Benjamin Gal-Or

Vectored Propulsion,
Supermaneuverability
and
Robot Aircraft
With 190 Figures

Springer-Verlag
New York Berlin Heidelberg
London Paris Tokyo Hong Kong
Professor Benjamin Gal-Or
Technion-Israel Institute of Technology
Department of Aeronautical Engineering
32000 Haifa
Israel

Library of Congress Cataloging-in-Publication Data

Gal-Or, Benjamin.
Vectored propulsion, supermaneuverability, and robot aircraft / by Benjamin Gal-Or.
P. em.
I. Airplanes - Jet propulsion.
2. Drone aircraft.
3. Jet planes, Military.
4. stealth aircraft. I. Title.
TL685.3.G23 1989
629. I 34'353-dc20 89-21797

© 1990 by Benjamin Gal-Or

All rights reserved. This work may not be translated or copied in whole or in part without the
written permission of the copyright holder, except for brief excerpts in connection with reviews
or scholarly analysis. Use in connection with any form of information storage and retrieval, elec-
tronic adaptation, computer software, or by similar or dissimilar methodology now known or
hereafter developed is forbidden.
The use of general descriptive names, trade names, trademarks, etc. in this publication, even if
the former are not especially identified, is not to be taken as a sign that such names, as under-
stood by the Trade Marks and Merchandise Marks Act, may accordingly be used freely by
anyone.

Typeset, Printed, and Bound by Keterpress, Jerusalem, Israel

ISBN-13: 978-0-387-97161-2 e-ISBN-13: 978-1-4613-8961-3

DOl: 10.1007/978-1-4613-8961-3
Dedicated to my family;
Leah, Amir and Gillad,
who, by their respective professions,
and by their encouragement,
have given me invaluable help
in preparing this volume,

and to the members of the

Jet Propulsion Laboratory Team,
Azar, B.;
Cohen, Z.;
Dekel, E.;
Friedman, E.;
Golijow, E.;
Mashiach, E.;
Rasputnis, A.;
Soreq, I.;
Turgemann M.;
and
Vorobeichik, S.,

to whom I am greatly indebted.

Mistakes!, Who may pretend to comprehend them? And for
the unknown ones, please forgive me."
Based on Psalms 19:12

"Our whole problem is to make the mistakes as fast as

possible ... "
John Archibald Wheeler

"All our knowledge grows only through the correcting of our

mistakes."
The Philosopher of Science
Sir Karl R. Popper

(Conjectures and Refutations; The Growth of Scientific

Knowledge, K.R. Popper, Routledge and Kegan Paul, Lon-
don, 1972).
 Foreword

With the advent of digital flight control and digital engine control technologies,
airframe and propulsion systems designers can now consider a much higher degree of
coupling between the aircraft and its engine than ever before to achieve revolutionary
new capabilities for high performance aircraft.
Professor Gal-Or's landmark book is, in my opinion, the most complete and
definitive treatment to date of the complex aerodynamic and control integration
associated with vectored-thrust propulsion, aircraft agility enhanced by multi-axis
thrust vectoring and reaction control systems, and provides valuable insight into
applications for piloted and robotic aircraft.
Although the military utility of enhanced agility using vectored propulsion is not
fully understood, it is clear that a better experimental and analytical technology base is
necessary to evaluate the concepts in realistic air combat scenarios. Professor Gal-Or's
book is a valuable reference in this regard, although the reader should be cautioned
that a text on such a rapidly-developing field may need to be updated as new
information becomes available. Professor Gal~Or addresses the key questions in the
Introduction, which are the subject of active research and development programs in
all major aerospace establishments.
The above comments should not be interpreted as an endorsement ofthis book, or
its contents, by the United States Air Force.

Dr. G. KEITH RICHEY

Technical Director of the Wright Research and Development Center,
Wright-Patterson Air Force Base,
Ohio, USA
TABLE OF CONTENTS

G. Keith Richey: Foreword ............................................. 5

Preface ............................................................... 9
Acknowledgements ..................................................... 15
Glossary and Notation.................................................. 17

Introduction ........................................................... 21

The main problems of thrust-vectored maneuverability • Vectored Aircraft:

Breaking the "Stall Barrier" • The New Standard Technology • Civil Appli-
cations............................................................. 25
Soviet and Western Concepts • Research tools • Cold propulsion • The
Fundamental Definitions of Pure Vectored Aircraft (PVA) .............. 29
Pure Sideslip Maneuvers (PSM) • Combat Applications • . . . . . . . . . . . . . . . 30
Partially-Vectored Aircraft ........................................... 33
Agility, Supermaneuverability and Supercontrollability ................. 34
Load factors ........................................................ 38
"Energy" Vs. "Angles" Fights • Time Lags • ........................... 40
The Overall Combat Picture ......................................... 43
Minimum Time Turns • Low Observability • Stealth RPVs ............ 49
The Time Gaps ..................................................... 50
New Educational Shifts.............................................. 52
Concluding Remarks 53

Lecture I:
Fundamental Concepts Revisited and Redefined ......................... 56

New Missions and Modified Design Trends ........................... 57

Early Conceptual Design Limitations ................................. 58
Internal Thrust Vectoring ............................................ 59
Engine Nozzle Redefined ............................................ 62
Performance Comparisons ........................................... 69
Subsonic Yaw-Pitch-Roll Thrust Vectoring ............................ 71
vii

Lecture II:
Vectored Aircraft and Supermaneuverability 78
Yaw-Pitch Vectoring and Supercirculation ............... ... ........... 78
Pure Jetborne Flight Control ......................................... 81
STOL or V/STOL Vectored Aircraft .................................. 82
Dimensionless Numbers for Pure Vectored Aircraft .................... 83
Emergency Landing ................................................. 88
Air-to-Air Operations ...... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
Air-to-Ground Operations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
Propulsion and Supermaneuverability ................................. 92
Maneuver Analysis..... ..... ....................... ... ... ........ ... 95
Controllability Limits. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
High-Alpha Flight Envelopes ......................................... 102
The Non-Availability of a "Vectored Inlet" ............................ 104

Lecture III:
The Matrix of Unknown Variables of Vectored Aircraft 105
Reservations and Precautions ........................................ 106
Technology Limits .................................................. 107
Cooling Limits .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 109
Nozzle Aspect Ratio Limits .......................................... 114
Aerodynamic Effects ................................................ 115
Nozzle Coefficients Revisited ........................................ 125
Thrust Vectoring Range . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 128
Induced drag and Supercirculaton .................................... 130

Lecture IV:
Vectored Aircraft as R&D Tools, or as Super-Agile Robotic Flying Systems 133
Cruise Missiles, RPV s, etc. ........................................... 134
Expanded Missions Vs. R&D tools ................................... 136
Cost-Effectiveness of Flying Vectored RPV s ........................... 139
Synergetic Studies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 145
Master Plan and New Programs ...................................... 151
Scaling Consideratons ............................................... 153

Lecture V:
Partially Vectored Aircraft . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 155
Canard-Configured Vectored Aircraft ................................. 156
Upgrading extant Aircraft to Become Vectored Fighters - The Vectored F-15
Fighter ........................................................... 163
 viii

Upgrading F-18 and F-16 Fighters.................................... 167

The Search for Design Philosophy in the Post-ATF Era .. . . . . . . . . . . . . . .. 168

Lecture VI:
The Pros and Cons of Internal Thrust Vectoring ........................ 177
Is ETV Effective in the RaNPAS-PST Domain? ....................... 178
The Propulsion System of the Vectored X-31, X-29, etc. ................ 181
Stability and Flight Control .......................................... 184
Power-Lift Aircraft .................................................. 186

Appendix A:
A Brief Historical Survey. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 188
Appendix B:
Powerplant Technology Limits ........................................ 194
Beyond the Year 2000/RCC/STOVL .................................. 194
Current-Technology Limits and Trends................................ 203
Whole-System Engineering Approach ................................. 209
European Fighter Engines ............................................ 209
New Fighter Engines: The ATF and Other New Engines ................ 210
The MiG-29 and Su-27 Propulsion Systems ........................... 211
Variable Cycle, Turbine Bypass, and Supersonic-Fan Engines ........... 212

Appendix C:
Nozzle Performance (Data-Base-l]
Appendix D
Nozzle Synergetic Effects (Data-Base-2]
Appendix E:
Temperature Distribution Test Facility. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 236
Appendix F:
Limiting Engine-Inlet Envelopes and IFPC . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 240
PST Inlet Design.................................................... 243
IFPC Transient Definitions 244
Appendix G:
On Some Other Vectored/Stealth Aircraft Consideratons ................ 251
Appendix H:
Thought-Provoking and Thought-Depressing Quotations ................. 251
References ............................................................ 259
Index................................................................. 271
 PREFACE

The hypotheses with which (a new development) starts,

become steadily more abstract and remote from experi-
ence ... Meanwhile the train of thought leading from the axi-
oms to the empirical facts of verifiable consequences gets
longer and more subtle.
Albert Einstein

This book is designed to fill a professional vacuum in the new field of advanced,
high-a, vectored-stealth aircraft. It contains a core of unclassified knowledge that,
according to the latest trends in military aviation, should be mastered by all aeronauti-
cal engineers and students, as well as by advanced pilots and various R&D people.
The subject matter of this book has never before been investigated, or presented, as
a unified field of study, partially because it covers an entirely new field of study, and
partially because specialized fragments of this unified field are scattered throughout
the literature on specific problems.
As a result, different engineers have approached the new integrated systems from
widely varying, and sometimes misleading viewpoints, employing disjointed concepts
to what should be a unified R&D-design methodology.
Consequently, this book departs from traditional, isolated texts on aerodynamics,
thermodynamics, propulsion, electrodynamics, radar, control, materials, flight
mechanics and design, in its emphasis on the interconnectedness of up-dated informa-
tion in vectored, agile, STOL, or V/STOL, manned or unmanned aircraft.
Thus, in trying to develop a new, integrated methodology, we refrain from repeating
various, well-known, isolated, mathematical-computational analyses, and, instead,
stress the physical and performance consequences from the end-user point-ofview.
For the sake of unification and simplicity, the physico-performance options are first
introduced in as simple a manner as possible, using many illustrated graphs and
diagrams.
Part of the text is an expanded version of three university courses on "Jet-Engine
Design", "Fluid Dynamics, Heat Transfer and the Performance of Jet Engines", and
"Technology of Jet Engines", given at this institute over a period of years. The other
part, and perhaps the major part of the text, is the exposition of the new technology
with every argument that can lead to testable consequences in the laboratory and in
flight testing.
9
10 Preface

Slmul.Uons .r Comb.l Err.cllvenes.

( CombIned Compuler/Veclored RPV, SImulations)

Fig. I. Proper use of vectored RPVs emerges as a highly-effective research tool, replacing some of the
more traditional roles of wind-tunnel simulations. The actual, feedback mechanisms employed in this
type of highly-integrated design methodology, are marked here as a preliminary definition of integrated
vectored aircraft.
This scheme demonstrates that an advanced propulsion/airframe system, as a whole, cannot be opti-
mized by improving each variable separately, and, then, superimposing the results on each other to get
the whole optimized system. It is, thus, imperative to strive for the overall design, i.e., to strive for an
Preface 11

Another viewpoint on the need of this technology is presented below.

* * *
All advanced fighter aircraft will be based on the new technology called vectored
propulsion.
This technology is the key element in enabling fighters to survive and win in battles
both beyond and within visual range of the enemy. It also provides the best design ele-
ments for highly agile, stealth, supermaneuverability aircraft and Remote Piloted
Vehicles (RPVs) of the future, including some vectored helicopters.
The first vectored flights in "the open history of aviation" were conducted in 1987
by our team, using various vectored RPVs. These flights were the final results of un pub-
lished work conducted during the past eight years.
The fundamental concepts and the design methodology required to understand this
field of engineering are detailed and explained here, using original drawings and fig-
ures which have never been published before.
Most of the experimental work conducted in. this laboratory on vectored engines,
and in flying "pure" vectored RPVs, is classified as the proprietary of our financial
sources, and is, consequently, unpublishable. However, the fundamental concepts and
the various methodologies described in this book stand out as a generic, non-
sponsored, academic investigation.
Another way to describe the aim and scope of this book is to examine the integrated
design ofvectored (manned and unmanned) aircraft as depicted in the following figure
on R&D cycles-feedbacks between theory and lab-flight testing and analysis. Thisfig-
ure, somewhat loose/y, defines the work cycles of many R&D and industrial develop-
ment programs, as expected for the 90s and beyond. It also defines the actual
work-cycles practiced in this laboratory over the past few years.

* * *
A few preliminary concepts for vectored aircraft R&D may be introduced now:
• Vectored RPVs emerge as highly-effective R&D tools, replacing some of the more
traditional roles of wind-tunnel testing and computational simulations.
• Mission definition and synthesis is a time-variant, testing-variant matrix, that
should not be specified prematurely as a "final cause".
• Design for low signatures should be of prime importance in conceptual and prelimi-
nary designs.
• Propulsion-laboratory testing of integrated wing-nozzle-engine-inlet systems is of
prime importance.

interconnected design philosophy in which the various variables are complementary within the frame-
work ofa reliable, whole-system-engineering methodology. However, such methodology does not exist
yet. Instead, one may use the feedback methodology depicted here and detailed in the text.
12 Preface

• The fundamental concepts of physics, and especially of aero-electro-thermo-

dynamics, must constantly serve as general guidelines for the emergence and testing
of novel ideas .
• As the complexity of highly-integrated vectored aircraft increases, the gap between
extant mathematical-computational tools and actual design needs widens, thereby
making the use of the former more and more ineffective, and, at times, even mis-
leading (except for the analysis of specific components). Consequently, during this
early evolutionary stage of the technology, one should avoid the introduction of pre-
mature mathematical tools into advanced design courses of highly integrated sys-
tems. Instead, physico-engineering insight, combined with proper integration
methods between preliminary test results and tentative mission definitions should
be given the first priority.

* * *
A Few Notes For Pilots, Engineers and Teachers.
This Volume contains a core of basic material that both the advanced researcher and
the desinger or test-pilot should master; namely, redefined and newly defined/unda-
mental concepts, components, variables, experimental data, basic unit operations,
and partially integrated systems.
It supplies two tracks through the subjects.
The first track focuses on general, "interdisciplinary" material, which is so ordered
as to help those readers who are either unfamiliar with, or uninterested in detailed for-
mulations and empirical data of a specific "discipline", but would rather consider the
general arguments involed without interruption. Hence, a substantial part of this Vol-
ume is available as a second track.
All "track two" material is marked with e, or included in the Appendices. It includes
the more specific data base and various experimental results of earlier designs, and of
some unorthodox subsystems. It also includes a few notes on the latest evolutionary
trends in this technology, including brief summaries on the expected PST/Vectoring
technology limits in the 90s and beyond.
The selection of the material for the appendices is intended to provide a greater
degree of competence in a few sub-systems. Nevertheless, researchers, teachers,
designers and pilots, especially those in need offurther enrichment material on experi-
mental data and limiting testing conditions, or on some highly specific designs and
missions that have recently gained special importance, are invited to read the original
references to those subjects that interest them the most.
All "track two" material can be understood by undergraduate students who have
taken at least one-semester courses on (compressible) fluid dynamics and on jet
propulsion.
With a few exceptions, "track one" material can be easily understood by readers
who have studied only "Freshman" physics, and, in addition, have participated in
basic pilot training courses.
Preface 13

Volume II, to be published later, is to stress the fundamental design methodology

and the flight-propulsion control of fully integrated stealth-vectored systems. Conse-
quently, it is designed to concentrate on new "figures of merit" and on integrated
propulsionljlight control theories and components of pure vectored aircraft.
Revisiting the fundamentals of electro-aero-thermodynamics is included in Volume
II, for proper component integration in pure vectored/stealth aircraft design, in terms
of new materials, control systems, inlets and powerplants, while simultaneously maxi-
mizing supermaneuverability, STOL, or V/STOL performance. All in all, it shows that
vectored aircraft design is not separable into disciplines.

* * *
The Lectures presented in Volume I may be adapted for a one-semester, undergrad-
uate course on "Design", or on "Advanced Flight/Jet Propulsion". Volume II, when
available, is designed to become useful in a combined, two-semester graduate/
undergraduate course, or as an independent, one-semester, graduate course.

* * *
It was in 1980 that I started to lecture on these topics.
Ever since I have been working on this book, which reproduces, if not the letter, then
the spirit of my lectures.
No series oflectures was ever a set-piece; they have remained in a state offlux, until
now, when the final writing and printing has "frozen" them. This "freezing" is very
apprehensive to me, but it cannot be helped. My hope is to complete the 2nd volume
within the next few years. Meanwhile, I hope that at least some of the readers will
"thaw out" the figures and printed lines, and give them greater dynamical force
through their own critical attention.

* * *
Indeed, it was the work on the present volume that gave me the opportunity to
develop more fully than in my earlier research projects, the general principles of this
field of study, as well as the integrated methodology of laboratorylflight testings of
post-stall, purely-vectored, research RPVs. Joining the test results obtained from the
jet-propulsion laboratory with the design and flight-testing results of vectored RPVs,
seemed, at times, like the generation of a vast jigsaw puzzle from scattered and (appar-
ently) unrelated pieces of test results. But the eventual emergence of new regularities
and unconventional design principles, which, at the beginning, I had not even sus-
pected to be part of the overall research picture, was often a rewarding surprise. I can
only invite the interested reader to join me in further work on some of these excursions
along trails leading to new fields of inquiry.

* * *
14 Preface

Verifiable evidence cannot yet be admitted to a number of topics presented in this

book. Therefore, if engineers (and pilots) are counted as "moderate skeptics", as no
doubt they should, they must adopt the practice of "suspense of judgment". By this we
do not mean radical empiricism, nor the denial of any rational addition to the
advancement of high-performance aircraft design.
 ACKNOWLEDGEMENTS

Part of the unclassified research work published here, as well as the integrated method-
ology of laboratorylflight testing developed by the author, has been financially spon-
sored by
Teledyne CAE,
The General Electric Co.,
The United States Air Force,
General Dynamics.
I wish to thank my research sponsors as well as the editors, publishers and boards of
the various periodicals, reports and books referenced in this volume for allowing me,
directly and indirectly, by policy or word, to include figures and pictures appearing in
this volume, as quoted below and in the respective references. Indeed, I am especially
indebted to AIAA, ASME, SAE, AHS, IDR, NASA, USAF-WPAFB, US Army, IDF,
IAF, Journal of Propulsion, International Journal of Turbo and Jet Engines, Journal
of Power and Gas Turbines, Journal of Aircraft, the General Electric Co., Pratt &
Whitney, R.R., Teledyne, Allison Gas Turbines of GM, Garrett, Bet Shemesh
Engines, Turbomeca, MBB, MTU, General Dynamics, Northrop, Lockheed, IAI,
NPT, TAT, IGTA, KHD, Micro-Turbo, Donaldson Co., Avco-Lycoming and the
McGraw Hill Co.
Most important, the evolution of this important field of aviation has been made
possible by the significant contributions of the many individuals whose works have
been selected in the preparation process of this book. A few of these pioneers are men-
tioned by name in Appendix A. However, our greatest gratitude is with the long list of
contributors as reflected in the References, and in the text itself.
In revising the text, I have been helped, in various ways, by Doron Bar Annann, Dr.
A. Rasputnis, Erez Friedman, Sara Voroveitchik, Mike Turgemann, Dan Erez,
Eduardo Golijow, Eran Liron, Prof. Yoram Tambour, and, indirectly, by General
David Ivry, the General Director ofIsraeJ's MoD, Dr. W. B. Herbst ofMBB, Mr. T. P.
McAtee, Mr. Jerry Murff, Mr. Trey Durham, Mr. Fran Ketter, Mr. Tom Barret, and
Mr. C. Porcher of General Dynamics, Mr. F. Ehric, Mr. Ed Rogala and Mr. Don
Dunbar of General Electric, Mr. E. Benstein of Teledyne, Mr. E. Albert and Mr. S.
Levy ofI.A.I., and Mr. D. Bowers and Mr. W. Lindsay, of the US Air Force. Indirectly,
I have also been helped by my students, who frequently contributed comments and
15
16 Acknowledgement

questions, especially by those who had selected to participate in my courses "Fluid

Dynamics, Heat Transfer, and the Performance of Jet Engines", "Design ofPropul-
sion Systems", "Jet-Engines Technology" and "Jet Engines-I". Erez Friedman has
given me invaluable, first-class professional help in constructing the first Pure Vec-
tored RPV s, and the first vectored F-15/F-16 RPV s, and in flight testing them in the
spring of 1987, and in the summer of 1989, respectively. Similar acknowledgement
and gratitude go to Mike Turgemann, who constructed our very first vectored RPV
and flight tested it in April and May 1987.
My deepest gratitude goes to my two sons, Amir Gal-Or and Gillad Gal-Or, who,
during our long, enthusiastic, professional discussions, have contributed many ideas,
and have also encouraged me to pursue these ideas in laboratory and RPV s flight
testing.
Finally I want to express my greatest indebtedness to the Publisher, Springer Verlag,
and especially to the Engineering Editors Dr. Z.Ruder, and A. Von Hagen, and to Zvi
Weller, Yaacov Polak, Ami Green and Hannah Lipschitz ofKeter Press, Jerusalem.

B. Gal-Or
Haifa, 1989
 GLOSSARY AND NOTATION

A - Dimension defined in Fig. III-lO

Ae - A9 - Exhaust nozzle exit area.
At - As - Exhaust nozzle throat area.
AC - Aerodynamic Center.
ACC-D - Acceleration-Deceleration dynamic response of Trim-IFPC-engine.
ACM - Advanced Cruise Missiles.
ADEN - A thrust-vectoring nozzle iIIustated in Appendix A.
AoA - Angle of Attack
As - Surface area of a wing with jet vectoring issuing from the airfoil's trailing edge.
AB - After-burning
ATA - Advanced Tactical Aircraft (A-12).
ATB - Advanced Tactical Bomber (B-2).
ATF - Advanced tactical fighter.
AVCM - Advanced Vectoring Cruise Missile.
b - wing span.
B - Two-dimensional nozzle height
BLS - Boundary Layer SeparationlThickness at each propulsion station.
BPR - By-Pass Ratio.
C - Defined in Fig. III-lO. An upper bar refers to average values.
C' - Defined in Eq. III-lO (Superscript h means "cos c5....component", Fig. III-lO).
CA - Angularity Coefficient, Eq. III-I3, accounting for Thrust Angularity Losses.
CD - Drag coefficient (cf. Fig. 2, Introduction), also - Cd
CDS - Nozzle flow Coefficient, Eq. III-7
Cf- 2D vectoring-nozzle flap width, (cf. Fig. I, Introduction and Fig. III-lO).
Cfg - Nozzle Thrust Coefficient, Eq. III-6
CO. - Center of gravity.
CL - Wing sectional lift coefficient. See also LiCL and (LiCdv
C-R - Circular-to-Rectangular transition ducts in vectoring nozzles (cf. Fig. III-I 7 and Figs. 1-13,
14).
Cv - Velocity Coefficient
q - Equivalent flap Length - see Fig. III-lO
Cp - Pressure Coefficient.
CP - Center of Pressure.
Cn - Yawing moment coefficient.
Cny - Increment in yawing moment per degree of deflection of both yawing vectoring nozzles (or of
both rudders).
Cr - Aerodynamic thrust coefficient, FglqSw
17
18 Glossary and Notation

CI' - Exhaust gross thrust blowing coefficient; Refers to the flight dynamic pressure and wing area Sj;
see Fig. III-IS and Eqs. III-29, 30.
D - Dimension of arm in pitching moment (Fig. 11-10 and Fig. I, Introduction). Drag.
Dv - Drag in the y-direction.
2D-CD - Two Dimensional-Converging-Diverging, vectoring exhaust nozzle. (Sometimes written
as 2-DIC-D).
Degr. - Engine Degradation Status/History.
DTV - Directional Thrust Vector
Dr - Induced drag force due to supercirculation
E - Dimension defined in Fig I, Introduction.
EGT - Exhaust Gas Temperature.
EW - Electronic Warfare.
ETV - External Thrust Vectoring.
Fg - Nozzle gross thrust.
FBL - Fly-By-Light.
FBW - Fly-By-Wire.
FCE - Canard Effects.
G - Supercirculation gain factor defined by Eqs. III-IS, 19
G' - Adjusted supercirculation gain factor - G/(Sj ISw). (Fig. III-IS).
G.E. - Ground Effects.
FQ - Fuel quality and caloric value and fuel-air ratio.
H - Altitude.
HIDEC - Highly Integrated Digital Engine Control.
IFPC - Integrated Flight-Propulsion Control.
Inlet - Aircraft air inlet geometric characterization + dynamic mode of operation and maximum dis-
tortion limits. It also depends on engine stall margins, to, Urn,x, etc. (see Appendix F).
IR - Infra Red
lTV - Internal Thrust Vectoring.
LTV - Longitudinal Thrust Vector
M - Gas Mass Flow Rate. Also Mach Number.
Ma - Aerodynamic yawing moment
NAR - Nozzle Aspect Ratio - A2/4As [nominally defined with As in S.L., static, "dry" or "AB" engine
conditions].
NG - Nozzle Geometry
NPR - Nozzle Pressure Ratio
Pa - Ambient Pressure.
PT8 - Total pressure at station S (i.e., at the nozzle throat).
PJV - Partial Jet Vectoring
PSM - Pure Sideslip Maneuvers
PST - Post-Stall Technology
PVA - Pure Vectored Aircraft
PVR - Pure Vectored RPV
Pw - Static pressure at the wall.
PT2 - Total pressure at compressor's inlet.
qo - Free-stream dynamic pressure
q - Jet dynamic pressure.
RA - Robot Aircraft
RaNPAS - Rapid-Nose-Pointing-and-Shooting.
RIC - Radio Control.
RCC - Reinforced Carbon-Carbon (Appendix B).
RCS - Radar Cross Section.
Glossary and Notation 19

Re - Reynolds Number.
RPM - Rounds Per Minute.
RPV - Remotely Piloted Vehicle, or, more generally, RA.
R, - Radius of nozzle flap surface at nozzle throat.
SAM - Surface to Air Missile
Sw - Wing reference, or total area.
S) - Supercirculation affected wing area (cf. Figs. III-I 8, 19, and II-I).
SERN - A type of thrust vectoring nozzle. see Fig. A-3
SIMTD - STOL and Maneuver Technology Demonstrator
SSW - Supersonic Waves interactions
STOL - Short Takeoff and Landing.
STOVL - Short Takeoff and Vertical Landing
( - time
(0 _ Dynamic Distortion & Turbulence, especially at station 2 (pressure and velocity).

T - Actual (total) Net Thrust.

Ta - Ambient Temperature.
Ti - Ideal isentropic thrust.
Tx. Ty. Tz - Thrust-vectored components in the X-. y-. z-directions
TR - Thrust Reversal mode of operation (cf. Fig. 14, Intr., Figs. 1.1,4 and 1.7).
T, - Total Temperature.
Trim - Engine Trim and Compressor stall margin (depend on altitude, M. Re, etc.).
Turb - Local Turbulence Degrees.
TV - Thrust Vectoring.
Tv - Vertical Thrust Component (normally equals to T=).
Voo - Free-stream velocity
V - Aircraft velocity.
V/STOL - Vertical and Short Takeoff and Landing.
VTOL - Vertical Takeoff and Landing.
VRA - Vectored Robot Aircraft.
VRT - Vectoring, Reversing, Targeting (or simultaneous use of pitch-yaw-reversal
thrust-vectoring).
Vs - Isentropic, or fully expanded, jet velocity.
V9i - Ideal velocity based OnA9/Ag.
WVR - Within Visual Range

Greek
a - Divergence angle of nozzle flap surfaces (cf. Fig. 1-1).
u:* - Engine throttle angle.
a - Angle of attack (AoA).
a- Time Rate of Change of a.
P - Angle of sideslip.
Ii - Time Rate of Change of p.
I> -AeIA, - Nozzle expansion ratio = A9/Ag.
Ydisl. - Circulation distribution over the wing, including wing's twist, etc.
8c - Canard angle of attack.
8, - Rudder deflection angle.
8e - Elevator deflection angle.
8v - Pitch Thrust Vectoring Angle (counted positive for downward flap/jet rotation)
8y - Yaw thrust Vectoring Angle, see Fig. 1, Introduction.
LlCL - Increased C L due to "down" thrust-vectored-induced lift and (LlCdl'
20 Glossary and Notation

(LiCdv - Increased CL due to direct engine lift Tv (or T z ).

LiH - Horizontal tail deflection angle.
p - Gas or air density. Also, in a few references, equals to Ii or 9.
9 - Cant angle of vertical tails/stabilizers
Jl - Missile ofT-boresight angle. Viscosity.

Subscripts/Superscripts
The subscript T refers to total conditions, C - to critical values, SC to supercirculation and 'a'to
ambient conditions. Superscript c to refers to canard. For the definitions of a few other parameters see
the figures and the text. Engine station numbers follow the conventional thermodynamic definitions.
 INTRODUCTION

"Powered lift is the biggest contributor to military aviation

today - and may be civil aviation tomorrow - since the
invention of the jet engine."
DR. JOHN FOZARD
British Aerospace Division Director (204)
THE MAIN PROBLEMS OF THRUST-VECTORED
MANEUVERABILITY

- Is thrust vectoring becoming the standard technology of fighter aircraft? Indeed,

how important it is to enhance maneuverability and controllability?
- Are the roads to thrust vectoring also the roads to Post-Stall Technology (PST), and
to low observability?
- What are the fundamental concepts of "pure" and "partial" vectored aircraft?; or of
"internal" and "external" thrust vectoring?
- What are the technology limits, and what is the state-of-the-art of vectored
propulsion/vectored aircraft?
- What is the lowest thrust-to-weight-ratio above which one can extract clear-cut
advantages of vectored fighters over conventional ones? I.e., which of the existing
fighters can be upgraded to become PST-vectored fighters?
- What are the most promising designs of PST-Stealth/vectored propulsion/vectored
aircraft? Do we have the proper design philosophy to handle these problems? Are
Soviet and Western thrust-vectoring methodologies similar?
- Does yaw-pitch-roll thrust vectoring constitute a basic requirement for survivability
and winning in the air combat arena? How can it contribute to the aircraft's STOL and
agility characteristics?
- Can an efficient PST-inlet be developed? Can such inlets be installed on, and flight-
tested by PST-vectored RPVs? Indeed, what should be the R&D tools for the evalua-
tion of vectored aircraft?
- How should PST-fighter agility be defined? What are the measurable parameters, or
"metrics", which define vectored aircraft agility?
- How can we identify maneuvers, missions, and flight regimes in which vectored air-
craft demonstrate advantages over conventional ones? Can one express these advan-
tages in terms of killing-ratios, or other measurable metrics?
- What should be the new flight/propulsion control rules for PST-vectored aircraft?
- What specific maneuvers, and what new pilot tactics are associated with "partial"
and "pure" PST-vectored aircraft?
22
Introduction 23

- What are the expected g-loads, and other limitations, associated with thrust-
vectored, PST-maneuvers?
- And, most important, how should vectorable engine nozzle and vectorable inlet geo-
metric design, and aspect ratio, be modified to meet a given set of mission needs, such
as low signatures, STOL-VTOL, air-to-air, or air-to-ground supermaneuverability and
supercontrollability? In particular, what are the engine/nozzle/inlet efficiency varia-
tions and limitations associated with these new design trends?
These are some of the main problems dealt with in this book.

1. VECTORED AIRCRAFT: BREAKING THE "STALL BARRIER"

Assessing these problems first in this introduction we will assert that in future aerial
combat, pointing the nose/weapon of the aircraft at the adversary first will be required
to win, since pointing first may mean having the first opportunity to shoot (§3.5). It is
also, under proper (post-stall) maneuverability rules, the required technology to dra-
matically increase survivability.
The availability of PST vectored fighters, helmet-sight-aiming systems, all-aspect
missiles and the new generation of EW systems, require reassessment of the optimal
balance between aircraft agility and effectiveness, and the agility and effectiveness of
missile/helmet-sight-aiming systems (§3.5).
Whatever is the aforementioned balance, high-performance fighter aircraft will
gradually be based on improved thurst vectored propulsion/maneuverability/con-
trollability.
Since future fighter aircraft would be thrust-vectored, and since thrust-vectoring
engines would be used for enhanced maneuverability and controllability, as well as for
conventional propulsion, one must first define new concepts and new measurable
"metrics", which would be employed in a realistic comparison of vectored aircraft
maneuverability-controllability with that of conventional fighter aircraft. For this to
be properly done one needs a new, highly-integrated methodology - a methodology
which does not exist yet.

THE NEW STANDARD TECHNOLOGY

Nevertheless, as will be demonstrated in this book, thrust vectoring is now becoming

the basic requirement and the standard technology for all future fighter aircraft. Yet,
as it stands now, this technology is still in its embryonic state. While the pitch/thrust-
reversal technology (see below) appears now to be maturing, the most critical technol-
ogy of simultaneous yaw-pitch-roll thrust vectoring (see below), is still far from this
stage.
In light of the prolonged time inherently associated with the advancement and
maturity of such an engineering field, one may expect its full exploitation only in the
24 Introduction

,..-z:,

Fig. I. A Schematic Representation of Pure Vectored Aircraft

Why the needs for Rapid-Nose-Pointing-And-Shooting dictate the need for vectored fighter aircraft?
The roles played by all-aspect missiles, and other, newer, combat-effectiveness factors, are evaluated
in detail in the main text. Altogether, this evaluation demonstrates why simultaneous yaw-pitch-roll
thrust vectoring is the basic requirement for all future fighter aircraft.
Vectored aircraft may be divided into those that are "pure" or "partial".
In pure thrust vectoring the flight-control forces generated by the conventional, aerodynamically-
affected, external control surfaces of the aircraft, are replaced by the stronger, internal, thrust forces of
the jet engine(s). These forces may be simultaneously, or separately, directed in all directions; i.e., in
the yaw, pitch, roll, thrust-reversal, and forward thrust coordinates of the aircraft.
Since engine forces (for PST-tailored inlets) are essentially independent ofthe external-flow angle-
of-attack and slip angle, the flight-control forces of PV A remain effective even beyond the maximum-
lift Angle-of-Attack (AoA), i.e., PV A are fully controllable even in the PST domain. Thus, vectored
flight provides the highest payoffs at the weakest domains of aerodynamically-controlled aircraft (i.e.,
at high-alpha-beta values, at very low (or zero) speeds, high altitude, high-rate spins, very-short run-
ways, and during all PST, Rapid Nose-Pointing-and-Shooting (RaNPAS) maneuvers).
Consequently, no rudders, ailerons, flaps, elevators and flaperons are required in pure vectored
flight, and even the vertical stabilizers may become redundant. Thus, by employing FBW, FBL and
Integrated Flight/Propulsion Control (lFPC) systems, vectored aircraft may resemble "flying wings",
or variously-shaped "body lift objects" which can provide Pure Sideslip Maneuvers (PSM), or even
PSM/RaNPAS maneuvers (cf. Lecture II).
In pure-vectored-takeoff-methodology one first turns the jet upward, and rotates (nose Iiftup) the
aircraft at a much lower speed (i.e., using a much shorter runway), than with conventional-technology
aircraft. Then, and under the automatic guidance of IFPC, the jets are turned down, adding direct lift
and supercirculation to proper wing sections (cf. Fig. 11-5). Approach and landing methodologies of
PYA (Fig. 11-6) also afford the use of very short runways. Thus, STOL, V/STOL, STOVL and VTOL
performance become the "natural" domains of PV A. These terms are defined in Appendix B and in
Lecture II.
NOTES: This figure is limited to a schematic representation of this class of aircraft. The canards, or
the vectored canards, are non-essential options (cf. Fig. II-I). The shaded area represents
supercirculation-affected wing sections (cf. Parag. III-7.5 for the definition of supercirculation effects
in vectored aircraft). PST is defined in Fig. 2. Pure sideslip, RaNPAS maneuvers are not consuming as
much energy as pure PST-maneuvers (cf. Fig. 3). Thrust reversal may be less effective in flight than
PST-maneuvers, cf. Fig. 21.
Introduction 25

tD\~
. __ _
,
~ "
!'--_ _ _ _~,_" ~ I I •
o 90 150

IUGlif PATII DUlltllDN

Fig. 2. The Definition of the PST Domain in which Rapid Nose-Pointing-and-Shooting (RaNPAS)
Provides the Highest Combat Payoffs. (See also Figs. 3 to 6).
By employing the change of the lift and drag coefficients with AoA (alpha), one may define the
domain of Post-Stall Technology (PST) as that marked in the upper two figures. In addition, one may
split alpha into alpha-conventional and alpha-PST, as shown in the lower figure. It should be stressed
that, in practice, AoA may be greater than 90 degrees. Vectored aircraft will provide combat effective-
ness in the STOL, PST and PSM/RaNPAS domains (cf. Fig. 3).

post-ATF era. Nevertheless, some of its proven elements may be gradually incorpo-
rated in such upgrading/transforming designs as those feasible for the new vectored
F-15 STOL Demonstrator, or for the proposed vectored F-16 and F-18 upgrading
programs.
Attempting the integration of some advanced propulsion concepts with vectored
aircraft's superagility concepts, is one of the central goals of this book. No doubt it will
also become the central goal of well-integrated engineering and pilot education, and of
research strategies for advanced propulsion/airframe programs.

PRELIMINARY REMARKS ON SAFETY AND CIVIL APPLICATIONS

This introduction defines and depicts the fundamental concepts associated with the
aforementioned new technologies. It is demonstrated, during this process, that
advanced propulsion research must be expanded now to include aircraft's thrust-
vectored agility, and, in particular, thrust-vectored maneuverability and controllabil-
ity. Thus, student instruction, as well as research and development injet propulsion
are gradually emerging as a highly integrated field of engineering.
At this early stage of the course one may also stress three points:
(i) - Thrust-vectored controllability of pure vectored aircraft is much safer and eas-
ier, as well as less complex and costly, than that of partially vectored aircraft (see the
26 Introduction

fundamental definitions below). Furthermore, maneuverability, or post-stall

supermaneuverability of pure vectored aircraft is considerably superior to that obtain-
able with partially vectored systems, or with conventional fighter aircraft.
(ii) - Vectored propulsion systems may be developed and tested both in the jet-
propulsion laboratory and in flight-test programs using small, vectored RPVs (Cf. Lec-
ture IV). Moreover, various missions/design options for PST-vectored RPVs are now
emerging as a highly promising field of study.
(iii) - The cargo and civil aircraft industries may also exploit the methodologies of
vectored propulsion, for instance, by introducing low-drag, cost-effective, STOL,
pure-vectoring propulsionlflight-control systems.

1.2 Vectored Aircraft:

»
Soviet and Western Concepts
Since thrust vectoring is now emerging as the pace-maker technology of all superagile
fighter aircraft, it is imperative to examine first the different design philosophies of
this important field of air defence. The growing importance of this field is well
reflected by the increasing number of heavily-financed R&D projects conducted now
by various governmental, industrial and academic bodies, as well as by the increasing
number of recent publications on this topic (see the two hundred and more references
at the end of this book). These papers reflect different international methodol-
ogIes.
For instance, the Central Institute of Aviation Motors in Moscow, has recently pub-
lished computer simulations of thrust-vectored aircraft (188). To start with, one may
note that the Soviet scientists present their analysis for aircraft propelled and con-
trolled by simultaneous yaw-pitch thrust vectoring.
Unlike the Soviets, who appear to be newcomers to this field, the American design-
ers had previously adopted a more conservative design philosophy, concentrating
their main R&D efforts only on pitch, or pitch/reversal, thrust-vectored aircraft. (E.g.,
the GE/PW-thrust-vectoring engines for the new F-15 STOL and Maneuver Technol-
ogy Demonstrator (S/MTD) and for the Advanced Tactical Fighter - the ATF.)
There is, nevertheless, a new US-German program to flight-test [see below] an
experimental yaw/pitch vectored plane - the X-31, as well as a new, comprehensive
NASA program (270-209) for simultaneous yaw-pitch (external) thrust vectoring.
Furthermore, a number of multi-axis (internal) thrust vectoring studies, e.g., the (now
cancelled) X-29A program employing the GE-ADEN and the GE-gimballed nozzle
(see Lecture VI), have been reported (220). A minor US program is also being con-
ducted now in this laboratory to evaluate the pros and cons of simultaneous yaw-
pitch-roll (internal) thrust vectoring (179). This program includes laboratory tests and
flight testing of vectored RPVs equipped with various two-dimensional nozzles, rang-
ing from 2 to 46.7 NAR. It also includes a comparison of axisymmetric thrust-
vectoring nozzles with 2D nozzles.
These design differences may be highly critical in the final assessment of fighter air-
craft combat-ability and effectiveness in the future. Hence, it is imperative, and
Introduction 27

_ =IIISSlll Off·BORISIG:'T ANGll

r;sJ"\ /IGIIIIR wmnl.llo

\..:.:::....-I IIANIUVIR

Fig. 3. Vectored Fighters Exploit the Nose Pointing Capability Achieved Through PST or PSM (Pure
Sideslip Maneuvering) (Tamrat, Northrop, 185).
This figure provides a simple example for the offensive exploitation of a PST-nose-pointing capabil-
ity of a vectored fighter. Here the conventional (8) and vectored (A) fighters are turning in a circle with
neither gaining any advantage. Then, at position A, an execution of, say, 90 degrees, post-stall maneu-
ver by the vectored fighter will make the conventional fighter vulnerable to missile/gun attack (in this
example, a successful acquisition of 8 in the missile off-boresight angle).
Combat applications of pure sideslip maneuvers (PSM) are numerous. They include air-to-ground
maneuvers (Fig. 15), escape maneuvers, and some highly promising alternatives to PST maneuvers.
The last category may be demonstrated by modification to the PST -maneuvering shown in this Figure:
Instead of the (energy degrading) pitch-PST maneuver depicted, one may consider a roll followed by
RaN PAS yaw/pure sideslip maneuver, which may be prolonged and/or repeated without significant
speed/energy degredation. The use of this option may become significant in close-combat, target-rich
environments. Thus, PSM/RaNPAS maneuvers are less dangerous than the PST/RaN PAS ones.
Combat agility requirements, when maneuvering well beyond the maximum lift angle of attack, are
treated differently from those for the conventional angle-of-attack regime, since at such extreme angles
of attack the fighter depends mainly on its thrust vectoring to maintain both lift and flight-control
power (185).
The very selection of the proper position, or the execution timing, is key to winning in PST maneu-
vers. For instance, a PST maneuver executed from post ion A' will fail to gain a missile/gun pointing
advantage. Thus, the vectored fighter, having missed the proper position/timing to shoot, will become
"a sitting duck", for it has lost speed during the PST maneuver, and, hence, the potential to change
flight path to evade a counterattack (185). It should be stressed, however, that this figure, as well as
Figs. 4 and 5, represent only "unit operations" which may be conducted within a complex time-space-
targets of actual air-combat.

timely, to define and compare the expected advantages of adding (axi or 2D) simulta-
neous yaw-pitch-roll thrust vectoring to the current F-15 (S/MTD) - ATF propulsion
methodology.
Secondly, one may note that a thrust-vectored version of the Su-27 is now being
developed. Thirdly, one must notice that the Soviet simulations of vectored
propulsion/flight control, have been reported by a propulsion institute, and not by a
28 Introduction

flight-dynamics institute, as is the tradition in the West. The reason behind this is,
probably, the realizaiton that vectored aircraft agility improvements require novel
integrated propulsion/flight-control programs. This philosophy is also reflected by a
new Russian book on thrust vectoring (226). This book combines aerogasdynamics,
flight/propulsion characteristics and powerplant design philosophy in a single text.

1.3 The Search for Vectored-Thrust/Super-Agility "Metrics"

The very definition of agility is still being debated (cf. parag. 3.5 and 5, and Figs. 9,19
and 20 below, Lecture II, and references 184, 195, and 196).
Anticipating the introduction of vectored aircraft technology, McAtee, ofGD (196),
has, in 1987, defined agility as composed of two complementary concepts: Maneuver-
ability and controllability. PST maneuverability is then called "supermaneuver-
ability", while PST controllability is named "supercontrollability" (see below). The
last two concepts, combined, may also be termed "super-agility" (see 3.5 below).

2. HOW TO ORDER THE DEFINITIONS OF NEW

INTERCONNECTED CONCEPTS

Following the definitions of "pure" and "partial" vectored aircraft, we shall stress the
higher combat effectiveness of pure, over partially-vectored aircraft. Within this ini-
tial framework, we shall also assess the pros and cons of the X-31 program, as well as
that of the F-15-S/MTD and other advanced programs.
During this process we shall distinguish between the advantages obtainable by
upgrading programs of existing fighter aircraft (e.g., vectored F-15, F-18 and vectored
F-16), and the radically-different task of designing high-performance, Rapid-Nose-
Pointing-and-Shooting (RaN PAS), Post-Stall-Technology (PST), low-signatures, pure
vectored fighter aircraft [with simultaneous yaw-pitch-roll vectoring capabilities].

2.1 PST-Vectored RPV as a Research Tool

We shall then introduce a new research tool: The mini, PST, RaNPAS, Pure Vectored
RPV (PVR). It has been demonstrated, on the basis of flight tests conducted by this
laboratory, that such low-cost, R&D&T tools, can replace some of the more expensive
roles of wind-tunnel and complex computer simulations of actual, thrust-vectored-
aircraft-maneuverability/controllability capabilities.

2.2 Cold Propulsion

Finally, we shall assess the pros and cons of "cold propulsion", e.g., of the effectiveness
of STOL, low-IR-signatures, inlet-distortion-free, cold-jet, PST-PVRs. Such PVRs
may be based on, say, multi-stage, multi-axis, ducted fans, ducted prop-fans, or sim-
Introduction 29

• _ MISSilE orr·
10RISIGIIl ANGLE

liME ON FI(UNG ,0SI1ION

fOR ,sr=& SIC

CONVINTIONAl /lCHIIA
,/" IGRn 111.111101
lAAJICl1lRY

Fig. 4. The Effect of Angle-of-Attack Rate Of Change On Close-Combat Effectiveness.

The combat scenario shown here is based on a point-mass trajectory model, as employed by Tamrat
of Northrop (185). At t=O sec., the vectored fighter pitches up, at an AoA rate of 50 deg./sec., to rapidly
reach a final AoA = 90 deg. The conventional adversary, which, for the sake of comparison, is assumed
to start at the same initial conditions, pulls up to the aircraft load-factor/lift-limit (apparently the only
option available), and climbs to positions 2 to 8. In such a theoretical comparison, according to
Tamrat's results, the vectored fighter may have the adversary within its missile/gun envelope for 6 sec.,
beginning at 2 sec. This "time within firing envelope" (TWIFE), may be considered as a nominal yard-
stick for PST combat advantage. (See, however, the remarks in the following figures and in the text.)
Alternatively, one may employ PSM/RaNPAS maneuvers.
The Pougachev's Cobra maneuver demonstrated by the Su-27 (without TV) may be considered as a
unit operation in this category. TV is now being added to the Su-27 (see also Fig. 11-7).

pIe, off-the-shelf, centrifugal compressors feeding sub- or supersonic cold-jet nozzles

with cold, compressed-air, for STOL, low-IR-signature, vectored propuslion. While
limited to specific applications, such [cold] vectored aircraft may be applicable as
STOL vehicles, or used as, say, low-signature, vectored, PST, cruise-missiles,
equipped with simple, (non-PST), semi-recessed, s-shaped, high-aspect-ratio inlets.

3. THE FUNDAMENTAL DEFINITIONS

3.1 Pure Vectored Aircraft

Vectored aircraft may be divided into those that are "pure" or "partial".
In pure thrust vectoring (Fig. I) the flight-control forces generated by the conven-
tional, aerodynamically-affected, external control surfaces of the aircraft, are replaced
by the stronger, internal, thrust forces of the jet engine(s). These forces may be simulta-
neously, or separately, directed in all directions, i.e., in the yaw, pitch, roll, thrust-
reversal, and forward thrust coordinates of the aircraft.
Since engine forces (for post-stall-tailored inlets), are less dependent on the external-
30 Introduction

,._MISSIU OfF.
OOR[SIGIIT ANGLE

0",
,-..--~

Fig. S. The TWIFE Yardstick and the AoA Rate Of Change (Tamrat, Northrop, 185).
Using similar modelling as those employed in Fig. 4, Tamrat estimates the relative combat advan-
tage of one vectored fighter over another with AoA rate difference of 20 deg./sec. The more agile
fighter has a TWIFE of about 2 sec. This value was increased to 10 sec., when the superior fighter was
engaged against an adversary having AoA rate capability of only 10 deg./sec. Again, PSM/RaNPAS
maneuvers may be considered instead.

flow, than the forces generated by the aerodynamic control surfaces, the flight-control
forces of Pure Vectored Aircraft (PVA), remain effective even beyond the maximum-
lift Angle-of-Attack (AoA), i.e., PVA are fully controllable even in the domain of Post-
Stall Technology (PST), as defined in Fig. 2. Thus, vectored flight provides the highest
payoffs at the weakest domains of aerodynamically-controlled aircraft (i.e., at high-
alpha-beta values, at very low (or zero) speeds, high altitude, high-rate spins, very-
short runways, and during all PST, Rapid Nose-Pointing-and-Shooting (RaNPAS)
maneuvers).
Consequently, no rudders, ailerons, flaps, elevators and flaperons are required in
pure vectored flight, and even the vertical stabilizers may become redundant. Thus, by
employig FBW, FBL and Integrated Flight/Propulsion Control (IFPC) systems, vec-
tored aircraft may resemble "flying wings"', or variously-shaped "body lift objects"
which can provide Pure Sideslip Maneuvers (PSM), or even PSM/RaNPAS maneu-
vers (cf. Figs. 1 and 3 and Lecture II).
In pure-vectored-takeoff-methodology one first turns the jets upward, and rotates
(nose liftup) the aircraft at a much lower speed (i.e., using a much shorter runway),
than with conventional-technology aircraft. Then, under the automatic guidance of
IFPC, the jets are turned down, adding direct lift and supercirculation to proper wing
sections (cf. Fig. 11-5). Approach and landing methodologies of PYA (Fig. 11-6) also
afford the use of very short runways. Thus, STOL, V/STOL, STOVL and VTOL per-
formance become the "natural" domains of PYA (cf. Appendix B).

3.2 Thrust Vectored Pure Sideslip Maneuvers (PSM)

Pure sideslip maneuvers are defined as sidewise translations with unchanged nose
pointing and without tilting the aircraft sideways (Le. no banking). When combined
Introduction 31

\
II

'ST GOIS POST·STAlI

SLOWS DOWN CONVINTIONAl
~OW. ATTACKIR
PUllS Al
/ N,ICL 1.I.u LIMIT

?/ 'ST ROllS
ABOUT fliGHT PATH.
UNIOAOS TO GAIN INIRGY

10 ,,"CONVINIION'I
AfTAClUI

Fig. 6. A Simplified Defensive, PST, Sub-Step Maneuver (after Tamrat of Northrop, 1988, Ref. 185).
In future aerial combat. according to Tamrat. pointing the nose/weapon ofthe aircraft at the adver-
sary first will be required to win. since pointing first means having the first opportunity to shoot (cf..
e.g .. Figs. 4 and 5). Thus. according to Tamrat. the point-and-shoot capability requirement. results
from the introduciton of the all-aspect missile. which. if pointed at the adversary. can be launched
from any aspect. including the head-on encounter.
At slower speeds. according to Tamrat. nose pointing capability offuture fighters is increased by uti-
lizing their capability to fly at large post-stall angles of attack (up to 90 degrees). In turn. post-stall
maneuvering capability is the SOLE RESULT of the new. potential availability of powerplant and
controls technologies for thrust vectoring. Consequently. Tamrat concludes. thrust vectoring is the
basic requirement for all future fighter aircraft.
An important use of various vectored-PST maneuvers is during the defensive portion of air combat.
This schematic figure shows an over-simplified situation in which the vectored fighter is initially
above and in front of the conventional adversary. At the initial position the vectored fighter starts a
vertical pitch-up vectoring to reach AoA up to 90 degrees. This PST-DECELERA nON forces the
adversary to overshoot.
When this is done. the adversary will be in front and high vs. the vectored fighter (role exchange).
After delivering its weapon. the vectored fighter may roll about its flight path. reduce AoA. and accel-
erate to regain energy. For a variant to this over-simplified sub-step maneuver see Fig. 3 and
Lecture II. including a PSM/RaNPAS maneuver.

with RaNPAS, it would be called PSM/RaNPAS-maneuvers (cf. Fig. 3). Such impor-
tant maneuvers (see below) become feasible with the introduction ofPVA technology.
There are a number of basic propulsionlflight-control rules and some technical solu-
tions to this kind of pure vectored flight (cf. parag. II-3 for detail and definitions of
preliminary-design criteria, control rules and associated dimensionless numbers).
The simplest pure sideslip flight (say, to the left hand side of the aircraft), may be
carried out by deflecting the right-engine jet to the yaw angle marked in Fig. 1 (until it
coincides with the Center of Pressure in the Y -direction - CPr ), while simultaneously
reducing the thrust of the left engine to the same level as that left over for the forward
thrust of the right engine [cf. Fig. 1]. This combined operation (which must be
32 Introduction

contolled by a proper IFPC system programming), keeps the nose-pomtmg

unchanged, while pushing the aircraft to the left without banking. In fact, pure side-
slips are feasible with and without canard-configured PV A. Most important, rapid
nose-pointing may be combined with PSM to provide endless PSM/RaNPAS
maneuvers.

3.2.1 Combat Applications of PSM or PSM/PST-RaNPAS

Combat applications of pure sideslip maneuvers are numerous. They include air-to-
ground maneuvers (Fig. 15), escape maneuvers, confusing-the-adversary maneuvers,
and some highly promising alternatives to pitch-only, PST maneuvers. The last cate-
gory may be demonstrated by Fig. 3: Instead of the (energy-degrading) pitch-PST
maneuver depicted, one may consider a 90-degrees-roll, and roll-stop followed by
RaNPAS yaw/pure sideslip maneuver, which may be prolonged and/or repeated for
other targets but without significant speed/energy degredation. (Pure, tailess, vectored
aircraft, of the type shown in Fig. I, do not incorporate vertical stabilizer(s). More-
over, their total sideslip drag is relatively low. Hence, during pure, 90-degrees
RaNPAS-sideslip, PSM-maneuvers, the instantaneous rate of energy degradation
should be substantially lower than that encountered during pitch-only, RaNPAS-PST-

Fig. 7. The PST, Thrust-Vectored X-31 Design Concepts.

This figure shows the yaw-pitch-thrust-pedals (beyond-nozzle-exit, free-jet deflectors), of the experi-
mental plane X-31. Expected to be flight-tested in 1990, this plane is to fly at up to 70 deg. AoA.
Designed for quick nose-pointing-and-shooting capabilities, such a propulsion-controlled aircraft may
significantly increase close-combat effectiveness. Consequently, by using such design concepts, and
especially the Herbst PST concepts (188), one may help to increase the probability to win, or the kill
ratio of PST fighter aircraft, etc. (cf. Table I).
While the X-31 is highly-promising and advanced, such a thrust-vectoring design methodology may
suffer from certain disadvantages (see Lecture VI for more detail).
Introduction 33

maneuvers.) Thus, the use of this option may become significant in close-combat,
target-rich environments, in which minimum degradation of energy becomes impera-
tive. A number of new RaNPAS-PST-PSM-maneuvers can be worked out on the basis
ofthis methodology. In fact, such maneuvers are being investigated now by this labo-
ratory, using vectored RPVs (Lecture IV).
However, the expected stability-IFPC problems associated with such potentlal
maneuvers should not be underestimated (179).

§
IMTHOUT
Tl4l\JST
§ VECTQnINQ

11:8
~
<
~'"

0- M

§
WITH
Tl-lAUSr
VECTOAtNO
~§

Fig. 8. The Introduction of Vectored Fighter Aircraft Causes Such Classical Concepts as 'Spin-Danger'
and 'Performance-Stall-Limit' to Become Obsolete, Thereby Demonstrating the Need for Curriculum
Redefinitions in Modern Pilot and Engineering Education.
This schematic drawing, taken from a recent report of the Central Institute of Aviation Motors in
Moscow, USSR, (180), demonstrates a simple performance payoff of vectored fighter aircraft as pre-
dicted by their numerical-mathematical analysis of the flight dynamics of Post-Stall Technology
(PST), simultaneous-yaw-pitch, Vectored Aircraft.

Finally, one may also stress here the expected combat potentials of this methodol-
ogy for partially-vectored fighter aircraft. This topic is taken-up next.

3.3 Partially-Vectored Aircraft

Rudders, elevons, ailerons, flaps and elevators are still used in Partial Jetborne Flight
(PJF). Thus, the elimination of one or more modes of the pure-vectored elements,
such as the elimination of yaw-vectoring and supercirculation in the Harrier, or in the
STOL F-15 demonstrator (cf. Lecture V), categorizes the aircraft as PJF. In fact, all
vectored aircraft employing external thrust-vectoring devices should also be classified
as partially-vectored aircraft which are only capable ofPJF, i.e., maneuverability and
controllability limits of these aircraft are still affected, to a degree, by the external-flow
34 Introduction

regime. Thus, this category includes a recent F-14 flight with a single, post-exist, yaw
vane, as well as the X-31, and one of the proposed designs for vectored F-16 and F-18
(see Lecture VI). Beyond a limiting AoA, the interaction of the external flow with
engine exhaust-jets may become prohibitive for establishing a reliable IFPC during
RaNPAS-PST maneuvers (Lecture VI).
Aside, and in addition to these inherent problems, one may stress the fact (which
was recently verified by flight-testing of our pure and partially-vectored RPVs), that
flight/propulsion control during PJF is much more complicated and cumbersome
than that feasible with pure vectored vehicles.
To start with, there are too many variables, most of which become redundant. Fur-
thermore, any reliable IFPC system for PJF suffers from lack of reliable DATABASE
for the very definitions of the relevant variables needed, their proper range, limits, and
coupling effects to each other during actual flight testing. (To verify this conclusion we
have established a PJF program, which compares the performance of pure vectored
RPVs with that of partially-vectored ones; e.g., prototype No.7 of this laboratory is a
flying, computerized, 9-feet-Iong, variable-canard-configured, vectored-F-15 RPV. It
is also a STOL-PST demonstrator.)

3.4 Thrust-Vectored, PST-Maneuvers

Figures 3 to 6 demonstrate some of the most-simplified, PST, or PSM, close-combat
maneuvers which may be carried-out by vectored fighters (in comparison with those
obtainable with conventional fighter aircraft). The conclusions extracted from these
simulations are depicted and detailed therein. Other, more fundamental conclusions,
are discussed next.

3.5 The Definitions of Agility, Supermaneuverability and

Supercontrollability
According to McAtee (196), the quality of agility is the combination of three (measura-
ble) tasks/abilities (Figs. 19 and 20):
1) -The ability to "outpoint" the opponent (pointing at him before he points at you).
This advantage must be such that the opponent does not have the opportunity to
launch his weapon before he is destroyed. Otherwise, with current launch-and-Ieave
weapons, mutual destruction would result. It is, therefore, the ability to point the nose/
weapon at the enemy quickly, while, simultaneously, computing and locking, so as to
minimize the total length of delay times associated with secure locking and obtaining the
shortest/optimal missile flight path/time from aircraft release point-attitude to the mov-
ing target. This requires aircraft superagility to be well-integrated with missile's high-
"g"/speed agility and initial vectoring conditions (cf. 11-7.4). It also dictates the develop-
ment of TV-missiles. This aircraft ability is measured in terms of Turn Rate/Bleed
Rate, as demonstrated in Figs. 19 and 20.
2) - The ability to continue maneuvering at high-turn rates over prolonged periods to
Introduction 35

I PITCH AGILlTV I Alrcr.11

Mach Nos.
Load Faclor

~I;~
~~
Spocilic Excess
Powor
TR. Tum Ralo
AOA' Angla 01 Allack
T.. Time 10 Roll
and SlOP

ITORSIONAL AGILITVI

~o
To

)~
P~
I ~ ~Mo
)

IAXIAL AGILlTV I

.t.P<..···MA.
......... A.
"

~
:J,
•
O
:.

.~ ~-83- M, M~

Fig. 9. Debated Agility Measures.

The current definitions of agility (cf. parag. 3.5), and of transient measures of merit are still being
debated. This figure shows some of which have been proposed. These measures may be considered as
the time-rate-of-change of the classical energy measures.
Unlike such simple analyses, Hodgkinson, et al. (184), have used complex battle simulation pro-
grams to reproduce many of the decisions and actions of air-to-air combat, and to compare aircraft
with various levels of performance and weaponry. Some of their results are shown in the next figure.
See also Figs. 19 and 20.

retain the potential for performing defensive maneuvers, or make multiple kills when
appropriate. I.e., to defend against attacks from other aircraft, or to accomplish multi-
ple kills if the opportunity exists, an "agile" aircraft must be able to continue maneu-
vering at high-turn rates over prolonged periods. This ability is measured in terms of
Residual Turn Rate, as shown by the middle drawing in Fig. 20.
3) - The ability to accelerate rapidly straight ahead so as to leave a flight at will, to
regain maneuvering speed when necessary, or to pursue a departing target when
appropriate. This includes the ability to disengage, or escape from a battle without
being destroyed in the process, as well as the acceleration necessary to "chase down"
an enemy that is trying to escape. This ability is measured by acceleration/speed plots
as those demonstrated in Figs. 19 and 20.
Using both his rich combat and (F-16) design experiences, as well as advanced ana-
lytical methods, McAtee concludes that these three measurable tasks/abilities are cru-
cial for success in modern close-in combat. Thus, McAtee states, the critical design
36 Introduction

features for modern fighters are those that enable the pilot to command very high max-
imum turn rates over prolonged periods, and to perform a l-g acceleration.
However, good maneuverability must be integrated with effective controllability,
i.e., the ability to change states rapidly (control power), and the ability to capture and
hold a desired state with precision (handling qualities) (196).
Traditionally, controllability was thought to be degraded at either of two condi-
tions: High Mach number, or high AoA (cf. Fig. 20). However, the introduction of PST
and vectored aircraft technology requires reassessment of the second condition. It also
requires the introduction of new standards and MIL specs (196).

3.6 Yaw-Pitch-Roll Thrust Vectoring for Breaking the "Stall Barrier".

Pitch and yaw control requirements increase with AoA. For a given roll rate, as AoA
increases, the requirements for pitch and yaw forces/moments (for non-thrust-
vectoring aircraft), increase exponentially. At the same time, with conventional aero-
dynamic controls, the forces/moments available decrease as airspeed decreases. Thus,
beyond a given limit, conventional control technology becomes obsolete. This technol-
ogy limit is reached when the size and weight of the aerodynamic control surfaces
needed to provide sufficient forces/moments bocome prohibitive. However, the intro-
duction of PST and vectored aircraft technology (together denoted by McAtee as the
new domain of "supercontrollability"), requires reassessment of all maneuverability
and controllability concepts and requirements.
Thus, according to McAtee, new point-and-shoot weapons have reduced engage-
ment times drastically, leaving aircraft with poor maneuverability and controllability
at the mercy of those that can use their agility to kill quickly during close-in combat (cf.
Fig. 2, Lecture II).
Vectored PST maneuvers, such as those shown in Figs. 3 to 6 and 16, or those dis-
cussed in Lectures II and V, may thus be defined as supermaneuvers.
There are a few dozen candidate supermaneuvers, half of which may demonstrate a
real combat promise (195). In Lecture II we provide a few examples for combat pay-
offs during the proper use, at the proper position/timing, of yaw-pitch-roll thrust vec-
toring during "angles" and "energy" tactics (see below). These tactics employ
supermaneuvers well beyond the current flight envelopes of conventional fighter
aircraft.

3.7 A Basic Thurst-Vectored, PST-RaNPAS, Turn-Back Maneuver

Fig. 16 schematically demonstrates a basic, PST, yaw-pitch-roll thrust vectoring for a
rapid, short-penetration radius, turn-back maneuver (E.g., an escape maneuver from,
say, a SAM envelope, or a RaNPAS, turn-back maneuver in head-on encounters). We
shall return below to this basic maneuver. In the meanwhile it should be stressed here
that in-flight thrust reversal is much less effective than the nose-up, PST-maneuver
shown (cf. Figs. 21).
Introduction 37
Nina SlarUng
1.4 Scanarloa
1.3

1.2 - ~........-:
0
~ 1.1
/ ~
II:
IU
Cl 1.0
/P
Z
c(
x
0 •• % ~
t1 ~ ~
:I
c(
•• ~
Allow Denol ••
Slarling DI"ollon
O' Combalanl
II: .7
w o Nominal
>
0
•• • • 100% Aglilly (Translonl)
l:). .30'1'. Turn Ralo
.5 o .30% Thrust

.4
I VI 2V2 eve

LEVEL Of AIR·TO·AIR CO .. DAT

Fig. 10. In a Target-Rich Environment, a 100% Increase In Transient Agility (in pitch, roll, and axial),
Is Similar to the Operational Payoff of a 30% Increase in Conventional Agility (i.e., Enhancement in
Turn Rate or Thrust). Moreover, the operational payoff of a given enhancement, is more pronounced
in a target-rich environment. Furthermore, of the three transient agilities shown in Fig. 9, torsional
(roll) agility has the largest impact on combat effectiveness.
These conclusions were recently reached by Hodgkinson, Skow, Ettinger, Lynch, Laboy, Chody and
Cord (184), using complex battle simulation programs. The simulated results obtained by these
authors, also indicate that the exchange ratio between highly capable, matched aircraft, tends to
increase as the target environment becomes richer, and, that intense, target-rich WVR (Within Visual
Range) combat, is an inevitable feature of tomorrow's war.
Definitions: This figure shows combat effectiveness (i.e., overall exchange ratio, cf. the definition
below), as a function of scenario size and level of 'blue' aircraft enhancement. "Nominal" aircraft in
the scenarios are F-16-like in performance.
The 'red' aircraft always had nominal characteristics. The 'blue' aircraft either had nominal charac-
teristics, or enhancement in transient and conventional agility. Scenario size was varied from 1vi, to
2v2, to 4v4, to progressively enrich the target environment.
The Exchange Ratio is defined as:
Number of 'red' aircraft killed
Number of 'blue' aircraft killed
Overall Exchange Ratio (OER) is defined as the aggregate from nine scenarios, each with a different
starting condition (see the arrows indicated on the upper-right portion ofthis drawing), so as to vary
the 'blue' positions from defensive to neutral to offensive. As such, OER measures combat effective-
ness of a 'blue' aircraft enhancement over a range of tactical positions from defensive to offensive.

3.8 A "Nominal", Three Steps, Vectored-PST Maneuver

In assessing various vectored maneuvers we have arrived at a common denomination;
PST-RaNPAS maneuvers nominally involve three-steps:
The first one is to acquire the PST AoA value (say, of 90 degrees), by employing a
38 Introduction

quick, maximum AoA rate of change, momentarily disregarding the energy level (see
also below the discussion of "angle" vs. "energy" tactics). The application of this
maneuver is highly risky as well as limited, for it is feasible only at the proper positionl
timing (cf. Fig. 3, and Lecture II). However, a less risky maneuver first involves a
90-degrees-roll, then roll-stop followed by 90-degrees RaNPAS/pure sideslip maneu-
ver (see parags, 3.2 and 3.2.1).
During the second step, the vectored fighter may roll about the velocity vector to
disengage, and/or yaw-vector its nose for starting the second part of a ISO-degrees
turn-back maneuver (cf. Lecture II).
The final step of such 'nominal' maneuvers involves unloading and acceleration to
regain the initial speed/energy level.

3.9 Low or High Load Factors During Post-Stall Maneuvers by Vectored

Aircraft: Intuiton vs. Physics
The lift coefficient as well as the effectiveness of all aerodynamic control surfaces are
drastically reduced during vectored, rapid-nose-pointing, high AoAlPost-Stall maneu-
vers (Fig. 2). On the other hand, the thrust-generated forces and moments of a
properly-designed, internal, thrust-vectoring system, remain almost unaffected by the

Fig. II. A Number of Preliminary (approximately 'I-feet') Models Tested in a Subsonic Wind-Tunnel
as Potential Candidates for the Final Vectored RPV Flight-Test Program. See also Fig. IV-3.
The finally-selected configurations for the design, construction and flight-testing of the earliest vec-
tored RPVs, were slightly influenced by the wind-tunnel results obtained for these models. However,
the first (approximately 7 x 4 ft.) flying prototypes were canard configured, and also had different aero-
dynamic characteristics, proportions, swept-wing angles, etc. These modifications were primarily due
to the need to incorporate optimal yaw-pitch-roll thrust vectoring systems, into the prototype final air-
frame design.
The optimal vectored propulsion system was first evaluated experimentally on a Mabore II, -400 kg
thrusC turbojet engine, using the altitude test chamber shown in Fig. IV -6. Test conditions included
pure yaw, pure pitch, pure roll, and simultaneous yaw-pitch and yaw-roll thrust vectoring. Maximum
tested vectoring angles were plus-minus 20 degrees, in both the pitching and the yawing planes. The
flying prototypes were simple, geometrically-scaled-down copies of the aforementioned optimal pro-
pulsion system.
Introduction 39

external-flow conditions, i.e., thrust vectoring is most effective at the weakest domain
of aerodynamically controlled aircraft.
Thus, as the angles of attack are increased beyond the maximum value, the load fac-
tor on a PST-maneuvering vectored aircraft, may not be as classically computed. In
general, the load factor on a PST-maneuvering vectored aircraft depends on the spe-
cific design of the thrust vectoring system, on the time-varying directions and values
of the vectored-jets deflected, on engine throttle, on the turn rate/radius, body-wing-
AoA/sideslip-angle, speed, altitude, the direction of the gravitational vector, canardl
elevatorslflaperons deflectionslloads, the time-variations in the proper drag
components, etc. Moreover, if the aircraft slows down just prior to a vectored-
controlled, turn-maneuver (with or without thrust-reversal), the load factor is reduced
during the turn performance (cf. Figs. 3 to 6).
Since the lift coefficient falls down at high-alpha values, a properly-designed
propulsionlflight control system should maintain the proper load-factor/acceleration-
force according to mission and the pilot's demands, using thrust vectoring forces and
moments to replace the loss in lift force and the loss in moments generated by the aero-
dynamic control surfaces. Furthermore, as the altitude is increased, the thrust, and,
hence, the vectoring moments and forces (and, thus, the total load factors), are
reduced, when other parameters remain unchanged.
Still further, one must distinguish between the different maximum "g"-components
which a pilot can sustain for a given duration (in the positive or negative pitch plane,
in the yaw plane and during head-on braking). One must also differentiate between
thrust-yaw, thrust reversal, and thrust-pitch forces for yaw, pitch, thrust-reversal, or
simultaneous yaw-pitch, yaw-pitch-roll, or yaw-pitch-roll/thrust-reversal maneuvers
(cf. Lectures I and II).
Consequently, for vectored aircraft performing PST-maneuvers, the instantaneous,
and the "time-averaged" load factors may be designed to be even lower, and shorter,
than those intuitively assumed initially for similar, but unvectored, rapid-nose-
pointing-and-shooting (RaNPAS) maneuvers.
Practically it means that thrust-vectored, post-stall maneuvers, can be safely
employed to increase survivability and killing ratios without surpassing human and
structural limitations.
It should also be stressed that high angles of attack can only be used for low-speed
flight increments, for instance, during a short-time, rapid turning for pointing-and-
shooting, or for a proper escaping maneuver from, say, a surface-to-air missile enve-
lope (cf. Figs. 11-2 and 11-3).
Nevertheless, such maneuvers do not require so high angles of attack when proper
thrust vectorization, say, during pure sideslips in the yaw plane, is used (cf. Figs II-la
and 1band parag. 11-3.2). Alternatively, higher speed maneuvers become possible
with vectorization at relatively moderate AoA combined with rapid yaw vectoring at
the proper timing (cf. Figs. 11-2 and 11-3). Restricting the vectorization method, as, for
instance, has been designed for the F-15 S/MTD, may, however, increase the load fac-
tor at low supersonic flight conditions (cf. Fig. V-9).
40 Introduction

Fig. 12. Two-dimensional/Converging-Diverging (2-D/C-D) nozzles may supply the same, or some-
what higher thrust levels than those of conventional (circular) nozzles. Moreover, the thrust level may
further increase during subsonic thrust vectoring (cr. Fig. 13 below and Figs. 17 and 18 in Lecture I).
I - GEIPW design for low aspect ratio, pitch-only, pitch/thrust-reversal nozzle for an upgraded
F-16 partial-vectoring program (Lecture I).(Axi-TV nozzles may also be available).
2 - The GEIPW nozzle during unvectored flight.
3 - The GEIPW nozzle during down-pitch vectoring.
4 - The GEIPW nozzle during full thrust reversal. The venetian-type vanes are oriented approxi-
mately 45 degrees forward. Note: During the approach phase (cf. Fig. 14), the venetian-type vanes are
oriented about 135 degrees to the back, the throat remains partially open, the engine throttle is fully
open, and the diverging flaps are vectored down. This type of thrust vectoring reduces the approach
speed, and, following touchdown, also the landing distance (for the engine spool-up time required in
conventional thrust reversing, has been saved). However, the cost, weight and complexity of this kind
of thrust reversal may be prohibitive (cf. Lecture I for detail).

4. "ENERGY" VS. "ANGLES" FIGHTS

Fighter pilots of the F-4 have been among the first flying-qualities tacticians who dis-
tinguish between the well-known combat payoffs of transient agility during "energy"
and "angles" fights (184).
In an "angles" fight, positional advantage is gained regardless of energy expense.
In an "energy" fight, the objective is to gain an energy advantage, and, then, at an
appropriate time, to trade that energy advantage for position.
According to Hodgkinson, Skow, Ettinger, Lynch, Laboy and Chody of Eidetics
International and Cord of AFWALlFIGC, Wright Patterson AFB, USAF (184), an
"angles" pilot tactician maneuvers his aircraft so that a snapshot opportunity presents
itself before he himself has spent his energy. Consequently, a typical "angles" maneu-
ver is to use altitude to sustain the time the aircraft spends close to the comer speed, so
that the "angles" tactician can maintain a small tum radius and a high tum rate, thus
achieving a snapshot.
On the other hand, a typical "energy" maneuver is a rolling scissors in which relative
Introduction 41

energy is sustained while a number of firing and disengagement opportunities present

themselves.
Following these notes the aforementioned authors introduce a number of important
criteria and definitions for transient, torsional, axial and pitch agility and for combat
payoffs of good flying qualities.

5. PITCH, TORSIONAL and AXIAL AGILITY CONCEPTS

A typical maneuver using transient agility might use high pitch rates beyond maxi-
mum lift, a missile launch, and a rapid unload and acceleration. Thus, the time scales
of engagements like these have been considerably compressed. Consequently, mea-
sures of agility or transient dynamics, which recognize an aircraft's ability to
"SUPERMANEUVER" beyond current flight envelopes for very short times, are
required.
Pitch agility, for instance, may be defined as time to pitch up to a designed AoA and
time to pitch down. There has, traditionally, been no explicit requirement for time-to-
pitch in the flying qualities requirements documents (184).
Torsional agility refers to the capability of an aircraft to change the plane of its
maneuver. Though this chiefly involves a rollng maneuver, the necessity to roll more
nearly about the wind axis at elevated AoA, or to perform a loaded roll, has led to pro-
posals to include in the definition of torsional agility times-to-bank and stop, and turn
rate divided by time to bank and stop. The latter expression is an attempt to augment a
traditional agility measure with a time function so that it would have the appearance
of a second derivative term.
Axial agility is the capability of an aircraft to change rapidly its thrust and drag.
Here the main contributors are engine thrust transient response and braking deploy-
ment time. However, current flying qualities specifications contain no requirement on
engine thrust responses (184). This brings us to an important difference between the
time lag in thrust reversing and that of yaw or yaw-pitch thrust vectoring.

6. TIME-LAGS IN THRUST VECTORING

The current engine specification - MIL-E-5007 - now contains only a requirement for
engine response to reach 95% of commanded thrust in five seconds. Since the afore-
mentioned definitions of agility strongly depend on this requirement, especially for
vectored aircraft, it is imperative to re-examine this specification as well as the near-
term, technology limits in t:·.1S field.
To start with one must stress the following points:
1) - To consider only the nominal time-lag for idle-to-maximum power transient may
be highly misleading, for such engine time-lags increase with altitude, inlet distortion,
nozzle position, engine life, etc.
42 Introduction

Ag

NPR >NPRc:

Se-paratlon
Bubble-

Enhance-d for NPR>NPR

1).,.0
[C'tJC'II]s.,s0
Supe-rsonlc
Domain 6:;!o

loweor turning losseos NPR<NPRc

in the-
Subsonic Domain [Cti> [Ct~
.. 0 6,,-0
(;y "'0
Sy ~O

Fig. 13. Thrust Vectoring May Increase the Subsonic Thrust Coefficient Beyond That of the Unvec-
tored Values.
During thrust vectoring, at a given value ofNPR, one must keep the values ofAs andA9 as a function
of (cos /iv) x (cos /iy) (cf. Lecture I, eqs. II to 13).
For NPR > NPR (critical), i.e., in the supersonic domain of the nozzle flow field, the expansion bub-
ble just downstream ofthe lower throat corner (in the specific case depicted), lowers the value of the
"effective" NPR. Consequently, the thrust coefficient during supersonic vectoring may be lower than
that for unvectored operation.
Nevertheless, for NPR < NPR (critical), the vectored thrust coefficient may be higher than that for
unvectored operation. This may result from the higher payoffs of the "straight" flow passing the upper
corner, than the (subsonic) losses associated with the lower corner. Indeed, some of the experimental
results reported in Lecture I, as well as some of those obtained by this laboratory, substantiate these
effects.
It should also be noted that during pitch vectorization the proper throat cross-sectional area moves
from 8 to 8', and the proper exit area, from 9 to 9'. Similar transformations (in the plane perpendicular
to the drawing page) take place during yaw vectoring (Lecture I). These effects require proper
variations/adjustments in all flap positions during pitch, yaw, or simultaneous yaw-pitch vectoring
(Lecture I, cf. the proper control rules for yaw-pitch thrust vectoring as given by eqs. I to 14).
Conclusion: Pitch-only, or Yaw-Pitch Vectoring Nozzles May Supply the Airframer With Approxi-
mately the Same, and in Some Cases Even Higher Thrust Levels Than Those Available From Conven-
tional (Circular, Unvectored) Engine Exhaust Nozzles.
 Introduction 43

2) - In design optimization an airframe manufacturer can trade off engine life for
improved axial, pitch and torsional agility capabilities.
3) - Yaw, pitch, or simultaneous yaw-pitch-roll thrust vectoring responses are faster
and more reliable than those for thrust reversing. This conclusion must be added to,
and combined with another conclusion discussed in this volume, namely, that the air-
frame designer can extract significantly higher combat-agility payoffs from yaw-pitch
vectoring, than from, say, pitch/thrust-reversing.

7. THE OVERALL COMBAT PICTURE

Unfortunately, the aforementioned conclusions may form only a slice through the
overall picture of many other practical factors such as: scenario size (e.g., number of
friendly aircraft vs. number of enemy aircraft, their initial positions, attitude, veloci-
ties, left-over weapons, etc.), low observability until WVR (Within Visual Range), the
technology levels of helmet-sight-aiming and EW systems on both sides, combat-
management methodology/systems used, the varying technology limits of all-aspect
missiles, the need to evade a missile launched previously by a now-destroyed oppo-
nent, etc.

Fig. 14. Takeoff, Approach and Landing of Vectored Aircraft.

Thrust vectoring during takeoff may be performed by first turning the jets upward, and, following nose
lift-up (rotation), turning the jets down using IFPC-FBW-yaw-vectoring system eliminates the conven-
tional need for vertical stabilizers. PV A lift is enhanced by the "direct lift" of the engines, as well as by
circulation (cf. Lecture V).
(below) Approach and landing. For detail see Fig. 12.
Good ground handling requires up to 45 degrees yaw thrust vectoring.
44 Introduction

8. THRUST VECTORING AND POST-STALL CAPABILITY IN AIR

COMBAT

Philippe Costes, of ONERA, has recently published the results of numerical simula-
tions of one-to-one, close-range, air-gunnery combat maneuvers, from two initial con-
ditions: High and low altitudes (182). His results show that thrust vectoring
considerably improves the aircraft maximum turn rate. Stating that such super-
maneuverability is a keypoint to the design of advanced combat aircraft, Costes con-
cludes that thrust vectoring is especially effective in close-range combat, below the
corner speed, and that its use, during post-stall maneuvers, allows a much quicker nose
pointing, and, consequently, first-shootig opportunity. For instance, the probability of
victory by vectored aircraft at high-altitude air combat is 0.78 vs. 0.22 for similar, but
unvectored aircraft. This gives the vectored aircraft a Kill-Ratio of3.55 (cf. Table 1).
During close-range dogfight, the probability of victory by vectored aircraft becomes
0.89, while that of the conventional combat aircraft is only 0.11. This gives the vec-
tored aircraft a Kill-Ratio of 8.1, under these conditons.

Table 1. Kill-Ratios: Vectored Vs. Conventional Aircraft

(After Costes, [182])

Type Altitude Probability

Kill Ratio
of (m) of
Pv(i)IPv(j)
aircraft Speed (M) Victory

Model A: Conventional 108000 Pv(A)=0.522

fighter aircraft M=0.9 Pv(A)=0.477 1.09
Model D: Vectored 10800 Pv(D) =0.78
fighter aircraft M=0.9 Pv(A)=0.22 3.55
(with PST capability)

Model D: Vectored 1500 Pv(D)=0.89

fighter aircraft M=0.5 Pv(A) =0. I I 8.1
(with PST capability
at close-range dogfight)

It should be noted that these simulations are limited to zero sideslip flight condi-
tions. Moreover, these models do not incorporate yaw-vectoring simulation capabil-
ity. Hence, these results should be viewed as the minimal baseline results for
pitch-only, partially-vectored aircraft.

9. MINIMUM TIME TURNS USING VECTORED THRUST

Schneider and Watt, of the US Air Force Institute of Technology at the Wright-
Patterson AFB, have recently published an interesting study (218) about optimal con-
Introduction 45

~." " »;; >;;;; >:. ; ;; ;; ;;

n rr-r/77"r/7"7

Fig. 15. The Simultaneous Use of Yaw Vectoring -T- (or "Targeting") with Pitch/Thrust-Reversal
-(VR)- (With and Without PST, or PSM), May revolutionize Air-to-Ground and Air-to-Air Multiple-
Target Performance. It should also be stressed that R is a non-essential option (cf. Fig. 3 and Lecture
II).

troIs and trajectories which minimize the time to turn for high performance aircraft
with thrust vectoring capability.
No constrains were placed on the angles through which the thrust was vectored in
order to determine how much range of thrust vectoring would be exploited ifit were
available. The determined controls and trajectories are then compared by the authors
against other methods of turning in minimum time, to determine the effects and
advantages of thrust vectoring.
The results obtained by Schneider and Watt indicate that the use of vectored thrust
to supplement the aircraft's lift, by directing the thrust into the turn, can substantially
reduce turning times and increase in-flight maneuverability. The greater the velocity at
which the turn is initiated, the larger the vectoring capability becomes, and the greater
the reduction in turning time.

10. VECTORED AIRCRAFT AND LOW OBSERVABILITY

To what extent should low observability affect the design of vectored propulsion/
airframe, and vice versa? Should low detectability be the over-riding principle in the
design of advanced, vectored propulsion system? Obviously, wing-integrated, high-
aspect-ratio, yaw-pitch-roll, vectoring nozzles/propulsion systems, help reduce ReS/
IRiOptical signatures, especially in the absence of vertical stabilizers (cf. Fig. 1).
46 Introduction

Designing engine inlets and nozzles to begin and to end, respectively, on the upper
wing skin, can reduce detectability from below.
Reduction of RCS by geometrical control of the winglbody shape may follow two
methodologies:
I) - Aircraft external geometry is a combination of many plane surfaces.
2) - Aircraft external geometry is dominated by constantly-changing radii of
curvature.
Thus, a design feature of the stealth F-117 A fighter aircraft, unlike that of the stealth
bomber B-2, is the absence of curves. Composed of many plane surfaces (cf. Fig. VI-I),
this plane drastically reduces the number of main-lobe reflections that a radar will be
able to seek, and track, during a given time period. Since the aircraft vibrates as it flies,
the probability of an enemy radar to pick-up an echo from one of these plane surfaces,

to maxImum
MckM,
(4)

TR+CSv
( 2)

Fig. 16. A Schematic Representationofa Rapid, 180-DegreesPST/PSM, Tum-Back Maneuver, Using

Simultaneous Yaw-Pitch-Roll Thrust Vectoring, Which Significantly Reduces Turn-Time Without Loss
of Energy (head-on encounters, or an escape supermaneuver from a SAM envelope).
States:
( I) - Maneuver initiation at speed MI.
(2) - Vectored, PST, nose-liftup against the gravitational vector (with or without the option ofthrust
reversal).
(3) - Aircraft is fully-controlled by vectored yaw-pitch jet-moments/forces in the post-stall domain
(Fig. 2).
(4) - Yaw vectoring, at low-speed, rapidly turns the aircraft nose down for the beginning of a rapid
acceleration downward (with the addition of the acceleration due to gravity). The timing of this yaw
(or pitch) vectoring is determined by the corresponding maximum load-component values allowed for
pilot and aircraft. The entire maneuver is carried out at full engine throttle. Note: The vectored fighter
aircraft is momentarily at a "sitting duck" situation. However, the conventional fighter still has its nose
pointing away from the vectored fighter. It should be stressed, from the very beginning, that this state
is a highly risky one in target-rich environments. Consequently, it may be considered only for lvi,
close-combat encounters, or for a short radius/rapid escape maneuver from, say, a SAM envelope.
(5) - Back to the initial speed/energy level, while saving about 4 seconds in comparison with a conven-
tional fighter aircraft (see Fig. 17).
Introduction 47

on two or more successive sweeps, are very low. Consequently, the probability of
detection, or the "averaged" RCS of this stealth aircraft is very low.
On the other hand, the B-2 shape (cf. Fig. VI-l) is constantly curved, so that it scat-
ters the electromagnetic waves over different spatial directions. Thus, in principle,
each point on the B-2 surface reflects a low-intensity electromagnetic radiation into a
different spatial direction. It should be stressed, however, that RCS depends on the
illuminating radar-wave length, and on the detector vs. aircraft-attitude angle, as well
as on numerous other factors (see below).
Adding to this design philosophy the elimination of vertical stabilizers, straight-line
wing-trailing-edges, straight-cut inlet lips, etc., and the absence of rudders, or
90-degrees-corners [cf., e.g., the B-2 all-wing shape], as well as the inclusion ofburried
engines, weapons and payloads, semi-burried, upper-wing, s-shaped engine inlets
(which also provide engine protection against bird ingestion), and fully-recessed, 2D
engine nozzles (cf. e.g., Fig. VI. I), further reduces RCS/IRloptical signatures. In fact,
the 2D nozzle's stealth benefits are greater than has been widely supposed. The 2D
nozzle also affords the use of radar-absorbing structures that cannot be incorporated
when the engine has a circular, axisymmetric nozzle.
Then, one may construct the aircraft primarily of RAM-composites, and especially
of carbon-based composites (cf. e.g., the RCC properties as given in Appendix B, Fig.
1), while eliminating material and surface discontinuities, improperly-designed
antennas, etc., and including special surface smoothness characteristics and skin treat-
ments, as well as highly-integrated active-passive-EW systems, adaptive management
of power, the use of internal, very-Iow-sidelobe antennas, constant variations in fre-
quency and waveform, and such internal, saw-type, wing-Ieading-edge-structures as in
the SR-71.
Thus, by employing such an over-riding design philosophy, sometimes in contradic-
tion with classical aerodynamic principles, one may further reduce detectibility.
eJI systems of stealthy aircraft also bring more integration of active and passive EW
systems across a broad frequency spectrum as well as of radio-frequency, electro-
optical, infrared and optical sensor systems. Battle management, flight/propulsion
control and intelligence gathering must also be well-integrated, using non-radiating
systems and a Common Integrating Processor - CIP.

10.1 RCS/IR Databases for Vectored Aircraft

Three different methods are being used to evaluate the RCS and IR signatures of a fly-
ing platform:
I) - Actual measurements from various viewing angles - a highly costly and time-
consuming method, especially for a flying RPV, or for a prototype. (Examples of the
temperature profiles inside hot jets emerging from high-aspect-ratio, two-
dimensional, engine exhaust nozzles, are given in Figs III-8 and 9. Note also, that even
the mixed, relatively low-temperature jets leaving the (ATB) B-2 propulsion system,
48 Introduction

SO
,j."
'A
Vm/S I '\ -Veoctoreod Aircraft
---Reogular Aircraft
~~
70

'0
'0
40 \
SO 00 '-L I-~-
20/-/' .....
-~

loa
80 ~ \ A'
'a / r----~ / '
'-..:..: At= 16s«
1,0
20 J
30
\
1-----
'"
20
V_oo...-·
10
I 10 IS ?J) t,seoc

.." .- ...... .

M.0.3 H -2000 m

Fig. 17. The Herbst PST Variations of AoA and Speed of the Vectored Fighter Aircraft Shown in Fig.
16, in Comparison With Those for a Conventional Fighter (upper drawing).
About 3.6 seconds have been saved. During the second leg of this supermaneuver, the vectored
fighter aircraft can acquire the conventional fighter in the missile off-boresight angle, while the reverse
aquisition, by the conventional fighter, is impossible.
(lower drawing) - The penetration radii for various AoA maneuvers as those depicted in Fig. 16
(after Herbst, 154).

are still detectable (but from a much-shorter range), by heat seeking sensors.)
2) - The measurements of static, scaled-down models - a method which may intro-
duce errors, but is much less expensive, or time-consuming. When scaled-down mod-
els are used for static or dynamic RCS measurements, the frequency of the
illuminating source must be increased to correspond to the scaled-down-ratio of the
model.
3) - Computational modeling - a method suffering from similar problems as those
associated with scaled-down modeling, plus the difficult task of exact modelling such
Introduction 49

parameters as curved, or multi-flat shapes, internal RAM/structural systems or

engineS/inlets/nozzles/jets, as well as material and surface discontinuities, engine inlet
lips (cf. Chapter VI), engine fan/compressor/turbine blades, weapons, payloa<Js, verti-
cal stabilizers, radoms, pilot canopies, aircraft deflected flaps, elevators, rudders, etc.,
including those contributed by passive and active antennas and various EW systems.
Obviously, there is no point to reducing aircraft RCS ifit is allowed to radiate (say,
by using improperly-designed, active, EW systems), or if it carries high-RCS weapons
and antennas.

10.2 Stealth RPVs

Currently, there are many low-flying, long-loitering, unvectored, EW-RPVs. Obvi-
ously, whatever can be well-done by a propeller-driven RPV, or by (unvectored) jet-

I
I
I
I -Ty
TR
-Tz
Fig. 18. Expanded Performance Envelopes Cor Vectored Fighter Aircraft.
50 Introduction

propelled RPV, does not require the extra capability of a vectored RPV. This leaves
out only a specific family of missions which require, at least in the final mission-
targeting phase, a highly-maneuverable, vectored RPV, including the R&D-type of
vectored RPVs discussed in Lecture IV. Another basic requirement may be VTOLI
stealth capability. The last topic is intended for Volume II.

11. THE TIME GAP BETWEEN NEW PROPULSION SYSTEMS

AND NEW e3I SYSTEMS

There is a time-difference between the pace of evolution, and maturity, of propulsion

systems, and that of integrated C3I systems. While the former changes about every ten
to twelve years, it takes the latter only five to six years to mature into a new technology.
This means that the designers of advanced airframe/propulsion systems can complete
the integration process of their system with advanced C3I systems, only during the
second half of the "evolution leg" of each new generation of advanced platforms/
propulsion systems. This constitutes a practical reason for the need to address the
design problems ofIFPC and low-signature systems only later, in a second volume, or
in a second edition to this volume. It also means that flight testing of vectored RPV s
may accelerate the build-up process of IFPC data bases.

12. WELL-INTEGRATED LABORATORY/FLIGHT-TESTING

METHODOLOGIES

The evolution of vectored-propul sion/vectored-aircraft requires the development ofa

well-integrated laboratory and vectored-RPV flight-testing methodology for data
extraction. Such a methodology is currently being developed by this laboratory. It con-
stitutes three, highly-integrated, feed-back phases:
1) - New ideas, or design modifications emerging from flight testing, are tested on a
small (about 1 kg/sec airmass flow rate), highly-instrumented, thrust-vectoring-nozzle
test rig (cf. Fig. IV-7), or on a PST-inlet test rig of a similar size.
2) - Pre-optimized thrust-vectoring nozzles, or inlets, emerging from phase one are
scaled-up and installed on a full-scale jet engine. Thrust-vectoring engine tests are then
carried out in a highly-instrumented (altitude) engine test rig. This rig provides pre-
optimization metrics during unvectored baseline tests, and during pitch, yaw, yaw-
pitch-roll thrust vectoring tests. All tests must be conducted with a proper engine inlet;
e.g., a standard bellmouth, or a Stealth/PST-tailored inlet. Thus, a complete Stealth/
PST-vectoring propulsion system must include thrust-vectoring nozzle well-
integrated with the engine/wing/inlet/I FPC systems, and tested under various
combinations ofNPR, altitude, air-speed, AoA, sideslip, and vectoring angles (cf. Fig.
IV-6).
3) - Pre-optimized nozzles emerging from phase two are scaled-down back to the
Introduction 51

• BASED ON EXISTING COMPARISONS

• F-15/F·16 VI F·4/F·5
.10,000 FT
• 50% FUEL

s
-tr--;±:==I:::::::;1 ~ II--------II--i
o - 101 liS

f--+r+-l~ : ;:: ~~~

o - 100 ITS
IlUO RAil 1.0
ITSIUt

Fig. 19. The McAtee Dynamic-Speed-Turn (DST) plots are Useful to Evaluate Fighter Aircraft
Maneuverability.
The combat tasks of generating higher turn rates (left figure), retaining residual maneuverability
(middle figure), and accelerating rapidly (figure at right), are relevant to the definition of agility
requirements (cf. parag. 3.5). However, good maneuverability must be integrated with effective con-
trollability, i.e., the ability to change states rapidly (control power), and the ability to capture and hold
a desired state with precision (handling qualities) (196).
Traditionally, controllability was thought to be degraded at either of two conditions: High Mach
number, or high AoA (cf. Fig. 20). However, the introduction of PST, and vectored-aircraft technol-
ogy, requires reassessment of the second condition. It also requires the introduction of new standards
and MIL specs for vectored engines, nozzles, inlets, agility metrics, and IFPC.

phase-one size, and installed on the proper powerplant of a vectored-RPY-prototype

of a vectored aircraft equipped with PST-inlets.
This phase includes flight-testing of thrust-vectored RPYs [which, in turn, have
been designed "around" the newly-developed, yaw-pitch-roll, propulsion systems -
cf. Figs. 1 and 11). Alternatively, they include flight testing of, say, 1/7-th scaled-down,
highly-instrumented, radio-controlled, vectored RPY-models of existing fighter air-
craft [e.g., a vectored F-15, or a vectored F-16, which are both being considered by the
USAF for various vectored-propulsion upgradig programs).
Using specially-developed, onboard-computers for vectored-propulsion/airframe
data extraction/recording, the aforementioned metrics are extracted during well-
defined thrust-vectored maneuvers, including PST-maneuvers. Performance-metrics
of the unvectored aircraft models are employed as Baseline-l (for a comparative study
of the payoffs obtainable with various vectored propulsion designs).
Secondly, pitch-only metrics are established as Baseline-2. Thus, using, for instance,
vectored F-l5 and vectored F-16 prototypes, Baseline-2 may constitute the metrics of
the current GEIPW-2D-CD "pitch-only" nozzle technology. (Such nozzle designs are
now installed on the F-l5 S/MTD, cf. Lecture Y.)
52 Introduction

____ ~ ROll RATE

~~A8 IAboll' Slitlilily a.ill

Ila. I

111.1 /
/
V
110.0 IOU UII

YAW UII 10'1' ,.<1 IGo.o

ID',' $IC!,'SY
/
V V'ID

---
10.0 t - - lOLL 'IItN CDU'Ula
'11LLlIlIT /. .

~ po--/
11.0

~
10.0
V/
G.D ~ I..---
1.0 \1.1 10.1 11.1 4U 11.1
uell Of ATlJCI'O...

Fig. 20. Pitch and Yaw Control Requirements Increase with AoA.
For a given roll rate, as AoA increases, the requirements for pitch and yaw forces/moments (for non-
thrust-vectoring aircraft), increase exponentially. At the same time, with conventional aerodynamic
controls, the forces/moments available decrease as airspeed decreases. Thus, beyond a given limit,
conventional-control technology becomes obsolete. This technology limit is reached, when the size
and weight of the aerodynamic control surfaces needed to provide sufficient forces/moments, become
prohibitive. (After McAtee ofGD, 196). However, the introduction of PST and vectored aircraft tech-
nology (together denoted by McAtee as the new domain of "supercontrollability"), requires reassess-
ment of all maneuverability and controllability concepts and requirements.
Thus, according to McAtee, new point-and-shoot weapons have reduced engagement times drasti-
cally, leaving aircraft with poor maneuverability and controllability, at the mercy of those that can use
their agility to point-and-shoot quickly during close-in combat (cf. Fig. 2, Lecture II).

Finally, simultaneous yaw-pitch-roll metrics are established, and compared with

Baseline-2 metrics, or with ETV metrics. Employing our newly-developed data extrac-
tion techniques for thrust vectoring, the next optimization step sometimes includes a
return to phase one, and repetition of the whole cycle (see also the figure depicted in
the Preface, and Lecture IV).

13. NEW EDUCATIONAL SHIFTS

The USAF Test Pilot School has recently revised its curriculum to place more empha-
sis on highly-integrated military aircraft entering the near-term active inventory. Con-
sequently, the classical flying-quality courses that had dominated the traditional
education of pilots, engineers and navigators, have now been significantly com-
pressed. Newly designed courses which integrate the various aspects of system engi-
neering with flying qualities and Command, Control, Communication and
Intelligence (C 3I) systems, have been introduced instead. This change is probably the
 I..--.-------it: ~
Introduction 53

'..,----------------"i:).3 ....,
w
U
........."....-----------iJ.2 ~
LL

TR
TR

2 3 4 5 6 7 8 9 10 1i 12
ALTITUDE r Km. )

Fig. 21. Flat Plate Drag Vs. Reversed Thrust Variations with Altitude and Speed: PST breaking is more
effective thanTR.

first important milestone in the educational shifts that are expected to take place in all
academic institutions whose curriculum deals with advanced aviation.

14. CONCLUDING REMARKS

- The availability of PST vectored fighters, helmet-sight-aiming systems, all-aspect

missiles and the new generation of E W systems, require reassessment of the optimal
balance between aircraft agility and effectiveness, and the agility and efectiveness of
missile/helmet-sight-aiming systems (§ 3.5).
- Whatever is the aforementioned balance, high-performance fighter aircraft will
gradually be based on improved thrust vectored propulsion/maneuverability/con-
trollability (Lectures I, IV and V).
- New point-and-shoot weapons have reduced engagement times drastically, leav-
ing aircraft with poor maneuverability and controllability at the mercy of those that
can use their agility to point-and-shoot quickly during close-in combat (§ 3.5).
- Since future fighter aircraft would be thrust-vectored, and since thrust-vectoring
engines would be used for enhanced maneuverability and controllability, as well as for
brute-force propulsion, one must first define new propulsion concepts and new meas-
urable "metrics", which would be employed in a realistic comparison of vectored air-
craft maneuverability-controllability with that of conventional fighter aircraft
(Lectures I, V).
54 Introduction

- The evolution of this highly-integrated propulsion technology depends on the

gradual replacement of partially thrust vectored aircraft by pure vectored aircraft.
- Controllability and supercontrollability of pure vectored aircraft is much safer
and easier than that of partially vectored aircraft.
- Maneuverability and supermaneuverability of pure vectored aircraft is considera-
bly superior to those obtainable with partially vectored aircraft and conventional
fighter aircraft (Lectures I to VI).
- In-flight thrust reversal is not required in PST/RaNPAS maneuvers (Fig. 21). It is
much less effective than nose-up, PST/thrust-vectoring deceleration, and is, therefore,
adding unrequired weight, complexity and cost to the vectored propulsion system and
to the aircraft. Nevertheless, it provides unmatched short landing runs.
- Pure sideslip thrust vectoring maneuvers constitute RaNPAS-PST/PSM method-
ology which does not dissipate as much energy as a similar, pitch-only, PST maneuver
(§ 3.2). Further investigations should assess this option in detail.
- Aerodynamic Flight Control is to be gradually replaced by Thrust-Vectoring
Flight Control. Pitch and yaw control requirements increase with AoA. For a given
roll rate, as AoA increases, the requirements for pitch and yaw forces/moments (for
non-thrust-vectoring aircraft), increase exponentially. At the same time, with conven-
tional aerodynamic controls, the forces/moments available decrease as airspeed
decreases. Thus, beyond a given limit, conventional control technology becomes obso-
lete. This technology limit is reached when the size and weight of the aerodynamic
control surfaces needed to provide sufficient forces/moments become prohibitive.
- The introduction of PST/PSM-vectored-aircraft technology requires reassess-
ment of all previous maneuverability and controllability concepts and powerplant
definitions/requirements (§ 3.2).
- In line with the new concepts of RaNPAS-PST-PSM-Vectored-Aircraft, the
aeroengine and airframe communities must search now for propulsion/airframe sys-
tems with the following concepts in mind:
(i) - A common set of engine-related agility-performance "metrics", or measurable
thrust-vectored maneuverability and controllability parameters (Lectures I-III).
Such metrics may be presented as three-dimensional depictions of powerplant
steady-state or dynamic responses, somewhat similar to those proposed recently to
depict aircraft agility. These depictions may include specific throttle/pitch/yaw/roll
transient capabilities for any practical combination of thrust-vectoring flight condi-
tion, i.e., one metrics map for each performance set of throttle, yaw-pitch-roll settings,
NPR, NAR, Y, D, T/W, AoA, sideslip rate, inlet geometry, speed, altitude, maneuver
mode, IFPC mode, etc.
(ii) - The cargo and civil aircraft industries may exploit some of the military metho-
dologies of vectored propulsion and controllability, for instance, by introducing low-
drag, cost-effective, STOL, pure-vectoring design methodologies and flight
training. Advanced TV systems can also increase flight safety.
(iii) - In the military domain, the aforementioned definitions, examples and con-
clusions form only a slice through the overall picture of many other relevant factors
Introduction 55

such as: scenario size (e.g., number of friendly aircraft vs. number of enemy aircraft,
their initial positions, attitude, velocities, left-over weapons, etc.), low observability
until WVR (Within Visual Range), the technology levels of EW systems and of helmet-
sight-aiming-systems on both sides, combat-management methodology/systems used,
the varying technology limits of all-aspect missiles, the methodology to evade a missile
launched previously by a now-destroyed opponent, fighter-strike missions, etc. (§ 3.5).
(iv) - One must also stress the importance of upgrading existing fighter aircraft hav-
ing (at least) TIW> 0.6. High-aspect-ratio, split-type, yaw-pitch-roll, vectoring noz-
zles, have the potential of replacing the horizontal and vertical tails of existing fighter
aircraft. Thus, by adding FBW, or FBL, to such propulsion systems, one may discard
the vertical stabilizer, and the elevators of, say, the F-15 and F-16 aircraft, and replace
them with "tailless", efficient, low-cost, yaw-pitch-roll vectoring nozzles (Lectures
IV-VI).
(v) - Integrating the propulsion, control, avionics, weapons, and airframe elements,
to maximize performance, must become the central goal of well-integrated R&D&T
strategies, and of new tactics for fighter pilots. Altogether, these new concepts amount
to a revolution in the educational process of engineers, and of pilots (§ 13).
(vi) - New integrated testing/design methodologies, involving cost-effective, inte-
gratedjet-propulsion-laboratory/vectored-RPVs-flight testing, may help advance this
field (§ 12).
(vii) - Automatic, "vectoring-PST-inlets" must be developed and laboratory/flight
tested for each PST aircraft, namely, for upgraded F-16, F-15 and F-18, for PV As, and
for post-ATF vectored aircraft.
(viii) - The 2D nozzle's stealth benefits are greater than has been widely supposed.
It also affords the use of radar-absorbing structures that cannot be incorporated when
the engine has a circular, axisymmetric nozzle. Thus, the STOL and super-
maneuverability benefits attainable by yaw-pitch 2D-ITV nozzles may be viewed only
as additional advantages to the provision for stealth that yaw-pitch, 2D-ITV nozzle's
blended integration into the airframe provides.
The "Why?" and "How?" of this forthcoming change are described in the main text.
 LECfURE I

FUNDAMENTAL CONCEPTS
REVISITED AND REDEFINED:
PRELIMINARY NOTES

"With the help of physical theories we try to find our way

through the maze of observed facts."
Leopold Infeld and Albert Einstein

"Revolutions are ambiguous things. Their success is gener-

ally proportionate to their power of adaptation and to the
reabsorption within them of what they rebelled against."
George Santayana
1.1 New Missions and Modified Design Trends
Traditionally the engine has been considered to have little influence in the flight-
control design. Consequently, aircraft dynamicists tended to develop sophisticated
airframe dynamic models in conjunction with only a rudimentary model for the
engine.
This was the so-called "Big-Airframe, little-engine" approach.
On the other hand, engine manufacturers had traditionally used the opposite
approach, almost ignoring the best integration methods that might be required by
future airframe designers.
The recent introduciton of new, vectored-airframe missions and performance has
drastically changed these attitudes (see below).
Almost suddenly it was realized that there is no unified approach, nor integrated
design tools and criteria to handle the new problems properly.
Simple additions of the two technologies, in some linear simulation methods, sim-
ply giving equal favor to both, have been quickly found to be inadequate, or even
misleading.
A new, really integrated methodology, had to be developed quickly, and, appar-
ently,from no verifiable base oflow-risk technology. This, in tum, was to revolutionize
the mode of thinking of pilots and of mission decision-makers, as well as the entire
approach to aerospace engineering education and practice.
In this lecture we first examine the origin of early conceptual design limitations. We
then review a new class of (patented) vectoring components and their inherent new
type of unit operations and potential new missions. While the new vectoring compo-
nents are quite simple in conceptual design (some would even say "trivia/,), their
recent introduction to the aviation industry has generated a substantial revolution.
This revolution has its origin in the earlier introduction of the well-known Harrier-
type technologies. However, the new revolution in thinking, research, design and test-
ing, has already generated its own momentum, with its own terminology of
fundamental concepts, design methodology, performance and mission definitions.
Consequently, we begin this lecture by revisiting and redefining the fundamental
concepts associated with the new vectoring nozzle components and their associated
57
58 Fundamental Concepts

unit operations. In the next Lectures we shall proceed to definitions and classifications
of cold propulsion, pure vectored aircraft, internal and external thrust-vectoring sys-
tems, supercirculation, and robot aircraft.
During this course we shall also examine the main "60 variables" associated with
vectored propulsion, and, especially, with integratedjlight-propulsion modes ofdesign,
control and performance payoffi.
Special attention will be given to the recent introduction of a new research tool: The
vectored RPV.
In tum, the newly designed vectored RPVs, or in their global engineering meaning,
'robot aircraft' (including Advanced Cruise Missiles (ACM) and drones), have already
generated a whole new class of special missions; missions unattainable by conven-
tional propeller-powered RPV s, or by current-technology, jet-powered robot aircraft.
This particular revolution in the design, performance and mission definitions of
robot aircraft will be examined in Lecture IV.
We tum now to the first subject, namely, the inherent prejudice toward innovative
designs in vectored propulsion and vectored aircraft.

1.2 Early Conceptual Design Limitations

Military aircraft exhaust nozzles have evolved from fixed, subsonic, convergent types,
to supersonic, variable-area, convergent-divergent systems. Traditional thinking,
combined with some nacelle/fuselage structural limitations of specific aircraft, have,
so far, dictated the restrictions of these nozzles to a circular cross-section and to a sim-
ple, unvectored, brute force 'forward'.
This situation has been changed in recent years, mainly because new survivability
improvements, over current-technology aircraft, have been stressed in a number of
categories:
• Injlight thrust vectoring.
• Drastic reductions of takeoffand landing distances by thrust vectoring and by thrust
reversal.
• Enhanced maneuverability through simultaneous roll-yaw-pitch thrust vectoring.
• Significant reductions of minimum jlight speeds by the combined unit operations of
vectoring (see below).
• Low observability (i.e., low RCS, optical, noise, and IR signatures associated, in
part, with this new technology).
Engine manufacturers had first tried to meet these requirements by simplified addi-
tions of existing, proven components. The result was large increases in weight and com-
plexity, and/or substantial reductions in performance.
Such early designs resulted in reduction in expected performance envelopes and
increases in weight and complexity. Even nowadays some designers simply add such
proven components as wedges, reverser clamshells, gimbal rings and plugs, while
Lecture I 59

retaining the old-fashioned control rules of current-technology exhaust nozzles and

powerplants.

1.3 Internal Thrust Vectoring (lTV)

It was in 1976, that PWA obtained a patent (No. 3,973,731) on a 2-DIC-D thrust-
vectoring nozzle shown in Fig. 1. This design contains only four major moving parts.
Yet it provides the best overall performance when yaw vectoring is not required. How-
ever, as was stressed in the Introduction, simultaneous yaw-pitch-roll thrust vectoring
is a basic requirement of all advanced, PST/PSM-RaNPAS, vectored aircraft.

-
(A)
r::b
fQlWAlO JUGHl

- -'---

-
1M.US' .IW.,tNG

Fig. I. The vectoring nozzle shown here is termed two-dimensional, converging-diverging, or 2D-CD. It
allows pitch vectoring and thrust reversing without thrust yawing. It may also be referred to as TVITR
nozzle. (See also Fig. 3)
60 Fundamental Concepts

- ------- -------~--.-----
Aa

(D)

-\I,

Pitch plane primary thrust vectoring of ± 20 degrees is available at all engine power settings, includ-
ing maximum AlB.
Independent controls of nozzle throat area, As, (see Fig. 11 for definitions) and nozzle exit area, A9,
are also provided, resulting in optimum area ratio matching for maximum internal performance (cf.
Fig. II and eqs. 5 and 6).
Lecture I 61

Closure of As by the convergent nozzle flaps, and simultaneous porting of the exhaust through
venetian-type, vectoring/reversing vanes, is available with engine throttle settings up to, and including
"Intermediate Power (or "full dry power").
Actuation of nozzle surfaces is accomplished with 6 actuators (3 inside each sidewall as shown in
Fig. 8). Two additional ones are required for the yaw vanes control in our new yaw-pitch designs.
Average actuation rate is 30 degrees per second. Nozzle reversing is possible throughout the dry
engine power range (excluding AlB thrust reversal). However, this design has been modified for the
F-15 S/MDT (cf. Fig. 3).
The porting of the exhaust flow through rotating vanes (Figs. 4 and 7) allows a wide range offorce
control versatility in the nozzle pitch plane, including axial force modulation to achieve longitudinal
moment. acceleration and velocity control. especially during landing approach phases (Fig. 11-6).
Reverse thrust is achieved while maintaining the engine throttle at the "intermediate" (or "dry")
power setting, or less. Up to 67% of the "forward dry" gross thrust is available for rapid (less than I sec)
deceleration on the ground and in flight.

Legend
I) Circular engine duct.
2) Circular-to-rectangularduct (this section may become AlB, as for instance, in Figs. 13 and 14).
3) Central hinges of converging flaps.
4) Nozzle throat hinges.
5) Diverging flaps (vectorized only in the pitching mode).

Subfigures:
A) 0,.=0 at full military power setting.
8) 0,.';'0 at full military power setting.
C) Full thrust reversal deployment.
D) Full AlB at 0,.=0.
E) Full AlB with pitching vectoring (throat area should be increased.
G) Sidewall, nozzle control actuators (0,.=0).

Notes:
I. Sidewalls may be truncated as in Figs. 15 and 16.
2. Throat area during vectoring is defined in Fig. II, and its control adjustments by eqs. 5 and 6.
3. For yaw thrust vectoring see Figs. II-I and 11-2.
4. Cooling methodology is illustrated in Fig. 8.

Legend for Fig. G:

5) Converging nozzle flaps actuator (symmetric mode oflinear displacement during VR flight con-
trol with I FPC).

6) Uper diverging flap actuator controls (0,.- a) pitch vectoring angles in VR-IFPC modes of control.
7) Lower diverging flap actuator controls (o,.+a) pitch vectoring angles in VR-IFPC modes of
control.
8) The design of this section may va~v considerably. First and foremost it includes circular-to-
rectangular ducts, _streamlined flow-dividerslstructural struts, etc. [Figs. 13 and 14]. In some
designs it includes Venetian-type vanes which are ful(v controlled during approach-landing (cf.
Fig. 7) and in air-to-air and air-to-surface maneuvers.
 62 Fundamental Concepts

The General Electric Co., PW A, and other research bodies, have since improved
this patent by various means, including the use of Venetian-type vanes during thrust
reversal (Fig. 7). Other improvements, including an Israeli Patent Application from
1987 (No. 80532), have introduced simultaneous yaw-pitch vectoring vanes/stators
inside the converging, or converging-diverging types ofexhaust nozzles. Combined with
numerous other improvements, to be described later, the introduction of simultane-
ous yaw-pitch-roll vectoring, or even yaw-pitch-reversal vectoring, has opened the
way to the design of pure vectored aircraft whose dependence on aerodynamic flight
control surfaces is minimal. Consequently, aircraft capable ofsuper-agility at post-stall/
PSM conditions may now become a real possibility.
The aforementioned inventions and improvements are readily adjustable to fit dif
ferent engine-cycles, or aircraft installation requirements. Their performance payoffs
begin at around 0.6 thrust-to-weight ratio. Hence, many older fighters may become
candidates for becoming vectored aircraft.

1.4 External Thrust Vectoring (ETV)

Figs. 4 and 5 show a comparison of yaw-pitch angle envelopes for three-vane and four-
vane External Thrust Vectoring (ETV) configurations. A comparison of lTV with
ETV is to be discussed in Lecture VI.

1.5 Subsonic-Supersonic Concepts of Engine Exhaust Nozzles Revisited

and Redefined
Fig. 1 is a schematic view of an engine exhaust nozzle during unvectored and vectored
operation. As represents the throat area, A9 the exit nozzle area, while d represents the
divergence (half-angle) deflection required to maintain optimaIA9IAs=Ac/A/=e area
ratios for achieving maximum thrust efficiency, C[~, during supersonic engine opera-
tion, i.e., for Nozzle Pressure Ratios (NPR) above a critical NPRc value.
Tables are available in standard textbooks, and in engine simulation programs, for
the calculation of e as a function ofNPR. However, as will be shown below, Cfg is not
the only coefficient to be considered for the evaluation of converging-diverging, or
converging, exhaust nozzles. For that purpose one needs at least four engine nozzle
coefficients: CfK' CDS, C A , and Cv (their definitions and physical meaning will be given
at a later stage of this course).
What should be stressed at this early stage of the course are the following prelimi-
nary notes .
• Pitch vectored, variable, two-dimensional (2D) engine exhaust nozzles, may be
formed with only four flat flaps (supersonic) [only two are required for converging-
type subsonic nozzles]. Cf. Figs. 1,4, and 7 to 14 .
• On the other hand, current-technology (axisymmetric) nozzles are formed, and con-
trolled, by a large number of movable metal flaps, which, being rotated together,
Lecture I 63

~/ 4®
\./ .':'N' ® .,.DCD

/
~
P.WAlNA~
=='~
~ \ "

"~ ~ P&WA
2·0 CO
( Reverser
Mode Shownl
/' P&WAlMCA~
Variable Incidence Plug (VIPI

Fig. 1a: A number of early-design, low aspect-ratio, vectoring nozzles proposed by GE and PWA. (Cf.
Appendix A for more details. The ones crossed were found to be less useful.)

form an almost circular (axisymmetric) nozzle. However, the large number of flaps
generate gas leakge and cooling (thrust) losses, cf Fig. 2. Moreover, they generate
velocity and thermal boundary layers, which deviate from simple 2D, or
axisymmetric forms. Consequently, vectored nozzles may demonstrate equal, or
even higher thrust efficiencies than those obtainable today with current-technology
nozzles, as shown in Fig. 3.
• During pure pitch vectoring (o,.~ 0, Oy=O), the geometry of the four 2D/CD flaps may
appear as in the scheme shown in Fig. 1 I.
• During supersonic operation (Le., when NPR> NPRc, where the subscript c refers
to the critical NPR which separates between subsonic and supersonic operations),
the expansion waves aft the throat corner are enhanced during vectoring. The char-
acteristics of these expansion waves depend on the 0" deflection value and they may
be enhanced by a 'separation bubble' formed downsteam of the hinge corner, well
inside the supersonic stream. A few compression lines, fore and aft the 'bubble', may
complicate the actual mechanism.
• The throat area variation is thus given by
A'
...! = coso (I)
As v

or, during simult(lneous pitch-yaw vectoring. by

A: = coso,.. coso
A'
y (2)
64 Fundamental Concepts

whereAs is the effective throat cross-sectional area defined by point 8' in Fig. 13,
Introduction. However, eqs. (1) and (2) neglect the effects ofthe engine, as described
below.
• To maintain a predetermined A~ IA; (=A9 lAs = £,.) ratio, the effective nozzle exit
area A9 should also be subject to the condition
A'
A: = cos 3" • cos3y (3)

• However, to maintain the same mass flow rate through the engine, for 8;, 0; deflec-
tions at a given NPR value, the effective throat area As must be subject to the
condition
As =As, or (A1)ov.ov=M(ov=oy-O) (4)

Consequently, the flaps in the throat area must be "opened" by a factor of

AO,.Dy
_s_ = ---:0:"'---:;'- (5) Control Rule No 1
As cos 0• • cos Oy
• Similarly, to maintain the required AelAt value, the flaps in the nozzle exit area
should be "opened" by a factor of

A3vDy
(6) Control Rule No 2
A;- = cos or . cos ~v

IDO -
;;----- tld~al

Pe-nalty
.96
efG
.94
I ~!COOlant
~--o..........
"'-0
U Bypass
Pt'nalty

~2~/__~__J I~_~I~~I
2 6 NPR6 10 12

Fig. 2. Typical variations of exhaust nozzle thrust efficiency, Cjg. with Nozzle Pressure Ratio (NPR) for
a constant AelAt (==A9IAs) value. The Straight and Cullom results shown are for the circular
(axisymmetric) nozzle of the J-85 engine [48]. Leakage and cooling (thrust) losses are mainly attributed
to current-technology, many-flaps nozzles. These losses are expected to,be minimized in the new, vec-
toring, 2D-CD nozzles manufactured with or without RCC materials (see text and Figs. C-I to C-6).
Thus, the thrust coefficient of vectoring nozzles may be higher than that for conventional (circular)
nozzles, even for current-technology (cooled) flaps, i.e., the leakage penalty shown would be substan-
tially reduced in properly designed 2D-CD thrust vectoring nozzles. This expectation has been recently
verified experimentally by our jet propulsion laboratory tests with a number of new thrust vectoring
nozzles.
Lecture I 65

• For "down" pure pitch vectoring the 'upper' divergent flap is tilted according to
Tilted angle = Ov - il, (7)

where (),. is counted positive for nose down pitching moments. This causes the sonic
line to move from point 8 to point 8'. Thus, point 8 remains in the subsonic domain
during supersonic vectoring operation (of the engine). Consequently, whatever is the
value of(0,. - il), expansion or compression waves would not be generated at point 8
on the "upper" hinge of the throat, while only enhanced expansion waves would be
generated aft the "lower" hinge, as shown in Figure I.
• The riet result is, therefore, asymmetric operation during which the enhanced
expansion waves cause the expansion ratio from throat to exit to be under lower
NPR values, than those computed by standard textbooks, i.e., in the supersonic
domain
NPRvec,oring < NPRnonvec,oring (8)
• The final result should, therefore, give the designer somewhat lower thrust efficiency
values for NPR > NPR .., i.e., for most of the supersonic domain.
• However, for a properly designed nozzle, the opposite conclusion may be expected
for subsonic operations/To start with, one may notice that the flow at point 8 would
be such as to improve the vectoring of the emerging jet, with minimal, or no throat-
comer losses on the "upper" hinge, while those on the "lower" hinge would not gen-
erate expansion-waves losses, but would be limited to known, well-calibrated
subsonic flows around a single comer. However, for subsonic operation the diver-
gent (half) angle Ii should normally be close to zero. Hence, the boundary-layer sepa-
ration tendency at the lower hinge would be less damaging than the payoffs supplied
by the "upper" hinge-flap geometry. We therefore conclude this section by five pre-
liminary control rules for primary and secondary effects in vectoring, i.e.,

[Cfg]vecloring < [Cfg]s •...o for NPR > NPR c (9)

6)/'"0 (supersonic)

[Cfg]vecloring ~ [Ctg]6....o for NPR < NPR( (10)

6)/'"0 (subsonic)

For primary thrust vectoring we may, tentatively, write

Tx= Cig ' T;. cos Ov' cos Oy CR. No.3 (II)
Tv= Cig ' T;. sin Ow' cos ()y CR. No. 4 (12)
Ty= Cfg ' T;. cos ov' sin ()y CR. No. 5 (13)
66 Fundamental Concepts

OryP""'o< OryP .....

1.00 100
O..{;}~-€t-§- @-0 ~'~:(j/(5::o:e
096 P'
/ 0.96
O·
0 I.'
I
o.n
(;]',-""
8
C rg Crg 092
Nou"
o e3
low M&c:h AllliIbulnlnQ Powet Lo", Mach .-.lte'OUrnlfl1) Power
1.00 100
~-o-b p.fY'O'X
0.96
BB' 0 .96 p ....
C fg 0.92
d Cfg 0.92 o..tf/\..Aal HOUle
I
0.88 , 0 .88 , ,
Noulo P,oS:tu,O Raho. NPR NoU" PrtlSsur. AaUG. NPA
A .u,ymm~J1nc::: NOllltli • Un'Wocloreci 2· 0/C · O NOUHt • Un'lldClQl ed

Fig. 3. Vectoring 2D-CD nozzles (at the unvectored position) may produce somewhat higher thrust than
conventional circular (axisymmetric) exhaust nozzles, up to NPRc: Rotatable converging flap (2) and
cylindrical seal-section (I) allow TV/TR modes A. B. C. D. Vectoring effects are shown in Fig. 6.

where T; is the ideal thrust without vectoring. given by (.)

_ .
T; - Mactual
[
RTT
2y
y_1 ( I- { PPT a }r-I/Y ] '11
) (14)

(*) For R&D projects at the Jet Propulsion Laboratory we employ enthalpy changes of a real gas
(air+fuel-reaction products) to estimate T; (without relying on the perfect-gas equations), i.e.,
"ideal thrust" does 110/ lI1('al1 "ideal gas". (f § 1-6 below.
Lecture I 67

B o· rf C

Fig. 4. External Thrust Vectoring (EVT) as advanced by Berrier and Mason of NASA (208, 209). A-
Sideview. B - Backview, 3 pedals. C - Backview, 4 pedals.
The pedals may be rotated asymmetrically during approach or ground-handling to trim pitching
moments, or to eliminate undesirable ground effects.

Fig. 5. ETV yaw-pitch envelopes as reported by Berrierand Mason of NASA (208, 209). (Cf. Fig. 4).
68 Fundamental Concepts

An experimental confirmation of eqs. 9 and lOis shown in Fig. 6.

The designer of vectored propulsion systems, whether limited to pitch-only vector-
ing, or not, may, therefore, expect higher thrust efficiency values up to about NPR=4,
i.e., in the subsonic and lower supersonic domains.
Apparently, the highest efficiency is obtainable around 8" = 10 degrees and NPR = 3
to 4. However,for initial design purposes one may consider the range 8. =:; 20 degrees.
No experimental results for the effects ofoy-vectoring on thrust efficiency have yet
been reported in the literature. One reason for this is the fact that only a few designs of
internal vectored propulsion systems with simultaneous yaw and pitch vectoring exist
today, and their geometrical and experimental characteristics are still not available for
publication.

1.00
Min Jet Are. p. Ma. Jet A,..

0.98
y, :J;:1 ~
-.x.
f./
~
~~

1'3w
0.96
Sym &
0.94 0 0_
t--

..
8t::. 5
'0
15
0.92 ,'---:
3 •
NPR

t"
0.8
NPR· ~!.
Jet Area ~= ;,,4~,..
- IS· t,,- ,..
0.6 ...l -"-
"",. ., r-c.
~
Upper Flap
~ L Upper Flap ~
~
0.4 ~ .:..

P'/P T
A'
"LJ..l:l -8 -8 ~.P.
....
0.2 P.I P,

LT'TP J:P
Lower Flap
.A..t>,.
l~
""081 Exil bit
0.0
0.4 -0.2 0.0 0.2 0.' 0.6 0.8 1.0 -0.4 -0.2 0.0 02 0.4 0.6 0.8 1.0
Tap Localion, X/L Tap location, X/L

Fig. 6. As demonstrated by Straight and Cullom of NASA (48), the thrust efficiency ofvectored exhaust
nozzles may increase during deflections, up to NPR=4 (i.e., mainly in the "subsonic" domain). Maxi-
mum efficiency is obtainable with 0" = 10 degrees. The experimental data shown are for a 2D-CD vec-
toring nozzle installed on a General Electric J-85 engine (cf. Table III-I and Figs. I1I-4. 5, 6, 7 for
additional details). The lower figures show the variations of wall static pressure on the "upper" and
"lower" diverging flaps at NPR = 4, 0" = 15 degrees. These highly interesting results were reported in
1983 by Straight and Cullom of the NASA-Lewis Research Center [48]. Cf. Fig. C-7.
Lecture I 69

1.6 How To Compare The Performance of Different Thrust-Vectoring

Nozzles
The central element in Internal Thrust Vectoring (lTV) is the nozzle design and its
integration with the wing and the flight control system.
Unfortunately, no systematic study, or design, has yet been reported in the literature
on a practical solution for simultaneous yaw-pitch-roll lTV design, control system and
integration potentials.

(;J
I
,
I
-_.

--
~
~
---l

---)
I

l\)
~.
~J
o
Fig. 7. An example of partially-vectored aircraft. This F-15 STOL fighter demonstrator has been flight
tested in 1989 (13, 18,62,70,71, III, 128,234).
Instead of the simplified TR design shown in Fig. 8, this design is based on venetian-type ofthrust
reversal vanes. These are closed during pitch vectoring (upper right). open in the downstream direction
during landing final approach (center). and opened in the upstream direction after touch-down (lower
right). However, thrust reversal is less important than thrust vectoring and may not be required (cf.
Fig. 21, Int.).
For various nozzle details see Figs. 13 and 14. Note that up-pitch vectoring is also useful in reducing
runway distance in takeoff, namely, in rotating (nose liftup) the aircraft at much lower speed than with
conventional aircraft.
Then. using [FPC. the jets are "vectored down" for getting the maximum (LlCdv+(dCdsc (cf.
Lecture III, Fig. III-I 8).
70 Fundamental Concepts

Furthermore, during experimental evaluations of the nozzle coefficients for such

yaw-pitch-roll systems, in comparison with pitch-only thrust-vectoring nozzles, one
needs an exact thermodynamic basis to perform the comparison properly. However,
in establishing such a basis one often encounters the following fundamental problem
with respect to the thrust coefficient efg :
IDEAL THRUST does NOT mean a permission to use ideal gas equations in
isentropic adiabatic expansion.
Instead, one must compute isentropic expansion of a REAL gas, namely, of air plus
the proper concentrations of the reaction products of burning a given jet fuel. Thus,
one must employ the proper ENTHALPY tables to determine the value of the ideal
thrust at each test point. This procedure, in itself, may cause two additional problems:
(i) Proper enthalpy tables may not be available for each chemical composition.
Hence, one may have to compute them by using well-tabulated enthalpy values of the
chemical components which must be known to the experimentalist as comprising the
fuel.

,-----------------------------------------

A V('clorlng- R.....E"sing Nonl ..

Fig. 8. An example of an early vectoring-reversing nozzle (no yaw vectoring) (49). Note the
impingement-film-cooling methods employed to cool the C-D flaps. These cooling flows generate two
boundary layers, one inside the flap wall (impingement cooling) and one outside it (film cooling). For
heat transfer rates in subsonic and supersonicjlow regimes see Eqs. Ill-I, 2 and Figs 111-1 to 6. Note also
the location of the actuators in the sidewalls (upper part of the drawing). Sidewall truncation effects on
nozzle efficiency are shown in Figs. 15 and 16. The design of the circular-to-rectangualr transition duct
may be done by following the superellipse equations (Figs. III-I 7).
However, for high NAR, like the ones shown in Figs. 13 and 14, this simplified design approach may
be misleading. Also misleading is the calculation of AR effects on nozzle weight in isolation [cf.,
Fig. 111-14].
Lecture I 71

------

2.0/C-O NoaJe

~ ~
~ P
Re'.... M yec;IOtIfO AUl)t'\Mled Or., Po.

Fig. 9 (upper left). Schematic view of current-technology axisymmetric nozzles.

Fig. 10 (right). Axisymmetric and 2-DIC-D nozzle geometry comparison (No yaw thrust vectoring) (ef.
Figs. 9, 10, 11, 12).
Fig. II (lower left). In pure pitch vectoring the throat cross-sectional area changes as shown. In yaw I"ec-
toring a similar change takes place in the perpendicular plane. (Cf. eqs. 5. 6).
Fig. 12 (lower right). Typical 2-DIC-D actuation system (49). Low AR vectoring nozzle without yaw
thrust vectoring (cf. Fig. 8).

(ii) Varying the throttle changes RPM, and, accordingly, the NPR values change.
Yet, during this change the chemical composition may change too, due to such factors
as combustion efficiency variations and an air-to-fuel ratio change, etc. (cf. Lecture
III).

1.7 Subsonic Yaw-Pitch-Roll Thrust Vectoring

The test results shown in Figs. 19 and 20 have been extracted from our early laboratory
test series of a subsonic, simultaneous yaw-pitch-roll, internal thrust-vectoring nozzle,
with NAR=46.6. Test results have been corrected to standard sea-level temperature
72 Fundamental Concepts

.,/ "

Fig. 13. G.E. High AR SERN (without yaw vectoring) (49).

Fig. 14 (right). NAR= 15 nozzle with flow dividers, but without yaw vectoring.

Uniform Flow
P ~ p. at Design

Possible Sidewall Cutback

Fig. 15. Figs. 8 and 12 show that the sidewalls of pitch-only, vectored, 2D-CD nozzles, may be shorter
than the diverging flaps. This is done to reduce nozzle weight (49).
For design cases where sidewall cutbacks are important, one may first examine the Mach lines in this
figure and, accordingly, design sidewall cutbacks.

and pressure. The turbojet engine employed delivered up to 350 kg. thrust with a well-
calibrated, axisymmetric nozzle. Yaw and pitch angles are the geometric angles of the
converging flaps. The fuel used is JET-A-!' The engine is equipped with a standard
bellmouth inlet. In general, no significant degredation of thrust levels, nor significant
SFC increases, were encountered during yaw, pitch, yaw-pitch and yaw-roll thrust vec-
toring tests.
During yaw-pitch-roll-vectoring flight tests (using vectored RPVs) the nozzle exit
area was maintained unchanged by a special control system.
Lecture 1 73

~FulIL8noth
_ _ Cl=-~
- - - C2 ~ 50·t. Cutback

... ~O~;-· C3 ~ m', Cutback

1.00

0.96
C~
rg

0.92 , 3 5 7 9 11
NPR

----.g;~
-.-.-.C8~

• - 1.797 1::::::::>'

1.00

0.96

Cfg
0.92 _ _....I011..._ _ _'--_ _......_ _--'_ _....,j
I 9 11
NPR

Fig. 16. Following design consideration shown in Fig. 15, one may examine the slight changes in the
thrust coefficient due to 50% and 75% cutbacks in diverging-section sidewalls.
The data are taken from Ref. 49 for a 2D-CD vectoring nozzle. Note a slight increase in CJg values in
subsonic operation. with truncated sidewalls. in comparison with untruncated ones.

Symmetry comparisons between LEFT-RIGHT-YAW vectoring is easily evaluated

by using such graphs. Moreover, such graphs may be used to verify the degree of
REPEATABILITY of the testing methodology and the test equipment.
74 Fundamental Concepts

Ae' At
•
0
Axi
2-D C-O
1.28
1. 65
0 SERN 1.15
c; Wedge 1.30

1.0

.9

T-D
T.
1
.8 0.60
t M = 0.90
-----

t
.7

.6
I I I I I I L.L __- ' - _.._L_~
I 2 4 6 8 10 I 2 4 6 8 10
NPR NPR

A/At Ae' At
• Axi 1.56 It Axi 1.56
o 2-D c-o 1.15 o 2-D C-O 1.65
o SERN 1.36 o SERN 1.36
o Wedge 1.40 o Wedge 1.40
.9

.8

.7 . M = 2.20

T-D .6

Ti
.5

.4

.3 L--l.__ .. __L _. L.__ ~

o 4 8 12 16 20

NPR NPR

Fig. 17. As the flight Mach number is increased, net thrust ratio (T-D)/T; variations demonstrate a
slight advantage for the 2D-CD vectoring nozzle over its competitors (cf. Fig. 1a). This advantage
applies through all NPR values [1591.
For additional details concerning weight advantages in comparison with other nozzles please turn to
the Appendices.
Lecture I 75

A/At
III Axi 1.28
o 2-0 C-O 1.65
o SERN 1.15
o Wedge 1.30
.016

M = 0_90

o, I I I I I I' ,
1 2 4 6 8 10 2 4 6 8 10
NPR NPR

"e1.5£
/A , 10,1",
•
0
IIxi
2-0 C-O 1.15
•
0
Axl
2-0 C~D
1.56
1.65
0 SERN 1.36 0 SERN 1.)6
0 Wcdf]c 1.-10 0 Wcdf]e 1.-10
.028
r
M 2_20

co.an
.016

.012

.008

.004 ,,
1 2
I
4
, ,
6 8
I
10
,
0
I
8
I
12
I
16
I
20

NPR NPR

Fig. 18. As the flight Mach number increases, 2D-CD nozzles demonstrate low drag penalty in compari-
son with current-technology nozzles throughout most NPR conditions tested [IS9). However, with some
airframes (e.g., the F-16), the rod-TV nozzle generates lower drag than the low AR 2D-CD nozzle. For
further details please turn to the Appendices.
76 Fundamental Concepts

N P R =

I
P R = 1. 6

::::::~~-
~ -,
';.1 0

-B
-12
-16
.4 Q .. .. ,~ .o!> Left
eo ' - -21)
bv Pi t.ch Angl e.d3g

Fig. 19. An example of our JPL graphical evaluation of thrust efficiencies during simultaneous yaw-
pitch thrust vectoring. AR = 46.6.
Lecture I 77

g .c

C1g

Fig. 20. How to report and compare different thrust efficiencies during simultaneous yaw-pitch thrust
vectoring? This figure provides a possible graphical procedure. AR=46.6
 LECTURE II

VECTORED AIRCRAFT AND

SUPERMANEUVERABILITY

Consider what effects that might conceivably have practical

bearings you conceive the objects of your conception to have.
Then, your conception of those effects is the whole of your
conception of the object.
C.S. Peirce (1878)
11.1 Yaw-Pitch Vectoring and Supercirculation vs. The Harrier
Technologies
Thrust reversal for vehicle deceleration on the runway, following touch-down, is in
common use in transport aircraft. The benefits of the Harrier (A V-8) are also well-
known. (Harrier-induced technologies are briefly reviewed in Chapter VI and in
Appendix B.)
However, the Harrier technologies do not provide efficient yaw thrust-vectoring, nor
the additional payoffs of supercirculation. High drag is also associated with these
technologies.
What is new is the continuous, highly integrated attempt to replace all, or most aero-
dynamic control surfaces, by the simultaneous use ofyaw, roll, and pitch thrust vector-
ing, (with, or without, simultaneous thrust reversing). The actual innovation is to
employ simultaneous yaw-pitch-roll vectoring to gain revolutionary enhancement of
performance throughout and "outside" the flight envelopes of conventional fighter
aircraft, while also reducing drag, radar cross section and infra red signatures. Other
benefits include drastic reductions in runway required distances (dry, wet, or icy). In
short, vectored aircraft means jetborne flight, or, simply, jetborne aircraft. It also
means post-stall maneuverability and controllability (Introduction).

* * *
The concept "vectored flight/propulsion" has added new dimensions to mission
definitions and to the design of high-performance fighter aircraft. Thrust vectoring
provides additionalforces and moment effectors for optimization and maximization of
performance at takeoff, cruise, air-to-air, and/or, air-to-ground missions, and in land-
ing. It can be used for augmentation of agility, stability, safety, and control at the very
weakest domains of aerodynamic-control surfaces. Thus, it can be used for the expan-
sion of flight envelopes, and for flight-control improvement and/or emergency-
redundancy systems.
Yaw and pitch vectoring, through multiple vectoring nozzles (for single as well as for
multiple-engine aircraft), or through external pedals/flaps, provides unprecedented
roll and turning rates. The simultaneous use of jet-yaw, pitch-and-thrust reversal fur-

79
80 Vectored Aircraft and Supermaneuverability

E
~

oo n CPY

Fig. 1a. Schematic view of a Pure Vectored Aircraft (PVA), in Pure Jetborne Flight Control. Note the
rolling arm Y, the yaw-directional jet angle ov,
and the arm D*for pitching moments of the vertical force
Tv = T . sin Ov . cos Sy. Yaw force is Tv = T . sin Oy cos 0,.. Axial thrust is Tx = T . cos Ov . cos Oy.
Shaded area represents wing area affected by supercirculation (cf. Figs. III-7 and III-8) of the vec-
tored jet in a pitching mode.
1 - Variable canard is differentially controlled. It becomes less effective at PST-maneuvers (see
however Fig. Ic).
2 - Wing's sectional area Sj where (L1Cdsc applies (cf. Fig. III-7).
Note the following cases:
(i) "Pure" Sideslips: The yawing thrust-vector ofone engine points in the sidewise center-ofpressure CPy
direction, causing "pure" sideslips, while Tx for the 2nd engine is reduced as dictated by eq. 4; (ii) Oy is
the same for both engines causing strong clockwise yawing-moments; (iii) the same as in (ii), but
anticlockwise, (iv) engine-out control: with Oy 0# 0 and Ov 0#.0 with a differentially-controlled canard and
Cjflaps operating as ailerons to maintain stability; (v) other combinations of pitch, yaw and thrust-
reversal vectoring in flight, and in landing (cf. Figs. 1b to 1f, and 6; also Fig. III-I).
Performance:
I) Minimal dependence on aerodynamic-flight-control-surfaces.
2) Surpasses stall limits of conventional flight envelopes without loss of control and agility.
3) The aircraft is able to point in one direction (to fire its weapons), while 'translating' in another direc-
tion (to improve survivability). Cf. eqs. 8 and 11 and the definitions of "direct lift" and "pure" side-
slips (11-3, 11-3.1).
4) The aircraft will be able to perform as, for instance, described in Fig. 2, thereby increasing survivabil-
ity potentials and Simultaneously providing an ofJ-boresight weapon aiming capability.
5) By rejecting other "direct-lift" type surfaces, tails, stabilizers, flaps, flaperons, elevators, etc., as well
as the canards, RCS and other signatures may be substantially reduced; cf Figs. Ib to If
6) STOL or V/STOL performance are substantially enhanced, thereby becoming the "natural" domain
of vectored aircraft.
Lecture" 81

----I

Canard

E <!i

D
-ltP'I
/)

_J~~
Fig. I b. Pure Vectored Aircraft Equipped With Vectored Canards.
The canards themselves may be fixed or variable. Oy should normally be less than o~ in pure sideslip
"translations" (see eq. 9).
Low pressure, cold/an air from turbofan engines may be used to control the vectored canards. Note
also the distance between CPy and CG.
Note: In the view of a few designers the optimal location for a canard is on someone else's airplane.
Those who subscribe to this assertion are invited to examine the next figures.

ther enhances performance (see, however, our reservations in regard to thrust-reversal

methodologies, as enumerated below).
Maximization of benefits to be derived from thrust vectoring requires a new, highly-
integrated approach to the design process, including entirely new applications of pro-
pulsion, aerodynamics, flight-mechanics, control, materials, and structures
technologies. However, the integrated approach to these disciplines is not available yet,
except for partially-vectored aircraft, to be defined next.

11.2 Pure Jetborne Flight Control

Vectored aircraft may first be divided into those that are "pure" or "partial". Vectored
propulsion, in tum, can be divided into internal or external thrust vectoring. In this
Lecture we consider the payoffs obtainable from internal thrust vectoring. External
thrust vectoring is discussed in Lecture VI.
82 Vectored Aircraft and Supermaneuverability

C.G. C.G.

Fig. I c. (left) A schematic view of PVA equipped with Extended Nose Vectoring Jet (ENVJ) for balanc-
ing moments and improving PST-Super-agility (without the use of canards)
Fig. I d. (right) A schematic view ofa VISTOL ENVJ-PV A. An example of a central VISTOL nozzle as
used by our team is shown in Fig. I h.

11.2.1 Pure Vectored Aircraft in Pure Jetborne Flight Control

In pure thrust vectoring, the conventional, aerodynamic-control-surfaces ofthe aircraft,
are replaced by the stronger, internal, engine forces. These forces may be simultane-
ously, or separately, directed in all directions, i.e., during pitch, yaw, roll, thrust rever-
sal and forward thrust-control of the aircraft.
Moreover, pure (internal) thrust vectoring may entail the use of exhaust nozzles-
wing structural-integration, so as to obtain optimal supercircuiation ( see below).
Thus, in pure vectored aircraft there is no need for flaperons, rudders, ailerons, ele-
vons and flaps, and even the vertical stabilizers may become redundant. However, to
improve maneuverability, a vectoring nose-jet, or differentially-controlled canards,
may be required to balance moments and improve rate of turns, dynamic stability, etc.
(see below, and Figs. la, to If).

1.2.2 STOL or V/STOL By Vectored Aircraft

A typical landing procedure is demonstrated in Fig. 6, with nozzle positions as in
Fig. 1-7. Lower approach speeds, as well as a rapid change of the Venetian-type vanes,
allow the use of very short runways. The internal structure of such nozzles is shown in
Fig. 1-8 with Figs. 1-13, 14 showing them with high NAR. Fig. 1-6 stresses the high
nozzle coefficients of 2D-CD vectoring nozzles and Fig. 1-8 demonstrates elaborate
cooling methods for non-RCC nozzles.
Pure vectored flight normally involves the simultaneous use of a few effects, e.g.,
Lecture II 83

Fig. 1e. A possible design for an advanced pitch-roll-yaw, thrust-vectored, cruise missile with low signa-
tures. Note the split-type 2D-CD thrust-vectoring nozzle which allows simultaneous yaw-pitch-roll
thrust-vectored maneuverability or even supermaneuverability. Axi-TV nozzles* cannot provide suf-
ficient roll thrust-vectoring moments, nor supercirculation lift gains and deflection angles greater than
13 to 15°.

thrust reversal combined with pitch and yaw vectoring. This may be highly useful in
close (viffing) combat engagements (Fig. 9), or in low-speed runway approach (Fig. 6).
Combined pitch and yaw vectoring is useful in high-maneuverability engagements, or
in landing sites with strong cross-winds (combined with roll moments). Vectored
propulsion/flight also allows decoupled flight maneuvers, in which the aircraft trans-
lates up, down, left or right, without banking, or changing attitude!

11.2.3 Partial Vectoring or Partial J etborne Flight (PJF).

Rudders, elevons, ailerons, flaps, canards, and elevators are still used in conjunction
with thrust vectoring during Partial Jetbome Flight (PJF). Thus, the elimination of
one or more modes of the pure-vectored elements, such as the elimination of yaw-
vectoring and supercirculation in the Harrier, or in the STOL F-15 demonstrator, cat-
egorizes the aircraft as PJF.
The flight experience of this laboratory, during 1987, employing pure vectored-
RPV s, has demonstrated, for the first time, the ability of a properly-spaced, high-
aspect-ratio nozzles**, to act as "elevons" in engine-out flight control, and in landing
situations (cf., A, C*, Yand D dimensions in Figs. 1 and III-I). Part of this course is
based on this experience, and also on the experience gained in testing various vectored
engines in the laboratory (see below).

11.3 Dimensionless Numbers For Pure Vectored Aircraft

Force, moment, energy, and Specific Excess Energy (SEP) equations must be derivedfor
the various types of PVAs shown in Figs. 1a to 19. These equations may become the
basis of various simulation programs of vectored aircraft performance.

* Cf. Fig. 1i.

**NAR=46.
84 Vectored Aircraft and Supermaneuverability

In addition, one may search for dimensionless numbers which allow scale-up and
scale-down design considerations. One may, therefore, search for dimensionless num-
bers which are, under certain conditions, independent of:
1) The thrust level of the engines.
2) The number of engines.
3) The external aerodynamic forces and moments acting on the aircraft.
4) The major aerodynamic design parameters of the aircraft.
For this purpose one may first define two dimensionless numbers for the type of vec-
tored propulsion shown in Fig. la:

Vectoring Yawing Moment

N,= (1)
Vectoring Pitching Moment

Vectoring Yawing Moment

N2 = (2)
Vectoring Pitching Moment

For the example shown one may further assume that:

a) The exhaust-nozzle-mass-flow-rate in single-engine PVAs is equally divided
between the "left" and "right" thrust-vectoring-nozzles, so that, during all vec-
toring angles,
-r = 'P-. (3)

b) In multiple engine PVAs eq. 3 holds for both right and left-hand engines (unless
pure sideslip flights are conducted - cf. the Introduction and eqs. 6 to 10 below).
c) During vectoring 0; = ~ and d; = d;, except in sideslips and rolling.
d) During rolling d; = -d;, or vice versa, while 0; = ~.

Let us consider now a steady-state, simultaneous yaw-pitch vectoring. Now, by

employing eqs. 1-11, 1-12 and 1-13, and by neglecting the coupling between yaw and
roll, as well as the canard effects, one obtains:

cos ov • sin or. (Control Rule No.6) (4)

N,=
sin ov' cos Oy

D cos ov' sin or. D

N2 = N, (Control Rule No.7) (5)
Y sin Ov • cos Oy Y

These numbers do not depend on the thrust level of the engine(s), nor on the number
of engines.
However, the couplings and the canard effects may not be negligible. Moreover,
large p, p, a and a. values may change engine-in take-performance (Appendix F), as well
Lecture II 85

as the left/right-wing-lift values, aerodynamic moments, etc. We shall return to this

problem below.

11.3.1 Pure Sideslip Translations

Various decoupled flight maneuvers, which allow the aircraft to "translate" "left",
"right", "up" or "down ", without banking, or changing attitude, are readily controlled by
well-designed vectored vehicles.
Consider, for instance, the following design option for a (steady-state) "left"

Fig. I f. An early concept of a vectored RPV proposed by this laboratory. The thrust-vectored roll rate of
this design was poor, due to the small yaw-arm length Y (Fig. la). Hence, its propulsion system was
replaced by a split-type roll-yaw-pitch TV nozzle.

REV£RSE TlIRUST 1,\ = -4~ol

BOUNDARY- lA'ltR
CONTROL BLOWING HIGH lin 1\ = 30")

kRUEGER LEAD INC EDCE

<::"~
;-HICH.lIfT CANARD d
! - ~'--SECONDARY
NOZZlE
CASCADE

Fig. Ig. A configuration proposed in 1983 by Paulson and Gatlin (Ref. 155) for an advanced (partially-
vectored) fighter. It shows a specific design for main-vectoring-nozzles-AR and location, as well as a
boundary-Iayer-control-blowing over high-lift canards [see also Lecture V and the notes in Figures I a
to Id).
86 Vectored Aircraft and Supermaneuverability

"translation" without changing vehicle's attitude, but with thrust reduction of the left-
hand engine, (see, however, § 11-3.5).
Consulting Fig. la first, one may choose to perform a pure "left" sideslip maneuver,
by keeping the throttle of the right-hand engine unchanged, but directing its yawing
thrust in the CPv direction, i.e.,

Dv(Right Engine) = sin-'[YID), (6)

where Y and D are defined in Fig. I a.

Yet, to balance the yawing moments, the left-hand thrust, T L , (with D;;=O) must be
reduced to
T.~= T.~- T: DIY + Fep(D*-D)/Y (7)

i.e., to

T~=CfgTi [ cosOvcos(sin-'{YID})- yD ..
(cosovsm(sm-'{YID}) ] +Fep (D*-D)
Y ,

(8)

where Fep is the aerodynamic drag force resulting from the (steady-state) sideslip
which operates at CPy'

Fig. I h. The Turgemann-Friedman-Cohen vectoring nozzle for pure vectored RPVs ofthe type described
in Fig.ld (was flight tested in 1987 on our No.5 V/STOL vectored RPV, with canards instead of
ENVJ).
Note that this nozzle divides "rear jet flow" (RJF) from "CG down flow" (CGF), and is normally
closed (forming a flat surface under the wing).
The dashed-line arrows indicate the respective flaps rotations during closure of the nozzle in normal
cruise flight.
Lecture II 87

A dimensionless number for such slideslip maneuvers may thus be defined by Con-
trol Rule No.8:

N3= LFy = T;-Dy (9)

L Fx T~ + T~ - D,

C{gT;[coso v ' sin {sin-I YIDl] - DI'

CjgT;[cosoycos(sin -I YID}] -~ /

. [cosovsin(sin- I YID}]+Fcp (D~-D) +C{gT;coso"cos(sin- 1 Y/D}-D,

where Dy and Dx are the drag components in the y and x-directions, respectively.
Small controlled variations in 0" and Oy may be required in actual sideslip transla-
tions (due, for instance, to variations in lift/drag forces on the left and right wings,
etc.). Moreover, other combinations of Ov and Oy are possible without changing engine
thrust of both engines (see § 11-3.5). A similar dimensionless number may be
derived for "upslip" translations,
N - LF:
4- "\"' F '
L x
(10)

as well as for the vectored-canard, and ENVJ-PVAs shown in Figures I b, lc and ld.

11-3.2 Pure Side-Slips in Vectored-Canard and ENVJ-PVAs

Vectored-canard and ENVJ-PVAs can perform sideslip maneuvers by rotating aI/jets
in one direction. In this case one must keep in mind that the rear oy-angles should nor-
mally be less than the canard o~-angles, due to smaller nose or canard thrust levels and
the different moment arms, i.e., under steady-state condition (cf. Fig. 1b);

T~ = D* (II)
Tv E
For this type of vectored aircraft the work-load on the pilot (or on the [FPC system) is
much less during pure sideslips. Consulting Figs. 1band 1c one may also conclude that,
unlike the previous case, no throttle-variations are required in sideslip "translations"
ofthisPVA.
The moment dimensionless numbers for this type of PV A become:

Yawing Moment Tv' D*+ T~. E

M= Pitching Moment Tv' D*+ T;'. E
(C.R. No.9) (12)

Yawing Moment Tr.' D*+ T~. E

~= (C.R. No. 10) (13)
Rolling Moment Tv' Y+T;'.E

Since normally Dv '" D~ and Dy '" D~, these numbers depend on the thrust, as well as on
the distribution of "forward" and "rear" thrust levels.
88 Vectored Aircraft and Supermaneuverability

Left/right-wing-lift-variations, aerodynamic moments and other considerations

may now be added, as in conventional aircraft design methodologies.

11-3.3 Vectored Flight-Control During One-Engine Flame-Out

Twin-engine vectored aircraft, like the one shown in Fig. 1, may be flight-controlled in
the following mode during one-engine flame-out:
The vectoring-nozzle-flap, sized as (CJ) x (AI2), of the powerless engine, is still used
as an aileron, while the thrust on the other engine is reduced to a predetermined mini-
mum and its yaw-thrust-vectoring set at
ov= sino' (YID) (Control Rule No. 11) (14)
This will cause a controlled, sideslip flight. However, depending on, ENJV, or on the
canard configuration and the dihedral and mode of control (e.g., differentially or non-
differentially), the degree of sideslip may be minimized, or eliminated altogether.

Emergency Landing, Or Restarts With Total Power-Failure

Both single and multiple engine vectored aircraft may be landed safely during total
thrust-failure.* Flight controllability, under these conditions, may, however be
regained for restarting in altitude flight.
A number of factors should be considered by the designer of a vectored aircraft for
this case:
I) Both vectored nozzle flaps, each sized at (CJ) x (AI2), can still operate in the conven-

Single- Twin
Installation Installation

44
Pitch Yaw
<i~
Pitch
G@-th,
Yaw Roll

Fig. I i. (upper drawing). Yaw-pitch, axisymmetric, thrust vectoring nozzles may provide the optimal mix
of low drag and simple modification of nozzle and engine controls. However, they are limited to low
deflection angles and provide no lift enhancement by superciculation. Their signatures may also be
larger than those characteristic of high aspect ratio, yaw-roll-pitch 2D-CD nozzles. (lower drawing)
Mini-, Midi-, and Maxi-RPV s may be designed with a central Vankel engine, or a piston engine driving
symmetric pairs of multiple-stage compressors for (cold), high-a, vectored aircraft (front view).
Such vectored RPVs need not rely on the performance of PST inlets (for these engines are not sensi-
tive to inlet distortion even at a values as high as 90 degrees). See also Lecture IV and Appendix F.

• This conclusion was verified recently in flying our vectored RPVs.

Lecture" 89

tiona I way as moments effectors for flight controllability, supplying pitch and roll con-
trol in the usual way of conventional technology flight-control.
2) The canards must be controlled differentially.
3) Canard configuration with dihedral, (cf. Fig. 7), or ENVJ are useful in gaining yaw
control (see also Lecture V).

11-3.4 C,4 Variations in Flight Control

An advanced IFPC system for PVA must employ 0. and Oy variations for flight control.
In turn, these variations would affect engine performance and engine thrust vectoring.
One may, therefore, distinguish between three vectoring flight-control domains:
1) Small, time-modulated, ov-, oy-variations affecting only the cruise conditions. Com-
pared with the large values of T, these variations are neglibily small, especially for
transient Dv, Dv deflections not exceeding a few degrees. Hence, these variations may
be considered as "engine-thrust-loss-variations" (see Lecture III).

+
Fig. Ij. How to By-Bass Inlet Distortion Problems in R&D, PST, Vectored RPVs? The answer is by
Cold Propulsion.
Shown here is the simplest multistage cold propulsion system which overcomes the problem of inlet
distortion at high AoA flight conditions. Note that the 2-stroke piston engines do not stall, as does ajet
engine equipped with a non-PST inlet at high AoA values. Moreover, high RPM values are always
maintained by the two compressors by separating the two spools and the two engines from each other.
In one-engine-out situations, the 2nd engine still performs well. Proper side covers allow individual
side-wise starting for each engine. Fuel tank is common to all engines.
The 2-stroke engines may be replaced by 4-stroke, or by Vankel-type engines. However, the need to
maintain very high RPM values dictate the need for a gearbox. This requirement presents special
design problems that are now investigated by this laboratory, using light-weight, aviation-type engines
in the range of 30 to 130 HP.
Alternatively, we investigate the use of high-by-pass-ratio turbofans and shrouded propfans, as well
as the design and laboratory/flight tests of variable PST inlets.
90 Vectored Aircraft and Supermaneuverability

2) STOL and super-agility ov-variations affecting pitch moments.

3) STOL and super-agility oy-variations affecting yaw moments.
Consequently, domain-l-type-of-variations may be included in the very definition
of CA , as explained in eqs. 1II-23-a, b. However, this incorporation is merely a matter
of convenience.

11-3.5 Pure Sideslip Without Changing Engines Throttles

Fig. lk, and eqs. 15 to 20, demonstrate a solution for pure sideslip flight, namely, the
elimination of the time-consuming throttle variations of one engine with respect to the
other. Thus, one may solve equations (15) to (18) to find the proper 0;:, 0: values for
obtaining pure sideslips when TR = TL. The procedure is given below:
L M y= (T:- T~) Y - (T;+ T/) D*+ Dcpy(D*-D) = 0 (15)

L Fx= T~+ T:- D x - Dcpx= 0 (16)

L Fy= T;+ T:- Dcpy= 0 (17)

T R= TL or [(T;)2+ (T~)2= (T:)2+ (T:)2] ( 18)
Solving (15) to (18),

,.CG
"CpV
....... OCpV
",
0 Dcpx
c
l~

T; V V Ty

~Dx
Fig. I k. What should be the values of 0; and off in pure sideslips involving "no-change" throttle settings?
Proper definitions of relevant vriables for steady-state, pure sideslip-flights involving "no-change"
of throttle settings (twin-engine, pure vectored aircraft). For additional definitions see Figs. la to Ic
and eqs. I-II to 1-13 and 11-15 to 11-20.
Note that the moments around C.g. are counted positive for counterclockwise rotations and CP is the
sidewise center-of-pressure for the steady-state sideslip motion at angle ~. Tx and Ty are the thrust
components during roll-yaw-pitch vectoring.
Lecture II 91

O!L
oy= tg
-I ( T;)
T; (19)

~= tg- I ( TTyR: ) (20)

11-4 Air-To-Air Operations

Air-to-air supermaneuverability operations involving, say (Figs. 2, 3, 4) deceleration
with or without thrust reversal, + climbing decelerations at high-a values (up to 90
degrees) + gravitational deceleration, and, then, at very low speed (when conventional
aerodynamic contol surfaces become ineffective) to yaw-turn the aircraft nose down,
by using full-jet-yaw-vectoring, (engine must stay at full thrust), and accelerating down,
and back, to the initial speed and altitude.
Combined with pure 'sidewise', or 'upwise' translations, such maneuvers result in the

Fig. 2. A typical PST/PSM by pure vectored flight to gain supermaneuverability in close-combat

between a vectored fighter and a conventional fighter (not shown) in head-on situations, or in firing mis-
siles at the edge ofan adversary's missiles envelope, and, simultaneously, turning back to escape from the
adversary's missile's envelope. In any case the load numbers are less than, or equal to those encoun-
tered nowadays with conventional fighters (see below).
(I) At t = 0, as in the Herbst PST-maneuver, the pilot decides to turn back for a rapid-pointing-and-
shooting, or for escaping the adversary's missile's envelope. (2) Increasingly higha values plus TR, plus
pitching vectoring 0,. may be applied simultaneously. (3) Aircraft enters the post-stall domain. (4) All
aerodynamic control surfaces become ineffective. (5) Yaw vectoring is applied for rapid nose-down,
ov-turn, for the start of diving and acceleration in the gravity vector. Speed is rapidly regained, while
the maximum load number allowed has not been surpassed.(8) Aircraft at the initial speed, altitude
and energy levels, ready to shoot at a much shorter time and at a much smaller turn-back radius than its
adversary. The proper use of a variable canard, or a nose-vectored jet, may enhance this performance.
Engine must stay at full thrust during the entire operation.
Note also that the use of thrust reversal (point 2) is not essential to gain superiority over the oppo-
nent. It is the combined use of pitch and yaw thrust vectoring which matters most! Also note the differ-
ence in aims and pilot's methodology between missile avoidance and close-combat situations.
Close-combat situations, in turn, are subdivided into a number of sub-categories, involving, say, I: I,
I :2, I :many fighters engaged in various, time-space-attitude-envelopes-weapons-speed situations.
Hence, to begin an exact analysis of such situations one must first examine the following figures and
the rest of the text.
92 Vectored Aircraft and Supermaneuverability

saving of time for shooting, and, simultaneously, in acquiring superiority-survivability

advantages in 1 to I, 1 to many, and many-to-many close-combat situations (see below).

11-5 Air-to-Ground and V/STOL Operations

Pure vectored propulsion may help not only in gaining superiority in air-to-air com-
bat, or in escaping missiles, or in aiding low-flying Advanced Cruise Missiles (ACM). It
also helps in reducing thermal, radar and optical signatures, and in reducing runway
distances, even on icy runways. Perhaps the most important innovations of vectored
propulsion are the new possibilities for its employment, without sacrifice in its per-
formance, in new V/STOL missions. This conclusion applies equally to manned or
unmanned aircraft.
Here, one must distinguish between single and multiple engine aircraft, and
between manned and Robot Aircraft (RA), such as RPVs, ACM and certain air-
breathing, vectored missiles. (The introduction of vectored propulsion into RA tech-
nology decreases the conceptual-technological differences between certain advanced
missiles and certain vectored RPV s.)*

11-6 Propulsion and Supermaneuverability

A primary factor in vectored aircraft R&D efforts is the design for survivability
through supermaneuverability and supercontrollability.
Supermaneuverability includes the capability ofa fighter to execute tactical maneu-
vers with controlled side-slipping, and at angles ofattack beyond maximum lift. As an
example of supermaneuverability one may re-examine Figs. 2 and 3 for head-on, or
escape maneuvers.
The tactical payoffs of such PSM-PST-maneuvers are:
• A higher survivability against adversary or missile attacks (counterfire denial).
• Increased ability to confuse adversary pilots and missiles by using decoupled flight
maneuvers that allow the aircraft sharp "translations"; "up", "down", "left" or
"right", without banking or a noticeable change of attitude.
• Increased first-shot opportunities.
• Quicker-into-firing position maneuver.
• Maintaining longer firing position.
• Quicker and easier to point-and-shoot in multiple-targets situations.
• Dictating tactics throughout the entire speed regime.

* Evaluating the feasibility of V/STOL and STOL RA, including our flight testing of single-and-
multiple-engine RA, has recently demonstrated the potential payoffs obtainable with 3-jet positions.
V/STOL RA, in which one of the "downward-pointing vectored jets" is located at the RA center of
gravity, while the other two are being operated as in vectored modes, but with newly balanced
moments. Cf. Figs. 10 and I h and Fig. 15 in the Introduction.
Lecture II 93

SNAP SHOl" 180' TURN REVERSAL

(Short Lived)
OPTIMUM
AOA

~~
.lOSE 'Ollno

~ -.~~
-=='~
ORAVIT'f ASSIST

AOA
RAPIO SLOW ExeURSIOI
OOWI

POSITIONING DEFENSIVE MANEUVER CAl CAUSE

(Ex. High AOA Barrel Roll) EIEMY-~
CREATES NOSE TAIL SEWlRATlON If OVERSHOOT

cr
ALLOWS NOSE POINTING ADVANTAGE -
MSLIGUN SHOT
q

:::r~~~/
fLIGHT PATH

Fig.II-2a. Herbst PST Excursions, even without thrust vectoring, can be advantageous (196). Accord-
ing to McAtee of G D, controllability during PST thrust vectoring ("supercontrollability"), presents the
industry with the greatest challenge.

• More missiles deployable.

• Lower g-loads on pilots and aircraft structure, and also less time spent at limiting
g-loads.
• Dictating the course of the engagement.
An early analysis of supermaneuverability has been reported by Herbst [154]. While
limited to zero side slipping, his conclusions are highly instructive for this course.
They are based on simplified maneuver elements, and are also aided by manned,
close-air-combat simulations. Using such simulations, Herbst defines some maneuver
cycles, which are consistent with unconventional air combat maneuvers with all-
aspect weapons. Inter alia, Herbst recommends a few optional ranges of speed and
(simple) vectoring maneuvers, as well as the required TIW ratios and control power to
attain super-agility status.
To start with, Herbst doubts the importance of thrust reversing and stresses mainly
the great advantages of pitch-yaw-vectoring in Post-Stall maneuvering. He does not,
however, deal with PSM.
Under these simulation conditions, a single PST-capable fighter, is able to neutralize
two conventional opponents with equal weapons.
Herbst employs computer simulations involving multiple opponents in which the
PST-capable fighter gains clear advantages in duel, and in multiple engagements.
His data are based on 3000 simulated engagements on 3 different manned combat
simulators operated by operational pilots of three airforces.
94 Vectored Aircraft and Supermaneuverability

,O~ ____ ~ __ ~ __ ~ ____- ,____--,

tI. '<m/S -Vectored Aircraft
.0 ---Regular Aircraft
701---IJ-t---\,

'oL-----~m----~'o~~~~~--~~
• IS 20 tsec
---
Fig. 3. Herbst Variations of AoA and aircraft speed for the vectored-supermaneuverability turn-back-
maneuver shown in Fig. 2.
Note the gain of about 4 seconds in comparison with a conventional-technology fighter (dashed line)
-Herbst [154]. This gain may become a key element in enabling vectored fighters to survive and win in
battles. both beyond and within visual range a/the enemy. Note also that the lowest speed zone of the
vectored fighter is passed through, while the conventional fighter's nose is still pointing in the opposite
direction (during a much larger radius of turn). This performance may be further improved by employ-
ing high NAR and by the design methods explained in the text .

....... ....
. .. ... .
:.
'

.
I- ••

. " ,,

M-o.3 H-2000m

Fig. 4. Simulated, turning radius values for the vectored-supermaneuverability maneuver shown in
Fig. 2.
This performance may be greatly improved by employing high NAR and supercirculation.
This simulation is limited to zero side-slipping, Herbst [154].
Any of these maneuvers can be designed with load-numbers less than or equal to those encountered
nowadays with conventional fighters.
Lecture II 95

Mostfiring opportunities occurredjust after finishing a PST-maneuver and returning

to conventional flight regime.
Pay-offs of such maneuvers are therefore based on a trade-off between a momentar-
ily loss of energy, versus the gaining of time and positional advantages for a first firing
(cf. Figs. 2, 3, 4). Moreover, with PST capability the fighter can roll around the veloc-
ity vector at a constant, very high-angle of attack. Consequently, PST fighters dictate
the tactical course of the engagement!

11-7 Maneuver Analysis

Limiting the analysis to zero sideslip conditions during PST-maneuvering, the results
shown in Figs. 11, 12 and 14 are calculated for a particular aircraft drag polar, at a par-
ticular altitude and for a maximum engine power, using a mass-point analysis [154].
Fig. 11 i describes two conditions of a vertical maneuver.
One may note first the zero Specific Excess Power (SEP) line, distinguishing
between accelerated and decelerated maneuvers, as well as the structural and maxi-
mum lift limits. Thus, any combination of turn rate and Mach number corresponds to
a particular radius of turn. These results indicate the following advantages:
• A turn rate advantage beyond maximum lift AoA at low speeds.
• Minimal turn radius advantages at PST-AoA beyond 50 degrees.
• Best results are obtained when engine thrust remains at its maximum setting during
the entire maneuver.

Fig. 5, Takeoff procedures of pure vectored aircraft. Note that the jets are first turned "up ", and,fo/low-
ing rotation, are turned "down". Using our vectored RPVs we have tested both vectored and
unvectored takeoffs, and found the first minimizing ground roll substantially, especially for high T /W
vectored RPV s.

Similar conclusions are obtained from horizontal maneuvers, at a time in which the
unvectored aircraft is still outside its firing cone.
Fig. lIe is another representation of the same engagement. It shows how the pilots
in the vectored aircraft (left hand side of Fig. Ie) and in the conventional aircraft (right
hand side of Fig. 11 e) would view their opponents as they look outside their cockpit
windows. The proper firing time for the vectored fighter is marked by the large arrow
in the middle of the given weapon firing cone.
The vectored fighter performs the low-speed, PST-maneuver far outside the
un vectored fighter weapon range. Yet it gets its first firing opportunity in a safe posi-
96 Vectored Aircraft and Supermaneuverability

Fig. 6. Landing procedure of pure vectored aircraft (STOL Vectored Aircraft). Note: Giving up thrust
reversal (TR) may increase landing ground roll, but would also save weight, cost and operational com-
plexity. Note also that pitch vectoring, combined with supercirculation, decreases the approach speed,
thereby decreasing landing ground roll. without the need to use thrust reversing. However, thrust rever-
sal potential deployment on approach - as shown here and as explained in Fig. 1-7 - allows the jet
engine to be maintained injull dry power, thus avoiding spool-up delays after touchdown (as encoun-
tered with civilian transport aircraft). Note also that in pure vectored aircraft there is no need to install
engines with a fixed pitch deflection, unless cruise drag considerations dictate it. In the following text
we assume that Ii,. is defined as in Fig. I-I and engine center-line/pitch deflection may be added or
substracted from the proper Ii" values.
The landing phase is an example for the simultaneous use oj vectoring jets in all directions, i.e., for a
partial reversal, a partial forward, yaw, pitch and roll control.

¥-~
I. /- WEIGHT \ ~bv
THRUSl\'

Fig. 6a. Glide-siope-force-equilibrium on (pitch-only) vectored aircraft.

Note that: T . cos (~)
a+uv = D rag- W'gh el t· sm. y

T· sin (a+liv) = Weight . cosy-Lift

However for future designs, the Vectoring inlet suction force should be separately added to these
calculations.

tion relative to the unvectored fighter. The overall result is that the vectored fighter
obtains a 'firing period' without facing the possibility of counterfire from the
unvectored fighter. In fact, the vectored fighter is not to be found in his opponent's fir-
ing cone until the vectored fighter's first missile hits the target. Moreover, the vectored
fighter rapidly recovers the energy lost during the PST-maneuver, as shown in
Figs. 11 a and 12. The area between the solid lines represents the energy of the vectored
fighter, while the region marked with dashed lines, illustrates that of the unvectored
fighter.

Vulnerability of the Vectored Fighter in PST-Maneuvers

There is no doubt that the vectoredfighter is vulnerable during the low-speed/high angle-
ofattack phase of the PST-maneuver, especially in multiple engagement situations.
Lecture" 97

5._608 11 '
.... -3.00
~.-O.I~
Au -"')-

Fig. 7. The Partially-Vectored F-15 Fighter.

The new F-15 S/MTD (STOL and Maneuver Technology Demonstrator) is intended to flight-
demonstrate six key technology areas holding promise for ATF-type aircraft. as well as for improved
derivatives ofcurrent-technology aircraft. These technology areas include (13, 18, 62, 70, 71, III, 128,
234):
1) [FPC systems.
2) 2D-CD. TV/TR nozzles.
3) Enhanced lift systems. including independently-variable-incidence-canards.
4) Enhanced pilot/vehicle interface/avionics.
5) Rough/soft field landing gear.
6) Engine-intake distortion avoidance designs for PST-maneuvers.
The new fighter is approximately 10% statically unstable with the canards fixed subsonically (see
also § V.4). Note; The Soviets are currently testing similar designs on the Su-27-1024 agile interceptor.
(The unvectored Su-27 can fly up to 120· AoA during the "Pougachev's Cobra" Maneuver.)
Inlets Suction forces/moments play increasing roles as AoA goes to 90· and beyond. Increasing the
throttle may affect attitude. AT AoA = 90· the inlet force is perpendicular to the wing, and the lift force
vanishes.

It is. therefore. left to the pilot's judgement. whether or not to perform a PST-
maneuver. or to employ other maneuvers, such as various decoupledflight maneuvers in
which the vectoredfighter translates uP. down. left or right, without banking or changing
attitude. Such maneuvers may also confuse the opponent and may be combined with
modified PSM -PST-maneuvers.

II -7.1 Powerplant, Structural and Controllability Limits

Fig. 13 marks the limits given by engine power, controllability and structural
constraints.
98 Vectored Aircraft and Supermaneuverability

There is a low-speed region of possible sustained maneuvers for fighters with

TIW> 1.0. However, in combat, to retain high SEP (specific excess energy) may be
preferred for acceleration, or to gain altitude, etc. See, however, DST in the
Introduction.
Nevertheless, as the T IW ratio ofhigh-performance fighters increases, one expects the
relative time maneuver in the PST-zone to decrease, thereby to decrease vectored-fighter
vulnerability during such PST-maneuvers!
Fig. 14 demonstrates that for TIW < 0.6 there is no tactical advantage during PST-
maneuvers. Thus, many old fighters may still be upgraded to become vectored aircraft.

11-7.2 Other Requirements for PST-Maneuvers

Herbst recommends the following rules for PST-capability:
I) Sufficient control power in pitch, yaw and roll at Mach numbers as low as 0.1 and
with an incidence of up to 70 degrees.
2) High-angle-of-attack compatibility of up to 70 degrees, at Mach numbers as high
as 0.6 (4,000m altitude), with respect to aircraft stability and/or engine intake
performance (i.e., flame-out limits, etc.).
3) TIW> 1.2.
4) PST-Maneuvers are characterized by high pitch rates and rotation in yaw and roll
at the same time. Coordinated flight with zero sideslip requires rotation around
the velocity vector. Since the pilot does not recognize the velocity vector, the con-
trol system has to be designed accordingly. Thus, a lateral stick input would have
to produce more yaw and less roll at increasing AoA. However, Herbst points out

'L1GHT
IHYILOPI

MACH NUMBER

Fig. 8. Simultaneous yaw-pitch-roll thrust vectoring may also be used to expand the various performance
and steady-state flight envelopes of advanced, vectored/stealth cruise missiles. Advanced Cruise Mis-
siles (ACM) could fly by body lift assisted by thrust vectoring. Such designs may be almost impossible
to detect. High-energy boron slurry fuel may be of interest in some missions.
Lecture II 99

VIFFING IN COMBAT

~-
/-
/'

/
~
I
/
~

VRT as AH.<k.,

...,.--......
Climb .nd Flip

I
:f \ \
I
I

/
f
/
I
I
I
~
Fig. 9. Examples of viffing performed by a vectored aircraft in combat with conventional fighter.
During these schematic maneuvers, a vectored fighter may use thrust reversing, or, more efficiently,
a PST-braking maneuver (cf. Introduction, Figs. 4 and 5 and Figs. 1-2, 3 and 4). Herbst PST may be
effectively combined with PSM.

;?
Releasl'd
ALT-Ft
o WlthTR

Di~e- With Speed Brake

Turn Entry 10 Angl
Point 30·
Release- ~~~ s
POlnt/
.,....;
, Dive Angle
0~~~~~~4~00~-=~
Circular Error Probable, Ft.

Fig. 10. The simultaneous use ofTR, yaw vectoring and combined pitch vectoring may revolutionize air-
to-surface performance of vectored aircraft. (Dive angle is increased, error decreased. Cf. Fig. 15,
Introduction).

that this has caused some confusion among the pilots participating in his simula-
tions (for the pilot tends to use body axis as reference).
Consequently, Herbst developed a requirement for velocity vector roll acceleration.
It is ploted in Fig. 11 f as a function of AoA for a speed of M = 0.2 at 6000 m altitude.
 100 Vectored Aircraft and Supermaneuverability

1,7

'nr-----,-----~-----.--

IOOO~

0.' 1--I--;'---1I+-'i--I-- ~1"-I--fo:: 1

0.' __I-_',II""'"
~ _1_+-_1 __1
10

•• I.
10 U 14 ,. I. 'D
I. ..
EI"'t'ud "'.roll"'"
lO
f,,"_ hI
f~'l't.d 1;"'0 "I

a. Energy management comparison for minimum d. Angle-of-Attack during a typical air combat engage-
time maneuvers. Results of trajectory optimizations. ment. Result of computer simulations.

b. Typical air combat engagement of a PST fighter e. Typical air combat engagement. Relative positions
against a conventional a-limited fighter with all aspect of opponents depicting their aspects and firing oppor-
weapons. Result of computer simulations. tunities. Result of computer simulations.

..... ,T... r/.1

I I I
~~~--~-,--~--~~~---r---r~

1\ ~A. 07 , 1'\"

o I
"'" ....... ---- -- --
01 0'2 O.J 04 0$ 06 n.1 0.. n,,!,! :II' <0
........ Nu .......

c. Maneuver characteristics of a typical air combat f. Control power requirements under PST conditions.
engagement. Elapsing time in sec. Result of computer Statistical results of manned combat simulations.
simulations.
Fig. 11. Herbst's simulation results (see text and Figs. 2, 3, 4).
Lecture II 101

nt .• &1 '\,I," "'a I

20

10 i-t-i1~9==:::::;:4:::::::::t==:t ',---+--

01 .2 0> o. ., .1 0,1 02 OJ 04 05 06
MflchnUn'lbe t

g. Maneuver states in a horizontal turn (aircraft i. Maneuver states in a vertical turn (aircraft depen-
dependent). dent). Upper curve for top, lower curve for bottom of a
vertical maneuver.

Halo 01 lurll 1'/51

Improvement 01

;,[
50 ?O Avorago
n~dlll'
\0' TII,n
~O
15 \ Rare 01
\
\ / Turn

30
1.0
\
/
"
20

/
/ "
"-
0.5
10 I ~~

nil'" I
0. 1 0 .2 0.3 o .~ to 20 30 ~o 50 60 10
Machnlllll\)m 1"1 Angle ·ol ·A\lack Lrrnll

h. Comparison of maneuver cycles for minimum time j. Results of manned simulations representing real air-
maneuvers. Results of trajectory optimizations. craft and control dynamics. Standardized PST maneu-
ver with 180' change of heading. Control augmentation
Fig. II. (continu.) by meansofa ± 10' vectored nozzle in pitch and yaw.

For a conventional fighter such a high demand for roll cannot be met by aerody-
namic control surfaces. Hence, the requirement for thrust yaw-pitch vectoring
becomes a necessity to improve future, close-air-combat effectiveness to a degree
unachievable by conventional performance.
102 Vectored Aircraft and Supermaneuverability

11-7.3 A 180 0 Change of Heading Maneuvers

A 180· change of heading, with the additional constraint of returning to the point of
departure at the same speed and altitude, is shown in Figs. 2, 3, 4, 11.
Fig. 11 b shows trajectories of a typical initial phase engagement depicting two
opponents with equal conventional performance. However, the aircraft moving to the
left is un vectored and limited to a given value of maximum lift, while the secondjighter
has yaw and pitch vectoring, PST-capability of up to 70· AoA. Both aircraft are
equipped with the same all aspect weapons.
The associated time history of turn rate vs. speed and AoA is plotted in Figs. 11 h
and 11 i. Starting at high speed and the same altitude, both aircraft pull to maximum
load factors, and then, slow down to achieve the best instantaneous turn rates by means
of gaining altitude.
At the same time, a smaller radius ofturn helps to get the opponent into own weapon-
off-boresight-cone and keeps the opponent from achieving a similar objective.
Thus, a properly performed penetration into the PST-regime enhances both mini-
mum radius and speed reduction, thereby giving the vectored aircraft the jirst jiring
opportunity.

11-7.4 High-Alpha Supermaneuverability & 2 Flight Envelopes

Let us re-examine now the high angle-of-attack (AoA) flight in combat by highly agile
fighters. According to current thinking a high AoA, beyond stalling, will be the inevita-
ble key element in enabling fighters to survive and win in battles both beyond and
within visual range of the enemy. Thus,futurejighters will have 2flight envelopes; one
for beyond visual range and one for close-in combat. The aircraft would be required to
perform rapid transitions between them, for some beyond-visual-range engagements
will inevitably turn into close-in combat.
The 2 envelopes would be governed by four categories:
Category 1:
Use of vectoring and IFPC to gain increased turn rates for beyond-visual-range
engagements.
Category 2:
Use of vectoring and IFPC to gain improved turn performance at high and very low
load factors, and at post-stall and transient overshoot under maximum lift conditions
(fighter survival vs. missiles, defensive break-away, etc.).

Category 3:
Use of vectoring and IFPC to reach a high pitch rate to large AoA, followed by a
velocity-vector rol/, or a PSM, to achieve a rapid pointing-angle change.

Category 4:
Use of vectoring and IFPC to momentarily lower speed at very high AoA, and very
high heading-angle rates.
Lecture II 103

En,.,y
12

I 'O .~~~

O,D

0 ,8

02 0 ,4 De 00 1.0

Fig, 12. Energy management in a typical air combat engagement. Result of computer simulations. Cf.
Figs. 1-2, 3 and 4. Herbst [154].

Fig. 13. The PST flight regime (aircraft dependent). Herbst [154].

10~------~----'-----~---;r----'

0.0

08 ·~~~--~---+----~--~----
0.1 OR , 0 , :z 1. 4

'hrurt·tu 'WI'VP'U A.ftc»

Fig. 14. Thrust dependence of the time advantage in a minimum time maneuver with limited PST capa-
bility. Results of trajectory optimizations. Herbst [154]. Consequently, vectored propulsion adds
maneuverability to fighters having TIW> 0.6. Many old aircraft may, therefore, be upgraded.
104 Vectored Aircraft and Supermaneuverability

These categories may be used quantitatively to obtain dominance over the threat, in
terms of maneuverability.
However, these categories convey a glaring contradiction to some prior tendencies
to design the fighter as a 'stand-off missile launcher', with no extra need to maneuver,
and, therefore, with no need ofvectoring, high AoA, and agility. Nevertheless, the unfor-
tunate experience with the F-4 during the Vietnam War has put serious doubts into the
latter mode of thinking. There is also a semi-intuitive resistance, especially among
pilots, to slow down, even momentarily, in close-in combat. However, the only alterna-
tive to properly-calculated, and performed combinations of the aforementioned cate-
gories may be to stay out of the "future air combat". Consequently, there is a
well-defined need 'to expand envelopes'by a proper interaction between external flow,
engine intake, vectoring nozzles, and multiple control means such as ENJV, or canards
and the 2D-nozzle flaps themselves. These should be interrelated to other advantages,
such as stealth and all-aspect missile maneuverability. Combined, they supply the vec-
tored aircraft pilot with unique maneuvers to rapidly point, and, during the turn, to allow
the computers to calculate and, then, to lock-on weapon (either gun or missile), put it
within its fire envelope, and fire. It is the respective time delays of each 'unit operation'
which must be properly minimized and well-integrated (cf. 3.5 in the Introduction).
Thus, RaNPAS time for PSM (or PST) must be shorter than the combined computing,
locking, releasing and (high-speed, high-g) turn times of the next-generation, all-aspect
missiles in offensive thrust-vectoring engagements. And it is here that PSM by "no-tail"
PVA would emerge as the pace-maker, determining technology.
Moreover, with the possibility of rapid sideslip, or "jump", with or without thrust-
reversal, while commanding very high nose rates, one can put the aircraft on such a "zig-
zag" trajectory which will make it less predictable to adversary missiles, and confuse the
opponent in defensive engagements .. Similar conclusions apply to future agile (vec-
tored) RPVs, SAMs, and ACMs (cf. Lecture IV).

11-7.5 A Major Propulsion Problem: The Nonavailability of a PST Inlet

A major obstacle to gain the aforementioned advantages is not wing stalling but engine
stalling (at very high AoA and turn rates, low speeds and high altitude). Engine
controllabiltiy must first be improved in comparison with current engines - both by
improving compressor stall margin and by use of HIDEC or IFPC. However, special
inlet doors, ducts and ramps are now needed. Since other requirements may dictate a
special inlet design, the industry requires the development of a variable inlet which
may, uniformly, pass and compress air both from the wing's upper and lower surfaces
and, in some designs, from inlet sidewalls. This can be done, alternately, or simultane-
ously, depending on the type of mission, take-off, maneuverability and landing needs.
Such inlets are presently being designed and tested in various laboratories, including
this one.
 LECTURE III

THE MATRIX OF UNKNOWN VARIABLES

OF VECTORED AIRCRAFT
111-1 Reservations and Precautions
Fundamental concepts, components and unit operations associated with vectored
propulsion and supermaneuverability have been introduced and illsutrated in the
Introduction and in lectures I and II.
However, a matrix of unknown variables emerges as one tries to proceed and
interconnect them into cost-effective, fully, or partially-integrated systems.
This difficulty may be best examplified by any attempt to define and analyze IFPC
variables for pure or for partially-vectored aircraft.
Thus, in seeking realistic design goals for vectored aircraft, one may first identify the
main parameters to be considered, evaluate their relative importance within each
'engineering discipline', as well as their proper 'groupings' and 'subgroupings' for
analysis and/or synthesis of their coupling effects to each other.
Traditional subgroupings into 'disciplines' involves at least 22 categories; e.g.:
1) PSM, PST/Stealth inlet variables and technology limits.
2) Thrust-vectoring nozzle variables and technology limits.
3) Engine technology limits (Appendix B).
4) Electro-Aero-Thermodynamics and Low observability technology limits.
5) Aero-thermodynamics, IR and heat-transfer variables.
6) Wing/body/propulsion system integrated variables.
7) Environmental variables.
8) Expected threat variables.
9) STOL, V/STOL, VTOL, STOVL variables.
10) The basic and revisited variables of flight mechanics and flight control.
11) Redefined and revisited mission variables and analysis .
• 12) Payload/weapons/Propulsion/airframe coupling effects .
• 13) Structural and materials technology limits.
14) Stability, maneuverability, and controllability technology limits .
• 15) Reliability, durability, safety, and maintenance .
• 16) Hardware manufacturing constraints .
• 17) Cost-Effective objectives .
• 18) R&D funding/management through realistic milestones.

106
Lecture III 107

.19) Time-table for a realistic program with respect to the threat.

.20) Governmental approval/budget/control effects and delays .
• 21) Constraints in restructuring engineering and pilot education and training
(including the delayed-time optimized vectored trainers will be available to
the services) .
• 22) Proprietary and classified constraints.

* * *
By examining this list one realizes that there is yet no verifiable methodology
regarding the following questions: How shoud one structure these categories? Or how
should one evaluate their relative importance, as well as their coupling effects to each
other?
Moreover, thanks to the subdivision of engineering science into fragmented
'disciplines', we may fail, a priori, to perceive the 'interconnectedness' between these
'self-centered' categories, to judge their collective priorities, and to estimate their
inherent structure and ordering in conceptual and preliminary design.
In trying to overcome these problems, one may proceed with the problem of
ordering the rest of these lectures and appendices.
Indeed, at this stage of the course, one is faced with empirical, logical, and
subjectivistic discourses; empirical, because a great number of pioneers have already
produced an enormous amount of empirical data on components, unit operations and
partially-integrated thrust-vectored systems; logical, because all rational engineering
singles out regularity and logical order; subjectivistic, because all thought is, to some
extent, ordered by personal experience.
All combined, the feedbacks of conceptual design and experimentation, as outlined
in the illustration shown in the Preface, must be considered first and foremost. What
order we choose next, it remains only a minor link within the feedback networks of this
illustration .

• 111-2 PSM/PST/Stealth Inlet Variables and Technology Limits

A brief introduction to this broad subject is given in Appendix F: "Limiting
Engine-Inlet Envelopes"
'Classical' and 'Non-Classical' intake-design approaches are discussed therein,
including the near-term possibility to employ inlet-distortion-insensitive cold-jet
powerplants for VRA (cf. Figs. II-I i and I f).
PSM/PST/Stealth engine-inlet-compatibility will be enumerated in Volume II
together with the variables required for IFPC systems of future PVA.
Volume II will also stress the inherent difficulty associated with the 'undefinable
matrix of intake distortions' and the different design options available for
PSM/PST-tailored inlets.
108 The Matrix of Unknown Variables

.111-3 Engine Variables and Technology Limits

A brief introduction to this central subject is given in Appendix B. This appendix
stresses RCC-technology potentials and present constraints.
In fact, much of what is discussed in this volume, will become feasible when RCC
materials would dominate the performance of powerplants, pushing their TIW ratio
from present 8-12 values to 20 and beyond.
The matrix of variables associated with this category is well-defined in standard
textbooks on gas-turbine theory, and is, therefore, not to be repeated here, unless
associated with the following:

• IFPC trim and engine margins (Appendix F and Volume II)

• Thrust-vectoring nozzle (Lectures I, II, III and appendix C)
• RCC electro-aero-thermodynamics (volume II)
• Non-RCC cooling and heat transfer rates (see below) .

• 111-4 Engine Variables and Technology Limits

Some of these variables are defined and identified below in connection with
inlet/wing/nozzle integration and PST-flight conditions .

• 111-5 Electro-Aero-Thermodynamics and Low Observability

The material available in the open literature is to be collected and published within the
framework of Volume II .

• 111-6 Jet-Aero-Thermodynamics and Heat-Transfer Rates

The variables associated with this subgroup are well-defined theoretically and
experimentally, as will be discussed below.

• 111-6.1 Subsonic and Supersonic-Flow Heat Transfer

A major limit to the development of advanced powerplants is the current
non-availability of Reinforced Carbon-Carbon (RCC) composite materials for
hot-sections of powerplants.
A distinct and unique advantage of RCC is the elimination of present-technology
performance penalties to cool turbine blades, disks, AlB ducts, nozzles, etc., as well as
weight and part-number reductions. RCC is also an excellent absorber of
electromagnetic waves.
Faced with such potentials, this laboratory has been co-working with another
laboratory in this institute, on the development of improved industrial processes to
manufacture well-impregnated RCC materials. At the same time, an effort has been
Lecture III 109

Fig. I. An early design of a film-cooled, 2D-CD, vectoring nozzle [tested in this laboratory during
1980-83 for the evaluation of temperature distributions during subsonic and supersonic, hot-gas
vectoring, up to 900°C, and up to 8. = =+= 23' ).
a. (Right) The converging and diverging flaps of this vectoring nozzles were produced from the
(film-cooled) combustor liner of a GE F-404 engine (3). or-variations: ± 23 degrees. Hot-gas
temperature: 900'C. (Left) Rear view of nozzle at 8,. = O. Throat AR = 6.5. Option: Thrust reversal.
I. Cooling air inlet; 2. Cooling-air flap cover; 3. Diverging-vectoring flap; 4. Converging flap.
Notes: The truncated sidewalls are also being film-cooled. The duct upstream of the nozzle is a
water-cooled transition duct between an Allison-GM T-56 combustor and the hot, air-cooled,
vectoring nozzle. [Vectoring nozzle angles were varied during the tests.) Cf. Fig. I, Appendix E.

made to develop methods to cool vectored nozzles made from non-RCC materials. This
effort is described below and in Figs. 1, 2, and 3.
The vectoring nozzles shown are of the early type employed in this laboratory in
1980/3. They are limited in size to 0.61 kg/sec airmass-flow rate, and in aspect ratio to
less than 7. However, this was one of the first published studies conducted about
subsonic and supersonic cooling effects on 2D-CD wall-temperature distributions (8.
was variable up to ± 23 degrees).
Another study, to be described next, was published in 1982 by Straight and Cullom
of the NASA Lewis Research Center [48] .

• 111 -6.2 The Straight-Cullom Heat-Transfer Correlation for Vectored

Nozzles.
These authors stress the fact that while many tests have been reported on vectoring-
nozzle payoffs, cooling performance data, in the full-scale engine environment, are
scarce. Accordingly, they present baseline thrust and cooling data for a 2D-CD nozzle
mounted on a 1-85 turbojet engine in an altitude chamber.
Straight-and-Cullom tests cover a broad range of pressure ratios, nozzle throat areas,
and internal expansion area ratios (see Figs. 4 to 7). Their nozzle configuration is
shown in Fig. 4, while the cooling methods are shown in Fig. 5.
Vectoring was conducted by two methods:
• Nozzle-flap vectoring (using interchangeable nozzle-flap cooling panels) .
• Pivoting the nozzle throat.
110 The Matrix of Unknown Variables

...
.2" Q9
...111
"
0-
E
"
t-
OO

"
N

111
E
....
o
Z
Q2 0.4
Dimensionless Length

W""1iF .~

'0
~ U
~
1! ...,~
Y i.J6
~i~ \1~\
-"
1\
N
111
nTgr'1
'\ 1/ -' V\ {1
z
...0E ~I
t- ~t;
Ol 0.4 0.6 0.8
Dim~nslonless Length

.Fig. 2a. The effects of2.5% cooling flow on the temperature distributions in the 2D-CD nozzles shown
in Fig. 1 (fiIm-cooled) and Fig. 3 (uncooled).
The 'sideline' distribution is a 'corner' temperature distribution (upper figures). The lower figures
are for nozzle's center-line.
Cooling-air temperature was 26.2°C; Tg = 683°C; A t = 11.2 cm 2 AelAt= I, 8. = 0, Ala =
0.669 kg/Sec (in the un cooled model); Ala = 0.612 kg/sec (in the cooled model). The test facility is
described in Fig.IV-7.

To keep costs down, their design philosophy was to use simple boilerplate struc-
tures, but keep realistic flow paths.
Simple, leaf-type, low-cost seals were used throughout to minimize leakage.
Lecture III 111

.111-6.3 Cooling and Leakage

The maximum performance potential of the nozzle was determined by correcting the
measured thrust coefficients for seal leakage and cooling airflows that bypassed the
nozzle throat. Their data corrections indicate that substantial thrust penalties could
result from coolant bypass flow and leakage, and that all significant leakage occurs
upstream from the nozzle throat. Moreover, they report that effective leakage area is
only a function of differential pressure (independent of whether the leak is subsonic or
sonic flow), and that the leakage area is equivalent to a simple-flow orifice (and not to a
series of flow restrictions).

0.5
Dlm.n$lonl.s. L.ngth

Q9

Convrrgrnt flip Dlv.rQ"nt Flip

-
~
t2
tl
~~I;~
....c ,
.... _-..(J"" .... I

0.70 ~
Dlm~n$lonl~s$ L.ngth

.Fig. 2b. Temperature distributions in the 2D-CD vectoring nozzle shown in Fig. 1 during supersonic
vectoring at 8. =23· and Tg =900·C, (upper figures) and during subsonic vectoring at 8. =21· and
Tg = 900·C (lower figures).

Notes:
(I) The throat is located close to the normalized length-value 0.5.
(2) Black points represent test results for the lower flaps, while the open marks those measured on
the upper flaps.

The experimental results demonstrate that there is a considerable difference between center-line
and nozzle-corner temperature distributions, and between heat-transfer rates in the subsonic and the
supersonic domains. This difference increases with NPR and Ov [103].
112 The Matrix of Unknown Variables

Fig. 3. The first 2D-CD vectoring nozzle tested in this laboratory in 1980 was uncooled.lts dimensions
are similar to the film-cooled nozzle shown in Fig. I. Hot-gas temperatures for this model were limited
to a maximum of 1300·C. Cf. Fig. 2a

.111-6.4 Flow Discharge Coefficient - em [Eq. 7 below].

This coefficient was determined by ratioing the effective area of the throat to the
actual measured area (cf. Eq. 111-7), after correcting the data for leakage and coolant
flows that by-passed the nozzle throat (Fig. 5).
The effective area was computed from the corrected values of throat flow rates and
mixed-hot-gas temperature and the measured values of total pressures. Similar effects
of cooling and leakage on efg have already been illustrated in Fig. 1-2 .

• 1I1-6.S Supersonic Hot-Gas Heat Transfer

The hot-gas-side heat-transfer coefficients were computed from the average bulk-to-
wall temperature differences for each cooling panel, and the heat flux from the temper-
ature rise of the coolant. The data are plotted in Figure 6 as a function of hydraulic
diameter of the flow path between the two divergent flaps, and also compared with
predicted values on the right-side of the figure. The data covered a range of Reynolds
numbers from 7.0x 10 5 to 2.0x 106, computed on a hydraulic-diameter basis. Film-
property values were determined from the average of hot-gas bulk and wall tempera-
ture. The equations used for the predicted values are:
Lecture III 113

Nu = h fH = 0.023 ReJ-8 Pr 04 (1)

j

where

Ref= DHVPr (2)

)1f

The experimental values are as much as 40 percent higher than predicted at the
higher heat-transfer rates. It is apparent from these data that a more rigorous analyti-
cal approach is needed to improve the predictions (which should include the effects of
shocks).

Fig. 4. The cooling methods employed by Straight and Cullom (48) using a research turbojet engine (the
GE J-8S) equipped with a 2D-CD pitch vectoring nozzle .

• 111-6.6 Accelerating Subsonic Heat Transfer

When the standard equation (1) was used for predicting heat transfer on the converg-
ing flap, the correlation was poor. Significant variations of heat transfer coefficient
occurred with change in the angle of the convergent flap to the hot-gas stream. The
data were then re-plotted as a function of the flap angle in Figure 6, where it grouped
about a single line. The empirically derived equation for the line shown is:

Nu = 0.023 Re~·8 PrO. 4

cf.8)
( 1 + -ts (3)

This equation is the standard equation (1) with the addition of a term containing the
angle of the flap relative to the hot-gas direction entering the convergent portion of the
nozzle. The form of the added term was chosen such that when the flap angle is zero
(parallel to the hot-gas flow), the equation reduces to the form of equation (1).* No
attempt has been made to date to explain the results analytically.

* Stoll and Straub [J.F. Turbomachinery, 110,57 (1988)] have recently reviewed some previous stud-
ies in film cooling and heat transfer in nozzles while adding their own studies of a water-cooled, low
AR, 2D-CD nozzle constructed from copper.
114 The Matrix of Unknown Variables

.Fig. 5. Straight and Cullom vectoring, 2D-CD nozzle during cruise or vectoring.

111-7 Wing/Body/Propulsion System Sub integration

111-7.1 The Primary Effects of Vectoring Nozzle Aspect Ratio
While NAR emerges as the most important parameter in pure vectored aircraft, other
parameters must be considered as well. Eqs. (3) to (31) below sum-up the main parame-
ters which must be defined for pure vectored flight-propulsion.
Let us first examine the effects of NAR on some preliminary design
considerations.
As NAR is increased, the relative 'heights' of the nozzle exit components decrease,
in proportion to the increase in width (cf. Figs. ILl a, 1b ). However, total exhaust
system's length may increase, because the circular-to-rectangular transition duct may
have to become longer to maintain uniform flow distribution and minimum
distortion/separation effects. Moreover, it may also include a rectangular afterburner
(an unorthodox, but feasible design!)
A whole-system-design-approach to the entire matrix of problems involved with
high NAR systems includes:
1) Use of streamlined-jlow-distributerslstruts inside high NAR nozzles.
2) Use of these distributors as structural elements in the wing-structure design for wing
weight reduction.
3) Use of the wing's skin and internal structure to reduce the weight penalty of high
aspect-ratio nozzles.
4) Use ofupper and lower uncurved panels to reduce production costs and improve inte-
gration with aft-wing sections.
5) Calculate transition-duct cooling loads due to increased heat-transfer area. This duct
may also form a new-type of rectangular afterburner.
6) Improve cooling loads and IR signatures by use ofinternal wing mixers and inlet pas-
sage for cooling airjlows.
Lecture III 115

Nu= h DH_Q023Re-U Pro,r. (,+ (/.Co,,)

K IS

.Fig. 6. Straight and Cullom impingement cooliing heat-transfer coefficients (upper drawing), super-
sonic. hot-gas. heat-transfer coefficients for cooled divergent flaps (drawing at center), and a heat-
transfer correlation for vectoring. 2D-CD nozzles. including the effects of converging flap angle in
accelerating hot-gas flow (cr. eq. 2) (48).

7) Use RCC materials (see below).

8) Reduce boattail drag by high NAR.
9) Use sidewall truncation (which does not reduce Cfg ) - see Figs. 1-15 and 1-16.

111-7.2 Aerodynamic Effects; Regrouping the Main "60-Variables"

Equations 2, I-II, 12, 13 and 14 specify and define the following variables: h, Du , k,•
V, PJ' JJJ' Cp , ae , D., D., Cli:' M acl ' T r , Pa, Pr and y. Aside from NAR, and the various
nozzle-wing variables defined in Fig. 10, one must also evaluate the dependence of
engine inlet, and C'i:' CDS' CA and C.(see below), on engine, and on 'externalflight con-
ditions'. The result is an unknown matrix ofcoupled variables. Thus, in trying to reduce
the confusion associated with this unknown matrix, one may first group the corre-
116 The Matrix of Unknown Variables

.Table 1. 2D-CD configurations tested by Straight and CuUom [48].

For further details see Appendix C.

Configuration Nozzle throat Nozzle Nozzle Convergent Divergent

designation area, A, area throat wall wall
expansion aspect angle angle
cm 2 in. 2
ratio ratio ·C ·C
AeIA, deg deg
AI 707.1 109.6 1.20 4.10 28.75 2.19
A2 1.50 5.66
A3 1.79 8.87
A4 2.29 14.60
BI 902.0 139.8 1.20 3.19 21.36 2.86
CI 1130.4 175.2 1.20 2.55 13.15 3.58
C2 1.50 8.91
C3 1.80 14.31
C4 2.30 23.72

____ IOtAL
_ _ IXPUIMENTAL
CONfIGURATIOtI ARtA RATIO. AJ"
0 CI LZ AI U
0 C2 1.5 A2 U
0 Cl La A) 1.&
A A C4 2.) I M 1.1
..! • L2 \.5 1.1 1.)
,U..... _J.L,..-. ____ .!:.'
lQO " (-''''-
;. ,.., ,~----:.o-:;::::--- ...
U

...
" ,. ..... '"")<.. --- --
, ..,~
0 ....... , --....
.:t" --~
'-...-
.'111 I '",'
I , ... .'111
.:t"
. I / / .., A ...............
• 96 I

I ' I /
I' ..
••
.94 I
I
fI II
'
I I .M
.92 0 II IILL-L-l I I I .n0
2 4 6 a 10 12 14 16

.Fig. 7. The NASA-Straight/Cullom test results for the 2D-CD vectoring nozzle show how C" depends
on NPR and A91A8 (AeIA,) values (48).
Note to the designer: Unlike the theoretical predictions, the actual test values presented have demon-
strated a negligible dependence of eJg on the expansion ratio AeIA"
Provided these results apply to other vectoring nozzles (the data presented here are for AR < 4.1
and the GE J-85 engine), these results may have a significant effects on installed (integrated nozzle-
wing structures) systems, especially in terms of reduced drag values associated with (Iower-than-
theoretical) AeIA, values at high NPR values. For further detail see Appendix C.

sponding variables, and, then, try to identify the more important ones, and/or select
those subject to experimental verification in wind tunnels, in the jet propulsion labo-
ratory, or in flying vectored RPVs (specially, and systematically designed to investi-
gate certain unknown variables). There are, nevertheless, a number of variables which
do not lend themselves to simple experimental (or theoretical) evaluation. One of
them is defined by Fig. 10 and the following equation.
Lecture III 117

Tolal- Temperoture Contours for

AD(N, NPR • 2 .16, BPR • 0

PftWA 20·CO NOULE. AR · u. •. NPA . J. I'A . 0

A·A B·B c·e
Cross-Sect lon.l Tot.l- Tempe.ature
Contours 1M ADEN, NPR • 2.16.
COIIIIlarlson of Tota 1-Tetnj>erature Contours BPR • O.
for 2D-CD Noules .

• Fig.8. Chu and Der 1911 calculated the temperature distributions inside high-aspect-ratio vectoring
exhaust jets and compared their results with experimental data (lower left). Their corresponding distri-
butions for low-AR nozzles are demonstrated on the upper left corner of this figure.
As expected. the high-AR nozzle lowers IR signatures. A similar study for low-AR ADEN nozzle is
demonstrated on the right-hand side of the figure.

c' = C + Crcos 0" + C~ COS 0,., COS 0,. (3)

c:
However, instead of the unknown variable (Eq.5), one may use the experimental
results of Figs. 18 and 19 below to estimate the main contributions to supercirculation
and direct lift. Yet. this would be a mistake. for one must first examine the whole inte-
grated system together. e.g.:

(L1Cdmax = (L1Cdsc+ (L1Cdv=

= f(NAR, C~. a, Ydist.> Ov. 0,.. CIl , {3, {3, M, G.E., H, Cr,
BLS, SSW, T T8, Ta, Re, IFPC, FCE, ... ) (4)

* In the x-direction. Unless otherwise specified we assume that the engine (i.e., its center line) is
installed with 0,. = 0 with respect to the aircraft zero lift angle of attack. Cruise drag considerations, for
instance, may add a few degrees in engine installation and the calculations given here should be
adjusted accordingly.
118 The Matrix of Unknown Variables

1200" ,CICIO'
.2
~ 1 .. . . t -= : : : : - - - - - - - - ,
1 •.&00
WITHOUT SWIRL

~t.200
~ 1,000

~ 100

~ ~L-~~~~~~~~~~J

Illustration of a 30% Reduction of

Comparison of Total-Temperature Contours Plume Total Temperature Difference
for a 2D-CD Nozzle with Bypass Air . Due to Engi ne Swi rl ,

'OO:t§5
14000 ,00000R 7000 R

::,2000
GOO·

~-
1100'
s>~ ,GOO"
'000" 1700" 1000·
SIDE VIEW 800"
1100"
r A rB r C
GOO'R
1200"

eooo

A-A B-B C-C

Total-Temperature Contours for ADEN, Pred; cted Total-Temperature Contours

NPR - 2. 76 , BPR = 0.25 . for ADEN (NPR·2.76 , BPR-0 . 25) Under
the Influence of Engl~e Swirl

l000 0 R
700'R

d'~'
800'
1GOO' 900·
11000
'000"
"00"
'200"
Il00'

A-A 8-B C-C

Cross-Secti ona 1 Tota 1- Temperature Measured Totifl- Temperature Contours

Contours for ADEN. NPR = 2.76. for ADEN (NPR=2.70, BPR-O.25) In the
BPR = 0. 25. Presence of Engine Swirl

.Fig.9. Various effects of engine swirl, bypass air and AR for 2D-CD and ADEN nozzles were also
reported in Ref. 91, by Chu and Der. Note that the IR signatures depend on the fourth power of the
absolute temperature.
Lecture III 119

Tv

c'
c
c'

Cf

Fig. 10. Preliminary definitions of an integrated wing/exhaust vectoring nozzle.

Notes: V-adjustments require (cos 8,)-1 and (cos 8.)-1 increases in throat area (cf. Eqs.I-5, and
Fig. 1-5) and A9 (Eq. 1-6). Analytic research in propulsion and aerodynamics must be integrative.
Thus, vectored aircraft design calls for full integration of engine exhaust nozzle, the wing, aerody-
namic control surfaces and the internal/external streamlined flow dividers (which become structural
nozzle-wing struts).

C:= lI'(a, NPR, Crg' C D8 , CA , C V , to, NAR" G.E., 0., 0Y' M, H,

T, /3, p, a, BLS, NG, Turb., ... ) (5)

Cfg = 'l'(NAR, NPR, 0,>,01' Cf , C-R ducting and splitting, flow dividers/
Structural struts, IFPC, ... ) = TITi =
= Actual thrust/Ideal thrust = (6)

CCA M 7 actual. V9 /g+ (Ps9i - Pa ) · A9 -LlC(g(Leakage+Cooling)

Vs' M7actual/g

(cf. Fig. 12)

120 The Matrix of Unknown Variables

Aes + Ae[Leakage] + Ae[Cooling]

(7)
As

(cf., e.g., Figs. 11, 16 and 12)

c.~ = Thrust Loss due to the nonaxial flow from the Nozzle (see eqs.

17,22,23 and 24 below and Fig. 13). (8)

C.= CfgpeakiCA (see also eq. 6) (9)

CfgP-rsAs· sin 0•. cos Oy. cJ>(Ys, P1'8 /Pa) Vs

'hUS•.T,S

(10)

NPR = f'(RPM, Pn,w, EGT, t, 0", OJ" TR, AB, M, IFPC, T a, Pa' NG, Degr., Trim,
ACC-D, FQ, Cooling, Leakage, fan-core mixing, ... ) (11)

Note: The thermodynamic stations are different for subsonic (converging), or

supersonic (converging-diverging [C-D)) nozzles. Thus, for a supersonic nozzle,

T= q>(M, H, t", Inlet, a, /3, Cfg , CDS, C A , C v, NAR, NG, W, Vs ' As, IFPC, AB, TR, T a,
Degr., Trim, ACC-D, FQ, Ys, P 1'8 ,Pa, ••• )

(12)

Substituting, one obtains

(.1cL )max= f (NAR, Cf , 0" Oy, w,M, H, t", Inlet, Cfg , Pn,P(p Cp ,P1'8 , Ys, C, CDS, CA , C v,
a,
RPM, NG, TR, EGT, IFPC, Vs , Ydist.> G.E., a, /3, 13, BLS, ssw, Re, FCE,
Fan-core Mixing, Turb., Degr., Trim, ACC-D, FQ, M s , T 1'8, As, ... )

(13)

One must also stress that the location of CP, for instance, varies with the degrees of
vectoring. Moreover, the evaluation ofFCE is a broad subject in itself(cf. Lecture V).
Lecture III 121

CIS
CD8
Q7

0.5

NPR

Fig. Ila. CDS as a function ofNPR for a subsonic, circular nozzle. (Data from the Turbo and Jet Engine
Laboratory-Technion-IIT, 1987).
Notes:
I) In developing and testing subsonic or supersonic vectoring nozzles, one must first compare their
performance to a circular (axisymmetric) nozzle, in order to asses their efficiencies with and without
vectoring.
Hence, such a Laboratory Data Base (LOB) should contain the dependence cf erg, eDS, and e,. on
NPR for an axisymmetric nozzle isntalled on the same type a/mission engine, and having the same
As and A91As values.
Facilities required for such LOB tests are shown in Figs. IV-6 and IV-7.
2) The next LOB should include a comparison between vectored-system performance with a standard
bell mouth inlet and that with high-alpha-beta-PST inlet designs that characterize subsonic or super-
sonic, partial or pure vectored aircraft.

1,00

.98
COS
,96 ~

.94:'0---:!""-,...-:-'::--~

.Fig. II b. Variations of CDS with NPR for low-AR 2D-CD vectoring nozzles with various A.IA, ratios
(48).
122 The Matrix of Unknown Variables

There are at least 45 parameters in Eq. (13). However,for preliminary design consid-
erations, the most important ones are only ten, namely:

(ACdmax ~f(NAR, oV, 0Y' CIl , M, H, Crg, CDS' G.E., TR) =

~ (ACdsc+ (ACdv [see also Figs. 18 and 19] (14)
p,
Inlet and nozzle geometric designs, as well as BLS, {3, etc., are important in analyz-
ing aerodynamic effects in PST maneuvers. Thus, a new family ofsubsonic and super-
sonic inlets must be designed and flight tested for PST angles of attack.
For NAR < 10 designs one may use (cf. Fig. 19):

(ACdmax
=4 (ov:$ 20) ( 15a)
(ACdv

(ACM)max
=2 (0,,:$ 20) (15b)
(ACM)v

(ACD)max
= 2.5-3.5 (0,,:$ 20) (15c)
(ACD)v
When Ae(Leakage) and Ae(Cooling) are negligible, eq. 7 reduces to

(16)

Eq. (16) is the ratio of effective to geometric cross-sectional, throat area of the nozzle
during unvectored or vectored operation (cf. eqs. 1-5, 6, and Fig. 13, Introduction).
Examples of CDS variations with NPR are given in Figs. 11 to 12. The effects of cooling
and leakage are shown in Fig. 1-2 and also in Figs. I1I-2a and I1I-2b.
Cfg accounts for all nozzle losses, while CA accounts for the thrust loss due to the
nonaxial exit of the exhaust gases from the nozzle.
In axisymmetric (rounded) nozzles the exit angle ai varies from zero, at the nozzle
center line, to i1 at the outer radius. Hence, for a constant Mi per unit cross-sectional
area

(17)

or, more generally, when swirl still exists,

i 2K ir-<pI2
/1-0 0
r cos ai (r, lJ) dr dO (17a)

However, whenever Mi varies with the unit cross-sectional area, as may be the case in
some high AR nozzles, its functional dependence should also be integrated.
It is also beneficial, to include streamlined flow-dividers/structural struts in rela-
tively high aspect ratio 2D nozzles. The streamlined struts may reduce overall weight
Lecture III 123

~~
~

I
C.
3J
4

I
C.

I 2 4 6 8
NPR

I) _0°
2) _7° pitch up
3) _7° pitch down
4) -20° pitch down

Fig. 12. Capone and Berrier of NASA have demonstrated that there is, in principle, fl difference between
geometric nap vectoring, B", and actual jet deflection 8t.ct. 1_0°, 2-7° pitch-up, 3-7° pitch-down, 4-20°
pitch-down. In SERN, and in other types of nozzles, this difference may be substantial. However. for
the 2D-CD type of nozzles the difference is quite small and. consequently. may be neglected in prelimi-
nary design estimations. For instance, the mistake in B~C! varies from 4 degrees. at internal subsonic
conditions. to I degree in supersonic conditions. The data presented here are for a 2D-CD vectoring
nozzle. [NASA, Capone and Berrier, 1980, Ref. 56]. This figure also shows a drastic reduction in CD8 at
Bv = 20 degrees.
Further data are given in Appendix C.
Notes for the Designer:
I) Constructing engine test rigs with multiple-degrees of freedom to simultaneously measure B,., By,
Tx. Ty and Tv is not only expensive, but time consuming.
2) Precalibration of o:C! with B", and atC! with By, combined with detailed experimental data on Cfgdur-
ing Bv. By vectoring angles (at each NPR value), allows the designer to employ engine test rigs which
measure only Tx , and, then to use eqs.I-II, 12 and 1-13 to calculate actual Tyand Tv values.
124 The Matrix of Unknown Variables

,~ ,Ie 1J .1 21»
A,/A •
• Fig. 13. An example of CA (axisymmetric. C-D nozzle angularity coefficient) variations with the diver-
gence angle II (see eqs.8, 17, 22, 23, 24 for details and explanations in regard to vectored, 2D-CD
nozzles).
a is defined in Figs. I-I and 1-5. Data taken from Kuchar's fundamental reference (138).

1.'

1..l--~

.Fig. 14. Stevens-Thayer-Fullerton's estimated effect of aspect ratio on exhaust system weight (49).
However, the actual weight may be considerably reduced by full integration ofhigh AR nozzles, such
as shown in Figs.II/·lO and II-I, with the wing structure.
This was recently demonstrated in this laboratory using yaw-pitch vectoring nozzles having AR
between 14 to 46.6.

of the nozzle in a well-integrated design with the external winglbody, while also reduc-
ing thermal signature and increasing Cfg and CA values.
The streamlined struts may be so designed as to distribute Mi evenly and reduce swirl
across the throat and exit areas. thereby allowing not only the use ofthe simplified equa-
tion (17) but. also. to increase CA and C. values (see, however, below).
Primitive transitions ducts may be designed by following eqs. 18-20 (cf. Fig. 17);
Lecture III 125

z' (x')/)(X') y'(x')I)(·t')

r'+ [~~~ rl'''"

+ =1 (18)
a(x') b(x')

;(x) - {[ ~;::) (19)

y'(x') = (x') sin ()

r' }'"'
z'(x') = (x') cos ()

I(x)· {[~;:) r' + [s~(:)

0 0
::5 ()::5 38
(20)

y'(x') = a(x') tan ()

z'(x') = a(x')
38 ::5 () ::5 90 0

y'(x') = b(x')

z'(x') = b(x')
tan ()

.111-7.3 Nozzle C" CDS' elK Coefficients Revisited.

Being related to the viscosity of the gases, C v is essentially a function of the "secondary
nozzle surface area" at stations 7, 8 and 9, and of the velocity distributions near the
solid walls, and, thus, of the Reynolds number.
However, experimental and analytical studies show that Reynolds number effects
are less than ± 0.2% for all exhaust nozzles and for all operating conditions, thereby
allowing easier 2D nozzle scale-up and scale-down development efforts.
The main difficulty associated with the development of high-aspect-ratio 2D vector-
ing nozzles, is the fact that, while CA and C,. may be estimated from analytical studies
of 2D nozzle geometries, it is extremely difficult, and virtually impractical, to deter-
mine them by experimental methods (using, say, very precise flow angle measure-
ments in subsonic and suersonic flow fields).
A practical solution to this problem may be the evaluation of either e,. or eA from
eq. (6) and the definition V9i= Vs when Ps9i= Pa, viz.;

Cfg(peak) = (C,C,A )Peak (21)

e
Since maximal fg values are readily determined experimentally for each vectored
position (at any pitch, yaw and thrust-reversal angle), the problem may be reduced to
calculating either CA or e,..
Comparing the difficulties and the associated reliabilities of various candidate theo-
126 The Matrix of Unknown Variables

retical approaches, one may realize that CA estimation is a practical approach, espe-
cially when CDS values are known experimentally.

~
CD
C.
,.00

u~ ::::
f 3.5
Min Jet Area

1.0_----01----...:.AR~.

I.'- ......L-_..l-_....J._ _
15.0

.l..-_-.L._=_.J.

Max Jet Area

0.99'[

0.9af..,·;.;;.0-----<~

0.97 - I -...!...
o 0.2 0.4 0.6 0.8 1.0 1.2
Log AR

.Fig. 15. A main design parameter is NAR (cf. eq. 14). While it affects supercirculation (Fig. 18). it
causes only minor changes in nozzle efficiency. Data taken from Ref. 49.

r-r-'-

- 1-. Side
View

~- f- ......

234 ~
I-r .....

-+--
~ Top
ViE'w

-r- ...J

.Fig. 16. An example of a C-R transition duct following the method demonstrated in Fig. 17 for low AR
vectoring nozzles. However. our recent laboratory test results with yaw-pitch vectoring nozzles having
AR between 14 to 46.6 have demonstrated the need to replace this method by more advanced methods
which do not add much weight. nor reduce Ck. CDS and other efficiencies of the HAR nozzles.
 Lecture III 127

-(X')

y'

"
-+___ ---¥'------L_ -.j, '
'->/ 111>iill1bMllr--- "
'(1 ')

.'
Fig, 17, Internal geometry of transition C-R ducts for 2D-CD vectoring nozzles characterized by low
aspect ratios may first be estimated by the equations of a superellipse. However, for high AR vectoring
nozzles this technique may not be suitable! Cf, the Notes in Fig, 16.
The superellipse equations may be expressed as in eqs, 18- 20.

TIt
40

a;
0
t;
20 \' r--... ?r-...

\I~.> ~.0-1 ,--II

~
z
--=c-. - ~
---·n
=i 1"'-- '
" '"
I\M
10
Ii: \: P ' ~ .:.-.,

. -,'
8
3
0
w 6
V ~
....... -- / ~
C .'-J ,~
I;; < ....... - -
a 4 _ s,
::l

<I:
~ AFFECTEO _ .........
.-
-3
~
VI-

.~
·'j'T~
s::.. <. <~
1
0.004 0.0060.0080.01
I 0.02 0.04 0,06 0.08 0.1 0.2 0.4 O,G
So - MOMENTUM COEFFICIENT

Fig. 18. Supercirculation Lift Gain Correlation 1164-1671. Cf. eq. 29 and Fig. 19.

Now, for RCC-type vectoring nozzles,

C _ M7Actual ~ A,'8
(22)
/)8 - M 8 Ideal A8 '
for no-cooling would be required and leakage losses may be substantially reduced (in
comparison with the multiple flaps leakage of conventional nozzles).
An example of C m dependence on NPR for a small turbojet engine used in this lab
128 The Matrix of Unknown Variables

(T ~ 350 kg(f) at S.L. standard conditions) is shown in Fig. Ila. These results were
evaluated from readily available experimental data on NPR, M 7 , T 1'8, and Ag. It
should be stressed, however, that the evaluation of M7 by a simple bell-mouth inlet
excludes the possibility oftesting [installed] vectored nozzles with various 2D [flush-
mounted] PST-tailored inlets.
Now, introducing small D,.- and Dy-flight-controlled variations, eq. (17) may be
employed to re-estimate the efficiency of vectored nozzles.
A
(CA)v = sin Dv • cos Dy • -A
1Z=8cos av(z)
_ dz (23)
9 z-O

(23a)

In rectangular vectoring nozzles the ducts are normally converging (subsonic noz-
zles) or converging-diverging (supersonic nozzles) only with respect to the upper and
lower surfaces.
Hence, when u,.(Y) = 0, i.e., in lTV-nozzles with fixed, parallel sidewalls, eq. (23a) is
reduced to

(24)

The split-up of CA into z(pitch) and y(yaw) components may be required in overall
moment calculations of vectored aircraft. (Crg • c,. and CDg may thus be split accord-
ingly.) It may be noted, however, that T itself varies with D,.• Dv. a:*, NPR, NG, NAR,
etc.
Typical data for CA are shown in Fig. 13.

III-7.4 The Range of Thrust Vectoring

Our laboratory tests show that Crg does not decrease much with increasing NAR. This
conclusion remains valid up to NAR ::;;;: 50. However, special, internal, flow struts and
turning curves must be designed for each NAR. We also found that relatively small var-
iations in Dv and Dy cause substantial effects on flight performance. Simple Dv-, Dy-effects
may be easily estimated from eqs. 1 and 2. However, the effects on Crg must be evalu-
ated on well-calibrated, thrust-vectoring test rigs. Such studies, as well as our flight
experience with pure, vectored RPVs, show that, with a variable canard:

-20:s D,.:S 20 (25)

-35:SDv :s 35 (26)

* The y and z-coordinates are the pitching and yawing axes, respectively.
Lecture III 129

However, most of the low altitude, high-tum-rates may be accomplished with

-10.:s 8v .:s 10 (27)

-10.:s 8y .:slO, (28)

or less.
The range of Eq. (26) has been found useful during runway taxiing without
front- wheel-steering.

By Supeorcirculatlon
0.8
0.7
C 0.6
L,TOTO.5
~~a::z::ZL..:~~
O.
n3.~ __ ~
By Dlr«t__Thrust
____~ ~ ____ ~

-0.30
-025
-020
CrT\}OT,,5 C
-0.15
-0.\0
-0.05
O~ __~__~~__- L_ _~
0.\8
0.16

OlO
OD8~iIII:iL,;;~""","""',",,",'-4I
OD6~ __~__~~__~__~
o 10 W 30 ~

Fig. 19. Supercirculation increases the maximal values of CL. Cm and Cf)o
Upper shaded areas represent supercirculation, i.e., (LiCdsc' while the lower shaded areas represent
direct lift generated by Tv= T· sin 0,.. cosoy , i.e., (LiCd,.(cf. eqs. 10 and 4). The "de1ta~ represents val-
ues beyond maximal values obtainable with (unvectored), current-technology wings. For a lift gain
correlation see Fig. 18 and eqs. 29 and 30.
The data represent computational and experimental data for relatively small NAR values.
Notes
I) Direct LiCM variations due to engine thrust are virtually the changes in pitch moment due to
cos Oy. sin 0,. engine variations (shown here only for Oy= 0).
2) Direct LiCD variations due to engine thrust are the changes in drag due to cos Oy . sin 0,. engine varia-
tions. For additional effects see eq. 31, Fig. 18 and the appendices.
130 The Matrix of Unknown Variables

111-7.5 Supercirculation in Vectored Aircraft.

For subsonic flight speeds the additional, supercirulation lift may first be estimated by
means of the 'adjusted gain factor G' of Fig. 18 defined by

,(Sw) _ (CII)Sin(a+ov)+(Cdsc (29)

G=G - -
Sj (CII)Sin(o+ov)

where G is the usual lift gain factor, Sw the total (or reference) wing area, Sj the
superciruc/ation affected wing area (which is defined as the portion of the wing area
included between chordwise lines at the inboard and outboard ends of each exhaust
nozzle, a the angle of attack (AoA), 0. the pitch-vectoring angle (counted positive in
downward rotations of the exhaust nozzle flaps), CII the exhaust gross thrust blowing
coefficent referred to the flight dynamic pressure and wing area, and (CL )sc the increased
lift coefficient due to supercirculation.
Apparently all rounded (axisymmetric) exhaust nozzles are potentially incapable of
supercirculation gains in lift. Hence eq. (29) is suitable only for rectangular-type (two-
dimensional, or non-axisymmetric) nozzles which are fully integrated with the trailing
edge of the wings as shown schematically in Figs. 11.1 and III. I O.
The available experimental data of references [72, 164-167] indicate that, for a
given aircraft configuration, G is relatively independent ofMach number over the sub-
sonic range, but is strongly dependent upon the momentum (blowing) coefficient and the
affected wing area Sj.
The experimental correlation of Fig. 18 succeeds in coalescing the majority of the
data points into a single line defined by
(30)

.111-7.6 Induced Drag Due To Supercirculation

Associated with the supercirculation lift is an induced drag force, Dr. The available
experimental data on this effect (References 165-167) has been analyzed and corre-
lated in terms of the efficiency factor er(which may be estimated as equal to 0.75 and
is relatively independent of CII and ~/SIV). Thus,

_ [(CLjO + (Cdsc)]2- CLo

CDr- 1 (31)
0.75nAR
where C LjO is the 'jet-ofT' lift coefficient.

.111-8 Engine Sizing Problems

An interesting study has been reported in 1986 by Brooke, Dusa, Kuchar and Romine
of the General Electric Co. While studying 2D-CD ejector cooling systems for vec-
tored propulsion they have raised the following design dilemma. Non-RCC, 2D,
Lecture III 131

.
7
• Test-5
~~
~

tf ll.
ll.
ll. Test-6
II

I
...
~ .=;.
ll.
~

-.
'b
\~
£-
u •
ll.i~
2
~~A&·t
LL ~q~~ ~.:l\
en l

lruD I4COJ 18(0)

Speed.RPM
Fig. 20. Engine SFC Variations (Data from this laboratory, 1986). Turbojet Engine.
Test-5: with a fixed, two-dimensional, subsonic vectoring nozzle, with Aspect-Ratio = 26.6
Test-6: with a fixed, circular (axisymmetric), subsonic nozzle with the same exit are as in Test-5.

exhaust systems inherently have more internal surface area than conventional,
axisymmetric nozzles, and, thus, require more cooling air to maintain structural integ-
rity during AlB operation. The trend in relation to engines for advanced multi-mission
systems is towards a low bypass ratio and high specific thrust cycles. This tends to
increase exhaust gas temperatures and nozzle size, because of higher nozzle pressure
ratios, and reduces the amount of exhaust nozzle cooling air available, thus making the
nozzle cooling task more difficult. Moreover, increased emphasis on survivability is
tending towards higher nozzle cooling flow needs. These increased cooling flow
requirements have a direct impact on nozzle gross thrust during AlB operation,
because the cooling air cannot be mixed with fuel and burned; thus, resulting in a
lower AlB thrust. If the engine size is established at a max AlB design point (which is
not likely for the new ATF for instance), then the exhaust system (which requires more
cooling flow) can directly influence the engine sizing and weight. Also, from a system
standpoint, it may become necessary to compromise engine cycle selection to provide
adequate nozzle cooling air.
An ejector system ma~' provide a solution to the cooling flow requirements of
nonaxisymmetric exhaust nozzles and may result in an improved propulsion system.
An ejector nozzle, however, significantly complicates the propulsion system installa-
132 The Matrix of Unknown Variables

8r--------r--------.--------.--------~~

CD • Test-5 ~fI:l.
~7r---------~--------~----------+-------~~~~
-Y ~ Tast-6 ~.~ ....
~ 6r---------~--------~----------~~~----~~
a
L AA
If-
3:5t------f-----+---:-;!-:le--+-----+----.j
o ~f-
l.L
4t-----;----~r~-----~---~~
a"
CD .1tJ.II'
CD
a MI..
~3r-------~~~.~-~--+-------~-------+~
~

L ..a~~
--
a::
2r-~-~·~··~_,r---------;---------~----------~~
~1A1i ~

HUll 14001 18CXXl

Speed.RPM
Fig. 21. Engine Air-Mass-Flow Rate Variations. (Data from this laboratory, 1986.) Tests 5 and 6 are
defined in Fig. 20.

tion and requires careful evaluation of the entire inlet and nozzle installation. The
impact on the installed propulsion system design and performance includes system
net thrust, component weights, inlet, bypass ducts, nozzle-bay cooling, and life cycle
cost. The system ne~ thrust is impacted by changes in inlet ram and spillage drag, bleed
and bypass system drag, nozzle afterbody drag, and nozzle internal performance.

111.9 Engine SFC and Air-Mass-Flow-Rate

Properly designed thrust-vectoring nozzles may cause only minor variations in engine
performance (cf., e.g., Figs. 20 and 21). Hence, transforming an un vectoring engine to a
vectoring one, is, relatively, a low-cost, short-period program, involving conventional
materials, cooling methods, and production means.
 LECTURE IV

VECTORED AIRCRAFT
AS R&D TOOLS,
OR AS SUPER-AGILE, ROBOTIC FLYING
SYSTEMS

"The more we multiply means the less certain and general is

the use we are able to make of them."
John Dewey
"Knowledge is one. Its division into subjects is a concession
to human weakness."
Sir. H.J. Mackinder
IV-l Which of the Available Jet-Powered RPVs, Cruise Missiles and Robot
Aircraft May be Considered for Thrust-Vectoring Upgrading?
One of the most notable shifts in missile propulsion is currently from rocket, or rocket-ramjet
power, to turbine power. Turbine power is now advancing in a number of technology fronts:
RCC-Low-RCS (App. B); Counter-Rotating Propfans and propellers; advanced, yet extremely
simple, low-cost, low-maintenance turbojet and turbofan engines (with and without TV);
stealthy, high-range propulsion systems with very hig/l terminal maneuverability (e.g., the anti-
radar Tacit Rainbow missile).
Most of the RPVs available now are driven by piston engines. The minority are Jet-powered.
Jet-powered RPVs may be divided into "turbojet", "turbofan", or "rocket" types. Similarly,
Cruise missiles (CM) and anti-radar missiles, may be divided into "turbojet" or "turbofan"
types. Thrust vectoring may be considered an attractive option for the last two categories, espe-
cially when terminal supermaneuverability (Lecture II), low-cost, low-weight, flexible, stealthy,
high-range, long-storage-time characteristics are required.

"Turbojet" Missiles
The "Turbojet CM" category includes the antiship CMs USN-McDD Harpon (surface-to-
surface R/UGM/-S4AC, or the air-to-surface AGM-S4A/C), and the USAF-Boeing-SRAM-
AGM-69A. These CMs are propelled by the Teledyne J402-
CA-400 turbojet engine (a scaled-down, improved version of the Teledyne-J-69 family of
engines, or of the Turbomeca Marbore older engines).
A (Microturbo-TRI-60) turbojet engine is the powerplant for the British ASM (Sea Eagle-P3T)
antiship missile, a (Turbomeca-Marbore) turbojet engine for the Saab-OSA-SSM-RBOSA, 110-
mi-range missile, and a (Turbomeca-Arbizon) turbojet engine for the French TH-CSFI
OtoMelara-SSM-Otomat/Mk.2, I 540-mi-range, antiship missile.
Also included in this category are the Soviet AS-15 (Kent) air-launched, IS50-mi-range, CMs
for launch by Blackjack, Backfire, Bear H, or by newer missile-carrying aircraft. It also includes
the AS-I (Kennel), 63-mi-range, antiship winged-CM, and the AS-2 (Kipper) CM with a range of
100 miles.
Thrust vectoring may add significant maneuverability, reliability and survivability merits to
this category. It may also help to simplify the various launching methodologies, lower costs, sig-
natures, and weight, increase range and improve flexibility and long-storage-time specifications
(see below).

134
Lecture IV 135

"Turbofan" Missiles
This category includes the (strategic) nuclear ALCM AGM-86B USAF-Boeing/Honeywell/
Litton, (l500+)-mi-range, air-launched CM (from the B-52, B-1, and B-2). These missiles are
propelled by Williams International (WI) FI07-WI turbofan engines.
Thrust vectoring may add significant maneuverability, reliability and survivability merits to
this strategic category. It may also help to simplify the launching methodology, reduce signa-
tures, weight and cost (Fig. 1).. Here the TV nozzle may deflect only the core (hot) jet, or both the
cold and hot jets.

"Turbojet" RPVs and Anti-Radar Missiles

This category includes the following subcategories:
I) Surveillance/Target acquisition and "Multi-Purpose" (MP), such as the Boeing Brave 3000
(propelled by Noel Penny NPT 171 turbojet engine), the German/Canadian AN/USD-502(CL-
289)-Canadair/Dornier (powered by KHD-117 turbojet engine).
2) Target (T), such as the Fuji Heavy Industry BQM-34A (powered by Teledyne J-69-T-29
engine) and the J/AMQ-2 (powered by MHI TJIM-3 engine); the People's Republic of China
Changkong IC (powered by WP-6 turbojet engine), the Australian/UK. Jindivik Mk. 4A (pow-
ered by R-R Viper Mk. 201 turbojet engine), the Aerospatiale CT-20 (powered by Turbomeca
Marbore engine), and C-22 (powered by Microturbo TRI-60 engine); and the Italian/Meteor Air-
craft & Electronics/Mirach 100, 300 and 600 (powered by Microturbo engines).
The US Target-RPVs include: the Beech Aircraft MQM-170A, MQM-107BID, 999F, and
BQM-126A (powered by Teledyne J402 and Microturbo engines), the Northrop AGM-136-A,·
the Chukar-II-MQM-74C, and the air-launchable Chukar-III-BQM-74C (powered by WI-24-7),
the Teledyne Ryan BQM-34A/S and -34E/T (powered by Teledyne CAE J-69 family of engines)
and the larger RPVs QF-100 D/F and QF-5 of Tracor (powered by P&W J-57-P21 and
GE-J85-GE-13 engines).
New turbojet engines are now being developed by various laboratories to fill the gap of low
thrust levels required for small, long-range, reconnaissance/stealth/MP RPVs, or CMs. (E.g., the
Teledyne CAE 305 engine [Fig. 1] with thrust levels in the range of 45 to 90 Ib.t., a diameter of
only 6.5 in. and a very low dry weight.)

"Turbofan" RPVs
New turbofan engines are now being developed by various laboratories to fill the gap of low
thrust levels required for small, long-range, reconnaissance/stealth/MP RPVs, or CMs. The
available candidates in the upper-level of this subcategory are WI F-44 (1800 Ib.t.), the Garrett
F-109-GA-100 (1330 Ib.t.), and the WI-107 turbofan engine (600 Ib.t.). Other engines have been
proposed and/or developed in this relatively new category.··

• Jet power is mandatory for loitering and final maneuverability of such anti-radar missiles as the
USAF/Northrop Tacit Rainbow AGM-136A, etc. It provides low cost, low weight, flexibility and high
degrees of terminal manerability even without TV .
•• E.g.. propfans. turbine low-pressure stage and four-stroke rotary high-pressure section on co-axial
shafts. cold-profulsion. etc.
136 R&D Tools and Robotic Flying Systems

IV-l.l The Thrust Vectored Barak Missile

The rocket motor of the antiship TV Barak missile provides 3 thrust levels:
I - Low thrust for launch and turnover. (The missile is vertically launched and the
aerodynamic fins are assisted during turnover by external thrust-vector control vanes
in the rocket efflux. These are jettisoned after turnover, which occurs 0.6 seconds
after launch.)
II - High acceleration boost.
III - Sustained thrust for short-range, high-maneuverability engagement.
The TV Barak missile has been developed jointly by Rafael-Israel's Armament
Development Authority, and IAI (Israel Aircraft Industries). It is to become opera-
tional in the early 90's.

IV-1.2 Expanded Missions for Vectored RPVs and Robot Aircraft

Much has been written on the mission philosophies and on the design of various
propeller-powered RPVs and on (un vectored) jet-powered, RPVs and cruise missiles. We
do not intend to reproduce this material here, except briefly, when a comparison is
required between those vectored and those unvectored.
To start with we note that (jet-powered) vectored RPVs may be subdivided into two
types, namely:
IV-I.3 Advanced R&D Tools
The emergence of vectored RPVs as R&D tools replaces some of the more traditional
roles of wind-tunnel testing techniques. This cost-effective methodology is capable of
reliable simulation of manned vectored aircraft performance under realistic condi-
tions involving, say, PSM or post-stall maneuvers, STOL and V/STOL. However, as in
wind-tunnel testing, it encounters scale-up and scale-down problems. (One may first
notice that materials and servos do not scale-up easily. RCS/IR signatures and
aerodynamic/propulsion systems present special scale-up/scale-down problems. A
few of these have already been discussed in 11-3, and in Lecture III. A few additional
ones will be reassessed below and in paragraphs IV-12 and V-l.l.)

IV-1.4 Unsurmountable VRA-Missions.

This class may be generally defined as Vectored Robot Aircraft (VRA): the term
includes, in principle, cruise missiles, drones, decoys, TV-missiles, flying vectorable
bombs, long-loitering target-designators, surveillance RPVs, etc.
However, the unique missions of VRA are not yet clear, nor well-defined. Some of
these are to be discussed below, while others are postponed to Volume II.

Assertion I.
VRA-technology opens the way to new, expanded missions with unsurmountable
performance in agility and launching or V/STOL capabilities (at low R&D, produc-
tion, and storage-maintenance costs).
Lecture IV 137

Assertion 2.
VRA-technology is one of the critical technologies required for high-performance,
stealth, [cruise and/or terminal supermaneuverability (cf. Lecture II)] robot
aircraft.
Assertion 3.
While the defence-role potential of VRA-technology has not yet been fully
exploited, it is already generating a revolution in the design philosophy ofsome unique,
air, land and marine missions.

Assertion 4.
The defence role potentials of vectored aircraft, in general, and of VRA-technology
in particular, depend on a new branch of research and development involving new vec-
tored components, new powerplants, new PST inlets, novel unit operations, and, as yet,
unknown, integrated flight/propulsion control theory.

* * *
In trying to define the practical meaning of the aforementioned assertions, one may
first examine the present state-of-the-art of un vectored, jet-powered, decoys, drones,
RPVs, and advanced cruise missiles (ACM), and compare it with that ofVRA. For this
purpose one may first concentrate on conventional, jet-powered-drones, and, then,
extend the comparison to other, less-known, applications.
Some of the more sophisticated missions of target RPVs, decoys, or drones, is in sim-
ulating subsonic or supersonic "stealthy" or "unstealthy" aircraft, operating under a
variety of different EW conditions, (using such devices as Lunaberg lenses, corner
reflectors, IR and shock-wave detectors, etc.). Indeed, in the general air-defence train-
ing role, the decoy-drone is sub-scale to the threat that it represents, and, consequently,
must impart performance and detection characteristics of the threat to be simulated.
Thus, if the threat is an advanced, vectored, cruise missile (A VCM), or a fighter, the
simulator must also be vectored.
Yet, the most important uses of VRA are to be found in the new roles that reflect
their own inherent advantages. These include:

1) Helicopter Killers (high closing speeds with final PSM-PST/RaNPAS vectoring

agility) and Anti-Radar Missiles (ARM).
2) AVCM and RPV Killers (high closing speeds with final, PSM-PST/RaNPAS vec-
tored agility).
3) Ship Protectors (V/STOL or VTOL with vectoring agility and very low speeds).
4) Unmanned Rescue (V/STOL or VTOL, stealth).
5) Multirole (land) (V/STOL or VTOL with vectoring agility).
6) Multirole (Air) (V/STOL or VTOL with vectoring agility and stealth).
7) Penetrators (low signatures).
8)Advanced Vectored Drones (AVD) and decoys.
138 R&D Tools and Robotic Flying Systems

9) Advanced Vectored Cruise Missiles (A VCM).

10) Vectored RPVs armed with hypervelocity all-aspect missiles.

Further classifications apply according to size and cost:

IV-2 Mini RPVs, mini AVCM, or mini-VRA.

Such vehicles are increasingly built now with stealth or enhanced-signatures configu-
rations and characteristics, IFPC (including autostabilization systems), sophisticated
avionics and weapon/surveillance subsystems. However, because of their small size,
and relatively low cost to develop and fly, these vehicles are often erroneously associated
with model aircraft.
A subdivision of this group inclues:

IV-2.t Hot (Jet-Powered) VRA.

The penalties associated with this technology are:
I) Jet-engine flame-out due to inlet distortion at PSM or during post-stall-maneuvers.
This limitation results in highly-limited agility, performance and mission charac-
teristics, unless special, PST-tailored inlets are employed.
2) Use of high-temprature jets and materials increases cost, weight, IR signatures,
cooling requirements, and complexity. It also reduces availability and reliability.

IV-2.2 Cold (Jet-Powered) VRA.

The payoffs associated with this technology are:
1) Supermaneuverability capability (cf. Lecture II) due to thrust vectoring uninhibited
by engine-intake distortion at high vehicle incidence (high-a-/3, PSM and
PST-maneuvers).
2) High T/W ratios afford the development of a bona fide VTOL, or V/STOL
capability.
3) Low IR signatures.
4) Low-weight materials increase the effective-payload capacity.
5) Low Specific Fuel Consumption.
6) Damage to launch-area/space equipment is minimized.
7) TBO, TBF, endurance, controllability and safety are maximized, while life cycle
costs and storage expenses are substantially decreased.
These systems may be based on newly investigated, combined-cycles powerplants
involving multiple-stage axial or centrifugal compressors (powered by, say, Wankel, or
by smaller types of beyond - 20,000 -RPM, 5 to 150 Hppiston engines). The compres-
sor supplies subsonic, or supersonic cold propulsion for un vectored, or vectored
vehicles.
[For small turbojet engines one may use here the new Teledyne, 6-inch-diameter,
Lecture IV 139

305 engine, or the SCAT, or the WR-24-7, 8, KHD 3171117, NPT, or Microturbo
series.]

IV-3 Midi-RPVs, Midi-ACM and Midi-VRA.

This size-cost category is limited, still, to hot-jet propulsion systems. Otherwise, it is
similar to the mini-class in terms of performance/missions. Jet engines for this cate-
gory are now available from Teledyne or WI.

IV-4 Maxi-RPVs and VRA.

This category includesfighter-size drones andj/ying bombs. The large drones are often
based on old fighters converted into target RPVs which employ various
autostabilization, EW gear and other specific systems.

MODEL 305 FAMILY

305-7E

Fig. IV.I. One of Teledyne's proposals to power small robot aircraft.

IV-5 Cost-Effectiveness of flying Vectored RPVs vs. Limited Wind-Tunnel

Testing
Our integrated laboratory tests and flight-testing efforts with pure vectored RPVs are
aimed at reducing the risk decision regarding propulsionlj/ight designs.
A near-term goal associated with these integrated studies is to demonstrate the lower
costs, time-condensed milestones, and high reliability-efficiency of this methodology,
namely, of designing, constructing, lab-testing and j/ight-testing pure vectored RPVs,
rather than relying only on the more expensive, scale-up methods associated with
vectored-models-suction-ejection in wind-tunnel model testing.
Thus, we proceed with the following integrated test stages:
STAGE 1:
Although we start the design of PV A with a wind-tunnel model testing, we refrain
140 R&D Tools and Robotic Flying Systems

Fig. IV-2. Control Rooms No.3 and S at the Jet Propulsion Laboratory (TIlT) help to substantiate the
proposed laboratory/flight-testing methodology (d. Figs. 6 and 7). The lower one controls Components
Research Test Room, while the upper one regulates the Engine Altitude-Speed Test Rig and R&D Facil-
ities shown in Figs. 6 and 7.

from testing these small models with internal suction-injection systems (so as to try
to simulate actual thrust-vectored conditions in flight).
STAGE 2:
New ideas are first tested on a component subscale test rig (Fig. 7). (Up to about
I kg/sec engine air-mass flow-rate-testing, and up to I ,OOO°C.) This small-size, com-
ponent category simulates supersonic and subsonic thrust-vectoring nozzles. The
results of these tests are employed directly, or are scaled-up for the full-scale testing
of STAGE 3.
Lecture IV 141

STAGE 3:
For up to 7 kg/sec air-mass-flow-rate-testing, we use the laboratory's altitude-speed,
full-scale, engine research system (cf. Fig. 6 and 7). This size category simulates vari-
ous fighter engines, provided certain guidelines and precautions are employed in the
scaling-up process, e.g., successfully passing first the component-test category
("STAGE 2"). The results of these tests are also useful for the next stage.
STAGE 4:
Laboratory and flight-testing of radio-controlled, pure vectored RPV s in the mini-
size category of 7x4 ft to 9x4 ft. The propulsion systems of these TV-RPVs are
scaled-down versions of optimized systems developed first through STAGE 3.
These flight tests are employed to get very high AoA and extremely sharp turns such
as those shown schematically in Fig. 11-2. The tests include short takeoffs and land-
ings and engine-out landing tests. Employing these results one may return to
STAGE 1 and proceed again to STAGE 3 (sometimes without passing again
through STAGE 2) [ef. the diagram in the Preface].

The experience gained by this laboratory through STAGES I to 4, has given verifia-
ble proofs to a new, cost-effective methodology in the design, testing, development
and flying of vectored propulsion/flight systems. So far the cost and time, in comparison
with the more traditional wind-tunnel methods, appear to be an order-of magn itude lower
(depending on the design, size and relevance of the laboratory and flight-testing data
required).

IV-6 Preliminary Phases of Vectored RPV Design.

In current design methodology the integration process is often limited to steps involv-
ing simple combinations of well-proven components with new ones; e.g.;
• An existing engine with a new nozzle.
• An existing engine with a new inlet.
• A new nozzle with an existing wing structure.
• A new inlet with existing, aerodynamic surfaces, and flight-control systems.
The results of such processes may only lead to partially-vectored aircraft, which are
characterized by less significant gains in performance and survivability.
Alternatively, one may first study the following:

• Expected technology limits of radars, IR and optical sensors, etc.

• Expected technology limits ofpowerplants, including PST inlets and thrust vector-
ing nozzles (cf Appendices B and F) ..
• Expected technology limits of supermaneuverability and PST-controllability.
• Expected threats.
• Expected missions.
• Financial limitations.
142 R&D Tools and Robotic Flying Systems

Fig. IV-3. Early windtunnel models (see also Fig. 11, Introduction). The one on the left was flight tested
(see Fig. IV-5). See also Fig. IV-4.

Then, as a first iteration, one may examine the methodology presented schemati-
cally in the Preface, and decide on a program which involves the relevant variables.
Next, one may gradually reduce the number of variables to the minimum required to
end up with ajirst prototype of a proofofconcept. vectored/stealth RPV. Thus, shape,
ReS. JR, materials, NAR, supercirculation, limits of yaw-pitch-reversing vectoring,
thrust-to-weight ratio, maximum engine intake distortion, cold or hot propulsion
powerplant, supermaneuverability, takeoff and landing factors may first be integrated
to adapt to a tentative mission. Next, a few tentative shapes may be constructed around
the best propulsion system available (including its optimal, laboratory tested, inlet-
core-engine-vectoring nozzle configuration). Trimmingfor low signatures. and proper
wing loads. payload. performance, weight, range, etc., one may jix the wing aspect ratio.
and the wing total surface area, sweep and range of incidence variations, etc.
A range of (unvectored) wind tunnel models may then be built and tested to evaluate
some preliminary variables, with or without canards, vertical stabilizers, tail, etc. The
data may then be fed to a highly simplified computer simulation program.
By evaluating the location of the aerodynamic center (and the center-of-pressure
variations), one may use simulation programs to decide on static stability limits, etc.,
for thejirstflying vectored RPV to be tested. A 5% positive static stability may be a good
starting guess in case the simulation is not sufficiently reliable.
A mock-up model may then be built in the laboratory, and all actual servos, batteries,
fuel vessels, inlet, engine, nozzle, telemetry-metry/flight computer installed in it, till the
required static stability limit is met. The mock-up, or the flyable prototype should be
constructed according to one-to-one scale drawings and in the following order:
I) Install (STAGE-3-tested) engine intake(s), and the selected (STAGE-3-tested)
Lecture IV 143

exhaust vectoring ducts and nozzles, with proper flaps for engine-out emergency
controls and internal, streamlined flow dividers/structural struts (cf. Fig. 1 in the
Introducton) in such a design that the inlet/engine/nozzle becomes the main carry-
through structure for control loads!
2) Following the low-signature, external skin lines, build the aircraft structure first
around the (laboratory-tested) 2-D nozzles shape, following each nozzle top and
side cross sections and integrating the wing structure with the exhaust system
internal structural struts.
3) Repeat stage 2 for the engine intake. This should be a PST intake for hot propul-
sion, or a simple intake for cold propulsion.
4) Introduce retractable landing/takeoff gear, engine auxiliary equipment/
instrumentation/control servos, fuel tanks, fuel and servo lines, on-board com-
puter, etc.
5) Construct and install the canard (if required) and its control servos.
6) Construct and install vertical stabilizer(s) (if required), and integrate them with
the nozzle-wing structure.
7) Install batteries, telemetry-metry data-acquisition/FBW computer, servos, ampli-
fiers, sensors, analog-to-digital converters, and other accessories, as required for
each particular prototype.
8) Hang the vehicle, without completing its external skin, and find its center of
gravity.
9) Trim the center of gravity location to meet predetermined static stability margin,
moments of inertia, etc., usually by moving the batteries and the com-
puters (fuel vessels must stay centered around the vehicle center of gravity).
10) Balance the vehicle around its lines of symmetry.
11) Complete vehicle's skin covering job.
12) Operate the engine(s) and evaluate static takeoff thrust, with and without vector-
ing, using the remote or FBW control commands.
13) Trim propulsion system and servos for remote, radio-controlled or FBW
operation.
14) Measure the empty weight of the prototype and calculate thrust-to-weight ratio
with fuel vessels full and empty, depending on the mission/research project aims.
15) Evaluate T/W ratios for various pay-loads.
16) Use the computer simulation program to estimate runway distance till rotation, at
various a and D,. settings and timing.
17) Final trimming of wheels height/propulsion-system control/Dr and D" etc.
18) Calibrate the on-board, data-extraction computer and freeze its software, mode of
operation and number and type of variables to be recorded during the flight tests.
IV-6.1 Flight Tests
One may now proceed to runway testing and trimming. and eventually to a well-
designed flight program. with proper monitoring ofperformance. including data acqui-
sition. At a given stage of this program one may also evaluate IR and ReS signatures,
144 R&D Tools and Robotic Flying Systems

II
I. I .
• I

-+I

--~~==~~~2F=~~~~~===~=---
Fig. IV -4. One of the early prototypes tested in our subsonic wind tunnel during STAGE I of our inte-
grated vectored propulsion/flight-testing program (cf. also Figs. II-la to II-lh, IV-3, IV-5).

and, accordingly, make variations in shape and materials of the vehicle, and in its pro-
pulsion system.
To obtain PST maximum-agility comparison-maneuvers the propulsion system
should first employ inlet-distortion-free. cold-jet vectoring, and, therefore, low-
temperature materials. Then, gradually, as the program proceeds into higher flight
speeds, a proper switch to hot-jet vectoring may be required. In turn, this requires new
materials and PST inlets to be integrated with the propulsion system, and repeating
the whole process from the very beginning at a much higher cost, lower program pace,
and much higher safety risks.
A cost-effective program should also stress the reliability of inflight data acquisition.
including reliable video-computer recording. In any case, low-speed maneuverability
and STOLlVTOL characteristics should be evaluated first.
Here the simplest and lowest-cost program uses only advanced PC computer
"cards" reconstructed in a new mode, which readily stores in its advanced PC RAM all
flight data during a/ew critical minutes of, say. flying in well-defined horizontal and ver-
tical sharp turns/loops involving PST conditions.
Then, following landing, the flight-borne computer is connected for data feeding to
another, ground-based computer, near the runway. Using suitable preprograms, the
ground computer may plot the results, compare them with previous flight data, and
with flight objectives, and, finally, may demonstrate the best conditions for the next
flight to be followed. The flight crew may then discuss the results to determine the next
flight procedures.
* * *
Lecture IV 145

The flight test data may once again be examined in the laboratory for incorporation
of some possible modifications in the basic design. For instance,following the analysis
ofrollingflight data ofa given prototype during high angles-ofattack maneuvers. a deci-
sion may be made to improve roll rate by increasing. say. the rolling moment arm Y (cf.
Fig. la in Lecture II), and, accordingly, to change the geometry of the exhaust nozzle.
and. consequently. also that of the entire shape ofthe vehicle. This may lead to the con-
struction of modified wind-tunnel models. The entire process may then be repeated,
ending up with a super-agile, vectored-stealth RPV.

* * *
Low-speed agility ofvectored-stealth RPVs must then be compared with that ofa simi-
lar. un vectored RPV, both having the same thrust-to-weight ratio. moments of inertia.
shape. size. etc. The incredible sharp turns obtained with the former are not usually the
objective, but only the baseline for proper comparisons with unvectored baseline per-
formance and agility metrics.
* * *
Scale-up problems are the next important problems to be treated. For instance, dur-
ing flight studies of engine-intake distortion (of a model RPV of the new STOL F-15),
an extensive use is made of various scale-up analogies involving similar Reynolds num-
bers in both model and actual systems. However, many design parameters do not scale-
up, e.g., materials and servos.

eIV-7 Synergetic Studies As Mid-Course Adjustments

Figure I in the Preface describes the feedback mechanism employed in this laboratory
during the stages described previously.
Some of the early conceptual design options are related to the iteration process
involving the emergence of "figures of merit " and configurations, and their testing in
wind tunnels, the jet propulsion laboratory, and in actual flight.
Some of the early configuratons tested in our subsonic wind-tunnel during STAGE I
of our vectored propulsion/aircraft program are shown, without canards, in Figs. 4
and 5. These are subsonic. single-engine prototypes.
However, whenever the entire vectored propulsion system is to be integrated with
such configurations (including, say, vectored nozzles and inlets as shown in Fig. II-I f),
a new family of "figures of merit" and configurations must be developed and tested.
Probably the best method to start such a testing is as follows:

PHASE-l

a) Determine the general aim and scope of the entire vectored aircraft program (i.e.,
R&D, or a given RPV-RA mission, or RPV simulation of available aircraft to be
146 R&D Tools and Robotic Flying Systems

Fig.IV-S. GAL-3; one ofthe early pure vectored RPV designed by this laboratory (1987). Cf. Fig. II-I.
Its first flights were conducted by (from left-to-right,) Erez Friedman, Mike Turgemann and Tzahi
Cohen, in Megiddo Airfield during I ge7, using radio controls. It is now equipped with on-board com-
puters. Its wind-tunnel model (Fig. IV-3) is shown on top. The (cold)jets emerge from the highAR (=46)
yaw-pitch-roll thrust vectoring nozzles in the back trailing edge (marked with strips). No rudders, ailer-
ons, flaps, etc. have been used. The vertical stabilizers will be removed during future flight tests with
IFPC. Using such RPVs, PSM/PST/RaNPAS maneuvers are being tested since 1989.

transformed into partially-vectored aircraft, [e.g., the F-I5 transformed into an

F-I5 STOL fighter)).
b) Determine the overall level of observability and reflectivity required (using different
nozzles, inlets, shapes, materials, etc.).
c) Determine the overall level of high-a-{3, STOL, VTOL, or V/STOL requirements.
d) Determine priorities, e.g., synergetic studies of high a, /3-inlets, vectored nozzles,
supermaneuverability, low observables, STOL, V/STOL, control links, or
PST-IFPC.
e) Use the material of this design course to evaluate:
e-I: Simplified computer simulation programs for takeoff and landing for various val-
ues of wing loads, T/W, L/W, -D/W, S, Cv Cm, CD' Ov, 0Y' On stability data, wing
shape, W, am, Y, D, TR, Cfg , CDS, CA , C., ground effects, inlet distortion limits,
emergency sizing of Cf(Fig. III-I), etc. (Cf. Lecture III for the matrix of unknown
variables involved).
e-2: Computer simulation programs for observables. (The physico-thermodynamic
fundamentals associated with this subject are to be treated in Volume II.)
Lecture IV 147

Side view
2 3

From com.2t'~!.sor

II i 7 100. 01
comprt'ssiNf
air ill \60alm
I;

.Fig. 6. The proposed integrated methodology oflaboratory/flight tests is based on "STAGE 3", i.e., on
the altitude-speed, full-scale, engine R&D system of this laboratory (cf. Figs. 2 and 7). I) Engine inlet
section, 2)Engine section, 3) Vectoring nozzle section, 4) Exhaust section (includes fans for S.L. tests),
5) Ejector system, 6) Gas turbine for low-altitude simulation, 7) Fuel system, 8) Large heat exchanger
for heating or cooling inlet air during altitude simulations, 9) Control Room No.5. Note: Engine inlet
section can be modified to fit R&D needs of various PST/Stealth engine inlets operating at various a,
{3, Ii, iJ conditions. Thrust vectoring nozzles tested are those scaled-up from component tests (cf.
Fig. 7).

e-3: Early candidate 'figures afmerit'to be tested in the wind tunnels, but without vec-
toring means, canards, gears, etc.
An example of such a preliminary family of wind-tunnel models is shown in
Figs. 4 and 11 in the Introduction.
f) Use the preliminary wind tunnel test results to verify and re-evaluate the computer
programs and to define preliminary PST-IFPC limits for various missions and con-
trol methods.

PHASE-2
a) Determine Statement of Work (SOW) and Program milestones, using the initial
data of Phase-I.
b) Determine limitations and proper interactions between computer simulation, wind-
tunnel tests, jet-propulsion tests, RCS-IR tests, telemetry-metry methods of flight-
control tests, flight tests, analysis, and conclusions.
c) Determine the flight-testing program methodology, including the landing proce-
148 R&D Tools and Robotic Flying Systems

dures, telemetry-metry feeding of flight data into a portable computer near the run-
way, video-recording/computer analysis of results, etc.
d) Re-examine the selected type of(radio, cable, fiber, or IR-Iaser beam)jlight control
system, safeguards, emergency procedures, number of channels, number of flyers,
altitude, maximum speed, airfield, etc.
e) Re-examine the materials, landing gear and propulsion method employed.
o Re-examine scale-up and scale-down design limitations.
g) Choose the type of core engine required (cold-jet, or hot-jet propulsion) (cf.
Figs. 1-1 i, 10.
h) Choose engine sizes and thrust levels required for laboratory testing of the entire
inlet-engine-nozzle propulsion system, or of its individual components, using
proper scale-up/scale-down techniques.

PHASE-3
1) Fix the overall size and shape ofthe vectored-stealth RPV, including its T IW, LIW,
C.G. - A.C. criteria, etc.
2) Determine the type, AR and shape ofthe PST-inlet to be tested first in the lab under
realistic operating conditions (i.e.,jlush, top or bottom mounted. mode ofvariation.
mode of control, etc.).
3) Repeat sub-stage 2 so as to fit the inlet AR, type of propulsion and shape ofRPV
and its size and space, and method of PST-IFPC.
4) Determine the degree ofsupercirculation and RCSIIR signatures. and. accordingly
fix the AR and shape of the vectoring nozzle. (CF. Fig. III-tO).
5) Repeat sub-stage 4 till exhaust nozzle fits with the configuration, and, simultane-
ously, determine the type of vectoring, i.e.:

• pitch vectoring only,

• pitch and yaw vectoring.
• pitch, yaw and TR vectoring.
• Simultaneous yaw-pitch-roll vectoring.
6) Design and construct a number of full-scale, laboratory models of inlets and
exhaust nozzles.
7) Perform full-scale laboratory tests of inlets and nozzles (cf., e.g., Figs. 6, 7. Note,
however, that this sub-stage may take a number of years).
E.g., install a bell-mouth inlet on a conventional type engine equipped with
rounded, un vectored nozzle and calibrate the propulsion system using full-scale,
laboratory test rigs, including the evaluation of Clk , CDS, CA , Cv, T, SFC, pressures,
temperatures, air-mass-flow rates, etc. (cf. Figs. 1-18 and 1-19 and Appendix A).
8) Replace the bell-mouth inlet by the required new inlet, and repeat the tests to verify
inlet-engine compatibility, low-level of distortion, and the expected limits of PST
maneuverability (cf. the Introduction and Appendix F).
9) Replace the rounded exhaust nozzle with the required vectored nozzle and repeat
Lecture IV 149

11

7
.Fig.7. The proposed integrated methodology of laboratory/flight tests is based on "STAGE-2"
(§ IV -5), i.e., on the component, subscale system for testing new ideas. I) Exhaust system. 2) 2D-CD,
simultanous yaw-pitch-roll thrust vectoring nozzle with high aspect ratio undergoing component tests
by a highly-instrumented, test rig before scaling-up to the "full-scale" tests (Fig. 6). 3) Circular-to-
rectangular duct. 4)No-swirl/uniform flow duct. 5) Combustor. 6) Fuel system controlled by the instru-
mentation of Control Room NO.3. 7) Air-mass-flow-rate monitoring. 8) Flow control valve. 9) Gas
turbine suplying up to I kg/sec air at up to 2.8 atm(g). 10) A connecting, flexible duct. (Cf, § IV-6,
IV-7).

the tests. (Note, the new nozzle should have the same As, A9 as those of the circular
conventional nozzle tested in 8).
10) Compare the test results, and, following additional iterations,jreeze the propulsion
system design. (Note: Before freezing the design one should take into account the
main design criteria enumerated in the text and in the appendices.)
11) Feed thefinal design data into the computer simulation program/or sizing routines
(e.g., the overall length of the inlet + engine + C-R duct + exhaust nozzle fixes the
minimal overall length o/the aircraft, or in scale-down calculations, o/the vectored-
stealth RPV to be ./light tested).
12) Scale-down the optimized propulsion system to fit the required size of the vectored
RPV and, accordingly, construct inlet(s) and exhaust nozzle(s).
13) Test the scaled-down propulsion system in the laboratory, including trim and
modifications. This includes two sub-stages, namely, testing the propulsion sys-
tem and testing the propulsion system installed in the vectored RPV, as described
below.
Thus, without changing engine-inlet-nozzle geometry, build the aft shape 0/ the
"RPV" around the exhaust vectoring nozzle, using maximum structural integration
ribs and supercirculation benefits. (Note: If impossible, modify the wind-tunnel
model and repeat the entire process.)
14) Install the vectored-nozzle servos, or actuators, and test them with the radio, cable,
150 R&D Tools and Robotic Flying Systems

fiber or IR-lazer-beam control system as in actual takeoff, flight and landing condi-
tions, with and without engine power.
15) Install canards, or ENVJ, and variable inlet servos or actuators. (Note: A
subprogram for the evaluation of proper canard or ENVJ systems should also be
conducted during the preliminary design phases).
16) Install retractable (or non-retractable) landing gear in relation to e.G., and, takeoff
and landing conditions, as chosen from the computer simulation program.
17) Complete the forebody and install the canards or ENVJ, while leaving the forebody
upper skin open for further work.
18) Install batteries, flight and data acquisition computer, and the payload,for an initial
e. G. evaluation.
19) Move the batteries, servos, computer and payloads, till the required stability margin,
the required moments ofinertia (for similarity comparisons with the full-scale air-
craft), and other criteria, are met.
20) Install fuel vessels in symmetry around the e. G. of the vehicle.
2l) Repeat e. G. tests with and without fuel.
22) Test axial symmetry balance of moments, and trim the vehicle accordingly.
23) Complete and close the upper skin of the vehicle while installing all inputs-outputs
connections (fuel, control, computer telemetry-metry, starters, etc.)
24) Start the engine and perform laboratory tests (including measurements of installed
thrust at various throttle power conditions, thrust-vectoring controls, canard-flap
controls, computer input-output commands and data extraction modes of opera-
tions, etc.).
25) Perform initial runway tests, including testing of the data-acquisition computer,
feeding into another computer, as well as the controllability of vehicle during
turning/taxiing, etc.
26) Freeze the first flight conditions and limits, including emergency procedures, and
data extraction metrics.
27) Place high-quality video camera(s), and a high-performance, portable computer
near the runway for recording ofturning radii, velocities, a, {3, high-a-{3 maneuvers,
takeoff distance, landing distance, canard/vectored flaps angles, etc., during or at
the end of each flight (i.e., telemetry or metry systems).

PHASE-4
1) Feed takeoff, flight and landing data into the stand-by computer and examine the
relevant video-recorders links and coordination procedures.
2) Use the portable computer simulation programs and the flight data to freeze the
next-flight procedures and conditions.
3) Repeat flights. Alternatively, one may perform mirror-like, sustained horizontal
turns, of both vectored and unvectored RPV s, simultaneously.
4) Evaluate and compare the performance-agility metrics (Figs. 19 and 20, Introduc-
tion) and rethink and reassess the entire program SOW/milestones.
Lecture IV 151

5) Design modified configurations and 'figures of merit' in light of the flight testing
results and performance metrics conclusions.
6) Design a modified vectored propulsion system as before.
7) Integrate the new design, as before, and reevaluate the control and data acquisition
modes.
8) Perform wind-tunnel testing with the new models.
9) Perform Jet-propulsion testing with the modified components.
10) Perform Jet-propulsion testing with the entire propulsion system.
11) Repeat Phases 3 and 4, starting from § 12 of Phase 3.

IV-8 Master Plan For Vectored RPVs and For Vectored Fighters
Funding for vectored aircraft may depend on individual initiations of new ideas and
also on a master plan. Indeed, to consolidate the multiple military programs, one must
first eliminate redundancy, obsolete designs and ineffective programs.
A number of existing programs would be curtailed and requests for proposals for a
family of new air vehicles must be issued. In the specific domain ofRPVs this includes
three missions: close-air operations; short-range, shipboard operations; and deep
operations. Other subdivisions of new missions have been enumerated previously in
this chapter.

IV-8.1 New Programs

NATO nations have recently begun sea-trial demonstrations to assess the capability of
RPVs operating from small naval vessels.
A few RPV builders have already agreed to demonstrate their systems on board
NATO vessels ranging from 4,000-ton frigates to 1,500-ton corvettes and 500-ton fast
patrol boats. Under the proposals, the companies have agreed to finance the demon-
strations, with the navies of the participating countries providing the ships and range
time. However, vectored RPVs have not yet been included in any of these programs,
except the Barak missile program (IV-1.1 ).
The DoD is interested in an RPV for close-in operations for which the prime candi-
date is a VTOL system, or a system capable of operating from a small ship.
The US Air Force will test RPVs to evaluate the use of unmanned vehicles to assess
runway damage.
Requests for proposals for a baseline family of RPVs would be issued in early Fiscal
1990 under the US plan, with testing and evaluation scheduled to begin in Fiscal 1992.
Under the plan, at least three airframes will be developed, one for each of the three
mission areas. Some will have unique subsystems, such as a signals intelligence pack-
age for deep operations, and others that will be common, such as a television or a
forward-looking infrared system. Another requirement could be development of two
data links that would be common to all the RPVs, as well as a single mission planning
and control system.
152 R&D Tools and Robotic Flying Systems

The US program would match off-the-shelf systems with the requirements to come
up with the definition of the most cost-effective system.

IV-9 STOL RPVs

Takeoffand landing distances are drastically reduced in vectored RPVs, depending on
the thrust-to-weight ratio of the RA; and on the positions and angles of the front
"downward-pointing vectored jet", as well as on its distance from the aircraft center of
gravity.
Automatic, or semiautomatic mechanisms, may then be designed and employed for
a gradual transition from V/STOL, or semi-hovering mode of operation, into regular
flight at maximum speed, and vice versa. Such vehicles have been recently tested by this
laboratory, using 3 engines (Prototype No.4), and a single engine (Prototype No.5).

IV-tO Fan and Core Flow Vectoring

In using turbofan engines to propel, maneuver and control vectored aircraft, the hot
and cold streams may be separately diverted into hot and cold nozzles, each with its
material and operational characteristics (cf. Fig. A-2).
Alternatively, the two streams may be mixed, as in conventional engines. However,
the elimination of cooling needs, and the possibility to introduce low-weight, low-
temperature materials in the "separated flows" design, appears more attractive.
Indeed, the evaluation ofNAR, (LlCdmax> T, 0., 0Y' Cjg , etc.Jor each stream would then
be different. Theoretically, (LlCdmax may first be estimated by computer simulations
programs. However, the errors involved in using this methodology should not be
underestimated. Empirical, flight-testing studies appear, therefore, to be the most reli-
able methodology, at least for the near term.

IV-ll Nozzle Length and Shape

Decreasing the 2D nozzle length reduces weight and increases overall T /W ratio. How-
ever, problems involving BLS, swirl, and non-uniformity may become inhibitive.
Cold fan air is preferably ducted to the outboard section of the wing, thereby involving
different shapes & BLS/mixing problems than those associated with hot nozzles. NPR
is also different for each vectored nozzle (cf. Fig. 2, Appendix A).

IV-t2 Scaling-Up and Scaling-Down of Aircraft Dynamic Characteristics

PST/StealthlVectored aircraft encounter unfamiliar dynamic characteristics (cf., e.g.,
Figs. 2 and 3 in Appendix D). Traditionally, the general equations associated with
maneuvering about an aircraft's axes are given by:
Lecture IV 153

Iyy ' q = liz • P . V 2SC· CM for the pitching moment

In' p' + Ixz';- = 0· p • v2 • S· b . Ca, for the rolling moment
Irz' P + I zz ';- = liz· P • V2 • S· b . CN, for the yawing moment

where I is the moment of inertia and

p,cj,;- represent the angular accelerations about the x, y and z axes, respectively;
CM' Ca , CN represent moment coefficients for pitch, roll and yaw, respectively; and
S,b,c are the wing area, span and average chord, respectively.
The latter two equations take into account the coupling between roll and yaw. How-
ever, none of the above-equations takes into account the gyroscopic effects presented
by the rotation of the engines during aircraft maneuvering.
When scaling a model for similar dynamic characteristics one is first faced with two
requirements:
I: To maintain a similar weight distribution.
II: To maintain the same relations between moment coefficients and moments of
inertia, e.g., for the pitching moment one would have:
I yy ' cj = liz • pv2 • S· C • CM,
or,

To proceed, one may add frequency scaling-down functions, and also read Ref. 179,
and the next paragraph.

eIV-13 On the Limitations of Wind-Tunnel Tests

Can reliable wind-tunnel tests of vectoring aircraft be conducted? Indeed, reliable
simulation of, say, PST-inlet suction and distortion, combined with 2D-nozzle pres-
sure ratio/integration/yaw-pitch-reversinglvectoring simulations of actual, dynamic,
in-flight conditions, encounter inhibitive technical and cost-effective problems, some of
which will be described below.
It is here that the designer offuture vectored aircraft may prefer to switch to actual
flight-testing of vectored RPVs to obtain optimized, actual, cost-effective test data for
scale-up design considerations. Consequently, the flight testing of vectored RPVs, as
described above, emerges as a highly cost-effective R&D tool.
Nevertheless, wind-tunnel testing remains an essential R&D tool during initial
design iterations of both research RPV s and manned vectored aircraft. Thus, wind-
tunnel tests, even without suitable suction/ejection flow simulations of PST-inlets and
PST-vectored propulsion systems, may give the designer of vectored aircraft an initial
estimation of Cv CD' CM' AC, Strouhal No., Reynolds No., Nusselt No., etc.
154 R&D Tools and Robotic Flying Systems

Canard and other control surfaces can be evaluated in such tests to trim nozzle
vectoring/reversing pitching moments. Indeed, various cold flow, performance tests
of 2-D nozzle models have been conducted in the past decade or so in wind tunnels,
and are reported in the open literature [1,4,6, 8, 13, 14, 17, 32, 34, 40, 45, 62, 67,69].
Such tests results have indicated that:
• At transonic and supersonic speeds, the 2D-CD nozzle exhibits higher thrust-
minus-afterbody-drag performance than the conventional, axisymmetric nozzle,
while at low subsonic speeds, the nozzle incurred up to 2 percent thrust-minus-
afterbody-drag penalty (cf. Figs. 1-17, 18).
• Thrust vectoring produced significant induced lift increases (cf. Fig. III-19).
• Thrust reversing is effective for static and in-flight operation involving high-a-p
maneuvers. However, PST-deceleration is much more effective than TR (Internal
Reports of this laboratory).
• At 0,. .:5 20· thrust vectoring, the thrust loss is as expected from cos (a + ov), depend-
ing on the nozzle design, i.e., on the difference between actual and geometric deflec-
tions during thrust vectoring.
• Other variables which can be evaluated by wind-tunnel tests include NAR, nozzle
type, nozzle pressure ratio and area ratio, horizontal tail deflection, nozzle thrust
vector angle and thrust reverser/tail interference, afterbody/nozzle force data, etc.
[cf. Appendices C and D].

eIV-14 IR Signatures
Design criteria for vectored aircraft should permit estimation of internal propul-
sion performance, weight, mixing and cooling requirements, and the corresponding
infrared signatures of various advanced concepts.
Such static (M = 0) subscale rig tests have been reported in the literature [6, 91,
103]. They provide the technical data base for the development of improved
prediciton techniques for IR signature, mixing and cooling requirements [103] and
also of internal performance of various nozzle types over a wide range of nozzle
aspect ratio (NAR) and operating conditions.
 LECTURE V

PARTIALLY VECTORED AIRCRAFT

"It is better to debate a question without settling it, than to

settle a questin without debating it."
Joseph Joubert

"To find the meaning of an idea we must examine the conse-

quences to which it leads in action; otherwise dispute about it
may be without an end, and will surely be impractical."
C.S. Peirce (1878)
V-I Canard-Configured Vectored Aircraft
Thrust vectoring may require a variable incidence canard or canard/flaps for counter-
balancing the large pitching moments which are incurred, especially during STOL and
high turning rates. During such PST-RaNPAS turns, the thrust vectoring nozzle, and
the canard work as a force couple', rotating the aircraft around its center ofgravity and
couple it properly with nozzle's dynamic vectoring deflections. Obviously, the canard
effectiveness declines sharply during PST maneuvers above, say, 70 degrees AoA.
Several wind-tunnel investigations have been conducted to examine this problem,
including optimized locations, shapes, etc., for canard integration with super-agile air-
craft (cf. Figs. 10 and 11 below and § VI-2.S.3).

V-I.l Preliminary Design of Partially-Vectored, Canard-Configured

Aircraft
The Sedgwick-Lockheed example (Sedgwick, 164) discussed below is highly instruc-
tive in arriving at some preliminary conclusions regarding canard-configured partially
vectored aircraft. To start with, the analysis is based on a comparison with a datum
fighter aircraft shown in Fig. 1 and detailed further in Tables 1 and 2.

Table 1. Dimensions of Wing and Control Surfaces (Sedgwick, 164)

Characteristics Wing Canard Vertical Tail
Area - ft2 400 87,5 54.0
Aspect Ratio 3.5 3.7 1.309
Span - ft 37.4 17.5 8.4
Root Chord - in. 216.9 88.9 118.5
Tip Chord - in. 39.7 31.1 35.5
Taper Ratio 0.183 0.35 0.30
Mean Aerodynamic Chord - in. 148.7 64.6 84.4
Leading Edge Sweep - degrees 35 29 50
Thickness Ratio - Root - % 5 4 4
Thickness Ratio - Tip - % 5 4 4

156
Lecture V 157

f--=~-----::J-'/

I
15 FT .. 3 IN.
0831N.1
~ .

-=====:::::!....--I
1.--------~-- 51 FT.· 91"1. - - - - -

Fig. I. The Datum Aircraft (Sedgwick, 164) is equipped with 2 conventional exhaust nozzles of two G E
JIOI engines (cf. Table I for additional data). No visible effort has been made to reduce outer-skin-
shape/geometry ReS and IR signatures. Yet, this early design concept is highly instructive for this
course. (Cf. Figs. 2 to 4.)

Secondly we note that supercirculation effects have been incorporated in the prelim i-
inary design (Fig. III-IS).
Thirdly we note that the primary emphasis is placed on performance and/or weight,
which are influenced by nozzle-aspect ratio selection and, accordingly, by inlet-engine
integration with exhaust nozzle and the wing structure (Figs. 2 to 4).
Fourthly, we note that whenever significant differences occur, the most favorable
design concept has been sought for each nozzle concept.
Finally we note that all four prototypes use the same stores and forebody arrange-
ments, have the same basic wing, canard, and vertical tail geometries as listed in
Table 1.
158 Partially Vectored Aircraft

V-1.2 The Datum, Unvectored Aircraft

The datum aircraft is shown in Fig. 1. It has close coupled, all movable, horizontal
canard stabilizer, and a single vertical tail. Its main characteristics are:
I) The canard is located forward, on the strake, so that the downwash from the canard
does not significantly unload the wing [cf. Figs. 10 and 11 for other types of canard
designs].
2) The inlets are partially shielded by the wing strakes from flow angularities produced
during high AoA flight. However, the inlets may be highly visible to enemy radars.
3) The engines are installed inside the fuselage at a minimum nozzle-to-nozzle spacing
consistent with small rolling inertia, structure, and engine maintenance
considerations.

·j'I-- - -- -20 FT.- 2IN.~

- - -- - 5 1 FT.· 9 IN. - - - - - - -- - - - - - - - <

Fig. 2. The Partially Vectored Aircraft thrust-vectoring nozzles. (Cf. Figs. 3 to 4.) (Sedgwick, 164)
Lecture V 159

4) The narrow nozzle spacing, the nozzle interfairing configuration, and the single ver-
tical tail, have been selected so as to yield the highest possible (installed) T -D force,
following the experimental and analytical methods employed at Lockheed - Cali-
fornia Co.
5) The main landing gears are located on the lower outboard side of the fuselage, and
retract aft.
6) The wing has leading- and trailing-edge flaps.
7) No visible effort has been made to reduce outer-skin-shape/geometry ReS and IR
signatures.

(12
20 FT. 21N. ~
1----- - - - - - 51 FT. 9 IN .

Fig. 3. The Partially-Vectored Aircraft with AR =6 supercirculation thrust vectoring nozzles (see also
Fig. 4) (Sedgwick, 164).
160 Partially Vectored Aircraft

V.1.3 The Partially-Vectored Aircraft with NAR =4

This fighter is shown in Fig. 2 and is identical with the datum fighter except for:
1) The replacement of the conventional, axisymmetric exhaust nozzles with the
ADEN vectoring nozzles. (See, however, our reservations concerning the ADEN
nozzle in Fig. 5 Appendix A, and in Chapter VI.) All other differences are a direct
result of this replacement, and are concentrated in the aftbody.
2) NAR = 4 refers to the exhaust nozzle throat aspect ratio at the Maximum Dry Power
Conditon (MDPC). It yields, with proper nozzle interfairing, a substantially lower
afterbody drag.
3) While the required canard lift force for trim during maneuvering has increased with
pitch thrust vectoring, the associated canard area requirement has not exceeded the
takeoff rotation requirement (which is essentially the same for both fighters).

V.1.4 The Vectored Aircraft With NAR = 6 and Low Supercirculation

This fighter is shown in Fig. 3.
The major changes with respect to the prior two fighters are:
1) The engines are installed in semi-nacelles just outboard of the fuselage, and
upstream of the inboard portion of the wing trailing edge.
2) A modified inlet is installed directly upstream of each engine. The modification
icnludes two-dimensional, horizontal-wedge (overhead ramp), fixed-geometry,
two-shock inlet (cf. Fig. 4, Appendix F).
3) To preclude adverse effects of the canard wake on inlet performance, the leading
edge of the inlet external compression surface is placed just upstream of the canard
leading edge.
4) The upper surface of the nozzle fairs smoothly into the wing upper surface, and the
upper side acts both as a thrust deflector and as a wing flap.

V.l.S The Vectored Aircraft with AR = 17 Medium Supercirculation

Nozzles
This fighter is shown in Fig. 4.
The significant differences between this design and the previousAR = 6 fighter are:
1) The cooling airflow limitations preclude installation of a conventional afterburner,
and, consequently, a C-R transition section has been added. Thus, AlB burning takes
place in this C-R ducting. (It is not yet known whether or not burning in rectangular
duct is feasible, especially with respect to cooling loads. Yet, new fighter engines may
need no AlB.)

2) Five stiffening rings are employed on the outside of the AlB, C-R duct.
3) The outboard offset of the AlB C-R duct and the large change in cross-sectional
shape necessitates internal streamlined turning vanes. These turning vanes and two
Lecture V 161

1,5 FT, 71N.

I:I

- -..>..;-L----
20 FT. 21N.
=:9-...1..j.i4-------lj'. l
' - - - -- ----- 51 FT. 9 IN.

Fig. 4. The Partially-Vectored Aircraft (Sedgwick, 164) with Aspect Ratio 17, thrust-vectoring
supercirculation nozzles.
Note the internal air-exhaust gas ducting for the vectoring propulsion system. Estimation of
supercirculation effects may be done with the help of Fig. III-IS. For the definitions of pure vectored
aircraft, see Figs. II-la to Ii, and the Introduction.

vertical struts upstream of the nozzle throat, are used to carry loads between the
upper and lower wing surfaces.
4) Increased wing area affected by supercirculation enhanced lift.
5) Main landing gear design differences.

V.2 Mission Definition vs. Aircraft Design Constraints

The subsonic, close-air-support mission for the aircraft analysed here, as an instruc-
tive example, is illustrated below.
162 Partially Vectored Aircraft

The estimated 5214 lb armament load includes (Sedgwick, 164):

• close-air-support missiles.
• air-to-air missiles.
• 20mm gun.
The design requirements for 60% of takeoff fuel include:
• Level flight acceleration from 250 KEAS (Mach 0.46) to 494 KEAS (Mach 0.9) in
35 seconds at 10,000 ft altitude, with full armament load.
• 2 g's sustained maneuver at 150 KEAS at 5000 ft altitude, with full armament load.
• 5.5 g's sustained maneuver at Mach 0.9 at 30,000 ft altitude, with combat arma-
ment load (full armament load less 3276lb of close air support missiles).
• Maximum speeds with full armament load and 60% fuel not less than Mach 0.9 at
10,000 ft altitude and Mach 1.6 at 20,000 ft altitude
• Structural design load factors are +8.00 and -3.0 at full armament load and 100%
internal fuel.

V-2.1 Leakage Estimates

The vectoring ADEN nozzle requires increased coolant leakage with increased NAR.
Since the lost flow is directly proportional to thrust, it becomes equivalent to a loss in
Crg . The overall result is therefore as follows:

Fighter Leakage Flow/Power Loss, Percents

Dry Power AlB Power
Datum 0.5 0.3
AR=4 0.2 0.2
AR=6 0.2 0.2
AR= 17 0.4 0.4
Similar estimations for cooling flow needs and pressure drops finally give the over-
all effect of AR on Crg (cf. Fig. III-I 5).

V.2.2 WeIght and Fuel Iteratiulls

The computerized weight routine used by Lockheed (Sedgwick, 164) starts with an ini-
tial specified takeoff gross weight (TOGW), engine weight, and geometric data, such as
wing area and control surface areas (Table I).
The flight design gross weight is derived, and, along with other inputs, is used to cal-
culate the group weights, which are summed to arrive at the aircraft operating weight.
The fuel weight is, therefore, the difference between TOGW and the sum of the
operating weight and the payload.
If this fuel weight is insufficient to perform the specified mission, TOGW is
increased, new groups weights are calculated, and a new fuel weight determined.
Lecture V 163

The computational loop is thus repeated until convergence takes place, as shown in
Table 2.
Table 2. Weight Summary (Sedgwick, 164)
INCREMENTAL WEIGHT
DATUM ASPECT ASPECT ASPECT
AIRCRAFT RATIO 4 RATIO 6 RATIO 17
WING 0 0 269 269
BODY 0 0 -411 -475
LANDING GEAR 0 0 208 208
NACELLE AND ENGINE
SECTION 0 11 636 650
ENGINES 0 402 492 976
FUEL SYSTEM 0 -23 -67 -91
MISCELLANEOUS 0 -7 5 -3
TOTALS 0 383 1132 1534

OPERATING WEIGHT 17,284 17,667 18,416 18,818

PAYLOAD 3,716 3,716 3,716 3,716
ZERO FUEL WEIGHT 21,000 21,383 22,132 22,534
FUEL 7,000 6,617 5,868 5,466
TAKEOFF GROSS WEIGHT 28,000 28,000 28,000 28,000

V.2.3 Synthesis and Design Reservations

Aircraft synthesis results for all four prototypes are presented in Figure 5 as plots of
TOGW vs. takeoff wing loading WIS (lb/ft2) and TIW ratios.
While the values obtained by this instructive example are to be treated with caution,
and reservation, the methodology employed here is highly instructive for the prelimi-
nary design cycles of partial, as well as pure vectored aircraft.
The reasons for caution emerge from the limitations mentioned earlier, and from
the updated information about improved components, unit operations and testing
methodologies as described in this course.

V-3 Upgrading Extant Aircraft to Partially-Vectored Fighters

Figs. 16 and 12 show feathers required to upgrade an extant fighter, like the F-15, into
a partially-vectored one. So far, the F-15 modifications have not included yaw vector-
ing of the jets (13, 18,62, 70, 71, III, 128,234). Yet, they provide some yaw vectoring
through thrust spoiling of one engine; a highly non-inspiring design (Fig. 12).
So far, all proposed upgradings of existing aircraft have been limited to 2D nozzles
with low aspect ratios, or to ETV. These limitations are due more to tradition and iner-
tia than to bona fide constraints, such as limited nozzle space area, adverse interfer-
164 Partially Vectored Aircraft

WIS
. W/S
7S 8S 55

,. r--+-T+----+t+-\-t-~~+_-+_-+ 6 \\ \.\\\ \ u

V
. ----' ,.----- 1\
, 1\ \-'" 1\ \ ,., TIW

\ \ ~ \:.
""'\ eH 1.6

CRAFT
'"
UIGN INT

\- .\' 1,\" '<\

D
T-Oe sG

"
MACHO
""
-\
~ 1.0
8

--' V
\ '\ /
6

Takeofr Oro.. WdCht • Datllll\ Aircraft Takeoff Gro.. Wd,ht - Aircr&ft "it~ "_speet
with AxiaJ'llllletrlc Noul.. , Ratio ~ Vectored. Thrun Montes

1 .v \ \\ ,\
16 55

,..
55 5 --
55

\ \
50
/' 50
.~~
ll" \ 8~ \:~
./
.IS
85
V 1\
\' '\ \ V!~·
5

\'" '\ \ "...,

5

co
\., '\ \
r-- 1\
\
7~
D
,n

.:s \~
TIW ~ CHI.6
,1.2

~
5 RAFTO StGNP( INT

\<----, ,...--' eH 1.6

'"
~
T-O. 5G

\; r----.\ \
MACH 930KF

~- --~- LD _____ LD

30 -
AIRCfI FT DES' NP01N f7 ;~H'O 3GKFT

Takl!Ott Gron Weight _ Aircraft ",Hi: .\s~"t

1'aJr.eott Oro.. Weight. Aircraft ... 1th ",peet
Ratio 6 Supercircul.ation "!buIes Ratio 17 S~pe,.d"culatlon Ho:zl .. s

.Fig. 5. Partially-Vectored Aircraft Synthesis Results (Sedgwick, 164).

ence with eXIstmg wings, tails, rudders, actuators, fuel vessels and cooling
requirements. Consequently, one may conclude that future upgradings of extant air-
craft into partially-vectored aircraft, would be much better in comparison with the
expected performance of the current F-15 S/MTD (13, IS, 62, 70, 71, Ill, l2S, 234).
To realize this conclusion let us first examine the expected increase in performance of
the F-15 S/MTD, in comparison with the standard F-15C (Table 3).
To start with, there would be a + 7S% increase in the maximum lift coefficient. This
value may be considered low in comparison with the potential increases in CL when
properly integrated 2D nozzle/wings are used (cf., e.g., Figs. III-IS).
Secondly, one may examine Fig. 12 to realize why the yaw control power on the
modified F-15 demonstrated would be quite poor. As shown it would be provided by
spoiling axial thrust on one engine, while maintaining full axial thrust on the other.
The utilization of this kind of yaw control has a number of drawbacks:
First, engine thrust variations are relatively time consuming. Secondly, the yawing-
moment arm in the proposed configuration is relatively small. Consequently, it is
Lecture V 165

TABLE 3
Improved Performance Expected for the Partially-Vectored (STOL) F-15
(The F-IS S/MTD fighter, in comparison with the standard F-lSC)
(13, 18, 62, 70, 71, 111, 128, 234)
% Change
Maximum Lift Coefficient .....................................................+78
Deceleration Rate in Hight ..................................................... +72
Landing Run .................................................................. -72
Roll Rate, Mach 1.4/40,0000 ft .................................................+53
Pitch Rate, Mach 0.3/20,000 ft .................................................+33
Take-off Roll Distance ......................................................... -29
Approach Speed (at a constant a) ...............................................-16 Kt
Acceleration Rate, Mach 1.4/40,000ft ...........................................+30
Cruise range ..................................................................+ 13

much less effective than in pure vectored aircraft configurations (cf., e.g., Fig. 11.1), or
in a newly-proposed, simultaneous yaw-pitch-roll thrust-vectored F-IS (Internal
Report, USAF/WPAFB-Program: "Vectored F-lS", This laboratory).
USAF specifications call for the STOL F-lS to operate at the F-lS's normal combat
weight and typical weapon loads from I,SOOft x SOft runway at night in a 30 kt
crosswind with a 200 ft eight-octa ceiling and O.S n.m. visibility with only passive
ground-based landing aids. The 2D nozzles will be capable of 20· pitch-up/down
(either together or differentially) in full afterburner at sea level. They should also be

o o
Fig. 6. The Upgraded, Partially-Vectored (pitch-only) F-IS S/MTD (128). (See Table 3 for details of
the expected improved performance).
This may be the first manned vectored aircraft to fly in the "open history of aviation" (the Harrier's
methodology is a forerunner to this technology. However, the Harrier does not employ
supercirculation, nor rectangular, vectored nozzles.).
166 Partially Vectored Aircraft

3r---r-~.-------------~
MOV'K!Of' I"l'J nrust

M6no "lCiCflAQ
Power 101 ').
Le1Iel F110ht J '
1 .~~
o 02 04 06 08 10 12
Mach Numoel

Fig. 7. Vectoring moments variations during maximum thrust maneuvers at various altitudes and pitch-
throttle settings (Mello and Kotansky of McDD, 128) (Altitude = 30,0000 ft). (Thrust vectoring and
enhanced agility are discussed in the Introduction.)

2.0
I I
f. Maximum Power

lO~lorlng
, 0
"-
--- -
I
----~----
~ -- b--'

l6no 1I1tlO'II'IO \.- pawler tor L"" ~ ~h~ r

05

o
02 04 06
I
08 10
I 1 ,

Macn Nu oe '

Fig. 8. Thrust vectoring in the new F-15 STOL demonstrator is also expected to enhance roll perform-
ance (Mello and Kotansky of McDD, 128) (Altitude in this example is maintaied at 10,000 ft. (See,
however, the notes in Figs. ILl, 2,11-14.)

capable of full thrust-reversing at 50% military power. Its selected canards are actually
F-18 tailplanes, and its pitch rate improvements may reach 100% (13, 18, 62, 70, 71,
111, 128, 234).

VA Upgrading Studies of F-15 by Vectored RPVs

Our jet-propulsion laboratory tests, as well as the vectored RPV program, including
flying F-15C and F-15 S/MTD 117th models with similar propulsion/aerodynamic
characteristics, high a, low speed performance and inlet distortion data are now being
monitored, using our low-weight, on-board, computerized metry systems. Perform-
ance metrics are being evaluated for:
Lecture V 167

o 91 g Canard
I e 33 g BOdy· Wong
I

o 34 9
I
aol
i 0.10 g ThrUSI Vee lor

Fig. 9. Structural load distributions on the new F-15 S/MTD and on the conventional F-15 baseline for
an equivalent 8.33g body-wing load at a low supersonic flight condition (Mello and Kotansky of McDD,
128). Target tracking of vectored F-15 should be much enhanced.
The conventional F-15 requires an equivalent counter productive Ig tail download, resulting in a
net 7.33g load factor capability.
The new F-15 S/MTD, with the forward canard load to trim-out the wing load pitching moment,
results in a much smaller tail loading and an increased 9.0g maneuver load factor capability.
The increase in load factor at supersonic speeds, as a function of vehicle gross weight, may be
achieved with no increase in baseline aircraft structural weight.

1) F-15C (baseline-I).
2) F-15 S/MTD (Baseline-2).
3) F-15 S/MTD with additional yaw thrust vectoring.
4) A tailless F-15C equipped with our NAR = 46.6, simultaenous yaw-pitch-roll thrust
vectoring system (cf. Lecture IV).

It should be noted that the Soviets are currently testing similar designs on the
Su-27-1024 agile interceptor. Moreover, the unvectored Su-27 can maneuver (the
"Pougachev's cobra maneuver") at up to 110-120° AoA. (Speed reduction is from 250
to 80 mph.)

eV-5 Upgrading F-18 and F-16 to Become Partially-Vectored Fighters

Capone and Berrier [56] have conducted a comprehensive study on partially-vectored
F-18 performance with 2D-CD vectoring/reversing nozzles using the Langley 16-foot
Transonic Tunnel and a O.l-scale model of the fighter aircraft. Higher T-D perform-
ance (untrimmed) has been demonstrated in comparison with current-technology
F-18. Figs. 15-18 provide a few views on this type of model testing. ETV has recently
been selected as means to investigate high-AoA performance, using ETV-FI8.
Similarly, lTV is employed by this laboratory for studying the PST performance of
F-16, using ITV-RPVs.
168 Partially Vectored Aircraft

"""'orm P,r.mllrit C.nrrd Location

s- 30 II I / Sode lO"901uO ,nal Selecled Confogura1ton
L.E . Sweep Aspecl ~at'o Taper Rat,e Locat'on
(40· 10 60·) (2 .0 103.0) (0 .25100.40) lii/c...-"1fS3 10 0.70)

Q~Q
. 111::::;::: :~
. 11:::;;;;;::;

o,nealal
10· to 20·'

Fig. 10. Independently variable, close-coupled, dihedraled canards, fully exploit the potential of vec-
tored exhaust nozzles to enhance maneuverability of pure and partially-vectored aircraft (cf. Fig. II-I)
(Mello and Kotansky of McDD, 128).
The canards may be variable in pitch through an angular range of -35 degrees to + 15 degrees.
Since no efficient thrust-yaw-vectoring is available for the new F- 15 STOL fighter, the canards must
be dihedraled to provide direct side force (and additional yawing moment) through differential deflec-
tion in flight.
In the subsonic and transonic speed range, the canard acts somewhat as a leading edge device, and is
scheduled symmetrically with vehicle Mach number and angle of attack.
At supersonic speeds the canard provides additional lift, and because of its forward location, has a
benefit in the reduction o!supersonic nose-down moments. The forward location also results in a power-
ful positive trimming device to reduce negative tail and vectoring-down loads at supersonic conditions.
The canard geometry and positioning on the vectored aircraft may be evaluated for optimal lift,
drag, stall characteristics, yawing moment and side-force generation.
The various plan form configurations shown on the left side of the figure had been evaluated. The
one selected for the F-15 STOL fighter is shown on the right. The one installed on the STOL F- I 5 is
actually an F- I 8's tail plane. Different longitudinal locations, heights and dihedral angles, have also
been considered (center) (128).

V.6 The Search for Design Philosophy in the Post-ATF Era

There are many wind-tunnel reports which examine, under subsonic and supersonic
conditions, a matrix of inlet, nacelle, and exhaust nozzle configurations (cf. the Miller-
review - 8, 9, 47). These include various external or internal propulsion components
which are normally incorporated into a wind-tunnel model, insuring proper geometric
simulation of aircraft-propulsion interference effects in a large range of Mach num-
bers. However, this method is limited from a number of viewpoints to be discussed
below.
Lecture V 169

- - Ci.ruro 011 ~nocll DCI\lflOI'loS.

- - - - c.; l'I ti rO -~ .. 51'10(\ .poSr'I'Oh!i
- Se D.lII IIIQ-n OOllJlr'l(l.u y

.Fig. II. Canard-Wing Interactions (Mello and Kotansky of McDD, 128). The favorable interaction
between the proposed canard and the F-15 wing upper surface is shown here for Mach number = 0.9.
The canard-induced flowfield improves the wing-flow separation characteristics. The improvement is
greatest at the higher CL values, thereby further enhancing the maneuver performance of the aircraft
(13, 18, 62, 70, 71 , I II, 128, 234).
The canards offer a combination of relaxed static stability, lifting trim loads and favorable flowfield
interactions on the wing drag-due-to-lift (cf. Fig. II-I).
Other interactions are demonstrated by Fig. 13.

Let us first consider the various high-performance propulsion systems which have
been proposed for different V/STOL, subsonic and supersonic aircraft.
V/STOL proposals variously stress propulsive lift systems ranging from remote lift
fans driven by rigid shafts, ducted high-energy gas, lift/cruise nacelles, swivel nozzles,
remote augmentors, reaction control/compressor bleed systems (cf. Figs. B-7, 8,
A-1-5), up to partial and pure vectoring propulsive systems (Lecture II).
The primary problem in the design and application of these systems is to retain their
performance without robbing the powerplant of its ability to perform safely and
economically. One of the highest interactions between airframe and these propulsion
systems occurs in the transition from wing-borne flight to thrust-vectored
maneuverability and controllabiltiy. Additional drawbacks are related to the slow
responses of these TV systems, as well as to their complexity and low cost-effec-
tiveness. Accordingly, yaw-roll-pitch lTV remains the only cost-effective choice for
vectored aircraft.
Here the propulsion system must be designed as the primary flight control in terms
170 Partially Vectored Aircraft

Long lluduU I
Force

Revers,ng vanes

PIICII
I"~ ~ _(d=)===1.b:.IWL: 'IJ--====-
Power

Symmel"c TV Symmelroc Vanes

(Nose UP) (No~e Down )

ROil
Power

D,fferent'al TV Asymmelroc Vanes

(Roll R, hI) (ROil R'ghl)

Yaw
Power
~.
Asymmeillc Spo,hng
IVaw R'gh l)
Asymmel"c Vanes
(V lW P,'g"!)

Fig. 12. The thrust-vectored yaw, roll and pitch control modes of the new F-15 STOL Demonstrator
(Mello and Kotansky, 128). During yaw-thrust vectoring the F-15 S/MTD employs thrust spoiling of
one engine. instead of direct yaw-vectoring of the jets themselves. The latter is expected. according to
our recent experience with vectored R,PVs. to be much more effective in terms of thrust efficiency and
response times. as well as in terms of the low actuator forces required for yaw control of the jets (cf. Fig.
II-I). Furthermore. it remains effective at the PST domain of supermaneuverability (cf. the Introduc-
tion and § V.5). It should also be stressed that. in general. the 2D nozzle considerably improves stealth
performance in comparison with circular nozzles.

of forces, moments, rates, responses, accuracy and reliability. IFPC technology must
therefore integrate various design methodologies and employ them to enhance per-
formance and revolutionize the design philosophy of fighter/attack aircraft in the
post-ATF period. The benefits include the definitions of:

I) Force/moment envelopes in takeoff, transition and hover modes.

2) Thrust vectoring/modulation envelopes and cross-coupling coefficient ranges.
3) Reaction control demand effects on bleed requirement and, consequently, on
engine performance degradation.
4) Definitions of IFPC variables, their ranges and their cross-coupling effects.
5) Verification of the aforementioned IFPC variables by flight testing manned and
unmanned vectored aircraft.
 Lecture V 171

'Or---
__ ,----r-~r--,----r----r--·'I---~----r---~

16r---t---t---+---~..-.-_~~~~~'0
11r
' ---r---r---r--~~-~,--+---~--+---+---4
C..... ' ,f.I~ ......... r--....
Ol~-+--;---r--+--;---r-~ .--.~_~~~~ ..
o,~--t---+---+---+---+---~ ---~--;----r'---~',
---
O~~~~~~~~~--L-~--~--L--J
o
01 O' 11 " 06 08 10 16 18 10
rI

Fig. 13. Blending of vectoring nozzles with close-coupled canards improves trimmed maximum lift coef-
ficient of the STOL F-15 in comparison with the base-line F-15 (Mello and Kotansky, 128). The gain is
27% at M = 0.8, and 58% at M = 1.6. Similar designs are being tested on the Soviet Su-27-1024.
The higher lift coefficient is due to synergistic combinations ofincreased lift and relaxed static stabil-
ity afforded by the canard and an increase in control power from both the canard and the 2D- CD vector-
ing nozzles.
These increases permit improvements in instantaneous maneuver capability in many regions of the
flight envelope.

INCItlASING CQIiIIII'WIl1Y

Inc'."'" a1ro".tt CCIIIPlezlty ••

9round roll 18 reduced.

Sketch ot F-lS configuration ahowlng

location of rotating .ane 'eve, •• thrult nozzle • •

.
.1

:'
c• . r
0
·.r
... J<~
01 .0
o ~ ...
'.'
Z. O ~ ·'1'10 -'-
1.1 ~ (.UOI ~,Oft
:;
c, :

.ffeet of power on the longitudinal Sketch of luper.onte oeu.l •• fighte,

••,odyn. . lca of tb. P-15 configuration "ith
conf1gutatln .hawing location at .dn vectoring
dUferenti_l vector lng rotat 1"9 ·... n. no:ulea,
non I •• and no•• jet.
a... 90'/50' .

• Fig. 14. NASA (Paulson and Gatlin, ISS) has proposed various interesting methods to trim and vector
aircraft, including "canard-blowing" for STOL fighters.
172 Partially Vectored Aircraft

~--~::::IH1====;~
o 10
JH~__=<t=·l=';. 1....""..."
4
--==--~
-c:::=:cs:::,'III. ID'

,,-?,
[] -10 c.
. )0 1Q -to

I'I.'O~
"I~
1.0
C,
.>
•
....
'.>
...
""aUlt:ro74H "1. 1 " & ', 4 II ,4 ,I I 2

'.
·.ta.t.IlU1014H
'
Effect of .Uitary power setting on
Effect or hodzontd taU detlection on the approach aerodyne.tce of an advanced f 19hter
the lonejitudinal aerodyn •• lca of the V-lS configuration with ISH - 40·, CT • 1.0, elJ • 0.23,
cont,wuratlon with differential vectorlnCj rotating Ie· 10·, IS fe • ]0·, and 61C • on.
nozzl •• , 6• • 90·/50·, MPa • 2.64.

HleH lifT lA, _PI

Fig. 14. Continued.

Fig. 15. Thrust vectoring model proposed for the F-1S, as tested by NASA's wind tunnels (Bare and
Pendergraft, 234).
Lecture V 173

Inlet dLvM' ur
blud . 10 '

)lose .traka

l'~ttlc br u k

.Fig. 16. The F-18 model during wind-tunnel tests (Capone and Berrier, 56). The lower figure shows
the details of the nozzles and instrumentation employed. The dimensions are in centimeters.
174 Partially Vectored Aircraft

+-----+
/6 000
----~
+-

...... -
---"~ -----

~ .. ~~
+ -- <... • +---

----_.
, '-
" ·w

~ ~-~"",.,,.
t----.... - -

-.
. " -- .~ "lO"

• Fig. 17. Details of the nozzles tested by using the system and instrumentation shown in Fig. 16.
Sketch showing composite view of nozzles tested and some afterbody cross sections.
Sketch of 2-DIC-D nozzle. Nozzle has diverging sidewalls
Lecture V 175

Fig. 18 The wedge-type vectoring nozzle. Picture shows the NASA wedge nozzle in a down pitching
vectoring, 0" = 20° (cf. Figs. 16 and 17),

Fig. 19. Hot-gas-ingestion into the inlet is encountered during thrust-reversal landing. The ingestion
begins below a minimal ground speed. Asymmetric flow-field interactions with the vertical stabilizers
are also encountered during in-flight thrust-reversing (cf. Fig. 10, Appendix D).
176 Partially Vectored Aircraft

Unfortunately no flight-tested IFPC variables exist today for pure vectored aircraft.
It is imperative, therefore, to start flight testing programs that would evaluate these
variables. We return to this subject in Appendix F.

V-7 The Need for Partially-Vectored Trainers (179)

Existing trainers, such as the F-5F and A-37/T-37, may be easily modified to become
partially-vectored trainers (without, however, expanding their flight envelopes). These
trainers are now powered by J-85 and J-69 engines for which a number of pitch-yaw-
roll thrust vectoring nozzles have already been developed by this laboratory. Young,
unexperienced pilots may thus employ these low-cost trainers to switch, in flight, from
normal aerodynamic control to (weak) vectored control, and back. Thus, until a pure
vectored trainer becomes available (probably not before the year 2000), a low-cost
vectored trainer may serve to fill the extant gap in pilot education. Here the F-5F may
also supply unlimited yaw thrust vectoring, while the A-37/T-37 may be somewhat
limited by the fuselage during yaw thrust vectoring.
 LECTURE VI

THE PROS AND CONS

OF INTERNAL OR EXTERNAL
THRUST VECTORING:
In this lecture we assess the pros and cons of two, post-Harrier, thrust-vectoring meth-
odologies: Internal Thrust Vectoring (lTV) vs. External Thrust Vectoring (ETV). In
trying to select the best technology, one must compare the technical methodologies,
systems and conclusions of this lecture, with the general methodologies presented in
the Introduction. However, this selection is not essential for the industry of future
fighters, for ETV may be regarded only as a cost-effective tool for investigating
RaNPAS-PST, high AoA performance.

1.1 IS ETV EFFECTIVE AT THE RaNPAS-PST DOMAIN?

Partial Jetborne Flight (PJF) has been defined in the Introduction as a flight in which
elevons, ailerons, flaps, canards, or elevators, or rudders, are still being used in con-
junction with a thrust-vectoring system. Most ofthe ETV-aircraft assessed below, as
we shall see, must, therefore, be classified as partially-vectored aircraft, Le., present
ETV-aircraft are generally restricted to PJF.
Practically this means that the maneuverability and controllability ofETV-aircraft
are still affected, to a degree, by the external-flow regime. Consequently, high AoA
and/or high sideslip flights in the PST domain may still suffer from some of the classi-
cal aerodynamic limitations. To comprehend this subtle point one may have to refer
to a few specific examples. This will be done next.

1.2 ETV-F-14 AS THE SIMPLEST EXAMPLE OF THE

TECHNOLOGY LIMITS OF ETV
An F-14 flight with a single, post-exit, yaw vane was first conducted in 1987. As the
yaw vane is deflected into the free jet, it provides external thrust-vectoring force/
moment for yaw control at AoA up to 40 degrees, and at very low subsonic speeds.
During this flight, it was observed, as might be expected, that beyond a limiting
AoA, the improved ETV-roll rate encounters a limit. This limit is, in part, due to the
aerodynamic interaction of the external air flow, at high AoA, with the two engine
exhaust-jets, and with the yaw-pedal deflector.
178
Lecture VI 179

B1llllUQ]
~_II 10]

----·J,r---L/

Fig. VI.l. Wing-Engine Integration.

Thus, the central questions before us are as follows: Does this limit constitute a par-
ticular design limit, or an authentic technology limit? And, secondly, if, indeed, this
limit is an authentic technology limit, do the newer, multiaxis ETV designs, suffer, in
principle, from the same limitation? I.e., can the addition of yaw-pitch, post-exit,
pedal deflectors, by-pass, or surpass this limit?
In trying to answer these questions one may first note the following assertion:
The closer the multi axis, ETV-flow passage, approaches the geometry of a shrouded,
deflected, internal-flow nozzle (cf. Fig. 2), the less it should be affected by adverse,
external aerodynamic effects during RaNPAS-PST maneuvers.
As an analogy to this complicated flow-field situation, but only as a highly-remote
analogy, one may first examine the flow-field in and around shrouded vs. unshrouded
180 Internal and External Thrust Vectoring

fans at very-high AoA. The differences are obvious; thrust-vectoring, in general, is

characterized by the reasonable boundary-layer behavior of expanding flows, while
the air-compression stage inside fans is dominated by boundary-layer separation
tendencies.
However, during PST-flights, the post-exit pedals (Fig. VI-2) encounter good
boundary-layer behavior on one side, and highly separating flow regimes, on the other
side. But what does it mean for the expected thrust-vectoring efficiency, and for the
expected forces/moments of muitiaxis-ETV during RaNPAS-PST maneuvers?
In trying to answer this question one may first examine the highly-instructive, and
significant, static-test results of Berrier and Mason from NASA (209). Secondly, one
may examine the design, and the flight-testing programs of the X-31. Thirdly, one may
flight-test ETV-RPVs vs. ITV-RPVs during comparison maneuvers.

"'>0"\ c:-\ YAW RIGHT

U N R O L L RIGHT
l R

PITCH UP n I"Q(
'-+>\. U
YAW leFT
ROLl. LEFT
l R
NOTE: ARROWS DENOTE omECTION OF FORCES FOR
INDICATED VANE DEFLECTION

FLOW \Ill

• O... M ON I.EADING EDGE

PERCEIVED INTERACTION OF BACK FLOW • EKISTENCE OF BACK FLOW
WITH VANE ACTUATION SYSTEM - SOME 0," CROSSED
ISCliEMATICI OVER !lAM

Fig. VI-2. ETV (Pedal Type) as advanced by Berrier and Mason of NASA (208, 209).
Lecture VI 181

2.1 A FEW ASSERTIONS ABOUT EXTERNAL

THRUST-VECTORING

ETV, or post-nozzle, thrust vectoring, is accomplished by single, or multiaxis, post-

exit "pedals" (cf. the X-31, Fig. 7, Int.), which provide yaw-pitch, or yaw-pitch-roll
controllability (by deflecting the free jet emerging from an axisymmetric (X-31) noz-
zle. This methodology is associated with relatively simple, readily-available, pedal/
flap external devices, on one hand, and with the absence of (high-aspect-nozzle-ratio)
supercirculation lift-gains (X-31); external-flow-dependent, jet-deflection propul-
sionlflight control laws/reliability; relatively high RCS/IR signatures (especially with
circular nozzles), and longer over-all propulsion-system length, on the other hand.
Both methodologies have been proposed recently for the (partially) vectored,
superagile F-16 and F-18 fighter aircraft. Hence, it is imperative to compare the two
different maneuverability/controllability/propulsion potentials and limitations
associated with them.
However, we must first pay our great respect to the British-American designers of
the Harrier.
There is no doubt, that, somewhat similar to the historical British-American evolu-
tion of the Whittle gas turbine/jet-engine, the Harrier technology has significantly
advanced the field of internal vectored propulsion, with some well-proven combat
effectiveness.
However, it had been a priori faced with some inherent technology limitations, e.g.,
the absence of the powerful yaw thrust vectoring and supercirculation lift gains, and
the introduction of high nozzle-drag, and heavy engine components, which reduce the
thrust-to-weight-ratio of both engine and aircraft (179). The pros and cons of this tech-
nology, as well as its newly-proposed improvements in terms of advanced STOVL
concepts, are available elsewhere (179), and in Appendix B.

2.2 THE PROPULSION SYSTEM OF THE VECTORED X-31

According to Herbst (188), there is no reason to use a rectangular nozzle configuration

when there is no preferred direction of jet deflection. While we share Herbst's skepti-
cism as to the priority that has been given to pitch/reversal over yaw-pitch vectoring,
we may differ in our assessment of circular vs. rectangualr, and internal vs. post-exit
yaw-pitch-roll vectoring nozzles. Accordingly, two, wing-integrated, rectangular,
internally-directed, yaw-pitch-roll thrust vectoring nozzles (Fig. 7, introduction), may
be superior to the post-exit, circular, yaw-pitch pedals, or to axi-TV nozzles.(Fig. 2).
Nevertheless, the X-31 constitutes one of the most important and most promising
aircraft in the evolution of vectored aircraft. Its flight testing in 1990 in the USA
would certainly become a significant milestone in aviation history. It would no doubt
become the 1st PST-ETV manned aircraft in aviation history.
182 Internal and External Thrust Vectoring

2.3 POST-EXIT PEDALS FOR FIXED 2D ENGINE NOZZLES

Another important contribution in yaw-pitch thrust vectoring was recently made in

NASA-Langley Research Center (207-209). In one of the most promising designs
(20S), post-exit vanes were mounted on the side-walls of a nonaxisymmetric 2D-CD
nozzle (20S). While the resultant yaw vector jet angles in this design are always smaller
than the geometric yaw vector angle, the widest post-exist vanes produce the largest
degree of jet turning. In fact this type of propulsion system combines some feathers of
2D nozzles, e.g., lower drag and RCS/IR signatures, with the simplicity of external
thrust vectoring pedals.
In fact, both internal and external, single, or multiaxis, thrust-vectoring systems
have recently been proposed for the upgraded versions of (partially) vectored F-lS,
F-16 and F-15. However, ETV in these studies is now considered only as a means to
investigate high-AoA flights.

2.4 CAN THRUST-FLAP VECTORING BE EMPLOYED?

Fig. VI-l describes a possibility to add external-flap thrust vectoring to, say, the
Advanced Tactical Bomber (ATB), the B-2. To start with, one must notice the follow-
ing concepts:
I) - The fixed nozzles of all four engines are located atop the wing.
2) - The side-by-side, nozzles of each couple of engines mix their emerging hot jets
with cold secondary air. The cold secondary air is venturi-pumped by the 2D nozzles
from the secondary inlet slot just below the main inlet. The two streams are mixed
inside a semi-confined, wing-recessed duct. This duct is open only on one side,
namely; in the upward direction. Hence, the IR signature of the mixed, semi-hot jets,
is drastically reduced in both the downward and sidewise directions. (However, the
semi-hot jets just behind the wing trailing edge are still radiating.)
3) - Two oblique thrust-vectoring flaps may be added downstream of the semi-
confined, wing-recessed ducts.
4) - Together, the two flaps form a Vee in the trailing edge ofthe wing. Now, both
flaps may be rotated together, or separately. (At the present design they appear to be
fixed.)
5) - An upward rotation of both flaps may supply nose-up pitch moment, while
their combined downward rotation can turn the jets downward, until a limiting angle,
by the Coanda effect. Consequently, downward rotation also provides thrust-
vectoring pitching moment, i.e., in this case, a nose-down moment. This downward jet
deflection also enhances lift, both by supercirculation effects (cf. Lecture IV), and by
the direct lift force of the down-deflected jets. Consequently, STOL and lower
approach speed payoffs may result, or TOGW may be increased.
6) - Thrust-vectoring roll may be carried out by rotating each of the two couples of
Lecture VI 183

thrust-vectoring flaps in opposite directions to each other, while pitch thrust-

vectoring propulsion/flight control is maintained by rotating all four thrust-vectoring
flaps in the same direction.
7) - Yaw thrust-vectoring control may be affected by rotating, in each engine
exhaust vee-bay, only one of the two thrust-vectoring flaps, say, only the right-hand
flap.
Note also that the existing outer, vee-shaped, aerodynamic-control surfaces, are
designed and controlled according to some similar reasonings, but without thrust-flap
vectoring, i.e., at very low speeds and high-AoA they become ineffective, while the
proposed thrust-flaps remain effective.
8) - Simultaneous yaw-pitch-roll thrust vectoring may become feasible by a prop-
erly designed IFPC system. Moreover, it remains effective even during very low
speeds and high AoA, approach, runway and ground handling. Consequently, STOL
and/or payload needs-specifications may be considerably revised by adding rotation
to these two trailing flaps.

2.5 INTERNAL THRUST-VECTORING FOR THE (Now Cancelled)

VECTORED X-29A

2.5.1 Previous Program Goals and Participants

Grumman's forward swept wing X-29A Program is a joint DARPA/NASA/USAF
project. The X-29 was to be flight-tested at up to 40 degrees AoA, but without thrust
vectoring.
Highly instructive simulations of thrust-vectored X-29A have, nevertheless, been
reported recently by Klafin of Grumman (220).
Grumman's goal was to evaluate the controllability of the X-29A with multi-axis
thrust vectoring for AoA up to +90 degrees. The 2-axis thrust vectoring was to be inte-
grated with conventional aerodynamic flight-control surfaces. Hence, the vectored
X-29A was to become a partially-vectored aircraft.

2.5.2 The Thrust-Vectoring Simulations

Klafin reports about piloted thrust-vectored simulation at 0.2 Mach, 15,000 feet and
33 degrees AoA. Without LTV the X-29A has adequate longitudinal control power for
flight up to approximately +60 deg. With LTV this value increases to +90 deg. (The
simulation was artificially stopped at +90 deg. AoA.) Thrust-vectored, lateral-
directional control was simulated up to +90 deg. AOA. Thrust vectored augmentation
was also found to reduce pilot workload with highly improved handling quality rat-
ings. Flight control envelope was expanded by thrust vectoring to lower speed, lower
dynamic pressures, and higher AoA.
Lateral-directional maneuvers are not possible at high AoA on the unvectored
184 Internal and External Thrust Vectoring

X-29A (even though ailerons remain effective to 90 deg.), due to fuselage masking of
the rudder. With Directional Thrust Vectoring (DTV), there was adequate lateral-
directional control power for all longitudinal maneuvers and ambitious lateral-
directional maneuvers, including 360 deg. rolls. Other benefits obtained with thrust
vectoring are enumerated below.

2.5.3 Aerodynamics, Stability and Flight Control

The X-29A canard aerodynamic design is integrated with the wing so that the canard
flow field interacts favorably with the wing and the wing flow field interacts favorably
with the canard.
This close-coupled, variable incidence canard design is advantageous because the
canard carries lift and also contributes to the aircraft being near neutral stability at
supersonic cruise for minimum drag. However, this also causes the aircraft to be up to
35% statically unstable subsonically. This relaxed static stability is controlled with
full-authority, digital FBW flight control system.
The normal flight control was employed during the simulations. Primary X-29A
longitudinal control was provided by blending the canards, inboard and outboard
wing flaperons, and strake flaps collectively as longitudinal control surfaces.
Aerodynamic surface control power decreases as dynamic pressure is reduced; this
is compounded by decreasing control surface effectiveness as AoA increases. As
expected, the addition of longitudinal and directional thrust vectoring has greatly
enhanced high-AoA, PST, control power.

2.5.4 The thrust-Vectoring Nozzles

The thrust-vectoring effectiveness of the ADEN and the GE gimballed nozzles (see
Appendix B), have been evaluated and simulated, using a variable-geometry inlet and
the F404-GE-400 engine. Articulated sidewalls were included in the simulation of the
ADEN to provide DTV (Directional Thrust Vector) in addition to its LTV (Longitudi-
nal Thrust Vector). (The gimballed nozzle provides DTV and LTV by its very design
concept - see Appendix B.)
During these studies, the gimballed nozzle was found a better choice than the
ADEN. Its characteristics include: lighter weight, minimum thrust magnitude effect
with deflection, thrust angle nearly equivalent to nozzle deflection angle, near-con-
stant thrust moment application point, and pitch/yaw angle application point coinci-
dence. Compared to the ADEN, its flow field is circular or eliptical, while the ADEN
flow field is rectangular.

2.5.5 LTV Benefits

Instantaneous pitch acceleration was increased significantly with LTV. Using only dry
thrust vectoring, the pitch acceleration of 0.5 radlsec 2 was expanded from a range of
Lecture VI 185

-9 to +5 degrees AoA (with only aerodynamic flight control), to +90/-50 degrees

(where the aerodynamic flight control power is practically ineffective).
The pitch acceleration increment due to LTV is essentially constant with airspeed at
fixed thrust levels. Moreover, LTV control moment surpasses that of the canards-
flaps-strakes combined (100%) below 124 knots, for nose-down, and below 109 knots,
for nose-up.
Furthermore, significant takeoff benefits were obtained with LTV: Liftoff speed
was reduced 17%, distance reduced 37%, and time reduced by 20%.
Using only aerodynamic control surfaces, the loop gain can hardly be increased.
However, when LTV was added, the loop gain was significantly increased.
The vectored-X-29A aircraft was found "perfectly capable of being flown to +90
degrees AoA (and higher) in the longitudinal axis, discarding any adverse inlet/engine
effects, which were not simulated" (220).

2.5.6 DTV Benefits

The baseline X-29A is unstable in roll at 0.2 Mach/15,000 feet. Moreover, with the roll
mode stabilized, effective rudder control is lost above 45 degrees AoA. However,
above this aerodynamic limit, DTV was found effective in providing directional con-
trol power.
Fighter aircraft typically roll about the stability-axis velocity vector to maintain
AoA and to minimize sideslip/lateral acceleration. Coordinated body-axis roll and
yaw rates are required to perform a stability-axis roll. For a given stability-axis roll
rate, the required body-axis yaw rate increases with AoA. Since rudder power
decreases with AoA, yaw thrust vectoring was found to greatly augment rudder power,
significantly increasing the roll rate/AoA capability as the aircraft enters into the PST
domain.

2.5.7 Flight-Control Changes Due to Thrust Vectoring

The application of thrust-vectoring requires new control/coupling rules and sub-
stantial flight-control system design changes in such categories as:
- DTV to ailron cross-feed to correct DTV coupling into roll.
- Lateral-directional cross-feed paths to provide stability-axis rolls with high AoA.
- LTV gains vs. the longitudinal system loop.

2.5.8 Piloted Simulations

Thrust vectored, simulated, piloted maneuvers were started with aircraft trimmed in
straight and level flight at 33 deg. AoA, with 34 Ib/fe dynamic pressure.
Longitudinal maneuvers included full aft control and hold, 80 deg. AoA capture and
hold, and specific pitch attitude capture sequences.
Lateral-directional maneuvers included roll attitude captures and 360 deg. rolls.
186 Internal and External Thrust Vectoring

These piloted simulations demonstrated that LTV improved pitch acceleration,

dynamic stability, and aircraft maneuverability. Significant handling quality
improvements were noted in pitch attitude and AoA tracking tasks.

2.5.9 Pilot Comments On Thrust Vectoring

Kalfin provides some pilot comments on thrust-vectoring simulated sessions, among
them: "Thrust vectoring helps to control AoA. Holding 60 deg. is possible with LTV,
and impossible with pure aerodynamic controls."

3.1 POWER-LIFT AIRCRAFT

According to USAF/MAC (204), powered lift is a leading technology contender for

low visibility, speed, and the ability to fly low and survive in an extreme EW
environment.
Thus, the US Air Force has been studying three different demonstrators: a generic
STOVL(cf. Appendix B); an assault transport; and a tactical transport aircraft (204).
Four propulsion concepts are being investigated under the US/UK ASTOVL
(Advanced Short Take-off/Vertical Landing) MOU [signed by the US DoD, NASA,
and the UK MoD in 1986]: Vectored thrust; ejector augmentation; tandem fan; and
the remote augmented lift system (cf. Fig. 8, Appendix B)
Upper Surface Blowing (USB) (Fig. 1, Appendix A), is being studied by Japan's
National Aerospace Laboratory and NASA Ames Research Center. This involves sub-
sonic STOL transport data, conceptual-designs, and flight-test results from the Asuke
STOL research transport, based on the Kawasaki C-l (204).
An ejector-lift STOVL fighter model (cf. Fig. 8, Appendix B) had been undergoing
windtunnel tests in General Dynamics during 1988 (204).
Franklin, Hynes, Hardy, Martin and Innis, of NASA Ames Research Center, have
recently flight-tested a power-lift, STOL, research aircraft during approach and land-
ingin a touchdown zone as small as 200x50 f1. (193). Using digital FBW flight control
system, a head-up display, a color head-down display and nonlinear, model-following
methods, the authors evaluated the influence of highly augmented control modes on
the ability of pilots to execute precision instrument flight operations in the terminal
area.

3.2 UNPOWERED MODELS FOR POST-STALL

MANEUVERABILITY: A GERMAN PROGRAM

Ransom, of MBB/VFW, has recently described the pre-development considerations

of a small, subsonic, delta-canard, highly-maneuverable, research aircraft as deter-
mined from the results of low-speed wind tunnel testing of a 1/7-scale model in MBB/
Lecture VI 187

VFW's facility in Bremen, and of a 1/ 13-scale model of a twin-engined variant at the

DFVLR, Braunschweig (194). The wind tunnel models were tested at a Mach number
of approximately 0.2, a Reynolds number of approximately 106 , and AoA from -5 to
+ 70 deg (VFW), and from -5 to +40 deg (DFVLR), and at angles of yaw between-1O
and +10 deg. The ultimate aim of this program is flight demonstration of direct-lift
with vectorable engine nozzles, direct side force control, and fuselage aiming modes
and post-stall modes, i.e., flight at AoA greater than that at which maximum lift is
obtained. (CF. the X-31 program.)
 APPENDIX A

A BRIEF HISTORICAL SURVEY

ON INDIVIDUAL AND
COLLECTIVE CONTRIBUTORS
A-I The Definitions and Acceleration of US Thrust-Vectored Aircraft Programs
The US programs during the period 1973-1980 are well-covered by the combined compre-
hensive review of Richey (from the USAF), Berrier (from NASA) and Palcza (from the Navy)
[73]. These authors stress that the contributions of USAF, NASA, and Navy and industry have
demonstrated that advanced, nonaxisymmetric nozzles, integrated with tactical fighter aircraft
configuration, have the potential for;

• increasing maneuverability and agility at high lift and/or low dynamic pressure conditions by
thrust vectoring with supercirculation,
• reducing subsonic/transonic cruise drag compared to close coupled axisymmetric nozzles
through better nozzle/airframe integration,
• improving longitudinal agility for air combat, and increased accuracy and survivability in air-
to-ground weapons delivery due to steeper dive angles and higher weapon release altitude by
incorporating an in-flight thrust reverser,
• reducing IR and RCS signatures due to nozzle configuration influence; also, the IR and RCS
observables tend to be highly directional, thereby having the potential of greatly increased air-
craft survivability against seeker missiles by maneuvers and application of thrust vectoring,
• improving take-off and landing performance and ground handling of high-thrust aircraft,
• reduced life cycle costs through projected lower fabrication costs (fewer moving parts),
reduced fuel costs to perform the same mission (due to lower cruise and acceleration drag),
• reduced fleet size, as a result of improved survivability and combat effectiveness.
These authors also stress that the proper utilization of this emerging technology in future air-
craft systems depends on;
• a coordinated attack on nonaxisymmetric nozzle technology to improve the data base on the
disciplines of propulsion, aerodynamics, structures and vehicle integration,
• definition and implementation of a meaningful. timely. flight research program to increase
confidence in the technology for transition to systems applications, to define and demonstrate
advanced fighter flight characteristics, to utilize design criteria for nonaxisymmetric nozzles
with thrust vectoring features, and to define full-scale performance and survivability
characteristics.

Accordingly, the authors describe a series of R&D programs, which, beginning in 1977, are
designed to substantiate the need for flight research programs, define their specific objectives,
evaluate candidate test aircraft against these objectives, and acquire data to find the cost-
effective approach to a 2D nozzle flight research program.

189
190 A Brief Historical Survey

VectOfed ,1VuS, RALS

,, /
, I

(CJ~~
I I ~
Lift-Fan or
Di rect-Life EIlS;..1.!!.£.L

Fig. I. The evolution of a few power lift concepts, including some early vectored thrust control designs.
For STOVL concepts see Figs. 7 and 8 in this Appendix.

A-2 Reduction in Intederence Transonic and Supersonic Wave Drag

The authors of this timely overview stress a number of additional points:
1) Exhaust nozzles have traditionally been circular in cross section to facilitate good integration
with the core engine, thereby neglecting the performance penalties associated with the
airframe.
2) A large part of "rounded" nozzle installation penalty for twin-engine fighters is due to the poor
integration of "round" nozzles into a "rectangular" afterbody, and the very neglection of
supercirculation potentials.
3) These "rounded" configurations generally have boattailed "gutter" interfairings, or base
regions on the afterbody, which result in interference drag at transonic flow conditions, and
wave drag at supersonic operaton. Thus, the installation of 2D nozzles reduces external flow
nozzle penalties by eliminating the separated interfairing and base regions.
Appendix A 191

I - -·-

~
.~- :> c-
~

- --~
---
----- -

Fig. 2. The evolution of a few "jetwing" design concepts.

upper left:
Dividing engine cold and hot streams for separate lift enhancement sectors involving cold and hot
materials, respecti vely.
upper right:
Core (hot) exhaust gas remains unchanged, but fan (cold) air IS used to enhance lift and
maneuverability.
lower Ie.ft:
Internal ducts are used for inboard and outboard blowing, utilizing the Coanda Effect [156].

A-3 Individual Contributions

Many names are associated with the subject matter of this book. Their collective and individual
contributions are well-reflected in the references given at the end of this book. What has been
described in the text is, therefore, only an introduction from this author's point of view. For a
full appreciation of the individual efforts one must turn to the original referencb.
By closely examining these references one can distinguish the collective contributions of such
bodies as USAF (and especially that of Bowers, Glidewell, Laughrey, Richey and Surber at the
Flight Dynamics Laboratory of WPAFB), NASA (and especially that of Banks, Berrier, Capone,
Leavitt, Mason, and Pendergraft), the US Navy, industry (especially that of Herbst and McAtee)
and the professional leadership demonstrated by other individuals.
Among the individuals who have significantly widened the scope and aims of this technology,
one may pay a particular attention to the valuable contributions of:
Banks [29, 148), Berndt [63, 142, 198, 205), Berrier [1, 6, 18,44,52,55,73,208,209,210), Bowers
192 A Brief Historical Survey

AXISYMMETRIC C·O NOZZLE

C. E. ALBEN.

Drt"'-'. M" - 0.&.1, -II", NPR -10 _ _ _---,

..<
z:
8
v

-0.02 -'-=-"""'
'''I:-
",--l~.O ( ·0 Mu1d · ell"'tC'd T~ft.
A.ld ( -0 ru,potl FlIP Thto.l
CO Plug Plul

VACS EXIIAUST SYSTEH

(lID • TK6PN

Fig. 3. The evolution of a few early concepts in thrust-vectoring nozzles. (The "thrust loss coefficient"
data were taken from Hiley, Wallace and Booz, Ref. 72. The VACS and CRD+ TRBPN nozzle figures
are from Berndt, Glidewell and Bums, Ref. 142.)

(4, 11, 13,24), Burley (33, 114, 162), Callahan (158), Capone (10, 17, 18,20,21,22,28,37,42,54,
56, 49, 60, 64, 67, 68, 69, 121), Chu (90, 92, 100), Costes (182), Glidewell (5, 14, 142, 198,
205), Goetz (41, 77), Herbst (154, 188], Hiley (11, 61, 66, 72, 78, 139], Joshi (124, 140, 161],
Klafin (220], Kotansky (128), Kucher (63,138], Laughrey (4,13), Leavitt (58,112,162], Mason (38,
39,40,52,57,208,209,215), Maiden 132, 34, 36, 40], McAtee 1196), Mello 1128), Miller 18, 9, 47),
Nash 145, 46), Palcza 16, 41, 43, 46, 73, 75), Paulson (15,16,27,29,148, ISS), Pendergraft (62,
234), Petit (40, 42, 68, 77), Re 128), Richey (6, 73, 210), Schneider (218), Sedgwick (164), Stevens
(49, 70), Surber (210), Tamrat (185, 213), Thomas 195, 96), Watt (218) and Well 1211).
A substantial portion of this book is actually based on the highly significant contributions of
these individuals.
Appendix A 193

@®
® GeACUtN Gt'OC·D

~ ~.;;:,::~"; M>.
P.W~:AS" ~~
loe 0
IR,.,... utl

....
Mci"- Slluwn'
'6wI\lMCAln

Fig. 4. A few early nozzle competitores in the thrust vectoring "race". The GE 2D-CD nozzle on the
upper-right is essentially the one shown in Figures I-I and 1-7. The ADEN nozzle is shown on the right
during various operating conditions (cf. Fig. 5).

ATA and ATFN (See also Appendix B)

Beyond the year 2000 the Navy is likely to replace the F-14 with a version of the USAF's
Advanced Tactical Fighter, known as the ATFN. Under a memorandum of understanding
signed in early 1986, both the (ATF) Northrop/McDonnell Douglas YF-23 and the Lockheed/
Boeing/GD YF-22 are being designed so that a navalised version can be developed without
excessive modification and the Navy will monitor the USAF program.
On the attack side, the Navy is developing the Advanced Tactical Aircraft (AT A), or A-12.
AT A appears to be running about two years ahead of ATF, so it may be operational in the
mid-90s. It is believed to be a subsonic aircraft powered by two uprated, nonafterburning
GE-F404 engine derivatives. It will be a two seater designed for low-level, all-weather, deep-
strike missions. AT A has been defined as a long-range, heavy-pay-Ioad aircraft, using stealth
technology.
The McDD/General Dynamics team has recently won the ATA/A-12 contract.

--
Z ~~,

-.....,.~..
Fig. 5. The ADEN and PWA Plug nozzles. (A particular analysis of ADEN/X-29A integration is given
in §YI-2.5.4. Its use on a STOYL aircraft is depicted in Fig. 8, Appendix B). Cf. Fig. 4.
 APPENDIX B

POWERPLANT TECHNOLOGY LIMITS

(A portion of this Appendix is based on an updated version

of a review published in 1984 in the Intern. J. Turbo and Jet
Engines, Ref. 119)

"Perfection of means and confusion of goals seem - in my

opinion - to characterize our age."
Albert Einstein

"Who but the Harrier pilot can thrust vector in forward flight
and confuse an attacker by killing forward speed, so that the
enemy overshoots into a lethal position?"
The Editor (JWRT)
Jane's All The World's Aircraft
1987-88 p. 43 [1751
B-1 Beyond the year 2000
R&D groups around the world are now shifting attention to a new generation of engine and
materials technologies to provide quantum leaps in performance, reliability and survivability
beyond those already on the drawing boards.
Until recently, much of this effort has been aimed at fighter engines, such as those for the A TF.
ATA and EFA. However, since the design definition of these engines has now been completed,
and since the various technology limits had been identified and set, advanced R&D and design
goals are now beginning to be redefined in terms of the next generation beyond.
The payoffs of these new powerplants are expected to come from a number of new technolo-
gies. These include:
I) RCC Materials (see below).
2) Vectored Propulsion Systems.
3) Advanced Engine Cores.
4) [FPC Technology.
Current-technology engines for the ATF, ATB and AT A are described below. The technology
limits ofthese engines are typified by a thrust-to-weight ratio ofaround 10: I to 12: I, while those of
the new-generation, beyond-the-year-2000, by a ratio of 20: I. These engines will include high
through-flow compression systems, to reduce the number of compressor stages, segmented com-
bustor liners and TT4 beyond 4000 F. Substitution of fiber-reinforced rings for compressor discs
is expected to reduce engine rotor weight by a factor of 3 or 4. A test compressor of this type may
be available in the I 990s for use in the Post-ATF engines. Ceramic bearings that can function
with reduced quantities of cooling loads may also be available by that time (see below).
The newer engines are thus expected to be made of much lighter, but higher-temperature mate-
rials. By itself, this will result in a substantial reduction in overall aircraft weight, and/or sizeable
increases in range and payload, and/or increased thrust-to-weight ratios beyond, say, 1.3: 1.
Engine part counts may also be substantially reduced, say, to 2000, in comparison with
15,000-20,000 on current-technology fighter engines. These, in turn, have up to 50% fewer
parts, 40% less supersonic fuel consumption and 20-30% reduction in life-cycle costs than the
FIOO fighter engine.

B-2 Uncooled Turbines/Nozzles at Up to 4000 F

Major gains in SFC and deSign simplicity are expected to emerge from materials such as RCC
and other ceramic/ceramic composite systems. These can operate uncooled at up to 4000 F (cf.
Figs. I and 2).

195
196 Powerplant Technology Limits

29
~Grtphl\4!' Poly
17

---- ---- ------

~4!'lnforc4!'CI fArbcr1 Carbor

8
I -I.
Str4!'ngth ~---1

-
toW.,gh1 \ , rTilanlum
rTI/AL
r lNCO 18/
Ratio 6
'<
,
1110'"
psVlb/ln J ~

I~,rAlumin~
....
'
--~~~\''-Waspaloy
---- \

-
rlNC0625
2
~ ! r olumblum
\
800 1600 3200 ,GOO
T4!'mperatur4!' of

Fig. I. Future fighter engines will be made from higher proportions ofRCC (Reinforced Carbon-Carbon
Composites).
Well-impregnated with oxidation-resistant ceramics, such as boron and silicon carbides, the new
materials demonstrate excellent performance at skin-temperatures in excess of 4000 F.
Eventually not only the thrust-vectored nozzles and the transition/augmentor's ducts will be manu-
factured from RCC, but the hot section of the engine as well.
A major difficulty in applying this technology today to all advanced fighter engines is the lack of effi-
cient manufacturing processes amenable to impregnating large-scale production parts (cf. Fig. 2)
[Bowers, 13, Gal-Or, 146].

double bl.de
ca!bon·corhun
pin coml)()nent rapid .00ldilicltion

/'4Qr'
0
aluminum Of litlnium
anoy

~ ~\\
\ I high lI,englh/welghl
\ / ,allo Compolill inll,l

/
"'-
\ - - - - - - - - - - - - t.
/
-,,",/

Fig. 2. A Reinforced-Carbon-Carbon (RCC) Turbine.

A wrap-around blade segment of RCC attached by a pin of an RCC disk may characterize future
engines.
The bladed ring, or "bling", shown here is much squatter than conventional designs. It is reinforced
by a composite insert to withstand the stresses of high rotational speeds.
Appendix B 197

Variable Iow-

Coannular
acoustJe nozzle

Fig. 3. The variable-cycle engine proposed by General Electric (cr. § B.12).

In recognition of the key role materials will play in future power-plant technology, the
so-called Integrated Technology Plan/or the 1990's, has been evolving from USAF Forecast II
Analysis [as adapted by joint Aero Propulsion Laboratory and the Materials Laboratory Projects
at USAF WPAFB, and in the US aero-engine industry]. These projects can lead to 20:1 TIW
ratio in future fighter engines with the same life expectancy as that of the ATF engine, (e.g.,
increasing inspection times with respect to around 1800 cycles for the ATF engine hot section
parts.)
Current emphasis is on damage tolerant designs, using materials that are configured to limit
propagation of cracks.
The new integrated design approach to turbine blades and disks, balances between what the
aerodynamicist may say he needs, what the thermal requirements are, and what the durability/
maintainability designer may figure as a reasonable interval for inspection.
Contra-rotating spools and single-stage high and low pressure turbines may characterize the
new engines. Both GE and PWA currently use single crystal superalloys in the turbine, improved
titanium alloys in the compressor system, and RCC as external skins for the thrust-vectoring
nozzles. For more details on the ATF engine see § B.ll.

B-3 RCC Thrust Vectoring Propulsion

RCC composite materials offer the best potential for drastic improvements in hot-section pro-
pulsion system performance, especially in optimal integration of wing and engine, and in reduc-
ing about 50% in the combined weight of a typical high-performance fighter engine.
Substantial increases in maximum thrust can be materialized, since practically no cooling
flow is required to protect RCC composites. Indeed, Figure 1 shows a dramatic reduction in
strength for "metals" at skin-temperatures in excess of 1600°F, while RCC composites maintain
their strength at temperatures in excess of 4000°F.
This fact, combined with almost negligible thermal expansion, in comparison to "metals",
also results in increased durability and reliability.
However, to prevent oxidation of the carbon fibers and of the inner matrix at elevated temper-
atures, one must improve the current-technology methods to impregnate the matrix porosity
with carbides, etc., using specially-developed new processes [146].
198 Powerplant Technology Limits

Fig. 4, Proposed configurations for turbojet/ramjet/scramjet modes of a future engine.

It should be stressed that no coating can offer the oxidation protection and durability available
by suitably-impregnated Carbon-Carbon composites.
Another important parameter of RCC is its electromagnetic capacity to absorb radar waves.
(This subject will be covered in more detail in Volume II.)
Hence, most future propulsion systems will contain higher proportions of RCC in their vari-
ous, highly-loaded, hot sections, and in their auxiliary systems.

B-4 USMC/STOVL Aircraft.

In the last decade, the US Marine Corps has taken a lead in the development ofSTOVL aircraft.
The AV-S8 Harrier II, developed by McDonnell Douglas for the USMC, has already demon-
strated an advance in performance and safety over the AV-SA.
In the long run, the USMC may want to replace all of its current combat aircraft-AV-SBs,
F-lSs, and A-6s - with a single STOVL Advanced Combat Aircraft (ACA). The "desired opera-
tional capabilities" include a 300n.m. intercept mission radius, with a speed of Mach 1.4 sus-
tained on the outbound leg, and a 450n.m. offensive counter-air mission radius.
The USMC may consider the STOVL as the key to control airpower, free from the dependence
on a Navy ship.
References 149, 150 and 197-201 provide comprehensive reviews on the new-type of
powerplants proposed for these aircraft. Fig. 7 shows the location of the engines and nozzles,
while Fig. 8 shows some additional details.

B-5 Supersonic STOVL Aircraft: The NASA Programs.

Wind tunnel tests using a full-scale model of the General Dynamics E-7 supersonic STOVL air-
craft design, fitted with a de Havilland ejector system, have been conducted in 1988 in the
NASA-Ames 40x 80-ft. wind tunnel. The nonflying E-7 model is equipped with a Rolls-Royce
Spey engine [cf. Figs. 6, 7, and 8].
In an arrangement unique to the E-7, engine fan flow is collected and fed through a plenum to
either a forward or aft duct. The forward duct directs the flow to an ejector augmentor in the
aircraft's wing, while flow directed to the aft duct is expelled through a thrust nozzle located in
the rear of the aircraft. Butterfly valves control the direction of the flow.
Appendix B 199

After the wind tunnel work at NASA-Ames is completed, full-scale ejector system compo-
nents will be tested on the NASA-Lewis powered lift facility. The facility was opened in late
1986 to support vertical lift propulsion tests that will be required under the joint US/UK
advanced STOVL program. The US/UK program is assessing several candidate propulsion tech-
nologies and airframe concepts that could lead to the development of a supersonic STOVL dem-
onstrator aircraft in the mid-I 990s.
The powered lift facility includes a triangular frame with 30-ft. long sides. The sides are sup-
ported 15 ft. off the ground by load cells that provide a six-component force measuring system.
The facility can measure vertical forces up to 20,000 lb., axial forces up to 30,000 lb. and lateral
forces up to 5,000 lb. in positive and negative directions.

Fig. 5. A variable stream control engine.

B-6 Ejector Tests: The NASA Programs.

Additional tests conducted at the NASA-Lewis powered lift facility were supportive of the
ejector approach. Designed to evaluate the thrust performance of the ejector system, these tests
indicated that the design requirements for a STOVL aircraft would be exceeded by the de
Havilland configuration.
The model is to be modified to incorporate a two dimensional converging/diverging aft nozzle
and a ventral nozzle that will provide vertical thrust.

Cj~
~ H " I\lflllt Spay Mk.202
. _~~l.I"', _::

· ~RBI99

C~~XG40
[ctm~q; [ACME
Fig. 6. Trends in British Powerplants.
200 Powerplant Technology Limits

The US and the United Kingdom are also studying airframes and powerplants based on the
remote augmentator fan, vectored thrust, ejector augmentor and tandem fan propulsion con-
cepts (Fig. 8). Related to this venture is a NASA and US Defense Dept. study. of lift/cruise
systems.
Proper propulsion design is clearly the key to a supersonic STOVL aircraft, and much of the
work performed under various agreements and contracts is aimed at developing such advanced
propulsion technologies.
Significant supersonic STOVL propulsion technology issues can be grouped into several areas
that include [cf. Fig. 8]:
• Development of high thrust-to-weight ratio engine cores with sufficient bleed capacity for
supersonic STOVL aircraft attitude-control systems.
• Reduction/avoidance of hot gas ingestion (Fig. I, Appendix D).
• Development of integrated flight/propulsion controls to ease pilot workload during takeoff,
transition and landing (Appendix F).
• Development of supersonic inlets that can be used in high angle-of-attack attitudes and
low-speed situations; thrust vectoring nozzles, and efficient low-loss ducts, valves and
collectors (Appendix F).
Under investigation are low-loss fan collectors, valves and ducting; hot gas ingestion avoid-
ance; short diffusers and supersonic inlets with high angle-of-attack capability, and integrated
flight/propulsion controls. Programs to explore thrust-deflecting and vectoring nozzles and
thrust augmentation are also in the planning stages (Lecture V).

B-7 Thrust Vectoring Control of Slow-Moving Airships

Design concepts for thrust vectoring of very slow-moving airships may serve as a simplified sim-
ulation for vectored-RPVs, or for VTOL-vectored aircraft at vertical takeoffs. Nagabhushan and
Faiss (192) have used a simulation using two ducted fans mounted one on either side of the air-
ship, each having the capability of tilting in pitch and roll to give vertical and lateral thrust for
control. An auxiliary thruster at the bow or stem of the airship, which augments its directional
control, was investigated.
It has been found that the tiltable ducted fans provide the airship with greater operational flex-

,\
\ ,
\
H REAR NOZZU ~

Fig. 7. The evolution of concepts of a STOVL fighter aircraft - (general lay-out of a delta-Canard and a
forward swept wing aircraft with canard (149, 150)).
Appendix B 201

ibility, especially during takeoff and landing. Thrust vectoring to provide roll control was found
to be effective while ground handling. The bow/stern thruster was found to give excellent direc-
tional control, which significantly improved lateral maneuverability. Thrust reversal and vec-
toring rates were found to be important design parameters.

B-8 Advanced Compressors

Compressor stall margins (cf. Fig. F-6) are expected to increase substantially to afford high-a-J3
maneuvers without engine stall/flame out. This increased margin will be combined with
advanced IFPC systems (Fig. F-7), new PST-inlets and new thrust vectoring systems as described
in the main text. Advanced vectored engines may also incorporate rectangular cross-section
afterburner [cf., eg., Fig. 1-13, 14].
These new compressors incorporate tip clearance control (cf. Table I) and low-aspect ratio
blades to gain beneficial compressor margins and characteristics without adding to the compres-
sor weight.

~ PERIPHERAL ,- A1V-ANNl,IlAR INVERTER VAlV£

'~L!T ~/ ,'- INTUI Duer \ :'t:z.:.:n

~ I
LfROH1 r~
.,
-.
.
\
\
-
'- fRONT FAN I'»IAU'ST PUNU'l1
1

.O
'- FROHT fAN NOlllL ATTAQiH(HI !lWOI

OMOER
OPOl~
A,= l 9 l
8 ~ ~9
8 A

,", .-0 swnOl A + I

AIV · TROHI AIV

engine schematic with periphelCll Inlet and annular Inverter valve.

< PERIPtirRAl ,"UT r RI}.R fAN IHUT PORTS

\ .. INITROUCI / ROIATING SUII'[ ClOSllt[ .. AOOl TYPE
'\ \ I ,- R£AIUAN \ NOUU

~
\.'~
--- \
~~
"\
~

\ \ '~RADl.\l Vol//[ BlOCm MRTICAU

fRONT fAN.J '- fROHT TAN NOZZIL ATTACllll.!NI ClWO)
SUMD a.OSlitE

~
;rPORTS

0° =
SlElVlVAJ.V!

.Fig. 8 The evolution of propulsion system concepts for STOVL fighter (149, ISO, 197-2011. For the
geometry of the ADEN see Fig. 5, Appendix A.
202 Powerplant Technology Limits

H'ClI Lav'AIS
OtaOl1U oue I
~~~
~~~
~~ lON
',,"onu:

~'''~
m['IUIUI,"e IIUlTU'IDoCOMP'E~50R
1 ~UI~""C CRU'S! S, ,(."CAL WillING • ~IOVlONl.YI
J8EllurlNKomprenor lift engln. propulsion modes lor super ..

~=::
AnOlIUKNINC TURe.OFMIAALS
.onl< SIOVL

lURIO~ ... tSSOI

.lI1ERIUlUlltlC
IURaO ·C\IMPR(SIOR

I, RUHWAuCaURAIKIN DRY. eot.IIIAI, SuPlRl""C caU'~1

~
DRY 1I(lIlUA.Q·ct.MP.(S~OR
-110Vl ANO caul5l1
. . ,v,tAlUR' ote
~~ ,) .. a!-;:I~·>""""',
11 ,• TT II
1, 1In-on
.. ~ I
~Wt--'-i;tit • t

1 5UI5"''' caUlI! \, VI.,ICAl VJlD,"C

TBEliurbo-Comprfssor lilt Jnd cruise engine propulsion mcdu
ror supersonic STOVL

l. SlIOIlIA1<£Cft
PAtAlLtl. - WnH All

t 14 ACctlIAAlIOH,
S....IIISO!1IC CRUISE, COMMT ~ V!llneAl lAND .....
SlIIlu· Willi "I 'ARAWl. · ORV
Serlts/PUBllel Tandem fan (SPTFJ propulsion for supersonic STOVL

AOtN IYP( NOtZu: 7

/

\,
I \ ',- INI[tDUC~
:
I
I \
'- RADIAL 'IIlV.( IlOCKf:R
-IRONl NOlIL[ UUlOI
ATlACK.\\UfI, VMl.D OR
SlEtVlD nos ...,
RALS Example Series/Parallel Tandem Fan (SPTF) In top Inlet cnnfigural!on.

Fig. 8 (continued)
Appendix B 203

rnboa rds Seclions

compared to
L +UC

L+UC ~-U#4f@@k+B< I

.JA S,etA A

IB
RALS ~~.ERT=
..J B S.et B A

"'l e
REX
~
... Je S'tl e A

-
Fan bleed (RALS)
....
~. ~~~'~~~.~_M'
\
flOl."
s;~-r ' --
~ ...i4-=-
...........::::-' "''':u·~~ • ./ t - - Hybrid (Iih + liIVcruise)
NoU" 1;t. 1DdI.1 ' ~...,

Fig. 8 (continued)

B-9 Current-Technology Limits and Trends

Much has been written on the advantages of the vectoring/reversing nozzles over current-
production nozzles. References I to 14 give general information and reviews concerning the
design and propulsion integration of various thrust vectoring schemes in fighter applica-
tions.
204 Powerplant Technology Limits

General aerodynamic considerations, including the effects of deflected thrust on the aerody-
namic characteristics of canard configurations and vertical tail loads, in single and twin engine
aircraft are covered in references 15 to 31.
Performance characteristics of various 2-D nozzles are dealt with in references 32 to 42, with
reference 42 devoted to the expected performance of the F-18 fighter. The performance of vari-
ous "ADEN" and "SERN" nozzles is given in references 43 to 47, with 47 devoted to the
expected performance of a forward swept-wing fighter utilizing thrust vectoring.
Altitude performance ofturbojet engines equipped with such nozzles is described in references
48 to 50, while reference 51 deals with the F-100 engine, and 52 with yaw vectoring.
A comprehensive comparison between different types of nozzles may be found in references
53 to 69, while engine integration with advanced fighter aircraft is covered in detail in references
70 to 80. Thus, F-15 integration studies are available, among other places, in references 70 to 71.
The mechanics offlight, as well asflight and engine control considerations, are dealt with in ref-
erences 81 to 87, while exhaust plume thermodynamic effects. infrared characteristics and signa-
tures are covered in references 88 to 92.
Computer programs and mathematical models associated with 2-D vectoring/reversing noz-
zles, especially with flow characteristics, are to be found in references 93 to 101.

B-9.1 Current-Technology Cooling Requirements

A major problem in the production of current nozzles is the cooling requirements of the
converging-diverging flaps and sidewalls of optimal nozzle configurations (cf. Lecture I). While
reference 102 describes the effects of shocks on film cooling of a full-scale turbojet exhaust noz-
zle having an external expansion surface, reference 103 is dealing with the temperature fields to
be found in such a nozzle, in subsonic and supersonic performance (cf. Lecture III).

Fig. 9. F-18 fighter with current-production nozzles.

Figs. 9 and 10 show the F-18 fighter with current and pitch-only vectoring nozzles,
respectively.
Being a non-symmetric flow channel, one expects the 2-D nozzle to develop non-uniformities
of the temperature field in mid flap, as compared to the nozzle corners. This is clearly indicated
in our recent test results (Lecture III). Moreover, C-D angles must be optimized to minimize
flow-separation phenomena (which cause temperature and thrust distortion).
Appendix B 205

Fig. 10. Pitch-only, 2-D, C-D nozzle, as planned for the F-18 fighter (Capone and Berrier, 56).

B-9.2 Current-Technology Engine Trends

In Table I, a comparison is made between F-lOO, F-404, F-IIO, PW-1128, F-118, and the
YF-120, and YF-II9 of the new ATF. A number of long-standing trends in engine R&D and
design methodologies can be drawn from this Table. These are enumerated below. Further
details are given in §B.ll.

Thrust to Weight Ratio Trends

According to Table I, the maximum thrust to weight ratios is expected to increase from 9 to 20
and beyond.

By-Pass Ratio Trends

By-pass ratios in fighter engines tend to be reduced, thereby transforming fighter turbofans into
"leaky turbo-jets" [Turbofans may be less suited to supersonic flight than turbo-jets, and, with
relatively large by-pass ratios, say 0.7: I, are more susceptible to the well-known "stagnation-
stall" phenomena, especially during AlB relight at high altitude, low Mach numbers and high
rates of turn]. However, V.c. engines bypass the bypass problem (Fig. 3).

Pressure Ratio and Turbine Temperature Trends in Current-Technology Fighter

Engines.
Engine pressure ratios remain almost constant, while Turbine Inlet Temperatures (TIT) keep
increasing. [Air pressure and temperature at inlet end increase with flight Mach number. Hence,
when mUltiplied by the compressor, they rapidly approach engine pressure-temperature limits,
whereby fuel flow must be cut back to prevent excessive TIT. This fuel cut-back limits fighter
performance speed and the thrust to weight ratio of a given engine design, unless, of course, its
hot part sustains higher TIT values. Higher thermodynamic efficiencies, combined with these
considerations, would, therefore, dictate high TIT values. The maximum TIT target for the ATF
engine may, therefore, be a few hundreds degrees F higher than that in the current engines,
thereby requiring the use of further improved blades and disks - see below.]
206 Powerplant Technology Limits

B-9.3 Current-Technology Integrated FADEC-Aircraft Controls

Fighter engines continue to change from hydromechanical control systems to various FADEC
systems (Full Authority Digital Electronic Control). Unlike the hydromechanical control systems
these schemes may be fully integrated with aircraft flight controls.
This trend is strongly affected by the increased number of engine variables to be controlled,
(especially in the new types of geometrically-variable engines) and by the increased capability of
FADEC systems to provide faster (stall-free) accelerations, lower SFC values, safer operation,
and higher degrees of maintainability and adaptability to engine failure or degeneration.
Fully-integrated FADEC-cockpit computer systems provide the best solution for advanced
fighters equipped with vectoring/reversing nozzles. They free the future fighter pilot from vari-
ous enginelflight working loads associated with current fighters during take-offs, cruise, combat
and landing. (For future trends see Appendix F: IFPC).

TABLE 1. Current-Technology Fighter Engine Trends (119]. (see also § 8.11 r

Current production engines Engines undergoing tests
Vectored Engines
Engine Designation F-IOO* F-404** F-IIO** PWl120 [GE37: YF-120.
(PW 220- F-118*· PW5000: YF-119J
229)
Engine Manufacturer PWA GE GE PWA PWA GE & PWA
Major applications F-15 F-18 F-16 ATF.ATFN
F-16 F-20 B-2 F-4?
117A F-15
Year of introduction 1972 1982 1985/8 1987 I 990s I 990s
Type T-F T-F T-F "leaky" T-J T-F 2-D nozzles. v.c. (GE)
TIW maximum (thrust to
weight ratio) [year] 7.8 7.5 7.3 7.23 8.5
Bypass ratio 0.69 0.35 0.78 0.15-0.19 -
SFC (specific fuel consump-
tion: intermediate (dry» 0.72 0.67 0.8 0.85
SFC at Max AlB 2.17 2.08 1.86 2.00
TIT(Maximum turbine inlet
temperature). F" [year] 2595 2530 2682 >2550 >2600 <3230 [4000/2000J
Turbine blade type [yearJ DS SC SC DS SC ([RSR/RCC/2000j)
PR (pressure ratio 24 25 29 27 27
Thrust (lb.): intermediate
1'-0,
SIL (dry) 14700 11000 19000 13570 16720 35000
Max AlB, T-O, SIL 23830- 16000+ 27500- 20600 -32000
29000 19000 29080
Nominal Airflow, Iblsec 225 140 254 178
Weight (lb.), dry 3040 2136 3830 2848 3225
Control System EECI H-M+ ACC& ACC& ACC& ACC & FADEC
FADEC Elect. H-M/E DEEC DEEC & IFPC
Includes a vectoring nozzle
Max. diameter, in
No···
46.5
No···
34.8
No···
46.5
No··*
40.2 47
yes

Length 191.2 158.8 181.9 162 209

+ All data presented have been accumulated from unclassified US public releases.
* Current production engines are given here for a comparison with the new fighter engine. F-100-PW-229 is an improved
version of the FIOO engine .
•• RM 12 is a derivative of the F404. It powers the Swedish SaabJAS 39 Grippen fighter. FII0-GE-129 is the improved suc-
cessorto the FII O-GE-I 00 used on the F-16 fighters. It is flight-tested on an F-15E. F-118 powers the B-2. It is based on the
FIOI/IIO core (cf. B-II). F404-F5D2 is to power the A-12 (ATA) .
••• Existing axisymmetric exhaust nozzles can, nevertheless, be replaced by Axi-TV nozzles, or by advanced 2-D vectoring!
reversing nozzles as stressed in the text.
SC = Single Crystal.
ACC = Active Clearance Control. RSR = Rapid Solidification Rate
DEEC = Digital Engine Electronic Control RCC = Reinforced Carbon-Carbon
FADEC = Full Authority Digital Engine Control
DS = Directionally Solidified
Appendix B 207

DEEC and FADEC abilities to diagnose engine problems are now being demonstrated in
actual service.
[For instance, a DEEC or a FADEC system may optimize engine performance with a failure in
engine fuel pump, or, in the nozzle, and allow the pilot to bring the aircraft safely to base.]

Turbine Blade Pyrometer Systems

Most engines use a set of thermocouples in the low-pressure turbine section to measure gas tem-
peratures. Then, using various schemes, the "calculated TIT" value is indicated and controlled.
General Electric, for instance, has recently introduced into the F-ll 0 engine an optical pyrom-
eter which directly monitors high-pressure turbine blade metal temperatures. However, the
blade metal temperature is monitored in the F-ll 0 only for exceedance, i.e., it is not governed as
a scheduled value.

Fig. II. Above: The ATF Engine. Below: The Post-ATF Engine is expected to double the T IW ratio (to
be more than 20:1) and to be much shorter than the ATF engine (as depicted in this comparative-size
scheme).
15 billion US dollars are expected to cause this change. Called Integrated High-Performance Tur-
bine Engine Technology (lHPTET), the program's time-table proceeds till the year 2003.
Note the increasing size of the nozzle in comparison with current-technology core-engines (cf.
Figs. 1-1,4,8, 12, 13 and 14, Table I, and § B.II).

Afterburner Light-Off Detectors (LOD)

An advanced AlB fuel management system, light-ofT detector, and a dual ignition system, have
been introduced by PWA into their PW-1128 engine. GE "LOD system", as used on its F-11O
engine, consists of an ultra-violet flame sensor, which views the innermost (flame-holder) "V"
ring in the afterburner duct.
To prevent hard AlB light-ofTs at high fuel flows during AlB ignition (which may cause fan
stall), the F-II 0 initially allows only a minimal flow rate of the AlB innermost "V" ring zone.
Only after the flame propagates around the innermost ring, does the LOD system allow higher
fuel flows to the afterburner. Hence, irrespective of the pilot's initial AlB throttle position, the
LOD system provides smooth AlB thrust increase.

Bypass Air Combustion Delays

The afterburner of the F-II 0, for intance, includes a scaled version of the F-I 0 I mixed flow type,
with convoluted flow mixer. Here, bypass air, and hot gas flows emerging from the turbines, mix
in the plane of the flame-holder, and 90% ofthe core flow is completely burned, before fuelling of
any bypass air is initiated. Such design methodologies provide improved altitude performance,
especially during AlB relight and acceleration.
208 Powerplant Technology Limits

Commonality
An increased degree of commonality with current-production engines is stressed by GE and
PWA.
For instance, the PW-1128 maintains about 80% parts commonality with the 1985 F-I 00 pro-
duction configuration.

Single Crystal Turbine Blades

Increased turbine durability, as well as improved cycle performance, are achieved by employing
single crystal turbine airfoil materials (superalloys), with advanced internal cooling
methodologies. .
Much has been written on these advanced cooling methodologies, and the most recent reviews
are available in No.1 and No.2 issues of the 1984 Volume of the Intern. J. o/Turbo and Jet
Engines.

Advanced Fracture Mechanics and Bearing Design

Gas turbine bearings operate at relatively high temperatures and rotational speeds. The new
trends in advanced aircraft engines require higher bearing speeds, which towards the next dec-
ade or so would probably reach 3 x 106 DN [bearing bore in mm multiplied by the shaft speed in
RPM].
Current production jet engine speeds are limited to 1.5-2.3 x 106DN (T-700, F-100, JT-9D,
F-IOI, F-404, CF6, etc.). [Jet lubrication is normally used to lubricate and cool the bearing mate-
rial. Here, the placement, direction and number of nozzles, jet velocity, lubricant type as well as
flow and removal rates into and out of the bearing and case, are all critical for obtaining safe
operation and long bearing life. Improved bearing structure, materials, coating, and hardness,
are also critical in increasing bearing reliability and load capacity. Even trace elements, nonme-
tallic inclusions, entrapped gases, and especially the processing method, affect bearing Iifespans
and reliability.]
Much research work is now being directed towards the improvement of fracture mechanics
methodology (see also a review by B.L. Koff of PWA in the aforementioned Journal volume on
materials and processes as well as in the pumping capacity of the bearing [which includes the
design of many radial holes in the race to cool and lubricate the bearing».

Durability and Survivability

Most technical design decisions associated with the development of advanced fighter engines are
based on the following categories:
l. Performance-including survivability, operability, observability and emissions.
2. Reliability, durability, maintainability, and safety.
3. Weight.
4. Cost (of both acquisition and ownership).
For instance, the durability development cycles in turbo engine design include:
I. Mission requirements definitions.
2. Duty cycle definitions (time vs. "Max", "Cruise" and "Idle" variations).
3. "Best" configuration selection (usually after many iterations and reviews involving critical
components).
4. Analysis and design.
5. Hardware manufacture.
6. Instrumented laboratory tests verifying the design intent.
Appendix B 209

7. Flight test evaluations.

8. Accelerated Mission Testing (AMT) and qualification.
9. Field data service evaluations.
The durability cycles are usually combined with such subcycles as Component Improved Pro-
grams (CIP), in which field data and proposals emerging from all formal (international) users are
combined to improve durability/maintainability/reliability/operability levels, as well as to
reduce ownership cost.
The trade-off between cost-effectiveness and durability/reliability/operability is certainly to
influence the development of the new ATF, and a whole new generation of ATF derivative
engines.

European Fighter Engines

A new (unvectored) European Fighter Aircraft (EFA) is now being planned by five European
nations (U.K., W. Germany, France, Italy and Spain). It may be propelled by the EJ200.
Early prototypes of the EFA will be propelled by the RB-199 engine. During these early phases
of the program the new EFA engine would be developed and tested. The new EFA engine may
also serve for re-engining the Tornado ADV fighter. However, the lack o/TV-maneuverability
may be a major cause to make the EFA obsolete.
Meanwhile France's Snecma is further developing its M-88 engine as a candidate future
fighter engine. [The M-88 has evolved from experience gained with the M-53 which propels the
Mirage 2000 fighter.] As the EJiOO, it is unvectored.
The unvectored EJ200 turbofan is based on RR-XG-40 demonstrator (Fig. 6), and is aiming at
TIW ratio of 10, a compression ratio of25, and a bypass ratio of approximately 0.4. In contrast
to the relatively-complex, 3-shaft RB199, the EJ200 is a 2-shaft engine with fewer parts and a
modular design for easier maintenance. It will include single-crystal turbine blades and
wide-chord fan blades. The powerplant will be the product of a multinational joint venture
known as Eurojet. It brings together RR, MTU, Fiat Aviazione and Sener.

B-IO Whole-System Engineering Approach [119]

More than ever before the new fighter engine design would require the "whole-system-
engineering" approach in engine and component R&D and production. It is required because of
the increasing need to integrate advanced engines with wing's design, aircraft flight control, 2D
nozzles and variable geometries, and, of course, with the aircraft structure.
Nevertheless, the methodology required for such a practice ha'!; not yet been fully developed.
Engine designers and manufacturers still tend to leave R&D of such systms to specialized bod-
ies and individuals. As a result, things tend to "fall between the chairs", with poor integration
and a resulting performance far below the optimal required.
Designing an engine, or an engine component, in isolation, ignoring these systems and their
drastic effects on aircraft and component design, performance/reliability/durability, is no longer
accepted around the world. Yet, the effects of these needs on component R&D and production
have not yet been fully understood.
Included in this approach are optimal and rapid integration of new results obtained in the var-
ious disciplines of engineering, and in materials science and computers. Hence, as TIT values are
increased beyond 3000F, new materials, new processes and new cooling methods are required. It
is, reasonable therefore, to expect continual government support in the development of critical
components such as yaw-pitch vectoring nozzles, FADEC-IFPC systems, RCC, and RSR (rapid
solidification rates) metallurgy. In fact, the list of subjects is unlimited. So there is no need to
specify here too closely how broad, or how narrow, this list should be at any given time in the
210 Powerplant Technology Limits

evolution of this technology. Instead, one may stress here only the need to develop the awareness
and the correct sensitivity to constantly re-educate ourselves towards increasing adaptations of
such practice.

B.II New Fighter Engines

The ATF Engine
As they currently stand, the design criteria for the thrust-vectoring ATF engine include a thrust
rating of 35,000 Ib, and 40% fewer parts than current-generation engines. The engine features
pitch-only, or pitch/reversal, partially-vectoring capabilities (Figs. 1-1,4, 5, 8 and 8 in the Intro-
duction). It is to sustain flight at Mach \.5, without recourse to afterburners. Nozzle materials
and cooling loads would thus become less demanding in comparison with current, AlB engines.
Moreover, if incorporated in later design versions of this engine, internal yaw-thrust-vectoring
vanes would have to sustain lower temperatures and use less cooling air than current AlB noz-
zles. In addition, the development of RCC materials for nozzle and AlB systems may remove
most, or all of the current cooling penalties (§ B.2).
The GE competing engine for the ATF is known as YF 120-GE-IOO (GE37). It began tests in
May 1987, and is scheduled to start flight testing in 1990. Thrust vectoring will be added later in
the flight program (as 2D-TV nozzles with low observables, or as Axi-TV nozzles). According to
GE the engine is of variable-cycle design, i.e., the bypass ratio may be set [relatively] high, for
subsonic, or low, for supersonic flights (cf. Table I and Fig. 3).
The PW competing engine is known as YF 119-PW-IOO (PW5000). It began tests in October
1986 (cf. Table I). The unvectored version will be flight tested in 1990.
The USAF expects first-production engines to be available by the mid-1990s. Post-ATF
Propulsion systems are, most likely, entirely different (see the main text).

The F-1l7A, B-2 and ATA (A-l2) Engines

Four, unreheated, FI18-GE-IOO turbofan engines propel the subsonic, stealth, "flying-wing",
Advanced Technology Bomber (cf. Fig. VI-I). These engines are believed to be unreheated
derivatives of the GE FlOl-FllO engine equipped with 2D nozzles.
The USAF F-117 A is believed to be powered by two General Electric F404 engines.
The other new (USN) fighter program, the subsonic, twin-engine, shipborne, attack aircraft,
the AT A, now known as A-12, has already been sewn up by GE. The firm will supply a derivative
version of its versatile F404 fighter engine. Another growth version of this engine, the RM 12, is
being developed by GE in association with Volvo Flygmotor to power Sweden's unvectored J-39
Gripen fighter.

The Up-Rated F-16/F-IS Engines

GE's F II 0-129, increased-performance fighter engine, has been recently flight-tested in a USAF
F-16C. Rated at 29,000 lb. thrust, the FI1O-129 is the successor to GE's FI1O-100/F-16 engine,
and is competing to power the F-16Cs, and F-15Es, scheduled for delivery to the USAF in 1991.
PW's FIOO-PWA229, increased-performance fighter engine, has been recently tested in a
USAF F-15. Together with GE-FI1O-129 it competes to powerthe F-16Cs, and the F-15Es, etc.

The Engines for the USM-AV-SB Harrier and the Rafale

The British RR Pegasus 11-61 (pitch-only) vectored engines, are to propel the USM-A V-8B Har-
rier II, while the new,' unvectored, French Snecma M88 engines are designed for a family of
common-core engines with thrust ratings of 7.5 to 10.5 tons, with overall compression ratio of
25 and bypass ratio ranges from 0.25 to 0.60. The first unvectored Rafale-D fighter aircraft pro-
Appendix B 211

totype is scheduled to be equipped with M88 engines. It is expected to be flight-tested in 1991.

(The present prototypes are propelled by the versatile F404-GE-400 engines.)

The MiG 29 and Su-27 Propulsion Systems

During full afterburning each of the two Tumansky R-33D produces 18,300 lb. of static thrust.
The engine nozzles are of contemporary, axisymmetric, variable, converging-diverging design.
Normal MiG-29 takeoff weight is 33,000 lb. (T/W> 1.0) while the maximum value is
39,000 lb. (T/W < 1.0).
The engine inlets have a multisegment ramp system that includes a hinged door, which closes
during takeoff, landing and ground operations to protect the engines from encountering FOD.
Simultaneously, overwing louvered vanes are opened to provide air to the engines. However,
additional air still enters the engines through a small gap just above each inlet upper lip. As the
aircraft reaches rotation speed, these doors/ramps are opened, i.e., rotate downward and the
overwing vanes are closed.
The Su-27 is powered by two Lyulka AL-3IF turbofan engines, rated at 27,5001bs each.
Thrust vectoring is current}.y added to these engines. (For a discussion of the "Pougachev's
Cobra" see Fig. 4 in the Introduction.)

F110-GE-100 Augmented Turbofan Engine

2 __ 12 STATOA SUPPORT fWISH8ONE)

I CENTERBOOY
'3 COMBUSTOR. CASE
• FAIl STATOA ,. COMBUSTOA
.. FAN ROTOR 1$ HPT NOZZLE
S FAN FRAME. ,s HPT SHROOO
IS ACCE.S$ORY 17 HPT ROTOA
" TURBINE_
~x
1 0\I1VI DUCT 11 LPT ROTOR llPT STATOR
• IotNER DUCTS
t COMPRESSOR STATOA - FWD >0 AUGMEHTOA
10 CON.PfESSOR ROTOR 2' E»t.AUST NOZZI..E
11 ccw.t.PAESSOR STATOR - AFt

Fig. II a. An example of current-technology, unvectored, fighter engine. See also § B.II and Table I.
212 Powerplant Technology Limits

B.12 THE NEW GE VARIABLE CYCLE ENGINE [See also the ATF engine in
parg. B-llJ
Very high exhaust jet velocities are required to maintain supersonic flight. For instance, the
exhaust jet velocity generated by the Concorde's Olympus engine is about 1800 ft/sec. However,
only about 1200 ft/sec may be recommended to reduce exhaust jet noise during takeoffs in popu-
lated areas.
The design approach to meet both high and low jet velocities requirements, as well as to
improve aeropropulsive efficiences and expand flight envelopes, is the Variable Cycle Engine
(VCE) which functions as a turbofan during subsonic mode of operation, and as a turbojet during
supersonic operation.
There are a number of methods to design VCE. A new GE-VCE design is shown in Fig. 3. It is
based on variable-area bypass injectors and a 'Coannular Acoustic Nozzle' [This nozzle may be
combined/replaced with a proper thrust vectoring nozzle for advanced, non-augmented, fighter
engines]. This GE-VCE functions as follows;

1. During the Subsonic Mode

In this mode the variable-area bypass injectors are open and fan cold-air bypasses the core
engine and, then, at the aft turbine section, injected into the core flow portion of the exhaust noz-
zle, emerging to the ambient air through an inner variable nozzle whose exposed geometry redi-
rects the flow in the axial direction [Fig. 3]. It is a variable area nozzle controlled by the
side-motion generated by the actuators in the far right of the drawing [inside the central nozzle
bluff-body]. The bypass ratio during this subsonic/turbofan mode may vary, say from 2 to 2.5.
An interesting phenomenon has been employed here: Out-directing the hot core gas flow and
in-directing the cold bypass air reduces noise [for the hot, external, less-dense, stream mixes bet-
ter with the internal cold air than the other way around]. The inner variable nozzle is therefore
referred to as 'Coannular Acoustic Nozzle' [Fig. 3]. Moreover, the internal cold air flow cools the
inner nozzle parts and reduces IR signatures.
For advanced military applications the 'Coannular Acoustic Nozzle' may be combined/
replaced with a proper [pitch-only, or pitch-yaw,or pitch-yaw-roll] thrust-vectoring nozzle. In any
design the new nozzles acount for about 50% of the overall length of the engine.

2 During Supersonic Cruise Mode

In this mode the [cold-air] bypass ducts are closed [see the vanes at the lower and upper aft tur-
bine section] and all fan air is forced through the core engine. The powerplant now functions as a
turbojet engine. Bypass ratio during this operational mode is, therefore, zero. The turbojet mode
is performed by closing the variable-area bypass injector 'doors'. Under these conditions the
'cold' flow through the 'Coannular Acoustic Nozzle' stops, while the external nozzle cross-
sectional area is being varied by engine controls to properly handle the hot gas efflux for optimal
conditions [Fig. 3]. Consequently, improved aeropropulsive efficiencies and expanded flight
envelopes characterizing turbojet engines are substantiated during this supersonic mode of
operation.

TURBINE BYPASS AND SUPERSONIC FAN ENGINES

Fig. 12 schematically shows two additional powerplants:
- The Turbine Bypass Engine [TBE].
- The Supersonic Fan Engine [SFE].
In the first type, during full-power conditions, a compressor discharge air flow is bypassed
Appendix B 213

Variable-cycle engine

~
(~ Turbine bypass engine

Fig. 12. VeE, Turbine Bypass and Supersonec Fan Engines.

around the high-pressure turbine and reintroduced so that it passes through the low-pressure tur-
bine. Thus, the HP turbine performs at near optimum speed throughout the throttle range,
which may significantly reduce SFC. This design concept was conceived by Boeing in the late
1970s and subsequently improved by PW A.
In the second type the long and heavy inlet system may be replaced/reduced by a supersonic
throughflow fan which, unlike conventional fans, can efficiently handle air at supersonic internal
flow regimes. This design may result in 25% lower engine weight and about 22% lower SFC val-
ues. The potential breakthrough of this design concept lies in the development of a reliable, low-
cost, supersonic throughflow fan and its integration with the core engine.
 APPENDIXC

DATA BASE-I:
VECTORING NOZZLE
INTERNAL/EXTERNAL PERFORMANCE
ec-l Vectoring Nozzle Internal Pedormance: Low NAR
Re and Leavit (28, 58, 112, 162) have used the NASA Langley static test facility to evaluate the
effects of geometric design parameters on 2D-CD nozzles at NPR up to 12 (see Figs. I to 7).
The (machined) nozzles had thrust deflection angles varying from 0° to 20.26°, throat aspect
ratios varying from 2.012 to 7.612, throat radii from sharp corners to 2.738 cm., expansion
ratios from 1.089 to 1.797, and various sidewall lengths (cf., e.g., Fig. I).
The results of this investigation (cf. Figs. 2 to 8) indicate that two-dimensional, low NAR
convergent-divergent nozzles have static internal performance comparable to axisymmetric
nozzles with similar expansion ratios. Nozzle expansion flap curvature (radius) at the throat had

I - - - - '•. u - - - - - I

~_._.~-.-•.-.~---r--.~I
.Fig. 1 Vectored-nozzle geometry as investigated by Re and Leavit of NASA (28, 58, 112, 162). Their
results are shown also in Figs. 2 to 7. All dimensions are in centimeters unless otherwise indicated.

AR = 3.696 and 7.612.

215
216 Vectoring Internal/External Nozzle Performance

C......' ..1on
U.........
AI'" '..... S_II
S_II

"."
L""'h_.... I,. C_ ',0 ca ."U' ca ' •. ~ eM t.. c. hI' c- ...... ., ...
, ,
J,7ii
MYIO 1.1110 U. SI LClIIO lUll - 11.l1li 11.601 1.6110 1541 .... IUO
ANIO 51 Uti

ANIO SJ .11' Uti 1t1'

""'10 51 .m 1.2,. -
AIVIJ LIM 1111 SI LOIIO
II."' - 11.1" I.tll -4,.,
ANIJ 52 I I 1,.,
AJIIIJ SJ ••11 Uti UJI

-
,
""'IJ 51 .m '.m
Alva LIllO .a SI 1.0lI0 lUll - lI.m .11' SlS.
AMI 52 f 1,.,
AMI 5J .11' .:., 1m
~IO 51 .m '.m -
AR • 2.012.

Sidewall
Unvectored Length Un'eelored
Contigurtltlon AR 6,. d"l xt' em Xs' em Xe, ualld Xe, d' em he -kd, em PU' d"l Pd' d"l
A/At ~
xe -xt
- 11.557 3.967 1.28 10.92
DIlV5 3.696 1.443 4.82 1.000 5.719 11.557

t -- C"-T :-;.239 T I
~
DI2V5 ~ I .253
+ T
f5V5 7.612 T 4.84 1.000 8.89D 11.557 ~ 1.926 1.33 11.00

AR = 3.696 and 7.612.

Fig. 1 cont.
Appendix C 217

little effect on the thrust coefficient, but the discharge coefficient, CD8 , decreased by as much as
3.5 percent when the radius was reduced to zero (sharp throat). Nozzle throat aspect ratio, NAR,
had little effect on thrust coefficient over the range of nozzle pressure ratio tested. A nozzle geo-
metrically vectored at angles up to 20.26° turned the flow at least as much as the design vector
angle, once nozzle pressure ratio was high enough to eliminate separation on the lower expan-
sion surface.

ec-2 The Limitations of Wing Maneuver Devices and Vectoring In Supersonic

Fighter Aircraft
Thrust vectoring technology may require incorporation of auxiliary trimming devices such as
canard surfaces, strakes, or nose jets (cf. Figs.lI-la to Id).
Aerodynamic improvements may be achieved through the use of wing maneuver devices such
as drooped leading-edge and trailing-edge flaps.
Vortex flow control may also be considered in the design of advanced fighter aircraft.
However, the interaction between these various technologies may negate individual perform-
ance gains.

1.00r------..----r-----.

.96~
Ctg .92
,8S
.84'--_ _..L.-_ _- ' -_ _...J

c 1.001
08 .96!---===~.
1 5
====:;:.9
NPR
==-J 13

Fig. 2. Re and Leavit's test results show variation of nozzle thrust and discharge coefficients with nozzle
pressure ratio for AR = 5.8 nozzle with Ae lA, = 1.09. (Cf. Fig. Ill-I I a.)

1.00r----r----.----......

.96

.84L....---~ _ __'__ ___J

100r-----~----- __

C08.961~1-==5~====9;::·::-=-~IJ
NPR
• Fig. 3. Variation of nozzle thrust and discharge coefficient with nozzle pressure ratio for AR =7.6
nozzle with A.IA,= 1.09 (Re and Leavit, NASA).
218 Vectoring Internal/External Nozzle Performance

l.00r----.r----=:::::::====~

.96

.92
Cfg
.88
.84L-_ _'---_ _"""---_ _...I
l00r-----------~------~

CO~96Lll..=:::;:===::;:::===-11
5 9 13
NPR
.Fig.4. Variation of nozzle thrust and discharge coefficients with nozzle pressure ratio for AR = 5.0
nozzle with A.IA,= 1.8 (Re and Leavit, NASA).

1.001~ .
.96 ,2,3 ~

Cfg .92
.S6
.84'--_ _~_ __I1~_..-J

00
COS 1.961
.
r======:=4r-:::::::i1
t 5 9
I-
~~~~~~J
-i---=F=-::::-=:-rll':i"- 13 1 5 9 13
NPR NPR
.Fig. 5. Effect of throat aspect ratio on variation of nozzle thrust and discharge coefficients with nozzle
pressure ratio for low and high expansion (cf. Fig.III-lla, Re and Leavit, NASA).
Key: Left Figure: AeIA,= 1.09. Right Figure: AeIAt=1.8. I)AR = 3.7; 2)AR = 5.8; 3)AR = 7.6

.84
1.00
-" . -
.\..4
.. ----.:.:...~31
: :*4~2
--
~
2r::-=:=-. -
COB .961 11 Cl

,
.92!-~-~-~-~~
3 5 7 9 11
. ,!,
357
,
9 11
NPR NPR
.Fig. 6. Effect of throat radius on the variation of nozzle thrust and discharge coefficients for low and
high expansion ratio nozzles. [AeIA,= 1.09 (left figure) and AeIA, = 1.8 (right figure), AR = 3.7).
Key: I) R t = 0; 2) R,= 0.68 cm; 3) R t = 1.59 em; 4) R t = 2.74 em. (Re and Leavit, NASA).
Appendix C 219

Low~r flap

P
PT7

1.8 .6 1.8

Fig. 7. Effect ofnozzle nap angle on nap centerline pressure distribution with full-length sidewalls (Re
and Leavit, NASA).
Key:
XIXjdimensionless distance along flap walls.
I) Pressure distribution on flaps with no vectoring (0,.= 0).
2) 0.- 9.8°, AeIA/= 1.3, NPR = 5.
3) 0.- 13.2°, AeIA/- 1.17, NPR = 5.
4) 0.= 20.3°, AeIA/= 1.3, NPR = 5.
(cf. Fig. 1-6).
--INT
-----MAXAID

0.99
._-::=--0.::::---------.
0.98

e,y 0.97 .

0.96

U.9& ' ! , I

1.0 2.0 3.0 4.0 S.O 6.0 7.0 8.0

NPR

0.99 _ _ _ INT
- - - MAXA/O

0.98

e'g 0.97
-..------- ...

0.96

0.95,.0 =':2:0=----:3~.0---:4:'::O--:5;":.0=---:6'O-=---:;7:'::~O--;:'B.O
NeR
.Fig. 8. Variation orthe GE F-404 axisymmetric nozzle thrust coefficient efg with NPR for static, S.L.
conditions, at "Intermediate" (INT) and maximum afterburning (Max A IB) power settings. The lower
graph represents similar tests but with the 2D ADEN nozzle shown in Fig. A-4.
220 Vectoring Internal/External Nozzle Performance

_ _ _ _ _ 0 ... 11", Axl, TV/fA

_ _ _ _ M3J!i2·0C·D
_ _ _ _ _ "" .22·0C·O • all.. l1n_ a;lll b. .-A 1.2 2·0 C·O
_ _ _ _ _ JllR12·DC·O o TVfTR IXI 0 oM 7.0 ' ·0 C·O
_ _ _ _ _ _ M1SEAN o IIR 3.6 2·0 C·O .. JIA 1,0 SEAN

1,02 0 ,024

1.00

-~
--
0.022
C 0.98
~~ r"-- 1--
fg
0 .96 ~ ~ t-...
~\
""-
~ ~ :-..
~
\.., ....
0 ,020
Y
~
....;::
u.,"" ...-:::
O.~

0.92

0.90
I""- [,-Subionic rru;" '"
0.015
0.88
0.028
1.02
(

1.00 0.02S

CO.98
J..- \
fg
0.06
I
1/ ~~,
/ J .~~~~~ ~
o.~
~\- >-
0.92
I
i SYpelionle Ofl~-""" 0.022
'-
--"0
-....,
I
0,90

II
0.020
3
0 .88 NPR
I 1 11 13 15
NPR

.Fig. 9. Hiley and Bower's test results show the effect of AR on CII for an AR range between 3.6 and 7.2
(11). The experimental results shown on the right-hand side are for a total configuration installed drag
for unvectoredjlight with various vectoring nozzles. The results depicted show that the drag values for
2D-CD nozzles are less than those of current-technology nozzles. However, this is probably not a
general rule.

- "T'"-- -- ______ .....,,........, I • .,

......

.. ,.."
.Fig. 10. NASA nacelle/nozzle installation. All dimensions are in centimeters [Capone, \983].
Appendix C 221

.Fig. II. Body arrangement and internal flow hardware in the NASA tests [Capone, 1983).

'.'111'

'. "W'

Fig. 12. The 2D-CD nozzle configurations in the NASA tests [Capone, 1983).

To evaluate such interactions Capone, of the NASA Langley Research Center, has conducted
the following instructive study [1983]:
The wing maneuver devices consisted of a drooped leading edge and a trailing-edge flap.
Thrust vectoring was accomplished with 2D-CD nozzles, located below the wing in two single-
engine podded nacelles. A canard was utilized for trim pitch-only. Thrust vector angles of 0° , 15°,
and 30° were evaluated in combination with a drooped wing leading edge and with wing trailing-
edge flap deflections up to 30°. The investigation was conducted at Mach numbers from 0.60 to
1.20, at angles of attack from 0° to 20°, and at nozzle pressure ratios from about I (jet om to 10.
Reynolds number based on mean aerodynamic chord varied from 9.24x 106 to 10.56x 106 .
The mutual interference effect of deployment ofthe drooped leading edge in conjunction with
thrust vectoring was beneficial to untrimmed drag-min us-thrust polars, because an additional
222 Vectoring Internal/External Nozzle Performance

drag reduction was obtained that was greater than the sum ofthe individual drag reductions due
separately to either the drooped leading edge or vectoring. However, deflection of the trailing-
edge flap in combination with the drooped leading edge and thrust vectoring caused an unex-
pected increase in incremental interference drag! The configuration with 15° pitch-only thrust
vectoring, the drooped wing leading edge, and 7° wing trailing-edge flap deflection had the best
performance at trimmed maneuver conditions. At 30° wing trailing-edge flap deflection, large
trim drag increments degraded the performance of this configutation, although it had the best
untrimmed performance.
The wind-tunnel models employed by Capone are shown in Figs. 10 to 12.
 APPENDIXD

DATA BASE - 2:
SYNERGISTIC EFFECTS
D-l Synergistic Effects of Thrust-Reversal
D-l.l Thrust Reversal Effects in Landing Procedures
Some aircraft make a constant AoA approach until touch-down, while others have to flare to
avoid an undercarriage disaster.
Thus, high lift, deceleration devices, and undercarriage strength, must be designed and
adjusted to the landing requirements.
Deceleration devices include thrust-reversal and aerodynamic braking. However, during the
ground roll the aerodynamic brakes are ineffective and at low landing speeds they become
unsafe.
Consequently, optimum use of thrust reversal may become an important design criterion in
the development ofSTOL aircraft, especially when short field landing techniques are required at
the low approach speeds and steep glidepath which characterize vectored aircraft. (See, however,
our reservations in Figs. 12 and 21, Introduction.)
The synergistic effects of thrust-reversal also include high vs. low pilot's work loads, and hot-
exhaust-gas reingestion as depicted in Fig. I. This figure demonstrates the interaction between
forward speed and the ground-reflected hot exhaust gases streaming from the lower part of the
thrust-reverser port to the ground.

Fig. I. Hot-gas reingestion through engine inlets is caused by ground effects in improper use of TR dur-
ing landing. Hence, one must shut-down TR below a given ground speed. Other asymmetric effects
caused by the upper and lower hot jets are discussed in the text and depicted in Figs. 10 to 12.

224
Appendix D 225

J.O

T
v 0

~ >.>

, , , ! ,

Time, sec.
, I , ! , ! ,

....,'"
~
'-
a. o Forward Thrusl
...
7.0
i 6 Full Reversp.

t,
,f,'"" )(')0"

Fig. 2. Typical pitch vectoring response times (51).

" Corrected Flow

1,t)

~... 2.'

T
x 5
il
.....
o tJ: 2.0 OB/MHO",.
Il 20- Veclored
c:

,., '-,.I...,.._ _...I""___ ,"",,J...,

Time, sec.
~ Corrl!Cloo Flow

Fig. 3. (left) Typical reversed thrust response times (minus values of Tx represent TR) (51) .
• Fig. 4 . (right) Reversed thrust (top) and vectored thrust (below) stabilities are high as shown by the
depicted performance with bell mouth inlets and Tx. Ty full-transient variations lSI).

Actuator load, lbs.

Actuator load, lbs.

o~
-1;;~
(1)

-3000~ , .
100 ~20 -10 0 +10 +20
Reverse PO!Illlon, "
(Melill Pitch Thrust-Vectored Angle

.Fig. 5. The variations in the force loads on 2D-CD actuators during pitch vectoring (right) and thrust
reversing (left) are shown to be quite moderate. (I) Fixed load calculated from wall pressures. (2) Aver-
age actuators load (51).
226 Synergistic Effects

As far as the aircraft ground roll speed Vg is subject to the condition

Vg > Vreflected forward jet' (I)
the TR device may remain operative.
However, when

Vg ;;:;; Vreflected forward jet' (2)

the ground-reflected hot gases may enter the engine intake and cause engine stall, instability, or
flame-out. This situation must be avoided, and, consequently, TR must be shut-down before the
condition described by eq. (2) is reached. Moreover, the stagnation point generated just below
the aircraft inlets may cause foreign objects (and rain water) to be air-suspended from the ground
and hit engine compressor blades, causing "foreign-objects" damage.

0-1.2 Hot Gas Ingestion

A one-tenth scale, vectored thrust McDonnell Douglas 279-3 supersonic STOVL model was
used in a wind tunnel in 1987 to investigate the effect of ground proximity and airframe-
mounted flow diverters on hot-gas ingestion.
To permit a wider range ofhot-gas-reingestion investigations, NASA-Lewis has prepard a new
system which allows variations in model height, angle of attack, pitch, roll and yaw. The opera-
ting temperatures in this system may be as high as 1000F.
Replacing vertical TRjets by horizontal (yaw) TRjets has been proposed recently (179) for the
elimination of the ingestion problem.
Another design factor is the deployment time during thrust vectoring and reversing (see
below).

0-2 Deployment Times Limitations During Thrust Vectoring and Reversing.

I-second maximum deployment time for full-range pitch thrust vectoring, and for thrust-
reversing, has emerged as the minimal requirement. However, shorter deployment times may be
demonstrated by a few designs (179).
Fig. 2 shows typical pitch vectoring response times. Thus, the vertical force component T,.
may be changed from 0 to -20 degrees in I second time interval, and from -20 degrees to +20
degrees in 2 seconds (51).
Fig. 3 shows typical reversed thrust response (minus values of T x represent thrust-reversal).
Thus, the transition from full forward thrust to full thrust-reversing should not exceed I second.
Note that, due to inherent space limitations, the thrust reversal efficiency is normally less than
67%.

0-2.1 Fan Pressure Ratio and Actuator Loads Variations During Thrust
Vectoring.
Fig. 4 shows that thrust vectoring variations, and thrust-reversing, may have negligible effects
on fan pressure ratio (51). This is demonstrated by the negligible fan pressure ratio change for the
thrust-vectoring variations and thrust-reversing shown in Figs. 2 and 3. Here fan-pressure stabil-
ity is demonstrated by a performance comparison between bellmouth (unvectored) and 20
degrees vectoring deployment operations.
An example of the variations expected in the force loads on 2D-CD actuators is depicted in
Fig. 5.
Appendix 0 227

0-3 Nozzle Efficiency During Thrust-Reversal.

Fig. 6 demonstrates decreasing thrust-reversing efficiencies with increasing NPR values. This
phenomenon is almost diametrically opposed to forward-thrust-nozzle efficiency changes with
NPR, up to a critical value of NPR (cf. Fig. 13, Introduction).
These characteristics depend on NAR and on other nozzle variables. However, they clearly
demonstrate internal nozzle stagnation-point effects associated with thrust-reversal turning of
the flow (normally about 135 degrees).

1.0
Reverser stow~d

,6

(1g .2

-
Reverser deployed
-.2 .------
,/ ""
-.6 0
4 8 12
NPR
Fig. 6. Thrust-reversing efficiency variations (NASA, Capone and Berrier, Ref. 56). Note the decrease
in thrust-reversing efficiency with increasing NPR values. (These findings should be compared with
those shown in Figs. 7a and 7b.)

Unlike the decrease of CfgR with increasing values of NPR, and the low efficiency values of
thrust-reversal reported by Ref. 56 (Fig. 6), Ref. II reports much higher CfgR with increasing
NPR (cf. Fig. 7a). Thus, at NPR = 3.5, C(gR= -0.7, and is not less than -0.6 during the entire
range of subsonic-supersonic mode of operation. Obviously, the efficiency depends on nozzle
geometry and internal design factors, such as NAR.

-0.2

u.'f. -0.3
1 1
u y
~ fa-
l1]:
",,1-30·
l!
'0
!i:
-0.4
~~:rR
8 -0.5

I I
~
,
t=
.,e -0.6

~0 -0.7
.b.. t- r<-'"
M3.0 ~"
~1-45·

z 2'DC'D~
'-0.0
1 11 13 15 17
NPR

Fig. 7a. Unlike the drastic decrease of C,gR with NPR, and the low efficiency values of thrust-reversal
reported by Ref. 56 (Fig. 6), Hiley and Bowers 1111 report much higher efficiency values, and only a mod-
erate decrease of C,gR with increasing NPR values 111). (see also Fig. 7b.)
228 Synergistic Effects

It should be noted, however, that the comparison made in Fig. 7 between axisymmetric nozzle
and an AR = 3.6 2D-CD nozzle, is slightly misleading, for the former deflection angle is limited
to 30 degrees, while that of the latter to 45 degrees. Further experimental work may therefore be
required to finally assess this problem.

-0.3

1i -o.~

~
8
u
e
-0.5

z:
I-
:I -0.6
II
i;
a:
-0.7 , 5
L-J
NPR

Fig. 7b. The experimental data of this reference are more in line with those of Hiley and Bowers (l1J
(Fig. 7a) than with those reported by Ref. S6 (Fig. 6). The data are for a 2D-CD vectoring/reversing
nozzle (Stevens, Thayer and Fullerton, 49]. (See, however, the text.)

0.9

0.85

0.80

0.15 ""o-_....J._--~--~--~~-~ .
• Fig. 8. Variations of the discharge-coefficient during transition from axial thrust to full thrust-
reversing (Callahan, IS8J, at different power levels.

• D-4 CDS-Variation With Temperature and Thmst-Reversing

The definition of CD8 is given by equation 7 in Lecture III. Various experimental values of CD8
during forward thrust with low AR. 2D-CD nozzles (and different A.JA, ratios), are shown in
Appendix C-9, including the effect of c5..-vectoring on CD8 values.
As expected, the values of CD8 should be lower during thrust-reversing, mainly because of
boundary-layer thickening effects near, and around, the internal stagnation point. This is dem-
Appendix 0 229

0.90

COB
0.85 Temperalure
Prolil.

0.80 - BRP NPR

0
•I
0.78 2.2
0.78 2.2

0.75'
a 20 40 60 80 100
0/. TR
.Fig. 9. Effect of temperature on the discharge coefficient (Stevens, Thayer and Fullerton, 49). Non-
uniform temperature profiles reduce the values of the discharge coefficient during various deploy-
ments of thrust-reversing at a constant NPR value.

onstrated in Figs. 6 and 7. Figs. 8 and 9 show the effects ofthrust-reversingon CD8 , including the
actual behavior in non-uniform exhaust-gas temperature distributions.
These figures show that the values of CD8 decrease with increasing NPR values (higher per-
centage of "full reverse").

eD-5 Asymmetric, Aircraft Tail/Thrust-Reversal Effects

Northrop low-speed wind tunnel facilities had been employed by Glezer, Hughes and Hunt
(157), to evaluate thrust-reverser effects on the aircraft-tail-surface aerodynamics of an F-18-
type configuration at conditions representing approach phases.

PITCH·UP
DUE TO
ENTRAINMENT

PITCH·DOWN
DUE TO UPWASH
FROM VERTICALS

Fig. 10. Reverser-induced, asymmetric effects contribute to changes in longitudinal stability (Miller, 8).
These asymmetric, aerodynamic forces generate pitch-up and pitch down moments on the aircraft.
IFPC rules must, thercfore, incorporate proper responses to these effects (see also Figs. 1, 11, 11a
and 12).
230 Synergistic Effects

A major feature of these tests was a systematic buildup of reverser jets and tail surface compo-
nents to identify and understand the major aerodynamic forces which the reversers generate on
the aircraft-tail surfaces .

• Fig. II. Reverser-induced yawing moments on vertical tails for a = 0° (157) .

. / ..... - ..... 6 ...

u06 "",,'
". WING 80Dy-vERTlCAI.
- ,. INFERAIED FROM 'fAwMOMENTS
004 .frO- "
.... - - - - - . -.
I
oJ..
.t./ ......
'._._._. __ .-t-.... -.-.... -.
'I

80 100 120 140 Qf

"l"o
.Fig. Ila. Asymmetric pitching moment as a function of q-ratio. a= 0·, {3= O· [Glazer, Hughes and
Hunt, 157).

"" I ···~.1:,.,~ tf:'

V I
A(ROOVNAMtr.
FORCE
ZERO SIDESLIP - - - CROSSWIND
SIDESLIP
lEFT Rl(jHT

1 ~ INC~:~ENr
~ • AEROOYNAMIC
fORCE

RUDDER EFFECTIVENESS
AT ZEROSIOESLIP

.Fig. 12. Sketch of jet asymmetric interactions with aircraft vertical tails, rear views [Glazer, Hughes
and Hunt, 157).
Appendix D 231

It is shown that the upper jets may produce an aerodynamic "blockage" between the two l'erti-
cal tails (Fig. 10). The induced aerodynamic effects generate pitch-up and pitch-down moments.
which. in turn. change the longitudinal stability of the aircraft.
In sideslip flight. or with the rudders deflected. this "blockage" is asymmetric. and results in
increases in directional stability and rudder effectiveness.
The reverser jets may induce a strong "entrainment flow" (Fig. 10) on the horizontal tails.
which amplifies the tail load. resulting in either a pitch-up. or a pitch-down effect, depending on
the tail setting (cf, Fig. II).
However, installed in a twin-engined fighter, with a single vertical tail, the reversers may cause
smaller effects on longitudinal stability. Nevertheless, they still produce an incremental pitching
moment, changes in directional stability, an increase in horizontal tail effectiveness and a
decrease in vertical tail effectiveness.
Since the vertical tails in the F-18 fighter aircraft are canted 20 degrees outwards, the yaH'
moments measured on the individual vertical tails may be associated with a downward force,
and hence, with a pitchup moment (Fig. 10).

D-6 Swirling Flows in Fixed, Rectangular Engine Nozzles

An interesting experimental and numerical investigation was reported recently by Sobota and
Marble [177]. In their experimental studies the authors employed the blowerltlow-conditioners
shown in Fig. 13, while the numerical study has been based on the computational grid shown in
Fig. 17. Their interesting results are shown in Figs. 14 to 16 and 18 to 23.

Tu,n,ft9 Vone
AUBmbl:r

" low
Condlfion.'$

Fig. 13. The Experimental System of Sobota & Marble (177).

As expected, the results indicate that the flow-field between the turbine discharge, and the
nozzle exit. is highly influenced by the vorticity distribution at the turbine exhaust. Such infor-
mation may help the engine/nozzle designer to "tailor" the vortex distribution at the nozzle exist
by a proper design of the turbine discharge and the intervening ducting and flow dividers. It may
be of great interest to see the expansion of this pioneering work into the
variable-vectoring-nozzle-domaln.
Some of the Sobota-Marble conclusions are given below:
(i) - The use of steady-state, laminar, incompressible, Navier-Stokes equations, on a digital
computer. can adequately model the qualitative features of the flowfield (under a proper set of
operating conditions).
(ii) - In the low and high swirl cases, the majority of the vorticity may be introduced in the neigh-
232 Synergistic Effects

Fig. 14. The Cross-Flow Velocity Vector Plot, 15 Degrees Blade Angle (Measured) (1771.

Fig. 15. The Cross-Flow Velocity Vector Plot, 30 Degrees Blade Angle (Measured) (1771.

Fig. 16. The Cross-Flow Velocity Plot, Split Blades (Measured) [1771.

borhood of the blade root and the blade tip in the boundary-layers on the centerbody and the
outer wall, respectively. The axial vorticity introduced at the blade-root may be convected along
the centerbody to the downstream tip, and, then, along the centerline of the transition section.
This results in the formation of a strong vortex along the central axis of the duct. The pressure
gradient established on the outer wall of the duct, as a result of the strong central vortex, may
promote the separation of the boundary-layer from the outer-wall, into the main stream. This
separation is evident in the formation of two streamwise vortices, one on each side of the central
vortex, which have a sense opposite to the sense of the central vortex.
(iii) - In another test case, a set of blades that introduce swirl into only the outer-half of the inlet-
annulus is used. This, in effect, introduces a cylindrical sheet of stream wise vorticity into the
flow at the inlet, at a radial-position halfway between the centerbody and the outer-wall. There is
also axial vorticity of the opposite sense introduced into the boundary-layer ofthe outer wall. In
this case the separation of the boundary-layer on the outer wall does not occur to a significant
extent, but is apparent in the formation of two small, streamwise vortices in a diagonally-
opposed comers of the rectangular, exit-cross-section. The cylindrical vortex sheet, when con-
vected through the annular to rectangular transiton-section, is distorted in such a way as to
conform to the shape of the outer-wall ofthe duct. In the regions where the vortex sheet is most
distorted, it tends to roll up. The result is the formation ofthe two axial, streamwise vortices, at
the exit plane of the duct.
Appendix 0 233

Fig. 17. The Outer Surface and Exit Plane of the Computational Grid employed by Sobota and Marble
[177).

.. -......
fI .. -

I I •••• - - - -
', ......
. . . . . . . . . . . . . . . . . . . . . . . . ,. "

- - - - _ • • , ........ _
/'
".....
,

- .
., ., - . - - . .... ,......

Il" ~!;: ,~ " ," : : : : ~ : : : : ~ : : ::

: • : : :: : : : : : : :: :::: :~..I~~~\
~
~~"
'\. \ '
........ -.-- ..... ,,"
-- -

,
.............. _ ................. -

• , •••••• - •••••• , "

.... ,

••

.1fr";"I".I~#"".<II'~'/
t I , I , I I • I I • , , , ~ , , ,
," I
..... /;'rp."""''''''''''II''I'
r .. " ., " ,
..-_ _ .... I I

---r.----.-._.
P .... #' " , ,

I_" ..... ___ .. ___ • __ •••1 \,,\~~~.

.... , \ \ " .." " .. Ito. "" ......... " " 'iii , , ,
V I~' ..................
_.~ " " " ,. #' ". , , , '
'I
,

I ................................ - .. ~ ....................~ • ... , 1 \ \ "" ........ .p , . , , , \. " . . . . . . . . . , I

" .... _ .......................... - - _ .,...", .... - - _ til r , , .# . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .,

Fig. 18. The Cross-Flow Velocity Vector Plot, 15 Degrees Blade Angle (Computed) [177).

..
.. '''' , "" ....
-"'."""-'''''

Fig. 19. The Cross-Flow Velocity Vector Plot, Split Blades (Computed) [177).

~---
................ J
-::::7--+-
I A

10 2022 30 40 44
Siollo""

------ ........,
------ 4 - - - ---
.-.
~
~
'
B ----- ...... - B

Fig. 20. Sections and Stations in the Sobota/Marble Study (I 77).

234 Synergistic Effects

Mh .Ii ,j , ..... n l n • • • • · -

a) Station 32

, H \ ~:~ • • ••• •••••

b) Station 38

c) Station 44

Fig. 21. The Cross-Flow Velocity Vector Plots and Axial Vorticity Contour Plots, 15 Degrees Blade
Angle (177).

II" • .....-_ _ _ _,...-_ __ _ _

Rodl'lol'

A, •••
~
lL..--,.,..---,o !--- ---.J
a) Tangential velocity b) Axial Von icily

Fig. 22. Inlet Conditions, 15 Degrees Blade Angle (177).

Appendix 0 235

/i
Rlip r------=~----_, RIiPr-----..,.------::,..-,

i
i
i
Radiu. i
i
Radius

R,oo, '--_-:-_ _-'-_ _ _ _-1

+ 0 R,OO"'-----+:---7
0 -----...J
a) Tangential Velocity b) Axial Vorticity

Fig. 23. Inlet Conditions, Split Blades [177/.

(iv) - The vortex patterns created in the low and high swirl cases appear to be highly conducive
to the mixing of the rectangular jet, particularly in the high swirl cases. The adjacent positioning
of strong vortices of opposite sense may tend to draw the cold ambient fluid between them. This
large-scale mixing may rapidly reduce the size of the potential core-region of the jet (cf.
Ref. 178).
 APPENDIX E

TEMPERATURE DISTRIBUTIONS INSIDE

VECTORING NOZZLES
This Appendix briefly describes a test facility for studying the cooling requirements for
components of vectoring nozzles.
To investigate the cooling requirements of the nozzle flaps and of the sidewalls, one may
design and construct the 2D-CD nozzles shown in Lectures I to III. The first one (Fig. III-3),
may be operated without cooling for the sake of comparison. Its task may include the evaluation
of maximum temperature-fields in subsonic or in supersonic nozzle gas flows. Secondly, one
may have to repeat these tests with a properly-cooled nozzle.
Using such a methodology, Figure III-I shows the same nozzle with film-cooled slots. (Both
models have the same size and internal-flow geometry.)

.The Test Facility

Model size is designed to match the maximum flow rate of a GTC PS5-ISO gas turbine available
in the JPL. This turbine provides air bleeding from a two-stage centrifugal compressor, regulated
at up to I kg/s air-mass flow rate (at up to 3.5 bar).
The test facility and the relevant instrumentation are schematically depicted in Fig. I.
Exhaust gas conditions have been simulated by means of a combustion chamber [using a
combustor and an ignition system of the T56 turbo-prop engine]. Alternatively, the air-mass
flow may be supplied to the combustor from a centrifugal fan (8), through a large (stainless-steel)
heat exchanger (7), (with air heating up to 250·C for better fuel evaporation).
Chormel-alumel thermocouples are connected to a universal data logger (24), which, in tum,
is linked to an interactive graphic computer system for test-results acquisition, computation and
graphing. Remote TV-control of the tests is conducted from a well-isolated control room
(Fig. IY-2).
The wall temperatures, calculated as TwiTg. may be expressed as

LTC= T .. - TA
Tg-TA
where TA is the air-cooling average temperature.
Tests may be conducted with various thrust vectoring angles. (In the example shown in
Figs. III-I, 2a, 2b: 6,. = -7, 0, 16, 17 and 23 degrees. At subsonic flows the local Mach number was
varied from 0.07 to 0.22, and at supersonic flow conditions from 1.0 to 1.19).
Conclusions: The test results indicate that in the subsonic flow regime the temperature
distribution may be approximately two-dimensional, i.e., the main temperature variation is
alongside the nozzle central axis, while in the supersonic domain, a three-dimensional
temperature distribution may be developed. The throat section ofthe nozzle is the most critical
region, in which the temperatures are maximized during all thrust vectoring conditions. Under

237
238 Temperature Measurement

Fig. 1. Test Rig for the Evaluaton of Temperature Profiles During Thrust Vectoring (with and without
nozzle cooling).
1,8 Electric motor-driven centrifugal fans.
3, 17 Fuel pumps.
5,16 Ignition systems.
7 Stainless-steel heat exchanger.
2, 9, 10 Butterfly valves.
II GTC P85-180 gas turbine with air bleeding from a two-stage centrifugal
compressor.
6, 13, 18, 22 Three TV-systems (closed circuit).
12 Thrust vectoring/reversing nozzle.
19 Air-mass-flow orifice panel.
23 GTC-engine control panel.
14 Water cooling system.
20, 21 , 24,25,26 Temperature, pressure and fuel-flow data loggers & indicators.

subsonic conditions thrust-vectoring generates temperature differences between the lower and
upper divergent flaps, by generating higher temperatures on the [ower flap during "downward "
thrust-vectoring, while under supersonic conditions, the thrust-vectoring causes lower
temperatures on the "convex" flow-path of the divergent flap i.e., a downward thrust-vectoring
causes lower temperatures on the lower divergent flap.
Appendix E 239

Fig. 2. Details of the JPL data loggers, indicators and controls depicted in Fig. 1.
 APPENDIX F

LIMITING ENGINE-INLET ENVELOPES

AND IFPC
F.l The Yet Undefinable Matrix of PST-Intake Variables
Intakes of hot-jet-powered aircraft require minimum variations in airflow qualities at station 2
(compressor inlet). This means minimum variations from uniformity in composition, tempera-
ture, pressure and axial velocity distributions at station 2. (Water, rocket, or jet-engine exhaust
gas characteristics, are typical distortion generators, in addition to high-a-fJ maneuvers, etc.)
However, no real engine intake can achieve full uniformity in flight. Thus, each engine manufac-
turer must specify the maximum allowable distortion limits for each relevant parameter, at any
engine inlet attitude throughout the flight envelope (with an 'uninstalled', or, preferably, with an
'installed' power system). Since compressor stall may become a limiting factor (at high altitudes,
small Mach numbers and at high-rates-of-turn), the problem of engine-intake compatibility dur-
ing PST/PSM/RaNPAS becomes quite complicated by a matrix of yet unknown variablesl
limiting conditions (235-237).
Consequently, there is no universally-accepted method to define, a priori, actual distortion
limits to 'installed' propulsion systems of highly agile aircraft. Moreover, there are, yet, no bona
fide PST inlets. (Cf. Fig. I).
On the other hand, there are various methods to avoid, or by-pass, some specific distortion
problems, e.g., to activate special engine control subroutines during weapon firing, icy or
stormy-cloud penetrations, or during operations in rain, or in highly-dusty environments (179).
Engine malfunctioning may first be subdivided into four categories:

• Aircraft operating conditions (e.g., M, Re, attitude in incidence and yaw).

• Intake design adequacy (e.g., fixed or variable geometric control, shape, size, bleed, relative
position to the aircraft, staggering angle, simple or multiple ducting, top, bottom or side-type,
etc.).
• Engine's margin variations (e.g., the margin between accelerating/decelerating paths and the
surge line during rapid acceleratin-deceleration transients at any point on the various agility
metrics flight envelopes). Cf. Fig. 6.
• External influences - such as weapon firing, ice, rain, dust, temperature, pressure and density
variations at station 2. (Cf. Fig. 2.)
• The simplest indication of airflow departure from uniformity, under constant temperature
and composition, is the variation of total pressure across the compressor inlet. These varia-
tions are subdivided into "steady" and "unsteady" (or "dynamic" and "time-variant").
• Intake spatial dynamic non-uniformity may be called "turbulence", while the "time-variant"
types may cause "buzz" (internal intake shock oscillations).
Ultimately all these variables must be brought together in a total operational system, i.e., in a

241
 242 Vectored Inlets and IFPC

0/

130 ...
I \
Izo-Su-27
110

lOa
I
Vrclort'd flghtt'r
90 ,... Aircraft (lnl.. t with
Varlabl .. geometry.
80 I blowing and Inj"c,1lon)

70
PST
Subsonic 60
50

40

suPt'rvsonlc(Vt'clort'd or Convt'ntlonal
d.. Aircraft)
20
/,.. 15\
<.. 0 J
\ 5/
I l. +po
10

Fig. 1. The maneuver envelopes of PST, vectored aircraft at subsonic and supersonic speeds (an exam-
ple). PST, vectored aircraft are expected to perform inside these envelopes, especially during high-a-fJ
maneuvers at subsonic speeds (cf. Introduction and Lecture II). Our vectored F-15 and F-16 RPVs,
and the PV As, now operate well into the subsonic PST domains depicted.

well-integrated, whole-system design. Unless some compromises in cost and performance are
considered, this is not an easy task.

F .1.1 Limited Maneuvers Envelopes

Vectored aircraft, especially highly-maneuverable, strike-fighters, have to operate close to the
limiting conditions imposed by the strength ofthe aircraft structure, by the ability ofthe pilot to
withstand high g-Ioads (caused by rapid braking and turning, as well as by strong sidewise varia-
tions - each with its particular limiting factor); by intake-engine performance at high-a-J3 and
high-a-J3 time-variations, and by a few other factors to be discussed later.
Two examples ofiimiting maneuver-envelopes for fighters, one for PST-subsonic maneuvers,
and one for supersonic speeds, are illustrated in Fig. I.
As enumerated in the main text, these examples represent maneuverability envelopes
expected for vectored aircraft, both manned and unmanned.
Let us consider for instance, TR braking at low subsonic speeds, which may be combined with
very high drag through a rapid increase of AoA (cf. Figs. 21 in the Introduction, as well as Figs.
11-2, 3, 4 and Figs. II-II to 14).
Under these conditions an unsuitable inlet may cause engine-surge or flameout, and rob the
aircraft of its ability to perform safely in line with its mission.
Appendix F 243

Consequently, PST-intakes must be developed prior to the flight-testing of manned vectoredl

PST fighter aircraft. However, as we shall see, this is not going to be a simple task.

F-2 'Classical' and PST Intake Designs

The design of advanced engine intakes for vectoredlPSTIStealth aircraft cannot follow the clas-
sical, "disciplinary" approach. Instead, it must follow a new, whole-system, highly-integrated
methodology (0 B.I 0).
Such a new methodology may extract relevant tools from classical (different) "disciplines"
such as conventional propulsion, aerodynamics, flight-control, materials, radar-waves-
absorption-reflection theories, computers, and wind-tunnel/propulsion-Iaboratories testing
methodologies. It may then attempt to integrate them into a new field of study and practice.
Thus, unlike the situation in classical wing-aerodynamics, where flow separation is generally
taken to mark the technology limit of operation, the new engine intakes for super-agile aircraft
may be required to function satisfactorily under separated-flow regimes, or to continuously vary
their geometry in response to variations in a, {3, etc. Thus, from the practical viewpoint, a few
preliminary conclusions may be considered from the very outset:
• F1ight-testedlexperimental-data-bases and exact-intake-flow-theories, especially for PSTI
vectored aircraft during high-a-{3 maneuvers, are not yet available.
• Flush, variable-geometrylsubsonic/supersonic, high-aspect-ratio, low RCS or PST-inlets must
be developed now (cf., for comparison, the B-2 inlets and the integrated methodology
discussed in the Preface and the Introduction).
• In wind-tunnel-model-testing of vectored aircraft, the vectored engine may not be well-
represented. (The internal flow undergoes only a loss of total pressure corresponding to the
pressure recovery assessment. This, in effect, is an "internal drag" of the model, and it must be
assessed as such in order to reduce the relevant external drag from measurements oftotal drag
forces on the model. Cold air-suction and injection further complicates the dynamic simula-
tion of vectored models in wind-tunelling.)
• The generally accepted definition of engine thrust leads to the inclusion of an intake drag

System
Tailwind
failure
Fall in static

(orrected
engine
speed

Intake Mach Number

Fig. 2. An example of uncontrolled responses of the propulsion system to the transients indicated.
Engine's surge-stall margin is the distance between the operating point and the surge line (Fig. 6). The
surge line varies, however, with altitude, engine deterioration rate and history, trim, and IFPC
characteristics [179, 133]. See also Lecture III.
 244 Vectored Inlets and IFPC

known as "pre-entry drag", or "additive drag". This concept must now be re-examined and
re-assessed, say, for AoA 2: 90° .
• The presence of the wing section wetted by the inflow, or the aircraft canard ahead of the
intake, must be taken into account under unorthodox flight conditions, including Post-Stall
(PST) performance during supermaneuverability turns as those shown in Figs. 11-2, 3 and
11-11-14, and discussed in the Introduction.
(For details on conventional intakes the reader is referred to such comprehensive texts as that
of Seddon and Goldsmith's "Inlet Aerodynamics", Collins, London, 1985) [133].)

Corrected Temperature change

°C per sec.
engine
speed
Uncontrolled
(rpm) 5400
response
Controlled
5200 L..-_ _.u...._--".I...+- response
1.9 2.0 2.1
Intake Mach Number
Fig. 3. Controlled system response to temperature transients (Concorde data) (133). Cf. Fig. 6.

F.2.1 Engine Inlet Scale-Up From Vectored/PST RPV Flight Testing.

In our current research programs, we employ a-9ft-Iong vectored F-15 demonstrator model,
equipped with variable canards, yaw-pitch-roll vectored nozzles and on-board computer-metry
systems. The aim of the on-board computer-metry systems is to evaluate inlet air distortion
under flight conditions otherwise unattainable in wind-tunnels, or with RPVs powered by hot jet
engines. To fly this model one must first install a couple of single-stage or multiple-stage ducted-
fans, and, later, as the PST inlet design and testing is completed, to install hot vectored jet
engines. Since jet engines are highly sensitive to inlet high-a-,B distortion, one cannot install them
on a PST/vectored RPV prior to the completion of STAGE IV; namely, the flight-testing of PST
inlets (IV-4).
This methodology supplies, under "scale·up" restrictions, inlet flight data for new PST/
variable-inlet designs to be employed in future vectored aircraft.

F-3 IFPC Transient Definitions·

In the following we refrain from well-documented designs of such systems as rotating engine-
nacelles, jet-induced-flow-over flap, and reaction-control systems. We concentrate, instead, on
the emerging technology of vectored aircraft as defined in the Introduction and in Lecture I.

* This subject will be taken up again in Vol. II.

 :~
a
;c.
"1'1

~ ttT1 ·f] [.
.:---

Fig. 4. An example of conventional technology, variable-geometry inlets. Such inlets must be modified and redesigned for PSt. (Depicted are the F-14 and the N
UI
8-1 engine inlets). Note the rotatable vanes at the lower wall of the 8-1 inlet, and at inlet lips.
.
246 Vectored Inlets and IFPC

A./B Is not r~quire'd

(n ntW fight., Engln~

Uo

Fig. 5. Conventional Engine Thermodynamic Stations and Main Controlled Variables.IFPC must also
include the thrust-vectoring control rules enumerated in the main text, as well as those depicted here,
those associated with advanced HI DEC (cf. Appendix B), and, most important, those enumerated in
Figs. 7 and 8.

To gain a better physical understanding of characteristic transients to be caused by PST/

RaNPAS maneuvers, one may first establish comprehensive laboratory and flight tests aimed at
the development of new variable PST-inlet geometries and control rules (§ IV-4). However, prior
to this effort, one may have to assess propulsion-aircraft responses to transient effects such as:
• A rapid fall in static temperatures increases both M2 demand and intake Mach number M J ,
before reaching the surge line. (Fig. 3 shows the difference between uncontrolled and con-
trolled responses to temperature transients.)
• A sudden headwind increases M J , but reduces M2 demand, and vice versa for a tail wind.
• To preserve the required margin between engine operating line and engine surge line must be
the time-invariant objective of advanced PST-IFPC (Figs. 5, 6). This objective includes
engine-status/history-deterioration rate, mechanical trim, IFPC trim, etc. (Lecture III).
• To shorten the acceleration time, the closer must be the engine operating transient path to the
surge line (without, however, causing engine surge in the entire flight envelope) (cf. Fig. 6).
• As one moves from civil to military engines the required engine transients must be signifi-
cantly shorter.
• Vectored strike/fighters may require the fastest transients. Hence, IFPC for strike/fighters
may have to be the most effective (at the smallest possible engine margins - cf. Fig. 6).
• Widening engine/IFPC margins is a major goal for any new vectored propulsion design.
• Widening IFPC margins requirements for vectored aircraft are entirely different from those
required for advanced conventional aircraft. This is due to a number of reasons as listed
below:
(i) Most current technology military engines are designed with improved margins over their
entire operational range, i.e., from start-up/idling, to fuJI military, and up to full AB power.
This is due, in part, to Pilot's throttle variations/demands during air-to-air, or air-to-
ground modes of current pilot tactics. Since margins decrease with increasing altitude and
Appendix F 247

A ComRressor MaR..

Operating line

Corrt'c ted flow

Fig. 6. Limiting factors in acceleration-deceleration of military and civil engines are affected by many
factors, but mainly by al~itude, aircraft speed, engine RPM, and, engine history and inlet distortion level.
The following definitions apply:
I) A characteristic rapid acceleration path of some military engines.
2) A moderate acceleration rate (e.g., a limiting regulation for civil engines).
3) Compressor-stall domain.
4) Same as (2) for deceleration.
5) A characteristic rapid deceleration path.
6) The problematic gap between operating line and stall line during starts and restarts in altitude.

with decreasing aircraft speed, the mission should be well-defined for any IFPC mode of
operation.
(ii) Most critical air-to-air and air-to-ground pilot's modes with vectored aircraft would prob-
ably be accomplished with (fixed) full-throttle (while rapidly changing the IFPC com-
mands (cf. Figs. IL2 and II.I 0).
(iii) While IFPC requirements may be relaxed as one moves down from full military power to
cruise conditions and below (AB modes may not be required in most vectored aircraft),
most PST maneuvers are expected to be carried-out with full throttle.
(iv) Variable, flush, top-bottom intakes must be developed for minimum inlet-air distortion
at extremely high-a-f3 transients. Here, the B-1 (Fig. 4), the MiG-29 and the Su-27 inlet
designs are of particular interest.
(v) The development of combined-cycle, distortion-insensitive, surge-line-insensitive, cold/
hot propulsion systems must be given priority (cf. Lecture IV).
(vi) The greatest innovation efforts for vectored-aircraft intakes should concentrate on low
and medium subsonic maneuverability, without relaxing transonic and supersonic
demands, and without relaxing stealth benefits.
(vii) Current fly-by-wire (FBW) aircraft, such as the F-16, use three or four redundant control
channels to prevent aircraft loss from a single failure. The channels are connected together
at control surfaces by complex and expensive actuators, and multiple surfaces are used for
aerodynamic redundancy. That redundancy must now be well-integrated with a new type
of F-\6/IFPC methodology.
(Traditional aerodynamic decoupling of control surfaces - the elevator provides only
pitch, the ailerons only roll, for example - helped keep mechanical flight control systems
248 Vectored Inlets and IFPC

Senscr
feedbacks

~-::""---I~--.
EIfIdar~_-M----.

FADECA
(a-n./ A)

FADEC B
(aIImeI B)

FADEC A

.baTo

FADEC 8

To
otIodIn

Fig. 7. IFPC Computer Architecture (145, 104, 110, 128,235 -239).

Appendix F 249

simple, but is no longer necessary with electronic flight controls. This gives the designer
new degrees of freedom in control surface layouts.
Various self-repairing, improved control concepts are now being tested by the USAF-
WPAFB Hight Dynamics Laboratory. For instance, simulations of emergency cases, such
as an aileron separation from a test RPV may have to be investigatged to evaluate how the
IFPC system reconfigures to employ the remaining control surfaces to cope with the emer-
gency. These tests may be intended for a few payoffs: Improve resistance to battIe damage,
reduce flight control purchase and operational costs, and improve combat effectiveness.)
(cf. Appendix G).
(viii) Integration of all on-board avionics is expected to facilitate the introduction of integrated
cockpit displays that will provide future pilots with significantly improved tactical situa-
tion awareness. Such integrated systems will include IFPC navigation and communica-
tions systems, identification friend-or-foe systems, sensors, weapons and electronic
warfare systems and provisions to provide future fighter pilots with three-dimensional
visual, audio and tactile displays and cues, while permitting voice, head, and eyesight acti-
vation of aircraft systems and rapid reconfiguration of cockpit controls and displays
(235-239).
(ix) The proposed integrated systems may also provide means for pilot state monitoring, so
that the aircraft's automated flight/propulsion controls could, temporarily, fly the aircraft
and recover from hazardous situation, in case of indications of pilot incapacitation due,
for example, to a g-induced loss of consciousness.

eF-4 Short Inlets

Short inlet development is another activity which must be supported by analytic modelling,
wind tunnel tests and flight-testing ofRPVs. Since weight and balance considerations in a super-

...
lolll_lItdll

~
EIItctor nOl tnt... c..UI ' : - ...l1li

5y_l,al PI"""y Jel VeClo".g X X X

Dlltere"lill "mary Jet Vettor1ftQ X X X
5ymm,,,,al RoUllno v... I I X
L/R DilI.".I;'1 AoIlllno v... X I I
Til Diltt".I;'I RoUlono VI" X X X
l/R. TlB Dille,e.11I1 RoI.Io.g v... X X X
Syrrvnel,iUl G,oss ThruSI X X X ~
D"'erenlll' Gross Thrust X X X ~
Symmet,ic.1 S.. bll.lo, X I X X
Ddt" ••lill SUbilllOl X X X X
Syrrvnel"ClI C...,d X X X X
Olfrerenl~1 Can,rd X X X
Symmelral A,Ie,on X
D,rt,renllli Aileron X X X I
Syrrvnel'al Rudd., X X X X
Ollf"enlll' AuIJder X
5y_l,al Fllpe,on X X X IX)
0.11....1..1 FII"",o. X X I
5PH<1"'"k. ~
Syrrvnellal ""n Gu, B,"kes X ~
()dI,rUliI' M'Ift Gil' Brakes X ~
"ose Gel' SIee,"'O X ~
llR It" 10 Roghl 1/8· 100 to Bottom
~ ......., Ope'.'oo. 0.1y

Fig. 8. Four Right modes of the Vectored-F-15-STOL IFPC (Mello and Kotansky, 128) (d. Fig. 7).
250 Vectored Inlets and IFPC

sonic STOVL aircraft may position the engine farther forward in the fuselage than is currently
normal, the distance between the powerplant and the engine air inlet may be reduced, shortening
the diffuser length. Short diffusers typically perform poorly and current work may have to focus
on this issue. The new needs of PST-inlets may cause similar problems.
Using computational methods and wind tunnel test data, NASA-Lewis researchers have
recently fabricated a supersonic short diffuser model configured with several types of boundary
layer controls. The model incorporates holes of varying porosity and distribution for suction,
discrete jets for blowing and distributed slots for blowing control. This topic will be taken up
again in Volume II.

Vectored PST
C I " - ' "

.1 . . -------------~~,
f old .1
20 .~ 60 80 100
AoA

ATF Nominal Thrust

T,lb
THRUST

o~--~--~~~~--~~~
o OltO 0.80 1 120 1.60 2
M

Fig. 9. Other Inputs to the IFPC System during Vectored PST }'light (Mello and Kotansky, 128). These
include the variation of (total) thrust with aircraft Mach number and with altitude.
Note that the ATF-engine delivers 35,000 lb. (0 at M = 0, H = O. Cf. Appendix B. Note also that the
very definition of net thrust at PST-AoA must be reassessed at the inlet station.
 APPENDIX G

ON SOME OTHER VECTORED/STEALTH

AIRCRAFT CONSIDERATIONS
 G.t Combat Effectiveness of Vectored Aircraft
The following simulations may be considered in association with the ones given in the main text,
or as a separate entity. They were recently performed in France by Costes, Ph. and H. T. Huynh,
for a schematic model of a canard-configured, delta-wing, pitch-only, partially-vectored fighter
aircraft (DRET, STPA Contracts-88/89-p. 8510NERA's Activities - see also Costes paper:
Ref. 182).

Takeoff
distance Iml
600
--
--------L Without! Th t

-
rus
--. vectoring
400 With

200
/-
Thrust/weight ratio
0
08 1.2 F/mg
Takeoff
distance 1m)
..... ,
... _ Without thrust vectoring
600 ...'fi......._/
Stable Unstable ........ _
400
_i-____....:..W:.:i.:..:th:...:t::.:;hrust vectoring
200
Center of gravity 1%1
0
40 45 50

Takeoff
distance 1m) Fig. I. Advantage of thrust vectoring
for takeoff distance. The influence of
600
...... ------
..eJ--------- the thrust-to-weight ratio, the center
of gravity position and wing loading:
400 thrust-to-weight ratio = 0.89, center
of gravity = 44%, wing loading =
200
Wing loading 302 kglm 2 (dry runway, altitude 0,
o L -______~------~------.-------~ temperature = 1ST). (Costes and
250 300 350 Ik9/m21 Huynh, 182).

252
Other Considerations 253

AI.i.udelml
:zooo
:zooo
1600

1600
1000

1000
600
r----.,:7,-----
- - - - --') 0
&~=::-~-~-- 500
---_. ____ J~_~_=_=_-_~~~~.~ __ 1000
500
------,.----- -=_ =-_-:."'2.::::-..::-....--,
o -500 o 500
.. I
.-'1
0
,oj.. .. ---_.. "
~
_____
,600
o ____ ..J _ _ _ ..:~:i~===:::L.. Jl1°OO
_ 1000 - 500 0 600

0.50~ I
J'T':~~~
o 10 20 30 40 T,mo III

! O.51--+-I-~F-----1I-+---i
o 10 20 30 40 Ti.... IsI o

Fig. 2 (left) Combat between two identical ain:raft witbout tbrust vectoring. Aircraft A flight path
optimization (the flight paths are monitored every lOs) (Initial conditions: altitude 1,500 m, speed
150 mls). (Costes and Huynh, 182)
Fig. 3 (right). Air combat between two aircraft. Flight path optimization of aircraft A with thrust
vectoring (the flight paths are monitored every lOs). (Costes and Huynh, 182)

The advantage of thrust vectoring was evaluated by comparing the "optimum" flight paths of
two versions of the model; with and without deflectable engine nozzle. Optimization was
achieved by a numerical code using a projected gradient method developed at ONERA. This
method allows the time histories of the aircraft control surfaces (elevator, thrust deflector,
canards, etc.), to be determined to minimize the takeoff distance, while complying with the vari-
ous constraints inherent in the aircraft, such as control surface excursion, maximum AoA, etc.
With (pitch-only) jet-deflection restricted to maximum 20 degrees, a reduction of more than
50% of the takeoff distance was obtained, delimited by clearing an obstacle 15 m high (Fig. I).
While the takeoff distance was linearly reduced with increasing thrust-to-weight ratio - T /W - of
the simulated fighter, the (takeoff) advantage of vectored over conventional fighter was found to
be almost unaffected by T /W. However, that margin increased with increasing wing loading.
254 Other Considerations

The advantage ofthrust vectoring for air combat was evaluated by comparing the "optimum"
maneuvers of two versions of the aircraft, with and without a deflectable (pitch-only) nozzle,
against the same adversary, who adopts an aggressive closed-loop pursuit-law, attempting to
cancel the off-boresight angle between the fuselage axes.
The "optimum" maneuvers are aimed at maximizing a pelformance criterion based on win-
ning and survival probabilities computed along the flight-path on the basis of the respective
instantaneous threats (cf. Ref. 182, and Table I in the Introduction).

--LEFT HOVER RIGHT-

~o.-Wl.ID 'UCHl 'i:lOf'WII.ID'UOH

Fig. 4. Partially-vectored helicopters may employ efficient directional/tail thrust-vectoring control and
"circulation control" as depicted. Note the air-jet blowing orientation into the down wash air stream.
Appendix G 255

Figs. 5. Front view of a pure vectored aircraft RPV model. A computer-generated model made by one of
the students at the Jet Propulsion Laboratory. TIlT. (1986). This topic will be discussed in Volume II.

I
1.0 I
Close With Yeetored
Combet 0.75 ". I
Su permueuverebHit"

I With Conventional
Effeetiv- o.:50

ness
-~ ( U nyeetored)
fighter Ai reraft

0 .2:5

o
0.6 0.8 1.0 1.2 1.4

T/W
Thrust-to-Weight Rotio

20 r------r::;::ii!!!~~~::l--Y4l..,-Pitch
Veotorin,
........._ - - - . , . . - ".,.odlln4lmics
Turn-Rate onlll

(Deg/sec)
o

-20
o 3 6

T I t1 E (sec)
Fig. 6. According to Herbst of MDD, all-round jet deflection generates speed-independent control
moments in the yaw-pitch planes, thus allowing safe maneuvering in the stall and post-stall regimes. This
figure schematically shows the simulation of a scissoring maneuver and the associated turning rates
which may be obtained with and without all-round vectoring flight/propulsion control at constant
AoA = 40 degrees.
Consequently, with a jet-vectoring range of only 10 degrees, the roll turn-around time, i.e., the roll
agility at high angles-of-attack, may be easily doubled. This effect becomes more prominent as AoA is
increased, i.e., as the aerodynamic effect becomes weaker. The close-combat effectiveness was esti-
mated on the basis of simulations of one-on-one combat against MiG-29. Cf. Introduction. (After
Herbst, 188).
256 Other Considerations

_ _ _ _/..-:7- Pelroleum

~
Ulbojel ./
SFC
TUlbolan ~././
Ramlet

====-------
\
. . .-.... ---_.l,:::~~

O~~~ ___ ~ __ ~~ __ ~~.~ __ ~

0.6070.1 1.0 2.0 J.O 4.0 SO 60 708.0 100 12.0

Fig. 7. Thrust vectoring is not limited to turbojet and turbofan engines, nor to air-breathing engines
using petroleum fuel. In fact thrust vectoring is extensively used in boating and in rockets. The next step
is to introduce this technology to PST-ramjets fueled by petroleum-based fuel, or by hydrogen.
STOL benefits may also be imperative in such TV systems.

These simulations have demonstrated, for a variety of initial conditions, that even a pitch-
only thrust-vectoring is significantly advantageous in terms of combat effectiveness.
For instance, in a one-on-one gunfight between two identical aircraft without thrust vectoring,
maximizing the winning probability (over a duration of 60 s) leads to the well-known yo-yo
maneuver, which gives a tactical advantage over the adv,ersary (at approximately 10 s, see
Fig. 2), thereby increasing the winning probability. Howevt:r, if the pilot does not exploit this
advantage in time, the evolution could become unfavorable (near 22 s, Fig. 2), especially in
target-rich situations, due to a greater loss in speed.
The most interesting results of this study are shown in Fig . 3 for aircraft with thrust vectoring
(aircraft A). Costes and Huynh noted, however, that the results depicted in Fig. 3 had been calcu-
lated with a pitch-down nozzle deflection of up to 60 degrees (and assuming that trimming of the
aircraft remained possible). They also stress the need for (flight-testing) confirmation of their
analysis and assumptions.

G.2 Vectored Helicopters

Fig. 4 demonstrates the main design features of the newly-introduced partially-vectored helicop-
ters which employ directly vectored jet-force, as well as the effects of "circulation control"
induced by the rotor down-wash.

G.3 StealthNectored RPVs

As a general rule: Properly designed TV systems reduce signatures.
Unclassified material concerning this subject is to be published in Volume II. The configuration
depicted in Fig. 5 may only represent a single set of "unit operations" required in the early con-
ceptual design philosophy, and during laboratory and flight tests of such vehicles, as schemati-
cally depicted in the Preface to this Volume.
 APPENDIX H

THOUGHT-PROVOKING
AND THOUGHT-DEPRESSING
QUOTATIONS
"The USA is relying heavily on thrust vectoring and stealth
to restore its diminished technological superiority."

The Editor (JWRT)

Jane's All The World's Aircraft
1987-88, p. 43.[1751

(This quotation is included here for its first part. This inclu-
sion does not mean that the author concurs with the second
part of the quotation).

* * *
"NATO would be unwise to believe that it still has the clear
superiority in conventional air power that it has enjoyed for
most of the period since the Second World War."

The Editor (JWRT)

Jane's All The World's Aircraft
1987-88, p. 43.[1751

* * *
"No style of thinking will survive which cannot produce a
usable product when survival is at stake."
Thomas Favill Gladwin

* * *
"If a professor thinks what matters most,
Is to have gained an academic post,
Where he can earn a living, and then,
Neglect research, let controversy rest,
He's but a petty tradesman at the best,
Selling retail the work of other men."
Kalidasa
REFERENCES

Let us come now to references to authors, which other books

contain and yours lacks. The remedy for this is very simple;
for you have nothing else do do but look for a book which
quotes them all from A to Z, as you say. Then you put this
same alphabet into yours ... And if it serves no other pur-
pose, at least that long catalogue of authors will be useful to
lend authority to your book at the outset.
Miguel de Cervantes

259
 I. B.L. Berrier, "A Review of Several Propulsion Integration Features Applicable to Supersonic
Cruise Fighters". NASA TMX-73991, 1976.
2. A.D. Wolfe and A.E. Fanning, "Advanced Nozzle Technology", AGARD Fighter Aircraft
Design, June 1978.
3. J.P. Werdner and W.A. Vah!, "A Preliminary Assessment ofthe Impact of Two-Dimensional
Exhaust Nozzle Geometry on the Cruise Range of a Hypersonic Aircraft. Top Mounted
Ramjet Propulsion", NASA TM-81481, Sept. 1980.
4. D.L. Bowers and J.A Laughrey, "Integration of Advanced Exhaust Nozzles", AGARD-
CP-301, Sept. 1981.
5. R.J. Glidewell and R.E. Warburton, "Advanced Exhaust Nozzles Technologies", AGARD·
CP-301, Sept. 1981.
6. B.L. Berrier, J.L. Palcza and G.K. Richey, "Nonaxisymmetric Technology Program - An
Overview", AIAA 77-1023, Aug. 1977.
7. J.e. Que, "Advanced Technology Thrust Vectoring Exhaust Systems", AIAA 73-1304, Nov.
1973.
8. E.H. Miller and J. Protopapas, "Nozzle Design and Integration on an Advanced Fighter",
AIAA 79-1813, Aug. 1973.
9. W.W. Hinz and E.H. Miller, "Propulsion Integration ofa Supersonic Strike Fighter", AIAA
79-0100, Jan. 1979.
10. F.J. Capone, "The Nonaxisymmetric Nozzle - It Is for Real", AIAA 70-1810, Aug. 1979.
II. P.E. Hiley and D.L. Bowers, "Advanced Nozzle Integration for Supersonic Strike Fighter
Applications", AIAA 81-1441, 1981.
12. D.W. Speri and J.T. Blozy, "Development of Exhaust Nozzle Internal Performance Predic-
tion Techniques for Advanced Aircraft Application", AIAA 81-1490, 1981
13. D.L. Bowers and J.L. Laughrey, "Application of Advanced Exhaust Nozzles for Tactical Air-
craft", AIAA Proceedings, Vol. I, p. 132, 141, 1982.
"New Nozzles Design Aimed for the F-15, F-16 Aircraft", Aviation Week and Space Tech-
nology, Vol. 117, p. 67, 71, 73, Sept. 13, 1982.
14. Glidewell, R.J., "Installation Trades for Axisymmetric and Nonaxisymmetric Nozzles",
AIAA 80-1084, 1980.
15. J.W. Paulson and J.L. Thomas, "Low Speed Power Effect on Advanced Fighter Configura-
tions with Two-Dimensional Deflected Thrust", NASA TMX-74010, 1977.
16. J.P. Yip and J.W. Paulson, "Effects of Deflected Thrust on the Longitudinal Aerodynamic
Characteristics of a Close-Coupled Canard Configuration", NASA TP-I090, 1977.
260
References 261

17. F.J. Capone, "Aerodynamic Characteristics Induced on a Supercritical Wing Due to Vector-
ing Twin Nozzles at Mach Numbers from .4 to .95", NASA-CR 78746, 1978.
18. B.L. Berrier and F.J. Capone, "Effect of Simulated In-Hight Thrust Reversing on Vertical
Tail Loads of F-18 and F-15 Airplane Modes", NASA TP 1890, 1981.
19. W.C. Schnell and G.W. Ordonez, "Axisymmetric and Nonaxisymmetric Exhaust Jet
Induced Effects on a V/STOL Vehicle Design", Part one, NASA CR 166146, 1981.
20. F.J. Capone and D.E. Reubusch, "Effects of Varying Podded Nacelle-Nozzle Installations on
Transonic Aeropopulsive Characteristics of a Supersonic Fighter Aircraft", NASA TP 2120,
1983.
21. F.J. Capone and D.E. Reubusch, "Effect of Thrust Vectoring and Wing Maneuver Devices
on Transonic Aeropropulsive Characteristics ofa Supersonic Fighter Aircraft", NASA TP
2119, 1983.
22. F.J. Capone, "Supercirculation Effects Induced by Vectoring a Partial Span Rectangular
Jet", Journal of Aircraft, Vol. 12, No.8, pp. 633, 638, 1975.
23. A.I. Rilov, "Design of Supersonic Asymmetric Nozzles", Fluid Dynamics, Vol. 12. No.3, pp.
414-420, 1978.
24. D.L. Bowers and F. Buchan, "An Investigation ofthe Induced Aerodynamic Effects of a Vec-
tored Nonaxisymmetric Exhaust Nozzle", AIAA 78-1082, 1978.
25. W.P. Henderson et al., "Canard Configurated Aircraft with Two Dimensional Nozzles",
AIAA 78-1450, 1978.
26. W.C. Schnell and R.L. Grossman, "Vectoring Nonaxisymmetric Nozzle Jet Induced Effects
on a V/STOL Fighter Model", AIAA 78-1080, 1978.
27. J.W. Paulson, J.L. Thomas and L.P. Yip, "Deflected Thrust Effect on a Close Coupled
Canard Configuration", Jounral of Aircraft, Vol. IS, No.5, pp. 287-292, 1978.
28. F.J. Capone, R.J. Re and E.A. Bar, "Thrust Reversing Effects on Twin Engine Aircraft Hav-
ing Nonaxisymmetric Nozzles", AlAA 81-2365, 1981.
29. P.W. Banks, P.F. Quinto and J.W. Paulson, "Thrust Induced Effects on Low Speed Aerody-
namics of Fighter Aircraft", AIAA 81-2612,1981.
30. P.D. Wutten, L.W.Woodkey and J.E. Hamer, "Application of Thrust Vectoring for STOL",
AIAA 81-2616, 1981.
31. R.E. La Froth, "Thrust Vectoring to Eliminate the Vertical Stabilizer", M.Sc. Thesis
(AD-A079852) Avail NTIS HC A07/MFAOI CSCL 0113, Dec. 1979.
32. D.L. Maiden, "Performance of an Isolated Two-Dimensional Variable Geometry Wedge
Nozzle with Translating Shroud and Collapsing Wedge at Speeds up to Mach 2.01", NASA
TN-D 7906, 1976.
33. R. Burley, "Hight Velocity Effects on Exhaust Nossle Installed on an Underwing Nacelle on
an F-106 Airplane", NASA TMX 3361, 1976.
34. D.L. Maiden, "Two Dimensional Wedge/Translating Shroud Nozzle", NASA Case
LAR-1I919-1, U.S. Patent Appl. SN-672221, 1976.
35. D.E.Harrington and J.J. Schoemer, "Thrust Performance oflsolated Two Dimensional Sup-
pressed Plug Nozzles with and without Ejectors at Mach 0, to 0.45", NASA TMX 3384,
1976.
36. D.L. Maiden, "Performance on an Isolated Two-Dimensional Wedge Nozzle at Fixed Cowl
and Variable WedgeCenterbodyat Mach Numbers upt02.01", NASA TN-D 8218,1976.
37. F. Capone, "Performance on Twin Two-Dimensional Wedge Nozzles Including Thrust Vec-
toring and Reversing Effects at Speeds up to Mach 2.2", NASA TN-D 8449, 1988.
38. G.T. Carson and M.L. Mason, "Experimental and Analitical Investigation of a
262 References

Nonaxisymmetric Wedge Nozzles at Static Conditions", NASA TD-1188, 1978.

M.J. Harris and M.D. Falarshi, "Static Calibration of Two-Dimensional Wedge Nozzle with
Thrust Vectoring and Spanwise Blowing", NASA TM-81161, 1980.
39. M.L. Mason and W.K. Abeyounis, "Experimental Investigation of Two Nonaxisymmetric
Wedge Nozzles at Free Stream Mach Numbers Up to 1.2", NASA TO 2054, 1982.
40. D.L. Maiden and J.E. Petit, "Investigation on Two Dimensional Wedge Exhaust Nozzles for
Advanced Aircrafts", AIAA 75-1317,1975.
41. G.F. Goetz, J.H. Yong and J.L. Palcza, "A Two Dimensional Airframe Integrated Nozzle
Design with Inflight Thrust Vectoring and Reversing Capabilities for Advanced Fighter Air-
craft", AIAA 76-626, 1976.
42. J.E. Petit and F. Capone, "Performance Characteristics of a Wedge Nozzle Installed on a
F-18 Propulsion Wind Tunnel Model", AIAA 79-1164.
43. J.L. Palcza, "Augmented Deflector Exhaust Nozzle (ADEN) for High Performance Fighter",
AGARD Variable Geometry and Multicycle Engine, 1976.
44. R.J. Re and B.L. Berrier, "Static Internal Performance of Single External Ramp Nozzle with
Thrust Vectoring and Reversing", NASA TP L962, 1982.
45. J.A. Lander and D.O. Nash, "ADEN Design for Future Fighters", AIAA 74-131, 1975.
46. D.O. Nash, T.G. Wakeman and J.L. Palcza, "Structural and Cooling Aspects of the ADEN
Nonaxisymmetric Exhaust Nozzle", ASME 77-GT-IIO, 1977.
47. E.H. Miller, "Performance of a Forward Swept Wing Fighter Utilizing Thrust Vectoring",
AIAA 83-2482, 1983.
48. D.M. Straight and R.R. Cullom, "Thrust Performance of a Variable Geometry,
Nonaxisymmetric, Two-Dimensional, Convergent-Divergent Exhaust Nozzle on a Turbojet
Engine at Altitude", NASA TP-2172, 1983.
49. J.D. Stevens. E.B. Thayer and J.F. Fullerton, "Development of a Multifunction 2D-CD
Nozzle", AIAA 81-1491,1981. Cf. NASA CR-145295, Cr-135252, 1978.
50. D.M. Straight and R.R.Cullom "Performance of a 2D/CD Nonaxisymmetric Exhaust Noz-
zle in a Turbojet Engine at Altitude", AIAA 82-1137, 1982.
51. G.H. McLafferty and J.H. Peterson, "Results of Test of a Rectangular Vectoring/Reversing
Nozzle on a F-100 Engine", AIAA 83-1285, 1983.
52. B.L. Berrier and M.L. Mason, "A Static Investigation of Yaw Vectoring on 2D/CD Nozzles",
AIAA 83-1288, 1983.
53. J. Pennington, "Simulation Study of the Effects of Thrust Vectoring on Combat Effective-
ness of a Fighter Aircraft", NASA TM-X 3202, 1975.
54. F.e. Capone, "Static Performance of Five Twin Engine Nonaxisymmetric Nozzles with Vec-
toring and Reversing Capabilities", NASA TP-1224, 1978.
55. B.L. Berrier and R.J. Re, "Effects of Several Geometric Parameters on the Static Perform-
ance of Three Nonaxisymmetric Nozzle Concepts", NASA TP-1468, 1979.
56. F.J. Capone and B.L. Berrier, "investigation of Axisymmetric and Nonaxisymmetric Noz-
zles Installed on a 0.1 Scale F-18 Prototype Airplane Model", NASA TP-1638, 1980.
57. M.L. Mason, L.E. Putnam and R.J. Re, "The Effect of Throat Contouring on Two Dimen-
sional Converging-Diverging Nozzles at Static Conditions", NASA TP-1704.
58. J .A. Yeter and L. Leavitt, "Effects of Sidewall Geometry on the Installed Performance of
Nonaxisymmetric Convergent Divergent Exhaust Nozzle", NASA TP-1771, 1980.
59. P.J. Capone, "Aeropropulsive Characteristics of Twin Nonaxisymmetric Vectoring Nozzles
Installed with Forward-Swept and After-Swept Wings", NASA TP-I778, 1981.
60. F.J. Capone, "Aeropropulsive Characteristics of Mach Numbers Up to 2.2 of Axisymmetric
References 263

and Nonaxisymmetric Nozzles Installed on an F-18 Model", NASA TP-2044, 1982.

61. P.E. Hiley, H.W. Wallace and D.E. Booz, "Study of Nonaxisymmetric Nozzles Installed on
Advanced Aircrafts", AIAA 75-1316, 1975.
62. O.C Pendergraft, "Comparison of Axisymmetric and Nonaxisymmetric Nozzles Installed
on the F-15 Configuration", AIAA 77-842, 1977.
63. D.E. Berndt and A.P. Kuchar, "Nonaxisymmetric Nozzle Configuration Development for a
Multimission Cruise Aircraft", AIAA 77-843, 1971.
64. CM. Willard, F.J. Capone, M. Konarsky and H.L. Stevens, "Static Performance of
Vectoring-Reversing Nonaxisymmetric Nozzles", AIAA 77-840, Journal of Aircraft, Vol.
16, No.2, pp. 116-123, 1979.
65. W.C Schnell, R.L. Grossman and G.E. Hoff, "Comparison of Nonaxisymmetric and
Axisymmetric Nozzles installed on a V/STOL Fighter Model", SAE 77-0983.
66. P.E. Hiley, D.E. Kitzmiller and CM. Willard, "Installed Performance of Vectoring!
Reversing Nonaxisymmetric Nozzles", AIAA 78-1022, 1978.
67. F.J. Capone, N.S. Gowadia and W.H. Wooten, "Performance Characteristics of
Nonaxisymmetric Nozzles Installed on the F-18 Aircraft", AIAA 79-010 I, 1979.
68, R.A. Hutchinson, J.E. Petit, F.J. Capone and R.W. Whittaker, "Investigation of Advanced
Thrust Vectoring Exhaust Systems for High Speed Propulsive Lift", AIAA 80-1159,1980.
69. F.J. Capone, B.L. Hunt and CE. Poth, "Subsonic/Supersonic Nonvectored Aeropropulsive
Characteristics of Nonaxisymmetric Nozzles Installed on a F-18 Model", AIAA 81-1445,
1981.
70. H.L. Stevens, "F-15 Nonaxisymmetric System Integration Study Support Program", NASA
CR-135252, 1978.
71. R.E. Martens, "F-15 Nozzle/Afterbody Integration", J. of Aircraft, Vol. 13, No.5, pp.
327-333, May 1976.
72. P.E; Hiley, H. W. Wallace and D.E. Booz, "Nonaxisymmetric Nozzles Installed on Advanced
Fighter Aircraft", Journal of Aircraft, Vol. 13, No. 12. pp. 1000-1006, 1976; AIAA-75-1316.
73. G.K. Richey, B.L. Berrier and J.L. Palcza, "Two-Dimensional Nozzle/Airframe Integration
Technology - An Overview", AIAA 77-839, 1977.
74. J.L. Mace, E.B. Thayer and D. Bergman, "Nonaxisymmetric Nozzle Concept for an F-III
Test Bed", AIAA 77-841, 1977.
75. H.R. Wasson, G.R. Hall and J.L. Palcza, "Results ofa Feasibility Study to Add Canards and
ADEN Nozzle to the YF-I T', AIAA 77-1227, 1977.
76. P.D. Whitten and R.W. Woodney, "Vectored Engine-Over-Wing Configuration Design",
AIAA 77-1228, 1977.
77. G.F. Goetz, J.E. Petit and M.B. Sussman, "Nonaxisymmetric Nozzle Design and Evaluation
for F-III Flight Demonstration", AIAA 78-1025, 1978.
78. P.E. Hiley et al., "An Air Vehicle Performance Evaluation Utilizing Nonaxisymmetric Noz-
zles", AIAA 79-1811.
79. D. Bergman, "Thrust Vectoring Applied to Aircraft Hang High Wing Loadens", AIAA
79-1812, 1979.
80. M.R. Robinson, "Enhanced Capabilities of Future Fighters as a Result of HiMat", AIAA
79-698, 1979.
81. L. Verhavert, "In Flight Thrust Vector Control", Ph.D. Thesis, Univ. Microfilm No.
75-211100, 1975.
82. G.B. Russel, "An Experimental Investigation of the Pressure Field Associated a Thrust Vec-
tor Control Device", M.Sc. Thesis, AL A021875, GAE/AE75D-17, 1975.
264 References

83. D.M. Davies and J.D. Hines, "The Effect of Thrust Vectoring on Aircraft Maneuvering",
AD-A027367 USAFA-TR-76-9, 1976.
84. c.J. Vi, R.L. Humbold, R.J. Miller and E. Rachovitzky, "Flight and Propulsion Control
Integration for Selected In-Flight Vectoring Modes", ASME 78-GT-79.
85. H.J. Kelley, E.M. Cliff and L. Lefton, "Thrust Vectored Differential Turns", Proceedings
Vol. 1, (A 814550221-63), 1980.
86. E. Lange, "Optimal Flight Path for Winged Supersonic Flight Vehicles Extension to the
Case Where Thrust Can Be Vectored", Aeronautical Journal, Vol. 85, pp. 343-348, 1981.
87. J.E Pennington et ai., "Performance and Human Factors Results from Thrust Vectoring
Investigations on Simulated Air Combat", Proceedings Vol. 1 (A81-4550221-63), Joint
Automatic Control Conf., S. Francisco, Aug. 1980.
88. C.E. Robinson, "Exhaust Plume Thermodynamic Effects on Nonaxisymmetric Nozzle
Afterbody Performance in Transonic Flow", AEDC-TR-78-24, 1978.
89. K.M. Gleason et aI., "Investigation of Infrared Characteristics of Three Generic Nozzle
Concepts", AIAA 80-1160, 1980.
90. C. W. Chu et ai., "A Simple Nozzle Plume Model for I.R. analysis", AIAA 80-1808, 1980.
91. C.W. Chu and J. Der, "Modelling ofa Two-Dimensional Nozzle Plume for an I.R. Signa-
ture Prediction Under Static Condition", AIAA 81-1108, 1981.
92. C.W. Chu and J., Der, "ADEN Plume Flow Properties forIR Analysis", Journal of Aircraft,
Vol. 19, pp. 90-92, 1982.
93. B.G. Jaeggy, "A FORTRAN Program for the Computation of Two-Dimensional or
Axisymmetric Nozzles", (in French), (ISL-R-IIOI75),Available: HCA04/MFAOI, 1975.
94. J.D. Hanck and N.O Stockman, "Computer Program for Calculating Two-Dimensional
Potential Flow Through Deflected Nozzles", NASA-TM 79144m 1979,
95. P.D. Thomas, "Numerical Method for Predicting Flow Characteristics and Performance of
Nonaxisymmetric Nozzles, Theory", NASA CR-3147, 1979.
96. P.D. Thomas, "Numerical Method for Predicting Flow Characteristics and Performance of
Nonaxisymmetric Nozzles, Part Two: Applications", NASA CR-3264, 1980.
97. R.K. Tagirov and I.M. Shikhman, "Calculation of Supersonic in Flat and Axisymmetric
Nozzles of a Given Geometry, Assuming Arbitrary Inlet Gas Properties", Fluid Mechanic-
Soviet Research, Vol. 5, pp. 1-16, May, June 1976.
98. A.G. Mikhail, W. L. Hankey and J.S. Shang, "Computation ofa Supersonic Flow Past an
Axisymmetric Nozzle Boattail with Jet Exhaust", AIAA 78-993, 1978.
99.1.L. Osipov, "Numerical Method for Constructing Two-Dimensional Nozzles", Fluid
Dynamics, Vol. 14, no. 2, pp. 312-318, 1979.
100. C.W. Chu, "Expedient Approach to Nonaxisymmetric Nozzle Performance Prediction",
Journal of Aircraft, Vol. 17, pp. 127, 128, 1980.
101. R.C. Swanson, "Navier-Stokes Solutions for Nonaxisymmetric Nozzle Flows", AIAA
81-1217, 1981.
102. D.M. Straight, "Effect of Shocks on Film Cooling of a Full Scale Turbojet Exhaust Nozzle
Having an External Expansion Surface", AIAA 79-1170, 1979.
103. B. Gal-Or, "2-D, Vectoring/Reversing Nozzle Cooling - An Experimental Study of the
Temperature Fields in Sub and Supersonic Performance", a Technical Report of the Turbo
and Jet Engine Laboratory, The Dept. of Aeronautical Eng., Technion - Israel Institute of
Technology, Haifa, 1984.
104. H.D. Stetson, "Designing for Stability in Advanced Turbine Engines", International J.
Turbo & Jet Engines, Vol. 1, No.3, 1984, pp. 235-245.
References 265

105. C.C. Wu, et aI., "Study of an Asymmetric Flap Nozzle as a Vector-Thrust Device",
AlAA-84-1360, 1984.
106. G.R. Barns, et. al., "Vectoring Exhaust Nozzle Technology", AlAA-84-1175, 1984.
107. E. Hienz, "Requirements, Definition and Preliminary Design for an Axisymmetric Vector-
ing Nozzle to Enhance Aircraft Maneuverability", AIAA-84-1212, 1984.
108. J.A. Cohn, et aI., "Axisymmetric Thrust Reversing/Thrust Vectoring Exhaust System for
Maneuver and Balanced Field Length Aircraft", AlAA-85-1466, 1985.
109. !DR, 1014, Aug. 1987.
110. K.E. Smith, et. al., "Aircraft Control Integration Methodology and Performance Impact",
AlAA-85-1424, 1985.
Ill. R.R. Popelewski, "Modified F-15 will Investigate Advanced Control Concepts" , Aviation
Week & Space Technology, Feb. 11, 1985, pp. 51-53. Also May 29,1989, pp. 44-47.
112. L.D, Leavitt, "Static Internal Performance of a Two-Dimensional Convergent Nozzle with
Thrust-Vectoring Capability up to 60·", NASA TP-239I , 1985.
113. M.M. Drevillon and R. Fer, "Tuyeres a pousee orientable pour avions de combat futurs",
L'Aeronautique et l'Astronautique, No. 108, 1984-5, pp. 3-15 (in French).
114. J.R. Burley & J.R. Carlson. "Circular-To-Rectangular Transition Ducts for High Ar
Nonaxisymmetric Nozzles", AIAA-85-1346, 1985.
115. Ramesh K. Agarwal and Jerry E. Deeze, "Euler Solutions for Airfoil/Jet/Ground-
Interaction Flowfields". J. of Aircraft, Vol. 23, No.5, May 1986.
116. Tavella D.A. and Karamcheti K., "Lift of an Airfoil with a Jet Issuing from Its Surface", J.
of Aircraft, Vol. 24, No.8, August 1987.
116. Tavella D.A. and Karamcheti K., "Lift of an Airfoil withaJet Issuing from its Surface" J. of
Aircraft, Vol. 24, No.8, August 1987.
118, Sinha, K., Kimberlin R. and WuJ.M., "Equivalent Flap Theory: A New look at the Aerody-
namics of Jet-Flapped Aircraft". AIAA 22nd Aerospace Sciences Meeting, Jan. 9-12, 1984,
Reno, Nevada (AlAA-84-0335).
119. Gal-Or, B., "New Fighter Engines - a Review", Part I includes 103 references relevant to
vectored engines, Intern. J. Turbo and Jet Engines, 1, 183-194 (1984).
120. Marcum, D.L. and Hoffman, J.D., "Calculation of Three-Dimensional Inviscid Flowfields
in Propulsive Nozzles with Centerbodies", AlAA-86-0449.
121. Capone, F.J., "A Summary of Experimental Research On Propulsive-Lift Concepts in the
Langley 16-Foot Transonic Tunnel", AIAA-75-I315.
122. Fair G. and Robinson, M.R., "Enhanced Capabilities of Future Fighters as a Result of
Himat", AIAA-79-0698.
123. Sweetman, B., "ATF", /DR, 172-174 (1983); Gilson, C. "Future Fighters for the US Air
Force", /DR, 165-171 (1983).
124. Joshi, P.B. and Hawthorne, C.A., "Approach and Landing Thrust Reverser Testing in
Ground Effects", AlAA-85-3075.
125. Schultz, J., "Stealth Takes Shape", De! Elect. 65, 1983.
126. Reneau, L.R., et aI., "Performance and Design of Straight, Two-Dimensional Diffusers", J.
of Basic Engineering, 141, 1967.
127. Kowalski, E.J., "A Computer Code For Estimating Installed Performance of Aircraft Gas
Turbine Engines", NASA CR 159691, 1979 (plus a few other NASA's computer programs
and tapes purchased through COSMIC, USE).
128. Mello, J.F. and Kotansky, D.R., "Aero-Propulsion Technology For STOL and Maneuver",
AIAA-85-4013.
 266 References

129. Mihaloew, J.R., "Flight Propulsion Control Integration for V/STOL Aircraft", (NASA),
Intern. J. Turbo and Jet Engines; In press.
130. Emami-Naeni, A; Anex, R.P.; Rock, S.M., "Integrated Control: A Decentralized
Approach" Proceedings of the IEEE Conference on Decision and Control.
131. Hawkins, J.E., "YF-16 Inlet Design and Performance", AIAA 74-1062.
132. Guo, R.A. and Seddon, J., "Swirl Characteristics of an S-Shaped Air Intake With Both Hor-
izontal and Vertical Offsets", Aeronautical Quarterly, May 1983.
133. Seddon, J. and Goldsmith, F.L., "Intake Aerodynamics", Collins, London, 1985.
134. Smith, Kenneth L.; Kerr, W. Bernie; Hartman, G.L.; Skira, charles, "Integrated Flight/
Propulsion Control-Methodology, Design and Evaluation", AIAA-85-3048, 23p.
135. Bill Sweetman, "Steatlh Aircraft", Airlife, England, 1986.
136. Mctycka, D.L., "Determination of Maximum Expected Instantaneous Distortion Patterns
From Statistical Properties of Inlet Pressure Data", AIAA 76-705.
137. Sedlock, D., "Improved Statistical Analysis Method for Prediction of Maximum Inlet Dis-
tortion", AIAA-84-1274.
138. Kuchar, A.P.,"Variable Convergent-Divergent Exhaust Nozzle Aerodynamics", Chart. 14
in Oates, Gorden, C., Editor: "The Aerothermodynamics of Aircraft Gas Turbine
Engines"; University of Washington, Seattle, Wash. 1978, US-DOC AD/A-059 784.
139. Hiley, P.E., D.E. Kiotzmiller and C.M. Willard, "Installed Performance of Vectoring!
Reversing Nonaxisymmetric Nozzles", AIAA-78-1022.
140. Joshi, P.B., "Unsteady Thrust Reverser Effects in Ground Proximity", AlAA-85-4035.
141. Cohn, J.A., Dusa, D.J., Wagenknecht, C.D. and Wolf, J.P., "Axisymmetric Thrust
Reversing-Thrust Vectoring Exhaust System for Maneuver and Balanced Field Length",
AIAA-85-1466.
142. Berndt, D.E., Glidewell, R. and Burnes, G.R., "Integration of Vectoring Nozzles in a STOL
Transonic Tactical Aircraft", AlAA-85-1285.
143. Stromecki, D.J., "An Assessment of Gas Turbine Engine Augment or Technology and
Needs for the 80's", AIAA-80-1200.
144. "Integrated Flight Propulsion Control (lFPC)" and related subjects: List of publications
found in DIALOG (available in Technion's Jet Lab.).
145. Landy, R.J., Yonke, W.A.; Stewart, J.F., McDonnell Aircraft Co., St. Louis, MO, USA,
"Development of Hi dec Adaptive Engine Control Systems", Journal ofEn gineering for Gas
Turbines and Power, Transactions of the ASME v109 N2 Apr. 1987, p. 146-151.
146. Gal-Or, L., "RCC Production/Processing By New Methods", Technion, 1988. A Research
Contract Funded By USAF.
147. Carr, J.E., "Aerodynamic Characteristics of a Configuration With Blown Flaps and Vec-
tored Thrust for Low Speed Flight", AIAA-84-2199.
148. Banks, D.W. and Paulson, J.W., "Approach and Landing Aerodynamic Technologies for
Advanced STOL Fighter Configurations", AIAA-84-0334.
149. Franciscus, L.c. and Luidens, R.W., "Supersonic STOVL Aircraft With Turbine Bypass/
Turbo-Compressor", AIAA-84-1403.
150. Zola, c.L.; Wilson, S.B. and Eskey, A., "Tundem Fan Applications in Advanced STOVL
Fighter Configurations", AIAA-84-1402.
151. Seginer, A. and Salomon, M., "Augmentation of Fighter Aircraft Performance by Span-
wise Blowing Over the Wing Leading Edge," Tech. Mem. 84330, NASA Ames Res. Center,
1983.
152. Englar, R.J. and Huson, G.G., "Development of Advanced Circulation Control Wing High
References 267

Lift Airfoils", David Taylor Naval Ship RD Center, Bethesda, Md. USA,
AIAA-83-1847.
153. Woan, c.J., "A Three-Dimensional Solution of Flows Over Arbitrary Jet-Flapped Configu-
rations Using a Higher-Order Panel Method", AIAA-83-1846.
154. Herbst, W.B., "Supermaneuverability", MBB/FEI/S/PUB/ 120, 7.10.1983; AGARD, FMP
Conference on fighter maneuverability, Florence, 1981; AGARD CP-319.
155. Paulson, J.W. and Gatlin, G.M., "Trimming High Lift for STOL Fighters",
AIAA-83-0168.
156. Kimberlin, R.D. and Sinha, A.K., "STOL Attack Aircraft Design Based on an Upper Sur-
face Blowing Concept". AIAA-86-1860.
157. Glazer, A., Hugh, R.V. and Hunt, B.L., "Thrust Reverser Effects on the Tail Surface Aero-
dynamics of an F-18 Type Configuration", AIAA-86-1860.
158. Callahan, c.J., "Tactical Aircraft Payoffs for Advanced Exhaust Nozzles",
AIAA-86-2660.
159. Banken, G.J., Cornetle, W.M. and Gleason, K.M., "Investigation ofInfrared Characteris-
tics of Three Generic Nozzle Concepts, AIAAISAEIASME 16th Joint Proulsion Conference,
June 30-July 2, 1980, Hartford, Conn.
160. Norain, J.P., "Viscous-lnviscidSimulation of Upper Surface Blown Configurations",
AIAA-84-2200.
161. Shaw, P.D., Blumberg, K.R. and Joshi, D.S., "Development and Evaluation of an Inte-
grated Flight and Propulsion Control System", AIAA-85-1423.
162. Leavitt, L.D. and Burley, J.R., "Static Internal Performance of a Single-Engine
Nonaxisymmetric-Nozzle Vaned-Thrust-Reverser Design With Thrust Modulation Capa-
bility", NASA Tech. Paper 2519 (1985).
163. Imlay, S.T., "Numerical Solution of 2-D Thrust Reversing and Thrust Vectoring Nozzle
Flowfields", AIAA-86-0203.
164. Sedgwick, T.A., "Investigation of Non-Symmetric Two-Dimensional Nozzles Installed in
Twin-Engine Tactical Aircraft", USAF, Flight Dynamics Laboratory, Contract
F3615-74-C-3051 to Lockheed, Burbank, California.
165. Lachmann, G.V., "Boundary Layer and How Control", Pergamon Press, N.Y., 1961.
166. Capone, F.J., "Lift Induced Due to Thrust Vectoring of a Partial-Span Two-Dimensional
Jet at Mach Numbers from 0.4 to 1.30", Langley Working Paper 116, July 1973.
167. Yoshihara, H., "Transonic Performance of Jet Flaps on an Advanced Fighter Configura-
tion", USAF, Flight Dynamics Laboratory-TR-73-97, 31 July, 1973.
168. Yonke, W.A, Terrel, L.A. and Myers, Larry P., "Integrated FlighUPropulsion Control:
Adaptive Engine Control System Mode", AIAA-85-1425.
169. Smith, K.L., Kerr. W.B., Hartmann, G.L. and Skira, Charles, "Aircraft Control
Integration-Methodology and Performance Impact", AIAA-85-1424.
170 Vizzini, R. W., "Integrated Flight/Propulsion Control System Considerations For Future
Aircraft Application", Journal of Engineering for Gas Turbines and Power, Transactions
ASME vl07 N4 Oct. 1985, p. 833-837.
171. Yonke, W.A., Landy, R.J.; Cushing, J.M., "Integrated Flight/Propulsion Control. Hidec
Modes", IEEE Proceedings of the 1984 National Aerospace and Electronic Conference V I,
p.472-478.
172. Emerson, L.D. and Davies, W.J., "Flight/Propulsion System Integration",
AIAA-83-1238.
173. Landy, R.J. and Kim, D.B., "Multivariable Control System Design Techniques: An Appli-
268 References

cation to Short Takeoff and Landing Aircraft", Proceedings of the AlAAllEEE 6th Digital
Avionics Systems Conference, Dec. 1984, p. 132-139.
174. Gal-Or, B., "Vectored Aircraft For the 90's" Intern. J. Turbo and Jet-Engines, 4, I (1987).
175. "Jane's All The World's Aircraft", 1987-88 Ed., "RPVs", "Israel", "TIlT";
"Jane's Yearrbooks", Jane's Publishing Inc., New York, p. 826. Also A W&ST, p. 21, May,
1987.
176. Gal-Or, B., Ed., "A Critical Review of Thermodynamics", Mono Book Corp., Baltimore,
Md. 1970, pp. 299, B-T Chue, Chapter on "Thermodynamics of Deformation of
Solids".
177. Sobota, T.H. and Marble, F.E., "An Experimental and Numerical Investigation of Swirling
Flows in a Rectangular Nozzle", AlAA-87-2108.
178. Von Glahn, U.H., "Secondary Stream and Excitation Effects on Two-Dimensional Nozzle
Plume Characteristics", AlAA-87-2112.
179. Gal-Or, B., "vectored Propulsion"; Internal Reports, The Jet Propulsion Laboratory, Aero
Eng.; Technion lIT, May, 1983; July, 1985; May 1986; May 1987; Aug., 1988 (see also Ref.
206).
180. Yugov, O.K., Selyvanov, O.D., Karasev V.N., and Pokoteeto P.L., "Methods ofIntegrated
Aircraft Propulsion Control Program Definition", AIAA-88-3268 (Central Institute of Avi-
ation Motors, Moscow, USSR).
i 81. Ogburn, M.E., Nguyen, L. T. and Hoffer, K.D., "Modeling of Large-Amplitude High-Angle-
of-Attack Maneuvers", AIAA-88-4357-CP.
182. Costes P., "Thrust Vectoring and Post-Stall Capability in Air Combat",
AIAA-88-4160-CP.
183. Beaufrere, H., "Integrated Flight Control System Design for Fighter Aircraft Agility",
AIAA-88-4503.
184. Hodgkinson, J., Skow, A., Ettinger, R., Lynch, U., Laboy, 0., Chody, J. and Cord, T.J.,
"Relationships Between Flying Qualities, Transient Agility, and Operational Effectiveness
of Fighter Aircraft", AIAA-88-4329-CP.
185. Tamrat B.F., "Fighter Aircraft Agility Assessment Concepts and Their Implication on
Future Agile Fighter Design", AIAA-88-4400.
186. Geidel, H.A., "Improved Agility for Modem Fighter Aircraft: PART II: Thrust Vectoring
Engine Nozzles", MTU Motoren-und Turbinen-Union, Munchen, Internal Report,
20.2.87.
187. Miller, S.c., "Future Trends - A European View" (Rolls-Royce), ISABE-87-7002.
188. Herbst, W.B., "Thrust Vectoring - Why and How?" ISABE-87-7061.
189. Adam, J.A., "How To Design an 'Invisible' Aircraft", IEEE Spectrum, April, 1988.
190. Knott, E.F. et al., Radar Cross Section, Artech House, Dedham, Mass., 1985.
191. Sweetman, B., "UAVs", Interavia, 775, 811988. IDR, 787, 711988.
192. Nagabhushan, B.L. and faiss G.D., J. AIRCRAFT, Vol. 21, No.6, 408, 1984.
193. Franklin, J.A., Hynes, C.S., Hardy, G.H., Martin, J.L. and Innis, R.C., J. GUIDANCE,
555, 1986.
194. Ransom, S., J. AIRCRAFT, Vol. 20, 59, 1983.
195. Foltyn, R.W., et al., "Development of Innovative Air Combat Measures of Merit for
Supermaneuverable Fighters", AFWAL-TR-87-3073, October 1987.
196. McAtee, T.P., "Agility - Its Nature and Need in the 1990s", Society of Experimental Test
Pilots Symposium, Sept. 1987.
197. Bucknell, R.L., "STOVL Engine/Airframe Integration", AIAA-87-171I.
References 269

198. Tape, R.F., Glidewell, R.J., and Berndt, D.E., "STOL Characteristics ofa Tactical Aircraft
with Thrust Vectoring Nozzles", AIAA-87-183S.
199. Barnes, G.R., et al., "Vectoring Exhaust Nozzle Technology", AIAA-84-117S.
200. Tape, F.F., et al., "Vectoring Exhaust Systems for STOL Tactical Aircraft", ASME
83-GT-212.
201. Curry, S.G., et aI., "Exhaust Nozzle Concepts for STOL Tactical Aircraft",
AIAA-83-1226.
202. Lewis, W.J. and P. Simkin, "A Comparison of Propulsion Systems for V/STOVL Super-
sonic Combat Aircraft", SAE-80-1141.
203. Adams, A. and S. Parkola, "Charting Propulsion's Future - The ATES Results",
AIAA-82-1139.
204. Lynn, N., "Powered Lift", Flight International, p. 29, April 9, 1988.
20S. Tape, R.F., Glidewell, R.J. and Berndt, D.E., "STOL Characteristics ofa Tactical Aircraft
with Thrust Vectoring Nozzles", Rolls-Royce Inc., 1987.
206. Janes "All-The-World-Aircraft", 'RPV';-TIIT, 1988/1989, AW&ST, May 18, 1987. See
also Ref. 212.
207. Bare, E.A. and Reubush, D.E., "Static Internal Performance of a Two-Dimensional
Convergent-Divergent Nozzle With Thrust Vectoring", NASA Tech. Paper, 2721, July,
1987.
208. Mason, M.L. and Berrier, B.L., "Static Performance of Nonaxisymmetric Nozzles With
Yaw Thrust-Vectoring", NASA Tech. Paper 2813, May 1988.
209. Berrier, B.L. and Mason, M.L., "Static Performance of an Axisymmetric Nozzle With Post-
Exit Vanes for Multiaxis Thrust Vectoring", NASA Tech. Paper, 2800, May 1988. See also
AIAA-87-1834.
210. Richey, G.K., Surber, L.E. and Berrier, 8.L., "Airframe-Propulsion Integration for Fighter
Aircraft", AIAA-83-0084.
211. Well, K.H., DFVLR, Report ASS2-7812, 1978.
212. Gal-Or, 8., "The Principles of Vectored Propulsion", International J. o/Turbo and Jet-
Engines, Vol. 6, 1989.
213, Tamrat, B.F. and D.L. Antani, "Static Test Results of an Externally Mounted Thrust Vec-
toring Vane Concept", AIAA-88-3221.
214. Papandreas, E. "SCAT: A Small Low Cost Turbojet for Missiles and RPVs",
AIAA-88-3248.
21S. Eames, D.J.H. and Mason, M.L., "Vectoring Single Expansion Ramp Nozzle (VSERN)
Static Model Test Program", AIAA-88-3000.
216. Miau, J.J., S.A. Lin, J.H. Chou, c.y. Wei and C.K. Lin, "An Experimental Study of Flow in
a Circular-Rectangular Transition Duct", AIAA-88-3029.
217. Humphreys, A.P., Paulson, J.W., and G.T. Kemmerly "Transient Aerodynamic Forces on
a Fighter Model During Simulate~ Approach and Landing With Thrust Reversers"
AIAA-88-3222.
218. Schneider, G.L. and G.W. Watt, "Minimum Time Turns Using Vectored Thrust",
AIAA-88-4070-CP.
219. Sobel, K.M. and F.J. Lallman, "Eigenstructure Assignment for a Thrust-Vectored High
Angle-of-Attack Aircraft", AlAA-88-4101-CP.
220. K1afin, J.F., "Integrated Thrust Vectoring On the X-29A", AIAA-88-4499.
221. Widdison, C.A., "Aircraft Synthesis with Propulsion Installation Effects",
AIAA-88-4404.
270 References

222. VanOverbeke, T.J. and J.D. Holdman, "A Numerical Study of the Hot Gas Environment
Around a STOVL Aircraft in Ground Proximity", AIAA-88-2882.
223. Oh, T.S. and J. A. Schatz, "Finite Element Simulation of Jets in a Crossflow with Complex
Nozzle Configurations for V/STOL Applications", AIAA-88-3269.
224. Gal-Or, 8., "The Fundamental Concepts of Pure Vectored Aircraft", ASME Conferences,
Belgium, in press.
225. Gal-Or, B., "The Fundamental Concepts of Vectored Propulsion", J. Propulsion, in press.
226. Pavlenko, V.F., "Powerplants with In-Flight Thrust Vector Deflection", Moscow, Izdatel'
stvo Mashinostroenie, 1987, 200p., 37 refs. In Russian.
227. Sweetman Bill, "B-2 Bomber for the 21st Century", Interavia, 111989, p. 22.
228. Cawthon, J.A., "Design and Preliminary Evaluation ofInlet Concepts Selected for Maneu-
ver Improvement", AIAA-76-70 I.
229. Hawkins, J.E., "YF-16 Inlet Design and Performance", AIAA-74-1062.
230.lmfield, Wiliam F., "The Development Program for the F-15 Inlet", AIAA-74-1061.
231. Wiliams, J. and Butler, S.F.J., "Aerodynamic Aspects of Boundary-Layer Control for High-
Lift at Low Speeds", J. of the Royal Aero. Soc. 67, No. 628 (April 1963), p. 201.
232. McLean, J.D., and Herring, H.J., "Use of Multiple Discrete Wall-Jets for Delaying
Boundary-Layer Separation", NASA CR-2389, June 1974.
233. Banks, D.W., "Aerodynamics in Ground Effect and Predicted Landing Ground Roll of a
Fighter Configuration With a Secondary-Nozzle Thrust Reverser", NASA TP-2834, Oct.
1988.
234. Bare, E.A., and O.c. Pendergraft, Jr., "Effect of Thrust Reverser Operation on the Lateral-
Directional Characteristics of a Three-Surface F-15 Model At Transonic Speeds", NASA
TP-2234, 1983.
235. Mattes, R. and W. Yonke, "The Evolution: IFPC to VMS", AIAA-89-2705.
236. Putnam, T.W. and R.S. Christiansen, "Integrated Control Payoff", AIAA-89-2704.
237. Tillman, K.D., T.I. Ikeler and R.A. Purtell, "The Pursuit of Integrated Control: A Real
Time Aircraft System Demonstration", AIAA-89-2701.
 INDEX
A number in brackets refers to a Reference number.
Otherwise a number refers to a page number.

A-12; 193,206,210 Aerodynamic Flight Control (see Thrust-

(see also ATA) Vectored Flight Control, IFPC)
AV-8; 57,81,79,181, 194, 198,210 AIAA 15
(see also Harrier and STOVL) Airframe-Nozzle-Integration 29, 119-122, 146,
ADEN thrust-vectoring engine nozzle 26, 118, 156-171
201 -and low observability (see Stealth Aircraft,
(see also Nozzle) Design)
Aerodynamics of Thrust-Vectored Aircraft -and drag (see Mach-Number effects, Nozzle
(see also Boundary-Layer control, Upper Drag, Design, R&D Cycles)
Surface Blowing, High-Alpha, Post-Stall Agiliy
Technology, Inlets, Nozzles, Wind-Tunnel (see also Supermaneuverability,
Simulations, Computer Simulations, Scal- Supercontrollability, Maneuver Analysis)
ing-Up, Canard Blowing, Lift, Cooling, Heat -definition/measurement debate 27, 34, 35,
Transfer, Agility, Supermaneuverability, 37-41,47-48,51-55,79-104,145
Stability, Flight Tests, Flight-Propulsion -and maneuverability 27, 34, 47-55,
Control, IFPC, Stall Barrier, RaN PAS, PSM, -and controllability 27,29,34,51-55
Supercirculation, Supercontrollability, Yaw- - and comparison of one aircraft to another 51
Pitch-Roll Thrust Control, Mach-Number Altitude Performance 49, 250
Effects) Angle-of-Attack during PST-TV 24, 25, 29, 32,
-of thrust-vectored aircraft 29-55, 88-104, 39,48, 52, 94, 95, 183-6,242,250
139,141,171,178-186,241-250 (see also Post-Stall, High Alpha)
-of thrust-vectored control surfaces 29, 52, ASME 15
54, 83-88, 171, 172 ATA (A-12) 193,195,206,210
-of engine inlets during post-stall flight 29, ATB (B-2) 46,47, 182,206,210,243
241-250, 48, 50, 89, 97, 107, 115-120, ATF (Advanced Tactical Fighter) 25,26,55,97,
143-4, 175,226 131,193, 195, 197,206,207,210
-of thrust-vectored engine nozzles 29, 42, -post ATF 29-55, 168, 195
59-97,115-120,157-171,180,197,199, Avionics for Thrust-Vectored Fighter Air-
212-214 craft 29,41, 50, 52-55, 225

271
272 Index

(see also IFPC, C3I, Time Lags) - Flight-Control & Data Extraction by 144,
Axi-TV Nozzles 40,71,75,83,88,181,206 150
(see also Nozzle) -installation in R&D-RPV for Flight
Testing 143
Controllability
Banks (29) 191
(see Supercontrollability, Agility, and
Bar-Annan 15
Supermaneuverability)
Barret, T. 15
Cooling 10, 64, 108-115, 204
Benstein, E. 15
(see also Heat Transfer)
Berndt (63, 142, 198,205) 191
Cost-Effectiveness 139, 189
Berrier, B. (1,6, 18,44,52,55,73,208,209,210)
Costes (182) 192, 252, 256
26,123,167,173,174,175,180,189,191,
Cruise Missiles, see Missiles
227
Cullom (48) 68, 109, 113-116
Boundary-Layer Control 85, 175, 190-1,
Current-Technology Limits & Trends 203
113-120,169,172
Bowers, D. (4, II, 13,24) 15,191,220,227,228
Burley (33, 114, 162) 192 Dekel, E. 3
By-Pass Ratio Trends 205,212-213 Der (91) 117, 118
Design of Thrust-Vectoring/Stealth/PST Air-
Callahan (158) 192, 228 craft 156-175, 10-13, 24, 29, 56-77,
Capone (10, 17, 18,20,21,22,28,37,42,54,56, 141-149, 189,209
49,60,64,67,48,69,121) 123, 167, 173, -philosophy 10-13, 56-59, 189, 209
174, 175, 192, 221, 222, 227 -criteria 24, 29, 56-59, 189, 209
Canard-Configured Vectored Aircraft 24, 29, Dewey 133
80-1, 156-172 Dimensionless Numbers 83-87
Canard Blowing 85, 171 Direct-Lift Engines 190
Chody (184) 35, 37, 40 Dunbar, D. 15
Chu (90, 92, 100) 117, 118, 192 Durham, T. 15
C 3 I 47, 50-55
Civil Applications 25 Educational Shifts 9, 12, 25, 52, 176
Cobra Maneuver 29,97, 167,211 Ehric, F. 15
Cohen,Z. 3, 146 Einstein, A. 9, 56, 194
Cold Propulsion 89, 138, 144 Ejector Lift/Thrust-Vectoring 190, 199
(see also Engines, Compound) Energy
Combat 24, 27, 34, 44, 48, 91-104, 162, 182, -Plots 95, 100, 101
249, 252, 255 - Maneuvers/Tactics 40
-scenario 22,24,29,34-7,44,46,93,96,99, - Management 40, 100-3
100-104, 252-3 -speed plots 103
-applicatons 34,35,48,91-104 (see also Mach-Number Effects)
-air-to-air 34, 35, 37-41,44,91, 104, 189, Engine
252-3 (see also Jet Engines)
-air-to-ground 34,45,92,99, 189 -technology limits 108, 195-213, 141,
-limitations 34, 38, 45, 48, 79-104 240-250
-close 91, 104, 162,252-3,255 -variable cycle 197,199,213
[see also Supermaneuverability and -ATF 25,26,55,97,131,193,195,197,206,
Supercontrollability1 207,210
Computer Simulations 10, 12, 48, 100-3, 146, -ATA 193,195,206,210
252-3 -ATB 206,210
Computer (Onboard) -EFA 199,209,210
(see also Flight Tests, RPVs, IFPC, -sizing 130, 149
Stability) - air-mass-flow 132
Index 273

-stall line 243,244,247 [see also Film Cooling]

-compound 88,89,135,138,141 Helicopters, Thrust Vectored 254
Ettinger (184) 35, 37, 40 Helmet-Sight-Aiming vs.
European Fighter Aircraft-EFA 209,210 Supermaneuverability 34, 43, 53
-engine 199,209,210 Herbst (154, 188) 32, 48, 91, 93, 94, 100-103,
181,255
FADEC 206 HIDEC 10, 104,246 [see also IFPC, FADEC]
Film Cooling 70, 109, 110, III High Alpha
Flight-Propulsion Control, see IFPC -research 22-55,88-89,92-104,178-186
Flight Testing [Integrated Methodology with Jet- -maneuverability [see PST]
Propulsion-Laboratory Testing] 10, 29, 38, - inlets [see Inlets]
48, 50-55, 72, 83, 85, 139, 143-151, 189, Hiley (11, 61,66, 72, 78, 139)192, 220, 227,228
244 Hodgkinson (184) 35, 37, 40
Flight Control Changes due to Thrust Vector- Hot-gas-ingestion 175
ing 185, 189 [see also inlet]
(see also IFPC, Combat, and flight
Testing) IHPTET 207
Flight Envelopes Changes due to Thrust vector- IFPC [Integrated Flight/Propulsion Control] 10,
ing 22,29,49,98, 102, 189 24,29, 30, 32, 33, 50, 51, 61, 69, 81-104,
(see also Supermaneuverability, Agility, 107, 115-120, 170, 189, 206, 244-250
Combat) Inertial Considerations 143, 153
Friedman, E. 3, 15, 16, 146 [see also Stability, Flight Testing,
Fundamental Concepts Scaling-Up]
-in propulsion revisited & redefined 24, 29, Inlet
56-77 -PST 24,29, 104, 143,240-250
-in combat Revisited & Redefined [see -vectorable 50-55, 240-251
Combat] -performance limitations 241-250, 24, 29,
- in agility and maneuverability/ 89, 107, 144, 175,226
controllability Revisited & Redefined [see -suction force/moments 96, 97
Agility, Supermaneuverability, IR Signatures 10,29,45-50,58, 143, 147, 154,
Supercontrollability] 189
Ivry David 15
Gal-Or, A. 3, 16
Gal-Or, B. (103, 119, 174, 176, 179,212,224, Jetborne Flight Control 24, 29, 80-104
225) 5, 6, 16, 26, 62, 110, Ill, 112, 121, 131, Jet Engines
132,140,142,146,149,153,154,176,181, -design limitations 9, 22-29, 89, 115-128,
194,209,238,239,241,243 194-213
Gal-Or, G. 3, 16 -flame-out vs. flight control 29,81-104
Gal-Or, L. (146) 3,196 -performance trends 115-128, 194-213, 9,
Gatlin (155) 85, 171 22-34
Glazer (157) 229, 230 [see also Engines]
Glidewell (5, 14, 142, 198, 205) 192 Joshi (124, 140, 161) 192
Goetz(41, 77) 192 Joubert 155
Goldsmith (133) 244 Jet Propulsion Laboratory of the TIlT 140,147,
Golijow, E. 3, 15 149, 238
-test rig systems 139-151,238
HarrierTypeTechnologies 57,81,181,194,198 Jet Wing 191
[see also STOVL]
Headwind Effects 243 Kalidasa 258
Heat Transfer 108-117, 204 Ketter, F. 15
274 Index

Killing Ratio 37, 44 -TV 10,29,40,42,51-55,57-71,126-127,

[see also Combat] 215-239
K1afin (220) 183-6, 192 - Nozzle-to-Nozzle Spacing 158
Kotansky (128) 165-167, 192,248,249 - Integration with airframe 10, 29, 40, 43,
Kucher (63, 138) 124, 130, 192 51-55,57-7I,119-12~146,151-17I,180
-aerodynamics (see Aerodynamics)
- coefficients reexamined 62-77, 112,
Laboy (184) 35, 37, 40 113-128,215-221
Landing of Pure Vectored Aircraft 30, 43, 96 -drag 74-75
Laughrey (4,13) 192 -control rules revisited 63-77, 112-120
Leakage losses in engine nozzles 64, III (see also Thrust vectoring and Vectored
Leavitt (58,112,162) 192,215-219 Aircraft)
Lift -cooling and leakage 64, 108-117, 120
-maximum 25, 101, 103, 127, 171 -temperature distributions 110, III
-line 101, 103 -aspect ratio 114-120, 157-163
-coefficient 25, 127, 169, 250 -length and shape 152
-gains by TV [see Supercirculation]
Lindsay, W. 15
Observability
Load factors 29, 38, 39, 97, 102, 103
(See stealth and RCS, IR and Thrust Vec-
(see also Structural Limitation)
tored Fighters)
Lynch (184) 35, 37, 40
ONERA 252
Optical Signatures
Mach-Number Effects 44,46,51,53,65,66,74, (See stealth and RCS, IR and Thrust Vec-
75, 98, 100, 101, 103, 117-123, 164, 165, tored Fighters)
166,171
Maiden (32, 34, 36, 40) 192 Pa1cza (6,41,43,46,73,75) 189, 192
Maneuver Paulson (15,16,27,29,148,155)85,171,192
-analysis 95, 114-151 Paugachev's Cobra 29,97, 167,211
- PST (see PST and Supermaneuverability) Peirce 78, 155
- PSM (see PSM) Pendergraft (62, 234) 192
-RaNPAS (see RaNPAS) Petit (40, 42, 68, 77) 192
Marble (177) 231-235 Pitch control, See Yaw-Pitch-Roll Control
Mashiach, E. 3 Porcher, C. 15
Mason (38, 39, 40, 52, 208, 209, 215) 26, 182, Popper, K.R. 4
192 Post ATF - See ATF
Mello (128) 165-167, 192,249 Pure sideslip Maneuvers - PSM 10,24,25,27,
Miller (8, 9, 47) 168,192,229 30,31,45,85-88,92-93, 102,241
Missiles 23,27,34,58,92, 133-154, 189 Post-Stall Technology - PST 10,22,24,25,27,
-cruise 45-50, 58, 83, 92, 98, 133-154 28,29,33,34,36,37,39,41,45,49,50-55,
-all aspect 23, 34 88-89,92-104,141,178-186,243
-anti-radar 134-137 PST Inlets 10, 22, 23, 29
-stealth 133-154 Power Lift 186, 190, 191
Mission Definition 29, 57, 161 Pure Vectored Aircraft - PYA 24, 29, 43,
-New 29,57, 134, 136 80-104, 139, 141
- Trends 57, 134, 145 (see also Combat)
Murff, J. 15 Ramjet/Scramjet Turbo Engine 198, 256
Rapid-Nose-Pointing And Shooting
Nash (45, 46) 192 (RaNPAJS) 10,24,25,27,29,30,31,32,34,
Nozzle 10,29,40,42,56-71,157-171,180,197, 36,38,39,44-55,178-180
212-214 Rasputnis, A. 3, 15
Index 275

RIC 10 -limits 97, 100-103

RCC 10,47, 108, 109, 115, 127, 130 SFC 10, 131-2, 195,206,213,256
R&D cycles 10,57, 139-151 Straight (48) 68, 109, 113, 115, 116
-& design (cf. Design) Stevens (49, 70) 124, 192, 229
-tools 10-11,22,136,138-151 Structural Limits 97, 103 (see also Load
-strategies 55,57, 139-151 Factors)
Richey (6, 73,210) 5, 189, 192 Su-27 29,97, 167, 211, 247
Robot Aircraft 133-154 Su-27 and Thrust Vectoring 97, 167, 171,211,
-as missiles (see missiles) 247
-as RPVs (see RPVs) Survivability 22, 92-104, 189
-as Cruise Missiles (see Missiles) -Swirling Flows in TV Nozzles 231-235
-design 106-107, 133-154 Synthesis 10, 164
Rogala, E. I 5
Roll Control (see Yaw-Pitch-Roll)
RPVs 9, 11,22, 33, 38, 58, 85, 104, 133-154, Tamrat (185,213) 27,30,31, 192
244, 255 Takeoff 30, 43, 95, 104, 252
-as R&D tool 9, 11,22,38,44-55,85 Tailwind Effects 243
-stealth 10, 11,29, 38, 49-55, 58, 85 Telemetry 10, 142
-flight testing (see Flight Testing) 139 Thrust 65, 66, 206, 250
RR 15, 199,201-203,209,210 -ideal 70, 119-120
-loss variations 89, 119-120
Santayana 56 -maximum dry 206, 162
Scaling-Up 152, 153, 244 -maximum AB 206, 162
(see also Dimensionless Numbers, Flight Temperature Profiles 110-118, 236-239
Tests, Inertia, Stability, Computer, Thermodynamic Stations 246
Simulations) Time Lags 41, 50, 225
SERN Nozzle 72 Thomas (95, 96) 192
Seddon (133) 244 Thrust Vectoring 23-55, 57-77, 128, 189,
Schneider (218) 44, 45, 192 215-239
Skow (184) 35, 37, 40 - Internal 22, 38, 57-71, 60-77, 78-104,
Specific Excess Power (SEP)95, 98, 101 178-183
Sedgwick (164) 192 -External 22,67,178-183
Sobota (177) 231-235 -range 128
Soviet Vectored Aircraft 26, 27, 33, 97 Thrust Reversal 40, 53, 59-61, 66, 69, 70, 71,
-R&D 26,27,33,97,167 91,93,97,99,119,224-231
Stability 142, 148, 229 Thrust Vectored Fighters 22,29, 37, 39,45-50,
(see also Flight Tests, Scaling, Inertia) 128, 189,242 (see also RPVs)
"Stall Barrier" 23, 36, 243, 247 -and low observability 45-50
ST()L 10,22,24,25,28,30,82,90,92,146,152, -Cruise Missiles 45-50, 58, 83, 92, 98,
165-7, 170, 186,249 133-154
ST()VL 10, 24, 25, 30, 79, 86, 146, 186, 193, -Anti-radar missiles 45-50, 134-137
198-199,200-203 -simulations 10,38, 183-187
Supercirculation 10, 29, 69, 80, 82, 117-122, -Nozzles 40
126-130, 157 -flight control 22, 29, 32, 51-55 (see also
Supercontrollability 29, 52-55 (see also Yaw-Pitch-Roll Control)
Supermaneuverability) -trainers 176
-Limits 97, 100-103, 178-180, 189 - helicopters 254, 256
Supermaneuverability 10, 22, 24, 25, 28 (see Training (see Educational Shifts and TV
also PST) trainers)
-&controllability 10,32,52-55,29,92-104, Turbofan-RAIMissiles 134-5
178-180, 189 Turbojet-RA/Missiles 135-6
276 Index

Turgeman, M. 3, 15, 16, 146 Vorobeichik, S. 3

Turn 48 Vorticity studies in TV Nozzles 231-235
- Minimum time 44, 45, 48 VTOL 10, 22, 30, 86
-rate 100-103,255 V/STOL 10, 30, 86, 92, 146
Vulnerability 96-97
Upper-Surface Blowing (USB) 190, 191
Upgrading Extant Fighters 163, 165-171 (see Watt (218) 44, 45, 192
also RPVs, R&D Cycles) Well (211) 192
Upgrading Extant Trainers 176 Wheeler, J .A. 4
Wind-tunnel simulations 9, 38, 51, 139,
Variable Cycle Jet engine 197,199,212-214 173-175
Vectored aircraft 29, 46-55, 56-77, 78-104, -limitations of 153
156-171, 206 (see also Pure Vectored Wing-nozzle integration 29, 119-122, 146,
Aircaraft-PV A) 51-71,43,156-171
- RPV (cf. RPV) 33, 49
-and stealth 45-50 Yaw-Pitch-Roll Thrust Vectoring, 23,26,29,33,
- canards 80-82, 156-171 36,71,76,79-104,167, 169
-supermaneuverability (see -control surfaces 29, 52, 76
supermaneuverability) - pitch-only TV 26, 51, 59-63, 70, 71, 85
Vectoring Angles 128,129,171,215,219 Yawing moments 84, 86, 90, 230 (see also
Viffing 83, 99 Yaw-Pitch-RolI)