Airplane Design Manual

Airplane
Design
Manual
FREDERICK K^TEICHMANN
Professor of Aeronautical Engineering
Assistant Dean, Day Division
College of Engineering, New York University
FOURTH EDITION
PITMAN PUBLISHING CORPORATION
NEW YORK TORONTO LONDON

Generated on 2012-05-30 00:39 GMT / http://hdl.handle.net/2027/mdp.39015000500895
Public Domain, Google-digitized / http://www.hathitrust.org/access_use#pd-google
Engin. Library
TL
G7/.2
3 SB Copyrioht, 1939, 1942, 1950, 1958
BT
PITMAN PUBLISHING CORPORATION
All rights reserved. No part of this book
may be reproduced in any form without
the written permission of the publisher.
4.1
cop. 2.
Associated Companies
Sir Isaac Pitman a Sons, Ltd.
London Melbourne Johannesburg
Sir Isaac Pitman & Sons (Canada), Ltd.
Toronto
Printed in the United States of America

Preface to the Fourth Edition
This book grew out of notes prepared for airplane design courses at the
Daniel Guggenheim School of Aeronautics, College of Engineering, New
York University more than twenty years ago. The field of airplane design
has undergone enormous changes in these years and will continue to do so.
Thus the book has been expanded to encompass recent developments,
thereby making it more effective and meaningful to the present-day
student.
In addition, an attempt has been made in this revision to amplify the
analytical approach to design problems as well as the purely empirical
approach. The student of airplane design is in this way offered a deeper
appreciation of the interplay of aerodynamics, structural analysis, human
considerations, and other such factors.
Although practical necessity restricts the treatment that can be given
to each of the various fields, it is hoped that the student may be stimulated
to refer to literature that is available elsewhere.
The objectives of aircraft design work in the college classroom are:
(1) To offer an integration of or focal point for applying the various
principles included in aerodynamics, structural design, installation require-
ments, and application of materials. If time permits, economics, perform-
ance calculations, and allied problems continue the understanding of the
design concept.
(2) To afford a basis for stress analyses.
(3) To provide some drafting experience.
(4) To develop an "engineering sense" in the student, enabling him to
evaluate various requirements, judge the necessity of compromise, and
know the amount of time needed to achieve a given objective.
This book has been designed to aid both the teacher and student of
airplane design to effectively meet and fulfill these important goals.
Frederick K. TeichmaHn
vii
viii
AIRPLANE DESIGN MANUAL
Preface to the Third Edition
Although airplane design has progressed tremendously since this book
was first published, the fundamental approach to the problem of airplane
design has not changed. However, the possibility of attaining transonic
and supersonic speeds with aircraft, a hope seemingly very remote only a
few years ago, has introduced additional considerations in designing even
the smallest detail. The student may not find all the needed answers in
the new edition but the way to his desired goal is indicated. It is up to
him to make the best use of material at hand, either in this book or col-
lateral reading.
The author greatly appreciates the work in redrawing and preparing
new sketches done by Messrs. Dong, Waxman, and Wood, his former
students.
Frederick K. Teichmann
Preface to the First Edition
This book has been written to fill what appears to the author to be a
gap in aeronautical literature, an introduction to the art of airplane de-
sign, with the needs of the student, the young engineer, the draftsman and
the student working on his own especially in view. While aerodynamics,
stress analysis and other aspects of airplane design have been covered many
times, experience in dealing with senior aeronautical students has shown
that such men experience considerable difficulty in coordinating their
knowledge and efforts in approaching the difficult problem of actually
beginning the design of a new machine, and carrying on the work system-
atically. From time to time notes have been prepared for student use
and these have gradually evolved into the present work.

PREFACE
ix
In view of the rapid growth and complexity of the subject, it is too much
to hope that the entire field has been adequately covered; still teaching
experience indicates that such a manual is helpful to instructors and stu-
dents alike.
It is of course expected that the student shall supplement the present
text by investigations of his own, by studying the latest designs at the
airport, or from descriptions in the technical press, or by study of the
numerous research publications published by the Government Printing
Office and the great engineering societies, even though an attempt has
been made to make each chapter of the book as complete in itself as
possible.
The author wishes to thank the following companies (among others) for
permission to use illustrations: The Pratt & Whitney Aircraft Co., Pioneer
Instrument Co., R.C.A. Manufacturing Co., Inc., The Cleveland Pneu-
matic Tool Co., The Goodyear Tire & Rubber Co., The B. F. Goodrich
Rubber Co., The Firestone Tire & Rubber Co., Bendix Products Corpora-
tion, Harrison Radiator Corporation, Eclipse Aviation Corporation, Kolls-
man Instrument Co.
Thanks are also due to Mr. Robert Boyer and Mr. Leonard Mihalov-
sky, N.Y.U.'37, who kindly permitted the use of their class designs, and
to Professor Alexander Klemin, who established the course in airplane de-
sign in 1924, in a form which has stood the test of time, for many valuable
suggestions and criticisms.
Frederick K. Teichmann
June, 1939
Contents
Prefaces
, Chapter I. Procedure in Design
Selecting Type of Airplane .
Power Plant
Payload and Crew
Performance Requirements
Step-by-Step Procedure .
Nomenclature
References
i Chapter II. Types of Airplanes
The Biplane
The Sesquiplane
The Multiwing Airplane .
The High-Wing Monoplane .
The Low-Wing Monoplane
Special-Purpose Airplanes
Factors Affecting Choice
Performance
Landing-Gear Retraction .
Structure
Special Features
Sample Airplane Data Sheets
Empirical Data
Airplane Data
8 Chapter HI. Airfoil Selection .
The Airplane in Rectilinear Flight
Horizontal Flight
Gliding Flight
The Dive
The Climb
Range
Other Airfoil Characteristics
Structural Considerations
Recapitulation
Aspect Ratio Corrections
Aerodynamic Section Characteristics .

Airfoil Construction
Sweepback for High-Speed Airplanes

Laminar-Flow Airfoils .
Compressible-Flow Airfoils
xi
xii AIRPLANE DESIGN MANUAL
Chapter IV. External Loads on an Airplane in Flight .... 50
Load Factor 50
The Airplane in a Maneuver 51
The Airplane in a Gust 52
Experimental Determination of the Load Factor .... 55
Airplane Categories 57
Empirical Load Factors 57
The Flight Envelope 58
Reference Axes 59
Generalized System of Forces 61
Other Dynamic Loads 63
Chapter V. Materials of Construction 64
Application of Available Materials 64
Aluminum and Aluminum Alloys 65
Classification and Nomenclature 66
Sheet 67
Tubing 68
Extruded Shapes 68
Forgings 68
Castings 70
Wire, Rod, and Bar 70
Rivets and Screw-Machine Products 70
Airframe Fabrication 71
Cutting and Blanking 72
Forming 72
Magnesium Alloys 74
Castings 74
Forgings 74
Extrusions 74
Sheet 74
Steel 74
Titanium Alloys 75
Hardware and Other Items 76
Chapter VI. Detail Design Considerations 81
Design Philosophies 81
Margin of Safety 81
Maintenance Requirements 82
Fool-Proof Operation 82
The "Fail-Safe" Principle 82
Safety 82
The "One-Horse Shay" Principle 83
Strength of Metal Aircraft Elements 84
Structural Behavior . . 84
Stiffeners or Stringers 86
Panels under Load 88

CONTENTS xiii
Panel Sizes 89
Strength of Sheet-Stringer Combinations 92
Equivalent Structure 93
Cut-Outs '94
Riveted and Bolted Joints 95
Fittings 97
Determining Bolt Diameter 99
Determining Thickness of Fitting Lug 99
Determining Value of R 99
Checking whether Dimension R is Sufficient 99
Sandwich Materials 99
Flooring 100
Thermal Problems 102
Miscellaneous 103
Lightening Holes 105
Beading 105
Stiffeners 105
Chapter VII. The Three-View 113 >
Steps in Assembling Preliminary Data for Three-View .114
Estimation of Gross Weight 114
Estimation of Wing Area 116
Determining Length of Span 117
Drawing Up the Wing Planform 118
Determination of the Mean Aerodynamic Chord . .118
Determining Length of Airplane 118
Locating the Engines 119
Planform of Horizontal Tail Surfaces 120
Completing the Top View 120
Completing the Side View 121
Completing the Front View 121
General Notes for Three-View 121
Engineering Studies 122
Final Three-View 122
Chapter VIII. Preliminary Weight Estimate 124
General Procedure 124
Estimated Weights 126

Calculated Weights .126
Actual Weights 127

Alternate Loads 127
Estimating the Gross Weight 127
Estimation of Structural Weight 128
Recapitulations 128
Form for Preliminary Weight Estimate 128
Variables Affecting Weight Estimation 132
Wing Weight Estimates 132

xiv
Fuselage Weight Estimates 136
Control Surface Weight Estimates 137
Landing-Gear Weight Estimates 138
Power Plant and Power-Plant Nacelle Weight Estimates . . 138
Weight Data Sources 139
Design Control of Weight 139
Empirical Formulas and Data 139
Wing Weight 139
Tail Surfaces 141
Fuselage Weight 142
Landing Gear 143
Total Fabricated Components 145
Weight Empty 146
Engine Nacelles 146
Fuel Weight 147
Oil Weight 147
Power Plant Weights 147
Gross Weight 148
Miscellaneous Weights 149
Propeller Weights 149
Chapter IX. The Balance Diagram 155
General Procedure 155
Center of Gravity Location 157
Practical Solution 160
Center of Gravity Movement 160
The Fuselage and Its Contents 161
The Wing and the Landing Gear 164
Supplementary Calculations 168
Ballast 169
Centers of Gravity of Individual Items ....... 169
Center of Gravity Estimation 170
Error in Balance Calculations 171
Chapter X. The Cockpit 172
Suggested Procedure in Design 174
General Requirements 175

Windows and Windshields 176
Characteristics of Glass 180

Vision and Visibility 182
Canopies 183
Seating 183
Exits 186
Protection for the Pilot 186
Ejection Equipment 187
Parachutes 188
Controls 188
Instrument Board 189

CONTENTS xv
Chapter XI. Instruments and Equipment 191
Instrument Board 193
Location 194
Grouping 194
Variety of Instruments 194
Selection of Instruments â€¢ 197
Electrical Equipment 198
Safety Equipment 199
De-Icing and Anti-Icing 199
Chapter XII. The Passenger Cabin 202
General Considerations 202
Comfort Factors 203
Cabin Dimensions 203
Passenger Seats 205
Seating Arrangements 206
Headroom for Small Airplanes 207
Headroom for Large Airplanes 209
Leg Room 209
Side-by-Side Arrangements 213
Staggered Arrangements 213
Back-to-Back Arrangements 213
Seating Facing Rearward 214
Vision 215
Seating Comfort 215
Seating Accessibility 215
Center of Gravity Considerations 216
Windows 216
Doors and Exits 217
Sleeping Accommodations 218
Lighting .218
Furnishings 218
Provision for Airsickness 218
Flooring 219
Toilets 219
Refreshments 220
Baggage Compartment 220

Chapter XIII. Air Conditioning 221
Physiological and Psychological Considerations 221

Air Movement for Comfort 222
Air Requirements 223
Pressure Considerations 224
Design Temperature Limits 226
Physical Conditions 226
Air Ducts 227
Boilers and Radiators 229

xvi AIRPLANE DESIGN MANUAL
Heat Sources 229
Pressure Cabin Equipment 229
Calculations 230
Heating Surface 232
Air Conditioning Problems at High Speeds 233
Chapter XIV. Soundproofing 235
Measure of Noise 235
Sources of Noise 236
Effect of Frequency on Soundproofing 236
Soundproofing Materials 237
Application of Materials 237
Noise Due to Jet Engines 239
V Chapter XV. The Propeller 241
General Propeller Characteristics 241
Aerodynamic Effects of Propeller 243
Gyroscopic Effect 243
Propeller Pitch 244
Number of Blades 245
Propeller Influence on Aircraft Configuration 247
Propeller Clearance 247
Asymmetrical Conditions 249
Effect of Engine Torque 250
Tandem Engines 251
Pusher Installations 252
Tail Installation 253
Spinners and Cuffs 254
Propeller Selection 254
Empirical Formulas 258
Chapter XVI. The Power Plant 263
Location 264
Submerged Engines 264
Number of Engines 265
Engine Rating 267
Engine Selection 268
Engine Nacelles 268

Fuel Consumption 268
Engine Mountâ€”Non-Jet Engines 269

Firewall 272
Cowling â€¢ 273
Exhaust Manifoldsâ€”Reciprocating Engines 276
Carburetor Scoop Design 277
Fuel Systems 278
Pumps 279
Tanks 279
CONTENTS xvii
Lubricating Systems 280
Tanks 281
Air Inlet Systemâ€”Turboprop Engines 281
Induction System Configuration 281
Design Considerations 281
Inlet Losses during Ground Operations 286
Engine Inlet Anti-Icing Provision 286
Anti-Icing Water Runback into Engine Inlet 286
Air Inlet Systemsâ€”Turbo-Jet Engines 287
The Wing-Root Inlet 289
Nose Inlet in the Fuselage or Nacelle 290
The Nacelle or Pod-Type Installation 290
The Annular Inlet 292
The External Scoop 292
The Flush Inlet 294
Rockets .298
Chapter XVII. Design of the Wing 299 1
Wing Layout Procedure 304
Mean Geometric Chord 306
Planforms and Taper Ratios 310
Sweepback for High-Speed Airplanes 313
Effect of Wing Sweepback 317
Dihedral 317
Combination Dihedral and Sweepback 318
Angle of Incidence 319
Wing Loading 320
Aspect Ratios 320
Airfoil Thickness and Thickness Ratios 321
Wing-Fuselage Configurations 323
Metal Wing Construction 323
Spars and Their Location 327
Spanwise Stringers and Their Location â€¢ 330
Ribs and Their Location 331
Wing-Fuselage Attachments 334
Fabric-Covered Wing Construction 337

Ailerons 341
Lift-Increase Devices and Spoilers 347

Wing Fences 354
Effect of Wing Mass Distribution 354
Effect of Wing Loading 355
Flutter Prevention 355

xviii AIRPLANE DESIGN MANUAL
Chapter XVIII. The Landing Gear 357
Dynamic Loads 358
Means for Landing-Speed Reduction 359
Dissipation of Energy 361
Shock Absorbers 362
Classification by Landing Contact 364
One-Point Contact 364
Two-Point Contact 364
Three-Point Contact 365
Four-Point Contact 365
Landing Gearâ€”Tail-Wheel Type 365
Landing Gearâ€”Nose-Wheel Type 367
Track-Type Landing Gear 370
Tandem Gear 371
Cross-Wind Landing Gear 371
Shimmy and Shimmy Dampers 373
Tread 375
Wheel and Tire Size 375
Size of Tail or Nose Wheel 376
Wheel Position 376
Retraction of Landing Gear 376
Special Problems 380
Chapter XIX. Tail Surfaces 382
Longitudinal Stability Considerations 382
Definitions 382
Preliminary Calculations for Static Longitudinal Stability . . 384
Adequacy of Static Stability 384
Flight Criteria for Stability and Controls 385
Longitudinal Stability and Control 385
General Requirements 386
Control Surfaces 387
Airfoil Sections 388
Aerodynamic Balance 389
Trailing-Edge Tabs and Other Devices 391
Static Balance 391

Dynamic Balance 392

Flutter Prevention 393
Construction 394
Horizontal Tail Surfaces 395
Location 395
Angular Deflection 397
Adjustable Stabilizer 397
Aspect Ratio 398
Angle of Incidence 398
Sweepback 400
CONTENTS xix
Dihedral 401
Area 401
Construction 401
Planform .401
Other Solutions 402
Clearances 403
Vertical Tail Surfaces . . - 403
Directional Stability and Control 404
Location 405
Angular Deflection 406
Aspect Ratio 406
Area 407
Planform 408
Sweepback 408
Tail Length 409
Butterfly or Vee Tail 409
Chapter XX. Control Systems 412
Typical Systems 412
Pulleys 414
Cables 414
Fairleads 415
Stops 415
Differential Ailerons 419
Adjustable Stabilizer 419
Tab Controls 420
Flap Controls 420
Wing Flaps 422
Tabs 422
Hinges 422
Flap- and Tab-Control Loads 423
Hydraulic and Pneumatic Systems 423
Travel of Controls 425
Irreversible Controls 426
Detail Requirements 427
Controls 427
Engine Controls 427

Electrical Systems 429
Chapter XXI. The Fuselage 430

Wing-Fuselage Considerations 430
Shape of Fuselage 432
Fuselage Length 435
Determining Fuselage Lines 435
Use of the Mockup 436
Analytical Studies of Fuselage Structure 437

XX
Frames and Their Location 440
Longitudinal Stringers and Their Location 444
Fuselage Skin 446
Windows 447
Pressurized Cabins 447
Flooring 449
Doors and Exits 449
Access Doors 449
Baggage Compartments 450
Twin Fuselages 450
Tail Booms 450
Alternate Type Structures 451
Effect of Fuselage Mass Distribution 453
Chapter XXII. Preliminary Performance Calculations. . 454
Source of Data 454
Calculations for Horsepower Required 455
Altitude Corrections 459
Arbitrary Standard Atmosphere 459
Parasite Resistance Data 460
The Engine 461
Horsepower Available 462
Maximum Speed 466
Rate of Climb 468
Absolute and Service Ceilings 468
Range 468
Performance Requirements 470
Take-off Performance with All Engines Functioning Normally . 470
Performance in Air with All Engines Functioning Normally . 471
Performance in Air with One Engine Dead 472
Landing 473
Empirical Formulas 474
Maximum Speed 474
Minimum or Stalling Speed 475
Rate of Climb at Sea Level 475
Absolute Ceiling 476

Range 476
Index 479

CHAPTER I
Procedure in Design
No task can be intelligently executed unless a definite goal has been set
and a line of attack or orderly form of procedure has been adopted. There
may be different ways of obtaining the same objective, but mistakes and
unnecessary work will be avoided if a definite plan is made before any real
work starts.
The responsibilities of the designer are many. Not only must he meet
the structural requirements, but also, by proper design, the operational
and performance specifications, and he must be able to produce an air-
plane that is economical and safe. For example, it is claimed that two
thirds of the responsibility for aircraft accident prevention lies within the
job of the aircraft designer. Such responsibilities require constant vig-
ilance in checking all phases of the design and in keeping abreast of all
the latest developments.
SELECTING TYPE OF AIRPLANE
It is not sufficient to say "Let's build an airplane." The question is:
What kind of airplaneâ€”an open or a closed type, a sleek racing mono-
plane, or a large flying boat? The first thing to be done is to write down
a set of such definite specifications that any designer who receives them
may be able to design an airplane which meets the original design pro-
poser's intentions. The procedure is much the same as that of buying a
family car. The term "family car" immediately sets one specification:
the buyer knows that he is not going to get a truck, or a roadster, or a
racing car. The price that the buyer can meet will set another specifica-
tion automatically, and so it goes.
Specifications for an airplane are far more comprehensive. Consider,
for example, the type of airplane it may be. The airplane to be designed
may be one of two conventional types, a monoplane or a biplane. If it is
3
4
a monoplane, the wing may be unsupported externally, in which case it
is known as a full cantilever monoplane; or the wing may be externally sup-
ported either by struts or by wires, in which case it is known as a semi-
cantilever monoplane. Moreover, the wing may be placed at the bottom
of the fuselage, when it is known as a low-wing monoplane; or the wing may
be placed halfway between the top and bottom of the fuselage, so that the
airplane is a midwing monoplane; or again, the wing may be at the top or
above the fuselage, in which case the airplane is known as a high-wing or a
parasol monoplane, respectively.
The same variables apply to a biplane. The two wings may not have
the same areas, or the same planform, or the same airfoil. There may be
large forward or positive stagger of the upper wing relative to the lower,
and perhaps more dihedral for one wing than for the other. The com-
binations are almost infinite especially when one considers that changes
may be made in structure, in materials, in planform, in stagger, in angle
of incidence, in airfoil sections, in decalage, in gap-chord ratios, in wing
placement relative to the fuselage, in distribution of wing areas, and a host
of other variables.
The variables just noted apply only to the wing. Consider the fuselage.
It may be round, oval, square, elliptical, rectangular, or a combination of
these cross sections. It may be shallow or deep; it may be wide or nar-
row; it may have an open cockpit or an enclosed cabin; it may be con-
structed of almost any material and in an infinite number of ways. For
each material and specific function, there is a definite, desirable shape of
fuselage.
The landing gear also offers enormous latitude in design. It may em-
ploy a landing gear having two wheels forward with a tail wheel rearward;
or the reverse order with a front or nose wheel and two wheels slightly
rearward, popularly known as the "tricycle" landing gear. The landing
gear may have a through-axle of the type used during the early period of
airplane design, or a split-axle type developed later. Moreover, it may
be nonretractable or retractable.
These are just a few indications of what the design trend might be.
Familiarity with different types of airplanes will help the potential designer
in determining the type best suited to a specific duty. The designer will
find it decidedly advantageous to read as widely as possible in the various
technical aeronautical publications and to collect, study, and correlate

design details. Thereby his facility in adaptation will be improved.
POWER PLANT
The power plant will be discussed in detail in a subsequent chapter.
However, a brief discussion of the place the power plant takes in the origi-
nal specifications may not be amiss here. In many cases, operating com-
PROCEDURE IN DESIGN
panies of aircraft may specify the type and number of enginesâ€”either be-
cause of known fuel economy, or efficiency and dependability under certain
operating conditions, or because of possible interchangeability with exist-
ing equipment.
An airline accustomed to maintaining and operating radial air-cooled
engines will be loath to use turboprop engines, for example, since its
personnel may not be trained or sufficiently experienced to handle the
new type of engine.
The reasons for choosing a certain engine may be many, and the section
on power plants should be studied before writing the specifications. The'
specifications may designate a particular engine, although it is more likely
that the number of engines will be designated, for it is quite possible to
obtain one engine or two engines delivering the same total horsepower.
PAYLOAD AND CREW
The payload includes all load from which revenue is obtained. It in-
cludes passengers, mail, baggage, and express. The crew includes pilot,
co-pilot, mechanics, navigators, radio men, stewards, and any other em-
ployee required for specialized work.
Military airplanes have a different type of payload, usually called fixed
equipment or disposable load, as the case may be. This consists of guns,
ammunition, bombs, and other military equipment. Special provision
must be made for these; therefore, these items have a definite bearing on
the airplane type as well as the weight permitted.
The gross weight of the airplane is largely dependent upon the require-
ments for payload and crew. It should be quite obvious that if a crew of
three (a pilot, a co-pilot and a radio man, for example) is required, some
provision must be made for it, and such provision will affect the size of
the cockpit as well as the fuselage and eventually the gross weight. Like-
wise, provision for mail and express will be entirely different from provi-
sion for passengers.
The larger the aircraft becomes, the larger the crew is likely to be.
Some indication of the eventual size of the airplane to be designed can be
gained by looking at the size of crew required. In the same way, the
number of passengers carried has a direct bearing on the size of the fuse-
lage and the gross weight: the greater the number of passengers to be car-
ried, the larger and wider and higher the cabin, and therefore the fuselage
will be larger. Not only does the increased number of passengers increase
the weight, but the structure will also weigh more because of increased size.
Actually, the gross weight of the airplane can be estimated if the weight
of the payload, crew, fuel, and oil are known since an analysis of a large
class of airplanes shows that there is a definite relationship between the
two weights.
6
It is very important to know as much as possible about the load the
airplane is to carry because these are the items for which the designer has
to make proper provision although he may have little or no control over
their weight, size, or location in the airplane.
PERFORMANCE REQUIREMENTS
Unless the airplane is designed for private use, the performance require-
ments are set by the ultimate purchaser. It takes but little thought to
realize that the keen competition among American airlines requires the
speed of the airplane to be as high as possible in order to obtain attractive
schedules. But where there is less competition, a far slower airplane
may be desirable because of the smaller horsepower and less fuel required
to carry practically the same load.
Likewise, an airplane operating over mountainous territory will need a
high service ceiling in order to clear the mountains, whereas a low service
ceiling would do over low level country.
The performance required for the airplane will have a direct bearing on
the number, type, and horsepower of the engines, as well as the type and
design of wing, fuselage, and perhaps landing gear. The ultimate criterion
of a good airplane is its performance in relation to the load carried and the
conditions to be met.
STEP-BY-STEP PROCEDURE
The foregoing discussion deals with specifications which are only part
of the work to be considered in designing the airplane. From the moment
a new design is contemplated until the final drawing leaves the drawing
board, a definite plan is followed in evolving the design. The individual
steps of the plan may not always be clear-cut, and sometimes several
phases are carried along in parallel sequence. The following procedure
may be gainfully employed.
1. Study of specifications to fix the more important items having im-
mediate import in the preliminary design steps.
2. Study of similar purpose airplanes to determine the possible types
which may be considered as meeting the over-all specifications.
3. Power plant survey for the selection of the likely engine or engines
to be used in the design.
4. Preliminary three-view, or views, to narrow down the number of

possible solutions and to present a framework on which the work that fol-
lows will depend.

5. Preliminary weight estimate based upon the data and information
obtained in the first four steps.
6. Airfoil selection to obtain the ultimate performance desired.

PROCEDURE IN DESIGN
7. Balance diagram to fix the items of equipment and structure in
proper relation to each other for purposes of design, stability, and effec-
tiveness.
8. Inboard profile to check upon installation of equipment and to pro-
vide studies of interior arrangement.
9. Structural layout (work on the various units usually carried on
simultaneously in order to take proper care of the interrelation of the
component parts) somewhat in the following order.
(a) Wing with reference to fuselage.
(b) Landing gear with reference to wing or fuselage.
(c) Tail surfaces with reference to fuselage.
(d) Fuselage.
(e) Power plant with reference to wing or fuselage, or both.
10. Final three-view from data obtained in the course of working on
balance diagram, inboard profile, and structural layouts.
11. Preliminary longitudinal, directional, and lateral stability calcula-
tions performed at time of airfoil selection, balance calculations, and three-
view conception.
12. Preliminary control calculations made along with preliminary sta-
bility calculations.
13. PreUminary performance calculations carried along simultaneously
with other calculations.
14. Preliminary stress analysis according to military or civil require-
ments.
15. Revised structural drawings with added information for the prepa-
ration of detailed design drawings.
16. Check all parts of the design and revise where necessary.
These steps are discussed in considerable detail in subsequent chapters.
More familiarity with the problem of airplane design will often suggest
alternative procedures.
In practice, practically all these steps are carried along simultaneously
because a number of men may be employed on the project, but even so the
initial work is usually done by one man. In practice, too, a mock-up of
the proposed design is made so that many design studies may be made on
the mock-up rather than on paper. In the classroom, suitable compro-

mises must be made.
NOMENCLATURE
In any discussion, it is necessary to understand the words used. Ordi-
narily, it would be assumed that the student is familiar with the names of
all the parts of the airplane; certainly he should be reasonably well in-
formed on matters dealing with aerodynamics and the internal-combustion

PROCEDURE IN DESIGN
engine. However, it may be that he is not too well informed on names of
structural details. Figure 1-1 will be useful for reference and in establish-
ing some standardization of terms.
REFERENCES
Much research and design material may be found in technical reports,
memoranda, and notes issued by the National Advisory Committee for
Aeronautics (NACA) and published by the Government Printing Office
in Washington, D. C.
Young engineers should become familiar with the current literature to
be found in various trade journals of the industry and in the journals of
the several engineering societies.
Much of airplane design is empirical, at least in the project stage, so that
all the research that may be brought to bear on the design will be particu-
larly fruitful.
CHAPTER II
Types of Airplanes
Many considerations enter into the selection of a particular type of
airplane to meet a given specification, and, therefore, it is well to know
the characteristics of each type before deciding definitely on any one type.
Airplanes may be classified in various ways, according to structure, meth-
od of construction, number of engines, type of landing gear, weight,
purpose, and any other variation which an airplane may have.
In the specific descriptions given here, it should be borne in mind that
the advantages and disadvantages indicated for the various types have to
be properly evaluated since, in all, certain compromises must be made.
In some cases the advantages, when properly considered, are more impor-
tant than any possible disadvantage. In many cases, the general "eye
appeal" is also a deciding factor in the selection of the final design. Any
statements made in the discussion that follows should not, therefore, be
applied immediately to any existing design.
THE BIPLANE
Historically the multibay biplane was favored because the art of aero-
dynamics had not yet progressed to the point where the thick airfoil was
favored, much less visualized. Economical and light design indicated the
truss as the most convenient type of structure, although the multibay lift
truss eventually gave way to the single-bay lift truss. In a few cases, even
the bracing between the upper and lower wing became more and more sim-
plified so that either only the interplane strut remained or none at all.
The biplane lift truss makes for efficient structural design, small over-all
dimensions, and lends itself to a variety of solutions as well as permitting
the use of relatively thin airfoil sections.
Even though the braced biplane may permit use of thinner and lower
drag airfoils, still the resistance offered by the additional bracing and in-
terferences may more than offset the lesser wing drag.
10
TYPES OF AIRPLANES
11
The biplane is open to an almost infinite number of variations since any
or all of the following geometric arrangements are possible.
1. Different airfoils for the upper and lower wings.
2. Varying decalage, that is, the upper wing at a different angle of in-
cidence than the lower.
3. Different dihedral for the upper and lower wings.
4. Different aspect ratios for the two wings.
5. Any degree of positive or negative stagger.
6. Varying gap-chord ratios.
7. Different planforms for the two wings.
8. Ailerons on either upper or lower wing, or on both.
9. Different sweepback for the two wings.
10. Various possible bracing arrangements.
Figure I1-1. A typical biplane is shown. The crossed single lines represent the
lift and landing wires. The over-all dimensions of the airplane are smaller than for
the monoplane and there are more design variables to consider. Generally, the biplane
has more parasite and interference drag than a monoplane designed for the same purpose
All these variations, however, can also introduce a great amount of
work in preparing the design and in the process of manufacture, so that
the more variations the design has, the more expensive it is likely to be.
THE SESQUIPLANE
A biplane that has a lower wing considerably smaller than the upper
is called a sesquiplane. The reverse order in the size of the two wings
has also been used. Such a design may be resorted to in order to provide
adequate landing-gear attachments and to afford an opportunity to in-
crease the tread of the wheels. Bracing between wings may be employed.
Such a design may be of particular usefulness in rugged terrain where
ground stability in landing and taxying is particularly important, and
where the high wing will be out of the way of underbrush and other
obstructions.
THE MULTIWTNG AIRPLANE
As the number of wings is increased, the less aerodynamically efficient
the airplane becomes. Although multiwing airplanes, such as triplanes,
have been built, they may be considered as curiosities rather than as sound
designs to be critically considered.

12
THE HIGH-WING MONOPLANE
The design that became popular after the biplane was the externally
braced high-wing monoplane. When externally braced, the wing is said
to be semicantilever. (See Figure II-2.) The lift strut supports the
wing and thereby reduces the bending moments sufficiently to make for a
lighter structure. A thin airfoil, although somewhat thicker than an air-
foil employed on a biplane, still offers less resistance and permits reason-
ably high-speed performance.
The struts are attached (as all external bracing should be) by a single
bolt at each end (with the head of the bolt facing forward), so that the
normal loads, acting either upward or downward, will impose bending
moments and shear on the wing structure proper but induce only axial
loads in the struts or wires.
Fig. II-2 Fiq. II-3
Figure II-2. A semicantilever high-wing monoplane braced by a V strut. This
arrangement permits a simplified fitting design at the apex of the V. The arrangements
of the struts may be varied. Such bracing permits use of thinner airfoils and efficient
structural configurations.
Figure I1-3. A full cantilever high-wing design with a power-plant installation in
the tail where the effect of propeller interference and slip stream do not affect the
aerodynamic qualities of the wing.
In order to reduce the column length of the external struts, so-called
jury struts are interposed between the wing and the lift strut, just below
the upper end of the lift strut. The juncture of the jury and the lift strut
is a hinge or pin joint, and the upper end of the jury strut is also a pin-
connected fitting attached to the wing spar.
The chord loads are applied to the internal drag structure of the wing
and are assumed not to act on the lift bracing.
The lift struts may be placed parallel to each other in the most common
arrangement, although carrying the lift struts down to the fuselage to
form a V is often done to eliminate one fitting and perhaps to offer better
access to the door which may be located at the rear strut.
The V may be so arranged also that the apex of the V is at the rear-
strut fitting instead of the front-strut fitting. There are also other pos-
sible variations with the struts both carried farther forward to a point of
TYPES OF AIRPLANES 13
attachment on the fuselage. The advantage of this arrangement lies in
the possibility of combining lift-strut fittings with landing-gear strut fit-
tings, or in the possibility of applying reactions to the wing-drag truss to
counteract the chord force components exerted on the drag truss.
Fig. II-6 Fig. II-7
Fiqure II-4. A low-wing monoplane with a twin vertical tail surface arrangement
which may be used to reduce the over-all height of the airplane; or, to operate more
effectively in the slip stream of a twin-engine design; or, to avoid the "blanketing"
effect of the fuselage. Twin vertical tail surfaces may help to increase the "apparent"
aspect ratio of the horizontal tail surfaces.
Figure II-5. A midwing design with a jet-engine installation. The dihedral in-
corporated in the horizontal tail surfaces installation brings these surfaces into a more
uniform downwash distribution across the span. The air scoops of the jet engines are
in the jet engine nacelle, with the exhaust in the rear.
Fiqube II-6. A so-called butterfly tail combines the functions of the vertical and
horizontal tail surfaces in this midwing monoplane, thereby simplifying the empennage
structure.
Fiqube II-7. The inverted gull wing designed to raise the propeller axis of the
centrally located engine while still bringing the wing down to permit a landing gear
with short members suitable for retraction. Acute angles are also avoided at the inter-
section of the wing with the fuselage, thus eliminating need for fillets. The dotted lines
show the position of the wing when partially folded for stowage purposes.
The cross sections of these lift struts are usually symmetrical airfoils of
small thickness ratio and large fineness ratio. It is possible to envelop
both lift struts in an airfoil in order to add to the lift, but such additional
lift is comparatively small in usual designs.

14
Figure II-8. A jet engine design
with air inlets along the fuselage. A
single nose inlet in the fuselage is also
used. The exhaust is in the extreme
end of the fuselage tail. The wing-tip
fuel tanks are droppable although their
expense may prohibit such procedure,
and are so located to reduce aerody-
namic resistance; they may affect the
stability of the airplane because of in-
creased moments of inertia about the
axes of the airplane.
Figure II-9. A high-wing mono-
plane with a pusher-type engine install-
ation designed to reduce noise in the
cabin, to obtain a more favorable center
of gravity location for certain designs,
and, perhaps primarily, to afford the
best possible view forward.
Figure 11-10. A small pusher air-
plane with twin booms supporting the
tail surfaces. The booms help to iso-
late the engine but are brought about
primarily by the engine location.
Figure 11-11. A canard pusher-
type airplane of unusual design with
the horizontal tail surfaces ahead of
the wing and the vertical tail surfaces
at the wing tips.
The high-wing monoplane affords excellent vision downward, which is
v especially useful in landing. Vision upward and toward the sides is, of
course, impeded.
It is difficult to obtain a wide tread for the landing gear unless a long
supporting strut is carried to the wing, or a sturdy structure is built to
support the landing gear.

The retraction of the landing gear becomes particularly difficult in this
type of design because the wheels cannot easily be retracted into wells
located in the fuselage. It is even more difficult if the landing gear is
retracted into the wings because the struts would have to be retracted as
well.
The incorporation of flaps and, in general, various lift-increase devices
is made easier in the high-wing design since ground clearance with de-
TYPES OF AIRPLANES
15
Figure 11-12. A proposed design
with a jet actuated propeller. Since
the blade tips operate at rather high
speeds, a jet engine may attain a rea-
sonable efficiency at the propeller tip
location. The air scoops for the jet
engine are in the leading edge of the
root section of the wing.
Figure 11-13. A multi-engine
monoplane incorporating a triple ver-
tical tail surface arrangement in order
to obtain the necessary directional con-
trol, especially when one or more en-
gines fail.
Figure 11-14. This multi-engine
design uses a pusher installation in or-
der to obtain a more favorable center
of gravity location, to keep the plane
of propeller rotation clear of personnel
in the cabin, to simplify the engine
nacelle construction and engine ar-
rangement by placing the units in a
straight row, and to move the control
cabin as far forward as possible for
visibility.
fleeted or extended flap does not become one of the primary design con-
siderations.
As far as wing locations on the fuselage are considered, the high wing is
aercdynamically superior, for it has both greater lift and less aerodynamic
resistance than if the same wing were placed at the bottom of the fuselage.
While the internally braced wing or full cantilever wing is aerody-
namically better because of the absence of lift struts (see Figure 11-13), the
landing gear offers a special problem in obtaining an adequate tread and
suitable shock-absorbing qualities. A wide tread would mean long land-
ing-gear struts at a shallow angle to the ground so that the shock-absorbing

medium is likely to function unsatisfactorily.
Struts may be carried from the landing gear to the wing to serve merely
as a support for the gear in order to obtain a reasonably wide tread.
Another variation of the high-wing monoplane is the parasol type where
the open cabane is interposed between the fuselage and the wing. Such
an arrangement is particularly suitable for an open cockpit airplane where
vision forward for the pilot, especially if he happens to be seated quite far
back, is desired.
Open cockpit airplanes do not have the general all-weather utility that
enclosed cockpits have and so are seldom seen, although there is a place
for them for special purposes.

16
Fig. 11-15
Fig. 11-17
Fig. 11-16
Fig. 11-18
Figure 11-15. A cargo airplane with twin booms supporting the tail surfaces high
and clear of the rear loading area of the fuselage.
Figure 11-16. A twin fuselage monoplane which makes use of twin-engine nacelles
and twin booms. It also suggests the possibility of coupling two monoplanes. Such
a design permits separate functions in the two fuselages without interference.
Figure 11-17. A twin fuselage design with a special compartment for personnel.
Figure 11-18. A monoplane with an unusually wide fuselage of airfoil cross section
designed to obtain certain flying wing advantages with more or less conventional air-
plane design.
The low-wing monoplane has variations similar to the high-wing mono-
plane. The design is excellent for short landing-gear structures, and also
affords a ready means for landing-gear retraction. Vision upward and
toward the sides is excellent, but poor down at the sides. It is often
claimed that low-wing monoplanes are not so stable as high-wing mono-
planes, but a properly designed airplane always has sufficient stability.
Struts may be replaced by wires but this requires a set above and below
the wing. In some racing designs for low-horsepower engines, such de-
signs have been used since the use of thin airfoils would offset, to some
degree, the resistance of the wires which offer less resistance than struts
do. The wires, of course, cause complications in rigging and mainte-
nance, and, therefore, are not considered favorably by the private flyer.
In general, external bracing, whether used for a biplane or a monoplane,
should not form too acute an angle at the intersection with the top or bot-
tom surface of the wing since it not only offers more aerodynamic re-
sistance when so located, but also is likely to affect adversely the airflow
THE LOW-WING MONOPLANE

TYPES OF AIRPLANES
17
over the wing and may, therefore, affect the longitudinal control and sta-
bility of the airplane.
To reduce the aerodynamic resistance resulting from the configuration
offered by the juncture of the wing and the fuselage, filleting is resorted to
in certain cases. However, filleting increases production complexity and
costs. Another solution to this problem is the so-called inverted gull
wing.
A gull-wing monoplane is one that has the root section of the wing
inclined at an angle to the fuselage so that the outer panels of the wing
are raised above the fuselage. (See Figure II-7 for example.)
The gull wing eliminates acute angles of intersection between the wing
and the fuselage and thus helps to reduce the parasite resistance. How-
ever, the design has some structural difficulties in that the construction
of the spars becomes complicated. Such a design improves the vision
upward and sideward over that of the conventional high-wing mono-
plane. In the case of multi-engine designs, when used for seaplanes and
flying boats, the wing is sufficiently raised to obtain propeller clearances.
For the low-wing monoplane, the inverted gull wing permits the fuselage
to be raised above the ground for a single-engine design in order to pro-
vide propeller clearance with the ground. This solution also has the
advantage of reducing the length of landing-gear supports so that re-
traction of the landing gear becomes unnecessary.
The midwing arrangement for the monoplane has definite aerodynamic
advantages in that its aerodynamic resistance is usually the lowest of the
various possible wing and fuselage arrangements. Structurally the design
offers complications because the spars should have "carry-through" mem-
bers in the fuselage. Such members interfere with the internal arrange-
ments. This design may also be braced externally, but because the
shallow angle between external struts and the wing causes large axial
loads in the spars, such bracing is usually not considered very desirable.
SPECIAL-PURPOSE AIRPLANES
Recent efforts have been directed toward developing aircraft requiring
short take-off and landing runs. Such aircraft are known as VTOL (for
vertical take-off and landing) and STOL (short take-off and landing).
VTOL aircraft must get off the ground without any roll and clear a 50
foot obstacle in 250 feet. STOL aircraft are allowed ground roll, but
must be able to clear a 50 foot obstacle 500 feet from the starting point.
Designs known as convertiplanes employ such means as (1) the com-
bination of helicopter rotors and normal wings, (2) tiltable wings with
counter-rotating propellers, (3) multi- or "Venetian-blind" flaps and
powerful slipstream effects, and (4) multi-purpose wings which when

18
rotating act as a helicopter and when stationary act as ordinary wings.
There are special craft such as "piggy-back," parasite, drone, and
towing, the names of which designate their respective functions.
FACTORS AFFECTING CHOICE
The discussion thus far has been primarily with reference to the lift truss
of the airplane because it is usually the first in the list of considerations in
choosing the type of airplane. However, the discussion has been general.
The student should not overlook the specific considerations entering in the
choice and in the design of the power plant, the landing gear, the fuselage
structure, the interior arrangements, and the wing. Each of these may
have a bearing on the whole.
In determining a suitable type of airplane, the factors affecting the final
design should be listed and then carefully considered with reference to per-
tinent parts of the airplane. Some of these factors will now be considered.
Performance
A high speed requires an aerodynamically "clean" airplane with the
minimum of struts and lifting surfaces. A monoplane seems to be the
answer.
A rapid-climbing airplane requires either a relatively lowipower or low-
> thrust loading, or wing loading, or both. If a low-power loadingâ€”is.se-
lected, then an externally-braced monoplane or biplane is desirable.
High ceiling may be obtained by a low-wing loading and a high aspect
ratio. A semicantilever monoplane may be the solution.
^ Low landing speeds may be obtained by means of low-wing loading or a
high maximum lift coefficient, or both. The lift coefficient may be con-
siderably increased by means of flaps or other lift-increase devices. A
high-wing monoplane can incorporate these better than a low-wing mono-
plane, but a low-wing monoplane may be a better solution in spite of that,
because of other considerations.
Landing-Gear Retraction
If the landing gear is to be retracted, an internally braced, low-wing de-
sign offers the best solution. The struts are shorter, the mechanisms sim-
pler, and in case of a forced landing due to impossibility of lowering the
landing gear caused by some mechanical difficulty, the low wing offers the
best protection.
Structure
^ Some types of structures are better adapted to one kind of airplane than
to another. A tubular steel fuselage is more efficient structurally for a
small airplane than for a very large one. A fabric covering may be satis-
TYPES OF AIRPLANES
19
factory for externally braced wings and for wings of airplanes whose top
speed is not much greater than 150 miles per hour.
Special Features
To help in deciding what type is best suited for your design, it is well to
list the special features the airplane may have, and then study them in the
light of a particular design being considered. Such features might be:
Engines
Air cooled or liquid cooled Pod installation
Radial or in-line Tractor or tandem
Single or multiple Cowled or concealed
Jet-propulsion type Maintenance requirements
Landing gear
Conventional or tricycle Fully cowled, partially cowled,
Retractable or nonretractable or uncowled
Full cantilever, split axle, or other Land or water type
Type of tires Maintenance requirements
Fuselage
Open cockpit or cabin Location of doors
Reinforced monocoque or tubular Type of cargo
steel Maintenance requirements
Wing
Performance requirements Space for landing gear
Fabric or metal covered Location of engine nacelles
Wood or metal spars High or low aspect ratio
Internally or externally braced Thickness ratio of airfoil
Lift-increase devices Maintenance requirements
Space for fuel tanks
Tail surfaces
Single or multiple surfaces Operating controls
Location above or below fuselage Maintenance Requirements
Tab controls
SAMPLE AIRPLANE DATA SHEETS
The design of airplanes is largely empirical. Thus it is advisable to
study as many airplanes as possible, catalog them under different cate-
gories, and collect as much data as possible on performance, weights,

structure, power, and all other items that may be useful as reference
material in future work. Data collected for a number of airplanes in the

20
same category may then be averaged to give empirical values for de-
termining ratios of power or thrust loadings, wing loadings, etc.
The following condensed airplane data sheet may be taken as a guide
to the type of information that should be collected. It may be expanded
to suit individual needs.
Airplane Data Sheet
Name of Company:
Name of Type:
Price:
1. Power Plant
Engine: Horsepower: Rev. per min.: Altitude:
Starter:
Design of exhaust:
Other engine accessories:.
Fuel, gallons:.â€”
Oil, gallons: â€”
Location of tanks: -
Type of engine controls:
Propellerâ€”Make:
Material:
Type:
Diameter:
Number of blades:
Angular range:
2. Wing1
Airfoil sectionâ€”Root:
Midspan:
Tip:
Wing area (including ailerons), Sw:
Span, b:
Chordâ€”Root, Cr\
'If biplane statistics are to be listed, repeat these entries for lower wing.
Tip, CT: p
Taper ratio, RT:
Aspect ratio, AR:
Mean aerodynamic chord, MAC:
Dihedral:
Sweepback:
Incidenceâ€”Root:
Midspan:'
Tip:
Length of cantilever tip:
Length of outer bay:
Length of inner bay:
Length of center section:
Location of wing spars in per cent of chordâ€”Front:
Rear:
Maximum rib spacing: â€”-
Aileron area, Sa:
Flap area, S/r.
Location of center of gravity when fully loaded, in per cent of mean
aerodynamic chord: .'
3. Tail surfaces
Stabilizer area, S,:
Elevator area, Sr:
Total horizontal tail surface area, Sh:
Distance from center of gravity loaded to elevator hinge:
Fin area, S/:
Rudder area, <Sr:
Total vertical tail surface area, Sv:
Distance from center of gravity loaded to rudder hinge: â€”
Equivalent fin area ahead of center of gravity loaded:
Stabilizer setting with reference to propeller axis:

22
4. Passenger accommodations
Dimensions of cabin:
Finish:
Seating (type and weight, size, spacing):....
Aisle width:
Aisle height:
Windows (type):
Doors and steps (emergency, etc.):
Heating:
Ventilation:
Lighting:
Baggage accommodation:
Toilets:
5. Pilot's cockpit
Single or dual controls:
Releasable controls:
Angles of visionâ€”Upward:
Downward:
Sideward:
Windows and their construction:.
Windshield:
Arrangement of seats:
Are seats adjustable?
Are rudder bars adjustable?
Instrument board:
Door (separate for pilot):
6. Auxiliary equipment and accessories
Night flyingâ€”radio:.
Direction finding:
Navigation instruments, etc.:
Special equipment:
7. Angles
Landing angle of airplane:

TYPES OF AIRPLANES
23
Angle in side elevation between vertical through axle and line connect-
ing center of gravity and axle:.
Angle in front elevation between vertical and line joining the center of
gravity and the point of contact with the ground at the outer wheel:
Angle between the ground and a line from the point of tangency of the
wheel with the ground to the wing tip on the same side of the plane
of symmetry of the airplane as the wheel:
Angle between the ground and a line joining the point of contact with
the ground of the deflected tail skid to the tip of the horizontal tail
surfaces:
8. Performance Sea Level Rated Altitude
High speed in mph:
Cruising speed in mph:
Landing speed in mph:
Climb in fpm:
Climb to 10,000 ft in min:
Normal range in miles:
Service ceiling in ft:
Fuel capacity in gal:
Normal fuel consumption at
cruising speed gal per hr:
9. Miscellaneous
Distance from front or rear face of propeller flange to center of gravity
of airplane loaded:
Type of fuselage construction:
Nose or tail wheel construction:
Type of chassis construction:
Design load f actorsâ€”Gust:.
Maneuvering:
Landing:.
Special Conditions:.
10. Weights
Weight empty (lb) with liquid if liquid-cooled engine is used:
Payload (lb):.â€”
Disposable load (lb):

24 AIRPLANE DESIGN MANUAL
Normal gross weight loaded (lb):
11. Over-all dimensions
Span, b:
Length, L:
Height, H:
Tread, T:
12. Ratios
Weight empty to gross weight (Wb/Wo) â€¢
Payload to gross weight (Wp/Wo) : â€”
Disposable load to gross weight (Wd/Wo) â€¢
Wing loading (WJSW):
Power loading (WJP):.
or
Thrust loading (WJT):
Aileron area to wing area (Sa/Sw):
Vertical tail surface area to wing area (Sv/Sw):
Fin to vertical tail surface area (S//Sv):
Horizontal tail surface area to wing area OS*/Sw):
Elevator area to horizontal tail area (Se/Sh):
Flap area to wing area 0S/i/jSâ€ž):
Tab area to aileron area (Si/Sa):
Tab area to rudder area (Si/ST):
Tab area to elevator area (St/Sc):
Tread to span (T/b):
Length to span (L/b):
Height to span (H/b):
Sometimes, it may be preferable to prepare a graph on which are plotted:
power and wing loadings versus gross weight; ratio of weight empty to gross
weight versus gross weight; ratio of control surface area to wing area versus
wing loading. From such graphs, the designer may select for his own de-
sign the information that will help him in developing the general details of
his design configuration and thus be reasonably assured of obtaining a
successful airplane. Of course, refinements are made after careful aero-

dynamic, performance, structural, and other careful analyses have been
made.
Fig. 11-21 Fig. 11-22
Figure 11-19. A design for a hypothetical 1,350 mph plane, prepared at NACA's
Flight Propulsion Laboratory. It relies on after-burner to double basic engine thrust,
thus allowing for small nacelles.
Figure 11-20. An aircraft capable of vertical take-off and landing, converting to
horizontal flight when sufficient altitude has been reached. The landing gear, consisting
of four casters, is located at the tips of the delta wing and fins. A contra-rotating
propeller is used to eliminate engine torque effects upon the flight characteristics at
take-off. The static thrust of the power plant has to be greater than the weight of the
airplane for take-off and attaining climb.
Figure 11-21. A jet-powered flying wing design. The flaps act in the dual capacity
of a lift-increase device and elevators. The ailerons can function differentially as
ailerons, or nondifferentially as flaps or elevators. Where ailerons and elevators are
combined in function, they are called "elevans." The air inlet is located in the leading
edge of the wing at the root close to the fuselage.
Figure 11-22. A supersonic airplane. Note the wing fences used to prevent the
drifting of the boundary layer spanwise. Note also the application of the "area-rule"
to the fuselage configuration.

26
EMPIRICAL DATA
The preceding Airplane Data Sheet has been used to list data on cur-
rent airplanes, and this information has been used to calculate the various
ratios which form part of the empirical data useful in projecting a new
design.
The empirical data are used, for example, in arriving at a preliminary
weight estimate and subsequent values for areas and dimensions necessary
in drawing up the preliminary three-view, from which a corrected weight
estimate and initial structural drawings may be made.
AIRPLANE DATA
Tables II-1 to II-4 list important data for a number of airplanes which
are identified merely by reference numbers. Many of these airplanes have
been in production, whereas a few are projected designs. The student
may derive his own empirical data by determining the average wing load-
ings, power loadings, and ratios of control surface areas to wing areas
which fall within each single category of airplane. Such averages are usu-
ally obtained for the same horsepower range: for example, for airplanes
powered with engines delivering horsepower from 65 to 100; another group
or category from 150 to 225; still another from 1500 to 2000.
Table II-1. Airplane data: power plant.
Airplane identification
Power plant
Rated
At
Propeller
Fuel
Ref.
No. of
No.
hp
altitude
diameter,
capacity
no.
Typef
seats
of engines
each, P
(ft.)
D(in.)
(gal.)
CLM
65
SL
72
15
CLM
65
SL
72
23
CLM
85
SL
74
22
CLM
2
TYPES OF AIRPLANES
27
Table II-2. Airplane data: dimensions and construction.
Dimensions
Landing gear
Wings*
Ref.
Span,
Length,
Height,
Wheel
Cov-
Hull or
no.
Type*
size
Spar
Ribs
ering
fuselage*
35'
21' 6"
9' 1"
Fx
6.00 X 6
WldStF
36'
20' 10"
9' 1"
Fx
6.00 X 6
W
WldStF
28' 8"
20'
7' 4"
FxTri
â€”
MSemiMo
33'
22'
8' 6"
FxTri
6.00 X 6
MSemiMo
30'
20' 9"
5' 11"
FxTri
6.00 X 6
A
28
Table II-4. Airplane performance data.
Performance
High
Cruising
Stalling
Normal
speed,
At
speed,
At
speed,
Climb
Service
Ref.
range
mph,
altitude
mph,
altitude
(mph)
(fpm)
ceiling
no.
(miles)
(ft)
Vcr
(ft)
V.
(ft)
270
100
SL
90
SL
38
500
10,000
420
100
SL
90
SL
38
500
10,000
400
118
SL
108
SL
49
610
11,000
500
125
SL
115
SL
50
750
14,000
500
CHAPTER
Airfoil Selection
In selecting a suitable airfoil or a combination of airfoils for the wing
of an airplane, it is necessary to study carefully both the aerodynamic
and structural characteristics required for the airplane. Typical aero-
dynamic data are shown in the curves presented. No attempt will be
made here to discuss in detail all the aerodynamic characteristics, but in
order to indicate the importance and the significance of some of these
characteristics in airplane design, elementary consideration of the airplane
in flight will be studied.
THE AIRPLANE IN RECTILINEAR FLIGHT
Figure III-JL shows the forces acting on the airplane along any flight
path through space in still air. The line of action of the relative wind is
along the flight path opposite to the direction of motion of the airplane.
Resolving forces perpendicular and parallel to the line of action of the
relative wind, the following equations result for the summation of forces,
since the airplane is in equilibrium.
For Sy = 0,
T sin (i + a) + Lw + Lt - W cos (8 = 0; (1)
For = 0,
-T cos (t + a) + Dp + Dw - W sin 0 = 0; (2)
For 2Me.â€ž. = 0,
Ma.e. + LwViC sin a + LwXiC cos a â€” DwyiC sin a +
DwXiC sin a â€” L<x2C cos at â€” Lty2C sin a, â€”
TysC cos (i + a) + TxÂ»C sin (i + a) = 0. (3)
29
AIRFOIL SELECTION
31
Also,
j8 = angle between the line of flight and the horizon. (Still-air
conditions are always assumed.)
a â€” angle of attack; angle included between the line of flight and
the mean aerodynamic chord of the wing.
i = angle of incidence, angle included between the line of thrust,
and the mean aerodynamic chord of the wing.
at = angle of attack of the horizontal tail surfaces; usually Lt sin at
is considered negligible with reference to Lâ€ž sin a, so that this
term is neglected. Likewise, it may be assumed that
cos a, = 1.
L = MpCLSv* and D = MpCdSv*,
where L = lift, in pounds;
D = drag, in pounds;
p = mass density of air, in slugs per cubic foot, which is 0.002378
for standard air at sea level;
S = wing area in square feet;
e; = speed in feet per second along the flight path of the airplane;
and
Cl and Cd are nondimensional lift and drag coefficients.
Case 1. Horizontal Flight. In horizontal flight the angle /3 is zero, and
since in normal flight attitudes a is small, the sine of (i + a) may be con-
sidered equal to zero (for example, even for a = 16Â°, sin a is only 0.17),
so that equation (1) becomes
Lw + Lâ€¢ - W = 0. (la)
Usually the load on the horizontal tail surfaces is also small compared with
the load on the wing in horizontal flight, so that Lw = W. In other words,
the lift of wings in horizontal flight may be assumed to be equal to the
gross weight of the airplane. This is the relationship which is useful in
determining the speed in horizontal flight since
Lw = tCJ3 - W = WClS, (4)
so that
-4
y2pcLs
(4a)'
Minimum speed is then fixed by Cl*., for a given wing loading, W/S.
The higher the wing loading for a given ClÂ«â€ž, the greater the minimum
or stalling speed. For all practical purposes this stalling speed is con-
sidered the landing speed. If the stalling speed is fixed and if the gross
weight of the airplane is fixed as well, then that airplane which had the
highest value of Cl^ would have the least wing area.

32
It is, therefore, desirable to compare values of Cl_, for a group of air-
foils. However, if lift-increase devices are used, then the maximum lift
coefficient of the basic airfoil is not so important since any airfoil with the
proper lift-increase device may attain the desired value of Ci^.,.
The maximum speed is determined by the maximum amount of horse-
power or thrust delivered by the propeller-engine combination or jet
engine at maximum speed. The horsepower required is equal to the
horsepower available..
The power required is
PT = (/>. + DP)00^; (5)
The power available is
Pa = 77BHP; (5a)
and since at maximum speed,
Pa = Pr (5b)
or,
rjBHP = (Z>â€ž + DP)00^0 0 (5c)
This relationship can be obtained from equation (2), for, again assuming
(a + 1) small and |3 = 0, then
- T + Dp + Dw = 0, (2a)
Â°r' T = DP + Dw (2b)
and multiplying both sides by v/550,
But, TV/550 is the thrust horsepower available, or
Â«S> = "bhp (6a)
where r) is the efficiency of the propeller and BHP is the brake horsepower
of the engine delivered to the propeller at the given airplane speed, and
the power required is given in equation (5) as
Pr = (DP + IWJL,
but
Dw = yiPv2cDs
Dp = y2Pv2cDrs,
where Co is the drag coefficient of the wing,

AIRFOIL SELECTION
33
CDp is the parasite resistance coefficient of the airplane (less the
wing) referred to the wing area.
Then,
Pr = P0 = â€žBHP = (CD + GDp), (7) .
or, \
T = p-f- (CD + CDp), (7a) V
and for any given lift coefficient the speed v can be determined from equa-
tion (3), the corresponding Cd of the wing will be known and the Cd, for
the airplane may be calculated,1 so that the horsepower required of the
engine-propeller combination can be determined. Then, if the horsepower
available is a certain value, it is obvious that the maximum speed that can
be obtained for a given airplane (whose Cd, is fixed) is one whose wing air-
foil has a minimum value of Cd0 0 It is important, therefore, to compare
the minimum values of the drag coefficients of a series of airfoils.
Since the range of speeds obtainable is determined by the values of the
maximum value of Cl (minimum speed) and of the minimum value of Cd
(maximum speed), then the ratio of CLm., to Co,,, is of importance and is
known as the speed range ratio.
Case 2. Gliding Flight. Again, assuming that the lift on the horizon-
tal tail surfaces is small and the angle (a + i) small, but that 0 is not un-
appreciable, equation (1) becomes Lw = W cos 0, and equation (2) be-
comes Dt â€” T = W sin 0.
Dividing the first equation by the second,
^Ty = cOt 0, (8)
when there is no thrust (that is, when the engine fails)
cot 0 - ^. (8a)
or the angle of glide is a function of the aerodynamic characteristics of the
airplane. This relationship may be rewritten
Lw qClS Cl Cl . a ,suN ('
-dâ€¢ = WnTs = c7, = cTfcZ = cot * (8b) A
Again, considering an airplane for which the parasite resistance can be
assumed constant for any angle of attack, the angle of glide /? will be
flattest for that airplane whose L/D for the airfoil alone is the largest.
1 See Chapter X XII, Preliminary Performance Calculations.

34
Also, the circle determined from
K = ffcot0 = ff(^. (9)
where R is the radius of the circle, and H the altitude from which the
glide takes place, is largest for the largest value of cot 13. The airfoils
should, therefore, be compared on the basis of the maximum L/D.
Case 3. The Dive. In this case, the angle /3 becomes 90 degrees, so
that equation (1) now becomes Lw + Lt = 0, assuming (i + a) is small.
Equation (2) becomes
- T + Dp + Dw - W = 0
when there is no thrust,
Dt = W
where
DP + DW = Dt!
or
WCDJ5 = W,
from which
/W
v V HpCdJS'
The maximum speed will then be obtained in the dive, unless the drag
coefficient is unusually large, and will be determined by the minimum
total drag coefficient of the airplane. This speed is of importance in con-
sidering local pressures on engine cowls and windshields which may be
pulled off the airplane by the "suction" pressure. The leading edge of
the wing would then be subjected to enormous pressure that would tend
to buckle it. Also, the highest load factor in flight is encountered when
pulling out of a dive. The higher the diving speed, the higher the load
factor encountered in the pull-out. (See Chapter IV, External Loads on
Airplane in Flight.) If the speed were to be limited, then means would
have to be provided to increase the drag.
Case 4. The Climb. In the case of the climb, equation (2), as derived
for the general case, is of importance. The angle /3 is now a negative
angle and the equation in question becomes, upon making the same as-
sumptions as to a and i,
T - Dt = W sin 0. (11)
But, sin /3 = a/v where v is the velocity along the flight path and a is
the vertical component of this velocity, or the rate of climb in feet per
second.
(10)
AIRFOIL SELECTION
35
Then, equation (11) becomes T â€” Dt = W(a/v), and by multiplying both
sides of the equation by v/550,
550 550 550' 1 J
But !Ty/550 = horsepower delivered by the propeller, or the horsepower
available = Pa. Therefore Tv/550 = 17 BHP where t] is the propeller effi-
ciency and BHP is the brake horsepower of the engine. Dtv/550 is the
horsepower required to overcome the total drag of the airplane at ve-
locity v and may be designated PT.
Rearranging terms
550(Pa - PT)
a= W'
W cos 0 = Lw = y2pv?SCL
or
(lib)
4.
y2pscL {10)
where ve is the velocity along the climb path.
Examination of equation (lib) indicates that the airplane which requires
the least amount of horsepower to overcome aerodynamic resistance will
have the greater climb. Since
Pr = (Dw + DP)(Â»/550),
then if the parasite resistance is kept constant, Pr is piimarily a function
of Dwv, but
Dwv = (y2PcDSv*)v = y2pcDSv\
and since
so that
or
Dwv
cD
Clw'
CD
Clw
also Cd/Cl*12 should be a minimum, or Clw/Cd a maximum in order to
maintain PT at a minimum. This ratio is sometimes called a "power co-
efficient" for the airfoil and is calculated for those angles of attack in the
region where the minimum drag coefficient Cj> and the maximum L/D of
the airfoil occur.

36
Case 5. Range. One of the simpler formulas for determining the range
of an airplane is the so-called Breguet's formula for weight with respect to
distance, which is derived from the differential expression,
dW = - TVc Â± - WVc (13)
dt 375tj 375(L/D)i7'
which, integrated between the limits of Wo and Wâ€ž gives
4*to*.S (13a)
where
R, range in miles = 863 ^ - logw
L Cl Cl
D CD, Cd + Cdf
r) = average propeller efficiency at cruising,
c = average fuel consumption in pounds per brake horsepower per
hour for the average cruising rpm,
Wo = gross weight in pounds at start of flight,
We = weight at end of flight after fuel has been consumed.
It will be noted that, all other things being equal, the higher the value of
the L/D, the longer the range. Therefore, if the parasite resistance of the
airplane is constant, the maximum value of the L/D = Cl/Cd of the airfoil
would be of interest.
This formula can be made to apply to jet engines by dropping the term
7j for propeller efficiency and considering c as the average fuel consumption
in pounds per pound of thrust per hour for cruising conditions.
Other Airfoil Characteristics
The slope of the lift curve, dCiJda, is one of the more important quan-
tities to know since it has an important bearing on the stability of the
airplane.
The angle at which zero lift curve occurs is also important since the
diving speed occurs very close to this angle.
The center of pressure movement over the normal flying range, between
the angle at which the minimum drag coefficients occur and the angle at
which the maximum lift coefficient occurs, is usually of interest since the
greater the movement the greater the load that will fall on the front spar
at high angle of attack, with very little load on the rear spar; the condi-
tions are reversed at low angle of attack.

Some data may not include the center of pressure but may give the mo-
ment coefficient, Cm., about the aerodynamic center instead. In such a

case a small value of Cm, is considered desirable, since
CP = a - ^
CV (14)
AIRFOIL SELECTION
37
v.
"I
Â©
16
*.
ft
"I
4 +.020
00
-.020
ae
420-
0=
~2Â°â€” | | I I I | | I I | I
0 20 40 60 80 100
Per cent of chord
cL
CP.
L-
4 8 12 16 20
Anqle of attack
(in degrees)
1.4 o
1.2 \ .24
00
10 $.20
0.8
16
25
0.6
.12
35
0.4
.08
45
0.2
.04
ft 2
rag eoeffieiem
1:
<Â» v.
, ft
-J Vi
Figure III-2. Representative curves of aerodynamic properties of an airfoil.
where CP = the center of pressure in fraction of the chord from the
leading edge,
a = the aerodynamic center expressed as a fraction of the chord
from the leading edge,
Cm. = the moment coefficient, a constant, about the aerodynamic
center, and
Cn = the normal force coefficient equal to (Cl cos a + Cd sin a)

38
and for small angles of attack, practically equivalent to CL.
Thus, for small values of Cm., the center of pressure will not vary much
from the aerodynamic center position.
STRUCTURAL CONSIDERATIONS
In selecting a suitable airfoil for a given design it is necessary to consider
the structural and aerodynamic characteristics required for the airplane.
A racing airplane, for example, will have an entirely different wing from a
transport airplane.
A full-cantilever wing, with an aspect ratio from 6 to 10, requires a thick-
ness ratio of at least 18 per cent for the root section. The larger the aspect
ratio, the greater the thickness ratio of the root section should be. It is
not desirable to use the same thickness ratio from root to tip, but to de-
crease it linearly to not less than 9 per cent at the tip.
A semicantilever wing, with an aspect ratio from 6 to 8, requires a thick-
ness ratio of at least 12 per cent for the root section. Again, the larger the
aspect ratio for the wing, the greater must be the thickness ratio of the
root section. The wing should be decreased linearly to about 6 per cent
thickness ratio for the tip chord.
A biplane would probably use an airfoil of thickness ratio from 6 to 9
per cent and use the same airfoil from root to tip.
The student should refer to Chapter XVII, Design of the Wing, for
further information.
Recapitulation
i When the thickness ratio for the airfoil has been tentatively established,
it is then desirable to compare a group of airfoils to determine which has
the best all-round characteristics.
Care should be taken when two airfoils are selected to choose two from
the same general family, since it may be difficult otherwise to avoid serious
double curvatures in the development of the skin. A concave lower sur-
face for one airfoil and a convex lower surface for another when used for a
wing of a tapered planform would usually not be desirable.
All comparisons should be made preferably for airfoils tested at the same
Reynolds or Mach numbers, in the same wind tunnel, and for the same
aspect ratio in order to eliminate, or at least reduce, certain experimental
variables. The characteristics of the airfoils to be compared have already

been discussed briefly. Table III-1 is a recapitulation of the quantities
mentioned in the discussions above.

ASPECT RATIO CORRECTIONS
With few exceptions, wind-tunnel tests of airfoils are presented for a
standard aspect ratio of infinity; but since the wing used for the design
AIRFOIL SELECTION
41
of the formula
Ci}
Cb = Cd, + Cd{ = Cd, +
where Cj> is the total drag coefficient for the airfoil used, while Cd, is the
profile drag, independent of the aspect ratio and constant for the airfoil.
The induced drag is expressed
â‚¬..-%. . CM)
As an example, to calculate the drag coefficient for aspect ratio 8 when
the characteristics for aspect ratio 6 are known, the following formula may
be derived:
where R = 8. Corrections for low values of d may be ignored. The
angle of attack must be corrected also for aspect ratio
= (16)
where clr and ao are in radians; or
57.3 Cl/1
where aK and a6 are in degrees and the known characteristics are for aspect
ratio 6. If the known characteristics are for infinite aspect ratio, then 6
is replaced by oo and l/oo becomes equal to zero. The lift coefficient, the
center of pressure, and the corrected drag coefficient correspond to the
corrected angle of attack. Table III-2 has been set up to expedite the
aspect ratio calculations. The various terms are self-explanatory. The
calculations are usually carried out in the form of such a table.
When an airfoil section used for the tip of a wing is different from that
used at the root, it is necessary to make wind-tunnel tests on a model of
the actual wing. Reasonably close approximation for preliminary cal-
culations may be obtained by averaging the characteristics of the root and
tip airfoils.
AERODYNAMIC SECTION CHARACTERISTICS
At the present time, much of the airfoil data such as the lift, drag, and
moment coefficients, the angle of attack, and the center of pressure, are
presented corrected to infinite aspect ratio. Such data are referred to as
section coefficients for infinite aspect ratio. To distinguish them from
similar coefficients obtained or given for airfoils of finite aspect ratio, the
42
coefficients are designated with lower-case letters (cj, ca, Cm..,., etc.) for
the infinite aspect ratio case and the upper-case letters (Cl, Cd, Cm...., Cdâ€ž
etc.) for finite aspect ratio.
These section characteristics are particularly useful in obtaining the
spanwise lift distribution as outlined in ANC-1 (1) Spanwise Airload
Distribution, a small volume obtainable from the U. S. Government Print-
ing Office, Washington, D. C. It is beyond the scope of this book to
discuss or to indicate the procedure in calculating the spanwise lift distri-
bution. However, for stress analysis purposes, and for more refined calcu-
lation of performance, the spanwise as well as the chordwise distribution
(the latter especially for wings equipped with flaps) should be carefully
calculated.
To obtain the necessary data for preliminary design purposes, the fol-
lowing discussion will indicate a practical procedure. When the design
of the wing is well advanced, more elaborate and refined calculations as
set forth in a number of textbooks on aerodynamics should be made.
The pertinent data of use at this stage are the variation of the lift and
drag coefficients with the angle of attack, and the moment coefficient re-
ferred to the aerodynamic center of the mean aerodynamic chord. As
pointed out elsewhere, the mean geometric chord may be taken as the
mean aerodynamic chord for practical purposes.
A wing having no aerodynamic twist2 would have its section angles for
zero lift arranged so that all sections would have no lift when the root sec-
tion was at zero lift. If aerodynamic twist were not zero, then the angles
of zero lift for the various sections spanwise would not be the same as for
the root. The root section is usually selected for reference. In order to
obtain its position when the lift coefficient of the wing is zero, it will be
necessary to determine when
The section angles corresponding to the section lift coefficients will then
make it possible to find the value of am for the root section.
To simplify the discussion and the presentation, a wing of zero aero-
dynamic twist will be considered. It is a comparatively easy matter to
work out the procedure for a wing with aerodynamic twist.
The equation of the lift curve (Cl versus a) at low values of the lift co-
efficient is
where the various terms have the significance indicated as follows:
Ci = the lift coefficient of the wing of finite aspect ratio, or may be

assumed equal for a wing of zero aerodynamic twist, as the aver-
age of the section lift coefficients, for the tip and root airfoils,
* For definition of this term, see under Airfoil Construction in this chapter.
ZCiCdy = 0.
(17)
CL = a(aR â€” ecu),
(18)
AIRFOIL SELECTION
43
a = the slope of the finite aspect-ratio wing,
Â«k = the angle of attack in degrees of the finite aspect-ratio wing at
which the lift coefficient Cl occurs,
c*lo â€” the angle of zero lift for the finite aspect-ratio wing; or may be
assumed equal to the average of the section zero-lift angles or
to the average for the root and tip airfoil; or may be determined
for the mean geometric chord assuming a linear variation span wise
of the section zero-lift angles.
The lift coefficient is usually considered the independent variable, and
the angle of attack as the dependent variable. It would, therefore, be
easier to use equation (18) in the form
aR = + (18a)
= KCL + Â«lo. (18b)
The slope a may be calculated from
a = i= 9.KB + 3)' (19)
where R is the aspect ratio of the wing and the other constants account for
planform and wing-tip corrections. These corrections vary for each type
of planform and wing tip, but the values given are sufficiently accurate
for preliminary work.
The maximum lift coefficient is not so easy to predict for the finite
aspect-ratio wing, but it may be assumed to be about the same as for in-
finite aspect ratio without incurring an error of more than from 3 to 7
per cent.
To determine the desired drag coefficient Cd for the corresponding lift
coefficient Cl, the following relationship is useful:
Ci}
Cd = CdÂ° + Â«00Â«(lp.5 - 0.322) (20) v
= CD, + KiCi? (20a)
where Cb0 is the average spanwise of the section drag coefficients Cd, or,
for a wing of zero aerodynamic twist, may be assumed to be the average
of the section drag coefficients of the tip and root airfoils. To make the
calculated values agree with experiment, the factor (10.5 â€” 0.3i?) has
been introduced to allow for planform variation and for tip effects.
The moment coefficient about the aerodynamic center of the mean aero-
dynamic chord may be calculated by means of the following equation:

Cm. = tt^ 2(cm.c2dy + ciXa.c.cdy), (21)

44
where the various terms have the significance indicated as follows:
Cm.â€ž. = theoretically, the mean aerodynamic chord, but for practical
purposes, the mean geometric chord (preferably stated in
inches);
S = wing area, in square inches if the linear dimensions are expressed
in inches;
C = chord, in inches, of the section under consideration;
dy = the span, in inches (usually one inch) of the section under con-
sideration;
Ci = the section-lift coefficient;
xa.c. = the distance, in inches, of the aerodynamic center behind the
leading edge of the section.
Instead of the aerodynamic center, the quarter-point on the chord may
be used, in which case the moment coefficients are about the quarter-point
of the chord.
Table III-3.
Section characteristics
Airfoil (or wing) characteristics
Root section
Tip section
<*Lo - (<*lo) root uelion
(1)
(2)
C,
(3)
Cd
(4)
aio
(5)
Ci
(6)
(7)
Cl
(8)
(9)
KCl
(10)
(11)
(12)
KlC,?
(13)
CD
-4Â°
-6Â°
Cdq
OR
Cl1
-2Â°
-4Â°
0Â°
-2Â°
2Â°
0Â°
4Â°
2Â°
6Â°
4Â°
8Â°
6Â°
10Â°
8Â°
12Â°
10Â°
14Â°
12Â°
16Â°
14Â°
Airfoil
AIRFOIL SELECTION
45
Table III-3 has been set up to illustrate the type of calculations required
for a wing having the wing-tip airfoil set at minus two degrees with refer-
ence to the root airfoil in order to obtain zero aerodynamic twist (assum-
ing linear span wise variation of all section characteristics).
Column 7 lists the average values of the section-lift coefficients given in
columns 2 and 5. Column 8 lists the average values of the section-drag
coefficients given in columns 3 and 6. Column 10 lists the values obtained
by adding values of column 9 to aLo. The angle ckb is referred to the root
airfoil in the calculations indicated, that is aa = (ttio) root sectionp The lift
coefficients in column 7, plotted against the values of aR in column 10, will
give the lift curve for the wing of finite aspect ratio R. Column 12 lists
the values of the drag coefficients of the wing obtained by adding the
values of KiC]}, listed in column 11, to the values of Ceo in column 8.
A summary of airfoil data is available in NACA Report No. 824. Subse-
quent reports should be consulted for airfoil data useful for transonic and
supersonic conditions. Some typical airfoil characteristics are shown in
Table III-4.
The wing characteristics are affected by the fuselage, engine nacelles,
and other aerodynamic bodies that may be attached to the wing. Such
effects will have to be accounted for if more precise calculations are
desired.
A small amount of wash-out, about 1 or 2 degrees, is usually desirable
to help in avoiding tip stall.
The ordinates of an airfoil are given in percentages of the chord so that
it is an easy matter to determine the depth of a section of a wing when the
airfoil and chord length are known.
When the wing employs a different airfoil at the tip than at the root, it
is possible to determine the ordinates of intermediate wing sections from
the known geometric relationships.
For example, it is desired to determine the ordinates of a section of a
wing, tapered in planform as well as in thickness, a distance x from the
root. (See Chapter XVII, Design of the Wing.) Corresponding stations
would then be connected by straight lines so that the ordinate for the
upper camber for chord C would be
AIRFOIL CONSTRUCTION
(22)
and the ordinate of the lower camber would be

(22a)
46
where hiu = upper ordinate for the root airfoil,
hu â€” lower ordinate for the root airfoil,
h2u = upper ordinate for the tip airfoil,
hu = lower ordinate for the tip airfoil,
b = distance in feet or inches between the root and tip airfoil,
y = distance in feet or inches between the root airfoil and the
airfoil at station y.
It is necessary to calculate the ordinate for each station along the chord
since the same relationship does not necessarily exist for all the stations
except in a few particular cases.
In general, it is undesirable to use a radically different airfoil at the tip
from that at the root because of the inherent structural difficulties.
Sweepback for High-Speed Airplanes
In a number of NACA technical notes (see TN 1032 and 1033) the plan-
forms of a number of wings designed for subsonic or supersonic speeds are
discussed. It is brought out that in case of an airfoil of infinite aspect
ratio moving at an angle of sideslip, the pressure distribution over the wing
is determined solely by the component of motion whose direction is normal
to the leading edge. From this fact it is further deduced that the pressure
drag of an airfoil may be reduced if the planforms utilize sweepback angles
greater than the Mach angle (the angle that the plane waves make with
the airfoil at sonic or supersonic speeds).
If j8 represents the angle of sweepback, then the velocity causing the
pressure distribution over the airfoil for an airplane flying at velocity V
would be V cos /J. If V cos /3 is less than the sound velocity Vc, then the
flow and distribution would be similar to those occurring in an incompres-
sible fluid or for subsonic speeds.
The section characteristics for the speeds where M% = [(V cos j8)/Fc]J
is less than 1 (that is, the lift, drag, and moment coefficients obtained for
low speeds) can be corrected to apply for speeds of V of the aircraft. The
following expressions permit calculating the corrections required:
ti .
VI - M2
to give the new lift coefficient;
a=
Vl - M2
to give the new slope of the lift curve. The subscript 0 refers to the sub-
sonic tests.
The drag coefficient has to be found for an equivalent airfoil having a
change in the thickness and the camber. The new equivalent "thickness"
is
-" aajSas jp aÄ‘o[s '.tu = treisuj jp e00ojs .'g
'Â»OTO = Â«*
p00bj ^paÄ‘sÂ« 00j wjÂ«s !V0VN JÂ° sauajuioqvj sjaij tefiuvj %v mu s^sa? HV Â«
%O
s00qs
O8
9f9
2OTs
90000
9O00
2Ol
90000
99'00
W20
OO-
2O0
pO
2OO
WO
fOO
00Ol
00OTs
8SZ
JO
%O
s0000
O0
*O9
0000
00T
OO
9O9
OOT
000
0
OO
OO9T
O00
20000
f00
00 OT
*TOT
OT9
9f9
0000
00OT
JO
%0Z
s0000
00
O00
Of-OT
*8*T
OOOT
2O8
WW
9OT0
92pOT
90OZ
OY\\
9*"T0
Op
8fT0
0
48
I to
Vl - M2'
For a cambered airfoil, the camber changes by the same ratio 1/Vl â€” M2.
The pitching moment coefficient Cma.c also changes by the same ratio.
In order to obtain aerodynamic characteristics at high speeds which
compare favorably with those at subsonic velocities, it is desirable to in-
troduce sweepback for the wing.
The section characteristics for an airfoil operating at supersonic speed
may be approximated for small angles of attack from subsonic data as in-
dicated in the following equations.
Under supersonic conditions, the Mach number M of the free stream is
greater than 1, and the following relationships obtain:
C = 4a
'Vm2 - 1
and
4a2
Cd =
VM2 - 1
where a is small, is expressed in radians, and is measured with reference to
a line connecting the leading and trailing edges of the airfoil.
These equations may be applied to thin airfoils of moderate camber.
Laminar-Flow Airfoils
It can be shown (consult a good book on aerodynamics) that if the
boundary layer could be maintained as laminar (nonturbulent) flow over
the airfoil, then the drag coefficients of the airfoil would be lower than for
airfoils not designed for maintaining laminar flow. While, theoretically,
laminar flow can be maintained, practically difficulties arise due to surface
irregularities, cut-outs, gaps in front of ailerons and flaps, and propeller
slipstream.
The commonly used laminar-flow airfoils have been the NACA 24 series
(NACA 2409, 2412, etc.), the NACA 44 series (NACA 4409, 4415, 4418),
and the NACA 230 series (NACA 23009, 23015, etc.). For an excellent
compendium of airfoil data see NACA Report No. 824 entitled "Summary
of Airfoil Data."
Compressible-Flow Airfoils
A series of airfoils were developed by NACA which attempted to prevent
local speeds of the airflow over the airfoil from exceeding the local speed
of sound. These airfoils are the NACA 16 series (NACA 16-009, 16-109,
16-209, etc.). The mean camber lines of these airfoils were designed to
AIRFOIL SELECTION 49
obtain uniform chordwise distribution. Then the thickness ordinates
were plotted perpendicular to the camber line.
The commonly used airfoils (NACA 24009, for example) had their maxi-
mum ordinate at about 30 per cent of the chord, whereas the NACA
16-009 had its maximum thickness at about 50 per cent of the chord.
CHAPTER IV
External Loads on an Airplane in Flight
The forces that normally act on the airplaneâ€”lift, thrust, drag, and
gravitational forcesâ€”are of importance in the calculation of the perform-
ance parameters of the airplane in rectilinear flight. However, these
forces are not of such magnitude that the structure would have to be
designed to withstand them; rather the structure is designed to with-
stand the loads incurred in accelerated flight. It is obvious, also, that
an airplane used for acrobatics will encounter greater loads than one
designed for cargo transportation. The designer must therefore be aware
of the nature and magnitude of such forces since the greater the forces
encountered, the greater the weight of the structure. It is not the pur-
pose of this book to give a full exposition of stress analysis methods or
the complete load determination required for such an analysis, but some
insight into the subject will be offered.
LOAD FACTOR
There are various ways of defining the load factor. It is the ratio of
the applied load to the weight of the airplane. More technically, it may
be described as the multiplying factor by which the steady flight forces
are multiplied to obtain equivalent static effect of dynamic forces acting
during the acceleration of the airplane. In common parlance, it is cus-
tomary to speak of the load factor as the number of <7's, where g refers to
the acceleration due to gravity.
The duration of the maximum dynamic loads on the airplane seldom
exceeds three seconds.
Strength requirements for aircraft stress analysis are specified in terms
of limit and ultimate loads. Limit loads are the maximum loads antici-
pated in service. Ultimate loads are equal to the limit loads multiplied
by the factor of safety.
50
EXTERNAL LOADS ON AN AIRPLANE IN FLIGHT 51
In considering dynamic loads on the airplane for flight conditions, it is
assumed that steady flight conditions prevailed just before the particular
maneuver or gust occurred. This assumption will then, usually, indicate
that some forces, such as thrust, will not have changed during the very
short time the critical condition is assumed to act. By the same token,
the speed of the aircraft will not have changed radically during the history
of a particular maneuver or gust condition.
THE AIRPLANE IN A MANEUVER
Some of the most severe loads encountered by the airplane in flight
occur in maneuvers, such as coming out of even a shallow dive.
Theoretically, the worst load condition should occur when the airplane
comes out of a vertical dive instantaneously. During the maneuver, the
airplane would change its angle of attack from one at which the lift co-
efficient was practically zero to an angle of attack where the lift coefficient
was maximum. The pertinent equations of interest are:
l = w = y2PvKin cLmaz s,
in horizontal flight at maximum lift coefficient; in the vertical dive
d = w = y2Pv^cDmins.
Immediately upon coming out of the dive, as in a sharp pull-out, the
speed would not have diminished and the maximum lift coefficient could
be attained. The lift on the airplane under this condition may be ex-
pressed as:
U = nW = y2Pv? cLmax s.
From these three relationships, it can be deduced that the load factor n
can be expressed in terms of the following ratios:
L1 CLmaz
n = t â€” â€”
L V% miâ€ž Cd min
By assuming possible values for CLmax and Cx>miâ€ž, we can gain some idea
of what magnitude such a load factor might have if other conditions did
not obtain. For example, 1.8 is a reasonable assumption for the value of
Cl max, and .025 is possible for the value of Cflm,â€ž; then the load factor
under these circumstances could be:
Â» = S = 720
However, the value is obviously fantastic. In the first place, the pilot
would probably not survive; and if the aircraft were designed for such a
load factor, it would be so heavy that, in all likelihood, it would never

get off the ground.

52
In actuality, the inertia of the airplane, as well as the air resistance,
prevent instantaneous reaction. Furthermore, the pilot's delay, caused
by his own psychological reaction, would prevent such an abrupt pull-out.
It is desirable to have a stick load that tends to resist the pilot's attempts
at a quick elevator pull-up.
The system of forces and moments acting on the airplane when coming
out of a dive is illustrated in Figure IV-I.
Figure IV-1. Forces acting on an airplane as it comes out of a dive.
THE AIRPLANE IN A GUST
Another flight situation that may produce extreme loads on the air-
plane occurs when an airplane encounters a vertical gust while in hori-
zontal flight. There is an immediate increase in the angle of attack,
which means there is also an increase in the lift coefficient and therefore
the lift. The resultant airplane speed is also increased. Referring to
Figure IV-2, it may be seen that the lift would be:
Li = L + AL = y2pD? (CL + ACL) S
when the gust struck the airplane, and
l = w = yfat cL s
just before the gust. Since even a vertical gust of about 50 per cent of
the aircraft speed would cause only a change of about 12 per cent in the
resultant velocity, vr may be assumed equal to v. Thus the load factor
expressed as the ratios of the two lifts would be:
Z,i L + AL _ CL + ACL _ ACL
X = -^- 1 + Anâ€”n CT~ = 1+ CT
HyH
Figure IV-2. Velocity and angle relationships at the time an airplane encounters
a vertical upward gust.
But since
C-W,
L y2psv*
and tan a = a radians = â€”'
and ACl â€” mAa Ta&umÂ»
where m is the slope of the lift curve, it follows that:
11/ i / W i i pmuv
n = 1 + m- Ty-sr> = 1 +
MpSv* a 1 2(W/S)
Since the gust encountered is seldom sharp-edged (i.e., the gust in-
tensity varies from a minimum to a maximum over a finite distance),
and since the aircraft is a flexible rather than a rigid structure, the gust
load factor is somewhat modified and may be written:
KmuV
n = 1 Â±
575 (W/S)
(since the vertical gust may act in either direction)
where V is in miles per hour,
u is in feet per second,
K is a function of the wing loading (See Figure IV-3),
1.467p _ J_
2 575'
1.4
1.2
1.0
tO.8
^ 0.6
(a
0.4
0.2
0 10 20 30 40 50 60
Wing loading- j?, lb/ft*
Fioubb IV-3. Graph for determining gust load factor.
Examination of the formula reveals that, all other factors remaining
the same, the load factor increases with:
1. increase in the aspect ratio of the wing,
2. increase in forward speed, V, of the aircraft,
3. intensity of the gust, u,
4. decrease in the wing loading, W/S.
Since for any given design, the slope of the lift curve is determined once
and for all upon selection of the airfoil and the aspect ratio, the wing
n * 1 + Ln
- / f KUVSm
575W
-/
1
1
1
loading is also fixed. The load factor attained by an airplane would
then be dependent upon the factors u and V. In order not to reach
excessive load factors, the airplane would either have to avoid severe
gust disturbances, or slow down when a gust was expected, as when
flying through a line storm.
The relative intensities and the altitudes at which gusts have been
encountered may be appreciated by examining the data given below.1
Velocity of gust, in
feet per second
Altitude limit,
in feet
Line Squalls
and Thunderstorms
40 to 120
20,000 or more feet
Obstructional
Interferences
10 to 30
varies with
type of
obstruction
Convection
Currents
5 to 30
4000
These gusts are usually vertical and may act either up or down, although
gusts may act in any direction, depending upon their provocation.
EXPERIMENTAL DETERMINATION OF THE LOAD FACTOR
Load factors are usually presented in an empirical form, but the student
should know the bases for the determination of these factors. In order
to ensure the structural integrity of the design, it is necessary to know
the magnitude of the load factors encountered under various conditions of
flight. For this purpose, airplanes used for flight tests are equipped with
instruments designed to obtain the pertinent data that engineers evaluate
and synthesize for the designers.
Restraining
spring
Smoked glass
target
Figure IV-4. A schematic diagram of a mechanically-operated accelerometer.
'The figures are approximate.

56
Loads in flight are recorded by means of a special accelerometerâ€”also
called a V-g recorderâ€”which is shown schematically in Figure
This recorder keeps a record of the velocity and the acceleration, and
when supplemented by the pilot's record, supplies a full history. A
calibrated grid (see Figure IV-5) is placed over the smoked-glass record
so that values of g, or the acceleration, may be read for corresponding
velocities. When removed from the recorder, the smoke film is fixed
with a thin fixing lacquer that protects it from damages.
Figure IV-5. A sketch of the calibration grid placed over the smoked-glass target.
The full scale deflection is of the order of three quarters of an inch. The horizontal
scale is in miles per hour; the vertical is in number of g'a or load factor.
By studying the data acquired in this manner, it is possible from time
to time to adjust the empirical formulas so that reasonable load factors
can be calculated mathematically.
This V-g recorder is usually placed at the center of gravity of the air-
plane and measures the vertical component of the acceleration of the
airplane. The horizontal component could also be measured, if desired,
by another accelerometer in which the mass is restrained vertically but
not horizontally. The horizontal and vertical axes here referred to are
those of the airplane.
It is usually necessary to measure only one componentâ€”the vertical or
normal component of the accelerationâ€”since, with this value known, the
horizontal component can be determined from the aerodynamic charac-
teristics of the airplane.
Correlation of the data will result in the necessary correction factor to
be applied in the theoretical formula evolved for the gust condition, for
example, or it may lead to either arbitrary assignment of load factors
for certain maneuvering conditions or even to empirical formulas using
various parameters such as. power loading, wing loading, or gross weight.
AIRPLANE CATEGORIES
Not every airplane must be designed for the same conditions since not
all airplanes are likely to be subjected to the same maneuvers. For the
purpose of stress analysis, airplanes are divided into certain categories.
The Civil Aeronautics Board establishes the regulations governing the
structural integrity as well as the airworthiness of the civil airplane.
These regulations may be obtained from the U. S. Government Printing
Office, Washington 25, D. C.
The categories of airplanes are generally listed as follows:
1. Normal. Non-acrobatic, non-scheduled passenger, and non-sched-
uled cargo operation airplanes are included in this category.
2. Utility. This refers to airplanes used in normal operations and
limited acrobatic maneuvers such as steep turns, spins, stalls, lazy eights,
and chandelles, but not snap or inverted maneuvers, or whip stalls.
3. Acrobatic. These have no specific restrictions as to type of maneu-
ver unless flight tests indicate certain limitations. The specified load
factors are highest for this category.
4. Transport. These are airplanes to be used in scheduled passenger
and/or cargo operations.
5. Restricted. This includes any airplane not covered by any other
specified category.
EMPIRICAL LOAD FACTORS
The maneuvering load factor for the air-transport airplane may be
found from the following empirical formula. It applies to power loadings
between 12 and 24 pounds per horsepower, provided the value of the load
factor does not exceed 4.33 or is less than 1.5.
n = 1 + An
_ [ 32,000 1 / [ 3.25 1
71 |Û.77 + w + g200j I | 0.435 J
For the other categories, the maneuvering load factors are specified
as follows:
XT 1 Ol 1 24000
Normal n = 2.1 +
W + 10,000
but n need not be greater than 3.8, and shall not be less than 2.5;
Utility n = 4.4;
Acrobatic n = 6.0.
The gust velocities of Â±30 feet per second are specified for speeds up
to cruising speed, and Â±15 feet per second for speeds up to the design
dive speed.
58
-I
-2 1
^4QV 80 120^160
|200 240 --28jp Speed,
in mph
Maneuver Gust
- Combined
Figure IV-6. A typical envelope for an airplane in the utility category. The con-
ditions indicated by the small circle are the ones usually investigated. In the illustration
shown, the gust conditions happen not to be critical.
The loads calculated for the maneuvering and gust load factors are
considered to be limit loads.
THE FLIGHT ENVELOPE
The flight envelope (V-n and V-g diagram are alternate names) estab-
lishes the relationships of limit load factors and airplane speeds. (See
Figure IV-6.) It represents, in empirical form, data similar to that
obtained in experimental flight tests from the record on the target plate
of the accelerometer.
While the limit load factor may be the same for several speeds and the
negative limit load factors may be numerically less, it must be remembered
that not only the magnitude of the applied forces, but their distribution
c,
Figure IV-7. Aerodynamic coefficients (forces) referred from the wind axes to the
geometric axis.
will affect the design of some part of the airplane structure. It is con-
ceivable that the front spar of the wing might be designed for one load
factor when the flaps, for example, are neutral, but the rear spar might
be designed for another but a lower load factor when the flaps are de-
flected. Again, while some flight loads may design no part of the wing,
certain parts of the fuselage or empennage may be affected by the
design loads.
REFERENCE AXES
For the aerodynamicist, it is usually convenient to use a set of so-called
"wind axes" where the line of action of the relative wind usually repre-
sents the abscissa. For the stress analyst, whose work concerns the
Table IV-1. Correcting aerodynamic coefficients for aspect ratio Ri.
1. Cl
-.6
-.4
-.2
.2
.4
.6
2.2
2. a Ro
3. Aa = ^ K Cl
4. Â«Rl = (2) + (3)
5. CoiorRo
7. CdBi = (5) + (6)
8. cos a = cos (4)
9. sin a = sin (4)
10. Cl cos a = (1) x (8)
11. Cd sin a = (7) x (9)
12. Cn = (10) + (11)
13. Cl sin a = (1) x (9)
14. Cd cos a = (7) x (8)
15. Cc = (14) - (13)
16. Cm.
K = - ir = W when Ro = 00
ill tt0 xli

60
geometric configuration of the structure where the reference line may be
the chord line, the thrust line, an axis of symmetry, or some other, it is
more convenient to use a geometric set of axes.
It is not difficult to transfer a set of aerodynamic coefficients from one
set of axes to another. Referring to Figure IV-7, it may be seen that,
using the chord line as the abscissa of the new set of geometry axes,
Cn = Cl cos a + Cd sin a,
and
Cc = â€”Cl sin a + Cd cos a,
where Cn is referred to as the normal force coefficient or beam component,
and Cc as the chord force coefficient or chord component. The calcula-
tions may be carried out in tabular form, as in Table IV-1, and presented
graphically, as shown for a typical case in Figure IV-8.
It should be noted that except for low angles of attack, the chord force
is towards the leading edge, not the reverse. This does not mean that
the local pressures, as at the nose of the airfoil, are negative, for such is
62
These forces may be either specified in terms of load factors or evaluated
from other relationships, as follows:
1. niW = applied limit load,
= qCrfS,
where ni is the specified vertical component of the limit load factor for
the appropriate gust or maneuvering condition for a given speed. CN
can therefore be calculated.
2. nxlW = qCJS,
where nxl is the horizontal component of the limit load factor. Cc cor-
responds to the same angle of attack as the Cn value above, so that when
Cn has been found, Cc can be found and nxl determined.
3. nx,W = -Tor - ^^-r-
where T or PT is the thrust or power required to fly at velocity V for which
the limit load factor has been specified.
4. n3W
is the tail balancing load and can be determined for the particular force
system.
5. n2W and nxlW
are the components of the inertia forces required to satisfy the equilibrium
conditions and can now be calculated from the equations satisfying
SF = 0 and SX = 0.
Since the forces may act in various directions and may have any magni-
tude, depending upon the conditions to be considered, it is convenient
to consider all forces and moments acting in the positive direction. By
substitution of the known or specified values with the proper sign (plus
or negative as the case may be), the unknowns may then be accordingly
evaluated as to magnitude and sign.
Comparing the force diagram for the airplane in a maneuver (Figure
IV-1) with the generalized system of forces (Figure IV-9), it will be
noted that some assumptions have been made. These may be explained
as follows:
1. The drag of the tail surfaces, the fuselage, etc., is usually considered
negligible for smaller airplanes, but some allowances should be made for
these forces for large airplanes.
2. The moment of inertia due to the angular acceleration is omitted for

many conditions since its magnitude is usually insignificant. If it is
specified, an additional term appears in the equation for SAf = 0.

The term n is called a load factor, for which the specific definitions are:
The load factor or acceleration factor n is then the ratio of a load to the
design weightâ€”generally the gross weight of the airplane. When the
load in question represents the net external load acting on the airplane
in a given direction, n represents the acceleration factor in that direction.
The limit load is a load (or load factor, or pressure) that, assumed
or known, may be safely experienced but may not be exceeded in opera-
tion. The factors wi and nxl are usually designated as the components
of the limit load factor.
The factors n2 and nxi are the components of the inertia load factor.
The factor n3 is known as the net tail load factor or tail balancing load.
OTHER DYNAMIC LOADS
The discussions so far have dealt primarily with flight loads. There
are many conditions, other than maneuvering or gust, that may impose
loads upon the structure.

CHAPTER V
Materials of Construction
APPLICATION OF AVAILABLE MATERIALS
Many structural materials have been used for building airplanes. Early
designs were built up primarily of wood elements with steel fittings and
wires, and the whole covered with doped fabric. It was not long, how-
ever, before tubular steel was used for the truss-type fuselage and landing
gears. The wing construction, however, was made of wood and fabric.
As the airplanes became larger, the old methods of construction were
not very economical, and it also became difficult to get the material in the
large sizes required. Reinforced monocoque, making use of aluminum
alloy sheet, was a natural evolution and is the most common form of con-
struction today.
Small airplanes, carrying from 1 to 5 passengers, may still employ a
wing with wooden spars, aluminum alloy ribs, steel wire drag bracing, all
enveloped in a doped fabric covering to give the structure the final airfoil
shape. The fuselage of such airplanes is generally a tubular steel truss-
type, with possibly wooden former strips to fair the lines, and the whole
covered with doped fabric.
The larger the airplane becomes, the more likely that all the structure,
the wing, fuselage, and tail surfaces, is composed of aluminum alloy sheet
and extrusions.
Another material that may be extensively used, especially for very large
airplanes where sizes are determined by allowable stresses rather than by
minimums established by fabrication or maintenance practice, is stainless
steel. Comparatively few airplanes have been made of this material to
date.
Wood, in the form of plywood, has been used for gliders and for air-
planes when strategic materials were lacking. However, the fabrication of
wooden structures of this type, while successful, still cause maintenance
and inspection difficulties so that deterioration due to extremes in tem-
64
MATERIALS OF CONSTRUCTION
65
perature, bacteria, moisture, and the like have not encouraged extensive
exploitation of this material.
Magnesium alloy may be used in place of aluminum alloy to some extent,
but as yet its use has been limited.
Experimental designs using glass fibers held together with synthetic
resins and impregnated fibers have been built and test flown. Although
these materials have certain merits, the wide use of a material depends
largely upon:
1. Availability of the raw material.
2. Manufacturing facilities for fabricating sheet, extrusions, and billets.
3. Cost.
4. Ease in fabrication of the final structure with the specific material.
5. Reliability of the mechanical properties of the material.
6. Ease of maintenance of the final product.
7. Susceptibility to deterioration due to moisture, chemicals, tempera-
ture, etc.
Magnesium alloys are particularly suitable for instrument casings, en-
gine housings, and the like.
Chrome molybdenum and chrome vanadium steels, and some low-carbon
steels are used for fittings, struts, and some hardware.
Stainless steels are suitable for exhaust stacks, heating units, firewalls,
and other specialized uses.
ALUMINUM AND ALUMINUM ALLOYS
The information listed here on aluminum and aluminum alloys is quoted
from various publications of the Aluminum Company of America.
Commercially pure aluminum has a minimum aluminum content of 99
per cent, the balance being other metals normally present, mainly iron
and silicon. It has a high degree of resistance to corrosion and can be
easily formed into intricate shapes. It is relatively low in strength, how-
ever, having a tensile strength of about 12,000 pounds per square inch in
its soft or annealed state. The pure metal, therefore, does not have the
strength required for structural airplane parts.
Higher strengths may be obtained by the process of alloying, that is, by
adding small amounts of one or more other metals to the aluminum. The
resulting alloys are less easily formed and, with some exceptions, they
have lower resistance to corrosion than the commercially pure metal.
Alloying is not the only way of increasing the strength of aluminum.

Like copper, brass, steel, and other metals, aluminum becomes stronger
and harder as it is rolled, formed, or otherwise cold-worked. Since the
hardness depends on the amount of cold work done, some of the wrought
aluminum alloys are available in several work-hardened tempers. As the

66
strengths are increased, there is some sacrifice in the metal's ability to be
worked easily and successfully into intricate commercial forms.
Heat treatment provides another means of increasing the strength of
some of the aluminum alloys. With but minor exceptions, the aluminum
alloys used for structural parts in aircraft are those which can be heat
treated to obtain higher strengths. When finally installed in the airplane,
such materials must be in the heat-treated condition.
Classification and Nomenclature
Aluminum alloys are obtainable in either cast or wrought forms. They
may be further classed as nonheat-treatable and heat-treatable alloys.
The different alloy compositions are distinguished by alloy numbers.
Wrought alloys are distinguished from cast alloys by the letter "S" fol-
lowing this number.
An alloy in the heat-treated temper (such as 24S-T) may be further
cold-worked, such as by rolling, to obtain still greater strengths. It is then
said to be in the "RT" temper. Although this further rolling reduces the
elongation and workability, 24S and Alclad 24S in the "RT" temper are
Table V-l. Temper designations.
Type
Description
Temper
designation
Example
Annealed condition
24S-0
Heat treated and quenched but
not aged completely
W*
61S-W
Heat-
treatable
alloys
Wrought
Full heat-treated condition

24S-T
Heat treated and subsequently cold
worked a prescribed amount
RT
24S-RT
Cast
Specific tempers attained by differ-
ent heat treatments are indi-
cated by number
T4, T6,
etc.
195-T4
195-T6
Annealed condition
3S-0
Nonheat-
Wrought
Intermediate tempers produced by
cold working various amounts
#H Y2H.
%H
53S-KH
treatable
alloys
Full hard temper obtained by max-
imum commercial degree of cold
working
2&-H
Cast
No temper designation
67
regularly corrugated and formed into other aircraft parts. Temper
designations for alloys are shown in Table V-l.
Sheet
A large percentage of the material in the modern airplane is aluminum
alloy sheet, used for the outer covering or skin and for various other struc-
tural parts. Since most of this sheet is subject to high stresses under some
flight conditions, high-strength alloy 24S-T is almost universally used.
(Since the war, most new designs are making use of 75S.) Nonstructural
parts are sometimes made from the alloys that are not heat-treated or
from one of the lower strength heat-treated alloys.
Approximately three-fourths of the 24S and 75S sheet used in aircraft is
in the form of Alclad sheet. This material has a high-strength core cov-
ered on each side with a coating of relatively pure aluminum which has a
high resistance to corrosion. This coating protects the core in two ways.
It not only covers most of the core and prevents contact with corrosive
agents, but also protects the core electrolytically. This electrolytic pro-
tection is given to the cut edges, to areas where the coating is abraded
away, and even to rivet heads.
Annealed coiled sheet is used for blanking out small parts which are
subsequently heat treated.
Table V-2. Flat sheet and plate (Alcoa standard sizes).
Thickness
(in.)
61S-T
24S
O and T
Alclad 24S
OandT
Alclad 75S
O and T
0.012
0.016
24 X
36 X
120
144
36 X 144
0.020
36 X
144
36 X
144
36 X 144
0.025
36 X 144
48 X
144
48 X
144
36 X 144
0.032
48 X 144
48 X
144
48 X
144
48 X 144
0.040
48 X 144
48 X
144
48 X
144
48 X 144
0.051
48 X 144
48 X
144
68
Tubing
Tubing is used in aircraft construction for fuel, oil, and other liquid lines;
instrument lines; and structural parts, either in the airframe itself or in
places such as the control system or power plant.
Extruded Shapes
An extruded shape is produced by forcing hot (but not molten) metal
through a die in which there is an opening corresponding in shape to the
desired cross section. This process frequently provides more efficient
utilization of metal than rolled shapes, and it permits production of many
shapes that cannot be rolled.
Extruded shapes are used as stringers and other important structural
parts of the airframe, and provide easy assembly as well as maximum
structural usefulness. The alloy 14S, which has been used extensively
for several years in forgings, has lately become available for extrusions in
thicknesses of inch and greater. Extrusions of this alloy, in general,
are significantly stronger than those of 24S alloy, which has been most ex-
tensively used up to the present for structural members of airframes.
(Again, new designs are making use of 75S, which is the strongest alumi-
num alloy available.) Extruded shapes not requiring high strength are
sometimes made from 63S or 61S in a suitable temper, or from 2S or 3S.
These last two alloys are used in the "as-extruded" condition which is very
slightly work-hardened, and not in a specific temper.
Table V-3. (Refer to Figure V-l.)
B&S
Thickness,
Range of dimension,
Sheet Gage
in inches
in inches
28-21
.012-.028
Vs-IH
20-18
.032-.040
V2-VA
17-16
.045-.051
1 -3
Vs-2
15-14
.057-.064
13-12
.072-.081
1^-4
1 -3
11-10
.091-.102
2 -5
2 -3^
R for all thick-
nesses
Forgings
Forgings make up a relatively small percentage of the total weight of an
airframe, but they serve a most important function as structural and other
fittings in addition to their well-known uses for making engine parts and
propeller blades. Most forgings are made in dies on hammers, but some
69
structure.
Extruded sections
Formed or drawn sections

Figure V-l. Typical formed, drawn, and extruded sections used in the aircraft
70
are pressed rather than struck. These are called press forgings. The
latter require less "draft" than hammer forgings and permit somewhat
closer tolerances. Many forgings are made from alloy A51S-T, whereas
14S-T is usually employed for highly stressed forged fittings. Alloy
75S-T forgings are also available, and are being used in several new de-
signs.
Castings
Aluminum castings are more extensively used in the airplane engine and
for accessories than in the airframe itself. Depending on the quantities
and properties required, such castings are produced by one of three basic
methods: sand, permanent-mold, or die. Sand castings are adaptable to
a wide range of products from the smallest to the largest. Permanent-
mold castings are used in order to obtain higher mechanical properties,
better surfaces, or more accurate dimensions. Die castings find applica-
tion in cases where relatively large production of a given part is involved,
since this will distribute the die costs over a large number of units, and at
the same time reduce machining costs because of closer tolerances and
better finish.
Any shape which can be forged can also be produced as a casting. While
forgings are sometimes preferred because of their greater strength and
homogeneity, it may be necessary to produce complicated shapes as cast-
ings. Many castings are furnished "as-cast," but some alloys are heat-
treated to obtain increased strength.
Wire, Rod, and Bar
The products in the group known as "Wire, Rod, and Bar" find their
application in aircraft principally in the form of structural fittings ma-
chined from rod or bar, usually 24S-T or 75S-T. Other uses include
screw-machine products. Rivets are made from special rivet wire.
Rivets and Screw-Machine Products
The most widely used rivet alloy is A17S-T. Joints requiring higher
strength are made with 17S-T or 24S-T rivets.
Nuts, bolts, screws, and various special screw-machine products for air-
craft use are made almost exclusively from 24S-T and 17S-T.
Bolts, rivets, and other types of fasteners used in aircraft structures
have been standardized by the military services. The drawings and

specifications for these parts are available in the AN Standard Aircraft
Parts book or in pertinent Federal Specifications. The designer may

therefore call for any of the standard parts simply by listing the AN
number on his drawing, rather than having to detail each feature of the
part.
71
AN bolts are coded for length by inches and eighths of an inch, and for
diameter by sixteenths of an inch. For all lengths below one inch, the
single figure means the number of eighths; above one inch, the firsi num-
ber of two digits indicates the number of inches and the second diît the
number of eighths.
Example: Aluminum Alloy Bolt
AN 5 DD â€” 6 Indicates aluminum alloy aircraft bolt:
5/16 in. dia.; 6/8 in. long.
length in eighths of inches
aluminum alloy
diameter in sixteenths
Army and Navy Standard Part
When the letters DD are omitted, the bolt is made of heat-treated nickel
steel with a minimum tensile strength of 125,000 psi.
Rivets are similarly coded, as shown by the following example:
Example: Rivet
AN 426 D 6 â€” 14
length in sixteenths of an inch
diameter in thirty-seconds of an inch
alloy used
indicates type of head
Army and Navy standard part
Steel bolts smaller than %-inch diameter or aluminum alloy bolts
smaller than 34-inch diameter are not used in the primary structure of
an airplane.
In general, rivets, bolts, and other pins should be subjected primarily
to bearing and shear rather than tension; and pinned connections, whether
rivet, bolt, or other devices, should be designed to fail in shear of the pin.
AIRFRAME FABRICATION
Aluminum alloys can be fabricated on the machinery and equipment
normally employed for other metals. The details of methods, tools, and
72
dies depend on the particular alloy and temper, the quantity of identical
parts involved, and the nature of the parts themselves.
Many of the manufacturing methods developed by the aircraft industry
have been the result of conditions peculiar to the industry itself, not to the
materials used. For example, aluminum alloys can be formed successfully
on the large mechanical presses which are so common in the automotive
field. However, the quantities involved in the aircraft industry have not.
until recently, warranted investment in the tools and dies required by
these presses.
Cutting and Blanking
One of the first fabricating operations involves the cutting of rectangular
sheets of metal into the sizes and shapes required for the individual parts
of the airframe. The methods used can be classified as shearing or cut-
ting.
Equipment for shearing includes guillotine shears (either power- or foot-
operated) for making straight cuts; roll shears for curves; male and female
blanking dies; and a variety of modifications of the foregoing. To obtain
sheared edges with a minimum of burr, the cutting edges must be smooth
and sharp and set with a clearance equal to about 7 per cent to 10 per cent
of the thickness of the metal being cut. The use of metal dies and rubber
pads for blanking sheet metal was very popular in the aircraft industry
before the present war because the tool cost was low and the small quanti-
ties did not justify the cost of male and female blanking dies.
Two types of cutting are used for sheet stock; sawing and milling (or
routing). Band saws are sometimes used for cutting stacks of sheet-metal
blanks, and of course, both band and circular saws are used for miscella-
neous work on sheet and extruded sections. High blade speeds (5000 feet
per minute for band saws and 10,000 feet per minute for circular saws) are
desirable for best results. Saw teeth should be fairly coarse with a slight
set.and front rake. A more recent method for cutting stacks of sheet-
metal blanks involves the use of routers. The cutter is a small-diameter
milling tool which turns at high speed on a vertical axis. It is guided by
a template clamped to the work and cuts a path all around the blanks
equal in width to the diameter of the cutter. This method leaves the
edges of the blanks in a smoother condition than when the metal is sawed.
The versatility of these machines and the speed with which templates for
new or altered shapes can be provided have made routing the standard
method for cutting dd-shaped blanks in many aircraft plants.
Forming
Most of the forming work in the aircraft industry is done on sheet. In
general, the equipment consists of basic machines that are fitted with spe-
73
cial tools or dies for each part they are to process. Thus a double-action
mechanical press (of which there is an increasing number in the aircraft
industry) is usually equipped with a number of sets of punches, dies, and
blankholders. The cost of such tools for large parts led the industry to
seek other methods prior to the war.
Single-action presses, usually hydraulic when large in size, are used ex-
tensively. The large hydraulic presses are usually fitted with a rubber
pad on the ram, both for blanking and forming. Because there is no tend-
ency to "iron" or draw out wrinkles when metal is formed over the edge
of a flat die (metal, wood, or synthetic material) with a rubber pad, it is
common practice to notch the flange and also to provide flutes or scallops
in the flange to use up the excess metal. Parts formed by this method
usually have a more or less flat surface surrounded by a flange (as, for
example, a curved channel), a wing rib, or a flanged bulkhead. Mechan-
ical presses equipped with metal dies are used for a number of forming
operations and simple draws, usually on small parts.
Drop hammers, using zinc or zinc alloy dies and lead punches cast in the
die cavity, are very extensively used. Many of the parts produced in this
manner could be drawn on a double-action press at a lower unit cost, but
a large number of parts is required to justify the higher cost of steel dies.
Sheet panels for wing tips, fins, and control surfaces as well as fairings and
fillets are made by this method.
Sheet panels having very slight double curvature are sometimes formed
by stretching them over a metal or hardwood form. The sheet is wrapped
over the form and is clamped along two opposite edges. The form is then
raised so that the sheet becomes tight over the higher parts. Further
motion of the form stretches the tight parts until the panel fits the form
at all points.
Apron brakes, folders, cornice presses, rolls, and draw-benches are used
in making the endless variety of parts requiring bends that are not curved.
These include corrugated sheet and sheet molding for structural uses in
the airframe. This equipment and its operation is typical of that used by
the better class sheet-metal shops fabricating other materials as well as
aluminum.
Considerable handwork is used in forming aircraft parts. Some of this

work is done by skilled power-hammer men who form sheet panels to ob-
tain the necessary curvature or crown. Other parts are made by ham-
mering over or into various wood forms. Even in production work where
tools have been provided, some hand straightening and adjusting may be
necessary. In general, larger quantities justify more nearly perfect tools
which, in turn, reduce or eliminate handwork. Experimental airplanes
sometimes warrant very little tooling.

74
MAGNESIUM ALLOYS
Magnesium alloys have been found useful for a number of applications
in the aircraft structure. The following information has been obtained
from publications of the Dow Chemical Company.
Castings
Magnesium alloy castings are used where light weight, high strength,
resistance to shock, and good machinability are desired. Magnesium
castings find widespread usage in aircraft engines and aircraft landing
wheels. Magnesium alloy permanent-mold castings are being used for
such parts as nose pieces on aircraft engines, fittings, wheel flanges, pis
tons, brackets, and various types of instrument and accessory housings.
Forgings
Magnesium alloy forgings include aircraft engine bearing-caps, bearing
housings, rocker arm supports, door and aileron hinges, engine mountings,
valve and pump bodies, hydraulic cylinders, miscellaneous control levers,
brackets, and fittings.
Extrusions
Magnesium extrusions include such applications as stiffener elements in
aircraft structures, floor beams, and moldings.
Sheet
Magnesium sheet and plate have been used for such parts as aircraft
wings, wing tips, ailerons, fairings, fuel and oil tanks, cases, floor plates,
ducts, and seats.
STEEL
The steel numbering system commonly used in the United States is that
of the Society of Automotive Engineers, which offers the following ex-
planation of the system:
A numeral index system is used to identify the compositions of the S.A.E.
steels, which make it possible to use numerals on shop drawings and blueprints
that are partially descriptive of the composition of material covered by such
numbers. The first digit indicates the type to which the steel belongs; thus
indicates a carbon steel; "2-" a nickel steel and "3-" a nickel chromium steel.
In the case of the simple alloy steels the second digit generally indicates the ap-
proximate percentage of the predominant alloying element. Usually the last two
or three digits indicate the average carbon content in "points," or hundredths of

1 per cent. Thus "2340" indicates a nickel steel of approximately 3 per cent
nickel (3.25 to 3.75) and 0.40 per cent carbon (0.35 to 0.45); and "71360" indicates
a tungsten steel of about 13 per cent tungsten (12 to 15) and 0.60 per cent carbon
(0.50 to 0.70).
In some instances, in order to avoid confusion it has been found necessary to
depart from this system of identifying the approximate alloy composition of a
steel by varying the second and third digits of the number. An instance of such
76
The application of titanium and titanium alloys is being constantly
expanded as the knowledge and the development of the material increase.
HARDWARE AND OTHER ITEMS
There are many items which the airframe designer does not design, but
which he instead selects to meet specific design requirements. Every
design office has the detailed dimensions, weights, and other specifications
pertaining to these parts available in the catalogs of the suppliers. Among
the items that may be treated in this manner are large ones such as engines,
propellers, radio and radar equipment, wheels, tires, and seats, and also
smaller items consisting mostly of hardware such as rivets, bolts and
nuts, washers, pulleys, cables, bearings, etc. In selecting these, the
designer must consider weight, maintenance, space, strength, and a host
of other factors pertinent to the particular problem involved.
Table V-5.
USS 18-8 Half Hard
PHYSICAL DATA (MINIMUM)
Yield Strength 110,000 psi
Tensile Strength 150,000 psi
Elongation in 2": Class
I II
0.015" and under 9% 15%
0.016" to 0.030" inc.... 10% 18%
0.031" and over 10% 18%
FINISH
Annealed, Pickled and Cold Rolled
APPLICATIONS
Fire Walls and Angles
Cowling
Columns
Spars
Air Intake Manifolds
Radiators
Fuel Tanks
Oil Tanks
Shear Webs
Fuselage Covering
Wing Cover and Wing Connection

Wing Radiator
Water Tanks
Tail Surfaces
Bomb Racks
Bomb Shackles
Bomb Chutes
Feed and Ejection Chutes
Gun Blast Cowl
Window Guides
Struts
Stainless steels.
USS 18-8 Full Hard
PHYSICAL DATA (MINIMUM)
Yield Strength 140,000 psi
Tensile Strength 185,000 psi
Elongation in 2": Class
I II
0.015" and under 3% 8%
0.016" to 0.030" inc.... 4% 9%
0.031" and over 4% 9%
FINISH
Annealed, Pickled and Cold Rolled
APPLICATIONS
Stabilizer, Horizontal Frame Ribs
Columns
Wing Spar (complete all parts)
Fuselage Cover and Structure Parts,
Corrugation
Struts
Wing Cover
Aileron Struts and Covering

77
Table V-6. Steel and steel alloys in aircraft.
APPLICATION S.A.E. APPLICATION
S.A.E.
STRUCTURE AND ACCESSORIES
Bearings, Roller
Bolts, Eye and Wing Attach-
ment
Clevises.
Clevis Pins
Engine Mount Structural Tub-
ing
Engine Mount Steel Forgings,
Strut Ends
Fittings, Brackets, Hinges,
Bushings
Fuselage Frame Tubing.
Nuts
Pins.
Racks, Bomb.
Rivets
4615
3312
52100
3140
4140
2335
X-3140
2330
1035
2330
X-4130
4640
X-4130
2330
X-4130
X-4340
X-4130
1025
2330
3140
6150
4140
2330
X-4130
2315
2340
3250
4620
52100
X-4340
3140
Screws.
1120
2330
3140
6150
Seat Frames X-4130
Shackles for Cable 2330
Spring Parts 1065
1095
Structural Tubing X-4130
4140
Tail Group Tubing X-4340
X-4130
Rudder:
Formed Shapes X-4130
Terminals 2330
Tie Rods 2330

Table V-7. Useful structural applications of nonferrous and ferrous alloys.
Commercial
designation
Available
form
Gas
weld
Use and limitations
28 (MH)
3S (HH)
Tubing
Sheet
Yes
Yes
Electrical conduit.
3S (HH)
3S (HH)
Bar
Tube
Yes
Yes
Welded fuel and oil tanks. Good weld, low fa-
tigue strength.
Tank flanges, bosses, etc.
On parts to be welded only.
24S (0)
248 (T)
248 (T)
248 (T)
Sheet
Sheet
Bar
Tube
No
No
No
No
Structuralâ€”Heat treat after forming.

Structural. Superseded by 75S.
General structuralâ€”Extrusions available.
Structural, 53ST and 61ST, may be used instead
for long columns. Although not commonly
carried in stock.
52S (MH)
52S (HH)
528 (0)
Sheet
Care
Care
Care
Cowling and fairing subject to vibration.
Spot Weld. Good on all 52S.
Fuel, oil, and instrument lines.
Bar
Tube
43 (Alcoa)
195 (T4 Alcoa)
220 (Alcoa)
356 (T4 Alcoa)
Casting
Casting
Casting
Casting
Yes
No
No
Yes
Nonstructural. May be cast in thin sections.
Pedals, gear boxes, camera hatches.

MATERIALS OF CONSTRUCTION 79
Table V-7. (Continued). Useful structural applications of nonferrous and ferrous alloys.
Commercial
designation
Available
form
Gas
weld
Use and limitations
SAE 1095
Sheet
Wire
No
No
Aircraft flat springs.
In all coil springs.
SAE 2330
Bar
No
Aircraft bolts, eyes, forks and terminals.
SAE X-4130
Sheet
Bar
Yes
Yes
Welded structural details. Purchased annealed.
High-strength welded. Wing and landing gear
fittings or machine parts under J4-inch thick.
Tubing
Yes
Structural. Excellent weld. Purchased normal-
ized.
SAE 4140
Bar
Yes
Yes
Welded structural parts. Over H-inch thick.

Forging
SAE 4340
Bar
No
High-strength forged or machine parts. Over 2
inches in thickness.
SAE 6150
Bar
No
No
Gears. Purchased annealed.
All important springs.
Wire
18-8 (Annealed)
18-18 (}A Hard)
Sheet
Sheet
No
No
Good "spot weld," easily formed, nonmagnetic.
Good "spot weld," structural. Ammunition chutes
and boxes.
18-8 (Annealed)
18-8 (Annealed)
Tube
Bar
No
No
Good "spot weld." For exhaust system.
Carpenter No. 8 free machining. Cannot be heat
treated.
Inconel
Sheet
80
Table V-8. Decimal equivalents of gauges used on aircraft material.
Birmingham
Brown & Sharpe
Wire Gauge
Aluminum and its alloys;
U. S. Standard
(all tubing)
Magnesium and its alloys
Steel Sheet
in inches
Sheet, in inches
in inches
.120*
.114*
.125*
.109
.102
.109
.095*
.091*
.094*
.083*
.081*
.078*
.072
.072*
.070*
.065*
.064*
.063*
.058*
.057
.056
.049*
.051*
.050*
.042*
.045
.044
.035*
.040*
.038*
.032*
.030
.034
.028*
.032*
.031*
.025
.028
.025
.022*
.025*
.022*
.020
.020*
.019
.018
.018
.017
.016
.016*
.016*
.014
.014
.014
(Note: On all materials, decimal equivalents of common fractions are generally used
for thicknesses over J00^-inch.)

CHAPTER VI
Detail Design Considerations
There are many guiding principles in design that are equally applicable
to the design of the wing and the design of the fuselage. The student has
been exposed to many of these principles in his initial drawing course.
There have been further additions made to his store of technical knowl-
edge through various courses preparing him for his first attempt at a
design problem. Therefore, it is not the purpose of this chapter to re-
view all of the aspects of detail design, but rather to point out some
specific helpful hints on the subject.
DESIGN PHILOSOPHIES
The approach to a specific design problem may often be influenced by
the initial design philosophy.
Margin of Safety
Aircraft structures, for example, are generally designed to obtain the
lightest structure for a given set of conditions. In keeping with this
design philosophy, each member should be chosen so that its strength
margin of safety is zero. (Margin of safety is the difference between the
maximum load or stress a member can withstand without failing and the
maximum load or stress that is likely to be imposed under any condition
in the normal operation of the aircraft.) However, the smallest margin
of safety may not always mean the lightest member, as the illustration in
Table VI-1 shows. If one were to select a tubular member to withstand
a torsional moment of 10,500 inch-pounds, either the 1% X .049 tube
with an Fty = 85,000 psi or a 1^ X .065 tube with an Fty = 75,000
would be satisfactory; yet, the one with the slightly higher margin of
safety is considerably lighter.
81
Table VT-1. Allowable torsional moment for SAE X 4130 steel tubes for length0
diameter.
L/D* Ratio = 20
Tube
Weight
Allowable
sise
lbs/100 in.
moment
Fty = 75,000 psi Ft, = 85,000 psi
\% x .049
7.42
10,180 in.-lb. 10,720 in.-lb.
IH x .065
9.05
10,650 in.-lb. 11,210 in.-lb.
IH x .095
9.76
10,870 in.-lb. 11,400 in.-lb.
*L = Length; D = Diameter
Maintenance Requirements
Sometimes maintenance may be an overriding consideration. It is
conceivable that a wheel brake assembly could be designed to withstand
the loads imposed upon it; but if it failed to stand up for more than 100
landings, it would not be considered an economical item and would,
therefore, be redesigned to meet certain maintenance requirements.
Fool-Proof Operation
A system that has to be operated must be designed so as to avoid dam-
age that may result from any deviation from correct operating procedure.
It is desirable also to avoid designs of assemblies in which it may be pos-
sible to interchange parts having similar outward appearances but different
characteristics.
The "Fail-Safe" Principle
A structure may be designed so as to be considered "fail-safe." In
other words, should one member of a complex structure fail, some other
member would be called upon to carry the load. Such design does not
always require a 100 per cent margin of safety, but entails a certain type
of structure or arrangement of members to compensate for the failure of
any one member.
Safety
Safety may require additional consideration in design. It may be
satisfactory, for most purposes, to have only one access to a cabin, but
in case of an emergency, this might be insufficient. A hydraulic system
to operate the retraction of a landing gear may be excellent, but if it fails
to function, can another system be used? Safety considerations should
always be taken into account by the designer to avoid situations which
may have unfortunate consequences. He must foresee certain prob-
lemsâ€”the movement of a control system chafing the insulation of an

DETAIL DESIGN CONSIDERATIONS
83
electrical system; a foreign object falling into some mechanism and hinder-
ing its operation; excessive deflections causing the malfunction of another
partâ€”and he must prevent them by a suitable and effective design.
Whatever structure is designed, stress analysis will be the final de-
terminant. One does not say that one structure is stronger than another,
but rather that one design is more efficient than another, meaning that
the material has been used more effectively. This efficiency refers not
only to individual elements but also to an assembly of elements.
The "One-Horse Shay" Principle
For optimum design, an assembly of elements (provided there is no
conflict with some other design philosophy) should be built according to
the principle of the "one-horse shay"â€”each part fails at the same time.
For example, in an assembly or panel, working stresses are so obtained
by the distribution of material between skin and stiffeners, that insta-
bility of the skin occurs simultaneously with the primary and secondary
instability of the stringers.
0.002 e-strain-inches per inch
Figure VI-1. Typical tensile stress-strain diagrams: (A) Material having a definite
yield point (such as plain low carbon steels); (B) Materials not having a definite yield
point (such as aluminum alloys-high carbon, alloy, and cold worked steels); (C) Alumi-
num-covered aluminum alloys.

84
STRENGTH OF METAL AIRCRAFT ELEMENTS
The strength properties of the most commonly used materials and ele-
ments in the aircraft industry may be found in a publication entitled
Strength of Metal Aircraft Elements, published by the U. S. Government
Printing Office.
Typical stress-strain curves are shown in Figure VI-1. It is generally
desirable to know the yield and the ultimate strength of materials, since
the basic allowable stress of a material should not exceed the yield. How-
ever, since the yield value is sometimes arbitrarily defined, the ultimate
stress may be used as a reference. Where this is done, the allowable
stress is assumed to equal the ultimate stress divided by the factor of
safety of 1.5. This may be stated as
where Fy = yield stress of the material, in tension or compression for
example,
Fu = ultimate stress of the material, in tension or compression
for example,
Jy = safety factor = 1,
Ju â€” safety factor = 1.5.
The allowable stresses may vary from the mechanical properties of the
material since appropriate allowances must be made for column effects,
method of fabrication, form factors and conditions, stress concentration,
fatigue, and the like.
A knowledge of the behavior of the structure under load is important
(1) for obtaining the optimum efficiency of the structure as to its allowable
load and strength-weight ratio, and (2) for designing the structure so
that excessive deflections do not interfere with the safe operation of
equipment and control systems that may be attached to it.
Stress analysis covers the determination of the stress state in a member,
and the selection of a material or member with mechanical properties
that can withstand the likely maximum stress. It is necessary to know
the00loads imposed upon the structure as well as the structural configura-
tions and materials employed, for the stresses caused by the limit loads
must not exceed the allowable stress of the material. In order to avoid
permanent sets, the applied stress must fall below the yield point of the
material.
In building up a complex structure composed of many open sections
and much thin material, it is important to understand not only the types
of failures possible in the complete structure, but also those to which
Eallowable
STRUCTURAL BEHAVIOR
85
individual members are subject. Failures of these members are generally
a result of instabilityâ€”either primary or local.
A column having a closed circular cross section may suffer a compres-
sion failure either through lateral deflection, usually at the middle of
the column (primary instability), or through collapse of the tube wall,
at a stress lower than that required to produce a general column failure
of the primary instability type.
Structural
Sectisn
Figure VI-2. Structural behavior.
Compression Bending
Efficient as a column
for slenderness ratios
less than 140.
Heavier than
round tube.
Equivalent
round tube.
to
Relatively efficient
as open section; bet-
ter when section is
closed.
Efficient for small
members and for D/t
ratios less than 50.
Better than a round
tube when bending
about the vertical
axis.
Equivalent
round tube.
to
Compares with
square tube when

closed.
Torsion
Efficient for small
members and for
D/t <70 and L/D
<100.
About equivalent
to a round tube.
Equivalent
round tube.
to a
Compares with
square tube when
closed.
Efficient when sup-
ported against lat-
eral deflection.
Buckles easily about
vertical axis when
unsupported.
Efficient when com-
pression side is sup-
ported. If unsup-
ported, inefficient
about vertical axis.
Somewhat ineffici-
ent in torsion.
Relatively inefficient
unless supported.
Buckles easily about
diagonal axis when
unsupported.
86
An I-beam may fail by sidewise deflection of the entire compression
flange extending over a comparatively large distance, or by local wrinkling
of one or both of the thin outstanding flanges.
In general, structural members should be investigated for (1) failure of
the material, (2) resistance to primary instability failure, and (3) re-
sistance to local instability failure. The type of failure that occurs at
the lowest stress should be used as the criterion in design.
STEFFENERS OR STRINGERS
The cross sections of the structural members by which sheet panels are
supported play an important role in the stability as well as in the strength
of the built-up structures. Such supports may be transverse ribs in a
wing, transverse frames in a fuselage, or longitudinal stringers (or stiffen-
ers) in either the wing or fuselage.
Some understanding of the behavior of these structures may be ob-
tained from the tabulation of representative cross sections of structural
members given in Figure VI-2. It will be noted that closed sections are
generally best, open symmetrical sections are next in line, and open un-
symmetrical sections are least efficient. However, local support, as
afforded by sheet covering, shear webs, or plates, increase the stability
and thus the maximum load-carrying ability of the individual members.
The effect of local supports and of the addition of a lip to a free edge is
shown comparatively in Figure VI-3. Open sections are generally pre-
ferred in reinforced monocoque stricture because of accessibility in rivet-
ing or bolting. The analysis is based upon the assumption that the
various component flanges of a structural member may be considered as
panels supported or free along the edges parallel to the direction of the
applied compressive load.
1.0
b â€ži, b
i jÂ±tÂ±L
5.2 4.3
Relative compressive strength
5.9
Figure VI-3. Compressive strengths of stiffeners. Assumptions: all loaded edges
(perpendicular to figure) were considered simply supported; edges considered freely

supported without lip; between simply and freely supported with lip; rivet line offers
condition between simple and clamped support condition.

87
To obtain an optimum design on the basis of strength-weight ratio, it
would be necessary to conduct a preliminary stress analysis for various
types of stringers. This would entail considering the different combina-
tions of skin thicknesses, stringer spacing and thicknesses, and transverse
frame or rib spacings. After obtaining such data, the next step would
be a comparison of the relative weights of the combinations. It would
be necessary to have stress data available on various cross sections of
stringers or stiffeners. With this, initial comparisons could be made on
the basis of crushing strength (defined as the end point of column curves,
projected to L/p = 0 drawn through data) obtained from tests on columns
of the section for a range of L/p values from 20 to 50, where L is the
length of the column, and p the radius of gyration.
In either the wing or the fuselage, much weight is often accumulated
by the addition of stringer splices. A closed section, for example, re-
quires heavier and more complicated splices than an open section to
^ =======a
Schematic representation
Possible deflection curves
Fixed or Free edges
clamped edges
Figure VI-4. Behavior of sheet under compression for different edge conditions.
Simply supported
edges
88
accomplish a satisfactory structure. It is therefore possible to save
weight with an open section, in spite of the fact that it can carry a lower
allowable stress.
PANELS UNDER LOAD
The behavior of a panel subjected to compression depends upon the
degree of flexibility or rigidity of the supports to which the panel is at-
tached. A schematic representation of the behavior of various edge
conditions is given in Figure VI-4, which also shows the possible deflec-
tion curves that a single panel could take if considered alone.
Figure VI-5 indicates the behavior of a panel attached to a series of
stringers or stiffeners. The sketches illustrate a portion of a wing, a
tail surface, or a fuselage under compression. For an originally curved
surface, the behavior would be similar. While at first it would seem that
a rigidly supported system (Figure B or Figure D) should result in the
lightest structure, this is not always the case, since the rigid supports may
contribute an inordinate amount of weight. It would be necessary to
Figure VI-5. Relative behavior of different types of metal sheet-stringer construc-
tion when under compressive stress. (A) The torsionally flexible stringers rotate and
carry the attached thin sheet with them. The stringers in this case are considered to
offer only simple support for the sheet between the stringers. (B) The torsionally stiff
longitudinal stringers and shear webs prevent rotation and offer clamped edge conditions
for the sheet covering. (C) The torsionally rigid longitudinal stringers rotate only
slightly and thus offer a condition between the simple and clamped edge conditions.
(D) The sheet, considerably thicker relative to the integrally incorporated longitudinal
stringers, takes over. The stringers, in this case, tend to reduce the spanwise undulation
of the sheet covering.

DETAIL DESIGN CONSIDERATIONS 89
make comparative studies for specific design conditions to ascertain the
optimum design. Even then, other factors, such as fabrication and pro-
duction, may cause the reconsideration of a design which was otherwise
thought to be optimum.
PANEL SIZES
The number of transverse frames or ribs can be determined in part by
the permissible dimensions of the skin panels of a given thickness.
The metal covering not only serves to give the aerodynamic form to the
wing, tail surfaces, or fuselage, but must do so under varying conditions.
It should not, for instance, develop wrinkles or an undulating surface
under normal air pressure as this would materially add to the aerodynamic
resistance. Furthermore, it should be strong enough to withstand the
stresses resulting from compression or shear. Usually, for preliminary
design, a stress analysis is either not available or is impossible to perform
without some idea of the likely configuration of the structure.
The dimensions recommended to prevent oil-canning provide a ready
means of determining not only the panel sizes, but also the spacing of
Figure VI-6. Chart for determining proportions of unsupported sheet covering for
exterior surfaces so as to prevent oil-canning and undesirable vibration leading to
fatigue cracking. R is the radius of curvature in inches. If the ratio of panel length
to panel width is greater than 3, either the panel width should be reduced from the
values obtained from the chart or the thickness of the sheet should be increased.
90
Figure VI-7. Chart for determining proportions of unsupported sheet covering for
interior surfaces so as to prevent oil-canning and undesirable vibration leading to
fatigue cracking. R is the radius of curvature in inches. If the ratio of panel length
to panel width is greater than 3, either the panel width should be reduced from the
values obtained from the chart or the thickness of the panel sheet should be increased.
frames, ribs, stringers, or other supports. Figures VI-6 and VI-7 repre-
sent charts for determining suitable panel sizes. Although subsequent
stress analysis may indicate a change in the skin thickness, this does not
necessarily entail a change in the spacing of supports. For aircraft of
the utility category, a skin thickness of not less than .025 inches for
aluminum alloy is recommended. This is usually sufficiently thick for
the stresses encountered at most of the airplane surface. Where greater
bending moments are likely to occur, as at the root of the wing or the
central portion of the fuselage, a thickness of .032 to .040 inches may be
assumed. An examination of Figures VI-8 and VI-9, which show the
variation of the coefficient K with variations in panel size and edge con-
ditions, reveals that the ratio of the two sides of a rectangular panel should
be less than 1 for panels in compression, and 1 for panels subjected to
shear. However, while this is true from a strength point of view, having
many stringers, frames, or other supports may not always be best from
the standpoint of weight and construction complexity. The optimum
relationship of the dimensions of a panel depends to some extent upon
the individual problem.

91
2345
Ratio sf panel dimensisns a/b
Figure VT-8. Envelope of values of K in the formula given on the chart. The
following conditions are represented: (1) ends and sides of panels clamped; (2) ends
simply supported, sides clamped; (3) ends clamped, sides simply supported; (4) ends
and sides simply supported; (5) ends simply supported, one side free, one side clamped;
(6) ends simply supported, one side free, one side simply supported. The values at the
right of the chart are the asymptotic values for each of the six conditions.
111
- n . tu
r 11 j i
I u,00^â€”.â€”â€”-,m
) ii U\
p*
( \\[ (
11
.n T
1,s
F-A
scr
h a â–
K$
10
9\
234567
Rath of long side to short side /+J
Figure VT-9. Graph for determining critical shear stress factor K,; a and 6 are
between adjacent skin supports, whether ribs, stringers, spars, or other supports.
92
It will also be noted that for a ratio of panel dimensions of 3 or greater,
the value of the coefficient K in the appropriate formulas remains con-
stant. A ratio of the sides of about 3 would seem reasonable for most
of the cases encountered.
Figure VI-10. The solid line in the top illustration represents the distribution of
the compressive stress on a reinforced panel after the sheet has buckled. The equivalent
stress distribution used for calculation on the basis of "effective widths" of the sheet is
shown by the dashed lines.
STRENGTH OF SHEET-STRINGER COMBINATIONS
When a combination sheet-stringer panel is subjected to compression,
the sheet will tend to buckle at a lower stress level than the stringer, and
will thereafter carry no more load. The stress distribution of such a
sheet-stringer combination is illustrated in Figure VI-10. The solid
line depicts the stress distribution chordwise at the time the C-stringers
have reached their maximum allowable stress. Since the sinusoidal
nature of the stress distribution is somewhat difficult to calculate, an
approximation is made by assuming a uniform stress distribution, as
shown by the dotted lines in the above figure. The extent of this uni-
form stress distribution is determined by the so-called "effective widths,"
as illustrated in Figure VI-11.
The effective widths may be calculated by means of the following
formula:
edge
.Â»W2*jÂ»WlÂ»0
edge
Li
<â€”H
Li-
edge
Â»Â»2| "2
"2 | "2
"2 | "2
-*
Jedge
Freej
-1
lu
u-
Fiodbe VI-11. The effective widths, u,i and tc2, are measured from the rivet center
line.
where w = effective width, in inches,
Cv> = .60 for wi, as shown in Figure VI-11,
= .85 for w2, as shown in Figure VI-11,
t = thickness of sheet, in inches,

E = modulus of elasticity, psi,
Fe = allowable column stress of the combination of the stringer

and the effective widths of the sheet.
It will be noted from Figure VI-11 that the stringer or stiffener re-
quiring two rows of rivets makes more of the skin effective than does the
stringer requiring only a single row. Moreover, the stringer having a
closed section with the skin is itself a more effective compression member.
However, the greater cost of two rivet rows must be0 0 considered against
the strength advantages.
Unless the proportion of the load carried by the effective skin is ap-
preciable, it is often impractical to determine in preliminary design the
true effective width of skin. For a first approximation, therefore, it is
satisfactory to consider an effective width of skin of 30< on each side of
the rivet line.
EQUIVALENT STRUCTURE
In considering a wing or fuselage cross section as a cross section of beam
in bending, it is immediately apparent that the effective centroid of the
cross section is not the center of gravity, for less material is effective on
the compression side of the beam than on the tension side.
In designing any beam for bending, it is desirable to distribute the
material so that it is located as far from the neutral axis as possible. A
sheet-stringer combination, however, involves another factor, as each

94
(o)
Actual location of skin
(b)
Figure VI-12. (A) A typical sheet-stringer construction; (B) the equivalent effec-
tiveness under compression. The solid line represents the shear elements, and the
solid circles the bending or compression elements.
sheet-stringer element under compression acts as a column under com-
pression. Since the stringer is restrained from buckling in a direction
parallel to the metal sheet covering but is only slightly restrained in the
direction perpendicular to the skin, it is desirable to have a large moment
inertia for that axis to resist any buckling tendencies. The effect of this
is shown schematically in Figure VI-12.
CUT-OUTS
Cut-outs caused by doors, windows, access or inspection plates, landing
gear retraction, and the like represent an interruption of the structural
integrity of the unit where they occur. Recourse to hydraulic analogy
(see Figure VI-13) indicates that high local stresses occur at the corners
of the cut-out, and that the structural elements ahead and behind the
cut-out are not carrying any stress unless some distortion takes place
elsewhere to transfer some of the load.
The condition of stress around a cut-out in a reinforced monocoque
structure is similar to that of a hole in plate, as illustrated in Figure VI-14.
Doors may be designed to carry loads through the structure if proper
provision is made for transmitting the loads. However, such factors as
loose fits and requirements for emergency operation prevent this from
being an easily-established condition.
Figure VI-13. Hydraulic analogy may offer an insight into the likely stress dis-
tribution around a cut-out. In the hydraulic analogy shown above, the obstruction in
the channel could represent a window in the side of a fuselage. The greatest stress con-
centration is at the corners. The flow ahead and behind the obstruction is practically
at a standstill. Translated to the structural counterpart, the structure ahead and
behind the cut-out is carrying no load or is not stressed.

95
Ui.
Figure VI-14.
jected to tension.
Stress distribution across the section through a hole in a plate sub-
Note the greater intensity adjacent to the hole.
Windows are not considered to carry loads through the structure. It is
true that pressure on them is transmitted, but the installation and the
material of the windows are such that they are not thought of as pro-
viding continuity to the structure for stress analysis purposes.
To enable the structure around the cut-out to adequately carry the
loads imposed upon it, provision must be made for reinforcements such
as additional doubler plates or rigid frames. Figures VI-15 and VI-16
show some possible solutions.
RIVETED AND BOLTED JOINTS
Should failure occur in riveted joints, it is generally better that it occur
by shearing of the connecting element, the rivet, rather than by bearing
and/or tearing of the sheet or plate.
Aluminum alloy rivets are available in Y%, % and % of an
inch diameters. The selection of the diameter size depends upon many
factorsâ€”the ease of upsetting the shop-head, the thicknesses to be joined,
the loads to be transmitted, and the size of rivet pattern possible. For
simple or straightforward sheet connections, a diameter approximately
equal to the total thicknesses of sheet to be joined may be assumed until
Figure VI-15. Treatment of cut-outs and openings. Rigid frames, rounded
corners, and doubler plates help to redistribute the stresses. The arrows indicate the
principal direction of stress.

96
(a) (b)
Figure VI-16. Two possible treatments of cut-outs such as windows, doors, and
access holes. (A) is appropriate for a section subjected to relatively high shear and
compression loads. Doubler plates or increased skin thickness may be added where
the stress would otherwise be high. (B) is suitable, for example, near the neutral axis
of a fuselage where the stress is low. The frame around the cut-out may be subject to
induced loads.
the detail design has progressed far enough to allow a study of the effect
of a different rivet size.
Bolted joints are used where the thickness of the plates or material to
be joined is large enough to warrant their use.
Whatever the method employed of joining two or more members to-
gether by means of pinning, consideration has to be given to making the
parts accessible in the course of assembly and to providing the necessary
clearance between adjacent pins (whether rivet or bolt) so that the tools
used to upset the rivet head or to tighten the nut on the bolt clear all
adjacent interferences.
[! ^
!A1
'a'J
it
6d
!00<M-
!! c
-2d
>
Figure VI-17. Rivet patterns. Spacing for single and double rows expressed in
terms of rivet diameters.

97
Figure VI-17 offers a guide to the development of a rivet row or rivet
pattern. Each manufacturer will have specific instructions available in a
drafting room manual or some similar reference. However for student
work, the simple instructions outlined in the figure should suffice.
The spacing of the rivets attaching the skin to the stiffener may often
determine the allowable load that the structural element consisting of
the stiffener and skin may carry (if buckling of any element determines
the upper design criterion) since buckling of the skin between rivets is
likely to occur first. The plastic buckling stress of the compressive skin
between the rivets may be obtained from the formula:
3.62 Eâ€¢
'(P/O2
where Fc = plastic buckling stress, psi,
Et = tangent modulus of the material,
p = rivet spacing or pitch, as measured between the center of ad-
jacent rivets, in inches,
t = thickness of skin in inches.
If the allowable stress should be considered too low, then obviously the
designer may change the value by either increasing the skin thickness or
changing the rivet spacing.
FITTINGS
Points at which concentrated loads are transferred from one complex
structure (landing gear members, engine mount supports, wing spars, tail
surface spars) to another (the fuselage or the wing), require a special study
of methods of transferring the loads. This must be taken not only from
the point of view of stresses incurred, but also from the standpoint of
secondary effects that may result from slight misalignments, initial buck-
ling, or excessive deflection of the adjacent supporting structure.
It is poor design to attach a rigid fitting to a thin sheet such as a spar
web or fuselage skin because of the instability of the thin sheet and the
high concentration of stresses. Figures VI-18 and VI-19 illustrate a
few simple applications of fitting attachments and splices.
For preliminary design, the point of departure in designing a fitting is
to consider the pin or pins to be used for joining the two or more members
together. After the selection of the pin or bolt, the necessary dimensions
for determining the configuration of the fitting become apparent.
It is advisable to have concentric fittings wherever possible. The

simplest fitting of this type would involve one pin subjected to double
shear, as shown in Figure VI-20. The minimum dimensions of the male
lug may be found by the method outlined below.

98
Figure VI-18. Illustration of a spar fitting. The steps are introduced to avoid
sudden changes in cross section. A tapered tang could be used but spot-facing would
be necessary to assure proper seating of bolt heads and nuts.
Figure VI-19. Examples of splices. Splicing members may be riveted or bolted,
depending upon the method of connecting members in other parts of the same structure.
An internal wrenching bolt may be used for the tension splice shown in bottom figure
at the right.
Figure VI-20. Principal dimensions of a fitting lug. The dimension W should not
be less than 2R-d. Where the shank behind the bolt hole is long, additional calculations
should be made for bending stresses, especially those that may be introduced by eccen-
tric loads, whether due to misalignment or applied.

99
Step 1. Determining Bolt Diameter.
where P. is allowable single shear strength of bolt obtained from Table
VI-1, and
P is tension or compression in pounds.
Step 2. Determining Thickness of Fitting Lug.
P = 2<F*,
where Ft* is allowable bearing stress of the plate. Select nearest stand-
ard thickness.
Step 3. Determining Value of R.
P = (2R - d)tFty.
Step 4. Checking whether Dimension R is Sufficient.
P = 2xtFâ€ž
where F, is the allowable shear stress of the plate and where x may be
assumed equal to R.
The illustration given should be helpful in determining proportions of
other fittings involving more bolts or plates.
SANDWICH MATERIALS
Two metal sheets with a relatively thick low-density material placed
between them is generally called a "sandwich material." This type of
construction is used to increase the greater rigidity and stability of thin
sheet material under load.
The low-density material or core may be balsa wood, foam rubber,
fiberglass, or another synthetic compound of low density and high volume.
This material serves primarily to give added support to the metal sheet
and thus increase the allowable stress that the material may take, whether
it is in compression or shear. The. core, when properly bonded to the
exterior sheets, prevents panel flutters.
The sandwich core may also be comprised of relatively high density
materials such as resin-impregnated paper or thin sheet metal, when
these materials are fabricated in honeycomb or a similar configuration.
(See Figure VI-21.) When this type of core is employed, the sandwich
has an added strength or stability normal to the outside sheets, and it
therefore lends itself to flooring or any purpose where normal loads are
likely to be high.
100
Figure VI-21. One form of sandwich construction is illustrated here. The core is
a honeycomb made of metal, fiber, or plastic. Such construction is used for flooring,
leading and trailing edges of tail surfaces and control surfaces.
The major disadvantages of sandwich construction are the difficulty of
bonding the faces to the core, the extensive tooling required by the compli-
cated configurations, and the local reinforcements and built-in fittings
necessitated by the application of concentrated loads.
FLOORING
Flooring, whether in the cabin, cockpit, or cargo compartment, requires
a type of structure that is rather rigid to the load normally applied to the
surface. For that reason, the cross section of the structure must have a
relatively high moment of inertia in order to reduce the deflections and
also a surface that is resistant to easy local penetration.
Consider, for example, a square plate of homogeneous material that is
simply supported along its edges (a conservative estimate) and subjected
to a uniformly distributed load of w pounds per square inch. The maxi-
mum deflection, occurring at the center, would be:
where b is the length of sides of the square,
t is the thickness of the plate, and
E is the modulus of elasticity.
For such a plate, it is obvious that steel would have approximately one
third of the deflection that aluminum alloy would have, in the ratio of
Eal. alloy/Eâi ,
but would weigh approximately three times more (in the ratio of specific
weight of steel to the specific weight of aluminum alloy). However, if
the thickness of the material used was increased, the deflections would
101
(a)
'U U U U LT^1
(b)
(c)
(d)
(e)
0â€”u hâ€”u uâ€”u uâ€”u uâ€”u
(f)
Figure VI-22. Cross sections of several types of flooring are illustrated: (A) an
example of sandwich construction; (B) an integrally stiffened plate; (C) a "beaded"
plate; (D) a corrugated sheet attached to a flat sheet on one side or to a flat sheet on
both sides; (E) a built-up section for a heavily loaded floor; (F) relatively closely-spaced
C stiffeners. The assemblies may be riveted, spot-welded, or bonded.
vary as the cube of the thickness and thereby accomplish more. Chang-
ing the dimension b would be helpful, provided that the transverse supports
were not the determining influence in the over-all deflection of a floor
comprised of many individual squares.
In reducing the deflection of a plate, much can be accomplished through
the incorporation of beads or integral stiffeners. Still more may be done
by using sandwich construction with a relatively stiff core, or a flat sheet
backed up by corrugations. (Figure VI-22 shows some representative
cross sections of floors.)
Further insight into the desirable characteristics of flooring construc-
tion will be gained through a consideration of the deflection formula of a
beam of uniform cross section loaded uniformly:
wl4
102
Obviously, the greater the moment of inertia of the cross section, the
smaller the deflection. Sandwich construction generally provides a ready
means of obtaining a high moment of inertia.
The requirement for floor loading depends in large part upon the type
of payload or cargo that is carried. For a distributed load, up to 40 or
more pounds per square foot is common; for concentrated loads, the
weight of a passenger over a square foot area may be assumed. For con-
centrated loads, such as is caused by a lady's high-heel shoe, the total
weight of the passenger over an area of one square inch would be ap-
propriate.
THERMAL PROBLEMS
In the internal combustion engine, the dissipation of excess heat was
accomplished either by liquid cooling and ultimate rejection of the heat
through the medium of air passing through the radiator through which
the liquid was circulated or, more directly, by using the circulating air to
absorb the heat directly, without a liquid intermediary. This type of
thermal problem is an ever-present one in any heat engine, and the engi-
neering means for alleviating it are readily available.
0l I I I I I I I
0 100 200 300 400 500 600 700
Exposure temperature, degrees F
Figure VI-23. Typical graph of strength characteristics of Aluminum Alloy 75S-T6
extrusions at elevated temperatures. Not only is the strength less, but the creep char-
acteristics of material at elevated temperatures will cause displacements, misalignments,
and other distortions.

Figure VI-24. Experimental structures for aircraft exposed to high-temperature-
producing conditions: (A) utilizes relatively closely spaced corrugated webs useful for
both heat dissipation and rigidity; (B) utilizes rigid trusses whose members serve to
drain the heat from the surface and function as a heat sink. In both figures, the top
skin has been removed to show the interior construction.
For short periods of excessive heating (as caused by braking), a tempo-
rary heat sink may be provided by the use of a large mass of material with
relatively high heat absorption. Since the time cycle of heating is short,
a considerable period may elapse afterwards for the cooling of the heat
sink.
A new thermal problem has presented itself with the attainment of
faster-than-sound speeds. At these high speeds, the aircraft surfaces
become heated due to aerodynamic friction. At Mach 2 (twice the speed
of sound), the metal sheet covering may reach a temperature of 275Â° F,
and at Mach 3, it may reach 650Â° F. At speeds as high as Mach 4 to 5,
the steady-state temperatures of airborne missiles may reach from 1000Â°
to 2000Â° F.
For internal spaces occupied by people and temperature-critical equip-
ment, air conditioners must be supplied. For the structure, materials
and methods of construction have to be used which:
1. retain their high strength characteristics at elevated temperatures,
such as titanium alloys (Figure VI-23 shows some typical strength charac-
teristics of a metal at various elevated temperatures), or
2. serve as a heat sink (requiring much material), or
3. allow dissipation of the heat by providing large surfaces about which
air may be circulated, under pressure if necessary.
Much research is presently being conducted to find an effective way of
meeting this new challenge. Figure VI-24 shows some possible solutions.
MISCELLANEOUS
Various methods are employed to strengthen a structural member made
of a relatively large area of flat sheet.

104
x^5=
iff
Â»j
\N If
iV III
J/ V
â™¦1 if
1â€”Â«-
.J
F=2ru
A 4d
mm. mm.
j mm. B00B
Figure VI-25. Edge distances for lightening holes. All bend radii are at least
three times the thickness of the material (St). The dimension d is the diameter of the
rivet used to attach a member having lightening holes to another. The lip around the
lightening hole should be at least It in height.
~* Aâ€”>J
<â€”Aâ€”Â»Â»
\,
ft
heâ€”A-
h-A-H
.. V5i
Figure VI-26. Minimum dimensions required for A = 3d for thicknesses of riveting
clearances sections less than .036"; A = Ad for thicknesses between .036" and .081".
105
Lightening Holes
In a sense this is a misnomer, for the decrease in weight due to removal
of material is secondary. The primary purpose of a lightening hole is
either (1) to increase the in-plane stability by means of the raised Up
around the hole, or (2) to provide a passageway for control systems, wir-
ing, tubing, and the like, or (3) to provide an access to another portion
of an interior.
Ridges called beads or beading are used to increase the rigidity of an
otherwise flat plate.
Stiffeners
Small, relatively short-length angles may be attached to an otherwise
large flat sheet to increase its shear carrying ability and to reduce or
eliminate any tin-canning tendencies.
Figube VI-28. Detail showing interconnection of a stiffener or any transverse
member and a heavier structural member. The section could represent the chord
member of a tension field spar or a heavy beam in a fuselage, with the stiffener
corresponding to a transverse frame.
Beading
min. min.
Figure VI-27. Recommended edge distances for bolts.

Table VI-2. Typical mechanical properties of wrought aluminum alloys.
Tension
Hardness
Shear
Fatigue
Alloy
and
temper
Yield
strength
Elongation
per cent in 2 in.
Brinell,
500-kg,
Endur-
Ultimate
strength,
lb/in.Â»
Shearing
strength,
lb/in.Â»
(set
0.2%),
lb/in.1
Sheet
specimen
Round
specimen
load
10-mm
ance
limit,
lb/in.J
(X in.
thick)
(H in.
diam.)
ball
2S-0
5,000
13,000
14,000
17,000
21,000
13,000
15,000
17,000
20,000
24,000
35
12
45
25
20
17
15
23
28
32
38
44
9,500
10,000
11,000
12,000
13,000
Table V00-00. Mechanical Ooserties of aluminum alloy.
2.0000-
3.000
5O000
3O000
i. 9,000,000000 0 000,000,000 Â»
5O000
3O000
36.. 93600,0000 0 000,000,000 >-
00,000
8O000
0.0000 lb/cu in. 0073 lb/cu ft
Clad 200O sheet and slate
Heat-treated by user
00.0000-
2.000
58,000
3O000
00000
00,000 1 58,000
3O000 3O000
i 0 0 p7000 0 00000
3O000
8O000
0.000-
00.000
00,000
3O000
i0 0 K00r
3O000
00,000
0.250-
0.0099
92,000
38,000
92,000
38,000
3O000
93,000
0.0 -
0.2009
9O000
38,000
9O000
38,000
3O000
36,000
<.0
58,000
3O000
Primarâ€¢
Oecond
58,000
3O000
Primar
Oecond
3O000
8O000
2.0000-
3.000
5O000
38,000
5O000
38,000
3O000
800,000
Bare 200O sheet and slate
Heat-t00ated by user
Table VI-5. Shear and tensile strengths of steel rivets, bolts and pins.
Low
Heat-treated steel
Material
carbon
Tensile strength, kips per
steel
Standard AN bolts
sq. in.
55
100
125
125
Shear strength, kips per
sq. in.
35
65
75
75
AN Std. bolt
Diameter
designation
of rivet,
Allowable single shear strength,
Tension, lb
bolt or pin,
in pounds
(in thread)
in inches
â€”
He
107
199
230
â€”
%
242
449
518
â€”
â€”
.112
345
640
739
â€”
â€”
430
798
920
â€”
â€”
.138
523
972
1,122
â€”
â€”
671
1,247
1,438
â€”
â€”
.164
739
1,372
1,584
Table VI-8. Properties of round tubing.
Out^
Weight,
side
Wall
Radius
Weight,
pounds
di-
thick-
Cross-
of
Moment
Section
pounds
100
ame-
ness
sectional
gyra-
of inertia,
modulus
D/t
100
inches,
ter
in
area,
tion,
/, tnÂ»
I/r, in'
ratio
inches,
steel
alumi-
in
inches
sq. in.
inches
num
inches
alloy
.022
.01576
.0810
.000103
.000825
11.38
.45
.16
.028
.01953
.0791
.000122
.000978
8.93
.55
.20
.028
.03053
.1231
.000462
.002466
13.39
.86
.31
.035
Table VI-8. (Continued.) Properties of round tubing.
Out^
Weight,
side
Wall
Radius
Weight,
pounds
di-
thick-
Cross-
of
Moment
Section
pounds
100
ame-
ness
sectional
gyra-
tion,
of inertia,
modulus,
D/t
100
inches,
ter
in
area,
/, in'
I/r, ,'â€žâ€¢
ratio
inches,
alumi-
in
inches
sq. in.
inches
steel
num
inches
alloy
VA
.035
.2023
.6507
.08565
.09136
53.60
5.73
2.04
.049
.2811
.6458
.11720
.12500
38.25
7.95
2.84
.058
.3311
.6427
.13677
.14589
32.30
9.38
3.35
.005
.3696
CHAPTER VII
The Three-View
The three-view of a projected design is composed of the top or plan view,
the front view or front elevation, and the side view or side elevation. It
corresponds to the photographs that might be taken of the top, front, and
side view of the completed airplane.
After the designer has his set of specifications, he makes a few sketches
of what he believes his final design should look like. It helps bim to visu-
alize arrangements more readily and it forms the basis of his detailed
weight estimate and subsequent balance calculations.
An experienced designer will first make a hasty "thumb-nail" three-
view sketch. It is sufficiently detailed to convey the basic ideas of .the
design and no further three-view is then made until the design has been
almost completely decided upon as to dimensions, correlation of wing,
engine, landing gear, and tail surfaces.
However, even the experienced designer usually finds it desirable to
make several three-viewsâ€”each one more accurate than its predecessor.
The "thumb-nail" sketch is excellent to determine the type, to initiate
the design, and to record particular features of exceptional nature that the
designer has in mind. However, the first design on which such work as
the weight estimate and balance diagram can be based is the preliminary
three-view.
Unless the airplane is radical in designâ€”departs wholly from conven-
tional designâ€”the data obtained from existing airplanes are the best
guides in proportioning the new project, and the new or uninitiated de-
signer should avoid "radical" or new designs until he has first worked
through a conventional design. The fundamental principles of airplane
design always hold, and there is no better way to understand them than by
working through conventional designs. "Radical" designs are usually
"radical" because they disregard fundamental principles with the result
that the designer will eventually be disappointed.
The outline which follows shows how standard data on existing airplanes
113
114
Â£
Figure VII-1. This illustrates how certain available data may be assembled in prep-
aration for the initial three-view. A represents the over-all length required for the
propeller hub, engine, engine accessories, etc., with sufficient clearance allowed for the
removal of the rearmost accessory; B the distance allowed for the pilot's cabin; C
the length of the cabin of about 30 (AT + 1) inches, where N represents the number of
rows of seats athwartship, and 30 inches the distance allowed between rows; D, equal
to about one-third over-all length, is the approximate location of the center of gravity
of the complete airplane in rear of the nose; E is equal to 2.5 to 3.0 times the mean
geometric chord length of the wing; F is allowance for desired headroom in cabin; G is
allowance for depth of the front-wing spar; L is the over-all length of the fuselage with
an engine located in its nose.
may be made of use in laying down the preliminary three-view. (See
Figure VII-1.)
In making a preliminary three-view, some knowledge of the installation
requirements of the power plant, cockpit and cabin is necessary. The
chapters on these subjects should be carefully studied before going too far
with the preliminary sketch, and chapters on wing design, landing gear,
power plants, and other parts should be referred to by the student.
The information necessary to draw up the first three-view can be as-
sembled easily, provided the specifications for the airplane have been given
and a study of existing airplanes which seem to meet the specifications is
made. From the information at hand, and a knowledge of the engine
specifications, it is usually simple to proceed along the following lines.
Table VII-1 presents important data to aid in drawing up the three-view.
Step 1. Estimation of Gross Weight.
The gross weight may be estimated in one of several ways:
(a) A first guess of the probable gross weight may be made, if the type
and number of engines are specified, from the engine power P which, mul-
STEPS IN ASSEMBLING PRELIMINARY DATA
FOR THREE-VIEW
THE THREE-VIEW
115
Table VII-1. Empirical data.
Wing Data
Area S, sq ft
Aspect ratio, R
Span 6, ft
Root chord Cr, ft
Tip chord Cr, ft_- â€”
Root airfoil
Tip airfoil
Mean geometric chord, MGC ft
Location of 25% of MGC projected on root chord, ft
Aileron area Sâ€ž, sq ft
Aileron span &Â«, ft.
Aileron chord C, ft
Flap area S/â€ž sq ft
Flap span, b/â€ž ft
Flap chord C/â€ž ft
Aileron area to wing area Sa/S %
Flap area to wing area S/,/S %
General dimensions (in feet or in inches)
Over-all span 6 -
Over-all dimensions
Over-all height to over-all span* H/b
Wheel tread to over-all span* T/b
Over-all length to over-all span* L/b
Over-all height H ~
Wheel tread, T
Over-all length L
Distance between center of gravity and tail post
Tail Surfaces
Vertical tail surfaces to wing area,* ST/S , %
Horizontal tail surfaces to wing area,* Sk/S . %
Fin area to vertical tail surfaces,* S//S,
Rudder area to vertical tail surfaces,* <Sr/S, %
Elevator area to horizontal tail surfaces,* S,/iSÂ» %

Stabilizer area to horizontal tail surfaces,* S,/Si, %
Horizontal tail surface area

Elevator area <S,
Stabilizer area S,
Aspect ratio of horizontal tail surfaces ftÂ»
Span of horizontal tail surfaces 6Â»
Chord of horizontal tail surfaces CÂ»
Vertical tail surface area Sv
[continued
* All obtained from averages of these same ratios compiled for existing airplanes of
similar type.
Table VII-1. (Continued).
Fin area S/
Rudder area Sr
Aspect ratio of vertical tail surfaces, R,
Span of vertical tail surfaces 6â€ž, ft
Chord of vertical tail surfaces, Câ€ž ft
Angles
Landing angle of airplane
Angle in side elevation between vertical through axle and line connecting
center of gravity, and line joining the center of gravity and the point
of contact with the ground at the outer wheel
Clearances with Ground
Propeller clearance, in.
Deflected elevator clearance, in.
Deflected flap clearance, in
tiplied by the average of power loading W/P for existing airplanes, will
give the value of the gross weight W.
(b) The gross weight of the proposed design may be calculated by de-
termining the disposable load as indicated by the specifications, and using
the known average ratios of disposable load to gross weight.
(c) A likely value of the gross weight may be arrived at by assembling
sufficient weight data, although this is usually difficult to do because the
three-view and its dimensions are required for such an estimate.
(d) If a design exists which meets the new specifications, then all the
necessary preliminary data may be obtained therefrom so that the subse-
quent steps outlined below become unnecessary.
Step 2. Estimation of Wing Area.
(a) The average ratios of wing loading are based upon the gross weight
and the actual wing area, which does not include that covered by the
fuselage.
The wing area may be determined if the wing loading is assumed from
the ratios calculated for existing airplanes, for:
S = WX^
where S = the wing area, to be found,
W = the gross weight, previously calculated,

W . â€¢
-g = the wing loading for airplanes of similar characteristicsâ€¢

(b) If the airfoil has been selected so that its characteristics are known,
THE THREE-VIEW
117
and if the landing speed has been either specified or assumed, then the area
of the wing could be calculated by formula:
where V = the specified or assumed stalling speed in feet per second;
p = 0.002378 at sea level where the maximum density of the air
is usually encountered; otherwise, the density corresponding
to the highest altitude at which landing takes place should
be taken;
= the maximum lift coefficient for the wing either with or
without a lift-increase device, depending upon whether the
latter is used or not. Lift-increase devices, such as the
flap, are used primarily to reduce the necessary wing area
required to meet the specified landing speed. Chapter
XVII, Design of the Wing, should be referred to for informa-
tion on selection of airfoils and more accurate determina-
tion of wing area.
If different airfoils are used along the span, a weighted average may be
used unless actual wind-tunnel tests on the particular wing are available.
For airfoil characteristics, refer to pertinent NACA reports.
Step 3. Determining Length of Span.
The ratios of dimensions for airplanes are calculated in terms of the
span of the wing since it is a convenient reference dimension and also has
some justification in the fact that other characteristics such as stability
and control are functions of the span of the wing.
In order to determine the span, it is first necessary to decide on the
aspect ratio R. Usually it has a value not less than 6, although aspect
ratios of 14 have been used for special purpose airplanes. Aspect ratios
larger than 14 may be usefully employed for gliders.
The aspect ratio of a wing is determined by the formula:
D (span)2 b2.
It = â€”: = -fj)
wing area b
where the wing area here includes that area of the wing blanketed by the
fuselage, whereas in the calculation of the wing loading W/S, the actual
wing area was used for the value of S. An approximation of the wing area,
blanketed by the fuselage (that blanketed by engine nacelles may usually

be neglected in these preliminary calculations), may be obtained by as-
suming a likely root chord for the wing and a likely width of the fuselage.
The area so calculated should be added to the area calculated in Step 2.
Using this total area, the span may be calculated from:
62 = RS or 6 = V~RS. Tr
118
Step 4. Drawing Up the Wing Planform.
With the wing area decided upon and the aspect fatio determined, a
suitable planform may be selected. If the wing is rectangular, the matter
is easy. If the planform is tapered, the tip chord may vary from to
% of the root chord.
When the approximate dimensions of the chord have been determined,
then other items relating to the planform may be decided upon, such as
whether (a) the leading edge should be perpendicular to the longitudinal
plane of symmetry of the airplane, (b) the trailing edge should be per-
pendicular to the longitudinal plane of symmetry, (c) the front spar (which
is usually located at constant percentage from the leading edge) should be
perpendicular to the longitudinal plane of symmetry, or (d) the locus of
the mean aerodynamic centers should be perpendicular to the longitudinal
plane of symmetry. A study of various planforms is presented in Chap-
ter XVII, Design of the Wing.
Step 6. Determination of the Mean Aerodynamic Chord.
Another useful dimension in proportioning the airplane is the mean
aerodynamic chord (MAC) which is determined for each wing planform.
Usually the mean geometric chord (MGC) is used, since it is more easily
calculated. Discussion of these two terms and how the values and loca-
tion of these chords may be obtained will be found in Chapter XVII.
The final center of gravity of the airplane is usually located at about
25 per cent of the MGC. It is desirable, therefore, to know the size and
location of this chord.
Step 6. Determining Length of Airplane.
Once the 25 per cent point of the MGC has been located, it is easy to
locate the tail post of the airplane since it is usually 2Yi to 3 times the
length of the MAC behind the quarter point.
Then, from the average of the ratios of length to span (L/b) for existing
airplanes, the over-all length of the airplane can be found since the span
of the wing was calculated in Step 3. The length of the new design is
then found from:
Measuring a distance forward from the tail post equal to L, less an allow-
ance of 1 or 2 feet for the elevator chord, will locate the approximate posi-
tion of the nose of the airplane.

where
L is to be determined,
b is obtained from Step 3,
L/b is ratio obtained for existing airplanes.

THE THREE-VIEW
119
Step 7. Locating the Engines.
For the purpose of discussion here, a twin-engine design will be con-
sidered. The diameter of the propeller to be used for each engine can be
found by formula (See Chapter XV, The Propeller). About a two-inch
clearance between the propeller tip and the side of the fuselage should be
allowed. If the nose of the fuselage is later faired even more, a greater
clearance will usually result, but this would not be undesirable. The
centerline of each engine would then be located outboard from the side of
Figure VII-2A. The top view of a three-view drawing of the airplane; 6Â» and 6â€ž
represent span dimensions of horizontal tail surfaces and ailerons, respectively; other
dimensions that could be shown are length, L, and location of mean aerodynamic
chord, C, and propeller diameter, D.
fuselage, a distance equal to [(D/2) + 2] inches. The fore and aft po-
sition of the propeller is located by taking into consideration that the
propeller plane, and vertical planes intersecting this plane at 5 degrees
ahead or 5 degrees behind at the center of rotation, should not cut through
the fuselage where any personnel is located. Usually, the placement of
the pilot as far forward as possible in the nose of the fuselage will locate
the critical personnel.
Once the propeller has been located, the rest of the engine nacelle can
be sketched in easily.
120
* Propeller clearance, d
Figure VII-2B. The side view of a three-view drawing of an airplane. L represents
the length of the airplane; d, the propeller clearance. The over-all height of the air-
plane, the distance between front and rear wheels, as well as the angle of incidence of
the root chord, and the position of the mean aerodynamic chord could also be shown
in this view.
Step 8. Planfonn of Horizontal Tail Surfaces.
The hinge line of the horizontal tail surfaces is usually located close to
the tail post. The approximate area of the horizontal tail surfaces may
be determined from the average ratios of horizontal tail surface area to
wing area (that is, Sh/S) and since the wing area has already been deter-
mined, the horizontal tail surface area can be calculated. Reference to
Chapter XIX, Tail Surfaces, will indicate the necessary procedure in
laying out a desirable planform.
Step 9. Completing the Top View.
The top view of the airplane can be completed by sketching in the miss-
ing lines. The width and length of that portion of the fuselage containing
personnel can be obtained from consideration of the data given for seating
in Chapters X and XII, The Cockpit and The Passenger Cabin.
Figure VII-2C. The front view of a three-view drawing of an airplane. A represents
the span; B, the wheel tread; C, the dihedral angle. The dihedral angle of the tail
surfaces and the propeller diameter could also be indicated in this drawing.
B
THE THREE-VIEW
121
Step 10. Completing the Side View.
The top view furnishes the necessary information as far as lengths are
concerned. The depth of the fuselage is determined from consideration
of personnel requirements. The depth of the engine nacelles is dependent
on the dimensions of the side view of the engine used.
The wheels are chosen on the basis of the static load placed on them,
and their placement in relation to the fuselage and the wing can be ob-
tained according to the design considerations given in Chapter XVIII,
The Landing Gear.
Step 11. Completing the Front View.
Most of the essential dimensions are now determined so that the front
view is easily completed.
GENERAL NOTES FOR THREE-VIEW
The propeller should have a minimum ground clearance of 9 inches
when the airplane is in the horizontal or level landing position with the
shock absorbers and tires deflected as they would be under the normal
gross weight of the airplane. For all normal considerations, the shock
absorber may be assumed to be deflected about two-thirds of its normal
travel and the tires about one-quarter of their normal travel. Proper
allowance must be made for the configuration of the landing gear when
the members are so disposed that the deflection of the shock absorber may
cause a greater deflection of the landing gear. Generally, it is not wise
to allow more than 9 inches of clearance for the propeller since this will
tend to increase the length of the landing-gear members and thereby
increase the weight of the landing-gear structure.
Propellers on seaplanes should clear the water by at least 18 inches
when the seaplane is at rest.
A clearance of at least two inches between the tips of propellers or any
part of the structure should be allowed also.
The elevators should clear the ground by at least 2 or 3 inches when the
airplane is in the three-point landing position or at rest.
For the location of the various parts such as the wing, tail surfaces and
landing gear, which are placed in relation to the center of gravity, it is
desirable to assume the center of gravity location. For a low-wing mono-
plane, the center of gravity is from 2 to 6 inches below the thrust line; for
a high-wing monoplane, it is about 2 to 4 inches above the thrust line for
airplanes up to about 20,000 pounds gross weight. In lieu of more accu-

rate information obtained from a balance diagram, such a location is suffi-
ciently accurate to assume for the purposes to be served by the preliminary
three-view.
122
ENGINEERING STUDIES
While the thumb-nail, three-view sketch will convey the type of air-
plane the designer desires to build, it is necessary in more complex designs
to make a series of subordinate studies first.
In a retractable landing-gear design, a careful study of the wing struc-
ture, recessing well, strut attachments, gasoline-tank location, control
systems, and the like should be carefully laid out because the wing design
may be considerably changed just by slight variations in the conception
and location of items incorporated within the wing, or attached to it. Not
only the wing but also the preliminary kinematic layout of the retractable
landing gear must be studied carefully. When copies are made of exist-
ing designs, the problem is not a difficult one, but a new conception re-
quires much attention to the details of operation and how they affect the
whole.
When production consideration is paramount, the method of attach-
ment of the wings to the fuselage may be very important, and the internal
structure of the fuselage is certainly dependent upon the method in which
loads are being transferred from one part of the structure to the other.
The fuselage structure affects window spacing, for example, and window
spacing affects the final appearance of the airplane.
In the case of large passenger and cargo arrangements, a preliminary
internal view is desirable even before the side view of the airplane is
sketched up.
Different types of lift-increase devices will affect the position of the wing
with relation to the fuselage. If the wing is placed at the bottom of the
fuselage, deflection of the flap may be limited by ground clearances; aero-
dynamic twist of the wing, or the incorporation of varying dihedral span-
wise, may cause difficulties in the mechanical operation of the lift-increase
devices.
These are only a few representative cases of engineering studies that
are necessary and desirable to assure a well coordinated design. These
studies should be made before detailed structural designs are considered.
Cursory readings of subsequent chapters on wing design, fuselage, land-
ing gear, power plant, and other pertinent chapters may also be helpful
in crystallizing ideas.
FINAL THREE-VIEW
The final three-view is based upon more accurate information than the
preliminary three-view since the weight estimate has been more accurately
determined and the final balance diagram has been completed. Both the
balance diagram and the final three-view depend in some degree on the
structural layout of the wing, landing gear, and fuselage, but once these
THE THREE-VIEW
123
difficulties in structural arrangements and the like have been ironed out,
it is possible to go ahead with the final three-view.
It may be found that the balance diagram has made changes in the fol-
lowing:
1. Position of the wing.
2. Location of the landing gear.
3. Location of the tail surfaces.
4. Location of the center of gravity.
The structural layouts may have caused changes in:
1. Planform of the wing, perhaps because the spar intersected the fuse-
lage where a bulkhead was found to be undesirable.
2. Arrangement of landing-gear members due to changes in spar loca-
tions.
3. Placement of tail surfaces with respect to planes of symmetry.
4. Vertical position of engine due to installation and vision require-
ments.
Only after all points have been considered should the final three-view
be made. Usually, it is desirable to wait until the control surfaces and
landing gear, as well as cabin installations, have been made in order to
incorporate the latest correction.

CHAPTER VIII
Preliminary Weight Estimate
Unless the beginner is copying an existing design, the first data that he
will need to begin his design after he has been given or has written his own
specifications are various estimates of the approximate gross weight of the
airplane. For on this gross weight will depend the wing area required to
meet the specified landing speed.
GENERAL PROCEDURE
Once the wing area has been determined, the length of the fuselage and
the areas of the horizontal and vertical tail surfaces can be obtained, since
modern airplanes have definite relationships of areas and dimensions of
fuselages, tail surfaces, and even the landing-gear tread, in terms of the
span and area of the wing. Subsequent aerodynamic studies and wind-
tunnel tests will have to be made to determine the governing proportions
exactly.
When the length of the fuselage is known, its weight can be estimated;
likewise, if the areas of the tail surfaces are known, their weight may be
estimated; and so on until all data are accumulated.
When an approximate figure for the gross weight has been decided upon,
the size and weights of the landing gear may be found, and so more and
more data accumulate until the structural weight of the complete airplane
has been determined. It may be only approximate, but a series of refine-
ments will finally give a weight estimate. When each part has been built
and weighed, this so-called estimate will be found to be surprisingly accu-
rate.
The methods by which the weight is estimated are empirical, on the
whole, and every bit of information that can be gleaned from the specifi-
cations and from existing similar designs will be helpful.
An experienced designer needs only to look at that portion of the speci-
fications which sets forth the payload and performance requirements in
124
PRELIMINARY WEIGHT ESTIMATE 125
order to be able to estimate within a very few pounds what the gross weight
of the airplane is likely to be, as well as what type of wing design is likely
to be most suitable. He is able to do this because he has studied similar
airplanes. Even then, his preliminary estimate must be broken down
into as many components as possible, because an intelligent designer can-
not afford to make any estimates without reasonable research to back up
those data.
It is necessary to make weight estimates as accurately as possible, for
if the airplane should weigh more than estimated, the performance might
be seriously affected by the greater weight. If this should happen, and
the weight is greater than estimated, it might be necessary to cut down on
the amount of payload, which certainly is not economical; or to cut down
on the fuel load which, in turn, will cut down the range.
The question may well be asked: "Why cut down on the payload or fuel
load if the airplane is overweight; isn't the airplane better if it carries more
load than that for which it was originally designed?" No, that is not the
case. In any given design, such items as the payload, crew, fuel and oil
are fixed in weight; that is, they have been set by the original specification
so that any increase in weight is undoubtedly due to the structure, which
has been either underestimated in weight or improperly designed. In both
cases the result is that the airplane will cost more to construct, and the
increased cost is undesirable. Moreover, the size of the wing is deter-
mined by the gross weight and is designed to be strong enough to carry
that load. Any increase in weight will mean also, since the wing area
cannot be changed, that the airplane will have to land faster. The faster
the landing speed, the harder it may be for the pilot to handle the airplane.
Too high a gross weight for the given wing area, airfoil section, and power
will also mean that the airplane will have to travel faster and longer along
the ground before it can take off. Unless the runways of the airport are
long enough, the airplane may not be able to take off at all. Also, the
heavier the airplane is for the given wing area and the horsepower of the
engines, the more difficult it will be to reach a high enough altitude to
clear high mountains.
Thus, when an airplane is overweight, in order to land slowly enough
and take off quickly enough, overloading must be reduced in some way.
The only way left to do this after the airplane has been built is to cut down
the amount of payload or the amount of fuel.
There is another serious factor regarding the overweight of an airplane.

When the structure is being designed, each part is designed to withstand
a certain load imposed on it under various flying and landing conditions,
but if the airplane is overweight some of these parts may fail.
Suppose an airplane has been estimated to weigh 10,000 pounds. The
landing gear (taking just one part of the airplane for illustration) is de-
126
signed to withstand this load without failure when landing. Suppose,
however, when the airplane is finally built, that it is found to weigh 11,000
pounds. Not all of this addition in weight may be due to improper struc-
tural design, perhaps, for the designer may have also decided to double
the mail or express load after the design was started. The landing gear,
originally designed to stand up under 10,000 pounds on landing, must now
stand up under 11,000 pounds. Will it do it? It may not.
Suppose the landing gear is designed originally to stand up under 11,000
pounds instead of 10,000 pounds, so that if the airplane does weigh more
it will be safe. Unfortunately, this is no remedy because it means design-
ing a structure which, even if not overweight, will be unnecessarily over-
strong. Overstrength of parts means fundamentally poor and very costly
design.
The estimates may be exceeded by a small percentage. A 5 per cent
increase in one item may be more than offset by a decrease in weight of
another item. In general, it is not desirable to exceed the estimate gross
weight by more than 2 or 3 per cent. When an entirely new type of air-
plane is being designed, the final gross weight may be very much larger
than the estimate, primarily because insufficient data were available to
make it possible to estimate more closely. An airplane designed to carry
250,000 pounds, for example, may weigh well in excess of the allowable
3 per cent, but suppose ten or twelve of these large airplanes were built
in a series and flown extensively for a few years so that all its character-
istics were known. Then the next designer to build an airplane of similar
size would have some data available and his design would meet the origi-
nal specification far more closely than the first prototype did.
The weight estimate goes through a series of trials until the added re-
finements indicate that it is sufficiently accurate for all subsequent cal-
culations. The weight estimate does not become really "final" until the
airplane has been built and each part weighed.
ESTIMATED WEIGHTS
To arrive at any basis for subsequent calculation, design, and develop-
ment, weights are determined by allotting certain weights to wing groups
per unit of wing area, to gas tanks per gallon capacity, to landing gear as
percentage of gross weight, etc. In such cases the weights are said to be
estimated and their accuracy depends upon the knowledge gained from
previous designs.
CALCULATED WEIGHTS
The weights of parts rray be found by calculation of volumes, length,
and areas of parts after the design has been sufficiently developed so that
definite sizes can be determined. Such calculated weights are used for
PRELIMINARY WEIGHT ESTIMATE
127
subsequent weight estimation and for setting up the final weight table and
balance diagram for the actual airplane.
ACTUAL WEIGHTS
After the parts or assemblies under consideration have been constructed,
they are weighed, as a matter of record, and also to check on estimated
weights previously arrived at. Very often this procedure helps in discov-
ering errors in methods of construction, or in methods of computation.
In either case, sometimes it becomes necessary, especially if the weights
are far out of line, to redesign the part or assembly in order to get closer
to the estimated or allotted weight.
ALTERNATE LOADS
Some designs may incorporate different engines having different weights,
or different fuel and oil loads with accompanying varying cargo. In such
a case alternate data should be included for calculating the weights.
ESTIMATING THE GROSS WEIGHT
The gross weight, before a more detailed weight breakdown has been
made, can first be estimated by means of rule-of-thumb methods.
One method is to determine the probable gross weight on the basis of the
power loading since the specifications for the design usually state a rated
engine horsepower. Since
, ,. gross weight W.
power loading = â€¢ . , â€”;â€”qâ€”2 = 77 ,
total engine horsepower P
then assuming a value for the power loading, and knowing the total rated
engine horsepower, a reasonable figure for the gross weight may be ob-
tained.
Another method is to estimate as closely as possible the detailed weights
for the
1. Power plant,
2. Crew and payload,
3. Combustible load,
4. Furnishings and equipment.
These four items constitute from 65 to 70 per cent of the gross weight,
or in equation form
. sum of items 1, 2, 3, and 4

gross weight = 0 7
Therefore, a second gross weight can be obtained, and then the difference
between the gross weight thus calculated and these four items will give the
amount allotted for the structural weight.

128
ESTIMATION OF STRUCTURAL WEIGHT
It is necessary to figure a likely value for the gross weight of the air-
plane in order to determine
1. The wing area and consequently the wing weight,
2. The size and weight of the landing gear,
3. The size and weight of the tail surfaces (once the wing area is tenta-
tively determined, the area of tail surfaces are likewise tentatively deter-
mined),
4. Fuselage length.
Manufacturers' ratings for wheels, tires, shock absorbers, and the like
are in terms of the static weight on these parts. Therefore, if the gross
weight of the airplane has been calculated, the static weight per wheel is
then one-half the gross weight of the airplane. The static weight on the
tail wheel varies from one-tenth to one-twelfth the gross weight of the
airplane. For these static weights, the sizes of the wheels and tires can
be found by consulting the catalogues of the manufacturers.
Once the wing area has been calculated tentatively, a tentative span
length can be set, and since the fuselage length varies from 60 to 70 per
cent of the span, the fuselage dimensions can be worked out readily. Just
as soon as the fuselage dimensions are at least tentatively established, the
weight can be estimated.
RECAPITULATIONS
After the structural weights have been listed, the weights are totaled to
give another gross weight which may vary considerably from the previous
values determined by rule-of-thumb methods. This gross weight will
change the wing and tail surface area and therefore the weights of these
items. The weight of the landing gear usually remains the same since the
difference in gross weight is usually not so large as to affect the ratings of
the landing gear components.
These corrections in the structural weight are continued with every new
total for the gross weight caused by these corrections until the changes are
sufficiently small to make further corrections unnecessary.
FORM FOR PRELIMINARY WEIGHT ESTIMATE
For convenience, the weights of the airplane are grouped as indicated in
the worksheet which follows. This form is shown in its preliminary stage.
It may be considerably expanded, and it is usually desirable to do so, espe-
cially in presenting the final weight estimate.

In setting up a weight table, group classifications are used because group
components are more easily weighed, and often a very detailed weight esti-
mate is not possible.

Table VTII-1. Weight control.
Power Plant
Engines, dry weight .?..13.
Per cent of power plant weight
Per cent of weight empty
Per cent of gross weight
Propellers and hubs
Starters.
Generators
Engine instruments
Engine controls
Fuel pumps
Scoops
Switches
Thermocouples
Deflectors .
Cowlings
Fire walls
Engine mounts
Exhaust manifolds
Fuel tanks
Fuel system
Oil tanks
Oil system
Miscellaneous
Total power plant...
Per cent of gross weight ..
Combustible load
Gasoline ........
Oil ill:.
Total combustible load
Per cent of disposable load
Crew
Pilot and co-pilot

Navigator
Radioman
Mechanic
Stewardess
Total crew
Per cent of gross weight ....
Payload
Passengers and baggage
Mail
Express
Total payload
Per cent of gross weight ....

Table VIII-1. (Continued).
â€¢ :.-.â€¢â€¢00â€¢.} <. I ~
Furnishings and equipment
Flooring .
Surface controls and wires . .
Seats and cushions
Safety belts LL*.â„¢..
Batteries
Instruments
Lighting .
Radio equipment
Upholstery
Flares
Soundproofing
Heating and ventilating.
First aid
Refreshments.
Miscellaneous
Total furnishings and equipment...
Structure
Wing group
Wing panels
Ailerons
Flaps.
Aileron controls â€”
Flap controls
Spoilers
Spoiler controls
Fillets
Struts
Miscellaneous
Total wing group
Pounds per square foot of wing area
Per cent of structure weight


Empennage group .
Elevators
Trimmers or tabs for elevators
Stabilizers
Rudders
Trimmers or tabs for rudders
Fins â€”
Struts
Control systems
Static balances
Total tail group

Body group
Fuselage, including covering
Doors
Windows
Fillets
Total body group
Per cent of weight empty ....
Landing gear group
Wheels
Tires
Brakes
Shock absorber struts
Additional struts
Retracting mechanism
Cowlings .
Tail wheel
Tail tire
Tail shock absorber strut ....
Retracting mechanisms
Total landing gear group .
Total
The following abbreviations and definitions are useful:
WE = weight empty = structure plus power plant plus furnishings.
UL = useful load = crew plus combustible load plus payload.
CL = combustible load = gasoline plus oil.
GW = gross weight = weight empty plus useful load.
DL = disposable load = crew plus easily removable equipment and
furnishings plus combustible l(jad.
Fittings, attaching bolts, built-in supports, and the like should be
charged to the group to which they are permanently fastened. Bolts

sometimes present a problem, but in most cases their weights are negli-
gible with respect to the whole, so that little difference will be caused if
they are included in one or another group.
In the case of a retractable landing gear the matter becomes somewhat
more involved since, undoubtedly, additional structure is necessary in the
wing (if the landing gear is to be retracted in it). If so, an attempt should
be made to obtain a reasonable distribution between the landing gear and
the wing.
When stubs of wings or tail surfaces are built integrally with the fuse-
lage the weights of such items should be included with the wings or tail
surfaces since it is obvious that the purposes which these parts serve should
132
be the criteria. In weighing up assemblies, however, it may be difficult to
ascertain the weights of such parts. One of the reasons for trying to keep
the various groups segregated is that subsequent comparisons with similar
planes, but with slightly different structural details, become easier.
Where difficulty is experienced in actual practice, the weight may be in-
cluded in the group to which it is attached but a special note should be
made of such a case for future reference.
VARIABLES AFFECTING WEIGHT ESTIMATION
As design practices change, and new materials and methods of construc-
tion are introduced, the weights of component parts of the airplane may
change considerably. It is, therefore, particularly important to re-evalu-
ate any formulas, and to re-examine any specialized methods in arriving
at certain weight estimates, in order to be sure that they still apply.
Wing Weight Estimates
To illustrate how the weight estimate of a wing might change, let us
consider a cantilever wing of a given area. It would be found that the
weight of this wing would be a function of the following parameters:
1. Wing loading W/S: The higher the wing loading, the greater the
shear and bending moments, and therefore the more material required to
meet the requirements for the allowable stress.
2. Aspect ratio R: The greater the aspect ratio for a given wing area,
the greater the span, and therefore the greater the bending moments;
again more material is required.
3. Another factor, which might be used in place of or in conjunction
with the aspect ratio factor, is the span loading W/b (the gross weight di-
vided by the span) since obviously the greater the span loading, the greater
the bending moment, and therefore the more material required.
4. Airfoil section used, or more specifically, thickness ratio t/C of the
airfoil used: Since the bending stress ft, = My/I, the maximum stress is
directly proportional to the section modulus I/y. For a symmetrical de-
sign (assumed for ease in illustrating the point to be made) y would be
one-half the depth of the wing section; I varies as the third power of the
depth so that the stress would vary inversely, all other factors being equal,
as the square of the depth of the wing; therefore, the smaller the depth,
the greater the stress so that more material would be required in order that
the allowable stress would not be exceeded.
5. Material used, or more properly, allowable ultimate stress permitted

for the material used: If one material has three-fourths the allowable ulti-
mate stress, Fu, of another, then four-thirds of the one material, a 33J^ per
cent increase, would be needed for the one over the other.
6. Limit load factor n for which the wing was designed: It should be
133
obvious that the greater the load factor, the greater the loads imposed,
and, therefore, the more material required.
7. Mass distribution of the wing component, since flight loads would be
counteracted by the inertia loads: If the weight of the gas tanks, engine
nacelles, bomb loads, and other items of equipment are distributed over
the span, the resultant loads in flight could be reduced materially. How-
ever, it might mean that landing conditions could become critical in the
design of the wing. For future reference, the mass distribution of the
wing may be considered proportional to wi/b where wi is the weight of
the wing assembly and b its span.
8. Taper ratio, Ctip/Croot = Ct/Cr, since the greater the taper ratio, for
the same aspect ratio and area of the wing, the more load would be con-
centrated near the root of the wing, and therefore the smaller the bending
moment.
9. Amount 'of sweepback, A: High-speed airplanes, flying at speeds ap-
proaching that of sound, employ large angles of sweepback for the wing
with the result that large torsional moments are introduced. Any struc-
ture subjected to torsional moments in addition to bending moments will
be heavier than one not subjected to torsional moments.
From the factors just considered, it should be possible to obtain an ex-
pression for the wing weight similar to the one that follows:
Such a formula might be considered as a purely theoretical formula, and
it would be necessary to obtain the exponents empirically by weighing as
many wings as possible where the various parameters were sufficiently dif-
ferent to permit an evaluation. In practice, however, such work would
not be done because of various difficulties that would be encountered.
The line of investigation suggested is useful nevertheless since it should
impress the student with the fact that there are many variables having an
effect upon the estimates to be made.
In addition to the various parameters mentioned, other factors enter
into the use of any formulas which would have to be restricted to certain
types of wings. The following factors also have some bearing on the wing
weight.
1. Type of construction: Just examining the historical development of
wing construction throws some light on this factor. Fabric-covered, two-

spar (perhaps of wood) wings were commonly employed. These were fol-
or possibly,
134
lowed by plywood-covered, two-spar wings where more of the load was
carried by the covering. Then the metal-covered (with the covering or
skin supported at intervals by stiffening members) wing with two metal
spars followed. As the wing and span loadings increased, more metal
spars were introduced. Now, for exceptionally high-speed airplanes, sand-
wich construction (wherein two thin metal faces with a soft core of organic
or synthetic material) and thick metal plates for skin with longitudinal
ribs and no longitudinal spars are being employed. The changes have
brought about the transfer of material from spars to the covering, and
should, therefore, affect the weight. Likewise, the change of covering
from fabric to plywood, to thin sheet metal, to thick plate, and to sand-
wich construction affect the wing weight since certain portions of the wing,
like the trailing edge, bear relatively light loads.
2. Use of lift-increase devices, whether they be nose or trailing-edge
flaps, boundary layer control, or other devices: These devices introduce
variations in the local pressure distribution; they add to the structural
weight by virtue of their control systems; they increase the torsional and
bending moments, particularly in the chordwise direction; they increase
the weight by virtue of rigidity requirements, quite apart from pure
strength considerations. Although a flat percentage increase may be
used when such lift-increase devices are used, yet the weight contributed
by such devices depends upon, among others, the following factors: their
size, conditions under which they operate, variation caused in local pres-
sure distribution, control forces and moments encountered, and rigidity
and strength requirements.
3. Use of anti-icing systems: While the proportion of weight caused by
wing and wing-control surface anti-icing and de-icing systems very often
can be allocated directly to these systems, there are certain items which
might be accounted for either under wing weight or anti-icing system
weights.
4. Introduction of ducts, cutouts for retractable landing gears, etc.,
will affect the wing weight: Sometimes these can be accounted for sepa-
rately and allocated against the items causing the additional weight.
5. Type and size of lateral control surfaces: Normally, when the con-
ventional aileron is used, its area is considered part of the wing area. If
the aileron has to be modified because of lift-increase devices, the change is
usually reflected in a change in weight. The change may be small. For

high-speed airplanes which employ a high degree of sweepback, the lateral-
control devices present a problem. Their design may have a different
effect on wing design than the conventional case.
6. Incorporation of jettisonable portions for the wing: Highly maneu-
verable military airplanes have used the device of making the outer wing-
tip portion quickly detachable (through failure at a critical section) when

the airplane encounters loads greater than that for which the major portion
of the airplane has been designed. While this stratagem is useful for
military airplanes, it might be uneconomical for commercial airplanes.
However, it is a factor to consider at the appropriate time.
7. Method of supporting .the wing; that is, whether internally or ex-
ternally braced; if externally braced, whether fixed or pinned at the fuse-
lage; if externally braced, the amount of wing-tip overhang: These factors
affect the moment distribution axial loads, and material distribution.
8. Purpose for which the airplane was designed: Climatic conditions
might affect detail design of component parts. Certainly, such a factor
would be difficult to incorporate in any formula.
9. Requirements for inspection and maintenance: If many inspection
points are necessary for inspection and maintenance purposes, the struc-
ture will have to be modified accordingly. Likewise, if requirements for
maintenance are such that it should be reduced to a minimum, especially
when the aircraft is to be located in certain areas of the globe, it may be
advisable "to beef up" certain parts of the structure.
10. Requirements for painting, anticorrosion protection, and the like:
The effect of these should be obvious. It may be accounted for in the
wing weight, or arrived at independently.
11. Relative distribution of structural material subject to normal
stresses with respect to the structural material used for transfer of shearing
forces; also the relative distribution of the primary structure, or load-
carrying material, as against the secondary structure, or nonload carrying
material provided to maintain the shape, or to cut off one section from
another, and the like.
12. Actual size of the aircraft: Relatively small airplanes (for example,
the lowest-powered airplanes) have a great part of their structure deter-
mined by practical considerations such as: availability of material, the
minimum size or dimensions of materials desirable for maintenance rea-
sons, the minimum sizes available for commercial reasons.
When the aircraft reaches a size outside the immediate knowledge of the
designing fraternity, then the question of weight estimation becomes even
more difficult since certain parts of the structure may not increase in weight
at the same rate as another part. For this reason any final formula
would be of the form:

Wi = w' + w" + w'" + w"" + â€¢â€¢â€¢
where w', w", w'", ..., etc., represent weights of different portions of the
wing whose estimation would be arrived at by different routes.
By analyzing the various factors involved, it would be possible, there-
fore, to arrive at an expression for determining the weight which could
cover all cases. It is more practical to obtain simple empirical formulas

136
for weight estimation. If those are lacking, or if they are suspect, then
the problem is reduced to fundamentals. This means estimating weights
by as many methods as are available, scrutinizing each assumption made,
and then building the airplane upon such assumptions, subject to any
corrections that later stress analysis may reveal and basing subsequent
weight analyses on the information gained in building the original pro-
totype.
Fuselage Weight Estimates
Many of the factors discussed in the analysis of wing-weight estimates
apply to any consideration of fuselage weight estimates. It may be worth
while to examine a few of the factors pertinent to the fuselage. The fol-
lowing factors would affect the fuselage weight:
1. Diameter D, or ratio of depth to width d/b, of the fuselage: Usually
this is taken at the median section and is proportional to the volume of
the fuselage. The greater this ratio, or the larger the diameter, the more
material involved.
2. Length L of the fuselage, since a long fuselage would indicate either
larger bending moments, or greater weight because of more material: Usu-
ally the parameter used is the ratio of the fuselage length to wing span, or
L/b.
3. Tail length, or distance between the center of gravity of the airplane
and the center of pressure of the tail surfaces: The parameter of interest
here would be the ratio of the tail length Lt to the mean aerodynamic
chord C, or Lt/C. If the value of this ratio is small, it is an indication
that the loads on the horizontal tail surfaces are large so that the trans-
verse shear may be greater than when the tail length is relatively large.
4. Center of gravity movement, since a relatively large movement usu-
ally means that the aircraft has to be designed for a number of flight
conditions and usually calls for increased weight: The center of gravity
movement is usually expressed in terms of the mean aerodynamic chord,
that is, xC.
5. Design load factors for both flight and landing conditions: No one
load factor or condition is usually critical for the entire fuselage so that a
problem is presented here. However, there is no doubt that the fuselage
weight is a function of some load factor, n.

6. Mass distribution, particularly of local concentrated weights caused
by cargo: This mass distribution could be represented as some function of

the average mass distribution and may be considered proportional to the
ratio oiv)//L where w/ is the fuselage weight and L the fuselage length.
137
The resulting formula, considering the factors mentioned, would be of
the order
However, there are other factors which would make necessary the
evaluation of such an expression so that it would apply to a certain cate-
gory, and thus a number of formulas would be necessary. For example,
the following categories are indicative:
1. Type of fuselage: whether tubular steel, reinforced monocoque,
sandwich construction; whether cabin or open cockpit; whether cargo, or
passenger; all would vary the weight.
2. Type of wing attachment: whether for an internally or an externally
braced wing, and if the latter, whether the wing is fixed or pinned at the
fuselage.
3. Type of landing gear and its configuration.
4. Extent of pressurization, if any.
5. Any special requirements as doors, exits, windows, windshields.
As for the wing, the weight of the fuselage can be broken down into
component parts, but here again, some estimate must first be made of the
weight, to proceed with the more detailed design. It is only after the more
detailed design has progressed to a certain point that it is possible to get the
required information as to dimensions, etc., so that the weight estimate
can be revised.
Control Surface Weight Estimates
Since the aileron is customarily regarded as part of the wing structure,
this discussion will apply primarily to the vertical and horizontal tail sur-
faces although some of the factors concerning them apply with equal force
in the consideration of the ailerons. Likewise, some of the factors affect-
ing estimation of the wing weight apply to control surfaces as well.
The weight of these tail surfaces would be affected by:
1. Aspect ratio R.
2. Proportion of movable surface area Se to the fixed Sf, or to the com-
plete surface area St, that is, Sc/S/, or Sc/St.
3. The ratio of the aerodynamic balance area <SÂ», ahead of the hinge
line to the movable surface, or Sh/Sc.
4. Proportion of the trimming tab surface Sta to movable surface Sc,
or SJSC.
5. Mass distribution, especially where static and dynamic balancing is
required: This distribution could be considered proportional to the average

unit weight of the tail surfaces or Wc/Sc where Wc represents the weight
of the movable portion of the tail surfaces and Se is its area.

138
6. Permissible hinge moment in terms of the stick or control wheel force:
This factor could be represented by Cc*.
7. Tail length Lt of the airplane, since the farther back, in terms of the
mean aerodynamic chord C, the tail surfaces are from the center of gravity
of the airplane, the lower the load on the tail surfaces.
The resulting formula would be similar to the following:
Such a formula could be broken down separately for the fixed and mov-
able portions of the control surfaces.
Landing-Gear Weight Estimates
The landing gear, perhaps more than any other structural assembly
with the exception of the power plant group, consists of a number of units
whose weight is not determined by the airplane designer so that the struc-
tural weight of those items depends primarily upon availability, although
they are designed to meet the conditions for which they are used.
Wheels, brakes, tires, and to some degree, the shock absorber, are de-
signed to satisfy the requirements of airplanes with a relative wide range
of gross weights. In the main, they are designed for a given gross weight
W and a given load factor n. However, their selection is simplified by
the manufacturers by rating them according to the static weight they are
supposed to carry.
The weights of rest of the structure, consisting of struts, including the
shock absorber member, and the retracting mechanism, depend upon a
number of factors such as:
1. Gross weight of the airplane;
2. Loads applied to the structure which depend upon the type of land-
ing gear, such as whether it is a conventional landing gear, or a tricycle
type;
3. Configuration of the members; that is, whether several members are
used to form a tripod, for example, or a pure cantilever single-member
type, etc.;
4. Length of the members;
5. Materials used.
Power Plant and Power-Plant Nacelle Weight Estimates
The estimation of the power-plant weight is relatively simple since once
the engine is selected, the remainder is comparatively easy. The weight

of the power-plant nacelle, consisting of the engine mount, cowling, ducts,
etc., is a function of:

1. Weight of the power plant,

2. Maximum load factor to be experienced by the airplane,
3. Position with relation to the fuselage or the leading edge of the wing,
4. Other factors such as landing-gear attachments, inspection provision,
disconnect features.
Weight Data Sources
To obtain a realistic weight estimate, it is necessary to know the various
sources from which such data may be obtained.
1. Accessories, power plant, hardware, wheels, etc. are usually of stand-
ard manufacture, and their weights are obtained from manufacturers'
catalogs.
2. AVeights of fabricated assemblies must be obtained from:
(a) published materials,
(b) data available from past experience, and
(c) actual weighings of dismantled or junked airplanes.
'A. Preliminary studies of structural design based upon empirical and/or
theoretical approaches.
Design Control of Weight
The aircraft designer has little control over the weight of such items
as accessories, power plant, and hardware, except insofar as he states
the original specifications for performance, material, or size.
The fabricated portion of the airplane may not amount to more than
15 to 35% of the gross weight. Even if the designer were able to devise
structures having weight differences of 40%, the final result would affect
the gross weight not more than about 10%. Nevertheless, he should
make every attempt to obtain the lightest structure possible, and his
selection of the component parts must always be considered in the light
of the possible saving of weight.
EMPIRICAL FORMULAS AND DATA
Some empirical formulas based upon various parameters are presented
here. The student may choose the ones he deems appropriate or use
several for comparison purposes.
Wing Weight
1. One formula for wing-weight estimation which may be used for air-
craft weighing up to about 50,000 pounds is stated as follows:
Ww = ^ (0.18756 + 53.50),
where Ww = weight of the wing, in pounds,

n = design load factor (see Chapter IV for a likely applicable
value),
140
b = span of the wing, in feet,
Wâ€ž = gross weight of the airplane, in pounds.
The limitations of the formula should be immediately apparent. The
gross weight Wâ€ž is itself an estimate at this stage; the load factor n is a
function of the aircraft weight and its performance (both still in their
formative stages); the span 6 of the wing is a function of the gross weight
(still an estimate) and of the airfoil characteristics, assuming that the
aspect ratio is known. The indications, therefore, are that any complex
formula derived will give values that are still unproved.
2. For commercial aircraft gross weights between 1,500 and 5,000
pounds:
3. For military airplanes (propeller-engine type) of gross weights be-
tween 10,000 and 100,000 pounds:
4. For commercial airplanes of gross weights between 20,000 to 70,000
pounds:
5. For gross weights between 70,000 and 150,000 pounds for commercial
airplanes:
6. For military airplanes of gross weights between 100,000 and 300,000
pounds:
7. Wing = 15% gross weight.
8. Weight may be obtained for metal wings by plotting (per inch of
span) the weight calculated from the formula
where n = highest design flight-load factor, and
C = chord in inches,
assuming that the wing is constructed of aluminum alloy.
9. Metal wings internally braced, no flaps: 1.6 lb to 2.6 lb per ft2.
Ww = .0763 Wâ€ž + 50.
W = 0.00002 (n) (C2),

141
10. Metal wings internally braced, with flaps and retracting mechanism:
2.6 to 4.6 lb per sq ft, with the higher value for wings incorporating high-
lift devices such as the Fowler flap.
11. Fabric-covered wings: 1.4 to 2.5 lb per ft2.
12. For many transonic and supersonic designs, elevators, as well as
the usual flaps and ailerons (either or both of which may have dual func-
tions), are incorporated into the trailing edge of the wing. The weights
of these wings tend to average on the high side per square foot compared
with the internally-braced wing of greater aspect ratio.
13. Ww = 0.00591 -
where Wâ€ž = gross weight, in pounds,
n = limit load factor, maximum positive,
b = wing span, in feet,
S = wing area, in square feet,
t = maximum thickness of root section, in feet.
Tail Surfaces
Empirical formulas applying to tail-surface weight estimation are:
1. (a) For rudders:
(b) For elevators:
(c) For fins:
WR = J^(S, + 340);
WB = ^(S* + 340);
Wf = Jb(s" + 490);
(d) For stabilizers:
Ws = OS*+ 475);
where WR, We, Wf, Ws represent the weights of the rudder, elevator,
fin, and stabilizer, respectively, and ST, Sâ€ž Sf, Sâ€ž Sâ€ž Sh represent the areas
of the rudder, elevator, fin, stabilizer, vertical tail surfaces, and horizontal
tail surfaces, respectively.

2. For commercial aircraft of 1,500 to 5,000 pounds gross weight:
WTa = .0163 Wg + 10,
for vertical and horizontal tail surfaces combined.
3. For commercial aircraft from 20,000 to 60,000 pounds gross weight:
w" - [M35 - S]w-
4. For commercial aircraft from 60,000 to 150,000 pounds gross weight:
w- - H - tsSst] w-
5. For military aircraft between 10,000 and 35,000 pounds gross weight:
wâ€ž-[am
w..
L 100,000.
WTS = |.0280 - 008 W>
W..
100.000J
8. For military aircraft between 90,000 and 300,000 pounds gross
weight:
wÂ» = [Â°246 - mk]w-
9. Vary from 1.0 to 2 lb per sq ft
(a) Near the lower figures for externally-braced tail surfaces.
(b) Figure allows for bracing.
(c) Near the higher figure for full-cantilever tail surfaces.
(d) Static balances increase weights.
Fuselage Weight
1. A reasonable empirical formula with which the fuselage weight can
be estimated in terms of the gross weight is given here for airplanes weigh-
ing up to about 100,000 pounds:
F 12
where WF is the fuselage weight, and
Wg is the gross weight of the airplane.

Such a formula is perhaps too simple, since it would indicate that the
fuselage weight for all airplanes is a little more than eight per cent of the
gross weight of the airplane. However, the value obtained is useful for
a first approximation.
2. Welded tubular type: 0.70 to 1.80 lb per in. of length. The length
of the fuselage is considered to be the distance from the fire wall to the
tail post.
3. Reinforced monocoque: 0.60 to 1.70 lb per in. of length.
4. Reinforced monocoque weight is obtained by the formula:
0.791 X (length) X (circumference of median section),
dimensions in feet.
5. For an airplane weighing about 8,500 pounds, a typical bare fuselage
weight (for a reinforced monocoque construction) is 650 pounds, including
firewall, stringers, covering, fittings, built-in supports, doors, and door
frames.
6. For commercial aircraft between 1,500 and 5,000 pounds gross weight:
WF = .205 Wg - 140.
7. For commercial aircraft of gross weights between 20,000 and 150,000
pounds:
WF = .073 Wâ€ž.
8. For military aircraft between 10,000 and 55,000 pounds:
w Tims 0309 Wj\ w
Wf = L1013 â€” lopoo"J W-
.004 w;
WF = .0853 -
100,000J
4Â«
w' -10833 - mm]w-
Landing Gear
1. An empirical formula which may be used is:
WLG = ^ (35 + 0.1 L),
where Wlo represents the landing-gear weight,
Wg represents the gross weight of the airplane, and

L represents the length, in feet, of the longest lartding-gear member.

Where there are several landing-gear assemblies (for example, a nose wheel
and the main landing gear), an approximation of L may be obtained by
averaging the lengths of the several assemblies involved.
2. The landing gear (including retraction mechanism) may be assumed
to be 5 to 6H% of the gross weight.
3. Wheels and tires are rated according to static weights imposed.
The static weight per wheel and tire is one half of the gross weight. The
static weight per tail wheel is from one tenth to one twelfth of the gross
weight. For tricycle landing gear, assume one sixth to one eighth on the
front wheel, the remainder on the rear wheel. With the static load
rating known, the required wheel size and corresponding weight may be
determined.
4. For commercial aircraft between 1,500 and 5,000 pounds gross
weight:
Wlg = .0521 Wâ€ž + 5.
wÂ» = [000817 - Siw-
wÂ» = [000567 â€” mM w-
T Â«,s., 000028 Wâ€ž~] Tâ€ž
9. For commercial aircraft between 20,000 and 60,000 pounds:
â€ž. T .0383 Wâ€žl â€ž
wÂ» = L 077 â€” lopoo J w"
10. For commercial aircraft between 60,000 and 150,000 pounds:
wÂ» = [06Â°â€” mM w*
11. Wheel, tire, and tube weight in pounds:
W = 10 + 00010 Wg.
12. Brake weight in pounds:
W = 10 + (7.2 X 10-5) (W>) (j^y
13. Shock absorber and axle assembly in pounds:
W = 5 + (2 X 10-4) (Wâ€ž) (L),
where L is length of shock absorber strut, in inches.
Total Fabricated Components
A check on the calculation of the weights for the wing, tail surfaces,
fuselage, and landing gear may be made with the aid of the following
formulas:
pounds:
WST = .368 W, - 100.
pounds:
.12 Wâ€ž
WST =
.36
100,000j
3. For commercial aircraft weighing between 50,000 and 200,000 pounds:
WST = .27 Wg.
4. For military aircraft weighing from 10,000 to 50,000 pounds:
wÂ» = [AU ~ mk]w-
5. For military aircraft weighing from 50,000 to 110,000 pounds:
6. For military aircraft weighing between 110,000 and 300,000 pounds:
w* = [0032 - mki\w-
7. For jet aircraft, the structural weight is about the same, being only
about Yi percent less. The above estimates may therefore be used for
preliminary calculations for jet-powered aircraft.
8. For turboprop airplanes equipped with Fowler flaps, wing aspect
ratios of 14 or less, wing loadings between 20 to 100 pounds per square
foot, the weight empty, and the structural (or airframe) weight may be
estimated as follows:
146
(a)
.054 - 0.003 4r W,
9i
(b)
WST
178 w {W
Weight Empty
Some idea of the reasonability of weight estimates for the structure,
power plant, and fixed equipment may be gained by comparing the total
with the values obtained for the weight empty from the appropriate
empirical formulas below, which are based on commercial airplane data.
1. For gross weights between 1,500 and 5,000 pounds:
Engine Nacelles
1. For power plant installations of the radial-type aircraft engine, a
reasonable estimate of the nacelle weight may be obtained from:
where Wâ€ža represents the nacelle weight, and
Wen represents the engine weight,
For other types of propulsion devices, such as jet-propulsion engines,
gas turbines, and similar power plants, the data on weights of the power-
plant housing in the airplane are still too meager to permit the derivation
of any reliable relationships.
2. The following formulas may be found useful:
(a) For aircraft weighing between 20,000 and 60,000 pounds, having
2 or 4 engine nacelles:
(b) For aircraft weighing between 60,000 and 150,000 pounds, having
4, 6, or 8 engine nacelles:
WB = &*Wg - 450.
2. For gross weights between 20,000 and 60,000 pounds:
3. Gross weights between 60,000 and 150,000 pounds:
Wâ€ž = 2 (WM)Â«*.
147
3. For jet installations, the weight of the power plant plus the nacelles
is from 7 to 8 per cent of the gross weight less.
4. Nacelles in the wings of small airplanes are from 1.5 to 2% of the
gross weight.
Fuel Weight
1. The specific fuel consumption (e) of propeller-operated engines is
usually given in terms of the number of pounds per horsepower per hour.
For jet engines, it is given in terms of number of pounds per pound of
thrust per hour. There may be considerable variation through the
operating regime of the engine, but for weight calculations, the specific
fuel consumption at cruising power is generally selected. The power for
cruising speed is usually 2/3 of the rated power rating. The following
formula, applicable to any type engine, may be used conveniently:
w _ (e) (2/3 rated power) (range in miles)
/ cruising speed in miles per hour
2. The weight of one gallon of gasoline is 6 pounds.
Oil Weight
1. This weight may be determined: by a formula similar to that for
fuel; or by allowing at least 1 gallon of oil for every 16 gallons of fuel; or
by allowing 10 gallons of oil plus an additional gallon for every 20 gallons
of fuel. The volume of one gallon is 231 in.8.
2. The weight of one gallon of oil is 8 pounds.
Power Plant Weights
1. Engine weights may be found in Chapter XVI, The Power Plant.
2. Engine controls for a single engine up to 500 hp. = 10-15 lb.
3. Engine mount up to 500 hp, including rubber bushings, bolts, etc.,
= 32-40 lb.
4. Engine mount = 3 to 3^% of gross weight.
5. Exhaust manifolds:
(a) short stacks = 12-18 lb;
(b) collectors = 30-50 lb;
(c) (700 hp capacity) = 33M lb to 49 lb.
6. Firewall = 7-10 lb.
7. Fuel system: piping, clips, etc. = 20-30 lb per engine.
8. Pumps (hand) = 2.5 lb.

9. Tanks (welded, riveted aluminum, or aluminum alloy): for oil or
gasoline 10-to 50-gal capacity = 9-30 lb (varies linearly);

50- to 100-gal capacity = 30-50 lb (varies linearly);
100- to 200-gal capacity = 50-80 lb (varies linearly).

148
10. Recommended sizes for fuel lines:
Qutside
diam. (in.)
Wall
thickness,
min. (in.)
Purpose
Fuel to be delivered
(a)
From tank to fuel pumps
Up to 60 gal/hr
From 60 to 100 gal/hr
0.040
(b)
(1) Pumps to carburetor,
Up to 30 gal/hr
0.032
0.040
(2) Carburetor to overflow
drains
(c)
Fuel pumps, relief lines
Same as for item 1
(d)
Primer tubing
0.032
(e)
Vents
0.032
11. Some representative weights for oil coolers:
Rating, hp Approximate tveight, lb
100 4.50
100-350 6.65
350-450 8.00
450-550 9.50
550 up 12.10 up
12. Power plant weights for commercial transport airplanes of gross
weights between 20,000 and 100,000 pounds:
w" = [lâ„¢-wmw-]w-
pounds:
wÂ» - [325â€” mm w>]w*
pounds:
wÂ»'[M7-v&w']w';
pounds:
w" = [â„¢-mmw-]w*
Gross Weight
In order to obtain some idea of the magnitude of the gross weight of an
airplane before a detailed breakdown is made of the weights of the various
components, an estimation on the basis of power loading (W/P) or thrust

loading (W/T) is often advisable. Multiplying the appropriate ratio by
the rated horsepower or the static thrust rating yields an approximate
value of the gross weight.
1. Personal and business airplanes have power loadings as follows:
(a) Single-engine planes with horsepower ratings of between 65 and
450 have power loadings (W/P) of between 10 and 16 pounds
per horsepower.
(b) Twin-engine airplanes with total horsepower of between 290 and
900 have power loadings of between 10 and 12 pounds per horse-
power. These airplanes usually have higher top speeds and
higher rates of climb than single-engine airplanes.
2. Civil and military transport airplanes with 2 or 4 engines, each
having between 1475 to 3500 horsepower, have power loadings of between
10 to 12.5 pounds per horsepower.
3. Private airplanes with single jet units, operating at speeds less than
500 miles per hour, have thrust loadings (W/T) of between 3.5 to 6.25
pounds of gross weight per pound of thrust.
4. Fighters and interceptors with single jet units, operating between
000 and 700 miles per hour, have thrust loadings (W/T) of between 2.5
and 3.85, with the majority at about 3.
5. Transport and bombers with two jet units, operating at speeds of
about 650 miles per hour, have weight-thrust ratios of between 3.5 and 3.85.
6. Transports and bombers with 4, 6, or 8 jet units, operating at speeds
around 650 miles per hour, have weight-thrust ratios of between 4 and
5.75.
7. Research aircraft operating at speeds between 650 and 1,650 miles
per hour or greater, whether utilizing one or two jet units, have had
weight-thrust ratios of between 2 and 3.
Miscellaneous Weights
Weights of equipment will vary with type and quantity. See Table
VIII-2 for some data on representative items.
Propeller Weights
For weights of propellers see Chapter XV, The Propeller.

150
Table VIII-2. Miscellaneous weights.
Item
Weight
(lb)
Item
Weight
(lb)
Ash trays, coat and hat racks00,
service call systems, etc., per
passenger 2.5
Axe, hand 2.0
Baggage, usual allowance per
passenger 40.0
Belts, safety 0.75 to 1.1
Berth equipment:
Bedsheet 0.5
Blanket 2.7
Mattress 4.8
Pillow 1.8
Pillowcase 0.2
Mattress slip 1.4
Upper-berth shock cords 0.1
Lower-berth curtain 1.0
Coat hanger 0.1
Lower-berth support rail 1.1
Upper-berth step 2.2
Table 4.0
Case, aircraft data 1.5
Case, airplane mooring 1.3
Case contents, airplane mooring
(quantity in parentheses):
Arrow, mooring anchor (18). . 0.2
Bag, mooring (1) 1.5
Eye assembly, mooring anchor
rod (3) 0.1

Pin, mooring ground breaking
(1) 0.7
Rod, mooring anchor (1) 1.2
Rod, mooring anchor (6) 0.3
Case, map: 1.09
Chairs, seats (upholstered):
Al. Alloy-reclining (single seat) 12.0
Al. Alloy-reclining (double
seat) 24.5
Steel tubing (nonreclining)... 8.5
Steel tubing (reclining) 13.0
Wicker (settee) 10.1
Wicker 14.4
Steward's seat 9.0
Combustion heaters 17.0
Cockpit enclosure and fairing
(small airplane) 80-100
Control system, including dual
rudder, pedals, dual wheel
control, stabilizer adjust-
ment mechansism, cables,
rods, pulleys, and supports
(small airplane) 110-120
Crew and passengers (each per-
son) 170
Curtain 1.25
Cushion, life preserver 2.00
Cushion, rubber inflated
(17x19 in.) 1.39
Cushion, rubber inflated
(17x29 in.) 2.14
Cylinders, Oxygen
Volume Over-all length

Table V11I-2. (Continued). Miscellaneous weights.
Item Weight
(ft)
Flashlight 1.75
Flit and spray gun 2.5
Floors, metal (per ft2) 1-1^
Floors, wood (per ft2) 1-1 Vi
Heating and ventilating system,
including cabin ducts (per
passenger) 0.5-10
Holder, flight report 0.86
Holder, paper cup 2.0
Holder, paper towel 2.0
Holder, magazines 4.0
Hydraulic operating system for
landing gear and wing flaps
(for airplane of gross weight
of 19,000 lb) 120
Ladder, entrance , 9.0
Lavatory and equipment 45.0
Machete 2.5
Magazine holders 4.0
Mail or express (allow at least
12VÂ£ cu ft for mail to be car-
ried for each) 200
Mail bag (with brass padlock) for
registered mail 6.5
Mail bag, iron locked 2.5
Mirror (7 x 10 inches) 1.3
Pail (10 quarts), metal 2.0
Parachutes
Irving, 24-ft lap type 18.0
Irving, 24-ft slat type 19.5
Irving, 24-ft back type 17.5
Irving, 28-ft back type 24.0
Pilot's seat cushion type 23.5

Russell, with cushion and back
pad 21.0
Switlik (weights in pounds; in-
clude complete equipment):
Over-
U-ft
size
Pack Type
Canopy
Canopy
Seat (fixed)
20H
Seat (quick attachable)
2iy2
23 H
Back (fixed)
18
20
Item Weight
(lb)
Back (quick attach-
able) 19 21
Chair chute 16^ 18
Chest 20 22
Pump:
Engine-driven fuel 3.0
Engine-driven gear type 5.9
Engine-driven vacuum 5.1
Engine-driven vane type 3.5
Gear-type hydraulic 5.9
Hand-operated hydraulic 3.8
1.50 cubic-inch hand hydraulic 3.3
2.00 cubic-inch hand hydraulic 3.7

152
Table VIII-3.
High pressure landing-wheel and tail-wheel tires.
Size tire
and wheel
Max
static
load
(lb)
Infl
press
(lbs
per
in.)
Wt in poundsâ€¢
subject to
change with-
out notice
Tire
only
Reg
tube
Dimensions (in.)
Over-
all
diam
Sec-
tion
diam
Static
loaded
radius
Landing-Wheel Tires
30x5
1,600
30
17.0
2.6
29.76
5.01
13.7
12.0
26x6
3,750
90
21.4
2.8
25.53
6.53
11.4
8.9
26x6
3,750
63
26.6
2.8
25.62
6.56
11.4
9.3
30x7
4,700
90
28.6
3.8
29.19
7.70
13.0
9.9
Low-profile nose-wheel tires.
Wt in pounds,
subject to
Size tire
and wheel
Max
static
load0
Ob)
Infl
press
(lbs
change with-
out notice
Dimensions (in.)
per
Over-
all
diam
Sec-
tion
diam
Static
loaded
radius
Flat
tire
radius
in.)
Tire
only
Reg
tube
19x6.80â€”10
1,600
55
12.4
2.4
19.38
6.76
8;7
7.1
22x7.25â€”11.50
2,000
55
14.5
3.2
22.08
7.29
9.8
7.8
26x9.00â€”13
3,750
50
30.3
4.2
25.92
8.85
11.6
9.2
30x10.50â€”15
5,000
50
46.2
6.3
29.98
10.30
154
Table VIII-5. Fabricated componentsâ€”Floats.
Estimated Added Weight
Gross Weight of Original of Floats Replacing
Land00plane (lb) Landing Gear (lb)
1,125 106
1,620 180
1,965 200
2,425 230
2,880 260
3,610 313
4,210 363
4,910 397
5,680 417
6,560 437
Table VIII-6. Hydraulic equipment.
Weight (lb)
Actuating Cylinders: .
Working
Strohe(in.) Pressure (psi)
Landing gear 50068 1,500 8
Landing gear 10.80 1,500 9
Wing flap 4.25 1,500 3.75
WngflaJ 4.75 1,500 8.75
Door operating 4.30 1,500 3.06
Pin pulling 2.69 150 3
Nose wheel â€” l>700 8du
Length (in.)
Extended Retracted
47 28 1,000 28
57 31 1.000 5
49 21.75 1,100 13
11.5 3 1,250 5.5
17.75 11.5 1,250 4.5
33 19.75 1,000 5.5
41.5 26.5 1,000 17.5

32 18 1,500 4.75
17 12 1,000 5.25
29.5 18.5 1,500 15
17.75 11.5 1,250 6
20 13.5 1,250 5.25
17.75 11.5 1,250 6.5
Accumulators:
9 in. dia., 240 in.Â» J*
10 in. dia., 532 in.Â»
5 in. dia., 69 in.Â» 3 50
Pump, metering....... . â€¢ â€¢ â€¢ â€¢....... :â€¢
Pump, hand (1,350 in.-lb at 1,500 psi less handle) 2.20

CHAPTER IX
The Balance Diagram
The balance diagram shows the location of the centers of gravity of the
components of the airplane as well as its contents. These centers of
gravity are usually indicated by reference to suitable horizontal and ver-
tical datum planes which may fall outside the drawing of the airplane
structure itself. For ease in calculation, these reference planes are so
chosen that all moment arms are of the same sign. For use in the field,
however, the balance diagram is redrawn to show the location of the cen-
ters of gravity with reference to a point (usually the leading edge of the
wing at the root) which may be easily located on the airplane.
GENERAL PROCEDURE
To be sure that the centers of gravity are located correctly, it is desir-
able to draw or refer to an inboard profile drawing showing the location of
all the main items drawn to scale.
Along with the balance diagram, there is a balance table listing all the
items contained in the weight estimate, with their weights and center of
gravity locations, as well as the vertical and horizontal moments. Calcu-
lations for various possible centers of gravity for the complete airplane
are included.
In an airplane the center of gravity, or "balance point," may change
while in flight. For example, the fuel and oil are being used up at a con-
stant rate and, therefore, some of the weight disappears so that the center
of gravity may change. Another case may occur when the balance is cor-
rect with the passenger cabin filled. What happens when there are no
passengers, or only a few? Suppose again, that there are only a few pas-
sengers in a large cabin and when the flight starts, they sit in the forward
portion of the cabin, but later decide to sit in the rear?
These possible changes caused by what may be called the "disposable
load" (although rightly the crew is included ordinarily in the term) have
to be considered when the design of an airplane is contemplated.
155
156
The balance diagram is such a study. It starts as a preliminary instal-
lation diagram, and therefore the fuselage becomes the first object in air-
plane design since it normally houses all the items that may change in
character, number, or location.
In a single-engine passenger airplane, for example, with the engine in
the nose of the fuselage, the procedure in obtaining a likely balance dia-
gram is to draw the engine and engine accessories accurately to scale, then
the firewall, then the pilot's cockpit, then the passenger cabin, in the order
in which they normally occur. They are arranged according to the de-
signer's wishes and drawn accurately to scale.
This arrangement or installation serves several purposes. It is similar
to the designing of a house by an architect who knows that he has to place
the kitchen, bedrooms, bathrooms, and the like to provide the maximum
of comfort and convenience. At the same time these arrangements help
to determine certain dimensions of the house. This principle applies also
to airplane design. The placement of the engine compartment, the pilot's
cockpit, and the passenger cabin will affect the length of the fuselage, its
size, and weight.
After the interior arrangements have been placed more or less as they
are wanted, a preliminary investigation is made to find out how much the
center of gravity of the fuselage and its contents vary when a certain num-
ber of passengers are removed. It may be found that because of first dic-
tates for passenger comfort, too much space has been allotted between
seats so that when the rear seats of the cabin are empty the center of
gravity is too far forward; or the reverse is true when the front passenger
seats are empty. Thus the seats should be a little closer so that having a
few seats empty will not make so great a difference.
Very often, the designer is overgenerous in allowing for passenger com-
fort and finds that giving too much fore and aft room makes for poor bal-
ance for the various possible flight-loading conditions. Likewise, allow-
ing too much height for the passengers may increase the cross-sectional
area of the fuselage so much that performance eventually will be impaired.
The top view is usually left to the last, unless the passenger accommo-
dations are not in rows.
It is usually wise not to give too much space for the engine compartment,
the pilot's cockpit, or the passenger cabin. Especially in the first solution
is it desirable to keep to minimum dimensions to avoid grief. Unfortu-

nately, generosity is denied the designer. He must learn to take advan-
tage of many small factors instead of a few large ones. To the uninitiated,
the balance diagram may seem simple, but it is really the crux of the entire
design. If the rules laid down subsequently in this chapter are obeyed,
there is comparatively easy sailing ahead.
After the preliminary three-view and preliminary weight estimate one

THE BALANCE DIAGRAM
157
proceeds to the preliminary balance diagram. This includes part of the
installation diagram or inboard profile as well, because arrangements have
to be substantially correct to assure a balance diagram of reasonable
accuracy.
CENTER OF GRAVITY LOCATION
The center of gravity location along the chord of the wing determines
the pitching moments of the airplane. In order to get the airplane to trim
at a given angle of attack, it is necessary to have the sum of the pitching
moments equal to zero at that angle of attack. In the conventional air-
plane, the pitciiing moments produced by the aerodynamic forces acting
on the wing are counterbalanced by the pitching moments of the tail sur-
faces. The pitching moments produced by the fuselage usually play but
a small part.
The forces and moments acting on a typical low-wing monoplane are
shown in the figure given in Chapter III, Airfoil Selection. The resultant
of the parasite resistance, which includes the drag of the tail surfaces, is
assumed to act through the center of gravity, primarily to simplify the
calculations to be made (although it is partially justified by wind-tunnel
measurements). For further simplification the thrust is assumed to act
through the center of gravity.
The pitching moments about the center of gravity may then be expressed
as follows:
MC.B. = Ma.c. + LwyiC sin a + LwXiC cos a â€” DwyiC sin a +
DuXiC sin a â€” LtXzC cos at â€” Lty2C sin at
qCSCif,.,. = qCSCM.p. + qCSCL(.Xi cos a + j/i sin a) + qCSCD (x sin a
â€” j/i cos a) â€” qCSCi.,St/8 cos at y2 sin at),
and simplifying further:
Cm.*. â€” Cm.*. + Xi(Cl cos a + Cd sin a) + j/i(Ct sin a â€”
CD cos a) â€” y2CuStia
where cos at has been considered equal to 1, sin at negligible, and the drag
of the tail surfaces included in the parasite drag, and where:
q =y*pv\
v = speed of the airplane at the trim angle in feet per second,
S = wing area in square feet,
C = mean aerodynamic chord in feet,

Cl, Cd, and Cm.... = aerodynamic characteristics of the wing,
Cl, = lift coefficient of the horizontal tail surfaces,

Cm.*. = moment coefficient of the wing airfoil about its aerodynamic
center,
158
Cm.*. =
Lw â€”
Da =
DP =
Ma.c. =
Me.,. =
Lt
pitching moment coefficient of the airplane about its center of
gravity,
qCLS,
qCDS,
qCD,S,
qCMâ€ž.cs,
qCLjSt,
7 moment arms of force vectors about center of gravity.
J Fig. III-l.
See
To account for the fact that the air velocity over the tail surfaces may
not be so great as that over the wing, the quantity in is usually
0.8
^04
.I
Â§
6-0.4
ac-<~
-y
1+
*1
â– 'i<
e
^+1
\
â€¢ â€”r
^+05
N\
0.5
Complete N
Wing nlane*^
airplane
\^
-4e
8*
12' 16Â°
Figure IX-1. Wing pitching momentsâ€”the effect of moving the center of gravity
along a line parallel to the wing chord and through the aerodynamic center.
THE BALANCE DIAGRAM
159
multiplied by a factor, often assumed as 0.8 to account for a decreased
velocity of about 0.9 over the wing.
The pitching moments due to the aerodynamic forces on the wing alone
have been calculated for a number of center of gravity locations and have
been plotted in Figures IX-1 and IX-2. It will be noted that moving the
center of gravity longitudinally along the chord has a greater effect upon
the pitching moments than moving the center of gravity perpendicularly
to the chord.
Superimposed upon these curves is another representing the desired
slope and angle of trim for the pitching moment curves for the complete
airplane. The difference in pitching moments for the complete airplane
and for the wing alone represents the pitching moments that would have
to be supplied by the horizontal tail surfaces.
Since the pitching moments of the horizontal tail surfaces are propor-
tional to XtCjJSt, it can be seen immediately that the larger the tail mo-
ment required, the larger either the tail length x2C should be, or the
greater the horizontal tail area St should be, or the combination of both.
Further, the slope of the pitching moment curve is a function of the slope
12'
16* 20Â°
Figure IX-2. Wing pitching momentsâ€”the effect of moving the center of gravity
perpendicular to the wing chord and through the aerodynamic center.

160
dd/da of the lift curve of the tail surfaces. This slope is determined by
the aspect ratio of the tail surfaces and the larger the aspect ratio, the
greater the slope.
Practical Solution
In order to obtain the right degree of stability with the necessary amoun t
of control, the center of gravity of the airplane is placed usually as close to
the aerodynamic center of the wing as possible while the horizontal tail
surfaces are placed from to 3 times the mean aerodynamic chord be-
hind the aerodynamic center of the wing.
Center of Gravity Movement
Since the airplane has to carry a variable payload from time to time, it
is necessary to know how much movement may be tolerated in the location
of the airplane center of gravity. An investigation can be carried out for
the given design by calculating the pitching moments for the complete
airplane for different center of gravity locations.
If the center of gravity is moved ahead of its design position, the air-
plane will be more stable, but will trim with the same horizontal tail sur-
face configuration as for the original center of gravity position at a smaller
angle of attack. This condition is referred to sometimes as the "nose-
heavy" condition. If the center of gravity is moved rearward, the air-
plane will be less stable and, with the same tail surface configuration, will
trim at a higher angle of attack. This is known as the "tail-heavy" con-
dition.
The general rule is that the center of gravity of the completed airplane
should not change more than 10 to 14 per cent of the mean geometric
chord of the wing between the fully loaded condition and a condition
corresponding to the most rearward or most forward position of the center
of gravity (not necessarily the fully empty condition). The center of
gravity travel could be restricted to about 7% of the chord ahead of the
center of gravity, and to about 7% of the chord behind the center of
gravity. For tailless designs the center of gravity movement is limited
to about 5 or 6 per cent of the mean aerodynamic chord. This means
that for an airplane weighing 10,000 pounds, fully loaded, which has a
mean geometric chord of 100 inches, the movement for a condition less
than fully loaded should not be more than 8 inches. When the fuselage
and its contents are considered alone (as they would be in preliminary
calculations) this movement for the same airplane may be greater, say
12 per cent, or 12 inches in this case.
For preliminary design, an eight per cent travel in the center of gravity
movement is used. If the airplane were highly stable for the fully-loaded
condition, then an eight per cent movement forward would increase the
THE BALANCE DIAGRAM
161
stability and would therefore be undesirable. Similarly, if the airplane
had the minimum stability for the loaded condition, then an eight per
cent movement rearward would decrease the longitudinal stability.
Therefore, in order to determine whether the airplane has sufficient
stability, it is incumbent upon the designer to carefully analyze the design
after the balance diagram and the general configuration have been set.
After detailed static longitudinal calculations have been made for the
fully loaded airplane, some restrictions can be set on the movement of the
center of gravity location for variously loaded conditions.
1. If longitudinal stability is more than adequate, the center of gravity
movement should be primarily to the rear.
2. If longitudinal stability is less than adequate, the center of gravity
movement should be forward.
3. If longitudinal stability is just adequate, the center of gravity move-
ment may be equally forward and rearward.
THE FUSELAGE AND ITS CONTENTS
Starting with the fuselage and its contents, each item from the nose of
the fuselage rearward is placed carefully to scale.
When the fuselage and its contents are located, the center of gravity of
the fuselage and its contents are determined. In the calculations to de-
cide upon the center of gravity, the horizontal and vertical tail surfaces
should be included as factors.
To determine the center of gravity, two datum lines are chosen: one
through the center of the propeller in case of a single-engine airplane, or
tangent to the foremost point on the airplane for the. horizontal arms of
the individual items; another, usually the ground line, for the vertical
arms. It is desirable to choose these datum lines so that the arms are all
of the same signs, thus avoiding possible errors in addition.
Sometimes a number of diagrams are made until the final center of
gravity of the whole group is located where the designer wants it. Figure
IX-3 shows one way in which a diagram may be presented.
The calculations should be set up in the form of a table, listing all items
included in the fuselage and its contents. Table IX-1 has been set up
as a guide for the balance calculations to be made.
It should be noted that it is convenient to lump together small items

located at about the same spot and determine the center of gravity of the
group by inspection. Otherwise, determination of the center of gravity

of the airplane becomes a series of tedious calculations. The error that
may be caused by such a method is certainly not large, as a few simple
calculations will demonstrate to the most skeptical.
Even in multi-engine designs, where the engines are attached to the

Table IX-1. Typical balance calculation data.
(1)
.(2)
Distance
(3)
(4)
Distance
(5)
from
vertical
datum
from
horizontal
datum
Weight
in lb
Horizontal
moment
Vertical
moment
Item
â€¢
2. Engine
^Starter
X Hotspot
/STÔiltank
9. Oil
11. Pilot
12. Co-pilot
13. Pilots' seats
14. Controls
16. Two seats
^STTwo seats
10. Two seats
20.' Two seats
24. Two passengers
28. Radio equipment

,29^ Lighting
31. Fuselage structure

32. Vertical tail surfaces
33. Horizontal tail surfaces
34. Tail wheel and gear...
^ Fuel tanks
37. Fuel
163
164
Column 1 = sum of all weight items;
Column 3 = sum of all horizontal moment calculations;
Column 2 = H
ZM/r totals in column 3
XW totals in column l'
Column 5 = sum of all vertical moment calculations;
â€ž . , Tr tMy totals in column 5
Column 4 = 7= â€”â€”0 0 = â€” .
2JV totals in column 1
wings, it is usually desirable to treat the engines with the fuselage as a unit,
since the fore and aft position of the engines is determined by the location
of the plane of rotation of the propeller with reference to the occupants of
the fuselage. The supporting structure making up the engine nacelle will
then be 00constructed later when the wing has been placed in its correct
position.
The reason that it is generally more desirable to follow this procedure
is that the wing planf orm and spar locations may undergo several revisions
in order to get suitable attachment points for the spars at the fuselage that
will not interfere with any internal arrangements. With each spar loca-
tion or wing planform, however, the original position of the engines re-
mains the same.
The mean aerodynamic chord is fixed with relation to the center of
gravity of the complete airplane, but any wing planform having the same
length and location of the mean aerodynamic chord may be used. (See
Figure IX-4.) The terms mean geometric chord and mean aerodynamic
chord are often used interchangeably.
Likewise, the engine is more or less fixed, either because of center of
gravity location requirements for the complete airplane or, more likely,
because the pilot and the primary-control units (excluding cables and
control rods) should be so located that none lies in the region between the
plane of rotation of the propeller and the surface generated by a line pass-
ing at an angle of 5 degrees through the axis of propeller rotation.
THE WING AND THE LANDING GEAR
It is usually easier to treat next the wing and the landing gear as a unit.
Since both the wing and the landing-gear positions are critical with refer-
ence to the final location of the center of gravity, it is appropriate to start
with a final center of gravity location. Figure IX-5 shows the complete
airplane, but in preliminary work it is unnecessary to treat more than the
wing and the landing gear.

166
tion of the propeller. The vertical position will have to be assumed. For
a low-wing monoplane, the center of gravity may be assumed at about
seat level; for a high-wing monoplane, at about shoulder level. Should
this assumption be found later to be a few inches off, it will usually not
affect the final configuration of the wing and landing gear.
A line drawn forward at an angle between 13 and 17 degrees with the
vertical drawn through the center of gravity will intersect the ground line
at the point of tangency of the wheel with the ground. Thus the wheel
is located.
The quarter point of the mean aerodynamic chord of the wing falls di-
rectly below (in the case of a low-wing monoplane) or directly above (in
the case of a high-wing monoplane) the final center of gravity. Therefore,
this requirement is observed by placing^the wing accordingly.
The centers of gravity of these component parts are easily established,
and the center of gravity of the group is then determined. Refer to
Figure IX-5 and Table IX-2.
Table IX-2.
(1)
(2)
(3)
Weight,
Vertical
Vertical
No.
Item designation
arm, y
moment, Wy
Wing
w,
Vi
wiVi
w,
2/2
3
w,
2/3
W,y,
Landing-gear struts
wt
2/Â«
Y _ Wlyi + W,y, + Wty, + Wtyt
Wl + Wt + W, +
It is now necessary to locate the horizontal position of the center of
gravity of this group. Refer to Figure IX-6 and Table IX-3.
Table IX-3.
(1)
(2)
(3)
Weight,
Horizontal
Horizontal
No.
Item designation
arm, x
moment, Wx
Wing
Wl
Xl
WlXl
wt
Xt
THE BALANCE DIAGRAM
167
Figube IX-tt. A simplified planview of the previous figure, showing horizontal
moment arms of the wing and landing-gear groups. The numbers 1, 2, 3, and 4 refer
to data shown in Table IX-2.

168
When these calculations are finished, the group may be located on the
fuselage. In Figure IX-7, 1 indicates the center of gravity of the fuselage
and its contents. Its horizontal movement is known and may be found
from Table IX-1, considering all except the last four items. The 2 indi-
cates the center of gravity of the wing and landing-gear group which has
just been determined. Its distance d behind the quarter point on the
mean geometric chord C is also known since
d = X - \iC.
The final position of the center of gravity of the complete airplane must
be in the plane y-y through the quarter point just mentioned. Let:
Wi = weight of fuselage and its contents,
Wt = weight of wing and landing gear group.
Since the horizontal moment of the complete airplane equals the sum of
the horizontal moment of the component parts, it follows that:
(JFi + W,) (L + x) = WJj + [W, (L + x + d)],
from which Wix = Wî is obtained so that the value of
Wtd
*= FT*
gives the necessary information to locate the wing on the fuselage.
If the engines and engine nacelles in a multi-engine design were consid-
ered with the wing and landing-gear group in these calculations instead of
with the fuselage group, it would be possible for the center of gravity of
the group (wing, engines, fuel tanks, landing gear, etc.) to fall ahead of the
quarter point of the mean aerodynamic chord. Corrections in the above
derivations can be made accordingly.
Before proceeding with the final form of the balance diagram, check the
following items for ground clearance:
1. Depressed elevator.
2. Depressed flap.
3. Propeller.
Also check the distance between the tail post and the center of gravity
in terms of the mean geometric chord. All weight items appearing in the
original weight estimate should appear in the balance table.
When the balance diagram has been made in its final form, the balance
table given in Table IX-1 should then be corrected accordingly.

SUPPLEMENTARY CALCULATIONS
Before the balance data are accepted, supplementary calculations are

necessary to determine the positions of the center of gravity for various
possible loading conditionsâ€”fully loaded, and partially or all empty. The

THE BALANCE DIAGRAM
169
extreme movement of the center of gravity should be limited. A total
horizontal movement of about 8 per cent of the mean geometric chord is
permissible and, if possible, half of this movement should be ahead and
half behind the center of gravity for the fully loaded position. A greater
movement may be permitted if static stability calculations for the various
locations of the center of gravity show that the airplane has sufficient
stability.
If the movement is extreme, rearrangement of variable, not fixed, items
such as passengers, baggage, fuel (but excluding the pilot in these calcu-
lations) may be made to reduce the movement. Should the center of
gravity movement still be beyond the desirable limits, ballast should be
provided.
BALLAST
Suggestions for the use of ballast:
Ballast may be used to enable aircraft to comply with the flight-test
requirements for longitudinal stability, balance, and landing, with the
following provisions:
(a) Ballast should not be used for this purpose in aircraft having a
gross weight of less than 5000 pounds, nor in aircraft providing a
total seating capacity of less than 7 persons.
(b) It should be demonstrated by the manufacturer that the specific
aircraft can be landed safely without ballast, passengers, or
cargo.
(c) The place or places for carrying ballast should be properly de-
signed and installedâ€¢ and plainly marked.
(d) The loading schedule which accompanies each license issued for
an aircraft may require that special loading of this type shall be
conspicuously posted either in the pilot's compartment or adja-
cent to the ballast compartments. Strict compliance should be
required of the aircraft operator.
These suggestions should be carefully checked against current govern-
mental regulations.
CENTERS OF GRAVITY OF INDIVIDUAL ITEMS
The centers of gravity of, the individual items may be determined by in-
spection, or rough calculations, or from known locations.

The following list will give some indication of the procedure to be fol-
lowed.
Propeller
Engine
Accessories
At centerline
Given by manufacturer
Estimate
170
Person seated At bottom vest button
Seat Seat level, forward of back
Furnishings, soundproofing At geometric center of cabin
Flooring At geometric centroid
Doors At geometric centroid
Windows At geometric centroid
Tail surfaces Center of gravity of each chord section is about
30-35% of chord
Wing At 40-42% of each chord section; make allow-
ance for flaps
Fuel and oil tanks At geometric centroid
Landing gear By inspection
If the wing is assumed to weigh the same per square foot from root to
tip, and if the center of gravity of each chord is located from 25 to 40 per
cent behind the leading edge, the center of gravity of the entire wing
would be at 25 or 40 per cent of the mean geometric chord, as the case may
be. Its vertical position may be assumed halfway between the top and
bottom camber at the mean chord, taking into account the location of the
mean chord due to dihedral and other structural factors.
If tail surfaces have irregular planforms, sufficient accuracy will be
obtained by resolving the areas in equivalent rectangles and triangles.
Fuel tanks, filled with fuel, of irregular cross section from one end to
the other may be resolved into a series of volumes whose centers of gravity
can be determined readily, and it is an easy matter to find the center of
gravity of the complete unit. The center of gravity of the empty tank,
unless more accurate data are available, may be assumed to coincide with
the center of gravity of the tank when filled.
CENTER OF GRAVITY ESTIMATION
One may occasionally be required to determine the center of gravity of
an object with which one is not familiar and for which there is no pertinent
information at hand. That would be the time to use some extra imagina-
tion and common sense.
Let us take a fuselage as an example. Where might one expect to find
the location of the center of gravity of the bare fuselage structure? A good
method of procedure would be to assume a cylinder of equal length and

uniform weight; then it is obvious that the center of gravity would be
located at the center. The center of gravity of the fuselage must then
lie ahead rather than behind the midpoint, unless some unusual condition
exists. The larger cross section ahead of the midpoint, the greater con-
centration of load, and the many more complexities of the structure caused
by windows and doors all indicate a center of gravity at about the mid-
point of this front half section alone, or at 25% of the total length. The
center of gravity of the complete fuselage structure would then lie some-
where between 25 and 50% of its length. As a first approximation, 35 to
40% would be a reasonable value.

THE BALANCE DIAGRAM
171
While such an approach assumes that the structure has some degree of
homogeneity, it still yields useful results for a first approximation. Later,
when more data is available as to location of frames, doors, floors, and
the like, a more careful and precise determination of the center of gravity
can be made.
ERROR IN BALANCE CALCULATIONS
Slight errors in estimating the exact location of the center of gravity
are not serious. Suppose, for example, that the center of gravity of an
item weighing 50 pounds was wrongly located by two inches, making for
a total error of 100 inch-pounds in moment. If the gross weight of the
airplane was 2000 pounds, then the center of gravity location would be off
by 100/2000, or one twentieth of an inchâ€”certainly not a large error, and
perhaps compensated for by other similar errors. The wrong location of
the center of gravity of the individual items may be more serious, how-
ever, when the local effects under dynamic conditions are considered.
It is certainly desirable to have data available that is as accurate as
possible, but the absence of such data, especially for non-critical items,
should not delay the progress of preliminary design.

CHAPTER
The Cockpit
The cockpit may be considered the most vital part of the airplane, for
in it all functions of the airplane and its power plant are coordinated and
directed by the pilot and his aides.
All control systems terminate in the cockpit; all operational and navi-
gational instruments are located here; all decisions regarding the flight of
the airplane, with the very few exceptions when the airplane is about to
take off and land, are determined here. As the airplane grows larger, the
cockpit assumes even greater importance. At present it is even more im-
portant than the captain's bridge of an ocean liner.
Since the pilot and his aides, such as the co-pilot or the navigational
officer or some other member of the crew, spend their entire time on a
flight in the cockpit, it is absolutely necessary that every means for com-
fort, for ease of operation of controls, for co-ordination of instruments, for
vision, and a host of other odds and ends which contribute to the efficiency
and well-being of the crew, be carefully planned and arranged. Too often
an airplane, otherwise satisfactory, cannot overcome the enormous sales
resistance caused by a poorly designed or arranged cockpit. The pilot
may find that he has not been given enough leg room or enough headroom
so that even on moderately short flights, he is easily tired due to a cramped
position. Or, he may find that his vision forward, sideward, upward, and
downward is very poor so that when he takes off from an airport or wants
to land, he must guess instead of being able to see how or where he is tak-
ing off or landing.
Another difficulty may be that the instruments are not arranged prop-
erly: that he cannot quickly see the instruments which he needs to guide
him in the proper operation of the engines and in the proper execution of
his flight maneuvers.
In some cases minor faults may be corrected to suit the flying personnel
but, for example, if there is insufficient headroom in the cockpit it may
172
THE COCKPIT
173
not be possible to make changes unless the airplane is completely rede-
signed.
Naturally, the thought arises: Why not standardize the cockpit in its
dimensions, appointments, and general arrangements? In other words,
design a satisfactory cockpit once and for all. Unfortunately it is not
possible to standardize cockpit design wholly until all the parts and equip-
ment located in the cockpit have been standardized.
New instruments are designed daily that record or indicate additional
data not measured before. Or, new equipment is designed that will take
care of some additional functions, but which may not wholly replace other
equipment of almost but not quite similar functions. The additions to
the available list of instruments and equipment are necessary in order to
lighten the burdens of the operating personnel, but their placement may
well upset the planned installation of the standardized cockpit. Each
new cockpit, therefore, offers its own problems.
However, there is standardization to some degree and this helps enor-
mously in reducing the many problems attendant in cockpit arrangement.
The military services are attempting to standardize cockpit designs for
certain types of airplanes.
The description below* is presented to give some idea and understand-
ing of the cockpit arrangements and procedures that a pilot or student
must follow, for example, in a jet trainer.
The side-by-side seating arrangement is favored for a trainer for two
reasons:
(1) Student confidence builder. With instructor alongside, student can be more
easily indoctrinated with jet techniques and procedures.
(2) Ease of instruction. Instructor can observe student at all times, be "ahead" of
him in all phases of flight training without having to resort to guesswork.
The general arrangement of instruments and controls are such that, with the student
sitting on left side, the instructor on the right, the navigation and flight instruments are
positioned on student's side, the engine instruments above throttle pedestal. The radio
controls are in front of the instructor, but are within reach of the student.
Positioned on the left, on student's side, are navigational and flight instruments
including directional and altitude indicators, altimeter, turn and slip, rate of climb and
airspeed indicators, course indicator (omni), and clock. Engine instruments, set over
the center quadrant, include tachometers, exhaust temperature indicators, fuel flow
indicators, oil pressure gages, loadmeters, accelerometer and fuel quantity indicator.
In front of the instructor, on the right but within reach of student, are radio controls,
circuit breaker and instrument panel, a.c. fuses, plus airspeed and turn and slip indicators
and altimeter.
Controls are easily accessible from either side of cockpit. Stick grips and throttle
quadrants are fighter type. Four throttles are provided, two for each quadrant.
Throttle movement mechanically actuates each engine fuel control by means of push-
pull rods and torque tubes.
Lift-type idle detents are included on instructor's quadrant to prevent inadvertent
positioning of either set of throttles from idle to cut-off. There is no de-icing equipment,
* Aviation Week, April 1, 1057, pp. 48-55.

174
but an engine ice-warning light, on instrument panel, will illuminate when ice forms over
the ice-detect probe in left engine air inlet duct.
Two automatic press-breathing, diluter-demand oxygen regulators for both student
and instructor are located on lower left and right side of instrument panel.
The trainer incorporates a simple cockpit check. Airplane can be started on its own
battery or external starter. Six to 10 starts can be made on former. Standard starting
procedure is: Battery and inverter on, fuel pump on, open fuel shut-off valve for left
engine, push up on starter switch and hold at 5% rpm, push up on ignition switch and
hold, bring throttle forward to idle, release starter and ignition switches at 22% rpm
Engine will idle at 35%. Same procedure is used to start right engine.
Power steering is activated by button on stick, and slight rudder pressure steers air-
plane neatly and quickly. Nose wheel will steer on free swivel to 50 deg. Large
transparent canopy allows excellent visibility during taxi and flight.
Before take-off there are two safety pins for the student to pull and one for the instruc-
tor. One of the pins is for canopy jettison and one for student's ejection seat. The
instructor's pin is for his ejection seat. Before aircraft is entered, ground crew removes
a fourth jettison pin for canopy behind and beyond reach of pilots.
A "hot" mike, geared to the helmet, enables instructor and student to talkâ€”without
button pushingâ€”from time electrical power is turned on until engines are shut down.
SUGGESTED PROCEDURE IN DESIGN
Since the pilot is the most important item in the design of his headquar-
ters, the cockpit is planned around him. To obtain a better conception
of his requirements, a celluloid figure with joints is made to scale to which
the cockpit installation is to be drawn. This jointed figure can then be
placed in various postures to see whether the pilot's seat is far enough up
from the floor, that the legs are not outstretched too far for comfort, and
that the control stick or wheel is not too far ahead so that it is awkward
to operate. Of course, slight adjustments are obtainable in raising or low-
ering the seat, or in moving it forward or rearward.
If there is insufficient headroom for the pilot to stand up in the cockpit,
then there should be at least enough room above his head so that he can
lean forward or back, or raise himself up slightly from the seat without
fear of bumping his head.
Once the pilot has been comfortably seatedâ€”and it is desirable that his
legs make an angle of not greater than 45 degrees with the floor to assure
comfort in flightâ€”it is necessary to locate the seat so that he may have
sufficient vision in all possible directions, with special attention to for-

ward, upward, downward, and sideward vision. When this has been done,
the windshield may be located. It should not be too far forward since the
ceiling will come too far ahead of the pilot and interfere with his line of
vision. Likewise, the windshield should not be curved in more than two
directions because distortion of the scene may result.

THE COCKPIT
175
Next, the instruments and equipment must be considered. Instru-
mentation is discussed in detail in a subsequent chapter, but reference to
a few required features may well be made here. All switches and controls
ivhich the pilot must operate should be located as close to him as possible
so that it will be unnecessary for him to stretch. Stretching reduces the
pilot's ability to operate the airplane properly at any time and especially
in an emergency. On the other hand, those instruments which do not
have to be operated manually may be farther away, but they must be
grouped according to functions. The more important functional group is
located in front of him. Others are slightly out of his prevailing line of
direct vision.
When the arrangements for distances and clearances have been settled
above and below the pilot, as well as behind and ahead of him, some atten-
tion should be given to "elbow" room. Here again, he should be given as
much room as possible. If there is a co-pilot, the width should be doubled
and an appreciable allowance made for the aisle between the two. These
allowances often determine the width of the fuselage unless the airplane is
very large. The result is that designers tend to skimp on width allowances
because the greater the width, the greater the eventual effect upon the
speed performance of the airplane. However, a difference of 2 or 3 inches
will be enormous as far as the comfort of the pilot is concerned and will
hardly affect the speed characteristics of the airplane.
In new designs it is usually the custom to lay out the pilot's cockpit on
paper, as discussed before, with the aid of additional information given
hereafter, and then to construct a "mockup" which is a full-scale represen-
tation of the actual installation in wood, metal, and cardboard that is
nailed, bolted, or screwed together. In the mockup, a man may sit in the
cockpit and test the arrangements before the airplane is constructed.
Changes are made until everything is as satisfactory as accompanying
conditions permit. The final solution is then drawn up and incorporated
into the final design.
GENERAL REQUIREMENTS
The following items must be considered in the design and study of ar-
rangements for the cockpit:
1. Windshield outline and construction.

2. Angles and field of vision.
3. Instruments and their location.

4. Power-plant controls and their location.
5. Pilot's and co-pilot's seats.
6. Primary control systems.
7. Brake systems.
8. Hydraulic controls for brakes, flaps, tabs, etc.

176
9. Automatic-pilot equipment.
10. Radio equipment.
11. Lighting.
12. Heating and ventilating.
13. De-icing equipment and controls.
14. Oxygen equipment.
15. Accessibility; emergency exits.
The cockpit for an airplane should be constructed to give the maximum
possible comfort, adequate vision, and accessibility to all controls. In a
cabin plane, consideration must be given to headroom, ventilation, and
vision.
The cockpit and primary control units, excluding cables and control
rods, should be so located with respect to the propellers that the pilot or
controls are not in the region between the plane of rotation of any propeller
and the surface generated by a line passing through the center of the pro-
peller hub and making an angle of 5 degrees forward or tfft of the plane of
rotation of the propeller.
Unless the airplane carries less than five passengers, the pilot or pilots
are located in a compartment separated from the cabin. Entrance to the
cockpit from the cabin is permitted, but passage through the cockpit
should not be considered as an emergency exit for the passengers.
When the pilot is entirely separated from the passengers, suitable means
for communication between the pilot or pilots and the passengers should
be provided.
Oxygen supply or a supercharged cabin should be provided for the crew
and the passengers above 18,000 feet altitude, or when operations are for
more than 15 minutes above an altitude of 14,000 feet.
The arrangement of controls, provision for adequate vision, and seating
arrangements, as well as numerous other problems, are best solved by con-
structing a mockup in which every item is represented.
WINDOWS AND WINDSfflELDS
The general requirements for windows and windshields of the pilot's
compartment cover the following items:
1. Satisfactory forward vision under all conditions.
2. Suitable protection under all conditions.

3. Windshields should be easily opened in flight.
4. Windshields should be so arranged that the air stream and snow or

rain are deflected across the opening when the windshields are open.
5. The compartment should be so arranged that glare and reflection are
avoided.
6. Leakage into the compartment during rain or snow should be avoided.

THE COCKPIT
177
Figure X-l. Two windshield arrangements
for military airplanes. The top windshield simi-
lar to that of the Fairchild "Cornell" may be
built in sections that slide rearward, one section
telescoping over the other. The bottom wind-
shield, similar to that of the North American
XF-JI, is an example of the bubble canopy,
which has practically no visual obstructions but
may have some optical distortions.
Figure X-2. Windshield
configurations for large trans-
port airplane. The top arrange-
ment may offer less aerodynamic
resistance while the lower may
give the better visibility.
For best vision, the windshield should be placed vertically although this
materially increases aerodynamic resistance. Hence, the rearward-slop-
ing windshield which tends to keep the streamline shape of the fuselage
is the best.
It has been proposed that in order to obtain a streamlined fuselage for
an airplane designed for greatest aerodynamic efficiency, the cockpit could
be made to retract into the fuselage just after take-off. This could be
done only with great complication.
For unusually large airplanes, it is possible to locate the pilot's cockpit
in the nose of the fuselage where, due to the enormous size, flat panes in-
serted in the surface hardly affect the smooth contour of the streamlined
body.
Windshields should be so installed that they may be cleaned or opened
easily in flight. Unless the windshields are such that rain, snow, or ice
will not stick to the surface and impede vision, they should be arranged so
that the air stream and snow or rain are deflected across the opening.
178
Figure X-3. Other possible
solutions to the windshield prob-
lem. These offer relatively little
aerodynamic resistance, but do
not offer the highest degree of
visibility to the pilots.
Figure X-4. Windshield
treatments for high-speed trans-
port airplanes. A special, so-
called retractable, windshield
covers the main windshield in
the lower solution. For take-
off and landing, the outer wind-
shield is retracted.
Figure X-5. This windshield arrangement provides better visibility for the pilot,
but offers greater aerodynamic resistance than any of the previous solutions shown.
179
Figure X-6. A typical nose-treatment for high-speed aircraft. Note the careful
streamlining of the windshield into the fuselage lines. The use of a periscope for the
landing operation has been proposed, with radar serving for the other situations of visual
observation. The nose section can be constructed in the form of a jettisonable capsule,
as indicated by the solid outlines.
Figure X-7. A pilot's canopy that hinges at the rear and is held up by a hydraulic-
ally-operated center jack. When in place, further engagement is obtained by means
of the three lugs on each side, through which pins are passed. All connections can be
released in time for the ejection of the pilot with his seat.
Figure X-8. One way of obtaining ice-free windshields.

180
Windshields fronting on the sides of the cabin may be kept free of ice by
means of a rotary windshield wiper that operates at relatively high speed.
One of the airlines has devised a double windshield, consisting of J^-inch
safety-glass pane in front and a J^-inch transparent plastic rear pane with
a J^-inch air space separating the two. Air, heated to 170Â° F, from the
cabin heating system, is forced through between the panes at a rate of
about 60 cubic feet per minute.
The windshield arrangement should be such as to give minimum interior
reflections and glare which might interfere with the pilot's vision, particu-
larly at night.
In a mockup, the windows may be represented by mirrors that show the
parts of the cockpit which may be reflected. By suitable rearrangement
of the windshield, these areas may be avoided or provision made to reduce
the reflection. Glare may be investigated by means of an outside source
of light that can be moved with respect to the cockpit.
By means of Figure X-9, the thickness of the glass for the windshield
may be determined. The safe pressure or load that may be sustained by
the glass may be calculated approximately by means of the formula
P = 21,000 fl/Af
where P = unit pressure in pounds per square inch (or square foot)
(assume P = ^pV2mox in lb/ft2),
t = thickness of glass in inches,
A â€” area in square inches (or square feet),
/ = safety factor, assumed equal to 10.
Characteristics of Glass
Coefficient of expansion of plate glass between â€”70Â° F to +100Â° F is
approximately 0.00000823 per degree centigrade or 0.00000457 per degree
Fahrenheit.
Compression 36,000 lb per in.2 Modulus of elasticity 10 X 106 to
Tension 6,500 lb per in.2 11 X 106 lb per in.2
Modulus of rup- 3,500 lb per in.2 Weight 3.29 lb per ft2 (M inch
ture thickness)
All the windows in the windshield should be made to open quickly, and
be removable, if desired. The glass should not lie too flat. An angle be-
tween 0 and 45 degrees between the vertical and plane of the glass when
the airplane is horizontal is recommended. If the glass lies too flat, dis-
tortion of vision or undesirable reflections from the sky above may result.
The thickness of the glass depends upon the type and size of plane.
In general, nonshatterable glass, at least % inch thick should be used.
Anything thinner is subject to accidental breakage or warping.

182
VISION AND VISIBILITY
The cockpit in jet-propelled aircraft can usually be placed well forward
in the fuselage, thus affording pilots a type of visibility during landing
approach that is not possible with propeller-driven aircraft. Downward
visibility of 17Â° or00more over the nose can be obtained, whereas the other
type of airplane permits as little as 3Â° and seldom more than 5Â°.
Downward visibility is limited by the type of equipment installed
ahead of the pilot.
The pilot at high altitudes, in addition to his other problems, is faced
with an inversion of the normal brightness direction. The sky and the
clouds at or near the earth's surface are normally brighter than the ter-
rain. Thus, at high altitude the pilot is covered with a black sky, while
the haze and clouds in the atmosphere below him are extremely bright.
Human beings are designed with eyes recessed under the frontal region of
the skull and with eyebrows superimposed. This is to protect the eyes
from brightness from above; but there is no protection from brightness
coming from belowâ€”a distinct handicap for the pilot.
Unobstructed forward vision for the pilot under all conditions is most
desirable. For easy landing, it should be possible to see the wheels at the
moment of contact. It is customary to allow an angle of not less than 5
degrees subtended by the eye below the horizontal to permit adequate
vision forward.
Figure X-10. The closer the obstruction to the eye, the greater the angle subtended
for impeding vision.
The twin-engine designs can usually provide for excellent vision forward
and sideways owing to the absence of such obstructions as the engine
cylinders.
A wing with sweepback usually does not permit as good a view of the
landing gear as a wing that does not incorporate a sweepback.
Obstructions such as posts to support the windshield do not offer as
much of a "blind spot" when located relatively far from the pilot's eyes.
Preliminary studies should be made to determine the most desirable
location of such supports as far as vision is concerned.
A larger field of vision is available to the pilot if the windshield or
windows are relatively close to the pilot's face. The type of windshield
that slopes rearward from top to bottom (Figure X-5) does not afford
THE COCKPIT
183
Figure X-12. Windshield arrangement for
a two-seater low-wing monoplane (top) simi-
lar to that of the Aeronca "Chum," and for a
small flying boat, (bottom) similar to that of
the Republic "Sea Bee."
much vision upward or sideways since the pilot is required to seat himself
farther back than if the windshield sloped the other way.
Canopies
The cleaner aerodynamic lines required at transonic and supersonic
speeds, as well as the difficulty of providing large transparent areas of
high-heat-resistant material may lead to the elimination of canopies.
One airplane now being built incorporates a periscope.
SEATING
The pilot should be comfortably seated. His seat should be adjustable
vertically, as well as fore and aft, and angularly. The back of the seat
should not be too high.
The seat should be provided with arms, so designed that the inside arm
can be swung out of the way when getting in or out of the seat.
When two seats in a side-by-side arrangement are provided in the pilot's
compartments, it is desirable to have equal access to the controls and in-
struments; where this is not possible, the left-hand seat should be favored.
This is known as the first pilot's seat, and the right-hand one is termed
the second pilot's seat.
When a wheel control is used, the height of the wheel should be such that
186
28 inches; across each knee, 6 to 8 inches; and above knees and thighs, 26
inches. For the occupant of the private airplane with an operational ceil-
ing below 10,000 feet and with provision of a reasonable heating system,
necessary clothing is not so heavy nor thick so that the dimensions speci-
fied in the listing may be reduced in some cases.
Before cockpit design has progressed to the mockup stage, it is desir-
able to construct a two-dimensional reduced scale model of the supposed
pilot. For this purpose, the dimensions of a man who is 6 feet tall are
presented in Figure X-14 as a practical compromise between the 5-foot
3-inch pilot and the 6-foot 3-inch pilot.
EXITS
If the cockpit is not accessible from the cabin, a separate door should be
provided for the cockpit. It is generally desirable to have an emergency
exit as well. If the windshield is large enough when opened, it may serve
as an emergency exit; otherwise, one should be provided.
When the airplane is not equipped to carry passengers, the pilot's com-
partment may be either open or equipped with a hatch of such size that it
can be used for an emergency exit for the crew with parachutes.
An opening for an emergency exit should be at least 17 X 24 inches
when rectangular in shape, or 24 inches in diameter if circular. Such
openings are generally not large enough, however, for a man equipped with
a parachute, and are usually intended for emergency exit on the ground.
PROTECTION FOR THE PILOT
Without a shoulder harness to hold the pilot in his seat, severe or fatal
injury is probable in a landing accident as a result of his head striking
the instrument panel. The danger can be avoided by making the space
in front of the occupant free of obstacles for a distance approximately
equal to the length of the torso from the hips to the top of the head,
plus seat-belt elongation. This would necessitate a 31 to 45 inch space
clear of any solid or unyielding protuberances which might injure not
only the head but also other parts of the body.
The pilot's head would be critically injured in a crash against a flat,
rigid surface at an average impact energy of about 600 in./lb. If low
density, energy-absorbing materials are used to pad the crash surfaces,
the head can take impacts five times greater.

In order to lessen or prevent injury, surfaces against which the pilot
might hit his head in a crash should have the following characteristics:
1. They should be flat or have large radii for curvature;
2. They should be rigid enough to help support the head but should
deflect under contact;

THE COCKPIT
187
3. They should be smooth, with no sharp edges to cause local high
pressure intensity;
4. The contact surface should be as large as possible for distributing
contact pressures;
5. Provision should be made for absorbing a relatively large amount
of impact or kinetic energy.
EJECTION EQUIPMENT
At low speeds, the customary method for a pilot to leave an airplane
under emergency conditions is to jump over the side. This method is
reasonably safe for aircraft speeds up to Mach 0.5. Beyond this limit,
various escape methods have been proposed and possible solutions arc
still being investigated.
It has been suggested that between speeds of Mach 0.5 and 1.5, an
escape chute located in the bottom of the fuselage may be suitable. Ejec-
tion of the pilot, complete with seat and head and face protection gear, is
another alternative.
From about Mach 1.0 up, a capsule holding the pilot's cockpit would
be detached from the rest of the aircraft structure by pneumatic or rocket
means.
Such problems as aerodynamic heating, stability in descent, and dy-
namic loads encountered make the design of suitable ejection equipment
and methods as challenging as the design of the entire aircraft.
Some appreciation of the difficulties encountered may be gained when
it is considered that from altitudes of 30,000 and above, tumbling and
spinning are serious hazards. Experiments with human subjects have
shown that a spin of 160 rpm for 12 seconds will produce unconsciousness,
and that visual, mental, and physical disorientation will take place at 90
rpm, or even less. Rotation at 200 rpm for two minutes has proved
fatal for animal subjects. Therefore, means will have to be provided to
make any self-contained system free from tumbling and spinning char-
acteristics.
It is believed that a streamlined capsule containing all of the cockpit
equipment would have greater mass and better streamlining than a body
in an ejection seat. Thus, it could reduce deceleration to within human
tolerance and give protection from tumbling, wind blast, extreme temper-
atures, and lack of oxygen. At altitudes outside the normal atmosphere,
the capsule would also protect the pilot from such conditions as dynamic
heating and sudden decelerations, which make re-entry into the normal
atmosphere quite dangerous. Furthermore, such a capsule would provide
an artificial environment capable of supporting the pilot for considerable
periods.
188
PARACHUTES
It is not customary to provide the crew with parachutes when passen-
gers are carried. When the airplane does not carry passengers, pilots
should be equipped with parachutes. Figure X-15 shows seating pro-
visions to be made for different type parachutes.
CONTROLS
All controls should be readily accessible.
The distance from a control wheel in its rearmost position to the back of
Figure X-16. Suggested configurations for the pilot's control wheel.

THE COCKPIT
189
the pilot should be at least 12 inches. The height of the wheel should
clear the pilot's legs with the seat in its highest position.
The maximum movement of the
controls should correspond to the
maximum movement of the con-
trol surfaces.
In a small airplane, seating 2
to 5 passengers, for example, it is
sometimes customary to place a
passenger alongside the pilot. In
such a case, the control column or
stick should not be located between
the pilot and passenger unless the
"throw-over" type of wheel con-
trol is incorporated.
The controls should be so arranged or constructed that neither the pas-
sengers, crew, nor cargo will interfere with the operation of such controls
during the course of flight of the airplane. Further, all controls must be
so located and designed that the pilot or co-pilot will not bump them when
moving in or out of his seat, or in and around the cockpit. Latest designs
of control handles simulate the mechanism which the handle operates.
This makes for faster action by the pilot and minimizes error in selecting
the proper control.
Figure X-17. Control levers are being
equipped with handles suggesting the mech-
anism each handle operates. (left) Landing
gear retractionâ€¢ (right) Flap lever.
INSTRUMENT BOARD
There is no agreement as to the best arrangement of the instrument
board. It is customary to place the automatic pilot in the top center of
the instrument board, with the flight instruments and power-plant instru-
ments in front of the pilot. The controls and switches should be readily
accessible, and in case of pilot and co-pilot arrangements, the engine con-
trols should be located between the two.
In a twin-engine airplane there are some 16 controls for the power plant
alone, about 11 flight-control instruments in addition to the flying controls,
landing-gear retraction, lights, heating and ventilating control, radio, and

brakes, as well as flaps and tab controls, for which provision must be made
on the instrument board or in the cockpit.
The instrument board should have sufficient room behind it to provide
easy access even to the largest instrument. It may be made of any mate-
rial, although aluminum alloy is commonly used.
In order to take additional factors into account, further reference should
be made to Chapters XI, XVI, and XX, entitled Instruments and Equip-
ment, The Power Plant, and Control Systems, respectively.

190
Figure X-18. Side and plan views of controls and instrument board for prone
position of pilot, which permits smaller cross section of fuselage for high-speed airplanes.
CHAPTER
Instruments and Equipment
From the first flight of the Wright brothers, and practically up to about
1928, pilots flew by feel and intuition. Today they cannot trust their own
senses at high altitudes or in fogs or in cross-country flights or in blind fly-
ing, since airplanes are becoming faster and more complicated with each
new design. Fogs, high altitudes, and night flying obliterate landmarks.
High headwinds, sleet, snow, and rain interfere with the intuitive senses
of the pilots. They must rely, for safety's sake, almost entirely on radio
communication, radio beacons, range compass findings, gyroscopic com-
passes, automatic pilots, turn and bank indicators, and at least 25 other
dials and gadgets essential to the safe operation of the airplane in all kinds
of weather.
In transoceanic flying the job becomes even more complicated. The
captain of a transoceanic flying boat must be also a first-class seaman, a
master mariner, a radio expert, an aeronautical engineer, an engine and
airplane mechanic, a celestial navigator, and a host of other things. But
no matter how well qualified the pilot may be along these lines, he would
be entirely helpless unless there were instruments or equipment to show
how the engines were performing, what speed the airplane was making, at
what altitude he was flying (so as to clear mountains), or what course he
was following, and the innumerable other indicators required.
This wide variety of information has to be given by reliable instruments,
and due to the diverse nature of information, the instruments, gadgets,
and mechanical aids become so numerousâ€¢ that the ideal arrangement of
the various instruments, for example, on the aircraft instrument board has
been the subject of considerable controversy among pilots since the pilot's
needs also vary with the type of flying he has to do. To this day, it has
been impossible for pilots to agree upon a definite location of even a few
instruments. This state of affairs may be better understood when it is
considered that the use of various instruments differs considerably for each
191
192
particular flying problem; certain instruments, for example, necessary to
make instrument landings (that is, making landings aided by instruments
only and not by actual view of the airport on which landing is taking place)
are rarely, if ever, used in normal landing. The best instrument board,
therefore, is the best compromise that has been worked out for the par-
ticular conditions the airplane has to meet.
Of course there are difficulties. It would be highly desirable for a pilot
to be able to step into a strange airplane and feel reasonably sure that he
will be able to locate any particular instrument in the usual place on the
instrument board. Such standardization, in addition to the reasons al-
ready pointed out, is also difficult owing to space and structural limita-
tions, as well as the size and shape of various airplanes. Some degree of
standardization is possible if the majority of the more important instru-
ments are placed in the same general location on the panel.
The military services prescribe some specifications, and the various air-
lines are accustomed to setting forth their own. Manufacturers of air-
craft are able to follow these in most instances except for the relatively
small airplane where some elasticity is necessary because of the small area
of the instrument board usually available.
The standardization, so far made, calls for the following arrangement.
The primary flight group is immediately in front of the pilot and near the
top of the panel. This group consists of the Sperry turn indicator and the
Sperry flight indicator, both on the same level. Below these is the second-
ary flight group consisting of the airspeed, bank and turn indicator com-
bined, and the rate of climb instruments. To the left, either in the same
row or as close as possible, the sensitive altimeter is located. In addition,
it is customary to locate the magnetic and radio compasses as conveniently
close to the other flight instruments as possible.
The engine instruments are usually grouped in the same general pattern,
depending upon their number, with the tachometer as close as possible to
the flight instruments.
There is often a series of other instruments placed according to the best
location available as well as the particular purpose for which the instru-
ments were designed.
It is estimated that in a subsonic jet fighter, the military pilot will

have checked his instruments at a rate of nearly 200 per minute. In
the presence of rain, air turbulence, and darkness, he will have inter-
preted these readings and applied control forces for definite corrections
in flight path and attitude at a rate of about 50 per minute. He will
have responded to, actuated, or concerned himself with over 200 cockpit
instruments, indicators, levers, and switches. At the point when he
stops his aircraft and clears the runway, he will have made more than
1,000 major and minor decisions and 2,000 lever, switch, and control
INSTRUMENTS AND EQUIPMENT 193
actuations or deflectionsâ€”all compressed within an uninterruptable
segment of time.
While the commercial pilot may not have quite so many operations,
with the increase in the speed of aircraft, the time for split decisions
decreases and the number of operations increases.
Qâ€”*- Aircraft in focus in visual field -Â«â€”O
Oâ€”*- Lag in perception - 0.1 second â€¢*â€”O
Oâ€”*â€¢ Time for recognition -0.5 second -*-O
O"*- Decision accomplished-0.1 second -*â€”O
0~*~Human reaction time-0.4 second-*â€”O
Machine reaction time plus time to
[ \)
deviate from flight path-2.0 seconds
q Flight path without countering action q
Time in seconds
4 3 2 10 12 3 4
3520 2640 1760 880 0 880 1760 2640 3520
Distance in feet
Figure XI-1. Estimated man-machine time-distance relationships of two aircraft
traveling a 180-degree collision course at a speed of 600 mph. An increase in the
amount of time spent in any phase of the time-response sequence would result in a
collision. Similarly, a marked increase in the relative aircraft speeds would make
avoidance impossible. From a distance of 20,000 feet, an airplane with a 100-foot
wing span would be visible as a point. Two 900-mph airplanes approaching head-on
would cover this distance in about 7.5 seconds.
INSTRUMENT BOARD
The instrument board may be of any convenient material although
aluminum or aluminum alloy is most commonly used. For small instru-
ment panels, plywood may be used. Instrument boards of 17ST or 24ST
aluminum alloy sheet are made in thicknesses varying from J^j; to }/$ inch.
Formica or bakelite instrument boards are from Y%\,o}/i inch thick.
The instrument board should be fully accessible from the rear so that
any instrument may be removed readily. Special attention should be

194
paid to the depth of the instrument, space required for removal, and extra
space for access to connections and the like.
Since instruments should not be subjected to vibration, the board should
be mounted on either felt or sponge rubber. Elastic stop nuts are excel-
lent for fastening down bolts which do not need extra lock washers. As
the board with the instruments has sufficient weight of its own to acquire
a vigorous vibration, both it and the individual instruments should be
mounted carefully with respect to vibration. The compass, particularly,
should be mounted as independently as possible.
Location
Instruments should be located far enough away from the pilot to permit
him to see all the instruments at a glance. If there are a great many in-
struments, the more important ones should be grouped so that they may
be seen readily.
The instrument board location is usually determined by cockpit limita-
tions, control stick clearances, and the like.
Grouping
The grouping of instruments depends upon the functions they perform.
Instruments relating to engine operation should usually be close to-
gether and preferably right in the line of sight of the pilot. Likewise,
instruments used for navigation, instruments used for communications,
and instruments and accessories used for heating and ventilating and
similar occasional operational functions should be grouped together.
Authorities differ as to the desirable arrangement of instruments in a par-
ticular group.
Variety of Instruments
The variety of instruments that may be included in the modern airplane
can best be realized from the following incomplete list.
1. Air distance recorderâ€”records the number of air miles.
2. Air speed indicatorâ€”registers the actual air speed, referred to stand-
ard air, at all times.
3. Altimeterâ€”a visual registering (but not recording) barometer, except
that instead of indicating the atmospheric pressure in equivalent inches of
mercury, the dial is graduated to read the altitude in feet corresponding to
the pressure.
4. Accelerometerâ€”may not be incorporated in the instrument board lay-
out. It is used to determine the maximum loads imposed upon the air-
plane in gusts or in landing. It is often carried to give research workers
additional data, but is not required or used by the pilot for his information.
5. Ammeters and voltmetersâ€”measure the electrical power used for vari-

ous instruments or equipment such as radios, telephones, and electrically
operated retraction gear.
6. Automatic pilotâ€”a device manufactured by the Sperry Company
and used to operate the control surfaces to maintain an even keel without
the intervention of the pilot. The automatic pilot relieves the pilot of
much strain of constantly "flying" the airplane so that he can devote more
attention to the behavior of the power plant and other equipment.
7. Bank indicatorâ€”a visual indicating device for registering the angle
of bank.
8. Climb indicatorâ€”registers the rate of climb in feet per minute, so
that the pilot has a ready means of determining whether he is climbing fast
enough. If it registers zero, the airplane is flying level; if it registers nega-
tively, the airplane is descending. The dial is marked to read "Up" and
"Down."
9. Clockâ€”is very important for navigation and maintaining schedules.
10. Compassâ€”may be one of two types, magnetic or radio. Both are
usually used for cross-checking purposes. For private airplanes, a compass
is required when the range is 100 or more miles, or when operating over
large bodies of water beyond sight of land.
11. Drift indicatorâ€”an instrument which measures the angle between
the actual flight path of the airplane and the longitudinal axis of the air-
plane (which indicates the direction in which the airplane is headed).
12. Fuel level gaugeâ€”a long-distance indicating gauge for registering the
actual fuel level in the fuel tank. If there are several fuel tanks, a sepa-
rate gauge for each tank is required.
13. Fuel pressure gaugeâ€”indicates the fuel pressure in the fuel line,
necessary for determining whether the fuel is brought to the engine in the
proper quantities.
14. Flap controlâ€”either a manually operated device for controlling the
flap or an electric switch for starting electric control. It may also be a
hydraulic pump handle.
15. Generatorsâ€”attached to the engine at locations provided for them.
These generators generate the power required for lighting and for motor-
driven appliances.
16. Manifold pressure gaugeâ€”registers the actual pressures in the engine
manifolds. Nowadays a close check is kept on the manifold pressures

which are changed from time to time by the pilot to give the necessary
power for take-off and cruising.

17. Octantâ€”used for celestial navigation. The instrument measures the
vertical angle between the horizon and a celestial body.
18. Oxygen equipmentâ€”consists of various regulators and indicators as
well as tanks and distributing units. Airplanes flying above 14,000 feet
for any great length of time are required to carry this equipment.
196
19. Propeller controls and indicatorsâ€”Propellers of today are usually of
the controllable pitch. Whether automatically, manually, or semi-auto-
matically controlled, there are certain control instruments and pitch indi-
cators present. These vary according to the type of propeller.
20. Pumpsâ€”Fuel pumps are either hand operated or engine driven;
there may be pumps for the hydraulic systems, and in some cases, oil
pumps may also be required. Pumps operated by hand, or used for in-
termittent duty only, are usually operated from the instrument board.
21. Radioâ€”Safe flying is dependent upon radio and telephone com-
munication. This equipment has a number of switches, dials, and con-
trols which must be located so as to be easily accessible to the co-pilot or
radioman.
22. Suction gaugesâ€”indicate the suction pressure in the lines leading to
the automatic pilot and other instruments depending upon suction for
their operation.
23. Sperry artificial horizonâ€”aâ€¢ visual indicating device used to indicate
the "horizon" when the actual horizon is obscured. It is of particular
importance in all blind flying operations when the intuitive sense of man
is decidedly unreliable.
24. Switchesâ€”ignition switches, radio switches, light switches, electrical
appliance and equipment switches in any number, found in the cockpit.
25. Startersâ€”Control for the operation of the engine starter may be
found in the cockpit unless the airplane is small.
26. Tachometerâ€”It is very important that the crankshaft speeds of the
engines be known at all times. Any sudden change is a warning that
something must be wrong. The revolutions per minute indicated by the
tachometer also serve as some indication of the actual power developed by
the engines on take-off; the indicated rpm is high since more power is re-
quired for the take-off than at any other time. During cruising, the rpm
of the engine is lower than for high speed since less power is required and
better fuel and oil consumptions are obtained at lower engine speeds.
27. Thermometersâ€”used to register oil temperatures; to register the
temperature of the coolant in case of liquid-cooled engines; to register
cabin temperatures for heating and ventilating purposes; to register the
outside temperatures (especially important when icing conditions are

likely to be present).
28. Turn indicatorâ€”often combined with the bank indicator; when the
two are used in conjunction with each other they are considered among
the most important flight-control instruments and should be located on
the instrument board in the direct and unobstructed forward view of the
pilot in the center of what is called the "primary flight group" consisting
of the airspeed indicator, the turn-and-bank, and the rate-of-climb indi-
cator.
29. Warning unitsâ€”may be special lights that flash on and off; or buzz-
ing signals; or any similar devices. Very often the positions of the hands
on the various indicating dials are sufficient warning signals. Since the
dials may be rotated, some instruments are so arranged that during the
most common flying regime, such as cruising, dial hands or pointers of all
the most important instruments point in the same general direction. If
one is out of alignment, it may be an indication that either the instrument
is not functioning or that the particular piece of apparatus with which it
is used is not functioning properly. Both cases are likely to be present.
Selection of Instruments
Instruments and equipment required or desirable for an airplane de-
pends greatly upon the purpose, and whether visual contact or instrument
flying is required. Recent airplane airworthiness regulations require
some or all of the following instruments and indicators for nonpassenger-
carrying airplanes.
1 airspeed indicator,
1 altimeter,
1 tachometer for each engine,
1 oil-pressure gauge when an oil-pressure system is used,
1 water thermometer for each water-cooled engine used,
^00t~oil thermometer for each air-cooled engine used,
1 manifold-pressure gauge, or equivalent, for each altitude engine,
1 fuel quantity gauge,
1 landing-gear position indicator for retractable main landing gear,
1 device for measuring or indicating the amount of oil in the tanks,
1 magnetic compass (for airplanes flying visual-contact, distance
greater than 100 miles).
For night visual-contact flying, the following should be added:
position lights,
landing lights,
landing flank,
storage battery for lights and radio,
radio range and weather broadcast receiver,
a set of spare fuses.
If instrument flying is contemplated, the following are added:
1 gyroscopic rate-of-turn indicator,

1 bank indicator,
1 sensitive altimeter,
1 clock with a sweep-second hand,
1 generator,
1 rate-of-climb indicator.
198
For airplanes carrying passengers, the following may be additional instru-
ments and equipment required, or may replace some of similar category
listed already:
1 electrically heated pitot tube, or equivalent for airspeed indicator,
1 additional portable fire extinguisher,
1 fixed fire extinguishing system for each engine compartment,
1 noncontinuous manual means for providing vision through windshield
in case of rain during take-offs and landings, instrument lights,
cabin lights in all passenger cabins and compartments,
1 gyroscopic direction finder,
1 free air thermometer of the distance type,
1 icing indicator.
ELECTRICAL EQUIPMENT
The electrical equipment and system of a large transport airplane is per-
haps more complex and larger than many reasonably sized communities.
All the storage and generation of the electrical power supply must be car-
ried by the airplane. Special precautions must be taken to protect the
electrical system from fuel, oil, water, and any other damaging substances.
Moreover, adequate clearances must be provided between the wiring and
such installations as oil and fuel tanks and systems, carburetors, exhaust
piping, and moving parts.
Storage batteries should be easily accessible, preferably from outside of
the airplane, and adequately isolated from fuel, oil, and ignition systems.
Suitable ventilation and isolation must be provided for each battery.
Fuses are a required protective feature, so installed to protect all elec-
trically operated equipment, and made readily accessible in flight.
In estimating the size of the generating system required, the engine-
driven generator should be designed to carry the running load with a
stand-by system if feasible. The running load comprises the electrical
consumption by all lights, radio equipment including radio-range signal re-
ceivers, and any other equipment which exacts a constant load. Radio
broadcasting equipment, landing lights, and electrically operated landing
gears and flaps are considered as causing only intermittent use. Landing
lights should be so installed on multi-engine aircraft so that at least one
light is not less than 10 feet either right or left of the first pilot's seat and
beyond the swept disk of the outermost propellers.
Instrument lights should provide sufficient illumination and should be

equipped with rheostat control for dimming unless the nondimming light
can be shown to be satisfactory.
Position lights are required for airplanes operating at night. In normal
flying position, the red forward-position light on the left and the green
forward-position light on the right should be spaced as far apart as possible

on the wing. Each should show unbroken light between two vertical
planes whose dihedral angle is 110 degrees when measured to the left and
right, respectively, of the airplane from dead ahead.
One rear-position light, either continuous white or alternate red and
white flashes (depending upon whether the airplane is a carrier or noncar-
rier type) should be placed as far back as possible. It should emit light
throughout a dihedral angle of 140 degrees bisected by a vertical plane
through the longitudinal axis of the airplane.
A master switch has to be provided in the cockpit for the entire elec-
trical equipment.
SAFETY EQUIPMENT
Every airplane is required to have some safety equipment, such as safety
belts for pilot, crew, and passengers. Private airplanes may carry a para-
chute for each occupant, but this item becomes a considerable weight fac-
tor where the number of passengers is reasonably large. Since the majority
of accidents, when they do occur, happen at altitudes where parachutes
are of no value, the presence of parachutes is more psychological than
practical.
Fire extinguishers are either portable or fixed installation type. At
least one portable fire extinguisher should be accessible to the passengers,
and another accessible to the crew. Airplanes carrying no passengers
need carry only one fire extinguisher.
Landing flares for use in emergencies should be releasable from the pilot's
compartment. Airplanes designed only for contact flying need carry no
landing flares.
DE-ICING AND ANTI-ICING
Included in the safety equipment which is required for transport air-
planes are de-icers for all the surfaces and some form of de-icing means for
the propeller. One of the positive methods of breaking up dry ice forma-
tion on the leading edges of the wing and tail surfaces are rubber overshoes
containing tubes that are inflated and deflated alternately to break up
any ice formation.
Heating the leading edge surfaces by conducting heat either from the
engine exhaust pipes or from especially designed heaters is also employed
on larger airplanes to prevent ice formation. This method is often referred
to as anti-icing.
The formation of ice on propellers may be avoided by suitable anti-icing
fluid which is distributed by slinger rings at the hub.

Some significant features suggested for an aircraft heating and thermal
anti-icing system, as proposed by the Stewart-Warner Corporation, are
indicated in the general description given here.

200
The safety features of the heater include a complete segregation of the
combustion and ventilating systems, with the combustion air being ob-
tained from a source separate from the ventilating air; the enclosure of the
heater fuel system so that no fuel connections are exposed within the
ventilating air stream; and, the construction of the heater so that all flame
and combustion gases are contained within a one-piece, hermetically
sealed, all-welded stainless steel assembly.
This heater is also constructed in such a manner that, if installed with
a 3-inch diameter exhaust pipe and with ventilating air ducts of the proper
restriction on the downstream side of the heater, the static pressure in the
ventilating air stream in the heater will be greater than the static pressure
in the combustion passages. Then, even in the event of a rupture of the
normally hermetically sealed heat exchanger, there will be no leakage of
combustion gases and fuel vapors into the ventilating air stream. Instead,
there will be a flow of air from the ventilating air passages into the com-
bustion chamber.
Instantaneous starting at altitudes as high as 41,000 feet, and greater,
can be obtained. Performance runs in which the heater is controlled by
rapid "cycling," as well as starting trials in which the heater is first turned
off and permitted to cool to ambient temperature condition, are part of
the qualifying tests usually specified.
One heater is intended to be mounted horizontally with the exhaust out-
let down so that the exhaust tube may act also as a drain. For that rea-
son, the tube should be installed so that its entire length slopes downward
from the outlet of the heater to the point where it discharges overboard.
Unless the exhaust tube is very short, some provision such as a flexible,
lined, stainless-steel bellows section should be provided to permit expan-
sion of the tube. The entire exhaust system should be designed to handle
exhaust gases at temperatures up to approximately 1000Â° F.
A combustion air regulator valve is needed only if the heaters are re-
quired to operate over a very wide range of air speeds and altitudes.
Usually, when the combustion chamber and fuel system are completely
enclosed, the heater may be installed with ventilating air flowing in either
direction (parallel flow or counter flow).
Heaters are designed for operation on a constant fuel pressure of ap-

proximately 15 pounds per square inch. If it is possible to obtain fuel at
that pressure from the main engine fuel pumps, the weight and potential
service troubles of a separate heater pump can be avoided. Moreover,
the heaters then benefit from all the high-altitude provisions made in the
engine fuel system to ensure delivery to the engine of fuel that is free of air
and vapor bubbles. If it is not possible to obtain fuel from the main engine
pumps, and if the heaters will be required to operate at altitudes in excess
of 15,000 feet, the inlet of the heater pump must be connected to a fuel
INSTRUMENTS AND EQUIPMENT
201
source downstream of the booster pump (customarily located at the fuel
tank) to remove vapor from the main fuel system. Whatever source is
provided, it must be capable of supplying a maximum fuel flow.
Controls include: an air pressure switch to prevent operation of the
heaters in the absence at the heater inlet of a static pressure sufficient to
provide an adequate air flow; a "cycling" thermal switch to maintain outlet
air temperatures at approximately 300Â° F by controlling a pair of fuel
shut-off valves at the heater; and, an overheat switch to close these valves
in the event outlet air temperatures reach 350Â° F. Since the "cycling"
switch normally prevents outlet air temperatures much in excess of 300Â° F,
the overheat switch is not called upon to operate except in the event of
malfunctioning of the "cycling" switch. It is, in short, a safety switch.
Two shut-off valves, arranged electrically in parallel and hydraulically
in series, are used at each group of heaters to provide a double precaution
in the event of sticking of one of the valves.

CHAPTER XII
The Passenger Cabin
It would seem to be an easy matter to design the passenger cabin, to pro-
vide proper seating arrangement, heating and ventilating, and vision out-
ward. After all, there are only a few rows of chairs placed rather closely
together and usually a window beside each chair. But, as in all parts of
the airplane, numerous considerations enter into the final solution which
is the result of many compromises of arrangement, structure, and balance,
as well as weight.
GENERAL CONSIDERATIONS
From the passenger's viewpoint, as much room as possible is desirable,
but if this consideration were given full weight the size of the cabin might
become unwieldy for the type of airplane to be built. Moreover, if all the
seats are not occupied, the center of gravity will not coincide with its posi-
tion for the fully loaded condition, and just as soon as the center of gravity
has moved too far from a given position, the pilot will find it difficult to
balance, and so to fly, the airplane. This has been discussed earlier, to
some extent, when the balance diagram was considered.
Therefore, to avoid a too-large cabin, it is best to start with the question:
"What is the minimum allowance for spacing of seats, for aisle widths,
and for headroom?" A student measured the distance between backs of
two church pews, one behind the other, and found it to be 28 inches, but
also reported that the spacing was not conducive to comfort. However,
one may say that distance is the absolute minimum spacing possible. Any
additional allowance adds to the comfort, but discretion must be used. If
there is a central fuselage with the engines located in the wings, greater
leeway may be permitted for both fully loaded and empty conditions and
still good balance obtained, because the movable or variable load may be
placed with its center of gravity coinciding with the center of gravity of
the complete airplane. For a single-engine airplane with the engine in
the nose, it is not usually possible to get this happy solution and so less
202
THE PASSENGER CABIN
203
allowance must be given to the spacing of seats than absolute comfort may
require.
There should be no obstructions overhead or on the floor. Sometimes,
to avoid too deep a fuselage, a designer will locate the wing spars so that
they extend above the floor level. Usually, in such a case, there is a mini-
mum of headroom so that the passenger is likely to stoop to avoid hitting
his head and while worrying about that matter, he is more than likely to
forget about the floor obstructions. If floor obstructions are necessary
(and that is debatable), a wall with a door in the center may be a good
solution since passengers usually expect to step high over a sill, especially
if in some way their attention is drawn to it.
Obstructions, even on the side walls, should be prohibited unless the
cabin is large enough to permit incorporation of compartments so that side-
wall obstructions can be hidden in the dividing partitions.
COMFORT FACTORS
The interior treatment of the cabin often may give an impression of
spaciousness without being actually large. White or light-colored ceilings
give an impression of height, so that the cabin may be made narrower or
shorter without affecting the passengers unfavorably. There is at present
much interest in interior decoration of aircraft cabins, particularly in the
psychological effects of various treatments.
The major factors affecting the comfort of passengers are:
1. Roominess, which has to be provided through proper seating; as large
cubical capacity per passenger as possible; and, a reasonable amount of
leg room, elbow room, and headroom.
2. Proper heating and ventilating, which is the subject of a subsequent
chapter.
3. Soundproofing, which is of great importance even for airplanes flying
for relatively short periods of time and much more so for transcontinental
and transoceanic airplanes which are in the air for many hours.
4. Vision outward so that the passengers may see the country over which
they are flying. Although on long flights the passengers may find the
scenery monotonous after a short while, there is a soothing effect produced
by being able to see below.
5. Conveniences such as toilet facilities, refreshments, reading material,

writing desks, and other small items, which make traveling more appeal-
ing.
The notes that follow will help the designer in planning the cabin layout.
CABIN DIMENSIONS
The cabin dimensions depend upon the type of chair, the number of
passengers, and the constructor's conceptions of comfort requirements.

THE PASSENGER CABIN
205
Two rows of chairs with a center aisle are common practice. Two seats
on one side and one on the other helps to concentrate the variable load and
is therefore better for stability and trim for empty and full conditions.
The latter arrangement requires a wider fuselage that may affect the per-
formance of the airplane by increasing the parasite resistance.
The main requirements for a comfortable cabin are:
1. An aisle wide enough to permit passengers to walk up and down with-
out disturbing passengers in seats. Aisle widths vary from 12 to 24 inches
with an average width of 16 inches.
2. An aisle ceiling high enough to permit a tall man with a hat on to
walk its length without stooping. Aisle heights vary from 51 inches for
the minimum value up to 75 inches maximum. Except for possible private
use, a passenger-carrying airplane should have an aisle height of at least
72 inches. Elsewhere it is not necessary to have such height as to permit
standing erect, but the use of members over the seats that may be struck
by a passenger rising should be avoided.
3. A minimum width for a cabin seating two passengers should be 66
inches with an additional 25 inches for each additional passenger across
the width of the plane. Cabin widths for two rows of passengers vary
from 52 inches for small cabins up to 66 inches for cabins seating 14 or
more passengers.
A wide aisle gives the impression of spaciousness and may permit some
reduction in the allowance for headroom.
The cabin should be free from structural members. If these are un-
avoidable, separate compartments provided for groups of passengers may
overcome this objection.
PASSENGER SEATS
Passenger seats should incorporate the following features:
1. Seats should be adjustable with respect to the angle of the seat with
the floor and the angle of the back of the seat to the vertical.
2. Seats should be reversible to permit the passengers to face each
other if they care to.
3. Seat adjustments should be made easily and quickly by the passen-
ger.
4. A foot rest should be provided in the form of an apron or shelf, pref-

erably adjustable, attached to the seat ahead.
5. Seats should be permitted to swivel for easy egress if cabin clear-

ances and weight allowances permit. Placing the chair at a slight angle
to the fore and aft axis of the cabin accomplishes practically the same
result.
206
6. Easy egress may be obtained also if the outside arm of the seat can
be dropped.
7. Cushions on seat bottom should be deep and provided with springs
as well as padding to reduce effects of bumps on landing and in the air.
8. Direct metallic contact of the seat with the structure may be avoided
by means of suitable rubber shock-absorbing connections.
9. Width of each seat should be at least 24 inches, with 1-inch clear-
ance between seat and inner lining of the cabin. A minimum width of 19
inches may be permitted.
10. For a narrow cabin, the backs of the seats should be tapered toward
the top, especially those paralleling the aisle. This design permits more
room for passengers going down the aisle.
11. Seats should not weigh more than 20 pounds. Seat weights vary
from about 12 to 20 pounds.
12. Seats may be made convertible into berths by lowering back to
form bottom of berth.
13. Ample leg room should be provided. Seats are spaced from 30 to
40 inches apart, with an average of 36-inch spacing for cabin passenger
capacities of less than 12 persons.
14. All passenger seats should be identical.
SEATING ARRANGEMENTS
The optimum seating arrangement for the occupants of an airplane may
be considered from several points of view, namely: (1) passenger comfort,
(2) minimum effect upon the center of gravity of the airplane as a whole
when one or more passengers are removed, (3) minimum frontal area of
the structure required to house the enclosed group, (4) ready access to the
seats and to the available means of ingress and egress. To these may be
added: (5) minimum over-all dimensions, (6) possible effect upon struc-
tural design, and (7) safety considerations. There are probably other
factors such as: (8) ease in providing suitable vision panels, (9) baggage
space requirements, and (10) multipurpose use of the airplane which
might be studied in relation to passenger seating. Obviously, all these
factors would affect the decision as to the seating arrangement finally
chosen. However, not all these factors can be given the same weight,
since all airplane design is the result of a series of compromises, and it

would be extremely difficult to set up a general method of evaluation sat-
isfactory to every one.

The seating arrangements are dependent upon the size of its occupants.
Of course, where space is limited for engineering reasons, it would be de-
sirable to know the most likely heights and weights one might expect
among the people who do the most flying. Figures XII-2 and XII-3 show
THE PASSENGER CABIN
207
to as
1 XTv^v
Figure XII-2. Various seating arrangements across the aisle for passenger airplanes.
various seating arrangements and measurements. But just as one would
not expect the average bed to accommodate comfortably a man 7 feet tall,
so one should not expect the cabin of a light airplane to provide headroom,
however measured, for the exceptionally tall person. The average dimen-
sions for a man 6 feet tall were assumed and all discussion hereafter will
be based upon this size.
Headroom for Small Airplanes
The first part of the problem resolves itself into consideration of head-
THE PASSENGER CABIN
209
bers either while moving about or in sudden jolts as in a rough landing or
in gusty air, a few more inches should be allowed (certainly not less than
2 inches) so that the total headroom required is about 39 to 40 inches.
This headroom could be reduced if the occupant of a seat could maneuver
into position either by crouching down and crawling in, or by sliding in
from the side. If the back of the seat were inclined so that the occupant
could recline as much as 45 degrees, a saving in headroom of 30 per cent
could be obtained, or for the figures cited, a total of 12 inches.
If there is an overhead canopy or opening, the passenger could drop into
his seat, and after adjusting himself into position, he could close the over-
head cover.
It is questionable whether passengers would care to remain for any great
length of time in such a reclined position, but such a position should cer-
tainly not be ruled out.
But even if one were to toy with the idea of reducing the headroom by
reclining the occupants, one must not forget that the pilot cannot conven-
iently operate the airplane in a reclined position. He must be able to sit
upright at times, at least. In order to provide the necessary headroom, a
special canopy could be designed to enclose the cockpit portion of the
airplane. While the frontal area of the fuselage might not be reduced
materially, still such a solution would offer less skin friction, and permit
better streamlining than if equal headroom had to be allowed for the pas-
sengers.
A progressive inclining of the backs of the seats rearward as shown in
Figure XI1-4 might be considered, where a number of passengers may be
seated one behind the other, in order to permit a gradual reduction in the
fuselage depth toward the tail of the airplane.
Headroom for Large Airplanes
For large airplanes, where one of the prime criteria for design is passen-
ger comfort, headroom for a standing person should be required. The
minimum headroom for a standing person would be 2 inches, so that the
distance between the floor and the ceiling in the aisle for a 6-foot man
would be 74 inches. Six inches would be more desirable. For those
seated next to the wall, the headroom over the seats may be reduced to a
value perhaps halfway between that provided for in the aisle and that
provided for the seated person.
Leg Room
In addition to the headroom as previously denned, there would be added
an amount to allow for the placement of the legs and feet. For every inch
increase in the height of the seat from the floor, there would be a corre-
sponding increase in the height between the ceiling and the floor.
THE PASSENGER CABIN
211
could be closed up about 33 inches; that is, two passengers would require
only 51 inches in length instead of 84 inches when the legs are kept together
due to the fact that the seat position of each person is some 8 inches less
when the legs are spread about 70 degrees apart, in addition to permitting
the legs of one man to overlap the seated position of the man in front of
him by some 17 inches. With the passengers seated thus, the feet would
be some 48 inches apart, so that the width of the fuselage would have to
have a minimum interior width of 48 inches. Therefore, what was ob-
tained in reducing the required length was gained somewhat at the expense
of the width required.
If elbow room is allowed each passenger, then a width of 24 to 26 inches
Fig. XII-6 Fig. XII-7
Figure XII-6. With the pilot seated normally, length of cabin space could be re-
duced by the passengers sitting on the floor as indicated, more or less toboggan fashion.
Figure X [1-7. Normal side-by-side seating.
is desirable. Seating passengers side by side, with no aisle width, would
require a width of from 48 to 52 inches, just about what would be required
for the toboggan method of seating. On the other hand, the width of the
fuselage could be kept to a minimum interior width of about 26 inches if
the wing root joining the fuselage at floor level had sufficient depth to pro-
vide the foot room for the "toboggan" seating arrangement.
In the three seating arrangements just considered, the seat level was
assumed to be the floor level so that the minimum distance between the
ceiling and the floor was somewhat less than 40 inches. Suppose the seat
level were raised. Then the possible available leg room would lie some-
what between the two extremes of having the legs and thighs extending
horizontally out from the body and that of having the lower legs perpen-
212
dicular to, and the feet in contact with, the floor. Two configurations
are possible for this seated position: (1) with the knees of one passenger
touching the back of the scat of the passenger ahead of him; and, (2) with
the knees spread apart, so that the passenger in back would straddle the
seat of the passenger in front. In both cases the height between the ceil-
ing and the floor would be a minimum of 59 inches. If the knees of one
touched the back of the seat of the other, a bare minimum of 24-inch spac-
ing would be required for each passenger; the passengers could be brought
together another 7 inches for the "pick-a-back" seating disposition and
therefore would require an average of about 17 inches for each man.
This latter arrangement might be considered for a combination of pilot
and one passenger, or instructor-pilot and student arrangement, where
communication between the two could be improved and whereby a com-
paratively small airplane, especially in so far as the length of the fuselage
is considered, could be designed.
Both of these arrangements do not provide the most comfortable posi-
tions for the passengers. If the legs are stretched out, then more room is
required. With the leg at a 45-degree angle, and the thigh horizontal or
nearly so, then two men would require 69 inches spacing and each passen-
ger after that an additional 28 inches.
Again, moving the passengers 7 inches forward by spreading the legs
apart would reduce the 28-inch spacing to about 21-inch spacing for each
passenger after the first one.
Figure XII-8. "Pick-a-back" seating arrangement.
Extending the legs out at about a 45-degree angle permits a lowering of
the seat level of about 63^ inches and, therefore, brings the ceiling height
down a corresponding amount.

THE PASSENGER CABIN
213
Side-by-Side Arrangements
In all these seating dispositions, a single-row tandem arrangement was
assumed side-by-side. The main value of doubling the number across
the width of the airplane is that the weight of
the passengers is kept near the center of gravity
of the airplane as a whole so that the condi-
tions of fully loaded and partially or all empty
do not affect the center of gravity location
materially. If the passengers are seated
shoulder-to-shoulder, a minimum interior
width of 52 inches would be required, with some
reduction if a sacrifice in comfort were made.
A possible alternative to the side-by-side
seating would be to reverse the position of one
of the passengers with one looking forward and
the other rearward. The allowance for the
width of the seat would be between 18 to 20
inches for each person with normal clothing,
so that a minimum interior width of about
46 inches (allowing an extra 6 inches for the
shoulders) might be obtained, or a reduction
of about 6 inches; perhaps not much in mag-
FiGuwt XII-9. Staggered nitude but still important enough to consider
seat arrangement. in some designs.
Another side-by-side arrangement might be
considered with the seating extending the length of the fuselage instead
of across the width. Allowing 26 inches for each person, a reduction of
2 inches might be obtained over that of the usual tandem arrangement.
However, the legs and feet have to be placed somewhere, with the result
that a minimum interior width from about 30 to 48 inches is required,
depending upon whether the legs are kept vertical or extended to some
Staggered Arrangements
A variation of the side-by-side arrangements is the staggered placement
of the seats, either across the width of the airplane or lengthwise. (See
Figure XII-9.) There is some reduction in the dimensions required but

not at the expense of comfort. As a matter of fact there may be greater
freedom of movement since there is freedom for the legs and arms in the
single-row arrangement.
Back-to-Back Arrangements
Instead of the "pick-a-back" suggestion, another alternative for keeping

214
the weight of the pilot and one passenger close together would be a possible
back-to-back arrangement, where each person would require a minimum
of 41 inches for the normal seating position, but where the centers of grav-
ity of the two would be about 25 inches apart instead of about 28 inches,
as in the tandem arrangement. Not much would be gained by such an
Figure XII-10. Back-to-back seat arrangement.
arrangement; however, for a two-seater, a back-to-back arrangement
would require sufficient headroom or a shorter length of the fuselage, so
that a better streamlining of the fuselage may be possible. See arrange-
ment in Figure XII-10.
Seating Facing Rearward
Recent investigations indicate that seating the passengers so that they
face rearward would be desirable as far as safety in even a minor crash is
concerned, since the acceleration of the passengers is forward in such a
case. If they were facing forward, the impact would be on the relatively
small area of the safety belts, or if no safety belts were used at the time,
they would be catapulted out of the seat. But, if the passengers were
seated the other way, the pressure would be against the back of the seats,
a very much greater area, and there would be relatively little danger of
being catapulted out of the seat unless there were some acceleration in a
vertical direction relative to the seat.
While there may be many people who object to riding backwards in sur-
face transportation, there should be no pressing reason other than custom
why this objection should hold in aircraft seating. The accelerations in
flight are normally not of the same order as in an automobile or a train,
nor of the same frequency of occurrence. Moreover, since flying is done
at relatively high altitudes (at least relative to the height of objects on the
ground) the same sensations of objects passing past the window do not pre-
vail. There might be objections to reversing the usual seating arrange-
ments when taking off or landing and in climbing or gliding steeply.
In trying to obtain minimum frontal area, care should be taken that the
passengers are not enclosed in the structure like mummies, for such designs
THE PASSENGER CABIN
215
build up sales resistance not only because of a feeling of claustrophobia but
also because of minor accidents that will constantly occur in bumping one's
Figure XII-11. Normal tandem seating arrangement.
head against the structure. Normal tandem seating is shown in Figure
XII-11.
VISION
In any of the tandem arrangements where the passenger faces either
forward or rearward, there should be no great difficulty in providing win-
dows for looking out, but where the passengers are seated athwartship,
the individual either has to look through the window on the opposite side
or has to crane his head to look out of a window in back of him. For large
airplanes, vision considerations are not so important as for the small air-
craft. (See Chapter X, The Cockpit.)
SEATING COMFORT
The degree of discomfort that may be tolerated is intimately tied up
with duration. A passenger may be willing to accept some discomfort for
a quarter of an hour, whereas he would be decidedly unhappy if he had to
be uncomfortable for an hour.
In general, any seating on the floor would not be acceptable in an air-
plane which stayed in the air any great length of time. It would be ques-
tionable whether such a position on the floor would be liked by passengers
in a highly maneuverable airplane subjected to even mild aerobatics.
The "pick-a-back" arrangement in a personal airplane would be decid-
edly disliked by a man who did not like someone breathing down his neck.
It would be permissible for a training airplane and perhaps moderately
acceptable in a normal single-seater personal airplane which needed pro-
vision for the occasional passenger.
SEATING ACCESSIBILITY
One of the greatest problems in trying to crowd several passengers into
a small place is the matter of getting each passenger settled in his place
216
with the least fuss and bother. Crowded seating arrangements require
jig-saw handling in that usually a particular order in filling and emptying
the seats is required, especially where the number of seats exceed two or
three. The more passengers the airplane is to hold and the more crowded
the seating is, the larger the door should be for getting into and out of the
airplane. (See also section on Doors and Exits.)
CENTER OF GRAVITY CONSIDERATIONS
The closer the passengers can be brought together, the less will be the
effect upon the center of gravity between the fully loaded, partially empty,
or empty condition.
h 26" H
Fig. XII-12 Fig. XII-13
Figure XII-12. Average dimensions for seating when the leg is vertical.
Figure XII-13. Average dimensions for seating when the lower leg is 45 degrees ts
the vertical.
The single-engine airplane with the engine in the nose of the airplane
will require closer spacing for the tandem arranged seats since this design
usually has the center of gravity so placed that the variable load (the pas-
sengers) is behind the center of gravity.
WINDOWS
A window should be located at each seat with the bottom of the window
at about shoulder level of the passenger when seated. The window should
also be slightly forward, that is, the right side of the window as one faces
218
In addition to this door, closed aircraft cabins carrying more than 5 pas-
sengers, but not in excess of 15, require at least one emergency exit whose
minimum dimensions are tangent to a 19 X 26 ellipse, on the side opposite
the main door. When more than 15 passengers are carried, an emergency
exit for every additional 7 passengers should be provided. Not more than
4 exits are necessary, and they may be located either in the top or side.
Doors may replace exits if desired. (See also section on Accessibility.)
SLEEPING ACCOMMODATIONS
Separate compartments are desirable when sleeping accommodations
are to be provided.
Berths vary in width from 28 to 32 inches with 30 inches a good average
value. The length of the berth varies from 72 to 76 inches with a prefer-
ence shown for the higher value.
Mattresses, blankets, pillows, and bed linen for a berth weigh from 12
to 18 pounds.
LIGHTING
Individual lights usually placed slightly overhead are provided for each
passenger. They should be subject to control by the passenger but a mas-
ter switch should be located in the pilot's cockpit so that the pilot may
shut off the light when necessary.
FURNISHINGS
The walls of the cabin are usually treated for soundproofing. The in-
terior may be finished in Fabrikoid, painted and doped aircraft cotton, or
some suitable upholstering material.
Inspection panels (in the case of fabrics, Talon fasteners are suitable)
must be provided to permit ready access to the structure underneath for
inspection purposes.
Curtains for windows should be installed so as to permit their removal
and replacement in one minute or less.
For larger transport airplanes, provision for attachment for small tables
for playing bridge or holding a typewriter should be made.
Since many trips are long, any device used to entertain the passengers
will be welcome. Airspeed indicators, altimeter, and probably a compass
located on the front wall of the cabin will draw attention. An illustrated
map of the country being traversed will add to the general interest of the
trip.
PROVISION FOR AIRSICKNESS

Paper bags are usually found in a pocket attached to the back of the
seat. Washroom facilities are provided, of course. The steward or stew-
ardess is also prepared to take care of such cases.

THE PASSENGER CABIN
219
FLOORING
The cabin flooring varies in treatment in various designs. It should be
light and soundproof, as well as sound and vibration damped. A free-
floating flooring made possible by the use of rubber or felt between the
floor and the primary structure is desirable.
A thickness of balsa wood between thin-face plies of birch or aluminum
alloy makes a very effective flooring which will take considerable local
loads. The floor may be further treated with a thin cork tile or "battle-
ship" linoleum for appearance and wear.
Floor panels should be provided with means for quick and easy access
to the structure below for inspection purposes.
No obstructions such as spars should extend above the floor level. It is
customary to rest the floor on the spars. Inasmuch as the airplane may
take any flight attitude, it is difficult to state what constitutes a "level"
floor. A good method is to select the angle at which the cruising speed of
the airplane occurs and make the floor level for this condition. A varia-
tion of a few degrees will not be serious.
Passenger airplanes are equipped with toilet and washroom facilities in
a compartment separated from, and usually in the rear of, the cabin.
The toilet (see Figure XlI-15) consists of a metal container with a water-
proof liner of sufficient capacity with a suitable chemical preparation. It
should be simply designed and easily cleaned.
Wash basins are small but useful. The water supply should be placed
in a tank so located that it can be filled readily from the outside, but should
be protected against freezing.
TOILETS
Figure XII-15. Toilet seating.

220
A water allowance of 2 to 3 pounds per passenger, depending upon the
range of travel, is ample for both drinking and washing purposes.
REFRESHMENTS
Since flying schedules do not permit stops at intermediate landing fields,
it is customary to provide passengers with a light lunch, often made up in
boxes. A weight allowance of to 3^ pound per person is ample for the
food, container, paper napkins, etc. On larger and newer equipment of
passenger airlines, a greater allowance may be made to permit furnishing
light, hot lunches in flight.
BAGGAGE COMPARTMENT
Since only coats and hats which the passengers are wearing are permitted
in the cabin, suitable baggage compartments must be provided. These
compartments are not accessible in flight.
To expedite the removal of luggage, mail, or express, these baggage com-
partments must be accessible from the outside by means of separate bag-
gage doors.
Usually, two or more baggage compartments are provided in the fuselage
to permit the limiting of center of gravity movement for the various load-
ing conditions possible. The baggage door carries a schedule giving the
load capacities of each baggage compartment and for fully loaded and
partially empty conditions of the airplane.
Luggage varies in size so that it is difficult to recommend exact cubical
capacities of baggage compartments. A standard suitcase measures
9 X 17 X 31 inches.
Mail compartments should allow 12J^ cubic feet for each 200 pounds of
mail.
CHAPTER XIII
Air Conditioning
The importance of heating in an airplane cabin for aircraft operating
at high subsonic speeds can be gauged easily when it is realized that the
temperature of the outside air decreases rapidly with altitude. When it
is some 60Â° F on the ground, it may be â€” 50Â° F at 30,000 feet. In these
days of substratosphere and stratosphere flying, the change in tempera-
ture that an airplane encounters from the moment it leaves the ground
until it reaches its required altitude only a few moments later is enormous.
A difference in temperature may also be experienced as the airplane flies
from a warm country to a cold country. While this change may not occur
in a few minutes as in the case above, it does occur during the same flight
so that a heating system has to be provided whether flying is during the
summer months or not.
It is not the heating alone, however, that has to be considered, but the
ventilation as well.
PHYSIOLOGICAL AND PSYCHOLOGICAL CONSIDERATIONS
The objects of ventilation are of course:
1. To supply oxygen,
2. To remove odors,
3. To remove toxic or poisonous substances,
4. To remove body heat, moisture, and heat from other sources.
At present there are no ready means available for controlling the hu-
midity of the air. It is generally believed that the relative humidity should
be about 40 per cent for good health conditions. However, the movement
of the air has some effect. In still air, the body is enveloped in a layer of
moist air so that even a moderate air movement tends to create the effect
of a draft.
Care should be taken to avoid possible overheating of the cabin. The
general effects of overheating may be summarized as follows:
1. Increased heart beat and blood flow.
2. Increased respiration.
3. Increased sweating with attack of cramps.
4. General lassitude and dizziness.
221
222
Poisonous ozone, which injures a man's lungs, corrodes metal, and rapidly
rots rubber, adds to the hazards of flying at altitudes from 80,000 to 95,000
feet. At higher and lower altitudes the ozone concentration is so low as
to be unimportant. As altitudes approach 120,000 feet, cosmic rays in
the atmosphere may become a factor in human flight. The heavy nuclei
of these rays can penetrate through a man without seriously injuring him,
but an unlucky hit on a critical nerve center could be serious.
Table XIII-1. Human requirements and performance limitations at various altitudes.
(Courtesy of Flight Safely Foundation, Inc.)
5,000 feetâ€”Maximum for normal night vision without supplemental oxygen.
8,000 feetâ€”Altitude at which supplemental oxygen should be used for routine flights.
10,000 feetâ€”Maximum without routine use of oxygen.
18,000 feetâ€”Maximum for emergency without use of oxygen.
20,000 feetâ€”Altitude at which consideration should be given to use of pressurized cabins.
23,000 feetâ€”Altitude at which there is evidence of depressurization sickness.
25,000 feetâ€”Approximate time of consciousness without oxygen is 116 seconds.
28,000 feetâ€”Maximum to avoid depressurization sickness. Approximate time of con-
sciousness without oxygen is 69 seconds.
30,000 feetâ€”Altitude above which slight positive pressure breathing should supplement
demand oxygen to avoid air leaks into oxygen mask. Approximate time
of consciousness without oxygen is 54 seconds.
35,000 feetâ€”Maximum for routine use of demand oxygen system. Approximate time
of consciousness without oxygen is 32 seconds.
40,000 feetâ€”Approximate time of consciousness without oxygen is 23 seconds.
42,000 feetâ€”Maximum for routine use of pressure breathing.
43,000 feetâ€”Maximum for emergency use of demand oxygen.
45,000 feetâ€”Bombardment aircraft having combat ceilings above this altitude require
provisions for use of partial pressure suit.
48,000 feetâ€”Fighter aircraft having combat ceilings above this altitude require pro-
visions for use of partial pressure suit.
50,000 feetâ€”Maximum for emergency use of pressure breathing demand oxygen.
AIR MOVEMENT FOR COMFORT
The air movement in the cabin should be such as to avoid the feeling of
drafts. To this end it is desirable, although not always possible, to meet
the following conditions:

1. Limit the maximum air velocity to about 2 feet per second for persons
at rest.
2. High-velocity currents of air should not be directed downward on
the occupants, nor strike them from behind at about neck level, nor come
up at about floor level.
3. The air should not be admitted to the cabin below the normal cabin
temperature.
The natural movement of the air within the cabin is usually from rear
to front. However, for various reasons it is desirable to reverse this flow.

AIR CONDITIONING
223
Figure XIII-1. A schematic diagram of the ventilating system of a flying boat
(Grumman Mallard). Fresh air is forced by ram pressure while in flight and by a
blower while on the ground. The heater (A) heats the air which is uniformly dis-
tributed through floor ducts, to windshield, and pilot's cockpit.
Normally, this reversal is accomplished easily by locating the exhaust at
the top rear and thus, due to the forced draft produced by the incoming
air and forced exhausting system usually provided, the air flow can be
easily controlled. Figure XIII-1 shows a schematic diagram of a venti-
lating system.
The circulation of air in a supercharged cabin is a problem not only of
maintaining a predetermined pressure but also of heating or cooling, and
ventilating. For hot weather, a separate cold-air system may be in-
corporated, although a by-pass for the heating unit in the system would
still permit use of the blower system as a pressure ventilator. The fresh
air is brought in through a duct, passed through a centrifugal rain separator
and then into the cabin when the heating system is by-passed; otherwise
the air is passed through a heating system, as described elsewhere, before
being carried into the cabin. For high speeds, where heating due to high
ram pressures appreciably raises the temperature of incoming air, cooling
has to be provided.
The toilet room should be made as airtight as possible, and a very posi-
tive system of ventilating should be provided for this room independent of
the cabin.
It has been found that cool air is most effective in keeping a passenger
comfortable in hot weather if an individual ventilator is located above and
in front of the seat.
All incoming air should be controlled by the pilot at its source. Win-
dows should not be used for ventilating purposes.
AIR REQUIREMENTS
For a uniform CO2 content in the air to be breathed, about 1 cubic foot
of air per passenger per minute is necessary, although 5 cubic feet per pas-
senger per minute is desirable when the air is recirculated, while 10 cubic
feet per passenger per minute without recirculation is ideal.

224
A person breathes on the average about 16 times per minute while at
rest on the ground for a total of 1J^ to 4 cubic feet per minute. This
amount would be increased, depending upon the activities of the individual
and the altitude at which he was working. To provide adequate ventila-
tion, the needs of each person in the cabin would have to be considered.
Normal air contains about 21 per cent of oxygen by volume, and 4 per cent
is removed by a person each time he breathes, so that the air could be
breathed about twice before it would reach the 12 per cent minimum
desirable. Recirculation of the air would, therefore, be possible provided
that odors can be removed and the necessary sanitary precautions can be
observed.
Aircraft cabins provide from 10 to 40 cubic feet per person per minute,
depending upon the individual aircraft, as well as the means employed in
circulating the air within the cabin.
Since the volume available per passenger in an aircraft cabin probably
varies from 40 to 60 cubic feet, regardless of the length of time the passen-
ger remains in the cabin, it is necessary to bring air into the cabin con-
stantly to replenish the supply. Some air will filter through seams and
cracks, but such air will have the temperature of the outside air which may
be very cold. It is necessary therefore to be able to bring air in from one
central inlet under control, and to remove it by suitable and positive
means.
This change of air every few minutes complicates the heating problem.
In a room there are steam radiators which will heat the air, and since the
air is removed very slowly by natural infiltration outward through win-
dows, doors, and cracks in the walls, it is a comparatively easy matter to
maintain a constant temperature. In an aircraft cabin, steam radiators
would be useless unless the walls, ceiling, and floors were radiators. It is
therefore necessary to heat the incoming air before it actually enters the
cabin.
PRESSURE CONSIDERATIONS
Both the oxygen supply and air pressure must be increased for high-alti-
tude flying. Above 35,000 to 40,000 feet, a mask supplying pure oxygen
is required, and the outside pressure must be increased to provide enough
lung pressure to permit breathing. At these extreme altitudes, pressure-

equipped or pressurized cabins are the most feasible solution for large com-
mercial airplanes. For small planes, pressurized suits and helmets are
necessary.
It has been found, experimentally, that if his oxygen supply is cut off at
20,000 feet, a man will pass out in 10 minutes; at 25,000 feet in 3 minutes.
Pressure is necessary to guard against "aeroembolism," the equivalent
of "diver's bends."
AIR CONDITIONING
225
No attempt is made to keep the cabin at sea-level pressure at all times.
In one installation, pressures vary up to the 8,000-foot level in accordance
with the usual atmospheric pressure relationships. From 8,000 to 14,700
feet the pressure within the cabin is maintained at the level normally exist-
ing at 8,000 feet altitude. Beyond 14,700 feet, a differential of 2.5 pounds
per square inch between the outside and the inside of the cabin is main-
tained with the result that the apparent altitude at 16,000 feet altitude is
about 9,000 feet; at 18,000 feet it is 10,600: and at 20,000 feet it is approxi-
mately 12,300 feet. (See Figure XIII-2.)
Indicated airplane altitude in thousands of feet
Figure XIII-2. Cabin pressure altitudes maintained by various pressure differ-
entials AP. The normal operating regime for pressurized cabins is from 6,000 to
12,000 feet equivalent altitudes.
Without supercharging, an altitude of 14,000 feet is assumed to be the
upper level for normal but limited commercial passenger flight.
Design for differential pressure depends upon structural limitations as
well as other considerations. If a pressure differential of 6.5 pounds per
square inch is used as the basis for structural design, and a safety factor of
2 is to be used, then the cabin would normally not be operated at pressures
exceeding 3.25 pounds per square inch.
The oxygen content of the air is the same up to usual altitudes at which
flights are made so that the air under pressure will have the necessary
oxygen supply.
226
DESIGN TEMPERATURE LIMITS
The temperature limits that the designer has to consider in dealing with
heating, cooling, and ventilating equipment are:
1. 100Â° F for human endurance at 50% humidity;
2. from 125Â° to 130Â° F for the engine accessory compartment;
from 150Â° to 160Â° F for hydraulic equipment;
4. from 175Â° to 185Â° F for electrical and electronic equipment.
PHYSICAL CONDITIONS
Summer temperatures greater than 100Â° F are seldom encountered even
in the warmest climates. Heating and ventilating systems have not yet
incorporated systems for cooling the
air except the cooling of cabins of
aircraft by means of an outside re-
frigerating system before take-off.
This cooling of the cabin disappears
within a short time after take-off.
By proper ventilation, moderate
temperatures above normal room
temperatures of 72Â° F are not un-
comfortable.
At the lower-temperature limit
the heating system must be designed
to keep both the passenger cabin and
the pilot's cockpit warm at a temper-
ature of 70Â° F when the outside
temperature is about â€” 20Â°, F. When
the airplane is used near the Arctic
or in other cold regions, the heating
system should be designed for air en-
tering at a temperature of â€”40Â° F.
The heating system should be designed to take care of these extremes in
temperature within five minutes after the engine is started, for warming
up, and must be readily controllable to take care of sudden changes in out-
side air temperatures encountered either in climbing or in passing from one
locality to another.
The incoming air should not be lower in temperature than that main-
tained in the cabin.

In order to regulate the temperature reasonably it is necessary, at times,
to deflect a sufficient amount of the incoming air from entering the cabin
by means of a diffusing or butterfly valve operated either by an electrically
driven gear box through a thermostat, or by manual operation. Either
method has a certain amount of time lag that is not possible to overcome.
Figure XIII-3. A cross section of
cabin showing the use of radiant heating for
keeping the interior uniformly heated.
A is the supply duct for the heated air,
and B is the exhaust duct.

AIR CONDITIONING
227
Water separation must also be provided in the system in case of rain in
the incoming air. Snow usually offers no problem since it is vaporized
easily in the heater system.
A dust separator is especially desirable when the airplane is to operate
from dusty airports.
As an additional means of keeping down temperatures of interiors, the
tops of fuselages are often painted white. For a normal skin temperature
of 152Â° F, painting the surface white reduces the temperature to 124Â° F,
with a possible reduction in cabin temperature of 15Â° F. White-painted
tops are most effective under high temperatures, zero wind conditions,
although some reduction can even be expected during flight. Zinc chromate
primer and du Pont White DeLux 83-508 is the treatment given the
surface.
AIR DUCTS
Air inlets should be so located as to be free from contamination of oil
fumes or other engine gases. Since the ducts should have as few bends as
possible to permit the normal air pressure in flight to force the air into the
cabin, the location of these inlet ports may offer difficulties.
In a single-engine plane, a duct
carried between the cowling and the
engine ahead of the collector ring
has been found satisfactory.
For a central fuselage having no
engine in its nose, the foremost point
on the fuselage is an excellent loca-
tion for this inlet port.
The leading edge of a wing in the
pressure region and within the pro-
peller slipstream, if possible, but
well outside the region of possible
contamination from oil or exhaust
gases, has also been found suitable.
The inlet ducts for leading the air
to the heater vary in cross section
from 3 to 5 inches in diameter. It

must be remembered that all open-
ings of this kind add to the aero-

dynamic resistance so that choice of
cross section should be guided by
the type of airplane; the larger the
craft, the larger these openings may
be permitted.
The ducts leading to the heater
Cabin wall
Filter
rs
Conduit
tori":::::::.
â€” ^
1 Floor
1 Fan
Louvre
nun
Figure XIII-4. A cross section of an
inlet air duct showing treatment for sound-
and dust-proofing. Filter has fine-mesh
screen. Conduits made of doped fabric,
lined with felt, held in place with coarse-
mesh aluminum screen. Individual fans
are located in each outlet. Louvres are
controllable for regulating volume of air.

228
are made either of aluminum, aluminum alloy, or stainless steel of a mini-
mum thickness of 0.025 inch.
The ducts leading away from the heater may be of metal, but usually
only for a short distance. The longer lengths of the ducts, after the metal
section, may be made of doped aircraft fabric. To prevent too rapid cool-
ing of the air while being transmitted to the cabin, and to aid in sound-
proofing, the fabric (or metal) ducts are lined with about 3^-inch thick felt
held in place by a light-mesh aluminum alloy screen. The outside diam-
eters of these ducts are about the same as the inlet ducts, but their inside
diameters become less when the felt has been applied.
Individual ducts are led to each seat. Since the amount of air leaving
each opening should be the same, it may be necessary to vary the opening
by inserting obstructions. Once the airflow has been regulated for each
opening, no changes need be made.
For warm weather, when the air at existing temperature is brought into
the cabin, relief for the passengers may best be obtained by diverting air
into an overhead duct in the ceiling of the cabin and forcing the air through
overhead grills and individual ventilators. At other times, it is more de-
sirable to have the air enter the cabin along the sides near floor level.
n ii
Figure XIII-5. Cross section of a Figure XIII-6. Cross section through
manually-operated swiveling nozzle to a conventional air scoop for auxiliary air
regulate individual cold-air supply. supply.
The air is taken out of the cabin through outlets located in the floor
although they may be located along the roof and exhaust near the top rear.
In some cases it may be more desirable to exhaust at intervals along the
side of the fuselage; however, each additional opening adds to the sound-
proofing problem.
A separate exhaust system should be provided for the toilet.

AIR CONDITIONING
229
The pressure of the incoming air and the slight suction that is usually
present on the top surface of the fuselage are enough to force exhaust of
the air. If the rate of exhaust is not fast enough, an exhausting fan may
be necessary.
A grill-work made of fine mesh wire should cover each opening so that
nothing may fall in the duct.
BOILERS AND RADIATORS
The safest type of system is one in which the exhaust gases heat a liquid
that is vaporized or circulated through a radiator. The air is forced
through the radiator and then led into the cabin. It is not feasible to
supply actual radiators for each passenger in the cabin since the volume of
air brought in must be heated quickly and effectively.
The liquids used in the boilers may be commercial ethylene glycol which
boils at 310Â° F to 340Â° F, specific heat of about 0.70; quenching oils of high
boiling points; or a mixture of water and alcohol. The requirements for
such liquids are that they boil at about 300Â° F, do not freeze or congeal at
the lowest temperature likely to be encountered, and have a flashpoint
high enough so as not to offer a fire hazard in case of a leak.
The radiators normally used are the cartridge type, or similar to the oil
coolers. It is desirable that the air be in contact with the radiator suffi-
ciently long to be heated, that the heat transfer of the radiator be very
rapid, and that the airflow speed be maintained. (See also discussion
under Calculations.) â€¢
HEAT SOURCES
The exhaust gases of the engine supply the main source for moderate-
size airplanes for heating the air, whether directly or through intermediate
means. In the case of the liquid-cooled engines, auxiliary radiators may
be used for heating the air. In any case, it is desirable to have a very hot
source if a light, efficient, and instantaneous system is to be obtained.
An aircraft engine operating at 500 hp rejects about 1,500,000 Btu of
heat per hour through the exhaust.
Because gas turbine engines provide great quantities of hot air, com-
bustion heaters are not used where such a hot air supply is available.
Electric "blankets" for the cabin side walls and certain areas of the
floor, and electric heaters in the cabin air-ventilation ducts may be used
for cabin heating systems in large airplanes.
PRESSURE CABIN EQUIPMENT

The equipment for pressurizing cabins is still undergoing extensive re-
search and design. In general, the necessary equipment for maintaining
passenger cabins at a desired pressure differential consists of:

230
1. A centrifugal air compressor, usually mounted on the rear of the fire
wall, driven either by a flexible shaft connected to the generator drive of
the engine, or by auxiliary power. This compressor incorporates an inte-
gral gear box and usually a weight-flow control as well as a valve permit-
ting the maintenance of an internal cabin pressure higher than the external
pressure.
2. A blower relief or surge valve, located in the air duct to the cabin. If
the pressure in the duct builds up due to a throttling down of the control
valve unit, the air is by-passed to the outside atmosphere and so prevents
surging or overheating of the blower. Such procedure would also be de-
sirable at low altitudes and in warm weather, although the blower may
also operate at low altitudes where it acts as a ventilating fan.
3. A supercharging control unit incorporating the air-inflow valves and
the spent-air discharge valves. The cabin pressure is regulated by chang-
ing the opening of the outlet valve. This valve is controlled by pressure-
sensitive mechanisms responsive to absolute pressure and to differential
pressure.
4. A valve opening inward which will not permit higher pressures to
exist outside the cabin than inside.
CALCULATIONS
A typical example will be given here to indicate the type of calculations
that may be made.
Temperature of incoming air â€”20Â° F
Temperature to be maintained 70Â° F
Number of passengers to be supplied with air 10
Volume of air per passenger to be supplied 25 cu ft per min
Volume of cabin 400 cu ft
Ethylene glycol for liquid-type system.
The number of changes required will be (10 X 25)/400 per minute, or
one change every 1.6 minutes.
Assuming 10 separate outlets, one for each passenger for the incoming
air, then air flows through each duct at the rate of 25 cubic feet per minute.
Restricting the air velocity to 2 feet per second, the area <S of each inlet
duct would be
2S
25
60'
25
120'
or
d = 0.515 feet, approximately 6.18 inches.

AIR CONDITIONING
231
For relatively small airplanes, such as the problem cited, the practical
diameter of the inlet duct to the cabin would be about 3 inches. To re-
duce the velocity of the airflow, it would be desirable to increase the num-
ber of openings. The duct leading up to the actual outlet into the cabin
is often larger and is provided with a valve for regulating the airflow.
The heat balance equation may be written as
Q = Qi + Q* - Qs,
where Q = heat required to raise 250 cubic feet of air per minute from
â€”20Â° F to an average of 70Â° F within the cabin, and
Qi = heat required to raise incoming air to maintain temperature
in cabin at 70Â° F, if there were no losses; then
Qi - WC,(h - <i),
where W â€” weight of air,
C,, = specific heat of air = 0.2375 at 70Â° F,
h = temperature of incoming air,
<i = outside temperature,
Q2 = heat lost by radiation from cabin through windows, fabric,
flooring, and the like,
Q3 = heat generated by the passengers. (This is usually neglected.)
The heat lost by radiation is difficult to ascertain accurately. With
well soundproofed and hermetically sealed cabins, this loss may be quite
negligible. A loss of 100 Btu per minute will be assumed.
Weight of 250 cu ft of air at 70Â° F = 250 X 0.076 = 19 lb,
Qi = 19(0.2375) (70 + 20) = 406 Btu/min,
Q2 = 100 Btu/min,
Q = 406 + 100 = 506 Btu/min to be supplied.
The temperature of the incoming air would then be
506 = 19(0.2375) (f, + 20),
U = 95Â° F, approximately.
This relatively high temperature indicates that the heat lost due to
radiation is high in proportion to the heat supplied. This heat loss should
be reduced if possible. If this is not practical, then the incoming heated
air should be so diffused that no passenger is located in a direct blast. In
cases where the incoming air is too hot, a compromise in the temperature
level to be maintained is desirable. By lowering the temperature level

within the cabin, and directing the warm air at the passenger at about
knee level, a reasonable degree of comfort may be obtained.

When a liquid system is used, a pump is required to circulate the liquid.
The capacity of the pump can be calculated from:
Q = WCp(t2 - h).
232
Assume a 50-degree drop through the radiator. The quantity of ethy-
lene glycol (Prestone) to be circulated is then
506 = TT(0.7) (300 - 250Â°),
W = 14.4 lb/min.
The weight of the liquid at 300Â° F is 9.3 pounds per gallon. Therefore,
the capacity of the pump would be 14.4/9.3 = 1.55 gallons per minute.
(See also discussion under Air Ducts.)
HEATING SURFACE
The heating surface required for the boiler can be calculated from the
formula:
Skdt = total heat to be supplied = Q,
where S = heating surface in square feet required,
k = mean coefficient of heat transmission,
dt = mean temperature difference between the liquid and the
exhaust gas.
In the last problem, 506 Btu were to be supplied per minute. The mean
coefficient of heat transmission may be obtained experimentally. For a
first approximation some value between 10 and 15 Btu per square foot per
hour may be assumed.
m - ta ~ U
~ log.tfo/*,)'
where to = the initial temperature difference between the liquid and the
gases,
tf = the final temperature difference between the liquid and the gas.
Assuming the initial temperature differences to be 1000Â° F, and the final
temperature difference between the liquid and the gases to be 850Â° F:
_ 1000 - 850 150 =
log. (1000/850) 0.162 ~ yZ0 r'
Therefore
S = ^ = (12 X5925)/60 = 20073 ft2,Say3-00ft0
If the exhaust pipe used for heating the liquid were 4 inches in diameter
the heater would have to be (3 X 144)/47r inches or about 35 inches long.
In a similar manner, the heat transfer surface of the radiator may be
calculated. The liquid temperatures and the air temperatures would be

used in the latter case in the above formulas, and calculations continued
as before.
AIR CONDITIONING
233
AIR CONDITIONING PROBLEMS AT HIGH SPEEDS
As aircraft speeds increase, heat is the inevitable result. Just below
the speed of sound, the temperature of the pilot's compartment may
reach 150Â° Fâ€”without air conditioningâ€”when the outside air tempera-
ture is zero. At Mach 2 (about 1600 mph) it may reach 280Â° F; at Mach
5, it is probably likely to reach 2000Â° F.
700
600
500
400
300
200
100
0L
-100
Aircraft Skin Temperatures
At sea level
At 10,000 feet0
At 20,000 feet 0
At 30,000 feet
At 35,000 feet-
200 400 600 800 1000 1200 1400
True airspeed in miles per hour
1600 1800
Figure XIII-7. The aircraft skin temperature gives an indication of the interior
temperature likely to occur if no cooling is provided. To the heat at these temperatures
might be added the heat developed by the power plant, by the electrical equipment,
and by the human occupants pf the craft.
Air conditioning at these speeds requires more attention to cooling
problems than to heating problems. For the jet engine, various methods
are being explored and are still subject to further development and im-
provement.
For the airplane equipped with a jet engine, high-pressure air may be
taken from the engine at a temperature much higher than the "ram" air
around the aircraft. The ram air can thus be used as the all-important
"heat sink" to which heat can be conducted. A heat exchange similar
to an automobile radiator transfers the heat from the engine bleed air to
the ram air. This cooled high-pressure air is then expanded through a
234
special air-cycle turbine, where the temperature is further reduced to well
below the ram air temperature by giving up energy in the form of work.
This work energy is used by the turbine to drive a fan that pulls air
through the heat exchanger. The cold air from the turbine is then em-
ployed to cool the aircraft's cabin and its occupants.
Gas-cycle refrigeration using Freon as the gas (similar to household
refrigeration units) has been used for cooling systems in airplanes.
When airplanes start flying regularly at altitudes where there is virtually
no air, it will no longer be practicable to bleed air from the jet compressor
in the large quantities required for pressurization.
Perhaps the pilot will carry his own air and pressurization supplyâ€”
possibly using compressed air kept in specially-designed storage tanks
within the airplane's structure, or the liquid oxygen breathing system
as a source of pressurization. Cockpit air would be recirculated through
an air-purifying dessicant canister to remove water vapor and odors.
Compressor air would be needed for cooling only, thus greatly reducing
bleed requirements. The air would pass through an air-cycle turbine,
the exhaust of which would go through a heat exchanger and then be
dumped overboard.
Air conditioning experts believe vapor-cycle cooling systems will be
used in very high-speed aircraft, mainly as auxiliary coolers, absorbing
cabin heat and giving it up to a heat sinkâ€”water or fuel on its way to
the engines.
Future cooling possibilities include extensive boiling of water and
boiling of ammonia in various mixtures. Another possibility is the use of
liquid oxygenâ€”commonly called "Lox"â€”as a stored coolant. It is pro-
posed, in one system, to spray Lox directly into the cabin to provide not
only cooling, but a breathable atmosphere and pressure level as well.
Under consideration for advanced aircraft is the drastic solution of having
the outer portion and leading edge of the airframe and airfoils made of a
fusible material that will melt at extremely high temperatures and thus
absorb some of the heat generated.
Where the top speed and altitude portions of a flight are of short dura-
tion, insulation can do much to prevent heat from reaching the critical
areas of the aircraft. This can be combined with reflective coating on

the inside of the cabin to provide further protection. Such measures can
bring reductions in the total weight of the cooling system required. Even
slight weight savings are vital, since in some cases a pound of such equip-
ment removed may permit as much as 100 pounds' reduction in the total
weight of the aircraft.

CHAPTER XIV
Soundproofing
The increasing demand for more passenger comfort has brought sound-
proofing very much to the fore, and no airplane passenger cabin has been
correctly designed unless every effort has been made to reduce noise to the
practical minimum.
The problem of soundproofing an auditorium is entirely different from
that of soundproofing an airplane cabin because, in the former, the source
of noise is inside so that the noise has to be absorbed and materials must
be chosen that will not only absorb the noise but also avoid reverberation.
In an airplane cabin the greatest portion of the noise comes from outside
so that the transmission of sound must be overcome. A very thick wall
is an excellent soundproofing medium but weight is important in an air-
plane so that materials must be chosen which will weigh the least for the
greatest amount of noise reduction.
MEASURE OF NOISE
The measure of noise level is the decibel. In equation form it is ex-
pressed as:
, . energy input in watts
N = 10 logw -â€”fâ€”7-0 0 rr0
energy output in watts
The acoustical energy is transformed into electrical energy by some suit-
able device, generally the "noise meter" or the "decibel meter." Since
the decibel is a logarithmic relationship, it can be shown easily that if an
airplane engine makes a noise equal to 100 decibels, two such engines
would cause a noise level of only 103 decibels. Therefore, to reduce the
sound level from 100 decibels to 70 decibels, for example, requires very
careful soundproofing.
235
236
SOURCES OF NOISE
The primary sources of noise encountered in the airplane are:
1. Engine exhaust.
2. Propeller.
3. Moving mechanical parts, such as gears.
4. Structural vibration.
5. Wall vibration or "drumming."
6. Whistling sounds caused by interferences.
7. Clatter due to loose structure, machinery, etc.
8. Conversation.
The noises which soundproofing attempts to alleviate are, primarily,
engine-exhaust noises, propeller noises, and wall vibration. The first two
cause the greatest disturbances by far.
A comparison of the noise of geared and ungeared airplanes is shown
in Table XIV-1.
Table XIV-1. Comparison of noise of geared and ungeared engines.
Location Geared Ungeared
Propeller tip speed (ft/sec) 830 685
Propeller diameter (ft) 9.75 10.75
Pitch (ft) 5.7 8.7
Airspeed (mph) 90-100 90-95
Average noise level above threshold in plane of propellers (db) 107 91
Average noise level above threshold in rear of cabin 95 85
These tests indicate that the reducing of the propeller-tip speed helps
appreciably in noise reduction. A reduction of 100 feet per second in tip
speed is supposed to give as much as a 10-decibel reduction in noise level.
It is not always practical to reduce the propeller-tip speed unless it is
already in the critical region (very close to the velocity of sound) when a
reduction in propeller-tip speed would also increase the propeller efficiency
as well as reduce noise.
Mufflers help in reducing exhaust noises, but the level is usually not suffi-
ciently reduced because so many other noises are present. Moreover,
mufflers may seriously reduce the engine horsepower and are not economi-
cal from that viewpoint.
EFFECT OF FREQUENCY ON SOUNDPROOFING

Experimentation with various materials reveals that all are not equally
effective in reducing the incoming noise. Thus it is necessary first to

analyze the frequency of the incoming noises.

SOUNDPROOFING
237
Table XIV-2. Frequency distribution of noise.
Frequency,
Causes of noise cycles per second
Propeller unbalance range 20-25
Engine unbalance range 30-40
Propeller beating against fuselage 60-80
Exhaust and engine explosions and vibration transmitted through
structure 200-270
After the range of frequency has been determined, sufficient materials
may be chosen to form a composite blanket or wall to reduce incoming
noises materially.
SOUNDPROOFING MATERIALS
A variety of materials is used for soundproofing. Choice is determined
by:
1. Location 7. Adaptability
2. Strength 8. Cost
3. Durability 9. Other qualities such as heat
4. Vermin-proof transmission, fire-resistance,
5. Availability hygroscopy, etc.
6. Compactness
The following materials have been used for soundproofing purposes:
"Dry Zero" airplane blanket Compressed cork
"Seapak" Linoleum
Balsam wool Micarta
Felt Wood
Asbestos Metal
Bakelite Doped fabric
Leather Upholstering material
Balsa wood Plywood
Rubberized hair Kapok fibers
The fibrous materials, in general, seem to be the best from the point of
view of noise reduction to weight ratio. It is not desirable to use metal
because of possible "drumming" and reverberation.
APPLICATION OF MATERIALS
Soundproofing materials are usually employed in the following applica-

tions.
1. Cabin walls. Alternate layers of fibrous material with air spaces

approximately three times the thickness of the material are used. One
layer of the material should be attached to the outer metal skin. The
inner layer may be faced with any suitable upholstery material.

238
Figure XIV-1. Suggested soundproofing treatments of cabin wall interiors. A, air
space; F, floating blanket; G, fiber glass; M, metal skin; O, organic material; H, organic
material, horizontal corrugated; V, organic material, vertical corrugated; T, cabin trim;
W, wire lacing.
Any fibrous material which is not very good in tension, or is likely to
pull apart due to its own weight when bung, may be reinforced by muslin
covers, preferably glued on in order to eliminate stitching holes as much
as possible.
1 fâ„¢!
1â€”n
Figure XIV-2. Suggested treatment of individual structural members to reduce
"drumming" and resonance effects. Unsupported panels may be similarly treated.

SOUNDPROOFING
239
Since all materials come in standard sizes, either in 36- or 72-inch widths,
wall seams in the three alternate layers should not be directly behind one
another.
2. Windows. One sheet instead of double panes of window glass is
sufficient since the windows are far better sound insulators than any
soundproofed portion of the cabin. The glass should be set in felt-cornered
window stripping and should be supported firmly over the entire perimeter.
3. Flooring. A fibrous material placed over the metal floor and then
covered with micarta or "battleship" linoleum forms a very effective floor-
ing. Micarta-covered panels may be made removable for inspection of
control cables and structure by means of cowl fastener attachments.
4. Doors, removable bulkheads, baggage compartments. Doors and re-
movable bulkheads may be covered with felt or some fibrous material and
then protected against ordinary wear by a doped and painted, aircraft-
linen cover. Large panels of metal in other parts of the airplane that are
subjected to drumming may be treated with small squares of fibrous mate-
rials cemented on. These are usually located in the center of the panels.
5. Control cables. Control cables should be carried through a duct, pref-
erably felt-lined. Inspection may be made by inserting suitable micarta
panels at intervals.
6. Ventilating system. Since the ventilating ducts are exposed to the
open atmosphere, special attention must be paid to them. Where the fire
hazard is negligible, the ducts may be made of doped fabric. The entire
lengths of the ducts, whether of metal or fabric construction, should be
lined with about 3^-inch layer of felt held in place by a lightweight, coarse-
mesh aluminum screen.
7. General. Additional precautions to take in keeping down the noise
level are:
(a) Placement of the exhaust stack exit below the wing, for example,
in a low-wing monoplane, so that the wing may act as a sound
"shadow."
(b) Making the gap between the propeller tip and the fuselage structure
as large as possible to avoid any possible dirunming effect.
(c) Avoiding all unobstructed holes, no matter how small. Even
stitching holes are undesirable.

Table XIV-3 is of interest in connection with the allocation of various
materials used for soundproofing.

NOISE DUE TO JET ENGINES
Since at take-off power 100% of all noise of the order of 116 to 120
decibels produced by a jet engine is generated behind the engine, all
muffling schemes are concerned with this noise, and any sound created
within the engine itself is discounted.

240
Table XIV-3.
Material
Allowance
(per cent)
Fibrous soundproofing materials
Interior fabric and dope
Sound deadening paint
Cement for attaching soundproofing material
Special reinforcements along walls at floor level
Cabin door
Talon fasteners (for inspection, removal, and replacements)
Moulding strips, screws, stiffeners, etc
Total soundproofing weight allowance
54
11
13
11
100
Noise power produced in the exhaust of a jet engine fitted with con-
ventional circular tail pipe varies with the 8th power of the exit velocity
of the exhaust gas. However, any slight increase in the tail pipe's perim-
eter transverse to the axis of the engine will result in an appreciable
reduction in noise because of the lowered velocity gradient between
exhaust and ambient air. The ideal way to increase the tail pipe's
perimeter is to flatten it into a long, thin orificeâ€”a "slit exhaust"â€”or
make it rectangular in shape. Depending on the design, tail pipe noise
suppressors may be expected to cut jet engine noise by about 6 to 18
decibles. The disadvantages of noise suppressors are that they will
impose a certain weight penalty and will create thrust losses ranging from
1 to 5%. No adequate jet exhaust muffler yet exists.
It will take about twice as much soundproofing (in terms of weight)

to quiet the roar in a commercial jet's cabin than is required for a piston-
engine plane of equivalent power.

CHAPTER
The Propeller
The aircraft designer is not likely to be concerned with the detailed
design of the propeller itself, but he is interested in obtaining the type of
propeller most suitable for his design from the point of view of eventual
performance of the aircraft, compatibility with the engine that operates
it, and the possible influence on the configuration of the aircraft.
GENERAL PROPELLER CHARACTERISTICS
The thrust of any propeller, as well as the power required to operate
it, may be determined analytically. The mathematical expressions ob-
tained are:
T - | CmW*,
Q = | CQnW*,
P=
\ CpnW;
also
P = 2iroQ;
where T = thrust in pounds,
Ct = non-dimensional thrust coefficient,
p = air density in slugs per cubic foot,
Q = torque in foot-pounds,
Cq = torque coefficient,
n = revolutions per second,
D = diameter of propeller in feet,
P = work in foot-pounds,
Cp = non-dimensional power coefficient.
241
242
While these equations are not normally of direct interest to the aircraft
engineer, they are helpful in achieving an understanding of the operation
of the propeller under certain circumstances. In plotting the thrust and
power coefficients for a typical propeller (Figure XV-1), it can be seen that
the propeller can operate as: a positive-thrust mechanism, or as a pro-
peller; or as a negative-thrust mechanism with power required to operate
it, and used therefore as a brake; or as a power-producing mechanism, or
as a windmill, with the negative thrust as an undesirable by-product.
Figure XV-1. Various operating states of the propeller made possible by varying
the pitch of the propeller blades.
Through efficient design, these different states can be extended so that
the propeller can be used as a positive-thrust mechanism for high speed
and climb, and as an air brake in the landing approach and upon landing.
Some designs may permit disengaging the propeller and rotating the indi-
vidual blades so that the entire propeller may windmill with the minimum
amount of negative thrust or drag and, at the same time, transmit no
power to the engine. Other designs may incorporate a brake on the
propeller blade when transmitting no power so that it offers the least
drag. (See Table XV-1.)

THE PROPELLER
243
Table XV-1. Relative drag of propeller under different conditions.
Condition Order of resistance
Stopped 1.00
Idling (no thrust) .73
Windmilling .44
Fully feathered .08
Instead of the equations for the propeller listed above, it is customary
to use a so-called "speed-power" relationship, derivable from the follow-
ing equation for power: 1/5
Ls â€” (bhp)i/5(Arrjm)2/5'
where C, is known as the speed-power coefficient.
AERODYNAMIC EFFECTS OF PROPELLER
The propeller, as viewed from behind, rotates toward the right. The
reaction of the forces acting on the propeller tends to turn the engine and
the airplane in an opposite direction. To counteract this, there must be
provided an increase in lift to produce an opposite rolling moment. This
may be done by establishing a difference in the angle of incidence between
the left and right wing so that more lift is produced on the left than on
the right. Another method is to rig the ailerons differentially or to in-
duce a differential action with thfe addition of a fixed or controllable tab
inset in one of the ailerons. \
Naturally, a difference in lift will also produce a difference in drag
which produces a yawing moment. To counteract the yawing moment
with its tendency to turn the airplane to the left, (1) a slight right rudder
may be employed, or (2) a slight right rudder may be rigged by use of a
tab, or (3) the vertical tail surfaces may employ an unsymmetrical air-
foil, or (4) the fin may be offset slightly from the plane of symmetry.
Propeller slipstream increases the air velocity over the wing and thereby
contributes to increased lift as well as drag. However, at high speeds
the effect of slipstream is considered negligible. At low speeds, the effect
may be appreciable and therefore has been given attention in converti-
planes, where the slipstream over the wing with its lift-increase devices
may become very important.
The propeller slipstream is of importance in the cooling of aircraft
engines while the plane is on the ground.

GYROSCOPIC EFFECT
When the airplane rotates about its vertical axis, gyroscopic forces are
brought into play. The moment about the propeller shaft may be ex-
pressed by the following formula:
T = o>Mh ~ h),
244
where coi = angular velocity of the airplane in the turn,
0)2 = angular velocity of the rotating propeller,
7i = mass polar moment of inertia of the propeller about its thrust
axis,
7 2 = mass polar moment of inertia of the propeller about an axis
parallel to the airplane's turning axis.
The effect of the gyroscopic couple is to force the airplane into a left-
hand turn when the airplane is pitched down from a horizontal path; or,
if the airplane is making a left-hand turn, the airplane will tend to pitch
up. The opposite effect is obtained when the airplane is controlled in
the reverse directions.
This gyroscopic couple not only affects the behavior of the craft, but
also introduces additional stresses in the propeller shaft and the aircraft.
For the normal propeller, the loads and consequent stresses produced
upon the aircraft are considered negligible.
PROPELLER PITCH
The geometric pitch of a propeller may be defined as (1) the advance
in a solid medium, or (2) as the tangent of the blade angle, measured be-
tween the chord of the airfoil section and the plane of rotation of the
propeller. Since the propeller blades have different blade angles from
root to tip, the geometric pitch is usually specified at three quarters of
'the radius.
The individual blade can be designed so that it can be rotated about its
axis, either manually or automatically, to change the blade angles. Since
variations are possible, several types will be briefly defined here.
1. A fixed-pitch propeller does not permit any rotation of the individual
blade after the propeller has been built. Such propellers are lighter than
any other type because of the simplicity of construction.
2. An adjustable-pitch propeller permits manual adjustment on the
ground so that the blade angle is changed to suit certain operating condi-
tions. After each such setting, the propeller operates as a fixed-pitch
propeller.
Both the fixed-pitch and adjustable-pitch propellers are used for rela-
tively low-horsepower engines, where the weight of the propeller becomes
an important factor in design considerations.

3. A variable- or controllable-pitch propeller is usually thought of as one
in which the blades may be rotated in flight. There are several possible
variations of this type of propeller:
(a) It may have a limited angular range and therefore not require as
complicated an operating mechanism as some others.
(b) It may permit a variety of blade angles so that the engine may
THE PROPELLER
245
operate at a constant number of revolutions per minute regardless ofthe_
path speed of the aircraft.
To reduce the resistance of a windmilling propeller after an engine has
ceased to deliver power, the propeller blades may be rotated and stopped
so that the engine itself may not be harmed and the drag of the propeller
may be reduced to a minimum.
To offer more resistance and thereby help to reduce the speed of the
aircraft, especially on landing, the blades may be sufficiently rotated to
act as a brake. See Table XV-2 for the effect of reversing propellers on
the length of the landing run.
Table XV-2. Effect of reversing propellers on a 4-engine airplane.
Number of
propellers reversed
Trial
Use of brakes
Landing run
None
100%
100%
130
100
71
92
100
52
6
50
48
65
45
100
45
Both the feathering and braking features may be incorporated in a
constant-speed propeller. The effect of the various types of propellers
upon the airplane's performance is illustrated in Figure XV-2.
NUMBER OF BLADES
The determination of the number of blades for a propeller requires the -
consideration of a variety of factors. Theoretically, a one-blade propeller
should have the highest propulsive efficiency since propeller blades inter-
fere with each other aerodynamically. However, while such a propeller /
can and has been manufactured, most propellers are produced with 2, 3,
or 4 blades, with combinations of 6 or 8 for dual-rotating propellers.
With the increase in the number of blades, the over-all efficiency de-
creases. However, as the amount of power to be absorbed by a two-
blade propeller, for example, increases, it would be necessary to increase
its diameter until either the physical dimensions would be too great for
246
1000
oi I I I I I I I
0 40 80 120 160 200 240 280
Airplane speed, miles per hour
Figure XV-2. The effect of different types of propellers on the performance of the
airplane:
1. Fixed-pitch propeller.
2. Limited controllable-pitch propeller.
3. Constant-speed (rpm) propeller.
4. Theoretical curve for a constant speed, variable-diameter propeller.
All propellers deliver the same power for the design speedâ€”in this case, the maximum.
installation in a reasonable airplane design configuration or the tip speed
would approach a velocity equal to that of sound, where the efficiency
would be adversely affected.
For the absorption of rather large horsepowers, rather than increasing
the number of blades on a single propeller, a better compromise may be
made with two counter-rotating propellers. The over-all diameter of
two counter-rotating propellers is likely to be considerably less than an
equally efficient multi-bladed propeller.
A large-diameter propeller will affect the height of the landing gear, for
example, or may require placement of the engine nacelles farther out-
board on the wing. Either or both would lead to structural complications
and consequent added weight. However, since the area swept out by
the propeller near its hub is not very efficient aerodynamically, especially
when the propeller is placed in front of a blunt body, the propeller of the
larger diameter is to be preferred over the smaller diameter. A large-
diameter propeller is generally considered good for a high rate of climb,
with consequent higher service and absolute ceilings, whereas the smaller-
diameter propeller is considered better for the high-speed capabilities of
the airplane. The metal propeller usually has good operating character-
istics throughout the entire performance range of the aircraft.

THE PROPELLER
247
Increasing the number of blades will increase the weight of the propeller
and will have some effect upon the airplane's performance.
Because of the different factors affecting the choice of a propeller, it is
particularly desirable to make a careful analysis of several types of pro-
pellers, if they happen to be available, in order to obtain the best engi-
neering compromise.
PROPELLER INFLUENCE ON AIRCRAFT CONFIGURATION
In order that the propeller may operate at its maximum efficiency, it is
desirable that it be placed as far ahead of a thick body as possible to avoid
the "blanketing" of the propeller itself and the resulting reduction in its
effective working radius.
The propeller would give its best performance if it were placed at the
front end of a well-streamlined long shaft. Obviously, such an installa-
tion is not always possible, since it would make difficult the proper mount-
ing of the shaft and increase the over-all weight of the installation.
Where the propeller tip operates close to a structure (at the side of the
fuselage, for example), allowance should be made for sufficient clearance
between the propeller tip and the structure. One to two inches usually
seems effective. However, the clearance is a function of the propeller-tip
speed and the power absorbed by the propeller blade. Too little clearance
causes "drumming" of the structure and possible vibration of the pro-
peller or the structure or both. The structure in the region of the pro-
peller tip should be reinforced in any case.
PROPELLER CLEARANCE
Nacelles are located outboard as far as allowance for propeller clearance
of at least 2 inches with the side of the fuselage and the tip of the propeller
will require, but usually no farther.
It may be desirable to place the axis of the nacelles at an angle to the
longitudinal axis for any of several reasons; the nacelle may be brought
closer to the fuselage if the thrust line is outward; or the plane of rotation
of the propeller may thus be made to pass through the fuselage ahead of
the cockpit. In some cases, it has been thought preferable to place the
nacelles at a slight outward angle to assure the slipstream effect over the
vertical tail surfaces when one engine quits, for then it is necessary to coun-
teract the yawing moment produced by the thrust of the other engine.
Any such angle incorporated in the design of the nacelle does not affect
the efficiency of the engine as far as forward thrust is concerned since the
cosine of an angle of 5 degrees would reduce the effective thrust forward
by less than Yi per cent. (See Figure XV-3.)

THE PROPELLER
249
4. Surfaces near the propeller tips should be suitably stiffened against
vibration.
5. The pilot and the primary control units, excluding cables and control
rods, should not be located in the region between the plane of rotation
of any propeller and the surface generated by a line passing through the
center of the propeller hub and making an angle of 5 degrees fore or aft
of the plane of rotation of the propeller.
6. Likewise, no passenger door should be located in the same region.
ASYMMETRICAL CONDITIONS
The propeller in a conventional installation, in the course of its rotation,
does not encounter the same airflow as it would if it were located at the
end of a long symmetrical shaft with no interferences behind it.
For example, in the propeller of a single-engine airplane, each blade
would pass by each half of the wing and the landing-gear members. (See
Figure XV-^t.) These interferences would set up vibrations in the pro-
peller. The frequency of the vibration may be found from the formula:
/ = nBc,
where / = frequency of the vibration,
B = number of blades,
n = number of revolutions per second of the propeller,
c = number of interferences.
Figure XV-4. This propeller has six interferences for each blade to set up vibration
responses.
If the natural frequency of the propeller were the same as the inducing
vibration frequency, resonance would occur, and the propeller would
operate roughly and would, in time, even disintegrate. Similarly, the
natural frequency of the propeller should not be in the realm of the fre-
quencies of engine explosions. Each propeller is therefore designed to
operate with specific engines.
Another asymmetrical condition may arise in multi-engine designs
when the airflow through one propeller disk interferes with the airflow
250
Figure XV-5. Flow interferences through the propeller disks lead to asymmetrical
conditions from which vibration in the propeller blades may result.
through an adjacent propeller disk (where the propeller disk is defined as
the area swept out in one rotation of the propeller blades). See Figure
When undesirable conditions arise which may be traced to asym-
metrical conditions, the solution may be found in a different propeller of
a different natural frequency obtained by either different blade design,
different diameter, or an increase in number of blades.
The normal propeller-engine combination imposes a torque upon the
airplane that gives it a tendency to turn about the longitudinal axis.
On the ground, the rotation of the airplane is prevented by the ground
reacting against the wheels. (See Figure XV-6.) As the airplane takes
off the ground, a countering torque or rolling moment must be produced
by the aileroTfST The airplane therefore has to have sufficient air speed
to obtain the necessary rolling moment. As the power used for the pro-
peller-type engine tends to become greater and the spans of the airplane
tend to become smaller, the air speed has to be quite appreciable at take-
off to prevent possible disaster, since engine torque is usually greater at
take-off than in normal flight.
XV-5.
EFFECT OF ENGINE TORQUE
Figure XV-6. Effect of engine torque on ground reactions.

THE PROPELLER
251
The torque present in the propeller-engine combination might be over-
come in multi-engine designs in several ways:
1. An equal number of left-hand and right-hand rotations of propellers
could be used. This solution is impractical, primarily from the standpoint
of economics.
2. The engines could be arranged in tandem. Such an arrangement
is seldom desirable due to the structural complications involved.
3. The use of counter-rotating propellers is the most practical solution
and the one that is most commonly employed.
For relatively high-speed airplanes using large-horsepower engines, the
counter-rotating or dual-rotation propeller is certainly necessary.
Another necessary application is for the vertical take-off airplane, since
there, the axis about which the torque would act is vertical. No easy
solution would be available in the design of the landing gear for counter-
ing the torque on the ground. The transition from the moment of take-
off to the time when sufficient air speed past ailerons was obtained would
include a period during which the aircraft rotated about its vertical axis,
unless secondary propellers were employed purely for the purpose of pro-
ducing a torque to counteract the engine torque.
TANDEM ENGINES
Tandem arrangements of engines are little used in multi-engine designs
although the increase in power of individual units may make such a solu-
tion preferable to some others. The advantage of the tandem setup is
that in case of one engine failure there is not so great a yawing moment
produced by the excess thrust on one side as would be the case if the outer-
most engine in a multi-engine design failed.
Auxiliary cooling, usually special blowers, must be provided for the rear
engine since the front engine partially blankets it. This could be over-
come by providing larger airscoops in the leading edge of the wing for ob-
taining the necessary cooling air.
The tandem arrangement may reduce aerodynamic resistance although
this may be open to question since scoops for cooling air will add to the
resistance. In designing the nacelle, a symmetrical airfoil shape with its
maximum thickness located approximately in the plane of the cylinders
(for radial engines) of the front engine is a good first solution. Slight alter-
ations may then be made as design conditions warrant.
The propeller for the rear engine, since it operates in the slipstream of
the front engine, may be designed with a smaller diameter than the front-
engine propeller. However, some difficulty may arise from vibration
caused by the irregular airflow.
For airplanes with the conventional type of landing gear, the propeller
clearance requirement (9 inches between the propeller tip and the ground)
252
will either force the placing of the engine nacelle comparatively far above
the normal position or require an increase in landing-gear length.
The torque of the tractor engine will be offset by the torque of the
pusher engine, so that aerodynamic means do not have to be resorted to
for providing the offsetting rolling moment.
The pusher propeller in a tandem design would be smaller than the
tractor propeller and also somewhat different from it since it works in the
slipstream of the latter. However, it may be blanketed to some extentâ€¢
by the structure ahead of it.
PUSHER INSTALLATIONS
When the propeller operates in the rear of the engine, relative to the
forward flight of the aircraft, it is known as a "pusher." Like any other
configuration, such an installation has certain advantages and disad-
vantages resulting not only from the placement of the propeller, but also
from the location of the engine.
In general, a pusher installation, since it is often located aft of the
trailing edge of the wing, will also be located aft of the center of gravity
of the airplane. Cargo and passenger items, then, are usually situated
ahead of the engine in order to get the proper location of the center of
gravity of the airplane in relation to the mean aerodynamic chord.
Some of the advantages of the pusher installation are:
1. Passengers and crew obtain unimpeded vision forward since there
are likely to be no obstructions ahead and below.
2. The installation is likely not to be as noisy as a tractor installation.
3. The pusher installation is likely to offer less aerodynamic inter-
ference to the wing, therefore enabling the wing to be more efficient.
4. With the power plant aft of the center of gravity, a longer fuselage
may be possible to house personnel and equipment since the power plant
represents a relatively heavy concentration of weight.
Figure XV-7. The pusher propeller may cause a higher location of the power plant
in order to allow for propeller-tip clearance in the three-point landing attitude.

THE PROPELLER
253
Against these advantages, the following disadvantages should be
considered:
1. Propeller clearance with the ground may necessitate a higher landing
gear or a higher location for the power plant since the three-point landing
attitude will govern the location of the propeller.
2. The propeller may be badly blanketed by the engine. See Figure
XV-7.
3. Doors and escape hatches may be located in the rear. In an emer-
gency exit in flight, such a location is considered to have an unfavorable
psychological effect as it may cause the people emerging to fear that they
may fall into the propeller.
4. A power plant behind the passengers is considered psychologically
unsound since passengers may fear that the heavy engines will crash in
on them in case of a bad landing.
5. The structure needed to absorb the energy caused in a crash by a
pusher installation may have to be sturdier than the structure needed
to absorb the energy of the cargo and crew in a tractor installation, where
there is no structure ahead of the power plant.
6. The pusher installation, especially for a single-engine design, is
likely to complicate the tail structure since booms may be required to
carry the tail surfaces and to protect the passengers from walking into
the propeller when embarking.
TAIL INSTALLATION
Desgins have been proposed and built with the propeller located di-
rectly aft of the tail surfaces and operating as a pusher. The engine is
connected to the propeller by an extended shaft.
In order to protect the propeller in landing, the vertical tail surfaces
are located below instead of above the horizontal tail surfaces, and
usually the tail wheel is attached to the rudder post.
Since the air is rotated with the propeller, the air hitting the vertical
tail surfaces on one side below the longitudinal axis of the airplane will
produce an aerodynamic torque or rolling moment which will, to some
degree, offset the engine torque.
In an installation of this kind, more than in the normal tractor installa-
tion, the gyroscopic couple may play an important part in the handling
behavior of the airplane in a turn. In any case, the moment of inertia
of the airplane about the vertical axis will be materially changed, so that
the response of the airplane may not be quite as quick as in the conven-
ventional designs.
Both the engine torque and the gyroscopic effects can be overcome by
the dual-rotating propeller.

254
SPINNERS AND CUFFS
To streamline the propeller hub into the engine nacelle, a nose spinner
is designed to conform with the contour of the nacelle. The maximum
diameter of these spinners (generally located at the base of the spinning
cone) should not extend beyond the aerodynamically inoperative blade
shank.
In addition to spinners, cuffs of an airfoil cross section are designed to
cover the shank of the propeller blades. The propeller blades, especially
those that are removable from the propeller hub, end in round shanks
at the root. These cuffs help to increase the thrust of the propeller and
improve the airflow in the region of the propeller hub. Since the aero-
dynamic loads and centrifugal forces on these cuffs are not appreciable,
they may be formed from sheet metal of suitable thickness.
Table XV-3. Propeller spinners.
Dime
asions
Manufacturer
Number
Weight (lb)*
01 d Jacies
Diameter
Length
Curtiss
21%
29%
21.5
21%
25Vâ„¢
23.0
24%
25%
18.5
24%
25%
22.0
Curtiss
24%
28%
25.5
Curtiss
30%
24K-
27.0
2Z%
23#
21.0
16
15
16.0
16
20
17.5
23K
14%
24.5
* On basis of per inch of diameter = 1 to 1.25 lb.
PROPELLER SELECTION
In selecting a propeller, the maximum speed as well as the rated horse-

THE PROPELLER
255
1000
800
0 600
Â«
1 400
200
Trial 2
Trial l<
\t
|j
p !j
0 40 80 120 160 200 240
Airplane speed, miles per hour
Figure XV-8. The proper selection of a propeller requires a number of trial and
error calculations to determine the point at which horsepower available is equal to the
horsepower required.
Example. The horsepower required for corresponding speeds has been
calculated for a given design and plotted as shown in Figure XV-8. The
engine chosen will deliver 850 horsepower at 2450 revolutions per minute
(crankshaft speed) at sea-level conditions. The propeller is geared at
3 to 2 ratio, so that the propeller speed is % of the crankshaft speed, or
will rotate at 1634 revolutions per minute. The brake horsepower is
given for the geared engine so that it is unnecessary to allow for the 2 or 3
per cent reduction in power due to gearing.
First, assume propeller efficiency at maximum speed equals 80 per cent.
This gives a thrust horsepower of 850 X ^0,8 = 680 horsepower. For a
required horsepower of 680, the speed in horizontal flight would be 222
miles per hour at sea level.
Next, the value of C, for the propeller is calculated. It is a convenient
"speed-power coefficient" which is used in an equation not involving the
propeller diameter.
= 2.14 Vmph </-
Â°' (bhp)1'6 (Nrpm)116 P0
0
For the conditions set forth:
c'= (^Zm^^^^ = Â°00008611F-*0

For 222 miles per hour:
C. = 1.91.
THE PROPELLER
257
V/nD at point C is 1.18. Since V and n are known, the diameter of the
propeller, D, may be solved for as follows:
X 1 io (227 X 1.467)
nD (1634/60)2>'
where D = 10.35 feet (say, 10 feet, 6 inches). The blade angle at 75 per
cent of the radius is about 30 degrees.
The critical propeller-tip speed is about 1000 feet per second and it
should not be exceeded if serious reduction in propeller efficiency is to be
avoided. The critical propeller-tip speed, Vc, may be calculated from
the following formula: & - 6 lf"
v. - VC^r25)' + ^
For the propeller diameter just calculated Ve becomes equal to
- (1000^1634) 2 + (227 X 1.467)2 = 960 feet per second.
The propeller should, therefore, give excellent performance at sea-level
conditions. If best cruising speed is desired at another altitude, say that
to which the engine is supercharged, then the corresponding values of
brake horsepower and air density must be substituted in the equation
given for calculating the value of the power coefficient C,.
Figure XV-9 represents propeller characteristics when the propeller
is placed in front of a cowled radial engine located in the leading edge of a
moderately thick wing. These characteristics depend upon obstructions
in the propeller slipstream. If more accurate values for efficiency and
blade angles are desired, pertinent NACA reports should be consulted.
If a 2-bladed propeller has too large a diameter so that the tip speed is
too high, or there is insufficient ground clearance, or other factors prevail-
ing which make it necessary to choose a smaller diameter, a 3-bladed pro-
peller may be required.
For the conditions set forth, again assuming an initial efficiency of 80
per cent:
c- - (mi%wÂ»*Â°^* = Â°000086Fp.
For 222 miles per hour:
C, = 1.91.
The corresponding maximum efficiency as found before was 86 per cent.
This efficiency is for a 2-bladed propeller whereas a 3-bladed propeller is
to be used, the maximum efficiency of which would be reduced from 86 to

83 per cent, which is still higher than initially assumed.

258
After successive trials, it is found that maximum speed is closer to 224
miles an hour, for which the value of C, = 1.92, and the propeller efficiency
for a 2-bladed propeller is about 86 per cent, or 83 per cent for a 3-bladed
propeller.
For this value of Câ€ž the corresponding V/nD, determined as before, is
1.15:
Vâ€” _ 224 X 100467
nD ~ (1634/60)Z)'
/ evaluating D = 10.47 (say, 10 feet, 6 inches). The blade angle at 75
/ per cent of the radius is approximately 29.5 degrees. Again this value
\ is for a 2-bladed propeller since all values have been based upon those
) obtained from Figure XV-9, which is for a 2-bladed propeller. However,
i these same values may be used, and the diameter so found may be re-
duced 7 per cent to obtain the corresponding diameter of a 3-bladed
propeller, if pertinent data for a 3-bladed propeller is not at hand. The
diameter of the 3-bladed propeller is then 10.5 feet less 7 per cent, or 9 feet
1^ 9 inches, roughly.
Alternative Method of Determining Propeller Diameter. The diameter
required for the 2-bladed propeller may be calculated directly by means
of the formula:
propeller diameter
For the example cited:
The same formula may be used to calculate the 3-bladed propeller by
using only 70 per cent of the value for the brake horsepower in the formula.
EMPIRICAL FORMULAS
Various empirical formulas may be devised, based on various parameters,
for the determination of propeller diameters. In all cases, D is the di-
ameter in feet, P is the rated horsepower, V is the maximum speed in
miles per hour, and N is the number of revolutions per minute of the
propeller.
K = 67 for a two-bladed propeller,
K = 47 for a three-bladed propeller.

THE PROPELLER 1 1 1
_l I Z-i I I L
0.8 0.9 1.0 1.1 1.2 1.3
Airplane Mach number
Figure XV-10. Propellers for turboprop engines.
2. For a single- or twin-engine design:
2 blades.
3. For a 4-engine design:
3 blades.
40 0 Â° = V(l^) (OTo) Xl^)'
2 or 3 blades.
Aâ€”
Figure XV-11. Typical propeller-blade shapes for high-speed applications.

260
100 h
Figure XV-12. Variation of the propulsive efficiency of a propeller with airplane
Mach number. Further reductions for the various Mach numbers may be expected due
to installation, number of blades, and other possible variations.
5. For horsepowers between 2,000 and 0,000:
K = 11.5 for 3-bladed propeller,
K = 9.5 for 4-bladed propeller,
K = 8.0 for G-bladed propeller.
Â»-Â«(VÂ£)(Vi>
For wooden propellers:
K = 48 for 2 blades,
K = 49 for 4 blades.
For metal propellers:
K = (34 for 2 blades,
X = 57 for 4 blades.
7. For horsepowers up to 3,000:
D = 480
2 or 3 blades.
THE PROPELLER
261
Table XV-4. Propeller data.
Num-
Maxi-
Maxi-
Material
ber
Diameter
mum
mum
Weight
of blades
of
Pitch
in inches
hp
rpm
in
blades
rating
rating
pounds
1. Wood
Fixed
63
40
2800
2. Wood
Adjustable
84
50
14
3. Al. Allov
Adjustable
84
54
41
4. Wood
Fixed
70
65
2800
10
5. Wood
Fixed
70
75
2600
6. Wood &Plastic
Variable
85
76
2800
25
7. Wood
Fixed
72
90
2700
11
262
Table XV-4. (Continued). Propeller data.
Num-
Maxi-
Maxi-
Material
ber
Diameter
mum
mum
Weight
of blades
of
Pitch
in inches
hp
rpm
in
blades
rating
rating
pounds
49. Al. Alloy
Controllable
151
2300
1520
494
50. Steel
Controllable
150 to 181
2500
1225
51. Al. Alloy

Controllable
114 to 180
2500
1430
52. Al. Alloy
(i
Dual-rotation
157
2700
1400
757
53. Al. Alloy
Dual-rotation
144
2700
1400
712
54. Hollow Steel
Variable
134
2800
1470
485
55. Hollow Steel
Variable
158
2800
1470
509
CHAPTER XVI
The Power Plant
The power plant consists of the engine, propeller, starting system, cool-
ing system, fuel and oil systems, cowling, engine mount, and miscellaneous
accessories. Each item requires a considerable amount of design study, for
the ultimate success of the airplane depends upon the proper selection and
proper functioning of every part of the power plant.
Suppose an engine is to be selected. There are air-cooled and liquid-
cooled engines, either radial or in-line, or inverted in-line, or Vee, as well
as a few others. If the horsepower required is rather large, the elimination
of a few types may be possible since there may be engines having a partic-
ular cylinder arrangement that do not develop the required horsepower.
Again, the air-cooled type may be preferred to a liquid-cooled type due to
less complications in the installation since no cooling system is required
for the former, yet the liquid-cooled engine may be preferred for its lesser
frontal area or greater reliability. The line of demarcation in considering
the advantages between air-cooled and liquid-cooled engines may be fine,
and one cannot say arbitrarily that one engine is better than the other
until all the facts have been considered.
The different possible arrangements of engines are to be considered also,
so that a complete book instead of this short chapter could be written on
the power plant alone. The engines may be tractors or pushers; that is,
they may be placed with the propeller in front of the engine, meeting the
air before the engine; or the propeller may be placed behind the engine.
Such arrangements are sometimes desired for compactness and, as in the
case of tailless airplanes, to obtain a center of gravity location reasonably
far back and relative to the fuselage length.
There are also tandem arrangements, or combinations of tractor and
pusher arrangements. These are to be considered especially when the
263
264
number of engines becomes so large that it may be desirable to concentrate
the engines as much as possible in order to reduce the length and complica-
tion of fuel, oil, and control lines.
The selection of the engine and the arrangement of a group of engines by
no means ends the problem, for even the best engine cannot function prop-
erly unless attention has been paid to proper installation of the cooling
system (whether it be the NACA cowl for air-cooled engines, or radiators
and pipe lines for liquid-cooled engines); to the correct installation of the
fuel and oil systems with special reference to the size of pipes or tubing; to
the location of pumps and relief valves; and, to the numerous little items
that go to make up the whole.
In the preceding paragraphs a brief survey of a few of the factors affect-
ing the power plant selection and design has been made. The following
material outlines the considerations to be taken into account in greater
detail.
LOCATION
.> To affect least the aerodynamic characteristics of the wing, it would be
desirable to locate the nacelle below the wing. To reduce torsional loads
imposed on the wing structure by the eccentric thrust-line position, it
would be desirable to locate the nacelle more or less with its axis in line
with the chord line. Usually, however, the governing condition for the
low-wing monoplane is the required propeller clearance with the ground.
For jet engines, the pod installation is preferred in this country. Since
the fuel is carried in the wing, the location of the jet pod below the wing
is a primary safety consideration. The torsional moment imposed on the
wing is desirable to offset the wash-out of the wing occurring at high
angles of attack and under accelerating conditions.
Unless the spar structure is cut away, the most rearward position usu-
ally possible is to have the engine as close to the front face of the front
spar as clearances for accessories will permit.
SUBMERGED ENGINES
Considerable attention has been devoted to enclosing engines within
the wing structure or within the fuselage. The chief reason for wanting
to enclose the engine within the wing is aerodynamic. Although this
arrangement eliminates the engine nacelle, which interferes with the air-
flow around the wing, there is still the possibility of increased resistance
caused by ducts within the wing. For the propeller-type engine, this
increased resistance may be partially offset by extending the propeller
shaft. Then, the propeller may work more efficiently because it is not
operating directly in front of a thick body. Such extensions, however,
have inherent disadvantages such as increased weight and vibration.

THE POWER PLANT
265
Engines located within the fuselage permit the placing of the pilot ahead
of the engine, giving the best possible vision forward. It is also possible
to obtain a better structural arrangement and center of gravity location.
Adequate air intakes may be provided by means of ducts whose open-
ings are located just at the leading edge of the wing where their position
causes the least effect upon the aerodynamic qualities of the wing, espe-
cially at very high speeds.
NUMBER OF ENGINES
Assuming that a certain horsepower is required for a given design, should
this horsepower be provided by one or more engines? Obviously, it is
simpler and cheaper to have only one engine since there is only one set of
controls, one engine mount; in short, one installation and all that it entails.
However, the one engine centrally located in the nose of the fuselage is a
source of undesirable noise, heat, and reduced vision. Moreover, in case
of engine failure due to any reason, the airplane must land immediately.
The other alternatives are two, three, or four engines, so that in case of
one engine quitting, the flight may be continued, provided that this is pos-
sible with the remaining engine or engines.
If flight is to be maintained with one engine in a twin-engine design,
there may be some penalty in allowable gross weight that the airplane
may carry because it is usually, although not always, difficult to maintain
rectilinear flight when the thrust vectors are not symmetrical and espe-
cially if the horsepower loading for the two engines is already initially
heavy. It may be possible to dump the fuel load in order to lighten the
load, but if the airplane is over rugged terrain it may not be advisable to
do so since the required landing might be dangerous if attempted immedi-
ately (as the loss of fuel would dictate).
The three-engine design, then, has the advantage over the twin-engine
design in such a case, since, with one engine not operating, it is usually still
possible to maintain rectilinear flight without loss of altitude with two
engines functioning. Moreover the power loading is increased only 50
per cent instead of 100 per cent.
The use of four or more engines may be necessary for large airplanes,
especially when the number of large capacity engines is limited.
Opposed to the safety factor of multi-engine design in possible engine

failure it is to be remembered that multi-engines offer many more com-
plications and require at least one co-pilot, in addition to the chief pilot,
to operate. In many cases, additional members are required in the crew
that is responsible for power plant operation.
In attempting to work out the best combination of engines from the
point of view of maintaining flight with one or more engines not operating,
266
it is possible to arrive at a mathematical analysis by means of the prob-
ability theory. If the probability of any engine failure is independent of
any other and is the same for every engine (as would be likely when the
engines have the same manufacturing and operating characteristics) then
the relative probability of 0, 1, 2 â€¢ â€¢ â€¢ failures is given by the successive
terms of the binomial
(5 + P)\
where p = the probability of a failure,
q â€” 1 â€” p, where q is the probability that this failure will not occur,
k = the number of engines involved.
The value of p may be assumed if no data are available, or may be based
upon existing flight data, and may then be calculated as follows:
/number of failures in annual number\ ,average j th of x
of miles flown H individual trips /
\ annual number of miles flown /
For example, if it is assumed that there were 10 single-engine failures in
1,000,000 miles flown and that the average trip length was 2,000 miles,
then from these data.
'= (imm)2'000 = 002'
q = l - p = l - 0.02 = 0.98.
An example will help to illustrate the calculation procedure and to inter-
pret the algebraic terms involved: for a three-engine airplane, expanding
(q + p)8 yields q* + Zq2j>1 + 3 g'p2 + P30 0 Each term of the expanded
binomial indicates the nonfailing and failing engine combination of the
three engines; for example, the exponent of q indicates the number of non-
failing engines of that combination and the exponent of p indicates the
number of failing engines in that combination. The term dqpp1 indicates
the number of failing and nonfailing combinations when any one of the
three engines may fail. The calculations for the three-engine design for
the values of p and q as determined, are tabulated in Table XVI-1.
Table XVI-1.
1.
Term of binomial
q*
3?'p
2.
Number of possible en-
gine failures
3.
Number of no engine
failures
4.
Evaluation of binomial
term (Probability of
number of no engine
failures indicated in
10,000 trips)
9,411.92 X 10-*
576.24 X 10â€”<
11.76 X lOâ€”4
0.08 X 10-Â«
THE POWER PLANT
267
Table XVI-2. Probable number of chances of failing and nonfailing engines in
10,000 trips.
Number of engine failures
of engines
9,800
200
9,604
392
â€”
â€”
â€”
â€”
9,412
576
12
(0)
â€”
â€”
â€”
9,224
753
23
(0)
(0)
â€”
â€”
9,039
922
38
(0)
(0)
â€”
8,858
1,085
55
(0)
(0)
(0)
From the calculations of possible engine failures it is possible to estimate
the possible number of successful nights if the flight of the airplane can be
maintained and controlled with one or more engines not operating. Refer
to Table XVI-3.
Table XVI-3. Probable total number of successful nights* in 10,000 trips.
With the following number of engines not operating
number
6
268
These ratings may be for different altitudes for the same engine, depend-
ing upon the certain design characteristics.
ENGINE SELECTION
Unless engines are specified in the original design specification, a study
of available engines should be made, keeping the following considerations
in mind for the propeller-type engine:
1.
Horsepower range
9.
Economy of fuel consump-
2.
Supercharged or not super-
tion
charged
10.
Economy of oil consumption
3.
Critical altitudes of super-
11.
Original cost
charged engine
12.
Ease of maintenance
4.
Normal revolutions per minute
13.
Type of cylinder arrangement
5.
Propeller gearing
14.
Over-all dimensions
6.
Weight per horsepower
15.
Method of cooling
7.
Dependability
16.
Department of Commerce ap-
proval
8.
Durability
17.
Availability
It is comparatively easy to determine the approximate maximum horse-
power required on the basis of horsepower loading and the estimated gross
weight of the airplane to be designed and built for the required horsepower
equals the ratio of gross weight to horsepower loading.
Some of the considerations listed may be difficult to ascertain, especially
for the student, and may be safely disregarded. It is well to keep these
various considerations in mind, however, for possible future use.
Similar considerations are applied in the selection of the jet-type engine.
ENGINE NACELLES
In multi-engine design, the power plant is located in a nacelle attached
to the wing. The engine mount and engine cowling, installation of acces-
sories, and the like are the same as before. The nacelle construction is
similar to a shortened fuselage. The main attachment of the nacelle is to
the front spar, with bracing or supports going back to the rear spar.
FUEL CONSUMPTION
The fuel consumption of an engine is one of the important criteria to
consider in engine selection. Just from simple considerations alone, it is
evident that the engine having the lowest fuel consumption would be the
desirable one to select. One of the useful formulas, the so-called Breguet
formula, may be used to indicate the importance of fuel consumption in
determining the range of an airplane. For the derivation of the formula
see page 36.
R (range in miles) = 863 If002 log,â€ž |P,
D c W,
THE POWER PLANT
269
where c = specific fuel consumption in pounds per brake horsepower
per hour.
Tj = propeller efficiency. (For pure jet-type engines, this term
disappears from the equation.)
L/D = lift-drag ratio of the complete airplane at speed V.
Wo = gross weight in pounds at start of flight.
We = W0 less oil and fuel weight.
For long ranges, it would be necessary to employ a step-by-step integra-
tion using successive values for Wo and We, in order to obtain more ac-
curate results.
It is not always possible to so coordinate the aerodynamic characteristics
of the airplane and the best operating efficiency of the power plant.
It should be noted that any appreciable change in the fuel consumption
has a greater effect than a change in propeller efficiency. Thus, choosing
an engine with a low fuel consumption is highly desirable and more ef-
fective than trying to improve the propeller efficiency. These observa-
tions apply to the non-propeller-type engine as well, as far as the fuel
consumption is concerned.
The engine-mount supports virtually the entire power plant in the usual
case, although for engines of very large horsepower, the engine mount
proper does not extend more than a few inches beyond the rearmost acces-
sory. It is a better engineering design to use tubular steel supports which
may be enclosed with a suitable cowling because of the necessity of gaining
access to various parts of the engine and its accessories.
Figure XVI-1. Engine mount for a radial engine of relatively low horsepower
capacity. The engine is attached to the ring by bolts through corresponding lugs on
engine mounting-plate and engine mount-ring.
ENGINE MOUNTâ€”NON-JET ENGINES
i
270
Figure XVI-2. An engine mount for large-horsepower radial engine. A detail of
the rubber-type vibration mount is shown on the right.
Tubular steel is used customarily for the entire mount whether the air-
plane is to be reinforced monocoque or not. The sizes of the members are
determined by stress analysis, although lj^-in. or lj^-in. outside diameter
tubing for engines of horsepower between 500 to 1500 may be used for
preliminary design until a stress analysis has been made.
The engine mount should be so constructed that the entire power plant,
including the oil-system cowling and manifold, may be readily detached
for replacement.
Figure XVI-3. Additional details of rubber mounts located at the points of attach-
ment of engine mount to the firewall frame station.

THE POWER PLANT
273
Figure XVI-8. A series of wind-tunnel tests on engine cowlings (see NACA TR
662) indicate that arrangement (2) is better than (1); (4) is better than (3).
6. It is important that the fire bulkhead fit closely to the cowling all
around the section of the fuselage.
COWLING
1. The type of cowl developed by the NACA for radial engines has met
with general favor by the industry. In designing the contours of any type
of engine cowling, special care should be exercised to avoid all abruptness
or reversal in curvature. A well-rounded contour will help effectively in
reducing parasite resistance. (See Figures XVI-8 and XVI-9.)
(8) (9)
Figure XVI-9. Tests indicated the exit slot of the cowling was too close to the
wing in (5), (7) was an improvement over the cowl in (6), and (9) was better than (8).
(See NACA TR 662.)

274
Figure XVI-10. A few governing dimensions are given in this illustration. Dimen-
sion A â€” 1 inch, B â€” 3 inches, C = 75 per cent of engine diameter (although 65% of
engine diameter may be permissible), D = 2 inches, and E = }4 inch.
2. The correct proportioning of the engine cowls depends much upon
the particular engine, engine nacelle, past experience, likely operating
conditions, etc.
3. Engine cowling should be fastened to the mount, and not to the
fuselage, and in such a fashion that the major portion may be removed in
a few minutes for maintenance, inspection, and minor repairs. Refer to
Figures XVI-10, XVI-11, and XVI-12.
4. The cowling may be made of aluminum alloy 450, 2S or 52S, vary-
ing in thickness from 0.032 to 0.050 with the larger thickness usually re-
Figure XVI-12. This attachment of
Figure XVI-11. Attachment of engine cowling permits adjustment in
cowling to engine crankcase by means flight that may be necessary to keep cowl-
of a vibration mount whose cross sec- ing in alignment and to reduce excessive
tion is shown. stresses.

THE POWER PLANT
275
served for the cowl ring proper and the smaller, or some intermediate thick-
ness, for the rest of the cowling.
5. The engine cowling should be suitably ventilated to prevent accu-
mulation of gases. Refer to Figure XVI-13.
Figure XVI-13. A typical cooling system for a horizontally opposed engine with
a wet sump oiling system.
6. All cowling around the power plant and on the engine side of the
firewall should be made of metal and so designed that any accumulations
of dirt, waste, or fuel may be observed without complete removal of the
cowling.
7. The cowling must fit tightly to the firewall, but openings may be
provided if the airplane surface within 15 inches is protected with metal
or other suitable material.
8. Unless small units of the cowling may be removed easily for inspec-
tion and repair purposes, it is desirable to provide properly secured, small
access doors in the cowling at suitable points.
9. The cowling should be completely drained in all attitudes of flight
and on the ground, with separate drains provided for the parts of the fuel
system liable to leakage.
10. All drains should be so located as to prevent fuel or oil from dripping
onto the exhaust manifold or any parts of the aircraft, or permeating any
cellular material.
11. Carburetor air intakes must open entirely outside of the cowling,
unless the emergence of back-fire flames is positively prevented. The air
intakes should be suitably drained.

276
EXHAUST MANIFOLDSâ€”RECIPROCATING ENGINES
The exhaust manifold carries the exhaust gases of the cylinders to some
point outside of the power-plant section. Usually it is circular in cross
Figure XVI-14. A typical exhaust and induction system for a downdraft radial
engine.
section and, since the volume of exhaust gases increases as the number of
cylinders are added, it becomes necessary to vary the cross section of the
exhaust manifold so that the product of velocity times the pressure remains
constant.
An empirical formula for the design of exhaust manifold is given by
A = (0.04) (P) |,
where A = minimum cross-sectional area required for the section under
consideration,
P = maximum engine horsepower at which the engine may be
operated,
n = number of cylinders exhausting into the manifold up to the
section under consideration, and
AT = total number of engine cylinders.
The number of outlets varies; there may be one or two. It is possible also
to have individual outlets which are comparatively short but are0 0 also dan-
gerous because the hot exhaust gases are not carried far enough away.
Manifolds with a single right-hand outlet with all gas flowing in a counter-
clockwise direction (that is, opposite to the engines firing order) are also
used in the belief that back pressures are reduced.
The following considerations enter into the design and installation of
the exhaust manifolds:
1. The exhaust manifolds, stacks, or collectors preferably should be
made of interchangeable sections, usually of 18-8 corrosion resisting steel
(Iconel) or carbon steel. The wall thicknesses vary from 0.035 to 0.049
THE POWER PLANT
277
inch with the greater thickness reserved for high-powered engines, partic-
ularly those using high-octane fuel.
2. The diameters of the manifolds vary from 2 to 4 inches, and may be
bent with an inside radius as small as 2 diameters, although a larger radius
is customarily used for ease in fabrication as well as to reduce undesired
back pressures. The cross-sectional area of the exhaust manifolds should
gradually increase until the cross section at the last cylinder is at least 50
per cent of the total exhaust-port area of the particular engine for which
the manifold is being designed.
3. The "downwind" clearance of the open end of the exhaust pipe should
be at least 4 feet. Any exhaust pipe which is not exposed to the outside
air should be either water- or air-cooled by a special cooling system sur-
rounding the unexposed pipe.
4. The exhaust pipe should be kept at least 3 inches away from any in-
flammable part of the airplane, and the exhaust end should be at least 5
inches away from any inflammable part.
5. Expansion joints should be provided for and such joints should be
capable of articulation to a certain degree to permit changes not only in
length but also in angle of alignment due to expansion.
6. Gases should be discharged clear of the airplane structure so that
they will not blow back on the carburetor air intake, the pilot or passen-
gers, nor cause a glare ahead of the pilot at night.
CARBURETOR SCOOP DESIGN
Engine performance depends greatly upon the carburetor air-induction
system, especially those using the pressure-type carburetor. These often
incorporate a 90-degree elbow relatively close to the entrance flange of the
carburetor.
Figure XVI-15. Governing dimensions for an air scoop. The figure on the left
shows an example of poor scoop design; the one on the right an improved scoop design.
278
Figure XVI-15 indicates both poor and good scoop design. The dimen-
sions may be obtained from the following information:
1. The length L of the straight portion of the pipe should be as long as
practical up to 4 diameters.
2. The radius r should be a minimum, preferably not over % inch and
may even be sharp.
3. The radius R should be as large as possible, up to such a value that
R/D = 2 when r is greater than % inch.
4. The ratio of W/D should approach 6 as the upper limit, although
generally it is more practical to keep W equal to the width of the carburetor
flange. When the length L is comparatively short, good distribution can
be obtained by introducing vanes in the elbow to direct the flow.
FUEL SYSTEMS
The fuel system consists of tanks, piping exterior to the engine, pumps
not integral with the engine, strainers, gauges, pipe fittings, and valves
and cocks.
The primer is operated from the instrument panel board in the pilot's
cockpit. A shut-off cock is located in the primer line.
For starting the flow of fuel, it is necessary to use the hand or "wobble"
pump which should be as far below the fuel tank as possible so that it is
flooded at all times. As soon as the engine is started, the engine-driven
gear pump will continue the pumping of fuel. The hand or wobble pump
is so arranged that fuel will flow through it even when it is not being oper-
ated. If its location is at some distance from the cockpit it may be oper-
ated through a suitable linkage system to a crank near at hand to the pilot.
The engine-driven gear pump has a by-pass so that, in case of failure,
the hand or wobble pump becomes the emergency pump and will force the
fuel past the gear pump through the by-pass.
In case the pump supplies the fuel at a greater pressure than the car-
buretors are designed to take, the adjustable relief valve permits the fuel
to flow back into the tank.
All tanks must be vented in order to prevent partial vacuums from form-
ing since these may reduce or prevent proper fuel flow. This vent is usu-
ally located at the topmost point, usually the filler unit of the gasoline tank.
A strainer must be included in any fuel line and is usually located at the
lowest possible point in the line and in a place where it is easily accessible;
for, the strainer collects the water, grit, and dirt which may have collected
in the fuel system even with the best of precautions.
The three-port, two-way cock is operated from the cockpit.
A pressure gauge is used in all fuel systems in order to indicate constantly
the pressure in the fuel lines just ahead of the carburetors. If the pressure
THE POWER PLANT
279
should fall below a predetermined value it is usually a sure sign of danger
â€”either the line is clogged or the fuel has given out.
Pumps
1. If a mechanical pump is used, an emergency hand pump is also re-
quired.
2. Hand pumps may be used for pumping fuel from an auxiliary to a
main tank.
3. The hand or wobble pump should be placed at least 50 per cent be-
low the top of the main fuel supply. In some cases, it may be desirable
to place it as far below the fuel system as the design will permit. Opera-
tion of the pumps may be done from the cockpit by means of a suitable
push-pull operating system.
4. The hand-operated pump must be so installed as to be operated
readily from the cockpit without requiring any opening of valves or cocks
in the system.
Tanks,
1. Fuel tanks should be capable of withstanding an internal test pres-
sure of ZYi pounds per square inch without failure or leakage. Fuel tanks
that have a maximum fuel depth greater than 2 feet should be investigated
for the pressure developed during the maximum applied acceleration with
full tanks.
2. No fuel tank may be closer to the engine than the remote side of the
firewall.
3. Surfaces of the tank or tanks should be so ventilated that the fuel
fumes cannot accumulate in case of leakage. At least inch of air space
should be allowed between the tank and the firewall.
4. Each tank must be suitably vented from the top portion of the air
space. Such air vents should be arranged so as to minimize the possibility
of stoppage by ice formation.
5. If two or more tanks have their outlets interconnected, the air space
in the tanks should also be interconnected. This will prevent the flow of
fuel between tanks that is often caused by differences in pressure at air
vents of each tank of sufficient magnitude.
6. Where large fuel tanks are used, the size of the vent tubes should be
so proportioned as to permit rapid changes in internal air pressure,

thereby preventing collapse of the tanks in a steep glide or dive.
7. Each fuel tank should be provided with a pump and a drain located
at the lowest point when the airplane is in the normal position on the
ground. The main fuel supply shall not be drawn from the bottom of this
pump.
THE POWER PLANT
281
shut-off cocks, and strainers. Typical oil systems are shown in Figures
XVI-16 and XVI-17.
Tanks
1. The oil capacity of the system should be at least 1 gallon for every
16 gallons of fuel but should not be less than the minimum specified for
safe operation of engine. Commercial operators use the formula: oil re-
quired equals 10 gallons plus 1 gallon of oil for every 20 gallons of fuel.
2. The oil tank should allow for at least 10 per cent volume over that
required for the oil alone to provide for expansion space.
3. Oil tanks must be capable of withstanding an internal test pressure
of 5 pounds per square inch without failure or leakage.
4. Oil tanks are made of the same materials as those used for the fuel
tanksâ€”2S, 3S, or 52S aluminum alloy with wall thicknesses varying from
0.040 to 0.065 inch.
AIR INLET SYSTEMâ€”TURBOPROP ENGINES'
The following discussion offers a number of design considerations
applicable to all turboprop-engine installations and gives an insight into
similar problems for other jet engines.
The induction system for the turboprop engine should be designed to
obtain the most uniform pressure distribution possible and the maximum
pressure recovery possible at the inlet to the compressor. Engine power
output and fuel consumption are affected by both of these factors, and
engine performance will suffer if the inlet distribution is poor, or the
pressure recovery is low.
Induction System Configuration
There are a number of general induction system arrangements which
may be considered for the turboprop engine. The more common ones are:
1. Hollow or ducted spinner arrangement, Figure XVI-18A.
2. Conventional cowl with a conventional spinner installed over the
propeller hub to provide an annular nose inlet, Figures XVI-18B, C,
and D.
3. Undercowl scoop and plenum chamber, Figure XVI-18E.
Design Considerations
1. Hollow or Ducted Spinner. The ducted spinner is believed to be the
best inlet design for the turboprop engine as far as internal airflow and
aerodynamic characteristics are concerned. Pressure recoveries of over
95 per cent and very uniform inlet pressure distributions have been ob-
tained in wind-tunnel and flight tests with ducted spinner configurations
designed specifically for the engine.
1 Information presented in this section is based upon material in the Installation
Handbook for the PTSG Turboprop Engine, by permission of Pratt and PPhitney Aircraft.
282
See text for


discussion. (Courtesy Pratt and Whitney Aircraft.) â€¢

Figure XVI-18A, B, C, D, and E. Air inlets for turboprop engines.
THE POWER PLANT
283
The ducted spinner has some disadvantages, however, which must be
considered. It is necessarily heavier than most of the other designs and
may be difficult to maintain in service. Also, suitable spinner de-icing
systems must be developed for this design. The decision as to whether
to use the ducted spinner for a given installation should be made by weigh-
ing the engine performance gains which will be obtained against the me-
chanical difficulties which may be encountered.
2. Conventional Propeller Spinner. The conventional propeller spinner
configuration is not as good as the ducted spinner from an airflow and
pressure recovery standpoint. It is usually considered to be mechanically
superior, however, because it" is lighter, simpler to construct and main-
tain, and easier to de-ice. The pressure recoveries which can be achieved
with a conventional spinner arrangement will vary, depending upon such
design factors as: the spinner contour; the propeller cuff thickness; the
width of the gap between the spinner surface and the bottom of the pro-
peller cuff; the distance from the trailing edge of the propeller cuff to the
cowl'inlet; and the ratio of the area at the cowl inlet to the area at the
compressor inlet. Wind-tunnel tests indicate that recoveries as high as
85 to 90% of the dynamic pressure can be obtained with a good design.
(a) Spinner Contour. A large percentage of the pressure losses which
occur ahead of the cowl inlet with a conventional propeller spinner occur
as the air flows through the propeller shanks or propeller cuffs. These
losses may be appreciably increased if a serious adverse pressure gradient
is encountered in the region of the propeller cuffs or if any serious separa-
tion occurs at the surface of the spinner. For this reason, care should be
taken to choose a spinner design which will make provision for the preven-
tion of these disturbing conditions. Wind-tunnel design information
(such as that supplied by the NACA or similar agencies) should be utilized
in choosing the spinner shape to help insure that the desired spinner con-
tour is satisfactory in this respect.
There are three principle spinner configurations which may be con-
sidered:
(1) The conical design, Figure XVI-18B.
(2) The elliptical or streamline designs such as those recommended by
the NACA, Figure XVI-18C. .

(3) The modified conical spinner with the shortened and rounded nose,
Figure XVI-18D.
For high-speed subsonic airplane applications, it is essential that the
cowl, spinner, and propeller shanks be designed for high critical Mach
numbers. In this respect, it is necessary to keep local velocities over the
spinner and propeller cuff platforms as low as possible.
From an aerodynamic standpoint, the conical-type spinner is slightly

284
better in certain respects than the streamline or elliptical spinner. The
principal advantages of the conical spinner are that (1) it can be operated
with somewhat lower cowl inlet-velocity ratios than the elliptical spinner
before serious separation occurs, and (2) the design of the conical spinner
is somewhat less critical than the elliptical type.
The elliptical or streamline spinner is usually preferred, however, since
it is difficult to design a conical spinner that will clear the propeller hub
without detracting from some other feature. Experience indicates that
when a conical design is used, in order to avoid interference with the
propeller hub, it is generally necessary to use either a low cone angle and
make the spinner quite long, or a high cone angle and make the cowl
diameter large. It is possible to avoid this difficulty by using a modified
conical spinner consisting of a low-cone-angle conical spinner cut short at
some point forward of the propeller blades and fitted with a rounded nose.
Limited wind-tunnel tests indicated that some loss in pressure recovery
will result from the nose of a conical spinner being shortened and rounded;
but these tests were very limited, and it is possible that this loss could be
avoided with an improved design. Any modified conical spinner design
should be based on wind-tunnel tests to be sure that the rounded nose
does not cause separation.
Wind-tunnel experience indicates that there is little difference in the
pressure recoveries obtained with streamline- and conical-type spinners
throughout most of the operating range, providing the contour of the
streamline spinner is properly chosen for the conditions involved. The
recoveries obtained with the streamline spinner will be low, however, if
the contour permits separation. Whatever the spinner contour chosen,
it is recommended that the design be based on wind-tunnel information,
and that the final design be checked by wind-tunnel tests whenever
possible.
(b) Inlet-Velocity Ratio. For any given cowl and spinner design,
there is an optimum inlet-velocity ratio, Vi/v0 (ratio of velocity at cowl
inlet to airplane velocity), at which maximum pressure recovery will be
obtained at the engine inlet. When the inlet-velocity ratio is increased
above the optimum, the inlet-duct losses increase, and the pressure re-
covery decreases. When the inlet-velocity ratio is decreased below the

optimum, the diffusion and separation losses ahead of the inlet increase,
causing a decrease in pressure recovery at the engine inlet. In working

out the inlet design, it is necessary to investigate the inlet-velocity ratios
at which the inlet will operate under all important flight conditions, and
to choose the inlet area and cowl and spinner design so as to obtain the
best possible performance compromise for the various operating condi-
tions. It is usually considered desirable to design the inlet for maximum
pressure recovery at the high-speed or normal-cruise condition (which-

THE POWER PLANT
285
ever is more important for the particular airplane involved) and to accept
some loss at climb conditions.
For any given design there is usually a critical inlet-velocity ratio below
which separation losses ahead of the inlet become quite severe and pres-
sure recoveries become relatively low. Care should be taken to design
the system in such a manner that this critical inlet-velocity range is not
encountered at any flight condition.
The engine air-consumption data needed to compute inlet-velocity ratios
may be obtained from the engine performance curves.
In designing the spinner inlet configuration for the engine, the following
points should be noted.
1. Wind-tunnel tests have shown that the optimum cone angle for a
conical spinner is about 40 degrees. This angle is not unduly critical and
may be varied plus or minus 10 to 15 degrees with only small losses in
performance.
2. When a conical spinner is used, the cone should be continued all the
way into the cowl inlet. The surface should not be curved ahead of the
cowl inlet as separation may occur and pressure recoveries will be reduced.
3. Pressure recovery is decreased as the propeller cuff or shank thick-
ness ratio is increased, particularly at high-speed conditions. The pro-
peller shank or cuff should be made as thin as practicable.
4. Round propeller shanks ahead of the cowl inlet are undesirable.
Pressure recoveries at critical Mach numbers will be lower with round
shanks than with faired shanks. Cuffs or fairings should be used.
5. The gap which is required between the spinner surface and the lower
edge of the propeller cuff to permit the blade to turn through its normal
pitch range has an adverse effect on pressure recovery which may become
quite serious when large gaps are used at high speeds. These adverse
effects may be minimized by attaching short fixed cuff sections to the
spinner surface to move the junction between the fixed and movable
portions of the propeller blades radially outward away from the boundary
layer area so that the wake from the junction does not enter the cowl
inlet under any design condition. Tests indicate that this allows good
pressure recoveries to be obtained even at relatively high air speeds.
For best results, the clearance gap between the platform and the base of
the propeller blade should be kept as small as possible.
6. Whenever possible, the cowl inlet area should be as large as or slightly

larger than the engine inlet area, i.e., a constant area or contracting duct
should be used between the cowl and engine inlets. A diffusing duct
should be avoided, as duct losses tend to be greater with diffusing systems.
For some applications, it may be necessary to make the cowl inlet smaller
286
than the engine inlet in order to keep inlet velocity ratios within the de-
sired range under high-speed conditions. Where this is necessary, good
diffuser design practice should be followed in this duct.
3. Undercowl Scoop and Plenum Chamber. There has been no experi-
ence with the undercowl scoop and plenum chamber type of induction
system with the engine. In general, it is believed that this type of system
would not be very satisfactory for this engine because the distance be-
tween the propeller and engine inlet is too short to permit a good duct
configuration, and, as a result, there are high internal losses. The inlet
flow conditions should be good with this design since the air which enters
the inlet does not flow through the propeller shanks, but it is difficult to
obtain good internal flow with this type of system.
Inlet Losses during Ground Operation
In designing an inlet for the turboprop induction system, care should
be taken to design the cowl lips to avoid separation and excessive inlet
losses during static ground operation. For some applications, it may be
difficult to design a cowl lip which will be satisfactory for both static and
high-speed operation. Static ground losses are important and should not
be overlooked since the power available for take-off will be reduced if
these losses are excessive. Present indications are that this problem is
not serious with good ducted-spinner or good conventional-spinner con-
figurations.
Engine Inlet Anti-Icing Provision
In order to prevent icing of the engine air inlet section, provisions are
made for bleeding compressor discharge air through passages in the com-
pressor inlet guide vanes and the four upper air inlet struts. The two
lower inlet struts are not subject to icing since oil circulates through
these during engine operation.
When icing conditions are anticipated, the anti-icing valve, located in
the bleed duct at the right of the compressor section, should be opened to
permit flow from the compressor discharge. The rate of flow is limited
to the amount sufficient for anti-icing purposes. The four anti-icing
air-discharge ports are located at the outer ends of the four air inlet
case struts.
Anti-Icing Water Runback into Engine Inlet

In considering an operative engine in flight icing conditions with engine
and airplane anti-icing system in operation, attention must be given to the

likelihood of a small amount of water runback into the engine inlet.
Engine testing on this has been confined to simulated icing conditions at
low temperatures of about 10Â° to 15Â° F. Results of such tests indicated

THE POWER PLANT
287
that no ice formation was evident on the inlet-passage outer wall between
struts or from struts aft to the inlet guide vanes; this was with engines
being run with no cowling. With an actual installation in an airplane,
some additional heat from within the nacelle will be available to this
outer wall. Anti-ice heat in engine struts and inlet guide vanes easily
prevented any buildup at these points. There was evidence of consider-
able heat transfer from the hot air and oil in struts to case areas between
and aft of the struts. Anti-icing of the inner passage wall was satis-
factory with transfer of heat from the propeller reduction gearing.
Simulated icing in these tests utilized a very high water-air concentra-
tion and, as noted above, no ice buildup was experienced. This con-
centration was considered to be greater than that which might be experi-
enced in actual flying during a rain at freezing temperatures, when the
ice accretion rate is at one of its highest values.
The aircraft manufacturer must take the responsibility for providing
anti-icing systems for the spinner, cowl, propeller equipment struts, and
propeller. De-icing systems wherein the ice is alternately allowed to
build up to a certain degree and then broken off should not be used for
these components since some of the ice particles may enter the engine
inlet and cause damage to the compressor blades.
AIR INLET SYSTEMSâ€”TURBO-JET ENGINES
The advent of the various types of jet engines requiring such enormous
amounts of air has increased the importance of the design of the induction
systems. Not only have important gains been made in obtaining good
ram recovery, but there has also been a marked impact on the amount
of drag (in relation to the entire design) that may be accepted.
Airplanes powered with jet engines obviously require well-coordinated
and integrated aerodynamic, thermodynamic, and structural designs.
From an aerodynamic standpoint, an ideal induction system would
deliver air to the jet engine:
1. Without reduction in available energy;
2. Without increasing the over-all aerodynamic resistance of the air-
craft and so detracting from the high-speed characteristics of the airplane;
3. Without interfering with a uniform air distribution at the point of
delivery.
The induction system performance may be adversely affected by any
or all of the following factors:

1. Separation of the airflow over the duct lips, due to stalling, through
a relatively wide range of angles of attack;
2. Local induced velocities approaching critical Mach numbers;
3. Thickening of the boundary layer, leading to an inherent total pres-

288
sure loss (a positive pressure gradient ahead of the inlet and continuing
throughout the length of the diffuser will result in an increasing thickness
of the boundary layer and may induce separation along the duct walls);
4. Changes in the direction of flow throughout the system;
5. Non-uniform flow, or cross-flow, or alternate pulsating flow caused
by more than one duct feeding into the system;
6. Unusual conditions leading to increased skin friction.
Inlets may be classified as:
1. Nose inlets, located in the nose of a fuselage or power plant, pod, or
nacelle;
2. Wing inlets, located along the leading edge of the wing, usually at
the root for single-engine installations;
3. Annular inlets, encircling, in whole or in part, the fuselage or power
plant, pod, or nacelle;
4. Scoop inlets, which project beyond the immediate surface of the
fuselage or nacelle;
5. Flush inlets, which are recessed in the side of a fuselage or nacelle.
For supersonic induction systems, additional classification may be
made, such as:
1. The external compression type, using either a ramp or a conical
inlet;
2. The internal compression type, employing either the so-called two-
dimensional or three-dimensional design.
The amount of air required for a jet power plant will he specified for
the engine and is a function of the amount of fuel consumed as well as
the amount of cooling air required.
For aircraft speeds up to about sonic, the air velocity in the induction
system varies from three tenths to nine tenths of the aircraft speed or
Vi _ air scoop inlet velocity ^
Vâ€ž airplane velocity
This ratio is known as the inlet-velocity ratio, and for preliminary design
purposes may be assumed to be about 0.5. For fighter aircraft, the ratio
is between 0.6 and 0.7.
To determine the quantity of air by volume required, the following
relationships are helpful:

V -Q
where Vi = air scoop inlet velocity, ft/sec,

Q = induction airflow, cu ft/sec,
A = air scoop inlet area, sq ft.

THE POWER PLANT
289
where Q
Q=-
pg
, cu ft/sec,
W = induction airflow, lb/sec,
p = inlet air density, slugs/cu ft,
g = acceleration of gravity, 32.2 ft/sec2.
W may be estimated from the following approximate air-consumption
values for the non-jet reciprocating type engines:
7.0 to 7.2 lb of air per horsepower hour for best economy mix-
tures,
6.1 to 0.8 lb of air per horsepower hour for best power mixtures,
G.6 to 7.0 lb of air per horsepower hour for rich best power
mixtures.
For the jet-type engine, the specific air consumption can be calculated
on the basis of the air-fuel ratio, which for complete combustion is about
15 to 1 by weight. The total weight of air required for combustion can
then be calculated:
where e is the specific fuel consumption (SFC) per pound of thrust or
equivalent shaft horsepower (eshp), as specified by the manufacturer.
It is given either as lbs/lb of thrust/hr or lbs/eshp/hr, depending upon
the type of engine used.
Some comments should be made on various types of air inlets for turbo-
jet engine installations in aircraft operating at less than supersonic speeds
in order to indicate the extent of the engineering compromises that have
to be made.
The Wing-Root Inlet
Advantages claimed for this type of installation are:
1. Excellent energy recoveries may be obtained through a satisfactory
range of angles of attack.
2. The critical Mach number for the inlet can be made as high as for
the wing.
3. Boundary-layer removal can be accomplished relatively easily.
4. The fuselage remains uncluttered by ducts and therefore allows

greater room for radar and cockpit installations.
5. There is practically no form drag added to the airplane.

W = 15(e) (T or P),
W = total weight of air in pounds required per hour,
T = thrust rating of jet engine,
P = eshp of the turboprop engine.

290
Figure XVI-19. The wing-root inlet.
Some of the unfavorable aspects of this type of installation are:
1. The wing structure is quite complicated, especially since the wing-
root section is a critical structural design problem even without ducts.
2. Unless engines are installed in the wing root, resulting bends in the
duct may lead to high energy losses.
3. Retraction space for the landing gear may be made less available
by the presence of the ducts.
Nose Inlet in the Fuselage or Nacelle
The apparent favorable features of this type are:
1. It is free of boundary layer effects.
2. Except for large angles of attack, there are few flow complications.
3. Excellent energy recoveries are possible.
4. High critical Mach numbers may be obtained.
5. It is considered an optimum design insofar as inlet aerodynamic
characteristics are concerned.
6. It is ideal for nacelle installation, especially when nacelles are below
the wing.
The disadvantages of this type of installation are:
1. The duct system may be excessively long.
2. The duct system may be of an undesirable shape.
3. The design may lead to excessive weight.
4. The design may lead to structural complications.
5. If it is used in the nose of a fuselage, vision is restricted since a more
aft location of the cockpit and wider nose of fuselage are usually required.
6. Radar equipment cannot be located in the fuselage.
The Nacelle or Pod-Type Installation
1. It is ideal, but there is more drag for the nacelle.

THE POWER PLANT
291
Figure XVI-20. Nose inlet in fuselage or nacelle. The nacelle or pod installation
is favored for multi-engine designs because it is considered less of a fire hazard, and
also keeps the exhaust heat away from the wing structure.
Figure XVI-21. The annular inlet. Such inlets may extend circumferentially only
partially or entirely around the fuselage.

292
2. Wing-nacelle interference is more difficult to change than air in-
duction, and therefore other means of introducing air may be better.
3. It makes it difficult to achieve high critical Mach numbers.
4. Angle of attack effects.2
The total pressure distribution at the inlet of the engine is usually very uniform with
a pod-type installation under normal level flight conditions. However, the distance
from the inlet to the engine is usually so short that disturbances generated at the inlet
by high angle of attack or yaw operation are not damped out before they reach the
engine, and relatively large disturbances can occur at high angles of attack. As a
general rule airplanes which use pod-type installations are not normally designed to
operate at extreme angles of attack so this problem may not be serious. However, in
designing pod installation it should be recognized that pressure distributions at the
engine inlet can be affected seriously by high angle of attack operation.
Every effort should be made to be sure that pressure distribution is satisfactory up
to the maximum angle at which the airplane will operate during maneuvers. The
adverse effect of angle of attack on inlet distribution can be reduced by using relatively
thick lips on the lower part of the inlet although this may be objectionable on high speed
aircraft because of drag penalties. In some cases it may be necessary to compromise
on lip design, even where some drag penalty is involved, in order to insure satisfactory
power plant operation during maneuver conditions.
The adverse effect of angle of attack on pressure distribution at the engine inlet also
can be minimized by making the distance from the cowl inlet to the face of the engine
long enough to permit distrubances created at the inlet to be damped out before reaching
the engine.
The Annular Inlet
Here, the advantages, especially when compared with the nose type, are:
1. Good energy recoveries are possible.
2. A high critical Mach number for the inlet may be achieved.
3. The boundary layer offers little trouble.
4. Visibility may be better than for the nose inlet.
5. Placement of equipment in the nose of the fuselage is not inter-
fered with.
The disadvantages considered are:
1. The design may lead to long and complicated duct configurations.
2. There are certain structural penalties.

3. More aerodynamic drag usually results.
4. Rearward location along the fuselage may be critical.

The External Scoop
The apparent advantages are:
1. May produce the lightest design.
2. May least affect internal arrangements and structure.
3. May be desirable for installations where speeds below transonic are
likely to be encountered.
2 Quoted from Pratt & Whitney JT3 Installation Handbook.

THE POWER PLANT 293
Figure XVI-22. The external scoop. Such a scoop may also be located on top or
bottom of the wing surface for multi-engine designs.
4. Can be located outside of the boundary layer of the structure to
which it may be attached.
The disadvantages encountered are:
1. An increase in form drag,
2. A relatively low critical Mach number.
3. Due to the changes in bends, high duct losses may be expected.

294
(a)
(b)
Figure XVI-23. The flush inlet. In order to suck away the boundary layer ahead
of the main inlet, a small secondary inlet is used, as shown in the schematic sketch in
Figure B.
The Flush Inlet
If properly installed, it may have these advantages:
1. Relatively low weight.
2. Structural simplicity.
3. Low form drag.
4. Relatively straightforward ducts.
The disadvantages are due primarily to:
1. Location in a region of thickened boundary layer, necessitating some
boundary layer control, or
2. Location outside of the normal air stream so that duct bends may
lead to energy losses and lower critical Mach numbers.
Matching the air inlet system of the airframe to the turbo-jet engine
requirements presents a more difficult problem as speeds move into the
supersonic region. Some of the problems encountered are:
1. Need for high inlet pressure recovery.
2. Undesirable flow instability at Mach numbers over 1.5.
3. Flow pattern distortions caused by the forward part of the induc-
tion system.
4. Duct rumble due to flow separation ahead of side inlets at low air
flows, separation at inlet lips at high angles of attack.
5. Excessive spillage.
6. Reasonable efficiency of airflow and energy recovery over a relatively
small portion of the operating range.

THE POWER PLANT
295
Research is still progressing in this area, and the following are some of
the suggestions which have been offered:
1. A freely rotating blade can be installed, under certain conditions, to
improve severe velocity gradients in an annulus at the compressor face.
The rotating blade transfers energy from regions of high axial velocity to
those where the velocity is low.
2. Screens placed across the inlet also are effective in reducing dis-
tortion, but the relatively large energy losses may make screens a last-
resort method for solving the problem.
3. A diffuser, when followed by a section producing a rapid accelera-
tion, reduces the distortion at the compressor face. Effectiveness appears
Table XVI-4. Representative reciprocating engines.
Fuel
Oil
Rated
con-
con-
Ratio of
No. of
Cylinder
horse-
Normal
sump-
sump-
Weight
weight
cylin-
arrange-
power
rpm
tion,
lb/hp/hr
tion,
lb/hp/hr
to horse-
ders
ment
power
65
2300
.49
.010
176
2.70
Horiz. Op.
85
2575
.51
.010
184
2.20
Horiz. Op.
90
2475
.52
.010
184
2.07
Horiz. Op.
100
2550
.50
.015
230
2.30
Table XVI-5. Representative gas turbine (jet) engines.
Maximum
Fuel con-
Maximum
Maximum
Dry
thrust, in
pounds
sumption,
diameter,
length,
weight,
lb/hr/lb
inches
inches
pounds
790
.55
â€”
â€”
315
920
1.13
â€”
â€”
364
1000
1.65
22
96
370
3250
1.06
24
120
1233
4600
1.14
49
94
1790
6970
â€”
37
145
2554
7650
â€”
37
288
3196
9000
â€”
37
148
â€”"
10000
â€”
48
191
3829
Jet engines with thrusts up to 25,000 lbs or more are available.
to depend partly upon the mixing process, although it involves energy.
In addition to the diffusion process, losses and space available for a dif-
fuser are usually limited.
4. A relatively straight cylindrical section of the duct, when placed
ahead of the compressor face, results in suitable mixing in the duct, thus
reducing distortion. Here again, space limitation is a factor in the
application.
298
ROCKETS
The rocket does not depend upon the surrounding air for its operation,
but carries its own, usually as a component of an unstable compound or
as a readily releasable element of a compound. The burned fuel pro-
duces high-velocity exhaust, producing thrust which increases with alti-
tude. It is considered the most effective of the extremely long-range and
high-altitude power plants.
For aircraft, it is used for the additional power required to take off
(and called "Rato," for Rocket-assisted take-off), as well as for other
short-duration booster requirements. As higher speeds are desired, the
limitations on high speed, all other factors being equal, may be the take-off
speed and take-off run, since there is a definite relationship for any given
design between the take-off speed and the high speed. Additional power
at take-off would be one way of decreasing the take-off distance and
attaining the necessary take-off speed in a very short space of time.
For each additional 1000 pounds of thrust at 60 miles per hour, there is
an additional 160 horsepower available for take-off. This is equivalent to
more than 200 brake horsepower for a propeller-operated aircraft, when
it is considered that the propeller efficiency is of the order of 80%.
There are two types of rocketâ€”one using a solid propellant, the other a
liquid propellant.
The solid-propellant type rocket is simple in construction and operation.
A solid fuel is contained in a steel bottle having an igniter such as black
powder for the pyrotechnic agent that furnishes hot gases and solid
particles to cause the solid fuel or propellant to ignite. The combustion
then proceeds at a controlled rate, governed by the area of the exposed
burning surface and the nozzle throat. The nozzle is used to control the
proper working pressure and to provide the rate of expansion of the hot
exhaust gases to the atmosphere.
This type of rocket is simple in construction since it has no moving
mechanical parts, but once in operation, it cannot be throttled or shut off
when desired.
The liquid-fuel rocket motor is more efficient but more complex in its
construction and operation. Two fuel systems are usedâ€”the mono-
propellant and the multipropellant.

A monopropellant, such as nitromethane or hydrogen peroxide, is in-
jected into the combustion chamber. These fuels are unstable, hard to
handle, and have relatively low performance characteristics.
In the multipropellant system, an oxidant and a fuel react in a com-
bustion chamber. For repeated operations, hyperbolic or self-igniting
liquids are preferredâ€”such as hydrogen peroxide with a calcium per-
manganate catalyst, or red fuming nitric acid with aniline. Non-hyper-
bolic fuel systems are liquid oxygen and alcohol, or gasoline and an acid.
Design of the Wing
CHAPTER XVII
Qln spite of its seeming simplicity, the wing of an airplane requires the y
most careful study in its design for it is the vital part of the airplane which
contributes to the lift. ) A slight change in any one of the factors affecting
the wing design is so critical that only one slight difference in two other-
wise identical designs would be apparent. These differences might be in
performance, such as the top speed, or landing speed, or in the rate of
climb. Again, there might be differences in stability00â€”one might be more
stable longitudinally, directionally, or laterally; or there might be differ-
ences in maneuverability.
Pilots who fly externally braced biplanes will attest to the fact that rig-
ging the wings differently by tightening lift wires or adjusting struts often
will change an otherwise beautifully maneuverable airplane to one that is
"logy." The change of rigging may have altered the angle of incidence
of the wings, and therefore the decalage, so that entirely different charac-
teristics of the combination are obtained.
From the foregoing statements, it should be apparent that it is not the
wing alone that should be considered but its relationship to the rest of the
airplane in combination with the fuselage and tail surfaces.
The aerodynamicist is concerned with the ultimate performance of the
airplane in flight, and therefore he is particularly interested in the aero-
dynamic characteristics of the wing. The structural designer of the wing
is concerned with the design of a structure that can meet the aerody-
namicist's requirements as well as withstand the loads imposed upon it.
It is therefore important to consider some general design aspects before
any specific structural details are studied.
The wing is essentially a beam that is subjected to shear, bending, and
torsion imposed upon it by aerodynamic and inertia loads. Figures
XVII-1 and XVII-2 indicate typical spanwise loadings and moments
imposed on the wing by the normal and chord components of these forces.
299
300
CN, Ce, or
Figure XVII-1. Reference diagram for the development of equations for bending
and torsional moments for a typical wing.
In Figure XVII-1,
or
AFâ€ž = qCN
AFâ€ž = q
144'
CnCx
144'
if dx = 1 inch, and represents the load at any station x inches from the
tip of the wing. Similarly,
Cxdx
AFC = qCc
144
DESIGN OF THE WING
301
or
and
AMac = qCmaCz2
in inch-pounds.
The spanwise loads AFn, and AFC cause shears and bending or torsional
moments which may be expressed as follows for any station n:
y=n v=n y=n
y=i=o y=i=o y=i=o
y~n y=n y=n
y=i=o y=i=o y=i=o
y=n y=n
Mz = ^(AFn)y = ^ ^CNCyydy;
y=i=o y=t = o
y-n y=n
M, = ^(AFc)y = j^^CcCvydy;
y=t=o y=i=o
y=n y=n
y=i=o y=i=o
Due to sweepback, the normal forces cause a torsional moment which
may be expressed as follows:
MJ = â€” ^\ &Fnxin = â€” AFâ€žyin tan A = -Mx tan A.
|fâ€”iâ€”o y=t=o
Due to dihedral, the chord forces cause a torsional moment which may
be expressed as:
y-n y=n
Â» = ÂFCZ<â€ž =^ÂFcyin tan r = M, tanr.

z or normal axis
Figure XVII-2. Representation of shears and moments, shown as positive, at any
spanwise station of the wing. The aerodynamic center is taken as the point of origin
of the system of axes shown. The subscripts of the moments are determined by the
designation of the axis about which the individual moment acts.
In Figure XVII-2, M,' is the total of all these torsional moments or:
M,' = Mâ€ž + MJ + Mac", or
M/ = Mac - Mx tan A + M, tan V,
for the configuration shown. Similar formulas may be derived for the
inertia load components.
It should be noted that the bending moments Mx and Mz are inde-
pendent of sweepback, dihedral, or the locus of the aerodynamic or elastic
centers of the wing.
The bending stress caused by the moments Mx and Mz at any station
may be found from the formula:
, (MJâ€ž - MJZ\ (M.Iâ€ž - MJ.\
fb = { u. - ij )x + { uu -1j ) *.
For purposes of subsequent discussion, let it be assumed that Mz is
insignificant with respect to Mx (which is usually the case, as can be seen
by comparing the numerical value of Cc with CV), and that the principal
axes of the cross section of the beam are known. Then, the above formula
becomes the simple bending formula:
t - Mxz
f" = â€”f
In a complex beam, such as the wing structure presents,
Ix = I0 + pLAz? = -Â£Az?,
DESIGN OF THE WING
303
Chord line
-Original or real location of skin
Principal
(b)
Figure XVII-3. (A) shows the principal components of a wing structure. (B)
represents the equivalent wing structure where the solid line segments represent the
shear elements and the solid circles the bending elements required to resist the transverse
and torsional shear and the bending moments.
from which it may be noted that all the longitudinal stringers, as well as
the chord members of the spars, will contribute to the strength of the
wing. The skin, if suitably supported so as to be able to sustain com-
pressive stresses, can be very effective in adding to the beam strength.
Figure XVII-3 gives a typical representation of the structural elements
important in the design of a suitable wing.
The torsional moments are likely to be quite severe for a swept-back
wing. If the wing has a negative dihedral, then the torsional moments
are all of the same sign and therefore of considerable magnitude. The
wing envelope must sustain these torsional moments.
In general, the bending and torsional moments are the design deter
minants for the wing structure.
<
The first three-view, the preliminary weight estimate, and the arrange-
ment of the balance diagram are the necessary steps in furnishing the data
for the type of wing for the position relative to the fuselage, and for its
size. The preliminary three-view has more or less set the shape of the
wing and determined whether it is to be a monoplane, a full cantilever or
an externally braced wing, or, for example, a biplane with wings of differ-
ent proportions.
The preliminary weight estimate is instrumental in determining the
approximate wing area needed, so that with at least this established, it

304
becomes a comparatively easy matter to select a suitable aspect ratio and
thus be able to fix the governing dimensions of span, chord, and taper.
The balance diagram is necessary to locate the wing relative to the cen-
ter of gravity; otherwise, difficulty might be experienced later in obtaining
suitable static longitudinal stability.
The wing planform may be changed considerably owing to certain re-
quirements of landing-gear retraction, flap attachments, and the like.
Suppose it is desired to retract the landing gear straight inboard towards
the fuselage without the necessity of swinging it back first and then in-
board in order to retract the gear fully into the wing without interfering
with the front spar? Such retraction requires that the root portion of
the wing be somewhat forward of the leading edge of the mean geometric
chord; or, in other words, the wing should have the leading edge swept
back so that the root chord will come forward along the fuselage. The
reason for this is that there are two original conditions that must be met
by the wing and the landing gear. (The wing, for example, should be placed
so that the 25 per cent point of the mean geometric chord falls directly
under the center of gravity. ^^The landing gear, if a rear tail-wheel type
is used, on the other hand, must be placed at a certain angle ahead of the
center of gravity to prevent nosing over] These conditions for the wing
and landing gear must be kept, and, unfortunately, these conditions may
play havoc with original ideas of wing planform and simple landing-gear
retraction.
The incorporation of flaps and ailerons often affects wing planform.
For simpler operating mechanisms, it may be desirable to have the hinge
lines perpendicular to the plane of symmetry of the airplane, or perhaps it
may be desirable to have a constant chord flap whose spanwise axis is per-
pendicular to the plane of symmetry. Both of these more or less arbitrary
conditions will affect the ultimate planform of the wing. It is a good plan,
therefore, to list at first all the various ideas that the designer wants to
incorporate, and then, perhaps, make preliminary sketches of possible
solutions to determine whether the various ideas are compatible.
Before the design of the wmg may be begun it is necessary to study
various features which affect its final design. The more important fea-
tures are considered here although it is impossible to point out all the pos-
sible effects of miscellaneous items such as landing lights, engine nacelles,
landing gear, and fuel tanks.

WING LAYOUT PROCEDURE
Since the wing design has to allow for so many factorsâ€”planform, spar
and stringer location, landing-gear attachment and retraction, power plant,
ailerons, flaps, and a host of othersâ€”it is desirable to make preliminary
studies to make sure that every design feature has been properly in-
DESIGN OF THE WING
305
corporated. The following suggestions may be followed to advantage.
The instructions are for a two-spar wing and may be modified for any
other type of design.
005
1. Draw planform of wing with the necessary dimensions, to scale, to
satisfy aspect ratio, area, and sweepback.
2. Determine the mean geometric chord and check that the relation
of the wing to the fuselage is such that the center of gravity lies in the
lateral plane perpendicular to the mean geometric chord at the mean
aerodynamic center.
3. Locate the front spar at a constant percentage of the chord, from
root to tip. The front spar is located at between 12 to 17 per cent of the
chord. Note that the constant percentage line of the chord is not parallel
to the leading edge of the wing. Indicate the spar location by its center
line.
4. Locate the rear spar similarly. The rear spar is located at from 65
to 75% of the chordâ€”usually 70% to accommodate a 25 per cent chord
aileron.
Neither the front nor rear spar need extend to the extreme wing tip,
since the extreme wing-tip structure is inherently rigid and capable of
transmitting tip loads to the spars and adjacent structure.
5. Mark out the aileron. The leading edge of the aileron may be
parallel to the rear spar centerline. If the rear spar is located at 70 per
cent of the chord, then the aileron chord cannot exceed about 27 per cent
of the chord, since some allowance must be made for rear spar width,
aileron gap, and the like.
6. If a flap is used for a lift-increase device, it may extend the entire
distance inboard of the aileron. Here some additional study may be
necessary if a considerable flap area is desired. In that case, the aileron
chord might be increased, even if such an aileron has some adverse charac-
teristics. Increasing the aileron chord may necessitate moving the rear
spar slightly forward to give sufficient clearances. The flap chord is
usually less than the aileron chordâ€”perhaps 15 to 20% of the wing chord.
A false spar, needed to support the flap hinges, may have to be added to
the wing structure.

7. The wing rib spacing may be spotted in next. There are certainly
likely to be ribs located at each aileron and flap hinge (unless a piano-type
hinge is possible). The aileron must have at least three hinges for sup-
port. The rib-spacing distance is determined from panel-size considera-
tions, to which reference should be made. Some adjustments in the rib
spacings may be desirable to get hinge-rib locations to coincide with the
rib stations. Reinforced ribs are also called for for engine-mount attach-
ments, landing-gear-member attachments, and fuel-tank supports. Also,

306
such ribs may suggest relocations of the other ribs in order to obtain a
more pleasing pattern.
8. Spanwise stringers may be located. These may be placed parallel
to each other or at constant percentages of the wing chord. These span-
wise stringers are not normally carried out to the tip, but are rather
discontinued at intervals inboard of the tip so that fewer and fewer stringers
are left from the mid-span outboard.
9. The main elements of the wing structure have now been located. If
the layout is considered satisfactory, other details may now be added.
Some of them may cause reconsideration of the structure locally. For
example, the well for the retraction of the landing gear may necessitate
some re-design.
10. The layout outlined above is shown primarily in outline form by
means of centerlines, with subsidiary sketches showing structural details.
Until the stress analysis is completed, such dimensions as thicknesses,
limiting dimensions, and the like should be omitted. After the stress
analysis is completed, a more detailed drawing is possible.
Before the structural layout has progressed too far, it may be wise to
check on the effect of the spar and spanwise stringer location on the
fuselage structure. If the front spar necessitates the installation of a
double frame just where the pilot has to be provided with the best vision,
it may be wise to relocate the spar, or even change the wing planform.
If the wing fuel tanks interfere with the landing-gear retraction, some
re-design considerations are in order.
In considering the wing layout, it is necessary to keep the requirements
of the fuselage, the power plant, and the landing gear constantly in mind.
MEAN GEOMETRIC CHORD
The center of gravity of the complete airplane is placed, usually, at the
maximum forward position of the center of pressure on the mean aero-
dynamic chord in order to get the desired stability.
The mean aerodynamic chord is difficult to determine unless the pressure
distribution for the wing being designed is definitely known. Moreover,
the pressure distribution varies with the angle of attack. It is customary,
therefore, to use the mean geometric chord of the wing instead.
The mean aerodynamic chord, or the mean geometric chord, is deter-

mined for only one-half of the wing, either up to the side of the fuselage
for a wing whose center portion is blanketed by the fuselage, or up to the

plane of symmetry as in the case of a parasol monoplane.
The mean geometric chord of a rectangular wing is located just halfway
out along the semispan out from the fuselage.

308
Figure XVII-6. Geometric determination of the mean geometric chord for a simple
trapezoidal wing.
If the tip of the wing is tapered or rounded off, an equivalent semispan
can be determined such that the area included in the equivalent semispan
is equal to the area excluded.
A geometrical method for finding the mean geometric chord of a trape-
zoidal planform wing is shown in Figure XVII-6.
The length of the mean geometric chord may also be calculated by
means of the formula:
The distance out from the side of the fuselage or plane of symmetry may
be calculated from
_ b (C + 2CQ
X ~ 3 (C + CO'
where b is the dimension as shown in Figure XVII-5, and the distance in
rear of the leading edge of the root chord
_ yi (C + 2d)
y ~ 3 (C + Ci)"
DESIGN OF THE WING 309
When the wing is made up of a rectangular and a tapered portion, the
procedure is to
1. Determine the mean geometric chord of the trapezoidal and the
rectangular section separately.
\ r* yn
it
<>
â€”r <o
*l
'2
"3
Figure XVII-7. Geometric determination of the mean geometric chord for a com-
bined rectangular and trapezoidal planform. If the wing is elliptical in planform, an
equivalent trapezoid may be used.
2. If the area of the rectangle is A, and the trapezoid A2 then the mean
geometric chord of the combination is
â€ž _ CR At + Ci A,
t2~ Al + A2'
Its distance from the mean geometric chord of the rectangular section is
If the root section is not rectangular, the method is similar.
The mean aerodynamic chord may be determined also by the elemental
strip method. The location of the chord out from the side of the fuselage
or plane of symmetry may be expressed mathematically:
fCixds _ fCixCdx _ 2(CixC)

fddS fCiCdx 2(C,C)'

Iâ€”*â€”I
h X -|
Figure XVII-8. Reference diagram for mathematical calculation of the mean aero-
dynamic chord.
The integration is best done mathematically by choosing an elemental
strip of width dx = unity. If the airfoil used is the same from root to tip,
and the angle of incidence of each strip is the same, Cl will cancel in the
numerator and the denominator. C is the chord at the distance x from
the reference plane.
PLANFORMS AND TAPER RATIOS
The taper ratio of wings varies greatly, depending upon the speed range,
span, aspect ratio and other similar parameters. For structural con-
siderations, it should be obvious that if no other factors need to be con-
sidered, a triangular planform assuming a uniform lift distribution per
square foot from root to tip would cause less bending moment at the root
than a rectangular wing of equal area. A triangular planform has an
aspect ratio of 26/C, where b is the span and C is the chord at the longi-
tudinal plane of symmetry; its geometric chord is equal to 2/3 C. To
obtain good performance for subsonic airplanes, an aspect ratio of at
least 6 is to be expected. Thus the result here would be a root chord of
one third the spanâ€”an unusually large chord at the root which brings
into its wake several problems of structural arrangement control and
stability. Certainly, in the conventional airplane using the conventional
horizontal and vertical tail surface arrangements, the aircraft would be-
come unusually large.

Leading edge of wing
perpendicular to plane
of symmetry.
Trailing edge perpendicular to
plane of symmetry.
Front spar perpendicular
to plane of symmetry
Locus of aerodynamic centers
perpendicular to plane of
symmetry
Rear spar perpendicular to
plane of symmetry
Figure XVII-9. Variation in wing planforms due to specified initial conditions.
Figure XVII-10. A wing planform employing two different angles of sweepback to
reduce the effect of tip stall.

312
Figure XVII-11. A double delta-wing design for obtaining optimum relative posi-
tions of the aerodynamic center and the center of gravity. Inner delta incorporates air
duct for jet engines. A low thickness ratio can be employed for the outer delta. In
spite of the low aspect ratio and small thickness ratio, the delta wing, because of its large
root chord, offers a relatively large volume and depth for ducts, structure, and retraction
of landing gear.
J Ideally, the wing planform should be planned so as to give an elliptical
1 lift distribution at subsonic speeds. All other things being equal, an
elliptical planform would therefore be in order. However, because of the
structural problems presented by such a planform, it is seldom used, except
for some fabric-covered wings. The closest compromise to the elliptical
planform for aerodynamic reasons, and to the triangular planform for
structural reasons is, then, the trapezoidal planform.
Figure XVII-12. Three planforms of wings with pronounced sweepback.

DESIGN OF THE WING
313
Typical taper ratios of root chord to tip chord are 2 to 1, 3 to 1, and
3 to 2, with intermediate values. Rectangular planforms are also in the
running.
To help solve some of the different aerodynamic problems posed by very
high-speed aircraft, triangular planforms and inverse taper ratios have
been proposed and used. Some of these are discussed under "Sweepback."
Some unusual planforms are illustrated in this chapter. See also Figures
XVII-9, 10, 11, 12.
SWEEPBACK FOR HIGH-SPEED AIRPLANES
For aircraft operating at high subsonic speeds, the use of the sweepback
in the planform of the wing is favored in order to increase the critical
Mach number of the wing. Since the thickness ratio t/C is measured for
the section parallel to the plane of symmetry, it can be made smaller.
If, for example, a rectangular wing planform of chord C and thickness t
were swept back an angle A, the apparent thickness ratio, instead of t/C,
would now be </(C/cos A) or t cos A/C. For a tapered wing planform,
the apparent thickness ratio would be even less. Therefore, by sweeping
the wing back, the critical Mach numberâ€”which is a function of the
thickness ratioâ€”can be increased, while still maintaining a reasonable
thickness or depth for structural considerations.
However, unless steps are taken in the design of the wing, a highly
swept-back wing is liable to experience tip-stalling effects. As the angle
Figure XVII-13. Suggested design limits for aspect-ratio determination based upon
effect of sweepback angle on tip-stall tendencies.

314
of attack of the swept-back wing is increased, the wing will stall at the
tips first, thereby reducing the lift at the tips. Since the lift on the swept-
back tip has a large moment arm from the center of gravity, the pitching
moment is materially reduced so that a "pitch-up" occurs, increasing the
angle of attack still further. In addition to stalling the entire wing, the
tip of the wing losing the lift first will drop, and a spiral dive will ensue.
Sweeping the wing back produces two aerodynamic results. For the
ordinary non-swept wing, the pressure distribution spanwise drops off
towards the tip, whereas for the swept wing, the reverse is true. The
increased pressure distribution indicates that the lift coefficient is greater
at the tip than inboard, and therefore the tip of the wing will stall before
the inboard section.
Concomitant with the pressure distribution, there is an outflow from
the longitudinal plane of symmetry or a drift of the boundary layer to-
wards the tip that tends to further complicate the problem of lift dis-
tribution.
The tip-stalling tendency seems to be more pronounced with high-
aspect-ratio wings than with low. Figure XVII-13 presents some design
limits.
Various compromises are incorporated in the wing design to ameliorate
the tip-stall tendency. Such compromises may include (1) low aspect
ratio with large angle of sweepback, which leads to poor performance, (2)
wing fences on the top wing surface to reduce or prevent the outflow of
the boundary layer, (3) wash-out of the wirig tip, (4) choice of special
airfoils, (5) design of wing-fuselage combinations to promote stalling of
the wing near the root rather than at the tip, and (6) special planforms
and aero-elastic designs such as the inversely-tapered planforms, the
"crescent" wing and the "isoclinic" wing, and the "cranked" wing.
Figure XVII-14. A variable-sweep wing would employ a normal, partially-swept
configuration for low subsonic-speed operations and a more swept-back version for higher
speeds. The aerodynamic center of the mean geometric chord would be used as the
approximate pivot point. A root cuff to accommodate the various positions of the wing
would be used to streamline the section. A complicated spar mechanism would have
to be constructed to permit variable sweep and to sustain the loads imposed upon the
structure.
Figure XVII-15. A wing planform incorporating reversed taper as well as sweep-
back.
To obtain better control, stability, and performance throughout the
flight range, designs have incorporated a variable sweep, making it possible
to attain anything from a wing planform with no sweepback to a planform
with the maximum sweepback. This compensates for shifts in center of
gravity and presents the optimum configuration for performance through-
out the flight regime. A forward position is desirable for take-offs and
landings, while a rearward position is best for high-speed flight. Figure
XVII-14 is a schematic illustration of the variable swept-back wing.
Figure XVII-15 demonstrates that an inversely-tapered wing counter-
acts the reduction in lift distribution spanwise by committing more area
Figure XVII-16. The "crescent" wing.

316
Locus of aerodynamic centers
Locus of elastic centers of beam
of high torsional rigidity
Locus of elastic centers of /
beam of low torsional A
Figure XVII-17. The "aero-isoclinic" wing.
to the tip section and by increasing the chord length at the tip to increase
the Reynold's number for a better lift coefficient. The distribution of
the lift farther outboard and the decreased depth of structure at the root
do involve some structural penalties. With the inversely-tapered wing,
there is less fuselage-wing interference at the root.
Figure XVII-16 illustrates the "crescent" design developed in England.
It employs three sections having different sweepback angles: 50Â° at the
leading edge for the root section; 40Â° for the middle section; 30Â° for the
tip section. The wash-out due to bending in the middle section is offset
by the wash-in due to torsion caused by the loads on the tip section. Be-
cause of the smaller sweep of the tip section, the possibility of aileron
reversal is minimized, since the torsion of that section alone is less.
Figure XVII-17, shows the so-called "aero-isoclinic" wing. It is similar
to the "crescent" wing except that the different sections of the wing
incorporate varying degrees of torsional rigidity in order to counteract
or reduce the wash-out effects produced by the torsional moments imposed
upon the wing due to sweepback and particularly tip-stalls.
Figure XVII-18. A "cranked" wing design proposed to counteract effects of "tip-
stall" and consequent "pitch-ups."

DESIGN OF THE WING
317
Effect of Wing Sweepback
Some discussion of sweepback has already taken place in considering
effective dihedral. In general, it may be said that sweeping of the wing
back decreases the maneuverability of the airplane since aileron control
is impaired and the damping moments in roll are impaired.
The effects of wing and fuselage mass distributions, wing loading, and
wing incidence and sweepback are increasingly important for high-speed
airplanes approaching or exceeding sonic speeds. They are also impor-
tant, although to a lesser degree, for lower-speed airplanes.
DIHEDRAL
The amount of dihedral that must be given to a wing must ultimately
be determined by wind-tunnel tests.
A wing equipped with flaps will probably require more dihedral than
one without since the span is relatively shorter for the former than for the
latter. However, it is not a question of lateral stability alone but of the
relation of lateral to directional stability as well.
It is customary to give at least 3 degrees up to as much as 6 degrees
dihedral to the wing. This may be given to the entire wing from the
fuselage outward or from the stub wing outward. Again, in the latter
case more dihedral would be required than for the former case.
It is difficult to determine what the effective dihedral is for a wing ta-
pered in planform and thickness. The angle measured between the plane,
passed through the mean aerodynamic chord and parallel to the locus of
the aerodynamic centers of the airfoils, and the horizontal plane, will give
the accepted value.
In Figure XVII-19, (A) illustrates a negative dihedral angle often used
to counteract the effect of a high degree of sweepback. Some airplanes
Figure XVII-19. Various configurations showing dihedral. The angles have been
exaggerated in magnitude to show contrast. The dihedral angle is measured between
the horizontal plane and a plane through the locus of the aerodynamic centers of the
airfoils.
318
show this droop appearance because of the aerodynamic twist of the wing
rather than the dihedral. The usual negative dihedral varies from 3
to 6 degrees.
(B) combines a negative dihedral for the root section and a positive
dihedral for the tip section. Such an arrangement may be employed to
obtain shorter landing-gear members or sufficient clearance for a nose
propeller. The positive angle of the dihedral in such a design is greater
than in a wing that is entirely at a dihedral angle. This serves both to
counteract the negative dihedral and to provide sufficient effectiveness for
the smaller portion of the wing affected. The positive dihedral would
vary from 5 to 10 degrees, depending upon the conditions involved.
(C) shows no provision for dihedral for the root section, but greater
dihedral for the tip section. Such a design might be employed to obtain
suitable space in the root section for landing-gear retraction or cargo,
fuel, or several of these.
(D) represents the dihedral angle applied to the full wing span. The
angle varies from 3 to 8 degrees.
COMBINATION DIHEDRAL AND SWEEPBACK
For high-speed designs whose maximum speeds are approaching the
velocity of sound, extreme sweepback is used to obtain lower effective
Mach numbers, as noted in Chapter III. The extreme sweepback of the
wings has necessitated a decrease in built-in dihedral so that in some cases
negative dihedral is required. The incorporation of a negative dihedral
gives the front view of an airplane a "droop wing" effect.
/ The dihedral for a conventional airplane improves the lateral stability
since in an asymmetrical gust, the airplane is rolled over into a sideslip
with the descending wing operating at a higher angle of attack, and there-
fore, greater lift than the ascending wing. The effectiveness of the dihe-
dral for a given airplane is measured by the rate of change of the rolling-
moment coefficient with the angle of sideslip.
When the wing is swept back, the effective dihedral increases rapidly
with change in the lift coefficient of the wing so that the effectiveness is
excessive. To decrease this, negative dihedral has to be employed in
combination with large sweepback angles. The combination results in
practically no effective dihedral at low angles of attack or high-speed flight

and reasonably effective dihedral at high angles of attack or low speed, or
take-off.
The effect on directional stability of the airplane (measured by the rate
of change with angle of sideslip) of increased negative dihedral is improved
with increasing angle of attack, while practically no effect is obtained on
directional stability by increasing positive dihedral when the wing incor-
porates sweepback.
DESIGN OF THE WING
319
Moreover, as far as pitching moments are concerned, increasing positive
dihedral with swept-back wings tends towards longitudinal instability,
while increasing the negative dihedral with a swept-back wing tends to
increase the longitudinal stability.
ANGLE OF INCIDENCE
Theoretically, an airplane designed for high speed should have its wing
set at such an angle to the fuselage that the combination will give the least
possible drag. This setting is difficult to determine without wind-tunnel
tests, due to unknown interference effects.
.^fa good compromise is to set the wing at an angle to the longitudinal axis
of the fuselage corresponding to the angle at which minimum drag occurs.
Other considerations may be the deciding factors such as flap clearance
in the deflected position, or the inconvenient ground angle for the fuselage
when in the three-point landing position. The former case usually requires
a smaller fixed angle of wing setting while the latter requires a larger angle
of incidence.
To obtain a desired lift distribution of the wing especially near the tip
where premature stalling is undesirable, as well as to change the stalling
characteristics of different sections, the various airfoil sections are given
different angles of incidence which result in the so-called aerodynamic
twist. Formerly, wash-in and wash-out was resorted to for the wing-tip
section only, but it is now customary to study every section along the
wing and to develop the change of angles of incidence carefully so that cer-
tain specific results may be obtained.
A wing is said to have zero aerodynamic twist when the zero lift lines
of all airfoils from the root of the wing outboard are parallel.
High-speed aircraft require low drag of the wing-fuselage combination,
especially at cruising speed at high altitude, with the result that the angle
of incidence of the wing with relation to the fuselage axis is relatively large.
The resulting angle happens to be useful in simplifying the landing-gear
design since the fuselage ground angle is reduced as well.
However, the large angle of incidence of the wing causes the airplane
fuselage to fly "nose-down" at high speeds. This negative inclination of
the fuselage tends to reduce the damping of oscillations resulting from a
sideslip, while the reverse is true when the inclination of the fuselage is
positive. The difference in the fuselage axis and the wing chord axis
causes coupling of inertia forces in such a manner that a rolling accelera-

tion produces a yawing moment and a yawing moment produces a rolling
moment.
For high-speed flight at high altitude, the thin wings have a relatively
larger angle of attack than might be expected for the subsonic designs.
In order to reduce the over-all drag of the wing and the fuselage, the wing
320
is set at a relatively larger angle of incidence to the fuselage so that at
high speeds the combination will present the optimum configuration.
The angle of incidence is dependent upon the airfoil chosen and the aspect
ratio of the wing. Whereas for subsonic designs, the angle of incidence
of the root chord to the wing is of the order of 1, 2, or 3 degrees, for tran-
sonic and supersonic designs, the angle of incidence is about twice those
values.
WING LOADING
For high rates of climb and high service ceiling, relatively low wing
loadings (W/S) are required, especially for the subsonic range.
If no lift-increase device is employed, the limitations on the wing load-
> ing are (1) the maximum lift coefficient obtainable, and (2) the permissible
landing speed, which is likely to be about the same as the stalling speed.
The personal or small executive airplane is likely to dispense with lift-
increase devices because of the additional cost and the reduction in the
margin left for cargo and pay load. These aircraft are likely to have
wing loadings between 10 and 18 pounds per square foot.
For commercial airplanes, high speedâ€”if any one performance charac-
teristic can be singled outâ€”is probably the determining factor in design.
Assuming the power or thrust is available, then in order to achieve the
high speed, relatively high wing loadings are necessary. Thus, for this
class of airplanes, one may expect to find wing loadings of from 18 to 80
pounds per square foot. To obtain reasonable landing speeds, lift-
increase devicesâ€”most of which are more commonly known as wing
brakesâ€”have to be incorporated into the wing design.
Military airplanes in the predominantly high-speed category are not
necessarily limited in landing speedsâ€”certainly not to speeds of the order
of magnitude of 65 to miles per hour. To obtain high top speeds, it is
also necessary to raise the landing speeds. Here, the wing loadings are
likely to be exceptionally highâ€”although lift-increase devices will prob-
ably not be usedâ€”so that values between 45 and 70 pounds per square
foot are to be expected. Wing loadings up to 150 pounds per square
foot are not uncommon.
ASPECT RATIOS
Aspect ratios for wings operating at subsonic speeds vary from 6 to 12,
) with the higher aspect ratio usually indicated for turboprop-powered
aircraft operating at high altitudes. In order to make large aircraft

economical, relatively high wing loadings are required, resulting in rela-
tively small wetted surface for the lifting surface as compared with the
fuselage and engine nacelles. To offset the larger proportion of the
parasitic drag, the induced drag can be materially reduced by incorpora-

DESIGN OF THE WING
321
tion of the high aspect ratio. Increasing the aspect ratio from 6 to 12
cuts the induced drag in half. A better lift over drag ratio (L/D) for the
complete airplane may be expected throughout the entire flying range,
and therefore better performance is a result.
For the personal or private business airplane, aspect ratios between 5
and 8 are common.
Aspect ratios for high-speed airplanes operating at transonic or greater
speeds vary with the designs. Some idea of the selection of the aspect
ratio may be obtained from Figure XVII-13 which indicates the relation-
ship between aspect ratio and sweepback angle. Based on data for
American transport airplanes, with gross weights between 5,000 and
150,000 pounds and operating at wing loadings between 18 and 80 pounds
per square foot, the aspect ratios vary between 6.5 and 12, with the higher
aspect ratios for the higher wing loadings. The following linear relation-
ship gives a reasonable approximation:
For speeds between 500 and 650 miles per hour, foreign fighter airplanes
have aspect ratios ranging between 4 and 7 for wing loadings between
44 and 72 pounds per square foot. The relationship between them may
be expressed as:
^ = 10.4 AR.
For aircraft operating at speeds between 400 and 800 mph, the follow-
ing empirical formula based on thrust loading seems reasonable:
AR = 3 + 1.175 ~
AIRFOIL THICKNESS AND THICKNESS RATIOS
In considering the relative suitability of an airfoil for locating the front
and rear spars of a two-spar system, the thickness ratios at various sta-
tions along the chord with reference to the maximum thickness ratio
may be examined.
Since
d
322
and
/ Â« bds,
where d/2 is the distance from the neutral axis to the outermost fiber of
the section under consideration, where in this case d is the depth of the
airfoil at station under consideration,
then
f ~L
h d2
A "measure" may be devised by obtaining the ratio of d2 at any station
to the value of d2 for the maximum thickness of the airfoil.
Table XVII-1. Data calculated for a few representative airfoils.
Chord
NACA
23018
NACA
0018
NACA
63018
NACA
643018
station
dydma2
d*/dmax2
%
5.0
10.72
35.8
10.68
35.0
8.724
23.4
8.372
22.8
7.5
12.61
49.5
12.60
48.8
10.616
34.3
10.152
33.7
10.0
14.05
61.3
14.04
60.7
12.136
45.5
11.606
43.7
15.0
16.04
80.1
16.04
79.2
DESIGN OF THE WING
323
WING-FUSELAGE CONFIGURATIONS
A discussion of various wing-fuselage configurations has been given in
Chapter II, Types of Airplanes. There is also some discussion of this
subject in Chapter XXI, The Fuselage. Every configuration has certain
advantages and disadvantages which the designer must evaluate.
In the delta-wing designs, the fuselage becomes an integral portion of
the wing structure over the entire root-chord length since a multi-spar
design is usually employed. Moreover, due to the length of the root
chord, the depth of the wing becomes appreciable and easily accommodates
the fuselage requirements.
METAL WING CONSTRUCTION r
Since the wing (considered as a structural unit) is essentially a beam,
different solutions are theoretically possible for a suitable internal struc-
ture. Figure XVII-20 presents a number of these likely solutions, ar-
ranged in order of increasing complexity of configuration.
Figure A may be used for high-speed aircraft employing relatively thin
wings. Its advantages are:
1. Distribution of material as far away from the neutral axis as possible.
2. May be tapered in cross section span wise, either through milling to
the proper thicknesses or through building up of excessive layers properly
bonded to each other.
3. Especially suitable for lifting surfaces having no attachments or
movable control surfaces.
The disadvantages may be listed as:
1. Difficulty in disposing control rods, cables, and the like in the interior.
2. High weight, which may be attributed to the small depth available,
since this type of construction would probably not be utilized for deeper
wing sections.
Figure B is a modification of Figure A. The spanwise stiffeners are
integrally incorporated with the skin. This type of construction may be
utilized for deeper wing sections where the skin may be somewhat less,
but may need spanwise support.
Figure C is a design long considered. Variations of it have been used,
particularly for gliders.
Its advantages are:

1. The vertical shear-web spar may be located at the chord station of
greatest depth rather than at a station of less depth and may therefore
sustain the bending moment most efficiently.
2. The torsional moment may be taken care of by the wing envelope.
3. The number of individual units is small.

Figure XVII-20. Different methods of wing construction.
The disadvantages of the type of construction shown in Figure C may be
enumerated as:
1. The addition of aileron and flaps necessitate a supporting spar which,
for optimum design, might as well work in conjunction with the main
spar. Thus, the ultimate design of this type of structure would lead to a
two-spar structure.
2. The wing envelope has to be attached to the fuselage side in order to
avoid transferring torsional moments to the spar, which, unless it is itself
of box construction, would be an inefficient torsional member. The skin

DESIGN OF THE WING
325
is of no benefit in carrying the wing bending moment. There may be
handling difficulties with thin skins.
3. The skin may form wrinkles when the spar bends and thus present a
poor aerodynamic form.
4. The skin is of no benefit for sustaining any bending moments, and
may therefore be a contributor of inefficient material leading to excess
weight.
Figure D is a two-spar or multi-rib design with the chordwise ribs spaced
relatively close together. There are no spanwise stringers or stiffeners,
and the skin is subject to torsional stresses only.
The favorable aspects of this structural configuration are:
1. No stringer cut-outs in ribs, thereby reducing fabrication difficulties
and the number of individual parts.
2. Minimum thicknesses, and therefore minimum weight of the sheet
covering.
3. Minimum weight penalty for cut-outs because the skin is less highly
stressed than in other forms of construction where it has to withstand
both shear and compression or shear and tension stresses.
4. Simplicity of construction of a two-spar system with load concentra-
tion in the spar chords leads to simpler fittings, whether for attachment
to the fuselage or a wing folding joint, or for landing-gear attachments.
There are some disadvantages, such as:
1. Relatively smaller effective depth for the spars, which are usually
located at about 15% and 70% of the chord.
2. Closer rib spacing is needed to carry the air loads into the spars
which are the main bending moment resistant members.
3. Undesirable wrinkling between ribs, leading to poor aerodynamic
form.
4. Possible handling difficulties with thin skin unless the skin thickness
is increased to enable it to carry some compressive stress.
Figure E illustrates one of the more popular types of construction used
in reinforced monocoque designs. Its advantages are:
1. Utilization of the skin in bending as well as in torsion.
2. Readily available means for resisting wrinkling or buckling of the
skin by supporting it in smaller panel sizes.

3. Relatively fewer ribs or formers and correspondingly fewer fabrica-
tion difficulties.
4. Greater opportunity for distributing stresses caused by cut-outs,
attachments, and local loads.
5. Rugged construction leading to easy maintenance and handling.
No structure is without some attendant disadvantages. For Figure
E, these are:
326
1. Tie-in with fuselage structure offers difficulties.
2. More riveting and bolting required than for other types.
Figure F depicts a multi-spar type which has these advantages:
1. Unusually rugged construction suitable for rather large aircraft.
2. Relative ease in taking care of cut-outs, attachments, and the like.
3. Possible "fail-safe" design since a local failure can be completely
localized.
4. Wing deflections are less than for any other type shown.
5. Wrinkling of skin is minimized.
6. Number of former ribs may be reduced.
Among the disadvantages are:
1. More fabrication.
2. Less accessibility to the interior due to greater compartmentization.
3. Possible greater structural weight due to inefficient use of material.
The final structure of the wing is dependent upon its final arrangement
on the location of the aileron, the flap, the fuel tanks, the retractable land-
ing gear, and any other items peculiar to the particular design.
Figure XVII-21. A cutaway sketch showing one type of metal wing construction.
In order to save weight, tapering of the wing cover as a function of the
loading must be considered for the thin wings operating at transonic or
supersonic speeds. Some of the recent designs have had thicknesses of
1% inches and more at the root section. For missile designs, such wing
covers may be cast from a magnesium alloy. However cast wing covers
are not used for manned aircraft receiving repeated use, primarily because
of the relatively low strength characteristics and relatively poor resistance
to vibration.
328
Figure XVII-24. The "tension-field" spar on the left, or modifications thereof,
are in common use. The one on the right shows a "shear-type" spar with lightening
holes.
DESIGN OF THE WING
329
the action of the limit load, the spar or beam would be considered a normal
shear-resistant beam; if allowed to buckle, it would be considered a ten-
sion-field beam. A pure tension-field beam is difficult to obtain, and
most airplane-spars fall in the semi-tension-field category.
Since the dimensions of a spar obviously depend upon the loads im-
posed and the depth of the wing available, it is difficult to give many
guiding dimensions. However in general, a tension-field beam would have
a shear web made of sheet material at least 0.025 inches thick and chord
members with thicknesses two or more times that of the sheet. For a
shear-resistant beam, extruded sections are common where the thickness
of the web is about the same order of magnitude as the chord members.
Truss-type spars (See Figure XVII-25) have been used only rarely in
recent designs. For small aircraft, the truss-type spar has some merit as
far as weight is concerned.
For a mono-spar wing, the spar would be located at the maximum
depth of the airfoil.
Figure XVII-25. Suggested truss-type spar construction. Trusses are used rela-
tively infrequently.
For a two-spar system, the front spar is located at from 12 to 17 per
cent of the chord, with the rear spar located at from 65 to 75 per cent of
the wing chord. While analyses could be made to determine the optimum
spacing of a two-spar system on the basis of center of pressure travel in
the normal flight regime and the load factors encountered at low and high
angles of attack (since not only the magnitude but also the distribution of
330
Figure XVII-26. A bolted fitting holding two sections of a spar together.
the loads has to be considered), the above percentages of chord locations
are common.
For a multi-spar design, it would be desirable to have the shear center
of the structural system coincide with the center of pressure. Since the
center of pressure of the wing varies with the angle of attack and there-
fore may vary as much as 25 per cent of the chord (more with trailing
edge flaps), it is difficult to pinpoint the desirable location of the shear
center.
Combination of spars and stringers, as well as the wing envelope, make
up the more common type of wing structures where the spars lose their
identity to some extent and are more commonly referred to as shear webs.
The spars vary not only in depth for the tapered wing, but also in the
cross-sectional area of the spar chords. If the spar chords are milled or
forged, the variation in cross-sectional area can be made without serious
discontinuities. In the built-up spar chords, plates may be added at the
root, and then discontinued at suitable intervals along the span. Since
the spar members will most likely have to be spliced, a reduction in thick-
ness can be made for the adjacent outboard member. If two angles back
to back form a Tee section for the inboard section of the spar, one of the
angles may be discontinued for the outboard section. In case of any
changes, special precautions must be taken to make sure that the dis-
continuities in cross-sectional area are not too abrupt; otherwise fatigue
failures may result.
SPANWISE STRINGERS AND THEIR LOCATION
In an earlier chapter, cross sections of typical spanwise stringers are
shown. In the discussion on "Metal Wing Construction," it is brought

DESIGN OF THE WING
331
out that spanwise stringers serve to add to the bending strength of the >
wing.
The spacing chordwise of these stringers is arrived atâ€”as it is for the ribs
â€”by the optimum proportions of the wing-skin panels, which may be
determined from oil-canning considerations before the stress analysis of
the wing has been completed.
The stringers may be located at a constant percentage loci or spaced
equidistantly apart. Not all the stringers are carried out to the extreme
tip since the wing-tip portion is relatively more rigidâ€”even for the mini-
mum structureâ€”than the inboard sections. Carrying out all the stringers
to the extreme tip would result in an unnecessarily strong wing portion
in that area. Accordingly, the spanwise stringers are discontinued, a
few at a station, from about the mid-span outboard so as to avoid extreme
discontinuities in cross-sectional area. When stringers are discontinued,
they should have their ends attached to a transverse rib.
Usually the positive load factor will cause greater compressive loads on
the top surface of the wing than the negative load factor will cause on the
bottom. Therefore the stringers on the bottom surface can be lighter in \â€¢
gauge or fewer in number.
RIBS AND THEIR LOCATION
The primary functions of the ribs in the wing are (1) to maintain the ~\
chordwise shape of the airfoil, (2) to act as supports for the panels of the
wing skin or envelope, and (3) to transmit the pressures on the wing to the
o V ^ V ^ \/
Figure XVII-27. Cross sections of a metal wing. In the top illustration, corru-
gated sheet with a smooth skin on the outside, serves the same function as spanwise
stringers. In the bottom illustration, truss-type ribs, spanwise C-sections, and tension
field spars show one type of metal wing construction. Plain transverse former ribs of
C-sections may be employed without resorting to any of the truss-type bracing.

332
spanwise bending members. In some cases, they also serve to act as the
supporting members to which landing-gear members or engine-mount
members are attached, or as supports for fuel tanks, control systems, and
localized loads. In turn, such ribs transmit the loads to the spanwise
spars.
The spacing of ribs may be determined by the need to prevent oil-
canning of the skin or by the optimum panel proportions, as discussed in
/ an earlier chapter. For preliminary considerations, rib spacings from 6
( to 18 inches may be assumed.
Ribs are located at or near centerlines of aileron and flap hinges of fitting
attachments and other supports.
Figures XVII-27, 28, 29 show some rib constructions. It is customary
to assemble the ribs in three sectionsâ€”the nose section, the center, and
the trailing edge sectionsâ€”in order to avoid any cut-outs of the spars.
The ribs are notched for the spanwise stringers, which should also be con-
tinuous members.
Unless the transverse ribs have to transmit concentrated loads, the
thickness of the sheet used in their construction may be one gauge lighter
than the skin covering of the wing.
Where concentrated loads are applied to transverse ribs, the rib struc-
ture has to be carefully designed to take the shears and bending moments
imposed.
DESIGN OF THE WING
333
Figure XVII-29. Various metal rib configurations. Flat sheet, beaded and light-
ening holes add to the rigidity of the wing. Truss-type ribs utilize U-sections for
added column strengths for the diagonals. Former ribs, top and bottom, are simplest
in form and are generally satisfactory for wings utilizing occasional ribs of the more
rigid type. The ribs shown are center-section ribs; the nose and tail portions are of
similar construction or utilize methods of other ribs shown for their construction. Ribs
are riveted, by means of integral flanges or separate brackets, to the faces of the spars
or shear webs. The top three ribs show cut-outs for transverse stringers. The skin
covering offers continuity of structure since it is riveted to the ribs and the spanwise
stringers.
334
V--""' (a)
WING-FUSELAGE ATTACHMENTS
The problem of providing attachments for the wing structure to the
fuselage merits considerable attention and detailed study, except for
the flying wing or some delta-wing designs where the fuselage becomes
practically an integral part of the wing structure.
If no other considerations but the structural were involved, it would
be desirable to support a wing not only at the fuselage, but at some point
outboard. The two-spar wing lends itself most conveniently to the ex-
ternal support.
The effects of different methods of wing support are illustrated in Figure
XVII-30.
Figure A shows two external configurations for the same wing sparâ€”
one full cantilever when the lift strut, shown dotted, is supplied.
Figure B illustrates a spanwise lift loading, making no allowance for tip
losses which would be the same regardless of the wing support.
Figures C, D, andâ€¢ E present the spanwise distribution of the bending

DESIGN OF THE WING
335
moment when (in C) the wing is full cantilever; when (in D) the wing is
externally supported and pinned at the fuselage; when (in E) the wing is
externally supported and fixed at the fuselage. The pinned condition is
obtained by the use of a single pin placed parallel to the wing chord. For
a pinned connection, no bending moment can be transmitted to the ad-
jacent structure. For a fixed condition, two or more horizontal pins
placed parallel to the chord are common. A vertical pin accomplishes the
same resultâ€”a many-riveted or bolted connection tying the wing spar to
the fuselage structure. In all cases where the lift struts are employed,
pin connections are used at the ends, with the pin parallel to the wing
chord or line of flight so that only axial loads are imposed upon the struts.
Schematic diagrams of various externally-supported wing configurations
are shown in Figure XVII-31.
Figure XVII-31. Various external support arrangements for a semi-cantilever wing.
The primary advantage of supporting a wing externally is that the re-
duction in bending moments imposed produces a lighter structure. If a
smaller bending moment is imposed, a thinner wing, (i.e., a wing of a lower
thickness ratio) could be used. If the wing can be pin-connected to the
fuselage, that structure will be subjected to axial or direct loads only,
and there will therefore be fewer complications in the design of the fuselage
structure at that point. Any gains in the wing construction may be
offset by the additional supporting structure.
Schematic diagrams of full cantilever wing attachments to the fuselage
are given in Figure XVII-32 and are described as follows:
Figure A. The fuselage transverse frame and wing spar assembled as

DESIGN OF THE WING
339
These spars are spaced at a constant percentage span00wise, although they
may be equally spaced provided the taper ratio of the wing planform is not
too great. In the latter case, equidistant spacing may cause relatively
too close spacing near the root and too large spacing near the tip for
economical design.
In some wings of tapered planform, the spars are not placed either
equidistant or at a constant percentage from the leading edge in order to
obtain some particular structural feature, such as suitable spar attach-
ments at the fuselage or to obtain enough space for a retractable landing-
gear well.
When spars are equidistant, it is possible to slip the ribs over the spars
from the tip end, but modern construction of ribs usually calls for three
sectionsâ€”nose, center, and trailing edge section so that another assembly
method may be used.
Wood or metal may be used for either the spars or ribs or both. Spars
and ribs may be made of the same material.
The ribs are spaced from about 6 to 14 inches apart depending upon the
speed of the airplane. Recommendations for rib spacing for fabric-covered
wings may be obtained from Figure XVII-37.
This spacing is based upon the gliding (or diving) speed and may be
assumed to be about V/i times the maximum speed attained (or expected)
Figure XVII-36. Samples of wooden rib construction for two-spar wings. These
could be made of metal with a little modification, as shown, for example, in the third
illustration from the top.

340
in horizontal flight. The chord line of the ribs is placed parallel to the
axis of symmetry (longitudinal axis) of the airplane. The ribs are usually
parallel to the longitudinal vertical plane of symmetry of the airplane.
A piece of sheet metal or plywood covers the nose section so that the
doped fabric maintains its shape properly. Where rib spacing is quite
large compared to the thickness of the nose covering, extra nose ribs be-
tween the standard ribs may be inserted.
Wherever compression ribs or members are located in the drag truss, or
where there are hinge connections, concentrated loads are applied to the
spar, so that local reinforcements are required. In the case of wood spars,
extra plies are glued to the sides, or a solid block is inserted in the box-type
spar. Metal spars have extra plates riveted or bolted to the sides to trans-
mit the load properly.
For fabric-covered wings, the maximum spacing of rib lacing may be
determined from the accompanying graph. These values are based on
wing designs incorporating leading-edge supports such as plywood, metal,
or subribs.
Figure XVII-37. Graph for determining spacing of fabric attachment. Va is the
design dive speed or design glide speed. (Adapted from C.A.M. 04.)
DESIGN OF THE WING
341
In the slipstream, use 75 per cent factor for lacing and rib spacings, ex-
cept that lacing spacing need not be less than 1 inch; aft of 30 per cent
chord point, when aft portion only is fabric covered, lacing and rib spac-
ings may be increased by 25 per cent. The lacing should be carried to
leading edge when velocity exceeds 275 miles per hour.
An externally braced wing could be either pin-ended or fixed-ended at
the juncture of the wing spar with the fuselage. For pin-type ends, a
male and female fitting with a single horizontal bolt is the usual solution.
No vertical bending moments can be transmitted through such a joint, a
necessary condition for a pin joint. For fixed ends, a male and female fit-
ting with a single vertical bolt or tapered pin for alignment purposes is a
common solution where the bearing loads are not excessive. Another
solution is the use of two horizontal bolts. For transmitting large bend-
ing moments (that is, fixed ends), continuity may be obtained by a series
of bolts, or rivets, or other pin connections.
The lift struts supporting externally braced wings are pin-connected
at their ends with one horizontal bolt parallel to the plane of symmetry at
each end of the strut. No bending moment from the wing should be
transmitted from the wing to the strut.
AILERONS
Figure XVII-38 shows a few possible planforms (A to F) for ailerons,
and some aileron sections (G to T). Figure A of this group shows an
aileron which diminishes in effectiveness at high angles of attack, although
very effective at low angles. Figure B shows the aileron slightly too far
inboard which is particularly undesirable when the maximum span for a
flap as a lift-increase device is required.
Figure C is a bad combination of a relatively good aileron and a bad
wing tip in that the same effect is obtained as for Figure A.
Figures D, E, and F show generally desirable planforms. An aileron
preferably should be not more than 25 per cent of the chord although 30
per cent is common when flaps are used for increased lift.
Figure G is a plain aerodynamic balanced aileron. Figure H shows a
balanced aileron of the Frise type, and Figure I a modified Frise type.
Figure J is the Handley-Page balanced type Figure K is a balanced
aileron with a deflector plate. Figures L and M, both shown in neutral

position, are distinguished respectively by variable and constant slot
widths. Figure N is the piano-hinged type. Figure 0 shows the slot-tip

type. Figure P illustrates the retractable-spoiler type, Figure Q the
floating wing-tip type, Figure R another retractable-spoiler type. Figure
S is a conventional narrow-chord aileron with the flap ahead of it. Figure
T is an upper-surface aileron.
342

DESIGN OF THE WING
343
The next question that arises is whether the aileron should be aerody-
namically balanced. For relatively slow-speed airplanes and possibly
for high-speed airplanes utilizing auxiliary mechanical, electrical, or hy-,
draulic aid, unbalanced ailerons may be used.
The leading edge preferably should not project beyond the normal con-
tour of the wing when deflected, since ice may form on the leading edge of
the control surface.
In order to reduce the manual effort to operate ailerons, a type of servo-
control has been devised whereby a small surface is deflected in order to
move a larger surface.
For control purposes, these trailing-edge flaps or tabs should have as
high an aspect ratio as possible and about 10 to 12 per cent of the total
movable area. A smaller percentage is sufficient when the tab is intended
for trim purposes. Reference to Chapter XIX, Tail Surfaces, should be
made for further explanations.
The aileron's primary function is to rotate the airplane about its longi-
tudinal axis. In general, any method that would increase or decrease the
lift on one side relative to the other would meet the objective. Various
devices have been invented and applied to flying designs. A few may be
briefly mentioned here.
1. The spoiler. A small slat usually raised near the top surface of the
wing near the leading edge. This is still used in conjunction with the flap-
type aileron.
2. Opening ports on one side to destroy the lift. Such a scheme is highly
effective for high angles of attack or low speeds, but ineffective at high
speeds. Moreover, the control system is not a continuous one in that the
control stick would have to disengage the control system for one side be-
fore engaging that of the other side.
3. Floating ailerons which are extensions spanwise at the wing tips
These smalf wings of symmetrical cross section are permitted to float at
zero lift, regardless of the wing's angle of attack, and are therefore effective
for any angle of attack. Such ailerons require relatively more area than
the flap type, and while they may increase the apparent aspect ratio of
the wing, this gain is offset by the parasite resistance of these surfaces.
4. The differentially-operated flap, which is the most commonly used

device.
5. Other devices can easily be envisaged that would meet the require-
ments for rotating the airplane.
The aerodynamic characteristics of the flap-type aileron are defined by
the following equations for:
1. Rolling moments
L = CiSwbq;
344
2. Yawing moments
N = ChSwbq;
3. Hinge moments
H = ChSaCaq;
where L, N, and H are moments expressed in foot-pounds,
C(, Câ€ž, and Ch are, respectively, the non-dimensional rolling, yawing,
and hinge-moment coefficients for the aileron,
Sw is the wing area in square feet,
Sa is the aileron area in square feet,
b is the wing span in feet,
q is the dynamic pressure, in pounds per square foot, at the velocity,
V, of the airplane, and
Ca is the aileron chord.
/ Ideally, the rolling-moment characteristics of the ailerons or equivalent
devices should be high, the yawing moment should be relatively small or
zero, and the hinge moments should be within the normal capacities of
the control system.
To explain further, it is obvious than efficient or most effective aileron
system is one that requires a relatively small proportion of the wing area.
The measure of effectiveness is gained in flight tests, and for the subsonic
region, combined lateral stability and control are usually specified as
follows:
With the rudder locked, and the stick or control wheel thrown hard over,
the airplane should:
1. Reach its maximum rolling velocity in not more than 0.2 seconds
later; *
2. Should attain a rolling velocity such that
g > 0.07,
where p is the maximum rolling velocity in radians per second,
b is the span of the wing, in feet, and
V is the true forward airspeed of the airplane in feet per second;
3. At any speed below 80 per cent of maximum level flight speed, should
attain the desired value of pb/2V without exceeding control forces of 80
pounds for the wheel, or 30 pounds for the stick-type of control.
With the rudder held fixed, and the airplane flying at 110 per cent of its
minimum speed, the craft should not sideslip more than 20 degrees due to
the yaw created by fully deflected ailerons.

DESIGN OF THE WING
345
In a sideslip:
1. The rolling moment obtained should vary smoothly with the sideslip
angle;
2. Aileron control should always be necessary to depress the leading
wing as sideslip is increased;
3. The tendency of the aileron to return to its trim position should in-
crease with increase of sideslip.
To obtain the desired lateral stability, low-wing monoplanes usually
require 4 to 8 degrees more dihedral than high-wing monoplanes.
The greater the lateral stability induced by the dihedral and/or sweep-
back effect of the wing, the larger the aileron area has to be to counteract
such effects.
Ailerons of the flap type produce an adverse yawing momentâ€”adverse
because the moment tends to turn the airplane out of its designated turn.
Variable aileron types produce different magnitudes of such yawing mo-
ments. Differentially-operated ailerons are those whereby the down-going
aileron travels through a greater angle than the up-going aileron.
Figure XVII-39 shows a typical case of the effectiveness of the aileron
changes with the angle of attack.
Figure XVII-39. Typical rolling and moment variation with anuglar deflection of
the ailerons and the angle of attack of the wing.

346
The hinge moments have to be overcome by the pilot at the control
stick or wheel. If they are too large, auxiliary boosting systems actuat-
ing hydraulic or electrical mechanisms have to be employed. Since the
greater the ratio of flap chord to wing chord, the greater the hinge moments
produced, the general compromise is not to exceed an aileron-chord to wing-
chord ratio of 25 per cent. The hinge moments may be reduced in magni-
tude either by forward aerodynamic balancing of the surface or by aero-
dynamic balancing through the use of trailing-edge tabs. The systems
employed for elevators and rudders are applicable here, and reference
should be made to Chapter XIX, Tail Surfaces, for further information
on this subject.
Aileron areas vary from 8 to 12 per cent of the total wing area (including
the aileron area, which is considered as part of the wing area).
Figure XVII-40 gives some guiding dimensions for determining the
cross section of an aileron located in the trailing edge of a wing. The
dimensions may have to be modified since the depth of airfoils vary.
0.005 C
0.1 C J
0.005 C
-M â€”0.25 C
- c-
Radius R variable
Figure XVII-40. Typical aileron (or flap) dimensions. The aerodynamic balance
may vary between 15 and 35 per cent of the aileron chord.

DESIGN OF THE WING
347
For delta-wing designs, particularly, the aileron function is combined
with the function of the elevators so that the control surfaces combining
both functions are known as "elevons." The control system has to be so
arranged that the surfaces can be operated differentially as ailerons, and
together in the same direction as elevators.
LET-INCREASE DEVICES AND SPOILERS
Figure XVII-41A represents a so-called plain flap of the type commonly
used for ailerons. An aileron-flap combination could be employed whereby
the aileron function would take place from any flap position.
In Figure B, a specially-designed slot has been incorporated just ahead
of the flap leading edge to provide a less turbulent airflow over the flap in
the deflected position.
In Figure C, a second slot has been incorporated in the design in order
to obtain a higher lift coefficient and less turbulent airflow aft of the flap.
In Figure D, the Handley-Page slot and flap are shown. This combina-
tion offers a number of possibilities. The slot can be brought out into
position automatically by virtue of the suction pressure at the leading
edge at high angles of attack. If the slot is connected to the flap by a
suitable linkage mechanism, the flap can be depressed at the same time.
The design must ensure fool-proof mechanical operation, since the front
slot, because of its shallow dimensions, is difficult to make reasonably
rigid. Under icing conditions it would have to be heated, lest ice interfere
with its operation.
In Figure E, the Wragg flap or external airfoil flap not only increases the
lift due to flap action, but also increases the effective wing area.
Figure F is a split flap in which the recessed portion may be deflected
downward while the top half remains fixed in place. A variation uses a
movable top split flap as well. This flap, moving up while the other
moves down, acts primarily as a spoiler to create more drag and is there-
fore useful in the landing approach.
Figure G is a split flap which moves rearwardly and down, thus increas-
ing the wing area and providing additional lift action when deflected.
The famous Fowler flap employs a recessed airfoil section for such a flap.
This type gives the highest increase in the maximum lift coefficient ob-
tainable.
There are other devices for increasing lift which are not shown. There
has been some study on a series of smaller recessed airfoils which when
extended give a "Venetian-blind" appearance.
Boundary layer control offers much promise both in increasing the
maximum lift coefficient and in decreasing the drag coefficients. The
advent of the turbo-jet engine promises to solve some of the problems
involved in getting enough air of sufficient pressure from a reliable source.

DESIGN OF THE WING
349
Table XVII-2. Maximum lift coefficients for various lift-increase devices.
(See Figure XVII-41 for reference.)
o, de-
â€¢
Description or
Flap
Maxi-
grees
L/D
NACA
diagram
data
mum
for
at
reference
Cl
CLmax
Clark Y airfoil
Basic airfoil .
1.29
15
7.5
-.085
TN459
Plain flap (Figure A)
Chord .30 Cw
L95_
12
4.0
â€”
TR427
deflected 45Â°
Slotted flap (Figure
Chord .30 Cw
1.98
12
4.0
â€”
TR427
B)
deflected 45Â°
Handley-Page auto-
1.84
28
4.1
â€”
TN459
matic slot
Handley - Page slot
Chord .30 Cw
2.18
19
3.7
â€”
TR427
and flap (Figure D)
deflected 45Â°
Handley - Page slot
Chord .40 Cw
3.36
16
3.7
-.740
TN459 v.
and Fowler flap
deflected 45Â°
Split flap (Figure F)

350
speed, with the result that the horizontal tail surfaces do not operate
under favorable conditions.
â€¢ Since the center of pressure moves rearward along the chord, the ef-
fective tail moment arm is decreased and the moment coefficient about
the original aerodynamic center is increased. The ratio of the horizontal
tail surfaces to the basic wing area is therefore greater for an airplane
using the flaps than for an airplane not using them. Accordingly, the
degree of static stability is different. An airplane designed to have ade-
quate stability when flaps are employed will have more than adequate
static longitudinal stability when the flaps are not in use.
Pertinent NACA reports should be referred to for a list of likely values
to be obtained for the lift, drag, and hinge moment coefficients. These
are given for full-span flaps on test models under two-dimensional testing
conditions. In the actual application, the flaps seldom extend over the
full span of the wing, nor is the wing likely to be identical with that used
in the test. Some qualifications in the data used are therefore always
necessary, but reasonable estimates are still possible until particular tests
are made.
In general, it may be said that the maximum lift coefficient obtainable
for a given flap configuration is possible for any airfoil, regardless of what
its maximum lift coefficient may be when no flap is used. Therefore, an
1.4
0 10 20 30 40 50 60
Flap deflection, in degrees
Figure XVII-42. Typical curve showing rate of increase of maximum lift coefficient
with flap deflection. Curve shown is for a split flap with a chord of 20 per cent of the
wing chord, Cw'.

352
For an airplane equipped with conventional size ailerons, and with flaps
devoted to the rest of the span not covered by the ailerons, the maximum
lift coefficient for the entire wing area is often arbitrarily assumed to have
been increased 40 per cent.
A more analytical method is to calculate the "effective" maximum lift
coefficient (refer to Figure XVII-44), as follows:
[(CLmaz)(areas A + D) +' (1.7 CLmax)(areas B+ C) + (0.1 CLmax)(area E)\
h- [areas (A + B + C + Â£>)] = effective CLmax,
where 1.7 Cimai represents a 70 per cent increase in lift due to the flap.
The area covered by the fuselage is normally not included in the wing
area, but if a flap is made continuous across the bottom of the fuselage,
additional lift results, depending upon the type of fuselage. This addi-
tion has been estimated at one-tenth of the maximum lift coefficient in
the empirical formula just given.
A recent development of a lift-increase device has been a so-called nose
flap or "droop snoot" whereby the forward portion of the wing nose is
rotated downward as much as 60 degrees. Several illustrations are shown
in Figure XVII-45. Wind-tunnel investigations of a laminar-flow air-
foil equipped with such a nose indicated that the lift coefficient could be
increased by about 30 per cent, and almost 60 per cent in conjunction
with a trailing-edge flap.
Figure XVII-45. Nose slats and flaps.
the wing chord.
The chord is usually about 10 per cent of

354
The nose flap also increases the angle of attack at which the maximum
lift coefficient is obtained over that for the basic airfoil, evidently because
the nose flap aligns the airflow better at the stall. However, even so, the
position of the flap is critical.
Another good feature of the nose flap is that the additional pitching
moment created by the nose flap is not so great, at least at low angles of
attack, as by the trailing-edge flap; at high angles of attack, the effective
pitching moment tends to decrease, thereby maintaining approximately
the same longitudinal stability for the airplane without the use of the nose
flap.
Figure XVII-48. "Wing fences" used to prevent drift of boundary layer towards
the tip. Note the "wing fences" on the vertical tail surfaces.
Wing fences have been employed for the transonic and supersonic air-
planes to prevent the outward flow of the boundary layer. These fences
consist of essentially simple vertical surfaces, about 4 inches or less in
height, attached to the top surface of the wing and extending from the
leading edge of the wing to the leading edge of the flap or aileron. Usually
two such fences are used on each half of the wing, sectioning the wing
semi-span into about three equal lengths. (See Figure XVII-48.)
The addition of engines, fuel, bombs, and other similar concentrated
weights along the wing of the airplane increases the rolling and yawing
moments of inertia so that the normal size ailerons provide reduced rolling
WING FENCES
EFFECT OF WING MASS DISTRIBUTION

DESIGN OF THE WING
355
accelerations and thereby require a longer time to attain a given angle of
bank. Likewise, the increased moments of inertia reduce the desired
oscillatory stability, and increase the time for the airplane to damp out
oscillations. It has been found that this effect was small for low values
of effective dihedral but became increasingly undesirable for high values
of effective dihedral, especially when the ratio of vertical tail surface area
to the wing area is small. The solution for high moment of inertia would
be either increased vertical tail surface area with relation to the wing, or
low effective or negative dihedral, or both. Gusts or turbulent air do not
affect an airplane with a high value of moment of inertia as much as one
with a low value of the moment of inertia; but due to poor damping, steady
flight is hard to obtain and rough rides may result.
EFFECT OF WING LOADING
While ordinarily the wing loading is given as the ratio of the gross weight
of the airplane to the wing area, it has recently been pointed out that the
relative density of the airplane (that is, the density of the airplane divided
by the density of the surrounding air) would be more appropriate. The
general effect of increasing the airplane relative density either by increas-
ing the wing loading at sea level or flying it at high altitudes where the air
density is less is the same. In both cases the degree of the lateral stability
is decreased.
The solution would be to increase the ratio of the vertical tail area to
the wing area in order to obtain an equivalent value of effective dihedral.
FLUTTER PREVENTION
Precautions must be taken in the design of the wing to prevent flutter.
Some recommendations are:
(a) Structural stiffness.
(b) Elimination of all play in hinges and control system joints.
(c) Rigid interconnection between ailerons.
(d) A relatively high degree of weight balance of control surfaces.
(e) A relatively low amount of aerodynamic balance.
(f) High frictional damping of control system.
(g) Adequate fillets.
(h) Proper fairing.
Features tending to create aerodynamic disturbance, such as sharp lead-

ing edges on movable surfaces, should be avoided. (These are also prone
to collect ice.)
356
When ailerons are attached to internally braced wings, they should be
statically balanced about their hinge lines. If a balancing weight is used,
it is usually more effective when located near the outboard end of the
aileron.
Partial static balance may be satisfactory when an irreversible and ex-
ceptionally rigid control system is used.

CHAPTER XVIII
The Landing Gear
The lac ding gear consists of the wheels, tires, brakes, shock absorbers,
struts, cowlings and, if retractable, the retracting mechanism. With only
minor exceptions, most of the items making up the landing gear are not
designed by the airplane designer but by the accessory manufacturer.
However, their selection and their relationship are determined by the air-
plane designer so that it is important for him to know the various condi-
tions which the landing gear must meet and the purpose for which each
part has been designed.
The landing of the airplane may be considered in terms of:
1. The approach glide. This is assumed to be a steady or uniform
rectilinear motion along a flight path, making an angle with the horizontal.
2. The transition from steady gliding flight to substantially level flight
before the point of contact. During this operation the rate of descent is
slowed to practically zero at contact; also the flight path speed along the
flight path is decelerated to substantially the stalling speed at contact.
3. The ground run during which the airplane is decelerated horizontally
from the contact speed to zero.
The landing gear is brought into the picture from the moment of contact
with the ground until the airplane comes to rest.
The landing gear must, of course, take the shocks when landing or when
going over an obstruction, and so it incorporates two means of absorbing
shock loadsâ€”the tire which absorbs minor shocks, and the shock absorber
which absorbs hard and sudden shocks. Not only must the landing gear
be able to take these shocks, but it must also be so arranged that the air-
plane will be prevented from nosing over when landing.
357
358
The proper design of the landing gear requires possibly as much or more
attention than any other component of the aircraft, even though the use
of the landing gear contributes such a very small proportion of the useful
life of the aircraft.
In coming in for a landing, there are generally three distinct phases:
the initial or steady-approach glide path; a transition phase of either a
"dead-stick landing" or a flared landing path; and the final ground run.
These are depicted in Figure XVIII-1.
Figure XVIII-1. Various stages in the landing approach of an airplane.
The difference in the landing load factor between the "dead-stick"
landing approach and the flared-path approach is insignificant. However,
the attitude of the airplane may affect the distribution of the loads, mak-
ing it necessary to investigate various conditions to make sure that the
landing-gear members are sufficiently strong. Such components as
wheels, tires, and brakes are not investigated by the airframe designer
since their selection on the basis of static load assures that strength re-
quirements have been amply met.
For the tail-wheel type of landing gear, the basic landing conditions
cover (1) the level landing attitude with inclined reactions at the main
wheels, and (2) the tail-down landing with vertical reactions at the wheels.
In addition (3) a brake-roll condition, and (4) a side-load condition are
investigated to cover the use of brakes and the landing occurring from a
side-slip.
DYNAMIC LOADS
Normal or "flared" landing path
Ground run
THE LANDING GEAR
359
For the nose-wheel type of landing gear, the basic landing conditions
are (1) level landing with inclined reactions at the nose and main wheels,
(2) level landing with inclined reactions at the main wheels, and the nose
wheel just off the ground, (3) the tail-down landing. Subsidiary condi-
tions take care of (4) braked landing, (5) obstruction load, and (6) side
loads, as well as several supplementary conditions for the nose wheel only.
The maximum load factor encountered in landing depends upon the
landing-gear characteristics and the method of landing. For the typical
commercial airplane, the limit vertical inertia load factor is one obtained
with a descent velocity, in feet per second, of
V = 4.4 (W7<S)1'Â«,
but need not exceed 10 feet per second nor be less than 6 feet per second.
Other restrictions may be imposed, such as that the inertia load factor
used for design purposes should not be less than 2.67. Since the aircraft
designer does not necessarily desire a load factor greater than 2.67, he
attempts to design the landing gear so that the load factor is not likely
to exceed the minimum specified value.
In the main, the load factor is determined by the rate of energy dissipa-
tion. To keep the load factor as low as possible, the designer would
ideally favor a long stroke for the shock absorber. The action of the
shock absorber is similar to that of catching a ball: in order to soften the
impact of the ball hitting his hand at a relatively high speed, the catcher
draws back his hand while catching the ball.
Compromises in the design of the shock-absorber stroke are inevitable.
If it is unusually long, the extended landing gear offers difficulties in re-
traction. Moreover, the motion of the airplane upon landing might be
uncomfortable for the passengers. Therefore, the stroke of the shock
absorber is generally limited to between 6 and 8 inches for normal air-
craft, and from 8 to 12 inches, or sometimes more, for helicopters.
MEANS FOR LANDING-SPEED REDUCTION
It would be desirable to reduce the approach speed and the landing-
run speed, thereby correspondingly reducing the kinetic energy and, in
turn, the work of the shock absorber and the wheel brakes. Some of the
means utilized to accomplish a possible reduction are:
1. Aerodynamic. To reduce the sinking velocity, high lift-increase

devices may be employed just before the landing run starts. Not only
is the sinking speed changed, but the drag is increased, thereby reducing
forward speed. High lift-increase devices, when employed in this manner,
are often referred to as air brakes.
It is also possible to increase the drag by having flaps added to the

360
fuselage or by splitting the vertical tail surfaces and opening them up to
present a greater flat-plate surface.
Boundary layer control methods are being extensively investigated for
possible use in obtaining high lift coefficients to reduce the approach
speed.
Various suggestions regarding the use of jet compressors for obtaining
high-volume, high-velocity air, available either as vertical thrust and thus
lift or for boundary layer control, seem to offer great possibilities for re-
ducing landing speeds to manageable proportions.
2. Reversible Propellers. Discussion of the use of reversible propellers
may be found in Chapter XV, The Propeller.
3. Drag Chute. For high-speed airplanes, primarily military aircraft,
a drag chute released from the tail of the airplane works effectively as a
braking device. The reasons for using a drag chute may be summarized
as follows:
(a) To provide positive decelerating forces under all conditions and
thus permit landing on a given field, regardless of the runway conditions.
(b) To provide emergency braking in case of the breakdown of landing
gear.
(c) To augment the normal braking system, thereby making possible
the use of shorter runways.
(d) To save tire wear.
The following data provide a specific example of the advantages in-
volved in the utilization of a drag chute in a landing operation.
Length of landing run utilized
Condition of runway: Dry Icy
Wheel brakes alone 100% 300%
Wheel brakes and drag chute 53% 113%
4. Jet Thrust Reversers. There have been various designs devoted to
applying the principle of thrust reversal to jet engines. These thrust
reversers are essentially cups placed in the exhaust jet stream to reverse
the direction of the exhaust gases and thereby produce a reverse thrust or
drag. It has been proposed that it might be worth while to employ small
jet engines, which would direct the thrust in a direction opposite to that
of the motion of the aircraft.

5. Rockets. Short-lived rockets, operating to produce a reverse thrust,
may also be employed to reduce the landing run.

6. Expendable Skids. Expendable skids or skis extended below the
fuselage to make contact with the ground at about the same level as the
wheels constitute another possible means of reducing the landing run.
7. Barriers. Barriers at the end of the landing run that make contact
with the wing, fuselage, and landing gear have been used to advantage
THE LANDING GEAR
361
in stopping the airplane. Such a device would not be desirable in pro-
peller-driven aircraft.
8. Cables. These are picked up by a landing hook. They have been
used on aircraft carriers and can be adapted for land use.
DISSIPATION OF ENERGY
The normal airplane upon landing has both a sinking and a forward
velocity. The kinetic energy of the airplane from the time the aircraft
first makes contact with the landing areaâ€”be it an airport, a ship's deck,
or a water surfaceâ€”until it comes to rest must be absorbed in some man-
ner. The various devices employed to accomplish this are termed shock
absorbers.
Since the airplane may glide to contact with the ground in landing, the
vertical and horizontal component of the landing speed can be determined.
(See Figure XVIII-2 and also Chapter III.)
Lw = W cos 0 = Y2pvgCLS,
v = ^jW cos (3
vg sin j3,
V2pCtS'
vg = gliding speed,
v, â€” sinking speed
Vh = Vg cos /3.
The energy of the airplane due to the sinking speed v, has to be absorbed
by the shock absorbers. The energy
of the airplane due to v>> is usually ab-
sorbed by the brakes of the airplane,
although air resistance also helps.
Therefore, when the airplane is about
ready to contact the ground, the pilot
brings it up to the maximum lift co-
efficient so that the speed may be
decreased by the high lift coefficient
and the large drag. In order to attain
this lift coefficient, the landing gear
must be so designed as to obtain the corresponding angle.
This energy may be dissipated in any one or a combination of the

following ways:
1. Hysteresis of a material. Rubber shock cords in tension or rubber

disks in compression dissipate energy by hysteresis. Steel in the form of
a coiled spring or a cantilever beam may also be used for the same purpose.
Figure XVIII-2. The velocity vec-
tor diagram for the airplane in the glide.

362
However, since such materials also store up some energy, "bouncing"
results, and landing with these shock absorbers may therefore be rough.
Light airplanes or single-place gliders may employ such means with
weight-saving results.
2. Friction
(a) Brake friction is the usual accepted method. Here the result
is heat, necessitating provision for adequate heat sinks.
(b) Fluid friction, whereby a fluid is forced through one or more
orifices, is the standard principle of the so-called hydraulic shock absorbers.
(c) Friction between adjacent leaves of a spring may also be employed
where the energy to be dissipated is not too great.
3. Compression of a gas. The usual aircraft shock absorber employs
both a fluid-friction device and the compression of air. The adiabatic
compression of the air represents a portion of the work done, and some of
the energy is therefore dissipated by this work. The amount of energy
thus dissipated is relatively small, but the compressed air acts as a spring.
Since the air is forced into the absorber under pressure, it tends to keep
the shock absorber (consisting of a piston and a cylinder) extended when-
ever the load is relieved.
A combination of these energy-dissipating media is present in every
airplane. The craft itself is a flexible structure, so that it may absorb
some of the energy. The wheels and tires are also shock-absorbing
mechanisms.
There has been some experimentation with starting landing wheels in
rotation before contact is made with the ground. In this way, the initial
landing shock could be lessened and the wear and tear on the tire reduced.
Such rotation has been accomplished by means of flaps attached to the
side of the tire acting like a wind-mill or by means of direct application of
power from a small electric motor.
Higher wing loadings generally result in faster landings, creating a need
for stronger struts and wheels, tires that can withstand higher ground-
impact loads and spin-up speeds, and brakes to absorb and dissipate
increasing amounts of kinetic energy and heat. Since high-speed air-
planes require high wing loadings to attain the desired speeds, the landing
speeds become correspondingly greater unless additional devices or means

are used to decrease the approach speed. Such devices and means would
be auxiliary and would usually serve only one specific purposeâ€”to reduce
the approach speed.
SHOCK ABSORBERS
All landing gears must be equipped with some shock-absorbing device in

THE LANDING GEAR
363
Inflation valve
Compressed air
Friction material
Rubber spring
Compression tube
order to avoid sudden shocks to the
passengers, and to avoid the enor-
mously high stresses that would be
caused even in a normal landing if
no shock absorbers were used.
There are various types of shock
absorbers. At the present time, the
popular type of shock absorber is
one that utilizes an oil which is made
to pass from one chamber to another
through one or more orifices. Com-
pressed air is used as a spring to
keep the shock absorber extended.
Shock absorbers usually have to
be designed for a particular airplane
although it is possible to obtain
some stock sizes.
These shock absorbers are de-
signed to absorb the energy created
by a free drop, in inches, of 0.36
times the calculated stalling speed
in miles per hour (but not greater
than 18 inches total for conventional
airplanes) and in doing so, should
not exceed the design load factors
for the landing conditions.
In designing shock absorbers it is usually necessary to consider the fol-
lowing:
1. Gross weight of airplane.
2. Static load on shock absorber.

3. Design load factors.
4. Design load for the principal landing conditions.

5. Required height of free drop in inches equal to 0.36 (stalling speed
in mph) but not greater than 18 inches.
6. Weight of chassis.
7^ Vertical movement of wheel center.
8. Maximum tire deflection.
9. Tire size.
10. Tire inflation.
11. Shock-absorber strut travel desired.
12. Required length of shock-absorber strut under static load in level-
landing and three-point landing position.
Figure XVIII-3. A shock absorber
developed by Firestone. The illustration
shows a suggested installation.

364
13. Minimum or maximum length, or both, required for fully com-
pressed condition.
14. Minimum or maximum length, or both, required for fully extended
condition.
It is often desirable to draw up a three-view line drawing of the landing
gear to show the configuration of the various members.
It is not generally desirable to place the shock absorbers at too great an
angle to the vertical for the best results. A limiting angle of 45 degrees
may be assumed although not more than half that would be more desirable.
CLASSIFICATION BY LANDING CONTACT
The landing gear could be initially classified according to the number of
contacts made by the wheels. Since variations are possible, it will be
helpful to discuss them briefly. Whatever the arrangement of wheels,
the primary goals are usually a minimum of weight installation difficulties
and of retraction complexities.
One-Point Contact
A single wheel has been used primarily for relatively light gliders or to
obtain an amphibious gear. In either case, the wheel is usually partially
recessed in the fuselage or in the keel of the flying boat hull. To obtain
stability on the ground and to protect the rest of the structure, a tail or
nose skid and wing skids are used. When the aircraft is under way,
either in landing or taking off, contact with the ground is made only
with the wheel.
Some modifications of this principle are possible through the use of a
dual-wheel installation (side by side), which would give the craft some
lateral stability on the ground; a tandem arrangement, which would give
some longitudinal stability of the craft on the ground; either a triangular
three-wheel or a rectangular four-wheel arrangement, which would give
both lateral and longitudinal stability on the ground and thus eliminate
skids.
Two-Point Contact
Two main wheels having a reasonable tread (distance between wheels
in the lateral direction) may be employed with the aid of a nose or tail
skid. Since this imposes some importance on the skids, there may be a
question as to whether such an arrangement really may be termed as

having two points of contact. This arrangement was used for relatively
light airplanes, but objections were raised against the wear and tear
caused by the skids.
A tandem arrangement of two wheels with an appreciable distance be-
tween them could be considered. Wing skids would have to be provided

THE LANDING GEAR
365
for lateral balance unless there was a dual installation of wheels at either
the forward or rearward location or both.
Three-Point Contact
This arrangement consists of two main wheels plus either a nose or tail
wheel and is described in greater detail elsewhere in this chapter It is
generally thought of as offering the best all-round compromise for weight,
ground stability, and control.
Four-Point Contact
This arrangement, simulating the arrangement of conventional surface
vehicles, has been proposed and used. While ground stability is excep-
tional, other aspects such as control, weight, and installation complexity
(especially retraction) are generally unfavorably affected.
LANDING GEARâ€”TAIL-WHEEL TYPE
The tail-wheel type landing gear consists of (1) two main wheels or
their equivalent, placed slightly ahead of the center of gravity, and (2)
a tail wheel or skid at some distance aft of the center of gravity.
While it may be desirable to have the main wheels reasonably far
ahead of the center of gravity of the airplane, if placed too far forward
there would be difficulty in taking off. In taking off, the tail of the air-
plane must be raised until the longitudinal axis of the airplane is practically
horizontal. In this position, the airplane accelerates quickly until it
reaches a climbing speed and is ready to take off. But, in order to reach
this horizontal attitude, there must be a lift on the horizontal tail surfaces
produced by the relative wind on these tail surfaces caused by the pro-
peller slipstream and the forward acceleration of the craft. This lift, mul-
tiplied by the distance from the center of pressure on the horizontal tail
surfaces to the point of contact of the wheel with the ground, is the mo-
ment which must be just equal to the moment produced by the weight of
the airplane times the distance from the center of gravity of the airplane
to the point of wheel and ground contact. When these two moments are
equal, the airplane will be in a horizontal attitude. As soon as the airplane
starts to accelerate, the elevators, which have been depressed up to this
time, are gradually neutralized. Otherwise, too much lift would be
created and the airplane would nose over.
If the front wheels were quite far ahead of the center of gravity of the
airplane, a greater moment would have to be produced by the horizontal
tail surfaces. Since the lift on the tail surfaces is proportional to the
square of the speed, it would be necessary to increase the speed to obtain
the necessary lift. However, it takes time to start at zero speed and ac-
celerate up to a particular speed, and the longer it takes to accelerate, the

370
to dig in at low angles of attack. By raising the nose of the airplane the
load on the nose wheel is relieved.
5. With increase in propeller thrust, the load on the nose wheel increases
since the propeller thrust line is above the nose wheel. Unless the nose
wheel is equipped with a larger "oleo" travel and a stronger structure,
severe shock loads are likely to be transmitted to the airplane structure,
for the "oleo" or shock absorber may be compressed long before any seri-
ous loads are imposed.
6. Difficulty may be encountered with the nose wheel in riding over
obstacles. The tail wheel seems to behave better under such circum-
stances.
7. The tail wheel has the advantage of protecting the tail surfaces. Un-
less a skid or special crash pad is provided for the rear portion of the fuse-
lage when the nose-wheel type of landing gear is used, the rear portion of
the fuselage may be damaged in case of an unusual "tail low" landing.
8. The nose wheel causes more difficulty in retraction because of its
location in the forward portion of the fuselage and because of its longer
shock-absorber travel.
TRACK-TYPE LANDING GEAR
Track-type landing gears have received serious consideration for large
transport planes.
The advantages of the track-type landing gear are:
1. Ability to operate from sand, dirt, and sod, since the track-type
gear sink is less than one third as much as the tires.
2. Ability of heaviest planes to operate from existing runways since
the aircraft load is spread over approximately 250 per cent greater run-
way surface.
3. "Bounce" eliminated from hot landings; the gear holds to the ground
after touch-down.
4. Virtual elimination of ground-loop dangers.
5. Elimination of possibility of blowouts and punctures.
6. Compactness enables retraction into smaller wing or nacelle well.
7. Greater stability when striking ground obstacles.
8. More efficient braking through larger ground-contact area; better
cooling of brakes through better heat dissipation.

9. Belt runs cooler than a tire so wears less; also, low unit ground
pressure of belt upon landing allows sufficient slippage at first contact to

practically eliminate wear as a factor in longevity.
The apparent disadvantages are:
1. Heavier installation.
2. May be very bulky and hard to retract. (See 6 above)

THE LANDING GEAR
371
3. At present, relatively easy to put out of commission with rocks and
sand. (See 5 above.)
4. Requires more maintenance. (See 9 above.)
Many of these disadvantages must be viewed in the light of the early
stage of development of track-type landing gear at present. Probably, with
more time, study, and ideas, most of these unfavorable aspects can be
lessened or even eliminated.
TANDEM GEAR
In place of the track-type landing gear, tandem wheel trains have been
devised, using either two wheels in tandem in place of the main wheel, or
two sets of dual wheels in tandem. These arrangements have been sug-
gested in order to:
1. Reduce the over-all size of the individual wheels and thus encourage
compactness.
2. Reduce the "foot-print" pressures of the tires by increasing their
contact area.
3. Obtain some of the favorable characteristics of the track-type land-
ing gear without the concomitant mechanical difficulties.
Some of the disadvantages encountered in such a landing gear are:
1. The speed range in which smooth landings can be made is more
limited. The airplane cannot be landed at a speed much higher than
that required for simultaneous contact of both gears. Touching the
front gear first with an appreciable rate of descent will result in an up-
ward bounce or will at least prevent sufficient load being applied to the
wheel to pemit effective braking.
2. Due to the large fore and aft spread of the tandem gears, the pilot
has virtually no control over the attitude on the ground. However, this
arrangement has been found satisfactory as it prevents the pilot from
varying the unstick speed improperly. The ground attitude is selected
for optimum take-off speed.
CROSS-WIND LANDING GEAR
In a cross-wind landing, the pilot approaches the airport with the air-
plane at a slight angle to the runway in order to counteract the drifting
force produced by the cross-wind. If his judgment is wrong, he may align
the airplane with the runway either too early or too late, thus imposing
severe loads on the landing gear that make ground-looping possible.
The cross-wind landing gear incorporates wheel installations that may

allow the wheels to swivel and to line up the airplane not with its longi-
tudinal axis, but in the direction of its inertia forces.

372
Several types of cross-wind landing gear have been proposed both here
and abroad.
One type provides for an assembly of wheel and wheel fork permitted
to swivel any amount under control of the pilot but locked by him in a
predetermined position. Where such a landing gear is employed, the
pilot must make a free run over the field to decide the amount of wind
correction needed.
The advantages claimed for this type are:
1. The pilot has accurate steering at all times. The airplane will taxi
just as with the normal configuration except that the longitudinal axis
of the airplane may not coincide with the direction of travel.
2. As the gear is locked at the desired angle of swivel, resistance to side
loads due to taxiing, propeller thrust, etc., will be exerted, regardless of the
position of the wheels.
3. As only a wheel-positioning attachment is needed, no changes in the
wheel, axle, and shock-absorber strut assembly are required.
4. A shimmy damper is not necessary.
5. The arrangement requires a minimum of added weight.
The disadvantages may be summarized as:
1. Additional operation required of the pilot.
2. Difficulty in exactly determining surface winds or in forecasting
gusts. The amount of locked-in swivel is therefore hard to gauge and
in any case cannot be accurately predetermined without practice.
Another type cross-wind landing gear is automatic in operation since
it employs the use of casterable wheels. Shimmy dampers would be
required.
The favorable aspects of this type may be listed as the following:
1. The pilot does not have to actuate the gear to the desired swivel
angle.
2. As the mechanism is simple, there is less chance of failure.
3. Fool-proof landings can be made if the runways are sufficiently wide.
The disadvantages involved in this type are:
1. The wheels are somewhat difficult to steer since the longitudinal axis
of the airplane must be turned through the entire swivel range of the
wheel (unless a strap is engaged) before the wheel starts to turn and the
direction of travel is changed.
2. There is the possible danger of wheel shimmy unless an adequate

shimmy damper is employed.
3. As the longitudinal axis of the airplane may vary greatly from the
direction of travel, instrument take-offs and landings may be difficult for
pilot-operated airplanes.
THE LANDING GEAR
373
4. If used on an aircraft with dual wheels, a device must be provided
to prevent any uncontrolled swivelling resulting from the unbalanced
moments caused by tire blowouts. To avoid this difficulty, each wheel
of a dual-wheel (side-by-side) arrangement would have to be independently
operated.
5. It is estimated that the weight increase in the landing gear for this
type of installation may be as much as 50 per cent above the normal
configuration.
Use of the cross-wind gear should shorten take-off runs in most cases
since during cross-wind operation there is an alleviation both of the early
separation of the flow over the down-wind wing in the region near the
root and the condition of boundary layer wash-out. This is especially
true for swept-wing designs.
shows how the wheels would operate in a crosswind landing. The wheels line up with
the direction of the inertia of the airplane, not with the longitudinal axis of the airplane.
The swivelling wheel, fork, and shock-absorber member may be preset in flight, and
then locked, to give the pilot steering control of the wheel; or wheel may be given a
slight trail (point of contact is behind the swivel axis) so as to give each wheel inde-
pendent and automatic castering ability.
Shimmy is a phenomenon commonly encountered in landing-gear
installations.
When castering rubber-tired nose or tail wheels are used, there is a
tendency for the wheel to oscillate violently about the spindle axis. This
phenomenon is popularly termed 'shimmy'. It may be corrected by (1)
the application of friction at the spindles of the tail wheels, (2) locking
the wheels while taxiing at high speeds, and (3) incorporation of shimmy
dampers.
Shimmy is common with the large nose wheels used in tricycle landing
gears. Because it is impossible to lock the wheels, friction for the spindle
has been the sole means of correction; but the larger the wheel and the
The illustration
SHIMMY AND SHIMMY DAMPERS

374
greater the load carried by the wheel, the larger the amount of spindle
friction necessary to prevent shimmy. Friction devices absorb energy by
clamping together two dissimilar materials. The resultant damping
action is maximum at zero velocity and decreases with motion, depend-
ing upon the materials and type of rubbing surfaces. Moreover, the damp-
Figure XVIII-10. A shimmy-damper installation. The wheel and fork assembly
may rotate in the upper bearing. The rotation is limited by the shimmy-damper
installation shown above and to the right of the axis of the fork strut. The shock
absorber is shown inclined upward to the right. The "scissors" or "nut-cracker"
attachment allows for only axial motion of the piston inside the shock-absorber cylinder.
The shimmy could be incorporated within the lower end of the shock-absorber strut.
The distance between the vertical through the axle center and the axis of rotation of
the wheel fork, at ground contact, is known as the caster length.

THE LANDING GEAR
375
ing is likely to fluctuate and to be undependable. The structure, further-
more, is large and unwieldy. Since friction damping is maximum at low
velocities, the nose wheel does not respond to steering and to finding its
true castering position.
Hydraulic damping means, on the other hand, have very little resistance
for slow wheel movements. However, as wheel motion due to shimmying
increases, the damping action increases also, since hydraulic energy ab-
sorption is proportional to the square of the velocity.
Hydraulic shimmy dampers may be of two typesâ€”the piston and the
vane type.
The piston damper is similar to the common hydraulic shock absorber.
The piston is connected by suitable means to the wheel fork and operates
in a cylinder attached to the shock strut. Through a system of torque
scissors and sliding collar, the wheel movements are transmitted to the
movement of the piston. The piston, in turn, reciprocates within the
cylinder and forces a fluid through appropriate orifices for energy ab-
sorption.
One vane-type damper consists of an oscillatory pair of vanes connected
to a shaft, plus a stationary pair of vanes connected to the housing or
stationary cylinder. This combination forms two pairs of diametrically-
opposite pressure chambers. Displacement of fluid and consequent re-
sistance are obtained by the movable vanes turning toward or in relation
to the stationary vanes. The fluid is passed through a valve orifice from
the chamber of decreasing volume to the adjacent chamber of increasing
volume. The fluids used are the same as for the shock absorber.
TREAD
The wheel tread is a function of (a) the height of the center of gravity
above the ground, (b) the wing span, and (c) the distance between the
front wheels and the tail wheel. As these dimensions increase in size, the
larger the wheel tread should be for good ground behavior.
WHEEL AND TIRE SIZE
The size of the wheels and tires in the conventional landing gear is de-
termined by the static weight equal to half the gross weight of the airplane
per wheel. It is not necessary to determine the load factors and loads im-
posed by various landing conditions since the wheels and tires are origi-
nally designed with ample margins of safety.
The size of the wheels for a tricycle landing gear depends upon their
position relative to the center of gravity. The two rear wheels may have
376
from 85 to almost 100 per cent of the load while the front wheel may have
from 10 to 25 per cent of the gross weight of the airplane as the static load.
The weight of the tail wheel for preliminary weight estimate and balance
determination can be estimated by assuming a static load of about one-fifth
to one-twelfth the gross weight of the airplane and then choosing the re-
quired nose or tail wheel, with the larger fraction for the nose wheel.
After the center of gravity has been found, the weight and size of the
nose or tail wheel may be corrected by finding the correct static load on
the wheel as follows:
Examination of the landing gears reveals that the wheel without any
load on it, as incorporated in the normal landing gear, may toe-in; or the
vertical centerline when viewed from the front is at an angle of several de-
grees from the vertical, commonly known as camber. These two features
are of course accentuated when the landing gear is in the fully extended
position due to the configuration of the particular members of a landing
gear. The condition for which to design is the static-loaded condition,
either in the level-landing position or the three-point landing position.
The wheels are given no toe-in for the normal condition unless the con-
figuration of the landing gear should be such as to cause an appreciable
toeing-out in the fully contracted position.
The camber given the wheels may be 1 or 2 degrees outward unless,
again, the configuration of the landing gear is such as to give an undesir-
able camber when in the fully contracted position.
Too much leeway either way may cause the tire to roll off when landing.
The position of the wheels with relation to the center of gravity is shown
in the illustration for the conventional landing gear and the tricycle land-
ing gear.
SIZE OF TAIL OR NOSE WHEEL
Refer to Figure XVIII-4 and Figure XVIII-7.
WHEEL POSITION
RETRACTION OF LANDING GEAR
No high-speed airplane should have a nonretracting gear if the best pos-
sible performance is desired.
The simplest form of retraction is one that employs only one motion,
THE LANDING GEAR
377
either in a plane parallel to the
line of flight or in a plane per-
pendicular to the line of flight.
For moderately small airplanes
of the single-engine type, retrac-
tion sideways is best since suf-
ficient room is available in the
root of the wing whereas retrac-
tion rearward will still leave at
least half of the wheel exposed.
The student should study the
schematic sketches of retractable
landing gear here presented in Figure XVIII-11. A landing gear re-
Figures XVIII-11, XVIII-12, tracted sideways into the fuselageâ€¢
and XVIII-13, in order to get a
good conception of the various methods that may be employed for re-
traction.
In manually operated retracting mechanisms, the force to operate the
crank or lever should not exceed 15 or 20 pounds. About sixty 12-inch
strokes per minute is a practical maximum. Large and heavy landing-
gear retraction systems incorporate a bungee.
To prevent creeping, the usual reduction ratios of screw and nut, or
Fiocbb XVIII-12. The landing gear is retracted directly rearward into the wing
between the front and rear spars. Similar linkages could be employed to retract the
gear sideways.
378
Figure XVIII-13. Another retractable landing gear design where space between
the front and rear wing spars is limited.
worm and wormwheel combinations, will insure irreversibility although
some positive stop device should be incorporated to assure safety.
Special attention should be paid to limit switches, valves, cable installa-
tion, latches, indicating systems, universal joints, and pressure devices to
reduce maintenance problems.
Table XVIII-1. Minimum radius of bend for hydraulic lines. Aluminum alloy or
stainless steel tubing.
Outside diameter of
Minimum bend radius
tube, in inches
measured to tube centerline
'/Â£
IX
3
THE LANDING GEAR
379
Figure XVIII-14. A simple brake installation. The arrows indicate the direction of
linkage when applying full brake. The brake pedal system is suspended from the
rudder bar.
Figure XVIII-15. A detail drawing of the landing gear used on the Ercoupe.
380
Figure XVIII-16. Representative proportions of a typical shock absorber are
shown in this figure. The piston has the smaller diameter.
Table XVIII-2. Representative dimensions of hydraulic-type shock absorbers.
Static load, in pounds,
Stroke,
Piston diameter,
Length,
per strut
in inches
in inches
in inches
500- 1000
1.50
19.5
1000- 1500
1.75
19.5-23.5
1500- 2000
2.00
19.5-24.0
2000- 2500
2.25
19.5-25.5
2500- 3000
2.50
19.5-26.0
3000- 4000
2.75
19.5-28.0
4000- 5000
6
3.125
20.0-30.5
5000- 6000
3.50
20.0-32.5
6000- 8000
4.125
20.0-32.5
8000-10,000
4.625
20.5-34.0
10,000-12,000
5.00
20.5-34.0
12,000-14,000
5.50
20.5-34.0
SPECIAL PROBLEMS
Speeds approaching hypersonic flight are posing special problems in the
design of landing gear and associated equipment. It is possible that
these equipment items will someday all but disappear.
Wings for transonic and supersonic flight are already too thin for
stowing complicated retractable landing gear. It is also difficult to find
room for the gear in the crowded fuselage, which must carry increasing
quantities of avionics, fuel, cooling systems, and other equipment. As
planes fly deeper into the thermal region, the landing-gear wheel well's
THE LANDING GEAR
381
Figure XVIII-17. Minimum clearance between wheel and supporting structure is
approximately 1/25 to 1/30 of the wheel diameter. There should be increasing allow-
ance away from the entrance point.

CHAPTER XIX
Tail Surfaces
The tail surfaces serve two functions. The fixed portions of the tail
surfaces, together with the movable, provide stability vhile the movable
portion in conjunction with the fixed portion provides a means for control.
LONGITUDINAL STABILITY CONSIDERATIONS
Imagine an airplane flying along in horizontal flight at constant speed
struck by a sudden gust of wind, either on the wing or on the tail surfaces,
so that the airplane is forced suddenly nose upward or nose downward.
Without any control exerted by the pilot, or any change made in the
throttle setting, the airplane should return to its original angle of attack
after cessation of the gust and continue to fly in its original direction. It
may not return to its original angle of attack instantaneously, but may
oscillate back and forth, with the oscillations gradually diminishing.
Definitions
Stability is that property which tends to return an airplane to its orig-
inal condition of equilibrium or line of motion when it has been disturbed
from that condition. This deviation from equilibrium or steady motion
may be caused by an external force, such as a gust, or it may be caused by
the pilot when he depresses the elevator and then returns it to its original
position almost immediately. When the airplane regains equilibrium or
line of motion (flies at a given angle of attack at constant speed) due to
forces and moments developed inherently and of such character as to
counteract the disturbing force or moments, it demonstrates this property
of stability. An example of longitudinal stability has been cited in the
previous paragraph. There are similar cases of stability about the longi-
tudinal axis, or lateral stability, and about the vertical or normal axis, or
directional stability.
TAIL SURFACES
383
There are two distinct types of stability, aside from their relationships
to particular axes. An airplane, for example, is so designed that at one
angle of attack the resulting moment of all forces about the center of
jrravity will be zero. This satisfies the condition of equilibrium together
with the fact that the sum of all forces along the 3 axes equals zero, but
this is known as static longitudinal stability. It was mentioned earlier
that the airplane does not come back to its original condition of equilibrium
immediately but continues to oscillate. This oscillation is a manifestation
of dynamic longitudinal stability.
There are also three different states of static stability and of dynamic
stability, whether applied to longitudinal stability, to lateral stability, or
to directional stability.
-4Â° 0Â° 4Â° 8Â° 12Â° 16Â° 20Â°
Figtob XIX-1. Pitching moment coefficients about the center of gravity plotted
for the wing, the tail, and the complete airplane. The diagram shows how and where
the slope of the pitching moment curve should be measured.
For example, an airplane may be neutrally stable. It may be disturbed
from its original attitude, and instead of coming back to it, will continue
in its new attitude.
Or, the airplane may be unstable: When disturbed from its original atti-
384
tude, instead of coming back to it, it will go farther and farther, not re-
maining at any attitude.
If the airplane does come back to its original attitude, then it is said to
be stable.
These three examples are cases of static stability. Similar cases of
dynamic stability would be:
1. Stable when the oscillations of the airplane gradually decrease to zero
after they are once started.
2. Neutral when the oscillations do not decrease or increase in amplitude
when once the oscillations have been set up.
3. Unstable when the oscillations increase with time after they have
been once set up.
Preliminary Calculations for Static Longitudinal Stability
Static longitudinal stability of an airplane may be determined with
reasonable accuracy by calculating the pitching moments about the center
of gravity of the fully loaded airplane. See Figure XIX-1.
The equation for the pitching moments has been derived in Chapter IX
and reads as follows:
Cmâ€ž. = Cmâ€ž. + Xi(CL cos a + CD sin a) + j/i(Ci sin a â€” Cd cos a) â€”
0.8y2 CL, f*. (1)
or simply,
Cm,.,. = Cm. + Cm,. (la)
This equation may be used for the initial calculations to determine whether
stability is adequate. When the designer is reasonably satisfied with his
design configuration, then a more accurate equation may be derived to
take into consideration effect of fuselage, landing gear, and the like.
The student may set up his own tables, which are always desirable where
repetitive calculations are called for, to evaluate numerically the various
values of the equation
Cm,.,. = Cm. + Cm,0
ADEQUACY OF STATIC STABILITY
To determine whether or not the longitudinal stability is adequate, the
degree of stability is determined by calculating the value of the so-called
Diehl's stability coefficient:

(dCM.Jda)
W/S
TAIL SURFACES
385
If, therefore, the value of Cm.*â€¢ is found for a series of angles of attack 0,
and plotted as the ordinate with values of a as the abscissa, then the value
of the slope of the resulting curve, particularly in the region in the angle
of trim will give the value of dCM..,./da. This value divided by the wing
loading W/S for the airplane will give the value of the stability coefficient.
For commercial airplanes of the passenger transport type, the value for
dCitâ€ž./da of the stability coefficient lies between â€”0.0005 and â€”0.0010.
FLIGHT CRITERIA FOR STABILITY AND CONTROLS
The designer attempts to build into the airplane such characteristics as
will insure adequate stability and control. The particular recommenda-
tions and procedures outlined in this book are relatively brief and are in-
tended to point out the high lights. They are believed to be all that a
student can cover in a reasonable time. The practicing engineer, how-
ever, has to make a careful study of all existing reports and research papers
in order to be able to select those features necessary to achieve the quali-
ties desired for a given design.
To give an idea of some of the free flight characteristics desired for air-
planes, the following synopsis based upon NACA reports (for example,
NACA Wartime Report L-276) may be useful.
Longitudinal Stability and Control
When the elevator is held in a given position, the airplane should main-
tain the attitude assumed:
1. With engines idling, flaps up or down, at all speeds above the stall;
2. With engines at cruising, flaps down, landing gear down, as for land-
ing approach, at all speeds above the stall;
3. With engines at full power, flaps up, at all speeds above 120 per cent
of the stalling speed;
4. At maximum and minimum speeds in level flight.
The elevator should be designed to
1. Obtain control forces such that "pull forces" are required at all
speeds below the trim speed, and "push forces" at all speeds above;
2. Obtain control forces always sufficient to return the airplane to its
trim position;
3. Obtain the allowable load factor (G's) of the airplane as well as the
maximum lift coefficient of the airplane by use of the elevator alone;

4. Obtain at least 4 inches of rearward movement of the stick for a
highly maneuverable airplane, such as a fighter, for a change in lift co-

efficient from 0.2 to the maximum lift coefficient of the wing;

386
5. Obtain control forces varying not more than 50 pounds for every G
attained in steady turning flight for heavy transports, but less than 6
pounds per G for highly maneuverable types, and at least 30 pounds of
stick force for every type to attain the allowable load factor;
6. Hold the airplane off the ground until the minimum speed of the air-
plane has been reached with the control force not exceeding 50 pounds on
control wheels, and 35 pounds on control sticks;
7. Hold the airplane level after one-half of the take-off speed has been
reached;
8. Be capable of trimming the airplane, after it has been trimmed in
level flight with zero stick forces, without obtaining forces greater than 50
pounds for the wheel and 35 pounds for the stick, regardless of the changes
in power or flap settings.
Elevator trim tabs must be capable of reducing stick forces to zero under
the following conditions:
1. Cruising condition at any speed between high speed and 120 per cent
of the stalling speed of the airplane;
2. In the landing condition at any speed between 120 to 140 per cent of
the landing speed;
3. Must maintain a given setting for an indefinite period of time.
Stick-free damping may be examined in flight by moving the stick sud-
denly and releasing it with the airplane in steady and level flight. The
resulting pitching of the airplane and movement of the elevator must dis-
appear completely after 1 cycle in order to insure stability in rough air
without aggravating the pitching of the airplane and thereby incurring
loss of control. Short period damping depends on:
1. Elevator hinge moment coefficients,
2. Mass balance of the control system,
3. Moment of inertia of the control system.
GENERAL REQUIREMENTS
It is very important that these tail surfaces be so located that they are
not blanketed by the fuselage. If the fuselage has a relatively large cross
section for the greater part of its length and then tapers suddenly near the
tail post, it is very likely that the horizontal tail surfaces will be blanketed
unless the aspect ratio of these surfaces is high.

The vertical tail surfaces are most likely to be blanketed not only by
the fuselage but also by the horizontal tail surfaces, especially when the
airplane is at a high angle of attack. In order to minimize this effect, it
would be desirable to get some of the vertical tail surfaces below the hori-
zontal tail surfaces.

TAIL SURFACES
387
Aerodynamically, it would be nice to have a large aspect ratio for both
the vertical and horizontal tail surfaces, but unfortunately the greater the
aspect ratio the more difficult it becomes to get an efficient structure that
will be rigid. Since the movable surfaces are a reasonably large propor-
tion of the total area, the fixed portion, which supports the movable sur-
faces, must contain all the necessary structure. If the aspect ratio is too
great for the area, there is relatively little depth with the result that
the fixed surfaces may deflect so much under a load that the hinges of
the movable surfaces bind.
The proportion of the movable surfaces to the whole depends upon the
degree of control desired. A large movable surface, for example, needs
less angular deflection than a smaller-sized surface. If the airplane is to
be very maneuverable, it is desirable to have relatively large movable sur-
faces. In any case it is necessary to have sufficient control at the slowest
and at the highest speeds the airplane will attain.
The elevator should be able to trim the airplane at the lowest, or stall
speed, at which time the elevator will usually have its maximum angular
deflection upward. The elevator should also be able to trim the airplane
practically at zero lift, at which time the elevator may have its maximum
angular deflection downward.
Control, however, is not measured only by the change in angle of trim
of the airplane caused by a definite angular deflection of the elevator (and
the discussion here applies equally to the vertical tail surfaces) but also by
the hinge moments produced. If for the same angular deflection of the
movable surface one has greater hinge moments than another, it should
be obvious that the one with the smaller hinge moments can be actuated
far more quickly, and the response of the entire airplane will be quicker
therefore.
CONTROL SURFACES
Typically, control surfaces are flapped surfaces located at the trailing
edge of a stationary surface. Therefore an aileron, a rudder, an elevator,
a tab, or a combination of these are control surfaces of the flap variety.
They are designed to change the lift of the entire surface (whether wing,
or horizontal or vertical tail surface) and thereby create a moment about
one of three axes of the airplane for motion or control about that axis.
Some discussion on flaps and ailerons in Chapter XVII, Design of the
Wing, dealt with increase in lift versus flap deflection and with hinge
moments. The student is referred to that discussion since it is equally
applicable to tail surfaces.
The magnitude of these hinge moments is becoming an increasingly im-
portant problem. There are several solutions available. In some cases,
(for example, the aileron) it helps materially in reducing the hinge mo-
merits to have a smaller chord so that the ratio of the chord of the movable
surface to that of the entire surface is 15 to 20 per cent. This necessitates
a larger span in order to get the same total control but, unfortunately, the
rudder or elevator seldom has the ratio of its chord to the chord of the com-
plete surface less than 45 or 50 per cent. In order to reduce the hinge
moments, the surfaces may be partially aerodynamically balanced either
by having the hinge line of the movable surface somewhat in rear of its
leading edge, or by having a small movable surface or tab attached near
the trailing edge of the main movable surface. This small surface has an
angular deflection opposite to that required for the main movable surface.
If the tab is small or its setting fixed and changed only when the load
conditions change the center of gravity, then its purpose is for trim only,
and is known as a trimming tab. It takes the place of the adjustable
stabilizer.
If the tab can be controlled from the cockpit, it may be used to operate
the larger surface and is then called a control tab or a servotab.
Aerodynamic balance is generally used, even if trailing-edge tabs are
present. The design of this balance is very critical and is still the subject
of much experimental work. For greater effectiveness, a slot in front of
the leading edge of the balance is provided. Although this slot helps to
increase the effectiveness of the movable surfaces, yet, more often than
not, the relatively large gap caused by the slot increases the parasite drag.
Great care should be taken in designing the leading edge of the aero-
dynamic balance so that it is not too sharp and does not project too far
above the upper or lower contour of the fixed surface when the movable
surface is deflected. Such projections collect ice very quickly under icing
conditions and may lead to unbalance of the control surface, or jamming
of the controls.
Design details of the tail surfaces are given under the various paragraph
heads immediately following.
For high-speed aircraft at or near transonic speeds, flaps for controls
produce an appreciable increase in drag, known as trim drag. For that
reason, instead of flapped surfaces, the entire surface is sometimes moved
(as on the Wright Brothers'original airplane).
AIRFOIL SECTIONS
Symmetrical airfoils are usually used for tail surfaces so that equal effec-
tiveness per degree of deflection may be obtained for both up and down
movements.
The airfoil section used should have a thickness ratio of from 6 to 9 per
cent and not more than 12 per cent. Unless the tail surfaces are ex-
ceptionally large, the same airfoil is used from tip to root. The NACA
0006, 0009 and 0012 series are recommended.

389
Figure XIX-2. Three forms of aerodynamic balance for movable surfaces. (A) is
called an unshielded horn balance; (B) is a shielded horn type; (C) shows the more
commonly used aerodynamic balance. All have a certain amount of area ahead of the
hinge pin.
AERODYNAMIC BALANCE
When the flap or control surface is deflected, the center of pressure or
the aerodynamic forces on the flapped surface is behind the hinge line,
producing hinge moments. These hinge moments divided by the lever
arm of the control arm in the cockpit give rise to the stick force that the
pilot has to exert in order to move the control surface.
Figures XIX-2 and XIX-3 show a number of means for obtaining
aerodynamic balance, each of which has special merits of its own. There
are various NACA reports which should be consulted for aerodynamic
details.
The ratio of aerodynamic balance area to the movable surface usually
is from .15 to .20 for the rudder and the elevator. It is generally not
desirable to overbalance the movable surface since it is considered psycho-
logically sound to have the stick forces that are to be overcome increase
with the angular deflection of the surface.
Figure XIX-3. The balancing tab reduces the hinge moments. The hinge moment,
H, for the illustration, is equal to Pia " PJ>.

390
Fig. XIX-* Fig. XIX-5
Figure XIX-4. Bulge contour for control surfaces. These tend to float, and
thereby to reduce the control forces.
Figure XIX-5. Pressure diaphragm balance, shown schematically. A flexible
membrane is used for the diaphragm.
Fig. XIX-6 Fig. XIX-7
Figure XIX-6. Cord or bead along the trailing edges of control surfaces may be
used in place of trimming tabs.
Figure XIX-7. Wedge or bevel contour for control surfaces may be used in place
of trim tabs.
Figure XIX-8. Different arrangements for the operation of trim tabs. The top
two diagrams illustrate fixed trim tabs adjustable, by trial and error, to the proper
degree on the ground, while the two lower diagrams show arrangements which may be
operated from the cockpit.

TAIL SURFACES
391
Figure XIX-9. Different arrangements of servo-control and boost-type tabs.
Trailing-Edge Tabs and Other Devices
Trailing-edge tabs on movable surfaces are another means of obtaining
aerodynamic balance.
A tab on the aileron is used to overcome engine torque. It may be used
with the rudder for the same purpose as the offset fin and on the elevator
for the same purpose as the adjustable stabilizer.
Trimming tabs have a chord varying from 5 to 10 per cent of the mov-
able surface chord and approaching 25 per cent of the chord if used as
servocontrol tabs. The aspect ratio should be as high as possible, varying
usually from as low as 5 to as high as 20.
A control surface is said to be statically balanced when the center of
gravity of the plane containing the hinge line acts through that hinge
line. Static balance is one of the design features required to avoid flutter
possibilities.
For low-speed aircraft having maximum speeds below about 200 miles,
static balances are generally not necessary. For greater speeds, static
balance for movable surfaces becomes a necessity.
For a surface that has its hinge line at the leading edge, static balance
can be obtained only with a concentrated weight ahead of the hinge line
to counterbalance the weight of the movable surface.
For the surface having its hinge line at some distance to the rear of the
leading edge, the structural weight can be so distributed that the center of
gravity falls on the hinge line. Usually, however, this is more easily ac-
STATIC BALANCE
Figure XIX-10. Three different ways by which static balance of a movable surface
could be obtained. By proper positioning of the added weight ahead of the hinge line,
dynamic balance may also be achieved.
complished with a fabric-covered surface than with one that is metal
covered. For the latter, some counterbalancing of the nose section may
become necessary.
DYNAMIC BALANCE
For a movable control surface to be fully balanced dynamically, the
principal mass axes of the surface in the plane containing the hinge line
must coincide with the hinge axis and an axis perpendicular to the hinge
line about which angular deflection could take place when subjected to
aerodynamic or inertia loads. For the rudder and elevator, this per-
pendicular axis would be the neutral or longitudinal axis of the fuselage.
For airplanes with a speed greater than 150 miles per hour, it is de-
sirable to have the rudder and elevator dynamically balanced. The
criterion for such balancing is a dynamic balance coefficient of not more
than 0.08.
The dynamic balance coefficient is nondimensional, which may be ex-
pressed by the formula:
iff p xydxdy) 4- (Sf p dxdy)
= Cdb0
0
TAIL SURFACES
393
Figure XIX-11. Designation of quadrants for rudder dynamic-balance calculations.
The hinge line is the F-axis, while the neutral axis of the fuselage is the X-axis.
Figure XIX-11 shows a typical surface with the axes designated and
the terms indicated for the equation above, in which the numerator is the
resultant product of inertia of the control surface, and the denominator
is the product of the mass and the aerodynamic area of the control sur-
face. In the formula expressed, p is the unit weight or mass referred to
the surface area, x is the distance between the center of gravity of the
mass and the F-axis, y is the distance between the center of gravity of
the mass and the X-axis, and dx dy is the area of the mass under con-
sideration, S = X dxdy.
The products of inertia in the first and third quadrants are considered
positive; the products of inertia in the second and fourth quadrants are
considered negative. For movable surfaces the hinge line is one axis,
and the principal or longitudinal axis the other.
FLUTTER PREVENTION
The Department of Commerce at one time suggested the following gen-
eral principles to be observed for flutter prevention on all airplanes in the
design of control surfaces and control systems.
1. Structural stiffness.
2. Elimination of all play in hinges and control-system joints.
3. Rigid interconnections between elevators.
4. A relatively low amount of aerodynamic balance.
5. High frictional damping.
6. Adequate wing fillets and fairing.
7. Sharp leading edges on movable surfaces should be avoided.

394
The following features are considered essential for flutter prevention.
8. When separate elevators are used, the interconnecting structure that
is required must be as stiff in torsion as practicable.
9. Trailing-edge controls should be irreversible, relatively rigid, and
the tab installation should be designed to prevent any development of free
motion of the tab. When the tabs are completely statically balanced,
irreversible controls are not required, provided that the tab control system
cannot be manipulated abruptly through a large range. A small amount
of overbalance of the tab, obtained by putting its center of gravity ahead
'of the hinge line, will help to offset harmful effects of any possible looseness
or tab-control system flexibility.
10. When trailing-edge tabs are used to assist in moving the main sur-
face, the areas and relative movements must be so proportioned that the
main surface is not aerodynamically overbalanced at any time.
11. Experimental determination of the natural frequency of vibration
of certain components of the airplane may be desirable in cases where dan-
gerously low frequencies or the coincidence of the natural frequencies of
two or more structural components exist. Where such cases do occur, re-
design is necessary to change such natural frequencies.
Figure XIX-12. Layout of the principal elements of the tail-surface structure.
Unlike the wing, the spars of the tail surfaces need not be located at a constant
percentage of the chord.
The structural details of the vertical and horizontal tail surfaces are
similar to a simplified version of the wing tip which includes the aileron.
A spar at or near the hinge line is desirable for the elevator or rudder.
This spar is supported by at least three hinge brackets, which in turn
are supported by the rear spar of the fixed surface, which usually has
another spar located at about 10 or 15 per cent of the chord.
Figure XIX-12 shows a schematic layout of the primary structure of
the tail surface. While spars in the wing are located at or close to a
constant percentage of the chord spanwise, such is not the case for tail
CONSTRUCTION
TAIL SURFACES
395
Figure XIX-13. The box structure at the hinge of this movable surface carries the
torsional moment around the hinge cut-out of the leading edge D-spar.
surfaces. Here, the location of the spars is dictated by other considera-
tions, such as hinge-bracket distances, location of attachment to the
fuselage, and rigidity requirements.
For aircraft operating at speeds less than 350 miles per hour or there-
abouts, the movable surfaces are usually fabric covered in order to obtain
a saving in weight and to reduce difficulties in static and dynamic balanc-
ing of the surface.
HORIZONTAL TAIL SURFACES
Location
The horizontal tail surfaces should be so located that any blanketing by
the wing or the fuselage is avoided. Partial blanketing usually exists,
however, but certain features may be incorporated to limit the effect of
blanketing.
In some cases, the location of the horizontal tail surfaces is determined
by clearance requirements for the elevator, as shown in Figure XIX-14,
when the elevator is deflected downward through its total angular range
and with the tail-wheel assembly fully deflected.
Conventional airplanes j>f today locate the tail surfaces about to 3
chord lengths (mean geometric chord of the wing) behind the center of
gravity so that the observance of this rule will assure reasonable static
longitudinal stability.
The horizontal tail surfaces may be ahead of or behind the vertical tail
surface or somewhere between these two extreme positions, as shown in
Figure XIX-15.
396
Figure XIX-14. Layout of the horizontal tail surfaces showing cut-out of elevators
to prevent interference with the movement of the rudder. Each movable surface should
be supported by at least three hinges.
For the small airplane, because of structural and dimensional considera-
tions, the rear stabilizer spar and the rear fin spar usually intersect and
are built in integrally with the fuselage frame at that station unless either
or both the fin and stabilizer are adjustable in flight or on the ground.
A position of the horizontal tail surfaces ahead of the vertical tail
surfaces may be considered if it results in less complication of the control
system. One disadvantage of such a location is the blanketing of the
vertical tail surfaces at high angles of attack.
Figure XIX-15. Location of horizontal tail surfaces. They may be located ahead
or behind the vertical tail surfaces. or located vertically anywhere between the two
extremes of the vertical tail surfaces at the bottom of the fuselage or at the top of the
vertical tail surfaces.

TAIL SURFACES
397
A position of the horizontal tail surfaces behind the vertical tail surfaces
would clear both, so that neither would be blanketed by the other.
Vertically there may also be some choice of location. The lowest
position would probably be determined by the ground clearance of the
deflected elevator, whether or not the nose-wheel or tail-wheel type landing
gear is used.
The location of the horizontal tail surfaces where they can be securely
attached to the fuselage structure is advantageous from the structural
point of view.
For the higher-speed airplanes, especially those having the exhaust
duct or ducts of the jet engines in .the tail of the fuselage, the horizontal
tail surfaces have to be located above the fuselage. Thus the horizontal
surfaces have to be secured to the vertical tail surfaces. Under asym-
metrical conditions, the aerodynamic load on the horizontal tail surfaces
produces a torsional moment upon the vertical tail surfaces. It there-
fore becomes necessary for those surfaces to be constructed more rigidly,
thereby entailing greater weight.
Consideration has been given to the use of a biplane set of horizontal
tail surfaces, whereby the smaller elevator could, in its operation, be used
as a booster control for the larger surface through suitable linkage.
A design used occasionally is the so-called canard type, which locates
the horizontal tail surfaces ahead of the wing. In that position, the
surfaces are acting in the up-wash, rather than the down-wash of the
wing. For this reason, the horizontal tail surfaces are apt to stall before
the wing, so that sufficient control may not be available at a critical time.
However, this feature has been considered an advantage since the tail
surfaces could act as a stall-warning for the wing. The location of the
tail surfaces ahead of the wing may interfere with the vision forward.
It is considered psychologically bad to be able to see the deflection of the
surfaces at practically all times. Another advantage, in addition to the
stall-warning device, is that the arrangement of the surfaces forward may
serve as a suitable crash-absorber in case of an accident.
Angular Deflection
Elevators are designed to have an equal angular movement up and
down from neutral. About 30-degree movement is considered maximum,

and, with efficient design, a 25-degree deflection up and a 25-degree
deflection down should be sufficient.

Adjustable Stabilizer
The stabilizer may be adjusted through a small angular displacement
either on the ground or in the air from the cockpit (usually the latter, if at
all, since trimming tabs are displacing adjustable stabilizers).

398
If an adjustable stabilizer is used, a total of 6- to 8-degree movement
(about 5 degrees up and 3 degrees down) is usually used.
The adjustable stabilizer is used to change the trim angle of the air-
plane without displacing the elevator. The elevator can then operate
from the new neutral position determined by the stabilizer. In the first
prototype to be flown, the range of angles provided may be greater than
is absolutely necessary until flight tests have determined the desirable
range.
For high-speed airplanes operating through the transonic regime, the
adjustable stabilizer is preferred over the elevator trim tab, since the trim
drag is becoming an important factor in the performance of the airplane.
Where power operation has been provided, a one-degree-per-second rate
of change through 11-degree travel has been used.
On small airplanes it has been customary to make the stabilizer adjust-
able through a limited angular range, about 3 degrees up and 3 degrees
down. This adjustment has been possible either on the ground or in the
air by means of a control located in the pilot's cockpit. The adjustment
in the air is preferable. On the large transport airplanes, variations in
trim (the object of the adjustable stabilizer) are obtained by means of
trailing-edge tabs.
Aspect Ratio
The aspect ratio of the tail surfaces should be as high as possible (usually
from 3 to 5) in order to avoid blanketing of the structure to which they
are attached. Aspect ratios greater than 6 are seldom used unless ttyey
can be adequately braced.
In proportioning the tail surfaces, it is not desirable to start with the
aspect ratio because the fuselage section increases the span of the tail sur-
faces seemingly beyond the desirable limit.
For correcting airfoil data from the given aspect ratio to that of the tail
surfaces, the aspect ratio is calculated on the basis of the square of the span
length from tip to tip divided by the area including that covered by the
fuselage. In other words, exactly the same procedure is followed as in
calculating the aspect ratio of the wing.
Angle of Incidence
The incidence of the horizontal tail surfaces is determined by the amount

of downwash from the wing, its relative location with respect to the wing,
and the moment required to obtain the required trim angle.

The angle of trim of the airplane is denned at that angle at which the
sum of the pitching moments about the center of gravity of the airplane is
zero. Normally, it is expected that the elevator be in the neutral position,
that is, undeflected for that attitude of the airplane or angle of trim for
TAIL SURFACES
399
which the flying time is the greatest. This attitude is usually for cruising.
In order to determine this angle, it would be desirable to have calculated
the preliminary performance. In lieu of the necessary data from such
calculations, it may be assumed that the desired angle of trim lies between
the angle of the wing at which the minimum drag coefficient occurs, and
the angle at which the maximum ratio of lift to drag occurs.
Once the angle of trim has been decided upon, the angle of incidence (or
setting) of the horizontal tail surfaces can be determined for at trim by:
Cjfw. = 0 = Cm. + Cm a or Cm, = â€”Cm.;
but
or
-0.8yi CL, f*.
â€” Cm.
â€” -0.8yASt/SK)&ttTim> (2)
where Cm. has been calculated for that angle of attack of the airplane cor-
responding to the angle of trim.
Once the lift coefficient of the tail surfaces at trim condition has been
determined, the required angle of attack of the tail surfaces can be deter-
mined from the aerodynamic characteristics of the tail surface airfoil.
Since the tail surfaces are affected by the downwash of the wing, correc-
tion for this downwash must be made not only to determine the angle of
incidence of the tail surfaces, but to determine the angle of attack of the
tail surfaces corresponding to the angle of attack of the wing.
The downwash angle is given sufficiently accurately by the following
modified Diehl's equation:
e = â^Â»ir*Â» (3)
where e is the downwash angle in degrees,
Cl is the lift coefficient of the wing,
R is the aspect ratio of the wing, "u0
x is the number of mean aerodynamic wing-chord lengths that
V the 20 per cent pomt of the mean geometric chord of the hori-
zontal tail surfaces is behind the aerodynamic center of the
wing, and
y is, similarly, the number of chord lengths the horizontal tail

surfaces are above or below the mean aerodynamic chord of the
wing.
For any given design, equation (3) could be written in terms of the angle
of attack of the wing since Cl â€” aa*, where aa is so measured that a = 0
when Cl = 0, and a is the slope of the lift curve. By further evaluation,
the formula would finally take the form

e = kaa, (3a)
where
0.40a
Â£-0.25 y-0.7I_
Then, if the wing has an angle of incidence measured with relation to a
fixed reference line on the fuselage, the effective angle of downwash meas-
ured with relation to this reference line would be e + iw. The angle of
Figure XIX-16. Reference diagram for angles referred to in equation (4a).
attack of the horizontal tail surfaces, when placed at zero angle of incidence
to the fuselage reference line, would be â€” (Â« + ij). Should the horizontal
tail surfaces have an angle of incidence of its own, the angle of attack of
the horizontal tail surfaces would be it â€” (Â« + iÂ«).
Recapitulating,
at = -(Â« + (4)
or
OLt = it â€” (e + im), (4a)
depending upon whether the horizontal tail surfaces have an angle of inci-
dence or not. This angle of incidence can now be determined for the re-
quired at, for Cl [as obtained from equation (2)] at trim gives the neces-
sary information. For subsequent calculations, it would be desirable to
determine at from a modified form of equation (4a) such as at = fcaâ€ž â€”
i* + it0
Sweepback
Normally, any sweepback that the horizontal tail surfaces may have is
due to the trapezoidal planform. However, just as sweepback has been
employed for the wing as speeds approach M = 1, so sweepback has to be
applied to the horizontal tail surfaces. The discussion relating to sweep-
back in Chapter XVII, Design of the Wing, is applicable to the design of
the horizontal tail surfaces.

TAIL SURFACES
401
In order to obtain relatively uniform down-wash conditions on the
horizontal tail surfaces, some sweepback may be employed to compensate
for the sweepback of the wing.
Dihedral
Normally horizontal tail surfaces are not given any dihedral, but it has
been found that the effectiveness of the horizontal tail surfaces can be in-
creased considerably, particularly at high angles of attack, by incorporat-
ing some dihedral in the horizontal tail surfaces. How large the dihedral
angle should be depends upon the down wash of the wing; for purposes of
symmetry, the span line of the tail surfaces may be made parallel to the
span line of the wings.
Area
Examination of airplanes of all sizes reveals that the jpatio of the hori-
zontal tail surfaces to the effective wing area varies from'20xo 25 per cent.
The greater the tail length is, in terms of the wing chord, the smaller per-
centage area is required. Wings equipped with lift-increase device usually
require that the percentage area of the horizontal tail surfaces be greater
than if the wings were not so equipped.
The elevator area varies from 35 to 45 per cent of the horizontal tail sur-
face area.
Construction
For ease in assembly and disassembly, the horizontal tail surfaces are
attached to the top of the fuselage, especially if tubular steel construction
is used for both the tail surfaces and the fuselage. When reinforced metal
monocoque construction is used, the horizontal tail surfaces may be located
nearer the longitudinal centerline of the rear portion of the fuselage and
still obtain the necessary rigidity.
Planform
Some indication of planforms used for horizontal tail surfaces may be
obtained from those shown in Figure XIX-17. Since the aspect ratios
are about the same for the horizontal tail surfaces as for the wing, and
some other considerations of design are similar, the planforms of the
horizontal tail surfaces look very similar to the wing planforms. Some
modification may be made due to the greater role that the movable sur-
face, which takes up almost half of the total area, plays in the design of
the structure.
402
Figure XIX-17. Typical planforms of horizontal tail surfaces. Only one half of
the surfaces are shown. Vertical tail surface planforms are similar.
Other Solutions
For some aircraft, especially those of the flying-wing type, the horizontal
tail surfaces disappear as separate entities, but their function is taken
over by trailing-edge flaps or wing-tip surfaces. The high-speed transonic
and supersonic designs employing the delta wing should be studied, es-
pecially with reference to the means employed in obtaining longitudinal
stability and control through flaps and other similar devices.
For other high-speed designs, there may be no elevator as such, but the
entire horizontal tail surface may be deflected angularly in order to elim-
inate the increase in drag caused by the deflected flap.

TAIL SURFACES 403
Figube XIX-18. Two illustrations showing the application of moveable wing tips
that, freely floating, would operate as an aileron on a normal wing, as in (A); or, for a
tailless airplane the tip surfaces could be operated differentially for aileron action or
unidirectionally for elevator action, as shown in (B).
Clearances
When the horizontal tail surfaces have been positioned on the fuselage
or some other component of the airplane structure, the student should
check allowable clearances with the ground when the elevator is deflected,
and with the vertical tail surfaces, especially the rudder.
VERTICAX TAIL SURFACES
The vertical tail surfaces consist of the fixed surfaces (the rudder), the
movable surface (the elevator), and the trim tab.
The function of these surfaces is to obtain the necessary directional
stability and control in flight.
Figure XIX-19. A "canard-type" airplane with the horizontal tail surfaces located
ahead of the center of gravity. The vertical tail surfaces are located at the wing tip.
404
Directional Stability and Control
Straight flight should be attainable at any speed above 140 per cent of
the minimum speed by sideslipping without the use of the rudder and with
a single engine (of a multi-engine design) being inoperative.
The yawing moment in a sideslip should be such that right-rudder de-
flection would be required for a sideslip toward the left, and left-rudder
deflection toward the right. For small angles of sideslip, the sideslip angle
should be proportional to the rudder angle.
The airplane should always tend to return from a sideslip without the
use of the rudder regardless of the angle of the sideslip.
Rudder control should be such that:
1. A control force of not more than 180 pounds is required to maintain
the airplane in rectilinear flight with one engine inoperative and the other
or others at full rated power at all speeds above the minimum take-off
speed;
2. To meet the spin recovery requirements of the airplane;
3. To overcome the adverse yawing moment caused by aileron deflec-
tions at any speed.
When all controls are released in flight, lateral oscillations of the air-
plane should always damp to one half amplitude within two complete
cycles. When the ailerons or rudder are moved and released quickly,
they should return to their neutral position and damp any oscillations of
the airplane in one cycle.
Both rudder and aileron may employ trimming systems. They should
be:
1. Used if the control forces for level flight are 10 per cent greater than
80 pounds for the aileron control wheel or 30 pounds for the aileron control
stick, or 180 pounds for the rudder pedals for any speed between the maxi-
mum speed and 120 per cent of the minimum speed;
2. Powerful enough to maintain rectilinear flight with one engine inop-
erative (for a multi-engine airplane) at speeds 140 per cent of the minimum
or above.
Further, the approaching stall of the airplane should develop gradually
and make itself felt by increasing "pull force" on the control column, and
by evidence of buffeting of the airplane. Recovery, after complete stall

has been attained, should be possible by normal use of the controls.
Rolling or yawing moments of the airplane at less than 2 degrees above

the attitude required for a 3-point landing should not cause the airplane to
stall.
TAIL SURFACES
405
From the flight criteria mentioned, the designer is able to work back to
the detail design so that the airplane will incorporate the characteristics
desired. Many features are arrived at through past experience or from
examination of existing designs. Other features are subject to research
in the wind tunnel, for there may be a number of variables that require
proper proportions to obtain the desired effect.
Location
The location of the vertical tail surfaces depends upon the type and
speed range of the aircraft. For the small, single-engine private airplane,
the vertical tail surfaces are, almost without exception, located above the
horizontal tail surfaces in order to centralize control systems and simplify
the supporting structure contained in the fuselage.
It is desirable to locate about half of the rudder below the axis of sym-
metry of the fuselage, but this may not be possible because of required
clearance with the ground.
Large airplanes employ multiple vertical tail surfaces (see Figure
XIX-20) for several reasons, although the primary one is to obtain the
advantage of slipstream effect over one of the surfaces when an engine of a
multi-engine design quits. The increased slipstream velocity helps in pro-
viding the greater yawing moment necessary to overcome that produced
by the offset thrust.
In large designs, it is often difficult to
prevent blanketing of the vertical tail
surfaces by the large fuselage so that
dividing the area into several smaller
ones and placing them at the ends of the
horizontal tail surfaces increases their
relative efficiency.
Where the required area is very large,
three instead of two sets of vertical tail
surfaces are used. This reduces the tor- Figure XIX-20. Arrangements
sional load imposed by the outrigger type of twin vertical tail surfaces.
of vertical tail surfaces which also complicates the control-system design.
The extension of the fin area ahead of the middle fin in triple vertical
tail-surface design has often been added after flight tests indicated that
additional vertical tail area was necessary. While inefficient aerodynami-
cally, such extensions do not impose very much additional torsion on the
fuselage.
To a less extent, multiple vertical tail surfaces are used in order to re-
duce the over-all height of the aircraft structure, especially in those designs
for which hangar door clearances are the determining factors.

406
For high-speed aircraft, the vertical tail surfaces may be displaced at
the wing tips of the swept-back wing, where they obtain a sufficiently long
moment arm to produce a satisfactory yawing moment.
The placing of the vertical tail surfaces at the end of the horizontal tail
surfaces helps, although to a very small degree, to increase the effective
aspect ratio of the horizontal tail surfaces.
In some cases the vertical tail surfaces are slightly tilted from the ver-
tical. This is usually the result of the dihedral incorporated in the hori-
zontal tail-surface design where dihedral is employed to account for the
downwash of the wing.
Angular Deflection
The rudder has an angular movement of a maximum of 30 degrees each
side of neutral. It is generally desirable to have all primary control
surfaces operate through the same angular range so that a pilot transferring
from one airplane to another is at least likely to be familiar with the
amount of control to be expected for the control-stick or rudder-pedal
movement.
The fin may be adjustable to offset the yawing moment induced by the
means used to produce a rolling moment of the wing to offset the torque
of the reciprocating engine.
Aspect Ratio
The aspect ratio of the vertical tail surfaces may be somewhat restricted
by the possible torsional moment imposed upon the fuselage structure
since the vertical tail surfaces are usually asymmetrical about the longi-
tudinal axis. Since the tail surfaces are at zero angle of attack throughout
the flight regime, aspect-ratio effects play a very small part, if any, in
Figure XIX-21. For aspect-ratio calculations, the surface area represented by the
side of the fuselage aft of the dotted line is included in the total area; but it is not con-
sidered as part of the vertical tail surface area in the evaluation of the ratios of tail
surface area to wing areaâ€”which ratio may vary from .08 to .12.
TAIL SURFACES
407
the over-all aerodynamic drag of the airplane. The main consideration
for the vertical tail surface is to locate it so that it is not made ineffective
by blanketing of the fuselage or horizontal tail surfaces. To minimize
such blanketing, a higher aspect ratio for the vertical tail surfaces often
becomes necessary.
The aspect ratio of the vertical tail surfaces should be between 2 and 4.
It is difficult to state exactly what the aspect ratio of the vertical tail sur-
faces may be, because the rear portion of the fuselage influences the vertical
tail surface effectiveness.
Area
The size of the tail surfaces is dependent upon the location; the greater
the distance between the center of gravity of the airplane and the center
of pressurôf the vertical tail surfaces in terms of the meanîerodynamic
chordôfThe wing^the smaller the area needs to be. Where the vertical
tail surfaces are attached to the wing tips, the relative distance is likely
to be smaller, so that the area would have to be proportionately greater
in order to obtain the same degree of directional stability and control.
The area of the vertical tail surfaces ranges from !10jto 15 per cent of
the wing area, with about 30 to 50 per cent of the area devoted to the
rudder. An exception is provided by those designs where dorsal or
ventral fins are employed, in which case the movable portion of the vertical
tail surfaces is likely to be no more than 30 per cent of the total area.
The dorsal fin (see Figure XIX-22) may be used to increase the fin area
in order to (1) increase the directional stability if the original surface
Figure XIX-22. The dotted lines show various ways of increasing the fin area.
The top diagram illustrates a dorsal fin, while the two bottom diagrams illustrate the
vertical-type fin.
408
proves inadequate, (2) transmit the loads on the fin to a greater number
of fuselage frames, (3) reduce the torsional moment about the longitudinal
fuselage axis, (4) reduce the over-all height dimensions of the vertical tail
surfaces, and (5) obtain a possible weight saving, although the total fin
area is likely to be greater for a dorsal fin than for a normal fin-type
surface.
The ventral fin (See Figure XIX-22) is another solution to increasing
the fin area. Since it is located below the fuselage, it is not blanketed
by any of the aircraft structure and is likely to be even more effective
than the dorsal fin.
Planform
The vertical tail surfaces have a variety of planforms, depending some-
what upon their location. A few likely designs are illustrated in Figure
XIX-23.
Figure XIX-23. Typical planforms of vertical tail surfaces.
Sweepback
For transonic and supersonic designs, the vertical tail surfaces must
incorporate sweepback. For lower-speed aircraft, on the other hand,
the sweepback observed in the design of the vertical tail surfaces is a result
of other design considerations such as planform, appearance, and position
of the fin-post.
TAIL SURFACES
409
Tail Length
For subsonic designs of conventional layout, the tail surfaces have been
so located that the distance from the center of gravity of the airplane to
the estimated center of pressure of the horizontal or vertical tail surfaces
is from 2.5 to 3.25 times the mean geometric chord of the wing.
As speeds approach the supersonic, the "trim drag", especially with
deflected surfaces, becomes an appreciable portion of the total drag of
the airplane. To ameliorate this condition, the tail length is made as
long as possible, so that smaller areas may be employed for the tail sur-
faces and relatively little deflection is necessary for the desired amount
of control. Both are attempts to reduce the drag. However, there is a
point of diminishing return in that a longer fuselage also implies a greater
skin drag, and thus a gain in one may be offset by a loss in the other.
|- A M
Figure XIX-24. The distance A for conventional airplanes should be from to 3
times the mean aerodynamic chord of the wing. The angle B, corresponding to the
maximum deflection of the elevator, should permit the tailing edge of the elevator to
clear the ground comfortably.
BUTTERFLY OR VEE TAIL
This type of tail surface combines the vertical and horizontal tail
surfaces in one, as shown in Figure XIX-25. The vertical component of
the lift corresponds to the normal tail-surface load, while the horizontal
Figure XIX-25. A "Vee" or "Butterfly" tail.

410
component of lift when both halves of the surfaces are neutral is zero,
unless both surfaces are at an angle to the normal direction of flight
when the horizontal component of lift produces the necessary yawing
moment to weather-cock the airplane.
The movable trailing-edge flaps act as elevators when both are deflected
in the same direction, and as rudders when operated in the opposite
direction. It is also possible to operate them in a combination of the two
motions so that simultaneous pitching and yawing moments are obtained.
The movable surfaces are called either ruddervaters or elerudders.
The cockpit controls are the same as for the normal type of rudders and
elevators, although the control mechanism from the cockpit to the tail
surfaces is a little more complicated.
The advantages claimed for such an arrangement are:
1. A saving in weight due to its simplicity of construction, its fewer
elements, and the smaller total area possible.
2. Higher maximum speed, due to less area and consequently less pro-
file drag as well as less interference drag.
3. Better spin recovery due to less blanketing of the tail surfaces.
The detailed design information is to be found in an NACA report.
For preliminary design purposes, the following relationships may be
considered:
The effective horizontal tail surface area
SH = S cos 0; (1)
oc
nOOOC
) O O Oil
DO
JUj
uihrj
â€” i
<
5â€”4
1= \
Figure XIX-26. Dive brakes are used to reduce speed of aircraft and may be used
as spoilers for additional aileron or directional control, depending upon the number
used and their location. When not in use, they recessâ€”flushâ€”into the top surface of
the wing.
TAIL SURFACES
411
The effective vertical tail surface area
Sv = S sin 0. (2)
Dividing (2) by (1),
tan /3 =
so that using the proportion of areas needed for the conventional vertical
and horizontal tail surfaces, the value of tan /J can be determined where /3
is the dihedral angle for the tail surfaces. Once the value of /3 has been
found, the total area of the Vee tail can be calculated from either equation
(1) or equation (2).

CHAPTER
Control Systems
The control system incorporates all the operating mechanisms located
in the cockpit, the connections between these mechanisms, and the control
surfaces.
TYPICAL SYSTEMS
The systems commonly employed are:
1. The push-pull system (Figure XX-1) consists of tubes which may
take tension and compression and whose direction may be conveniently
changed by suitable mechanical devices, usually bell-cranks.
The push-pull system is likely to be heavy since any member subjected
to compression is larger than one subjected to tension. Where the struc-
ture, as the wing for example, has large deflections, the push-pull control
system may be bent with the wing causing difficulty of operation. It may
be used for part of the system in combination with others, and for engine
controls, and rudder and elevator controls. Its advantage is that it has
no appreciable stretch to cause lag in operation of the movable controls.
2. The cable system, most extensively employed because deflections of
the structure to which it is attached do not affect its operation, consists of
a continuous system of cables operating over pulleys and through fairleads.
Since cables have a tendency to stretch, they are prestretched before in-
stallation to about 60 per cent of their rated strength; some manufacturers
install them with an initial tension of 100 to 150 pounds. Long cables,
particularly, may have a large amount of stretch unless these special
precautions are taken to limit it.
3. The torque-tube system, which converts lineal motion into rotary mo-
tion, has relatively few members and usually its installation is compara-
tively simple structurally. Long members under torsion have consider-
able angular deflection which is liable to cause a time lag in operation of
412
CONTROL SYSTEMS
413
controls or, if stops are provided in the system, the angular deflections
may be large enough to reduce seriously the angular depression of the con-
trol surface it is to operate. To limit the angular deflections, relatively
rigid members (offering greater weight) have to be employed.

414
In designing control systems, especial care should be given to assure
that there is no binding, change in tension, or interference of the various
components comprising the system as well as possible interference with
the attaching structure. Extreme weather conditions require considera-
tion of thermal expansion and contraction.
Adjustable stabilizer-elevator combinations, especially when the stabi-
lizer is adjusted to its extreme position, may cause either interference with
other parts of the control system or excessive tension in the rigging of the
surfaces or in the control system.
PULLEYS
Cables passing over pulleys are kept in place by guards. These should
be close-fitting to prevent jamming or to prevent cables from slipping off
when they slacken due to temperature variations.
Figure XX-2. Schematic diagram of the tension cable system for control surface
operation. Motion caused by rotation of control wheel for aileron, and by fore and
aft motion of the control column. Rudder operation is obtained by another control
system employing a cross-bar or a pedal assembly actuated by the foot.
The pulleys should be located in such a way that the plane of rotation
of the pulley lies in the plane determined by the cables. There should be
no rubbing of the cable against the side of the pulley flanges.
CABLES
Control cables are designated by the number of strands such as 6 X 19
or 7 X 19 extra flexible. These are generally preferred because of their
greater sturdiness, but 6 X 7 or 7 X 7 flexible cable of ^2-inch diameter
may be used where especial care is taken to provide against wear. Cable
smaller than %a-inch diameter may not be used in the primary control
CONTROL SYSTEMS
415
system except for tab-control systems, provided the airplane may still be
controlled should such cables fail.
In general, cable sizes are determined by considerations of control-system
deflections previously discussed as well as by strength requirements.
Table XX-1 shows weights and lengths of cables.
FAIRLEADS
Cable systems, or chains, or connecting links of the control system
should be supported at inter-
vals to prevent chafing of the
system against adjacent struc-
ture or possible interference
with other components of the
same system. For such pur-
poses, fairleads are employed.
These should not be used to
change the direction of the
cable, although for relatively
low cable loads, a change up to
3 degrees is permitted.
Wherever guards or fair-
leads are used, means should
be available for easy inspec-
tion and maintenance.
Figure XX-3. A fairlead block attached to
a metal bracket to prevent chafing of the metal
control cable.
In "open" systems for the
aileron-control system where
the balance cable between the
ailerons is the only system for returning the ailerons to neutral, the effects
of opposite loads on the wheels or sticks of a dual-control system may
cause such deflections (stretch) in the control cables or chains that jamming
may occur unless close-fitting fairleads or guards are used.
STOPS
Adjustable stabilizers require stops at their extreme limits in case the

adjusting mechanism should fail. Elevator trailing-edge tab systems also
should be provided with stops to limit the travel. It is generally wise to

incorporate some form of stops on all movable surfaces to avoid interfer-
ence with the adjacent structure. It is also desirable to have additional
stops located close to the operating force in order to avoid a "springy"
control.
416
si
03 g
Â° 8
CP
iS
*j
C Â»H
S5 S
11
Ex
Weight ser
thouOnd
feet (lb)
noiusOHqoqqooqo
Minimum
b00aking
strength (lb)
rtrtHNft
lOOOQOOOOOOOQOÂ©
OO00OOOOQOOOOOOQQ
i-T pa to to oo" o cf h--
Size
(in.)
'$ '% 'g k 'S 'Â§ k pÂ§ > 'St '% f E >
Weight ser
thouOnd
feet Ob)
m Â« Â« Â« n .* xoo n Â«o eÂ« m â€¢ sÂ«Â»
Minimum
b00aking
strength (lb)
SSSSSoSqoS0
Oize
(in.)
Weight ser
thouOnd
feet Qb)
Â»-H l-( N
Minimum
b00aking
strength Oh)
S- <n Â«f ^- oo~ of eÂ£
Size
(in.)
Â£ g
.a I
St-
418
Figure XX-5. Aileron control unit. The mechanism employed here may be used
with equal effectiveness for the operation of any movable surface. (Courtesy Convair.)
422
time. Positive means should also be incorporated for providing positive
retention of the flap in the fully retracted position.
WING FLAPS
In addition to the usual air loads acting upon the deflected wing flaps,
high local loads may also be caused by the impact of water, especially
when low-wing airplanes are operated from wet fields or water. For these
reasons, it is desirable to make the flaps sturdy. When the piano-type
hinge is used, a series of short sections rather than one long section should
be used not only for ease of maintenance but also for better operation.
Ground clearance of the flap for low-wing installation should be carefully
considered, especially in the initial design stages because subsequent flight
tests may indicate that a greater angular deflection of the flap may be de-
sirable. The ground clearance is least for the largest angular depression
of the flap. For a start, 12 inches of ground clearance is assumed reason-
able.
TABS
Tabs for control surfaces should be rigid and well anchored to prevent
damage or misalignment from handling. For small airplanes, such tabs
usually consist of thin sheets attached to the trailing edge. In this case,
care should be taken that successive bendings of these tabs do not interfere
with the performance of the main surface or do not damage the tab itself.
HINGES
Hinges for control surfaces need careful study. It is usually desirable
to provide at least three hinges for control surfaces to assure operation
of the surfaces. The deflections of the surfaces and the structure to which
they are attached should be studied.
Usually it is desirable to so space these hinges that the load for each
hinge is the same. However, a study of deflections may indicate another
disposition. For initial design, hinges may be spaced equidistant since
subsequent stress analysis is necessary to indicate likely deflections.
Bearings usually allow a few degrees for misalignment, but if deflections
cause too much misalignment, the surface cannot be deflected at all.
Closed piano-type hinges are acceptable when they are positively at-
tached (other than wood screws) as close to the hinge line as possible.
Several sections should be used to reduce inspection and maintenance

problems and they should not be located near the control horns where the
load is concentrated.
Strap-type hinges are generally not desirable since the wear is appre-
ciable, and their repair relatively difficult.
Provision for lubrication should be provided wherever self-lubrication
methods or sealed bearings are not used.

CONTROL SYSTEMS
423
FLAP- AND TAB-CONTROL LOADS
For the upper limit, a torque of 133 inch-pounds applied to the control
knob in "twist" controls may be considered a design factor. Where con-
trol wheels, cranks, levers, or handwheels with a well-defined rim are used,
a torque of T = 100 R (where R is the radius of the wheel) should be con-
sidered.
HYDRAULIC AND PNEUMATIC SYSTEMS
Hydraulic and pneumatic systems are increasingly used to replace
manual operation of all devices which are changed from one position or
condition to another for which the time element may not be important
and where rapid reversals of such motions are not required.
The widest application so far has been for cases requiring lineal motion
although rotary motion can also be obtained through the use of hydraulic
motors employing a variable-volume hydraulic pump as a pressure source.
Hydraulic systems have been used extensively for actuating:
1.
Retractable landing gears
8.
Fuel dump valves
2.
Wheel brakes
9.
Tail wheels
3.
Automatic pilot
10.
Hydraulic motors for starting
4.
Trailing-edge wing flaps
auxiliary engines
5.
Engine-cowling flaps
11.
Remotely-mounted fuel
6.
Windshield wipers
pumps
7.
Full-feathering hydromatic
12.
Folding wings or wing tips.
propellers
All these could also be operated by electrical means and for military pur-
poses. Such means would be preferable because of the vulnerability of
the hydraulic systems. However, weight saving is always an important
matter in commercial design and the hydraulic systems are considerably
lighter than a comparable electrical system.
Hydraulic systems are generally preferred because of:
1. Flexibility of control permitting immediate and positive starting and
stopping of operating mechanisms.
2. Simple variation of output by modification of bore and stroke.
3. Ease of installation.
4. Ease of inspection and maintenanceâ€”both of which are done by any
mechanic more quickly than for the equivalent electrical system.
5. Reserve power available (when pressure accumulators are used)
when the engine quits.
In large airplanes several hydraulic systems may be required since pres-
sure and fluid requirements vary. Usually one system is installed for use
with engine oil for operation of full-feathering hydromatic propeller, and

424
another for use with Lockheed hydraulic fluid for operating brakes, re-
tractable mechanisms, flaps, and the like.
Hydraulic systems for aircraft may operate from 1000 to 1500 pounds
per square inch for which 52SO aluminum alloy tubing is usually employed.
Occasionally the systems may be required to operate at 3000 pounds per
square inch or higher, for which stainless-steel tubing may be employed.
Hydraulic
tubing
Support
bracket
Support
bracket
Figure XX-10.
cylinder.
Actuating cylinder
Sketch showing one method of installing a hydraulic actuating
The diameter of the tubing is selected on the basis of fluid flow required.
The wall thickness is determined on the basis of allowable stress, required
load factor, and commercial thickness available. The average fluid flow
for certain tubing sizes is listed here.
Nominal Tubing,
Outside Diameter (in.)
y*
Average Flow Desired
(gal per min)
1.2
2.3
3.5
6.0
10.5
16.0
29.0
Barlow's formula for wall thickness is
KPD
t=
2F
lu
CONTROL SYSTEMS
425
where /
= wall thickness in inches,
= safety factor required (usually about 5),
= nominal operating pressure, pounds per square inch,
= outside diameter of tube in inches,
= ultimate tensile stress of the material.
Pneumatic systems, because of the availability of a large air supply and
the absence of extreme changes in viscosity, are used for aircraft powered
with jet engines.
Above 70,000 feet, air compressors will probably not be used since the
ambient air is too thin to compress efficiently. Furthermore, the turbo-
jet engine will need all the air it can get, leaving none for such functions
as cabin pressurization or operation of controls.
A pneumatic source will probably still be necessary, but the use of air
will most likely be restricted to emergency operations such as canopy or
stores jettisoning. Seals and joints of the entire operating system would
have to be completely leak-free. The power or supply source may be
high-pressure air bottles or specially designed compressed-air storage
areas within the aircraft structure.
Another source of pneumatic povrer could be nitrogen or another gas in
liquid form. This would require an insulated, high-strength container.
"Pseudo-pneumatics"â€”still another source of powerâ€”makes use of
the expanding gases produced by the firing of explosive cartridges to
operate certain pneumatic components, such as starters, and could be
adapted for other functions requiring reliability and instantaneous action.
TRAVEL OF CONTROLS
It is generally desirable to standardize on the travel allowed for the
various controls so that a pilot, in transferring from one type of airplane
Figure XX-11. A schematic diagram of a simple irreversible control mechanism.
(See also detail in Figure XX-5 for irreversible control.)

426
to another, does not have to become accustomed to a new set of controls
and lose much valuable time in the learning process. The travel of pri-
mary control elements seems to be dependent to a great degree on the size
Figure XX-12. Pulleys guiding and supporting control cables. Pulleys are also
used to change the angular direction of the cables.
of the aircraft. Stick travel at the grip may vary 18 inches fore and aft,
and 18 inches side to side, to somewhat smaller dimensions for the light
airplane. Angular travel of the control wheel from neutral may vary cor-
respondingly from 270 degrees to 90 degrees. The usual pedal travel
amounts to a total of 6 inches.
Adjustment in the controls, especially in the rudder controls, seems 'to
be preferred to fore and aft adjustment of the pilot's seat.
Irreversible Controls
In subsonic designs where the control forces caused by hinge moments
are not too severe, irreversible controls are used primarily for reducing
the chances of flutter. For higher-speed designs, especially those employ-
ing boost controls, irreversible controls are required.
Above Mach 0.8, effectiveness of the control surfaces drops off since
the upstream surfaces are no longer influenced by the control surface
deflections. However, the stick forces, and the hinge moments that
cause them, increase rapidly as the speed of the aircraft increases. Ir-
reversible power controls are therefore required for such craft. An
artificial feel corresponding to the actual pressures exerted is usually
introduced in such systems so that the pilot may also have some psycho-
logical response to the increased forces.

CONTROL SYSTEMS
427
DETAIL REQUIREMENTS
Detail requirements specified for a design may be of interest.
Controls
1. A dual set of controls should be readily removable. All controls not
duplicated should be so located as to be conveniently accessible to the
co-pilot but no control should be so
located as to constitute a hazard when
the front seat is occupied by a pas-
senger.
2. All control surfaces should be
adequately balanced.
3. Adequate controllable trimming
tabs should be fitted to the elevator
and rudder.
4. Ball- or roller-bearing hinges
should be used throughout.
5. The rudder pedals should be ad-
justable fore and aft.
6. The rudder pedals should con-
trol the differential action of the
brakes. (Note: The brake controls
preferably should be mounted on the
rudder pedals and provided with ad-
ditional parking brake lock.)
7. A suitable parking brake ca-
pable of holding the plane against
maximum thrust should be provided.
8. Full operation of high-lift de-
vices should not require an independ-
ent readjustment of the trim control
for trim.
9. Electric, hydraulic, or mechan-
ical operation of high-lift devices and
retractable landing gear, if provided,
is required and full operation in either direction should not require more
than one minute.
Figure XX-13. A detail of the con-

trol wheel: 1-2 elevator up cables; 3-4
elevator down cables; 5-6 aileron cables.
Engine Controls
1. Engine controls consist of all mechanisms exterior to the engine
required for controlling spark, throttle and mixture adjustments, as well
as cowling shutter controls, and the like.

428
Figure XX-14. Extruded components of a piano hinge (A); machined (B);
assembled (C). The bottom figures show the top view and cross section of such a hinge.
2. Engine controls should be placed at the left of the pilot's seat, except
for side-by-side seating of pilots, when they should be located between the
pilots' seats.
3. All engine controls should be marked plainly to show their function
and method of operation.
4. Throttle control and ignition switches should be easily accessible to
the pilot and so arranged as to afford a positive means of controlling all
engines separately or simultaneously.
5. A positive means for shutting off all ignition must be readily acces-
sible to the pilot.
6. Throttle controls may be operated by an approved positive-action
cable or wire control system.
7. The controls should be positively operated and springs should not
be relied upon to actuate the controls in either direction. Push-and-pull
rods should be used wherever possible.
8. Engine controls should be so designed as to avoid the undoing of a
large number of bolts or unions to take the rods apart.
9. Ball-bearing supports should be provided wherever possible.
10. Each control system should have one lever of adjustable length.
11. All rod lengths should be adjustable and stiff enough to take 70
pounds on the control handle without failure.
12. Adjustable stops should be provided.

CONTROL SYSTEMS
429
Figure XX-15. Piano hinge, unbalanced control surface.
ELECTRICAL SYSTEMS
By use of motors and suitable linkages, and mechanisms, the hydraulic
system may be completely replaced by an electrical system.

CHAPTER XXI
The Fuselage
Even though comparatively little thought seems to have been given to
the size, shape, and structure of the fuselage up to now, it has been practi-
cally designed in most of its major elements while the pilot's cockpit, the
passenger cabin, the placement of the wing, and the balance diagram were
under consideration.
For conventional designs, the fuselage and its contents usually represent
the largest concentration of mass items. Under acceleration, the relative
effect of loads due to inertia forces is therefore greatest on this structure.
Furthermore, the conventional fuselage is the connecting link for the
various appendages that transmit the aerodynamic and landing loads to
it. Because of these loads of significant proportions, the structure must
be strong and efficiently designed.
The fuselage serves another important function: it represents the most
basic- part of the airplane, for it houses the crew, cargo, instrumentation,
etc. This serves to present problems in such important areas as heating,
ventilating, and air supply.
The fuselage design therefore requires the consideration of a great many
factors and the adaptation of many compromises. In arriving at a final
design, it is advisable to prepare a large number of detail designs before
the final compromise is accepted. Once a likely arrangement of struc-
tural members, openings, and attachments has been agreed upon, a more
detailed study of the component parts is possible.
WING-FUSELAGE CONSIDERATIONS
In discussing various aspects of wing design, the matter of wing-fuselage
configurations was included in some detail, primarily from the standpoint
430
THE FUSELAGE
431
of the wing structure. However, the suggestions brought forth there
should be referred to again when considering the design of the fuselage.
For airplane speeds of less than 350 miles per hour, the problem of
reducing the drag of wing-fuselage configurations has been met primarily
by careful streamlining. In general, a fuselage that by itself represents
a minimum drag form, when combined with the wing produces an optimum
combination. Filleting at the juncture of the wing and the fuselage helps
to reduce the drag.
NACA TR 540 entitled "Interference of Wing and Fuselage" covers
extensive wind-tunnel tests made on various wing and fuselage combina-
tions. Parasol arrangements with round fuselages seemed to be best if
drag of supporting struts were neglected; otherwise, a slightly higher than
midwing arrangement was next best; forward positions of the wing with
respect to the fuselage seemed favorable; low-wing positions are unfavor-
able, except by means of adequate filleting. Efficient airfoils of moderate
thickness and low camber are most susceptible to adverse interference
effects.
In general, acute angles included between the fuselage and attaching
wing or tail surfaces should be avoided.
When multi-engine designs are considered, care has to be taken that
the nacelle shapeâ€”particularly if it is long and of a cross section comparable
in magnitude to the fuselage cross sectionâ€”does not cause interference
effects when the nacelle is placed too close to the fuselage or to another
nacelle. Wind-tunnel tests are necessary to determine the optimum
configuration. The interference effects are manifested not only in a drag
that is higher than would be expected from consideration of the drag
characteristics of the individual components alone, but also in longitudinal
and directional stability problems.
With transonic speeds, the problem of wing-fuselage integration becomes
more acute. For the large delta wing, the fuselage practically disappears
within the contour of the root airfoil and therefore presents no serious
problem. For other designs, a more careful study must be made of the
spanwise cross-sectional area distribution along the longitudinal axis of
the airplane according to what is known as the Whitcomb's Area Rule.
The method is illustrated in Figure XXI-1. The side and top views of a
proposed design are shown in the top two figures. The cross-sectional
areas normal to the line of flight are plotted against the longitudinal axis
of the airplane, as shown in the bottom figure. Curve ABCEF represents
the cross-sectional area of the fuselage and the tail surfaces. Curve BDE
is the cross-sectional area of the fuselage normal to the longitudinal axis
of the airplane, plotted on top of the fuselage curve. An "ideal" com-
posite curve, approximating that of a volume of revolution with good
transonic (and speeds beyond) drag characteristics, is shown in curve

432
Figure XXI-1. Application of Whitcomb's Area Rule.
ACDG. The station where the pilot's cockpit would be located was taken
as the determining diameter. As a result of this "ideal" curve, for the
cross-sectional area distribution, it would be necessary to reduce the cross-
sectional area of the fuselage by the amount indicated by the cross-hatched
portion of the curve ABDCA, and to increase the cross-sectional area of
the rear portion of the fuselage as indicated by the cross-hatched portion
of curve DGFE.
Since the fuselage has to be reduced in cross-sectional area in the region
of the wing and power-plant nacelles, it is likely that for airplanes flying
at transonic or supersonic speeds and required to carry many passengers,
the fuselage may become very long in order to have the cargo and pas-
sengers ahead of the wing. To compensate for the large mass ahead of
the center of gravity, the power plant will be carried somewhat aft of the
center of gravity.
The student should study the latest NACA reports on this subject in
order to obtain a further insight into the basic aerodynamic problems
involved in the design. Design of high-speed aircraft today requires
careful integration of the aerodynamic, structural, and power plant
aspects.
SHAPE OF FUSELAGE
The length of the fuselage is determined by the cockpit and cabin con-
siderations in the front and the location of the tail surfaces and the tail
wheel at the rear; the over-all depth and width by cabin requirements.
THE FUSELAGE
433
For efficient structural design, a circular cross section is desired, but this
shape is inefficient for accommodating a rectangular cross section required
by the cabin so that the natural compromise is an oval, perhaps with greater
width at the top than at the bottom. Figure XXI-2 shows three basic
fuselage shapes.
The various cross sections of the fuselage may vary from a circular sec-
tion at the engine mount to an oval section and finally to a circle or a rec-
tangle at the tail post. It is, therefore, very important to fair one section
into another very carefully, and to avoid double curvatures (two curvatures
at an angle to each other) wherever possible. Circular cross section for
the fuselage offers theoretically greater strength for shell structures, but
the inscribed rectangle indicates the available useful space. Rectangular
cross section permits the most economical use of the space but is not suit-
able for shell structures. Oval or elliptical outline for the reinforced
monocoque or shell-type fuselage is the best compromise between the cir-
cular and rectangular cross section.
The fuselage structure is determined gradually, and often, when it is
not possible to compromise, radical changes in the wing planform or in
the interior arrangement may result. For example, it may be desirable
to have the front-spar double frame intersect the fuselage at the front
cabin wall; but in order to do so, the wing has to be given an appreciable
sweepback in order to satisfy not only the particular condition just men-
tioned but also to obtain the proper location of the mean geometric chord
with respect to the center of gravity.
Figure XXI-3. Cross sections of two fuselage shapes designed for large volume
capacity. The two intersecting circular shapes lend themselves to two compartments
without too great an increase in structural wieght.

434
Since all these items affect the size and shape of the fuselage, it is obvious
that the arrangement and general dimensions of the pilot's cockpit and
the passenger cabin are the primary considerations. Only after all this is
done should the contours of the fuselage be settled. The cross sections of
the fuselage will vary from circular at the nose to slightly elliptical at the
pilot's cockpit, and then to an elliptical section somewhat modified by the
intersection of the root of the wing, and then to an elliptical or oval section,
sometimes even rectangular (with the long side vertical) at the tail post.
The object of approaching the elliptical section is that it represents the
best compromise between the square or rectangular section which is most
efficient for space utilization (as cabins, cockpits, etc.) and the circular
section which is most efficient from a structural point of view for reinforced
monocoque structure, and for aerodynamic reasons.
If the fuselage is to be designed for a single-engine airplane, the type of
engine and its cowling determine the nose treatment of the fuselage. The
pilot's cockpit governs the development of the windshield and that section
of the fuselage directly behind the engine. If there is no nose engine, as
in multi-engine designs, the pilot's cockpit has even a greater influence on
the development of the front section of the fuselage.
The main section of the fuselage is built around the passenger cabin, and
then the rear portion of the fuselage is gradually tapered back to the tail
post.
Other considerations affecting fuselage shapes have been referred to
under "Wing-Fuselage Considerations."
For a preliminary approach, the student will find the following pro-
cedures useful in arriving at a reasonable solution.
1. Lay out the cockpit to scale, showing the important items to assure
that proper clearances have been provided. Since the relationship of the
cockpit to the cabin may vary, it is helpful to draw the layout of the
cockpit separately, so that it may readily be moved up or down or forward
or rearward, as the case may be, before being traced on the fuselage
drawing.
2. Lay out the cabin and cargo areas to scale so that the limiting interior
dimensions can be ascertained. This layout, too, should be a separate
drawing.
3. Draw the root airfoil at the fuselage intersection as a separate draw-
ing, indicating the principal structural members which will subsequently

point out the location of the frames.
4. Draw the empennage, if there is one, as a separate drawing.
5. These separate drawings can now be assembled, by tracing, in a
final drawing of the fuselage interior, and an inboard profile can be ob-
tained. If the immediate goal is to obtain the fuselage shape, many of

THE FUSELAGE
435
nn
(a) (b)
Figure XXI-4. Rear fuselage cross sections just ahead of the tail surfaces often
look like this. If the cross section in (A) is used, the horizontal and vertical tail surfaces
are more likely to be blanketed than if the cross section in (B) is used.
the interior arrangements may now be omitted, since the immediate
purpose was to determine the limiting interior dimensions.
FUSELAGE LENGTH
For subsonic airplanes having wings of aspect ratios between 6 and 14,
the ratio of fuselage length to wing span is from .60 to .70; for supersonic
airplanes having a wing aspect ratio varying from 2.5 to 4.5, the ratio of
fuselage length to wing span becomes materially greater, not only because
of the shorter wing span but also because the fuselage must be longer.
In general, there is as much or more of the fuselage length ahead of the
wing as there is behind it, and ratio of fuselage length to wing span varies
from 1.2 to 2.5.
The optimum fineness ratio (length to maximum diameter) is 16, with
the volume of revolution such that both ends are pointed. However,
"boat-tail" bodies are the practical compromises.
DETERMINING FUSELAGE LINES
The contour of a fuselage has to satisfy not only certain aerodynamic
minimums, but also production requirements. These vary with each
design. In any case, the development of the fuselage lines entails the
establishment of second-degree curves. Various methods available for
determining fuselage lines are:
1. The mockup, from which the lines may be transferred to suitable
templates. These pose problems in storing and handling but are never-
theless necessary, even in conjunction with other methods.
2. Lofting of curves requiring full-size layouts on large metal sheets
painted a dull white. This method has been adopted from the marine
field, where it is common practice to lay out the lines on a large lofting
436
floor. In this method, curves are laid in by long flexible splines and are
faired by eye. To determine the agreement of curve sections, different
angles through the body have to be taken to determine how well the
curves faired; the method is long and tedious. Again, permanent records
require large steel plates which have to be filed.
3. Mathematical techniques have been developed to achieve a more
uniform and accurate resultâ€”one that is easily reproducible and where
storage presents no problem. The second-degree curve technique based
on curves defined analytically by an equation of the form
Ax* + Bxy + Cy* + Dx + Ey + F = 0
is the one most often used. By properly assigning values to the con-
stants A, B, C, D, E, and F, any useful curve can be defined.
USE OF THE MOCKUP
Since it is difficult to visualize a structure as complex as the fuselage
and the arrangement of its interior, it is customary to build a full-scale
mockup, usually made of materials at hand that require little fabrication.
In such a mockup the transverse rings may be represented by plywood
forms that may be easily sawn to shape and to which attachments may be
easily nailed, bolted, or screwed.
The covering may be represented by thin plywood sheets or even heavy
cardboard. Where double curvatures exist, thin sheet metal, properly
formed, may be inserted. Seats, equipment, instrumentation, and such
may be represented by the actual prototypes or reasonable facsimiles.
The mockup can be used to determine external line, even if a more
analytical approach may have been used in determining the lines of the
fuselage, since some solutions are arrived at only by trial and error methods.
By building a mockup, a fuselage form may be evolved which may use
flat sheet material without resorting to double curvatures, and so help in
materially reducing the production costs. If the covering has a curvature
in one direction only, a flat sheet may be used. If, however, there is a
curvature at an angle to the main radius of curvature, the covering has
what is known as a double curvature which must be produced by forming
the sheet either by gradual "bumping" or by hydraulic presses over special
dies.
The mockup also helps to indicate whether any other modifications in

arrangements are desirable. This method is a simple and inexpensive way
to solve many problems usually too intricate for the drafting board, and it
also helps to visualize a design far better than a series of drawings could
possibly do.
Figure XXI-5. The fuselage structure, consisting of the skin and longitudinal
stringers, has to withstand the bending moment caused by the inertia and tail loads.
The above diagram illustrates the bending moment distribution for a fuselage to which
a two-spar wing is attached. To best sustain the bending moment, the fuselage, con-
sidered as a beam, has to be deep with bending elements as far away from the neutral
axis as practicable.
ANALYTICAL STUDIES OF FUSELAGE STRUCTURE
Assuming, then, that the outside contour has been definitely established,
the next step is to locate the structure. Wherever concentrated loads
have to be transmitted to the fuselage (as from the wing spars, or landing
gear members, or tail surface spars), a double or reinforced frame has to be
located. For example, there would be double frames at the front and rear
spar locations of the wing. Between these double frames, single frames
would be located. The spacing of these frames may be governed by the
window locations. How often do the front spar frames intersect a window
which has been painstakingly placed so as to give the passenger the
proper vision! It is generally not desirable to break or distort these main
frames just to accommodate a window. In such a case the difficulty
may be adjusted by relocating the window, or relocating the spar so that
the frames may be changed.
Naturally there will be frames at each side of the door, for such frames
will provide convenient hinge supports as well as door frame supports.
The front and rear cabin wall will determine another station where frames
should be located to provide anchorage for the cabin. Such considerations
gradually "build up" the structure until a few intermediate frames are
added just to cut down the unsupported length between frames. Similar
considerations determine the location of the longitudinal members.
Some insight may be gained into the desirable sheet, stringer, and
frame-spacing combinations by investigating, for bending stresses, a
fuselage section close to the rear wing spar or shear web. The approxi-
mate bending moment at this station could be found from previously-
determined tail load and from the inertia loads of the structure and items
of equipment that the tail portion of the fuselage contains.

438
The investigation may be carried out in the following steps:
1. Select one or more longitudinal stringer sections that are considered
to have good crushing-strength characteristics. A closed section may be
desirable at times, but production considerations may make an open
section more fitting.
2. Determine the bending stresses for a number of combinations of skin
thicknesses, frame spacings (which determine stringer column lengths), and
stringer spacings. Reject any combinations for which the applied stress
exceeds the allowable stress.
3. Compare the weights of the various combinations investigated.
4. Check the more promising combinations for stresses imposed by the
transverse shear and torsion.
Such studies can result in optimum designs for the combinations chosen.
They would have to be modified by other considerations in the most highly-
used section of the fuselage, where cut-outs, floors, equipment, cargo, and
the like impose other problems.
The main portion of the fuselage structure, aside from the frames just
discussed, is the skin and supporting longitudinal structure. Much de-
pends upon the proper proportioning of the frames and stringers.
For example, if the fuselage is without any reinforcement longitudinally,
the wall thickness of the shell would have to be considerably greater to
avoid local buckling. Such a fuselage structure could be employed with
profit for small, light airplanes where the thickness, originally determined
by maintenance considerations, would be strong enough.
If the skin is partially reinforced longitudinally by a few members, the
skin contributes little to the strength of the fuselage.
When enough longitudinal members are used, the skin will add mate-
rially to the bending strength. Proper balance between stringers and skin
will provide the best combination from a strength-weight ratio.
Instead of stringers, corrugated sheet covered with flat sheet will give a
very rigid structure but, unfortunately, somewhat unsuitable where double
curvatures are required.
A unique form of construction is the so-called geodetic structure which,
essentially, employs stringers that wind around the fuselage contour.
Figure XXI-6. The fuselage structure may "sit" on top of the wing structure with
connections at the wing spars (see chapter on Design of the Wing). The leading and
trailing edge portions of the wing are discontinued at the side of the fuselage.
THE FUSELAGE
439
Figure XXI-7. Longitudinal stringers may be omitted at the bow or nose section
of a fuselage since the curvature of the skin enhances the pressure-supporting ability of
the structure. Simple frames, such as shown above, are sufficient to maintain the
cylindrical shape. Similarly, in the tail section of the fuselage, the longitudinal stringers
may be omitted so that rigidity is obtained through the curved skin and the transverse
frames.
Figure XXI-8. Where transverse frames allow the longitudinal stringers by suitable
cut-outs, additional rigidity may be obtained by small brackets riveted to both the
stringer and the frame. A frame like the one above may support brackets or attachment
fittings for control systems, electrical conducts, and other light equipment.
440
FRAMES AND THEIR LOCATION
Just as for the ribs in the wing structure, the primary functions of the
transverse fuselage frames are (1) to maintain the shape of the fuselage,
(2) to sustain concentrated loads imposed, (3) to serve as attachments for
equipment, flooring, and the like, and (4) to transmit the loads to ad-
jacent structural members. The frames may therefore be roughly classi-
fied into three categories:
1. The frames which serve merely to maintain the shape of the fuselage.
These will not be subjected to stress unless distortion of the entire ad-
jacent structure has taken place. They are often just one gauge thinner
than the metal sheet covering, and the developed width of the frame,
formed out of sheet material, varies from 2 to 4 inches for aircraft weighing
from 2000 pounds to 50,000 pounds. It should be realized that in addi-
tion to the developed width, the other variable available is the thickness of
the sheet from which the frame is fashioned. For purposes of later
identification, these forms will be referred to as simple frames.
2. The frames which serve to act as anchorage for medium-weight
equipment, control systems, and the like. These are similar to the simple,
single-member, transverse frame, but must be reinforced locally to carry
the load and reduce deflections to a minimum. Such frames are subjected
to severe loads only locally and therefore may differ little from the simple
frames. Additional brackets may have to be introduced and tied in with

THE FUSELAGE
441
Figure XXI-10. Transverse frames located where additional rigidity is desired or
where the depth of the fuselage is comparatively large may add to the rigidity of the
fuselage either by the larger depth of the frame ring around its entire circumference or
by the added depth of the bottom half of the frame as shown in the sketch above.
Figure XXI-11. Certain frames may be modified to incorporate a beam for the
support of flooring or as a continuation of the wing spar. In the latter case, two frames
may be placed back to back, as the cross section indicates. If the flooring alone is to
be supported, one frame may suffice. Longitudinal stringers may pass through the
frame and be tied into the frame structure more securely by special brackets.
442
Figure XXI-12. In order to make the spanwise stringers of the wing effective, such
stringers would have to tie into the frame structure of the fuselage either through
separate fittings or through the aid of local reinforcements or gusset plates. The span-
wise stringers, furthermore, could be riveted or bolted to the longitudinal stringers of the
fuselage through intermediate gusset plates or fittings. Only the spars, not the stringers
of the wing structure, would pass through from one side of the fuselage to the other.
the longitudinal stringers as well as the frames. For purposes of later
identification, these will be referred to as intermediate frames.
3. The primary frames, to which large external loads are supplied
through the landing gear, power plant, or wing structure. These are
usually two in number, spaced a small distance apart, and designed so as
to take fittings to serve as carry-through members and, in short, act as
the main transverse load carrying members. For purposes of later
identification, these will be referred to as main frames.
Figures XXI-7, 8, 9, 10, 11, 12, 13, 14 show some sketches of typical
frames serving different functions.
Figure XXI-13. For a two-spar wing, the spars could pass through the fuselage and
be supported by two transverse frames, back to back. The spars may be at an angle
to the perpendicular to the wing chord if the inclination will permit the frames to clear
the windshield of the cockpit or cabin. The wing profile and the structure other than
the two spars stop at the side of the fuselage.

THE FUSELAGE
443
Figure XXI-14. The transverse frames of the fuselage are perpendicular to the
longitudinal axis of the airplane even though the spars are at an angle to that axis.
The frames, whether circular rings or reinforced frames, are located
from 10 to 24 inches apart with the closer spacing near the cabin and cock-
pit and gradually increasing towards the tail post.
The frames may rest on the longitudinal members or on the outside
covering with cutouts for the longitudinal members.
Reinforced frames are located wherever concentrated loads are applied
as at attachment of wing spars or landing-gear members. In order to
reduce structural weight, it is decidedly advisable to locate several external
attachment points on the same frame wherever possible.
In general, the locating of frames will be determined first by the location
of spars, landing gear, doors, and windows. Reinforced frames are re-
quired at every one of these locations of external attachments in order to
distribute the loads into the metal covering evenly. After these reinforced
frames have been placed, the intermediate frames can be placed at proper
intervals.
Reinforced frames may be similar to intermediate frames in construction
except for the thicker material or local reinforcements; or they may be two
frames, back to back, but spaced a few inches apart, with tying members
between them to obtain rigidity and continuity of structure.
The spacing of the simple and intermediate frames is dependent to some
extent on (1) tin-canning of the sheet covering, (2) the allowable shear
stresses of the covering, and (3) the allowable column compressive stress
of the longitudinal stringers and adjacent effective widths of the covering.
The discussion on selecting panel dimensions in an earlier chapter should
be referred to.
444
Figure XXI-15. Main frames, to which external loads and bending moments are
applied, have transverse shear, axial load, and bending moment varying in intensity all
along the centerline of the frame. The representation above illustrates the mathe-
matical approach to the solution of the stresses in the indeterminate frame. Because
these frames have to withstand severe stresses, their construction is somewhat different
from the non-load-carrying frames.
The main frames are subjected to a variety of concentrated loads and
require a stress analysis for the determination of dimensions. Figure
XXI-15 illustrates the stress-analysis problem that is involved. It will
be noted that the frame is subjected to shear, axial load, and bending
moments, making it necessary to provide an adequate structure for these
conditions.
LONGITUDINAL STRINGERS AND THEIR LOCATION
Since the fuselage is essentially a beam, the longitudinal stringers serve
an important function in that they, along with the effective width of sheet
covering, are the main bending elements of the structure.
For small aircraft of gross weights of about 2500 pounds or less and
fuselage depths of less than 48 inches, relatively few stringers are neces-
sary, since the minimum thickness of sheetâ€”0.025 inchesâ€”selected for
handling and maintenance reasons, usually contributes markedly to the
bending strength. The curvature of the sheet adds to the allowable
compressive strength of the sheet as well.
The cross sections of the longitudinal stringers are the same as for the
wing.
The stringer spacing is determined by the number required for the
loads imposed. An analytical approach is possible to determine the light-
est combination of stringer-sheet combination.

THE FUSELAGE
445
Figure XXI-16. Figure (A) shows a partially reinforced monocoque construction
where the skin contributes little to the strength. Figure (B) shows a pure monocoque
structure where the skin is made sufficiently thick to avoid local buckling.
Since the fuselage shape is not likely to be very regular, the spacing
from frame to frame is apt to vary. A pattern of stringer spacing when
projected upon the plane for symmetry may show parallel lines. Further
examination, however, may show that the longitudinal stringers have to
be twisted about their longitudinal axis to achieve such a pattern. It
might therefore be better to obtain a location at which the stringers would
Figure XXI-17. Figure (A) shows an increase in longitudinal stiffening members
so that the skin is amply supported and contributes materially to the fuselage strength.
The longitudinal members have been replaced by corrugated sheet, as shown in Figure
(B), to form a very rigid structure but causing difficulties in production.

446
have little twist but would follow a faired line. Another method would
be to use proportional distances for the spacing from frame to frameâ€”i.e.,
the larger spacing at the largest cross sectionâ€”and to reduce the spacing
for the smaller cross sections in proportion to the radius.
Since openings have to be reinforced along the edges, it is desirable to
have the top and bottom of all frames for such openings rest on longi-
tudinal stringers. It is customary to spot in these stringers first* in order
to avoid unnecessarily close spacings, should the desired stringer spacing
and the required stringer spacings not coincide.
Longitudinal members are spaced from 6 to 12 inches apart around the
periphery of the largest cross section. Since the cross sections gradually
diminish in size, the spacing is closer towards the tail post so that alternate
members may be stopped at a forward frame. It is desirable not to end
all longitudinal members at the same frame.
To assure a continuous structure, it is customary to make the longitu-
dinal members continuous rather than breaking them at every frame.
Since the nose of the fuselage is not heavily subjected to bending mo-
ments, some of the longitudinal stringers may be discontinued in that
section by ending a few stringers at each frame starting at a frame for-
ward of the front spar.
The same procedure would be in order in the rear portion of the fuselage,
where the bending moments are less severe and where the available circum-
ferential space is likely to be very limited.
In general, since the stringers are the main bending elements, they
should be continuous and therefore pass through the transverse frames.
Where local stresses are likely to be high, brackets tying the frames and
stringers together are added for greater rigidity and load-carrying capa-
bilities.
FUSELAGE SKIN
The gauge of metal sheet used for the fuselage will vary according to
the region of application. The normal pressures likely to be encountered
are not apt to be too critical. Where the structure is more or less normal
to the line of flight, such as the windshield and fuselage nose, the posi-
tive pressures, even at 800 miles an hour, may be of the order of 350 to
600 pounds per square inch; whereas curved surfaces along the top
bottom, or sides of the fuselage may have normal negative pressures vary-
ing from 60 to 300 pounds per square inch. At lower speeds, the pres-
sures may be assumed to vary as the square of the speed. Obviously,
in the region of high pressures, the skin thickness will have to be relatively
greater in order to avoid tin-canning of the surface.

THE FUSELAGE
447
In the region of propeller-tip passage, as in multi-engine designs, the
skin thickness is often increased to reduce the noise caused by drumming
and to avoid fatigue failures resulting from the induced vibration.
In areas where rough handling is likely to occur (e.g., around doors and
cargo compartments), again, a heavier gauge is probably necessary.
The smallest gauge of aluminum alloy sheet or aluminum-coated
aluminum alloy sheet is 0.025 inches, since a thinner sheet is easily punc-
tured and is rather hard to handle. For airplanes of less than 10,000
pounds gross weight, the predominant gauge thicknesses are 0.025 and
0.032 inches. For heavier aircraft, the thicknesses are proportionately
greater.
In regions such as jet exhausts, where excessive heat is likely to be en-
countered, nickel steel or titanium alloys replace the aluminum alloy sheet.
WINDOWS
Since windows are not considered part of the primary structure, they
are not designed to serve as carry-through members for the transmission
of stresses in the primary structure. Therefore they are carried by a
rigid frame, which must bear the stresses around the cut-outs caused
by the windows.
The normal pressures are likely to be high for the material used, espe-
cially for pressurized cabins. A differential pressure of 8.15 pounds per
square inch at an altitude of 20,000 feet is a common design value.
For generous vision, the proportions of the window should be reason-
ably large. However, since windows have been known to fail in flight,
there is a tendency to make the openings smaller. In this way, in case
of failure resulting in explosive decompression, a passenger seated next
to the window would not be blown out.
For additional safety, window materials are shatterproof and windows
for pressurized cabins are three layers thick, with each layer constructed
of two or more plies.
The student should also refer to Chapter XII, The Passenger Cabin.
PRESSURIZED CABINS
To avoid the use of individual oxygen supplies, the passenger cabins of
today's transport airplanes are pressurized. To simplify the control of
pressurization, to avoid structural interferences, and to reduce leaks, it is

common to provide a separate structure for the pressurized cabin.
For optimum weight, it is desirable to construct the pressurized cabin

as a cylinder with hemispherical ends. The stresses imposed by the

448
differential pressure can then be calculated by means of the standard
formula
Jt 2t
The pressurized cabin floats inside the fuselage on flexible mounts so
that the stresses and distortions of the fuselage are not transmitted to the
cabin.
Structurally, the pressurized cabin is similar to the fuselage in that
frames and stringers are used.
Since the cabin is not subjected to the airstream, stringers and other
structural members may be used either on the inside or the outside of the
skin or metal envelope.
To make cabins pressure-tight, special attention must be paid to all
seams, openings such as windows and doors, and control-system operation
through nominally airtight openings.
Skin seams of the main structure may be sealed effectively by inserting
flannel tape impregnated with a sealing compound (much the same as for
water-tightening a metal float or hull) between the seam laps prior to rivet-
ing. In addition, the rivets are spaced from 4 to 6 diameters apart to form
a double row of staggered rivets.
Soft rubber gaskets, so arranged that they are pressed together when
subject to internal pressure, are used to seal all doors and hatches. Win-
dows are sealed in rubber channels. Sliding windows are best mounted
in steel frames seated against rubber seals.
Control cables, passing through the pressurized section, should pass
through a special gland which reduces air leakage to a minimum.
A spherical shape is ideal for a structure designed to withstand internal
pressure most efficiently. In the case of the fuselage, a spherical shape
has to be modified. A cylinder with partly spherical ends is the best com-
promise.
Unless the airplane is very large, so that the weight involved becomes
negligible, the pressure cabin forms an integral part of the fuselage.
Where weight considerations are not so vital, the pressure cabin is a sepa-
rate compartment "floating" inside the fuselage. Airtight openings for
control systems and the like are not a serious problem in such cases.
Cabin windows should be double with a dead air space between, or pro-
vision may be made to heat and circulate the heated air at high altitudes,
because the moisture in the air within the cabin has a tendency to frost
the windows.
THE FUSELAGE
449
FLOORING
A general analytical approach to the design of flooring is given in the
chapter on Detail Design Considerations.
Flooring may float: it may not be structurally tied to frames and string-
ers. In this way, loads on the fuselage structure are not transmitted to
the flooring. However, the loads on the flooring may be transmitted to
the rest of the structure through flexible spring or rubber supports. In
other words the floating floor may be shock-mounted. The primary
advantages of such a type of support is that vibration is not transmitted
to the flooring. Since the flooring is not an integral part of the main
structure, it may be altered without worrying about the structural in-
tegrity of the primary structure.
Flooring may be tied in with the primary structure and so become part
of it. The advantage is likely to be a saving of weight since the flooring
would serve a dual purpose. Whether or not the flooring is tied into the
main structure will to some extent determine its detailed construction.
The flooring would extend only over that section of the fuselage where
the presence of passengers, crew, and cargo make it necessary. The
floors in the various compartments need not all be placed at the same
level, nor necessarily at the same angle.
Since the flooring will transmit loads to the fuselage structure, it should
rest on longitudinal stringers, which may have to be larger than those
that are not so loaded. Usually the transverse frames brought up to the
floor level will help to support the flooring as well.
Reference should be made to Chapter XII, The Passenger Cabin, which
contains some additional discussion on this subject.
DOORS AND EXITS
The number of doors and exits included in an airplane will depend upon
the number of persons it must carry. It is desirable to have separate
exits for the crew and for the passengers, although the main entrance to
the cockpit may be through the passenger cabin. (See Chapter XII.)
The closures for doors and exits are usually rigid in construction so as
to avoid possible distortions that would lead to interference in opening
the doors; yet they are seldom designed to act as part of the primary
structure of the fuselage unless the locking device can be depended upon
to provide a continuous type structure.
Access Doors
Where access from the outside is required, readily removable access
doors, which usually employ some form of cowl fastener for quick opening,
should be provided in the fuselage design.

450
Batteries should be removable from the outside and not from the inside
of the cabin or cockpit; likewise, the water tank should be finable from
the outside. Since these two items are not removable and access is not
required at every landing, the access doors do not have to be as easily
opened as the doors to the baggage compartments.
Baggage Compartments
Baggage compartments and the like are lined with a secondary structure
of metal sheet, either flat or corrugated (may be covered with canvas or
felt for soundproofing purposes). This secondary structure is attached
to the frames or longitudinal members at appropriate points.
To avoid the added complexity and higher costs involved in the design
of a larger fuselage, the use of twin fuselage designs has been proposed.
This suggestion may be worth while provided that different functions
are assigned to each fuselage. It could also prove effective in cases where
the heavy concentrated load of the larger fuselage would hinder the ful-
fillment of center of gravity requirements. It must be recognized, how-
ever, that the maneuverability of such a craft might be curtailed by the
increased moment of inertia resulting from the dispersion of large units.
The design considerations for the tail booms are very similar to those
for the aft section of a fuselage. The boom is subjected primarily to
TWIN FUSELAGES
TAIL BOOMS
Figure XXI-18. A twin fuselage airplane.

THE FUSELAGE
451
bending and torsional moments, with very little transverse shear. For
this reason, a cross section that is fundamentally circular in form is the
most efficient one.
For the nose-wheel type of landing gear, the tail boom or booms would
seem to have considerable merit for the small airplane design.
For a pusher-propeller installation, twin tail boomsâ€”one on each side
of the engineâ€”serve as protection from the propeller.
There is some weight penalty involved with twin tail booms. On the
basis of skin alone, it can be seen that 2 ir di > v d, where di is the di-
ameter of each of the twin booms, and d is the diameter required for a
single boom. The single-boom diameter would be a little larger than the
twin-boom diameter.
Figure XXI-19. A centerline sketch of a truss-type fuselage for an in-line engine in
the nose. Only the top and front side trusses are shown. The Pratt truss is shown for
the top and rear portion of the fuselage, and the Warren truss for the front side. The
side trusses may be either Pratt or Warren, or a combination of the two, while the top
and bottom trusses are always Pratt type since they tie in the side trusses, which are
usually symmetrical with each other.
ALTERNATE TYPE STRUCTURES
The use of truss-type structures, especially for small airplanes, should
not be ruled out entirely. In some cases, a desirable solution is found in
composite structure utilizing a reinforced monocoque structure for the
cockpit and cabin section and a steel truss structure for the tail section,
or vice versa.
Some of the truss arrangements are shown in Figures XXI-19 and
XXI-20. The two truss types employed are (1) the Warren truss, in
which three adjacent members in the side truss are placed to resemble a
letter W, and (2) the Howe truss, in which every two members in the side
truss are parallel to each other.

Figure XXI-20. Possible arrangements for structural members around a door for a
truss-type fuselage.
In general, the truss structures are easily produced. The stress analysis
of such a structure is also relatively uncomplicated. No prototype struc-
ture has to be tested statically, as in a reinforced monocoque structure,
since the structural analysis is well authenticated.
In order to provide a streamlined fuselage, a fairing structure is built
around the trusses. Although this satisfies the aerodynamic requirements,
the structural weight may be high since separate structures are usedâ€”one
for primary purposes to carry the load, the other for secondary purposes
only; whereas in the reinforced monocoque structure, the two purposes or
functions are combined in one.
Another design approach is to use one or more keel beams to serve the
needs of the primary structure. The remainder of the structure, consist-
ing of transverse frames, stringers, and covering, would serve a secondary
role.
Such a structure would be particularly useful in situations in which
local concentrated loads were extremely high, such as carrying heavy
machinery or driving tractors or tanks into the cargo compartment.
The structure has been proposed for small airplanes in which the seats
and control systems would be attached to the longitudinal fuselage beam.
The structural advantages of using one or more beams of this kind are (1)
that they are easily stress-analyzed, (2) they are easy to manufacture, and
(3) their deflections and distortions are reasonably easy to predict.

THE FUSELAGE
453
EFFECT OF FUSELAGE MASS DISTRIBUTION
When the mass is distributed primarily in the fuselage of the airplane,
the moment of inertia about the normal axis of the airplane is increased.
It was found that in extreme cases, as long as the principal or longitudinal
axis was kept inclined at a positive angle of attack to the flight path, the
oscillatory stability was improved, but at zero angle of attack, there was
a decrease in this stability.

CHAPTER XXII
Preliminary Performance Calculations
Performance calculations are necessary at several stages of the design to
determine whether or not the original specifications set forth for the design
are being met. The first set of preliminary performance calculations is
based upon empirical formulae that have been established on the basis of
past performance of similar airplanes and which are predicated upon simi-
lar power and wing loadings, gross weights, and general design. These
calculations are wholly rule-of-thumb and are used by the engineer for his
general information and to decide whether his design is likely to meet the
performance specifications at all.
SOURCE OF DATA
The second form of preliminary performance calculations is set forth
briefly in this chapter. The data for determining the horsepower required
to fly in horizontal flight may be obtained in several ways:
1. From scattered wind-tunnel tests made by various testing agencies.
2. From wind-tunnel tests on individual components of the design.
3. From wind-tunnel tests on the complete design.
4. From wind-tunnel tests and full flight corrections obtained on similar
previous designs.
These calculations can be considered only as preliminary since their
accuracy depends upon the judgment of the engineer using the particular
data. Performance data must be obtained finally from actual full flight
tests.
Preliminary performance estimates for minimum or stalling speed,
maximum speed, maximum rate of climb, service and absolute ceiling,
and range may be obtained by a series of calculations for horsepower re-
quired at various speeds and the corresponding horsepower available.
These calculations are given here in briefest form, and their accuracy de-
pends upon the accuracy of the various assumptions made. However,
they may be considered sufficiently accurate until extensive wind-tunnel
tests on the design have been made.
454
PERFORMANCE CALCULATIONS
455
CALCULATIONS FOR HORSEPOWER REQUIRED
The horsepower required at different speeds in horizontal flight is calcu-
lated by means of the equation:
Pr = (D. + D,) X,
where Dâ€ž = wing drag in pounds at speed V,
Dp = sum of parasite and interference drag in pounds at speed V
and
V = speed in feet per second of the airplane in horizontal flight
at the corresponding angle of attack.
The wing-drag coefficients are usually obtainable for an infinite aspect
ratio and must be corrected for the aspect ratio used. These corrections
are outlined in Chapter III, Airfoil Selection.
The parasite resistance includes all the resistance caused by the fuselage,
tail surfaces, surface irregularities, landing gear, supports and the like, and
all parts of the airplane except the wing proper.
Parasite-resistance coefficients are obtainable from various sources and
are often referred to unit length as in the case of struts; or, per unit as in
the case of wheels; or, per square foot of maximum cross section in the case
of the fuselage. It is usually desirable to obtain the total drag for a speed
of 1 foot per second and then obtain a new coefficient referred to the wing
area by dividing by the wing area.
The interference drag is more difficult to appraise. Usually it is ac-
counted for in the calculations by arbitrarily increasing the parasite re-
sistance by a small percentage. An airplane with a fuselage well faired
into the wing, with landing gear retracted, offers far less interference drag
than one not so well faired. The arbitrary factor may be as much as 5
per cent.
These calculations are best carried out by means of a table, such as
Table XXII-1.
Table XXII-1.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
Dâ€¢
Dl,
Dvts
Dbts
A>
D, X 1.05
Com
Cd,
CL
HP
456
Computations Derived from Table XXII-1.
Column 1. Here are entered the angles of attack of the airplane or the
wing, whichever is chosen for reference. The angular range should start
slightly above the angle at which zero lift of the wing occurs and should
stop at a few angles beyond that at which maximum lift of the wing occurs.
It is usually satisfactory to use 2-degree intervals.
Column 2. Record here the total drag, at 1 foot per second, of the fuse-
lage. The fuselage drag usually varies with angle of attack. If this
change is considered small, it may be neglected and thus be assumed con-
stant for all angles of attack.
Column S. The total drag of the landing gear, at 1 foot per second, of
the wheels, struts, and cowling. This drag usually remains constant for
all angles of attack.
Column 40 0 The vertical tail surfaces do not change their angle of attack
as the airplane changes its angle of attack; therefore, the drag remains
constant for all angles of attack. Any blanketing effect of the fuselage on
the vertical tail surfaces is neglected. The drag of the vertical tail sur-
faces at 1 foot per second is therefore constant.
Column 6. The drag of the horizontal tail surfaces changes with angle
of attack so that the drag at 1 foot per second varies with increasing angle
of attack of the airplane. The drag coefficients of the horizontal tail sur-
faces must be corrected for aspect ratio. These corrections are outlined
in Chapter III. Usually no account is taken of difference in the drag of
the horizontal tail surfaces due to downwash of wing.
Column 6. If columns 2, 3, 4, and 5 account for all items (outside of
the wing) contributing to the parasite drag, then the addition of these col-
umns for each angle of attack will give the total parasite resistance, desig-
nated Dp, at 1 foot per second.
Column 7. As previously explained, the interference drag may be ac-
counted for by multiplying the parasite resistance by a factor varying
from 1.00 to 1.05. This may be called a "bogey factor" by which any
omissions in the calculation of parasite resistance may readily be accounted
for if so desired.
Column 8. The value of Cdâ€ž the parasite and interference resistance co-
efficient in the absolute system referred to the wing area, may be found by
dividing the values found in column 6 by Yi p Su v2 where Sw represents
the wing area; since v = 1 foot per second (the speed for which calculations
were made, or if desired, calculations may be made for 100 feet per second)
the expression becomes p Sa (or 5000 pSâ€ž).
Column 9. Here are recorded the drag coefficients of the wing corrected
for aspect ratio, of the airfoil used in the design corresponding to the angles
of attack listed in column 1.

Table XXII-2. Standard atmospheres.
Altitude
(ft)
<(Â°F)
p/po
Ob/ftÂ«)
gp
p/po
(in.)
59.
0.002378
1.
0.07651
29.92
1.
1,000
55.434
0.002309
0.9710
0.07430
28.86
0.9644
2,000
51.868
0.002242
0.9428
0.07213
27.82
0.9298
3,000
48.301
0.002176
0.9151
0.07001
26.81
0.8962
4,000
44.735
0.002112
0.8881
0.06794
25.84
0.8636
5,000
41.169
0.002049
0.8616
0.06592
24.89
0.8320
6,000
37.603
0.001988
0.8358
0.06395
23.98
0.8013
7,000
34.037
0.001928
0.8106
0.06202
23.09
0.7716
8,000
30.471
0.001869
459
where D and corresponding v are found in columns 13 and 12, respectively.
This procedure comprises the calculations for horsepower required. It
is customary to plot the values of horsepower versus corresponding speed
(usually given in miles per hour); then, with the aid of horsepower avail-
able curves, almost all the performance figures required may be calculated
readily. Refer to Figure XXII-1.
The calculations outlined are usually calculated for sea-level conditions.
They may be extended to include calculations for other altitudes by mul-
tiplying the square roots of the relative densities by both the velocity and
horsepower found at another altitude as follows:
where vi represents the new speed at the density pi at the new altitude,
vo represents the originally calculated speed at the density po,
hpi represents the new horsepower corresponding to Vi, and
hpo represents the originally calculated horsepower corresponding
The density relationships of any altitude to that of sea level may be
found in Table XXII-2.
The table for standard atmosphere, as normally presented, gives the
density relationships of air at any altitude at 45 degrees latitude. This
table does not usually extend beyond the stratosphere, or above 50,000
feet. Additional data has been made available for use in extreme high-
altitude considerations. The information presented in Tables XXII-3
and XXII-4 covers upper atmosphere characteristics under the most
probable conditions and the arbitrary standard atmosphere, and may be
of interest even though it may not be of immediate use to the student.
Aside from performance calculations, a number of other problems are
introduced by increase in altitude. Under special conditions, people can
live in altitudes up to 18,000 feet without the need of oxygen, but oxygen
has to be supplied beyond that altitude. Pressure has to be supplied, as
well, for the human system to absorb the required oxygen, but if no other
difficulties arose, it would be found that the human blood would boil at its
own temperature at 63,000 feet altitude. One hundred per cent oxygen
would be required from altitudes above 40,000 feet.
ALTITUDE CORRECTIONS
tO V0.
ARBITRARY STANDARD ATMOSPHERE

Table XXII-3. Upper atmosphere characteristics.*
Altitude
(ft)
Altitude
(miles)
Weight of air
(lb/cu ft)
Mean free path
between mole-
cules (in.)
Molecules
(per in.*)
500,000
94.6
370
43 X 10"
450,000
85.2
â€”
65
25 X 10"
400,000
75.8
â€”
12
15 X 10"
350,000
66.3
0.2 X 10-Â»
1.9
84 X 10"
300,000
56.8
2.6 X 10â€”7
0.3
48 X 10"
250,000
47.4
t2.9 X 10-*
5.8 X 10-*
27 X 10"
200,000
37.9
1.6 X 10-Â»
1.0 x io-Â«
16 X 10"
150,000
28.4
9.5 X 10-*
1.7 X 10-'
96 X 10"
100,000
18.9
0.000107
2.7 X 10â€”Â«
62 X 10"
50,000
9.5
0.0116
3.6 X 10-*
47 X 1018
0.0
0.0765
3.5 X 10-*
44 X 10"
* From "Developments in High Speed Aircraft," by E. H. Heinemann, Mechanical
Engineering, October 1947.
f This is another way of saying 2.9 pounds per million cubic foot.
Table XXII-4. Arbitrary standard atmosphere.*

461
Item
Vertical tail surface cantilever per square foot. If the minimum drag
coefficient of the vertical tail surfaces is known, this value may be
used with an increase of about 25 per cent to account for gap at hinges.
If the surfaces are externally braced, increase the drag by about 50
per cent.
Horizontal tail surfaces, cantilever, use drag coefficients, corrected for
aspect ratio, for each angle of attack. Add about 25 per cent for
gap at hinges. If horizontal tail surfaces are externally braced, in-
crease the drag coefficients by about 50 per cent. If the external
bracings are struts of streamlined tubing, allow additional foot for
each end fitting, and estimate drag of bracing on basis of projected
length of these struts when thrust line is horizontal.
Fuselage plus cowled engine per square foot of projected maximum frontal
area. If the fineness ratio (ratio of length to depth) is rather small,
use high figure. The variation of fuselage drag (for approximately
elliptical cross section) may be obtained by means of the formula
Da = D0 + 0.0015 Do a1
where Do = value given (for angle of attack a = 0 degrees).
THE ENGINE
The horsepower available at different speeds of the airplane in horizon-
tal flight is determined from the brake horsepower curves of the engine
used for the design, and the propeller characteristics.
The brake horsepower, usually plotted against corresponding revolu-
tions per minute of the engine is generally furnished by the manufacturer.
This power curve is given for sea-level conditions unless the engine happens
to be supercharged when the power curve corresponds to the rated altitude
limit of the supercharger. See Figure XVIII-6.
The brake horsepower of nonsupercharged engines and of supercharged
engines above the rated altitude limit of the supercharger varies with the
altitude. This variation may be determined easily from the following
formula:
P = Pâ€ž (p/po)1 \
where P = brake horsepower at altitude considered,
Po = brake horsepower at sea level or at limiting supercharged alti-

tude,
p = density at altitude under consideration,

po = density at sea level or at limiting supercharged altitude.
The following should be borne in mind also when the engine horsepower
is desired for other than the usual standard for the engine in question.
1. Propeller gearing will reduce the engine brake horsepower 2 or 3 per
cent.
2. For engines equipped with turbo supercharger, the power decreases
Full scale drag
in pounds at
100 mph
0.40
6-8.5
to about 80 per cent of the normal sea-level power of the unsupercharged
engine.
3. For direct-drive superchargers, the power at the limiting altitude of
the supercharger is the same as the normal sea-level power developed by
the unsupercharged engine.
A typical power curve for an engine is shown in Figure XXII-2.
900
I 800
* 700
Â«s
5 600
500
1400 1600 1800 2000 2200
Revolutions per minute
Figure XXII-2. Variation of brake horsepower of the engine with the propeller
shaft speed in revolutions per minute.
Horsepower Available
The modern constant-speed propellers are so designed as to keep the
same revolutions per minute regardless of the forward speed of the air-
plane, at least within the permissable angular range of the propeller blades.
It is easy, therefore, to determine the horsepower available by calculating
the value of C, and determining the corresponding efficiency and blade
angle. It is desirable to determine the blade angle also, since the con-
stant-speed propellers are usually limited to 8 to 10 degrees in angular
range. The method of determining the horsepower available may be
illustrated best by the results of calculations presented in Table XXII-5.

463
Table XXII-5. Determining horsepower available. Design speed 224 mph; pro-
peller diameter 9.75, 3 blades; engine 850 horsepower at 2450 rpm; sea level = 0.002378;
gear ratio 2:3.
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
Propeller
Brake
HP
(mph)
V*/nD
C.
0*
n*
li
(rpm)
hp
available
224
1.11
1.92
29.5Â°
86
83
1,634
850
706
200
1.025
1.72
28.5
85
82
1,634
850
697
180
0.922
1.55
27.0
83
80
1,634
850
680
160
0.821
1.375
26.5
81
78
1,634
850
664
140
0.717
1.202
25.5
78
75
1,634
464
40 50 60 70 80 90 100
Per cent of design airplane speed
Figure XXII-4. The variation of engine power with change in aircraft speed when
a fixed-pitch propeller is used.
Column 40 0 The blade angles at 75 per cent of the radius are determined
for the values of C, given in column 3. See Chapter XV, The Propeller.
Column 5. The propeller efficiency and blade angle corresponding to
the calculated values of C, are determined with the aid of curves such as
those to be found in Chapter XV.
Column 6. Since the efficiencies tabulated in column 5 are for 2-bladed
propellers, a subtraction of 3 per cent is made from those values to give the
efficiencies for a 3-bladed propeller. See column 6.
Columns 7 and 8. Since the controllable-pitch propeller is a constant-
speed propeller, the engine brake horsepower and corresponding revolu-
tions per minute remain constant throughout the angular range.
Column 9. Values of brake horsepower given in column 8, multiplied by
the corresponding propeller efficiencies given in column 6, give the horse-
power available at full throttle at the speed listed in column 1. These
values are plotted in Figure XXII-3.
In case of a fixed-pitch propeller, the calculations are slightly different
since the engine revolutions per minute and power do not remain the same.
Refer to Figure XXII-4. Suppose the controllable-pitch propeller used
could not change its pitch angle below 20.0 degrees so that the propeller
would be, in reality, a fixed-pitch propeller at speeds below 140 miles per
hour. The calculations are set forth in Table XXII-6.

PERFORMANCE CALCULATIONS 465
Table XXII-6.
(1)
(2)
Per
cent
of
(3)
Per
cent
(4)
(5)
(6)
Pro-
(7)
(8)
(9)
(10)
(11)
HP
of
design
rated
Brake
RPM
peller
avail-
speed
hp
hp
rpm
C.
1*
ii
able
140
100
100
850
2,450
1,634
1.202
25.5
78
75
638
120
85.7
94.5
803
2,300
2,185
1,535
1.071
25.5
74
71
570
100
71.4
90.0
765
1,456
0.919
25.5
72
69
528
80
466
by the corresponding efficiencies 1/1 in column 10 will give the values of
horsepower available in column 11.
The calculations for other altitudes are similar to those given for sea-
level conditions. The horsepower-available values for the critical alti-
tude of the supercharged engine are tabulated in Table XXII-7.
The upper limit of the controllable-pitch propeller of 29.5 degrees has
been determined by the design sea-level condition and therefore remains
the same.
Table XXII-7. Horsepowerâ€”available values for critical altitude. Design speed
238 mph; altitude 7000p = 0.001928; engine 850 horsepower at 2450 rpm; propeller
diameter 9.75 feet, 3 blades; gear ratio 2:3.
Propeller
Brake
Available
(mph)
C.
VÂ»/nD
V*
Â»?i
rpm
hp
hp
238
1.982
1.22
29.5
86
83
1,634
850
705
220
1.815
1.13
29.0
85
82
1,634
850
697
200
1.65
1.025
28.5
84
81
1,634
850
687
180
1.485
0.922
28.0
83
80
1,634
850
680
160
1.321
0.821
27.5
82
79
1,634
850
672
467
55700
(a
v.
*>
^ 80
v.
<i
Â£ 60
< 40
*.
Critical
20,0
altitude
00 ft/
OO ft.'
level'
- /Â£,0<
sea
0 4 8 12 16
Altitude above critical altitude of
supercharged engine (thousand ft.)
Figure XXII-5. Variation of brake horsepower of a normal supercharged engine
above the critical altitude.
800\
5:
200\
HP i
1O,0(
ivailable
OO ft. a
at
1titude
Ma*
HP
: excess
385
i|
>
HP ret
at 1O,t
juired
700 ft.
? 40 80 120 1t
iO 200 2*
to
Speed, miles per hour
Figure XXII-6. Available and required horsepower curves obtained for an altitude
of 10,000 feet for the supercharged engine whose critical altitude was 7000 feet.
468
airplane. In Figure XXII-1, the maximum speed at sea level is 224 miles
per hour. Figure XXII-3 indicates that the maximum speed at 7000
feet is 238 miles per hour, and Figure XXII-6 indicates that the maximum
speed at 10,000 feet is 237 miles per hour.
RATE OF CLIMB
The rate of climb is calculated by means of the formula
Â»,.,.-. . Â± 33,000 (excess horsepower)
rate of climb in feet per minute = â€”1 â€”. , â€¢ . câ€”-,â€”â€”
gross weight in pounds
The maximum difference between the curves of the horsepower available
and the horsepower required at the same speed along the flight path will
determine the maximum rate of climb at the particular altitude for which
the calculations are made.
In Figure XXII-1, the maximum excess power is 435 horsepower; the
rate of climb is then
435 X 33,000 , , .
^ â€” = 1436 feet per minute
at sea level. Similarly, the rate of climb at 7000 feet is 1419 feet per min-
ute, and at 10,000 feet the maximum rate of climb is 1270 feet per minute.
Refer to Figure XXII-7.
ABSOLUTE AND SERVICE CEILINGS
The absolute ceiling is that altitude at which the rate of climb Ja zero
feet per minute. The service ceiling is that altitude at which the rate of
Islimb is 100 feet per minute^ ~~~~ . â€”
The absolute and service ceilings may be obtained graphically by deter-
mining the rates of climb for two different altitudes (outside of the super-
charged region and extending the straight line drawn through these points
to intersect the zero ordinate).
This has been done in Figure XXII-7. The service and absolute
ceilings, of course, also could have been calculated by means of trig-
onometric relationships for two different altitudes. Actually, the rate of
climb does not vary directly with the change in altitude but the assump-
tion is sufficiently close for all practical purposes.
RANGE
The range may be calculated on the basis of the known horsepower, fuel
consumption, and speed. This method is likely to give too conservative

469
40
I32
24
16
bsolute
?rvice c
ceiling
eiling
Si
rv
1\
â€” TOOL
7 feel
U-l 1
400 800 1200 1600 2000
Rate of climb (feetper minj
Figure XXII-7. A typical rate-of-climb curve for an airplane equipped with a
supercharged engine.
values although more refined formulae are available. Breguet's formula
is also used:
R = range in miles = 863 (J^J ^ logw (^r^)'
where L/D = maximum over-all effectiveness of the complete airplane,

rj = average propeller efficiency at cruising,
c = average fuel consumption in pounds per brake horsepower

per hour at cruising,
Wo = gross weight of airplane at start of flight, and
W, = Wo less oil and fuel weight.
Extensive flight tests are made on the prototype airplane to establish
the necessary data on the basis of which the usefulness of the airplane for
certain operations can be evaluated. The outline presented here gives an
indication of the type of data desired.

470
PERFORMANCE REQUIREMENTS
With fixed equipment as listed in the following text, and a useful load
consisting of
1 pilot 170 lb
5 passengers 850 lb
baggage 200 lb
cargo 200 lb
fuel and oil sufficient for a cruising
range of 500 miles at any altitude
between sea level and 5000 feet,
the minimum performance acceptable when using fuel of not more than
80 octane will be:
1. * High speed in level flight 175 mph
2. f Landing speed with power off 65 mph
3. Ceiling (one engine dead) 6000 ft
4. Distance from start to clear 50-foot obstacle 1500 ft
5. Distance to stop after clearing 50-foot obstacle 1000 ft
* At any altitude from sea level to 5000 feet altitude (standard air),
t Standard air.
Take-off performance with all engines functioning normally.
1. Ground run to minimum take-off speed.
Weight
Min take-off
speed (mph)
Distance
(ft)
Engine
manifold
pressure (in.)
Propeller
rpm
Max gross
Light
2. Horizontal distance required for acceleration from standing start to
minimum speed required for level flight with one engine dead.
Weight
Min
dead engine
speed (mph)
Distance
(ft)
Engine
manifold
pressure (in.)
Propeller
rpm
Max gross
Light
471
3. Horizontal distance required for acceleration from standing start to
speed of best angle of climb.
Engine
manifold
pressure (in.)
Weight
Best climbing
speed (mph)
Distance
(ft)
Propeller
rpm
Light
4. Best angle of climb.
Angle
(degrees)
Engine manifold
pressure (in.)
Propeller
rpm
Weight
Performance in air with all engines functioning normally.
1. High speed in level flight (at maximum gross weight).
Altitude
(ft)
â€¢ High speed
(mph)
Engine manifold
pressure (in.)
Propeller
rpm
Sea level
5,000 ft
Critical altitude of engine
2. Service ceiling (where best climb is 100 feet per minute).

Engine manifold
pressure (in.)
Propeller
rpm
Weight
Ceiling (ft)
Max. gross at start
Light
472
3. Cruising speed (approximately 70 per cent rated power, dependent
upon operating instructions, maximum gross weight only).
Altitude (ft)
Cruising speed
(mph)
Engine manifold
pressure (in.)
Propeller
rpm
Sea level
Critical altitude of engine
4. Fuel and oil consumption at cruising speed, maximum gross weight.
Fuel consumption (gals per hr) Oil consumption (gals per hr)
5. Minimum level flight speed.
Speed with flaps
retracted (mph)
Speed with flaps
extended (mph)
Engine manifold
pressure (in.)
Propeller
rpm
Performance in air with one engine dead.
1. Service ceiling (where best climb is 100 feet per minute).
Engine manifold
pressure (in.)
Propeller
rpm
Weight
Ceiling (ft)
Max. gross
Light
PERFORMANCE CALCULATIONS 473
2. High speed in level flight with engines at maximum power authorized
for continuous operation, maximum gross weight only.
Engine manifold
pressure (in.)
Propeller
rpm
Altitude
Speed (mph)
(a) Sea level
(b) Engine ceiling....
3. Fuel and oil consumption for conditions 2a and 2b.
Fuel consumption (gals per hr) Oil consumption (gals per hr)
4. Best angle of climb.
Angle
(degrees)
Engine manifold
pressure (in.)
Propeller
rpm
Weight
Max cross
light
Landing.
1. Steepest gliding angle practicable for landing approach, flaps ex-
tended, maximum gross weight, no power.
Angle (degrees)
Speed (mph)
474
2. Length of ground roll from 3-point landing, maximum gross weight
only, brakes used, landing for conditions as in 1.
Flap position Distance (ft) Landing speed (mpb)
EMPIRICAL FORMULAS
It is sometimes desirable to calculate performance data from a few basic
assumptions, such as wing loading and power or thrust loading, in order to
determine whether a set of proposed performance requirements is realistic
and can be reasonably met.
The formulas listed below have been based upon published data and
can be used to determine reasonable values.
Maximum Speed
1. Maximum speed, in miles per hour, in level flight at critical altitude
of power plant for power loadings (W/P) between 5 and 12 and wing
loadings (W/S) between 20 and 90 may be found from the formula:
'â€”Â» + *(?)(?).
This equation holds for propeller-equipped aircraft for speeds between
200 and 400 mph.
2. For aircraft powered with jet engines, with speeds of between 450
and 700 miles an hour, the following formulas have reasonable validity:
V^, - 450 + 1.28 (^r)(y): (a)
Vâ€” = 780 - 1.28 (b)
For the upper values of speed, the following formula gives closer values:
Vmax = 820 - 1.28 (jf)0 0 (c)
The available data for jet-powered airplanes are still too sparse to en-
able one to determine reliable values for the constants in these empirical
equations. Therefore, any such formulas should be checked with existing
data and the constants should be re-evaluated.

475
3. For aircraft powered with reciprocating engines, the following
formula may be used:
where K has a value from 135 to 145 for small personal aircraft employing
fixed-pitch propellers;
K has a value from 125 to 135 for small flying boats and amphibians;
K has a value from 200 to 225 for transport airplanes employing
constant-speed propellers.
4. Another formula that may be usefully employed in determining
maximum speed reads as follows:
for which K may be evaluated from existing values of Vmaz, Vm,â€ž, and
power loading, (W/P).
5. When enough data become available for specific categories of air-
planes equipped with jet power, an empirical formula such as the following
can be used:
Minimum or Stalling Speed
The following formula may be used with good results to determine the
minimum or stalling speed in miles per hour:
where K varies from 15 to 18 for small airplanes without the use of lift-
increase devices and with fixed-pitch propellers, for wing loadings (W/S),
between 5 and 25,
K varies from 10 to 13 for transport airplanes equipped with partial
span flaps and constant-speed propellers for W/S values from 30 to 80.
Rate of Climb at Sea Level
The rate of climb, in feet per minute, may be calculated from the fol-
lowing empirical formulas:
R = 18,000 ~ - 45 ^-
476
for personal airplanes equipped with fixed-pitch propellers, and power
loadings (W/P) varying from 15 to 25;
PW
Rc = 18,000 w- 10 -^
for aircraft using constant-speed propellers with wing loadings (W/8)
from 30 to 80, and power loadings (W/P) from 5 to 15.
Absolute Ceiling
1. The theoretical formula for reciprocating-engine-powered airplanes is
H = 40,000 logio (jr)<
where Pao is the horsepower available at sea level, and
Prâ€ž is the horsepower required to fly at speed V at sea level.
2. From the formula above, an empirical formula such as the following
may be derived:
H = 40,000 log,o ,^v* y
\p)\s)
where K may have a value from 105 to 120 for
â€¢ jw
values ranging from 45 to 55 for personal airplanes with unsupercharged
engines.
3. For supercharged reciprocating engines, the formula takes the form
H = Hc + 40,000 logio
where Hc is the critical altitude to which the engine is supercharged, and
K has a value between 215 and 230, for values for W/PVW/S from 40
to 90.
4. Turbo-jet-powered aircraft have a practical ceiling between 50,000
and 75,000 since the number of stages of the turbo-supercharger deter-
mines the practical limits of air compression.
Range
1. Brequet's formula is the customary range formula used for re-

477
ciprocating engines. When applied for successive increments of decreas-
ing gross weights, it will lead to results that are quite reasonable.
2. A modification of the formula for personal aircraft with fixed-pitch
propellers reads as follows:
where W0 = initial gross weight,
We = weight without fuel, and
K varies from 10,500 to 12,000.
3. For aircraft employing controllable-pitch propellers and reciprocating
engines:
where K\ varies from 50 to 70 and K2 varies from 12,000 to 13,000.
R = K logio j^2'
Absorbers, shock, 359-361, 362-364
dimensions, 380
energy considerations, 361-362
proportions, 380
types, 361-362
Aerodynamic twist, 319
Ailerons, 341-347, 387-388, 417-
419, 420
Air, arbitrary standard, 458
inlet systems, 281-296
movement, 222-223
requirements, 223-224
standard atmosphere, 458, 459-
460
upper atmosphere, 460
Air conditioning, air ducts, 227-
229, 281-296
air movement, 222-223
air requirements, 222, 223-224
boilers, 229
calculations, 230-232
equipment, 229-230
heat sources, 200-201, 229, 232
high speeds, 233-234
importance, 221-223
physical conditions, 226-227
pressure considerations, 224-225
radiators, 229
schematic system, 223, 226
soundproofing, 227
temperature limits, 226
Air ducts, 227-229, 281-296 (See
also Air induction systems)
Airfoil, aspect ratio corrections, 38,
39-41
characteristics, 36, 37, 47
compressible flow, 48-49

computation table, 40
construction, 45-46
criteria, 39
laminar flow, 48
ordinates, 45-46
power coefficient, 35
section characteristics, 41-45
selection, 29-49
structural considerations, 38
supersonic speed, 46-48
tail surfaces, 388-389
thickness, 321-322
thickness ratios, 38, 45
Air induction systems, annular inlet,
292
anti-icing, 284-285
conventional spinner, 283-284
ducted spinner, 281-283
flush inlet, 294-296
inlet-velocity ratio, 284-286, 288-
289
losses, 286
nose inlet, 290-292
scoop, 292-293
undercowl scoop, 286
wing-root inlet, 289-290
Airplanes, choice, 18-25
data, 26-28
forces in flight, 30
front view, 120
length, 118
performance, 28
482
Airplanes (Continued):
rectilinear flight, 29-36
side view, 120-121
three-quarter view, 8
three-view, 113-123
top view, 120
types, 10-28
Airsickness, cabin provisions, 218
Aisle dimensions, 205
Aluminum alloy, (See Materials of
construction)
Angle, aerodynamic twist, 319
dihedral, horizontal tail, 401
dihedral, wing, 317-319
downwash, 399-400
incidence, horizontal tail, 398-
400
incidence, wing, 319-320
sweepback, horizontal tail, 400-
401
sweepback, wing, 313-317, 318-
319
trim, 399
Anti-icing (See also De-icing)
engine inlet, 286-287
equipment, 199-201
Areas, control surfaces, 401, 407-
408
wing, 116-117
Aspect ratio, corrections, 38-41
tail surfaces, 398, 406
wing, 320-321
Axes, reference, 59-61
Baggage compartment, 220
Balance, aerodynamic, 389-391

ballast, 169
calculations, 163, 168-169

center of gravity, 157-171
clearances, 168
defined, 155
diagram, 155-157
dynamic, 392-393
fuselage, 161-164
power plant, 164
procedures, 155-157
static, 391-392
wing, 164-167
worksheet, 163
Ballast, 169
Beading, 105
Berths, cabin, 206, 217, 218
weights, 151
Biplane, 10-11
Boilers, 229
Bolts, 109
Brake, installation, 379
Cabin, airsickness, 218
aisles, 205
baggage compartment, 220
berths, 206, 217, 218
center of gravity, 216
comfort, 203
considerations, 202-203
dimensions, 203-205
doors, 217-218
exits, 217-218
flooring, 219
furnishings, 218
INDEX
483
Cables (Continued):
strengths, 416
supports, 415, 426
weights, 150, 416
Canopies, 183
Castings, aluminum, 70
magnesium, 74
Center of gravity, error, 171
estimation, 170-171
fuselage, 161
individual items, 169-170
movement, 160
seating, 170-216
Chord, mean aerodynamic, 306-310
mean geometric, 306-310
Clearances, flap, 351
landing gear, 369, 381
propeller, 247-249
Cockpit, canopies, 179, 183
controls, 188-189
design, 174-176
dimensions, 184
exits, 186
instrument board, 189
parachutes, 188
pilot dimensions, 185
prone seating, 190
seating, 183-186
vision and visibility, 182-183
windows, (See Windshields)
windshields, 176-181
Comfort, air movement (See Air
conditioning)
factors, 203
seating (See Seats and seating)

Compartments, baggage, 220
soundproofing, 239
Compressible flow, airfoils, 48-49
Control surfaces, areas, 387-388
travel, 425
Control systems, 412-428
adjustable stabilizer, 414, 419
aileron control, 420, 421
boost system, 419
cable, 412, 414-415, 416
control travel, 425-426
engine, 427-428
fairleads, 415
flap, 189, 420, 422
hinges, 422, 428, 429
hydraulic, 423-425
irreversible, 425, 426
pneumatic, 423-425
pulleys, 414, 426
push-pull, 412, 413
stops, 415
tabs, 420, 422
torque tube, 412
weights, 150
wheels, 188
Controls, cockpit, 188-189
engine, 427-428
wheel, 188
Cooling, at high speeds, 233-234
engine, 274-275
484
Electrical equipment, 198-199
heat sources, 229
Engines (See also Power plants)
carburetor scoop, 277-278
controls, 427-428
cooling system, 274-275
cowling, 273-275
firewall, 272-273
fuel consumption, 268-269
fuel systems, 278-279
installations, 251-253
jet types, 296
location, 119, 264
lubricating systems, 280-281
manifolds, 276-277
mount, 269-272, 297
nacelles, 268
number, 265-267
piston, 295
power rating, 267-268, 295
probability theory, 265-267
pusher installation, 252-253
rating, 267-268
selection, 268
submerged, 264-265
tandem, 251-252
thrust ratings, 296
weights, 295-296
Equipment, electrical, 198-199
hydraulic, 419, 423-425
safety, 199
Exits, cabin, 217-218, 449
cockpit, 186
Extrusions, 68, 74, 78
Fabricated components, weights,

145-146
Fabrication, 71-73
blanking, 72
cutting, 72
forming, 72-73
Fail-safe, 82
Fairleads, 415
Fool-proof, 82
Firewall, 272-273
Fittings, 97-99
Flaps, (See also Lift-increase de-
vices)
clearance, 351
controls, 420
wing, 422
Flight, climbing, 34-35
diving, 34
envelope, 58-59
gliding, 33-34
horizontal, 31-33
range, 36
Floats, weights, 153
Flooring, 100-102
cabin, 219, 449
soundproofing, 219
Flutter, prevention, 355-356, 393-
394
Forgings, 68-70
Frames, fabrication 439-442
spacing, 440-442
Fuel, consumption, 268-269
pumps, 279
systems, 278-279
INDEX
485
Fuselage (Continued):
shape, 432-435
skin thickness, 90, 446-447
stringers, 439, 442, 444-446
structure, 451-452
tail boom, 450-451
twin, 450
weight, 136-137, 142-143
windows, 447
wing configuration, 323, 430-432
Gauges, metal sheet, 80
tubing, 111-112
Headroom, cabin, 207-209
Heating, calculations, 230-232
sources, 229
surface, 232
Hinges, 422
Hydraulic equipment, 419, 423-425
actuating cylinders weight, 154
bend radii, 378
Instrument board, 189, 193-194
Instruments, cockpit, 191-201
grouping, 194
importance, 191
location, 194
selection, 197-198
variety, 194-197
Jet engines (See Power plant)
Joints, bolted and riveted, 95-97
Landing gear, brakes, 359-361
clearances, 381
considerations, 304, 357-358
cross-wind, 371-373
description, 357
dynamic loads, 358-359

energy dissipation, 361-362
engine torque, 250-251

landing contact, 364-365
location, 364-371
nose-wheel type, 367-370
retraction, 376-378
shimmy, 373-375
shock absorbers, 362-364
special problems, 380
speed reduction, 359-361
tail-wheel type, 365-367
tandem, 371
tire size, 152-153
track type, 370-371
tread, 375
weights, 138, 143-145
wheel position, 376
wheel size, 375-376
Legroom, cabin, 206, 209-215
cockpit, 184
Lift-increase devices, 347-354
Lighting, 218
Lightening holes, 105
Load factor, definition, 50-51
determination, 55-56
empirical, 57-58
gust, 52-55
maneuver, 51-52
Loads, 50-63
dynamic, 50-53
factors, 50-58
generalized system, 61-63
gust, 52-55
486
Materials (Continued):
aluminum alloys, 65-71
application, 64-65, 76-79
considerations, 65
fabricated, 71-73
magnesium alloys, 74
mechanical properties, 106-110
steel, 74-75
titanium alloys, 75-76
Mechanical properties, 106-110
Mockup, cockpit, 175
fuselage, 436
Noise, engine, 236, 239-240
frequency, 236-237
measure, 235
soundproofing, 235-240
sources, 236, 239-240
Oxygen, requirements, 221-224
cylinder weights, 150
Panels, sizes, 89-92
under load, 88-89
Parachutes, 188
weights, 151
Payload, 129
Performance, absolute ceiling, 468,
476
altitude corrections, 459
calculations, 455-459, 462, 469
data, sources, 454, 460-461
empirical formulas, 474-477
engine, 461-462
landing run, 245, 359-361
maximum speed, 466-468, 474-
475
parasite resistance, 460-461

power available, 246, 462-466
power required, 455-459

propeller characteristics, 242-243
propeller selection, 254-260
range, 268-269, 468-469, 476-477
rate of climb, 468, 475
service ceiling, 468, 471
stalling speed, 475
take-off, 470-J71
Pilot, average dimensions, 185
ejection equipment, 187
protection, 186-187
seating, 183-186
Power plants, (See also Engines)
air inlet systems, 281-296
defined, 263
fuel systems, 278-280
general considerations, 263-264
lubricating systems, 280-281
weights, 138-139, 146, 147-148
Pressurization, cabin equipment,
229-230
structure, 447-448
Propeller, aerodynamic effects, 243
asymmetrical conditions, 249-250
characteristics, 241-243
clearances, 247-249
cuffs, 254
diameter, 254-262
gyroscopic effect, 243-244
influence, 247
INDEX
487
Rockets, 298
Safety, design philosophy, 82
equipment, 199
margin, 81
Scoop, carburetor, 277-278
Seating, accessibility, 215-216
arrangements, 206-207
back-to-back, 213, 214
canoe, 210
cockpit, 183-186
dimensions, 204, 207
pick-a-back, 212
prone, 190
rearward, 214-215
side-by-side, 211, 213
staggered, 213
tandem, 215
toboggan, 210-211
Seats, dimensions, 184, 208
passenger, 205-206
Sheet, sizes, 67, 74, 78
Sheet-stringer, 92-93
Shock absorbers (See Absorbers,
shock)
Soundproofing (See also Noise)
application, 237-239
choice, 237
flooring, 219
materials, 227-230
noise measure, 235
weights, 151
Span (See Wing)
Spar, fittings, 334-337
wing, 327-331
Spinners, propeller hub, 254

Spoilers (<See Lift-increase devices)
Stability, criteria, 385

directional, 404-405
longitudinal, 382-386
mass distribution, 453
Stabilizer, adjustable, 419
Steel, 74-75, 77, 78, 79
Stiffeners, 86-88, 105, 330-331
Stops, 415
Stringers, 86-88, 330-331
with sheet, 92-93
Sweepback, 313-316
high-speed airplanes, 46-48
stability effects, 317
wing, 46
Tabs, tail surface, 391, 420, 422, 423
Tail surfaces, 382-411
airfoils, 388-389
areas, 401, 407-408
aspect ratio, 398, 406-407
balance, 389-393
butterfly, 409-411
clearances, 403
construction, 394-395, 401
control, 385-386, 404, 405
control surfaces, 387-388
dihedral, 401
dive brakes, 410
flutter prevention, 393-394
horizontal, 395-403
incidence, 398-400
location, 395-397, 405-406
movement, 397-398, 406

488
Three-view, engineering studies, 122
final, 122-123
front view, 121
general notes, 121
side view, 121
steps in assembly, 114-116
top view, 120
Tires, weights, 152-153
Toilets, 219
Tubing, round, 111-112
weights, 111-112
Ventilation (See Air conditioning)
Vision and visibility, cabin, 215
cockpit, 182-183
Weights, actual, 127
alternate, 127
calculated, 126
control surfaces, 137-138
empty, 146
engines (See Power plant), 295-
296
estimates, 114-116, 126, 128-132
fabricated components, 145-146
floats, 154
fuel, 147
fuselage, 136-137, 142-143
general procedure, 124-126
gross, 127, 148-149
hydraulic equipment, 154
landing gear, 138, 143-145
miscellaneous, 150-154
nacelles, 146-147
oil, 147
oil coolers, 148
power plants, 138-139, 146-148

propeller, 261-262
sources, 139
spinners, 254
structural, 128
tail surfaces, 141-142
tires, 152-153
variables, 132-139
wheels, 152-153
wing, 132-136, 139-141
worksheets, 129-131
Wheels, camber, 376
position, 376
size, 375-376
toe-in, 376
weights, 152-153
Windows, 176-181,216-217, 447
Windshields, 176-181
Wing (See also Airfoil)
aerodynamic twist, 319
aero-isoclinic, 316
ailerons, 341-347, 387-388
area, 116-117
balance calculations, 164-167
bending moments, 301
construction, 327
cranked, 316
crescent, 315-316
design, 299-356
dihedral, 317-319
fabric-covered, 337-341
fences, 354
flap construction, 353-422
flutter prevention, 355-356

INDEX
489
Wing (Continued):
ribs, 331-334, 339
rib spacing, 332, 341
skin thickness, 90
span, 117
spar construction, 327-331
spar fittings, 334-337
spoilers, 347-354
stringers, 330-331
structural considerations, 327
sweepback, 313-317, 318-319
taper ratio, 310-317
torsion, 301-302
weight, 132-136, 139-141


Airplane Design Manual - Teichmann

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Airplane Design Manual - Teichmann

Uploaded by

Copyright:

Available Formats

Airplane

Professor of Aeronautical Engineering

Assistant Dean, Day Division

College of Engineering, New York University

PITMAN PUBLISHING CORPORATION

NEW YORK TORONTO LONDON

3 SB Copyrioht, 1939, 1942, 1950, 1958

PITMAN PUBLISHING CORPORATION

All rights reserved. No part of this book

may be reproduced in any form without

the written permission of the publisher.

Sir Isaac Pitman a Sons, Ltd.

London Melbourne Johannesburg

Sir Isaac Pitman & Sons (Canada), Ltd.

Printed in the United States of America

Daniel Guggenheim School of Aeronautics, College of Engineering, New

Thus the book has been expanded to encompass recent developments,

thereby making it more effective and meaningful to the present-day

In addition, an attempt has been made in this revision to amplify the

analytical approach to design problems as well as the purely empirical

approach. The student of airplane design is in this way offered a deeper

appreciation of the interplay of aerodynamics, structural analysis, human

considerations, and other such factors.

Although practical necessity restricts the treatment that can be given

to refer to literature that is available elsewhere.

The objectives of aircraft design work in the college classroom are:

(1) To offer an integration of or focal point for applying the various

principles included in aerodynamics, structural design, installation require-

ments, and application of materials. If time permits, economics, perform-

ance calculations, and allied problems continue the understanding of the

(2) To afford a basis for stress analyses.

(3) To provide some drafting experience.

(4) To develop an "engineering sense" in the student, enabling him to

evaluate various requirements, judge the necessity of compromise, and

know the amount of time needed to achieve a given objective.

airplane design to effectively meet and fulfill these important goals.

Preface to the Third Edition

Although airplane design has progressed tremendously since this book

was first published, the fundamental approach to the problem of airplane

design has not changed. However, the possibility of attaining transonic

few years ago, has introduced additional considerations in designing even

The author greatly appreciates the work in redrawing and preparing

Preface to the First Edition

gap in aeronautical literature, an introduction to the art of airplane de-

the student working on his own especially in view. While aerodynamics,

times, experience in dealing with senior aeronautical students has shown

that such men experience considerable difficulty in coordinating their

knowledge and efforts in approaching the difficult problem of actually

and these have gradually evolved into the present work.

experience indicates that such a manual is helpful to instructors and stu-

It is of course expected that the student shall supplement the present

text by investigations of his own, by studying the latest designs at the

airport, or from descriptions in the technical press, or by study of the

numerous research publications published by the Government Printing

been made to make each chapter of the book as complete in itself as

Instrument Co., R.C.A. Manufacturing Co., Inc., The Cleveland Pneu-

tion, Harrison Radiator Corporation, Eclipse Aviation Corporation, Kolls-

man Instrument Co.

to Professor Alexander Klemin, who established the course in airplane de-

suggestions and criticisms.

, Chapter I. Procedure in Design

Selecting Type of Airplane .

Payload and Crew