Definition

-Data Correlations," The Boeing Company, Dayton, USA, 1972.

[11] S. F. Hoerner, Aerodynamic Drag, Dayton, USA: The Otterbein Press, 1951.

[12] H. McDonald and P. Hughest, "A Correlation of High Subsonic Afterbody Drag in the
Presence of a Propulsive Jet or Support Sting," Journal of Aircraft, vol. 2, no. 3, pp. 202-
207, 1965.

[13] M. W. Petersen and G. C. Tamplin, "Experimental Review of Transonic Spillage Drag of

Rectangular Inlets," North American Aviation Inc. /Los Angeles Divisions, Ohio, 1972.

[14] G. R. Center, "Beginner's Guide to Aerodunamic (BGA)," NASA- Glenn Research Center,
05 May 2015. [Online]. Available: https://www.grc.nasa.gov/www/k-
12/airplane/short.html. [Accessed September to December 2015].

[15] Y. A. Cengel and J. M. Cimbala, Fluid Mechanics: Fundamentals and Applications, 3rd
ed., Singapore: McGraw-Hill Education, 2014.

60
[16] J. D. Anderson, Modern Compressible Flow With Historical Perspective, 3rd ed., New
York, USA: McGraw-Hill Education, 2004.

[17] Y. A. Cengel and B. A. Michael, Thermodynamics: An Engineering Approach, 7th ed.,

Boston, US: McGraw-Hill Higher Education, 2011.

[18] M. Daly, Jane's Aero-Engines, 29th ed., Surrey, UK: IHS, 2011.

[19] J. D. Mattingly, Elements of Propulsion; Gas Turbines and Rockets, AIAA Education
Series, 2006.

[20] H. Cohen, G. F. C. Rogers and H. I. H. Saravanamuttoo, Gas Turbine Theory 3rd ed., New
York, USA: Longman Scientific & Technical, 1972.

[21] W. J. Devenport, "Java Applets for Engineering Education," National Science Foundation,
[Online]. Available: http://www.engapplets.vt.edu/. [Accessed 11 2015].

[22] J. Seddon and E. L. Goldsmith, Intake Aerodynamics, London, UK: Collins Professional
and Technical Books, 1985.

[23] B. Prof. Roy and A. M. Prof. Pradeep, Jet Aircraft Propulsion, Lecture 28, Bombay:
Department of Aerospace Engineering, IIT Bombay.

[24] o. t. A. D. E. T. C. The Design Guide Subcommittee, AIAA Aerospace Design Engineers

Guide, 5th ed., Reston, USA: American Institute of Aeronautics and Astronautics , 2007.

[25] J. D. Anderson Jr, Introduction to Flight, 3rd ed., New York, US: McGraw-Hill Book
Company, 1989.

[26] E. J. Kowalski, "A Computer Code for Estimating Installed Performance of Aircraft Gas
Turbine Engines, Vol. 1 -Final report," National Aeronautics and Space Administration ,
Seattle, USA, 1979.

[27] W. B. I. Compton and J. F. Runckel, "Jet Effects on the Boattail Axial Force of Conical
Afterbodies at Subsonic and Transonic Speeds," National Aeronautics and Space
Administration, Washington D.C, USA, 1970.

[28] S. F. Hoerner, "Base Drag and Thick Trailing Edges," Journal of the Aeronautical
Sciences, vol. 17, no. 10, pp. 622-628, 1950.

[29] M. S. Group, "Guide to In-Flight Thrust Measurement of Turbojets and Fan Engines,"
AGARDograph, vol. 237, 1979.

[30] K. Stenholm, "Creating model of turbojet engine GE J79," GKN Aerospace, Trollhättan,
2015.

[31] J. Kurzke, GaTurb12 -Design and Off-Design Performance of Gas Turbines, Germany:
GasTurb GmbH, 2013.

61
[32] J. Kurzke, GasTurb Details 5 -An Utility for GasTurb 11, Germany: J. Kurzke, 2007.

[33] E. J. Kowalski and R. A. Atkins Jr., "A Computer Code for Estimating Installed
Performance of Aircraft Gas Turbine Engines, Vol. III -Library of Inlet/Nozzle
Configurations and Performance Maps," National Aeronautics and Spcae Administration ,
Seattle, USA, 1979.

[34] J. Roskam, Airplane Design, Part VI: Preliminary Calculation of Aerodynamic, Thrust and
Power Characteristics, Lawrence, USA: Design, Analysis and Resarch Corporation , 2004.

[35] S. G. MIDAP, "Guide to In-Flight Thrust Measurement of Turbojets and Fan Engines,"
AGARD, London, UK, 1979.

62
A PPENDIX 1
The project will be implemented through different phases to facilitate the comprehension
of the work and provide a simple step-by-step concept. The different phases and their
meaning are presented below in relative order.

 The Pre-Study Phase (PSP) - The first phase where literature studies will be
accomplished to obtain a wider understanding of the engine performance area.
 The Prioritized Development Phase (PDP) - Here the equations, formulas and
tables for the priority installation effects will be configured.
 The Prioritized Implementation Phase (PIP) - The equations, etc. for the priority
effects will be implemented into the engine performance simulation tool Gasturb 12.
 The Prioritized Validation Phase (PVP) – The results from the previous phase
will be validated through modeling data for an existing aircraft.

If time is provided:

 The Secondary Development Phase (RDP) - Here the equations, formulas and
tables for the residual installation effects will be configured.
 The Secondary Implementation Phase (RIP) - The equations, etc. for the residual
effects will be implemented into the engine performance simulation tool Gasturb 12.
 The Secondary Validation Phase (RVP) - The results from the previous phase
will be validated through modeling data for an existing aircraft.

 Report writing (RW) – This phase will proceed during the entire project.
 Final Presentation (FP) – A final presentation of the project for GKN Aerospace
and Luleå University of Technology.

A Gantt-chart for the different phases and their running time is presented in

Appendix_figure 1. The Gantt-chart for the project.

A PPENDIX 2
Boattail drag coefficient at nozzle pressure ratio 2.5 for Mach numbers M≤0.95.
A PPENDIX 3
Boattail drag correction for nozzle pressure ratios other than 2.5.
A PPENDIX 4
Annular supersonic base pressure correlation.
A PPENDIX 5
Reference base pressure ratio.
A PPENDIX 6
Interference drag coefficient for Neng number of engines.
A PPENDIX 7
Code with the implemented equations in GasTurb12 for calculating the spillage drag.
Units]
SI=1
Kelvin=1
Rankine=1

[Composed Values]
cp_val1=F4J_inlet=1
cp_val2=Normal_shock_inlet=0
cp_val3=Supersonic_external_comp_inlet=0
cp_val4=pi=3.14129265
cp_val5=R_air=287.058
cp_val6=GM=1.4
cp_val7=GM1=GM-1
cp_val8=Ac=6.820*0.09290304
cp_val9=Pstd=101.325*1000
cp_val10=Tstd=288.15
cp_val11=m2corr=82.37
cp_val12=RF=0.824
cp_val13=error=-99999
cp_val14=Q_dyn=0.5*1.4*(Pamb)*1000*(XM)^2
cp_val15=Coeff2=(GM/(1+((GM1/2)*(XM)^2))^((GM+1)/(GM1)))^0.5
cp_val16=Ainf=(((R_air*Tstd)^0.5*m2corr)/((XM)*Pstd*Coeff2))*
RF
cp_val17=A0Ac=(Ainf)/(Ac)
cp_val18=Undefined
cp_val19=F4j_bleed=f18[x=cp_val17;par=XM]
cp_val20=Table_BN=0
cp_val21=Ex_bleed=f17[x=cp_val17;par=XM]
cp_val22=F4JI=F4J_inlet
cp_val23=NSI=Normal_shock_inlet
cp_val24=SECI=Supersonic_external_comp_inlet
cp_val25=ABAc=(F4JI)*(F4j_bleed)+(NSI)*(Table_BN)+(SECI)*(Ex_
bleed)
cp_val26=A0IAc=A0Ac+ABAc
cp_val27=Undefined
cp_val28=Cd_f4j_spill=f15[x=cp_val26;par=XM]
cp_val29=Cd_norm_spill=f16[x=cp_val26;par=XM]
cp_val30=Cd_ex_spill=f14[x=cp_val26;par=XM]
cp_val31=Undefined
cp_val32=lowMach=IFaGTbxELSEy[0.8;(XM);error;(Cd_f4j_spill)]
cp_val33=lowbleed_f4j=IFaGTbxELSEy[0.45;(A0Ac);(error);(lowMa
ch)]
cp_val34=highbleed_f4j=IFaGTbxELSEy[(A0Ac);0.65;(error);(lowb
leed_f4j)]
cp_val35=f4j_Cd=IFaGTbxELSEy[(Cd_f4j_spill);0.45;(error);(hig
hbleed_f4j)]
cp_val36=norm_Cd=IFaGTbxELSEy[(Cd_norm_spill);0.7;(error);(Cd
_norm_spill)]
cp_val37=bleed_ex=IFaGTbxELSEy[(ABAc);0.08;(error);(Cd_ex_spi
ll)]
cp_val38=ex_Cd=IFaGTbxELSEy[(Cd_ex_spill);0.4;(error);(bleed_
ex)]
cp_val39=Undefined
cp_val40=Cdspill=(F4JI)*(f4j_Cd)+(NSI)*(norm_Cd)+(SECI)*(ex_C
d)
cp_val41=Undefined
cp_val42=Fdspill=Cdspill*Q_dyn*Ac
A PPENDIX 8
Spillage drag coefficients for the three different aircraft inlets.

Appendix_figure 2. Spillage drag coefficient for an F-4J inlet configuration.

Appendix_figure 3. Spillage drag coefficient for an normal shock inlet configuration.
Appendix_figure 4. Spillage drag coefficient for an mach 2.5 external compression inlet
configuration.
A PPENDIX 9
Bleed mass flow ratio for the F-4J inlet and the 2.5 Mach external compression
inlet.

Appendix_figure 5. Bleed mass flow ratio for an F-4J inlet configuration.

Appendix_figure 6. Bleed mass flow ratio for a 2.5 Mach external compression inlet
configuration.
A PPENDIX 10
Code with the implemented equations in GasTurb12 for calculating the boattail drag.
[Units]
SI=1
Kelvin=1
Rankine=1

[Composed Values]
cp_val1=Area=A9
cp_val2=Factor=IFaGTbxELSEy[Area;0.754;0;1]
cp_val3=Factor
cp_val4=Gap=Factor*((F_A9)/(FN))
cp_val5=AaB=IFaGTbxELSEy[Factor;0.5;1;0]
cp_val6=Sap=(Gap+1)*AaB
cp_val7=Hej=(-1)*(Factor-1)*((A9)/0.755)
cp_val8=Iter=Sap+Hej
cp_val9=(F_A9)/(FN)+1
cp_val10=Undefined
cp_val18=Undefined
cp_val19=Undefined
cp_val20=pi=3.14159265359
cp_val21=Dmax=38.6*0.0254
cp_val22=Lbeta=23.4*0.0254
cp_val23=Space=53.8*0.0254
cp_val24=Q_dyn=0.5*1.4*(Pamb)*1000*(XM)^2
cp_val25=Thick_base=0.0
cp_val26=Undefined
cp_val27=Undefined
cp_val28=A8
cp_val29=Pamb
cp_val30=A9
cp_val31=Dexit=((4*(A9))/pi)^0.5
cp_val32=Amax=(pi/4)*Dmax^2
cp_val33=Db=Dexit+2*Thick_base
cp_val34=beta=arctan((Dmax-Db)/(2*Lbeta))
cp_val35=beta_deg=beta*(180/pi)
cp_val36=Pt9Pf=(St9_P)/(Pamb)
cp_val37=Undefined
cp_val38=Dj2DbDm=(Dexit^2)/(Db*Dmax)
cp_val39=DbDm=Db/Dmax
cp_val40=Undefined
cp_val41=D9Dm=(1-((Dexit/Dmax)^2))
cp_val42=Cdbet1=((1.4*tan((beta)))/((XM)^1.53))*D9Dm
cp_val43=Undefined
cp_val44=CorrBT=f2[x=P8qamb;par=cp_val35]
cp_val45=BTCd25=f1[x=cp_val35;par=XM]
cp_val46=Over025=(BTCd25+CorrBT)
cp_val47=Undefined
cp_val48=BTgrad=f9[x=cp_val35;par=none]
cp_val49=Cpbetaz=f10[x=cp_val39;par=cp_val35]
cp_val50=Grad=f6[x=cp_val36;par=St9_T]
cp_val51=Zerojet=f7[x=cp_val36;par=cp_val35]
cp_val52=DeltaCpb=Grad*Dj2DbDm+Zerojet
cp_val53=Under025=Cpbetaz+(BTgrad*DeltaCpb)
cp_val54=Cdbet0=IFaGTbxELSEy[(Dj2DbDm);0.25;Over025;Under025]
cp_val55=Undefined
cp_val56=Cdbeta=IFaGTbxELSEy[(XM);1;(Cdbet1);(Cdbet0)]
cp_val57=FDbeta=Cdbeta*Q_dyn*Amax
A PPENDIX 11
Code with the implemented equations in GasTurb12 for calculating the base drag.

[Units]
SI=1
Kelvin=1
Rankine=1

[Composed Values]
cp_val1=Area=A9
cp_val2=Factor=IFaGTbxELSEy[Area;0.754;0;1]
cp_val3=Factor
cp_val4=Gap=Factor*((F_A9)/(FN))
cp_val5=AaB=IFaGTbxELSEy[Factor;0.5;1;0]
cp_val6=Sap=(Gap+1)*AaB
cp_val7=Hej=(-1)*(Factor-1)*((A9)/0.755)
cp_val8=Iter=Sap+Hej
cp_val9=(F_A9)/(FN)+1
cp_val10=Undefined
cp_val11=Undefined
cp_val12=Undefined
cp_val13=Undefined
cp_val14=Undefined
cp_val15=Undefined
cp_val16=Undefined
cp_val17=Undefined
cp_val18=Undefined
cp_val19=Undefined
cp_val20=pi=3.14159265359
cp_val21=e_bas=2.71828182846
cp_val22=Dmax=38.6*0.0254
cp_val23=Lbeta=23.4*0.0254
cp_val24=Space=53.8*0.0254
cp_val25=Qdyn=0.5*1.4*(Pamb)*1000*(XM)^2
cp_val26=Thick_base=0.025
cp_val27=GM=1.3
cp_val28=Mdj=1.3
cp_val29=St9_Ps
cp_val30=St9_P
cp_val31=St9_T
cp_val32=Pamb
cp_val33=A9
cp_val34=Dexit=((4*(A9))/pi)^0.5
cp_val35=Amax=(pi/4)*(Dmax^2)
cp_val36=beta=arctan((Dmax-Dexit)/(2*Lbeta))
cp_val37=beta_deg=beta*(180/pi)
cp_val38=Dj=Dexit
cp_val39=Db=Dexit+2*Thick_base
cp_val40=DbDj=Db/Dj
cp_val41=DbDm=Db/Dmax
cp_val42=PtPf=(St9_P)/(Pamb)
cp_val43=GM1=GM-1
cp_val44=PtPj=(1+((GM1)/2)*Mdj^2)^(GM/GM1)
cp_val45=Mj=((2/GM1)*((PtPf)^((GM1/GM))-1))^0.5
cp_val46=DD=((Db^2)-(Dj^2))/(Dmax^2)
cp_val47=DbDj2=(Db/Dj)^2
cp_val48=DmDj2=(Dmax/Dj)^2
cp_val49=PjePf=((St9_Ps)/(Pamb))
cp_val50=Dj2DbDm=(Dj^2)/(Db*Dmax)
cp_val51=Undefined
cp_val52=Coeff1=Mdj^((0.2*(DbDj)-0.15)*Mdj)
cp_val53=CC=(((0.37*(XM)+0.62)*Coeff1)/(DbDj2))*PjePf
cp_val54=BB=f4[x=cp_val53;par=none]
cp_val55=Klarge=((XM)*(DbDj)+Mj)/(((DbDj)-1)^0.5)
cp_val56=Ksmall=(XM)
cp_val57=KK=IFaGTbxELSEy[(PtPf);1;(Klarge);(Ksmall)]
cp_val58=PbPfK=f3[x=cp_val57;par=none]
cp_val59=PbPfM=(PbPfK*BB)/((XM)*e_bas^(0.815-1.15*ln(KK)))
cp_val60=Undefined
cp_val61=Cpbz=f5[x=cp_val41;par=cp_val37]
cp_val62=Grad=f6[x=cp_val42;par=St9_T]
cp_val63=Zerojet=f7[x=cp_val42;par=cp_val37]
cp_val64=DeltaCpb=Grad*Dj2DbDm+Zerojet
cp_val65=Less025=(-DD)*(Cpbz+DeltaCpb)
cp_val66=Undefined
cp_val67=TablePbPf=f8[x=cp_val42;par=none]
cp_val68=More025=((2*(DbDj2-1)*(1-TablePbPf))/((GM*(XM)^2)*DmDj2))
cp_val69=Undefined
cp_val70=Cdb1=((2*(DbDj2-1)*(1-PbPfM))/((GM*(XM)^2)*DmDj2))
cp_val71=Cdb0=IFaGTbxELSEy[(Dj2DbDm);0.25;More025;Less025]
cp_val72=Undefined
cp_val73=Cdbase=IFaGTbxELSEy[(XM);1;(Cdb1);(Cdb0)]
cp_val74=FDbase=Cdbase*Qdyn*Amax
A PPENDIX 12
Matlab code for calculating spillage drag.
clc
clear all
%Spillage drag Matlab code

%Choose inlet type by putting equal to 1, while the rest is set to

0.
F4J_inlet=1;
Normal_shock_inlet=0;
Supersonic_external_comp_inlet=0;

%Insert data
XM=2.0;
R_air=287.058; %Specific air constant for air
[J/(kg*K)]
GM=1.40; %Inlet heat capacity ratio
Ac=6.820*0.09290304; %Acpture area [m^2]
Pamb=23.2946; %Ambient pressure (free stream
pressure) [kPa]
Tstd=288.15; %Standard temperature [K]
Pstd=101.325*1000; %Standard pressure [a]
RF=0.904; %Presssure recovery factor of the
inlet
m2corr=63.407; %Corrected airflow at Station 2 [kg/s]
%m2=91.7841; %Airflow at Station 2 (Gasturb vs.
Matlab)
%T2=393.744; %Tot. temp. at Station 2 (Gasturb vs.
Matlab)
%P2=186.683; %Tot. press. at Station 2 (Gasturb
vs. Matlab)

%Equations
%m2corr=(m2*sqrt(T2/Tstd)/((P2*1000)/Pstd)) %(Gasturb vs.
Matlab)
GM1=GM-1;
Q_dyn=0.5*1.4*(Pamb)*1000*(XM)^2;
Coeff2=sqrt(GM/(1+((GM1/2)*(XM)^2))^((GM+1)/(GM1)));
Ainf=((sqrt(R_air*Tstd)*m2corr)/((XM)*Pstd*Coeff2))*RF
A0Ac=(Ainf)/(Ac);

% Bleed mass flow ratio

XM
A0Ac
Quest='What is A0BLC/Ac?'; %Get data from Figure 41 in Vol 3
(F4j)
ABAc=input(Quest);

%Calculating inlet capture mass flow ratio:

A0IAc=A0Ac+ABAc;

%Spillage drag coefficient for the F-4J inlet

if F4J_inlet>=1
XM
A0IAc
Quest='What is Cd_f4j_spill?'; %Get data from Figure 39
in Vol 3.
Cd_f4j_spill=input(Quest);
 %Sending error messase if extrapolted value is used
if Cd_f4j_spill>0.45
f4j_Cd=NaN
elseif 0.45>A0Ac
f4j_Cd=NaN
elseif A0Ac>0.65 && XM<2.0
f4j_Cd=NaN
elseif XM<0.8
f4j_Cd=NaN;
else f4j_Cd=Cd_f4j_spill;
end
norm_Cd=0;
ex_Cd=0;
end

%Spillage drag coefficient for the normal shock inlet

if Normal_shock_inlet>=1
XM
A0IAc
Quest='Cd_normal_spill?'; %Get data from Figure 81 in Vol
4.
Cd_norm_spill=input(Quest);

%Sending error messase if extrapolted value is used

if Cd_norm_spill>0.7
norm_Cd=NaN;

elseif XM<0.6
norm_Cd=NaN;
else
norm_Cd=Cd_norm_spill;
end
f4j_Cd=0;
ex_Cd=0;
end

%Spillage drag coefficient for the normal shock inlet

if Supersonic_external_comp_inlet>1
XM
A0IAc
Quest='Cd_external_spill?'; %Get data from Figure 145 in
Vol 4.
Cd_ex_spill=input(Quest);

%Sending error messase if extrapolted value is used

if Cd_ex_spill>0.4
ex_Cd=NaN;
elseif ABAc>0.08
ex_Cd=NaN;
elseif XM<0.6
ex_Cd=NaN;
else
ex_Cd=Cd_ex_spill
end
norm_Cd=0;
f4j_Cd=0;
end
Cdspill=(f4j_Cd)+(norm_Cd)+(ex_Cd)
Fdspill=Cdspill*Q_dyn*Ac
A PPENDIX 13
Matlab code for calculating boattail drag.
%Boattail drag script in SI-units
clear all
clc
%Insert data
XM=0.4; %Mach number
Dmax=38.6*0.0254; %Maximum diameter [m]
Lbeta=23.4*0.0254; %Boattail length [m]
S=53.8*0.0254; %Nozzle spacing [m]
%A8=693.72*0.0254^2; %Throat area [m^2]
%A9=719.614*0.0254^2; %Exit area [m^2]
A8=0.407431; %[m^2] Gasturb vs.
Matlab
A9=0.429852; %[m^2] Gasturb vs.
Matlab
Pamb=57.1819; %Ambient pressure [Pa]
NPR=2.809; %Nozzle pressure ratio
Thick_base=0.0; %Base thickness [m]
Q_dyn=6404.38; %Dynamic pressure
[Pa]

%Equations
%Q_dyn=0.5*1.4*(Pamb)*1000*(XM)^2; %Dynamic pressure
(matlab vs boe)
Amax=(pi*Dmax^2)/4; %Maximum area [m^2]
D9=sqrt(4*A9/pi); %Exit diameter [m]
Db=D9+2*Thick_base; %Base diameter [m]
beta=atan(( Dmax-Db)/(2*Lbeta)); %Boattail angle [rad]
beta_deg=(beta*180)/pi %Boattail angle [deg]

Dj2DbDm=(D9^2)/(Db*Dmax)
DbDm=Db/Dmax;

%Supersonic flight, M>=1

if XM>=1
Cdbeta=((1.4*tan(beta))/(XM^1.53))*(1-((D9/Dmax)^2))
DFbeta=Cdbeta*Q_dyn*Amax %Boattail drag [N]

elseif XM<0.9
if Dj2DbDm<=0.25
DbDm
beta_deg
Quest='What is Cdbetaz?'; %Get data from Figure 5 in
McDonald
Cdbetaz=input(Quest);

beta_deg
Quest='What is BTgrad?'; %Get data from Figure 7 in
McDonald
BTgrad=input(Quest);

Quest='What is Grad?'; %Get data from Figure 10 in

McDonald
Grad=input(Quest);

NPR
 beta_deg
Quest='What is Zerojet?'; %Get data from Figure 10 in
McDonald
Zerojet=input(Quest);

DeltaCpb=Grad*Dj2DbDm+Zerojet

Cdbeta=Cdbetaz+(BTgrad*DeltaCpb)
DFbeta=Cdbeta*Q_dyn*Amax

else Dj2DbDm>0.25
beta_deg
Quest='What is Cdbeta_in_table_low_mach?';
Cdbetatable=input(Quest);

%Correction for NPR > 2.5

if (beta_deg >= 0.0 && beta_deg <= 6.0)
betacorr = 0.0;
elseif (beta_deg > 6.0 && beta_deg <= 12.0)
if (NPR >= 2.0 && NPR <= 8.0)
%betacorr = 0.000014163207*NPR^6 -
0.000414028046*NPR^5 + 0.004781165346*NPR^4 - 0.027363378516*NPR^3
+ 0.078264922366*NPR^2 - 0.102978319861*NPR+ 0.046305633214;
betacorr= 0.00000668851937946080*NPR^6 -
0.00019127760711692200*NPR^5 + 0.00210968496962697000*NPR^4 -
0.01091116230599670000*NPR^3 + 0.02355113910771290000*NPR^2 -
0.00999233804356577000*NPR - 0.01665662115521540000;
elseif (NPR > 8.0)
NPR = 8.0;
%betacorr = 0.000014163207*NPR^6 -
0.000414028046*NPR^5 + 0.004781165346*NPR^4 - 0.027363378516*NPR^3
+ 0.078264922366*NPR^2 - 0.102978319861*NPR+ 0.046305633214;
betacorr=NaN
end
elseif (beta_deg > 12.0)
if (NPR >= 2.0 && NPR <= 8.0)
%betacorr = 0.000009378612*NPR^6 -
0.000275451321*NPR^5 + 0.003190312464*NPR^4 - 0.01816913699*NPR^3
+ 0.050125973403*NPR^2 - 0.057388440799*NPR + 0.013904579871;
betacorr= 0.00000599124504052639*NPR^6 -
0.00017086868938080900*NPR^5 + 0.00188724418426841000*NPR^4 -
0.00981615455624801000*NPR^3 + 0,02119204649239710000*NPR^2 -
0.00616911504728309000*NPR - 0.02220674123446150000
elseif (NPR > 8.0)
NPR = 8.0;
%betacorr = 0.000009378612*NPR^6 -
0.000275451321*NPR^5 + 0.003190312464*NPR^4 - 0.01816913699*NPR^3
+ 0.050125973403*NPR^2 - 0.057388440799*NPR + 0.013904579871;
betacorr=NaN
end
end

Cdbeta=Cdbetatable+betacorr
DFbeta=Cdbeta*Q_dyn*Amax
end
%Subsonic flight, XM<1
else 0.9<=XM<1
if Dj2DbDm<=0.25
DbDm
beta_deg
 Quest='What is Cdbetaz?'; %Get data from Figure 5 in
McDonald
Cdbetaz=input(Quest);

beta_deg
Quest='What is Gradbeta?'; %Get data from Figure 10 in
McDonald
Gradbeta=input(Quest);

Cdbeta09=Cdbetaz+Gradbeta*DeltaCpb;

else Dj2DbDm>0.25
beta_deg
Quest='What is Cdbeta_in_table_trans?';
Cdbetatable=input(Quest);

NPR
Quest='What is betacorr?';
betacorr=input(Quest);

Cdbeta09=Cdbetatable+betacorr;
end
Cdbeta1=(1.4*tan(beta))*(1-((D9/Dmax)^2));

Cdbetatrans=Cdbeta09+((XM-0.9)/0.1)*(Cdbeta1-Cdbeta09)
DFbeta=Cdbetatrans*Q_dyn*Amax
end
A PPENDIX 14
Matlab code for calculating base drag.

clear all
clc
format long
%Script for base drag at subsonic and supersonic flight.

%Insert engine data

XM=0.8;
e_bas=2.71828182846;
Dmax=100*0.0254;
Lbeta=100*0.0254;
Space=53.8*0.0254;
Thick_base=0.1;
GM=1.3;
Mdj=1.3;
Pamb=23.842;
Qdyn=10681.3;
A9=0.493370;
St9_P=98.08;
St9_T=1938.74;
St9_Ps=23.9844;

%Variables created from engine data

Dexit=((4*(A9))/pi)^0.5;
Amax=(pi/4)*Dmax^2;
beta=atan((Dmax-Dexit)/(2*Lbeta));
beta_deg=beta*(180/pi);
Dj=Dexit;
Db=Dexit+2*Thick_base;
DbDj=Db/Dj;
DbDm=Db/Dmax;
PtPf=(St9_P)/(Pamb);
GM1=GM-1;
P9Pj=(1+((GM1)/2)*Mdj^2)^(GM/GM1);
Mj=((2/GM1)*(((PtPf)^(GM1/GM))-1))^0.5;
DD=((Db^2)-(Dj^2))/(Dmax^2);
DbDj2=(Db/Dj)^2;
DmDj2=(Dmax/Dj)^2;
PjePf=(St9_Ps)/(Pamb);
Dj2DbDm=(Dj^2)/(Db*Dmax);

%Base drag for supersonic flight

if XM>=1
Coeff1=Mdj^((0.2*(DbDj)-0.15)*Mdj);
CC=(((0.37*(XM)+0.62)*Coeff1)/(DbDj2))*PjePf;

if (CC >= 0.08 && CC < 0.922)

BB = 4.548297655701*CC^4 - 14.490577165021*CC^3 + ...
17.700266056446*CC^2 - 9.113164438123*CC +
2.667029906043
elseif (CC > 0.922 && CC <= 3.0)
BB = 0.733515691006*CC + 0.564530150955
else
BB = 0.733515691006*CC + 0.564530150955
end

Klarge=((XM)*(DbDj)+Mj)/(((DbDj)-1)^0.5);
 Ksmall=XM;

if PtPf>=1
KK=Klarge;
elseif PtPf<1
KK=Ksmall;
end

PbPfK=0.917092698296*KK^-0.907452120588
PbPfM=(PbPfK*BB)/((XM)*e_bas^(0.815-(1.15*log(KK))))

Cdbase=((2*(DbDj2-1)*(1-PbPfM))/((GM*(XM)^2)*DmDj2))

%Base drag for subsonic flight

else XM<1
if Dj2DbDm<=0.25
DbDm
beta_deg
Quest='What is Cpbz?'; %Get data from Figure 5 in McDonald
Cpbz=input(Quest);

PtPf
St9_T
Quest='What is Gradbase?'; %Get data from Figure 10 in
McDonald
Gradbase=input(Quest);

PtPf
beta_deg
Quest='What is Zerojet?'; %Get data from Figure 9 in
McDonald
Zerojet=input(Quest);

DeltaCpb=Gradbase*Dj2DbDm+Zerojet;
Cdbase=(-DD)*(Cpbz+DeltaCpb)

else Dj2DbDm>0.25

PtPf_vec=[1.5:0.5:4.5];

PbPf_data=[0.95;
0.86;
0.79;
0.73;
0.69;
0.67;
0.65];

PbPfM=interp1(PtPf_vec,PbPf_data,PtPf);

Cdbase=((2*(DbDj2-1)*(1-PbPfM))/((GM*(XM)^2)*DmDj2))
end

end

FDbase=Cdbase*Qdyn*Amax
A PPENDIX 15
Code with the implemented equations in GasTurb12 for calculating the interference drag.
[Units]
SI=1
Kelvin=1
Rankine=1

[Composed Values]
cp_val1=Area=A9
cp_val2=Factor=IFaGTbxELSEy[Area;0.754;0;1]
cp_val3=Factor
cp_val4=Gap=Factor*((F_A9)/(FN))
cp_val5=AaB=IFaGTbxELSEy[Factor;0.5;1;0]
cp_val6=Sap=(Gap+1)*AaB
cp_val7=Hej=(-1)*(Factor-1)*((A9)/0.755)
cp_val8=Iter=Sap+Hej
cp_val9=(F_A9)/(FN)+1
cp_val10=Undefined
cp_val18=Undefined
cp_val19=Undefined
cp_val20=pi=3.14159265
cp_val21=Dmax=38.6*0.0254
cp_val22=Space=53.8*0.0254
cp_val23=A8
cp_val24=Pamb
cp_val25=noe=2
cp_val26=Qdyn=0.5*1.4*(Pamb)*1000*(XM)^2
cp_val27=Amax=(pi/4)*(Dmax^2)
cp_val28=Dexit=((4/pi)*(A9))^0.5
cp_val29=SDexit=(Space)/(Dexit)
cp_val30=Undefined
cp_val31=intcd=f13[x=cp_val29;par=XM]
cp_val32=CdNIA8=IFaGTbxELSEy[(SDexit);1;(intcd);0]
cp_val33=FDNIAP=CdNIA8*A8*Pamb*1000*4.34
cp_val34=FDin=FDNIAP*((noe-1)/noe)
cp_val35=Cdin=FDin/(Qdyn*Amax)

Matlab code for calculating interference drag.

%Interference drag script in SI-units
%Table data
clear all
clc
XM=2.0; %Mach number
Dmax=38.6*0.0254; %Maximun diameter of the nozzle
Space=53.8*0.0254; %Spacing between the centre of two
engines
A8=835.164*0.0254^2; %Nozzle throat area
A9=1170.214*0.0254^2; %Nozzle exit area
%A8=0.400486; %Nozzle throat area (Matlab vs.
Gasturb)
%A9=0.453303; %Nozzle exit area (Matlab vs.
Gasturb)
Pamb=23.2946; %Ambient pressure
noe=2; %Number of engines
Q_dyn=66565.6; %Dynamic pressure, used for Matlab vs.
Boeing
%Equations
%Q_dyn=0.5*1.4*(Pamb)*1000*(XM)^2 %Used for Matlab vs.
GasTurb
Amax=(pi/4)*(Dmax^2);
Dexit=((4/pi)*(A9))^0.5;
SDexit=(Space)/(Dexit);

if SDexit>=1;
XM
SDexit
Quest='What is CdNIA8?'; %Get data from Figure 44 in Vol.4
CdNIA8=input(Quest);
else
CdNIA8=0;
end

DNIAP=CdNIA8*A8*Pamb*1000*4.34;
FDin=DNIAP*((noe-1)/noe)
Cdin=FDin/(Q_dyn*Amax)
A PPENDIX 16
The name and numbers of the implemented tables in GasTurb12.

Name in Table Found in

Variable in report
GasTurb12 number Appendix
for the F-4J inlet F4j_bleed 18 9

for the 2.5 Mach external

Ex_bleed 17 9
compression inlet
for the F-4J inlet Cd_F4j_spill 15 8
for the normal shock inlet Cd_norm_spill 16 8
for the 2.5 Mach external
Cd_ex_spill 14 8
compression inlet
CorrBT 2 3
BTCd25 1 2

BTgrad 9 Ref.

Cpbetaz 10 Ref.
Grad 6 Ref.

Zerojet 7 Ref.
Correlation parameter, B BB 4 4
PbPfK 3 5
Cpbz 5 Ref.
TablePbPf 8 Page.
intcd 13 6
A PPENDIX 17
Existing Matlab code for calculating the boattail drag coefficient at nozzle pressure ratio
2.5 for Mach numbers M≤0.95.
function [ out ] = CD_BoatTail(Mach,BoatAngle,QD9Dmax,NPR)
%UNTITLED4
Mach_vec = [0.4 0.5 0.6 0.7 0.8 0.85 0.9 0.925 0.95];
BoatAngle_vec = [0 2 4 6 7 8 10 12 14 16 18 20];

CD_Boat_Data = [0.001740056 0.001741796 0.001743536 0.001745276

0.001747016 0.001826627 0.001888198 0.00189 0.002367852;
0.002063000 0.002063000 0.002063000 0.002063000
0.002063000 0.004406801 0.006697329 0.009863992 0.012848086;
0.005096000 0.005096000 0.005096000 0.005096000
0.005096000 0.00752284 0.011238244 0.017311772 0.021878185;
0.011530000 0.011530000 0.011530000 0.011530000
0.011530000 0.015078831 0.019474464 0.027854651 0.040824728;
0.015950000 0.015950000 0.015950000 0.015950000
0.015950000 0.018860000 0.023720000 0.032940000 0.051380000;
0.023290000 0.023290000 0.023290000 0.023290000
0.023290000 0.024748849 0.030344904 0.041196567 0.065871497;
0.029730000 0.029730000 0.029730000 0.029730000
0.029730000 0.032822274 0.041165988 0.055297961 0.089676356;
0.033240000 0.033240000 0.033240000 0.033240000
0.033240000 0.037226783 0.049863981 0.067975543 0.108266237;
0.034330000 0.034330000 0.034330000 0.034330000
0.034330000 0.03872901 0.056037531 0.077927936 0.1231702;
0.036150000 0.036150000 0.036150000 0.036150000
0.036150000 0.040312886 0.060850427 0.085187226 0.138790604;
0.040398 0.040398 0.040398 0.040398
0.040398 0.044735655 0.065754567 0.090996391 0.138929395;
0.045971534 0.045971534 0.045971534 0.045971534
0.045971534 0.050261558 0.070043143 0.097112617 0.139068186];
%BT angle check:
if (BoatAngle > 20.0)
BoatAngle = 20.0
End

%Subsonic:
if (Mach >= 0.4 && Mach <= 0.95)
CD_BT_LS =
interp2(Mach_vec,BoatAngle_vec,CD_Boat_Data,Mach,BoatAngle);
CD_BT_NPR25 = CD_BT_LS;
elseif (Mach >= 0.0)
Mach = 0.4
CD_BT_LS =
interp2(Mach_vec,BoatAngle_vec,CD_Boat_Data,Mach,BoatAngle);
CD_BT_NPR25 = CD_BT_LS;
%Supersonic
elseif (Mach >= 1.0)
CD_BT_HS = (1.4*tand(BoatAngle)/(Mach^1.53))*(1.0-QD9Dmax^2);
CD_BT_NPR25 = CD_BT_HS;
%Narrow Transonic, 0.95 < Mach < 1.0
elseif (Mach > 0.95 && Mach < 1.0)
CD_BT_M095 =
interp2(Mach_vec,BoatAngle_vec,CD_Boat_Data,0.95,BoatAngle);
CD_BT_M100 = (1.4*tand(BoatAngle)/(1.0^1.53))*(1.0-QD9Dmax^2);
CD_BT_INT = interp1([0.95 1.0],[CD_BT_M095 CD_BT_M100],Mach);
CD_BT_NPR25 = CD_BT_INT;
end
A PPENDIX 18
Matlab code for calculating the theoretical maximum base drag.
clear all
clc
%Script for valdidation of supersonic base drag if vacuum was
created

%Input data
%XM=0.2:0.1:2;
XM=2.0;
GM=1.3;
A9=0.755;
Dmax=38.6*0.0254;
Thick_base=0.025;
Qdyn=66758.3;

Dexit=((4*(A9))/pi)^0.5;
Db=Dexit+2*Thick_base;
Abase=pi*((Db^2)/4);
Ab=Abase-A9;
Cdbmax=2./(GM*(XM).^2)

%Maximun base drag at supersonic speeds

Dbasemax=Cdbmax*Qdyn*Ab

y=Dbasemax;
x=XM;
plot(x, y)
title('Maximun base drag');
ylabel('Dbasemax');
xlim([0.2 2]);
ylim([0 2000]);
grid on
A PPENDIX 19
Existing Matlab code for calculating the interference drag coefficient.
function [CD_IF]= CD_IF(Mach,QSD9)
Mach_vec = [0.55 0.75 0.85 0.95 1.0 1.2 1.5 1.8 2.4];
S_Dexit = [1.0:0.1:3.6];

CD_IF_Data=[0.004756455 0.004756832 0.005463852 0.006116279

0.006509353 0.019031 0.007119 0.004649 0.004649;
0.006374032 0.006404292 0.007201526 0.008764435
0.011068436 0.019031 0.007119 0.004649 0.004649;
0.008146711 0.008111185 0.009072273 0.011416978
0.015581049 0.019031 0.007119 0.004649 0.004649;
0.010074492 0.010073753 0.010957301 0.014073908
0.020047193 0.019031 0.007119 0.004649 0.004649;
0.012157376 0.012156834 0.012980607 0.016735224
0.024466866 0.019031 0.007119 0.004649 0.004649;
0.009143468 0.013608718 0.01516368 0.019400928
0.028840069 0.019031 0.007119 0.004649 0.004649;
0.005923203 0.01004848 0.017080209 0.022071019
0.033166802 0.019031 0.007119 0.004649 0.004649;
0.004030000 0.007307566 0.01701684 0.024745497
0.037447065 0.019031 0.007119 0.004649 0.004649;
0.002827000 0.005385974 0.013597244 0.027424362
0.041680858 0.019031 0.007119 0.004649 0.004649;
0.002219681 0.004815201 0.010837793 0.030107613
0.045868181 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.00915401 0.032795252
0.050009033 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.008297742 0.035487278
0.054103416 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.00794644 0.037400000
0.058151329 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007805369 0.033034553
0.062152772 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007709812 0.029916967
0.057590582 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007700 0.027632100
0.040811138 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007700 0.026063385
0.036913963 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007700 0.025094254
0.03390659 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007700 0.02460814
0.031821167 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007700 0.0245801
0.030415218 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007700 0.024552
0.029487882 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007700 0.024552
0.028879913 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007700 0.024552
0.028473678 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007700 0.024552
0.02819316 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007700 0.024552
0.028003954 0.019031 0.007119 0.004649 0.004649;
0.0020339 0.004649 0.007700 0.024552
0.02791327 0.019031 0.007119 0.004649 0.004649;
 0.0020339 0.004649 0.007700 0.024552
0.027969935 0.019031 0.007119 0.004649 0.004649];

if QSD9 > 3.6

QSD9 = 3.6
end

if Mach > 2.4

Mach = 2.4;
CD_IF = interp2(Mach_vec,S_Dexit,CD_IF_Data,Mach,QSD9);
elseif Mach < 0.55
Mach = 0.55;
CD_IF = interp2(Mach_vec,S_Dexit,CD_IF_Data,Mach,QSD9);
else
CD_IF = interp2(Mach_vec,S_Dexit,CD_IF_Data,Mach,QSD9);
end
end