NASA/TP–2015-218751

NDARC
NASA Design and Analysis of Rotorcraft

Theory

Appendix 5
Release 1.12
August 2017

Wayne Johnson
NASA Ames Research Center, Moffett Field, CA

August 2017
 The NASA STI Program Office . . . in Profile

Since its founding, NASA has been dedicated to the • CONFERENCE PUBLICATION. Collected
advancement of aeronautics and space science. The papers from scientific and technical conferences,
NASA Scientific and Technical Information (STI) symposia, seminars, or other meetings sponsored
Program Office plays a key part in helping NASA or cosponsored by NASA.
maintain this important role.
• SPECIAL PUBLICATION. Scientific, technical,
The NASA STI Program Office is operated by or historical information from NASA programs,
Langley Research Center, the Lead Center for projects, and missions, often concerned with
NASA’s scientific and technical information. The subjects having substantial public interest.
NASA STI Program Office provides access to the
NASA STI Database, the largest collection of • TECHNICAL TRANSLATION. English-
aeronautical and space science STI in the world. language translations of foreign scientific and
The Program Office is also NASA’s institutional technical material pertinent to NASA’s mission.
mechanism for disseminating the results of its
research and development activities. These results Specialized services that complement the STI
are published by NASA in the NASA STI Report Program Office’s diverse offerings include creating
Series, which includes the following report types: custom thesauri, building customized databases,
organizing and publishing research results . . . even
• TECHNICAL PUBLICATION. Reports of providing videos.
completed research or a major significant phase
of research that present the results of NASA For more information about the NASA STI
programs and include extensive data or theoreti- Program Office, see the following:
cal analysis. Includes compilations of significant
scientific and technical data and information • Access the NASA STI Program Home Page at
deemed to be of continuing reference value. http://www.sti.nasa.gov
NASA’s counterpart of peer-reviewed formal
professional papers but has less stringent limita- • E-mail your question via the Internet to
tions on manuscript length and extent of graphic [email protected]
presentations.
• Fax your question to the NASA Access Help
• TECHNICAL MEMORANDUM. Scientific Desk at (301) 621-0134
and technical findings that are preliminary or of
specialized interest, e.g., quick release reports, • Telephone the NASA Access Help Desk at
working papers, and bibliographies that contain (301) 621-0390
minimal annotation. Does not contain extensive
analysis. • Write to:
NASA Access Help Desk
• CONTRACTOR REPORT. Scientific and tech- NASA Center for AeroSpace Information
nical findings by NASA-sponsored contractors 7115 Standard Drive
and grantees. Hanover, MD 21076-1320
NASA/TP–2015-218751

NDARC
NASA Design and Analysis of Rotorcraft

Theory

Appendix 5
Release 1.12
August 2017

Wayne Johnson
NASA Ames Research Center, Moffett Field, CA

National Aeronautics and

Space Administration

Ames Research Center

Moffett Field, California 94035-1000

August 2017
Available from:

NASA Center for AeroSpace Information National Technical Information Service

7115 Standard Drive 5285 Port Royal Road
Hanover, MD 21076-1320 Springfield, VA 22161
(301) 621-0390 (703) 487-4650
 iii

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2. Nomenclature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3. Tasks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4. Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
5. Solution Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
6. Cost . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
7. Emissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
8. Aircraft . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
9. Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95
10. Fuselage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
11. Landing Gear . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
12. Rotor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
13. Wing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 139
14. Empennage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
15. Propulsion System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
16. Fuel Tank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
17. Propulsion Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
18. Engine Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
19. Jet Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
20. Charge Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
21. Referred Parameter Turboshaft Engine Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
22. Reciprocating Engine Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
23. Compressor Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
24. Motor Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
25. Referred Parameter Jet Engine Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221
iv Contents

26. Fuel Cell Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227

27. Solar Cell Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
28. Battery Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
29. AFDD Weight Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247
30. Other Weight Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
 Chapter 1

Introduction

The NASA Design and Analysis of Rotorcraft (NDARC) software is an aircraft system analysis
tool that supports both conceptual design efforts and technology impact assessments. The principal tasks
are to design (or size) a rotorcraft to meet specified requirements, including vertical takeoff and landing
(VTOL) operation, and then analyze the performance of the aircraft for a set of conditions. For broad and
lasting utility, it is important that the code have the capability to model general rotorcraft configurations,
and estimate the performance and weights of advanced rotor concepts. The architecture of the NDARC
code accommodates configuration flexibility, a hierarchy of models, and ultimately multidisciplinary
design, analysis, and optimization. Initially the software is implemented with low-fidelity models,
typically appropriate for the conceptual design environment.
An NDARC job consists of one or more cases, each case optionally performing design and analysis
tasks. The design task involves sizing the rotorcraft to satisfy specified design conditions and missions.
The analysis tasks can include off-design mission performance calculation, flight performance calcula-
tion for point operating conditions, and generation of subsystem or component performance maps. For
analysis tasks, the aircraft description can come from the sizing task, from a previous case or a previous
NDARC job, or be independently generated (typically the description of an existing aircraft).
The aircraft consists of a set of components, including fuselage, rotors, wings, tails, and propulsion.
For each component, attributes such as performance, drag, and weight can be calculated; and the aircraft
attributes are obtained from the sum of the component attributes. Description and analysis of conven-
tional rotorcraft configurations is facilitated, while retaining the capability to model novel and advanced
concepts. Specific rotorcraft configurations considered are single-main-rotor and tail-rotor helicopter,
tandem helicopter, coaxial helicopter, and tiltrotor. The architecture of the code accommodates addition
of new or higher-fidelity attribute models for a component, as well as addition of new components.

1–1 Background

The definition and development of NDARC requirements benefited substantially from the ex-
periences and computer codes of the preliminary design team of the U.S. Army Aeroflightdynamics
Directorate (AFDD) at Ames Research Center.
In the early 1970s, the codes SSP-1 and SSP-2 were developed by the Systems Research Integration
Office (SRIO, in St. Louis) of the U.S. Army Air Mobility Research and Development Laboratory.
SSP-1 performed preliminary design to meet specified mission requirements, and SSP-2 estimated the
performance for known geometry and engine characteristics, both for single-main-rotor helicopters
(ref. 1). Although similar tools were in use in the rotorcraft community, these computer programs were
independently developed, to meet the requirements of government analysis. The Advanced Systems
Research Office (ASRO, at Ames Research Center) of USAAMRDL produced in 1974 two Preliminary
2 Introduction

Systems Design Engineering (PSDE) studies (refs. 2 and 3) using SSP-1 and SSP-2. These two codes
were combined into one code called PSDE by Ronald Shinn.
The MIT Flight Transportation Laboratory created design programs for helicopters (ref. 4) and
tiltrotors (ref. 5). Michael Scully, who wrote the helicopter design program and was significantly
involved in the development of the tiltrotor design program, joined ASRO in 1975, and ideas from the
MIT programs began to be reflected in the continuing development of PSDE. An assessment of design
trade-offs for the Advanced Scout Helicopter (ASH) used a highly modified version of PSDE (ref. 6).
A DoD Joint Study Group was formed in April 1975 to perform an Interservice Helicopter Com-
monality Study (HELCOM) for the Director of Defense Research and Engineering. The final HELCOM
study report was published in March 1976 (ref. 7). A result of this study was an assessment by ASRO
that PSDE needed substantial development, including better mathematical models and better technical
substantiation, more flexible mission analysis, and improved productivity for both design and analysis
tasks. Thus began an evolutionary improvement of the code, eventually named RASH (after the devel-
oper Ronald A. Shinn, as a consequence of the computer system identification of output by the first four
characters of the user name). RASH included improvements in flight performance modeling, output
depth, mission analysis, parametric weight estimation, design sensitivity studies, off-design cases, and
coding style. The code was still only for single-main-rotor helicopters.
In the early 1980s, tool development evolved in two separate directions with the Preliminary
Design Team at ASRO. RASH was developed into the HELO (or PDPAC) code, for conventional
and compound single-main-rotor helicopters. With the addition of conversion models and wing weight
estimation methods (refs. 8 and 9), RASH became the TR code, for tiltrotor aircraft. The JVX Joint
Technology Assessment of 1982 utilized the HELO and TR codes. A special version called PDABC,
including a weight estimation model for lift-offset rotors (ref. 10), was used to analyze the Advancing
Blade Concept. The JVX JTA report (ref. 11) documented the methodology implemented in these codes.
Work in support of the LHX program from 1983 on led to a requirement for maneuver analysis
of helicopters and tiltrotors, implemented in the MPP (Maneuver Performance Program) code by John
Davis. The core aircraft model in MPP was similar to that in TR and HELO, but the trim strategy in
particular was new. A design code does not require extensive maneuver analysis capability, but MPP had
an impact on the design code development, with the MPP performance and trim methods incorporated into
TR87. The sizing analysis of TR88 and the aircraft flight model from MPP were combined into the VAMP
(VSTOL Design and Maneuver Program) code. VAMP combined the capability to analyze helicopters
and tiltrotors in a single tool, although the capability of HELO to analyze compound helicopters was
not replicated.
In the early 1990s, the RC (RotorCraft) code emerged from the evolution of VAMP, with John
Preston as the lead developer (refs. 12 and 13). Some maneuver analysis capabilities from MPP were
added, and the analysis capability extended to helicopters. The models were confirmed by comparison
with results from TR and HELO. RC was operational by 1994, although HELO and TR continued to be
used into the mid-1990s. RC97 was a major version, unifying the tiltrotor and helicopter analyses. The
RC code introduced new features and capabilities, and productivity enhancements, as well as coding
standards and software configuration control. Special versions of RC were routinely produced to meet
the unique requirements of individual projects (such as ref. 14).
NASA, with support from the U.S. Army, in 2005 conducted the design and in-depth analysis of
rotorcraft configurations that could satisfy the Vehicle Systems Program technology goals (ref. 15). These
Introduction 3

technology goals and accompanying mission were intended to identify enabling technology for civil
application of heavy-lift rotorcraft. The emphasis was on efficient cruise and hover, efficient structures,
and low noise. The mission specified was to carry 120 passengers for 1200 nm, at a speed of 350 knots
and 30,000 ft altitude. The configurations investigated were a Large Civil Tiltrotor (LCTR), a Large
Civil Tandem Compound (LCTC), and a Large Advancing Blade Concept (LABC). The results of the
NASA Heavy Lift Rotorcraft Systems Investigation subsequently helped define the content and direction
of the Subsonic Rotary Wing project in the NASA Fundamental Aeronautics program. The design tool
used was the AFDD RC code. This investigation was an example of the role of a rotorcraft sizing code
within NASA. The investigation also illustrated the difficulties involved in adapting or modifying RC
for configurations other than conventional helicopters and tiltrotors, supporting the requirement for a
new tool.

1–2 Requirements

Out of this history, the development of NDARC began in early 2007. NDARC is entirely new
software, built on a new architecture for the design and analysis of rotorcraft. From the RC theoretical
basis, the parametric weight equations and the Referred Parameter Turboshaft Engine Model were used
with only minor changes. Use was also made of the RC component aerodynamic models and rotor
performance model. The current users of RC, informed by past and recent applications, contributed
significantly to the requirements definition.

The principal tasks are to design (size) rotorcraft to meet specified requirements, and then analyze
the performance of the aircraft for a set of flight conditions and missions. Multiple design requirements,
from specific flight conditions and various missions, must be used in the sizing task. The aircraft
performance analysis must cover the entire spectrum of aircraft capabilities, and allow general and
flexible definition of conditions and missions.

For government applications and to support research, it is important to have the capability to model
general rotorcraft configurations, including estimates of the performance and weights of advanced rotor
concepts. In such an environment, software extensions and modifications are routinely required to meet
the unique requirements of individual projects, including introduction of special weight and performance
models for particular concepts.

Thus the code architecture must accommodate configuration flexibility and alternate models, in-
cluding a hierarchy of model fidelity. Although initially implemented with low-fidelity models, typical
of the conceptual design environment, ultimately the architecture must allow multidisciplinary design,
analysis, and optimization. The component performance and engine models must cover all operat-
ing conditions. The software design and architecture must facilitate extension and modification of the
software.

Complete and thorough documentation of the theory and its software implementation is essential,
to support development and maintenance, and to enable effective use and modification. Most of the
history described above supports this requirement by the difficulties encountered in the absence of good
documentation. Documentation of the methodology was often prompted only by the need to substantiate
conclusions of major technology assessments, and occasionally by the introduction of new users and
developers. For a new software implementation of a new architectures, documentation is required from
the beginning of the development.
4 Introduction

fixed model or
previous job or
previous case
DESIGN ANALYZE
Airframe
Aerodynamics Map

Sizing Task
size iteration Aircraft Engine
Description Performance Map

Mission Analysis

design
conditions design Flight
missions Performance
Analysis

Mission
Flight Condition
adjust & fuel wt iteration
max GW
max takeoff GW

each segment

Flight State
max effort / trim aircraft / flap equations

Figure 1-1. Outline of NDARC tasks.

1–3 Overview

The NDARC code performs design and analysis tasks. The design task involves sizing the rotorcraft
to satisfy specified design conditions and missions. The analysis tasks can include off-design mission
performance analysis, flight performance calculation for point operating conditions, and generation of
subsystem or component performance maps. Figure 1-1 illustrates the tasks. The principal tasks (sizing,
mission analysis, and flight performance analysis) are shown in the figure as boxes with heavy borders.
Heavy black arrows show control of subordinate tasks.
The aircraft description (fig. 1-1) consists of all the information, input and derived, that defines
the aircraft. The aircraft consists of a set of components, including fuselage, rotors, wings, tails, and
propulsion. This information can be the result of the sizing task; can come entirely from input, for a
fixed model; or can come from the sizing task in a previous case or previous job. The aircraft description
information is available to all tasks and all solutions (indicated by light green arrows).
Introduction 5

The sizing task determines the dimensions, power, and weight of a rotorcraft that can perform a
specified set of design conditions and missions. The aircraft size is characterized by parameters such as
design gross weight, weight empty, rotor radius, and engine power available. The relationships between
dimensions, power, and weight generally require an iterative solution. From the design flight conditions
and missions, the task can determine the total engine power or the rotor radius (or both power and radius
can be fixed), as well as the design gross weight, maximum takeoff weight, drive system torque limit,
and fuel tank capacity. For each propulsion group, the engine power or the rotor radius can be sized.

Missions are defined for the sizing task and for the mission performance analysis. A mission
consists of a number of mission segments, for which time, distance, and fuel burn are evaluated. For
the sizing task, certain missions are designated to be used for engine sizing, for design gross weight
calculations, for transmission sizing, and for fuel tank sizing. The mission parameters include mission
takeoff gross weight and useful load. For specified takeoff fuel weight with adjustable segments, the
mission time or distance is adjusted so the fuel required for the mission (burned plus reserve) equals the
takeoff fuel weight. The mission iteration is on fuel weight or energy.

Flight conditions are specified for the sizing task and for the flight performance analysis. For the
sizing task, certain flight conditions are designated to be used for engine sizing, for design gross weight
calculations, for transmission sizing, for maximum takeoff weight calculations, and for antitorque or
auxiliary-thrust rotor sizing. The flight condition parameters include gross weight and useful load.

For flight conditions and mission takeoff, the gross weight can be maximized, such that the power
required equals the power available.

A flight state is defined for each mission segment and each flight condition. The aircraft performance
can be analyzed for the specified state, or a maximum effort performance can be identified. The maximum
effort is specified in terms of a quantity such as best endurance or best range, and a variable such as
speed, rate of climb, or altitude. The aircraft must be trimmed, by solving for the controls and motion
that produce equilibrium in the specified flight state. Different trim solution definitions are required for
various flight states. Evaluating the rotor hub forces may require solution of the blade flap equations of
motion.

1–4 Terminology

The following terminology is introduced as part of the development of the NDARC theory and
software. Relationships among these terms are reflected in figure 1-1.

a) Job: An NDARC job consists of one or more cases.

b) Case: Each case performs design and/or analysis tasks. The analysis tasks can include off-design
mission performance calculation, flight performance calculation for point operating conditions, and
generation of airframe aerodynamics or engine performance maps.

c) Design Task: Size rotorcraft to satisfy specified set of design flight conditions and/or design missions.
Key aircraft design variables are adjusted until all criteria are met. The resulting aircraft description can
be basis for the mission analysis and flight performance analysis tasks.

d) Mission Analysis Task: Calculate aircraft performance for one off-design mission.

e) Flight Performance Analysis Task: Calculate aircraft performance for point operating condition.
6 Introduction

f) Mission: Ordered set of mission segments, for which time, distance, and fuel burn are evaluated.
Gross weight and useful load are specified for the beginning of the mission, and adjusted for fuel burn
and useful load changes at each segment. Missions are defined for the sizing task and for the mission
performance analysis.
g) Flight Condition: Point operating condition, with specified gross weight and useful load. Flight
conditions are specified for the sizing task and for the flight performance analysis.
h) Flight State: Aircraft flight condition, part of definition of each flight condition and each mission
segment. Flight state solution involves rotor blade motion, aircraft trim, and perhaps a maximum-effort
calculation.
i) Component: The aircraft consists of a set of components, including fuselage, rotors, wings, tails, and
propulsion. For each component, attributes such as performance, drag, and weight are calculated.
j) Propulsion: A propulsion group is a set of components and engine groups, connected by a drive
system. An engine group consists of one or more engines of a specific type. The components define
the power required. The engine groups define the power available. A jet group consists of one or more
systems that produce a force on the aircraft. A charge group consists of systems that generate energy
for the aircraft. Fuel tank systems are associated with the engine groups, jet groups, and charge groups.
Fuel quantity is measured as either weight or energy.

1–5 Analysis Units

The code can use either English or SI units for input, output, and internal calculations. A consistent
mass-length-time-temperature system is used, except for weight and power:

length mass time temperature weight power

◦
English: foot slug second F pound horsepower
◦
SI: meter kilogram second C kilogram kiloWatt

Weight in the design description is actually mass, with pounds converted to slugs using the reference
gravitational acceleration (9.80665 m/sec2 ). Gravitational force is the product of the mass and the actual
acceleration due to gravity. In addition, the default units for flight conditions and missions are: speed
in knots, time in minutes, distance in nautical miles, and rate of climb in feet-per-minute. The user can
specify alternate units for these and other quantities.

1–6 Outline of Report

This document provides a complete description of the NDARC theoretical basis and architecture.
Chapters 3–5 describe the tasks and solution procedures, the cost model is described in chapter 6, the
emissions model is described in chapter 7, and chapters 8–20 present the models for the aircraft and
its components. The propulsion system models are described in chapters 15–20. The engine, jet, and
charger models are described in chapters 21–28; and the weight models in chapters 29and 30. The
accompanying NDARC Input Manual describes the use of the code.
Introduction 7
1–7 References

1) Schwartzberg, M.A.; Smith, R.L.; Means, J.L.; Law, H.Y.H.; and Chappell, D.P. “Single-Rotor
Helicopter Design and Performance Estimation Programs.” USAAMRDL Report SRIO 77-1, June
1977.
2) Wheatley, J.B., and Shinn, R.A. “Preliminary Systems Design Engineering for a Small Tactical Aerial
Reconnaissance System-Visual.” USAAMRDL, June 1974.
3) Shinn, R.A. “Preliminary Systems Design Engineering for an Advanced Scout Helicopter.” US-
AAMRDL, August 1974.
4) Scully, M., and Faulkner, H.B. “Helicopter Design Program Description.” MIT FTL Technical Memo
71-3, March 1972.
5) Faulkner, H.B. “A Computer Program for the Design and Evaluation of Tilt Rotor Aircraft.” MIT
FTL Technical Memo 74-3, September 1974.
6) Scully, M.P., and Shinn, R.A. “Rotor Preliminary Design Trade-Offs for the Advanced Scout Heli-
copter.” American Helicopter Society National Specialists’ Meeting on Rotor System Design, Philadel-
phia, Pennsylvania, October 1980.
7) “Interservice Helicopter Commonality Study, Final Study Report.” Director of Defense Research and
Engineering, Office of the Secretary of Defense, March 1976.
8) Chappell, D.P. “Tilt-rotor Aircraft Wing Design.” ASRO-PDT-83-1, 1983.
9) Chappell, D., and Peyran, R. “Methodology for Estimating Wing Weights for Conceptual Tilt-Rotor
and Tilt-Wing Aircraft.” SAWE Paper No. 2107, Category No. 23, May 1992.
10) “Weight Trend Estimation for the Rotor Blade Group, Rotor Hub Group, and Upper Rotor Shaft of
the ABC Aircraft.” ASRO-PDT-83-2, 1983.
11) “Technology Assessment of Capability for Advanced Joint Vertical Lift Aircraft (JVX), Summary
Report.” U.S. Army Aviation Research and Development Command, AVRADCOM Report, May 1983.
12) Preston, J., and Peyran, R. “Linking a Solid-Modeling Capability with a Conceptual Rotorcraft
Sizing Code.” American Helicopter Society Vertical Lift Aircraft Design Conference, San Francisco,
California, January 2000.
13) Preston, J. “Aircraft Conceptual Design Trim Matrix Selection.” American Helicopter Society
Vertical Lift Aircraft Design Conference, San Francisco, California, January 2006.
14) Sinsay, J.D. “The Path to Turboprop Competitive Rotorcraft: Aerodynamic Challenges.” American
Helicopter Society Specialists’ Conference on Aeromechanics, San Francisco, California, January 2008.
15) Johnson, W.; Yamauchi, G.K.; and Watts, M.E. “NASA Heavy Lift Rotorcraft Systems Investiga-
tion.” NASA TP 2005-213467, December 2005.
8 Introduction
 Chapter 2

Nomenclature

The nomenclature for geometry and rotations employs the following conventions. A vector x is a
column matrix of three elements, measuring the vector relative to a particular basis (or axes, or frame).
The basis is indicated as follows:
a) xA is a vector measured in axes A;
b) xEF/A is a vector from point F to point E, measured in axes A.
A rotation matrix C is a three-by-three matrix that transforms vectors from one basis to another:
c) C BA transforms vectors from basis A to basis B, so xB = C BA xA .
The matrix C BA defines the orientation of basis B relative to basis A, so it also may be viewed as rotating
the axes from A to B. For a vector u, a cross-product matrix u is defined as follows:
⎡ ⎤
0 −u3 u2
 = ⎣ u3
u 0 −u1 ⎦
−u2 u1 0

such that uv is equivalent to the vector cross-product u × v. The cross-product matrix enters the rela-
tionship between angular velocity and the time derivative of a rotation matrix:

Ċ AB = −  BA/B
ω AB/A C AB = C AB ω

(the Poisson equations). For rotation by an angle α about the x, y , or z axis (1, 2, or 3 axis), the following
notation is used: ⎡ ⎤
1 0 0
Xα = ⎣ 0 cos α sin α ⎦
0 − sin α cos α
⎡ ⎤
cos α 0 − sin α
Yα = ⎣ 0 1 0 ⎦
sin α 0 cos α
⎡ ⎤
cos α sin α 0
Zα = ⎣ − sin α cos α 0 ⎦
0 0 1
Thus for example, C BA = Xφ Yθ Zψ means that the axes B are located relative to the axes A by first
rotating by angle ψ about the z -axis, then by angle θ about the y -axis, and finally by angle φ about the
x-axis.
10 Nomenclature

Acronyms
AFDD U.S. Army Aeroflightdynamics Directorate
ASM available seat mile
CAS calibrated airspeed
CG charge group
CPI consumer price index
CTM Cost Too Much (cost model)
EG engine group
GW gross weight
IAS indicated airspeed
IGE in ground effect
IRP intermediate rated power
IRS infrared suppressor
ISA International Standard Atmosphere
ISO International Organization for Standardization
JG jet group
MCP maximum continuous power
MCT maximum continuous thrust
MJ Mega-Joule
MRP maximum rated power
NDARC NASA Design and Analysis of Rotorcraft
OEI one engine inoperative
OGE out of ground effect
PG propulsion group
RPJEM referred parameter jet engine model
RPTEM referred parameter turboshaft engine model
SDGW structural design gross weight
SI Système International d’Unités (International System of Units)
SLS sea level standard
TAS true airspeed
WMTO maximum takeoff weight

Weights
WD design gross weight
WE empty weight
WM T O maximum takeoff weight
WSD structural design gross weight
WG gross weight, WG = WE + WU L = WO + Wpay + Wfuel
WO operating weight, WO = WE + WF U L
WU L useful load, WU L = WF U L + Wpay + Wfuel
Wpay payload
Wfuel fuel weight
WF U L fixed useful load
Wburn mission fuel burn
Wvib vibration control weight
Wcont contingency weight
χ technology factor
Nomenclature 11

Fuel Tanks
Wfuel−cap fuel capacity, maximum usable fuel weight
Efuel−cap fuel capacity, maximum usable fuel energy
Vfuel−cap fuel capacity, volume
Nauxtank number of auxiliary fuel tanks
Waux−cap auxiliary fuel tank capacity (weight)
Eaux−cap auxiliary fuel tank capacity (energy)

Power
PreqP G power required, propulsion group; Pcomp + Pxmsn + Pacc
PreqEG power required, engine group
PreqCG power required, charge group

PavP G power available, propulsion group; min( fP PavEG , (Ωprim /Ωref )PDSlimit )
PavEG power available, engine group; (Neng − Ninop )Pav
PavCG power available, charge group; (Nchrg − Ninop )Pav
Pcomp component power required
Pxmsn transmission losses
Pacc accessory power
Ninop number of inoperative systems, engine group or jet group or charge group
PDSlimit drive system torque limit (specified as power limit at reference rotor speed)
PESlimit engine shaft limit
PRSlimit rotor shaft limit

Engine
Peng sea level static power available per engine at specified takeoff rating
Neng number of engines in engine group
Pav power available, installed; min(Pa − Ploss , Pmech )
Pa power available, uninstalled
Preq power required, installed; Pq − Ploss
Pq power required, uninstalled
Ploss installation losses
Pmech mechanical power limit
SP specific power, P/ṁ (conventional units)
sfc specific fuel consumption, ẇ/P (conventional units)
ṁ mass flow (conventional units)
ẇ fuel flow (conventional units)
Ė energy flow
FN net jet thrust
Daux momentum drag
N specification turbine speed
SW specific weight, P/W
12 Nomenclature

Jet
Teng sea level static thrust available per jet at specified takeoff rating
Njet number of jets in jet group
TavCG thrust available, jet group; (Njet − Ninop )Tav
Tav thrust available, installed; min(Ta η, Tmech )
Ta thrust available, uninstalled
TreqJG thrust required, jet group
Treq thrust required, installed; Tq η
Tq thrust required, uninstalled
η installation losses (efficiency)
Tmech mechanical thrust limit
ST specific thrust, T /ṁ (conventional units)
sfc specific fuel consumption, ẇ/T (conventional units)
ṁ mass flow (conventional units)
ẇ fuel flow (conventional units)
Daux momentum drag
SW specific weight, T /W

Charger
Pchrg sea level static power available per charger at specified takeoff rating
Nchrg number of chargers in charge group
η installation losses (efficiency)

Tip Speed and Rotation

Vtip−ref reference tip speed, propulsion group primary rotor; each drive state
r gear ratio; Ωdep /Ωprim for rotor, Ωspec /Ωprim for engine
Ωprim primary rotor rotational speed, Ω = Vtip−ref /R
Ωdep dependent rotor rotational speed, Ω = Vtip−ref /R
Ωspec specification engine turbine speed
Nspec specification engine turbine speed (rpm)

Mission
T mission segment time
D mission segment distance
dR mission segment range contribution
E endurance
R range
ẇ fuel flow
Nomenclature 13

Environment
g gravitational acceleration
h altitude
cs speed of sound
ρ density
ν kinematic viscosity
μ viscosity
T temperature, ◦ R or ◦ K
τ temperature, ◦ F or ◦ C
Vw wind speed

Axis Systems
I inertial
F aircraft
A component aerodynamic
B component
V velocity

Geometry
SL, BL, WL fixed input position (station line, buttline, waterline)
positive aft, right, up; arbitrary origin
x/L, y/L, z/L scaled input position; positive aft, right, up; origin at reference point
L reference length (fuselage length, rotor radius, or wing span)
x, y , z calculated position, aircraft axes; positive forward, right, down;
origin at reference point for geometry,
origin at center of gravity for motion and loads
zF component position vector, in aircraft axes, relative reference point
length
Swet wetted area
14 Nomenclature

Motion
φF , θF , ψF roll, pitch, yaw angles; orientation airframe axes F relative inertial axes
ψ̇F turn rate
θV , ψV climb, sideslip angles; orientation velocity axes V relative inertial axes
F
vAC aircraft velocity
F
ωAC aircraft angular velocity
F
aAC aircraft linear acceleration
n load factor
V aircraft velocity magnitude
Vh horizontal velocity
Vf forward velocity
Vs sideward velocity
Vc climb velocity
Vcal calibrated airspeed
Vind indicated airspeed

Aerodynamics and Loads

v component velocity relative air (including interference)
q dynamic pressure, 1/2ρ|v|2
α angle of attack, component axes B relative aerodynamic axes A
β sideslip angle, component axes B relative aerodynamic axes A
f ratio flap chord to airfoil chord, cf /c
δf flap deflection
F force
M moment
D, Y , L aerodynamic drag, side, lift forces (component aerodynamic axes A)
Mx , My , Mz aerodynamic roll, pitch, yaw moments (component aerodynamic axes A)
cd , c section drag, lift coefficients
CD , CY , CL component drag, side, lift force coefficients
C , CM , CN component roll, pitch, yaw moment coefficients
D/q drag area, SCD (S = reference area of component)

Aircraft
DL disk loading, WD /Aref

Aref reference rotor area, fA A; typically projected area of lifting rotors
WL wing loading, WD /Sref

Sref reference wing area, S ; sum area all wings
cAC aircraft control
T control matrix
c component control, c = ST cAC + c0
αtilt tilt control variable 
M aircraft hover figure of merit, W W/2ρAref /P
De aircraft effective drag, P/V
L/De aircraft effective lift-to-drag ratio, W V /P
Nomenclature 15

Rotor
W/A disk loading, W = fW WD
2
CW /σ design blade loading, W/ρAVtip σ (Vtip = hover tip speed)
R blade radius
A disk area
σ solidity (ratio blade area to disk area)
Tdesign design thrust of antitorque or auxiliary-thrust rotor
r direction of rotation (1 for counter-clockwise, −1 for clockwise)
r blade span coordinate
ψ blade azimuth coordinate
μ advance ratio
λ inflow ratio
Vtip tip speed
Mtip tip Mach number
Mat advancing tip Mach number
ν blade flap frequency (per-rev)
γ blade Lock number
CT /σ thrust coefficient divided by solidity, T /ρA(ΩR)2 σ
βc , βs longitudinal, lateral flapping (tip-path plane tilt relative shaft)
θ0.75 blade collective pitch angle (at 75% radius)
θc , θs lateral, longitudinal blade pitch angle)
H, Y , T drag, side, thrust force on hub (shaft axes)
Mx , My roll, pitch moment on hub
Q shaft torque
Pi , Pt , Po , P p induced, interference, profile, parasite power
κ induced power factor, Pi = κPideal
cdmean profile power mean drag coefficient, CP o = (σ/8)cdmean FP
M rotor hover figure of merit, T fD v/P
L/De rotor effective lift-to-drag ratio, V L/(Pi + Po )
η propulsive efficiency, Pp /P = −XV /P

Wing
W/S wing loading, W = fW WD
S area
b span
c chord, S/b
AR aspect ratio, b2 /S
16 Nomenclature
 Chapter 3

Tasks

The NDARC code performs design and analysis tasks. The design task involves sizing the rotorcraft
to satisfy specified design conditions and missions. The analysis tasks can include mission performance
analysis, flight performance calculation for point operating conditions, and generation of subsystem or
component performance maps.

3–1 Size Aircraft for Design Conditions and Missions

3-1.1 Sizing Method

The sizing task determines the dimensions, power, and weight of a rotorcraft that can perform a
specified set of design conditions and missions. The aircraft size is characterized by parameters such as
design gross weight (WD ) or weight empty (WE ), rotor radius (R), and engine power available (Peng ).
The relationships between dimensions, power, and weight generally require an iterative solution. From
the design flight conditions and missions, the task can determine the total engine power or the rotor
radius (or both power and radius can be fixed), as well as the design gross weight, maximum takeoff
weight, drive system torque limit, and fuel tank capacity. For each propulsion group, the engine power
or the rotor radius can be sized:
a) Engine power: Determine Peng , for fixed R. The engine power is the maxi-
mum of the power required for all designated sizing flight conditions and sizing
missions (typically including vertical flight, forward flight, and one-engine inopera-
tive). Hence the engine power is changed by the ratio max(PreqP G /PavP G ) (exclud-
ing flight states for which zero power margin is calculated, such as maximum gross
weight or maximum effort). This approach is the one most commonly used for the
sizing task.
b) Rotor radius: Determine R for input Peng . The maximum power required for all
designated sizing flight conditions and sizing missions is calculated, and then the
rotor radius determined such that the power required equals the input power available.

The change in radius is estimated as R = Rold PreqP G /PavP G (excluding flight
states for which zero power margin is calculated, such as maximum gross weight or
maximum effort). For multi-rotor aircraft, the radius can be fixed rather than sized
for some rotors.
Alternatively, Peng and R can be input rather than sized. For each engine group that consumes power,
the design power can be sized:
Determine Peng . The design power is the maximum of the power required for all
designated sizing flight conditions and sizing missions. Hence the design power
18 Tasks

is changed by the ratio max(PreqEG /PavEG ) (excluding flight states for which zero
power margin is calculated).

For each jet group, the design thrust can be sized:

Determine Tjet . The design thrust is the maximum of the thrust required for all
designated sizing flight conditions and sizing missions. Hence the design thrust
is changed by the ratio max(TreqJG /TavJG ) (excluding flight states for which zero
thrust margin is calculated).

For each charge group, the design power can be sized:

Determine Pchrg . The design power is the maximum of the power required for all
designated sizing flight conditions and sizing missions. Hence the design power
is changed by the ratio max(PreqCG /PavCG ) (excluding flight states for which zero
power margin is calculated).

Aircraft parameters can be determined by a subset of the design conditions and missions:

a) Design gross weight WD : maximum gross weight from designated conditions

and missions (for which gross weight is not fixed).
b) Maximum takeoff gross weight WM T O : maximum gross weight from designated
conditions (for which gross weight is not fixed).
c) Drive system torque limit PDSlimit : maximum torque from designated conditions
and missions (for each propulsion group; specified as power limit at reference rotor
speed).
d) Fuel tank capacity: maximum fuel weight Wfuel−cap or energy Efuel−cap from
designated missions (without auxiliary tanks).
e) Antitorque or auxiliary thrust rotor design thrust Tdesign : maximum rotor thrust
from designated conditions and missions.

Alternatively, these parameters can be fixed at input values. The design gross weight (WD ) can be fixed.
The weight empty can be fixed or scaled with WD , which is implemented by changing the contingency
weight.

A successive substitution method is used for the sizing iteration, with an input tolerance . Re-
laxation is applied to Peng or R, Tjet , Pchrg , WD , WM T O , PDSlimit , Wfuel−cap or Efuel−cap , and Tdesign .
Convergence is tested in terms of these parameters, and the aircraft weight empty WE . Two successive
substitution loops are used. The outer loop is an iteration on performance: engine power or rotor radius,
jet thrust, charger power. The inner loop is an iteration on parameters: WD , WM T O , PDSlimit , Wfuel−cap
or Efuel−cap , and Tdesign . Either loop can be absent, depending on the definition of the size task.

For each flight condition and each mission, the gross weight and useful load are specified. The
gross weight can be input, maximized, or fallout. For flight conditions, the payload or fuel weight can be
specified, and the other calculated; or both payload and fuel weight specified, with gross weight fallout.
For missions, the payload or fuel weight can be specified, the other fallout, and then time or distance of
mission segments adjusted; or fuel weight calculated from mission, and payload fallout; or both payload
and fuel weight specified (or payload specified and fuel weight calculated from mission), with gross
weight fallout. For each flight condition and mission segment, the following checks are performed:
Tasks 19

a) The power required does not exceed the power available: PreqP G ≤ (1 + )PavP G
(for each propulsion group).
b) The torque required does not exceed the drive system limit: PreqP G /Ω ≤
(1 + )PDSlimit /Ωprim (for each propulsion group). Rotor shaft torque and engine
shaft torque are also checked.
c) The jet thrust required does not exceed the thrust available: TreqJG ≤ (1 + )TavJG
(for each jet group).
d) The charger power required does not exceed the power available: PreqCG ≤
(1 + )PavCG (for each charge group).
e) The fuel weight does not exceed the fuel capacity: Wfuel ≤ (1 + )(Wfuel−cap +

Nauxtank Waux−cap ) (including auxiliary tanks).

These checks are performed using an input tolerance .

Sizing flight conditions typically include takeoff (hover or specified vertical rate of climb), one-
engine inoperative, cruise or dash, perhaps transmission, and perhaps mission midpoint hover. Sizing
missions typically include a design mission and a mission to determine fuel tank capacity.

3-1.2 Component Sizing

3-1.2.1 Propulsion System

The engine size is described by the power Peng , which is the sea-level static power available per
engine at a specified takeoff rating. The number of engines Neng is specified for each engine group.
If the sizing task determines the engine power for a propulsion group, the power Peng of at least one

engine group is found (including the first engine group). The total power required is PP G = r Neng Peng ,

where r = max(PreqP G /PavP G ). The sized power is Psized = PP G − fixed Neng Peng , where the sum is
over the engine groups for which the power is fixed. Then the sized engine power is Peng = fn Psized /Neng

for the n-th engine group (with fn an input ratio and f1 = n=1,sized fn for the first group).
The jet size is described by the thrust Tjet , which is the sea-level static thrust available per jet at
a specified takeoff rating. The number of jets Njet is specified for each jet group. If the sizing task
determines the jet thrust for a jet group, the thrust Tjet is scaled by the factor r = max(TreqJG /TavJG ).
The charger size is described by the power Pchrg , which is the sea-level static power available. The
number of systems Nchrg is specified for each charge group. If the sizing task determines the charger
power for a charge group, the power Pchrg is scaled by the factor r = max(PreqCG /PavCG ).

3-1.2.2 Main Rotor

The main-rotor size is defined by the radius R or disk loading W/A, thrust-weighted solidity σ ,
2
hover tip speed Vtip , and blade loading CW /σ = W/ρAVtip σ . With more than one main-rotor, the disk
loading and blade loading are obtained from an input fraction of design gross weight, W = fW WD . The
air density ρ for CW /σ is obtained from a specified takeoff condition.
If the rotor radius is fixed for the sizing task, three of (R or W/A), CW /σ , Vtip , and σ are input, and
the other parameters are derived. Optionally the radius can be calculated from a specified ratio to the
radius of another rotor.
If the sizing task determines the rotor radius (R and W/A), then two of CW /σ , Vtip , and σ are input,
20 Tasks

and the other parameter is derived. The radius can be sized for just a subset of the rotors, with fixed
radius for the others. The radii of all sized rotors are changed by the same factor.

3-1.2.3 Antitorque or Auxiliary Thrust Rotor

For antitorque and auxiliary thrust rotors, three of (R or W/A), CW /σ , Vtip , and σ are input, and the
other parameters are derived. Optionally the radius can be calculated from a specified ratio to the radius
of another rotor. Optionally the radius can be scaled with the main-rotor radius. The disk loading and
blade loading are based on fT Tdesign , where fT is an input factor and Tdesign is the maximum thrust from
designated design conditions and missions.

3-1.2.4 Wing
The wing size is defined by the wing area S or wing loading W/S , span (perhaps calculated from
other geometry), chord, and aspect ratio. With more than one wing, the wing loading is obtained from
an input fraction of design gross weight, W = fW WD .
Two of the following parameters are input: area (or wing loading), span, chord, and aspect ratio;
the other parameters are derived. Optionally the span can be calculated from the rotor radius, fuselage
width, and clearance (typically used for tiltrotors). Optionally the span can be calculated from a specified
ratio to the span of another wing.

3-1.2.5 Fuel Tank

The fuel tank capacity Wfuel−cap (maximum usable fuel weight) or Efuel−cap (maximum usable fuel
energy) is determined from designated sizing missions. The maximum mission fuel required, Wfuel−miss
or Efuel−miss (excluding reserves and any fuel in auxiliary tanks), gives
Wfuel−cap = max(ffuel−cap Wfuel−miss , Wfuel−miss + Wreserve )
Efuel−cap = max(ffuel−cap Efuel−miss , Efuel−miss + Ereserve )
or
Wfuel−cap = dfuel−cap + ffuel−cap (Wfuel−miss + Wreserve )
Efuel−cap = dfuel−cap + ffuel−cap (Efuel−miss + Ereserve )
where ffuel−cap ≥ 1 is an input factor. Alternatively, the fuel tank capacity Wfuel−cap or Efuel−cap can
be input. Optionally the maximum mission battery discharge power gives Pcap , from which Efuel−cap =
max(Efuel−cap , (etank /πtank )Pcap ) (MJ from kW).

3-1.2.6 Weights
The weight empty WE can be calculated, or input, or scaled with design gross weight: WE =
dW E + fW E WD . Fixed or scaled weight empty is implemented by adjusting the contingency weight.

The structural design gross weight WSD and maximum takeoff weight WM T O can be input, or
specified as an increment d plus a fraction f of a weight W :
dSDGW + fSDGW WD
WSD = dSDGW + fSDGW W = dSDGW + fSDGW (WD − Wfuel + ffuel Wfuel−cap )
dSDGW + fSDGW WM T O
dW M T O + fW M T O WD
WM T O = dW M T O + fW M T O W = dW M T O + fW M T O (WD − Wfuel + Wfuel−cap )
dW M T O + fW M T O WSD
Tasks 21

This convention allows the weights to be input directly (f = 0), or scaled with a design weight. For
WSD , W is the design gross weight WD , or WD adjusted for a specified fuel state (input fraction of fuel
capacity), or the maximum takeoff weight WM T O . Alternatively, WSD can be calculated as the gross
weight at a designated sizing flight condition. For WM T O , W is the design gross weight WD , or WD
adjusted for maximum fuel capacity, or the structural design gross weight WSD . Alternatively, WM T O
can be calculated as the maximum gross weight possible at a designated sizing flight condition.

3-1.2.7 Drive System Limit

The drive system limit is defined as a power limit, PDSlimit . The limit is properly a torque limit,
QDSlimit = PDSlimit /Ωref , but is expressed as a power limit for clarity. The drive system limit can be
specified as follows (with flimit an input factor):

a) Input PDSlimit .

b) From the engine takeoff power limit, PDSlimit = flimit Neng Peng (summed over
all engine groups).
c) From the power available at the transmission sizing conditions and missions,

PDSlimit = flimit (Ωref /Ωprim ) Neng Pav (largest of all conditions and segments).
d) From the power required at the transmission sizing conditions and missions,

PDSlimit = flimit (Ωref /Ωprim ) Neng Preq (largest of all conditions and segments).

The drive system limit is a limit on the entire propulsion system. To account for differences in the
distribution of power through the drive system, limits are also used for the torque of each rotor shaft
(PRSlimit ) and of each engine group (PESlimit ). The engine shaft limit is calculated as for the drive system
limit, without the sum over engine groups. The rotor shaft limit is either input or calculated from the
rotor power required at the transmission sizing flight conditions. The power limit is associated with a
reference rotational speed, and when applied the limit is scaled with the rotational speed of the flight
state. The rotation speed for the drive system limit PDSlimit is the hover speed of the primary rotor of
the propulsion group (for the first drive state). The rotation speed for the engine shaft limit PESlimit
is the corresponding engine turbine speed. The rotation speed for the rotor shaft limit PRSlimit is the
corresponding speed of that rotor.

The drive system limits can be specified for several levels, analogous to engine ratings. The limit
PDSlimit is associated with the maximum continuous rating (MCQ or MCP). An alternate rating changes
the torque limit by the factor x. Typically x > 1 for ratings associated with short duration operation.
The torque limit is calculated from Qlimit = Q/x for the flight condition or mission segment. The torque
limit is applied as Q = xQlimit .

3–2 Mission Analysis

For the mission analysis, the fuel weight or payload weight is calculated. Power required, torque
(drive system, engine shaft, and rotor shaft), and fuel weight are then verified to be within limits.
Missions can be fixed or adjustable.

3–3 Flight Performance Analysis

For each performance flight condition, the power required is calculated or maximum gross weight
is calculated. Power required, torque (drive system, engine shaft, and rotor shaft), and fuel weight are
22 Tasks

then verified to be within limits.

3–4 Maps

3-4.1 Engine Performance

The engine performance can be calculated for a specified range of power, altitude, and speed.

3-4.2 Airframe Aerodynamics

The airframe aerodynamic loads can be calculated for a specified range of angle of attack, sideslip
angle, and control angles. The aerodynamic analysis evaluates the component lift, drag, and moments
F
given the velocity. The aircraft velocity is vAC = C F A (v 0 0)T ; interference velocity from the rotors is
not considered. From the angle of attack α and sideslip angle β , the transformation from wind axes to
airframe axes is C F A = Yα Z−β (optionally C F A = Z−β Yα can be used, for better behavior in sideward
flight). The loads are summed in the airframe axes (with and without tail loads), and then the wind axis
loads are: ⎛ ⎞ ⎛ ⎞
−D Mx
F A = ⎝ Y ⎠ = C AF F F M A = ⎝ My ⎠ = C AF M F
−L Mz
The center of action for the total loads is the fuselage location zfuse . The ratio of the loads to dynamic
pressure is required, so a nominal velocity v = 100 (ft/sec or m/sec) and sea level standard density are
used.
 Chapter 4

Operation

4–1 Flight Condition

Flight conditions are specified for the sizing task and for the flight performance analysis. For each
condition, a flight state is also defined. For the sizing task, certain flight conditions are designated for
engine sizing, design gross weight calculations, transmission sizing, maximum takeoff weight calcula-
tions, or rotor thrust sizing. The flight condition parameters include gross weight and useful load. The
gross weight can be specified as follows, consistent with the sizing method:
a) Design gross weight, WD (calculated or input).
b) Structural design gross weight, WSD , or maximum takeoff weight, WM T O (which
may depend on WD ).
c) Function of WD : W = d + f WD (with d an input weight and f an input factor).
d) Function of WSD (W = d + f WSD ); or function of WM T O (W = d + f WM T O ).
e) Input W .
f) Gross weight from specified mission segment or flight condition; or function of
source (W = d + f Wsource ).
g) Gross weight maximized, such that power required equals specified power:
PreqP G = f PavP G + d, with d an input power and f an input factor; in general,
min((f PavP G + d) − PreqP G ) = 0, minimum over all propulsion groups; default
d = 0 and f = 1 gives zero power margin, min(PavP G − PreqP G ) = 0.
h) Gross weight maximized, such that thrust required equals specified thrust:
TreqJG = f TavJG + d, with d an input thrust and f an input factor; in general,
min((f TavJG + d) − TreqJG ) = 0, minimum over all jet groups; default d = 0 and
f = 1 gives zero thrust margin, min(TavJG − TreqJG ) = 0.
i) Gross weight maximized, such that transmission torque equals limit: zero torque
margin, min(Plimit − Preq ) = 0 (mininum over all propulsion groups, engine groups,
and rotors).
j) Gross weight maximized, such that power required equals specified power, or
thrust required equals specified thrust, or transmission torque equals limit (most
restrictive).
k) Gross weight fallout from input payload and fuel weights: WG = WO + Wpay +
Wfuel .

Only the last five options are available for WD design conditions in the sizing task. The gross weight can
be obtained from a mission segment only for the sizing task. Optionally the altitude can be obtained from
the specified mission segment or flight condition. The secant method or the method of false position is
used to solve for the maximum gross weight. A tolerance  and a perturbation Δ are specified.
24 Operation

The useful load can be specified as follows, consistent with the sizing method and the gross weight
specification.

a) Input payload weight Wpay , fuel weight fallout: Wfuel = WG − WO − Wpay .

b) Input fuel weight Wfuel , payload weight fallout: Wpay = WG − WO − Wfuel .
c) Input payload and fuel weights, gross weight fallout (must match gross weight
option): WG = WO + Wpay + Wfuel .

The input fuel weight is Wfuel = min(dfuel + ffuel Wfuel−cap , Wfuel−cap ) + Nauxtank Waux−cap . For fallout
fuel weight, Nauxtank is changed (optionally only increased). If the auxiliary tank weight is greater than
the increment in fuel weight needed, then the fallout fuel weight Wfuel = WG − WO − Wpay can not be
achieved; in such a case, the fuel weight is capped at the maximum fuel capacity and the payload weight
changed instead. The fixed useful load can have increments, including crew weight increment; equipment
weight increment; and installed folding, wing, wing extension, and other kits. These increments are
reflected in the fallout weight. If the motive device burns energy not weight, then the fuel weight is zero

and the input fuel energy is Efuel = min(dfuel + ffuel Efuel−cap , Efuel−cap ) + Nauxtank Eaux−cap .

4–2 Mission

Missions are defined for the sizing task and for the mission performance analysis. A mission
consists of a specified number of mission segments. A flight state is defined for each mission segment.
For the sizing task, certain missions are designated for engine sizing, design gross weight calculations,
transmission sizing, or fuel tank sizing. The mission parameters include mission takeoff gross weight
and useful load. The gross weight can be specified as follows, consistent with the sizing method:

a) Design gross weight, WD (calculated or input).

b) Structural design gross weight, WSD , or maximum takeoff weight, WM T O (which
may depend on WD ).
c) Function of WD : W = d + f WD (with d an input weight and f an input factor).
d) Function of WSD , W = d + f WSD ; or function of WM T O , W = d + f WM T O .
e) Input W .
f) Gross weight maximized at specified mission segments, such that power required
equals specified power: PreqP G = f PavP G + d, with d an input power and f an input
factor; in general, min((f PavP G + d) − PreqP G ) = 0, minimum over all propulsion
groups; default d = 0 and f = 1 gives zero power margin, min(PavP G − PreqP G ) = 0.
g) Gross weight maximized at specified mission segments, such that thrust required
equals specified thrust: TreqJG = f TavJG + d, with d an input power and f an input
factor; in general, min((f TavJG + d) − TreqJG ) = 0, minimum over all jet groups;
default d = 0 and f = 1 gives zero thrust margin, min(TavJG − TreqJG ) = 0.
h) Gross weight maximized at specified mission segments, such that transmission
torque equals limit: zero torque margin, min(Plimit − Preq ) = 0 (mininum over all
propulsion groups, engine groups, and rotors).
i) Gross weight maximized at specified mission segments, such that power required
equals specified power, or thrust required equals specified thrust, or transmission
torque equals limit (most restrictive).
j) Gross weight fallout from input initial payload and fuel weights: WG = WO +
Operation 25

Wpay + Wfuel .
k) Gross weight fallout from input initial payload weight and calculated mission
fuel weight: WG = WO + Wpay + Wfuel .

If maximum gross weight is specified for more than one mission segment, then the minimum takeoff
gross weight increment is used; so the power or torque margin is zero for the critical segment and positive
for other designated segments. Only the last six options are available for WD design conditions in the
sizing task. The secant method or the method of false position is used to solve for the maximum gross
weight. A tolerance  and a perturbation Δ are specified.

The useful load can be specified as follows, consistent with the sizing method and the gross weight
specification:

a) Input initial payload weight Wpay , fuel weight fallout: Wfuel = WG − WO − Wpay .
b) Input fuel weight Wfuel , initial payload weight fallout: Wpay = WG − WO − Wfuel .
c) Calculated mission fuel weight, initial payload weight fallout: Wpay = WG −
WO − Wfuel .
d) Input payload and fuel weights, takeoff gross weight fallout (must match gross
weight option): WG = WO + Wpay + Wfuel .
e) Input payload weight and calculated mission fuel weight, takeoff gross weight
fallout (must match gross weight option): WG = WO + Wpay + Wfuel .

The input fuel weight is Wfuel = min(dfuel + ffuel Wfuel−cap , Wfuel−cap ) + Nauxtank Waux−cap ; if the
fuel weight is fallout, then this is the initial value for the mission iteration. If the fuel weight is not
calculated from the mission, then the mission is changed. The fixed useful load can have increments,
including installed folding kits; other increments are specified for individual mission segments. If
the motive device burns energy not weight, then the fuel weight is zero and the input fuel energy is

Efuel = min(dfuel + ffuel Efuel−cap , Efuel−cap ) + Nauxtank Eaux−cap .

The takeoff gross weight is evaluated at the start of the mission, perhaps maximized for zero power
margin at a specified mission segment (either takeoff conditions or midpoint). Then the aircraft is
flown for all segments. For calculated mission fuel weight, the fuel weight at takeoff is set equal to the
fuel required for the mission (burned plus reserve). For specified takeoff fuel weight with adjustable
segments, the mission time or distance is adjusted so the fuel required for the mission (burned plus
reserve) equals the takeoff fuel weight. The mission iteration is thus on mission fuel weight or energy.
Range credit segments (defined below) can also require an iteration. A successive substitution method
is used if an iteration is required, with a tolerance  specified. The iteration to maximize takeoff gross
weight could be an outer loop around the mission iteration, but instead it is executed as part of the mission
iteration. At the specified mission segment, the gross weight is maximized for zero power margin, and
the resulting gross weight increment added to the takeoff gross weight for the next mission iteration.
Thus takeoff gross weight is also a variable of the mission iteration.

Each mission consists of a specified number of mission segments. The following segment types
can be specified:

a) Taxi or warm-up (fuel burned but no distance added to range).

b) Distance: fly segment for specified distance (calculate time).
c) Time: fly segment for specified time (calculate distance).
d) Hold: fly segment for specified time (loiter, so fuel burned but no distance added
26 Operation

to range).
e) Climb: climb or descend from present altitude to next segment altitude (calculate
time and distance).
f) Spiral: climb or descend from present altitude to next segment altitude (fuel
burned but no distance added to range).
g) Fuel: use or replace specified fuel amount (calculate time and distance).
h) Burn: use or replace specified fuel amount (calculate time but no distance added
to range).
For each mission segment a payload weight can be specified; or a payload weight change can be specified,
as an increment from the initial payload or as a fraction of the initial payload. If the payload is calculated
from the number of passengers, then for each mission segment a change in the number of passengers
can be specified.
The number of auxiliary fuel tanks can change with each mission segment: Nauxtank is changed
based on the fuel weight (optionally only increased relative to the input number at takeoff, optionally fixed
during mission). For input fuel weight, Nauxtank is specified at takeoff. For fallout fuel weight, the takeoff
fuel weight is changed for the auxiliary fuel tank weight given Nauxtank (fixed WG −Wpay = WO +Wfuel ).
If the auxiliary tank weight is greater than the increment in fuel weight needed, then the fallout fuel
weight Wfuel = WG − WO − Wpay can not be achieved; in such a case, the fuel weight is capped at the
maximum fuel capacity and the takeoff payload weight changed instead. For fuel tank design missions,
Nauxtank and fuel tank capacity is determined from Wfuel . Optionally the aircraft can refuel (either on the
ground or in the air) at the start of a mission segment, by either filling all tanks to capacity or adding a
specified fuel weight. Optionally fuel can be dropped at the start of a mission segment. The fixed useful
load can have changes, including crew weight increment, equipment weight increment, and installed
wing extension and other kits.
For calculation of the time or distance in a mission segment, a headwind or tailwind can be specified.
The wind velocity is a linear function of altitude h: Vw = ±(max(0, dwind + fwind h)), with the plus sign
for a headwind and the minus sign for a tailwind. For example, California-to-Hawaii 85th percentile
winter quartile headwind profile is Vw = 9.59 + 0.00149h (with altitude h in ft).
Mission fuel reserves can be specified in several ways for each mission. Fuel reserves can be
defined in terms of specific mission segments, for example 200 miles plus 20 minutes at speed for best
endurance. Fuel reserves can be an input fraction of the fuel burned by all (except reserve) mission
segments, so Wfuel = (1 + fres )Wburn or Efuel = (1 + fres )Eburn . Fuel reserves can be an input fraction
of the fuel capacity, so Wfuel = Wburn + fres Wfuel−cap or Efuel = Eburn + fres Efuel−cap . If more than
one criterion for reserve fuel is specified, the maximum reserve is used. Time and distance in reserve
segments are not included in endurance and range.
To facilitate specification of range, range calculated for a group of segments (typically climb and
descent segments) can be credited to a designated distance segment. For mission analysis, missions can
be fixed or adjustable. In an adjustable mission, the fuel is input, so the time or distance in specified
segments is adjusted based on the calculated fuel burned. If more than one segment is adjusted, all must
be distance or all must be time or hold. Each segment can have only one special designation: reserve,
adjustable, or range credit.
A segment with a large distance, time, or altitude change can be split into several segments, for
more accurate calculation of the performance and fuel burned. The number of segments n can be input,
Operation 27

or calculated from an input increment Δ: n = [x/Δ] + 1, where the brackets indicate integer truncation,
and x is the total distance, time, or altitude change. Then the change for each split segment is Δ = x/n.
Table 4-1 summarizes the time T , distance D, and range dR calculations for each segment. The
segment fuel burned is dWburn = T ẇ, where ẇ is the fuel flow; and dEburn = T Ė , where Ė is the energy
flow. The horizontal velocity is Vh , and the vertical velocity (climb or descent) is Vc . The altitude at the
start of the segment is h, and at the end of the segment (start of next segment) hend . The wind speed is
Vw , and the ground speed is Vh − Vw . The air distance is calculated from the time and speed (D/Vh ),
without the wind speed.
To use or replace fuel (weight or energy), the increment is specified in terms of the capacity
(dWburn = dtank + ftank Wfuel−cap ) or the current fuel (dWburn = dtank + ftank Wfuel ). Alternatively, a
target fuel is specified and the increment calculated from the current fuel. The tank is charged if the
energy rate is negative. The segment time T is the minimum of dWburn /ẇ or dEburn /Ė for all fuel tank
systems.
In an adjusted mission, the distances or times are changed at the end of the mission such that
the sum of the fuel burned increments will equal the difference between takeoff fuel weight (plus any
  
added fuel) and the calculated mission fuel: dWburn = ẇ dT = ẇ dD/(Vh − Vw ) = ΔWfuel . The
increments are apportioned among the adjusted segments by the factor f , determined from the ratio of
 
the input distances or times: dD = f ΔD or dT = f ΔT . Hence ΔD = ΔWfuel / f ẇ/(Vh − Vw ) or
 
ΔT = ΔWfuel / f ẇ . The approach is similar if fuel energy is used, not weight. For a segment that
is a source of range credit, the range increment is set to zero and the distance D is added to Dother of
the destination segment. For the destination segment, the range contribution remains fixed at the input
value, but the time and hence fuel burned are calculated from (dist − Dother ). It is necessary to separately
accumulate Dother from earlier segments and Dother from later segments; Dother from later segments
are estimated initially from the last iteration. At the end of the mission, the times and fuel burned are
recalculated for all range credit destination segments.

Table 4-1. Mission segment calculations.

segment kind time T distance D range dR
taxi time 0 D
distance D/(Vh − Vw ) dist D
time time T (Vh − Vw ) D
hold time 0 D
climb (h − hend )/Vc T (Vh − Vw ) D
spiral (h − hend )/Vc 0 D
fuel dWburn /ẇ or dEburn /Ė T (Vh − Vw ) D
burn dWburn /ẇ or dEburn /Ė 0 D
range credit
source T D 0
destination D/(Vh − Vw ) dist − Dother dist
adjusted
distance T + dD/(Vh − Vw ) D + dD = D + f ΔD Dnew
time T + dT = T + f ΔT D + dT (Vh − Vw ) Dnew
hold T + dT = T + f ΔT 0 Dnew
28 Operation

The segment time, distance, and fuel burned are evaluated by integrating over the segment duration.
This integration can be performed by using the horizontal velocity, climb velocity, and fuel flow obtained
for the flight state with the gross weight and altitude at the start of the segment; or at the middle of the
segment; or the average of the segment start and segment end values (trapezoidal integration). The gross
weight at the segment middle equals the gross weight at the segment start, less half the segment fuel
burned (obtained from the previous mission iteration). The gross weight at the segment end equals the
gross weight at the segment start, less the segment fuel burned. With trapezoidal integration, for the
output the flight state is finally evaluated at the segment middle.
 
The mission endurance (block time), range, and fuel burned are E = T , R = dR, Wburn =

dWburn (sum over all non-reserve segments). The reserve fuel from mission segments is Wres =

dWburn (sum over all reserve segments). Optionally the reserve fuel is the maximum of that from
mission segments and the fraction fres Wburn , or the fraction fres Wfuel−cap . The calculated mission fuel
is then Wfuel = Wburn + Wres .
A fuel efficiency measure for the mission is the product of the payload and range, divided by the
fuel weight: e = Wpay R/Wburn (ton-nm/lb or ton-nm/kg). A productivity measure for the mission is
p = Wpay V /WO (ton-kt/lb or ton-kt/kg), where WO is the operating weight and V the block speed; or
p = Wpay V /Wburn (ton-kt/lb or ton-kt/kg). The Bréguet range equation R = RF ln(W0 /W1 ) is obtained
by integrating dR = −RF (dW/W ) for constant range factor
L/De W V /P
RF = =
sfc sfc
  √
The endurance E = EF 2 W0 /W1 − 1 is obtained by integrating dE = −EF W0 (dW/W 3/2 ) for
constant endurance factor  
L/De W/W0 W/P W/W0
EF = =
sfc V sfc
Constant RF implies operation at constant L/De = W V /P . Constant EF implies operation at constant
√ 3/2
(L/De ) W /V = W 3/2 /P (or constant CL /CD for an airplane). It follows that overall range and
endurance factors can be calculated from the mission performance:
R
RF =
ln W0 /W1
E
EF =  
2 W0 /W1 − 1

where W0 = Wto is the takeoff weight, and W1 = Wto − Wburn . If energy is burned, not weight, the
efficiency metrics can be based on the equivalent fuel burned Eburn /eref and the range factor can be
based on the equivalent specific fuel consumption. But since the weight does not change as energy is
used, the equation dR = −RF (dW/W ) integrates to R = RF (Wburn /W ).

4–3 Takeoff Distance

The takeoff distance can be calculated, either as ground run plus climb to clear an obstacle or
accelerate-stop distance in case of engine failure. The obstacle height ho is typically 35 ft for commercial
transport aircraft, or 50 ft for military aircraft and general aviation. This calculation allows determination
of the balanced field length: engine failure at critical speed, such that the distance to clear the obstacle
Operation 29

horiz
ontal obstacle
γ G ground slope hO
ground
x inertial
z inertial

b
im
cl
R TR
γ
γ relative ground
n
sitio
ground run rotation tran h TR
distance sG sR s TR s CL
ground V=0 VEF V1 VR VLO VTR VCL
or climb
sta

en

de

ro

lif

tra

cl

ob
speed

im
ta

to
gi

ci

sta
ns
rt

tio
sio

ff
ne

b
iti

cl
n

on
n

e
fa
ilu
re

accelerate stop
distance sA sS
ground V=0 VEF V1 V=0
speed

Figure 4-1. Takeoff distance and accelerate-stop distance elements.

equals the distance to stop. Landing and VTOL takeoff calculations are not implemented, as these are
best solved as an optimal control problem.

The takeoff distance consists of a ground run, from zero ground speed to liftoff speed VLO , perhaps
including engine failure at speed VEF ; then rotation, transition, and climb; or decelerate to stop. Figure
4-1 describes the elements of the takeoff distance and the accelerate-stop distance, with the associated
speeds. The ground is at angle γG relative to the horizontal (inertial axes), with γG positive for takeoff up
hill. The takeoff profile is defined in terms of ground speed or climb speed, input as calibrated airspeed
(CAS). The aircraft speed relative to the air is obtained from the ground speed, wind, and ground slope.
The aircraft acceleration as a function of ground speed is integrated to obtain the ground distance, as
well as the time, height, and fuel burned. Usually the speed increases from the start to liftoff (or engine
failure), but the calculated acceleration depends on the flight state specification. The analysis checks for
consistency of the input velocity and the calculated acceleration (on the ground), and for consistency of
the input height and input or calculated climb angle (during climb).

The takeoff profile consists of a set of mission segments. The first segment is at the start of the
takeoff, V = 0. Subsequent segments correspond to the ends of the integration intervals. The last
segment has the aircraft at the required obstacle height, or stopped on the ground. The mission can
consist of just one takeoff, more than one takeoff, or both takeoff and non-takeoff segments. Takeoff
segments contribute to the mission fuel burned, but do not contribute to the mission time, distance, or
range. The takeoff distance calculation is performed for a set of adjacent segments, the first segment
specified as the takeoff start, and the last segment identified as before a non-takeoff segment or before
30 Operation

another takeoff start. The takeoff distance is calculated if a liftoff segment (with VLO ) is specified;
otherwise the accelerate-stop distance is calculated. Table 4-2 summarizes the mission segments for
takeoff calculations. There can be only one liftoff, engine failure, rotation, and transition segment
(or none). The engine failure segment must occur before the liftoff segment. Rotation and transition
segments must occur after liftoff. All ground run segments must be before liftoff, and all climb segments
must be after liftoff. Takeoff segments (except start, rotation, and transition) can be split, in terms of
height for climb and in terms of velocity for other segments. Splitting the takeoff or engine failure
segment produces additional ground run segments. Separately defining multiple ground run, climb, or
brake segments allows configuration variation during the takeoff.

Each takeoff segment requires that the flight state specify the appropriate configuration, trim option,
and maximum effort. In particular, the number of inoperative engines for a segment is part of the flight
state specification, regardless of whether or not an engine failure segment is defined. The engine failure
segment (if present) serves to implement a delay in decision after failure: for a time t1 after engine
failure, the engine rating, power fraction, and friction of the engine failure segment are used (so the
engine failure segment corresponds to conditions before failure). The number of inoperative engines
specified must be consistent with the presence of the engine failure segment. The takeoff is assumed
to occur at fixed altitude (so the maximum-effort variable can not be altitude). The flight state velocity
specification is superseded by the ground or climb speed input for the takeoff segment. The flight state
specification of height above ground level is superseded by the height input for the takeoff segment.

The ground distance, time, height, and fuel burned are calculated for each takeoff segment. The
takeoff distance or accelerate-stop distance is the sum of the ground distance of all segments. Takeoff
segments do not contribute to mission time, distance, or range.

Table 4-2. Mission segments for takeoff calculation.

takeoff distance accelerate-stop distance
start V =0 start V =0
ground run V ground run V
engine failure VEF engine failure VEF
ground run V brake V
liftoff VLO brake V =0
rotation VR
transition VT R
climb, to h VCL
climb, to ho VCL

4-3.1 Ground Run

The takeoff starts at zero ground speed and accelerates to liftoff ground speed VLO (input as CAS).
Possibly an engine failure speed VEF < VLO is specified. Start, liftoff, and engine failure segments
designate events, but otherwise are analyzed as ground run segments. The decision speed V1 is t1
seconds after engine failure (typically t1 = 1 to 2 sec). Up to t1 after engine failure, conditions of the
engine failure segment are used (so the engine failure segment corresponds to conditions before failure).
The aircraft acceleration is obtained from the thrust minus drag (T − D in airplane notation), plus a
Operation 31

friction force proportional to the weight on wheels (W − L in airplane notation):

 
M a = T − D − μ(W − L) = Fx − μ Fz

from the force components in ground axes (rotated by the ground slope angle γG from inertial axes).
Table 4-3 gives typical values of the friction coefficient μ. The velocity of the aircraft relative to the
air is obtained from the ground velocity V , wind velocity Vw (assumed parallel to the ground here),
and the ground slope: Vh = (V + Vw ) cos γG and Vc = (V + Vw ) sin γG . The takeoff configuration is
specified, including atmosphere, in ground effect, gear down, power rating, nacelle tilt, flap setting,
and number of inoperative engines. An appropriate trim option is specified, typically fixed attitude
with longitudinal force trimmed using collective, for a given longitudinal acceleration. Perhaps the net
aircraft yaw moment is trimmed with pedal. The maximum-effort condition is specified: maximum
longitudinal acceleration (ground axes) for zero power margin. The aircraft acceleration as a function
of ground speed is integrated to obtain the segment time, ground distance, height, and fuel burned:
  
dv  1  1
dt 1 
tG = dt = =dv = + (v2 − v1 ) = Δt
dv
a seg
2 a2 a1 seg
   
dt v dv d(v 2 )  1  1 1  2   v2 + v1 
sG = v dt = v dv = = = + (v2 − v12 ) = Δt
dv a 2a seg
2 2a2 2a1 seg
2

hG = 0
   ẇf 2 + ẇf 1 
wG = ẇf dt = Δt
seg
2

Trapezoidal integration is used; each segment corresponds to the end of an integration integral.

Table 4-3. Typical friction coefficient μ.

surface rolling braking
dry and hard 0.03–0.05 0.30–0.50
grass 0.08 0.20
ice 0.02 0.06–0.10

4-3.2 Brake
After engine failure, the aircraft can decelerate to a stop. The operating engines are at idle. Reverse
thrust is not permitted for the accelerate-stop distance calculation. The braking configuration is specified.
Typically no trim option is executed; rather the aircraft has fixed attitude with controls for zero rotor
thrust (such as zero collective and pedal). The aircraft acceleration as a function of ground speed is
integrated, as for ground run.

4-3.3 Rotation
Rotation occurs at speed VR ; usually VR = VLO is used. The duration tR is specified, then sR = VR tR ,
hR = 0, and wR = ẇf tR are the ground distance, height, and fuel burned. Typically tR = 1 to 3 sec.
32 Operation

4-3.4 Transition
Transition from liftoff to climb is modeled as a constant load factor pull-up to the specified climb
angle γ , at speed VT R . Usually VT R = VLO is used, and typically nT R ∼ = 1.2. From the load factor
nT R = 1+VT2R /gRT R , the flight path radius is RT R = VT2R /(g(nT R −1)) and the pitch rate is θ̇ = VT R /RT R .
Then
RT R VT R
tT R = γ/θ̇ = γ =γ
VT R g(nT R − 1)
sT R = RT R sin γ
hT R = RT R (1 − cos γ)
wT R = ẇf tT R
are the time, ground distance, height, and fuel burned.

4-3.5 Climb
Climb occurs at an angle γ relative to the ground and air speed VCL , from the transition height
hT R to the obstacle height ho (perhaps in several climb segments). The climb configuration is specified,
including atmosphere, in ground effect, gear down or retracted, power rating, nacelle tilt, flap setting,
and number of inoperative engines. An appropriate trim option is specified, typically aircraft force and
moment trimmed with attitude and controls. The climb angle and air speed can be fixed or a maximum-
effort condition can be specified. The maximum-effort options are fixed air speed and maximum rate of
climb for zero power margin; or airspeed for best climb rate or best climb angle with maximum rate of
climb for zero power margin. Not implemented is a maximum-effort calculation of maximum flight path
acceleration for zero power margin, for specified climb angle; this calculation would require integration
of the acceleration as a function of flight speed. For the climb segment, the input VCL is the magnitude
of the aircraft velocity relative to the air, and the climb angle relative to the horizon is θV = γ + γG .
Hence from the maximum-effort calculation, the climb angle relative to the ground is γ = θV − γG and
the ground speed is Vground = VCL cos γ − Vw (from the wind speed Vw ). Then

tCL = sCL /Vground

sCL = (h − hlast )/ tan γ
hCL = h
wCL = ẇf tCL

are the time, ground distance, height, and fuel burned.

4–4 Flight State

A flight state is defined for each flight condition (sizing task design conditions and flight performance
analysis), and for each mission segment. The following parameters are required:
a) Speed: flight speed and vertical rate of climb, with the following options:
1) Specify horizontal speed (or forward speed or velocity magnitude), rate
of climb (or climb angle), and sideslip angle.
2) Hover or vertical flight (input vertical rate of climb; climb angle 0 or
±90 deg).
3) Left or right sideward flight (input velocity and rate of climb; sideslip
Operation 33

angle ±90 deg).

4) Rearward flight (input velocity and rate of climb; sideslip angle 180
deg).

b) Aircraft motion:

1) Pitch and roll angles (Aircraft values or flight state input; initial values
for trim variables, fixed otherwise).
2) Turn, pull-up, or linear acceleration.

c) Altitude: For mission segment, optionally input, or from last mission segment;
climb segment end altitude from next segment.

d) Atmosphere:

1) Standard day, polar day, tropical day, or hot day at specified altitude.
2) Standard day, polar day, or tropical day plus temperature increment.
3) Standard day, polar day, or tropical day and specified temperature.
4) Input density and temperature.
5) Input density, speed of sound, and viscosity.

e) Height of landing gear above ground level. Landing gear state (extended or
retracted).
f) Aircraft control state: input, or conversion schedule.
g) Aircraft control values (Aircraft values or flight state input; initial values for trim
variables, fixed otherwise).
h) Aircraft center-of-gravity position (increment or input value).

For each propulsion group, the following parameters are required:

i) Drive system state.

j) Rotor tip speed for primary rotor:

1) Input.
2) Reference.
3) Conversion schedule or function speed.
4) Default for hover, cruise, maneuver, one engine inoperative (OEI), or
transmission sizing condition.
5) From input rotor advance ratio μ, or rotor advancing tip Mach number
Mat .

For each engine group (which is associated with a propulsion group):

k) Number of inoperative engines.

l) Infrared suppressor state: off (hot exhaust) or on (suppressed exhaust).
m) Engine rating, fraction of rated engine power available, and drive system rating.

For each jet group:

n) Number of inoperative jets.

o) Jet rating, fraction of rated thrust available.
34 Operation

For each charge group:

p) Number of inoperative chargers.

q) Charger rating, fraction of rated power available.

Aircraft and rotor performance parameters for each flight state:

r) Aircraft drag: forward flight drag increment, accounting for payload aerodynam-
ics.
s) Rotor performance: induced power factor κ and profile power mean cd .

The aircraft trim state and trim targets are also specified.

The aircraft performance can be analyzed for the specified state, or a maximum-effort performance
can be identified. For the maximum effort, a quantity and variable are specified. The available maximum-
effort quantities include:

a) Best endurance: maximum 1/ẇ.

b) Best range: 99% maximum V /ẇ (high side), or low side, or 100%.
c) Best climb or descent rate: maximum Vz or 1/P .
d) Best climb or descent angle: maximum Vz /Vh or V /P .
e) Ceiling: maximum altitude.
f) Power limit: zero power margin, min(PavP G − PreqP G ) = 0 (minimum over all
propulsion groups).
g) Torque limit: zero torque margin, min(Qlimit − Qreq ) = 0 (minimum over all
propulsion groups, engine groups, and rotors; Qlimit expressed as power at reference
rotation speed).
j) Jet thrust limit: zero thrust margin, min(TavJG − TreqJG ) = 0 (minimum over all
jet groups).
i) Power limit or torque limit or thrust limit: most restrictive.
j) Battery limit: zero power margin, min(Pmax − |Ėbatt |) = 0 (minimum over all fuel
tanks).
k) Rotor stall: zero rotor thrust margin, (CT /σ)max − |CT /σ| = 0 (for designated
rotor, steady or transient or equation limit).
l) Wing stall: zero wing lift margin, CLmax − CL = 0 (for designated wing).

Here ẇ is the aircraft fuel flow (or equivalent fuel flow Ė/eref ), and P is the aircraft power. The available
maximum-effort variables include:

a) Horizontal velocity Vh or vertical rate of climb Vz (times an input factor).

b) Aircraft sideslip angle.
c) Altitude.
d) Aircraft angular rate, θ̇ (pull-up) or ψ̇ (turn).
e) Aircraft linear acceleration (airframe, inertial, or ground axes).
f) Aircraft controls.
g) Aircraft orientation, θ (pitch) or φ (roll).
h) Propulsion group tip speed or engine speed.

If the variable is velocity, first the velocity is found for the specified maximum effort; then the performance
is evaluated at that velocity times an input factor. For endurance, range, or climb, the slope of the quantity
Operation 35

to be maximized must be zero; hence in all cases the target is zero. The slope of the quantity is evaluated
by first-order backward difference. For the range, first the variable is found such that V /ẇ is maximized
(slope zero), then the variable is found such that V /ẇ equals 99% of that maximum. Two maximum-
effort quantity/variable pairs can be specified, and solved in nested iterations. The secant method or
the method of false position is used to solve for the maximum effort. The task of finding maximum
endurance, range, or climb is usually solved using the golden-section or curve-fit method. A tolerance
 and a perturbation Δ are specified.

Given the gross weight and useful load (from the flight condition or mission specification), the
performance is calculated for this flight state. The calculated state information includes weight, speed
and velocity orientation (climb and sideslip), aircraft Euler angles, rotor tip speeds, and aircraft controls.
A number of performance metrics are calculated for each flight state. The aircraft effective drag is
De = P/V , hence the effective lift-to-drag ratio is L/De = W V /P . For these metrics, the aircraft power
is the sum of the engine group power and jet group propulsive power: P = Preq + V Tjet . The specific
range is the ratio of the speed to the fuel flow: V /ẇ (nm/lb or nm/kg). From the Bréguet range equation,
it follows that the range for which the fuel equals 1% of the gross weight is
 
L/De 1
R1%GW = ln
sfc .99

A fuel efficiency measure is the product of the payload and specific range: e = Wpay (V /ẇ) (ton-
nm/lb or ton-nm/kg). A productivity measure is p = Wpay V /WO (ton-kt/lb or ton-kt/kg), where WO is
the operating weight. If fuel energy is used, not weight, these performance metrics are based on the
equivalent ẇ and equivalent sfc.
The aircraft weight statement defines the fixed useful load and operating weight for the design
configuration. For each flight state, the fixed useful load may be different from the design configuration,
because of changes in auxiliary fuel tank weight or kit weights or increments in crew or equipment
weights. Thus the fixed useful load weight is calculated for the flight state; and from it the useful load
weight and operating weight. The gross weight, payload weight, and usable fuel weight (in standard
and auxiliary tanks) completes the weight information for the flight state.

4–5 Environment and Atmosphere

The aerodynamic environment is defined by the speed of sound cs , density ρ, and kinematic viscosity
ν = μ/ρ of the air (or other fluid). These quantities can be obtained from the standard day (International
Standard Atmosphere), or input directly. Polar day, tropical day, and hot day atmospheres can also be
used. The following options are implemented:
a) Input the altitude hgeom and a temperature increment ΔT . Calculate the temper-
ature from altitude and pressure from temperature for the standard day (or polar
day, tropical day, hot day), add ΔT , and then calculate the density from the equa-
tion of state for a perfect gas. Calculate the speed of sound and viscosity from the
temperature.
b) Input the pressure altitude hgeom and the temperature τ (◦ F or ◦ C). Calculate the
pressure (from temperature vs. altitude) for the standard day (or polar day, tropical
day, hot day), and then calculate the density from the equation of state for a perfect
gas. Calculate the speed of sound and viscosity from the temperature.
36 Operation

c) Input the density ρ and the temperature τ (◦ F or ◦ C). Calculate the speed of sound
and viscosity from the temperature.
d) Input the density ρ, sound speed cs , and viscosity μ. Calculate the temperature
from the sound speed.
e) For a fluid that is not air on earth, input the density ρ, sound speed cs , and viscosity
μ.

Here hgeom is the geometric altitude above mean sea level. The sources of the atmosphere descriptions
are references 1–6.
The gravitational acceleration g can have the standard value or an input value.
The International Standard Atmosphere (ISA) is a model for the variation with altitude of pressure,
temperature, density, and viscosity, published as International Standard ISO 2533 by the International
Organization for Standardization (ISO) (ref. 1). The ISA is intended for use in calculations and design
of flying vehicles, to present the test results of flying vehicles and their components under identical
conditions, and to allow unification in the field of development and calibration of instruments. The ISA
is defined up to 80 km geopotential altitude and is identical to the ICAO Standard Atmosphere up to 32
km.
Dry air is modeled in the ISA as a perfect gas with a mean molecular weight, and hence a gas
constant R, defined by adopted values for sea level pressure, temperature, and density (p0 , T0 , ρ0 ). The
speed of sound at sea level cs0 is defined by an adopted value for the ratio of specific heats γ . The variation
of temperature with geopotential altitude is defined by adopted values for vertical temperature gradients
(lapse rates, Lb ) and altitudes (hb ). The variation of pressure with geopotential altitude is further defined
by an adopted value for the standard acceleration of free fall (g ). The variation of dynamic viscosity μ
with temperature is defined by adopted values for Sutherland’s empirical coefficients β and S .
The required parameters are given in table 4-4, including the acceleration produced by gravity, g .
The temperature T is in ◦ K, while τ is ◦ C (perhaps input as ◦ F); T = Tzero + τ . The gas constant is
R = p0 /ρ0 T0 , and μ0 is actually obtained from S and β . The standard atmosphere is defined in SI units.
Although table 4-4 gives values in both SI and English units, all the calculations for the aerodynamic
environment are performed in SI units. As required, the results are converted to English units using the
exact conversion factors for length and force.
The ISA consists of a series of altitude ranges with constant lapse rate Lb (linear temperature change
with altitude). Thus at altitude hg , the standard day temperature is

Tstd = Tb + Lb (hg − hb )

for hg > hb . The altitude ranges and lapse rates are given in table 4-5. Note that h0 is sea level, and h1 is
the boundary between the troposphere and the stratosphere. This altitude hg is the geopotential height,
calculated assuming constant acceleration due to gravity. The geometric height h is calculated using an
inverse square law for gravity. Hence hg = rh/(r + h), where r is the nominal radius of the Earth. The
standard day pressure is obtained from hydrostatic equilibrium (dp = −ρg dhg ) and the equation of state
for a perfect gas (p = ρRT , so dp/p = −(g/RT )dhg ). In isothermal regions (Lb = 0) the standard day
pressure is
pstd
= e−(g/RT )(hg −hb )
pb
Operation 37

and in gradient regions (Lb = 0)

 −g/RLb
pstd T
=
pb Tb
where pb is the pressure at hb , obtained from these equations by working up from sea level. Let T0 , p0 , ρ0 ,
cs0 , μ0 be the temperature, pressure, density, sound speed, and viscosity at sea level standard conditions.
Then the density, sound speed, and viscosity are obtained from
   −1
p T
ρ = ρ0
p0 T0
 1/2
T
cs = cs0
T0
 
(T /T0 )3/2
μ = μ0
α(T /T0 ) + 1 − α

where μ0 = βT03/2 /(T0 + S) and α = T0 /(T0 + S). For the cases using input temperature, T = Tzero + τ .
The density altitude and pressure altitude are calculated for reference. From the density and the standard
day (troposphere only), the density altitude is:
  1/(g/R|L0 |−1) 
T0 ρ
hd = 1−
|L0 | ρ0

From the pressure p = ρRT and the standard day,

  1/(g/R|L0 |)    1/(g/R|L0 |) 
T0 p T0 ρ T
hp = 1− = 1−
|L0 | p0 |L0 | ρ0 T 0

is the pressure altitude.

The polar and tropical days are atmospheric models that describe realistic profiles of extremes
of temperature and density, needed to calculate performance in near-worst-case conditions (ref. 6).
These atmospheres are hydrodynamically balanced and can be used in calculations involving engine
performance and aerodynamic characteristics, including calculations of true vertical velocity (ref. 4).
Tables of air properties for the polar and tropical days are given in MIL-STD-3013A, based on MIL-
STD-210A. The break points in the lapse rates are evident in MIL-STD-210A, in terms of degrees C
as a function of altitude in ft (geopotential altitude in MIL-C-5011B). The pressure ratio at sea level is
δ = 30.268/29.92 = 1.0116 for the polar day, and δ = 1 for the tropical day. The altitude ranges and lapse
rates are given in table 4-6 for the polar day and in table 4-7 for the tropical day. Figure 4-2 compares
the temperature profiles. The air properties for the polar and tropical days are calculated from hb , Lb ,
and Tb using the equations of hydrostatic equilibrium, as for the standard day. The primary data for the
polar and tropical days are the altitude in ft and temperature in ◦ C; hb in km and Lb are derived using
the exact conversion factor for length.
The conditions of the standard day, polar day, and tropical day are applicable to free air conditions.
Temperatures close to the surface of the earth, even at high elevations, can be considerably higher
than those for free air. The hot day is a model of ground-level atmospheric conditions, to be used for
takeoff and other ground operations at elevations up to 15,000 ft (ref. 6). The hot day properties are
statistically sampled and are not hydrodynamically balanced, hence should be used for approximately
constant altitude conditions (ref. 4).
38 Operation

The origins of the hot day are in ref. 2, which has 103◦ F for sea level, and a lapse rate −3.7◦ F per
1000 ft (geometric) to 40,000 ft. The pressure in the table is from the equations for equilibrium, which
gives the pressure altitude; these data are not used further. According to MIL-STD-210A (paragraph
3.1.1), the pressure (hence pressure altitude) as a function of altitude was obtained from statistics. Up
to 15,000 ft, the ratio of geometric altitude to pressure altitude is 1.050. The −3.7 geometric lapse rate
plus the mapping of geometric to pressure altitude gave ◦ F vs. pressure altitude, rounded to 1 decimal
place. Then ◦ C was calculated from ◦ F, rounded to 1 decimal place. The pressure ratio δ followed
from the pressure altitude. The geometric altitude (to 35,000 ft) in MIL-STD-210A was calculated from
the temperature and the −3.7 lapse rate, rounded to 100s. MIL-C-5011B has the same data as MIL-
STD-210A, but does not give geometric altitude, and truncates the table at 15,000 ft. MIL-STD-3013A
took ◦ C (already rounded to 1 decimal place) vs. pressure altitude (ft) from MIL-STD-210A as the
temperature profile up to 15,000 ft, and calculated ◦ F (2 decimal places, so no more loss of information)
from ◦ C. The geopotential altitude was calculated from ◦ C and −3.7◦ F lapse rate, rounded to 100s; but
this altitude information is not meaningful, since the atmosphere is not in equilibrium; only the pressure
altitude is used. Thus the hot day model has a sea level temperature of 39.4◦ C (102.92◦ F). Curve fitting
the data to 15,000 ft gives a lapse rate of −7.065◦ C per 1000 m = −2.1534◦ C per 1000 ft (table 4-8).
Given the pressure altitude h, the hot day temperature and pressure ratio are

Thot = Tb + Lb (h − hb )
 −g/RLb
phot L0
= 1+ h
p0 T0

using the standard day L0 = −6.5. A constant lapse rate fits the tabular data for the hot day atmosphere
to only about 0.1◦ C. Thus the tabular data can be used directly instead (table 4-9, from MIL-C-5011B),
with linear interpolation to the specified pressure altitude. MIL-STD-3013A has the same data, but only
for altitudes a multiple of 1000 ft.

4–6 References

1) International Organization for Standardization. “Standard Atmosphere.” ISO 2533-1975(E), May

1975.
2) Theiss, E.C. “Proposed Standard Cold and Hot Atmospheres for Aeronautical Design.” Wright Air
Development Center, United States Air Force, Technical Memorandum Report WCSE 141, June 1952.
3) Department of Defense Military Specification. “Climatic Extremes for Military Equipment.” MIL-
STD-210A, August 1957.
4) Department of Defense Military Specification. “Charts: Standard Aircraft Characteristics and Perfor-
mance, Piloted Aircraft (Fixed Wing). Appendix IC, Atmospheric Tables.” MIL-C-005011B(USAF),
June 1977.
5) Department of Defense Military Specification. “Climatic Information to Determine Design and Test
Requirements for Military Systems and Equipment.” MIL-STD-210C, January 1987.
6) Department of Defense Military Specification. “Glossary of Definitions, Ground Rules, and Mission
Profiles to Define Air Vehicle Performance Capability.” MIL-STD-3013A, September 2008.
Operation 39

standard day
40.
polar day
tropical day
temperature (deg C) 20. hot day

0.

-20.

-40.

-60.

-80.
0. 5. 10. 15. 20. 25. 30.

altitude (km)

Figure 4-2. Temperature as a function of altitude.

Table 4-4. Constants adopted for calculation of the ISA.

parameter SI units English units
units h m ft
◦ ◦
units τ C F
m per ft 0.3048
kg per lbm 0.45359237
T0 288.15 ◦ K 518.67 ◦ R
Tzero 273.15 ◦ K 459.67 ◦ R
p0 101325.0 N/m2 2116.22 lb/ft2
ρ0 1.225 kg/m3 0.002377 slug/ft3
cs0 340.294 m/sec 1116.45 ft/sec
μ0 1.7894E-5 kg/m-sec 3.7372E-7 slug/ft-sec
S 110.4 ◦ K
β 1.458E-6
γ 1.4
g 9.80665 m/sec2 32.17405 ft/sec2
r 6356766 m 20855531 ft
40 Operation

Table 4-5. Temperatures and vertical temperature gradients: standard day.

level base altitude hb lapse rate Lb temperature Tb
◦ ◦ ◦ ◦
km K/km K C F
troposphere –2 –6.5 301.15 28 82.4
0 troposphere 0 –6.5 288.15 15 59
1 tropopause 11 0 216.65 –56.5 –69.7
2 stratosphere 20 +1.0 216.65 –56.5 –69.7
3 stratosphere 32 +2.8 228.65 –44.5 –48.1
4 stratopause 47 0 270.65 –2.5 27.5
5 mesosphere 51 –2.8 270.65 –2.5 27.5
6 mesosphere 71 –2.0 214.65 –58.5 –73.3
7 mesopause 80 0 196.65 –76.5 –105.7

Table 4-6. Temperatures and vertical temperature gradients: polar day.

base altitude hb lapse rate Lb temperature Tb pressure ratio
◦ ◦ ◦ ◦
ft km * K/km * K C F* δ = p/p0
0 0 6.326 246.15 –27 –16.6 30.268/29.92
3111.871 0.948 –1.083 252.15 –21 –5.8
9172.604 2.796 –5.511 250.15 –23 –9.4
28224.543 8.603 –0.476 218.15 –55 –67.0
83363.393 25.409 0 210.15 –63 –81.4
* derived

Table 4-7. Temperatures and vertical temperature gradients: tropical day.

base altitude hb lapse rate Lb temperature Tb pressure ratio
◦ ◦ ◦ ◦
ft km * K/km * K C F* δ = p/p0
0 0 –6.687 305.25 32.1 89.78 1
55000 16.764 4.374 193.15 –80 –112.00
70000 21.336 2.401 213.15 –60 –76.00
100745 30.707 235.65 –37.5 –35.50
* derived

Table 4-8. Temperatures and vertical temperature gradients: hot day.

base altitude hb lapse rate Lb temperature Tb pressure ratio
◦ ◦ ◦ ◦
ft km * K/km K C F* δ = p/p0
0 0 –7.065 312.55 39.4 102.92 1
15000 4.572 280.25 * 7.10 * 44.78
* derived
Operation 41

Table 4-9. Temperature table for hot day (from MIL-C-5011B).

pressure altitude temperature
◦ ◦ ◦
ft km * K C F*
0 0 312.55 39.4 102.92
500 0.152 311.55 38.4 101.12
1000 0.305 310.45 37.3 99.14
1500 0.457 309.45 36.3 97.34
2000 0.610 308.35 35.2 95.36
2500 0.762 307.25 34.1 93.38
3000 0.914 306.25 33.1 91.58
3500 1.067 305.15 32.0 89.60
4000 1.219 304.05 30.9 87.62
4500 1.372 302.95 29.8 85.64
5000 1.524 301.85 28.7 83.66
5500 1.676 300.75 27.6 81.68
6000 1.829 299.65 26.5 79.70
6500 1.981 298.55 25.4 77.72
7000 2.134 297.45 24.3 75.74
7500 2.286 296.35 23.2 73.76
8000 2.438 295.25 22.1 71.78
8500 2.591 294.15 21.0 69.80
9000 2.743 293.05 19.9 67.82
9500 2.896 291.95 18.8 65.84
10000 3.048 290.85 17.7 63.86
10500 3.200 289.85 16.7 62.06
11000 3.353 288.85 15.7 60.26
11500 3.505 287.75 14.6 58.28
12000 3.658 286.75 13.6 56.48
12500 3.810 285.65 12.5 54.50
13000 3.962 284.55 11.4 52.52
13500 4.115 283.55 10.4 50.72
14000 4.267 282.45 9.3 48.74
14500 4.420 281.35 8.2 46.76
15000 4.572 280.35 7.2 44.96
* derived
42 Operation
 Chapter 5

Solution Procedures

The NDARC code performs design and analysis tasks. The design task involves sizing the rotorcraft
to satisfy specified design conditions and missions. The analysis tasks can include off-design mission
performance analysis, flight performance calculation for point operating conditions, and generation of
subsystem or component performance maps. Figure 5-1 illustrates the tasks. The principal tasks (sizing,
mission analysis, and flight performance analysis) are shown in the figure as boxes with dark borders.
Dark black arrows show control of subordinate tasks.

The aircraft description (fig. 5-1) consists of all the information, input and derived, that defines
the aircraft. The aircraft consists of a set of components, including fuselage, rotors, wings, tails, and
propulsion. This information can be the result of the sizing task; can come entirely from input, for a
fixed model; or can come from the sizing task in a previous case or previous job. The aircraft description
information is available to all tasks and all solutions (indicated by light green arrows).

Missions are defined for the sizing task and for the mission performance analysis. A mission consists
of a specified number of mission segments, for which time, distance, and fuel burn are evaluated. For
specified takeoff fuel weight with adjustable segments, the mission time or distance is adjusted so the
fuel required for the mission (burned plus reserve) equals the takeoff fuel weight. The mission iteration
is on fuel weight or energy.

Flight conditions are specified for the sizing task and for the flight performance analysis.

For flight conditions and mission takeoff, the gross weight can be maximized such that the power
required equals the power available.

A flight state is defined for each mission segment and each flight condition. The aircraft performance
can be analyzed for the specified state, or a maximum-effort performance can be identified. The
maximum effort is specified in terms of a quantity such as best endurance or best range, and a variable
such as speed, rate of climb, or altitude. The aircraft must be trimmed, by solving for the controls
and motion that produce equilibrium in the specified flight state. Different trim solution definitions are
required for various flight states. Evaluating the rotor hub forces may require solution of the blade flap
equations of motion.

The sizing task is described in more detail in chapter 3. The flight condition, mission, and flight
state calculations are described in chapter 4. The solution of the blade flap equations of motion is
described in chapter 11. The present chapter provides details of the solution procedures implemented
for each iteration of the analysis.

The nested iteration loops involved in the solution process are indicated by the subtitles in the
boxes of figure 5-1, and illustrated in more detail in figure 5-2. The flight state solution involves up
44 Solution Procedures

fixed model or
previous job or
previous case
DESIGN ANALYZE
Airframe
Aerodynamics Map

Sizing Task
size iteration Aircraft Engine
Description Performance Map

Mission Analysis

design
conditions design Flight
missions Performance
Analysis

Mission
Flight Condition
adjust & fuel wt iteration
max GW
max takeoff GW

each segment

Flight State
max effort / trim aircraft / flap equations

Figure 5-1. Outline of NDARC tasks.

to three loops. The innermost loop is the solution of the blade flap equations of motion, needed for an
accurate evaluation of the rotor hub forces. The next loop is the trim solution, which is required for
most flight states. The flight state optionally has one or two maximum-effort iterations. The flight state
solution is executed for each flight condition and for each mission segment. A flight condition solution
or any mission segment solution can optionally maximize the aircraft gross weight. The mission usually
requires an iterative solution, for fuel weight or for adjustable segment time or distance. Thus each flight
condition solution involves up to four nested iterations: maximum gross weight (outer), maximum effort,
trim, and blade motion (inner). Each mission solution involves up to five nested iterations: mission
(outer), and then for each segment maximum gross weight, maximum effort, trim, and blade motion
(inner). Finally, the design task introduces a sizing iteration, which is the outermost loop of the process.
Solution Procedures 45

Sizing Task Flight Condition

Size Iteration Maximum GW

method: successive method: secant or false position

substitution
Flight State
Missions

Mission
Flight Conditions
Mission Iteration
fuel weight, adjust time/distrance
Mission Analysis
method: successive substitution

Missions
Maximum GW

Flight Performance Analysis method: secant or false

position

Flight Conditions Flight State

Flight State

Maximum Effort

method: golden section search for maximum endurance,

range, or climb; otherwise secant or false position

Trim

method: Newton-Raphson

Component Performance
Blade Flapping

method: Newton-Raphson

Figure 5-2. Design and analysis tasks, with nested loops and solution methods.
46 Solution Procedures
5–1 Iterative Solution Tasks

5-1.1 Tolerance and Perturbation

For each solution procedure, a tolerance  and a perturbation Δ may be required. Single values are
specified for the task and then scaled for each element tested or perturbed.
The scaling is based on a reference weight W (design gross weight, or derived from aircraft
CT /σ = 0.07), a reference length L (fuselage length, rotor radius, or wing span), and a reference power

P (aircraft installed power, or derived from P = W W/2ρA). Then the force reference is F = W ,
the moment reference is M = W L/10, and the angle reference is A = 1 deg. The velocity reference is
V = 400 knots. The angular velocity reference is Ω = V /L (in deg/sec). The coefficient reference is
C = 0.6 for wings and C = 0.1 for rotors. Altitude scale is H = 10000 ft. Acceleration scale is G = g
(acceleration due to gravity). The range scale is X = 100 nm. These scaling variables are referred to in
the subsections that follow, and in tables 5-1 to 5-4.

5-1.2 Size Aircraft

The sizing task determines the dimensions, power, and weight of a rotorcraft that can perform a
specified set of design conditions and missions. The aircraft size is characterized by parameters such as
design gross weight, weight empty, rotor radius, and engine power available. The relationships between
dimensions, power, and weight generally require an iterative solution. From the design flight conditions
and missions, the task can determine the total engine power or the rotor radius (or both power and radius
can be fixed), as well as the design gross weight, maximum takeoff weight, drive system torque limit,
and fuel tank capacity. For each propulsion group, the engine power or the rotor radius can be sized.
A successive substitution method is used for the sizing iteration, with an input tolerance . Relax-
ation is applied to Peng or R, Tjet , Pchrg , WD , WM T O , PDSlimit , Wfuel−cap or Efuel−cap , and Tdesign . Two
successive substitution loops are used. The outer loop is an iteration on performance: engine power
or rotor radius, jet thrust, charger power. The inner loop is an iteration on parameters: WD , WM T O ,
PDSlimit , Wfuel−cap or Efuel−cap , and Tdesign . Either loop can be absent, depending on the definition of the
size task. Convergence is tested in terms of these parameters, and the aircraft weight empty WE . The
tolerance is 0.1P  for engine power and drive system limit; 0.01W  for gross weight, maximum takeoff
weight, fuel weight, jet thrust, and design rotor thrust; and 0.1L for rotor radius.

5-1.3 Mission
Missions consist of a specified number of segments, for which time, distance, and fuel burn are
evaluated. For calculated mission fuel weight, the fuel weight at takeoff is adjusted to equal the fuel
required for the mission (burned plus reserve). For specified takeoff fuel weight with adjustable segments,
the mission time or distance is adjusted so the fuel required for the mission (burned plus reserve) equals
the takeoff fuel weight. The mission iteration is thus on fuel weight or energy. Range credit segments
can also require an iteration.
A successive substitution method is used if an iteration is required, with a tolerance  specified.
The principal iteration variable is takeoff fuel weight, for which the tolerance is 0.01W . For calculated
mission fuel weight, the relaxation is applied to the mission fuel value used to update the takeoff fuel
weight. For specified takeoff fuel weight, the relationship is applied to the fuel weight increment used
to adjust the mission segments. The tolerance for the distance flown in range credit segments is X. The
relaxation is applied to the distance flown in the destination segments for range credit.
Solution Procedures 47

5-1.4 Maximum Gross Weight

Flight conditions are specified for the sizing task and for the flight performance analysis. Mission
takeoff conditions are specified for the sizing task and for the mission analysis. Optionally for flight
conditions and mission takeoff, the gross weight can be maximized, such that the power required equals
the power available, min(PavP G −PreqP G ) = 0 (zero power margin, minimum over all propulsion groups);
or such that the power required equals an input power, min((d + f PavP G ) − PreqP G ) = 0 (minimum over
all propulsion groups, with d an input power and f an input factor; this convention allows the power to
be input directly, f = 0, or scaled with power available). Similarly, the gross weight can be maximized
for zero jet thrust margin, or zero torque margin.

The secant method or the method of false position is used to solve for the maximum gross weight.
A tolerance  and a perturbation Δ are specified. The variable is gross weight, with initial increment of
W Δ, and tolerance of 0.01W . Note that the convergence test is applied to the magnitude of the gross
weight increment.

5-1.5 Maximum Effort

The aircraft performance can be analyzed for the specified state or a maximum-effort performance
can be identified. The secant method or the method of false position is used to solve for the maximum
effort. The task of finding maximum endurance, range, or climb is usually solved using the golden-
section or curve-fit method. A tolerance  and a perturbation Δ are specified.

A quantity and variable are specified for the maximum-effort calculation. Tables 5-1 and 5-2
summarize the available choices, with the tolerance and initial increment used for the variables. Note
that the convergence test is applied to the magnitude of the variable increment. Optionally two quantity/
variable pairs can be specified, solved in nested iterations. The two variables must be unique. The two
variables can maximize the same quantity (endurance, range, or climb). If the variable is velocity, first
the velocity is found for the specified maximum effort; the performance is then evaluated at that velocity
times an input factor. For endurance, range, or climb, the slope of the quantity to be maximized must be
zero; hence in all cases the target is zero. The slope of the quantity is evaluated by first-order backward
difference. For the range, first the variable is found such that V /ẇ is maximized (slope zero), and then
the variable is found such that V /ẇ equals 99% of that maximum; for the latter the variable perturbation
is increased by a factor of 4 to ensure that the solution is found on the correct side of the maximum.

Table 5-1. Maximum-effort solution.

maximum-effort variable initial increment tolerance
horizontal velocity Vh 0.1V Δ 0.1V 
vertical rate of climb Vz 0.1V Δ 0.1V 
aircraft velocity β (sideslip) 100AΔ 100A
altitude HΔ H
aircraft angular rate θ̇ (pullup), ψ̇ (turn) ΩΔ Ω
aircraft linear acceleration ax , ay , az GΔ G
aircraft control angle 100AΔ 100A
aircraft orientation θ (pitch), φ (roll) 100AΔ 100A
propulsion group tip speed Vtip VΔ V
propulsion group engine speed Nspec 10ΩΔ 10Ω
48 Solution Procedures

Table 5-2. Maximum-effort solution.

maximum-effort quantity
best endurance maximum 1/ẇ
best range 99% maximum V /ẇ high or low side, or 100%
best climb or descent rate maximum Vz or 1/P
best climb or descent angle maximum Vz /V or V /P
ceiling maximum altitude
power limit power margin, min(PavP G − PreqP G ) = 0 over all propulsion groups
torque limit torque margin, min(Qlimit − Qreq ) = 0 over all limits
thrust limit thrust margin, min(TavJG − TreqJG ) = 0 over all jet groups
power, torque, thrust limit power, torque, thrust margin most restrictive
battery limit power margin, min(Pmax − |Ėbatt |) = 0 over all fuel tanks
rotor stall thrust margin, (CT /σ)max − |CT /σ| = 0 for designated rotor
wing stall lift margin, CLmax − CL = 0 for designated wing

Table 5-3. Trim solution.

trim quantity target tolerance
aircraft total force x, y , z components 0 F
aircraft total moment x, y , z components 0 M
aircraft load factor x, y , z components Flight State 
propulsion group power Flight State P
power margin PavP G − PreqP G Flight State P
torque margin PDSlimit − PreqP G Flight State P
engine group power Flight State P
power margin PavEG − PreqEG Flight State P
jet group thrust Flight State F
thrust margin TavJG − TreqJG Flight State F
charge group power Flight State P
charge power margin PavCG − PreqCG Flight State P
fuel tank energy flow Ėbatt Flight State P
battery power margin Pmax − |Ėbatt | Flight State P
rotor force lift, vertical, propulsive Flight State, component schedule F
rotor thrust CT /σ Flight State, component schedule C
rotor thrust margin (CT /σ)max − |CT /σ| Flight State C
rotor flapping βc , βs Flight State A
rotor hub moment x (roll), y (pitch) Flight State M
rotor torque Flight State M
wing force lift Flight State, component schedule F
wing lift coefficient CL Flight State, component schedule C
wing lift margin CLmax − CL Flight State C
tail force lift Flight State F
Solution Procedures 49

Table 5-4. Trim solution.

trim variable perturbation
aircraft control angle 100AΔ
aircraft orientation θ (pitch), φ (roll) 100AΔ
aircraft velocity Vh (horizontal velocity) VΔ
aircraft velocity Vz (vertical velocity) VΔ
aircraft velocity β (sideslip) 100AΔ
aircraft angular rate θ̇ (pullup), ψ̇ (turn) ΩΔ
propulsion group tip speed Vtip 20V Δ
propulsion group engine speed Nspec 10ΩΔ

5-1.6 Trim
The aircraft trim operation solves for the controls and motion that produce equilibrium in the
specified flight state. A Newton–Raphson method is used for trim. The derivative matrix is obtained by
numerical perturbation. A tolerance  and a perturbation Δ are specified.
Different trim solution definitions are required for various flight states. Therefore one or more
trim states are defined for the analysis, and the appropriate trim state selected for each flight state of
a performance condition or mission segment. For each trim state, the trim quantities, trim variables,
and targets are specified. Tables 5-3 and 5-4 summarize the available choices, with the tolerances and
perturbations used.

5-1.7 Rotor Flap Equations

Evaluating the rotor hub forces may require solution of the flap equations E(v) = 0. For tip-
path plane command, the thrust and flapping are known, so v = (θ0.75 θc θs )T . For no-feathering plane
command, the thrust and cyclic pitch are known, so v = (θ0.75 βc βs )T . A Newton–Raphson solution
method is used: from E(vn+1 ) ∼
= E(vn ) + (dE/dv)(vn+1 − vn ) = 0, the iterative solution is

vn+1 = vn − C E(vn )

where C = f (dE/dv)−1 , including the relaxation factor f . The derivative matrix for axial flow can be
used. Alternatively, the derivative matrix dE/dv can be obtained by numerical perturbation. Convergence
of the Newton–Raphson iteration is tested in terms of |E| <  for each equation, where  is an input
tolerance.

5–2 Theory

The analysis uses several methods to solve nonlinear algebraic equations. Such equations may be
written in two forms: (a) fixed point x = G(x), and (b) zero point f (x) = 0; where x, G, and f are vectors.
The analysis provides operations that implement the function G or f . Solution procedures appropriate
for the zero point form can be applied to equations in fixed point form, by defining f (x) = x − G(x). In
this context, f can be considered the iteration error.
Efficient and convergent methods are required to find the solution x = α of these equations. Note
that f  (α) = 0 or G (α) = 1 means that α is a higher-order root. For nonlinear problems, the method will
50 Solution Procedures

successive substitution iteration

save: xold = x
evaluate x
relax: x = λx + (1 − λ)xold
test convergence: error = x − xold  ≤ λtolerance × weight
Figure 5-3. Outline of successive substitution method.

be iterative: xn+1 = F (xn ). The operation F depends on the solution method. The solution error is:

n+1 = α − xn+1 = F (α) − F (xn ) = (α − xn )F  (ξn ) ∼

= n F  (α)

Thus the iteration will converge if F is not too sensitive to errors in x: |F  (α)| < 1 for scalar x. For
x a vector, the criterion is that all the eigenvalues of the derivative matrix ∂F/∂x have magnitude less
than one. The equations in this section are generally written for scalar x; the extension to vector x
is straightforward. Convergence is linear for F  nonzero, quadratic for F  = 0. Iterative methods
have a relaxation factor (and other parameters) to improve convergence, and a tolerance to measure
convergence.
The following subsections describe the solution methods used for the various iterations, as shown
in figure 5-2.

5-2.1 Successive Substitution Method

The successive substitution method (with relaxation) is an example of a fixed point solution. A
direct iteration is simply xn+1 = G(xn ), but |G | > 1 for many practical problems. A relaxed iteration
uses F = (1 − λ)x + λG:
xn+1 = (1 − λ)xn + λG(xn ) = xn − λf (xn )

with relaxation factor λ. The convergence criterion is then

|F  (α)| = |1 − λ + λG | < 1

so a value of λ can be found to ensure convergence for any finite G . Specifically, the iteration converges
if the magnitude of λ is less than the magnitude of 2/(1 − G ) = 2/f  (and λ has the same sign as
1 − G = f  ). Quadratic convergence (F  = 0) is obtained with λ = 1/(1 − G ) = 1/f  . Over-relaxation
(λ > 1) can be used if |G | < 1. Since the correct solution x = α is not known, convergence must be
tested by comparing the values of two successive iterations:

error = xn+1 − xn  ≤ tolerance

where the error is some norm of the difference between iterations (typically absolute value for scalar x).
Note that the effect of the relaxation factor is to reduce the difference between iterations:
 
xn+1 − xn = λ G(xn ) − xn

Hence the convergence test is applied to (xn+1 − xn )/λ, in order to maintain the definition of tolerance
independent of relaxation. The process for the successive substitution method is shown in figure 5-3.
Solution Procedures 51

initialize
evaluate h
test convergence: error = |hj − htargetj | ≤ tolerance × weightj
initialize derivative matrix D to input matrix
calculate gain matrix: C = λD−1
iteration
identify derivative matrix
optional perturbation identification
perturb each element of x: δxi = Δ × weighti
evaluate h
calculate D
calculate gain matrix: C = λD−1
increment solution: δx = −C(h − htarget )
evaluate h
test convergence: error = |hj − htargetj | ≤ tolerance × weightj
Figure 5-4. Outline of Newton–Raphson method.

5-2.2 Newton–Raphson Method

The Newton–Raphson method (with relaxation and identification) is an example of a zero point
solution. The Taylor series expansion of f (x) = 0 leads to the iteration operator F = x − f /f  :
−1
xn+1 = xn − [f  (xn )] f (xn )

which gives quadratic convergence. The behavior of this iteration depends on the accuracy of the
derivative f  . Here it is assumed that the analysis can evaluate directly f , but not f  . It is necessary to
evaluate f  by numerical perturbation of f , and for efficiency the derivatives may not be evaluated for
each xn . These approximations compromise the convergence of the method, so a relaxation factor λ is
introduced to compensate. Hence a modified Newton–Raphson iteration is used, F = x − Cf :

xn+1 = xn − Cf (xn ) = xn − λD−1 f (xn )

where the derivative matrix D is an estimate of f  . The convergence criterion is then

|F  (α)| = |1 − Cf  | = |1 − λD−1 f  | < 1

since f (α) = 0. The iteration converges if the magnitude of λ is less than the magnitude of 2D/f  (and λ
has the same sign as D/f  ). Quadratic convergence is obtained with λ = D/f  (which would require λ
to change during the iteration however). The Newton–Raphson method ideally uses the local derivative
in the gain factor, C = 1/f  , so has quadratic convergence:
f f 
F  (α) = =0
f 2

since f (α) = 0 (if f  = 0 and f  is finite; if f  = 0, then there is a multiple root, F  = 1/2, and the
convergence is only linear). A relaxation factor is still useful, since the convergence is only quadratic
sufficiently close to the solution. A Newton–Raphson method has good convergence when x is suf-
ficiently close to the solution, but frequently has difficulty converging elsewhere. Hence the initial
52 Solution Procedures

initialize
evaluate f0 at x0 , f1 at x1 = x0 + Δx, f2 at x2 = x1 + Δx
iteration
calculate derivative f 
secant: from f0 and f1
false position: from f0 , and f1 or f2 (opposite sign from f0 )
calculate gain: C = λ/f 
increment solution: δx = −Cf
shift: f2 = f1 , f1 = f0
evaluate f
test convergence
Figure 5-5. Outline of secant method or method of false position.

estimate x0 that starts the iteration is an important parameter affecting convergence. Convergence of the
solution for x may be tested in terms of the required value (zero) for f :

error = f  ≤ tolerance

where the error is some norm of f (typically absolute value for scalar f ).
The derivative matrix D is obtained by an identification process. The perturbation identification
can be performed at the beginning of the iteration, and optionally every MPID iterations thereafter. The
derivative matrix is calculated from a one-step finite-difference expression (first order). Each element
xi of the vector x is perturbed, one at a time, giving the i-th column of D:
   
∂f f (xi + δxi ) − f (xi )
D = ··· ··· = ··· ···
∂xi δxi

Alternatively, a two-step finite-difference expression (second order) can be used:

   
∂f f (xi + δxi ) − f (xi − δxi )
D = ··· ··· = ··· ···
∂xi 2δxi

With this procedure, the accuracy of D (hence convergence) can be affected by both the magnitude and
sign of the perturbation (only the magnitude for a two-step difference).
The process for the Newton–Raphson method is shown in figure 5-4. A problem specified as
h(x) = htarget becomes a zero point problem with f = h − htarget . A successive substitution problem,
x = G(x), becomes a zero point problem with f = x − G. At the beginning of the solution, x has an
initial value. The perturbation identification can optionally never be performed (so an input matrix is
required), be performed at the beginning of the iteration, or be performed at the beginning and every
MPID iterations thereafter.

5-2.3 Secant Method

The secant method (with relaxation) is developed from the Newton–Raphson method. The modified
Newton–Raphson iteration is:

xn+1 = xn − Cf (xn ) = xn − λD−1 f (xn )

Solution Procedures 53

initialize
evaluate f0 at x0 , f1 at x1 = x0 + Δx, f2 at x2 = x1 + Δx
bracket maximum: while not f1 ≥ f0 , f2
if f2 > f0 , then x3 = x2 + (x2 − x1 ); 1,2,3 → 0,1,2
if f0 > f2 , then x3 = x0 − (x1 − x0 ); 3,0,1 → 0,1,2
iteration (search)
if x2 − x1 > x1 − x0 , then x3 = x1 + W (x2 − x1 )
if f3 < f1 , then 0,1,3 → 0,1,2
if f3 > f1 , then 1,3,2 → 0,1,2
if x1 − x0 > x2 − x1 , then x3 = x1 − W (x1 − x0 )
if f3 < f1 , then 3,1,2 → 0,1,2
if f3 > f1 , then 0,3,1 → 0,1,2
test convergence
Figure 5-6. Outline of golden-section search.

where the derivative matrix D is an estimate of f  . In the secant method, the derivative of f is evaluated
numerically at each step:
f (xn ) − f (xn−1 )
f  (xn ) ∼
=
xn − xn−1
It can be shown that then the error reduces during the iteration according to:

|n+1 | ∼
= |f  /2f  | |n | |n−1 | ∼
= |f  /2f  |.62 |n |1.62

which is slower than the quadratic convergence of the Newton–Raphson method (2n ), but still better than
linear convergence. In practical problems, whether the iteration converges at all is often more important
than the rate of convergence. Limiting the maximum amplitude of the derivative estimate may also be
appropriate. Note that with f = x − G(x), the derivative f  is dimensionless, so a universal limit (say
maximum |f  | = 0.3) can be specified. A limit on the maximum increment of x (as a fraction of the x
value) can also be imposed. The process for the secant method is shown in figure 5-5.

5-2.4 Method of False Position

The method of false position is a derivative of the secant method, based on calculating the derivative
with values that bracket the solution. The iteration starts with values of x0 and x1 such that f (x0 ) and
f (x1 ) have opposite signs. Then the derivative f  and new estimate xn+1 are

f (xn ) − f (xk )
f  (xn ) ∼
=
xn − xk
xn+1 = xn − λD−1 f (xn )

using k = n − 1 or k = n − 2 such that f (xn ) and f (xk ) have opposite signs. The convergence is
slower (roughly linear) than for the secant method, but by keeping the solution bracketed convergence
is guaranteed. The process for the method of false position is shown in figure 5-5.

5-2.5 Golden-Section Search

The golden-section search method can be used to find the solution x that maximizes f (x). The
problem of maximizing f (x) can be attacked by applying the secant method or method of false position to
54 Solution Procedures

initialize
evaluate f0 at x0 , f1 at x1 = x0 + Δx, f2 at x2 = x1 + Δx
bracket maximum: while not f1 ≥ f0 , f2
if f2 > f0 , then x3 = x2 + (x2 − x1 ); 1,2,3 → 0,1,2
if f0 > f2 , then x3 = x0 − (x1 − x0 ); 3,0,1 → 0,1,2
fmax = f1
curve fit
fmax = f1 , xmax = x1
evaluate f for x = xmax + nΔx and x = xmax − nΔx
least-squared error solution for polynomial coefficients
solve polynomial for x at peak f
Figure 5-7. Outline of curve-fit method.

the derivative f  (x) = 0, but that approach is often not satisfactory as it depends on numerical evaluation
of the second derivative. The golden-section search method begins with a set of three values x0 < x1 < x2
and the corresponding functions f0 , f1 , f2 . The x value is incremented until the maximum is bracketed,
f1 ≥ f0 , f2 . Then a new value x3 is selected in the interval x0 to x2 ; f3 evaluated; and the new set of
x0 < x1 < x2 determined such that the maximum is still bracketed. The new value x3 is a fraction
√
W = (3 − 5)/2 ∼ = 0.38197 from x1 into the largest segment. The process for the golden-section search
is shown in figure 5-6.
5-2.6 Curve-Fit Method
The curve-fit method can be used to find the solution x that maximizes f (x), by fitting the solution
to a polynomial. If the function f is flat around the maximum and the inner loop tolerances are not tight
enough, the golden-section search can become erratic, particularly for best range and best endurance
calculations. Curve fitting the evaluated points and then solving the curve for the maximum has the
potential to improve the behavior. The curve-fit method begins with a set of three values x0 < x1 < x2
and the corresponding functions f0 , f1 , f2 . The x value is incremented until the maximum is bracketed,
f1 ≥ f0 , f2 , giving a course maximum fmax at xmax . Next a set of x and f values are generated by
incrementing x above and below xmax , until f < rfit fmax is found (typically rfit = 0.98 for best range).
This set of points is fit to the cubic polynomial f = c3 z 3 + c2 z 2 + c1 z + c0 , z = x/xmax − 1 (or to a
   
quadradic polynomial). Let cT = c0 c1 c2 c3 and ξ T = 1 z z 2 z 3 . Then the least-squared-error solution
for the coefficients is  −1  
c= ξi ξiT fi ξi
i i

where the sums are over the set of points to be fit. For a quadratic polynomial fit, the solution is then
  
√ 2
x = xmax 1 − c1 /2c2 ± 1−r (c1 /2c2 ) − c0 /c2

where r = 1 for the maximum, or r = 0.99 for 99% best range. For a cubic polynomial fit, the maximum
is at     
c2 3c3 c1 ∼ c1 1 3c3 c1
z=− 1− 1− =− 1+
3c3 c22 2c2 4 c22
It is simplest to search the cubic for the peak (z where df /dz = 0), and then if necessary search for the
99% range point (f = 0.99fpeak ) The process for the golden-section search is shown in figure 5-7.
 Chapter 6

Cost

Costs are estimated using statistical models based on historical aircraft price and maintenance
cost data, with appropriate factors to account for technology impact and inflation. The aircraft purchase
price (CAC , in dollars) covers airframe, mission equipment package (MEP), and flight control electronics
(FCE) costs. The direct operating cost (DOC, in cents per available seat mile (ASM)) is the sum of
maintenance cost (Cmaint , in dollars per flight hour), flight crew salary and expenses, fuel and oil cost,
depreciation, insurance cost, and finance cost.

Inflation factors can be input, or internal factors used. Table 6-1 gives the internal inflation factors
for DoD (ref. 1) and CPI (ref. 2). For years beyond the data in the table, optionally the inflation factor
is extrapolated based on the last yearly increase.

6–1 CTM Rotorcraft Cost Model

The CTM rotorcraft cost model (refs. 3–6) gives an estimate of aircraft purchase price, maintenance
cost, and direct operating cost. The model was developed for shaft-driven helicopters and turboprop
aircraft. If the aircraft shaft power is zero, P/WAF = 0.25 is used in the equation for purchase price.

6-1.1 Aircraft Purchase Price

Aircraft purchase price is estimated from the statistical relationship of Harris and Scully (ref. 3,
updated in 2001), based on a 1994 database of mostly civil aircraft (plus EH-101, UH-60L, CH-47D, CH-
53E, and MV-22). The model starts with a function of aircraft weight and power; has several complexity
factors; a factor for rotorcraft or turboprop aircraft; and a country or industry factor (specifically U.S.
military). The model includes (as $/lb) separate calculations of a composite construction increment
(increase or decrease), mission equipment package cost, and flight control electronics cost. The model
accounts for inflation and includes an overall technology factor. With these equations, the purchase price
is predicted within 20% for 96% of 128 rotorcraft (figs. 6-1 and 6-2), implying a standard deviation of
10%. For the five military aircraft, price is predicted within ±10% or less.

The basic statistical relationship for airframe purchase price is:

1.0619
cAF = 739.91 KET KEN KLG KR WAF (P/WAF )0.5887 Nblade
0.1465

with WAF = WE + ΔWkit − WMEP − WFCE , including airframe kits ΔWkit (the wing and wing extension
kits, and optionally the folding kit). The configuration factor Kconfig = KET KEN KLG KR has the
56 Cost

factors:

KET = 1.0 for turbine aircraft 0.557 for piston aircraft

KEN = 1.0 for multi-engine aircraft 0.736 for single-engine aircraft
KLG = 1.0 for retractable landing gear 0.884 for fixed landing gear
KR = 1.0 for single main-rotor 1.057 for twin main-rotors, 1.117 for four main-rotors

The number of blades and the configuration factor are essentially measures of complexity. In particular,
retractable/fixed landing gear is a surrogate for general complexity. The term Ccomp = rcomp Wcomp
accounts for additional costs for composite construction (negative for cost savings); Wcomp is the com-
posite structure weight, obtained as an input fraction of the component weight, with separate fractions
for body, tail, pylon, and wing weight. The MEP and FCE costs are obtained from input cost-per-weight
factors: CMEP = rMEP WMEP and CFCE = rFCE WFCE .
The statistical cost equation for cAF is based on 1994 dollars and current technology levels. In-
cluding an inflation factor Fi and technology factor χAF gives the purchase price CAC :
CAC = χAF (Fi cAF ) + Ccomp + CMEP + CFCE

In addition to technology, χ accounts for calibration and industry factors; for example, χAF = 0.87
for U.S. Military (ref. 3). This equation also estimates turboprop airliner purchase price by setting
Nrotor = Nblade = 1 and using the additional factor 0.8754 (pressurized) or 0.7646 (unpressurized). The
purchase price in $/lb or $/kg is
rAF = (χAF (Fi cAF ))/WAF
rAC = CAC /(WE + ΔWkit )
for the airframe and the aircraft. Parameters are defined in table 6-2, including units as used in these
equations.

6-1.2 Maintenance Cost

Total maintenance cost per hour (dollars per flight hour) is estimated from the statistical relationship
of Harris and Scully (ref. 3, updated in 2001). The maintenance cost per hour is:
cmaint = 0.49885 WE0.3746 P 0.4635

This equation is based on 1994 dollars and current technology levels. Including an inflation factor Fi
and technology factor χmaint gives the maintenance cost per flight hour Cmaint :
Cmaint = χmaint (Fi cmaint )

Parameters are defined in table 6-2, including units as used in these equations.
Alternatively, the maintenance cost cmaint can be calculated from separate estimates of labor, parts
(airframe, engine, and avionics), engine overhaul, and major periodic maintenance costs. The equations
for these maintenance cost components are from Harris (ref. 6), based on a 2011 civil database. The
contributions to the dollars per flight hour are:
Clabor = rlabor (MMH/FH)
cparts = Mparts (CAC /106 )0.68
cengine = Mengine P 0.67
cmajor = Mmajor (CAC /106 )
Cost 57

where rlabor is the maintenance labor rate (dollars per hour). The maintenance-man-hours per flight-hour
is estimated from
MMH/FH = Mlabor WE0.78

or specified directly. These equations are based on 2011 dollars and current technology levels. For
current best practice (bottom of data), the constants are Mlabor = 0.0017, Mparts = 34, Mengine = 1.45,
Mmajor = 18; while for current average practice Mlabor = 0.0027, Mparts = 56, Mengine = 1.74, Mmajor =
28. Finally, the maintenance cost per flight hour is

Cmaint = χmaint Fi (cparts + cengine + cmajor ) + Clabor

including an inflation factor Fi and technology factor χmaint .

6-1.3 Direct Operating Cost

Given specifics of the operation (including available block hours, non-flight time, spares fraction,
and financial numbers), the direct operating cost is calculated for each mission. The direct operating
cost includes maintenance, fuel, crew, depreciation, insurance, and finance costs. The contributions to
the yearly operating cost are:
 
Cfuel = Gfuel (Wfuel /ρfuel ) + Genergy Efuel Ndep
0.4
Ccrew = 2.84Fi Kcrew WM T OB
1+S
Cdep = CAC (1 − V )
D
Cins = 0.0056 CAC
1 + S 2L + 1 i
Cfin = CAC
D 4 100

There is a separate credit (Gcredit , perhaps zero) for mission energy generation (Efuel < 0). The crew
factor Kcrew = 1 corresponds to low-cost, domestic airlines (1994 dollars). The fuel burn Wfuel , block
time Tmiss , and block range Rmiss are obtained for a designated mission. The number of departures per
year is Ndep = B/Tmiss . The flight time per trip is Ttrip = Tmiss − TN F . The flight hours per year are
TF = Ttrip Ndep . Alternatively, the sum of the crew, insurance, and depreciation costs can be estimated
from the purchase price:
 
Ccrew + Cdep + Cins = Kcdi 175000 + 29.8(CAC /1000)

where Kcdi is a calibration factor (ref. 6). The yearly operating cost COP and DOC (cents per available
seat mile) are then:
COP = TF Cmaint + Cfuel + Ccrew + Cdep + Cins + Cfin
DOC = 100 COP /ASM
where the available seat miles per year are ASM = 1.1508Npass Rmiss Ndep (range in nm).

6–2 References

1) “National Defense Budget Estimates for FY 1998/2017.” Office of the Under Secretary of Defense
(Comptroller), March 1997/2016. Department of Defense Deflators, for Total Obligational Authority
(TOA), Procurement.
58 Cost

2) “Consumer Price Index.” U.S. Department of Labor, Bureau of Labor Statistics, 2016. All Urban
Consumers (CPI-U), U.S. city average, All items.
3) Harris, F.D., and Scully, M.P. “Rotorcraft Cost Too Much.” Journal of the American Helicopter
Society, Vol. 43, No. 1, January 1998.
4) Harris, F.D. “An Economic Model of U.S. Airline Operating Expenses.” NASA CR 2005-213476,
December 2005.
5) Coy, J.J. “Cost Analysis for Large Civil Transport Rotorcraft.” American Helicopter Society Vertical
Lift Aircraft Design Conference, San Francisco, California, January 2006.
6) Harris, F.D. “Introduction to Autogyros, Helicopters, and Other V/STOL Aircraft.” NASA SP
2011-215959, volume 2, 2012.
Cost 59

Table 6-1. DoD and CPI inflation factors.

inflation factors inflation factors inflation factors

year DoD CPI year DoD CPI year DoD CPI
1911 1951 17.02 17.54 1991 93.99 91.90
1912 1952 16.26 17.88 1992 96.21 94.67
1913 6.68 1953 16.31 18.02 1993 98.16 97.50
1914 6.75 1954 16.02 18.15 1994 100.00 100.00
1915 6.82 1955 17.29 18.08 1995 101.70 102.83
1916 7.35 1956 17.58 18.35 1996 103.18 105.87
1917 8.64 1957 18.17 18.96 1997 104.32 108.30
1918 10.19 1958 18.02 19.50 1998 105.40 109.99
1919 11.67 1959 18.11 19.64 1999 106.81 112.42
1920 13.50 1960 18.34 19.97 2000 108.39 116.19
1921 12.08 1961 18.31 20.18 2001 109.92 119.50
1922 11.34 1962 18.84 20.38 2002 111.62 121.39
1923 11.54 1963 19.09 20.65 2003 113.95 124.16
1924 11.54 1964 19.76 20.92 2004 116.92 127.46
1925 11.81 1965 20.36 21.26 2005 120.10 131.78
1926 11.94 1966 21.97 21.86 2006 123.06 136.03
1927 11.74 1967 22.77 22.54 2007 125.56 139.91
1928 11.54 1968 24.03 23.48 2008 127.43 145.28
1929 11.54 1969 25.10 24.76 2009 128.95 144.76
1930 11.27 1970 25.76 26.18 2010 131.66 147.14
1931 10.26 1971 27.24 27.33 2011 133.84 151.78
1932 9.24 1972 28.98 28.21 2012 136.07 154.92
1933 8.77 1973 31.34 29.96 2013 138.34 157.19
1934 9.04 1974 34.13 33.27 2014 139.54 159.74
1935 9.24 1975 38.10 36.30 2015 141.61 159.93
1936 9.38 1976 42.14 38.39 2016 144.03 161.95
1937 9.72 1977 43.75 40.89 2017* 146.76
1938 9.51 1978 47.82 43.99 2018* 149.63
1939 9.38 1979 53.02 48.99 2019* 152.63
1940 9.45 1980 58.52 55.60 2020* 155.68
1941 9.92 1981 63.80 61.34 2021* 158.79
1942 11.00 1982 68.35 65.11
1943 11.67 1983 71.92 67.21
1944 11.88 1984 74.57 70.11
1945 12.15 1985 76.86 72.60
1946 13.16 1986 79.19 73.95
1947 15.05 1987 81.87 76.65
1948 16.26 1988 85.01 79.82
1949 16.06 1989 88.19 83.67
1950 15.37 16.26 1990 91.30 88.19
* projected
60 Cost

Table 6-2. Cost model parameters.

parameter definition units
WE weight empty lb
WM T O maximum takeoff weight lb
Nblade number of blades per rotor
P rated takeoff power (all engines) hp
WMEP fixed useful load weight, mission equipment package lb or kg
WFCE fixed useful load weight, flight control electronics lb or kg
rMEP cost factor, mission equipment package $/lb or $/kg
rFCE cost factor, flight control electronics $/lb or $/kg
rcomp additional cost for composite construction $/lb or $/kg
Fi inflation factor, relative 1994
Wfuel mission fuel burned lb or kg
Tmiss mission time hr
Rmiss mission range nm
Gfuel fuel cost $/gallon or $/liter
Genergy energy cost $/MJ
Gcredit energy credit $/MJ
B available block hours hr
S spares per aircraft (fraction purchase price)
D depreciation period yr
V residual value (fraction)
L loan period yr
i interest rate %
TN F non-flight time per trip hr
Npass number of passengers
ρfuel fuel density (weight per volume) lb/gal or kg/liter
Ndep number of departures per year
Kcrew crew factor
TF flight hours per year
Cost 61

1000.
no error
±10%
900.
±20%
aircraft
800.

700.
predicted base price (1994$/lb)

600.

500.

400.

300.

200.

100.

0.
0. 100. 200. 300. 400. 500. 600. 700. 800. 900. 1000.
actual base price (1994$/lb)

Figure 6-1. Statistical estimation of rotorcraft purchase price ($/lb).

62 Cost

100.0
predicted base price (1994 $M)

10.0

1.0

0.1
0.1 1.0 10.0 100.0
actual base price (1994 $M)

Figure 6-2. Statistical estimation of rotorcraft purchase price ($M).

 Chapter 7

Emissions

Evaluating the environmental impact of an aircraft begins with calculating the engine emissions for
the missions flown. It is best to use metrics that account for all relevant aircraft emissions, although met-
rics based on a single species can be useful. From the engine emissions, the atmospheric concentrations
are evaluated, then radiative forcing (RF ). Radiative forcing is a measure of the amount of heat trapped
in the atmosphere by a particular pollutant, and is expressed in terms of trapped energy per unit area
(W/m2 ). Finally, climate changes and the resulting impacts and damages can be determined. The steps
in this causal chain become increasingly important and increasingly uncertain. Any metric that is used
to evaluate new aircraft concepts must balance uncertainty with relevance. The present environmental
impact models were developed by Russell (ref. 1), based on the work of Dallara, Kroo, and Waitz (ref. 2).

The important aircraft emission species are the direct greenhouse gas carbon dioxide (CO2 ), nitrogen
oxides (NOx ), water vapor (H2 O), aerosols of soot and sulphate (SO4 ), and aviation-induced cloudiness
(AIC). Carbon dioxide has a long lifetime relative to chemical processes in the atmosphere, hence aviation
CO2 impacts do not vary with altitude and can be calculated as for other sources. NOx emissions cause
changes in RF indirectly through chemical processes in the atmosphere. Increases in NOx lead to
increases in atmospheric ozone by a short lifetime process (O3S ), a warming effect, and over long time-
scales reductions in the greenhouse gas methane (CH4 ), a cooling effect. The methane reduction has a
secondary effect of reducing ozone (O3L ), a cooling effect. Water vapor, short-lived ozone, soot, and
sulfate aerosols have lifetimes much shorter than one year, hence have impacts for only a short time after
emission. Aviation-induced cloudiness (AIC), including the effects of linear contrails and altered cirrus
cloudiness, have a short-lived warming effect. Total radiative forcing from aircraft can be several times
greater than forcing due to CO2 alone, depending on altitude and AIC conditions. The impact of NOx
emissions is similar in magnitude to that of CO2 , but there is greater uncertainty in the NOx RF values.
Estimates of the RF due to induced cirrus are particularly uncertain. The uncertainty of the models and
parameters for the various emission species is discussed in reference 2.

The environmental impact of the aircraft operation is estimated in terms of two metrics: emissions
trading scheme (ETS) credits, and average temperature response (ATR). The impact of energy used is
also modeled.

7–1 Emissions Trading Scheme Credits

The emissions trading scheme (ETS) is the European Union system to curb the effects of greenhouse
gas emissions on global climate change, by limiting the amount of carbon dioxide that can be produced
by large polluters such as energy and industrial installations. The ETS was extended to the aviation
sector in early 2012. Under the ETS, each member nation has an emission cap that is used to allocate
allowable carbon emissions to their industrial operators. Operators that do not use their entire allocation
64 Emissions

can sell their unused “carbon credits” on the open market, while operators exceeding their allocation
must purchase credits on the market. The price of carbon credits has varied widely, ranging from below
10 to above 30 Euro/ton CO2 . The metric used here is the weight of CO2 produced:
WCO2 = Kfuel Wburn + Kenergy Eburn

per mission. Typically Kfuel = EICO2 ∼ = 3.75 (lb/lb or kg/kg) and Kenergy ∼
= 0.14 (kg/MJ). Different
aircraft designs or operations can be compared in terms of WCO2 . Accounting for the influence on
operating cost requires the price of carbon credits.
Kfuel depends on the combustion process and the chemical composition of the fuel (ref. 3). For
combustion of conventional jet fuel, Kfuel = 3.16 kg/kg = 74 g/MJ = 266 g/kWh (using 42.8 MJ/kg).
Accounting for emissions from production of the fuel (principally carbon dioxide, methane, and nitrogen
oxides), gives an equivalent CO2 value perhaps 20% larger: Kfuel = 3.75 kg/kg = 88 g/MJ = 315 g/kWh.
For combustion of alternative jet fuels, typically Kfuel = 3.01 kg/kg = 70 g/MJ = 263 g/kWh. Account-
ing for biomass credit (CO2 absorbed during biomass growth), the equivalent CO2 value might be 50%
lower than for conventional fuel (ref. 3).
Kenergy depends on the energy source. Table 7-1 (from ref. 4) gives typical values of equivalent
CO2 for electricity from various sources. Worldwide average values for equivalent CO2 range from 6
to 1100 g/kWh. The current average for the United States is about Kenergy = 500 g/kWh = 139 g/MJ.
Optionally there is no emissions credit for energy generation (Eburn < 0).

Table 7-1. Electricity green-house-gas intensity (equivalent CO2 , g/kWh).

source median range
coal 860 780-1100
gas 500 400-600
small wind turbine 35 20-50
nuclear 22 10-30
large wind turbine 20 15-25
hydroelectric 18 10-20

7–2 Average Temperature Response

The average temperature response (ATR) quantifies the lifetime global mean temperature change
caused by operation of an aircraft, as a measure of climate change (ref. 2). ATR can be used with a
number of different climate models, but simple linear climate models are appropriate for the conceptual
design of rotorcraft. While atmospheric processes are nonlinear, the globally averaged responses to
small perturbations (in magnitude similar to commercial aircraft emissions) of CO2 , NOx , and AIC
perturbations are nearly linear (ref. 2). The ATR metric is based on the radiative forcing (RF ) generated
by each emission species. Radiative forcing measures the net imbalance of incoming and outgoing
energies in the Earth-atmosphere system caused by a perturbation. RF is an instantaneous measure
that quantifies the change in energy that produces changes in climate properties, including temperature.
Many climate change metrics, such as Global Warming Potentials, rely on RF , but do not specifically
target emissions due to aviation. The total RF for all emitted pollutants is used to calculate the global
temperature response. The use of an altitude-sensitive climate model captures the effects of aircraft
operating conditions. In addition, ATR includes parameters such as usage rates and operating lifetime
of the aircraft to determine the total climate impact.
Emissions 65

The average temperature response is calculated by integrating the temperature change during and
after H years of sustained operation of an aircraft:
 ∞
1
ATR = ΔTH (t) w(t) dt
H 0

ATR has units of temperature. The metric can also be used in relative terms, by dividing by the value
for a baseline design. ΔTH (t) is the global mean temperature change resulting from the operation of the
given design, where emissions are assumed to be constant for the first H years and zero thereafter. This
temperature change is a function of the radiative forcing caused by the emission of a number of different
pollutant species. There are multiple models available with varying levels of fidelity for calculating
ΔTH . Here linear climate models and functions are used, based on reference 2. One limitation of this
model is that the RF of NOx and AIC is based on data that only goes down to 17500 ft. Below this
altitude the effects of NOx and AIC are assumed constant.
There are climate effects from perturbations remaining in the Earth-atmosphere system after the
aircraft operating lifetime has ended. Hence ΔTH increases for time up to H , and thereafter decays, but
is not zero. The weighting function w(t) allows discounting of temperature change effects in the years
following H , so that long-term effects such as CO2 warming do not dominate ATR. The function
⎧
⎪ 1 t≤H
⎪
⎨ 1
w(t) = H < t ≤ tmax
⎪
⎪ (1 + r)t−H
⎩
0 t > tmax

has unit weighting during the operating lifetime, and exponential devaluation thereafter, in terms of the
discount rate r (typically r = 2 to 5%). A rate of zero means that post-operation impacts are equally
important compared with impacts during operating years; a rate of infinity means that post-operation
impacts have no importance; and a positive, finite rate means that the temperature change each post-
operation year is less important than the temperature change experienced the previous year (by the factor
1/(1 + r)).

Also contributing to the ATR metric is aviation induced cloudiness (AIC), which includes both
contrails and aviation-induced cirrus clouds. Following the methodology of reference 2, the impact of
AIC is assumed to be a function of cruise altitude and distance, and does not account for changes in
water vapor emissions or exhaust temperature or atmospheric conditions. The consequence of these
assumptions is that for a given rotorcraft mission, the radiative forcing due to AIC may be greatly
overestimated. Also, strategies are currently being developed to reduce the effects of AIC by rerouting
around areas in the atmosphere prone to aircraft contrails, so future aircraft may have significantly
reduced climate impacts due to AIC. Since there is a high degree of uncertainty in radiative forcing due
to AIC, calculations of ATR should be made both with and without its effects.
Design for minimum ATR depends on the time horizon, discount rate, operating altitude and speed,
and the engine technology. If there is no discounting of long-term effects (r = 0), radiative forcing due
to CO2 dominates the ATR, favoring a higher cruise altitude. When long term effects are included and
engine technology for low NOx is assumed, ATR can be generally minimized by designing for low fuel
burn. If only short-term impacts are considered (r = ∞), then NOx emissions have a greater impact,
and a lower cruise altitude results in decreased climate impact, despite increased fuel burn and CO2
emissions. The inclusion of AIC in the metric calculations also has a large impact on the design, and
favors a low cruise altitude.
66 Emissions

7-2.1 Aircraft Mission

ATR is calculated assuming constant emissions during the first H years of operation and zero
emissions thereafter. For a single operating mission, ΔTH is the sum over all mission segments, times a
utilization rate of U missions per year. The information for each mission segment includes: fuel burned
Wburn , energy used Eburn , altitude h, air distance flown d, engine fuel flow ẇ, and operating condition
(M , θ, δ ).

7-2.2 Engine Emissions

Engine emissions can be quantified in terms of an emissions index (EI , lb/lb or kg/kg) that relates
the amount of a pollutant species to the amount of fuel burned. For some emission species, such as CO2
and water vapor, the EI is nearly constant, assuming ideal or near-ideal fuel combustion. For others,
particularly NOx , the EI varies widely throughout the operating range of the engine. The emission of
species i is thus obtained from the fuel consumption summed over all mission segments:

ei = EIi Wburn
seg

The time-varying annual emissions E is found by summing over all operations conducted in a year.
Assuming a single operating mission, Ei = U ei , where U is the aircraft utilization rate (number of
missions per year).
For each fuel, EI is required for CO2 , H2 O, SO4 , soot, and NOx . Optionally there is no emissions
credit for energy generation (Eburn < 0).

7-2.3 Turboshaft Engine Emissions

For carbon dioxide, water vapor, sulfates, and soot the emissions indices are assumed to be constant.
The EI of CO2 , H2 O, or SO4 depends solely on the composition of the fuel. The EI for soot can vary
with engine operating condition, but because soot comprises a small fraction of total climate impacts
(typically less than 5%), this index can also be assumed constant. The values used for the constant EI
emissions are given in table 7-2 (from ref. 2).

Table 7-2. Emission indices for constant EI species.

species emission index EI (lb/lb or kg/kg)
jet fuel methane hydrogen
CO2 3.16 2.75 0
H2 O 1.26 2.25 8.94
SO4 2.0 × 10−4
soot 4.0 × 10−5

The EI for oxides of nitrogen (including NO and NO2 , collectively called NOx ) depends on operat-
ing conditions, including engine throttle setting, Mach number, and altitude (temperature and pressure).
Hence EINOx must be evaluated for each mission segment. Engines are required to comply with NOx
emissions regulations during landing and takeoff. The International Civil Aviation Organization (ICAO)
measures and publishes certification data relating fuel flow, thrust, and EINOx at four sea-level static
throttle settings that simulate taxi, takeoff, climb, and approach operating conditions. However, ICAO
Emissions 67

does not measure NOx at cruise conditions, because these emissions are not currently regulated. There-
fore, a model is needed to compute EINOx at specific flight conditions including cruise. Fuel flow
correlation methods have been developed for this purpose. While there is a large amount of published
turbofan NOx emissions data and established methods for estimating variation with altitude, there is
little public information for turboshaft engines. Russell (ref. 1) developed a method to estimate EINOx
based on engine performance information from the NDARC engine models, ICAO turbofan emissions
data, and the German Aerospace Center (DLR) fuel flow method described in reference 5. The DLR
semi-empirical method models the relationship between engine throttle setting and EINOx at varying
ambient temperature and pressure conditions. The DLR method has been shown to predict EINOx to
within approximately 10% of measurements at typical cruise conditions.
The emission index for NOx is calculated from the fuel flow ẇ as follows. The fuel flow is corrected
√
to referred conditions at the inlet: ẇcorr = ẇ/δt θt ), where the temperature and pressure ratios are
 
γ−1 2  
θt = 1 + M θ = 1 + 0.2M 2 θ
2
  γ
γ − 1 2 γ−1  3.5
δt = 1 + M δ = 1 + 0.2M 2 δ
2
(for ratio of specific heats γ = 1.4). The engine model gives the fuel flow ẇ as a function of Preq = fP Pto
at SLS static conditions, which can be interpolated to find fP such that ẇ = ẇcorr . The corrected emission
index is scaled to SLS takeoff power, hence

EINOx corr = KEI0 + KEI1 fP

The constants depend on the engine design and technology. Typically 1000KEI0 = 3.6739 and 1000KEI1
= 7.48 for a low-emissions engine (based on CF34); 1000KEI0 = 2.4392 and 1000KEI1 = 23.26 for a
high-emissions engine (based on HTF7000). Finally, the emission index is

EINOx = EINOx corr θt3 δt0.4 FH

 
with the humidity correction factor FH = exp (6.29 − e−0.000143(h−12900) )/53.2 , altitude h in ft.
There are two assumptions in this method. The first assumption is that the variation of EINOx for
a turboshaft is the same as that of a turbofan engine. Since both are gas turbine engines, employing the
same thermal cycle, this assumption should be valid. The second assumption is that the RPTEM engine
model is producing variations in fuel flow with speed and altitude in a manner consistent with the DLR
fuel flow method. Since both the fuel flow method and the engine model in NDARC are based on real
engine data, this assumption should be valid as well. Two different engines are used as references. The
GE CF34-3B and the Honeywell HTF7000 have fuel flows similar to what is calculated for a turboshaft
engine. Both engines are relatively modern, high bypass ratio, small turbofans. The HTF7000 and CF34
also represent the upper and lower bounds on NOx emissions for this category of engine, so they should
bracket the expected quantity of NOx emissions for a turboshaft engine. Compared to the differences
between the EINOx values of the two reference engines, the speed and altitude corrections are relatively
small.
An alternative approach is to calculate the emission index for NOx from the engine power required.
Reference 6 gives the parametric relation
 0.5677
Preq
EINOx = KEIs
1000
68 Emissions

with Preq in hp, and 1000KEIs = 4 to 14, depending on the engine design and technology.
With more than one turboshaft engine group using fuel from a tank, the emission index for a
 
segment is weighted by the fuel flow, hence by the fuel burned: EINOx = EI ẇ/ ẇ (sum over
engine groups).

7-2.4 Energy Emissions

The environmental impact of energy used is quantified in terms of constants (K , kg/MJ) that relate
the amount of a pollutant species to the amount of energy burned. These constants depend on the source
of the energy, and are required for CO2 , H2 O, SO4 , soot, and NOx . The emission of species i is thus
obtained from the energy consumption summed over all mission segments:

ei = Ki Eburn
seg

The time-varying annual emissions are Ei = U ei , which is added to the engine emissions. In the climate
model, the altitude corresponding to these emissions is zero.

7-2.5 Climate Model and Radiative Forcing

A climate model is required to determine the temperature change resulting from the emissions
produced by an aircraft operation. The simplest approach (ref. 2) is to use a linear climate model,
so the time-varying radiative forcing of each species is obtained from the emissions and a climate
impulse response function. Although derived from the globally averaged results of some complex,
three-dimensional, global climate model, a linear model simplifies the physics and chemistry of aircraft-
induced climate change, and captures first-order effects. The climate model used, developed in reference
2 specifically for aircraft operations, includes the altitude variation for NOx and AIC impacts. Different
models are required for long-lived gases, short-lived pollutants, and AIC. Models for RF from all
emissions except CO2 , which has a very long atmospheric lifetime, depend on the assumed geographical
distribution of emissions. The models used are based on average impacts from fleetwide routing in
a single year within the last decade, quantifying the average forcing caused by emissions spatially
distributed according to routing similar to current traffic.
For long-lived emission species i (CO2 , CH4 , O3L ), the radiative forcing RFi is obtained by
integrating the product of the Ei (kg/yr) and the impulse response Gi (W/m2 /kg):
 t
RFi (t) = si (h) Gi (t − τ )Ei (τ ) dτ
0
    t
=U si (h) EIi Wburn + Ki Eburn Ai (Gi (t − τ )/Ai )u(τ ) dτ
seg 0

=U si (h)ai Ai Xi (t)
seg

where ai = (EIi Wburn + Ki Eburn ), u(t) is the emission time history (unit amplitude), and for t > 0
  
GCO2 (t) = ACO2 1 + Σ3j=1 αcj e−t/τcj − 1
GCH4 (t) = ACH4 e−t/τn
GO3L (t) = AO3L e−t/τn
Emissions 69

EINOx is used for CH4 , O3L . The parameters are given in table 7-3. The forcing factors si (h) are given
as a function of altitude in table 7-4; s = 1 for carbon dioxide.
For short-lived emission species (H2 O, O3S , soot, SO4 ), RF is assumed to be proportional to the
emissions:
RFi (t) = si (h) (RFref /Eref )i Ei (t)
  
=U si (h) EIi Wburn + Ki Eburn (RFref /Eref )i u(t)
seg

=U si (h)ai Ai Xi (t)
seg

where Ai = RFref /Eref and Xi (t) = u(t). EINOx is used for O3S . The parameters (RFref /Eref ) are given
in table 7-3. The forcing factor s = 1 except for O3S , which is given in table 7-4.
Aviation-induced cloudiness (AIC) effects are assumed to be proportional to the distance traveled:
RFAIC (t) = sAIC (h) fAIC (RFref /Lref )AIC L(t)

=U sAIC (h)dfAIC (RFref /Lref )AIC u(t)
seg

=U sAIC (h)aAIC AAIC XAIC (t)
seg

where L = U seg d is the stage length flown per year, aAIC = d; AAIC = fAIC RFref /Lref and XAIC (t) =
u(t). The parameter (RFref /Lref ) is given in table 7-3, and the forcing factor sAIC in table 7-4; fAIC is
an input correction factor.
These forcing factors (table 7-4) characterize globally and annually averaged RF s based on current
flight routes. Forcing factors, particularly for AIC impacts, are expected to vary with time of day, season,
and earth latitude, effects which are averaged in this model.
The normalized RF for each species is based on the RF that would result from a doubling of carbon
dioxide:
RFi (t)  fi
RFi∗ (t) = fi =U si (h)ai Ai Xi (t)
RF2xCO2 seg
RF 2xCO2

for CO2 , CH4 , O3L , O3S , H2 O, soot, SO4 , and AIC. The efficacy factors fi are given in table 7-3;
fCO2 = 1. From the total normalized RF and a climate impulse response function, the time-varying
global mean temperature change is evaluated:
 t 
ΔT (t) = GT (t − τ ) RFi∗ (τ ) dτ
0 i
   t
fi
= U si (h)ai Ai S (GT (t − τ )/S)Xi (τ ) dτ
i seg
RF2xCO2 0
  fi
= U si (h)ai Ai SYi (t)
i seg
RF2xCO2

where  
αt −t/τt1 1 − αt −t/τt2
GT (t) = S e + e
τt1 τt2
The time constants in GT describe the thermal response of the earth system to an energy perturbation,
and S is the steady-state temperature change produced by a constant annual forcing of RF2xCO2 . The
parameters are given in table 7-3.
70 Emissions
 
The corresponding weight emitted for each species is Wi = seg (EIi Wburn + Ki Eburn ) = ai
 seg
per mission (Wi = HU seg ai over the operating duration H ).

7-2.6 ATR Evaluation

The average temperature response (ATR) is calculated from ΔT = ΔTH , assuming constant emis-
sions during the first H years of operation and zero emissions thereafter. Hence the shape function
u(t) = 1 for 0 < t < H , and zero at other times. For a single operating mission, ΔTH is the sum over all
mission segments (each giving fuel and energy burned and distance traveled), with a utilization rate of
U missions per year. The radiative forcing factors are:
 t  
XCO2 (t) = 1 + Σj αcj e−(t−τ )/τcj − 1 u(τ ) dτ
0
 
t(1 − Σj αcj ) + Σj αcj τcj 1 − e−t/τcj t<H
=  
H(1 − Σj αcj ) + Σj αcj τcj e−(t−H)/τcj − e−t/τcj t>H
 t
XCH4 (t) = XO3L (t) = e−(t−τ )/τn u(τ ) dτ
0
 
τn 1 − e−t/τn t<H
=  
τn e−(t−H)/τn − e−t/τn t>H
!t
and Xi (t) = u(t) for the other species. The temperature change factors Yi (t) = 0
(GT (t − τ )/S)Xi (τ ) dτ
are: ⎧ " 
⎪
⎪ αt t − τt1 (1 − e−t/τt1 ) (1 − Σj αcj )
⎪
⎪
⎪
⎪ # $
⎪
⎪ τt1 τcj
⎪
⎪ +Σj αcj τcj 1 + e −t/τt1
− e−t/τcj
⎪
⎪ τcj − τt1 τcj − τt1
⎪
⎪
⎪
⎪ "  t<H
⎪
⎪ + (1 − αt ) t − τt2 (1 − e−t/τt2 ) (1 − Σj αcj )
⎪
⎪
⎪
⎪ # $
⎪
⎪
⎪
⎪ τt2 −t/τt2 τcj −t/τcj
⎪
⎪ +Σ α τ
j cj cj 1 + e − e
⎪
⎪ τcj − τt2 τcj − τt2
⎪
⎪
⎪
⎪
⎪
⎪ " 
⎪
⎪
⎨ αt H − τt1 (e−(t−H)/τt1 − e−t/τt1 ) (1 − Σj αcj )
YCO2 (t) = #  
⎪
⎪ τt1
⎪
⎪ +Σj αcj τcj − e−(t−H)/τt1 − e−t/τt1
⎪
⎪ τcj − τt1
⎪
⎪
⎪
⎪  $
⎪
⎪ τcj −(t−H)/τcj −t/τcj
⎪
⎪ + e −e
⎪
⎪ τcj − τt1
⎪
⎪ "  t>H
⎪
⎪ −(t−H)/τt2 −t/τt2
⎪
⎪ + (1 − α ) H − τ (e − e ) (1 − Σj αcj )
⎪
⎪
t t2
⎪
⎪ #  
⎪
⎪ τt2 −(t−H)/τt2 −t/τt2
⎪
⎪ +Σ α τ − e − e
⎪
⎪
j cj cj
τcj − τt2
⎪
⎪
⎪
⎪  $
⎪
⎪ τcj
⎩ + e−(t−H)/τcj − e−t/τcj
τcj − τt2
Emissions 71

⎧  
⎪
⎪ τt1 −t/τt1
−
τn −t/τn
⎪
⎪ αt τn 1 + e e
⎪
⎪ τn − τt1 τn − τt1
⎪
⎪   t<H
⎪
⎪ τt2 τn
⎪
⎪ + (1 − αt )τn 1 + e−t/τt2
− e−t/τn
⎪
⎪ τn − τt2 τn − τt2
⎪
⎪
⎪
⎪
⎪
⎪ 
⎪
⎪  
⎪
⎨ αt τn − τt1 e−(t−H)/τt1 − e−t/τt1
YCH4 (t) = YO3L (t) = τn − τt1
⎪
⎪
⎪ τn  
⎪
⎪ −(t−H)/τn
−e −t/τn
⎪
⎪ +
τn − τt1
e
⎪
⎪
⎪  t>H
⎪
⎪
⎪
⎪ τt2  −(t−H)/τt2 
⎪
⎪ + (1 − αt )τn − e − e−t/τt2
⎪
⎪ τn − τt2
⎪
⎪  
⎪
⎪ τn
⎪
⎩ + e−(t−H)/τn − e−t/τn
τn − τt2

and for the other species

⎧ " % " %
⎪
⎨ αt 1 − e
−t/τt1
+ (1 − αt ) 1 − e−t/τt2 t<H
Yi (t) =
⎪ " % " %
⎩ α e−(t−H)/τt1 − e−t/τt1 + (1 − α ) e−(t−H)/τt2 − e−t/τt2 t>H
t t

Then the average temperature response is


1 tmax
ATR = ΔTH (t) w(t) dt
H 0
  
fi 1 tmax
= U si (h)ai Ai S Yi (t)w(t) dt
i seg
RF2xCO2 H 0
  fi
= U si (h)ai Ai SZi
i seg
RF2xCO2

= ATRi
i

Given H and r, Zi is evaluated by numerical integration. For the special cases of r = 0 (no discounting)
and r = ∞ (no effects beyond H ):
72 Emissions

⎧  
⎪ H
⎪
⎪ − αt τt1 − (1 − αt )τt2 (1 − Σj αcj ) + Σj αcj τcj
⎪
⎪ 2
⎪
⎪ " τ  
⎪
⎪
⎪
⎪ + α τ
t1
1 − e−H/τt1
(1 − Σj αcj )
⎪
⎪
t t1
H #
⎪
⎪ $
⎪
⎪
⎪ τt1 τt1  −H/τt1
 τcj τcj  −H/τcj
⎪
⎪ +Σ α
j cj cjτ 1 − e − 1 − e r=∞
⎪
⎪ τcj − τt1 H τcj − τt1 H
⎪
⎪ " τ  
⎪
⎪
⎪ 1 − e−H/τt2 (1 − Σj αcj )
t2
⎪
⎪ + (1 − αt ) τt2
⎪
⎪ H
#
⎪
⎪ τt2   τcj  $
⎪
⎪ τt2 −H/τt2 τcj −H/τcj
⎪
⎪ +Σj αcj τcj 1−e − 1−e
⎪
⎪ τcj − τt2 H τcj − τt2 H
⎪
⎪
⎪
⎪
⎪
⎪  
⎪
⎪
⎨ tmax − H − αt τt1 − (1 − αt )τt2 (1 − Σj αcj ) + Σj αcj τcj
ZCO2 = 2
⎪
⎪ " τ  
⎪
⎪ + α τ
t1
e −(tmax −H)/τt1
− e −tmax /τt1
(1 − Σj αcj )
⎪
⎪ t t1
⎪
⎪ H #
⎪
⎪ τt1  −(tmax −H)/τt1 
⎪
⎪ τt1
⎪
⎪ +Σj αcj τcj e − e−tmax /τt1
⎪
⎪ τcj − τt1 H
⎪
⎪ $
⎪
⎪
⎪ τcj τcj  −(tmax −H)/τcj −tmax /τcj
⎪
⎪ − e − e r=0
⎪
⎪ τcj − τt1 H
⎪
⎪ " τ  
⎪
⎪
⎪
⎪ + (1 − α ) τ
t2
e−(tmax −H)/τt2
− e −tmax /τt2
(1 − Σj αcj )
⎪
⎪
t t2
H
⎪
⎪ #  
⎪
⎪ τt2 τt2 −(tmax −H)/τt2
⎪
⎪ +Σ α τ e − e −tmax /τt2
⎪
⎪ j cj cj
τcj − τt2 H
⎪
⎪
⎪
⎪
⎪ τ τcj  −(tmax −H)/τcj $
⎪
⎩ −
cj
e −e −t max /τ cj
τcj − τt2 H

⎧  
⎪
⎪ τt1 τt1  −H/τt1
 τn τn  −H/τn
⎪
⎪ τ n + α τ
t n 1 − e − 1 − e
⎪
⎪ τn − τt1 H τn − τt1 H
⎪
⎪     r=∞
⎪
⎪
⎪ τt2 τt2 −H/τt2 τn τn  −H/τn
⎪
⎪ + (1 − α )τ
t n 1 − e − 1 − e
⎪
⎪ τn − τt2 H τn − τt2 H
⎪
⎪
⎪
⎪
⎪
⎪ 
⎪
⎪
⎨ τn + αt τn τt1 τt1  −(tmax −H)/τt1 
e − e−tmax /τt1
ZCH4 = ZO3L = τn − τt1 H
⎪
⎪
⎪ τn τn  −(tmax −H)/τn 
⎪
⎪ − −e−t /τ
⎪ e max n
⎪
⎪ τn − τt1 H
⎪
⎪  r=0
⎪
⎪
⎪
⎪ τt2 τt2  −(tmax −H)/τt2 −tmax /τt2

⎪
⎪ + (1 − αt )τ n e − e
⎪
⎪ τn − τt2 H
⎪
⎪  
⎪
⎪ τ τ
⎪
⎩ −
n n
e−(tmax −H)/τn
−e−tmax /τn
τn − τt2 H
and for the other species:
⎧ " τ  % " τ  %
⎪
⎪ 1 + α −
t1
1 − e −H/τt1
+ (1 − α ) −
t2
1 − e−H/τt2
r=∞
⎪
⎪
t
H
t
H
⎪
⎪
⎨
" τ  %
Zi =
e−(tmax −H)/τt1 − e−tmax /τt1
t1
⎪
⎪ 1 + αt −
⎪
⎪ H "
⎪
⎪
⎩ τt2  −(tmax −H)/τt2 % r=0
+ (1 − αt ) − e − e−tmax /τt2
H
Emissions 73

Since Σj αcj = 1, the radiative forcing for CO2 is finite at long times (XCO2 → H(1 − Σj αcj )), and CO2
dominates the temperature response. It is necessary to introduce tmax so that ZCO2 (and ATR) is finite
in the absence of discounting (r = 0).

7–3 References

1) Russell, C., and Johnson, W. “Application of Climate Impact Metrics to Civil Tiltrotor Design.”
AIAA Paper No. 2013-1087, January 2013.
2) Dallara, E.S.; Kroo, I.M.; and Waitz, I.A. “Metric for Comparing Lifetime Average Climate Impact
of Aircraft.” AIAA Journal, Vol. 49, No. 8 (August 2011).
3) Stratton, R.W.; Wong, H.M.; and Hileman, J.I. “Life Cycle Greenhouse Gas Emissions from Alter-
native Jet Fuels.” PARTNER Project 28 report, PARTNER-COE-2010-001, June 2010.
4) Thompson, T.R., and Stouffer, V. “Greenhouse Gas Impacts of On-Demand Electric Aircraft.” AIAA
Paper No. 2015-2388, June 2015.
5) Deidewig, F.; Döpelheuer, A.; and Lecht, M. “Methods to Assess Aircraft Engine Emissions in
Flight,” 20th International Council on Aeronautical Sciences Congress, Sorrento, Italy, 1996.
6) Rindlisbacher, T. “Guidance on the Determination of Helicopter Emissions.” Federal Office of Civil
Aviation, Swiss Confederation, March 2009.

Table 7-3. RF factor s for NOx impacts and AIC.

altitude (ft) AIC CH4 , O3L O3S
<17500 0.000 0.868 0.470
17500 0.029 0.868 0.470
19500 0.000 0.924 0.557
21500 0.000 0.959 0.620
23500 0.171 0.965 0.711
25500 0.399 0.947 0.713
27500 0.798 0.933 0.809
29500 1.254 0.930 0.931
31500 1.710 0.942 1.009
33500 2.109 0.977 1.131
35500 1.824 1.136 1.431
37500 1.539 1.213 1.630
39500 0.969 1.204 1.802
41500 0.798 1.201 1.937
74 Emissions

Table 7-4. Radiative forcing models.

CO2 parameters
ACO2 1.80 × 10−15 (W/m2 )/(kg CO2 )
αc1 0.259
αc2 0.338
αc3 0.186
τc1 172.9 yr
τc2 18.51 yr
τc3 1.186 yr
NOx parameters
ACH4 −5.16 × 10−13 (W/m2 )/(kg NOx )
AO3L −1.21 × 10−13 (W/m2 )/(kg NOx )
τn 12.0 yr
fCH4 1.18
fO3 1.37
(RFref /Eref )O3S 1.01 × 10−11 (W/m2 )/(kg NOx )
H2 O, SO4 , soot, and AIC parameters
fH2 O 1.14
fSO4 0.9
fsoot 0.7
fAIC 0.59
(RFref /Eref )H2 O 7.43 × 10−15 (W/m2 )/(kg H2 O)
(RFref /Eref )SO4 −1.00 × 10−10 (W/m2 )/(kg SO4 )
(RFref /Eref )soot 5.00 × 10−10 (W/m2 )/(kg soot)
(RFref /Lref )AIC 2.21 × 10−12 (W/m2 )/nm
Temperature change model parameters
RF2xCO2 3.70 W/m2
S 3.0 K
αt 0.595
τt1 8.4 yr
τt2 409.5 yr
 Chapter 8

Aircraft

The aircraft consists of a set of components, including rotors, wings, tails, fuselage, and propul-
sion. For each component, attributes such as performance, drag, and weight can be calculated. The
aircraft attributes are obtained from the sum of the component attributes. Description and analysis of
conventional rotorcraft configurations is facilitated, while retaining the capability to model novel and
advanced concepts. Specific rotorcraft configurations considered include: single-main-rotor and tail-
rotor helicopter, tandem helicopter, coaxial helicopter, tiltrotor, compound helicopter, multicopter, and
airplane.
The following components form the aircraft:
a) Systems: The systems component contains weight information (fixed useful load, vibration, contin-
gency, and systems and equipment) for the aircraft.
b) Fuselage: There is one fuselage for the aircraft.
c) Landing Gear: There is one landing gear for the aircraft.
d) Rotors: The aircraft can have one or more rotors, or no rotors. In addition to main-rotors, the
component can model tail-rotors, propellers, proprotors, and ducted fans.
e) Wings: The aircraft can have one or more wings, or no wings.
f) Tails: The aircraft can have one or more horizontal or vertical tail surfaces, or no tails.
g) Fuel Tanks: There are one or more fuel tank systems for the aircraft. Fuel tank systems are associated
with the engine groups, jet groups, and charge groups. Fuel quantity is measured as either weight or
energy. There can be one or more sizes of auxiliary fuel tanks.
h) Propulsion Groups: The aircraft can have one or more propulsion groups, or none. Each propulsion
group is a set of components (rotors) and engine groups, connected by a drive system. The components
define the power required. The engine groups define the power available.
i) Engine Groups: An engine group consists of one or more engines of a specific type. An engine group
transfers power by shaft torque, so it is associated with a propulsion group. For each engine type an
engine model is defined. The engine model describes a particular engine, used in one or more engine
groups.
j) Jet Groups: The aircraft can have one or more jet groups, or none. A jet group produces a force on
the aircraft. A jet model describes a particular jet, used in one or more jet groups.
k) Charge Groups: The aircraft can have one or more charge groups, or none. A charge group generates
energy for the aircraft. A charge model describes a particular charger, used in one or more charge groups.
76 Aircraft
8–1 Loading

The aircraft disk loading is the ratio of the design gross weight and a reference rotor area: DL =

WD /Aref . The reference area is a sum of specified fractions of the rotor areas, Aref = fA A (typically
the projected area of the lifting rotors). The disk loading of a rotor is the ratio of a specified fraction of
the design gross weight and the rotor area:
T fW WD fW WD
(DL)rotor = = =
A A A/Aref Aref

where probably rotor fW = 1, and the lifting rotors are all rotors not designated antitorque or auxiliary-
thrust. If there are N lifting rotors of the same area, with no overlap, then fA = 1, Aref = N A,
fW = A/Aref = 1/N , and (DL)rotor = DL. For rotors designated antitorque or auxiliary-thrust, the disk
loading is calculated from the design rotor thrust: (DL)rotor = Tdesign /A.
For coaxial rotors, the default reference area is the area of one rotor: fA = 1/2, Aref = A, fW = 1/2,
and (DL)rotor = 1/2DL. For tandem rotors, the default reference area is the projected area: Aref =
(2 − m)A, where mA is the overlap area (m = 0 for no overlap, m = 1 for coaxial). Then fA = 2−m 2 ,
2−m
fW = /2, and (DL)rotor = 2 DL. Optionally, the reference area for tandem rotors can be total rotor
1

area instead: Aref = 2A.

The aircraft wing loading is the ratio of the design gross weight and a reference wing area: WL =

WD /Sref . The reference area is a sum of the wing areas, Sref = S . The wing loading of an individual
wing is the ratio of a specified fraction of the design gross weight and the wing area:
W fW WD fW WD
(WL)wing = = =
S S S/Sref Sref

where probably wing fW = 1. If there are N wings of the same area, then fW = S/Sref = 1/N , and
(WL)wing = WL.

The aircraft power loading is the ratio of the design gross weight and the total installed takeoff

power: W/P = WD / Neng P eng, where the sum is over all engine groups.

8–2 Controls

A set of aircraft controls cAC are defined, and these aircraft controls are connected to the component
controls. The connection to the component control c is typically of the form c = ST cAC + c0 , where T
is an input matrix and c0 the component control for zero aircraft control. The connection (matrix T ) is
defined for a specified number of control system states (allowing change of control configuration with
flight state). The factor S is available for internal scaling of the matrix. The control state and initial
control values are specified for each flight state. Figure 8-1 illustrates the control relationships.
Typical (default) aircraft controls are the pilot’s controls: collective stick, lateral and longitudinal
cyclic sticks, pedal, and tilt. Units and sign convention of the pilot’s controls are contained in the matrix
T . For the single-main-rotor and tail-rotor configuration, it is often convenient for the collective and
cyclic stick motion to equal the collective and cyclic pitch input of the main-rotor, and the pedal motion to
equal the collective pitch input of the tail-rotor. The aircraft controls should be scaled to approximately
the same amplitude, by appropriate definition of the matrix T and scale factor S .
These aircraft controls are available for trim of the aircraft. Any aircraft controls not selected for
trim will remain fixed at the values specified for the flight state. Thus by defining additional aircraft
controls, component controls can be specified as required for a flight state.
Aircraft 77

aircraft controls component controls

c AC c = T c AC + c0
c 0 = zero, constant, f(V)
collective
lateral cyclic
long cyclic ROTOR
pedal collective
tilt lateral cyclic
other T long cyclic
incidence
(e.g. flap,
elevator, cant
rudder, gear) diameter
gear

TAIL
control state control
(or conversion (elevator
schedule) WING or rudder)
flap incidence
flight state value flaperon
or zero, constant, f(V) aileron
or conversion schedule incidence
or trim JET
GROUP
amplitude
mode
ENGINE incidence
trim option GROUP yaw
amplitude
mode
incidence CHARGE
yaw GROUP
amplitude
mode
incidence
PROPULSION yaw
GROUP
rotational speed

Figure 8-1. Aircraft and component controls.

Each aircraft control variable cAC can be zero, constant, or a function of flight speed (piecewise
linear input). The flight state input can override this value of the aircraft control. The input value is an
initial value if the control is a trim variable.
Each component control variable c0 (value for zero aircraft control) can be zero, constant, or a
function of flight speed (piecewise linear input). Optionally the use of c0 can be suppressed for a flight
state. The component control from aircraft control (T cAC ) is a fixed value, or a function of speed, or a
linear function of another control (perhaps a trim variable).
78 Aircraft

The tilt control variable αtilt is intended for nacelle tilt angle or conversion control, particularly for
tiltrotors. The convention is αtilt = 0 for cruise, and αtilt = 90 deg for helicopter mode. If αtilt exists as
a control, it can be zero, constant, or a function of flight speed (piecewise linear input).
An optional control conversion schedule is defined in terms of conversion speeds: hover and
helicopter mode for speeds below VChover , cruise mode for speeds above VCcruise , and conversion mode
between. The nacelle angle is αtilt = 90 in helicopter mode, αtilt = 0 in airplane mode, and it varies
linearly with speed in conversion mode. The tip speed is Vtip−hover in helicopter and conversion mode,
and Vtip−cruise in airplane mode. Control states and drive system states are defined for helicopter, cruise,
and conversion mode flight. The flight state specifies the nacelle tilt angle, tip speeds, control state,
and drive system state, including the option to obtain any or all of these quantities from the conversion
schedule.
The flight speed used for control scheduling is usually the calibrated airspeed (CAS), hence variation
with dynamic pressure. Velocity schedules are used for conversion, controls and motion, rotor tip speed,
landing gear retraction, trim targets, and drive system ratings. Optionally these velocity schedules use
indicated airspeed Vind , calibrated airspeed Vcal , or true airspeed V .
The control matrices T can be defined based on the configuration. Let cAC0 , cACc , cACs , cACp be
the pilot’s controls (collective, lateral cyclic, longitudinal cyclic, and pedal). For the helicopter, the first
rotor is the main-rotor and the second rotor is the tail-rotor; then
⎛ ⎞ ⎡ ⎤⎛ ⎞
TMcoll 1 0 0 0 cAC0
⎜ TMlat ⎟ ⎢ 0 −r 0 0 ⎥ ⎜ cACc ⎟
⎝ ⎠=⎣ ⎦⎝ ⎠
TMlng 0 0 −1 0 cACs
TTcoll 0 0 0 −r cACp

where r is the main-rotor direction of rotation (r = 1 for counter-clockwise rotation, r = −1 for clockwise
rotation). For the tandem configuration, the first rotor is the front rotor and the second rotor is the rear
rotor; then ⎛ ⎞ ⎡ ⎤⎛ ⎞
TFcoll 1 0 −1 0 cAC0
⎜ Flat ⎟ ⎢ 0
T −rF 0 −rF ⎥ ⎜ cACc ⎟
⎝ ⎠=⎣ ⎦⎝ ⎠
TRcoll 1 0 1 0 cACs
TRlat 0 −rR 0 rR cACp
For the coaxial configuration:
⎛ ⎞ ⎡ ⎤
T1coll 1 0 0 r1
⎛ ⎞
⎜ T1lat ⎟ ⎢ 0 −r1 0 0 ⎥ cAC0
⎜ ⎟ ⎢ ⎥
⎜ T1lng ⎟ ⎢ 0 0 −1 0 ⎥ ⎜ cACc ⎟
⎜ ⎟=⎢ ⎥⎝ ⎠
⎜ T2coll ⎟ ⎢ 1 0 0 r2 ⎥ cACs
⎝ ⎠ ⎣ ⎦
T2lat 0 −r2 0 0 cACp
T2lng 0 0 −1 0

For the tiltrotor, the first rotor is the right rotor and the second rotor is the left rotor; then
⎛ ⎞ ⎡ ⎤
TRcoll 1 −1 0 0
⎜ TRlng ⎟ ⎢ 0 0 −1 1 ⎥⎛ ⎞
⎜ ⎟ ⎢ ⎥ cAC0
⎜ TLcoll ⎟ ⎢ 1 1 0 0 ⎥
⎜ ⎟ ⎢ ⎥⎜c ⎟
⎜ TLlng ⎟ = ⎢ 0 0 −1 −1 ⎥ ⎝ ACc ⎠
⎜ ⎟ ⎢ ⎥ cACs
⎜ Tail ⎟ ⎢ 0 −1 0 0 ⎥
⎝ ⎠ ⎣ ⎦ cACp
Telev 0 0 1 0
Trud 0 0 0 1
Aircraft 79

with cyclic stick and pedal connected to rotor controls only for helicopter mode. For the multicopter
configuration:
Ticoll = cAC0 − cACc sin ψi + cACs cos ψi + cACp ri

where ψi is the angle of the i-th rotor, measured clockwise from the forward longitudinal axis. Alternative
multicopter control strategies use rotor rotational speed instead of collective, or cyclic for yaw control.
The sign conventions for the pilot’s controls are collective stick positive up, lateral cyclic stick
positive right, longitudinal cyclic stick positive forward, and pedal positive nose right. The rotor controls
are a positive Fourier series, with azimuth measured in the direction of rotation.

8–3 Trim

The aircraft trim operation solves for the controls and motion that produce equilibrium in the
specified flight state. In steady flight (including hover, level flight, climb and descent, and turns),
equilibrium implies zero net force and moment on the aircraft. In general, there can be additional
quantities that at equilibrium must equal target values. In practice, the trim solution can deal with a
subset of these quantities. Usually it is at least necessary to achieve equilibrium in the aircraft lift and
drag forces, as well as in yaw moment for torque balance. The basic purpose of the trim solution is to
determine the component states, including aircraft drag and rotor thrust, sufficient to evaluate the aircraft
performance.
Different trim solution definitions are required for various flight states. Therefore one or more
trim states are defined for the analysis, and the appropriate trim state selected for each flight state of a
performance condition or mission segment. For each trim state, the trim quantities, trim variables, and
targets are specified. The available trim quantities include:
aircraft total force and moment; aircraft load factor;
propulsion group power;
power margin PavP G − PreqP G ; torque margin PDSlimit − PreqP G ;
engine group power; power margin PavEG − PreqEG ;
jet group thrust; thrust margin TavJG − TreqJG ;
charge group power; charge power margin PavCG − PreqCG ;
fuel tank energy flow Ėbatt ; battery power margin, Pmax − |Ėbatt |;
rotor force (lift, vertical, or propulsive);
rotor thrust CT /σ ; rotor thrust margin (CT /σ)max − |CT /σ|;
rotor flapping βc , βs ; rotor hub moment, roll and pitch; rotor torque;
wing force; wing lift coefficient CL ; wing lift margin CLmax − CL ;
tail force.
Targets for aircraft total force and total moment (including inertial loads in turns) are always zero. The
available trim variables include:
aircraft controls;
aircraft orientation, θ (pitch), φ (roll);
aircraft horizontal velocity Vh ;
aircraft vertical rate of climb Vc ; aircraft sideslip angle;
aircraft angular rate, θ̇ (pullup), ψ̇ (turn);
propulsion group tip speed or engine speed.
80 Aircraft

axes for description of aircraft coordinate

aircraft geometry system
(arbitrary reference) (origin at CG)

WL starboard

BL y
forward

SL x aft

port
z

Figure 8-2. Aircraft geometry.

The aircraft orientation variables are the Euler angles of the body axes relative to inertial axes. The
aircraft controls (appropriately scaled) are connected to the component controls.
A Newton–Raphson method is used for trim. The derivative matrix is obtained by numerical
perturbation. A tolerance  and a perturbation Δ are specified.

8–4 Geometry

The aircraft coordinate system has the x-axis forward, y -axis to the right, and z -axis down, measured
from the center of gravity (fig. 8-2). These aircraft axes are body axes (x is not aligned with the wind),
the orientation determined by the convention used for the input geometry. The center of gravity is the
appropriate origin for describing the motion of the aircraft, and summing the forces and moments acting
on the aircraft.
Layout of the geometry is typically in terms of station line (SL, positive aft), buttline (BL, positive
right), and waterline (WL, positive up), measured relative to some arbitrary origin (fig. 8-2). The x-y -z
axes are parallel to the SL-BL-WL directions. One or more locations are defined for each component
of the aircraft. Each component will at least have a location that is the point where component forces
and moments act on the aircraft. Each location is input in fixed or scaled form. The fixed form input
is SL/BL/WL (dimensional). The scaled form input is x/L (positive aft), y/L (positive right), and
z/L (positive up), based on a reference length L, from a reference point. The reference length is the
rotor radius or wing span of a designated component, or the fuselage length. The reference point can
optionally be input, or the location (hub) of a designated rotor, or the location (center of action) of a
designated wing component, or the location (center of action) of the fuselage, or the location of the
center of gravity. Fixed input can be used for the entire aircraft, or just for certain components.
From this fixed or scaled input and the current aircraft size, the actual geometry (x, y , z ) can be
Aircraft 81

calculated for each location. There are also options to calculate geometry from other parameters (such
as tiltrotor span from rotor radius and clearance). This calculated geometry has the sign convention of
the aircraft axes (x positive forward, y positive right, z positive down), but has the origin at the reference
point (which may or may not be the center of gravity). All input uses the same sign convention; all
internal calculations use the same sign conventions. Table 8-1 summarizes the conventions.
The location of the aircraft center of gravity is specified for a baseline configuration. With tilting
rotors, this location is in helicopter mode. For each flight state the aircraft center of gravity is calculated,
from the baseline location plus any shift due to nacelle tilt, plus an input center-of-gravity increment.
Alternatively, the aircraft center-of-gravity location for the flight state can be input. Any change of the
center-of-gravity position with fuel burn during a mission is not automatically calculated, but could be
accounted for using the flight state input.
The aircraft operating length and width are calculated from the component positions and dimensions:
total = xmax − xmin and wtotal = ymax − ymin , where the maximum and minimum dimensions are for the
fuselage and all rotors, wings, and tails. The corresponding footprint area is then Stotal = total wtotal .

Table 8-1. Geometry conventions.

layout scaled input calculated motion and loads
origin arbitrary reference point reference point center of gravity
x SL (+ aft) x/L (+ aft) x (+ forward) x (+ forward)
y BL (+ right) y/L (+ right) y (+ right) y (+ right)
z WL (+ up) z/L (+ up) z (+ down) z (+ down)

8–5 Aircraft Motion

The aircraft velocity and orientation are defined by the following parameters: flight speed V ; turn
rate; orientation of the body frame relative to inertial axes (Euler angles); and orientation of the velocity
frame relative to inertial axes (flight path angles). Aircraft conventions are followed for the direction
and orientation of axes: the z -axis is down, the x-axis forward, and the y -axis to the right; and a yaw-
pitch-roll sequence is used for the Euler angles. However, the airframe axes are body axes (fixed to the
airframe, regardless of the flight direction) rather than wind axes (which have the x-axis in the direction
of the flight speed). The orientation of the body frame F relative to inertial axes I is defined by yaw,
pitch, and roll Euler angles, which are rotations about the z , y , and x axes, respectively:

C F I = XφF YθF ZψF

So yaw is positive to the right, pitch is positive nose up, and roll is positive to the right. The flight path
is specified by the velocity V , in the positive x-axis direction of the velocity axes. The orientation of the
velocity axes V relative to inertial axes I is defined by yaw (sideslip) and pitch (climb) angles:

C V I = YθV ZψV ZψF

Sideslip is positive for the aircraft moving to the right, and climb is positive for the aircraft moving up.
Then
C F V = C F I C IV = XφF YθF Z−ψV Y−θV
82 Aircraft

In straight flight, all these angles and matrices are constant. In turning flight at a constant yaw rate, the
yaw angle is ψF = ψ̇F t; the turn radius is RT = Vh /ψ̇F ; and the nominal bank angle and load factor are
√
tan φF = n2 − 1 = ψ̇F Vh /g . Then the forward, sideward, and climb velocities are:

Vf = V cos θV cos ψV = Vh cos ψV

Vs = V cos θV sin ψV = Vh sin ψV
Vc = V sin θV = Vh tan θV

where Vh = V cos θV is the horizontal velocity component. The velocity components in airframe axes are
F
vAC = v F I/F = C F V (V 0 0)T (aircraft velocity relative to the air). The calibrated airspeed is calculated
from the true airspeed V :

* +2/7  
√ δ((1 + 0.2M 2 )7/2 − 1) + 1 −1 ∼ √ 1 3
Vcal = V σ = V σ 1 + (1 − δ)M 2
+ (1 − 10δ + 9δ 2
)M 4
0.2M 2 δ 8 640

where σ = ρ/ρ0 is the density ratio, δ = p/p0 is the pressure ratio, and M is the Mach number. The
indicated airspeed Vind is obtained from calibrated airspeed using a specified conversion table. The
aircraft angular velocity is
⎛ ⎞ ⎡ ⎤⎛ ⎞
φ̇F 1 0 − sin θF φ̇F
F
ωAC = ω F I/F = R ⎝ θ̇F ⎠ = ⎣ 0 cos φF sin φF cos θF ⎦ ⎝ θ̇F ⎠
ψ̇F 0 − sin φF cos φF cos θF ψ̇F

For steady state flight, θ̇F = φ̇F = 0; ψ̇F is nonzero in a turn.

Accelerated flight is also considered, in terms of linear acceleration aFAC = v̇ F I/F = gnL and pitch
rate θ̇F . The nominal pullup load factor is n = 1 + θ̇F Vh /g . For accelerated flight, the instantaneous
equilibrium of the forces and moments on the aircraft is evaluated, for specified acceleration and angular
velocity; the equations of motion are not integrated to define a maneuver. Note that the fuselage and
wing aerodynamic models do not include all roll and yaw moment terms needed for general unsteady
flight (notably derivatives Lv , Lp , Lr , Nv , Np , Nr ).
The aircraft pitch and roll angles are available for trim of the aircraft. Any motion not selected for
trim will remain fixed at the values specified for the flight state. The pitch and roll angles each can be
zero, constant, or a function of flight speed (piecewise linear input). The flight state input can override
this value of the aircraft motion. The input value is an initial value if the motion is a trim variable.
The aircraft never-exceed speed VNE (knots TAS) can be specified by a combination of tabular,
stall, compressibility, and Mach number limits. The tabular limit VNEt is a function of density altitude
hd and weight ratio rW = WG /WD , and perhaps temperature τ (◦ C). The stall limit is evaluated for
each rotor from the steady or transient or equation definition of the rotor thrust capability, (CT /σ)max
as a function of advance ratio μ. Given CT /σ , the steadyor transient function is interpolated to μ; or
the equation (CT /σ)max = K0 − K1 μ2 is solved for μ = (K0 − |CT /σ|)/K1 ; then VNEs = μVtip . The
compressibility limit, typically obtained from flight test, is the indicated airspeed (knots) as a function
of density altitude hd , temperature τ (◦ C), main rotor tip speed Vtip , and weight ratio rW = WG /WD :

VNEc = C1 − C2 hd + C3 τ − C4 Vtip − C5 rW

The Mach number limit is evaluated for each rotor from a specified advancing tip Mach number:
VNEm = Mat cs − Vtip . The aircraft never-exceed speed is the minimum of the active limits: VNE =
Aircraft 83

min(VNEt , VNEs , VNEc , VNEm ). Finally, minimum and maximum speeds can be specified, so VNE =
max(Vmin , min(Vmax , VNE )); where Vmin and Vmax are each the minimum of true airspeed and indicated
airspeed values (expressed as true airspeed).

8–6 Loads and Performance

For each component, the power required and the net forces and moments acting on the aircraft can
be calculated. The aerodynamic forces F and moments M are typically calculated in wind axes and then
resolved into body axes (x, y , z ), relative to the origin of the body axes (the aircraft center of gravity).
The power and loads of all components are summed to obtain the aircraft power and loads. Typically
the trim solution drives the net forces and moments on the aircraft to zero.
The aircraft equations of motion, in body axes F with origin at the aircraft center of gravity, are the
equations of force and moment equilibrium:
 F I/F v F I/F ) = F F + Fgrav
m(v̇ F I/F + ω F

 F I/F I F ω F I/F = M F
I F ω̇ F I/F + ω
F
where m = W/g is the aircraft mass; the gravitational force is Fgrav = mC F I g I = mC F I (0 0 g)T ; and
the moment of inertia matrix is ⎡ ⎤
Ixx −Ixy −Ixz
I F = ⎣ −Iyx Iyy −Iyz ⎦
−Izx −Izy Izz
For steady flight, ω̇ F I/F = v̇ F I/F = 0, and ω F I/F = R(0 0 ψ̇F )T is nonzero only in turns. For accelerated
flight, v̇ F I/F can be nonzero, and ω F I/F = R(0 θ̇F ψ̇F )T . The equations of motion are thus
m(aF AC
AC + ω
F F
vAC ) = F F + Fgrav
F

AC
ω F
I F ωAC
F
= MF
The body axis load factor is n = (C F I g I − (aFAC + ω
AC
F F
vAC ))/g . The aF
AC term is absent for steady flight.
The forces and moments are the sum of loads from all components of the aircraft:
     
F F = Ffus
F
+ F
Frotor + F
Fwing + F
Ftail F
+ Ftank + F
Fengine + F
Fjet + F
Fcharge
     
MF F
= Mfus + F
Mrotor + F
Mwing + F
Mtail F
+ Mtank + F
Mengine + F
Mjet + F
Mcharge

Forces and moments in inertial axes are also of interest (F I = C IF F F and M I = C IF M F ). A particular
component can have more than one source of loads; for example, the rotor component produces hub
forces and moments, but also includes hub and pylon drag. The equations of motion are Ef = F F +
F
Fgrav − Finertial
F
= 0 and Em = M F − Minertial
F
= 0.

The component power required Pcomp is evaluated for all components (rotors, motors, and com-
pressors) of the propulsion group. The total power required for the propulsion group PreqP G is obtained
by adding the transmission losses and accessory power. The power required for the propulsion group
must be distributed to the engine groups. The fuel flow is calculated from the engine power, jet thrust,
charger power, and equipment power required. The total fuel flow is the sum from all components of
  
the aircraft: ẇ = ẇreqEG + ẇreqJG + ẇreqCG + ẇeq .

8–7 Aerodynamics

Each component has a position z F in aircraft axes F, relative to the reference point; and orientation
of component axes B relative to aircraft axes given by the rotation matrix C BF . It is expected that the
84 Aircraft

component axes are (roughly) x forward and z down (or in negative lift direction). The aerodynamic
model must be consistent with the convention for component orientation. Acting at the component
F
are interference velocities vint (velocity of air, in F axes), from all other components. Then the total
component velocity relative to the air is

v F = vAC
F
AC
+ω F
Δz F − F
vint

where Δz F = z F − zcg F
. Then v B = C BF v F is the velocity in component axes. The aerodynamic
environment is defined in the component axes: velocity magnitude v = |v B |, dynamic pressure q = 1/2ρv 2 ,
angle of attack α, and sideslip angle β . The angle of attack and sideslip angle provide the transformation
between airframe axes and velocity axes:
C BA = Yα Z−β

This is the conventional aircraft definition, corresponding to yaw-then-pitch of the airframe axes relative
to the velocity vector. By definition, the velocity is along the x-axis in the A axes, v B = C BA (v 0 0)T ;
from which the angle of attack and sideslip in terms of the components of v B are obtained:

α = tan−1 v3B /v1B

β = sin−1 v2B /|v B |

This definition is not well behaved for v1B = 0 (it gives α = 90 sign(v3B )), so for sideward flight a
pitch-then-yaw definition can be useful: C BA = Z−β Yα . Then

α = sin−1 v3B /|v B |

β = tan−1 v2B /v1B

which gives β = 90 sign(v2B ) for v1B = 0.

The component aerodynamic model may include coefficient values for sideward flight, but not have
equations for a continuous variation of the coefficients with sideslip angle. For such cases, sideward
flight is defined as |β| = 80 to 100 degrees.
From v , q , α, and β , the aerodynamic model calculates the component force and moment, in wind
axes acting at z F : ⎛ ⎞ ⎛ ⎞
−D Mx
FA = ⎝ Y ⎠ M A = ⎝ My ⎠
−L Mz
where D, Y , and L are the drag, side force, and lift; Mx , My , and Mz are the roll, pitch, and yaw moments.
The aerodynamic loads in aircraft axes acting at the center of gravity are then:

F F = C F B C BA F A
,F F F
M F = C F B C BA M A + Δz

where Δz F = z F − zcg F
. In hover and low speed, the download is calculated: FzI = kT (C IF F F ), the
downward component of the aerodynamic force in inertial axes. Download can be expressed as a fraction
of the total rotor vertical force or as a fraction of gross weight. The aerodynamic model also calculates
F
the interference velocities caused by this component at all other components: vint = C F B vint
B
.
Equations for the aerodynamics models are defined for all angles in radians. Input values of angles
are, however, in degrees.
Aircraft 85

The aircraft neutral point is calculated from the airframe aerodynamics with all controls set to zero.
The neutral point is here defined as the longitudinal position about which the derivative of the pitch
moment with lift is zero. Hence SLna = SLcg − ΔM/ΔL, with the change in lift and moment calculated
from the loads at angles of attack of 0 and 5 deg.

8–8 Compressible Aerodynamics

The aerodynamic coefficients of lifting surfaces (wing and empennage) can be corrected for com-
pressibility effects. Based on Prandtl-Glauert, the incompressible lift-curve slope is multiplied by the
factor ⎧ 
⎪ 1
⎪
⎪ √ M < Mdiv
⎪
⎪ 1 − M2
⎪
⎪ - .
⎪
⎪
⎪
⎪ 1−M
⎨  Mdiv < M < Mdiv + .1
Fc = (1 − Mdiv ) 1 − Mdiv
2
⎪
⎪ - .
⎪
⎪
⎪
⎪ 1−M M − Mdiv − .1
⎪
⎪  + Mdiv + .1 < M < 1
⎪
⎪ (1 − Mdiv ) 1 − Mdiv
2 1 − Mdiv − .1
⎪
⎩
1. M >1
∼
Typically the lift-divergence Mach number Mdiv = 0.75, about 0.1 less than the drag-divergence Mach
number. A compressible drag increment is modeled following ref. 1 (fig. 6):
 
CDc = CDcc 6.5x2 + 28x4 + 2870x8

where x = (M/Mcc − 0.7) (with M/Mcc restricted to the range 0.7 to 1.1) and CDcc ∼
= 0.0011 is the drag
increment at M = Mcc . The critical Mach number depends on the lift coefficient:
 
Mcc = Mcc0 − Mcc1 0.03CL + 0.19CL2 + 0.78CL3

with typically Mcc0 ∼

= 0.74, Mcc1 ∼
= 0.31, and CL restricted to the range 0 to 1.

8–9 Trailing-Edge Flaps

The lifting surfaces have controls in the form of trailing-edge flaps: flap, flaperon, and aileron for
wings; elevator or rudder for tails. The aerodynamic loads generated by flap deflection δf (radians) are
estimated based on two-dimensional aerodynamic data (as summarized in refs. 2 and 3). Let f = cf /c
be the ratio of the flap chord to the wing chord. The lift coefficient is c = cα (α + τ ηδf ), where
η ∼
= 0.85 − 0.43δf is an empirical correction for viscous effects (ref. 2, eq. 3.54 and fig. 3.36). Thin
airfoil theory gives
θf − sin θf ∼  π n
τ =1− = sin( f)
π 2
with θf = cos−1 (2 f − 1) (ref. 2, eq. 3.56 and fig. 3.35; ref. 3, eq. 5.40). The last expression is an
approximation that is a good fit to the thin airfoil theory result for n = 1/2, and a good approximation
including the effects of real flow for n = 2/3 (ref. 3, fig. 5.18); the last expression with n = 2/3 is used
here. The increase of maximum lift coefficient caused by flap deflection is less than the increase in lift
coefficient, so the stall angle of attack is decreased. Approximately

Δcmax ∼
= (1 − f ) (1 + f −5 2
f + 3 3f )
Δc
86 Aircraft

(ref. 2, fig. 3.37). Thin airfoil theory gives the moment coefficient increment about the quarter chord:
    
Δcm = −0.85 (1 − f ) sin θf δf = −0.85 (1 − f )2 (1 − f) f δf

(ref. 2, eq. 3.57; ref. 3, eq. 5.41); with the factor of 0.85 accounting for real flow effects (ref. 3, fig. 5.19).
The drag increment is estimated using

Sf
ΔCD = 0.9 1.38
f sin2 δf
S

for slotted flaps (ref. 2, eq. 3.51). In summary, the section load increments are:
cf
Δc = cα Lf ηf δf
c
Δcmax = Xf Δc
cf
Δcm = Mf δ f
c

The decrease in angle of attack for maximum lift is

Δc − Δcmax Δc

Δαmax = − = −(1 − Xf )
cα cα

The coefficients
ηf = 0.85 − 0.43|δf | = η0 − η1 |δf |
1  π 2/3
Lf = sin( f )
f 2
Xf = (1 − f ) (1 + 3 3f )
+ f −5 2
f

1  
Mf = −0.85 (1 − f )2 (1 − f ) f
f
1.38
Df = 0.9 f

follow from these equations.

For three-dimensional aerodynamic loads, these two-dimensional coefficients are corrected by
using the three-dimensional lift-curve slope, and multiplying by the ratio of flap span to wing span bf /b.
Then the wing load increments caused by flap deflection, in terms of coefficients based on the wing area,
are:
Sf
ΔCL = CLα Lf ηf δf
S ΔCLmax = Xf ΔCL
Sf
ΔCM = Mf δ f ΔCL
S Δαmax = −(1 − Xf )
Sf CLα
ΔCD = Df sin2 δf
S
where Sf /S is the ratio of flap area to wing area.

8–10 Drag

Each component can contribute drag to the aircraft. A fixed drag can be specified as a drag area
D/q ; or the drag can be scaled, specified as a drag coefficient CD based on an appropriate area S . There
may also be other ways to define a scaled drag value. For fixed drag, the coefficient is CD = (D/q)/S
(the aerodynamic model is formulated in terms of drag coefficient). For scaled drag, the drag area
Aircraft 87

is D/q = SCD . For all components, the drag (D/q)comp or CDcomp is defined for forward flight or
cruise; typically this is the minimum drag value. For some components, the vertical drag ((D/q)V comp or
CDV comp ) or sideward drag ((D/q)Scomp or CDScomp ) is defined. For some components, the aerodynamic
model includes drag due to lift, angle of attack, or stall.

Table 8-2 summarizes the component contributions to drag, and the corresponding reference areas.
If no reference area is indicated, then the input is only drag area D/q . An appropriate drag reference
area is defined for each component, and either input or calculated. Wetted area is calculated for each
component, even if it is not the reference area. The component wetted areas are summed to obtain the
aircraft wetted area. Some of the weight models also require the wetted area. The component drag
contributions must be consistent. In particular, a rotor with a spinner (such as on a tiltrotor aircraft)
would likely not have hub drag. The pylon is the rotor support and the nacelle is the engine support. The
drag model for a tiltrotor aircraft with tilting engines would use the pylon drag (and no nacelle drag),
since the pylon is connected to the rotor shaft axes; with non-tilting engines it would use the nacelle
drag as well.

Table 8-2. Component contributions to drag.

component drag contribution reference area
fuselage fuselage fuselage wetted area
fuselage vertical fuselage projected area
fittings and fixtures fuselage wetted area
rotor-body interference fuselage wetted area
contingency (aircraft) —
payload increment (flight state) —
landing gear landing gear —
rotor hub, hub vertical rotor disk area
duct, duct vertical duct wetted area
pylon, pylon vertical pylon wetted area
spinner spinner wetted area
wing wing, wing vertical wing planform area
wing-body interference wing planform area
tail tail, tail vertical tail planform area
fuel tank auxiliary tank (flight state) —
engine nacelle, nacelle vertical nacelle wetted area
momentum drag —
jet nacelle, nacelle vertical nacelle wetted area
momentum drag —
charger nacelle, nacelle vertical nacelle wetted area
momentum drag —

Optionally the aircraft drag can be fixed. The quantity specified is the sum (over all components)
of the drag area D/q (minimum drag, excluding drag due to lift and angle of attack), without accounting
for interference effects on dynamic pressure. The input parameter can be D/q ; or the drag can be scaled,
specified as a drag coefficient based on the rotor disk area, so D/q = Aref CD (where Aref is the reference
88 Aircraft

rotor disk area); or the drag can be estimated based on the gross weight, D/q = k(WM T O /1000)2/3
(where WM T O is the maximum takeoff gross weight; units of k are ft2 /k-lb2/3 or m2 /Mg2/3 ). Based
on historical data, the drag coefficient CD = 0.02 for old helicopters, CD = 0.008 for current low-drag
helicopters. Based on historical data, k = 9 for old helicopters, k = 2.5 for current low-drag helicopters,
k = 1.6 for current tiltrotors, and k = 1.4 for turboprop aircraft (English units). If the aircraft drag is
input, then the fuselage contingency drag is adjusted so the total aircraft D/q equals the input value.

Optionally the aircraft vertical drag (download fraction) can be fixed. The quantity specified is
the sum over all components of the vertical drag area (D/q)V . The input parameter can be (D/q)V , or
k = (D/q)V /Aref (where Aref is reference rotor disk area). Approximating the dynamic pressure in the
wake as q = 1/2ρ(2vh )2 = T /Aref , the download fraction is DL/T = q(D/q)V /T = k. If the aircraft
vertical drag is input, then the fuselage contingency vertical drag is adjusted so the total aircraft (D/q)V
equals the input value.

The nominal drag areas of the components and the aircraft are part of the aircraft description and
are used when the aircraft drag is fixed. The nominal drag area is calculated for low-speed helicopter
flight, for high-speed cruise flight, and for vertical flight. An incidence angle i is specified for the
rotors, wings, and nacelles, to be used solely to calculate the nominal helicopter and vertical drag areas.
The convention is that i = 0 if the component does not tilt. Table 8-3 summarizes the contributions
to the nominal drag areas, with D for the drag in normal flow and DV for the drag in vertical flow.
While vertical drag parameters are part of the aerodynamic model for the hub, duct, pylon, and nacelle,
aerodynamic interference at the rotor and at the propulsion components is not considered, so these terms
do not contribute to download. In the context of download, only the fuselage, wing, tail, and contingency
contribute to the nominal vertical drag.

From the input and the current aircraft size, the drag areas D/q and coefficients CD are calculated.
The aerodynamic analysis is usually in terms of coefficients. If the aircraft drag is fixed for the aircraft
model, then the fuselage contingency drag is set:


(D/q)cont = (D/q)fixed − (D/q)comp

and similarly for fixed vertical drag. Note that this adjustment ignores changes caused by interference
in the dynamic pressure and the velocity direction, which will affect the actual component drag.

The component aerodynamic model calculates the drag, typically from a drag coefficient CD , a

reference area, and the air velocity of the component. The drag force is then D = qcomp Sref CD , where
the dynamic pressure qcomp includes interference. From the aerodynamic forces and moments in wind
axes, the total force and moment in body axes (F F and M F ) are calculated. For reference, the aircraft
total drag and total drag area are

DAC = eTd Faero
F

(D/q)AC = DAC /q

where the aircraft velocity (without interference) gives the direction ed = −vAC
F F
/|vAC | and dynamic
pressure q = /2ρ|vAC | ; and Faero is the component aerodynamic force. An overall skin friction
1 F 2 F

drag coefficient is then CD AC = (D/q)ACwet /SAC , based on the aircraft wetted area SAC = Swet
and excluding drag terms not associated with skin friction (specifically landing gear, rotor hub, and
Aircraft 89

contingency drag).

Table 8-3. Component contributions to nominal drag area.

component drag contribution cruise helicopter vertical
fuselage fuselage D D DV
fittings and fixtures D D D
rotor-body int D D D
landing gear landing gear D D 0
retractable 0 D 0
rotor hub D D cos2 i + DV sin2 i 0
duct D D cos2 i + DV sin2 i 0
pylon D D cos2 i + DV sin2 i 0
spinner D D 0
wing wing D D cos2 i + DV sin2 i D sin2 i + DV cos2 i
wing-body int D D D
tail horizontal tail D D DV cos2 φ
tail vertical tail D D DV sin2 φ
engine nacelle D D cos2 i + DV sin2 i 0
jet nacelle D D cos2 i + DV sin2 i 0
charger nacelle D D cos2 i + DV sin2 i 0
contingency D D DV

8–11 Performance Metrics

The following

performance metrics are calculated for the aircraft. The aircraft hover figure of merit
is M = W W/2ρAref /P . The aircraft effective drag is De = P/V , hence the effective lift-to-drag ratio
is L/De = W V /P . For these metrics, the aircraft power is the sum of the engine group power, jet group
propulsive power, and charge group power: P = Preq + V Tjet + Pchrg . The aircraft power loading is
W/P (lb/hp or kg/kW). Isolated rotor performance metrics are described in Chapter 12.

8–12 Weights

The design gross weight WD is a principal parameter defining the aircraft, either input or determined
by the sizing task from designated design conditions and missions. Typically WD is the takeoff weight
for the primary design mission. The design gross weight is used by the analysis to calculate the rotor disk
loading and blade loading, wing loading, power loading, and thrust loading; to obtain aircraft moments
of inertia from the radii of gyration; for tolerance and perturbation scales of the solution procedures;
optionally to define structural design gross weight and maximum takeoff weight; and optionally to
specify the gross weight for missions and flight conditions.

The aircraft weight statement defines the empty weight, fixed useful load, and operating weight
for the design configuration. The aircraft weight statement is the sum of the weight statements for all
the aircraft components, the component weight determined by input or by parametric calculations with
90 Aircraft

technology factors. The definitions of the weight terms are as follows:

gross weight WG = WE + WU L = WO + Wpay + Wfuel

operating weight WO = WE + WF U L
useful load WU L = WF U L + Wpay + Wfuel

where WE is the weight empty; WF U L the fixed useful load; Wpay the payload weight; and Wfuel the
usable fuel weight. Aircraft weight definitions are given in SAWE RP7D (ref. 4), including:
Payload is any item which is being transported and is directly related to the purpose
of the flight as opposed to items that are necessary for the flight operation. Payload
can include, but is not limited to, passengers, cargo, passenger baggage, ammo,
internal and external stores, and fuel which is to be delivered to another aircraft or
site. Payload may or may not be expended in flight.
Operating weight is the sum of aircraft weight empty and operating items. Operating
weight is equivalent to takeoff gross weight less usable fuel, payload, and any item
to be expended in flight.
Weight empty is an engineering term which is defined as the weight of the complete
aircraft as defined in the aircraft specifications, dry, clean, and empty except for
fluids in closed systems such as a hydraulic system.
The weight empty consists of structure, propulsion group, systems and equipment, vibration, and con-
tingency weights. The weight empty WE can be calculated, or input, or scaled with design gross weight:
WE = dW E + fW E WD . Fixed or scaled weight empty is implemented by adjusting the contingency
weight so WE equals the required value. If the design gross weight is input, then the payload or fuel
weight must be fallout.
The structural design gross weight WSD and maximum takeoff weight WM T O can be input, or
specified as an increment d plus a fraction f of a weight W :

dSDGW + fSDGW WD
WSD = dSDGW + fSDGW W = dSDGW + fSDGW (WD − Wfuel + ffuel Wfuel−cap )
dSDGW + fSDGW WM T O
dW M T O + fW M T O WD
WM T O = dW M T O + fW M T O W = dW M T O + fW M T O (WD − Wfuel + Wfuel−cap )
dW M T O + fW M T O WSD

This convention allows the weights to be input directly (f = 0), or scaled with a design weight. For
WSD , W is the design gross weight WD , or WD adjusted for a specified fuel state (input fraction of fuel
capacity), or the maximum takeoff weight WM T O . Alternatively, WSD can be calculated as the gross
weight at a designated sizing flight condition. The structural design gross weight is used in the weight
estimation. For WM T O , W is the design gross weight WD , or WD adjusted for maximum fuel capacity,
or the structural design gross weight WSD . Alternatively, WM T O can be calculated as the maximum
gross weight possible at a designated sizing flight condition. The maximum takeoff weight is used in
the cost model, in the scaled aircraft and hub drag, and in the weight estimation.
The design ultimate load factor nzult at the structural design gross weight WSD is specified, in
particular for use in the component weight estimates. The structural design gross weight WSD and
Aircraft 91

design ultimate load factor nzult are used for the fuselage, rotor, and wing weight estimations. The
maximum takeoff weight WM T O is used for the cost and drag (scaled aircraft and hub), and for the
weights (system, fuselage, landing gear, and engine group).
The gross weight WG is specified for each flight condition and mission, perhaps in terms of the
design gross weight WD . For a each flight state, the fixed useful load may be different from the design
configuration because of changes in auxiliary fuel tank weight, or kit weights, or increments in crew or
equipment weights. Thus the fixed useful load weight is calculated for the flight state; and from it the
useful load weight and operating weight are calculated. The gross weight, payload weight, and usable
fuel weight (in standard and auxiliary tanks) complete the weight information for the flight state.
For each weight group, fixed (input) weights can be specified; or weight increments dW added to
the results of the parametric weight model. The parametric weight model includes technology factors
χ. Thus typically a component or element weight is obtained from W = χWmodel + dW . Weight of
individual elements in a group can be fixed by using dW and setting the corresponding technology factor
χ = 0. With χ = 0, the increment dW can account for something not included in the parametric model.

For scaled weights of all components, the AFDD weight models are implemented. The user can
incorporate custom weight models as well.
The operating weight is composed of scaled and fixed weights, so the design gross weight can be
written WD = WO + Wpay + Wfuel = WOfixed + WOscaled + Wpay + Wfuel . The growth factor is the change
in gross weight due to a change in payload:
 
∂WD ∂WOscaled ∂Wfuel ∂WOscaled ∂Wfuel ∂WD
=1+ + =1+ +
∂Wpay ∂Wpay ∂Wpay ∂WD ∂WD ∂Wpay
 
∼ W Oscaled W fuel ∂W D 1
=1+ + =
WD WD ∂Wpay 1 − φOscaled − φfuel

in terms of the weight fractions φ = W/WD .

The aircraft operating weight can be divided into core vehicle weight and military load. The core
vehicle weight is the weight in minimum airworthy state, with the aircraft capable of normal flight
throughout the envelope, but not mission capable. Military load is the sum of fixed useful load and
military features in weight empty. Thus

weight empty = core vehicle weight + military features

military load = fixed useful load + military features in weight empty
operating weight = WE + WF U L = core vehicle weight + military load

In terms of the weight breakdown used here, military features in weight empty consist of folding weight
(wing, rotor, tail, fuselage terms), crashworthiness weight (fuselage, landing gear terms), marinization
weight (fuselage), rotor brake (drive system), avionics group (mission equipment), armament group,
furnishings and equipment group, anti-icing group (including electrical group term), and load and
handling group.

8–13 Weight Statement

Aircraft weight information is stored in a data structure that follows SAWE RP8A Group Weight
Statement format (ref. 5), as outlined in figure 8-3. The asterisks designate extensions of RP8A for
92 Aircraft

the purposes of this analysis. Typically only the lowest elements of the hierarchy are specified; higher
elements are obtained by summation. Fixed (input) weight elements are identified in the data structure.
A weight statement data structure exists for each component. The aircraft weight statement is the sum
of the structures from all components.

8–14 References

1) Krabal, R.J.; Leo, R.N.; Ruble, J.R.; Dienstberger, C.F., Jr. “Advanced Propulsion Technology
Assessment for an Externally Blown Flap Transport.” AFAPL TR 72-1, April 1972.
2) McCormick, B.W. Aerodynamics, Aeronautics, and Flight Mechanics. New York: John Wiley &
Sons, Second Edition, 1995.
3) Kuethe, A.M., and Chow, C.-Y. Foundations of Aerodynamics. New York: John Wiley & Sons, Fifth
Edition, 1998.
4) “Mass Properties Management and Control for Military Aircraft, Revision D.” Society of Allied
Weight Engineers, Recommended Practice Number 7, May 2004.
5) “Weight and Balance Data Reporting Forms for Aircraft (including Rotorcraft), Revision A.” Society
of Allied Weight Engineers, Recommended Practice Number 8, June 1997.
Aircraft 93

WEIGHT EMPTY
STRUCTURE
wing group
basic structure
secondary structure
fairings (*), fittings (*), fold/tilt (*)
control surfaces
rotor group
blade assembly
hub & hinge
basic (*), fairing/spinner (*), blade fold (*), shaft (*)
rotor support structure (*), duct (*)
empennage group
horizontal tail (*)
basic (*), fold (*)
vertical tail (*)
basic (*), fold (*)
tail rotor (*)
blades, hub & hinge, rotor supports, rotor/fan duct
fuselage group
basic (*)
wing & rotor fold/retraction (*)
tail fold/tilt (*)
marinization (*)
pressurization (*)
crashworthiness (*)
alighting gear group
basic (*), retraction (*), crashworthiness (*)
engine section or nacelle group
engine support (*), engine cowling (*), pylon support (*)
air induction group
PROPULSION GROUP
engine system
engine
exhaust system
accessories (*)
propeller/fan installation
blades (*), hub & hinge (*), rotor supports (*), rotor/fan duct (*)
fuel system
tanks and support
plumbing
drive system
gear boxes
transmission drive
rotor shaft
rotor brake (*)
clutch (*)
gas drive
Figure 8-3a. Weight statement (* indicates extension of RP8A).
94 Aircraft

SYSTEMS AND EQUIPMENT

flight controls group
cockpit controls
automatic flight control system
system controls
fixed wing systems
non-boosted (*), boost mechanisms (*)
rotary wing systems
non-boosted (*), boost mechanisms (*), boosted (*)
conversion systems
non-boosted (*), boost mechanisms (*)
auxiliary power group
instruments group
hydraulic group
fixed wing (*), rotary wing (*), conversion (*)
equipment (*)
pneumatic group
electrical group
aircraft (*), anti-icing (*)
avionics group (mission equipment)
armament group
armament provisions (*), armor (*)
furnishings & equipment group
environmental control group
anti-icing group
load & handling group
VIBRATION (*)
CONTINGENCY
FIXED USEFUL LOAD
crew
fluids (oil, unusable fuel) (*)
auxiliary fuel tanks
other fixed useful load (*)
equipment increment (*)
folding kit (*)
wing extension kit (*)
wing kit (*)
other kit (*)
PAYLOAD
USABLE FUEL
standard tanks (*)
auxiliary tanks (*)

OPERATING WEIGHT = weight empty + fixed useful load

USEFUL LOAD = fixed useful load + payload + usable fuel

GROSS WEIGHT = weight empty + useful load

GROSS WEIGHT = operating weight + payload + usable fuel
Figure 8-3b. Weight statement (* indicates extension of RP8A).
 Chapter 9

Systems

The systems component contains weight information (fixed useful load, vibration, contingency,
and systems and equipment).

9–1 Weights

The weight empty consists of structure, propulsion group, systems and equipment, vibration, and
contingency weights. The vibration control weight can be input, or specified as a fraction of weight
empty: Wvib = fvib WE . The contingency weight can be input, or specified as a fraction of weight empty:
Wcont = fcont WE . However, if the weight empty is input, then the contingency weight is adjusted so
WE equals the required value. The weights of all components are evaluated and summed, producing the
aircraft weight empty less vibration and contingency weight, WX . Then:
a) Fixed or scaled weight empty: Wvib input or Wvib = fvib WE , Wcont = WE −
(WX + Wvib ).
b) Both fractional: WE = WX /(1 − fvib − fcont ), Wvib = fvib WE , Wcont = fcont WE .
c) Only vibration weight fractional: Wcont input, WE = (WX + Wcont )/(1 − fvib ),
Wvib = fvib WE .
d) Only contingency weight fractional: Wvib input, WE = (WX + Wvib )/(1 − fcont ),
Wcont = fcont WE .
e) Both input: Wvib and Wcont input, WE = WX + Wvib + Wcont .
Finally, the operating weight WO = WE + WF U L is recalculated.
Systems and equipment includes the following fixed (input) weights: auxiliary power group, in-
struments group, pneumatic group, electrical group (aircraft), avionics group (mission equipment),
armament group (armor and armament provisions), furnishings and equipment group, environmental
control group, and load and handling group. Systems and equipment includes the following scaled
weights: flight controls group, hydraulic group, electrical group (anti-icing), and anti-icing group.
Flight controls group includes the following fixed (input) weights: cockpit controls and automatic
flight control system. Flight controls group includes the following scaled weights: fixed wing systems,
rotary wing systems, and conversion or thrust vectoring systems. Rotary wing flight control weights can
be calculated for the entire aircraft (using rotor parameters such as chord and tip speed for a designated
rotor), an approach that is consistent with parametric weight equations developed for conventional two-
rotor configurations. Alternatively, rotary wing flight control weights can be calculated separately for
each rotor and then summed. The fixed wing flight controls and the conversion controls can be absent.
The fixed useful load WF U L consists of crew (Wcrew ), trapped fluids (oil and unusable fuel, Wtrap ),
auxiliary fuel tanks (Wauxtank ), equipment increment, kits (folding, wing, wing extension, other), and
96 Systems

other fixed useful load (WF U Lother ). Wcrew , Wtrap , and WF U Lother are input. For a each flight state, the
fixed useful load may be different from the design configuration because of changes in auxiliary fuel
tank weight, kit weight, and crew or equpment weight increments.

Folding weights can be calculated in several weight groups, including wing, rotor, empennage, and
fuselage. These weights are the total weights for folding and the impact of folding on the group. A
fraction ffoldkit of these weights can be in a kit, hence optionally removable. Thus, of the total folding
weight, the fraction ffoldkit is a kit weight in the fixed useful load of the weight statement, while the
remainder is kept in the component group weight.

9–2 Detailed Weight Definition

Figure 9-1 presents a more detailed description of the weights for the systems and equipment, and
the useful load. The input information for a group can be just the total weight (ΔW in the following
equations), or can include all the terms.

The electrical group consists of power supply, power conversion, power distribution and controls,
lights and signal devices, and equipment supports; plus a term associated with the anti-icing group. The
total weight is

Welectrical = Welect aircraft + WDIelect

Welect aircraft = Wsupply + Wconv + Wdistrib + Wlights + Wsupport + ΔWelectrical

The avionics group consists of equipment and installation; here the equipment weights include installa-
tion. The equipment items are communications, navigation, identification, control and display, aircraft
survivability, and mission system equipment; plus armament electronics. The total weight is

WMEQ = Wcom + Wnav + Wident + Wdisplay + Wsurvive + Wmission + ΔWMEQ + Warm elect

The armament group consists of armament provisions (gun provisions, turret systems, expendable
weapons provisions) and armor. Armament electronics weight Warm elect (such as targeting, sights,
radar) is part of the avionics group.

Warm prov = Wgun + Wturret + Wexpend + ΔWarm prov

Warmor = Uarmor floor Scabin floor + Uarmor wall Scabin wall + Uarmor crew Ncrew seat + ΔWarmor

Here Uarmor floor , Uarmor wall are the armor weights per surface area; Scabin floor and Scabin wall are the
cabin floor and wall areas; Uarmor crew is the armor weight per crew; and Ncrew seat is number of crew
seats.

The furnishings and equipment group consists of accommodation for personnel, miscellaneous
equipment, furnishings, and emergency equipment. Accommodation for personnel consists of seats,
miscellaneous accommodation (including galleys, toilets), and the oxygen system. Miscellaneous equip-
ment includes cockpit displays. Furnishings includes floor covering, trim, partitions, crash padding, and
acoustic and thermal insulation; but excludes vibration absorbers. Emergency equipment consists of fire
detection and extinguishing, and other emergency equipment (including first aid, survival kit, and life
Systems 97

raft). The total weight is

Wfurnish = Waccom + Wmisc + Wfurn + Wemerg + ΔWfurnish

Waccom = (Useat crew + Uaccom crew + Uox crew )Ncrew seat

+ (Useat pass + Uaccom pass + Uox pass )Npass seat
Wfurn = Wtrim + Uinsulation Scabin
Wemerg = Wfire + Wother

Here Uzz crew , Uzz pass are the weights per crew and passenger; Ncrew seat , Npass seat are the number of
crew and passenger seats; Uinsulation is the acoustic and thermal insulation weight per area; and Scabin is
the total cabin surface area. The load and handling group consists of aircraft handling and load handling
(cargo handling, hoist, external load provisions). The total weight is

Wload = Waircraft + Wcargo + ΔWload

Wcargo = Uhandling Scabin floor + Whoist + Wext prov

Here Uhandling is the cargo handling weight per cabin floor area, and Scabin floor is the cabin floor area.
The crew weight is
Wcrew = Ucrew Ncrew + ΔWcrew

Here Ucrew is the weight per crew, and Ncrew is the number of crew. A crew weight increment for a
flight condition or mission is given by δNcrew and δWcrew . Other fixed useful load consists of various
categories, such as baggage, gun installations, weapons provisions, aircraft survivability equipment
(chaff, flares), survival kits, life rafts, and oxygen.
An equipment increment can be defined for a flight condition or mission, in terms of δNcrew seat ,
δNpass seat , and δWequip :

Wequip inc = Ucrew seat inc δNcrew seat + Upass seat inc δNpass seat + δWequip

The default weights per crew and passenger seats are Ucrew seat inc = Useat crew + Uaccom crew + Uox crew +
Uarmor crew and Upass seat inc = Useat pass + Uaccom pass + Uox pass .

The payload consists of passengers or troops, cargo (internal and external), ammunition, and
weapons:

Wpayload = Upass Npass + Wcargo + Wext load + Wammo + Wweapons + ΔWpayload

Here Upass is the weight per passenger, and Npass is the number of passengers. A payload increment is
defined by ΔWpayload . For fallout payload, the value of ΔWpayload is adjusted.
98 Systems

WEIGHT EMPTY
SYSTEMS AND EQUIPMENT
electrical group
aircraft
power supply
power conversion
power distribution and controls
lights and signal devices
equipment supports
anti-icing
avionics group (mission equipment)
equipment
installation
armament group
armament provisions
armor
furnishings & equipment group
accommodation for personnel
seats
miscellaneous accommodation
oxygen system
miscellaneous equipment
furnishings
emergency equipment
fire detection and extinguishing
other emergency equipment
load & handling group
aircraft handling
load handling
USEFUL LOAD
FIXED USEFUL LOAD
crew
other fixed useful load
various categories
equipment increment
PAYLOAD
passengers/troops
cargo
ammunition
weapons
Figure 9-1. Details of weight descriptions (based on RP8A).
 Chapter 10

Fuselage

There is one fuselage component for the aircraft.

10–1 Geometry

The fuselage length fus can be input or calculated. The calculated length depends on the longitudinal
positions of all components. Let xmax and xmin be the maximum (forward) and minimum (aft) position
of all rotors, wings, and tails. Then the calculated fuselage length is

fus = nose + (xmax − xmin ) + aft

The nose length nose (distance forward of hub) and aft length aft (distance aft of hub) are input, or
calculated as nose = fnose L and aft = faft L. Typically faft = 0 or negative for the main-rotor and
tail-rotor configuration, and faft = 0.75 for the coaxial configuration. The fuselage width wfus is input.
The reference length L is the rotor radius or wing span of a designated component, or the input fuselage
length.
The fuselage wetted area Swet (reference area for drag coefficients) and projected area Sproj (ref-
erence area for vertical drag) are input (excluding or including the tail boom terms); or calculated from
the nose length:
Swet = fwet (2 nose hfus +2 nose wfus + 2hfus wfus ) + Cboom L
Sproj = fproj ( nose wfus ) + wboom L

using input fuselage height hfus , and factors fwet and fproj ; or calculated from the fuselage length:

Swet = fwet (2 fus hfus +2 fus wfus + 2hfus wfus ) + Cboom L

Sproj = fproj ( fus wfus ) + wboom L

Using the nose length and the tail boom area is probably best for a single-main-rotor and tail-rotor
helicopter. Here Cboom is the effective tail boom circumference (boom wetted area divided by reference
length), and wboom is the effective tail boom width (boom vertical area divided by reference length).
Alternatively, the wetted area and projected area can be estimated based on the gross weight:

Swet = kwet (WD /1000)2/3

Sproj = kproj (WD /1000)2/3

where WD is the design gross weight; units of k are ft2 /k-lb2/3 or m2 /Mg2/3 .
Cabin areas are required for weight estimates: total cabin surface area Scabin for acoustic and
thermal insulation weight (furnishings and equipment group); cabin floor area Scabin floor for armor and
100 Fuselage

cargo handling weights; and cabin wall area Scabin wall for armor weight. These areas are input, or
calculated from the fuselage dimensions:

Scabin = fcabin (2 fus hfus +2 fus wfus )

Scabin floor = ffloor ( fus wfus )

Scabin wall = fwall (2 fus hfus )

with typically f ∼
= 0.6.

The fuselage contribution to the aircraft operating length is xfus + fref fus (forward) and xfus − (1 −
fref )fus (aft). Here fref is the position of the fuselage aerodynamic reference location aft of the nose, as
a fraction of the fuselage length. If the fuselage length is input, then fref is input; if the fuselage length
is calculated, then fref = (xmax + nose − xfus )/ fus .

10–2 Control and Loads

The fuselage has a position z F , where the aerodynamic forces act; and the component axes are
aligned with the aircraft axes, C BF = I . The fuselage has no control variables.

10–3 Aerodynamics

The aerodynamic velocity of the fuselage relative to the air, including interference, is calculated in
component axes, v B . The angle of attack αfus , sideslip angle βfus (hence C BA ), and dynamic pressure q
are calculated from v B . The reference area for the fuselage forward flight drag is the fuselage wetted area
Swet , which is input or calculated as described previously. The reference area for the fuselage vertical
drag is the fuselage projected area Sproj , which is input or calculated as described previously.

10-3.1 Drag
The drag area or drag coefficient is defined for forward flight, vertical flight, and sideward flight.
In addition, the forward flight drag area or drag coefficient is defined for fixtures and fittings, and for
rotor-body interference. The effective angle of attack is αe = αfus − αDmin , where αDmin is the angle
of minimum drag; in reverse flow (|αe | > 90), αe ← αe − 180 signαe . For angles of attack less than a
transition angle αt , the drag coefficient equals the forward flight (minimum) drag CD0 , plus an angle of
attack term. Thus if |αe | ≤ αt
CD = CD0 (1 + Kd |αe |Xd )

and otherwise
CDt = CD0 (1 + Kd |αt |Xd )
   
Sproj π |αe | − αt
CD = CDt + CDV − CDt sin
Swet 2 π/2 − αt
and similarly for the transition of payload drag (D/q)pay and contingency drag (D/q)cont . Optionally
there might be no angle-of-attack variation at low angles (Kd = 0), or quadratic variation (Xd = 2). With
an input transition angle, there will be a jump in the slope of the drag coefficient at αt . For a smooth
transition, the transition angle that matches slopes as well as coefficients is found by solving
 
2Xd (Sproj /Swet )CDV − CD0
− 1 αtXd − Xd αtXd −1 + =0
π Kd CD0
Fuselage 101

This calculation of the transition angle is only implemented with quadratic variation, for which
⎛  ⎞
1 ⎝ (Sproj /Swet )CDV − CD0 ⎠
αt = 1+ 1−a
a Kd CD0

with a = (4/π) − 1; αt is however required to be between 15 and 45 deg. For sideward flight (vxB = 0) the
drag is obtained using φv = tan−1 (−vzB /vyB ) to interpolate between sideward and vertical coefficients:

Sproj
CD = CDS cos2 φv + CDV sin2 φv
Swet

Then the drag force is

    
D = qSwet CD + CDfit + CDrb + q (D/q)pay + (D/q)cont

including drag coefficient for fixtures and fittings CDfit and rotor-body interference CDrb (summed over
all rotors); drag area of the payload (specified for flight state); and contingency drag area.

10-3.2 Lift and Pitch Moment

The fuselage lift and pitch moment are defined in fixed form (L/q and M/q ), or scaled form (CL
and CM , based on the fuselage wetted area and fuselage length). The effective angle of attack is
αe = αfus − αzl , where αzl is the angle of zero lift; in reverse flow (|αe | > 90), αe ← αe − 180 signαe . Let
αmax be the angle-of-attack increment (above or below zero lift angle) for maximum lift. If |αe | ≤ αmax

CL = CLα αe
CM = CM 0 + CM α αe

and otherwise  
π/2 − |αe |
CL = CLα αmax signαe
π/2 − |αmax |
 
π/2 − |αe |
CM = (CM 0 + CM α αmax signαe )
π/2 − |αmax |
for zero lift and moment at 90 deg angle of attack. In sideward flight, these coefficients are zero. Then
L = qSwet CL and M = qSwet fus CM are the lift and pitch moment.

10-3.3 Side Force and Yaw Moment

The fuselage side force and yaw moment are defined in fixed form (Y /q and N/q ), or scaled form
(CY and CN , based on the fuselage wetted area and fuselage length). The effective sideslip angle is
βe = βfus −βzy , where βzy is the angle of zero side force; in reverse flow (|βe | > 90), βe ← βe −180 signβe .
Let βmax be the sideslip angle increment (above or below zero side force angle) for maximum side force.
If |βe | ≤ βmax
CY = CY β βe
CN = CN 0 + CN β βe
and otherwise  
π/2 − |βe |
CY = CY β βmax signβe
π/2 − |βmax |
 
π/2 − |βe |
CN = (CN 0 + CN β βmax signβe )
π/2 − |βmax |
102 Fuselage

for zero side force and yaw moment at 90 deg sideslip angle. Then Y = qSwet CY and N = qSwet fus CN
are the side force and yaw moment. The roll moment is zero.

10–4 Weights

The fuselage group consists of the basic structure; wing and rotor fold/retraction; tail fold/tilt; and
marinization, pressurization, and crashworthiness structure.
 Chapter 11

Landing Gear

There is one landing gear component for the aircraft. The landing gear can be located on the body
or on the wing. The landing gear can be fixed or retractable; a gear retraction speed is specified (CAS),
or the landing gear state can be specified in the flight state.

11–1 Geometry

The landing gear has a position z F , where the aerodynamic forces act. The component axes are
aligned with the aircraft axes, C BF = I . The landing gear has no control variables. The height of the
bottom of the landing gear above ground level, hLG , is specified in the flight state. The landing gear
position z F is a distance dLG above the bottom of the gear.

11-1.1 Drag
The drag area is specified for landing gear extended, (D/q)LG . The velocity relative to the air at
z gives the drag direction ed = −v F /|v F | and dynamic pressure q = 1/2ρ|v F |2 (no interference). Then
F

F F = ed q(D/q)LG

is the total drag force.

11–2 Weights

The alighting gear group consists of basic structure, retraction, and crashworthiness structure.
104 Landing Gear
 Chapter 12
Rotor

The aircraft can have one or more rotors, or no rotors. In addition to main-rotors, the rotor component
can model tail-rotors, propellers, proprotors, ducted fans, thrust vectoring rotors, and auxiliary-thrust
rotors. The principal configuration designation (main-rotor, tail-rotor, or propeller) is identified for
each rotor component, and in particular determines where the weights are put in the weight statement
(summarized in table 12-1). Each configuration can possibly have a separate performance or weight
model, which is separately specified. Antitorque rotors and auxiliary-thrust rotors can be identified, for
special sizing options. Other configuration features are variable diameter and ducted fan, and reaction
drive.
Multi-rotor systems (such as coaxial or tandem configuration) are modeled as a set of separate rotors,
in order to accommodate the description of the position, orientation, controls, and loads. Optionally
the location of the center of the rotor system can be specified and the rotor locations calculated based
on input separation parameters. The performance calculation for twin and multiple rotor systems can
include the mutual influence of the induced velocity on the power.
The main-rotor size is defined by the radius R or disk loading W/A, thrust-weighted solidity σ ,
2
hover tip speed Vtip , and blade loading CW /σ = W/ρAVtip σ . With more than one main-rotor, the disk
loading and blade loading are obtained from an input fraction of design gross weight, W = fW WD . The
air density ρ for CW /σ is obtained from a specified takeoff condition. If the rotor radius is fixed for
the sizing task, three of (R or W/A), CW /σ , Vtip , and σ are input, and the other parameters are derived.
Optionally the radius can be calculated from a specified ratio to the radius of another rotor. If the sizing
task determines the rotor radius (R and W/A), then two of CW /σ , Vtip , and σ are input, and the other
parameter is derived. The radius can be sized for just a subset of the rotors, with fixed radius for the
others.
For antitorque and auxiliary-thrust rotors, three of (R or W/A), CW /σ , Vtip , and σ are input, and
the other parameters are derived. Optionally the radius can be calculated from a specified ratio to the
radius of another rotor. The disk loading and blade loading are based on f T , where f is an input factor
and T is the maximum thrust from designated design conditions. Optionally the tail-rotor radius can be
scaled with the main-rotor radius: R = f Rmr (0.1348 + 0.0071W/A), where f is an input factor and the
units of disk loading W/A are lb/ft2 . Figure 12-1 shows the basis for this scaling.

Table 12-1. Principal configuration designation.

configuration weight statement weight model performance model
main-rotor rotor group rotor rotor
tail-rotor empennage group tail-rotor rotor
propeller propulsion group rotor, aux thrust rotor
106 Rotor

0.26
aircraft
equation
0.24

0.22

0.20
R tr/Rmr

0.18

0.16

0.14

0.12
0. 2. 4. 6. 8. 10. 12. 14. 16.

disk loading (lb/ft 2)

Figure 12-1. Tail-rotor radius scaling.

12–1 Drive System

The drive system defines gear ratios for all the components it connects. The gear ratio is the ratio
of the component rotational speed to that of the primary rotor. There is one primary rotor per propulsion
group (for which the reference tip speed is specified); other components are dependent (for which a gear
ratio is specified). There can be more than one drive system state, in order to model a multiple-speed or
variable-speed transmission. Each drive system state corresponds to a set of gear ratios.
For the primary rotor, a reference tip speed Vtip−ref is defined for each drive system state. By
convention, the “hover tip speed” refers to the reference tip speed for drive state #1. If the sizing
task changes the hover tip speed, then the ratios of the reference tip speeds at different engine states
are kept constant. By convention, the gear ratio of the primary rotor is r = 1. For dependent rotors,
either the gear ratio is specified (for each drive system state) or a tip speed is specified and the gear
ratio is calculated (r = Ωdep /Ωprim , Ω = Vtip−ref /R). For the engine group, either the gear ratio is
specified (for each drive system state) or the gear ratio is calculated from the specification engine
turbine speed Ωspec = (2π/60)Nspec and the reference tip speed of the primary rotor (r = Ωspec /Ωprim ,
Ωprim = Vtip−ref /R). The latter option means the specification engine turbine speed Nspec corresponds
to Vtip−ref for all drive system states. To determine the gear ratios, the reference tip speed and radius are
used, corresponding to hover.
Rotor 107

The flight state specifies the tip speed of the primary rotor and the drive system state, for each
propulsion group. The drive system state defines the gear ratio for dependent rotors and the engine
groups. From the rotor radius, the rotational speed of the primary rotor is obtained (Ωprim = Vtip /R);
from the gear ratios, the rotational speed of dependent rotors (Ωdep = rΩprim ) and the engine groups
(N = (60/2π)reng Ωprim ) are obtained; and from the rotor radius, the tip speed of the dependent rotor
(Vtip = Ωdep R) is obtained. The flight state specification of the tip speed can be an input value, the
reference tip speed, a function of flight speed or a conversion schedule, or one of several default values.
These relationships between tip speed and rotational speed use the actual radius of the rotors in the flight
state, which for a variable-diameter rotor may not be the same as the reference, hover radius.
A designated drive system state can have a variable speed (variable gear ratio) transmission, by
introducing a factor fgear on the gear ratio when the speeds of the dependent rotors and engines are
evaluated. The factor fgear is a component control, which can be connected to an aircraft control and
thus set for each flight state.
An optional conversion schedule is defined in terms of two speeds: hover and helicopter mode for
speeds below VChover , cruise mode for speeds above VCcruise , and conversion mode for speeds between
VChover and VCcruise . The tip speed is Vtip−hover in helicopter and conversion mode, and Vtip−cruise in
airplane mode. Drive system states are defined for helicopter, cruise, and conversion mode flight. The
flight state specifies the nacelle tilt angle, tip speeds, control state, and drive system state, including the
option to obtain any or all of these quantities from the conversion schedule.
Several default values of the tip speed are defined for use by the flight state, including cruise,
maneuver, one-engine inoperative, drive system limit conditions, and a function of flight speed (piecewise
linear input). Optionally these default values can be input as a fraction of the hover tipspeed. Optionally
speed can be calculated from μ = V /Vtip , so Vtip = V /μ; or from Mat = Mtip (1 + μ)2 + μ2z , so
the tip 
Vtip = (cs Mat )2 − Vz2 − V . Optionally the tip speed can be the minimum of the input value or that for
Mat .

The sizing task might change the hover tip speed (reference tip speed for drive system state #1),
the reference tip speed of a dependent rotor, a rotor radius, or the specification engine turbine speed
Nspec . In such cases the gear ratios and other parameters are recalculated. Note that it is not consistent
to change the reference tip speed of a dependent rotor if the gear ratio is a fixed input.
An increment on the primary rotor rotational speed (or primary engine group, if there are no rotors)
is a control variable of the propulsion group.

12–2 Geometry

The rotor rotation direction is described by the parameter r: r = 1 for counter-clockwise rotation
and r = −1 for clockwise rotation (as viewed from the positive thrust side of the rotor).
The rotor solidity and blade mean chord are related by σ = N c/πR; usually thrust-weighted values
are used, but geometric values are also required by the analysis. The mean chord is the average of
the chord over the rotor blade span, from root cutout to tip. The thrust-weighted chord is the average
of the chord over the rotor blade span r, from root cutout to tip, weighted by r2 . A general blade
chord distribution is specified as c(r) = cref ĉ(r), where cref is the thrust-weighted chord. Linear taper is
specified in terms of a taper ratio t = ctip /croot , or in terms of the ratio of thrust-weighted and geometric
chords, f = σt /σg = c.75R /c.50R .
108 Rotor
F
The rotor hub is at position zhub . Optionally, a component of the position can be calculated,
superseding the location input. The calculated geometry depends on the configuration. For a coaxial
rotor, the rotor separation is s = |kT C SF (zhub1
F
− zhub2
F
)/(2R)| (fraction rotor diameter), or the hub
locations are calculated from the input separation s, and the input location midway between the hubs:
⎛ ⎞
0
F
zhub F
= zcenter ± CF S ⎝ 0 ⎠
sR

For a tandem rotor, the rotor longitudinal overlap is o = Δ /(2R) = 1 − /(2R) (fraction rotor diameter),
or the hub locations are calculated from the input overlap o, and the input location midway between the
hubs:
xhub = xcenter ± R(1 − o)

For a tail-rotor, the longitudinal position can be calculated from the main-rotor radius R, tail-rotor radius
Rtr , and tail-rotor/main-rotor clearance dtr :

xhubtr = xhubmr − (Rmr + dtr + Rtr )

For a multicopter, the longitudinal and lateral position can be calculated from the rotor radius R, the
scaled arm length i of the i-th rotor, and the angle ψi (measured clockwise from the forward longitudinal
axis):
xhub = xcenter + R i cos ψi
yhub = ycenter + R i sin ψi
For a tiltrotor, the lateral position can be calculated from the rotor radius R (cruise value for variable-
diameter rotor), fuselage/rotor clearance dfus , and fuselage width wfus :

yhub = ± (f R + dfus + 1/2wfus )

with the pivot, pylon, and nacelle center-of-gravity lateral positions adjusted to keep the same relative
position to the hub. The calculated clearance between the rotor and fuselage is dfus = |yhub |−(R+ 1/2wfus ).
Alternatively, for a tiltrotor the lateral position can be calculated from the wing span, yhub = ± b/2, so
the rotors are at the wing tips, or from a designated wing panel edge, yhub = ± ηp (b/2).
For twin rotors (tandem, side-by-side, or coaxial), the overlap is o = Δ /(2R) = 1 − /(2R)
(fraction of diameter; 0 for no overlap and 1 for coaxial), where the hub-to-hub separation is =
[(xhub1 − xhub2 )2 + (yhub1 − yhub2 )2 ]1/2 ( = 2R for no overlap and = 0 for coaxial). The overlap area
is mA, with A the area of one rotor disk and
2 " −1  %
m= cos ( /2R) − ( /2R) 1 − ( /2R)2
π

The vertical separation is s = |zhub1 − zhub2 |/(2R).

The reference areas for the component drag coefficients are the rotor disk area A = πR2 (for hub
drag), pylon wetted area Spylon , duct wetted area 2Sduct , and spinner wetted area Sspin . The pylon wetted
area is input, or calculated from the drive system (gear box and rotor shaft) weight, or from the drive
system plus engine system (engine, exhaust, and accessories) weight:
 2/3
Spylon = k w/Nrotor
Rotor 109

where w = Wgbrs or w = Wgbrs + WES and the units of k are ft2 /lb2/3 or m2 /kg2/3 . The pylon area
is included in the aircraft wetted area if the pylon drag coefficient is nonzero. The duct wetted area is
twice the duct area Sduct . The duct area is input, or calculated from the rotor circumference and the duct
length duct = kR:
Sduct = (2πR) duct

The spinner wetted area is input, or calculated from the spinner frontal area:

2
Sspin = k(πRspin )

where Rspin is the spinner radius, which is specified as a fraction of the rotor radius.

The rotor contribution to the aircraft operating length and width is calculated from the locus of
the rotor disk: zdisk = zhub + RC F S (cos ψ sin ψ 0)T . The longitudinal distance from the hub position is
√
Δx = R(a cos ψ + b sin ψ), so the maximum distance is Δx = ±R a2 + b2 . The lateral distance from the
√
hub position is Δy = R(c cos ψ + d sin ψ), so the maximum distance is Δy = ±R c2 + d2 .

12–3 Control and Loads

The rotor controls consist of collective, lateral cyclic, longitudinal cyclic, and perhaps shaft inci-
dence (tilt) and cant angles. Rotor cyclic control can be defined in terms of tip-path plane or no-feathering
plane command. The collective control variable is the rotor thrust amplitude or the collective pitch angle.

The relationship between tip-path plane tilt and hub moment is M = N2 Ib Ω2 (ν 2 − 1)β = Khub β ,
where N is the number of blades, Ω the rotor speed, and ν the dimensionless fundamental flap frequency.
The flap moment of inertia Ib is obtained from the Lock number: γ = ρacR4 /Ib , for seal level standard
(SLS) density ρ, lift-curve slope a = 5.7, and thrust-weighted chord (or from the blade weight, or from
an autorotation index). The flap frequency and Lock number are specified for hover radius and rotational
speed. The flap frequency and hub stiffness are required for the radius and rotational speed of the flight
state. For a hingeless rotor, the blade flap spring is Kflap = Ib Ω2 (ν 2 − 1), obtained from the hover
quantities; then Khub = N2 Kflap and
Kflap
ν2 = 1 +
Ib Ω2
For an articulated rotor, the hinge offset is e = Rx/(1 + x), x = 23 (ν 2 − 1) from the hover quantities; then

3 e/R
ν2 = 1 +
2 1 − e/R

and Khub = N2 Ib Ω2 (ν 2 − 1), using Ib from γ (and scaled with R for a variable diameter rotor) and Ω for
the flight state.

Optionally the rotor can have a variable diameter. The rotor diameter is treated as a control, allowing
it to be connected to an aircraft control and thus set for each flight state. The basic variation can be
specified based on the conversion schedule, or input as a function of flight speed (piecewise linear input).
For the conversion schedule, the rotor radius is Rhover for speeds below VChover , Rcruise = f Rhover for
speeds above VCcruise , and linear with flight speed in conversion mode. During the diameter change, the
chord, chord radial distribution, and blade weight are assumed fixed; hence solidity scales as σ ∼ 1/R,
blade flap moment of inertia as Ib ∼ R2 , and Lock number as γ ∼ R2 .
110 Rotor

12-3.1 Control Variables

The collective control variable is direct command of rotor thrust magnitude T or CT /σ (in shaft
axes), from which the collective pitch angle can be calculated; or rotor collective pitch angle θ0.75 , from
which the thrust and inflow can be calculated.
Shaft tilt control variables are incidence (tilt) and cant angles, acting at a pivot location.
Tip-path plane command is direct control of the tip-path plane tilt, hence tilt of the thrust vector.
This control mode requires calculation of rotor cyclic pitch angles from the flapping. The control
variables are longitudinal tilt βc (positive forward) and lateral tilt βs (positive toward retreating side).
Alternatively, the cyclic control can be specified in terms of hub moment or lift offset, if the blade
flap frequency is greater than 1/rev. The relationship between tip-path plane tilt and hub moment is
M = Khub β , and between moment and lift offset is M = o(T R). Thus the flapping is
     
βs 1 rMx TR ox
= =
βc Khub −My Khub −oy
for hub moment command or lift offset command, respectively.
No-feathering plane command is control of rotor cyclic pitch angles, usually producing tilt of the
thrust vector. This control mode requires calculation of rotor tip-path plane tilt from the cyclic control,
including the influence of inflow. The control variables are longitudinal cyclic pitch angle θs (positive
aft) and lateral cyclic pitch angle θc (positive toward retreating side).

12-3.2 Aircraft Controls

Each control can be connected to the aircraft controls cAC : c = c0 +ST cAC , with c0 zero, constant, or
a function of flight speed (piecewise linear input). The factor S can be introduced to automatically scale
2
the collective matrix: S = a/6 = 1/60 if the collective control variable is CT /σ ; S = ρVtip Ablade (a/6)
if the collective control variable is rotor thrust T ; S = 1 if the collective control variable is pitch angle
θ0.75 . For cyclic matrices, S = 1 with no-feathering plane command, and S = −1 for tip-path plane
command.

12-3.3 Rotor Axes and Shaft Tilt

F
The rotor hub is at position zhub , where the rotor forces and moments act; the orientation of the rotor
shaft axes relative to the aircraft axes is given by the rotation matrix C SF . The pivot is at position zpivot
F
.
The hub or shaft axes S have origin at the hub node; the z -axis is the shaft, positive in the positive thrust
direction; and the x-axis downstream or up. The rotor orientation is specified by selecting a nominal
direction in body axes (positive or negative x-, y -, or z -axis) for the positive thrust direction; the other
two axes are then the axes of control. For a main-rotor, the nominal direction would be the negative
z -axis; for a tail-rotor, it would be the lateral axis (ry -axis, depending on the direction of rotation of
the main-rotor); and for a propeller, the nominal direction would be the positive x-axis. This selection
defines a rotation matrix W from F to S axes. The hub and pivot axes have a fixed orientation relative
to the body axes:
hub incidence and cant: C HF = Uθh Vφh
pivot dihedral, pitch, and sweep: C P F = Xφp Yθp Zψp
where U and V depend on the nominal direction, as described in table 12-2. The shaft control consists
of incidence and cant about the pivot axes, from reference angles iref and cref :
Ccont = Ui−iref Vc−cref
Rotor 111

For a tiltrotor aircraft, one of the aircraft controls is the nacelle angle, with the convention αtilt = 0 for
cruise, and αtilt = 90 deg for helicopter mode. The rotor shaft incidence angle is then connected to
αtilt by defining the matrix Ti appropriately. For the locations and orientation input in helicopter mode,
iref = 90. Thus the orientation of the shaft axes relative to the body axes is:

C SF = W C HF C F P Ccont C P F

or just C SF = W C HF with no shaft control. From the pivot location zpivot

F
and the hub location for the
F
reference shaft control zhub0 , the hub location in general is

F
zhub F
= zpivot + (C F P Ccont C P F )T (zhub0
F
− zpivot
F
)

Similarly, the pylon location and nacelle center-of-gravity location can be calculated for given shaft
control. The shift in the aircraft center of gravity produced by nacelle tilt is

F
W (zcg − zcg0
F F
) = Wmove (znac − znac0
F
) = Wmove (C F P Ccont
T
C P F − I) (znac0
F
− zpivot
F
)

where W is the gross weight and Wmove the weight moved. Table 12-2 summarizes the geometry options.

Table 12-2. Rotor shaft axes.

nominal thrust incidence cant
z S -axis xS -axis W + for T + for T Uθh Vφh
main-rotor −z F up −xF aft Y180 aft right Yθ Xφ
other zF down −xF aft Z180 aft right Y−θ X−φ
propeller xF forward −z F up Y90 up right Yθ Zφ
other −xF aft −z F up Z180 Y−90 up right Y−θ Z−φ
tail-rotor (r = 1) yF right −xF aft Z180 X−90 aft up Zθ X−φ
tail-rotor (r = −1) −y F left −xF aft Z180 X90 aft up Z−θ Xφ

⎡ ⎤ ⎡ ⎤
−1 0 0 0 0 −1
Y180 =⎣ 0 1 0 ⎦ Z180 Y−90 =⎣ 0 −1 0 ⎦
0 0 −1 −1 0 0
⎡ ⎤ ⎡ ⎤
−1 0 0 −1 0 0
Z180 = ⎣ 0 −1 0 ⎦ Z180 X−90 =⎣ 0 0 1⎦
0 0 1 0 1 0
⎡ ⎤ ⎡ ⎤
0 0 −1 −1 0 0
Y90 = ⎣0 1 0 ⎦ Z180 X90 =⎣ 0 0 −1 ⎦
1 0 0 0 −1 0

12-3.4 Hub Loads

The rotor controls give the thrust magnitude and the tip-path plane tilt angles βc and βs , either
directly or from the collective and cyclic pitch. The forces acting on the hub are the thrust T , drag H ,
and side force Y (positive in z -, x-, y -axis directions, respectively). The hub pitch and roll moments are
112 Rotor

proportional to the flap angles. The hub torque is obtained from the shaft power Pshaft and rotor speed
Ω. The force and moment acting on the hub, in shaft axes, are then:
⎛ ⎞ ⎛ ⎞
H 0
FS = ⎝ Y ⎠ + ⎝ 0 ⎠
T −fB T
⎛ ⎞ ⎛ ⎞
Mx Khub (rβs )
M S = ⎝ My ⎠ = ⎝ Khub (−βc ) ⎠
−rQ −rPshaft /Ω

The force includes a term proportional to the rotor thrust and an input blockage factor fB = ΔT /T ≥ 0.
This term accounts for blockage or download, as an alternative to including the drag of the fuselage
or a lifting surface in the aircraft trim. For example, fB can model the tail-rotor blockage caused by
operation near the vertical tail. The rotor loads in aircraft axes acting at the center of gravity are then:

F F = CF SF S
,F F F
M F = C F S M S + Δz

where Δz F = zhub
F
− zcg
F
.
The wind axis lift L and drag X are calculated from the net rotor hub force F F and the rotor velocity
v . The velocity relative to the air gives the propulsive force direction ep = v F /|v F | (no interference)
F

and the velocity magnitude V = |v F |. The drag and lift components of the force are X = −eTp F F and
L = |(I − ep eTp )F F |, respectively. Thus XV = −(v F )T F F and L2 = |F F |2 − |X|2 . The rotor contribution
to vertical force is the z -axis component of the force in inertial axes, FV = −kT C IF F F .

12-3.5 Download
Rotor-induced download on the airframe can be modeled in several ways. The first approach is
to calculate the aerodynamic force on components (fuselage, wings, and tails), produced by the rotor
wake-induced interference velocity. Optionally the aircraft vertical drag (D/q)V can be fixed. Aircraft
trim increases the rotor thrust to counter the download force.
The second approach is to introduce a blockage factor fB = ΔT /T ≥ 0 to produce an additional
force fB T on the rotor hub, opposing the rotor thrust. Again, aircraft trim increases the rotor thrust to
counter the blockage force.
The third approach is to use a download factor fDL = 1/(1 − ΔT /T ) ≥ 1 and evaluate the rotor
induced and profile power at the increased thrust T fDL . The download will not be reflected in the rotor
thrust that trims the aircraft, but rather in higher rotor power for a given thrust.
The blockage factor and download factor for hover, out of ground effect are given by B = ΔT /T
and DL = ΔT /T , respectively. These values are reduced with forward speed and in ground effect:
fB = Bfμ fz , fDL = 1/(1 − DL fμ fz ), where

fμ = (1 − m)(1 − aDL m)
fz = (1 − ζ 2 )(1 − bDL ζ 4 )

with m = μ/μDL and ζ = (zg /D)DL /(zg /D). The blockage and download are zero for μ > μDL or
zg /D < (zg /D)DL .
Rotor 113
12–4 Aerodynamics

The rotor velocity relative to the air is v F = vAC

F
AC
+ω F
Δz F in aircraft axes. The velocities in shaft
axes are ⎛ ⎞ ⎛ ⎞
−μx rα̇x
v S = C SF v F = ΩR ⎝ rμy ⎠ ω S = C SF ωAC
F
= Ω ⎝ α̇y ⎠
μz rα̇z
where ΩR is the rotor tip speed. The advance ratio μ, inflow ratio λ, and shaft angle of attack α are
defined as 
μ= μ2x + μ2y
λ = λi + μz
α = tan−1 (μz /μ)
The

blade velocity relative to the air has the maximum amplitude (advancing tip velocity) of μat =
(1 + μ)2 + μ2z , from which the advancing tip Mach number is Mat = Mtip μat , using the tip Mach
number Mtip = (ΩR)/cs . The rotor thrust coefficient is defined as CT = T /ρA(ΩR)2 . The dimensionless
ideal induced velocity λi is calculated from μ, μz , and CT ; then the dimensional velocity is vi = ΩR λi .
The ideal induced power is then Pideal = T vi . Note that for these inflow velocities, the subscript “i”
denotes “ideal.”

12-4.1 Ideal Inflow

The ideal wake-induced velocity is obtained from Glauert’s momentum theory:
CT sλ2
λi =  = h
2 λ2 + μ2 λ2 + μ2
where λ = λi + μz , λ2h = |CT |/2 (λh is always positive), and s = sign CT . This expression is generalized
to
λi = λh s F (μ/λh , sμz /λh )
If μ is zero, the equation for λi can be solved analytically. Otherwise, for non-axial flow, the equation
is written as follows:
sλ2h
λ=  + μz
λ2 + μ2
Using λ instead of λi as the independent variable simplifies implementation of the ducted fan model. A
Newton–Raphson solution for λ gives:
2
/in =  sλh
λ
λ2n + μ2
λn − μz − λ/in
λn+1 = λn − f
/in λn /(λ2 + μ2 )
1+λ n

A relaxation factor of f = 0.5 is used to improve convergence. Three or four iterations are usually
sufficient, using
sλ2h
λ∼
= + μz
(sλh + μz )2 + μ2
to start the solution. To eliminate the singularity of the momentum theory result at ideal autorotation,
the expression  
0.373μ2z + 0.598μ2
λ = μz − 0.991
λ2h
114 Rotor

is used when
1.5μ2 + (2sμz + 3λh )2 < λ2h

The equation λ = μz (aμ2z − bλ2h + cμ2 )/λ2h is an approximation for the induced power measured in the
turbulent-wake and vortex-ring states. Matching this equation to the axial-flow momentum theory result
√ √
at μz = −2λh and μz = −λh gives a = 5/6 = 0.3726780 and b = (4 5 − 3)/6 = 0.9907120. Then
matching to the forward-flight momentum theory result at (μ = λh , μz = −1.5λh ) gives c = 0.5980197.
For axial flow (μ = 0) the solution is:
⎧ 0 
⎪
⎪ μz μz 2
+ λ2h −λh < sμz
⎪
⎪ +s
⎪
⎪ 2 2
⎪
⎨  0.373μ2 
λ = μz z
− 0.991 −2λh < sμz < −λh
⎪
⎪ λ2h
⎪
⎪ 0 
⎪
⎪
⎪
⎩ μz − s μz 2
− λ2h sμz < −2λh
2 2

Note that λi and vi are the ideal induced velocities; additional factors are required for the wake-induced
velocity or induced power calculations.

12-4.1.1 Ducted Fan

Rotor momentum theory can be extended to the case of a ducted fan. Consider a rotor system with
disk area A, operating at speed V , with an angle α between V and the disk plane. The induced velocity
at the rotor disk is v , and in the far wake w = fW v . The far wake area is A∞ = A/fA . The axial velocity
at the fan is fV z Vz , with fV z accounting for acceleration or deceleration through the duct. The edgewise
velocity at the fan is fV x Vx , with fV x = 1.0 for wing-like behavior, or fV x = 0 for tube-like behavior of
the flow. The total thrust (rotor plus duct) is T , and the rotor thrust is Trotor = fT T . For this model, the
duct aerodynamics are defined by the thrust ratio fT or far wake area ratio fA , plus the fan velocity ratio
fV . The mass flux through the rotor disk is ṁ = ρAU = ρA∞ U∞ , where U and U∞ are respectively the
total velocity magnitudes at the fan and in the far wake:

U 2 = (fV x V cos α)2 + (fV z V sin α + v)2

2
U∞ = (V cos α)2 + (V sin α + w)2

Mass conservation (fA = A/A∞ = U∞ /U ) relates fA and fW . Momentum and energy conservation give

T = ṁw = ρAU∞ w/fA = ρAU fW v

1  w
P = ṁw (2V sin α + w) = T V sin α +
2 2

With these expressions, the span of the lifting system in forward flight is assumed equal to the rotor
diameter 2R. Next it is required that the power equals the rotor induced and parasite loss:

P = Trotor (fV z V sin α + v) = T fT (fV z V sin α + v)

In axial flow, this result can be derived from Bernoulli’s equation for the pressure in the wake. In
forward flight, any induced drag on the duct is being neglected. From these two expressions for power,
Vz + fW v/2 = fT (fV z Vz + v) is obtained, relating fT and fW . With no duct (fT = fV x = fV z = 1), the
far wake velocity is always w = 2v , hence fW = 2. With an ideal duct (fA = fV x = fV z = 1), the far
Rotor 115

wake velocity is fW = 1. In hover (with or without a duct), fW = fA = 2fT , and v = 2/fW vh . The
rotor ideal induced power is Pideal = T w/2 = fD T v , introducing the duct factor fD = fW /2.
For a ducted fan, the thrust CT is calculated from the total load (rotor plus duct). To define the duct
effectiveness, either the thrust ratio fT = Trotor /T or the far wake area ratio fA = A/A∞ is specified (and
the fan velocity ratio fV ). Thewake-induced velocity is obtained from the momentum theory result for
a ducted fan: λ2h = (fW λi /2) (fV x μ)2 + (fV z μz + λi )2 . If the thrust ratio fT is specified, this can be
written
sλ2h /fT μz
fV z μz + λi =  +
(fV z μz + λi )2 + (fV x μ)2 fT
In this form, λi can be determined using the free-rotor expressions given previously: replacing λ2h , μz ,
μ, and λ with λ2h /fT , μz /fT , fV x μ, and fV z μz + λi , respectively. Then from λi the velocity and area
ratios are obtained:  
μz
fW = 2 fT − (1 − fT fV z )
λi

μ2 + (μz + fW λi )2
fA =
(fV x μ)2 + (fV z μz + λi )2
If instead the area ratio fA is specified, it is simplest to first solve for the far wake velocity fW λi :
sλ2h 2fA
μz + fW λi =  + μz
(μz + fW λi )2 + μ2

In this form, fW λi can be determined using the free-rotor expressions given previously: replacing λ2h
and λ with λ2h 2fA and μz + fW λi , respectively. The induced velocity is
1 * 2 +
(fV z μz + λi )2 = 2 μ + (μz + fW λi )2 − (fV x μ)2
fA

The velocity ratio is fW = (fW λi )/λi , and

μz + fW λi /2
fT =
fV z μz + λi

is the thrust ratio. However, physical problems and convergence difficulties are encountered with this
approach in descent, if an arbitrary value of fT is permitted. From the expression for fT , fT should
approach 1/fV z at high rates of climb or descent. To avoid problems with an arbitrary value of fT , it is
assumed that the input value of fT defines the velocity ratio fW = 2fT in descent. So in descent μz is
not replaced by μz /fT .

12-4.1.2 Ground Effect

The wake-induced velocity is reduced when the rotor disk is in the proximity of the ground plane.
√
Ground effect in hover can be described in terms of the figure of merit M = (T 3/2 / 2ρA)/P as a function
of scaled rotor height above the ground, zg /D = zg /2R. Usually the test data are given as the ratio of
the thrust to OGE thrust, for constant power: T /T∞ = (M/M∞ )2/3 = κg ≥ 1. The effect on power at
constant thrust is then P = P∞ fg , where fg = κ−3/2
g ≤ 1. Ground effect is generally negligible at heights
above zg /D = 1.5 and at forward speeds above μ = 3λh .
The ground plane is assumed to be perpendicular to the inertial frame z -axis. The ground normal
(directed downward) is kgF = C F I k in airframe axes, or kgS = C SF kgF in rotor shaft axes. The height
116 Rotor

of the landing gear above ground level, hLG , is specified in the flight state. The height of the rotor hub
above ground level is then
zg = hLG − (kgF )T (zhub
F
− zLG
F
) + dLG
F
where zLG is the position of the landing gear in the airframe, and dLG is the distance from the bottom of
the gear to the location zLG . From the velocity
⎛ ⎞
μx
v S = ⎝ −rμy ⎠
−λ

the angle  between the ground normal and the rotor wake is evaluated: cos  = (kgS )T v S /|v S
| ( = 0 for

hover,  = 90 deg in forward flight). Note that if the rotor shaft is vertical, then cos  = λ/ μ2 + λ2 (see
ref. 1). The expressions for ground effect in hover are generalized to forward flight by using (zg / cos )
in place of zg . No ground effect correction is applied if the wake is directed upward (cos  ≤ 0), or if
−3/2
zg / cos  > 1.5D. From zg /D cos , the ground effect factor fg = κg is calculated. Then

(λi )IGE = fg (λi )OGE

is the effective ideal induced velocity.

Several empirical ground effect models are implemented: from Cheeseman and Bennett (ref. 1,
basic model and using blade-element (BE) theory to incorporate influence of thrust), from Law (ref. 2),
from Hayden (ref. 3), a curve fit of the interpolation from Zbrozek (ref. 4), and from University of
Maryland (ref. 5):
⎧ 3/2
⎪
⎪ 1
⎪
⎪ 1 − Cheeseman and Bennett
⎪
⎪ (4zg /R)2
⎪
⎪
⎪
⎪
⎪
⎪  −3/2
⎪
⎪ σaλi 1
⎪
⎪ 1 + 1.5 Cheeseman and Bennett (BE)
⎪
⎪ 4CT (4zg /R)2
⎪
⎪
⎪
⎪  3/2
⎪
⎪
⎪
⎪ 1.0991 − 0.1042/(zg /D)
⎪
⎨ 1 + (C /σ)(0.2894 − 0.3913/(z /D)) Law
T g
fg =
⎪
⎪  −1
⎪
⎪ 0.03794
⎪
⎪ 0.9926 + Hayden
⎪
⎪ (z g /2R)
2
⎪
⎪
⎪
⎪ - .−3/2
⎪
⎪
⎪
⎪ 0.0544
⎪
⎪ 0.9122 +  Zbrozek
⎪
⎪
⎪
⎪ (zg /R) CT /σ
⎪
⎪
⎪
⎪"         %
⎪
⎩ 0.146 + 2.090 zg − 2.068 zg 2 + 0.932 zg 3 − 0.157 zg 4
R R R R University of Maryland

These equations break down at small height above the ground, and so are restricted to zg /D ≥ 0.15;
however, the database for ground effect extends only to about z/D = 0.3, except for the Maryland data,
which goes to z/D = 0.1. Also, fg ≤ 1 is required. Figure 12-2 shows T /T∞ = κg = fg−2/3 as a function
of z/R for these models (CT /σ = 0.05, 0.10, 0.15), compared with test data from several sources.
Alternatively, a table can be specified for either κg = T /T∞ or fg = P/P∞ , as a function of CT /σ
and h/D = zg /2R; or as a function of CT /σ , h/D, and Mtip .
The influence of the ground on tiltrotor power required is stronger than for an isolated rotor. This
further reduction of power is probably due to a reduction of wing and fuselage download when operating
Rotor 117

Hayden Cheeseman & Bennett

Rabbott Cheeseman & Bennett (BE)
1.3 Cerbe Law
Cheeseman Hayden
Zbrozek Zbrozek
Maryland Maryland
1.2
T/T∞

1.1

1.0

0.9
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

z/R

Figure 12-2. Ground effect models (hover).

near the ground. For quad tiltrotors, an upload on the airframe has been measured at low heights. These
effects can be modeled by using an effective distance above the ground: ze = Cg zg , with typically
Cg = 0.5 for a tiltrotor.

12-4.1.3 Inflow Gradient

As a simple approximation to nonuniform induced velocity distribution, a linear variation over the
disk is used: Δλ = λx r cos ψ + λy r sin ψ . There are contributions to Δλ from forward flight and from
hub moments, which influence the relationship between flapping and cyclic. The linear inflow variation
caused by forward flight is Δλf = λi (κx r cos ψ + κy r sin ψ), where λi is the mean inflow. Typically κx
is positive, and roughly 1 at high speed, and κy is smaller in magnitude and negative. Both κx and κy
must be zero in hover. Based on references 6–9, the following models are considered:

15π 15π μ
Coleman and Feingold: κx0 = fx tan χ/2 = fx 
32 32 μ2 + λ2 + |λ|
κy0 = −fy 2μ
√ √ μ
White and Blake: κx0 = fx 2 sin χ = fx 2 
μ + λ2
2

κy0 = −2fy μ
118 Rotor

where tan χ = |λ|/μ is the wake angle. Extending these results to include sideward velocity gives
κx = (κx0 μx + κy0 μy )/μ and κy = (−κx0 μy + κy0 μx )/μ. For flexibility, the empirical factors fx and fy
have been introduced (values of 1.0 give the baseline model). There is also an inflow variation produced
by any net aerodynamic moment on the rotor disk, which can be evaluated using a differential form of
momentum theory:
fm
Δλm =  (−2CM y r cos ψ + 2CM x r sin ψ) = λxm r cos ψ + λym r sin ψ
μ2 + λ2

including empirical factor fm . Note that the denominator of the hub moment term is zero for a hovering
rotor at zero thrust; so this inflow contribution should not be used for cases of low speed and low thrust.

12-4.2 Rotor Forces

When direct control of the rotor thrust magnitude is used, the rotor collective pitch angle θ0.75 must
be calculated from the thrust CT /σ . If the commanded variable is the collective pitch angle, then it is
necessary to calculate the rotor thrust, resulting in more computation, particularly since all quantities
depending on the thrust (inflow, induced power factor, and mean drag coefficient) are also unknown.
There may be flight states where the commanded thrust can not be produced by the rotor, even with
stall neglected in the section aerodynamics. This condition manifests as an inability to solve for the
collective pitch given the thrust. In this circumstance the trim method can be changed so the required
or specified thrust is an achievable value, or commanded collective pitch control can be used.
Cyclic control consists of tip-path plane command, requiring calculation of the rotor cyclic pitch
angles from the flapping; or no-feathering plane command, requiring calculation of the tip-path plane tilt
from the cyclic control angles. The longitudinal tip-path plane tilt is βc (positive forward) and the lateral
tilt is βs (positive toward retreating side). The longitudinal cyclic pitch angle is θs (positive aft), and the
lateral cyclic pitch angle is θc (positive toward retreating side). Tip-path plane command is appropriate
for main-rotors. For rotors with no cyclic pitch, no-feathering plane command must be used.
The forces acting on the hub are the thrust T , drag H , and side force Y (positive in z -, x-, y -
axis directions, respectively). The aerodynamic analysis is conducted for a clockwise rotating rotor,
with appropriate sign changes for lateral velocity, flapping, and force. The analysis is conducted in
dimensionless form, based on the actual radius and rotational speed of the flight state. The inplane hub
forces are produced by tilt of the thrust vector with the tip-path plane, plus forces in the tip-path plane,
and profile terms (produced by the blade drag coefficient). The orientation of the tip-path axes relative
to the shaft axes is then C P S = Xrβs Y−βc . Then
⎛ ⎞ ⎛ ⎞ ⎛ ⎞ ⎛ ⎞
CH 0 CHtpp CHo
⎝ CY ⎠ = C SP ⎝ 0 ⎠ + ⎝ rCY tpp ⎠ + ⎝ rCY o ⎠
SP
CT CT /C33 0 0

The inplane forces relative to the tip-path plane can be neglected, or calculated by blade element theory.
Note that with thrust and tip-path plane command and CHtpp and CY tpp neglected, it is not necessary to
solve for the rotor collective and cyclic pitch angles. In general the inplane forces relative to the tip-path
plane are not zero, and may be significant, as for a rotor with large flap stiffness. Figures 12-3a and b
show, respectively, the tip-path plane tilt and thrust vector tilt with cyclic pitch control (no-feathering
plane tilt), as functions of flap stiffness (frequency), for several rotor thrust values. The difference
between tip-path plane tilt (fig. 12-3a) and thrust vector tilt (fig. 12-3b) is caused by tilt of the thrust
vector relative to the tip-path plane.
Rotor 119

The profile inplane forces can be obtained from simplified equations, or calculated by blade element
theory. The simplified method uses:
   
CHo σ μx /μ
= cdmean FH
CY o 8 −μy /μ

where the mean drag coefficient cdmean is from the profile power calculation. The function FH ac-
counts for the increase of the blade section velocity with rotor edgewise and axial speed: CHo =
! 1 ! 1 2 2 2 1/2
2 σcd U (r sin ψ + μ)dr = 2 σcd (uT + uR + uP ) (r sin ψ + μ)dr; so (from ref. 10)
 
1 2π 1  1/2
FH = 4 (r + μ sin ψ)2 + (μ cos ψ)2 + μ2z (r sin ψ + μ) dr dψ
2π 0 0

    -√ .
 1 3 V2−1 3 3 1+V2+1
∼
= 1+V 2 3μ + μ 2
+ μμz + μ ln
4 (1 + V 2 )2 4 V

with V 2 = μ2 + μ2z .

12-4.3 Blade Element Theory

Blade element theory is the basis for the solution for the collective and cyclic pitch angles (or thrust
and flap angles) and evaluation of the rotor inplane hub forces. The section aerodynamics are described
by lift varying linearly with angle of attack, c = cα α (no stall), and a constant mean drag coefficient
cdmean (from the profile power calculation). The analysis is conducted in dimensionless form (based on
density ρ, rotor rotational speed Ω, and blade radius R of the flight state). So in the following σ , ν , and
γ are for the actual ρ, Ω, and R; and a = 5.7 is the lift-curve slope used in the Lock number γ . The blade
section aerodynamic environment is described by the three components of velocity, from which the yaw
and inflow angles are obtained, and then the angle of attack:

U 2 = u2T + u2P
uT = r + μx sin ψ + μy cos ψ 
cos Λ = U/ u2T + u2P + u2R
uR = μx cos ψ − μy sin ψ
φ = tan−1 uP /uT
uP = λ + r(β̇ + α̇x sin ψ − α̇y cos ψ) + uR β
α=θ−φ

In reverse flow (|α| > 90), α ← α − 180 signα, and then c = cα α still (airfoil tables are not used). The
blade pitch consists of collective, cyclic, twist, and pitch-flap coupling terms. The flap motion is rigid
rotation about a hinge with no offset, and only coning and once-per-revolution terms are considered:

θ = θ0.75 + θtw + θc cos ψ + θs sin ψ − KP β

β = β0 + βc cos ψ + βs sin ψ

where KP = tan δ3 . The twist is measured relative to 0.75R; θtw = θL (r − 0.75) for linear twist. The
mean inflow is λ0 = κλi , using the induced velocity factor κ from the induced power model. The inflow
includes gradients caused by edgewise flight and hub moments:

λ = μz + λ0 (1 + κx r cos ψ + κy r sin ψ) + Δλm

fm
= μz + λ0 (1 + κx r cos ψ + κy r sin ψ) +  (−2CM y r cos ψ + 2CM x r sin ψ)
μ2 + λ2
120 Rotor

1.1
1.0 C T/σ = 0.14
0.9 C T/σ = 0.10
C T/σ = 0.06
TPP tilt / cyclic magnitude

0.8
C T/σ = 0.02
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0.0
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0
flap frequency ν (per-rev)

Figure 12-3a. Tip-path plane tilt with cyclic pitch.

1.1
1.0
0.9
thrust tilt / cyclic magnitude

0.8
0.7
0.6
0.5
C T/σ = 0.02
0.4
C T/σ = 0.06
0.3
C T/σ = 0.10
0.2 C T/σ = 0.14
0.1
0.0
1.0 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2.0

flap frequency ν (per-rev)

Figure 12-3b. Thrust vector tilt with cyclic pitch.

Rotor 121

From the hub moments    

−CM y σa ν 2 − 1 βc
=
CM x 2 γ βs
the inflow gradient is
 
fm σa ν2 − 1 ν2 − 1
Δλm =  (rβc cos ψ + rβs sin ψ) = Km (rβc cos ψ + rβs sin ψ)
μ +λ 8
2 2 γ/8 γ/8

The constant Km is associated with a lift-deficiency function:

1 1
C= =   
1 + Km 1 + fm σa/ 8 μ2 + λ2
The blade chord is c(r) = cref ĉ(r), where cref is the thrust-weighted chord (chord at 0.75R for linear
taper). Yawed flow effects increase the section drag coefficient, hence cd = cdmean / cos Λ. The section
forces in velocity axes and shaft axes are
1 2 1
L= ρU cc Fz = L cos φ − D sin φ =
ρU c(c uT − cd uP )
2 2
1 1
D = ρU 2 ccd Fx = L sin φ + D cos φ = ρU c(c uP + cd uT )
2 2
1 2 1
R = ρU ccr = D tan Λ Fr = −βFz + R = −βFz + ρU ccd uR
2 2
These equations for the section environment and section forces are applicable to high inflow (large μz ),
sideward flight (μy ), and reverse flow (uT < 0). The total forces on the rotor hub are

T =N Fz dr

H=N Fx sin ψ + Fr cos ψ dr

Y =N −Fx cos ψ + Fr sin ψ dr

with an average over the rotor azimuth implied, along with the integration over the radius. Lift forces
are integrated from the root cutout rroot to the tip loss factor B . Drag forces are integrated from the root
cutout to the tip.
2
In coefficient form (forces divided by ρAVtip ) the rotor thrust and inplane forces are:

CT = σ F/z dr 1
F/z = ĉU (c uT − cd uP )
 2
1
CH = σ F/x sin ψ + F/r cos ψ dr F/x = ĉU (c uP + cd uT )
2

1
CY = σ −F/x cos ψ + F/r sin ψ dr F/r = −β F/z + ĉU cd uR
2

(and the sign of CY is changed for a clockwise rotating rotor). The terms ΔF/x ∼
= F/z β̇ and ΔF/r = −F/z β
produce tilt of the thrust vector with the tip-path plane (CH = −CT βc and CY = −CT βs ), which
are accounted for directly. The section drag coefficient cd produces the profile inplane forces. The
approximation uP ∼ = μz is consistent with the simplified method (using the function FH ), hence
 
1 σ
F/xo = ĉU0 cd uT CHo = σ F/xo sin ψ + F/ro cos ψ dr = ĉU0 cd (r sin ψ + μx ) dr
2 2
 
1 σ
F/ro = ĉU0 cd uR CY o = σ −F/xo cos ψ + F/ro sin ψ dr = − ĉU0 cd (r cos ψ + μy ) dr
2 2
122 Rotor

where U02 = u2T + μ2z , and cd = cdmean / cos Λ. Using blade element theory to evaluate CHo and CY o
accounts for the planform (ĉ) and root cutout. Using the function FH implies a rectangular blade and no
root cutout (plus at most a 1% error approximating the exact integration). The remaining terms in the
section forces produce the inplane loads relative to the tip-path plane:
1 1
F/xi = F/x − F/z β̇ − F/xo = ĉU c (uP − uT β̇) + ĉU cd ((1 − U0 /U )uT + uP β̇)
2 2
1
F/ri = F/r + F/z β − F/ro = ĉU cd (1 − U0 /U )uR
2

CHtpp = σ F/xi sin ψ + F/ri cos ψ dr

CY tpp = σ −F/xi cos ψ + F/ri sin ψ dr

(including small profile terms from U0 = U ).

Evaluating these inplane forces requires the collective and cyclic pitch angles and the flapping
motion. The thrust equation must be solved for the rotor collective pitch or the rotor thrust. The
relationship between cyclic pitch and flapping is defined by the rotor flap dynamics. The flap motion is
rigid rotation about a central hinge, with a flap frequency ν > 1 for articulated or hingeless rotors. The
flapping equation of motion is

γ
β̈ + ν 2 β + 2α̇y sin ψ + 2α̇x cos ψ = F/z r dr + (ν 2 − 1)βp
a

including precone angle βp ; the Lock number γ = ρacref R4 /Ib . This equation is solved for the mean
(coning) and 1/rev (tip-path plane tilt) flap motion:

γ
ν 2 β0 = F/z r dr + (ν 2 − 1)βp
a
      
βc γ 2 cos ψ 2α̇x
(ν 2 − 1) = F/z r dr +
βs a 2 sin ψ 2α̇y

with an average over the rotor azimuth implied. The solution for the coning is largely decoupled by
introducing the thrust:

γ 6CT γ
ν02 β0 = + (ν02 − 1)βp + F/z (r − 3/4) dr
8 σa a

A separate flap frequency ν0 is used for coning, in order to model teetering and gimballed rotors. For
an articulated rotor, βp = 0 should be used.
The thrust and flapping equations of motion that must be solved are:

6 6CT
Et = F/z dr −
a σa
        
Ec 8 2 cos ψ ν 2 − 1 βc 16 α̇x
= F/z r dr − +
Es a 2 sin ψ γ/8 βs γ α̇y

The solution v such that E(v) = 0 is required. The variables are v = (θ0.75 θc θs )T for thrust and tip-path
plane command; v = (θ0.75 βc βs )T for thrust and no-feathering plane command; v = (CT /σ θc θs )T for
collective pitch and tip-path plane command; v = (CT /σ βc βs )T for collective pitch and no-feathering
Rotor 123

plane command. Note that since c = cα α is used (no stall), these equations are linear in θ. However,
if ∂T /∂θ0.75 is small, the solution may not produce a reasonable collective for commanded thrust. A
Newton–Raphson solution method is used: from E(vn+1 ) ∼ = E(vn ) + (dE/dv)(vn+1 − vn ) = 0, the
iterative solution is
vn+1 = vn − C E(vn )

where C = f (dE/dv)−1 , including the relaxation factor f . The derivative matrix dE/dv is obtained by
numerical perturbation. Convergence of the Newton–Raphson iteration is tested in terms of |E| <  for
each equation, where  is an input tolerance.

12–5 Power

The rotor power consists of induced, profile, parasite, and adjustment terms: P = Pi + Po + Pp + Px .
The parasite power (including climb/descent power for the aircraft) is obtained from the wind axis drag
force: Pp = −XV = (v F )T F F .
The induced power is calculated from the ideal power: Pi = κPideal = κfD T videal . The empirical
factor κ accounts for the effects of nonuniform inflow, non-ideal span loading, tip losses, swirl, blockage,
and other phenomena that increase the induced power losses (κ > 1). For a ducted fan, fD = fW /2 is
introduced. The induced power at zero thrust is zero in this model (or accounted for as a profile power
increment). If κ is deduced from an independent calculation of induced power, nonzero Pi at low thrust
will be reflected in large κ values.
The profile power is calculated from a mean blade drag coefficient: Po = ρA(ΩR)3 CP o , CP o =
(σ/8)cdmean FP . The function FP (μ, μz ) accounts for the increase of the blade section velocity with rotor
! !
edgewise and axial speed: CP o = 12 σcd U 3 dr = 12 σcd (u2T + u2R + u2P )3/2 dr; so (from ref. 10)
 
1 2π 1  3/2
FP = 4 (r + μ sin ψ)2 + (μ cos ψ)2 + μ2z dr dψ
2π 0 0

  
∼ 5 3 4 + 7V 2 + 4V 4 9 μ4
= 1 + V 2 1 + V 2 + μ2 −
2 8 (1 + V 2 )2 16 1 + V 2
  - √ .
3 4 3 2 2 9 4 1+V2+1
+ μ + μ μ + μ ln
2 z 2 z 16 V

with V 2 = μ2 + μ2z . This expression is exact when μ = 0, and fP ∼ 4V 3 for large V .

Two performance methods are implemented, the energy method and the table method. The induced
power factor and mean blade drag coefficient are obtained from equations with the energy method, or
from tables with the table method. Optionally κ and cdmean can be specified for each flight state,
superseding the performance method values.
Rotor power adjustments (Px ) are associated with wing interference, propulsive force efficiency,
and climb efficiency.

12-5.1 Energy Performance Method

12-5.1.1 Induced Power

The induced power is calculated from the ideal power: Pi = κPideal = κfD T videal . Reference values
of κ are specified for hover, axial cruise (propeller), and edgewise cruise (helicopter): κhover , κprop , and
124 Rotor

1.30

1.25

1.20

1.15
κ

1.10

1.05

1.00
0.00 0.03 0.06 0.09 0.12 0.15 0.18
C T/σ

Figure 12-4. Induced power factor for rotor in hover.

κedge . Two models are implemented: constant model and standard model. The constant model uses
κ = κhover if μ = μz = 0; or κ = κprop if |μ| < 0.1|μz |; or κ = κedge otherwise.

The standard model calculates an axial flow factor κaxial from κhover , κclimb , and κprop . Let Δ =
CT /σ − (CT /σ)ind . For hover and low-speed axial climb, including a variation with thrust, the inflow
factor is
2 " %
tan−1 ((|μz |/λh )/Maxial ) axial
X
κh = κhover + kh1 Δh + kh2 |Δh |Xh2 + (κclimb − κhover )
π

where |μz |/λh = Maxial is the midpoint of the transition between hover and climb, and Xaxial is large
for a fast transition. Figure 12-4 illustrates κ in hover (with a minimum value). Figure 12-5 shows the
behavior of this function for a helicopter in climb (Xaxial = 0.65). A polynomial describes the variation
with axial velocity, scaled so κ = κh at μz = 0 and κ = κp at μz = μzprop . Including variations with
thrust and shaft angle:

κp = κprop + kp1 Δp + kp2 |Δp |Xp2 + kpα μ2z |μ|Xpα

κaxial = κh + ka1 |μz | + S(ka2 μ2z + ka3 |μz |Xa )

where S = (κp − (κh + ka1 μzprop ))/(ka2 μ2zprop + ka3 |μzprop |Xa ); S = 0 if ka2 = ka3 = 0 (not scaled); and
κaxial = κh if μzprop = 0. A polynomial describes the variation with edgewise advance ratio, scaled so
κ = κaxial at μ = 0 and κ = κe at μ = μedge . Thus the induced power factor is

κe = fα foff (κedge + kt1 Δt + kt2 |Δt |Xt2 )

κ = κaxial + ke1 μ + S(ke2 μ2 + ke3 μXe )

where S = (κe − (κaxial + ke1 μedge ))/(ke2 μ2edge + ke3 |μedge |Xe ); S = 0 if ke2 = ke3 = 0 (not scaled); and
κ = κaxial if μedge = 0. The function fα = 1 − keα μz accounts for the influence of angle of attack (μz /μ)
or rotor drag (CX ). The function foff = 1 − ko1 (1 − e−ko2 ox ) accounts for the influence of lift offset,
ox = rMx /T R = (Khub /T R)βs . Figure 12-6 illustrates κ in edgewise flight. Minimum and maximum
values of the induced power factor, κmin and κmax , are also specified.
Rotor 125

1.14

1.12

1.10

1.08
κ

1.06
Maxial = 1.176
1.04 Maxial = 0.5
1.02 Maxial = 2.0

1.00
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0

μz/λh

Figure 12-5. Induced power factor for rotor in axial flight.

4.50
C T/σ = 0.08
4.00 C T/σ = 0.14
3.50 (μedge, κedge )
3.00
κ

2.50
2.00
1.50
1.00
0.00 0.10 0.20 0.30 0.40 0.50

Figure 12-6. Induced power factor for rotor in edgewise flight.

12-5.1.2 Profile Power

The profile power is calculated from a mean blade drag coefficient: CP o = (σ/8)cdmean FP . Since the
blade mean lift coefficient is c ∼
= 6CT /σ , the drag coefficient is estimated as a function of blade loading
CT /σ (using thrust-weighted solidity). With separate estimates of the basic, stall, and compressibility
drag, the mean drag coefficient is:

cdmean = χS (cdbasic + cdstall + cdcomp )

where χ is a technology factor. The factor S = (Reref /Re)XRe accounts for Reynolds number effects
on the drag coefficient; typically XRe = 0.2 for a turbulent boundary layer. Re is based on the thrust-
weighted chord, 0.75Vtip , and the flight state; Reref corresponds to the input cd information. The following
models are implemented for the basic drag:
126 Rotor

a) Array model: The basic drag cdbasic is input as a function of CT /σ ; the array is linearly interpolated.
b) Equation model: The basic drag cdbasic is a quadratic function of CT /σ , plus an additional term
allowing faster growth at high (sub-stall) angles of attack. Let Δ = |CT /σ − (CT /σ)Dmin |, where
(CT /σ)Dmin corresponds to the minimum drag and Δsep = |CT /σ| − (CT /σ)sep . Values of the basic
drag equation are specified for helicopter (hover and edgewise) and propeller (axial climb and cruise)
operation:
cdh = d0hel + d1hel Δ + d2hel Δ2 + dsep ΔX sep
sep

sep + dpα μz |μ|

cdp = d0prop + d1prop Δ + d2prop Δ2 + dsep ΔX sep 2 Xpα

The separation term is present only if Δsep > 0. The helicopter and propeller values are interpolated as
a function of μz :
2
cdbasic = cdh + (cdp − cdh ) tan−1 (|μz |/λh ) + df 1 μ + df 2 μXf + dz1 μz + dz2 |μz |Xz
π
so |μz |/λh = 1 is the midpoint of the transition. The last terms are the effect of edgewise and axial flow.
The stall drag increment represents the rise of profile power caused by the occurrence of significant
stall on the rotor disk. Let Δs = |CT /σ| − (fs /fα foff )(CT /σ)s (fs is an input factor). The function
fα = 1 − dsα μz accounts for the influence of angle of attack (μz /μ) or rotor drag (CX ). The function
foff = 1 − do1 (1 − e−do2 ox ) accounts for the influence of lift offset, ox = rMx /T R = (Khub /T R)βs . Then
cdstall = ds1 ΔX
s
s1
+ ds2 ΔX
s
s2
(zero if Δs ≤ 0). The blade loadingat which the stall affects the entire rotor
power, (CT /σ)s , is an input function of the velocity ratio V = μ2 + μ2z .
The compressibility drag increment depends on the advancing tip Mach number Mat , and the tip
airfoil thickness-to-chord ratio τ . The following models are implemented:
a) Drag divergence model: Let ΔM = Mat − Mdd , where Mdd is the drag divergence Mach number of
the tip section. Then the compressibility increment in the mean drag coefficient is

cdcomp = dm1 ΔM + dm2 ΔM Xm

(ref. 11). Mdd is a function of the advancing tip lift coefficient, c(1,90) . The advancing tip lift is estimated
from α(1,90) = (θ.75 + 0.25θL + θs − (λ − βc )/(1 + μ)) ∼= 1.6(1 − 2.97μ + 2.21μ2 )(6CT /σa) + 0.25θtw (zero
above μ = 0.6). Then the Korn expression (ref. 12) gives Mdd for small lift coefficient:

Mdd = κA − κ|c | − τ = Mdd0 − κ|c |

where Mdd0 is the drag divergence Mach number at zero lift, and typically κ = 0.16.
b) Similarity model: From transonic small-disturbance theory (refs. 13–14), the scaled wave drag must
2
be a function only of K1 = (Mat − 1)/[Mat
2
τ (1 + γ)]2/3 . An approximation for the wave drag increment
is
τ 5/3 τ 5/3
Δcd = 2 (1 + γ)]1/3
D(K 1 ) = 2 (1 + γ)]1/3
1.774(K1 + 1.674)5/2
[Mat [Mat
(constant for K1 > −0.2). Integration of Δcd over the rotor disk gives the compressibility increment in
the profile power. Following Harris, the resulting compressibility increment in the mean drag coefficient
is approximately:
cdcomp = 1.52f (K1 + 1)2 [(1 + μ)τ ]5/2 (1 + γ)1/2

including the input correction factor f ; cdcomp is zero for K1 < −1, and constant for K1 > −0.2.
Rotor 127

transient limit
steady limit
0.20
high stall
low stall
0.16

0.12
(CT/σ)s

0.08

0.04

0.00
0.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70
μ

Figure 12-7. Stall function.

0.0350
cd mean (low stall)
0.0300 cd mean (high stall)
cdh (with separation)
0.0250
cdh (quadratic)
cd mean

0.0200

0.0150

0.0100

0.0050

0.0000
0.00 0.03 0.06 0.09 0.12 0.15 0.18

C T/σ

Figure 12-8. Mean drag coefficient for rotor in hover.

c) Tip Mach number model: From the tip Mach number and thrust coefficient:
2.5 2.5 2.5 2.5
ΔCP o = dmt (Mtip − Mtip limit ) (CT − CT limit ) − dmt (Mtip ref − Mtip limit ) (CT − CT limit )

and then cdcomp = ΔCP o /((σ/8)FP ). The compressibility increment is zero if Mtip < Mtip limit or
CT < CT limit . Mtip ref is the tip Mach number corresponding to the basic profile power model.

Figure 12-7 shows typical stall functions (CT /σ)s for two rotors with different stall characteristics,
128 Rotor

C T/σ = 0.14
0.0350
C T/σ = 0.12
0.0300 C T/σ = 0.10
C T/σ = 0.08
0.0250
cd comp
cd mean

0.0200

0.0150

0.0100

0.0050

0.0000
0.00 0.10 0.20 0.30 0.40 0.50

Figure 12-9a. Mean drag coefficient for rotor in forward flight, high-stall.

C T/σ = 0.14
C T/σ = 0.12
C T/σ = 0.10
0.0350 C T/σ = 0.08
cd comp
0.0300

0.0250
cd mean

0.0200

0.0150

0.0100

0.0050

0.0000
0.00 0.10 0.20 0.30 0.40 0.50

Figure 12-9b. Mean drag coefficient for rotor in forward flight, low-stall.

designated high-stall and low-stall, resulting from design features such as different airfoils, Figure 12-7
also shows, for reference, typical helicopter rotor steady and transient load limits. Figure 12-8 illustrates
the mean drag coefficient in hover, showing cdh with and without the separation term, and the total for
the high-stall and low-stall cases. Figure 12-9 illustrates the mean drag coefficient in forward flight,
showing the compressibility term cdcomp , and the growth in profile power with CT /σ and μ as the stall
drag increment increases.
Rotor 129

Rotor thrust capability (CT /σ)max as a function of advance ratio μ can be specified for steady and
transient operation. Alternatively, the thrust limit can be obtained from an equation:

(CT /σ)max = K0 − K1 μ2

For example, (CT /σ)max = 0.17 − 0.25μ2 corresponds to the steady limit, and (CT /σ)max = 0.20 − 0.25μ2
corresponds to the transient limit. One of these limits can be used to calculate the aircraft never-exceed
speed. The rotor thrust margin (CT /σ)max − |CT /σ| is available for the maximum effort or trim solution.

12-5.1.3 Twin and Multiple Rotors

For twin rotors, the induced power is determined by the induced velocity of the rotor system, not
the individual rotors. The induced power is still obtained using Pi = κPideal = κfD T videal for each rotor,
but the ideal induced velocity is calculated for an equivalent thrust CT e based on the thrust and geometry
of both rotors. The profile power calculation is not changed for twin rotors.

In hover, the twin rotor induced velocity is vi = κtwin T /2ρAp , from the total thrust T and the
projected disk area Ap = (2 − m)A. The overlap fraction m is calculated from the rotor hub separation
. A correction factor for the twin rotor ideal power is also included. For a coaxial rotor, typically
κtwin ∼
= 0.90. So the ideal inflow is calculated for CT e = (CT 1 + CT 2 )/(2 − m).

In forward flight, the induced velocity vi = κtwin T /2ρAV is calculated from the equivalent thrust
Te = x1 T1 + x2 T2 , which gives vi = κtwin (x1 v1 + x2 v2 ) = κtwin (x1 T1 /2ρAV + x2 T2 /2ρAV ). The induced
velocity of a coaxial rotor is vi = κtwin T /(2ρAV ), from the total thrust T and a span of 2R. The
ideal inflow is thus calculated for CT e = CT 1 + CT 2 . The correction factor for ideal induced power
(biplane effect) is κtwin ∼ = 0.88 to 0.81 for rotor separations of 0.06D to 0.12D. The induced velocity
of side-by-side rotors is vi = κtwin T /(2ρAe V ), from the total thrust T and a span of 2R + , hence
Ae = A(1 + /2R)2 . The ideal inflow is thus calculated for CT e = (CT 1 + CT 2 )/(1 + /2R)2 . If there
is no overlap ( > 2R), it is assumed that there is no performance impact of interference between the
rotors. The induced velocity of tandem rotors is vF = κtwin (TF /(2ρAV ) + xR TR /(2ρAV )) for the front
rotor and vR = κtwin (TR /(2ρAV ) + xF TF /(2ρAV )) for the rear rotor. For large separation, xR ∼ = 0 and
xF ∼= 2 ; for the coaxial limit xR = xF = 1 is appropriate. Here x R = m and xF = 2 − m is used.
In summary, the model for twin rotor ideal induced velocity uses CT e = x1 CT 1 + x2 CT 2 and the
correction factor κtwin . In hover, xh = 1/(2 − m); in forward flight of coaxial and tandem rotors, xf = 1
for this rotor and xf = m or xf = 2 − m for the other rotor; in forward flight of side-by-side rotors,
xf = 1/(1 + /2R)2 (x = 1/2 if there is no overlap, /2R > 1). The transition between hover and forward
flight is accomplished using
xf μ2 + xh Cλ2h
x=
μ2 + Cλ2h
with typically C = 1 to 4. This transition is applied to x for both rotors, and to κtwin .
With a coaxial rotor in hover, the lower rotor acts in the contracted wake of the upper rotor.
Momentum theory gives the ideal induced power for coaxial rotors with large vertical separation (ref. 15):
√
Pu = Tu vu , vu2 = Tu /2ρA for the upper rotor; and P = (ᾱs/ τ )T v , v2 = T /2ρA for the lower rotor.
Here τ = T /Tu ; ᾱ is the average of the disk loading weighted by the induced velocity, hence a measure
of nonuniform loading on the lower rotor (ᾱ = 1.05 to 1.10 typically); and the momentum theory solution
is  
ᾱs 1
√ = 3/2 1 + 4(1 + τ )2 ᾱτ − 1
τ 2τ
130 Rotor

The optimum solution for equal power of the upper and lower rotors is ᾱsτ = 1, giving τ = T /Tu ∼
= 2/3.
Hence for the coaxial rotor in hover the ideal induced velocity is calculated from CT e = CT u for the
√
upper rotor and from CT e = (ᾱs/ τ )2 CT  for the lower rotor, with κtwin = 1. Thus xh = 1/(2−m) = 1/2
and the input hover κtwin is not used, unless the coaxial rotor is modeled as a tandem rotor with zero
longitudinal separation.

For multiple rotors, the model for ideal induced velocity is extended to use CT e = xn CT n (sum
over all rotors, scaled with square of tip speed), and the correction factor κtwin . The thrust factors xn are
input for hover and forward flight.

12-5.2 Table Performance Method

The induced power is calculated from the ideal power: Pi = κPideal = κfD T videal . The induced
power factor κ is obtained from an input table (linearly interpolated) that can be a function of up to three
independent variables. Optionally κ from the equations can be retained as an increment to the table
value.
The profile power is calculated from a mean blade drag coefficient: Po = ρA(ΩR)3 CP o =
ρA(ΩR)3 σ8 cdmean FP . The mean drag coefficient cdmean , or alternatively cdmean FP = 8CP o /σ , is ob-
tained from an input table (linearly interpolated) that can be a function of up to three independent
variables. Optionally cdmean from the equations can be retained as an increment to the table value.
The independent variables can be velocity magnitude V /Vtip , horizontal velocity Vh /Vtip , edgewise
advance ratio μ (hub plane or tip-path plane), axial velocity ratio μz (hub plane or tip-path plane), shaft
angle of attack α = tan−1 (μz /μ) (hub plane or tip-path plane), blade loading CT /σ , lift offset Mx /T R,
tip Mach number Mtip , or advancing tip Mach number Mat . The thrust value used in the table can
include download. The thrust value used in the table can include ground effect (thrust divided by κg ),
or the power result from the table can include ground effect (power multiplied by fg ).

12-5.3 Power Adjustments

Rotor power adjustments are associated with wing interference (Pw ), propulsive force efficiency
(Pd ), and climb efficiency (Pv ).
The interference power can be produced by interactions from the wing. The component of the
F
wing interference velocity vind parallel to the rotor force vector F F (velocity roughly normal to the rotor
disk) produces a power change Pw ∼ T
= vind F . The component of the interference velocity perpendicular
to the rotor force vector (velocity roughly in the plane of the rotor disk) produces interference through
the swirl, hence Pw ∝ (V /ΩR)|vind ||F |. Thus the interference power is calculated from

V
Pw = −Kintn vind
T
F + Kintp (|vind ||F |)2 − (vind
T F )2
ΩR

Separate interference factors Kint are used for the two terms. Kintp is negative for favorable interference.
An effect of propulsive force on the rotor induced and profile power can be included in terms of an
efficiency η :  
1
Pd = V ΔD −1
η
 
where ΔD = q D/q − (D/q)ref , and the reference propulsive force (D/q)ref corresponds to the baseline
profile and induced power models. In terms of the rotor wind axis forces, the propulsive force is here
Rotor 131

D = −X (or D = −X cos αc − L sin αc in climb, tan αc = Vc /Vh ). The efficiency is a function of rotor
speed and blade loading, η(CT /σ, V /Vtip ). This increment is intended for use with the table model for
the reference propulsive force.
The main rotor propulsive power Pp = −XV includes the climb power Pc = WG Vc , and the influence
of climb on the induced power is contained in videal and the appropriate terms of κ. Alternatively, the
influence of climb on induced and profile power can be obtained from an efficiency factor fclimb as a
function of Vc /vh : P − Plevel = WG Vc fclimb . In high speed flight, climb or descent does not change the
induced velocity much, so the net change in induced power due to climb would be WG Vc (fclimb − 1).
In low speed flight or axial climb, the induced velocity is influenced by climb rate. Let the difference
between the ideal induced velocity and the level flight induced velocity be videal − vlevel = vh x. Then
the net change in power due to climb is Pv = WG Vc (fclimb − 1) − κT vh x. In axial climb
   
videal − vlevel = −Vc /2 + (Vc /2)2 + vh2 − vh = vh −Vc /vh 2 + (Vc /vh 2)2 + 1 − 1 = vh x

In forward flight, vlevel is evaluated as for videal , but with μz = 0. With multiple main rotors, Pv for each
rotor is obtained from WG /Nrotor . This climb power increment is intended for use with the table model
for level flight power.

12–6 Power Required and Reaction Drive

The total rotor power required is Preq = Pi + Po + Pp + Px . In most helicopter designs the power
is delivered to the rotor by a mechanical drive, through the rotor shaft torque. Such designs require a
transmission and a means for balancing the main-rotor torque. The shaft power is Pshaft = Preq , which
contributes to the propulsion group power required, PreqP G , and produces a torque on the aircraft.
An alternative is to supply the power by a jet reaction drive of the rotor, using cold or hot air
ejected out of the blade tips or trailing edges. Helicopters have also been designed with ram jets on the
blade tips, or with jet flaps on the blade trailing edges that use compressed air generated in the fuselage.
Since there is no torque reaction between the helicopter and rotor (except for the small bearing friction),
no transmission or antitorque device is required, resulting in a considerable weight saving. With a jet
reaction drive, the propulsion system is potentially lighter and simpler, although the aerodynamic and
thermal efficiency are lower. The helicopter must still have a mechanism for yaw control. With reaction
drive the shaft power is Pshaft = Preq − Preact , where the reaction power Preact contributes to the engine
group or jet group power required. The reaction drive produces a force Freact on the rotor blades at
effective radial station rreact , so Preact = Ωrreact Freact . Momentum balance gives the total force
*   +
F = ṁ Vreact − Ωrreact eψ − ṁV ex

where eψ and ex are unit vectors perpendicular to the blade and in the free stream direction; and the
sum is over all blades. The average force in the nonrotating frame is the drag of the inlet momentum
(ṁreact V ), which is accounted for in the engine group or jet group model. The mean in the rotating
frame gives the total jet force required
 
Freact = ṁreact Vreact − Ωrreact

The engine group or jet group performance includes the blade duct and nozzle, perhaps even with tip
burning. Optionally the reaction power can be set equal to the rotor power required (Preact = Preq ), so
132 Rotor

Pshaft = 0. If the reaction drive is turned off (Preact = 0 and Freact = 0), then the rotor must be trimmed
such that Preq = 0.

12–7 Performance Metrics

Several performance metrics are calculated for each rotor. The induced power factor is κ = Pi /Pideal .
The rotor mean drag coefficient is cd = (8CP o /σ)/FP , using the function F (μ, μz ) given previously. The
rotor effective lift-to-drag ratio is a measure of the induced and profile power: L/De = V L/(Pi + Po ).
The hover figure of merit is M = T fD v/P . The propeller propulsive efficiency is η = Pp /P = −XV /P .
These two metrics can be combined as a momentum efficiency: ηmom = T (V + w/2)/P , where w/2 =
fW v/2 = fD v .

12–8 Interference

The rotor can produce aerodynamic interference velocities at the other components (fuselage, wings,
and tails). The induced velocity at the rotor disk is κvi , acting opposite the thrust (z -axis of tip-path
P
plane axes). So vind = −k P κvi , and vind
F
= C F P vind
P
. The total velocity of the rotor disk relative to the
air consists of the aircraft velocity and the induced velocity from this rotor: vtotalF
= v F − vind
F
. The
direction of the wake axis is thus ew = −C vtotal /|vtotal | (for zero total velocity, ew = −k is used).
P PF F F P P

The angle of the wake axis from the thrust axis is χ = cos−1 |(kP )T ePw |.
F F
The interference velocity vint at each component is proportional to the induced velocity vind (hence
is in the same direction), with factors accounting for the stage of wake development and the position of
the component relative to the rotor wake. The far wake velocity is w = fW vi , and the contracted wake
area is Ac = πRc2 = A/fA . The solution for the ideal inflow gives fW and fA . For an open rotor, fW = 2.
For a ducted rotor, the inflow and wake depend on the wake area ratio fA , or on the ratio of the rotor
thrust to total thrust: fT = Trotor /T . The corresponding velocity and area ratios at an arbitrary point on
the wake axis are fw and fa , related by

μ2 + (μz + fw λi )2
fa =
(fV x μ)2 + (fV z μz + λi )2

Vortex theory for hover gives the variation of the induced velocity with distance z below the rotor disk:
 
z/R
v = v(0) 1 + 
1 + (z/R)2

With this equation the velocity varies from zero far above the disk to v = 2v(0) far below the disk. To
use this expression in edgewise flow and for ducted rotors, the distance z/R is replaced by ζw /tR, where
ζw is the distance along the wake axis, and the parameter t is introduced to adjust the rate of change (t
small for faster transition to far wake limit). Hence the velocity inside the wake is fw vi , where
⎧
⎪
⎪ ζw /tR
⎪
⎪ 1+  ζw < 0
⎨ 1 + (ζw /tR)2
fw = fW fz =
⎪
⎪ ζw /tR
⎪
⎩ 1 + (fW − 1) 
⎪ ζw > 0
1 + (ζw /tR)2
√
and the contracted radius is Rc = R/ fa .
Rotor 133

The wake is a skewed cylinder, starting at the rotor disk and with the axis oriented by ePw . The
F
interference velocity is required at the position zB on a component. Whether this point is inside or
outside the wake cylinder is determined by finding its distance from the wake axis, in a plane parallel to
P
the rotor disk. The position relative to the rotor hub is ξB = C P F (zB
F
− zhub
F
); the corresponding point on
P P P P
the wake axis is ξA = ew ζw . Requiring ξB and ξA have the same z value in the tip-path plane axes gives

(k P )T C P F (zB
F
− zhub
F
)
ζw =
(k P )T eP
w

from which fz , fw , fa , and Rc are evaluated. The distance r from the wake axis is then
 P 2
  P 2

r2 = (iP )T (ξB
P
− ξA ) + (j P )T (ξB
P
− ξA )

The transition from full velocity inside the wake to zero velocity outside the wake is accomplished in
the distance sRc , using
⎧
⎨1 r ≤ Rc
fr = 1 − (r − Rc )/(sRc )
⎩
0 r ≥ (1 + s)Rc

(s = 0 for an abrupt transition, s large for always in wake).

F F
The interference velocity at the component (at zB ) is calculated from the induced velocity vind ,
the factors fW fz accounting for axial development of the wake velocity, the factor fr accounting for
immersion in the wake, and an input empirical factor Kint :
F F
vint = Kint fW fz fr ft vind

An additional factor ft for twin rotors is included. Optionally the development along the wake axis
can be a step function (fW fz = 0, 1, fW above the rotor, on the rotor disk, and below the rotor disk,
respectively); nominal (t = 1); or use an input rate parameter t. Optionally the wake immersion can use
the contracted radius Rc or the uncontracted radius R; can be a step function (s = 0, so fr = 1 and 0
inside and outside the wake boundary); can be always immersed (s = ∞ so fr = 1 always); or can use
an input transition distance s. Optionally the interference factor Kint can be reduced from an input value
at low speed to zero at high speed, with linear variation over a specified speed range.
To account for the extent of the wing or tail area immersed in the rotor wake, the interference
velocity is calculated at several points along the span and averaged. The increment in position is
ΔzBF
= C F B (0 Δy 0)T , Δy = (b/2)(−1 + (2i − 1)/N ) for i = 1 to N ; where b is the wing span. The
average interference is calculated separately for each wing panel (left and right), by interpolating the
interference velocity at N points along the wing to N/2 points along the panel span.
For twin main-rotors (tandem, side-by-side, or coaxial), the performance may be calculated for
the rotor system, but the interference velocity is still calculated separately for each rotor, based on
its disk loading. At the component, the velocities from all rotors are summed, and the total used to
calculate the angle of attack and dynamic pressure. This sum must give the interference velocity of
the twin rotor system, which requires the correction factor ft . Consider differential momentum theory
to estimate the induced velocity of twin rotors in hover. For the first rotor, the thrust

and area in the
non-overlap region are (1 − m)T1 and (1 − m)A, hence the induced velocity is v1 = κ T1 /2ρA; similarly

v2 = κ T2 /2ρA. In the overlap region the thrust and area are mT1 + mT2 and mA, hence the induced
134 Rotor

velocity is vm = κ (T1 + T2 )/2ρA. So for equal thrust, the velocity in the overlap region (everywhere
√
for the coaxial configuration) is 2 larger. The factor KT is introduced to adjust the overlap velocity:
√ 
vm = κ(KT / 2) (T1 + T2 )/2ρA. The interference velocities are calculated separately for the two
 
rotors, with the correction factor ft : vint1 = ft κ T1 /2ρA and vint2 = ft κ T2 /2ρA. The sum vint1 + vint2
must take the required value. Below the non-overlap region, the component is in the wake of only one
of the rotors, so the interference velocity from the other rotor is zero, and thus ft = 1. Below the overlap
region, the component is in the wake of both rotors, and the sum of the interference velocities equals vm
if √
KT / 2
fth = √ √
τ1 + τ 2
√
where τn = Tn /(T1 + T2 ) is the thrust ratio. For equal thrusts, fth = KT /2; or fth = 1/ 2 for the
nominal velocity. The expression ft = fth cos2 χ + sin2 χ gives the required correction factor, with ft = 1
√
in edgewise flight. Optionally the correction for twin rotors can be omitted (ft = 1); nominal (KT = 2);
or use an input velocity factor in the overlap region (KT ).

12–9 Drag
F
The rotor component includes drag forces acting on the hub and spinner (at zhub ) and on the pylon
F
(atzpylon ). The component drag contributions must be consistent. In particular, a rotor with a spinner
(such as on a tiltrotor aircraft) would likely not have hub drag. The pylon is the rotor support and the
nacelle is the engine support. The drag model for a tiltrotor aircraft with tilting engines would use the
pylon drag (and no nacelle drag), since the pylon is connected to the rotor shaft axes; with non-tilting
engines it would use the nacelle drag as well.
The body axes for the drag analysis are rotated about the y -axis relative to the rotor shaft axes:
BF
C = C BS C SF , where C BS = Y−θref . The pitch angle θref can be input, or the rotation appropriate for
a helicopter rotor or a propeller can be specified.
a) Consider a helicopter rotor, with the shaft axes oriented z -axis up and x-axis
downstream. It is appropriate that the angle of attack is α = 0 for forward flight and
α = −90 deg for hover, meaning that the body axes are oriented z -axis down and
x-axis forward. Hence θref = 180 deg.
b) Consider a propeller or tiltrotor, with the shaft axes oriented z -axis forward and
x-axis up. It is appropriate that the angle of attack is α = 0 in cruise and α = 90 deg
for helicopter mode (with a tilting pylon), meaning that the body axes are oriented
z -axis down and x-axis forward. Hence θref = 90 deg.

The aerodynamic velocity relative to the air is calculated in component axes, v B . The angle of attack
α and dynamic pressure q are calculated from v B . The reference areas for the drag coefficients are the
rotor disk area A = πR2 (for hub drag), pylon wetted area Spylon , ducted wetted area 2Sduct , and spinner
wetted area Sspin ; these areas are input or calculated as described previously.
The hub drag can be fixed, specified as a drag area D/q ; or the drag can be scaled, specified as a
drag coefficient CD based on the rotor disk area A = πR2 ; or the drag can be estimated based on the
gross weight, using a squared-cubed relationship or a square-root relationship. Based on historical data,
the drag coefficient CD = 0.004 for typical hubs, CD = 0.0024 for current low-drag hubs, and CD =
0.0015 for faired hubs. For the squared-cubed relationship: (D/q)hub = k((WM T O /Nrotor )/1000)2/3
(where WM T O /Nrotor is the maximum takeoff gross weight per lifting rotor; units of k are ft2 /k-lb2/3 or
Rotor 135

m2 /Mg2/3 ). Based on historical data, k = 1.4 for typical hubs, k = 0.8 for current low-drag
hubs, and
k = 0.5 for faired hubs (English units). For the square-root relationship: (D/q)hub = k WM T O /Nrotor
(where WM T O /Nrotor is the maximum takeoff gross weight per lifting rotor; units of k are ft2 /lb1/2 or
m2 /kg1/2 ); based on historical data (ref. 16), k = 0.074 for single rotor helicopters, k = 0.049 for tandem
rotor helicopters (probably a blade number effect), k = 0.038 for hingeless rotors, and k = 0.027 for faired
hubs (English units). To handle multi-rotor aircraft, the scaling weight w = WM T O /Nrotor is calculated
as for disk loading: w = fW WM T O for main-rotors or w = f T for antitorque and auxiliary-thrust rotors.
The hub vertical drag can be fixed, specified as a drag area D/q ; or the drag can be scaled, specified
as a drag coefficient CD based on the rotor disk area A = πR2 .
The pylon forward flight drag and vertical drag are specified as drag area or drag coefficient, based
on the pylon wetted area. The duct forward flight drag and vertical drag are specified as drag area or drag
coefficient, based on the duct wetted area. The spinner drag is specified as drag area or drag coefficient,
based on the spinner wetted area.
The drag coefficient for the hub or pylon or duct at angle of attack α is
CD = CD0 + (CDV − CD0 )| sin α|Xd
Optionally the variation can be quadratic (Xd = 2). For sideward flight, CDhub = CD0 for the hub,
CDpylon = CDV for the pylon, and CDduct = CDV for the duct. Then the total component drag force is
D = qACDhub + qSpylon CDpylon + qSduct CDduct + qSspin CDspin
The dynamic pressure and position of the hub are used for the duct and spinner. The force and moment
produced by the drag are 
FF = ed D

MF = ,F F F
Δz
where Δz F = z F − zcg
F
(separate locations are defined for the rotor hub and for the pylon), and ed is the
drag direction. The velocity relative to the air gives ed = −v F /|v F | (no interference).

12–10 Weights

The rotor configuration determines where the weights occur in the weight statement, as summarized
in table 12-3. The rotor group consists of blade assembly, hub and hinge, fairing/spinner, blade fold
structure, inter-rotor shaft, rotor support, and duct. The tail-rotor (in empennage group) or the propeller/
fan installation (in propulsion group) consists of blade assembly, hub and hinge, rotor support, and duct.
There are separate weight models for main-rotors, tail-rotors, and auxiliary-thrust systems (pro-
pellers). The tail-rotor model requires a torque calculated from the drive system rated power and
main-rotor rotational speed: Q = PDSlimit /Ωmr . The auxiliary-thrust model requires the design max-
imum thrust of the propeller. The engine section or nacelle group includes the engine support weight
and pylon support weight; these must be consistent with the use of the rotor support structural weight.
Table 12-3. Principal configuration designation.
configuration weight statement weight model performance model
main-rotor rotor group rotor rotor
tail-rotor empennage group tail-rotor rotor
propeller propulsion group rotor, aux thrust rotor
136 Rotor

The flap moment of inertia Ib and the Lock number γ = ρacR4 /Ib are required for the blade motion
solution. Several options are implemented to calculate Ib . The Lock number can be specified, and
then Ib = ρacR4 /γ used, independent of the blade weight; this is the only option for the tail-rotor and
auxiliary-thrust weight models, which do not give separate blade and hub weight estimates. The moment
of inertia Ib can be calculated from the blade weight and the weight distribution. The Lock number
can be specified, hence Ib = ρacR4 /γ , and then mass added to the blade to achieve this value. An
autorotation index AI = KE/P = 12 N Ib Ω2 /P can be specified, hence the required Ib , and then mass
added to the blade to achieve this value. Reference 17 describes this and other autorotation indices;
AI = KE/P ≥ 3 sec gives good autorotation characteristics for small helicopters.

In order to increase the moment of inertia, a tip weight Wt can be added to each blade at radial
station rt . Thus the total blade weight is Wb = χwb + dWb + (1 + f )Wt N (lb or kg); where wb is the blade
weight estimate, χ the technology factor, dWb a specified weight increment; and the factor f accounts for
the blade weight increase required by the centrifugal force due to Wt . The mass per blade is Mb = Wb /N
(slug or kg), or Mb0 without the tip weight. The blade moment of inertia is

Ib = R2 (r22 (Mb0 + f Mt ) + rt2 Mt ) = Ib0 + R2 (rt2 + f r22 )Mt

√
where r2 is the radius of gyration of the distributed mass. Typically r2 ∼ = 0.6; r2 = 1/ 3 = 0.577
for uniform mass distribution. If the required moment of inertia Ib is greater than Ib0 , the tip mass
Mt is needed. Additional mass is required inboard to react the centrifugal force increase due to Mt .
This additional mass is less effective than Mt at increasing Ib . Assume a fraction a of the blade mass
!
reacts the centrifugal force F , so ΔM/Mb0 = aΔF/F0 . The reference values are Mb0 = R m dr and
!
F0 = Ω2 R2 rm dr = Ω2 Rr1 Mb0 , where r1 ∼ = 12 . Then
 
Mb0 1  2 2
 rt art /r1
ΔM = a ΔF = a 2 Ω Rr1 ΔM + Ω Rrt Mt = a ΔM + Mt = Mt = f Mt
F0 Ω Rr1 r1 1−a

With a ∼= 13 , f ∼
= 1. The tip mass required to produce ΔIb = Ib − Ib0 is Mt = ΔIb /(R2 (rt2 + f r22 )), and
the total blade weight increment is ΔWb = (1 + f )Mt N (lb or kg).

12–11 References

1) Cheeseman, I.C., and Bennett, W.E. “The Effect of the Ground on a Helicopter Rotor in Forward
Flight.” ARC R&M 3021, September 1955.
2) Law, H.Y.H. “Two Methods of Prediction of Hovering Performance.” USAAVSCOM TR 72-4,
February 1972.
3) Hayden, J.S. “The Effect of the Ground on Helicopter Hovering Power Required.” American
Helicopter Society 32nd Annual National V/STOL Forum, Washington, D.C., May 1976.
4) Zbrozek, J. “Ground Effect on the Lifting Rotor.” ARC R&M 2347, July 1947.
5) Schmaus, J.; Berry, B.; Gross, W.; and Koliais, P. “Experimental Study of Rotor Performance in
Deep Ground Effect with Application to a Human-Powered Helicopter.” American Helicopter Society
68th Annual Forum, Fort Worth, TX, May 2012.
6) Coleman, R.P.; Feingold, A.M.; and Stempin, C.W. “Evaluation of the Induced-Velocity Field of an
Idealized Helicopter Rotor.” NACA ARR L5E10, June 1945.
Rotor 137

7) Mangler, K.W., and Squire, H.B. “The Induced Velocity Field of a Rotor.” ARC R & M 2642, May
1950.
8) Drees, J.M. “A Theory of Airflow Through Rotors and Its Application to Some Helicopter Problems.”
Journal of the Helicopter Association of Great Britain, Vol. 3, No. 2, July–September 1949.
9) White, T., and Blake, B.B. “Improved Method of Predicting Helicopter Control Response and Gust
Sensitivity.” Annual National Forum of the American Helicopter Society, May 1979.
10) Harris, F.B. “Rotor Performance at High Advance Ratio; Theory versus Test.” NASA CR 2008-
215370, October 2008.
11) Gessow, A., and Crim, A.D. “A Theoretical Estimate of the Effects of Compressibility on the
Performance of a Helicopter Rotor in Various Flight Conditions.” NACA TN 3798, October 1956.
12) Mason, W.H. “Analytic Models for Technology Integration in Aircraft Design.” AIAA Paper No.
90-3262, September 1990.
13) Ashley, H., and Landahl, M. Aerodynamics of Wings and Bodies. Reading, Massachusetts: Addison-
Wesley Publishing Company, Inc., 1965.
14) Spreiter, J.R., and Alksne, A.Y. “Thin Airfoil Theory Based on Approximate Solution of the
Transonic Flow Equation.” NACA Report 1359, 1958.
15) Johnson, W. “Influence of Lift Offset on Rotorcraft Performance.” NASA TP 2009-215404,
November 2009.
16) Keys, C.N., and Rosenstein, H.J. “Summary of Rotor Hub Drag Data.” NASA CR 152080, March
1978.
17) Wood, T.L. “High Energy Rotor System.” American Helicopter Society 32nd Annual National
V/STOL Forum, Washington, D.C., May 1976.
138 Rotor
 Chapter 13

Wing

The aircraft can have one or more wings, or no wings.

13–1 Geometry

The wing is described by planform area S , span b, mean chord c = S/b, and aspect ratio AR = b2 /S .
These parameters are for the entire wing. The geometry is specified in terms of two of the following
parameters: S or wing loading W/S , b (perhaps calculated from other geometry), c, AR = b2 /S . With
more than one wing, the wing loading is obtained from an input fraction of design gross weight,
W = fW WD . Optionally the span can be calculated from a specified ratio to the span of another wing; or
the span can be calculated from a specified ratio to the radius of a designated rotor, b = 2f R. Optionally
the wing span can be calculated from an appropriate specification of all wing panel widths.
Optionally for the tiltrotor configuration, the wing span can be calculated from the fuselage and rotor
geometry: b = 2(f R + dfus ) + wfus , where R is the rotor radius (cruise value for variable-diameter rotor),
dfus the rotor-fuselage clearance, and wfus the fuselage width. Note that the corresponding option for the
rotor hub position is yhub = ±(f R + dfus + 1/2wfus ). Optionally the wing span can be calculated from the
rotor hub position: b = 2|yhub | (regardless of how the rotor position is determined). As implemented,
symmetry is not assumed; rather the radius or hub position of the outermost designated rotors is used.
The wing is at position z F , where the aerodynamic forces act. The component axes are the aircraft
body axes, C BF = I .
The wing planform is defined in terms of one or more wing panels (fig. 13-1). Symmetry of the
wing is assumed. The number of panels is P , with the panel index p = 1 to P . The wing span station η is
scaled with the semi-span: y = η(b/2), η = 0 to 1. Each panel is a trapezoid, with a straight aerodynamic
center and linear taper. The aerodynamic center locus (in wing axes) is defined by sweep Λp ; dihedral δp ;
and offsets (xIp , zIp ) at the inboard edge relative to the aerodynamic center of the previous panel. The
wing position z F is the mean aerodynamic center. The offset (x̄A , z̄A ) of the mean aerodynamic center
from the root chord aerodynamic center is calculated (so the wing planform can be drawn; typically the
aerodynamic center is drawn as the quarter-chord). Outboard panel edges are at ηEp (input or calculated).
A panel is characterized by span bp (each side), mean chord cp , and area Sp = 2bp cp (both sides). The
taper is defined by inboard and outboard chord ratios, λ = c/cref (where cref is a panel or wing reference
chord, depending on the options for describing the geometry).
The span for each panel (if there are more than two panels) can be a fixed input; a fixed ratio of the
wing span, bp = fbp (b/2); or free. The panel outboard edge (except at the wing tip) can be at a fixed input
position yEp ; at a fixed station ηEp , yp = ηEp (b/2); calculated from the rotor radius, yp = f R; calculated
from the fuselage and rotor geometry, yp = f R + dfus + 1/2wfus (for a designated rotor); calculated from
140 Wing

centerline

panel p
(x A , z A ) aero center locus

mean aero
center
(wing location)

inboard outboard
edge edge
ηEp outboard panel edge, η = y/(b/2)
ηIp ηOp wing station
λIp λOp chord ratio, λ = c/cref
cIp cOp chord

Λp sweep (+ aft)
δp dihedral (+ up)
x Ip aero center offset (inboard, + aft)
zIp aero center offset (inboard, + up)

bp span (each side)

cp mean chord
Sp area = 2 b p cp

Figure 13-1. Wing geometry (symmetric, only right half-wing shown).

the hub position, yp = |yhub | (for a designated rotor); or adjusted. An adjusted station is obtained from
the last station and the span of this panel, yp = yp−1 + bp or yp = yp−1 + fbp (b/2); or from the next station
and the span of the next panel, yp = yp+1 − bp+1 or yp = yp+1 − fb(p+1) (b/2). The specification of panel
spans and panel edges must be consistent, and sufficient to determine the wing geometry. Determining
the panel edges requires the following steps:

a) Calculate the panel edges that are either at fixed values (input, or from width, or from hub
position) or at fixed stations; root and tip edges are known.
b) Working from root to tip, calculate the adjusted panel edge yp if panel span bp or ratio fbp
is fixed, and if previous edge yp−1 is known.
c) Working from tip to root, calculate the adjusted panel edge yp (if not yet known) if panel
span bp+1 or ratio fb(p+1) is fixed, and if next edge yp+1 is known.

At the end of this process, all edges must be known and the positions yp must be unique and sequential.
If this geometry is being determined for a known span, then there must not be a fixed panel span or span
Wing 141

ratio that has not been used. Alternatively, if the wing span is being calculated from the specification of
all panel widths, then the process must leave one and only one fixed panel span or span ratio that has not
been used. Since the wing span is to be calculated, each panel edge is known in the form yp = c0 + c1 b/2.
Then the unused fixed panel span gives the equation (c0 + c1 b/2)O − (c0 + c1 b/2)I = bp (subscript O
denotes outboard edge, subscript I denotes inboard edge), or the unused fixed panel span ratio gives the
equation (c0 + c1 b/2)O − (c0 + c1 b/2)I = fp b/2, which can be solved for the semispan b/2.
To complete the definition of the geometry, one of the following quantities is specified for each
panel: panel area Sp ; ratio of panel area to wing area, fs = Sp /S ; panel mean chord cp ; ratio of panel
mean chord to wing mean chord, fc = cp /c; chord ratios λI = cI /cref and λO = cO /cref (taper); or free.
The total wing area equals the sum of all panel areas:
     1
S= Sp + S fs + 2 bp cp + 2c bp fc + 2cref bp (λI + λO )
2
If there is one or more taper specification (and no free), then cref is calculated from this equation for S ,
and the mean chord is cp = 12 (cI + cO ) = cref 12 (λI + λO ), Sp = 2bp cp . If there is one (and only one) free
specification, then Sp is calculated from this equation for S , and the mean chord is cp = Sp /(2bp ), with
cI = 2cp /(1 + λO /λI ), cO = 2cp − cI .

Since the panels have linear taper (c = cref λ), the mean aerodynamic chord is
 b/2  1
2
Sc̄A = c dy = b c2ref λ2 dη
−b/2 0
 1 1
=b c2ref (λ2I + λI λO + λ2O ) Δηp = (c2 + cI cO + c2O ) 2bp
3 3 I
 b/2  1
S= c dy = b cref λ dη
−b/2 0
1  1
=b
cref (λI + λO ) Δηp = (cI + cO ) 2bp
2 2
These expressions are evaluated from panel cI and cO , as calculated using λI and λO , or using the ratio
λO /λI (cref may not be the same for all panels).

The mean aerodynamic center is the point where there is zero moment due to lift: x̄A CL S =
! !
x̄A c c dy = xc c dy , with cc = (y) the spanwise lift distribution. Thus
 1
(η)(x̄A − xAC (η)) dη = 0
0
The locus of section aerodynamic centers xAC is described by the panel sweep Λp and the offset xIp at
the inboard end of the panel. These offsets can be a fixed input,

a fraction of the root chord, or a fraction
of the panel inboard chord. Assuming elliptical loading ( = 1 − η 2 ) gives
 1    
π b
x̄A = (η)xAC dη = 1 − η2 x/Ip + tan Λp η dη
4 0 2
 1    b 1
ηO
= /Ip η 1 − η 2 + sin−1 η − tan Λp (1 − η 2 )3/2
x
2 2 3 ηI
p  
where x/Ip = q=2 xIq + (b/2) tan Λq−1 (ηO(q−1) − ηI(q−1) ) − (b/2) tan Λp ηIp . The vertical position of
the mean aerodynamic center is obtained in a similar fashion, from panel dihedral δp and offset zIp at
the inboard edge of the panel. Assuming uniform loading ( = 1) gives
 1   b   b
η
1 2 O
z̄A = zAC dη = z/Ip + tan δp η dη = z/Ip η + tan δp η
0 2 2 2 ηI
142 Wing

Then (x̄A , z̄A ) is the offset of the mean aerodynamic center from the root chord aerodynamic center.
Finally,
 b 
Λ = tan−1
p
tan Λp
b/2
 b 
δ = tan−1
p
tan δp
b/2
2c
λ= −1
croot
are the wing overall sweep, dihedral, and taper.
The wing contribution to the aircraft operating length is xwing + (0.25c) cos i (forward), xwing −
(0.75c) cos i (aft), and ywing ± b/2 (lateral).

13–2 Control and Loads

The control variables are flap δF , flaperon δf , aileron δa , and incidence i. The flaperon deflection
can be specified as a fraction of flap deflection, or as an increment relative to the flap deflection, or
the flaperon can be independent of the flap. The flaperon and aileron are the same surface, generating
symmetric and antisymmetric loads, respectively, hence with different connections to pilot controls.
With more than one wing panel, each panel can have control variables: flap δF p , flaperon δf p , aileron
δap , and incidence ip . The outboard panel (p ≥ 2) control or incidence can be specified independently,
or in terms of the root panel (p = 1) control or incidence (either fraction or increment).
Each control is described by the ratio of the control surface chord to the wing panel chord, f = cf /cp ;
and by the ratio of the control surface span to wing panel span, fb = bf /bp , such that the control surface
area is obtained from the panel area by Sf = f fb Sp .

13–3 Aerodynamics

The aerodynamic velocity of the wing relative to the air, including interference, is calculated in
component axes, v B . The angle of attack αwing (hence C BA ) and dynamic pressure q are calculated from
v B . The reference area for the wing aerodynamic coefficients is the planform area, S . The wetted area
contribution is twice the exposed area: Swet = 2(S − cwfus ), where wfus is the fuselage width.
The wing vertical drag can be fixed, specified as a drag area (D/q)V ; or the drag can be scaled,
specified as a drag coefficient CDV based on the wing area; or calculated from an airfoil section drag
coefficient (for −90 deg angle of attack) and the wing area immersed in the rotor wake:

1  
CDV = cd90 S − Scenter − fd90 bF cF (1 − cos δF ) − fd90 bf cf (1 − cos δf )
S

The term Scenter = c(wfus + 2dfus ) (where wfus is the fuselage width and dfus the rotor-fuselage clearance)
is the area not immersed in the rotor wake, and is used only for tiltrotors. The last two terms account
for the change in wing area due to flap and flaperon deflection, with an effectiveness factor fd90 .
From the control surface deflection and geometry, the lift coefficient, maximum lift angle, moment
coefficient, and drag coefficient increments are evaluated: ΔCLf , Δαmaxf , ΔCM f , and ΔCDf . These
increments are the sum of contributions from flap and flaperon deflection, hence weighted by the control
surface area. The drag coefficient increment includes the contribution from aileron deflection.
Wing 143

13-3.1 Lift

The wing lift is defined in terms of lift-curve slope CLα and maximum lift coefficient CLmax (based
on wing planform area). The three-dimensional lift-curve slope is input directly or calculated from the
two-dimensional lift-curve slope:

cα
CLα =
1 + cα (1 + τ )/(πAR)

where τ accounts for non-elliptical loading. Optionally the input incompressible lift-curve slope is
multiplied by the compressibility factor Fc . The effective angle of attack is αe = αwing + i − αzl , where
αzl is the angle of zero lift; in reverse flow (|αe | > 90), αe ← αe − 180 signαe . Let αmax = CLmax /CLα be
the angle-of-attack increment (above or below zero lift angle) for maximum lift. Including the change of
maximum lift angle caused by control deflection, Amax = αmax + Δαmaxf and Amin = −αmax + Δαmaxf .
Then ⎧
⎪ CLα αe + ΔCLf Amin ≤ αe ≤ Amax
⎪
⎪
⎪
⎪  
⎪
⎨ (C A f π/2 − |αe |
Lα max + ΔCLf ) max 0, αe > Amax
CL = f π/2 − |Amax |
⎪
⎪  
⎪
⎪
⎪
⎪ f π/2 − |αe |
⎩ (CLα Amin + ΔCLf ) max 0, αe < Amin
f π/2 − |Amin |

(for zero lift at f 90 deg angle of attack, f = .9). Note that CLα Amax + ΔCLf = CLα αmax + ΔCLmaxf . In
sideward flight, CL = 0. Finally, L = qSCL is the lift force.

13-3.2 Pitch Moment

The wing pitch moment coefficient is CM = CM ac +ΔCM f . Then M = qScCM is the pitch moment.

13-3.3 Roll Moment

The only wing roll moment considered is that produced by aileron control. Typically the flaperon
and aileron are the same surface, but they are treated separately in this model. The aileron geometry is
specified as for the flaperon and flap, hence includes both sides of the wing. The lift coefficient increment
ΔCLa is evaluated as for the flaperon, so one-half of this lift acts up (on the right side) and one-half
acts down. The roll moment is then Mx = 2(ΔLa /2)y , where y is the lateral position of the aileron
aerodynamic center, measured from the wing centerline (defined as a fraction of the wing semi-span).
y 1
The roll moment coefficient is C = − b/2 2 ΔCLa . Then Mx = qSbC is the roll moment.

13-3.4 Drag

The drag area or drag coefficient is defined for forward flight and vertical flight. The effective
angle of attack is αe = αwing + i − αDmin , where αDmin is the angle of minimum drag; in reverse flow
(|αe | > 90), αe ← αe − 180 signαe . For angles of attack less than a transition angle αt , the drag coefficient
equals the forward flight (minimum) drag CD0 , plus an angle of attack term and the control increment.
If the angle of attack is greater than a separation angle αs < αt , there is an additional drag increase.
Thus if |αe | ≤ αt , the profile drag is

CDp = CD0 (1 + Kd |αe |Xd + Ks (|αe | − αs )Xs ) + ΔCDc + ΔCDf

144 Wing

where the separation (Ks ) term is present only for |αe | > αs ; and otherwise

CDt = CD0 (1 + Kd |αt |Xd + Ks (|αt | − αs )Xs ) + ΔCDc + ΔCDf

 
π |αe | − αt
CDp = CDt + (CDV − CDt ) sin
2 π/2 − αt

ΔCDc is the compressibility drag increment. Optionally there might be no angle of attack variation at low
angles (Kd = 0 and/or Ks = 0), or quadratic variation (Xd = 2), or cubic variation for the separation term
(Xs = 3). For sideward flight (vxB = 0) the drag is obtained using φv = tan−1 (−vzB /vyB ) to interpolate
the vertical coefficient: CD = CD0 cos2 φv + CDV sin2 φv . The induced drag is obtained from the lift
coefficient, aspect ratio, and Oswald efficiency e:

(CL − CL0 )2
CDi =
πeAR

Conventionally the Oswald efficiency e represents the wing parasite drag variation with lift, as well as the
induced drag (hence the use of CL0 ). If CDp varies with angle of attack, then e is just the span efficiency
factor for the induced power (and CL0 should be zero). The wing-body interference is specified as a
drag area, or a drag coefficient based on the wing area. Then
 
D = qSCD = qS CDp + CDi + CDwb

is the drag force. The other forces and moments are zero.

13-3.5 Wing Panels

The wing panels can have separate controls, different incidence angles, and different interference
from the rotors. Thus the lift, drag, and moment coefficients are evaluated separately for each panel,
based on the panel area Sp and mean chord cp . The coefficient increments due to control surface deflection
are calculated using the ratio of the control surface area to panel area, Sf /Sp = f fb . The lateral position
of the aileron aerodynamic center is ηa bp from the panel inboard edge, so y/(b/2) = ηE(p−1) + ηa bp /(b/2)
from the wing centerline. Then the total wing loads are:

L= qp Sp CLp

M= qp Sp cp CM p

Mx = qp Sp bCp

D= qp Sp CDpp + qS (CDi + CDwb )

The sums are over all panels (left and right). The reference area is S = Sp , accounting for possible

absence of wing extensions. The total wing coefficients are based on qS = qp Sp . The three-
dimensional lift-curve slope CLα is calculated for the entire wing and used for each panel. The induced
drag is calculated for the entire wing, from the total CL . Since C BF = I ,
⎛ ⎞
 −CDpp − CDi − CDwb
F F = C F B C BA F A = CpBA qp Sp ⎝ 0 ⎠
−CL

is the wing aerodynamic force.

Wing 145

13-3.6 Interference
With more than one wing, the interference velocity at other wings is proportional to the induced
F F
velocity of the wing producing the interference: vint = Kint vind . The induced velocity is obtained
from the induced drag, assumed to act in the k direction: αind = vind /|v B | = CDi /CL = CL /(πeAR),
B

F
vind = C F B k B |v B |αind . For tandem wings, typically Kint = 2 for the interference of the front wing on
the aft wing, and Kint = 0 for the interference of the aft wing on the front wing. For biplane wings, the
mutual interference is typically Kint = 0.7 (upper on lower, and lower on upper). The induced drag is
then
(CL − CL0 )2 
CDi = + CL αint
πeAR
other wings
 
CL
αint = Kint αind = Kint
πeAR other wing
The induced velocity from the rotors is included in the angle of attack of the wing. The rotor interference
must also be accounted for in the wing induced power:
⎡ ⎤
(CL − CL0 )2  
CDi = + CL ⎣ Kint αind + Cint αind ⎦
πeAR rotors
other wings

The angle αind = vind /V is obtained from the rotor induced velocity λi . If the interference is wing-like,
vind = ΩRκλi (so vind ∝ L/ρb2 V ∝ T /2ρAV ). If the interference is propeller-like, vind = V κλi (so
vind ∝ Γ/b ∝ T /2ρAΩR). The interference factor can be evaluated from wing induced drag calculations:
Cint = ΔCDi /(CL κλi /μ) or Cint = ΔCDi /(CL κλi ). Typically for tiltrotors the interference is wing-like,
with Cint ∼
= −0.06.

The wing interference at the tail produces an angle-of-attack change  = E(CL /CLα ), where
E = d/dα is an input factor determined by the aircraft geometry. Then from the velocity v B of the
wing, ⎛ ⎞
−vzB
F
vint =C FB ⎝ 0 ⎠
vxB
is the interference velocity at the tail.
The wing interference at the rotor can produce interference power. The induced velocity at the rotor
F
disk is vint F
= Kint vind F
, with vind = C F B k B |v B |αind again. Separate interference factors Kint are used
for the components of the interference velocity parallel to and perpendicular to the rotor force vector
(roughly normal to and in the plane of the rotor disk).

13–4 Wing Extensions

The wing can have extensions, defined as wing portions of span bX at each wing tip. For the
tiltrotor configuration in particular, the wing weight depends on the distribution of wing area outboard
(the extension) and inboard of the rotor and nacelle location. Wing extensions are defined as a set of

wing panels at the tip. The extension span and area are the sum of the panel quantities, bX = ext bp

and SX = ext Sp . The inboard span and area are then bI = b − 2bX , SI = S − SX . Optionally
the wing extensions can be considered a kit, hence the extensions can be absent for designated flight
conditions or missions. As a kit, the wing extension weight is considered fixed useful load. With wing
extensions removed, the aerodynamic analysis considers only the remaining wing panels. The total wing
146 Wing

coefficients are then based on the area without the extensions. For the induced drag and interference,
the effective aspect ratio is then reduced by the factor (bI /b)2 , since the lift and drag coefficients are still
based on total wing area S .

13–5 Wing Kit

The wing can be a kit, the kit weight an input fraction of the total wing weight. The wing kit
weight can be part of the wing group, or considered fixed useful load. With the kit removed, there are
no aerodynamic loads or aerodynamic interference generated by the wing, and the wing kit weight is
omitted.

13–6 Weights

The wing group consists of: basic structure (primary structure, consisting of torque box and spars,
plus extensions); fairings (leading edge and trailing edge); fittings (non-structural); fold/tilt structure;
and control surfaces (flaps, ailerons, flaperons, and spoilers). There are separate models for a tiltrotor
or tiltwing configuration and for other configurations (including a compound helicopter).
The AFDD wing weight models are based on parameters for the basic wing plus the wing tip
extensions (not the total wing and extensions). The tiltrotor wing model requires the weight on the wing
tips (both sides), consisting of: rotor group, engine system, drive system (except drive shaft), engine
section or nacelle group, air induction group, rotary wing and conversion flight controls, hydraulic group,
trapped fluids, and wing extensions. An adjustment of this calculated weight can be used; a negative
increment is required when the engine and transmission are not at the tip location with the rotor.
 Chapter 14

Empennage

The aircraft can have one or more tail surfaces, or no tail surface. Each tail is designated as
horizontal or vertical, affecting some parameter definitions.

14–1 Geometry

The tail is described by planform area S , span b, chord c = S/b, and aspect ratio AR = b2 /S . The
tail volume can be referenced to rotor radius and disk area, V = S /RA; to wing area and chord for
horizontal tails, V = S /Sw cw ; or to wing area and span for vertical tails, V = S /Sw bw . Here the tail
length is = |xht − xcg | or = |xvt − xcg | for horizontal tail or vertical tail, respectively. The geometry
is specified in terms of S or V ; and b, or AR, or c. The elevator or rudder is described by the ratio of
control surface chord to tail chord, cf /c; and the ratio of control surface span to tail span, bf /b.

The tail contribution to the aircraft operating length is xtail + 0.25c (forward), xtail − 0.75c (aft), and
ytail ± (b/2)C (lateral), where C = cos φ for a horizontal tail and C = cos(φ − 90) for a vertical tail.

14–2 Control and Loads

The tail is at position z F , where the aerodynamic forces act. The scaled input for tail position can
be referenced to the fuselage length, or to the rotor radius.

The horizontal tail can have a cant angle φ (positive tilt to left, becomes vertical tail for φ = 90 deg).
Thus the component axes are given by C BF = X−φ . The control variables are elevator δe and incidence
i.

The convention for nominal orientation of the vertical tail is positive lift to the left, so aircraft
sideslip (positive to right) generates positive tail angle of attack and positive tail lift. The vertical tail
can have a cant angle φ (positive tilt to right, becomes horizontal tail for φ = 90), so the component axes
are given by C BF = X−90+φ . The control variables are rudder δr and incidence i.

14–3 Aerodynamics

The aerodynamic velocity of the tail relative to the air, including interference, is calculated in
component axes, v B . The angle of attack αtail (hence C BA ) and dynamic pressure q are calculated from
v B . The reference area for the tail aerodynamic coefficients is the planform area, S . The wetted area
contribution is Swet = 2S . From the elevator or rudder deflection and geometry, the lift coefficient,
maximum lift angle, and drag coefficient increments are evaluated: ΔCLf , Δαmaxf , and ΔCDf .
148 Empennage

14-3.1 Lift
The tail lift is defined in terms of lift-curve slope CLα and maximum lift coefficient CLmax (based
on tail planform area). The three-dimensional lift-curve slope is input directly or calculated from the
two-dimensional lift-curve slope:
cα
CLα =
1 + cα (1 + τ )/(πAR)

where τ accounts for non-elliptical loading. Optionally the input incompressible lift-curve slope is
multiplied by the compressibility factor Fc . The effective angle of attack is αe = αtail + i − αzl , where
αzl is the angle of zero lift; in reverse flow (|αe | > 90), αe ← αe − 180 signαe . Let αmax = CLmax /CLα be
the angle-of-attack increment (above or below zero lift angle) for maximum lift. Including the change of
maximum lift angle caused by control deflection, Amax = αmax + Δαmaxf and Amin = −αmax + Δαmaxf .
Then ⎧
⎪ CLα αe + ΔCLf Amin ≤ αe ≤ Amax
⎪
⎪
⎪
⎪  
⎪
⎨ (C A f π/2 − |αe |
Lα max + ΔCLf ) max 0, αe > Amax
CL = f π/2 − |Amax |
⎪
⎪  
⎪
⎪
⎪
⎪ f π/2 − |αe |
⎩ (CLα Amin + ΔCLf ) max 0, αe < Amin
f π/2 − |Amin |
(for zero lift at f 90 deg angle of attack, f = .9). Note that CLα Amax + ΔCLf = CLα αmax + ΔCLmaxf . In
sideward flight (defined by (vxB )2 + (vzB )2 < (0.05|v B |)2 ), CL = 0. Finally, L = qSCL is the lift force.

14-3.2 Drag
The drag area or drag coefficient is defined for forward flight and vertical flight. The effective angle
of attack is αe = αtail + i − αDmin , where αDmin is the angle of minimum drag; in reverse flow (|αe | > 90),
αe ← αe − 180 signαe . For angles of attack less than a transition angle αt , the drag coefficient equals
the forward flight (minimum) drag CD0 , plus an angle-of-attack term and the control increment. Thus
if |αe | ≤ αt , the profile drag is

CDp = CD0 (1 + Kd |αe |Xd ) + ΔCDc + ΔCDf

and otherwise
CDt = CD0 (1 + Kd |αt |Xd ) + ΔCDc + ΔCDf
 
π |αe | − αt
CDp = CDt + (CDV − CDt ) sin
2 π/2 − αt
ΔCDc is the compressibility drag increment. Optionally there might be no angle-of-attack variation at
low angles (Kd = 0), or quadratic variation (Xd = 2). In sideward flight (defined by (vxB )2 + (vzB )2 <
(0.05|v B |)2 ), the drag is obtained using φv = tan−1 (−vzB /vyB ) to interpolate the vertical coefficient:
CDp = CD0 cos2 φv + CDV sin2 φv . The induced drag is obtained from the lift coefficient, aspect ratio,
and Oswald efficiency e:
(CL − CL0 )2
CDi =
πeAR
Conventionally the Oswald efficiency e can represent the tail parasite drag variation with lift, as well as
the induced drag (hence the use of CL0 ). Then
 
D = qSCD = qS CDp + CDi
Empennage 149

is the drag force. The other forces and moments are zero.

14–4 V-Tail

A V-tail is modeled as a pair of horizontal and vertical tails, each sized by area or tail volume. This
model follows references 1 and 2. The aerodynamic loads are calculated separately for each tail. The
weight is calculated only for the second tail, using the V-tail area and aspect ratio. The V-tail area and
span are the sum of the two tails:
SV = Sht + Svt
bV = bht + bvt

and the aspect ratio is AR = b2V /SV . The V-tail dihedral angle is δ = tan−1 Svt /Sht , and the cant angle
φ = 0 for both. The location (which should be the same for the two tails) is the midpoint of the V-tail,
vertically and laterally. An upward V-tail and a downward V-tail are identical in this model.

14–5 Weights

The empennage group consists of the horizontal tail, vertical tail, and tail-rotor. The tail plane
weight consists of the basic structure and fold structure. The tail weight (empennage group) model
depends on the configuration: helicopters and compounds, or tiltrotors and tiltwings. Separate weight
models are available for horizontal and vertical tails.
The AFDD tail weight model depends on the design dive speed at sea level (input or calculated).
The calculated dive speed is Vdive = 1.25Vmax , from the maximum speed at the design gross weight and
sea level standard conditions.

14–6 References

1) Raymer, D.P. Aircraft Design: A Conceptual Approach. Fourth Edition. Reston, Virginia: American
Institute of Aeronautics and Astronautics, 2006.
2) Purser, P.E., and Campbell, J.P. “Experimental Verification of a Simplified Vee-Tail Theory and
Analysis of Available Data on Complete Models with Vee Tails.” NACA Report 823, 1945.
150 Empennage
 Chapter 15

Propulsion System

The aircraft propulsion system can be constructed from a number of components: propulsion
groups, engine groups, jet groups, charge groups, and fuel tank systems. Figure 15-1 illustrates the
configuration possibilities.
The aircraft can have one or more propulsion groups, or none. Each propulsion group is a set of
components (rotors) and engine groups, connected by a drive system. The components define the power
required, and the engine groups define the power available. There are one or more drive states, with a
set of gear ratios for each state. The power required equals the sum of component power, transmission
losses, and accessory losses.
An engine group consists of one or more engines of a specific type. An engine group transfers
power by shaft torque, so it is associated with a propulsion group. For each engine type an engine
model is defined. The engine model describes a particular engine, used in one or more engine groups.
The models include turboshaft engines (perhaps convertible, for turbojet operation or reaction drive),
reciprocating engines, compressors, electric motors (perhaps with fuel cells), electric generators, and
generator-motors.
The aircraft can have one or more jet groups, or none. A jet group produces a force on the aircraft.
A jet model describes a particular jet, used in one or more jet groups. The models include turbojet and
turbofan engines (perhaps convertible, for reaction drive), reaction drive, and a simple force. A reaction
drive supplies a blade force that provides the rotor power required.
The aircraft can have one or more charge groups, or none. A charge group generates energy for the
aircraft. A charge model describes a particular charger, used in one or more charge groups. The models
include fuel cells and solar cells.
There are one or more fuel tank systems for the aircraft. Fuel tank systems are associated with
the engine groups, jet groups, and charge groups. Fuel quantity is measured as either weight or energy.
Fuels considered include jet fuel, gasoline, diesel, and hydrogen. Storage systems considered include
batteries, capacitors, and flywheels.

15–1 Referred Performance

Referred performance parameters are used for propulsion system components that operate with air.
The operating condition and atmosphere give the standard conditions (temperature Tstd and pressure
pstd ) for a specified pressure altitude; the sea-level standard conditions (temperature T0 and pressure p0 );
and the operating temperature T and pressure p. Here the temperatures are ◦ R or ◦ K. The performance
characteristics depend on the√
temperature ratio θ = T /T0 and pressure ratio δ = p/p0 . The flight Mach
number M = V /cs = V /cs0 θ is obtained from the aircraft speed V .
152 Propulsion System

power flow fuel flow

each fuel tank can serve

one or more components

engine group fuel tank charge group fuel tank

turboshaft energy
jet fuel solar cell
engine
propulsion reciprocating
group gasoline fuel cell fuel
engine
rotors energy
drive train compressor

compressor
rotor react drive

motor + fuel
drive fuel cell
train
electric energy
motor
jet group fuel tank
generator- energy turbojet
motor jet fuel
turbofan
drive generator energy simple weight or
train
force energy

rotor reaction
fuel
drive

Figure 15-1. Propulsion system components.

The referred or corrected performance parameters are:

P ẇ
power √ fuel flow √
δ θ δ θ
ṁ F
mass flow √ force
δ/ θ δ
P/ṁ N
specific power rotational speed √
θ θ

where P is power, ṁ is mass flow, ẇ is fuel flow, F is a force, and N is a rotational speed. The
performance at sea-level-standard static conditions is indicated by subscript 0.
Propulsion System 153
15–2 Engine Ratings

Engine performance depends on the engine rating. Each engine rating has specific operating
limitations, most importantly an operating time limit intended to avoid damage to the engine.
The power available from a turboshaft engine depends on the engine rating. Typical engine ratings
are given in table 15-1. Engine power is generally specified in terms of sea level standard (SLS) static
MCP. Takeoff typically uses MRP. CRP or ERP is restricted to use in one-engine inoperative (OEI)
emergencies.
Table 15-1. Typical turboshaft engine ratings.
rating description time limit
MCP maximum continuous power ∞
IRP intermediate rated power 30 min
MRP maximum rated power 10 min
CRP contingency rated power 2.5 min
ERP emergency rated power 1.0 min

The thrust available from a turbojet or turbofan engine depends on the engine rating. Typical engine
ratings are given in table 15-2.

Table 15-2. Typical jet engine ratings.

rating description time limit
MCT maximum continuous thrust ∞
MTO maximum takeoff thrust 5 min

15–3 Efficiency from Equivalent Circuit

The efficiency of an electrical device can be expressed in terms of its power by considering an
equivalent circuit, defined by internal resistance R and current I0 . The total voltage is Vx = V + IR and
the total current is Ix = I + I0 . Then the useful power is

P = V I = (Vx − IR)I = Vx Ix − I 2 R − Vx I0 = Ptotal − Ploss

and the efficiency is

P P P
η= = =
Ptotal P + Ploss P + P 2 R/V 2 + P0
In terms of a reference power, let  
1 1
R/V 2 = −1
Pref ηref
P0 = cPref
Then  
1 P 1 Pref
= 1 + P (R/V 2 ) + P0 /P = 1 + −1 +c
η Pref ηref P
−1
So η = (1/ηref + c) at P = Pref . The efficiency decreases with P because of the internal resistance,
but is zero at P = 0 because of the internal current term.
154 Propulsion System
15–4 Control and Loads

Geometry and control are defined for engine groups, jet groups, and charge groups. The group
amplitude A and mode B are control variables:
A = A0 + TA cAC
B = B0 + TB cAC

with A0 and B0 zero, constant, or a function of flight speed (piecewise linear input). The amplitude can
be power (engine group), thrust (jet group), or power (charge group). The mode can be mass flow (for
convertible engines), or power flow (for generator-motor).
The group orientation is specified by selecting a nominal direction ef 0 in body axes (positive or
negative x-, y -, or z -axis; usually thrust forward, hence positive x-axis); then applying a yaw angle ψ ;
and then an incidence or tilt angle i (table 15-3). The yaw and incidence angles can be connected to the
aircraft controls cAC :
ψ = ψ0 + Tψ cAC
i = i0 + Ti cAC
with ψ0 and i0 zero, constant, or a function of flight speed (piecewise linear input). Hence the incidence
and yaw angles can be fixed orientation or can be control variables. Optionally the lateral position of
the group can be set equal to that of a designated rotor (useful for tiltrotors when the rotor hub lateral
position is calculated from the clearance or wing geometry).
The group produces a force T , acting in the direction of the group; and a drag D, acting in the wind
direction. The group is at location z F . The force and moment acting on the aircraft in body axes are
thus:
F F = ef T + ed D
,F F F
M F = Δz
where Δz F = z F − zcgF
, ef is the force direction, and ed is the drag direction. The velocity relative to the
air gives ed = −v F /|v F | (no interference). The group axes are C BF = Ui Vψ , where U and V depend on
the nominal direction, as described in table 15-3. The force direction is ef = C F B ef 0 .
For a tiltrotor aircraft, one of the aircraft controls is the nacelle angle, with the convention αtilt = 0
for cruise and αtilt = 90 deg for helicopter mode. The incidence angle is then connected to αtilt by
defining the matrix Ti appropriately. If the nominal direction is defined for airplane mode (+x), then
i = αtilt should be used; if the nominal direction is defined for helicopter mode (−z ), then i = αtilt − 90
should be used.

Table 15-3. Group orientation.

nominal (F axes) ef 0 incidence, + for force yaw, + for force C BF = Ui Vψ
x forward i up right Yi Zψ
−x aft −i up right Y−i Z−ψ
y right j aft up Zi X−ψ
−y left −j aft up Z−i Xψ
z down k aft right Y−i X−ψ
−z up −k aft right Yi Xψ
Propulsion System 155
15–5 Nacelle Drag

An aerodynamic model is defined for engine groups, jet groups, and charge groups. The group
includes a nacelle, which contributes to the aircraft drag. The component drag contributions must be
consistent. The pylon is the rotor support and the nacelle is the engine support. The drag model for a
tiltrotor aircraft with tilting engines would use the pylon drag (and no nacelle drag), since the pylon is
connected to the rotor shaft axes; with non-tilting engines it would use the nacelle drag as well.
The nacelle drag acts at the group location z F . The nacelle axes are the group axes, hence C BF
is calculated as described previously (see table 15-1). For the nominal direction forward (+x-axis), the
nacelle z -axis is downward and the x-axis is forward; zero incidence angle corresponds to zero angle of
attack; and 90 deg incidence angle corresponds to 90 deg angle of attack (vertical drag). The velocity,
angle of attack, and dynamic pressure are calculated at the nacelle (without interference). The reference
area for the nacelle drag coefficient is the nacelle wetted area. The wetted area is input, or calculated
from the weight w:
 2/3
Swet = k w/N

where N is the number of engines, jets, or chargers; and the units of k are ft2 /lb2/3 or m2 /kg2/3 . The
reference area is then Snac = N Swet . The nacelle area is included in the aircraft wetted area if the drag
coefficient is nonzero. The drag area or drag coefficient is defined for forward flight and for vertical
flight. The drag coefficient at angle of attack α is

CD = CD0 + (CDV − CD0 )| sin α|Xd

typically using Xd = 2. In sideward flight, CD = CD0 is used. The nacelle drag is Dnac = qSnac CD .
156 Propulsion System
 Chapter 16
Fuel Tank

The fuel quantity stored and burned can be measured in weight or energy. Each component (engine
group, jet group, charge group, or equipment) that uses or generates fuel is associated with a fuel tank
system of the appropriate type. The unit of fuel energy is Mega-Joules (MJ). For reference, 1 British
Thermal Unit (BTU) = 1055.056 Joule and 1 kW-hr = 3.6 MJ.

16–1 Fuel Tank System Store and Burn Weight

For fuel use measured by weight, the fuel properties are density ρfuel (weight per volume, lb/gal or
kg/liter) and specific energy efuel (MJ/kg). Table 16-1 gives the properties of a number of aviation fuels,
based on military and industry specifications (refs. 1–3). Fuels considered include jet fuel, gasoline,
diesel, and hydrogen. From the fuel weight Wfuel , the energy is Efuel = efuel Wfuel (MJ) and the volume
is Vfuel = Wfuel /ρfuel (gallons or liters). A motive device has a fuel flow ẇ (lb/hour or kg/hour), and its
specific fuel consumption is sfc = ẇ/P or sfc = ẇ/T .

16-1.1 Fuel Capacity

The fuel tank capacity Wfuel−cap (maximum usable fuel weight) is determined from designated
sizing missions. The maximum mission fuel required, Wfuel−miss (excluding reserves and any fuel in
auxiliary tanks), gives
Wfuel−cap = max(ffuel−cap Wfuel−miss , Wfuel−miss + Wreserve )

or
Wfuel−cap = dfuel−cap + ffuel−cap (Wfuel−miss + Wreserve )
where ffuel−cap ≥ 1 is an input factor. Alternatively, the fuel tank capacity Wfuel−cap can be input. The
corresponding volumetric fuel tank capacity is Vfuel−cap = Wfuel−cap /ρfuel .
For missions that are not used to size the fuel tank, the fuel weight may be fallout, or the fuel weight
may be specified (with or without auxiliary tanks). The fuel weight for a flight condition or the start of
a mission can be specified as an increment d, plus a fraction f of the fuel tank capacity, plus auxiliary
tanks: 
Wfuel = min(dfuel + ffuel Wfuel−cap , Wfuel−cap ) + Nauxtank Waux−cap
where Waux−cap is the capacity of each auxiliary fuel tank. The fuel capacity of the wing can be estimated
from 
Wfuel−wing = ρfuel f ctb tw bw
where ctb is the torque box chord, tw the wing thickness, and bw the wing span; and f is the input fraction
of the wing torque box that is filled by primary fuel tanks, for each wing. This calculation is performed
in order to judge whether fuel tanks outside the wing are needed.
158 Fuel Tank

if Wfuel > Wfuel−max

for designated auxiliary tank
Nauxtank = Nauxtank + n
if fixed total weight: ΔWfuel = −nfauxtank Waux−cap
ΔWfuel−max = nWaux−cap
repeat if Wfuel > Wfuel−max
if fixed total weight and Wfuel ≤ Wfuel−max − nWaux−cap
Nauxtank = Nauxtank − n
ΔWfuel−max = −nWaux−cap
Wfuel = Wfuel−max (capped)
else if Wfuel < Wfuel−max
for designated auxiliary tank (then for last nonzero Nauxtank )
Nauxtank = Nauxtank − n
if fixed total weight: ΔWfuel = nfauxtank Waux−cap
ΔWfuel−max = −nWaux−cap
repeat if Wfuel < Wfuel−max
undo last increment
Nauxtank = Nauxtank + n
if fixed total weight: ΔWfuel = −nfauxtank Waux−cap
ΔWfuel−max = nWaux−cap

Figure 16-1. Outline of Nauxtank calculation.

16-1.2 Fuel Reserves

Mission fuel reserves can be specified in several ways for each mission. Fuel reserves can be
defined in terms of specific mission segments, for example 200 miles plus 20 minutes at Vbe . Fuel
reserves can be an input fraction of the fuel burned by all (except reserve) mission segments, so Wfuel =
(1 + fres )Wfuel−miss . Fuel reserves can be an input fraction of the fuel capacity, so Wfuel = Wmiss−seg +
fres Wfuel−cap . If more than one criterion for reserve fuel is specified, the maximum reserve is used.

16-1.3 Auxiliary Fuel Tank

Auxiliary fuel tanks are defined in one or more sizes. The capacity of each auxiliary fuel tank,
Waux−cap , is an input parameter. The number of auxiliary fuel tanks on the aircraft, Nauxtank for each
size, can be specified for the flight condition or mission segment. Alternatively (if the mission is not
used to size the fuel tank), the number of auxiliary fuel tanks at the start of the mission can be determined
from the mission fuel.

Figure 16-1 describes the process for determining Nauxtank from the required fuel weight Wfuel

and the aircraft maximum fuel capacity Wfuel−max = Wfuel−cap + Nauxtank Waux−cap . The fuel weight
adjustment ΔWfuel is made if fuel weight is fallout from fixed gross weight and payload, accounting
for the operating weight update when Nauxtank changes. If the auxiliary tank weight is greater than
the increment in fuel weight needed, then the fallout fuel weight Wfuel = WG − WO − Wpay can not be
achieved; in such a case, the fuel weight is capped at the maximum fuel capacity and the payload weight
adjusted instead. The tanks changed can be the first size, the first size already used, or a designated size.
The tanks can be added or dropped in groups of n (n = 2 for pairs).
Fuel Tank 159

The weight and drag of Nauxtank tanks are included in the performance calculation. Optionally the
number of auxiliary tanks required can be calculated at the beginning of designated mission segments
(based on the aircraft fuel weight at that point), and tanks dropped if no longer needed. The weight of the

auxiliary fuel tanks is an input fraction of the tank capacity: Wauxtank = fauxtank Nauxtank Waux−cap .

16-1.4 Auxiliary Fuel Tank Drag

The auxiliary fuel tanks are located at position z F . The drag area for one auxiliary tank is specified,
(D/q)auxtank . The velocity relative to the air gives the drag direction ed = −v F /|v F | and dynamic
pressure q = 1/2ρ|v F |2 (no interference). Then
F F = ed q Nauxtank (D/q)auxtank
is the total drag force, calculated for each auxiliary tank size.

16-1.5 Weights
The fuel system consists of the tanks (including support) and the plumbing. For the fractional
model, the fuel tank weight is Wtank = χtank ftank Wfuel−cap .
The weight of the auxiliary fuel tanks is part of the fixed useful load; it is an input fraction of the

tank capacity: Wauxtank = fauxtank Nauxtank Waux−cap .
The AFDD weight model for the plumbing requires the fuel flow rate (for all engines), calculated
for the takeoff rating and conditions.

16–2 Fuel Tank System Store and Burn Energy

For fuel use and storage measured by energy, there is no weight change as energy is used. The
energy storage (tank) is characterized by specific energy etank (MJ/kg) and energy density ρtank (MJ/liter).
Table 16-2 gives the properties of a number of systems. The tank weight and volume are obtained from
the fuel energy Efuel−cap (MJ). The fuel weight Wfuel is zero. A motive device has an energy flow Ė
(MJ/hr), and its specific fuel consumption is sfc = Ė/P (inverse of efficiency). An equivalent fuel flow
is ẇeq = Ė/eref , based on the specific energy eref (MJ/kg) of the first fuel tank that burns weight (or
GP-4, eref = 42.8 MJ/kg). The corresponding equivalent specific fuel consumption is sfceq = ẇeq /P .
The specific power is πtank (kW/kg).
Storage systems considered include batteries, capacitors, and flywheels. A battery (or capacitor)
stores charge (A-hr), so the capacity is expressed as energy for a nominal voltage. Variation of the voltage
with operation affects the efficiency of the relation between useful power and the rate of change of the
energy stored. Each fuel tank system that stores and burns energy has a battery model for computation
of the charge/discharge efficiency. The components associated with a fuel tank system define the total
energy flow Ėcomp (charge or discharge). Accounting for battery capacity, efficiency, and losses gives
the effective energy flow Ėeff .

16-2.1 Fuel Capacity

The fuel tank capacity Efuel−cap (maximum usable fuel energy) is determined from designated
sizing missions. The maximum mission fuel required, Efuel−miss (excluding reserves and any fuel in
auxiliary tanks), gives
Efuel−cap = max(ffuel−cap Efuel−miss , Efuel−miss + Ereserve )
160 Fuel Tank

if Efuel > Efuel−max

for designated auxiliary tank
Nauxtank = Nauxtank + n
ΔEfuel−max = nEaux−cap
repeat if Efuel > Efuel−max
else if Efuel < Efuel−max
for designated auxiliary tank
(then for last nonzero Nauxtank )
Nauxtank = Nauxtank − n
ΔEfuel−max = −nEaux−cap
repeat if Efuel < Efuel−max
undo last increment
Nauxtank = Nauxtank + n
ΔEfuel−max = nEaux−cap

Figure 16-2. Outline of Nauxtank calculation.

or
Efuel−cap = dfuel−cap + ffuel−cap (Efuel−miss + Ereserve )

where ffuel−cap ≥ 1 is an input factor. Alternatively, the fuel tank capacity Efuel−cap can be in-
put. The corresponding fuel tank weight is Wtank = χtank Efuel−cap /etank (lb or kg) and the fuel
tank volume is Vtank = Efuel−cap /ρtank (gallons or liters). The corresponding power capacity is
Pcap = (πtank /etank )Efuel−cap (kW and MJ). Optionally the maximum mission battery discharge power
gives Pcap , from which Efuel−cap = max(Efuel−cap , (etank /πtank )Pcap ) (MJ from kW).
For missions that are not used to size the fuel tank, the fuel energy may be fallout, or the fuel energy
may be specified (with or without auxiliary tanks). The fuel energy for a flight condition or the start of
a mission can be specified as an increment d, plus a fraction f of the fuel tank capacity, plus auxiliary
tanks: 
Efuel = min(dfuel + ffuel Efuel−cap , Efuel−cap ) + Nauxtank Eaux−cap

where Eaux−cap is the capacity of each auxiliary fuel tank.

16-2.2 Fuel Reserves

Mission fuel reserves can be specified in several ways for each mission. Fuel reserves can be defined
in terms of specific mission segments, for example 200 miles plus 20 minutes at Vbe . Fuel reserves can be
an input fraction of the fuel burned by all (except reserve) mission segments, so Efuel = (1+fres )Efuel−miss .
Fuel reserves can be an input fraction of the fuel capacity, so Efuel = Emiss−seg + fres Efuel−cap . If more
than one criterion for reserve fuel is specified, the maximum reserve is used.

16-2.3 Auxiliary Fuel Tank

Auxiliary fuel tanks are defined in one or more sizes. The capacity of each auxiliary fuel tank,
Eaux−cap , is an input parameter. The number of auxiliary fuel tanks on the aircraft, Nauxtank for each
size, can be specified for the flight condition or mission segment. Alternatively (if the mission is not
used to size the fuel tank), the number of auxiliary fuel tanks at the start of the mission can be determined
from the mission fuel.
Fuel Tank 161

Figure 16-2 describes the process for determining Nauxtank from the required fuel energy Efuel and

the aircraft maximum fuel capacity Efuel−max = Efuel−cap + Nauxtank Eaux−cap . The tanks changed
can be the first size, the first size already used, or a designated size. The tanks can be added or dropped
in groups of n (n = 2 for pairs).
The drag of Nauxtank tanks is included in the performance calculation. Optionally the number of
auxiliary tanks required can be calculated at the beginning of designated mission segments (based on the
aircraft fuel energy at that point), and tanks dropped if no longer needed. The weight of the auxiliary fuel

tanks is obtained from eauxtank (MJ/kg) and the tank capacity: Wauxtank = Nauxtank Eaux−cap /eauxtank .

16-2.4 Weights
The fuel system consists of the tanks (including support) and the plumbing. The plumbing weight
here means the power distribution (wiring). The fuel tank weight is Efuel−cap /etank + WBM S (lb or kg)
and the fuel tank volume is Vfuel−cap = Efuel−cap /ρtank (gallons or liters). The battery management
system (BMS) weight is a fraction of the basic tank weight: WBM S = fBM S (Efuel−cap /etank ). The
wiring weight is a fraction of the basic tank weight: fwire (Efuel−cap /etank ). Alternatively, the wiring
weight can be input as part of the electrical group weight.
The weight of the auxiliary fuel tanks is part of the fixed useful load; it is obtained from eauxtank

(MJ/kg) and the tank capacity: Wauxtank = Nauxtank Eaux−cap /eauxtank .

16–3 Equipment Power

Systems that supply equipment power can directly use fuel. Equipment power (similar to drive
train accessory losses) is calculated as the sum of an input constant; terms that scale with air density and
temperature; a term that scales aircraft weight; deice power loss (if deice system is on); and an increment
specified for each flight state:

Peq = Peq0 + Peqd σ + Peqt θ + Peqw + Peqi + dPeq

where σ = ρ/ρ0 is the density ratio and θ = T /T0 is the temperature ratio. The power scaling with aircraft
Xeq
weight is Peqw = Keq WM T O (where WM T O is the maximum takeoff gross weight). The fuel flow or
energy flow is then obtained from an input specific fuel consumption: ẇeq = (sfc)Peq or Ėeq = (sfc)Peq .
The energy sfc is the inverse of an efficiency: sfc=3.6/η for energy flow in MJ/hr and power in kW.

16–4 References

1) Department of Defense Military Specification. “Glossary of Definitions, Ground Rules, and Mission
Profiles to Define Air Vehicle Performance Capability.” MIL-STD-3013A, September 2008.
2) Department of Defense Detail Specification. “Turbine Fuel, Aviation, Grades JP-4 and JP-5.” MIL-
DTL-5624U, September 1998.
3) Department of Defense Detail Specification. “Turbine Fuel, Aviation, Kerosene Type, JP-8 (NATO
F-34), NATO F-35, and JP-8+100 (NATO F-37).” MIL-DTL-83133H, October 2011.
162 Fuel Tank

Table 16-1. Fuel properties.

fuel specification density specific energy energy dens.
lb/gal kg/L MJ/kg BTU/lb lb/hp-hr MJ/L
gasoline MIL-STD-3013A 6.0* 0.719 43.50 18700* 0.136 31.3
diesel 7.0 0.839 43.03 18500 0.138 36.1
6.84– 0.820– 43.0 18487 0.138 35.8
7.05 0.845
Jet A/A-1 MIL-STD-3013A 6.7* 0.803 42.80 18400* 0.138 34.4
6.84/ 0.820/ 42.8 18401 0.138 34.8
6.71 0.804
JP-4 6.5 0.779 42.80 18400 0.138 33.3
MIL-DTL-5624U 6.23– 0.751*– 42.8* 18401 0.138 32.2
6.69 0.802*
JP-5 MIL-STD-3013A 6.6* 0.791 42.57 18300* 0.139 33.7
alternate design 6.8* 0.815 42.91 18450* 0.138 35.0
MIL-DTL-5624U 6.58– 0.788*– 42.6* 18315 0.139 34.8
7.05 0.845*
JP-8 MIL-STD-3013A 6.5* 0.779 42.80 18400* 0.138 33.3
alternate design 6.8* 0.815 43.19 18570* 0.137 35.2
MIL-DTL-83133H 6.45– 0.775*– 42.8* 18401 0.138 34.6
7.01 0.840*
LPG (methane) 3.75 0.45 50. 21496 0.118 111.
hydrogen (700 bar) 0.328 0.03930 120. 51591 0.0493 4.72
high heating value 0.328 0.03930 140. 60190 0.0423 5.50
hydrogen (liquid) 0.592 0.07099 120. 51591 0.0493 8.52
*specification value

Table 16-2. Energy storage properties.

tank specific energy tank energy density efficiency power
3
MJ/kg kW-hr/kg MJ/L kW-hr/m kW/kg
lead-acid battery 0.11–0.14 0.03–0.04 0.22–0.27 60–75 70–90% 0.18
nickel-cadmium battery 0.14–0.20 0.04–0.06 0.18–0.54 50–150 70–90% 0.15
lithium-ion state-of-art 0.54–0.90 0.15–0.25 0.90–1.30 250–360 ∼99% 1.80
+5 years 1.26 0.35 1.80 500
+15 years 2.34 0.65 2.25 625
ultracapacitor 0.01–0.11 0.004-0.03 0.02–0.16 6–45 1.00
flywheel steel 0.11 0.03 ∼90%
graphite 0.90 0.25
 Chapter 17

Propulsion Group

The propulsion group is a set of components and engine groups, connected by a drive system. The
engine model describes a particular engine, used in one or more engine group. The components (rotors)
define the power required. The engine groups define the power available. Figure 17-1 illustrates the
power flow.

17–1 Drive System

The drive system defines gear ratios for all the components it connects. The gear ratio is the ratio
of the component rotational speed to that of the primary rotor. There is one primary rotor per propulsion
group (for which the reference tip speed is specified); other components are dependent (for which a gear
ratio is specified). There can be more than one drive system state, in order to model a multiple-speed or
variable-speed transmission. Each drive system state corresponds to a set of gear ratios.

For the primary rotor, a reference tip speed Vtip−ref is defined for each drive system state. By
convention, the “hover tip speed” refers to the reference tip speed for drive state #1. If the sizing
task changes the hover tip speed, then the ratios of the reference tip speeds at different engine states
are kept constant. By convention, the gear ratio of the primary rotor is r = 1. For dependent rotors,
either the gear ratio is specified (for each drive system state) or a tip speed is specified and the gear
ratio is calculated (r = Ωdep /Ωprim , Ω = Vtip−ref /R). For the engine group, either the gear ratio is
specified (for each drive system state) or the gear ratio is calculated from the specification engine
turbine speed Ωspec = (2π/60)Nspec and the reference tip speed of the primary rotor (r = Ωspec /Ωprim ,
Ωprim = Vtip−ref /R). The latter option means the specification engine turbine speed Nspec corresponds
to Vtip−ref for all drive system states. To determine the gear ratios, the reference tip speed and radius are
used, corresponding to hover.

The flight state specifies the tip speed of the primary rotor and the drive system state, for each
propulsion group. The drive system state defines the gear ratio for dependent rotors and the engine
groups. From the rotor radius, the rotational speed of the primary rotor is obtained (Ωprim = Vtip /R);
from the gear ratios, the rotational speed of dependent rotors (Ωdep = rΩprim ) and the engine groups
(N = (60/2π)reng Ωprim ) are obtained; and from the rotor radius, the tip speed of the dependent rotor
(Vtip = Ωdep R) is obtained. The flight state specification of the tip speed can be an input value, the
reference tip speed, a function of flight speed or a conversion schedule, or one of several default values.
These relationships between tip speed and rotational speed use the actual radius of the rotors in the flight
state, which for a variable-diameter rotor may not be the same as the reference, hover radius.

A designated drive system state can have a variable speed (variable gear ratio) transmission, by
introducing a factor fgear on the gear ratio when the speeds of the dependent rotors and engines are
164 Propulsion Group

POWER POWER
AVAILABLE REQUIRED
ENGINE Pa Pq
UNINSTALLED
installation losses
Ploss
mechanical (N/N spec) Pmech
power limit

ENGINE SHAFT Pav Preq

INSTALLED sum
all operable engines dis
trib
ute
flight state fp
power fraction

engine
shaft (Ωprim / Ωref) PES limit
limit

ENGINE GROUP PavEG PreqEG

sum
all engine groups dis
trib
ute
drive
system (Ωprim / Ωref) PDS limit
limit

PROPULSION PavPG PreqPG

GROUP
transmission losses Pxmsn
accessory power Pacc
dis
trib
ute
all components su
rotor m
shaft (Ωprim / Ωref) PRS limit
limit

ROTORS Pav comp Preq comp = Pi + Po + Pp

Max Gross Weight Options Max Effort or Trim Options

PreqPG = fP avPG + d PreqPG = PavPG
Qreq ≤ Q limit Qreq ≤ Q limit

Figure 17-1. Power flow.

Propulsion Group 165

evaluated. The factor fgear is a component control, which can be connected to an aircraft control and
thus set for each flight state.
An optional conversion schedule is defined in terms of two speeds: hover and helicopter mode for
speeds below VChover , cruise mode for speeds above VCcruise , and conversion mode for speeds between
VChover and VCcruise . The tip speed is Vtip−hover in helicopter and conversion mode, and Vtip−cruise in
airplane mode. Drive system states are defined for helicopter, cruise, and conversion mode flight. The
flight state specifies the nacelle tilt angle, tip speeds, control state, and drive system state, including the
option to obtain any or all of these quantities from the conversion schedule.
Several default values of the tip speed are defined for use by the flight state, including cruise,
maneuver, one-engine inoperative, drive system limit conditions, and a function of flight speed (piecewise
linear input). Optionally these default values can be input as a fraction of the hover tipspeed. Optionally
speed can be calculated from μ = V /Vtip , so Vtip = V /μ; or from Mat = Mtip (1 + μ)2 + μ2z , so
the tip 
Vtip = (cs Mat )2 − Vz2 − V . Optionally the tip speed can be the minimum of the input value or that for
Mat .

The sizing task might change the hover tip speed (reference tip speed for drive system state #1),
the reference tip speed of a dependent rotor, a rotor radius, or the specification engine turbine speed
Nspec . In such cases the gear ratios and other parameters are recalculated. Note that it is not consistent
to change the reference tip speed of a dependent rotor if the gear ratio is a fixed input.
An increment on the primary rotor rotational speed (or primary engine group, if there are no rotors)
is a control variable of the propulsion group.

17–2 Power Required

The component power required Pcomp is evaluated for a specified flight condition, as the sum of
the power required by all the components of the propulsion group. The total power required for the
propulsion group is obtained by adding the transmission losses and accessory power:

PreqP G = Pcomp + Pxmsn + Pacc

The transmission losses are calculated as an input fraction xmsn of the component power, plus windage
loss:
Pxmsn = xmsn fxmsn |Pcomp | + Pwindage (Ωprim /Ωref )
The factor fxmsn can equal 1, or can include a function of the drive shaft limit (increasing the losses at
low power): ⎧ 1 1
⎪
⎨ 2 PXlimit  Q< 4
fxmsn |Pcomp | = 3 − 3 Q |Pcomp |
7 4 1
<Q<1
⎪
⎩
4
|Pcomp | 1<Q
where Q = |Pcomp |/PXlimit , PXlimit = rPDSlimit , and r = N/Nspec = Ωprim /Ωref . Accessory losses are
calculated as the sum of an input constant; terms that scale with air density and rotor speed; a fraction
of power required (such as environmental control unit (ECU) losses); infared suppressor fan loss (if IRS
system is on); deice power loss (if deice system is on); and an increment specified for each flight state:

Pacc = Pacc0 + Paccd σ + Paccn σ(Ωprim /Ωref ) + acc (|Pcomp | + Pxmsn ) + IRfan σNeng Peng + Pacci + dPacc

where σ = ρ/ρ0 is the density ratio.

166 Propulsion Group

The power required for the propulsion group must be distributed to the engine groups. With only one
engine group, PreqEG = PreqP G . An engine group power can be fixed at PreqEG = (Neng −Ninop )A, where
A is the input power amplitude; or fraction A of engine power available, PreqEG = (Neng − Ninop )APav ;
or fraction A of engine rated power PreqEG = (Neng −Ninop )APeng . The power required for the remaining
(perhaps all) engine groups is distributed proportional to the engine rated power:
   (Neng − Ninop )Peng
PreqEG = PreqP G − PreqEG 
fixed notfixed (Neng − Ninop )Peng

omitting engine groups that do not supply shaft power. If the sum of the fixed PreqEG exceeds the
propulsion group power required, or if the power is fixed for all engine groups, then each is scaled by

the ratio PreqP G / fixed PreqEG . The fuel flow of the propulsion group is obtained from the sum over

the engine groups: ẇreqP G = ẇreqEG .

17–3 Geometry

The length of the drive system DS can be input or calculated. The calculated length is the sum
of the longitudinal, lateral, and vertical distances from the primary rotor hub to the other hub locations,

for all rotors in the propulsion group: DS = f (|Δx| + |Δy| + |Δz|), where f is an input factor.
Alternatively, the drive system length can be scaled with the radius of the primary rotor: DS = f R.

17–4 Drive System limit

The drive system limit is defined as a power limit, PDSlimit . The limit is properly a torque limit,
QDSlimit = PDSlimit /Ωref , but is expressed as a power limit for clarity. The drive system limit can be
specified as follows (with flimit an input factor):
a) Input PDSlimit .

b) From the engine takeoff power limit, PDSlimit = flimit Neng Peng (summed over
all engine groups).
c) From the power available at the transmission sizing conditions and missions,

PDSlimit = flimit (Ωref /Ωprim ) Neng Pav (largest of all conditions and segments).
d) From the power required at the transmission sizing conditions and missions,

PDSlimit = flimit (Ωref /Ωprim ) Neng Preq (largest of all conditions and segments).

The drive system limit is a limit on the entire propulsion system. To account for differences in the
distribution of power through the drive system, limits are also used for the torque of each rotor shaft
(PRSlimit ) and of each engine group (PESlimit ). The engine shaft limit is calculated as for the drive system
limit, without the sum over engine groups. The rotor shaft limit is either input or calculated from the
rotor power required at the transmission sizing flight conditions. The power limit is associated with a
reference rotational speed, and when applied the limit is scaled with the rotational speed of the flight
state. The rotation speed for the drive system limit PDSlimit is the hover speed of the primary rotor of
the propulsion group (for the first drive state). The rotation speed for the engine shaft limit PESlimit
is the corresponding engine turbine speed. The rotation speed for the rotor shaft limit PRSlimit is the
corresponding speed of that rotor.
The drive system limits can be specified for several levels, analogous to engine ratings. Typically
higher torque limits are associated with short duration operation. For each drive system rating, a torque
factor x is specified. Then from the rating for a flight condition or mission segment, the torque limit is
Propulsion Group 167

xQlimit , with Qlimit obtained from PDSlimit . The torque factor x is used for engine shaft limits and rotor
shaft limits as well. Optionally the drive system rating can be scheduled with flight speed.

17–5 Weights

The drive system consists of gear boxes and rotor shafts, drive shafts, rotor brakes, clutches, and
gas drive. The drive system weight depends on the rotor and engine rotational speeds, evaluated for
the propulsion group primary rotor and a specified engine group, at a specified drive system state (gear
ratio).
The AFDD drive system weight model depends on fQ , the second (main or tail) rotor rated torque
as a fraction of the total drive system rated torque; and on fP , the second (main or tail) rotor rated power
as a fraction of the total drive system rated power. These parameters are related by the rotational speeds
of the two rotors: fP = fQ Ωother /Ωmain . Typically fP = fQ = 0.6 for twin rotors (tandem, coaxial, and
tiltrotor configurations). For the single-main-rotor and tail-rotor configuration, typically fQ = 0.03 and
fP = 0.15 (0.18 with a two-bladed teetering main-rotor).
168 Propulsion Group
 Chapter 18

Engine Group

The engine group consists of one or more engines of a specific type. An engine group transfers
power by shaft torque, so it is associated with a propulsion group. For each engine type an engine model
is defined. The engine model describes a particular engine, used in one or more engine groups.

The models include turboshaft engines (perhaps convertible, for turbojet operation or reaction
drive), reciprocating engines, compressors, electric motors (perhaps with fuel cells), electric generators,
and generator-motors.

18–1 Engine Group Performance

The engine size is described by the power Peng , which is the sea-level static power available per
engine at a specified takeoff rating. The number of engines Neng is specified for each engine group.

If the sizing task determines the engine power for a propulsion group, the power Peng of at least one

engine group is found (including the first engine group). The total power required is PP G = r Neng Peng ,

where r = max(PreqP G /PavP G ). The sized power is Psized = PP G − fixed Neng Peng . Then the sized

engine power is Peng = fn Psized /Neng for the n-th engine group (with f1 = n=1,sized fn for the first
group). If an engine group does not consume power (compressor or generator) or does not contribute to
shaft power (converted), the size is scaled with r = max(PreqEG /PavEG ).

The propulsion group power available is obtained from the sum over the engine groups: PavP G =

PavEG .

The propulsion group component power Pcomp includes compressor power, generator power re-
quired, and generator-motor power when it is producing energy.

The flight condition information includes the altitude, temperature, flight speed, and primary rotor
speed; a power fraction fP ; and the states of the engine, drive system, and infrared suppressor (IRS). The
engine turbine speed is N = (60/2π)reng Ωprim , where Ωprim is the current rotor speed and reng is the gear
ratio (depending on the drive system state, including a factor fgear for a variable speed transmisson).
If the reference primary rotor speed Ωprim corresponds to the specification turbine speed Nspec , then
reng = Ωspec /Ωprim ; alternatively, the engine gear ratio can be a fixed input.

The drive system limit at the flight condition is rxPDSlimit , where r = Ωprim /Ωref and x is the rating
factor. Optionally this limit is applied to the propulsion group power: PavP G = min(PavP G , rxPDSlimit ).
Similarly the engine shaft limit at the flight condition is optionally applied to the engine group power:
PavEG = min(PavEG , rxPESlimit ).
170 Engine Group
18–2 Turboshaft Engine

Turboshaft engine performance is obtained from the Referred Parameter Turboshaft Engine Model
(RPTEM).

18-2.1 Power Available

Given the flight condition and engine rating, the power available Pa is calculated from the specific
power SPa = Pa /ṁa and mass flow ṁa :
SPa
= SP0 gsp (θ, M, n)
θ
ṁa
√ = ṁ0 gm (θ, M, n)
δ/ θ
Pa
√ = P0 gp (θ, M, n)
δ θ
as functions
√
of temperature ratio θ = T /T0 , Mach number M , and referred engine turbine speed n =
N/ θ.

In the engine model, installation losses Ploss are subtracted from Pa (Pav = Pa − Ploss ), and then the
mechanical limit is applied: Pav = min(Pav , rPmechR ), r = N/Nspec . The mechanical limit is properly a
torque limit, Qmech = Pmech /Nspec , but is expressed as a power limit for clarity.
The engine model gives the performance of a single engine. The power available of the engine
group is obtained by multiplying the single engine power by the number of engines operational (total
number of engines less inoperable engines):

PavEG = fP (Neng − Ninop )Pav

including a specified power fraction fP .

18-2.2 Performance at Power Required

The engine performance (mass flow, fuel flow, and gross jet thrust) is calculated for a specified
power required Pq and flight condition:
ṁreq
√ = ṁ0C gm (q, θ, M, n)
δ/ θ
ẇreq
√ = ẇ0C gw (q, θ, M, n)
δ θ
Fg
= Fg0C gf (q, θ, M, n)
δ
√
as functions of q = Pq /(P0C√δ θ), temperature ratio θ = T /T0 , Mach number M , and referred engine
turbine speed n = N/(Nspec θ).
The engine model deals with a single engine. The power required of a single engine is obtained
by dividing the engine group power by the number of engines operational (total number of engines less
inoperable engines):
Preq = PreqEG /(Neng − Ninop )

In the engine model, installation losses Ploss are added to Preq : Pq = Preq + Ploss .
Engine Group 171

The engine model gives the performance of a single engine. The performance of the engine group
is obtained by multiplying the single engine characteristics by the number of engines operational (total
number of engines less inoperable engines):

ṁreqEG = (Neng − Ninop )ṁreq

ẇreqEG = (Neng − Ninop )ẇreq Kf f d
FN EG = (Neng − Ninop )FN
DauxEG = (Neng − Ninop )Daux

The fuel flow has also been multiplied by a factor Kf f d accounting for deterioration of the engine
efficiency.

18-2.3 Installation

The difference between installed and uninstalled power is the inlet and exhaust losses Ploss : Pav =
Pa − Ploss and Preq = Pq − Ploss . The inlet ram recovery efficiency ηd is included in the engine model
calculations. The inlet and exhaust losses are modeled as fractions of power available or power required:
Ploss = ( in + ex )Pa or Ploss = ( in + ex )Pq . The installed gross jet thrust is FG = Kf gr Fg , where Kf gr
accounts for exhaust effects. The net jet thrust is FN = FG − ṁreq V . The momentum drag of the
auxiliary air flow is a function of the mass flow ṁaux = faux ṁreq :

Daux = (1 − ηaux )ṁaux V = (1 − ηaux )faux ṁreq V

where ηaux is the ram recovery efficiency. Exhaust losses ( ex ) and auxiliary air flow parameters (ηaux ,
faux ) are defined for IR suppressor on and off. Inlet particle separator loss is added to the inlet losses
( in ).

18-2.4 Convertible Engine: Turbojet/Turbofan

The engine mode B is the mass flow fraction diverted for a convertible engine: B = 0 for all mass
flow to the power turbine (turboshaft operation), and B = 1 for all mass flow to the jet exhaust or a fan
(turbojet/turbofan operation).

A separate engine model defines the performance for turbojet/turbofan operation (B = 1). The
engine group power PreqEG is prescribed, as a measure of the jet thrust, and this engine does not
contribute to the propulsion group shaft power available. The turbojet/turbofan thrust is the engine
group net jet thrust, FN = FG − ṁreq V .

18-2.5 Convertible Engine: Reaction Drive

The engine mode B is the mass flow fraction diverted for a convertible engine: B = 0 for all mass
flow to the power turbine (turboshaft operation), and B = 1 for all mass flow to the rotor (reaction jet
operation).

A separate engine model defines the engine performance for reaction jet operation (B = 1). The
engine group power is fixed by amplitude input, or obtained from the rotor power (PreqEG = Preact ),
and this engine does not contribute to the propulsion group shaft power available. The gross jet thrust
is zero, so the net thrust is the momentum drag, FN = −ṁreq V .
172 Engine Group
18–3 Turboshaft Engine Tabular Model

Tabular models of the engine performance are implemented, suitable for use with data from an
exercise of an engine deck. Two table formats are defined (E and H). Input factors Kp , Kw and Kf can
account for technology level, respectively for power available, fuel flow, and jet thrust. The tables are
used with linear interpolation.
The tables are usually for an installed engine, including losses. Hence Ploss = 0 ( in = ex = 0)
with this model, and Pav = Pa , Preq = Pq . Mechanical limits are included in the power available data.
Engine mass flow is not considered, so Daux = 0, and the table is for net jet thrust. The fuel flow is
multiplied by the factor Kf f d , accounting for deterioration of the engine efficiency. The engine is not
scaled. The engine weight Wone eng is fixed. The turbine speed Nspec is fixed.

18-3.1 Table Format E

The tables are for power available Pa , fuel flow ẇ, and net jet thrust FN as a function of altitude h,
flight speed V , and rating R. The altitude influence is for a selected atmosphere, hence for a temperature
variation with altitude. The ratings reflect engine throttle setting, including partial power. Power turbine
speed variation is not considered.
Given the altitude, speed, and rating, the power table Tp (h, V, R) is interpolated. Then the power
available is Pa = Kp Tp . For fixed altitude and speed, the table Tp (h, V, R) implies a variation of power
with rating; and the tables for fuel flow and jet thrust can be interpreted as Tw (h, V, Pq ) and Tf (h, V, Pq ).
Given altitude, speed, and power required, the tables are interpolated; then the fuel flow is ẇ = Kw Tw
and the net jet thrust is FN = Kf Tf .

18-3.2 Table Format H

The tables are for power available Pa and fuel flow ẇ as a function of altitude h, temperature τ ,
flight speed V , and rating R. The power available is the product of static power P0 and a ram factor
√
fRAM : Pa = Kp P0 (h, τ, R)fRAM (V, τ, h). Optionally, fRAM = (δM θM )Xpa , where δM = (1 + 0.2M 2 )
and δM = (1 + 0.2ηd M 2 )3.5 .
The reference fuel flow ẇref = ẇ/δ Xdf θXrf is a function of reference power required Pqref =
Xdp Xrp
Pq /δ
√
θ : ẇ = Kw δ Xdf θXrf ẇref (Pqref ). Note that referred power and referred fuel flow are based on
δ θ. Alternatively, the fuel flow is a function of power, altitude, and temperature: ẇ(Pq , h, τ ). The net
jet thrust is zero.

18–4 Reciprocating Engine

Reciprocating engine performance is obtained from the Reciprocating Engine Model.

18-4.1 Power Available

Given the flight condition and engine rating, the power available Pa is calculated from:

Pa = P0 gp (σ, θ, M, r)

as a function of density ratio σ = ρ/ρ0 , temperature ratio θ = T /T0 , Mach number M , and engine speed
ratio r = N/Nspec .
Engine Group 173

Installation losses Ploss are subtracted from Pa (Pav = Pa − Ploss ), and then the mechanical limit is
applied: Pav = min(Pav , rPmechR ). The mechanical limit is properly a torque limit, Qmech = Pmech /Nspec ,
but is expressed as a power limit for clarity. The power available of the engine group is obtained by
multiplying the single engine power by the number of engines operational (total number of engines less
inoperable engines): PavEG = fP (Neng − Ninop )Pav , including a specified power fraction fP .

18-4.2 Performance at Power Required

The engine performance for a specified power required Pq is:

ẇreq = ẇ0 gw (q, r)

ṁreq = ṁ0 gm (q, r)
Fg = Fg0 gf (q, r)

where q = Pq /P0 and r = N/Nspec . Installation losses Ploss are added to Preq : Pq = Preq + Ploss ,
Preq = PreqEG /(Neng − Ninop ).

The performance of the engine group is obtained by multiplying the single engine characteristics
by the number of engines operational (total number of engines less inoperable engines):

ṁreqEG = (Neng − Ninop )ṁreq

ẇreqEG = (Neng − Ninop )ẇreq Kf f d
FN EG = (Neng − Ninop )FN
DauxEG = (Neng − Ninop )Daux

The fuel flow has also been multiplied by a factor Kf f d accounting for deterioration of the engine
efficiency.

18-4.3 Installation
The difference between installed and uninstalled power is the installation loss: Pav = Pa − Ploss and
Preq = Pq −Ploss . This loss is modeled by an efficiency factor: Ploss = (1−ηloss )Pa or Ploss = (1−ηloss )Pq ,
where ηloss = in + ex . The installed gross jet thrust is FG = Kf gr Fg , where Kf gr accounts for exhaust
effects. The net jet thrust is FN = FG − ṁreq V . The momentum drag of the auxiliary air flow is a
function of the mass flow ṁaux = faux ṁreq :

Daux = (1 − ηaux )ṁaux V = (1 − ηaux )faux ṁreq V

where ηaux is the ram recovery efficiency.

18–5 Compressor

A compressor converts input shaft power to a jet velocity and thrust. The shaft power contributes
to the propulsion group power required. The compressor does not use fuel.

18-5.1 Power Available

Given the flight condition and engine rating, the power available Pa is calculated from the specific
power SPa = Pa /ṁa and mass flow ṁa :
174 Engine Group

SPa
= SP0 gsp (θ, M, n)
θ
ṁa
√ = ṁ0 gm (θ, M, n)
δ/ θ
Pa
√ = P0 gp (θ, M, n)
δ θ
√
as functions of temperature ratio θ = T /T0 , Mach number M , and referred compressor speed n = N/ θ.
Installation losses Ploss are subtracted from Pa (Pav = Pa − Ploss ), and then the mechanical limit
is applied: Pav = min(Pav , rPmechR ), r = N/Nspec . The mechanical limit is properly a torque limit,
Qmech = Pmech /Nspec , but is expressed as a power limit for clarity.

The compressor model gives the performance of a single compressor. The power available of
the engine group is obtained by multiplying the single compressor power by the number of engines
operational (total number of engines less inoperable engines):

PavEG = fP (Neng − Ninop )Pav

including a specified power fraction fP .

18-5.2 Performance at Power Required

The compressor performance (mass flow and gross jet thrust) is calculated for a specified power
required Pq and flight condition:

ṁreq
√ = ṁ0C gm (q, θ, M, n)
δ/ θ
STreq
√ = ST0C gst (q, θ, M, n)
θ
Fg
= Fg0C gf (q, θ, M, n)
δ
√
as functions of√q = Pq /(P0C δ θ), temperature ratio θ = T /T0 , Mach number M , and referred compressor
speed n = N/ θ). The specific thrust gives the gross thrust, Fg = (ST )ṁ.
The power required of a single compressor is obtained by dividing the engine group power by the
number of engines operational (total number of engines less inoperable engines):

Preq = PreqEG /(Neng − Ninop )

Accounting for installation losses gives the uninstalled power required Pq = Preq + Ploss .
The compressor model gives the performance of a single compressor. The performance of the
engine group is obtained by multiplying the single compressor characteristics by the number of engines
operational:
ṁreqEG = (Neng − Ninop )ṁreq
FN EG = (Neng − Ninop )FN
DauxEG = (Neng − Ninop )Daux
Pcomp = (Neng − Ninop )Pq Kf f d
Engine Group 175

The component power is the product of the uninstalled power required and the number of operational
engines, and a factor Kf f d accounting for deterioration of the engine efficiency.

18-5.3 Installation
The difference between installed and uninstalled power is the inlet and exhaust losses Ploss : Pav =
Pa − Ploss and Preq = Pq − Ploss . The inlet and exhaust losses are modeled as fractions of power available
or power required: Ploss = ( in + ex )Pa or Ploss = ( in + ex )Pq . The installed gross jet thrust is
FG = Kf gr Fg , where Kf gr accounts for exhaust effects. The net jet thrust is FN = FG − ṁreq V . The
momentum drag of the auxiliary air flow is a function of the mass flow ṁaux = faux ṁreq :

Daux = (1 − ηaux )ṁaux V = (1 − ηaux )faux ṁreq V

where ηaux is the ram recovery efficiency.

18-5.4 Compressor for Reaction Drive

If the compressor supplies the jet force for rotor reaction drive, then the engine group power required
is fixed by ampitude input, or obtained from the rotor power (PreqEG = Preact ). The gross jet thrust is
zero, so the net thrust is the momentum drag, FN = −ṁreq V .

18–6 Electric Motor or Generator

A motor converts electrical energy (fuel) to shaft power. A generator converts input shaft power to
electrical energy, and the shaft power contributes to the propulsion group power required.
The power available is related to the size Peng . Given the flight condition and motor rating, the
uninstalled power available Pav is calculated, including a torque limit. The motor model gives the
performance of a single engine. The power available of the engine group is obtained by multiplying the
single engine power by the number of engines operational:

PavEG = fP (Neng − Ninop )Pav

including a specified power fraction fP .

From the engine group power required PreqEG , the power required of a single engine is

Preq = PreqEG /(Neng − Ninop )

18-6.1 Motor
The motor power required determines the energy flow from the fuel tank. The energy flow is
calculated for Preq and a specified flight condition:

Ėreq = Ė0C ge (q, n)

as a function of q = Preq /Peng and engine speed n = N/Nspec . The motor model gives the performance
of a single engine. The performance of the engine group is obtained by multiplying the single engine
characteristics by the number of engines operational:

ĖreqEG = (Neng − Ninop )Ėreq Kf f d

176 Engine Group

The energy flow has also been multiplied by a factor Kf f d accounting for deterioration of the engine
efficiency.

18-6.2 Generator
The generator energy flow to the fuel tank defines the power required. The energy flow is calculated
for Preq and a specified flight condition:

Ėreq = Ė0C ge (q, n)

as a function of q = Preq /Peng and engine speed n = N/Nspec . The motor model gives the performance
of a single engine. The performance of the engine group is obtained by multiplying the single engine
characteristics by the number of engines operational:

ĖreqEG = (Neng − Ninop )Ėreq

Pcomp = (Neng − Ninop )Preq Kf f d

The component power is the product of the power required and the number of operational engines, and
a factor Kf f d accounting for deterioration of the engine efficiency.

18-6.3 Generator-Motor
The engine mode B is the direction of power flow for a generator-motor: B positive for motor
operation, and B negative for generator operation. Separate motor models are used for the two modes.

18-6.4 Motor and Fuel Cell

A motor with a fuel cell burns a fuel (typically hydrogen) to produce electrical energy, which is
converted to shaft power. The device can also be modelled as separate motor and fuel cell components,
with a battery (fuel tank) to transfer the electrical energy.
The engine performance (mass flow and fuel flow) is calculated for Preq and a specified flight
condition:
ṁreq = ṁ0C gm (q, n) = Kmf ẇreq
ẇreq = ẇ0C gw (q, n) = sfc Preq
as a function of q = Preq /Peng and engine speed n = N/Nspec . Inlet and exhaust losses are included in
the specific fuel consumption. The net thrust is the inlet momentum drag, FN = −ṁreq V . The motor
model gives the performance of a single engine. The performance of the engine group is obtained by
multiplying the single engine characteristics by the number of engines operational:

ṁreqEG = (Neng − Ninop )ṁreq

ẇreqEG = (Neng − Ninop )ẇreq Kf f d
FN EG = (Neng − Ninop )FN
DauxEG = (Neng − Ninop )Daux

The fuel flow has also been multiplied by a factor Kf f d accounting for deterioration of the engine
efficiency. The momentum drag of the auxiliary air flow is a function of the mass flow ṁaux = faux ṁreq :

Daux = (1 − ηaux )ṁaux V = (1 − ηaux )faux ṁreq V

Engine Group 177

where ηaux is the ram recovery efficiency.

18–7 Control and Loads

The engine power amplitude A and mode B are control variables. The mode can be mass flow (for
convertible engines), or power flow (for generator-motor). In distributing the propulsion group power
required to the engine groups, an engine group power can be fixed at PreqEG = (Neng − Ninop )A; or
fraction A of engine power available, PreqEG = (Neng − Ninop )APav ; or fraction A of engine rated power
PreqEG = (Neng − Ninop )APeng .

The engine orientation is specified by selecting a nominal direction ef 0 in body axes (positive or
negative x-, y -, or z -axis; usually thrust forward, hence positive x-axis); then applying a yaw angle ψ ;
and then an incidence or tilt angle i. The yaw and incidence angles can be control variables.
The engine group produces a jet thrust FN , acting in the direction of the engine (actually, FN =
FG − ṁV , with the gross thrust in the direction of the engine and the momentum thrust in the wind
direction); a momentum drag Daux (if the component has mass flow), acting in the wind direction; and
a nacelle drag Dnac , acting in the wind direction. The engine group is at location z F . The force and
moment acting on the aircraft in body axes are thus:
F F = (ef FG + ed ṁV ) + ed Daux + ed Dnac
,F F F
M F = Δz
where Δz F = z F − zcg
F
, ef is the engine thrust direction and ed is the drag direction.

18–8 Nacelle Drag

The engine group includes a nacelle, which contributes to the aircraft drag. The component drag
contributions must be consistent. The pylon is the rotor support and the nacelle is the engine support.
The drag model for a tiltrotor aircraft with tilting engines would use the pylon drag (and no nacelle drag),
since the pylon is connected to the rotor shaft axes; with non-tilting engines it would use the nacelle
drag as well.
The reference area for the nacelle drag coefficient is the nacelle wetted area. The wetted area per
engine is input, or calculated either from the engine system (engine, exhaust, and accessories) weight
or from the engine system plus drive system weight:
 2/3
Swet = k w/Neng
where w = WES or w = WES + Wgbrs /NEG , and the units of k are ft2 /lb2/3 or m2 /kg2/3 . The reference
area is then Snac = Neng Swet .

18–9 Weights

The component weight consists of engine system, engine section or nacelle group, and air induction
group. The engine system consists of engine, exhaust system, and accessories. The engine section or
nacelle group consists of engine support, engine cowling, and pylon support. These weights are for the
engine group, consisting of Neng engines. The engine system weight WES = Weng + Wexh + Wacc is used
for the engine nacelle wetted area, the rotor pylon wetted area, and the rotor moving weight. The rotor
group includes the rotor support structural weight; this must be consistent with the use of the engine
support weight and pylon support weight.
178 Engine Group
 Chapter 19

Jet Group

A jet group produces a force on the aircraft, possibly used for lift, propulsion, or control. A jet
model describes a particular jet, used in one or more jet groups. The models include turbojet and turbofan
engines (perhaps convertible, for reaction drive), reaction drive, and a simple force. A reaction drive
supplies a blade force that provides the rotor power required.

19–1 Jet Group Performance

The jet size is described by the thrust Tjet , which is the sea-level static thrust available per jet at a
specified takeoff rating. The number of jets Njet is specified for each jet group.

If the sizing task determines the jet thrust, the total thrust required is TJG = rNjet Tjet , where
r = max(TreqJG /TavJG ).

The flight condition information includes the altitude, temperature, and flight speed; and a thrust
fraction fT .

19–2 Turbojet or Turbofan

The thrust of a turbojet is

 
T = ṁ (1 + f )Ve − V + (pe − patm )Ae

where ṁ is the mass flow; f = ẇ/ṁ is the fuel-air ratio; and Ve , pe , and Ae are the velocity, pressure,
and area at the exit. The pressure term is zero or small, and the fuel-air ratio is small, so the net thrust
is approximately
T = ṁ(Ve − V )

from the gross thrust TG = ṁVe and the inlet-momentum or ram drag ṁV .

The thrust of a turbofan is

    
T = ṁ (1 + f )Ve − V + ṁfan Vefan − V = ṁ (1 + f )Ve + βVefan ) − ṁ(1 + β)V

with bypass ratio β = ṁfan /ṁ.

Turbojet or turbofan performance is obtained from the Referred Parameter Jet Engine Model (RP-
JEM), based on references 1 and 2. The referred thrust, specific thrust ST = T /ṁ, and specific fuel con-
180 Jet Group
√
sumption sfc = ẇ/T are functions of the flight Mach number M and compressor speed n = N/(N0 θ):

T
= Gt (M, n)
δ
T /ṁ
√ = Gst (M, n)
θ
ẇ/T 1
√ = Gsfc (M, n)
θ ηb

The independent variable can be the compressor speed, or the turbine inlet temperature, or the fuel flow.
The combustion efficiency ηb depends on the atmosphere (altitude and temperature), hence the specific
fuel consumption is not a function of just M and n.

19-2.1 Thrust Available

Given the flight condition and engine rating, the thrust available Ta is calculated from the specific
thrust STa = Ta /ṁa and mass flow ṁa :

STa
√ = ST0 gst (θ, M )
θ
ṁa
√ = ṁ0 gm (θ, M )
δ/ θ
Ta
= T0 gt (θ, M )
δ

as functions of temperature ratio θ = T /T0 and Mach number M .

In the jet model, installation losses Tloss are subtracted from Ta (Tav = Ta − Tloss ), and then the
mechanical limit is applied: Tav = min(Tav , TmechR ).
The jet model gives the performance of a single jet. The thrust available of the jet group is obtained
by multiplying the single jet thrust by the number of jets operational (total number of jets less inoperable
jets):
TavJG = fT (Njet − Ninop )Tav

including a specified thrust fraction fT .

19-2.2 Performance at Thrust Required

The thrust required of a single jet is obtained by dividing the jet group thrust by the number of jets
operational (total number of jets less inoperable jets):

Treq = TreqJG /(Njet − Ninop )

In the jet model, installation losses Tloss are added to Treq (Tq = Treq + Tloss ).
The jet performance (mass flow and fuel flow) is calculated for a specified thrust required Tq and
flight condition:
ṁreq
√ = ṁ0C gm (t, θ, M )
δ/ θ
ẇreq
√ = ẇ0C gw (t, θ, M )
δ θ
Jet Group 181

as functions of t = Tq /(T0C δ) (or referred gross thrust), temperature ratio θ = T /T0 , and Mach number
M.

The jet model gives the performance of a single jet. The performance of the jet group is obtained
by multiplying the single jet characteristics by the number of jets operational:
ṁreqJG = (Njet − Ninop )ṁreq
ẇreqJG = (Njet − Ninop )ẇreq Kf f d
DauxJG = (Njet − Ninop )Daux

The fuel flow has also been multiplied by a factor Kf f d accounting for deterioration of the jet efficiency.

19-2.3 Installation
The difference between installed and uninstalled thrust is the inlet and exhaust losses Tloss : Tav =
Ta − Tloss and Treq = Tq − Tloss . The inlet and exhaust losses are modeled as fractions of thrust available
or thrust required: Tloss = ( in + ex )Ta or Tloss = ( in + ex )Tq . The momentum drag of the auxiliary
air flow is a function of the mass flow ṁaux = faux ṁreq :

Daux = (1 − ηaux )ṁaux V = (1 − ηaux )faux ṁreq V

where ηaux is the ram recovery efficiency. Exhaust losses ( ex ) and auxiliary air flow parameters (ηaux ,
faux ) are defined for IR suppressor on and off.

19-2.4 Convertible Engine: Reaction Drive

The jet mode B is the mass flow fraction diverted for a convertible engine: B = 0 for all mass flow
to the exhaust (turbojet operation), and B = 1 for all mass flow to the rotor (reaction jet operation).
A separate jet model defines the engine performance for B = 1. The jet thrust required is fixed by
amplitude input, or obtained from the rotor power (TreqJG = Freact = Preq /Ωrreact ). The mass flow and
fuel flow follow. The jet group net thrust is the inlet momentum drag, FN = −ṁreq V .
Thrust, specific thrust, and mass flow are defined by the jet model for turbojet operation. The
specific fuel consumption of the jet model for reaction jet operation is scaled proportional to the sfc
scaling for turbojet operation.

19–3 Reaction Drive

Rotor power can be supplied by a reaction drive, using cold or hot air ejected out of the blade tips
or trailing edges. Helicopters have also been designed with ram jets on the blade tips, or with jet flaps
on the blade trailing edges that use compressed air generated in the fuselage. The jet group performance
includes the blade duct and nozzle, perhaps even with tip burning.
The net jet thrust required is fixed by amplitude input, or obtained from the rotor power. The rotor
power required Preq gives the required force on the rotor blade
 
Freact = Preq /Ωrreact = ṁreact Vreact − Ωrreact

at effective radial station rreact . The net jet group thrust required is

TreqJG = Freact = TG − ṁreqJG Ωrreact

182 Jet Group

TG is the gross thrust. From TreqJG , the jet performance (mass flow and fuel flow) is calculated. The jet
group net thrust is the inlet momentum drag, FN = −ṁreq V .

19–4 Simple Force

For the simple force model, the design maximum thrust is Tmax (per jet). The thrust available is
thus Ta = Tmax . The force generation can use fuel as weight or as energy.
If the component burns fuel weight, the fuel flow is calculated from an input thrust-specific fuel
consumption: ẇreq = Tq (sfc) = ẇ0C q , where q = Tq /Tmax . Units of sfc are pound/hour/pound or
kilogram/hour/Newton. Then
ẇreqJG = (Njet − Ninop )ẇreq Kf f d

The fuel flow has also been multiplied by a factor Kf f d accounting for deterioration of the jet efficiency.
If the component uses fuel energy, the energy flow is calculated from an input thrust-specific
fuel consumption: Ėreq = Tq (sfc) = Ė0C q , where q = Tq /Tmax . Units of sfc are MJ/hour/pound or
MJ/hour/Newton. Then
ĖreqJG = (Njet − Ninop )Ėreq Kf f d

The energy flow has also been multiplied by a factor Kf f d accounting for deterioration of the jet efficiency.
The simple force weight is calculated from specific weight S plus a fixed increment: W = STmax +
ΔW . This weight is identified as either engine system or propeller/fan installation weight, both of the
propulsion group; or tail-rotor, of empennage group.

19–5 Control and Loads

The jet thrust amplitude A and mode B are control variables. The mode can be mass flow (for
convertible engines). The jet group thrust can be fixed at TreqJG = (Njet − Ninop )A; or fraction A of jet
thrust available, TreqJG = (Njet −Ninop )ATav ; or fraction A of jet rated thrust TreqJG = (Njet −Ninop )ATjet .
The jet orientation is specified by selecting a nominal direction ef 0 in body axes (positive or negative
x-, y -, or z -axis; usually thrust forward, hence positive x-axis); then applying a yaw angle ψ ; and then
an incidence or tilt angle i. The yaw and incidence angles can be control variables.
The jet group produces a jet thrust T , acting in the direction of the engine (actually, the gross thrust
in the direction of the engine and the momentum thrust in the wind direction); a momentum drag Daux ,
acting in the wind direction; and a nacelle drag Dnac , acting in the wind direction. The jet group is at
location z F . The force and moment acting on the aircraft in body axes are thus:

F F = (ef TG + ed Tmom ) + ed Daux + ed Dnac

,F F F
M F = Δz

where Δz F = z F − zcg
F
, ef is the jet thrust direction and ed is the drag direction.

19–6 Nacelle Drag

The jet group includes a nacelle, which contributes to the aircraft drag. The reference area for the
nacelle drag coefficient is the nacelle wetted area. The wetted area per jet is input, or calculated from
Jet Group 183

the engine system (engine, exhaust, and accessories) weight:

 2/3
Swet = k w/Njet

where w = WES , and the units of k are ft2 /lb2/3 or m2 /kg2/3 . The reference area is then Snac = Njet Swet .

19–7 Weights

The component weight consists of engine system, engine section or nacelle group, and air induction
group. The engine system consists of engine, exhaust system, and accessories. The engine section or
nacelle group consists of engine support, engine cowling, and pylon support. These weights are for the
jet group, consisting of Njet engines. The engine system weight WES = Weng + Wexh + Wacc is used for
the engine nacelle wetted area.

19–8 References

1) Sanders, N.D. “Performance Parameters for Jet-Propulsion Engines.” NACA TN 1106, July 1946.
2) Hill, P.G., and Peterson, C.R. Mechanics and Thermodynamics of Propulsion. Reading, MA:
Addison-Wesley Publishing Company, Inc., 1965.
184 Jet Group
 Chapter 20
Charge Group

A charge group generates energy for the aircraft. A charger model describes a particular charger,
used in one or more charge groups. The models include fuel cells and solar cells.

20–1 Charge Group Performance

The charger size is described by the power Pchrg , which is the sea-level static power available per
charger at a specified takeoff rating. The number of chargers Nchrg is specified for each charge group.
If the sizing task determines the charger power, the charger power required is rPchrg , where r =
max(PreqCG /PavCG ).

The flight condition information includes the altitude, temperature, and flight speed; and a power
fraction fP .
Given the flight condition and the charger rating, the charger power available is
Pav = Ėacell = P0 gp (δ, M )

as a function of pressure ratio δ = p/p0 and Mach number M . The power available is obtained by
multiplying the single charger power by the number of chargers operational (total number of chargers
less inoperable chargers):
PavCG = fP (Nchrg − Ninop )Pav
including a specified power fraction fP .
The energy flow to the fuel tank gives the charge group power required: PreqCG = ĖreqCG . The
power required of a single charger is
Preq = PreqCG /(Nchrg − Ninop )

The cell energy flow required Ėqcell is Ėqcell = P0C ge (q), where q = Preq /P0C . Then
Preqtotal = (Nchrg − Ninop )Ėqcell

is the total cell power required.

20–2 Fuel Cell

A fuel cell burns a fuel (typically hydrogen) and generates electrical energy. For a specified flight
condition, the cell power available and power required are
Pav = Ėacell = P0 gp (δ, M )
Ėqcell = P0C ge (q)
186 Charge Group

where q = Preq /P0C . The fuel cell performance (mass flow and fuel flow) is

ṁreq = ṁ0C gm (q, δ, M )

ẇreq = ẇ0C gw (q, δ, M )

Installation losses are included in the specific fuel consumption. The net thrust is the inlet momentum
drag, FN = −ṁreq V . The charger model gives the performance of a single charger. The performance of
the charge group is obtained by multiplying the single charger characteristic by the number of chargers
operational:
ṁreqCG = (Nchrg − Ninop )ṁreq
ẇreqCG = (Nchrg − Ninop )ẇreq Kf f d
FN CG = (Nchrg − Ninop )FN
DauxCG = (Nchrg − Ninop )Daux
The fuel flow has also been multiplied by a factor Kf f d accounting for deterioration of the engine
efficiency. The momentum drag of the auxiliary air flow is a function of the mass flow ṁaux = faux ṁreq :

Daux = (1 − ηaux )ṁaux V = (1 − ηaux )faux ṁreq V

where ηaux is the ram recovery efficiency.

20–3 Solar Cell

The power available from solar radiation is approximately 1.36 kW/m2 , reduced by atmospheric
effects (absorption, reflection, and scattering) to about 1.00 kW/m2 . The average solar radiation in the
continental United States is 3.5–7.0 (kW/m2 )(hour/day), hence approximately 15 to 25% of the available
power. Typical efficiencies of solar cells are 10–35%, with some sensitivity to temperature. The solar
cell is characterized by power density (W/m2 ) and weight density (kg/m2 ).
For a specified flight condition, the cell energy flow is

Pav = Ėacell = P0 gp (δ, M )

Ėqcell = P0C ge (q)

where q = Preq /P0C .

20–4 Control and Loads

The charger power amplitude A and mode B are control variables. The charge group power can
be fixed at Ė = PreqCG = (Nchrg − Ninop )A; or fraction A of charger rated power Ė = PreqCG =
(Nchrg − Ninop )APchrg .

The charger orientation is specified by selecting a nominal direction ef 0 in body axes (positive or
negative x-, y -, or z -axis); then applying a yaw angle ψ ; and then an incidence or tilt angle i. The yaw
and incidence angles can be control variables.
The charge group produces a jet thrust FN (actually, FN = FG − ṁV , with the gross thrust in the
direction of the engine and the momentum thrust in the wind direction), acting in the direction of the
charger; a momentum drag Daux (if the component has mass flow), acting in the wind direction; and
Charge Group 187

a nacelle drag Dnac , acting in the wind direction. The charge group is at location z F . The force and
moment acting on the aircraft in body axes are thus:

F F = (ef FG + ed ṁV ) + ed Daux + ed Dnac

,F F F
M F = Δz

where Δz F = z F − zcg
F
, ef is the charger thrust direction and ed is the drag direction.

20–5 Nacelle Drag

The charge group includes a nacelle, which contributes to the aircraft drag. The reference area for
the nacelle drag coefficient is the nacelle wetted area. The wetted area per charger is input, or calculated
from the engine system (engine, exhaust, and accessories) weight:
 2/3
Swet = k w/Nchrg

where w = WES , and the units of k are ft2 /lb2/3 or m2 /kg2/3 . The reference area is then Snac = Nchrg Swet .

20–6 Weights

The component weight consists of the engine system. The engine system consists of the charger
weight. These weights are for the charge group, consisting of Nchrg chargers. The engine system weight
WES = Nchrg Wone chrg is used for the charger nacelle wetted area.
188 Charge Group
 Chapter 21

Referred Parameter Turboshaft Engine Model

Aircraft gas turbine engine performance capabilities are formally specified by computer programs
known as engine decks, which are created by engine manufacturers in an industry-standard format.
Engine decks are typically based on thermodynamic cycle analysis using real engine component per-
formance maps. The most important performance maps for turboshaft engines are compressor, gas
generator turbine, and power turbine. These component performance maps are critical to obtaining
realistic off-design engine performance. Design and analysis codes calculate aircraft performance for
a very wide range of operating conditions. Thus engine performance must be realistic even far from
the engine design point. A simple thermodynamic cycle analysis that assumes design point component
efficiencies everywhere is not realistic for such an application. Rather than developing models for com-
ponent performance, the approach taken is to use a model for the total engine performance. The engine
is not being designed.
The Referred Parameter Turboshaft Engine Model (RPTEM) is based on curve-fits of performance
data for existing or projected engines over a range of operating conditions. The curve-fits are typically
obtained by exercising an engine deck. The use of referred parameters tends to collapse the data,
and provides a basis for scaling the engine. The operating condition is described by pressure altitude,
ambient air temperature, flight Mach number, power turbine speed, exhaust nozzle area, and either
engine rating or engine power required. These curve-fits, typically based on real engines, are scaled to
the required size and adjusted to the appropriate technology level to represent a notional engine. Engine
size is represented by mass flow. Engine technology is represented by specific power available and
specific fuel consumption at maximum continuous power (MCP), sea level/standard day (SLS), static
(zero airspeed) conditions. Engine installation effects (inlet and exhaust losses) are also modeled.
The use of referred parameters to curve-fit engine performance data was suggested by David
Woodley from Boeing during the JVX program (1983). The RPTEM was developed and documented
by Michael P. Scully and Henry Lee of ASRAO, U.S. Army Aeroflightdynamics Directorate (AFDD),
with a subsequent implementation written by Sam Ferguson (1995).

21–1 Operating Environment

The operating condition and atmosphere give the standard conditions (temperature Tstd and pressure
pstd ) for a specified pressure altitude; the sea-level standard conditions (temperature T0 and pressure
p0 ); and the operating temperature T and pressure p. Here the temperatures are ◦ R or ◦ K. The engine
characteristics depend on the temperature ratio θ = T /T0 and pressure ratio δ = p/p0 .
√
The flight Mach number M = V /cs = V /cs0 θ is obtained from the aircraft speed V .
190 Referred Parameter Turboshaft Engine Model

The inlet ram air temperature ratio and pressure ratio are obtained then from M and the inlet ram
recovery efficiency ηd :
 
γ−1 2  
θM = 1+ M = 1 + 0.2M 2
2
  γ−1
γ
γ−1  3.5
δM = 1+ ηd M 2 = 1 + 0.2ηd M 2
2

where the ratio of specific heats γ = 1.4.

21–2 Performance Characteristics

The engine performance is described by: the power available Pa , at each engine rating and the
specification engine turbine speed Nspec ; the mass flow ṁ and fuel flow ẇ required to produce power
required Pq at engine turbine speed N ; and the gross jet thrust F at a given power required Pq . Then the
specific power is SP = P/ṁ, and the specific fuel consumption is sfc = ẇ/P .
The reference performance is at sea-level-standard static conditions (subscript 0), and MCP (sub-
script C ). For each rating R, the performance is characterized by the following quantities for sea-level-
standard static conditions: power P0R , specific power SP0R , and mechanical power limit PmechR . The
mass flow is then ṁ0R = P0R /SP0R . The gross jet thrust Fg0C is given at MCP. These characteristics
are at the specification turbine speed Nspec .
The installed power required Preq and power available Pav > Preq are measured at the engine output
shaft. In addition to shaft power, the engine exhaust produces a net jet thrust FN , from mass flow that
goes through the engine core. The fuel flow and mass flow are the total required to produce the shaft
power and jet thrust. The forces produced by mass flow that does not go through the engine core (such
as infrared suppressor or cooling air) are treated as momentum drag Daux .
The difference between net and gross jet thrust is the momentum drag: Fn = Fg − ṁreq V =
ṁreq (Vj − V ), where Vj is the engine jet exhaust velocity. Note that traditional units for mass flow are
pound/sec (pps), while this equation requires slug/sec (ṁreq /g replaces ṁreq ).
The uninstalled power required is Pq , the power available Pa , the gross jet thrust Fg , and net jet thrust
Fn . The engine model calculates Pa as a function of flight condition and engine rating; or calculates
engine mass flow, fuel flow, and jet thrust at Pq .

21–3 Installation

The difference between installed and uninstalled power is the inlet and exhaust losses Ploss :

Pav = Pa − Ploss
Preq = Pq − Ploss

The inlet ram recovery efficiency ηd (through δM ) is included in the engine model calculations. The inlet
and exhaust losses are modeled as fractions of power available or power required: Ploss = ( in + ex )Pa
or Ploss = ( in + ex )Pq . So
Pav = Pa (1 − in − ex )

Pq = Preq /(1 − in − ex )
Referred Parameter Turboshaft Engine Model 191

The engine model gives uninstalled power and the gross thrust Fg for a nominal exhaust nozzle area. The
gross jet thrust FG and exhaust power loss ( ex ) are both functions of the exhaust nozzle area. Smaller
exhaust nozzle areas increase exhaust losses and increase gross thrust. Thus the ratio of installed to
uninstalled thrust is approximated by a function of the exhaust power loss:

2 3
FG /Fg = Kf gr = Kf gr0 + Kf gr1 ex + Kf gr2 ex + Kf gr3 ex

so the net installed jet thrust is

FN = Kf gr Fg − ṁreq V

The momentum drag of the auxiliary air flow is a function of the mass flow ṁaux = faux ṁreq :

Daux = (1 − ηaux )ṁaux V = (1 − ηaux )faux ṁreq V

where ηaux is the ram recovery efficiency. Exhaust losses ( ex ) and auxiliary air flow parameters (ηaux ,
faux ) are defined for IR suppressor on and off. Inlet particle separator loss is added to the inlet losses
( in ).

21–4 Power Turbine Speed

The shaft power available is a function of the gas power available PG and the power turbine efficiency
ηt : Pa = ηt PG . Generally the power turbine speed N has a significant effect on ηt , but almost no effect
on PG . The model used for the efficiency variation is ηt ∼ = 1 − |(N/Nopt ) − 1|XN η , where Nopt is the
speed for peak efficiency, hence

P (N ) ηt (N ) 1 − |(N/Nopt ) − 1|XN η
= =
P (Nspec ) ηt (Nspec ) 1 − |(Nspec /Nopt ) − 1|XN η

Two approximations
√
for the optimum turbine speed are used. The first is a linear expression in p =
P/(P0R δ θ):  
√  KN optB
Nopt = Nspec θ KN optA θM + p
δM
√
and the second is a cubic function of p = P/(P0C δ θ):
√   X
Nopt = Nopt0C θ KN opt0 + KN opt1 p + KN opt2 p2 + KN opt3 p3 [θM ] N opt

The second expression is based on a larger data sample. For power available calculations, P = Pa (Nspec );
for power required calculations, P = Pq (N ).

21–5 Power Available

Given the flight condition and engine rating, the power available Pa is calculated as follows. The
specific power and referred mass flow (at Nspec , relative to SP0 and ṁ0 for this rating) are approximated
by functions of the ambient temperature ratio θ and inlet ram air ratios:
"  %Xspa
SPa (Nspec ) = SP0 θ Kspa δM θM
 √  "  %Xmf a
ṁa (Nspec ) = ṁ0 δ/ θ eKmf a δM θM
192 Referred Parameter Turboshaft Engine Model

where the static lapse rate (Kspa , Kmf a ) and ram air exponents (Xspa , Xmf a ) are piecewise linear
functions of θ. The power available is then
SPa (Nspec ) ṁa (Nspec )  √  "  %Xspa +Xmf a
Pa (Nspec ) = P0 = P0 δ θ Kspa eKmf a δM θM
SP0 ṁ0
This expression for ṁa is used only to calculate Pa ; elsewhere the ṁq expression below (for performance
at power required) is used to obtain the mass flow at a power Pq . Finally
1 − |(N/Nopt ) − 1|XN η
Pa (N ) = Pa (Nspec )
1 − |(Nspec /Nopt ) − 1|XN η

is the power available at turbine speed N . Installation losses Ploss are subtracted from Pa (Pav =
Pa − Ploss ), and then the mechanical limit is applied: Pav = min(Pav , rPmechR ), r = N/Nspec . The
mechanical limit is properly a torque limit, Qmech = Pmech /Nspec , but is expressed as a power limit for
clarity.
For a simple model,
√
the power available can be constant, Pa = P0 ; or the referred power can be
constant, Pa = P0 (δ θ); without any effect of power turbine speed.

21-5.1 Piecewise Linear Parameters

Several parameters K are input as piecewise linear functions of the temperature ratio θ. One format
of the input is a set of I regions, with the function in the i-th region given by K = K0i + K1i θ. The break
point between the i and i − 1 regions is
K0i − K0(i−1)
θbi = −
K1i − K1(i−1)
Kbi = K0i + K1i θb

for i = 2 to I . Another format is a set of I + 1 break points, with the values at the i-th point (θbi , Kbi ).
Then the coefficients between i and i + 1 are
Kbi θb(i+1) − Kb(i+1) θbi
K0i =
θb(i+1) − θbi
−Kbi + Kb(i+1)
K1i =
θb(i+1) − θbi

for i = 1 to I . The interpolation is performed using K = K0i + K1i θ for θ in the range θbi to θb(i+1) .
Outside the defined regions, the linear expression is continued; hence θbi is used only for i = 2 to I .

21–6 Performance at Power Required

The engine performance (mass flow, fuel flow, and gross jet thrust) is calculated for a specified power
required Pq (which might equal the power available), flight condition, and engine rating. Installation
losses Ploss are added to Preq (Pq = Preq + Ploss ). The referred quantities
√
(relative to SLS static MCP
quantities) are approximated by cubic functions of q = Pq (Nspec )/(P0C δ θ):
 √   −X
ẇreq = ẇ0C δ θ Kf f q0 + Kf f q1 q + Kf f q2 q 2 + Kf f q3 q 3 [θM ] f f q
 √   X
ṁreq = ṁ0C δ/ θ Kmf q0 + Kmf q1 q + Kmf q2 q 2 + Kmf q3 q 3 [θM ] mf q
  X
Fg = Fg0C (δ) Kf gq0 + Kf gq1 q + Kf gq2 q 2 + Kf gq3 q 3 [θM ] f gq
Referred Parameter Turboshaft Engine Model 193

at Nspec , with ẇ0C = sfc0C P0C . The mass flow and fuel flow are primarily functions of the gas power PG ,
and are assumed to be independent of ηt , hence independent of turbine speed. However, these equations
are functions of Pq (Nspec ), obtained from Pq (N ) using

1 − |(Nspec /Nopt ) − 1|XN η

Pq (Nspec ) = Pq (N )
1 − |(N/Nopt ) − 1|XN η

Then the installed net jet thrust FN and momentum drag Daux are calculated.
For a simple model, the referred performance can be constant:
√
ẇreq = ẇ0C (δ θ)q = sfc0C Pq
√
ṁreq = ṁ0C (δ/ θ)
Fg = Fg0C (δ)

without any effect of power turbine speed.

21–7 Scaling

The parameters of the engine model can be defined for a specific engine, but it is also necessary to
scale the parameters as part of the aircraft sizing task, in order to define an engine for a specified power.
In addition, advanced technology must be represented in the model. Scaling and advanced technology
are handled in terms of specific power and specific fuel consumption (at SLS static conditions, MCP, and
Nspec ). Figures 21-1 through 21-3 present historical data for engine specific fuel consumption, weight,
and specific power; the variation at a given power reflects technology insertion with time.
The engine model includes reference values of the engine performance parameters: P0R , SP0R ,
PmechR , sfc0C , SF0C , Nspec , and Nopt0C . Mass flow and fuel flow are obtained from ṁ0R = P0R /SP0R and
ẇ0C = sfc0C P0C . The reference power at each engine rating R defines a ratio to MCP: rp0R = P0R /P0C .
Similarly for specific power and mechanical limits: rs0R = SP0R /SP0C and rm0R = PmechR /P0C . These
ratios are kept fixed when the engine is scaled.
The engine size is specified as takeoff power Pto = Peng , which is the power at rating R, for SLS
static conditions and specification turbine speed Nspec . Hence the MCP is P0C = Pto /rp0R , and the
power at all other ratings follows. If P0C is not equal to the reference value of the engine model, then
the engine is scaled. To reflect advanced technology, the specific power, specific fuel consumption,
and specification turbine speed can be specified: SP0C = SPtech , sfc0C = sfctech , and Nspec = Ntech
(replacing the engine model reference values). The default values are the reference values of the engine
model. In the following paragraph, the subscript “tech” refers to these quantities; the subscript “ref ”
means the engine model reference values.
The engine technology parameters SP0C and sfc0C are assumed to vary linearly with mass flow ṁ0C
up to a limit ṁlim , and constant thereafter at SPlim and sfclim . The mass flow at the technology condition
is ṁtech = Pref /SPtech , with the technology values SPtech and sfctech . The intercept values are projected
from the technology values: Ksp0 = SPtech − Ksp1 ṁtech , Ksp1 = (SPlim − SPtech )/(ṁlim − ṁtech ); and
similarly for sfc. Then for ṁ0C < ṁlim

SP0C = Ksp0 + Ksp1 ṁ0C

sfc0C = Ksf c0 + Ksf c1 ṁ0C
194 Referred Parameter Turboshaft Engine Model

and for ṁ0C ≥ ṁlim

SP0C = SPlim
sfc0C = sfclim

These equations are used if ṁlim > ṁtech . Otherwise the model sets Ksp1 = Ksf c1 = 0, so there is no
variation with scale: SP0C = SPtech and sfc0C = sfctech . Usually the effect of size gives Ksp2 ≥ 0 and
Ksf c2 ≤ 0. The power at the limit is Plim = SPlim ṁlim . Figure 21-4 illustrates the scaling of SP , sfc,
and SW .

Using ṁ0C = P0C /SP0C , the specific power equation can be solved for the mass flow given the
power:
⎧
⎪
⎪
P0C /Ksp0 Ksp1 = 0
⎪
⎨
ṁ0C = P0C /SPlim P0C ≥ Plim
⎪
⎪ 
⎪
⎩ ( K0 )2 +
2K1
P0C
K1 − K0
2K1 otherwise

From this mass flow, SP0C and sfc0C are calculated, hence the fuel flow ẇ0C = sfc0C P0C . The specific
thrust available at MCP is assumed to be constant, and the specification power turbine speed decreases
with the mass flow:

Fg0C = SF0C ṁ0C

    
Nspec = Nspec − KN s2 / ṁ0C + KN s2 / ṁ0C = KN s1 + KN s2 / ṁ0C
tech
(Nopt0C )ref
Nopt0C = Nspec = KN o Nspec
Ntech

Then the power and specific power at all ratings R are obtained from the ratios: P0R = rp0R P0C ,
SP0R = rs0R SP0C , and PmechR = rm0R P0C .

The actual (perhaps scaled) values of the performance parameters are available for the engine group:
P0R , SP0R , PmechR , sfc0C , Fg0C , Nspec , and Nopt0C .

21–8 Engine Speed

The model as described in the previous sections may not adequately account for variation of engine
performance with engine speed, so it is also possible to define the parameters corresponding to a set
of engine speed ratios r = N/Nspec . Then the engine performance and power available quantities are
linearly interpolated to obtain the values at the required engine speed N . If this option is used, then the
correction based on P (N )/P (Nspec ) = ηt (N )/ηt (Nspec ) is not applied.

21–9 Weight

The engine weight can be a fixed input value, calculated as a function of power, or scaled with
engine mass flow. As a function of power, the weight of one engine is:

Wone eng = K0eng + K1eng P + K2eng P Xeng

where P is the installed takeoff power (SLS static, specified rating) per engine. A constant weight per
power W/P is given by using only K1eng . Alternatively, the specific weight SW = P/W can be scaled
Referred Parameter Turboshaft Engine Model 195

0.8

specific fuel consumption (lb/hp-hr)

0.7

0.6

0.5

0.4

0.3
0. 1000. 2000. 3000. 4000. 5000. 6000. 7000.
takeoff power (hp)

Figure 21-1. Historical data for turboshaft engine specific fuel consumption.

with the mass flow ṁ0C . The scaling is determined from the specific weight SWref at the mass flow
ṁtech , and the limit SWlim at ṁlim . Then

Ksw0 + Ksw1 ṁ0C ṁ0C < ṁlim

SW =
SWlim ṁ0C ≥ ṁlim

(if ṁlim < ṁtech , then Ksw1 = 0, so SW = SWref ) and Wone eng = P/SW .

21–10 Units

In this engine model, only the reference values and scaling constants are dimensional. Conventional
English units and SI units are shown in table 21-1. Units of specific power and specific fuel consumption
follow from these conventions.

Table 21-1. Conventional units.

power P mass flow ṁ fuel flow ẇ force F turbine speed N
English: horsepower pound/sec pound/hour pound rpm
SI: kiloWatt kilogram/sec kilogram/hour Newton rpm

21–11 Typical Parameters

Typical values of the principal parameters describing the engine and its performance are given in
table 21-2 for several generic engine sizes. These values represent good current technology. Advanced
196 Referred Parameter Turboshaft Engine Model

technology can be introduced by reducing the specific fuel consumption and weight, and increasing the
specific power. Typical ratios of the power, specific power, and mass flow to the values at MCP are
given in table 21-3 for several ratings. Figure 21-5 shows typical performance characteristics: fuel flow
ẇ, mass flow ṁ, and net jet thrust Fg variation with power P and speed (referred quantities, normalized,
at Nspec ). Figure 21-6 shows typical variation of the power available with engine turbine speed. Figures
21-7 to 21-12 show typical power available characteristics: specific power SP , mass flow ṁ, and power
P variation with temperature ratio θ, for static and 200 knots conditions and several engine ratings
(referred quantities, normalized, at Nspec ).

Table 21-2. Typical engine performance parameters.

power (MCP) P0C 500 1000 2000 4000 8000 16000 hp
specific power SP0C 116 125 134 143 153 164 hp/lb/sec
mechanical limit Pmech 750 1500 3000 6000 12000 24000 hp
specific fuel cons. sfc0C 0.54 0.48 0.44 0.40 0.38 0.35 lb/hp-hr
specific jet thrust SF0C 5.9 7.3 8.9 11.0 13.6 16.7 lb/lb/sec
gross jet thrust Fg0C 25 58 134 308 707 1625 lb
mass flow ṁ0C 4.3 8.0 15.0 27.9 52.1 97.3 lb/sec
turbine speed Nspec 35600 26100 19100 14000 10200 7500 rpm
weight W/Pto 0.34 0.23 0.18 0.16 0.16 0.15 lb/hp

Table 21-3. Typical parameter ratios for various ratings (percent).

IRP MRP CRP Pmech
power P0 /P0C 120 127 133 150
specific power SP0 /SP0C 117 123 128
mass flow ṁ0 /ṁ0C 102.6 103.2 103.9
Referred Parameter Turboshaft Engine Model 197

0.7

0.6

weight (lb/hp) 0.5

0.4

0.3

0.2

0.1

0.0
0. 1000. 2000. 3000. 4000. 5000. 6000. 7000.
takeoff power (hp)

Figure 21-2. Historical data for turboshaft engine weight.

250.

200.
SP (hp/lb/sec)

150.

100.

50.

0.
0. 1000. 2000. 3000. 4000. 5000. 6000. 7000.

takeoff power (hp)

Figure 21-3. Historical data for turboshaft engine specific power.

198 Referred Parameter Turboshaft Engine Model

. . . .
m lim> mtech m lim≤ mtech

lim lim
SP0C SP0C
tech tech

tech lim lim tech

. .
m 0C m 0C

sfc 0C tech sfc 0C tech

lim lim

tech lim lim tech

. .
m 0C m 0C

lim lim
SW SW
ref ref

tech lim lim tech

. .
m 0C m 0C

Figure 21-4. Turboshaft engine scaling; SP = P/ṁ, sfc = ẇ/P , and SW = P/W .
Referred Parameter Turboshaft Engine Model 199

2.0

1.5

1.0

· · · ·
w / w0C static w / w0C 200 knots
0.5 · · · ·
m / m0C static m / m0C 200 knots
F g /Fg0C static F g /Fg0C 200 knots

0.0
0.0 0.5 1.0 1.5 2.0
P(Nspec)/P 0C

Figure 21-5. Fuel flow, mass flow, and net jet thrust variation with power.

1.2

1.1

1.0
P(N)/P(Nspec)

0.9

0.8

0.7

0.6
0.4 0.6 0.8 1.0 1.2

N/Nspec

Figure 21-6. Power variation with turbine speed.

200 Referred Parameter Turboshaft Engine Model

1.8 MPC
IRP
1.6
MRP
1.4 CRP

1.3
SP/SP0

1.1

0.9

0.8

0.6

0.4
0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15
temperature ratio θ

Figure 21-7. Specific power variation with temperature ratio, static.

1.8 MPC
IRP
1.6 MRP
CRP
1.4

1.2
SP/SP0

1.0

0.8

0.6

0.4
0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15

temperature ratio θ

Figure 21-8. Specific power variation with temperature ratio, 200 knots.
Referred Parameter Turboshaft Engine Model 201

1.20 MPC
IRP
1.15 MRP
CRP
1.10
/ m0
· ·

1.05
m

1.00

0.95

0.90
0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15
temperature ratio θ

Figure 21-9. Mass flow variation with temperature ratio, static.

1.20 MPC
IRP
1.15 MRP
CRP
1.10
/ m0
· ·

1.05
m

1.00

0.95

0.90
0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15

temperature ratio θ

Figure 21-10. Mass flow variation with temperature ratio, 200 knots.
202 Referred Parameter Turboshaft Engine Model

1.8 MPC
IRP
1.6 MRP
CRP
1.4

1.2
P/P0

1.0

0.8

0.6

0.4
0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15
temperature ratio θ

Figure 21-11. Power variation with temperature ratio, static.

1.8 MPC
IRP
1.6 MRP
CRP
1.4

1.2
P/P0

1.0

0.8

0.6

0.4
0.80 0.85 0.90 0.95 1.00 1.05 1.10 1.15

temperature ratio θ

Figure 21-12. Power variation with temperature ratio, 200 knots.

 Chapter 22

Reciprocating Engine Model

A reciprocating engine generates power by means of reciprocating pistons that convert pressure
into a rotating motion. Each piston is inside a cylinder, into which a gas is introduced, either already
under pressure (external combustion, as in a steam engine) or heated inside the cylinder by ignition
of a fuel-air mixture (internal combustion). The hot gases expand, pushing the piston to the bottom
of the cylinder. Here only internal combustion engines are considered, perhaps more appropriately
identified as intermittent combustion engines, to distinguish them from the continuous combustion of
gas turbines and jet engines. An internal combustion engine is either a spark-ignition engine, where a
spark plug initiates the combustion; or a compression-ignition engine, where the air within the cylinder
is compressed, heating it and thereby igniting fuel. Internal combustion engine performance is described
in reference 1.

Design and analysis codes calculate aircraft performance for a very wide range of operating con-
ditions. Thus engine performance must be realistic even far from the engine design point. A simple
thermodynamic cycle analysis is not realistic for such an application. Rather than developing high
fidelity analyses, the approach taken is to use a model for the total engine performance. The engine is
not being designed.

The Reciprocating Engine Model is based on curve-fits of performance data for existing or projected
engines over a range of operating conditions. The operating condition is described by pressure altitude,
ambient air temperature, flight Mach number, engine speed, and either engine rating or engine power
required. These curve-fits, typically based on real engines, are scaled to the required size and adjusted
to the appropriate technology level to represent a notional engine. Engine size is represented by power.
Engine technology is represented by specific fuel consumption and power-to-weight ratio. Engine
installation losses are also modeled.

22–1 Operating Environment

The operating condition and atmosphere give the standard conditions (temperature Tstd and pressure
pstd ) for a specified pressure altitude; the sea-level standard conditions (temperature T0 and pressure
p0 ); and the operating temperature T and pressure p. Here the temperatures are ◦ R or ◦ K. The engine
characteristics depend on the density ratio σ = ρ/ρ0 and the temperature ratio θ = T /T0 .
√
The flight Mach number M = V /cs = V /cs0 θ is obtained from the aircraft speed V .

The inlet ram air temperature ratio and pressure ratio are obtained then from M and the inlet ram
204 Reciprocating Engine Model

recovery efficiency ηd :
 
γ−1 2  
θM = 1+ M = 1 + 0.2M 2
2
  γ−1
γ
γ−1  3.5
δM = 1+ ηd M 2 = 1 + 0.2ηd M 2
2

where the ratio of specific heats γ = 1.4. The ram density ratio is σM = δM /θM . The engine mass flow
and power available depend on (σ + ΔσM ), where ΔσM = σM − 1

22–2 Reciprocating Engine Performance

The engine work per cycle for a reciprocating engine is W = P nc /R; where P is the power,
R = N/2π the rotational speed (rev/sec, N in rad/sec), nc the revolutions per cycle (nc = 2 for a 4-stroke
engine, nc = 1 for a 2-stroke engine), and R/nc is the cycles per second. The piston area is Ap , bore
b, stroke k ; the engine displacement Vd = Ap k (swept volume). The mean piston speed is defined as
s = 2Rk (1/2R is the time for one stroke). Thus Ap s = 2RVd . The mean effective pressure (mep) is
defined as
W P nc P Qnc 2π
mep = = = =
Vd Vd R Ap s/2nc Vd
from the power P = N Q = 2πRQ; so mep is the specific torque. The power and torque can be written
in terms of mep:
P = (mep)Vd R/nc = (mep)Ap s/2nc
Q = (mep)Vd /2πnc = (mep)Ap k/2πnc
The specific output is P/Ap = (mep)s/2nc .
The engine output is the brake horsepower (BHP), which equals indicated power less friction power:

BHP = IHP − FHP = IHP − (MHP + PHP + CHP + AHP − THP)

The friction power is composed of losses due to mechanical friction (MHP), pumping (PHP, the work of
the piston during inlet and exhaust strokes), compressor or supercharger (CHP), auxiliary or accessories
(AHP, such as oil pump, water pump, cooling fan, generator), and exhaust turbine (THP, treated as
negative friction). The mechanical efficiency is η = BHP/IHP.
The sum of airflow (ṁ = ẇa ) and fuel flow (ẇ = ẇf ) is the charge flow: ẇc = ẇf + ẇa . The fuel-air
ratio is F = ẇf /ẇa . The stoichiometric (chemically correct) fuel-air ratio is about Fstoich = 0.068 for
both gasoline and diesel. Rich mixture (fuel-air ratio) means extra fuel and lean mixture means extra
air. The fuel-air equivalence ratio is φ = F/Fstoich . The mass flow is

ṁ = ev ṁideal = ev ρi Vd (R/nc ) = ev ρi Ap s/2nc = σi N (ev ρ0 Vd /(2πnc ))

where ρi is the density at the intake and ev = ṁ/ṁideal is the volumetric efficiency. Including the
influence of forward speed, ρi = ρ0 (σ + ΔσM ). Typically ev = 0.8 to 0.9. The fuel flow is then ẇ = F ṁ.
The indicated power Pi is the product of the fuel flow, fuel specific energy (efuel = JQc , from the heat
of combustion Qc and Joule’s constant J relating work and heat), and thermal efficiency:

Pi = efuel ηth ẇ = (efuel ηth F )ṁ = σi N (efuel ηth F )(ev ρ0 Vd /(2πnc ))

Reciprocating Engine Model 205

The specific fuel consumption is

sfc = ẇ/P = 1/(efuel ηth )

and the power to mass flow ratio is

P/ṁ = efuel ηth F = F/sfc = (mep)/(ev ρi )

These equations all have constant factors when conventional units are used.
From ṁ ∼ ev ρi and P ∼ ev ρi ηth F , it follows that mass flow and power are proportional to the
density ratio σ = δ/θ. For unsupercharged engines at constant fuel-air fraction and constant throttle,
√
typically ev ρi ∼ p/ T (ref. 1). Similarly, operating curves give ±1% power change for ∓6◦ C at constant
√
pressure altitude and manifold pressure, implying again√P ∼ 1/ T . The implication is that mass flow,
√
indicated power, and imep are proportional to δ/ θ = σ θ.
The engine has a limit on bmep (hence torque) corresponding to detonation limits (spark-ignition)
or thermal and loads limits (compression-ignition) at maximum power. This limit is defined in terms
of power at a specified engine speed. Load limits associated with mean piston speed give an engine
speed limit. Maximum throttle, encountered at a critical altitude or critical density, can be expressed as
a power limit that varies with engine speed. These limits depend on the rating, fuel-air ratio (lean or
rich), and supercharger speed.

22–3 Performance Characteristics

The engine performance is described by the uninstalled power available Pa , as a function of flight
condition and engine rating; the mass flow ṁ and fuel flow ẇ required to produce uninstalled power
required Pq ; and the gross jet thrust. The installed power required Preq and power available Pav > Preq
are measured at the engine output shaft. In addition to shaft power, the engine exhaust can produce a
net jet thrust FN , from mass flow that goes through the engine. The difference between net and gross
jet thrust is the momentum drag: Fn = Fg − ṁreq V = ṁreq (Vj − V ), where Vj is the engine jet exhaust
velocity. The forces produced by mass flow that does not go through the engine (such as cooling air)
are treated as momentum drag Daux .
The engine performance depends on the rating. Typical ratings include maximum continuous power
(MCP) and takeoff power. The rating structure of the model includes dependence on fuel-air ratio (lean
or rich) and supercharger speed.
The reference performance is at sea-level-standard static conditions (subscript 0). For each rating R,
the performance is characterized by the following quantities: power P0R and specific fuel consumption
sfc0R ; fuel-air ratio F0R ; specific thrust SF0R ; mean effective pressure (mechanical) power limit PmepR
and critical power PcritR ; and a reference engine speed N0R . Then the mass flow is ṁ0R = ẇ0R /F0R , and
the fuel flow ẇ0R = sfc0R P0R . The gross jet thrust is Fg0R = SF0R ṁ0R (SF0R = Vj0R ). The specification
engine speed is Nspec .

22–4 Installation

The difference between installed and uninstalled power is the installation loss: Pav = Pa − Ploss and
Preq = Pq −Ploss . This loss is modeled by an efficiency factor: Ploss = (1−ηloss )Pa or Ploss = (1−ηloss )Pq ,
where ηloss = in + ex . The engine model gives the gross thrust Fg . The installed gross jet thrust is
206 Reciprocating Engine Model

FG = Kf gr Fg , where Kf gr accounts for exhaust effects. The net jet thrust is FN = FG − ṁreq V . The
momentum drag of the auxiliary air flow is a function of the mass flow ṁaux = faux ṁreq :

Daux = (1 − ηaux )ṁaux V = (1 − ηaux )faux ṁreq V

where ηaux is the ram recovery efficiency.

22–5 Power Available

Given the flight condition and engine rating, the power available Pa is calculated from:

Pa = P0 Kp (σ + Kram ΔσM )rXpN θXpθ

where the engine speed ratio is r = N/N0 . The factor Kp is included so P0 is a relevant reference power,
instead of the power extrapolated to sea-level conditions. Kram permits adjustment (or suppression)
of the effect of flight speed. The exponents XpN and Xpθ are obtained by fitting the engine operating
curves. The reference engine speed N0 corresponds to P0 .
Maximum throttle, encountered at a critical altitude or critical density, is expressed as a power
limit that varies with engine speed: Pa = min(Pa , PcritR rXcrit ), where r = N/N0 . The exponent Xcrit is
obtained by fitting the engine operating curves.
Installation losses Ploss are subtracted from Pa (Pav = Pa − Ploss ), and then the mechanical limit
(mean effective pressure) is applied: Pav = min(Pav , rPmechR ), r = N/Nspec . The mechanical limit is
properly a torque limit, Qmech = Pmech /Nspec , but is expressed as a power limit for clarity. The mep
limit PmepR is specified at the reference engine speed N0R , hence PmechR = PmepR (Nspec /N0R ).

22–6 Performance at Power Required

The engine performance (fuel flow, mass flow, and gross jet thrust) is calculated for a specified power
required Pq (which might equal the power available), flight condition, and engine rating. Installation
losses Ploss are added to Preq (Pq = Preq + Ploss ). The fuel flow and fuel-air ratio are approximated by
cubic functions of q = Pq /P0 :
 
ẇreq = ẇ0 Kf f q0 + Kf f q1 q + Kf f q2 q 2 + Kf f q3 q 3 rXf f q = ẇ0 Kf f q (q)rXf f q
 
Freq = F0 KF q0 + KF q1 q + KF q2 q 2 + KF q3 q 3 rXF q = F0 KF q (q)rXF q

where the engine speed ratio is r = N/N0 . The reference engine speed N0 corresponds to P0 . Alterna-
tively, Kf f q (q) and KF q (q) can be specified as piecewise-linear functions of q . The constants (Kf f qn ,
KF qn ) and exponents (Xf f q , XF q ) are obtained by fitting the engine operating curves for sfc and F .
For constant specific fuel consumption, Kf f q (q) = q . For constant fuel-air ratio, KF q (q) = 1; for fuel-
air ratio proportional to power, KF q (q) = q . If Fstoich is used for F0 , the fuel-air equivalence ratio is
φ = F/Fstoich = KF q (q)rXF q .

The mass flow is ṁ = ẇ/F . The gross jet thrust is Fg = SF0R ṁ. The installed net jet thrust FN
and momentum drag Daux are calculated from the mass flow.

22–7 Scaling

The parameters of the engine model can be defined for a specific engine, but it is also necessary to
scale the parameters as part of the aircraft sizing task, in order to define an engine for a specified power.
Reciprocating Engine Model 207

The engine model includes reference values of the engine performance parameters: P0R , sfc0R ,
F0R , SF0R , PmepR , PcritR , N0R , Nspec . The actual (perhaps scaled) values are available to the engine
group.
The engine size is specified as takeoff power Pto = Peng , which is the power P0X at rating R = X ,
for SLS static conditions. If P0X is not equal to the reference value of the engine model, then the engine
is scaled. The reference power at each engine rating R defines a ratio to MCP: rp0R = P0R /P0C . The
reference engine speed gives the ratio rN 0R = N0R /Nspec . These ratios are kept fixed when the engine
is scaled. Hence the MCP is P0C = Pto /rp0R , and the power at all other ratings follows.
The size of geometrically scaled engines can be represented by the cylinder bore b. The specific
output, mean piston speed, and specific fuel consumption are relatively weak functions of size (ref. 1).
It is assumed that P/Ap ∼ b−Xo , s ∼ b−Xs , sfc ∼ b−Xf . Hence P ∼ b2−Xo , Nspec ∼ s/k ∼ b−1−Xs , and

sfc0R = sfc0Rref p−Xf /(2−Xo )

Nspec = Nspec−ref p−(1+Xs )/(2−Xo )

where p = Peng /P0Xref . For example, Xo ∼

= 0.2, Xs ∼
= 0.3, Xf ∼
= 0.1 for diesel engines (ref. 1).

Pmep and Pcrit scale with power:

P0R = P0C rp0R

PcritR = P0R rcritR
PmechR = PmepR (Nspec /N0R ) = P0R rmepR

where rcritR = Pcrit−refR /P0Rref and rmepR = (Pmep−ref /P0Rref )(Nspec−ref /N0Rref ). The engine speed
scales with Nspec : N0R = Nspec rN 0R .

22–8 Weight

The engine weight can be a fixed input value, or calculated as a function of power. As a function
of power, the weight of one engine is:

Wone eng = K0eng + K1eng P + K2eng P Xeng

where P is the installed takeoff power (SLS static, specified rating) per engine. A constant weight per
power W/P is given by using only K1eng .

22–9 Units

In this engine model, only the reference values and scaling constants are dimensional. Conventional
English units and SI units are shown in table 22-1. Units of specific power and specific fuel consumption
follow from these conventions.

Table 22-1. Conventional units.

power P mass flow ṁ fuel flow ẇ force engine speed N mep
English: horsepower pound/sec pound/hour pount rpm pound/in2
SI: kiloWatt kilogram/sec kilogram/hour Newton rpm kN/m2
208 Reciprocating Engine Model

22–10 Typical Parameters

The maximum brake mean effective pressure is 125–300 lb/in2 (850–2050 kN/m2 ). Reciprocating
aircraft engine mean piston speed at maximum power is typically 10-16 m/sec for geared engines.
Direct drive aircraft engine speeds are typically limited by propeller tip speed. Stoichiometric (exact
chemical) fuel-air ratios are given in table 22-2 for a number of fuels. The fuel-air equivalence ratio is
φ = F/Fstoich .

Spark-ignition engines can run at stoichiometric fuel-air ratio (φ = 1). Typical operating fuel-
air ratios are φ = 0.8 to 1.5, with rich values up to φ = 1.8 for takeoff power. Modern “lean-burn”
spark-ignition engines use Gasoline Direct Injection (GDI) to get φ below 0.3 at low power. Highly
supercharged engines at maximum power run rich. From reference 1, the highest power for each throttle
setting is at φ ∼
= 1.1, with best brake economy at φ = 0.9 (full load) to φ = 1.1 (zero load).

Compression-ignition engines run lean, limited by incomplete combustion. Typical operating fuel-
air ratios are φ = 0.2 to 0.8, with bmep increasing with φ. Best break economy is near φ = 0.5 (ref. 1).

Table 22-2. Stoichiometric fuel-air ratios.

fuel Fstoich fuel Fstoich
gasoline 0.0678–0.0685 propane 0.0638
diesel 0.0690–0.0697 ethanol 0.111
hydrogen 0.0292 methanol 0.155
natural gas 0.058 methane 0.0581

22–11 References

1) Taylor, C.F. The Internal-Combustion Engine in Theory and Practice. Volume 1: Thermodynamics,
Fluid Flow, Performance. Second edition. Cambridge, MA: MIT Press, 1966.
 Chapter 23
Compressor Model

A compressor converts input shaft power to a jet velocity and thrust. The shaft power contributes
to the propulsion group power required. The compressor does not use fuel.
The operating condition is described by pressure altitude, ambient air temperature, flight Mach
number, and either compressor rating or power required. The parametric model is scaled to the required
size and adjusted to the appropriate technology level to represent a notional compressor. Compressor
size is represented by mass flow. Technology is represented by specific power available at maximum
continuous power (MCP), sea level/standard day (SLS), static (zero airspeed) conditions. Installation
effects are also modeled.

23–1 Operating Environment

The operating condition and atmosphere give the standard conditions (temperature Tstd and pressure
pstd ) for a specified pressure altitude; the sea-level standard conditions (temperature T0 and pressure p0 );
and the operating temperature T and pressure p. Here the temperatures are ◦ R or ◦ K. The characteristics
depend on the temperature ratio θ = T /T0 and pressure ratio δ = p/p0 .
√
The flight Mach number M = V /cs = V /cs0 θ is obtained from the aircraft speed V .
The inlet ram air temperature ratio and pressure ratio are obtained then from M and the inlet ram
recovery efficiency ηd :
 
γ−1 2  
θM = 1+ M = 1 + 0.2M 2
2
  γ−1
γ
γ−1  3.5
δM = 1+ ηd M 2 = 1 + 0.2ηd M 2
2

where the ratio of specific heats γ = 1.4.

23–2 Performance Characteristics

The uninstalled power required is Pq , the power available Pa , the gross jet thrust FG , and net jet
thrust FN . The compressor model calculates Pa as a function of flight condition and rating; or calculates
jet thrust and mass flow at Pq . The specific power is SP = P/ṁ, the specific thrust is ST = FG /ṁ. The
forces produced by mass flow that does not go through the core are treated as momentum drag Daux .
The reference performance is at sea-level-standard static conditions (subscript 0), and MCP (sub-
script C ). For each rating R, the performance is characterized by the following quantities for sea-level-
standard static conditions: Power P0R , specific power SP0R , and mechanical power limit PmechR . The
mass flow is then ṁ0R = P0R /SP0R .
210 Compressor Model

The difference between net and gross jet thrust is the momentum drag: FN = FG − ṁreq V . Note
that traditional units for mass flow are pound/sec (pps), while this equation requires slug/sec (ṁreq /g
replaces ṁreq ).

23–3 Installation

The difference between installed and uninstalled power is the inlet and exhaust losses Ploss :
Pav = Pa − Ploss
Preq = Pq − Ploss

The inlet and exhaust losses are modeled as fractions of power available or power required: Ploss =
( in + ex )Pa or Ploss = ( in + ex )Pq . So

Pav = Pa (1 − in − ex )

Pq = Preq /(1 − in − ex )

The momentum drag of the auxiliary air flow is a function of the mass flow ṁaux = faux ṁreq :

Daux = (1 − ηaux )ṁaux V = (1 − ηaux )faux ṁreq V

where ηaux is the ram recovery efficiency.

23–4 Power Available

Given the flight condition and engine rating, the power available Pa is calculated as follows. The
specific power and referred mass flow (relative to SP0 and ṁ0 for this rating) are approximated by
functions of the ambient temperature ratio θ, here just:
"  %Xspa
SPa = SP0 θ δM θM
 √  "  %Xmf a
ṁa = ṁ0 δ/ θ δM θM

The power available is then

 √  "  %Xspa +Xmf a
Pa = SPa ṁa = P0 δ θ δM θM

This expression for ṁa is used only to calculate Pa ; elsewhere the ṁq expression below (for performance
at power required) is used to obtain the mass flow at a power Pq . The influence of compressor rotational
speed is not considered.

23–5 Performance at Power Required

The compressor performance (mass flow and gross jet thrust) is calculated for a specified power
required Pq , flight condition, and rating. Installation losses Ploss are added to Preq (Pq = Preq + Ploss ).
The referred
√
quantities (relative to SLS static MCP quantities) are approximated by functions of q =
Pq /(P0C δ θ):
 √   X
ṁreq = ṁ0C δ/ θ Kmf q0 + Kmf q1 q + Kmf q2 q 2 + Kmf q3 q 3 [θM ] mf q
√ X
STreq = ST0C θ [θM ] stq
Compressor Model 211

The gross jet thrust is then

  X +X
FG = STreq ṁreq = Fg0C δ Kmf q0 + Kmf q1 q + Kmf q2 q 2 + Kmf q3 q 3 [θM ] stq mf q

Then the installed net jet thrust FN and momentum drag Daux are calculated. The influence of compressor
rotational speed is not considered.

23–6 Scaling

The parameters of the compressor model can be defined for a specific compressor, but it is also
necessary to scale the parameters as part of the aircraft sizing task, in order to define a compressor for
a specified power. In addition, advanced technology must be represented in the model. Scaling and
advanced technology are handled in terms of specific power and specific thrust (at SLS static conditions
and MCP).
The compressor model includes reference values of the performance parameters: P0R , SP0R , and
PmechR . Mass flow is obtained from ṁ0R = P0R /SP0R . The reference power at each rating R defines a
ratio to MCP: rp0R = P0R /P0C . Similarly for specific power and mechanical limits: rs0R = SP0R /SP0C
and rm0R = PmechR /P0C . These ratios are kept fixed when the compressor is scaled.
The compressor size is specified as power Peng , which is the power at takeoff rating R, for SLS
static conditions. Hence the MCP power is P0C = Pto /rp0R , and the power at all other ratings follows.
If P0C is not equal to the reference value of the compressor model, then the compressor is scaled.

23–7 Weight

The compressor weight can be a fixed input value, or calculated as a function of power. As a
function of power, the weight of one compressor is:

Wone eng = K0comp + K1comp P + K2comp P Xcomp

where P is the installed power (SLS static, specified rating) per compressor. A constant weight per
power W/P is given by using only K1comp .
212 Compressor Model
 Chapter 24
Motor Model

A motor converts electrical energy (fuel) to shaft power. A generator converts input shaft power to
electrical energy, and the shaft power contributes to the propulsion group power required. The model
follows references 1 to 4.

24–1 Motor Characteristics

The motor or generator size is defined by the maximum power available, Pmax = Peng , and a
peak torque Qpeak . For clarity, the torque limit is expressed as a power limit Ppeak , such that Qpeak =
Ppeak /Nspec . So Ppeak is the torque limit at Nspec , and Ppeak (N/Nspec ) is the torque limit at engine rotation
speed N . The ratio of maximum power and peak torque is the base rotational speed: Pmax /Qpeak = Nbase ;
at Nbase , the power and torque limits coincide. It follows that Nbase = (Pmax /Ppeak )Nspec . Figure 24-
1 shows the maximum power and peak torque for a number of motors. The base rotational speed
is typically Nbase = 700–7000 rpm, lower for high-torque motors. The ratio of maximum power to
continuous power is typically MRP/MCP = 1.25–3 (fig. 24-2).
Motor weight depends primarily on the peak torque. Figure 24-3 shows the weight for a number of
motors. For a given size, there is a wide range of Q/W values. Here high torque-to-weight is defined as
Q/W > 3.5 ft-lb/lb. Motor electrical controller weight is typically 10–30% of the basic motor weight,
up to 70% for small motors. The density is typically 100–250 lb/ft3 (fig. 24-4).
The engine performance is described by the power available Pav at each engine rating, and the
energy flow Ė required to produce the power required Preq . The specific fuel consumption is sfc = Ė/P
(inverse of efficiency). For each rating R, the maximum power is P0R , and the torque limit is PpeakR .

24–2 Power Available and Performance at Power Required

Given the flight condition and engine rating R, the power available is
 
Pav = min P0R , rPpeakR

where r = N/Nspec , and N is the motor rotational speed.

The performance is calculated for a specified power required (which might equal the power avail-
able), flight condition, and engine rating. The motor power required determines the energy flow:

Ėreq = Ė0C ge (q, n) = Preq /ηmotor

where q = Preq /Peng , n = N/Nspec , and ηmotor is the motor efficiency. The generator energy flow to the
fuel tank is related to the power required:

Ėreq = Ė0C ge (q, n) = Preq ηmotor

214 Motor Model

where ηmotor is the generator efficiency.

For a motor and fuel cell, the performance (mass flow and fuel flow) is calculated for a specified
power required:
ẇreq = ẇ0C (q/η) = sfc0C (Preq /η)
ṁreq = ṁ0C (q/η) = Kmf ẇreq
Inlet and exhaust installation losses are included in the specific fuel consumption. The efficiency
η = ηcell ηmotor includes both motor and fuel cell losses. The ratio of mass flow and fuel flow (Kmf )
follows from the chemistry of the reaction. This is a simplified model for the fuel cell. For a better fuel
cell model, use separate motor and fuel cell components, with a battery between to transfer the electrical
energy.

24–3 Efficiency

Motor loss sources include copper (internal resistance, proportional to current-squared hence
torque-squared), iron core (eddy current and hysteresis, proportional to rotational speed), and me-
chanical (friction, proportional to speed, and windage, proportional to speed-cubed). Equivalent circuit
and efficiency map models are implemented. The efficiency can also be a fixed value.

24-3.1 Equivalent Circuit

The motor or generator efficiency as a function of power is estimated considering an equivalent
circuit, defined by internal resistance R and current I0 . The voltage is V = Vo − IR. The efficiency is
ηmotor = P/(P + Ploss ), from the power loss Ploss = I 2 R + P0 . For small loss, I ∼
= P/Vo . In terms of
Peng , let  
1 1
R/Vo2 = −1
Peng ηref
P0 = cPeng
Then  
1 P 1 Peng
= 1 + P (R/Vo2 ) + P0 /P = 1 + −1 +c
ηmotor Peng ηref P
So ηmotor = (1/ηref +c)−1 at P = Peng . The efficiency decreases with P because of the internal resistance,
but is zero at P = 0 because of the internal current term.

24-3.2 Efficiency Map

The motor or generator power loss is described as a polynomial in the motor torque and rotational
speed:

3 
3
Ploss = Peng floss Cij ti nj
i=0 j=0

where q = Preq /Peng , n = N/Nspec , and t = q/n. The factor floss allows adjustment of the peak efficiency.
Controller losses, including power conversion and conditioning, are represented by an efficiency ηcont .
Then
Preq q
ηmotor = ηcont = ηcont
Preq + Ploss q + Ploss /Peng
is the motor or generator efficiency. Constant efficiency implies just C11 = 1/η − 1. The copper, iron,
 
and windage losses imply Ploss = Kc Q2 + Ki N + Kw N 3 + K0 = Peng C20 t2 + C01 n + C03 n3 + C00 .
Motor Model 215

Consider an efficiency map with a peak η0 at t0 = Q0 /Qeng and n0 = N0 /Nspec . Taking the
 
derivatives of C20 t2 + C01 n + C03 n3 + C00 = nt(1 − η)/η with respect to t and n, and evaluating at t0
and n0 so ∂η/∂t = ∂η/∂n = 0, gives

t0 1 − η0 3C00
C01 = −
4 η0 2n0
t0 1 − η0 C00
C03 = 2 + 3
4n0 η0 2n0
n0 1 − η0
C20 =
2t0 η0

(ref. 2). Note that C01 > 0 requires C00 < (n0 t0 /6)(1 − η0 )/η0 .
In general, expanding the efficiency in t and n about the peak gives

1 Ploss Ploss 1 1 
= +1= +1∼
= 1 + a(
t − 1)2 + b(
t − 1)(
n − 1) + c(
n − 1)2
η QN Peng tn η0

where t = t/t0 and n

 = n/n0 . Hence
   
Ploss 1 1 
= tn −1 = tn 1 + a(
t − 1)2 + b(
t − 1)( n − 1)2 − 1 + d(
n − 1) + c( t − 1)2 (
n − 1)2
Peng η η0

The last term is introduced so the efficiency is zero at zero torque or zero speed.
Alternatively, consider an equivalent circuit defined by internal resistance R, no load current I0 ,
motor constant Kv , and friction Kf . So V = Vm + IR (back emf Vm = N/Kv ), I = Im + I0 (Im =
Qm N/Vm = Kv Qm ), Q = Qm −Qf (output torque, Qf = Kf N ). The input power is V I = (N/Kv +IR)I ,
with I = Kv Qm + I0 = Kv (Q + Kf N ) + I0 . The shaft power is QN . Then the power loss is
 
Ploss = V I − QN = N + (I/Kv )Kv2 R (I/Kv ) − QN
   2
= N Q + Kf N + I0 /Kv + Q + Kf N + I0 /Kv Kv2 R − QN
   2
= N Kf N + I0 /Kv + Q + Kf N + I0 /Kv Kv2 R

This expression does not show a peak efficiency as a function of torque and speed.

24–4 Weight

The motor or generator weight can be a fixed input value, or calculated as a function of power or
torque. As a function of power, the weight of one engine is:

Wone eng = K0motor + K1motor P + K2motor P Xmotor QXqmotor

where P = Peng , and Q = PpeakR /Nspec (ft-lb or m-N) for takeoff rating R. Note then P X QXq =
P X+Xq N −Xq = N X QX+Xq .

Motor weight depends primarily on the peak torque. For the NASA15 model, the weight is:

Wone eng = 0.5382fdesign Q0.8129

216 Motor Model

where Q = PpeakR /Nspec (ft-lb) for takeoff rating R. The structural design factor fdesign = 1.0 for high
torque-to-weight motors (Q/W > 3.5 ft-lb/lb), and fdesign = 2.5606 for others. Based on 64 motors, the
average error is 27.5% (fig. 24-5); 25% for high torque-to-weight and 29% for others. Including either
maximum rotational speed or maximum power does not reduce the weight estimation error significantly.
Considering only the high torque-to-weight motors, the weight is
Wone eng = 0.3928Q0.8587
Based on 25 motors, the average error is 21.8% (fig. 24-5). This sensitivity of motor weight to output
torque capability is somewhat greater than the drive system trend.

24–5 Scaling

The parameters of the motor model can be defined for a specific motor, but it is also necessary to
scale the parameters as part of the aircraft sizing task, in order to define a motor for a specified power.
In addition, advanced technology must be represented in the model.
The motor model includes reference values of the performance parameters: P0R , PpeakR . The
reference power at each rating R defines a ratio to MCP: rp0R = P0R /P0C . Similarly for torque limits:
rm0R = PpeakR /P0C . These ratios are kept fixed when the motor is scaled. The specification motor
speed and base rotational speed are scaled with the power:
 KN s
Pref
Nspec = Nspec−ref
Peng
 KN b
Pref
Nbase = Nbase−ref
Peng
The speed ratio N/Nspec influences the efficiency map and the torque limit. Then PpeakR scales as
 KN s −KN b
Peng Nspec /Nspec−ref Pref
PpeakR = PpeakR−ref = rm0R Peng
Pref Nbase /Nbase−ref Peng
So PpeakR just scales with Peng if KN s = KN b . Depending on the design approach, KN b ranges from
about 0.8 to zero (constant Nbase ) to about −1.2 (Nbase increase with power).
The motor size is specified as power Peng , which is the power at takeoff rating R, for sea level
standard (SLS) static conditions. Hence the MCP power is P0C = Pto /rp0R , and the power at all other
ratings follows. If P0C is not equal to the reference value of the motor model, then the motor is scaled.

24–6 References

1) McDonald, R.A. “Electric Motor Modeling for Conceptual Aircraft Design.” AIAA Paper No.
2013-0941, January 2013.
2) McDonald, R.A. “Electric Propulsion Modeling for Conceptual Aircraft Design.” AIAA Paper No.
2014-0536, January 2014.
3) Sinsay, J.D.; Alonso, J.J.; Kontinos, D.A.; Melton, J.E.; and Grabbe, S. “Air Vehicle Design
and Technology Considerations for an Electric VTOL Metro-Regiional Public Transportation System.”
AIAA Paper No. 2012-5404, September 2012.
4) Datta, A., and Johnson, W. “Powerplant Design and Performance Analysis of a Manned All-Electric
Manned Helicopter.” Journal of Propulsion and Power, Vol. 30, No. 2 (March-April 2014).
Motor Model 217

100000.

high Q/W = 100

other e rpm
bas
10000.
peak torque (ft-lb)

0
100

1000.

00
100

100.

10.
1. 10. 100. 1000.
maximum power (hp)

Figure 24-1. Motor maximum power and peak torque.

5.

high Q/W
maximum power / continuous power

other
4.

3.

2.

1.

0.
1. 10. 100. 1000.
maximum power (hp)

Figure 24-2. Motor maximum power and continuous power (MRP/MCP).

218 Motor Model

10000.

high Q/W
other
1000.
weight (lb)

100.

10.

1.
10. 100. 1000. 10000. 100000.
peak torque (ft-lb)

Figure 24-3. Motor weight.

350.

high Q/W
300.
other

250.
density (lb/ft3)

200.

150.

100.

50.

0.
10. 100. 1000. 10000. 100000.
peak torque (ft-lb)

Figure 24-4. Motor density.

Motor Model 219

high Q/W
other
80.
only high Q/W
60.

40.

20.
error (%)

0.

-20.

-40.

-60.

-80.
1. 10. 100. 1000. 10000.

actual weight

Figure 24-5. Motor weight (NASA15).

220 Motor Model
 Chapter 25

Referred Parameter Jet Engine Model

Design and analysis codes calculate aircraft performance for a very wide range of operating con-
ditions. Thus the jet performance model must be realistic even far from the design point. A simple
thermodynamic cycle analysis that assumes design point component efficiencies everywhere is not real-
istic for such an application. Rather than developing models for component performance, the approach
taken is to use a model for the total turbojet or turbofan performance. The jet is not being designed.
The Referred Parameter Jet Engine Model (RPJEM) is based on curve-fits of performance data for
existing or projected jets over a range of operating conditions. The use of referred parameters tends
to collapse the data and provides a basis for scaling the jet. The operating condition is described by
pressure altitude, ambient air temperature, flight Mach number, and either jet rating or jet thrust required.
The parametric model is scaled to the required size and adjusted to the appropriate technology level to
represent a notional jet. Jet size is represented by mass flow. Jet technology is represented by specific
thrust available and specific fuel consumption at maximum continuous thrust (MCT), sea level/standard
day (SLS), static (zero airspeed) conditions. Jet installation effects are also modeled.

25–1 Operating Environment

The operating condition and atmosphere give the standard conditions (temperature Tstd and pressure
pstd ) for a specified pressure altitude; the sea-level standard conditions (temperature T0 and pressure p0 );
and the operating temperature T and pressure p. Here the temperatures are ◦ R or ◦ K. The characteristics
depend on the temperature ratio θ = T /T0 and pressure ratio δ = p/p0 .
√
The flight Mach number M = V /cs = V /cs0 θ is obtained from the aircraft speed V .
The inlet ram air temperature ratio and pressure ratio are obtained then from M and the inlet ram
recovery efficiency ηd :
 
γ−1 2  
θM = 1+ M = 1 + 0.2M 2
2
  γ−1
γ
γ−1  3.5
δM = 1+ ηd M 2 = 1 + 0.2ηd M 2
2

where the ratio of specific heats γ = 1.4.

25–2 Performance Characteristics

The uninstalled thrust required is Tq , the thrust available Ta . The jet model calculates Ta as a
function of flight condition and engine rating; or calculates mass flow and fuel flow at Tq . The specific
thrust is ST = T /ṁ, and the specific fuel consumption is sfc = ẇ/T . A turbofan engine has a bypass
222 Referred Parameter Jet Engine Model

ratio β = ṁfan /ṁ. The forces produced by mass flow that does not go through the core or fan are treated
as momentum drag Daux .
The reference performance is at sea-level-standard static conditions (subscript 0), and MCT (sub-
script C ). For each rating R, the performance is characterized by the following quantities for sea-level-
standard static conditions: thrust T0R , specific thrust ST0R , and mechanical thrust limit TmechR . The
mass flow is then ṁ0R = T0R /ST0R .
The difference between net and gross thrust is the momentum drag: TN = TG − ṁ(ST )mom ,
where (ST )mom = (1 + β)V for turbojet or turbofan, or (ST )mom = Ωrreact for reaction drive. Note
that traditional units for mass flow are pound/sec (pps), while this equation requires slug/sec (ṁreq /g
replaces ṁreq ).

25–3 Installation

The difference between installed and uninstalled thrust is the inlet and exhaust losses Tloss : Tav =
Ta − Tloss and Treq = Tq − Tloss . The inlet and exhaust losses are modeled as fractions of thrust available
or thrust required: Tloss = ( in + ex )Ta or Tloss = ( in + ex )Tq . The momentum drag of the auxiliary
air flow is a function of the mass flow ṁaux = faux ṁreq :

Daux = (1 − ηaux )ṁaux V = (1 − ηaux )faux ṁreq V

where ηaux is the ram recovery efficiency. Exhaust losses ( ex ) and auxiliary air flow parameters (ηaux ,
faux ) are defined for infrared suppressor on and off.

25–4 Thrust Available

Given the flight condition and jet rating, the thrust available Ta is calculated as follows. The gross
specific thrust and referred mass flow (relative to ST0 and ṁ0 for this rating) are approximated by
functions of the ambient temperature ratio θ, here just:
√ "  %Xsta
STa = ST0 θ δM θM
 √  "  %Xmf a
ṁa = ṁ0 δ/ θ δM θM

The thrust available is then

"  %Xsta +Xmf a
Ta = STa ṁa − ṁa (ST )mom = T0 δ δM θM − ṁa (ST )mom

This expression for ṁa is used only to calculate Ta ; elsewhere the ṁq expression below (for performance
at thrust required) is used to obtain the mass flow at a thrust Tq .

25–5 Performance at Thrust Required

The jet performance (mass flow and fuel flow) is calculated for a specified thrust required Tq , flight
condition, and jet rating. The referred quantities (relative to SLS static MCT quantities) are approximated
by functions of referred gross thrust t = (Tq + ṁ(ST )mom )/(T0C δ):
 √   −X
ẇreq = ẇ0C δ θ Kf f q0 + Kf f q1 t + Kf f q2 t2 [θM ] f f q
 √ 
X
ṁreq = ṁ0C δ/ θ tKmf q [θM ] mf q
Referred Parameter Jet Engine Model 223

The mass flow solution is

⎧
ṁreq ⎨1 Kmf q = 0
√ = b/(1 − a)  Kmf q = 1
X
ṁ0C (δ/ θ) [θM ] mf q ⎩
−(a/2) + (a/2)2 + b Kmf q = 1/2

√
where a = (ST )mom ṁ0C [θM ]Xmf q /T0C θ and b = Tq /T0C δ .

25–6 Scaling

The parameters of the jet model can be defined for a specific turbojet or turbofan, but it is also
necessary to scale the parameters as part of the aircraft sizing task, in order to define a jet for a specified
thrust. In addition, advanced technology must be represented in the model. Scaling and advanced
technology are handled in terms of specific thrust and specific fuel consumption (at SLS static conditions
and MCT). Figures 25-1 through 25-3 present historical data for jet specific fuel consumption, weight,
and specific thrust.

The jet model includes reference values of the performance parameters: T0R , ST0R , TmechR , and
sfc0C . Mass flow and fuel flow are obtained from ṁ0R = T0R /ST0R and ẇ0C = sfc0C T0C . The reference
thrust at each rating R defines a ratio to MCT: rt0R = T0R /T0C . Similarly for specific thrust and
mechanical limits: rs0R = ST0R /ST0C and rm0R = TmechR /T0C . These ratios are kept fixed when the
jet is scaled.

The jet size is specified as takeoff thrust Tto = Teng , which is the thrust at rating R, for SLS static
conditions. Hence the MCT is T0C = Tto /rt0R , and the thrust at all other ratings follows. If T0C is not
equal to the reference value of the jet model, then the jet is scaled. To reflect advanced technology,
the specific thrust and specific fuel consumption can be specified: ST0C = STtech and sfc0C = sfctech
(replacing the jet model reference values). The default values are the reference values of the jet model.
In the following paragraph, the subscript “tech” refers to these quantities; the subscript “ref ” means the
jet model reference values.

The jet technology parameters ST0C and sfc0C are assumed to vary linearly with mass flow ṁ0C up
to a limit ṁlim , and constant thereafter at STlim and sfclim . The mass flow at the technology condition
is ṁtech = Tref /STtech , with the technology values STtech and sfctech . The intercept values are projected
from the technology values: Kst0 = STtech − Kst1 ṁtech , Kst1 = (STlim − STtech )/(ṁlim − ṁtech ); and
similarly for sfc. Then for ṁ0C < ṁlim

ST0C = Kst0 + Kst1 ṁ0C

sfc0C = Ksf c0 + Ksf c1 ṁ0C

and for ṁ0C ≥ ṁlim

ST0C = STlim
sfc0C = sfclim

These equations are used if ṁlim > ṁtech . Otherwise the model sets Kst1 = Ksf c1 = 0, so there is no
variation with scale: ST0C = STtech and sfc0C = sfctech . Usually the effect of size gives Kst2 ≥ 0 and
Ksf c2 ≤ 0. The thrust at the limit is Tlim = STlim ṁlim . Figure 25-4 illustrates the scaling of SP , sfc, and
SW .
224 Referred Parameter Jet Engine Model

1.4

1.2
specific fuel consumption (lb/lb-hr)

1.0

0.8

0.6

0.4

0.2

0.0
0. 20000. 40000. 60000. 80000. 100000.
takeoff thrust (lb)

Figure 25-1. Historical data for turbojet and turbofan specific fuel consumption.

0.6

0.5

0.4
weight (lb/lb)

0.3

0.2

0.1

0.0
0. 20000. 40000. 60000. 80000. 100000.

takeoff thrust (lb)

Figure 25-2. Historical data for turbojet and turbofan weight.

Referred Parameter Jet Engine Model 225

80.

ST (lb/lb/sec) 60.

40.

20.

0.
0. 20000. 40000. 60000. 80000. 100000.
takeoff thrust (lb)

Figure 25-3. Historical data for turbojet and turbofan specific thrust.

Using ṁ0C = T0C /ST0C , the specific thrust equation can be solved for the mass flow given the
thrust: ⎧
⎪
⎪
T0C /Kst0 Kst1 = 0
⎪
⎨
ṁ0C = T0C /STlim T0C ≥ Tlim
⎪
⎪ 
⎪
⎩ ( K0 )2 +
2K1
T0C
K1 − K0
2K1 otherwise
From this mass flow, ST0C and sfc0C are calculated, hence the fuel flow ẇ0C = sfc0C T0C . Then the thrust
and specific thrust at all ratings R are obtained from the ratios: T0R = rt0R T0C , ST0R = rs0R ST0C , and
TmechR = rm0R T0C .

The actual (perhaps scaled) values of the performance parameters are available for the jet group:
T0R , ST0R , TmechR , and sfc0C .

25–7 Weight

The jet weight can be a fixed input value, or calculated as a function of thrust. As a function of
thrust, the weight of one jet is:

Wone jet = K0jet + K1jet T + K2jet T Xjet

where T is the installed takeoff thrust (SLS static, specified rating) per jet. A constant weight per thrust
W/T is given by using only K1jet .
226 Referred Parameter Jet Engine Model

. . . .
m lim> mtech m lim≤ mtech

lim lim
ST0C ST0C
tech tech

tech lim lim tech

. .
m 0C m 0C

sfc 0C tech sfc 0C tech

lim lim

tech lim lim tech

. .
m 0C m 0C

lim lim
SW SW
ref ref

tech lim lim tech

. .
m 0C m 0C

Figure 25-4. Turbojet or turbofan scaling; ST = T /ṁ, sfc = ẇ/T , and SW = T /W .

25–8 Units

In this jet model, only the reference values and scaling constants are dimensional. Conventional
English units and SI units are shown in table 25-1. Units of specific thrust and specific fuel consumption
follow from these conventions.

Table 25-1. Conventional units.

thrust T mass flow ṁ fuel flow ẇ
English: pound pound/sec pound/hour
SI: Newton kilogram/sec kilogram/hour
 Chapter 26

Fuel Cell Model

A fuel cell burns a fuel (typically hydrogen) and generates electrical energy, which is stored in a
fuel tank system or used directly by a motor. The energy flow defines the power required. The power
available is related to the size Pchrg . The model follows reference 1.

26–1 Operating Environment

The operating condition and atmosphere give the standard conditions (temperature Tstd and pressure
pstd ) for a specified pressure altitude; the sea-level standard conditions (temperature T0 and pressure
p0 ); and the operating temperature T and pressure p. Here the temperatures are ◦ R or ◦ K. The engine
characteristics depend on the temperature ratio θ = T /T0 and pressure ratio δ = p/p0 .
√
The flight Mach number M = V /cs = V /cs0 θ is obtained from the aircraft speed V .

The inlet ram air temperature ratio and pressure ratio are obtained then from M and the inlet ram
recovery efficiency ηd :
 
γ−1 2  
θM = 1+ M = 1 + 0.2M 2
2
  γ−1
γ
γ−1  3.5
δM = 1+ ηd M 2 = 1 + 0.2ηd M 2
2

where the ratio of specific heats γ = 1.4.

26–2 Fuel Cell Performance

The performance of a fuel cell is characterized by the single cell behavior: voltage vc as a function
of current density ic . The corresponding power density is pc = vc ic . The maximum possible voltage
is the cell ideal reversible voltage Eh , obtained when all the energy available is converted to electrical
energy. The efficiency of the cell is η = vc /Eh . From the fuel specific energy efuel (MJ/kg), the specific
fuel consumption is then sfc = ẇ/P = efuel /η = efuel Eh /vc .

Figure 26-1 shows the typical performance of a proton exchange membrane fuel cell. For hydrogen,
Eh = 1.472 volts. The voltage is reduced by activation losses (low current), Ohmic losses, and mass
transport losses (high current). The performance depends on the cell pressure δc , increasing as the
operating pressure increases. A voltage reduction corresponds to a reduction in efficiency. There is a
maximum power density pmax at vmax , where dvc /dic = −vc /ic . The system is designed to operate at a
current below the maximum power.
228 Fuel Cell Model

The fuel cell system consists of nc individual cells in series, with total area A. Then V = nc vc ,
I = Aic , and the total power is P = V I = nc Apc . The total fuel flow is proportional to the current,
ẇ = sfcP = efuel Eh nc Aic .

Consider a fuel cell with maximum continuous power PD (MCP), designed for current density
id , from which vd and pd are determined. The design point is sea level static conditions, nominal
temperature, and design cell pressure. The controllers, wiring, compressor, and other subsystems are
designed for this current, so the total system weight will depend on PD . As id decreases, less fuel is
used (larger efficiency) but the weight is more (larger nc A). The fuel cell can operate for a short time at
a higher power (MRP). The power density pc = (MRP/MCP)pd gives the corresponding current density
iMRP . There is also an absolute maximum on the power. The cell efficiency at the design point is
ηD = vd /Eh , hence sfcD = (efuel Eh /vd )/ηchrg and the fuel flow is ẇD = sfcD PD . The system specific
fuel consumption can include additional losses in terms of an efficiency ηchrg .
The cell pressure depends on the atmospheric pressure ratio δ , inlet ram pressure ratio δM , and
compressor pressure ratio πC : δc = δδM πC . It is assumed that the cell temperature is maintained at
the design value. The cell performance vc (ic ) is required at the design conditions (δc = πC ) and at the
operating conditions of the mission. The performance dependence on cell pressure can be captured by
scaling the current with δcXf c : given the reference characteristics vref (iref ) for δc = 1,

Xf c
vc = vref (iref = ic /δc )

Typically Xf c = 0.38; increasing δc then reduces iref , increasing vc and the efficiency. This scaling leaves
the voltage at maximum power unchanged. The reference characteristics are only needed to maximum
power. From nominal performance at the design pressure ratio, δnom = πC , the reference characteristics
Xf c
are vref = vnom at iref = inom /δnom .
The ratio of mass flow and fuel flow follows from the chemistry of the reaction. For hydrogen and
air
λ A mA λA
Kmf = = 68.59
λH xO mH λH
The molar masses of hydrogen and air are mH = 2.016 and mA = 28.97 g/mole; xO = 0.2095 is the molar
fraction of oxygen in air. The supply ratio λA /λH = 1/2 from stoichiometry, and typically λA /λH ∼
= 1.25
in practice.
The fuel cell is described by the static power P0R at rating R, the MCP specific fuel consumption
sfc0C , the design current id and compressor pressure ratio πC , the mass flow ratio Kmf , and the cell
characteristics vref (iref ). The cell characteristics are only used as ratios to the design values.

26–3 Power Available

Given the flight condition and charger rating, the power available Pav = Ėacell is
pa
Pav = P0C
pd

where pa is obtained from the rated current iR . From vR = vref (iref = iR /δcXf c ), pR = vR iR . The
maximum power density pmax = vmax imax is obtained from imax = δcXf c iref (vref = vmax ). Then pa =
min(pR , pmax ).
Fuel Cell Model 229
26–4 Performance at Power Required

The fuel cell performance (fuel flow and mass flow) is calculated for a specified power required
Preq = Ėqcell and flight condition. The current iq required at pq = pd (Preq /P0C ) is obtained from
X
pc = ic vref (iref = ic /δc f c ) as a function of ic . Then

ẇreq = ẇ0C (iq /id )

ṁreq = Kmf ẇreq

where ẇ0C = sfc0C P0C .

26–5 Simple Model

Constant vref implies constant efficiency and specific fuel consumption, and so pc proportional to
ic with no maximum power density. Then the rated power available and specific fuel consumption are
constant: Pav = P0R , ẇreq = ẇ0C (Preq /P0C ) = sfc0C Preq , ṁreq = Kmf ẇreq .

26–6 Weight

The fuel cell weight can be a fixed input value, or calculated as a function of power. As a function
of power, the weight of one charger is:

Wone chrg = K0cell + K1cell P + K2cell P Xcell

where P is the installed power (SLS static, specified rating) per charger. A constant weight per power
W/P is given by using only K1cell .

26–7 References

1) Datta, A., and Johnson, W. “Powerplant Design and Performance Analysis of a Manned All-Electric
Manned Helicopter.” Journal of Propulsion and Power, Vol. 30, No. 2 (March-April 2014).
230 Fuel Cell Model

1.0

0.8
voltage vc (V)

0.6

0.4 1 atm
2 atm
3 atm
0.2
4 atm

0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2

0.5

0.4
power density pc (W/cm2)

0.3

0.2

0.1

0.0
0.0 0.2 0.4 0.6 0.8 1.0 1.2
current density ic (A/cm2 )

Figure 26-1. Characteristic performance of a proton exchange membrane fuel cell.

 Chapter 27

Solar Cell Model

A solar cell generates electrical energy, which is stored in a fuel tank system. The energy flow
defines the power required. The power available is related to the size Pchrg . The solar cell is characterized
by power density esolar (W/m2 ) and weight density σsolar (kg/m2 ). From the size Pchrg , the area is
Asolar = Pchrg /esolar ; and then the weight is Wone chrg = Wsolar = Asolar σsolar .

Given the flight condition and the charger rating, the cell power available is Pav = Ėacell = P0 . The
power required Preq is defined by the charge group energy flow. The cell power required is

Ėqcell = P0C ge (q) = Preq /ηchrg

where q = Preq /P0C , and ηchrg is the solar cell efficiency.

The solar cell efficiency as a function of power is estimated considering an equivalent circuit,
defined by internal resistance R and current I0 . The voltage is V = Vo − IR. The efficiency is
ηchrg = P/(P + Ploss ), from the power loss Ploss = I 2 R + P0 . For small loss, I ∼
= P/Vo . In terms of Pchrg ,
let  
1 1
R/Vo2 = −1
Pchrg ηref
P0 = cPchrg
Then  
1 P 1 Pchrg
= 1 + P (R/Vo2 ) + P0 /P = 1 + −1 +c
ηchrg Pchrg ηref P
So ηchrg = (1/ηref +c)−1 at P = Pchrg . The efficiency decreases with P because of the internal resistance,
but is zero at P = 0 because of the internal current term. Alternatively, the efficiency can be a fixed
value.
A simple model requires the power density esolar and weight density σsolar , with efficiency included
in esolar (ηchrg = 1).
232 Solar Cell Model
 Chapter 28

Battery Model

A battery is a fuel tank system for which the fuel quantity stored and burned is measured in energy.
The unit of fuel energy is Mega-Joules (MJ). For reference, 1 kW-hr = 3.6 MJ. The operating state affects
the efficiency of the relation between useful power and the rate of change of the energy stored. The
battery model produces the charge/discharge efficiency. The battery model can be used for capacitors
and flywheels as well. References 1–6 provide background for the model.
The components associated with a fuel tank system (motor, generator, fuel cell, and solar cell)
define the total energy flow Ėcomp (charge or discharge). Accounting for efficiency and losses gives the
battery energy flow Ėbatt . Accounting for battery capacity then gives the effective energy flow Ėeff . The
change in stored energy is calculated from Ėeff and time. The convention is that energy flow Ė > 0 for
discharge, and Ė < 0 for charge. Power and current are positive for both discharge and charge.
The battery capacity is Efuel−cap (maximum usable fuel energy). The battery is characterized
by specific energy etank (MJ/kg) and energy density ρtank (MJ/liter), so the tank weight and volume
are obtained from the capacity. The current amount of energy stored is Efuel . The state-of-charge is
s = Efuel /Efuel−cap ; the depth-of-discharge is d = 1 − s.

The charge capacity is C (A-hr). C is the usable capacity at low current and a reference temperature.
Typically 20–30% of the stored charge is not usable at nominal conditions, which is accounted for in
the capacity through the specific energy and energy density values. The corresponding energy capacity
is obtained for a reference voltage, E = CVref (W-hr, expresssed in MJ). Then the specific energy
(conventional units W-hr/kg) gives the weight, and the energy density (conventional units W-hr/liter)
gives the volume. The current is measured in terms of the charge capacity: I = xC (A), corresponding
to a discharge time of 1/x hours. The units of x are 1/hr. The maximum burst discharge current (mbd)
is xmbd .
The usable capacity decreases with time and duty cycles, depending on the temperature and current.
This factor is assumed to be constant during a mission or flight condition. Capacity fade is accounted
for using a factor ffade < 1 on the available capacity or on the energy flow. Thus the effective energy
flow is obtained by dividing the actual energy flow by ffade .
The capacity depends on the discharge current. The Peukert model assumes I k T = constant, where
T is the discharge time for current I = xCref . The Peukert coefficient k = 1.2 to 1.3 for lead-acid batteries,
and k = 1.01 to 1.05 for lithium-ion batteries (weak dependence). Thus the capacity C = IT ∼ 1/xk−1 .
An increase in current by a factor of 10 means a 2–11% reduction of capacity for lithium-ion batteries.
For a larger current, the battery reaches a specified discharge voltage sooner, hence effectively has a
reduced capacity.
234 Battery Model

The specific power is πbatt (kW/kg). The power capacity is obtained at the maximum burst discharge
current:
Pcap = Imbd Vref = xmbd CVref = xmbd Efuel−cap

(for Pcap in W and Efuel−cap in W-hr). Hence given xmbd , πbatt = xmbd etank /3.6 (kW/kg from MJ/kg).
Typically xmbd is smaller for large capacity, implying a trade between high specific power and high
specific energy.

28–1 Equivalent Circuit Model

The discharge or charge efficiency as a function of power is estimated considering an equivalent

circuit, defined by internal resistance R and current I0 . The voltage is V = Vo − IR. The efficiency is
ηbatt = P/(P + Ploss ), from the power loss Ploss = I 2 R + P0 . For small loss, I ∼
= P/Vo . In terms of a
reference power Pref = Pcap , and battery power P = |Ėcomp |, let
 
1 1
R/Vo2 = −1
Pref ηref
P0 = cPref

Then  
1 P 1 Pref
= 1 + P (R/Vo2 ) + P0 /P = 1 + −1 +c
ηbatt Pref ηref P
−1
So ηbatt = (1/ηref + c) at P = Pref . The efficiency decreases with P because of the internal resistance,
but is zero at P = 0 because of the internal current term. Alternatively, the efficiency can be a fixed value.
The current is x = I/C ∼ = P/V0 C ∼= xmbd P/Pref = xmbd |Ėcomp |/Pcap . Given the maximum current xmax ,
the maximum battery power is Pmax = (xmax /xmbd )Pcap .
For discharge, the battery energy flow is Ėbatt = Ėcomp /ηbatt = Ėcomp + Ploss . Then the effective
energy flow is Ėeff = Ėbatt /ffade . For charge (negative energy flow), the battery energy flow is Ėbatt =
Ėcomp ηbatt = Ėcomp + Ploss . Then the effective energy flow is Ėeff = Ėbatt /ffade . The battery power
margin is Pmax − |Ėbatt |.

28–2 Lithium-Ion Battery Model

The capacity and power of currently available lithium-ion batteries are shown in figure 28-1. The
power shown corresponds to the maximum burst discharge current, Pcap = xmbd Efuel−cap . The cell
reference voltage is typically Vref = 4.2 V. The maximum continous discharge current is in the range
xmcd = 1–30 (fig. 28-2). The maximum burst discharge current (which equals the power divided by
capacity) is in the range xmbd = 10–50 (fig. 28-3), with xmbd ∼ = 10 for large capacity. Values of xmbd
above 100 are achievable, but such high-power/short-duration capability will not have much application
to aircraft. The maximum charge current is typically xCCmax = 1–5 (fig. 28-4). Figure 28-5 shows
the corresponding specific energy etank and specific power πbatt = xmbd etank . The specific energy is
typically etank = 50–200 W-hr/kg = 0.2–0.7 MJ/kg. Figure 28-6 shows the specific energy etank (W-
hr/kg) and the energy density ρtank (W-hr/liter). The battery density is 2–3 kg/liter for all configurations
(cylindrical, pouch, prismatic).
The battery temperature is likely controlled, for safety and efficiency. The operating temperature
depends on the environment ambient temperature, internal heat production, and active heating or cooling.
Internal heat production includes that due to internal resistance (I 2 R), and radiation and convection
Battery Model 235

(roughly proportional to the difference between cell and ambient temperatures). The cell temperature
Tc is defined for a flight condition, or at the start of a mission. Without active temperature control,
during a mission the cell temperature changes: dTc /dt is proportional to the internal heat production.
With temperature control, there is a power loss associated with the control system, depending on the cell
temperature relative to ambient, hence on the internal heat production. As a simple model, it is assumed
that this loss is a fraction fT C of the total power.

28-2.1 Discharge Characteristics

The lithium-ion battery voltage during discharge depends on the discharge capacity (depth-of-
discharge d), current (I = xC ), and cell temperature (Tc ). Figure 28-7 shows typical variation of the
voltage V at constant current and constant temperature. The nominal energy capacity E = CVref is
obtained from the nominal charge capacity C and a reference voltage Vref . The open-circuit voltage Vd
is the voltage at zero current, a reference temperature Tref , and 100% charge. Typically the cell voltage
is Vd = 4.2 V, and Vd is used as the reference Vref . The actual battery capacity is determined by the
voltage reaching a critical value Vcrit (typically 2–3 V). Usually V decreases gradually with d, so the
voltage provides a measure of depth-of-discharge. The decrease of V with current is primarily due to
internal resistance. The voltage also decreases as temperature decreases (fig. 28-7). There is a maximum
discharge current (continuous or burst), which gives the rated power: Pcap = Imbd Vref = xmbd CVref . The
actual battery power supplied is P = IV .

The voltage variation with current is primarily due to the internal resistance: V = Vo − IR. The
open circuit voltage decreases with depth-of-discharge: Vo = Vd FV (d), such that FV = 1 at d = 0
and FV = fcrit = Vcrit /Vd at d = 1; and Vd = fd Vref . For large discharge (d approaching 1), there
is a secondary influence of current, which can be modelled by adjusting the discharge by a factor k.
Then the capacity is given by FV (kd) = fcrit . Variations of the voltage with current are scaled using
I = xC = (xmbd C)(x/xmbd ). The influence of temperature on the open circuit voltage Vo is reasonably
accounted for by an increment proportional to the cube of the temperature difference ΔT = Tc − Tref .

The model for the discharge voltage is

xmbd CR
V = Vd FV (kd) + kV T ΔT 3 − IR = Vd FV (kd) + kV T ΔT 3 − Vref (x/xmbd )
Vref

with
k = 1 + kdI I − kdT ΔT = 1 + (kdI xmbd C) (x/xmbd ) − kdT ΔT

The open circuit voltage function FV (d) is derived from V (d) as a function of I and Tc : FV = (V −
kV T ΔT 3 + IR)/Vd at kd. The model parameters R, kV T , kdI , and kdT are adjusted for a good fit to the
measured discharge characateristics. Figure 28-8 shows FV (d) obtained for a number of lithium-ion
batteries. Typically fcrit ∼
= 0.6. That d = 1 at FV = fcrit reflects the definition of nominal capacity for
the battery characteristics. Table 28-1 gives a typical FV (d) function. The function extends below fcrit
since high temperature can increase the voltage. Figure 28-9 shows the current parameters xmbd CR/Vref
and kdI xmbd C for a number of batteries. Generally xmbd CR/Vref = 0.05–0.20 and kdI xmbd C = 0–0.25.
Figure 28-10 shows the temperature parameters kV T and kdT for a number of batteries.

The battery power available to the components is

 
Ėcomp = P = IV = I Vd FV (kd) + kV T ΔT 3 − I 2 R − PT C = Ėbatt − Ploss
236 Battery Model

including power loss for temperature control. Hence

 
Ėbatt = Pcap ξ fd FV (kd) + (kV T /Vref )ΔT 3
/ 2 + fT C |Ėcomp |
Ploss = Pcap Rξ
k =1+/
kdI ξ − kdT ΔT

where ξ = x/xmbd , R/ = xmbd CR/Vref , and /kdI = kdI xmbd C . The battery capacity for a given current
and temperature is obtained by finding kdcrit such that

V = Vd FV (kdcrit ) + kV T ΔT 3 − IR = Vd fcrit

or
/ − (kV T /Vref )ΔT 3 )/fd
FV (kdcrit ) = fcrit + (Rξ

Then dcrit = (kdcrit )/k, and the effective capacity is dcrit ffade Efuel−cap . Rather than change the fuel tank
capacity, an effective fuel flow is used: Ėeff = Ėbatt /(dcrit ffade ).
For a flight condition or mission segment, the total energy flow required is Ėcomp . The temperature
Tc is specified. The depth-of-discharge is d = 1 − s = 1 − Efuel /Efuel−cap . The discharge current
x = xmbd ξ is calculated from the required energy flow:

(Ėcomp + Ploss )/Pcap

ξ=
fd FV (kd) + (kV T /Vref )ΔT 3

Successive-substitution solution for ξ , with fcrit as the minimum for FV , converges in about 10 iterations.
Since FV is a monotonically decreasing function of d, dcrit is obtained from x by interpolating d(FV ).
Then Ėbatt = Ėcomp + Ploss , and the battery efficiency is ηbatt = Ėcomp /Ėbatt . The effective energy flow
is Ėeff = Ėbatt /(dcrit ffade ). The maximum battery discharge power is Pmax = Ėbatt evaluated at xmbd ,
or at a specified discharge current. The battery power margin is Pmax − |Ėbatt |.

Table 28-1. Lithium-ion battery open circuit voltage.

depth-of-charge d FV depth-of-charge d FV
0.00 1.000 0.92 0.842
0.10 0.970 0.93 0.835
0.20 0.950 0.94 0.826
0.30 0.930 0.95 0.815
0.40 0.915 0.96 0.800
0.50 0.900 0.97 0.780
0.60 0.890 0.98 0.750
0.70 0.880 0.99 0.700
0.80 0.870 1.00 0.600
0.90 0.850 1.01 0.400
0.91 0.847 1.02 0.000

28-2.2 Charge Characteristics

Lithium-ion battery charging is controlled for safety, with limits on minimum and maximum voltage,
maximum current, and minimum and maximum temperature. Standard charging procedure consists of
Battery Model 237

constant current (CC) followed by constant voltage (CV). Figure 28-11 shows the voltage, current, and
capacity variation with charging time. The charging current is ICC = xCC C ; the maximum charging
current is typically xCCmax = 1–5 (fig. 28-4). The charging voltage is Vc = fc Vref ; typically Vc = Vref = 4.2
V. The state-of-charge is s.
During the CC phase, the voltage starts at a fraction kcV (typically 0.05–0.20) below Vc and increases
with time, reaching Vc at time tc = σ/xCC . The charging model assumes linear variation of the voltage.
During the CV phase, the voltage is fixed at Vc , while the current decreases rapidly. The charging model
assumes the current varies inversely with the square of the time increment since tc . The charge is then
!t
the integral of the current, Q = 0 I dt; and the state-of-charge s = Q/C . The charging power is P = IV .
Table 28-2 gives the charging model as a function of normalized time τ = t/tc − 1 = txCC /σ − 1.

Table 28-2. Lithium-ion battery charging model.

CC phase, t < tc , τ = −1 to 0, s < σ CV phase, t > tc , τ > 0, s > σ
xCC C
current I = xC = xCC C I = xC =
(1 + kcI τ )2
voltage V = Vc (1 + kcV τ ) V = Vc
1 + (kcI + 1)τ
charge Q = xCC Ct = σC(1 + τ ) Q = σC
1 + kcI τ
1 + (kcI + 1)τ
state-of-charge s = xCC t = σ(1 + τ ) s=σ
1 + kcI τ
1
power P = xCV = xCC CVc (1 + kcV τ ) P = xCVc = xCC CVc
(1 + kcI τ )2
s
−1
time τ= s
−1 τ= σ
σ 1 − kcI ( σs − 1)

At t = tc (τ = 0), the current and voltage have the CC/CV values (I = xCC C , V = Vc ), the state-of-charge
s = σ , and the charging power is at the peak (P = xCC CVc = Pcap (xCC /xmbd )fc ). For s = 1 at large
time, kcI must be related to σ : 1/σ = 1 + 1/kcI ; the measured CV capacity is matched better with smaller
values for kcI . Figure 28-12 shows the charging parameters kcV and kcI for a number of batteries.
For small charging current xCC , the voltage Vc is reached at the nominal charging time tc = 1/xCC .
So the CC phase is the entire process, and s = xCC t. For higher current, the voltage Vc is reached sooner,
at tc = σ/xCC with σ < 1. Then the CV phase is required to complete the charging. For a given battery,
σ = 1 at low charge current, and σ decreases with increasing xCC . Figure 28-13 shows 1/σ and 1 + 1/kcI
for a number of batteries. An approximation for the variation of σ is
1 1
=1+ = 1 + kσ (xCC − 0.2)
σ kcI

above xCC = 0.2, with kσ depending on the battery.

For a flight condition or mission segment, the total energy flow delivered to the battery is Ėcomp . The
temperature Tc is specified. The state-of-charge is s = Efuel /Efuel−cap . The charge current x = xmbd ξ
238 Battery Model

is calculated from the required energy flow, |Ėbatt | = P = IV = ξxmbd CV = Pcap ξV /Vref . In the CC
phase (τ < 0, s < σ ):  

ξ= |Ėcomp | − Ploss /Pcap /fc (1 + kcV τ )

and xCC = x. In the CV phase (τ > 0, s > σ ):

  
ξ= |Ėcomp | − Ploss /Pcap /fc

and xCC = x(1 + kcI τ )2 . Successive-substitution solution for ξ converges in about 10 iterations. First
the CC solution is found, then if τ > 0 the CV solution is found (with an upper limit on xCC ). Then
Ėbatt = Ėcomp + Ploss , and the battery efficiency is ηbatt = Ėbatt /Ėcomp . The effective energy flow is
Ėeff = Ėbatt /ffade . The maximum battery charging power is Pmax = P evaluated at current xCCmax , or
at a specified charge current xCC . The battery power margin is Pmax − |Ėbatt |.

28–3 References

1) Gao, L.; Liu, S.; and Dougal, R.A. “Dynamic Lithium-Ion Battery Model for System Simulation.”
IEEE Transactions on Components and Packaging Technologies, 25:3 (September 2002).
2) Ramadass, P.; Haran, B.; White, R.; and Popov, B.N. “Mathematical Modeling of the Capacity Fade
of Li-Ion Cells.” Journal of Power Sources, 123:2 (September 2003).
3) Chen, M., and Rincón-Mora, G.A. “Accurate Electrical Battery Model Capable of Predicting Runtime
and I-V Performance.” IEEE Transactions on Energy Conversion, 21:2 (June 2006).
4) Erdinc, O.; Vural, B.; and Uzunoglu, M. “A Dynamic Lithium-Ion Battery Model Considering the
Effects of Temperature and Capacity Fading.” Second International Conference on Clean Electrical
Power, June 2009.
5) Datta, A., and Johnson, W. “Powerplant Design and Performance Analysis of a Manned All-Electric
Manned Helicopter.” Journal of Propulsion and Power, Vol. 30, No. 2 (March-April 2014).
6) Barth, A.; Feil, R.; Konkak, K.; and Hajek, M. “Conceptual Study for an Autonomous Rotorcraft for
Extreme Altitudes.” Fortieth European Rotorcraft Forum, Southampton, UK, September 2014.
Battery Model 239

100000.

10000.

1000.
power (W)

100.

10.

1.
1. 10. 100. 1000. 10000.
capacity (W-hr)

Figure 28-1. Lithium-ion battery capacity (W-hr) and power (W).

35.

30.

25.
xmcd (1/hr)

20.

15.

10.

5.

0.
0.1 1. 10. 100. 1000. 10000.
capacity (W-hr)

Figure 28-2. Lithium-ion battery maximum continuous discharge current.

240 Battery Model

150.

xmbd = power/capacity (1/hr) 125.

100.

75.

50.

25.

0.
0.1 1. 10. 100. 1000. 10000.
capacity (W-hr)

Figure 28-3. Lithium-ion battery maximum burst discharge current.

16.

12.
xCCmax (1/hr)

8.

4.

0.
0.1 1. 10. 100. 1000. 10000.
capacity (W-hr)

Figure 28-4. Lithium-ion battery maximum charge current.

Battery Model 241

100000.

specific power (W/kg) 10000.

1000.

100.

10.
0. 50. 100. 150. 200. 250.
specific energy (W-hr/kg)

Figure 28-5. Lithium-ion battery specific energy (W-hr/kg) and specific power (W/kg).

700.

600. cylindrical
energy density (W-hr/liter)

other
500.

400.

300.

200.

100.

0.
0. 50. 100. 150. 200. 250. 300.
specific energy (W-hr/kg)

Figure 28-6. Lithium-ion battery specific energy (W-hr/kg) and energy density (W-hr/liter).
242 Battery Model

charge current ( 25 οC )
4.2
0.3C
0.5C
4.0 1.0C
2.0C
3.8 3.0C
4.0C
voltage (V)

3.6

3.4

3.2

3.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
depth-of-discharge

4.2 temperature (0.3C)

40 οC
4.0 25 οC
0 οC
3.8 -20 οC
-40 οC
voltage (V)

3.6

3.4

3.2

3.0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1
depth-of-discharge

Figure 28-7. Lithium-ion battery discharge characteristics.

Battery Model 243

1.1

1.0

open circuit voltage F V 0.9

0.8

0.7

0.6

0.5
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1

depth-of-discharge d

Figure 28-8. Lithium-ion battery open circuit voltage function.

0.25

0.20

0.15
kdIxmbdC

0.10

0.05

0.00
0.00 0.05 0.10 0.15 0.20 0.25

internal resistance, xmbdCR/Vref

Figure 28-9. Lithium-ion battery discharge model, current parameters.

244 Battery Model

0.010

0.008

0.006
kdT

0.004

0.002

0.000
0.0 0.3 0.6 0.9 1.2 1.5E-5

kVT

Figure 28-10. Lithium-ion battery discharge model, temperature parameters.

 Battery Model 245

1.05 1.2

1.0
1.00 xCC = 0.5
0.8 xCC = 1.0

current x=I/C
voltage V/Vc

0.95
0.6
xCC = 0.5
0.90
xCC = 1.0 0.4

0.85
0.2

0.80 0.0
0.0 1.0 2.0 3.0 0.0 1.0 2.0 3.0
time (hr) time (hr)

1.2 1.2

1.0 1.0
state-of-charge s

0.8 0.8 xCC = 0.5

power P/CVc

xCC = 1.0
0.6 0.6

0.4 0.4
xCC = 0.5
0.2 xCC = 1.0 0.2

0.0 0.0
0.0 1.0 2.0 3.0 0.0 0.2 0.4 0.6 0.8 1.0
time (hr) state-of-charge s

Figure 28-11. Lithium-ion battery charge characteristics.

246 Battery Model
0.25

0.20

0.15
kcV

0.10

0.05

0.00
0. 3. 6. 9. 12. 15.

kcI

Figure 28-12. Lithium-ion battery charge model parameters.

1.8
kσ = 0.8

1.6 1 /σ
1 + 1 / k cI

1.4
kσ = 0.1

1.2

1.0
0.0 0.5 1.0 1.5 2.0 2.5 3.0
charge current xCC

Figure 28-13. Lithium-ion battery charge model, CV parameters.

 Chapter 29

AFDD Weight Models

This chapter presents the rotorcraft weight models developed by the U.S. Army Aeroflightdynamics
Directorate (AFDD). For some weight groups several models are available, designated AFDDnn. The
weights are estimated from parametric equations, based on the weights of existing turbine-powered
helicopters and tiltrotors. The figures of this chapter compare the weights calculated from these equations
with the actual weights. The results of these equations are the weight in pounds, and the units of the
parameters are noted in the tables. Technology factors χ are included in the weight equations. The input
usually includes a weight increment dW that can be added to the results of the weight model. Thus
typically a component or element weight is obtained from W = χwmodel + dW . Weight of individual
elements in a group can be fixed by using dW and setting the corresponding technology factor χ = 0.
With χ = 0, the increment dW can account for something not included in the parametric model.

The weight models are implemented as part of the aircraft components. The weights are entered
into the weight statement data structure (extended RP8A format) for each component, reflected in the
organization of this chapter.

29–1 Wing Group

The wing group consists of: basic structure (primary structure, consisting of torque box and spars,
plus extensions); fairings (leading edge and trailing edge); fittings (non-structural); fold/tilt structure;
and control surfaces (flaps, ailerons, flaperons, and spoilers). There are separate models for a tiltrotor
or tiltwing configuration and for other configurations (including compound helicopter).

29-1.1 Tiltrotor or Tiltwing Wing

Wing weight equations for a tiltrotor or tiltwing aircraft are based on methodology developed by
Chappell and Peyran (refs. 1 and 2). The wing is sized primarily to meet torsional stiffness requirements.
The primary structure weight is calculated from torque box and spar weights:

Wbox = Atb ρtb bw /etb

Wspar = Ct Asp ρsp bw /esp
wprim = (Wbox + Wspar )funits
Wprim = χprim wprim

A consistent mass-length-time system is used in the equations for Wbox and Wspar , which therefore have
units of slug or kg. The primary structure weight Wprim however has units of lb or kg, hence a conversion
factor funits = g is required for English units. The wing fairing (leading edge and trailing edge), control
248 AFDD Weight Models

surface (flaps, ailerons, flaperons, and spoilers), fittings (non-structural), and fold/tilt weights are:

wfair = Sfair Ufair Wfair = χfair wfair

wflap = Sflap Uflap Wflap = χflap wflap
wfit = ffit Tcap Wfit = χfit wfit
wfold = ffold (Wprim + Wfair + Wflap + Wfit + Wtip ) Wfold = χfold wfold

The control surface area Sflap for a tiltrotor wing is the sum of the flap and flaperon areas. The fairing
area is
Sfair = (bw − wattach ) cw (1 − wtb ) − Sflap

The wing extension weight is:

wext = Sext Uext Wext = χext wext

wefold = fefold Wext Wefold = χefold wefold

and these terms are added to Wprim and Wfold . The tiltrotor wing weight (and wing folding weight in
fuselage group) depends on Wtip , the weight on the wing tips; which is the sum of rotor group, engine
section or nacelle group, air induction group, engine system, drive system (except drive shaft), rotary
wing and conversion flight controls, hydraulic group, trapped fluids, and wing extensions. An adjustment
of this calculated weight can be used; a negative increment is required when the engine and transmission
are not at the tip location with the rotor. The weight on wing tip is used as the fraction ftip = Wtip /WSD ;
the mass on the wing tip is Mtip (slug or kg).
To estimate the wing weights, the required stiffness is scaled with input frequencies (per rev)
of the wing primary bending and torsion modes. First the torque box is sized to meet the torsional
stiffness (frequency) requirement. Next spar cap area is added as required to meet the chord and beam
bending frequency requirements. Finally spar cap area is added if necessary for a jump takeoff condition.
Wing section form factors, relating typical airfoil and torque box geometry to ideal shapes, are input or
calculated from the thickness-to-chord ratio and the torque box chord to wing chord ratio:

FB = 0.073 sin(2π(τw − 0.151)/0.1365) + 0.14598τw

+ 0.610 sin(2π(wtb + 0.080)/2.1560) − (0.4126 − 1.6309τw )(wtb − 0.131) + 0.0081
FC = 0.640424wtb2
− 0.89717wtb + 0.4615τw + 0.655317
  
FT = ((0.27 − τw )/0.12)0.12739 −0.96 + 3.32 + 94.6788wtb − (wtb /0.08344)2
− 2.7545wtb
2
+ 5.1799wtb − 0.2683
FV H = 0.25 sin(5.236wtb ) + 0.325

for beam bending, chord bending, torsion, and spar cap vertical/horizontal bending. The ideal shape for
torsional stiffness is a tube of radius tw , so the torsional stiffness J = FT Atb t2w /4. The ideal shape for
chord bending is two caps ctb apart, so ICtb = FC Atb c2tb /4. The ideal shape for beam bending is two caps
tw apart, so IBsp = FV H Asp t2w /4 and IBtb = FB Atb t2w /4. The torque box cross-sectional area is obtained
from the wing torsion frequency;

1 1
GJ = (ωT Ω)2 (bw − wattach ) Mtip rpylon
2
2 2
Atb = 4GJ/(Gtb FT t2w )
AFDD Weight Models 249

It is assumed that between the points of attachment to the fuselage, the torque box is very stiff in torsion,
but does deflect in bending. So the effective length is 12 (bw − wattach ) for torsion, and 12 bw for bending.
The spar cap cross-sectional area (in addition to torque box material) is obtained from beam and chord
bending frequencies:
1 3 1
EIC = (ωC Ω)2 b Mtip fmode
24 w 2
1 1
EIB = (ωB Ω)2 b3w Mtip fmode
24 2
EICtb = Etb FC Atb c2tb /4
EICsp = EIC − EICtb
ACsp = EICsp /(Esp c2tb /4)

EIBtb = Etb FB Atb t2w /4

EIV H = Esp FV H ACsp t2w /4
EIBsp = EIB − EIBtb − EIV H
ABsp = EIBsp /(Esp t2w /4)
EIsp = EIV H + EIBsp

Asp = ACsp + ABsp

where EICsp and EIBsp are replaced by zero if negative (no additional spar material required). The factor
fmode = 1 − ftip is a mode shape correction for fuselage motion. Next the primary structure, fairing,
flap, and fitting weights are calculated as before; and the sum Wwing = Wprim + Wfair + Wflap + Wfit .
Additional spar cap material for a jump takeoff condition is obtained from the ultimate applied bending
moment at the wing root:
 
MU = Tcap w 0.75(1 − ftip ) − 0.375( w /bw )(Wwing /WSD )

where Tcap is the maximum thrust capability of one rotor, equal to the greater of njump WSD /Nrotor or
2
(CT /σ)ρAb Vtip (from an input CT /σ at the jump takeoff condition, sea level standard, and hover rotor
speed). The bending moment capacity of the wing is

Mtb = 2EIBtb U /tw

Msp = 2Cm EIsp U /tw

Then the additional cross-section area is obtained from the moment deficit:
ΔM = MU − (Mtb + Msp )
ΔAsp = 2ΔM/(U Esp tw )
ΔWspar = Cj ΔAsp ρsp bw /esp

where ΔM is replaced by zero if negative (no additional spar material required). If ΔWspar is positive, it
is added to Wspar , and the primary structure, fairing, flap, and fitting weights are recalculated. Parameters
are defined in table 29-1, including units as used in these equations. Here a consistent mass-length-time
system is used, producing Wbox and Wspar in slug or kg. Typically the input uses conventional English
units for density (lb/in3 ) and modulus (lb/in2 ).
250 AFDD Weight Models

Table 29-1. Parameters for tiltrotor wing weight.

parameter definition units
Wbox , Wspar wing torque box and spar weights slug or kg
WSD structural design gross weight lb
(CT /σ)jump rotor maximum thrust capability (jump takeoff)
njump load factor at WSD (jump takeoff)
Nrotor number of rotors
bw wing span (length of torque box) ft or m
w = bw − wf us wing length (span less fuselage width) ft or m
cw wing chord ft or m
wtb torque box chord to wing chord ratio
ctb = wtb cw torque box chord ft or m
τw wing airfoil thickness-to-chord ratio
tw = τw cw wing thickness ft or m
2
rpylon pylon radius of gyration (pitch inertia = rpylon Mtip ) ft or m
Ω rotor speed for wing weight design condition rad/sec
ωT wing torsion mode frequency (fraction rotor speed) per rev
ωB wing beam bending mode frequency (fraction rotor speed) per rev
ωC wing chord bending mode frequency (fraction rotor speed) per rev
ρtb density of torque box material slug/ft3 or kg/m3
ρsp density of spar cap material slug/ft3 or kg/m3
Gtb torque box shear modulus lb/ft2 or N/m2
Etb torque box modulus lb/ft2 or N/m2
Esp spar modulus lb/ft2 or N/m2
U ultimate strain allowable (minimum of spar and torque box)
Ct weight correction for spar taper (equivalent stiffness)
Cj weight correction for spar taper (equivalent strength)
Cm strength correction for spar taper (equivalent stiffness)
etb structural efficiency factor, torque box
esp structural efficiency factor, spars
Ufair unit weight of leading and trailing-edge fairings lb/ft2 or kg/m2
Uflap unit weight of control surfaces lb/ft2 or kg/m2
Sfair area of leading and trailing-edge fairings ft2 or m2
Sflap area of control surfaces ft2 or m2
ffit wing fittings and brackets (fraction maximum thrust of one rotor)
ffold wing fold/tilt (fraction total weight excluding fold, including weight on tips)
wattach width of wing structural attachments to body ft or m
Uext unit weight of wing extension lb/ft2 or kg/m2
Sext area of wing extensions (span bext times mean chord cext ) ft2 or m2
fefold wing extension fold/tilt (fraction extension weight)
AFDD Weight Models 251

29-1.2 Aircraft Wing

There are two models intended for the wing of a compound helicopter: area method and parametric
method. For the area method (based on weight per unit area), the total wing weight excluding folding
is:
wwing = Sw Uw
and fprim = 1 − ffair − fflap − ffit . Typically Uw = 5 to 9 lb/ft2 . For the parametric method (AFDD93),
the total wing weight including folding is:
 0.847
WSD fL
wwing = 5.66411fLGloc n0.39579
z
0.21754 0.50016
Sw Aw
1000 cos Λw
(1 − bfold )−0.14356
0.09359
((1 + λw )/τw )
with fprim = 1 − ffair − fflap − ffit − ffold ; here fLGloc = 1.7247 if the landing gear is on the wing, and
1.0 otherwise. The lift factor fL accounts for wing relief factor and wing/rotor lift share. Based on 25
fixed-wing aircraft, the average error of the aircraft wing equation is 3.4% (fig. 29-1). Then the primary
structure, secondary structure, and control surface weights are:
Wprim = χprim fprim wwing
Wfair = χfair ffair wwing
Wflap = χflap fflap wwing
Wfit = χfit ffit wwing
Wfold = χfold ffold wwing
The wing extension weight is:
wext = Sext Uext Wext = χext wext
wefold = fefold Wext Wefold = χefold wefold
and these terms are added to Wprim and Wfold . Parameters are defined in table 29-2, including units as
used in these equations.
Table 29-2. Parameters for aircraft wing weight.
parameter definition units
Sw wing planform area (theoretical) ft2
Uw unit weight of wing planform lb/ft2 or kg/m2
WSD structural design gross weight lb
fL lift factor
nz design ultimate flight load factor at WSD g
Λw wing sweep angle deg
Aw wing aspect ratio
λw wing taper ratio (tip chord/root chord)
τw wing airfoil thickness-to-chord ratio
bfold fraction wing span that folds (0 to 1)
ffair fairings (fraction total wing weight)
fflap control surfaces (fraction total wing weight)
ffit fittings (fraction total wing weight)
ffold fold/tilt (fraction total wing weight)
Uext unit weight of wing extension lb/ft2 or kg/m2
Sext area of wing extensions (span bext times mean chord cext ) ft2
fefold wing extension fold/tilt (fraction extension weight)
252 AFDD Weight Models
29–2 Rotor Group

The rotor group consists of: blades, hub and hinge, spinner, and blade fold structure. The blade
and hub-hinge weights for the AFDD82 model are:

0.6592 1.3371 0.9959 0.6682 2.5279

wblade = 0.02606Nrotor Nblade R c Vtip νblade Wblade = χblade wblade
0.2807 1.5377 0.4290 2.1414
whub = 0.003722Nrotor Nblade R Vtip νhub (Wblade /Nrotor )0.5505 Whub = χhub whub

Based on 37 aircraft, the average error of the blade equation is 7.7% (fig. 29-2). Based on 35 aircraft,
the average error of the hub equation is 10.2% (fig. 29-3). The blade and hub-hinge weights for the
AFDD00 model are:
0.53479 1.74231 0.77291 0.87562 2.51048
wblade = 0.0024419ftilt Nrotor Nblade R c Vtip νblade Wblade = χblade wblade
0.20373 0.60406 0.52803 1.00218
whub = 0.0061182Nrotor Nblade R Vtip νhub (Wblade /Nrotor )0.87127 Whub = χhub whub

where ftilt = 1.17940 for tilting rotors; 1.0 otherwise. Based on 51 aircraft, the average error of the blade
equation is 7.9% (fig. 29-4) and the average error of the hub equation is 12.2% (fig. 29-5). This equation
for hub weight was developed using the actual blade weight. Using the blade weight equation instead
gives
0.16383 0.19937 0.06171 0.46203
whub = 0.18370Nrotor Nblade R Vtip νhub (wblade /Nrotor )1.02958
1.02958 0.71443 1.99321 0.79577 0.96323 0.46203 2.58473
= 0.00037547ftilt Nrotor Nblade R c Vtip νhub νblade

which really just adds chord and ftilt to the regression parameters. Based on 51 aircraft, the average
error of this equation is 9.2% (fig. 29-5). Thus using the blade weight equation results in a lower average
error, and best represents legacy rotor systems. The hub weight equation using the actual blade weight
is best for advanced technology rotors with blades lighter than trend. For teetering and gimballed rotors,
the flap frequency ν should be the coning frequency. If the weight is evaluated separately for each rotor,
then Nrotor = 1 should be used in the equations.

The fairing/spinner and blade fold weights are:

2
Wspin = χspin 7.386Nrotor Dspin

Wfold = χfold ffold Wblade

The blade weight is for all blades of the rotors. Typically ffold = 0.04 for manual fold and ffold = 0.28
for automatic fold. The rotor shaft, rotor support, and duct weights are:

Wshaft = χshaft Nrotor shaft Ushaft

 
Wsupt = χsupt fsupt WM T O + Usupt R

Wduct = χduct Nrotor Sduct Uduct

The rotor shaft length is shaft = fshaft R. Parameters are defined in table 29-3, including units as used
in these equations.
AFDD Weight Models 253

Table 29-3. Parameters for rotor weight.

parameter definition units
Nrotor number of rotors
Nblade number of blades per rotor
R rotor radius in hover ft
c rotor mean geometric blade chord ft
Vtip rotor hover tip velocity ft/sec
νblade , νhub flap natural frequency (for weight estimate) per rev
Dspin spinner diameter ft
ffold blade fold weight (fraction total blade weight)
WM T O maximum takeoff weight lb
fshaft rotor shaft length (fraction rotor radius)
Ushaft unit weight of rotor shaft lb/ft or kg/m
fsupt rotor support structure weight (fraction maximum takeoff weight)
Usupt unit weight of rotor support structure lb/ft or kg/m
Sduct area of duct ft2 or m2
Uduct unit weight of duct lb/ft2 or kg/m2

29-2.1 Lift-Offset Rotor

For lift-offset rotors, the blade and hub weights can be calculated based on the methodology of
reference 3. The blade weight is estimated based on the beam stiffness required to maintain the clearance
s when the blade is loaded by the lift offset. The blade tip deflection is proportional to δ ∝ P R3 /EI ,
where EI is the bending stiffness. The beam loading is P ∝ nz WSD L/Nblade . With Axs the blade
cross-section area, the moment of inertia I ∝ Axs t2 . The criterion is δ = h − s. Hence the blade weight
is Wblade ∝ ρNblade RAxs ∝ (ρ/E)nz WSD LR4 /(t2 (h − s)) with E the elastic modulus, and ρ here the
material density. The hub weight is estimated based on the structure in upper and lower hub plates
required to react a tensile force F = Ccent + Mbend /(x/2) due to combined centrifugal force and bending
moment at the root; where the hub plate separation x scales with the blade thickness t. The centrifugal
force Ccent ∝ (Wblade /Nblade )Vtip2
/R. The bending moment Mbend ∝ nz WSD R. The limit tensile stress
σ ∝ F/Axs gives a criterion for the total hub arm area Axs . The radius of the hub scales with the blade
 
thickness t. Hence the hub weight is Whub ∝ ρNblade Axs ∝ (ρ/σ) Wblade Vtip 2
t/R + Knz WSD RNblade .
The distribution factor K is determined from the calibration weights. The inter-rotor shaft weight
is estimated based on the structure to react the hub moment caused by lift offset. The hub moment
M ∝ nz WSD LR. The shaft diameter d scales with the blade thickness t. The shaft length scales with
the rotor separation h. The ultimate bending stress σ = M/(I/c) gives a criterion for the area moment
I/c ∝ d2 w, hence for the shaft wall thickness w. Then Wshaft ∝ ρ dw ∝ (ρ/σ)nz WSD LRh/t is the shaft
weight.
The blade, hub-hinge, and inter-rotor shaft weights are:
wblade = Nrotor 0.000083770 wL R3 /(2(h − s)t2.2R ) Wblade = χblade wblade
 
2
whub = Nrotor 0.17153 w RNblade + 0.000010534(Wblade /Nrotor )Vtip t.2R /R Whub = χhub whub
wshaft = Nrotor 0.081304 wL R2 2h/t.2R Wshaft = χshaft wshaft
where w = nz WSD /1000, and  
0.8 + 0.2λ
t.2R = τ.2R c
0.5 + 0.5λ
254 AFDD Weight Models

is the blade thickness at 20%R. These equations were developed for the coaxial rotor configuration
(Nrotor = 2), and calibrated to the XH-59A weights. The material factors (ρ/E and ρ/σ ) are included
in the technology factors χ. These weight terms can be used separately, with other hub and blade and
shaft weight models. Parameters are defined in table 29-4, including units as used in these equations.

Table 29-4. Parameters for lift-offset rotor weight.

parameter definition units
Nrotor number of rotors
Nblade number of blades per rotor
WSD structural design gross weight lb
nz design ultimate flight load factor at WSD g
λ blade taper ratio (tip chord/root chord)
τ.2R blade airfoil thickness-to-chord ratio (at 20%R)
R rotor radius ft
c blade mean chord ft
h coaxial rotor separation (fraction rotor diameter)
s coaxial rotor tip clearance (fraction rotor diameter)
L lift offset (Mroll /T R)
Vtip rotor hover tip velocity ft/sec

29–3 Empennage Group

The empennage group consists of: horizontal tail, vertical tail, and tail-rotor. The tail plane weight
consists of the basic structure and fold structure. There are two models for tail plane basic weight: area
method and parametric method. For the area method (based on weight per unit area), the weight is
Wtail = Stail Utail . The parametric weight model depends on the aircraft configuration. The helicopter
or compound model is AFDD82. The horizontal tail weight is:

tiltrotor or tiltwing 0.2

Wht = χht Sht (0.00395Sht Vdive − 0.4885)
1.1881 0.3173
helicopter or compound Wht = χht 0.7176Sht Aht

Based on 13 aircraft, the average error of the helicopter horizontal tail equation is 22.4% (fig. 29-6).
The vertical tail weight is:

tiltrotor or tiltwing 0.2

Wvt = χvt Svt (0.00395Svt Vdive − 0.4885)
0.9441 0.5332
helicopter or compound Wvt = χvt 1.0460ftr Svt Avt

where ftr = 1.6311 if the tail-rotor is located on the vertical tail; 1.0 otherwise. Based on 12 aircraft, the
average error of the helicopter vertical tail equation is 23.3% (fig. 29-7). Vdive is the design dive speed,
calculated or input; Vdive = 1.25Vmax , where Vmax is the maximum speed at design gross weight and sea
level standard conditions. The fold weight is a fraction of the basic weight: Wfold = χfold ffold Wbasic ,
where Wbasic = Wht or Wvt .
The tail-rotor weight is:

0.0897
Wtr = χtr 1.3778Rtr (PDSlimit R/Vtip )0.8951
AFDD Weight Models 255

Based on 19 aircraft, the average error of the helicopter tail-rotor equation is 16.7% (fig. 29-8). The
tail-rotor weight is calculated by the rotor component model, including rotor support and duct weights.
Rotor weight models can be used for the tail-rotor. Parameters are defined in table 29-5, including units
as used in these equations.

Table 29-5. Parameters for tail weight.

parameter definition units
Sht horizontal tail planform area ft2
Svt vertical tail planform area ft2
Aht horizontal tail aspect ratio
Avt vertical tail aspect ratio
Vdive design dive speed at sea level kts
Rtr tail-rotor radius ft
PDSlimit drive system power limit (MCP) hp
R main-rotor radius in hover ft
Vtip main-rotor hover tip velocity ft/sec
ffold fold (fraction basic weight excluding fold)
Stail horizontal or vertical tail planform area ft2 or m2
Utail unit weight of tail planform lb/ft2 or kg/m2

29–4 Fuselage Group

The fuselage group consists of: basic structure; wing and rotor fold/retraction; tail fold/tilt; and
marinization, pressurization, and crashworthiness structure. The AFDD84 model is a universal body
weight equation, used for tiltrotor and tiltwing as well as for helicopter configurations. The AFDD82
model is a helicopter body weight equation, not used for tiltrotor or tiltwing configuration.
For the AFDD84 (UNIV) model, the basic structure weight is
 0.4879  0.2075
WM T O nz WSD 0.1676 0.1512
wbasic = 25.41fLGloc fLGret framp Sbody
1000 1000
Wbasic = χbasic wbasic

where fLGloc = 1.1627 if the landing gear is located on the fuselage, and 1.0 otherwise; fLGret = 1.1437
if the landing gear is on the fuselage and retractable, and 1.0 otherwise; framp = 1.2749 if there is a cargo
ramp, and 1.0 otherwise. Based on 35 aircraft, the average error of the body equation is 6.5% (fig. 29-9).
The tail fold, wing and rotor fold, marinization, pressurization, and crashworthiness weights are:
wtfold = ftfold Wtail Wtfold = χtfold wtfold
wwfold = fwfold (Wwing + Wtip ) Wwfold = χwfold wwfold
wmar = fmar Wbasic Wmar = χmar wmar
wpress = fpress Wbasic Wpress = χpress wpress
wcw = fcw (Wbasic + Wtfold + Wwfold + Wmar + Wpress ) Wcw = χcw wcw

Typically ftfold = 0.30 for a folding tail, and fcw = 0.06. For wing folding the weight on the wing tip
(Wtip ) is required (calculated as for the wing group). Parameters are defined in table 29-6, including
units as used in these equations.
256 AFDD Weight Models

Table 29-6. Parameters for fuselage weight (AFDD84 model).

parameter definition units
WM T O maximum takeoff weight lb
WSD structural design gross weight lb
Sbody wetted area of body ft2
R main-rotor radius ft
nz design ultimate flight load factor at WSD g
length of fuselage ft
ftfold tail fold weight (fraction tail weight)
fwfold wing and rotor fold weight (fraction wing/tip weight)
Wtail tail group weight
Wwing + Wtip wing group weight plus weight on wing tip
fmar marinization weight (fraction basic body weight)
fpress pressurization (fraction basic body weight)
fcw crashworthiness weight (fraction fuselage weight)

For the AFDD82 (HELO) model, the basic structure weight is

 0.4908
WM T O
wbasic = 5.896framp n0.1323
z
0.2544
Sbody 0.6100
1000
Wbasic = χbasic wbasic
where framp = 1.3939 if there is a cargo ramp, and 1.0 otherwise. Based on 30 aircraft, the average error
of the body equation is 8.7% (fig. 29-10). The tail fold, wing and rotor fold, marinization, pressurization,
and crashworthiness weights are:
wtfold = ftfold Wbasic Wtfold = χtfold wtfold
wwfold = fwfold (Wbasic + Wtfold ) Wwfold = χwfold wwfold
wmar = fmar Wbasic Wmar = χmar wmar
wpress = fpress Wbasic Wpress = χpress wpress
wcw = fcw (Wbasic + Wtfold + Wwfold + Wmar + Wpress ) Wcw = χcw wcw
Typically ftfold = 0.05 for a folding tail, and fcw = 0.06. Parameters are defined in table 29-7, including
units as used in these equations.

Table 29-7. Parameters for fuselage weight (AFDD82 model).

parameter definition units
WM T O maximum takeoff weight lb
Sbody wetted area of body ft2
R main-rotor radius ft
nz design ultimate flight load factor at WSD g
length of fuselage ft
ftfold tail fold weight (fraction basic structure)
fwfold wing and rotor fold weight (fraction basic structure and tail fold)
fmar marinization weight (fraction basic body weight)
fpress pressurization (fraction basic body weight)
fcw crashworthiness weight (fraction fuselage weight)
AFDD Weight Models 257

29–5 Alighting Gear Group

The alighting gear group consists of: basic structure, retraction, and crashworthiness structure.
There are two models, parametric (AFDD82) and fractional. The basic landing gear weight is:
0.6662 0.5360
parametric wLG = 0.4013WM T O NLG (W/S)0.1525
fractional wLG = fLG WM T O
and WLG = χLG wLG . Typically fLG = 0.0325 (fractional method). Based on 28 aircraft, the average
error of the parametric equation is 8.4% (fig. 29-11). The retraction and crashworthiness weights are:
wLGret = fLGret WLG WLGret = χLGret wLGret
wLGcw = fLGcw (WLG + WLGret ) WLGcw = χLGcw wLGcw
Typically fLGret = 0.08, and fLGcw = 0.14. Parameters are defined in table 29-8, including units as used
in these equations.

Table 29-8. Parameters for landing gear weight.

parameter definition units
WM T O maximum takeoff weight lb
fLG landing gear weight (fraction maximum takeoff weight)
W/S wing loading (1.0 for helicopter) lb/ft2
NLG number of landing gear assemblies
fLGret retraction weight (fraction basic weight)
fLGcw crashworthiness weight (fraction basic and retraction weight)

29–6 Engine Section or Nacelle Group and Air Induction Group

The engine section or nacelle group consists of: engine support structure, engine cowling, and
pylon support structure. The weights (AFDD82 model) are:
Wsupt = χsupt 0.0412(1 − fairind )(Weng /Neng )1.1433 Neng
1.3762

1.3476
Wcowl = χcowl 0.2315Snac
Wpylon = χpylon fpylon WM T O
Based on 12 aircraft, the average error of the engine support equation is 11.0% (fig. 29-12). Based on
12 aircraft, the average error of the engine cowling equation is 17.9% (fig. 29-13). The air induction
group weight (AFDD82 model) is:
Wairind = χairind 0.0412fairind (Weng /Neng )1.1433 Neng
1.3762

Typically fairind = 0.3 (range 0.1 to 0.6). Based on 12 aircraft, the average error of the air induction
equation is 11.0% (fig. 29-14).
Alternatively, the nacelle group weight (including support, cowling, and pylon) scales with power
or thrust, and the air induction group weight scales with the nacelle area:
engine group: jet group:
Wcowl = χcowl Knac P Xnac Wcowl = χcowl Knac T Xnac
258 AFDD Weight Models

Wairind = χairind Snac Uairind

Parameters are defined in table 29-9, including units as used in these equations.

Table 29-9. Parameters for engine section, nacelle, and air induction weight.
parameter definition units
WM T O maximum takeoff weight lb
Weng weight all main engines lb
Neng number of main engines
Snac wetted area of nacelles and pylon (less spinner) ft2
fairind air induction weight (fraction nacelle plus air induction)
fpylon pylon support structure weight (fraction WM T O )
P engine group power (SLS static, specified rating) hp
T jet group thrust (SLS static, specified rating) lb
Uairind unit weight of air induction group lb/ft2 or kg/m2

29–7 Propulsion Group

The propulsion group consists of the engine system, fuel system, and drive system.

29-7.1 Engine System

The engine system consists of the main engines, the engine exhaust system, and the engine acces-
sories. The engine system weights are:
engine group: jet group:
Weng = χeng Neng Wone eng Weng = χjet Njet Wone jet

Wexh = χexh Neng (K0exh + K1exh P ) Wexh = χexh Njet (K0exh + K1exh T )
Wacc = χacc 2.0088 flub (Weng /Neng )0.5919 Neng
0.7858

where flub = 1.4799 if the accessory weight includes the lubrication system weight, 1.0 if the lubrication
system weight is in the engine weight. The exhaust system weight is per engine, including any infrared
suppressor. The accessory weight equation is the AFDD82 model. Based on 16 aircraft, the average error
of the accessories equation is 11.5% (fig. 29-15). The engine system weight WES = Weng + Wexh + Wacc .
Parameters are defined in table 29-10, including units as used in these equations.

Table 29-10. Parameters for engine system weight.

parameter definition units
Neng number of main engines
Njet number of jets
P installed takeoff power (SLS static, specified rating) per engine hp
T installed takeoff thrust (SLS static, specified rating) per jet lb
K0exh , K1exh engine exhaust weight vs. power or thrust, constants
AFDD Weight Models 259

29-7.2 Propeller/Fan Installation

The auxiliary propulsion or propeller weight is:

AFDD82 1.04771
Wat = χat 0.0809484Nat Tat (Tat /Aat )−0.07821
0.91996 −0.48578 −0.45904 0.15690
AFDD10 Wat = χat 9.90350Nat Pat Nblade Ωprop Dprop fm

where Tat is at maximum speed, design gross weight, and SLS conditions, calculated or input. The
material factor fm = 1 for composite construction; 1.20 for wood; 1.31 for aluminum spar; and 1.44 for
aluminum construction. Based on 16 aircraft, the average error of the AFDD10 equation is 10.5%. The
propeller weight is calculated by the rotor component model, including rotor support and duct weights.
Rotor weight models can be used for the propeller. Parameters are defined in table 29-11, including
units as used in these equations.

Table 29-11. Parameters for propeller weight.

parameter definition units
Nat number of auxiliary thrusters
Tat thrust per propeller lb
Aat auxiliary thruster disk area ft2
Pat power per propeller hp
Nblade number of blades per propeller
Ωprop propeller rotation speed at Pat rpm
Dprop propeller diameter ft

29-7.3 Fuel System

The fuel system consists of tanks and support structure (including fuel tanks, bladders, supporting
structure, filler caps, tank covers, and filler material for void and ullage), and fuel plumbing (including
fuel system weight not covered by tank weight). There are two models for tank weight, parametric
(AFDD82) and fractional. The fuel system weights (AFDD82 model) are:

0.7717 0.5897 1.9491

Wtank = χtank 0.4341Cint Nint fcw fbt
* +
Wplumb = χplumb K0plumb + K1plumb (0.01Nplumb + 0.06Neng )(F/Neng )0.866

where fcw = 1.3131 for ballistically survivable (UTTAS/AAH level) and 1.0 otherwise. The ballistic
tolerance factor fbt = 1.0 to 2.5. Based on 15 aircraft, the average error of the fuel tank equation is 4.6%
(fig. 29-16). The fuel flow rate F is calculated for the takeoff power rating at static SLS conditions.
K1plumb is a crashworthiness and survivability factor; typically K1plumb = 2. K0plumb is the sum of
weights for auxiliary fuel (typically up to 120 lb), in-flight refueling (up to 150 lb), pressure refueling
(up to 150 lb), inerting system (up to 20 lb), etc.; typically K0plumb = 50 to 250 lb. Alternatively, for the
fractional model the fuel tank weight is Wtank = χtank ftank Wfuel−cap . The fractional model for the fuel
plumbing weight is Wplumb = χplumb (Wtank /χtank )fplumb /(1 − fplumb ). Parameters are defined in table
29-12, including units as used in these equations.
260 AFDD Weight Models

Table 29-12. Parameters for fuel system weight.

parameter definition units
Nint number of internal fuel tanks
Cint internal fuel tank capacity gallons
fbt ballistic tolerance factor
Nplumb total number of fuel tanks (internal and auxiliary) for plumbing
Neng number of main engines
K0plumb , K1plumb plumbing weight, constants lb
F fuel flow rate lb/hr
ftank fuel tank weight (fraction fuel capacity)
fplumb fuel plumbing weight (fraction total fuel system weight)

29-7.4 Drive System

The drive system consists of gear boxes and rotor shafts, drive shafts, and rotor brake. This
distribution of drive system weights is based on the following functional definitions. Gearboxes are
parts of the drive system that transmit power by gear trains, and the structure that encloses them. Rotor
shafts are the structure (typically a shaft) that transmits power to the rotor. Drive shafts are the structure
(typically a shaft) that transmits power in the propulsion system, but not directly to the rotor or by a gear
train. The rotor brake weight encompasses components that can prevent the rotor from freely turning.
The gear box and rotor shaft weights for the AFDD83 model are:
0.8195 0.0680 0.0663
wgbrs = 57.72PDSlimit fQ Ngb (Ωeng /1000)0.0369 /Ω0.6379
rotor

Wgb = χgb (1 − frs )wgbrs

Wrs = χrs frs wgbrs

Based on 30 aircraft, the average error of the gear box and rotor shaft equation is 7.7% (fig. 29-17). The
gear box and rotor shaft weights for the AFDD00 model are:
0.38553 0.78137 0.09899
wgbrs = 95.7634Nrotor PDSlimit Ωeng /Ω0.80686
rotor

Wgb = χgb (1 − frs )wgbrs

Wrs = χrs frs wgbrs

Based on 52 aircraft, the average error of the gear box and rotor shaft equation is 8.6% (fig. 29-18).
Typically frs = 0.13 (range 0.06 to 0.20). Parameters are defined in table 29-13, including units as used
in these equations.
The drive shaft (AFDD82 model) and rotor brake weights are:

Wds = χds 1.166Q0.3828 1.0455 0.3909

DSlimit xhub Nds (0.01fP )0.2693
Wrb = χrb 0.000871Wblade (0.01Vtip )2

where fP = fQ Ωother /Ωmain . Based on 28 aircraft, the average error of the drive shaft equation is 16.0%
(fig. 29-19). Based on 23 aircraft, the average error of the rotor brake equation is 25.1% (fig. 29-20).
The clutch weight in the weight statement is associated with an auxiliary power unit, and is a fixed input
AFDD Weight Models 261

value. The conventional rotor drive system clutch and free wheeling device weights are included in the
gear box and rotor shaft weight equations. Parameters are defined in table 29-14, including units as used
in these equations.
Typically fP = fQ = 60% for twin main-rotors (tandem, coaxial, and tiltrotor); for a single main-
rotor and tail-rotor, fQ = 3% and fP = 15% (18% for 2-bladed rotors).

Table 29-13. Parameters for drive system weight.

parameter definition units
PDSlimit drive system power limit (MCP) hp
Nrotor number of main-rotors
Ngb number of gear boxes
Ωrotor main-rotor rotation speed rpm
Ωeng engine output speed rpm
fQ second (main or tail) rotor torque limit* %
frs rotor shaft weight (fraction gear box and rotor shaft)
*input as fraction of total drive system torque limit

Table 29-14. Parameters for drive shaft and rotor brake weight.
parameter definition units
QDSlimit PDSlimit /Ωrotor (MCP) hp/rpm
Nds number of intermediate drive shafts
xhub length of drive shaft between rotors ft
fP second (main or tail) rotor power limit* %
Vtip main-rotor tip speed ft/sec
*input as fraction of total drive system power limit

29–8 Flight Controls Group

The flight controls group consists of cockpit controls, automatic flight control system, and system
controls. System controls consist of fixed wing flight controls, rotary wing flight controls, and conversion
(rotor tilt) flight controls. The weight equations model separately non-boosted controls (which do not
see aerodynamic surface or rotor loads), boost mechanisms (actuators), and boosted controls (which are
affected by aerodynamic surface or rotor loads). The load path goes from pilot, to cockpit controls, to
non-boosted controls, to boost mechanisms, to boosted controls, and finally to the component.
The cockpit controls weight Wcc is fixed (input), or scaled with design gross weight: Wcc =
Xcc
Kcc WD . The automatic flight control system weight Waf cs is fixed (input).
Fixed wing flight controls consist of non-boosted flight controls and flight control boost mechanisms.
The weights are:
0.6
full controls w = 0.91000WM TO
0.64345 0.40952
only stabilizer controls w = 0.01735WM T O Sht

and then
WF W nb = χF W nb fF W nb w
WF W mb = χF W mb (1 − fF W nb ) w
262 AFDD Weight Models

For a helicopter, the stabilizer control equation is used. Parameters are defined in table 29-15, including
units as used in these equations.

Table 29-15. Parameters for fixed wing flight control weight.

parameter definition units
Sht horizontal tail planform area ft2
WM T O maximum takeoff weight lb
fF W nb fixed wing non-boosted weight
(fraction total fixed wing flight control weight)

Rotary wing flight controls consist of non-boosted flight controls, flight control boost mechanisms,
and boosted flight controls. The non-boosted flight control weight (AFDD82 model) is:

fraction method WRW nb = χRW nb fRW nb (1 − fRW hyd )wf c

0.3999 1.3855
parametric method WRW nb = χRW nb 2.1785fnbsv WM T O Nrotor

where fnbsv = 1.8984 for ballistically survivable (UTTAS/AAH level), 1.0 otherwise. The parametric
method assumes the rotor flight controls are boosted and computes the weight of the non-boosted portion
up to the control actuators. Based on 20 aircraft, the average error of the non-boosted flight controls
equation is 10.4% (fig. 29-21). The flight control boost mechanism weight and boosted flight control
weight (AFDD82 model) are:

wf c = 0.2873fmbsv (Nrotor Nblade )0.6257 c1.3286 (0.01Vtip )2.1129 fRW

0.8942
red

WRW mb = χRW mb (1 − fRW hyd )wf c

WRW b = χRW b 0.02324fbsv (Nrotor Nblade )1.0042 Nrotor
0.1155 2.2296
c (0.01Vtip )3.1877

where fmbsv = 1.3029 and fbsv = 1.1171 for ballistically survivable (UTTAS/AAH level); 1.0 otherwise;
and fRW red = 1.0 to 3.0. Typically fRW nb = 0.6 (range 0.3 to 1.8); fRW hyd = 0.4. Based on 21 aircraft,
the average error of the boost mechanisms equation is 6.5% (fig. 29-22). Based on 20 aircraft, the
average error of the boosted flight controls equation is 9.7% (fig. 29-23). Parameters are defined in table
29-16, including units as used in these equations.

Table 29-16. Parameters for rotary wing flight control weight.

parameter definition units
WM T O maximum takeoff weight lb
Nrotor number of main-rotors
Nblade number of blades per rotor
c rotor mean blade chord ft
Vtip rotor hover tip velocity ft/sec
fRW nb rotary wing non-boosted weight (fraction boost mechanisms weight)
fRW hyd rotary wing hydraulics weight
(fraction hydraulics plus boost mechanisms weight)
fRW red flight control hydraulic system redundancy factor
AFDD Weight Models 263

The conversion controls consist of non-boosted tilt controls, and tilt control boost mechanisms;
they are used only for tilting rotor configurations. The weights are:

wCV mb = fCV mb WM T O WCV mb = χCV mb wCV mb

wCV nb = fCV nb WCV mb WCV nb = χCV nb wCV nb

Parameters are defined in table 29-17, including units as used in these equations.

Table 29-17. Parameters for conversion control weight.

parameter definition units
WM T O maximum takeoff weight lb
fCV nb conversion non-boosted weight (fraction boost mechanisms weight)
fCV mb conversion boost mechanisms weight (fraction maximum takeoff weight)

29–9 Hydraulic Group

The hydraulic group consists of hydraulics for fixed wing flight controls, rotary wing flight controls,
conversion (rotor tilt) flight controls, and equipment. The hydraulic weight for equipment, WEQhyd , is
fixed (input). The weights (AFDD82 model) are

WF W hyd = χF W hyd fF W hyd WF W mb

WRW hyd = χRW hyd fRW hyd wf c
WCV hyd = χCF hyd fCV hyd WCV mb

Typically fRW hyd = 0.4. Parameters are defined in table 29-18, including units as used in these equations.

Table 29-18. Parameters for hydraulic group weight.

parameter definition units
fF W hyd fixed wing hydraulics weight (fraction boost mechanisms weight)
fRW hyd rotary wing hydraulics weight
(fraction hydraulics plus boost mechanisms weight)
fCV hyd conversion hydraulics weight (fraction boost mechanisms weight)

29–10 Anti-Icing Group

The anti-icing group consists of the anti-ice system. The electrical system for anti-icing is part of
the electrical group. The weights are obtained from the sum over all rotors, all wings, and all engines:

WDIelect = χDIelect kelec Ablade
    
WDIsys = χDIsys krotor Ablade + kwing wing + kair Weng + kjet Wjet

Parameters are defined in table 29-19, including units as used in these equations.
264 AFDD Weight Models

Table 29-19. Parameters for anti-icing group weight.

parameter definition units
Ablade total blade area of rotor, from geometric solidity ft2 or m2
wing wing length (wing span less fuselage width) ft or m
kelect electrical system weight factor
krotor rotor deice system weight factor
kwing wing deice system weight factor
kair engine air intake deice system weight factor
kjet jet air intake deice system weight factor

29–11 Other Systems and Equipment

The following weights are fixed (input) in this model: auxiliary power group; instruments group;
pneumatic group; electrical group (aircraft); avionics group (mission equipment); armament group
(armament provisions and armor); furnishing and equipment group; environmental control group; and
load and handling group. Typical fixed weights are given in table 29-20, based on medium to heavy
helicopters and tiltrotors.

Table 29-20. Other systems and equipment weight.

group typical weight (lb)
SYSTEMS AND EQUIPMENT
flight controls group
cockpit controls 100–125
automatic flight control system 35–200
flight control electronics, mechanical 35–100
flight control electronics, fly-by-wire 250
auxiliary power group 130–300
instruments group 150-250
hydraulic group
equipment 50–300
electrical group 400–1000
avionics group (mission equipment) 400–1500
furnishings & equipment group 600–1000
crew only 100–200
environmental control group 50–250
anti-icing group 50–300
load & handling group
internal 200–400
external 150–300
FIXED USEFUL LOAD
crew 500–800
fluids (oil, unusable fuel) 50–150
AFDD Weight Models 265

29–12 Folding Weight

Folding weights are calculated in a number of groups: wing Wfold (including extensions), rotor
Wfold , tail Wfold , fuselage Wtfold and Wwfold . These are the total weights for folding and the impact of
folding on the group. A fraction ffoldkit of these weights can be in a kit, hence optionally removable.
Thus of the total folding weight, the fraction ffoldkit is a kit weight in the fixed useful load of the weight
statement, while the remainder is kept in the wing, rotor, or fuselage group weight.

29–13 Parametric Weight Correlation

Table 29-21 summarizes the statistics of the parametric weight estimation equations. Figure 29-24
shows the error of the calculated weight for the sum of all parametric weight, accounting on average
for 42% of the empty weight. This sum is composed of the structural group (based on the AFDD00
equation for rotor blade and hub weights, and the AFDD84 equation for body weight), the propulsion
group (based on the AFDD00 equation for drive system weight), and the flight controls group. Based
on 42 aircraft, the average error of the sum of all parametric weight is 5.3%. The corresponding average
error is 6.1% for the structural group (8.6% for the rotor group alone), 10.9% for the propulsion group,
and 8.7% for the flight controls group.

Table 29-21. Statistics of parametric weight equations.

group number of aircraft average error (%)
wing 25 3.4
rotor blade AFDD82 37 7.7
rotor hub AFDD82 37 10.2
rotor blade AFDD00 51 7.9
rotor hub AFDD00 51 12.2
horizontal tail 13 22.4
vertical tail 12 23.3
tail-rotor 19 16.7
fuselage AFDD82 30 8.7
fuselage AFDD84 35 6.5
alighting gear 28 8.4
engine support 12 11.0
engine cowling 12 17.9
air induction 12 11.0
accessory 16 11.5
propeller AFDD10 16 10.5
fuel tank 15 4.6
gear box + rotor shaft AFDD83 30 7.7
gear box + rotor shaft AFDD00 52 8.6
drive shaft 28 16.0
rotor brake 23 25.1
rotary wing flight controls non-boosted 20 10.4
rotary wing flight controls boost mechanisms 21 6.5
rotary wing flight controls boosted 20 9.7
266 AFDD Weight Models
29–14 References

1) Chappell, D., and Peyran, R. “Methodology for Estimating Wing Weights for Conceptual Tilt-Rotor
and Tilt-Wing Aircraft.” SAWE Paper No. 2107, Category No. 23, May 1992.
2) Chappell, D.P. “Tilt-rotor Aircraft Wing Design.” ASRO-PDT-83-1, 1983.
3) “Weight Trend Estimation for the Rotor Blade Group, Rotor Hub Group, and Upper Rotor Shaft of
the ABC Aircraft.” ASRO-PDT-83-2, 1983.
AFDD Weight Models 267

15.

10.

5.
error (%)

0.

-5.

-10.

-15.
0. 10000. 20000. 30000. 40000. 50000. 60000.
actual weight

Figure 29-1. Wing group (AFDD93).

25.
20.

15.
10.
5.
error (%)

0.

-5.
-10.
-15.
-20.
-25.
0. 500. 1000. 1500. 2000. 2500. 3000. 3500.

actual weight

Figure 29-2. Rotor group, blade weight (AFDD82).

.
268 AFDD Weight Models

25.
20.

15.
10.
5.
error (%)

0.

-5.
-10.
-15.
-20.

-25.
0. 1500. 3000. 4500. 6000. 7500.
actual weight

Figure 29-3. Rotor group, hub weight (AFDD82).

25.
20.

15.
10.
5.
error (%)

0.

-5.
-10.
-15.
-20.
-25.
0. 1500. 3000. 4500. 6000. 7500.

actual weight

Figure 29-4. Rotor group, blade weight (AFDD00).

AFDD Weight Models 269

25.
using actual blade weight
20.
using calculated blade weight
15.
10.
5.
error (%)

0.

-5.
-10.
-15.
-20.

-25.
0. 1500. 3000. 4500. 6000. 7500.
actual weight

Figure 29-5. Rotor group, hub weight (AFDD00).

50.
40.

30.
20.
10.
error (%)

0.

-10.
-20.
-30.
-40.
-50.
0. 10. 20. 30. 40. 50. 60. 70. 80. 90. 100. 110. 120.

actual weight

Figure 29-6. Empennage group, horizontal tail weight (AFDD82).

270 AFDD Weight Models

50.
40.

30.
20.
10.
error (%)

0.

-10.
-20.
-30.
-40.

-50.
0. 10. 20. 30. 40. 50. 60. 70. 80. 90. 100. 110. 120.
actual weight

Figure 29-7. Empennage group, vertical tail weight (AFDD82).

50.
40.

30.
20.
10.
error (%)

0.

-10.
-20.
-30.
-40.
-50.
0. 50. 100. 150. 200. 250. 300. 350. 400. 450. 500.

actual weight

Figure 29-8. Empennage group, tail-rotor weight (AFDD82).

AFDD Weight Models 271

30.
rotorcraft
fixed wing
20.
other RC
other FW
10.
error (%)

0.

-10.

-20.

-30.
0. 2500. 5000. 7500. 10000. 12500. 15000.
actual weight

Figure 29-9. Fuselage group, fuselage weight (AFDD84).

50.
40. rotorcraft
fixed wing
30.
20.
10.
error (%)

0.

-10.
-20.
-30.
-40.
-50.
0. 5000. 10000. 15000. 20000. 25000. 30000.

actual weight

Figure 29-10. Fuselage group, fuselage weight (AFDD82).

272 AFDD Weight Models

50. rotorcraft
40. fixed wing
other RC
30.
20.
10.
error (%)

0.

-10.
-20.
-30.
-40.

-50.
0. 2000. 4000. 6000. 8000. 10000. 12000.
actual weight

Figure 29-11. Alighting gear group, landing gear weight (AFDD82).

25.
20.

15.
10.
5.
error (%)

0.

-5.
-10.
-15.
-20.
-25.
0. 50. 100. 150. 200. 250. 300.

actual weight

Figure 29-12. Engine section or nacelle group, engine support weight (AFDD82).
AFDD Weight Models 273

30.
25.
20.
15.
10.
5.
error (%)

0.
-5.
-10.
-15.
-20.
-25.
-30.
0. 50. 100. 150. 200. 250. 300. 350. 400. 450.
actual weight

Figure 29-13. Engine section or nacelle group, cowling weight (AFDD82).

25.
20.

15.
10.
5.
error (%)

0.

-5.
-10.
-15.
-20.
-25.
0. 50. 100. 150. 200. 250. 300.

actual weight

Figure 29-14. Air induction group, air induction weight (AFDD82).

274 AFDD Weight Models

25.
20.

15.
10.
5.
error (%)

0.

-5.
-10.
-15.
-20.

-25.
0. 50. 100. 150. 200. 250. 300. 350. 400. 450.
actual weight

Figure 29-15. Propulsion group, accessories weight (AFDD82).

25.
20.

15.
10.
5.
error (%)

0.

-5.
-10.
-15.
-20.
-25.
0. 200. 400. 600. 800. 1000. 1200. 1400. 1600.

actual weight

Figure 29-16. Propulsion group, fuel tank weight (AFDD82).

AFDD Weight Models 275

25.
20.

15.
10.
5.
error (%)

0.

-5.
-10.
-15.
-20.

-25.
0. 1000. 2000. 3000. 4000. 5000. 6000. 7000. 8000. 9000. 10000.
actual weight

Figure 29-17. Propulsion group, gear box and rotor shaft weight (AFDD83).

25.
20.

15.
10.
5.
error (%)

0.

-5.
-10.
-15.
-20.
-25.
0. 1000. 2000. 3000. 4000. 5000. 6000. 7000. 8000. 9000. 10000.

actual weight

Figure 29-18. Propulsion group, gear box and rotor shaft weight (AFDD00).
276 AFDD Weight Models

50.
40.

30.
20.
10.
error (%)

0.

-10.
-20.
-30.
-40.

-50.
0. 100. 200. 300. 400. 500. 600. 700.
actual weight

Figure 29-19. Propulsion group, drive shaft weight (AFDD82).

75.
60.

45.
30.
15.
error (%)

0.

-15.
-30.
-45.
-60.
-75.
0. 20. 40. 60. 80. 100. 120. 140. 160. 180. 200.

actual weight

Figure 29-20. Propulsion group, rotor brake weight.

AFDD Weight Models 277

25.
20.

15.
10.
5.
error (%)

0.

-5.
-10.
-15.
-20.

-25.
0. 50. 100. 150. 200. 250. 300. 350. 400. 450. 500.
actual weight

Figure 29-21. Flight controls group, rotor non-boosted control weight (AFDD82).

50.
40.

30.
20.
10.
error (%)

0.

-10.
-20.
-30.
-40.
-50.
0. 100. 200. 300. 400. 500. 600. 700.

actual weight

Figure 29-22. Flight controls group, rotor boost mechanisms weight (AFDD82).
278 AFDD Weight Models

50.
40.

30.
20.
10.
error (%)

0.

-10.
-20.
-30.
-40.

-50.
0. 200. 400. 600. 800. 1000. 1200. 1400. 1600. 1800. 2000. 2200.
actual weight

Figure 29-23. Flight controls group, rotor boosted control weight (AFDD82).

all parametric weight

25. structure group
propulsion group
20.
flight control group
15.

10.
5.
error (%)

0.
-5.
-10.
-15.
-20.
-25.
0. 5000. 10000. 15000. 20000. 25000. 30000. 35000. 40000.

weight

Figure 29-24. Sum of all parametric weight.

 Chapter 30

Other Weight Models

This chapter presents aircraft weight models developed by a number of organizations: Boeing
(refs. 1 to 4), GARTEUR (Aerospatiale and MBB, ref. 5), Tishchenko (refs. 6 and 3), Torenbeek (ref. 7),
and Raymer (ref. 8). The weights are estimated from parametric equations, based on the weights of
existing aircraft. In addition, generic weight estimation equations are implemented for some groups.

The results of these equations are the weight in pounds, and the units of the parameters are noted
in the tables. Technology factors χ are included in the weight equations. The input usually includes a
weight increment dW that can be added to the results of the weight model. Thus typically a component
or element weight is obtained from W = χwmodel + dW . Weight of individual elements in a group can
be fixed by using dW and setting the corresponding technology factor χ = 0. With χ = 0, the increment
dW can account for something not included in the parametric model.

30–1 Wing Group

The wing group consists of: basic structure (primary structure, consisting of torque box and spars,
plus extensions); fairings (leading edge and trailing edge); fittings (non-structural); fold/tilt structure;
and control surfaces (flaps, ailerons, flaperons, and spoilers).

The total wing weight is:

Boeing
- 0 .0.585
√ 1 + λw
wwing = 0.0680 nz WD fL Sw log(bw /wf ) log Vdive log Aw
2τw
GARTEUR
 0.4
0.4 0.52 0.7 0.47 2 0.695
wwing = 0.00715fT R (1 + 0.12fflap ) (nz WD fL ) Sw Aw (1.1 + λw /2)
τw

The factor fT R = 1.116 for tiltrotors, 1.0 otherwise.

Torenbeek (light aircraft, up to 12500 lb)

 0.75    b / cos Λ t 0.3
bw  w w w
wwing = 0.00125(n0.55
z WM T O fL ) 1+ 6.25 cos Λw /bw
cos Λw WM T O fL /Sw
Torenbeek (transport catagory aircraft)
 0.75    b / cos Λ t 0.3
bw  w w w
wwing = 0.0017(n0.55
z WM ZF fL ) 1+ 6.25 cos Λw /bw
cos Λw WM ZF fL /Sw
280 Other Weight Models

Raymer (cargo/transport)
(1 + λw )0.1 Sc0.1
wwing = 0.0051(nz WD fL )0.557 Sw
0.649 0.5
Aw
τw0.4 cos Λw
Raymer (general aviation)
λ0.04 0.012
w Vdive
wwing = 0.0087(nz WD fL )0.49 Sw
0.758 0.6
Aw
τw cos Λ0.9
0.3
w

The lift factor fL accounts for wing relief factor and wing/rotor lift share.
Then the primary structure, secondary structure, and control surface weights are calculated as for
the AFDD aircraft wing model. The wing extension weights are calculated as for the AFDD model.
Parameters are defined in table 30-1, including units as used in these equations.

Table 30-1. Parameters for wing weight.

parameter definition units
WD design gross weight lb
WM T O maximum takeoff weight lb
WM ZF maximum zero fuel weight lb
fL lift factor
Sw wing planform area (theoretical) ft2
bw wing span ft
Aw wing aspect ratio
wf maximum fuselage width ft
λw wing taper ratio (tip chord/root chord)
Λw wing sweep angle deg
τw wing root airfoil thickness-to-chord ratio
tw wing root airfoil thickness ft
nz design ultimate flight load factor g
Vdive dive speed kts
Sc control surface area ft2
fflap ratio maximum lift coefficient with and without flaps

30–2 Rotor Group

The rotor group consists of: blades, hub and hinge, spinner, and blade fold structure. The blade
and hub-hinge weights are:

Boeing
" %0.438
wblade = 0.0378Nrotor (nz WD ftb /Nrotor )R2 (R − r)Nblade cfb fdr
" %0.358
whub = 0.00704Nrotor (Wblade /Nrotor )RΩ2 (Pto fth /Nrotor )r1.82 Nblade
1.5
famd

with fb = 1.0 for articulated rotor, 2.2 hingeless or teetering rotor; ftb = 0.247 and fth = 0.577 for
tandem, 1.0 single main rotor; famd = fa fm fd , design concept factor fa = 0.53 for hingeless rotor and
1.0 otherwise, material factor fm = 1.0 for steel and 0.56 titanium and 0.354 composite, development
factor fd = 1.0 early, 0.62 developed. The droop factor is fdr = max(1, R1.6 /(1200t0.25 )). The blade
attachment radius is typically r/R = 0.08 to 0.11.
Other Weight Models 281

GARTEUR
wblade = 4.199Nrotor froot fcb (RNblade c)
" %0.36
2
whub = 0.004111Nrotor (Wblade /Nrotor )Vtip (Pto /Nrotor )R0.82 Nblade
1.5

with froot = 1.17 for hingeless and teetering blades, 0.88 for composite articulated blades. The original
model is extended to other configurations by introducing fcb = 0.56 for tandem helicopter, 0.65 for
tiltrotor.
Tishchenko
" %
wblade = 1.23Nrotor fcb (σR2.7 /λ̄0.7 ) 1 + 0.011R̄(λ − λ0 )
" %
whub = 0.0000777Nrotor fch Nblade (CFblade )1.35 1 + 0.05(Nblade − 4)

where λ̄ = λ/18, blade aspect ratio λ = R/c0.7 , R̄ = R/52.5, R̄λ0 = 12.4, and the last factor in the
blade weight equation has a minimum value of 1. For the hub weight, the centrifugal force per blade
!R
is CFblade = 0 Ω2 rm dr ∼ = Ω2 21 R(Mblade /(Nblade Nrotor )). The original model is extended to other
configurations by introducing fcb = 0.45 for tandem helicopter, 0.80 for tiltrotor; and fch = 1.11 for
tandem helicopter, 0.34 for tiltrotor, 0.53 for teetering rotor, 0.67 for hingeless rotor.
generic
XbldN XbldR Xbldc XbldV Xbldν
wblade = Kblade Nrotor Nblade R c Vtip νblade (WM T O /Nrotor )XbldW
XhubN XhubR Xhubc XhubV Xhubν
whub = Khub Nrotor Nblade R c Vtip νhub (WM T O /Nrotor )XhubW
using system units. If the the rotor is not a main rotor, Tdesign is used for WM T O /Nrotor .
The blade weight is for all blades of the rotors. If the weight is evaluated separately for each rotor,
then the leading Nrotor = 1 in these equations (with blade weight, design gross weight, and takeoff
power per rotor). The fairing/spinner, blade fold, rotor support, and duct weights are calculated as for
the AFDD model. Parameters are defined in table 30-2, including units as used in these equations.

Table 30-2. Parameters for rotor weight.

parameter definition units
Nrotor number of rotors
Nblade number of blades per rotor
R rotor radius in hover ft
r center of rotation to blade attachment ft
c rotor mean geometric blade chord ft
σ rotor solidity
t0.25 blade thickness at 25%R ft
WD design gross weight lb
nz design limit flight load factor g
Ω rotor rotation speed rpm
Vtip hover tip speed ft/sec
Pto takeoff power hp
νblade , νhub flap natural frequency (for weight estimate) per rev
WM T O maximum takeoff weight
Tdesign rotor design thrust
282 Other Weight Models
30–3 Empennage Group

The empennage group consists of: horizontal tail, vertical tail, and tail-rotor. The tail plane weight
consists of the basic structure and fold structure. The horizontal tail weight is:
GARTEUR
2.44 0.35
Wht = χht 0.00000518Vmax WD

Torenbeek (Vdive up to 250 knots)

 2
0.75
Wtail = χht 0.04 nz Stail
Torenbeek (transport catagory aircraft)
  
0.2
Wht = χht fh Sht 0.00346 Sht Vdive / cos Λht
where fh = 1.0 for fixed stabilizer, 1.1 for variable-incidence tail.
Raymer (cargo/transport)
(1 + Se /Sht )0.1
Wht = χht 0.0379fh n0.10
z
0.639 0.75 0.166 0.704
WD Sht Aht Ky
(1 + wf /bht )0.25 Lt cos Λht
Raymer (general aviation)
0.336
Vdive cos Λ0.034
Wht = χht 0.00354(nz WD )0.414 Sht
0.896 0.043
Aht 0.02
ht
0.12
λht τht
where fh = 1.143 for all-moving horizontal tail, 1.0 otherwise.
The vertical tail weight is:
GARTEUR
0.508 0.647
Wvt = χvt 0.0253Svt WD

Torenbeek (Vdive up to 250 knots)

 2
0.75
Wtail = χvt 0.04 nz Stail
Torenbeek (transport catagory aircraft)
  
0.2
Wvt = χvt fv Svt 0.00346 Svt Vdive / cos Λvt
where fv = 1.0 for fuselage-mounted horizontal tail, fv = 1 + 0.15(Sht ht )/(Svt bvt ) for fin-mounted
stabilizer (T-tail).
Raymer (cargo/transport)
1
Wvt = χvt 0.0026fv1 n0.536
z
0.556 0.5 0.35 0.875
WD Svt Avt Kz 0.5 L0.5 cos Λ
τroot t vt
Raymer (general aviation)
0.244
Vdive
Wvt = χvt 0.00382fv2 (nz WD )0.376 Svt
0.873 0.357
Avt
λvt τvt cos Λ0.224
0.039 0.49
vt
where fv1 = 1.0 for conventional tail, 1.169 for T-tail; fv2 = 1.0 for conventional tail, 1.2 for T-tail.
The tail-rotor weight is:
Boeing
" %0.67
Wtr = χtr 0.0297 r0.25 Ptr
0.5
Vtip RNblade c
GARTEUR
 1.372
0.556
Wtr = χtr 12.48WD Rστ0.7
Other Weight Models 283

Fold weight is calculated as for the AFDD model. Parameters are defined in table 30-3, including units
as used in these equations.

Table 30-3. Parameters for tail weight.

parameter definition units
Vmax maximum speed kts
Vdive dive speed kts
S tail planform area ft2
b tail span
A tail aspect ratio
Λ sweep angle deg
λ taper ratio (tip chord/root chord)
ht /bvt height horizontal tail above fin root (fraction span)
Ptr tail rotor transmission power limit hp
Vtip hover tip speed ft/sec
R radius ft
Nblade number of blades
c mean chord ft
r rotor blade attachment radius ft
σ solidity
τ0.7 airfoil thickness-to-chord ratio at 0.7R
WD design gross weight lb
nz design ultimate flight load factor g
wf /bht fuselage width at horizontal tail (fraction span)
Lt tail length (wing to horizontal tail) ft
Sc control surface area ft2
Ky aircraft pitch radius of gyration ft
Kz aircraft yaw radius of gyration ft

30–4 Fuselage Group

The fuselage group consists of: basic structure; wing and rotor fold/retraction; tail fold/tilt; and
marinization, pressurization, and crashworthiness structure. The basic structure weight is

Boeing
" %0.8
0.5
wbasic = 0.198 (nz WSD Sbody ( c + r + ΔCG)) log Vdive

where ( c + r + ΔCG)/ body

∼
= 1.0 for tandem, and 0.3–0.6 for single main rotor (0.7–0.8 with ramp).

GARTEUR (airplane)
0.44
wbasic = 0.149 (nz WD Sbody body )

GARTEUR (helicopter)
0.67 0.335
wbasic = 0.0280RWD nz
Tishchenko
0.25 0.88 0.16(1+α)
wbasic = 0.251WD Sbody Lrr
284 Other Weight Models

where α = 0.0 for single main rotor, 0.2 for tandem helicopter, 0.05 for side-by-side.
Torenbeek
 0.5
1.2 Lt
wbasic = 0.021Sbody Vdive
wbody + hbody
Raymer (cargo/transport)
wbasic = 0.3280fdoor fLG (nz WD )0.5 Sbody
0.302 0.25 0.04
body xw ( body /hbody )
0.10

Raymer (general aviation)

0.482
Vdive
wbasic = 0.0103(nz WD )0.177 Sbody
1.086
0.051
Lt ( body /hbody )0.072
 0.271
wpress = 11.9 Vpress Pdelta
where fdoor = 1.0 for no cargo door, 1.06 for one side cargo door, 1.12 for two side cargo doors or aft
clamshell door, 1.25 for two side cargo doors and aft clamshell door (fdoor = 1+.06ndoor ); fLG = 1.12 for
fuselage-mounted main landing gear, 1.0 otherwise; and xw = 1+0.75[(1+2λw )/(1+λw )]bw tan Λw / body .
For the weight penalty due to pressurization, Vpress ∼= body wbody hbody , and typically Pdelta = 8 psi.
generic
 XfusW  Xfusn
WM T O nz WSD XfusS Xfus
wbasic = Kfus Sbody body
1000 1000
using system units.
The tail fold, wing and rotor fold, marinization, pressurization, and crashworthiness weights are
calculated as for the AFDD84 model. Parameters are defined in table 30-4, including units as used in
these equations.

Table 30-4. Parameters for fuselage weight.

parameter definition units
WM T O maximum takeoff weight
WSD structural design gross weight lb
WD design gross weight lb
nz design ultimate flight load factor at WSD g
Sbody wetted area of body ft2
c length of cabin (nose to end of cabin floor) ft
r length of rampwell ft
ΔCG center of gravity range at WD ft
Vdive dive speed kts
body fuselage length ft
wbody fuselage width ft
hbody fuselage height ft
Lrr rotor-rotor longitudinal separation ft
Lt tail length (wing to horizontal tail) ft
ndoor number of cargo doors
Λw wing sweep angle deg
λw wing taper ratio (tip chord/root chord)
bw wing span
Vpress pressurized volume ft3
Pdelta cabin pressure differential psi
Other Weight Models 285
30–5 Engine Section or Nacelle Group and Air Induction Group

The engine section or nacelle group consists of: engine support structure, engine cowling, and
pylon support structure. The weights (engine mounts and nacelle structure) are:

Boeing
" %0.41
Wsupt = χsupt Neng (Weng /Neng )nclf
Wcowl = χcowl 1.25Snac

The crash load factor is typically nclf = 8g for civil rotorcraft, 20g for military rotorcraft.

Raymer (cargo/transport)
0.10 0.294 0.119 0.611 0.224
Wcowl = χcowl 0.6724fn nac wnac nz Weng Snac
where fn = 1.017 for pylon-mounted nacelle, 1.0 otherwise. The Raymer model includes the air
induction group weight. The nacelle length and width are calculated from f = wnac / nac , assuming
Snac = πwnac nac . Parameters are defined in table 30-5, including units as used in these equations.

Table 30-5. Parameters for engine section, nacelle, and air induction weight.
parameter definition units
Weng weight all main engines lb
Neng number of main engines
Snac wetted area of nacelles and pylon (less spinner) ft2
nac nacelle length ft
wnac nacelle width ft
nz design ultimate flight load factor g

30–6 Propulsion Group

The propulsion group consists of the engine system, fuel system, and drive system.

30-6.1 Propeller/Fan Installation

The auxiliary propulsion or propeller weight is:

Boeing
" %0.67
Wat = χat 0.0297Nat r0.25 Pat
0.5
Vtip RNblade c
GARTEUR
Wat = χat 0.5320Nat R1.544 Nblade
0.471 0.376
Pat
Torenbeek
 0.5
0.78174
Wat = χat 0.186Nat fe RPat Nblade

where fe = 1.0 for turboprop, 1.333 for reciprocating engine. The propeller weight is calculated by the
rotor component model, including rotor support and duct weights. Rotor weight models can be used for
286 Other Weight Models

the propeller. Parameters are defined in table 30-6, including units as used in these equations.

Table 30-6. Parameters for propeller weight.

parameter definition units
Nat number of auxiliary thrusters
Pat power per propeller hp
Vtip hover tip speed ft/sec
R radius ft
Nblade number of blades
c mean chord ft
r rotor blade attachment radius ft

30-6.2 Fuel System

The fuel system consists of tanks and support structure (including fuel tanks, bladders, supporting
structure, filler caps, tank covers, and filler material for void and ullage), and fuel plumbing (including
fuel system weight not covered by tank weight). The fuel system weights are:

Torenbeek (integral tanks)

" %
Wtank = χtank 80(Neng + Nint − 1) + 15Cint
0.333 0.5
Nint
Torenbeek (generic)
Xtank
Wtank = χtank Ktank Cint

with Ktank = 2 and Xtank = 0.667 for single-engine reciprocating; Ktank = 4.5 and Xtank = 0.60 for
multi-engine reciprocating; Ktank = 3.2 and Xtank = 0.727 for bladder tanks, turbofan or turboprop.

Raymer (cargo/transport)
0.606 0.5 1 + fp
Wtank = χtank 2.405Cint Nint
1 + fi
Raymer (general aviation)
Nint Neng (1 + fi )−0.363
0.726 0.242 0.157
Wtank = χtank 2.49Cint

Parameters are defined in table 30-7, including units as used in these equations.

Table 30-7. Parameters for fuel system weight.

parameter definition units
Nint number of internal fuel tanks
Cint internal fuel tank capacity gallons
fi integral tank capacity (fraction total)
fp protected (self-sealing) tank capacity (fraction total)
Neng number of main engines
Other Weight Models 287

30-6.3 Drive System

The drive system consists of gear boxes and rotor shafts, drive shafts, and rotor brake. This
distribution of drive system weights is based on the following functional definitions. Gearboxes are
parts of the drive system that transmit power by gear trains, and the structure that encloses them. Rotor
shafts are the structure (typically a shaft) that transmits power to the rotor. Drive shafts are the structure
(typically a shaft) that transmits power in the propulsion system, but not directly to the rotor or by a gear
train. The rotor brake weight encompasses components that can prevent the rotor from freely turning.
The gear box and rotor shaft weights are:
Boeing
" %0.67
0.25
wgbrs = 250 (PDSlimit /Ωmr )zmr ft
Boeing (alternate)
" %0.8
wgbrs = 300 1.1P/Ω
where ft = 1.0 for single main rotor, ft = 1.3 × 1.2 = 1.56 for tandem main rotors. The number of
drive stages is typically zmr ∼
= WD /5000. For the alternate model applied to the entire drive system,
P/Ω = PDSlimit /Ωmr . The alternate model is also applied to the tail rotor only, with P/Ω = Ptr /Ωtr .
GARTEUR (helicopter)
 0.74
wgbrs = 252.3 PDSlimit /Ωmr
GARTEUR (tiltrotor)
 0.8
wgbrs = 345.6 PDSlimit /Ωmr
Tishchenko
 0.8  0.8
wgbrs = 199.3 PDSlimit /Ωmr + 643 Ptr /Ωtr
with Ptr = ftr PDSlimit , typically ftr ∼
= 10%.
generic
X X
gsP
wgbrs = Kgbrs PDSlimit ΩX gsr
eng Ωrotor
gse

using system units. These models are all implemented with an optional tail rotor drive weight increment
Δwgbrs = Ktrgb 270(Ptr /Ωtr )0.8 . Then
Wgb = χgb (1 − frs )wgbrs
Wrs = χrs frs wgbrs
Parameters are defined in table 30-8, including units as used in these equations.

Table 30-8. Parameters for drive system weight.

parameter definition units
PDSlimit drive system power limit hp
Ωmr main-rotor rotation speed rpm
Ωrotor main-rotor rotation speed rpm
Ωeng engine output speed rpm
zmr number of stages in main-rotor drive
Ptr tail rotor drive system power limit hp
Ωtr tail-rotor rotation speed rpm
ftr tail-rotor power (fraction main-rotor power)
288 Other Weight Models

30–7 Flight Controls Group and Hydraulic Group

The flight controls group consists of cockpit controls, automatic flight control system, and system
controls. System controls consist of fixed wing flight controls, rotary wing flight controls, and conversion
(rotor tilt) flight controls. The weight equations model separately non-boosted controls (which do not
see aerodynamic surface or rotor loads), boost mechanisms (actuators), and boosted controls (which are
affected by aerodynamic surface or rotor loads). The load path goes from pilot, to cockpit controls,
to non-boosted controls, to boost mechanisms, to boosted controls, and finally to the component. The
hydraulic group consists of hydraulics for fixed wing flight controls, rotary wing flight controls, and
equipment.
Fixed wing flight controls consist of non-boosted flight controls and flight control boost mechanisms.
The weights are:

GARTEUR
 −0.36
w = 0.0350WD WD /Sw
Raymer (cargo/transport)
Nf0.554 0.20 0.07
w = 55.47 S I
1 + fm cs y
WF W hyd = χF W hyd 0.2673Nf ( body + bw )0.937
Raymer (general aviation)
w = 0.053(nz WD /10000)0.8 1.536 0.371
body bw

generic
XF W
w = KF W WM TO

with typically Nf = 4 to 7, fm Nf = 0 to 2. Then

WF W nb = χF W nb fF W nb w
WF W mb = χF W mb (1 − fF W nb ) w
WF W hyd = χF W hyd fF W hyd WF W mb

as for the AFDD model. Parameters are defined in table 30-9, including units as used in these equations.

Table 30-9. Parameters for fixed wing flight control weight.

parameter definition units
WM T O maximum takeoff weight
WD design gross weight lb
Sw wing planform area (theoretical) ft2
Nf number of functions performed by controls
fm number of mechanical functions (fraction Nf )
Scs total area of control surfaces ft2
Iy yaw moment of inertia (WD Ky2 ) lb-ft2
body fuselage length ft
bw wing span ft
Other Weight Models 289

Rotary wing flight controls consist of non-boosted flight controls, flight control boost mechanisms,
and boosted flight controls. The Boeing weight equations model the main rotor controls (from and
including the power actuators, swashplate, up through the pitch links; identified as boost mechanisms
plus boosted controls) and the rotor systems (between cockpit controls and rotor controls; identified as
non-boosted controls) and hydraulics. The weights are:

Boeing
" %0.41
Wcc = χcc 1.7 WD
" %1.11
0.5
wmb+b = 0.562Nrotor c (RWblade /Nrotor )
" %0.84
wnb+hyd = 0.0906Nrotor WD /Nrotor

WRW mb = χRW mb (1 − fRW b )wmb+b

WRW b = χRW b fRW b wmb+b
WRW nb = χRW nb (1 − fRW hyd )wnb+hyd
WRW hyd = χRW hyd fRW hyd wnb+hyd

or
GARTEUR
1.0136
wf c = 0.0324WD
Tishchenko
wf c = 0.824Nrotor Nblade c2 R
generic
wf c = KRW (Nrotor Nblade )XRW N RXRW R cXRW c (WM T O /Nrotor )XRW W (Wblade /Nrotor )XRW b

WRW mb = χRW mb fRW mb wf c

WRW b = χRW b fRW b wf c
WRW nb = χRW nb (1 − fRW mb − fRW b − fRW hyd )wf c
WRW hyd = χRW hyd fRW hyd wf c
using system units for the generic model. Parameters are defined in table 30-10, including units as used
in these equations.
Table 30-10. Parameters for rotary wing flight control weight.
parameter definition units
WM T O maximum takeoff weight
WD design gross weight lb
c rotor mean blade chord ft
R rotor radius ft
Wblade weight rotor blades lb
Nrotor number of main rotors
Nblade number of blades
290 Other Weight Models
30–8 Parametric Weight Correlation

Tables 30-11 and 30-12 summarize the statistics of the parametric weight estimation equations, for
the AFDD data base and a data base from Boeing (ref. 4), respectively. The technology factor χ was
adjusted for each equation to minimize the average error. Except for the wing group, the aircraft in the
data base are all helicopters, in particular for the fuselage group and empennage group. The fuel system
weight is for just the tanks; the technology factors are about 70% larger for both tanks and plumbing.
The drive system weight includes the tail rotor.

30–9 References

1) Davis, S.J., and Wisniewski, J.S. “User’s Manual for HESCOMP, The Helicopter Sizing and Perfor-
mance Computer Program.” NASA CR 152018, September 1973.
2) Davis, S.J.; Rosenstein, H.; Stanzione, K.A.; and Wisniewski, J.S. “User’s Manual for HESCOMP,
The Helicopter Sizing and Performance Computer Program.” NADC Report No. 78265-60, October
1979.
3) Stepniewski, W.Z., and Shinn, R.A. “A Comparative Study of Soviet vs. Western Helicopters.”
NASA CR 3579, CR 3580, March 1983.
4) Wisniewski, J.S. “The Boeing Model 360 Advanced Technology Demonstrator Helicopter.” SAWE
Paper No. 2050, May 1992.
5) Esculier, J.; Lambert, O.; Gmelin, B.; Humpherson, D.; Jones, A.; Lepretre, A.; Philippe, J.-J.; Rauen,
A.; Schmidt, U.; and Stevens, J. “Advanced Rotorcraft Evaluation. Preliminary Design Study of Tilt
Rotor Aircraft and Compound Helicopter.” GARTEUR Report No. TP-036 (HC-AG 05), November
1987. Schmidt, U. “Tilt Rotor and Compound Helicopter Mass Modeling.” Annex C
6) Tishchenko, M.N.; Nagaraj, V.T.; and Chopra, I. “Preliminary Design of Transport Helicopters.”
Journal of the American Helicopter Society, 48:2 (April 2003).
7) Torenbeek, E. Synthesis of Subsonic Airplane Design. Delft, Holland: Delft University Press, 1982.
8) Raymer, D.P. Aircraft Design: A Conceptual Approach. Fourth Edition. Reston, Virginia: American
Institute of Aeronautics and Astronautics, 2006.
Other Weight Models 291

Table 30-11. Statistics of parametric weight equations, AFDD data base.

number of average technology
model aircraft error (%) factor
wing group
Boeing 23 14.1 χ = 0.73
GARTEUR 23 20.9 χ = 1.07
Torenbeek 23 20.3 χ = 1.04
Raymer (cargo/transport) 23 16.9 χ = 1.24
Raymer (general aviation) 23 16.8 χ = 1.31
rotor group, blade
Boeing 51 10.1 χ = 0.90
GARTEUR 51 11.8 χ = 0.98
Tishchenko 51 18.8 χ = 0.99
rotor group, hub
Boeing 45 27.5 χ = 1.07
GARTEUR 45 27.6 χ = 1.34
Tishchenko 51 22.8 χ = 0.89
horizontal tail
GARTEUR 13 37.3 χ = 0.65
Torenbeek (transport) 13 24.7 χ = 1.80
vertical tail
GARTEUR 12 26.7 χ = 0.79
Torenbeek (transport) 12 29.2 χ = 1.92
tail rotor
Boeing 18 28.5 χ = 0.64
GARTEUR 18 18.0 χ = 0.67
fuselage group
GARTEUR (airplane) 28 13.0 χ = 0.89
GARTEUR (helicopter) 28 21.5 χ = 0.94
Tishchenko 28 15.0 χ = 0.90
engine section or nacelle group
Boeing (support) 14 24.9 χ = 1.30
Boeing (cowl) 17 23.0 χ = 0.84
Raymer (cargo/transport) 16 23.7 χ = 0.64
propeller
Boeing 13 27.9 χ = 1.13
GARTEUR 13 30.0 χ = 1.18
Torenbeek 13 24.4 χ = 0.77
fuel system
Raymer (cargo/transport) 34 24.8 χ = 1.75
Raymer (general aviation) 34 27.2 χ = 1.10
Torenbeek (integral tanks) 32 29.2 χ = 0.60
Torenbeek (generic) 34 29.3 χ = 0.90
drive system
Boeing (alternate) 81 12.1 χ = 0.61
GARTEUR (helicopter) 81 15.6 χ = 0.90
GARTEUR (tiltrotor) 2 4.3 χ = 0.75
Tishchenko 81 12.1 χ = 0.99
tail rotor 24 19.5 Ktrgb = 1.00
rotary wing flight controls
Boeing 24 19.0 χ = 0.91
GARTEUR 23 20.1 χ = 0.81
Tishchenko 22 32.8 χ = 1.35
292 Other Weight Models

Table 30-11. Statistics of parametric weight equations, Boeing data base.

number of average technology
model aircraft error (%) factor
rotor group, blade
Boeing 20 6.0 χ = 1.05
GARTEUR 20 12.8 χ = 0.93
Tishchenko 20 16.3 χ = 0.90
rotor group, hub
Boeing 20 10.6 χ = 0.93
GARTEUR 18 11.3 χ = 1.28
Tishchenko 19 20.5 χ = 0.99
fuselage group
Boeing 26 12.9 χ = 1.04
GARTEUR (airplane) 23 16.6 χ = 0.89
GARTEUR (helicopter) 26 16.5 χ = 0.98
Tishchenko 23 12.9 χ = 0.87
propeller
Boeing 12 19.1 χ = 0.95
GARTEUR 12 18.3 χ = 0.97
Torenbeek 12 21.0 χ = 0.51
drive system
Boeing 25 6.3 χ = 0.98
Boeing (alternate) 27 10.2 χ = 0.73
GARTEUR (helicopter) 27 14.4 χ = 1.07
Tishchenko 27 10.2 χ = 1.18