NASA/TM-20220015470

Design of an Open Rotor, Braced-Wing Electric Transport

Jeffrey J. Berton
Glenn Research Center, Cleveland, Ohio

December 2022
 NASA STI Program . . . in Profile

Since its founding, NASA has been dedicated • CONTRACTOR REPORT. Scientific and
to the advancement of aeronautics and space science. technical findings by NASA-sponsored
The NASA Scientific and Technical Information (STI) contractors and grantees.
Program plays a key part in helping NASA maintain
this important role. • CONFERENCE PUBLICATION. Collected
papers from scientific and technical
conferences, symposia, seminars, or other
The NASA STI Program operates under the auspices
meetings sponsored or co-sponsored by NASA.
of the Agency Chief Information Officer. It collects,
organizes, provides for archiving, and disseminates • SPECIAL PUBLICATION. Scientific,
NASA’s STI. The NASA STI Program provides access technical, or historical information from
to the NASA Technical Report Server—Registered NASA programs, projects, and missions, often
(NTRS Reg) and NASA Technical Report Server— concerned with subjects having substantial
Public (NTRS) thus providing one of the largest public interest.
collections of aeronautical and space science STI in
the world. Results are published in both non-NASA • TECHNICAL TRANSLATION. English-
channels and by NASA in the NASA STI Report language translations of foreign scientific and
Series, which includes the following report types: technical material pertinent to NASA’s mission.

• TECHNICAL PUBLICATION. Reports of For more information about the NASA STI
completed research or a major significant phase program, see the following:
of research that present the results of NASA
programs and include extensive data or theoretical • Access the NASA STI program home page at
analysis. Includes compilations of significant http://www.sti.nasa.gov
scientific and technical data and information
deemed to be of continuing reference value. • E-mail your question to help@sti.nasa.gov
NASA counter-part of peer-reviewed formal
• Fax your question to the NASA STI
professional papers, but has less stringent
Information Desk at 757-864-6500
limitations on manuscript length and extent of
graphic presentations.
• Telephone the NASA STI Information Desk at
757-864-9658
• TECHNICAL MEMORANDUM. Scientific
and technical findings that are preliminary or of • Write to:
specialized interest, e.g., “quick-release” reports, NASA STI Program
working papers, and bibliographies that contain Mail Stop 148
minimal annotation. Does not contain extensive NASA Langley Research Center
analysis. Hampton, VA 23681-2199
NASA/TM-20220015470

Design of an Open Rotor, Braced-Wing Electric Transport

Jeffrey J. Berton
Glenn Research Center, Cleveland, Ohio

National Aeronautics and

Space Administration

Glenn Research Center

Cleveland, Ohio 44135

December 2022
 Acknowledgments

This study was performed with support from NASA’s Advanced Air Transport Technology Project of the Advance Air Vehicles
Program. Special thanks go to William Haller, Jesse Quinlan, and Eric Hendricks of NASA for establishing the project’s
advanced concept studies, and to Mark Guynn and James Felder of NASA for reviewing this manuscript.

This report is a formal draft or working

paper, intended to solicit comments and
ideas from a technical peer group.

This report contains preliminary findings,

subject to revision as analysis proceeds.

This work was sponsored by the Advanced Air Vehicle Program

at the NASA Glenn Research Center

Trade names and trademarks are used in this report for identification
only. Their usage does not constitute an official endorsement,
either expressed or implied, by the National Aeronautics and
Space Administration.

Level of Review: This material has been technically reviewed by technical management.

This report is available in electronic form at https://www.sti.nasa.gov/ and https://ntrs.nasa.gov/

 Design of an Open Rotor, Braced-Wing Electric Transport
Jeffrey J. Berton
National Aeronautics and Space Administration
Glenn Research Center
Cleveland, Ohio 44135

Abstract
A notional, 19-passenger, battery-powered, fully electric airplane is investigated in this study. It is
designed to serve the regional aviation market of the near future. Three strategies are used to reduce
energy demands placed on its batteries: 1) a high aspect ratio, braced wing, 2) two efficient open rotor
propellers, and 3) flying at comparatively low cruise speeds. It has an otherwise conventional airframe
architecture. Parametric system weight, aerodynamic, propeller, and mission performance models are
developed. Single-rotation and contra-rotation propellers are studied. Implementation of a novel selective
noise reduction system is also investigated. During sizing and optimization of its design variables, battery
cell specific energy is treated as a technology parameter that is varied to determine its influence on
mission range. To achieve a minimum range success criterion of 250 nmi with reserves, it is found that
battery cell specific energy must be at least 600 W-h/kg, more than twice the capability of today’s
lithium-ion cells.

1.0 Introduction
Electric airplanes are potentially disruptive innovations enabled by rapidly emerging battery and
electronic technologies. Better airplanes are promised by clever, synergistic integration of electric
propulsion with airframe structures and control systems. Concepts could be either entirely electric or
electrical hybrids. Transformative new airplane architectures may be possible by exploiting “distributed
electric propulsion” (e.g., Refs. 1 to 4).
But, if battery technologies arrive sooner than transformative aircraft technologies (or if
transformative aircraft technologies fail to emerge or become popular), then it is possible that electric
airplanes will have more conventional architectures. In the near term, at least, advanced electric airplanes
could appear with otherwise ordinary-looking features. This study is an attempt to define an airplane in
this scenario, and to determine roughly the battery characteristics needed to support it.
Although battery technology is progressing quickly, researchers usually cite battery performance as
the most significant barrier to the viability of fully electric aircraft. In particular, most studies identify
shortfalls in battery cell specific energy. Some vehicle concepts require specific energies well beyond the
capability of today’s lithium-ion cells, with the problem becoming more severe as payload or range
requirements increase (e.g., Refs. 5 to 11).
Thus, it is of interest to see if a small, fully electric transport can be made viable when the demands
placed on its batteries are reduced. Three strategies that reduce energy requirements are considered.
Improving propulsive efficiency, improving vehicle aerodynamics, and optimizing cruising airspeed are
all effective means of reducing energy requirements.
Cruising relatively slowly (if it is acceptable to do so) directly reduces power requirements, and it can
result in less energy to fly a mission by cruising closer to a maximum lift-to-drag ratio condition. High
levels of propulsive efficiency can be achieved with efficient, open rotor propellers. Contra-rotation
propellers that convert swirl losses into useful forward thrust are investigated. Electric motors may

NASA/TM-20220015470 1
improve the implementation prospects of contra-rotation simply by being more compact than gas turbine
engines. Wings with large spans have very low lift-dependent drag. A classic means of enabling high
wingspan is through use of a wing brace. The brace relieves the bending moment at the wing root and
enables high-span designs without the weight penalty that would result from a cantilever design. Using a
wing brace for this concept capitalizes on the rekindled interest in braced wings (Ref. 12), though at a
much lower airspeed than other braced concepts under investigation (e.g., Ref. 13).
Mission energy per passenger-mile is often used as a measure of merit in aircraft design. Maximizing
the number of passengers allowed by type category regulations is hypothesized to minimize this metric.
The highest number of passengers permitted by U.S. FAA Part 23 (Ref. 14) and by European CS-23
(Ref. 15) regulations for normal-category airplanes is 19. So to state the problem formally, the goal of this
study is to determine the minimum battery specific energy required to enable a 19-passenger commuter
transport while using airframe technologies that might be available by about 2035. Alternative designs
having fewer passengers is investigated as a sensitivity in a later section.
The Open Rotor, Braced-Wing Electric Transport studied here is dubbed “ORBET.” Though it is
fully electric (and is therefore advanced in many ways), it is otherwise conventional and uses relatively
near-term technologies wherever possible. For example, ORBET does not rely on distributed propellers to
enhance lift via a “blown wing” effect, nor does it relocate its propellers to the wing tips to recover vortex
energy. In other words, ORBET is not intended to be a larger version of NASA’s experimental X-57
Maxwell (Ref. 1). ORBET does not rely on superconductor materials (high-temperature or otherwise) to
reduce electrical system losses and weight, and it does not rely on boundary-layer ingestion to reduce
drag, as is proposed for other futuristic electric airplane concepts (e.g., Refs. 2 and 16). The lack of
advanced features results in an otherwise conventional appearance.
ORBET is proposed to serve the so-called “Regional Air Mobility” (RAM) aviation market of the
future. Although a common nomenclature has yet to emerge, RAM refers usually to low-demand routes
characterized by shorter distances. They differ from the transcontinental or intercontinental trunk routes
serviced by large-capacity airplanes involving at least one large airport, and they also differ from the
regional or interregional routes serviced by smaller jets and turboprops. The RAM market consists of
short, intraregional connections and point-to-point service between airports in urban, suburban, or rural
areas. Service can be scheduled or on-demand, and payloads can be either passengers or freight. RAM
service has also been referred to as “thin-haul” aviation, since the frequency of flights for any individual
route is low, or “thin.”
The need for a RAM transportation system that is separate – to the extent possible – from the long-
haul system was recognized as early as 1971 (Ref. 17). Today, RAM routes are served by small-capacity
airplanes carrying typically 7 to 19 passengers or small cargo loads. Their airworthiness directives are
given in the U.S. by (Ref. 14) and in Europe by (Ref. 15). RAM airplanes are often multi-role utility
transports that target the commuter or VIP passenger market, small package delivery, and mail delivery.
RAM operations are a subset of what NASA refers to more broadly as Advanced Air Mobility missions
that are characterized by routes of 300 nmi or less (Ref. 18). Mostly, these routes have been abandoned by
high-volume commercial operators because their larger aircraft are costlier to operate with reduced
payloads. RAM routes are served usually by boutique commuter operators using smaller equipment.
It can be challenging for RAM services to operate profitably. Operating costs per available seat mile
for RAM carriers are more than three times higher than for large air carriers (Ref. 19). And RAM
operators must compete with surface modes of travel such as automobiles, buses, trucks, and rail. This is
true especially within densely traveled corridors, where customers have multiple mode and schedule
choices. Still, though the frequency of flights for any individual route is low, the cumulative number of
flights across the entire RAM network is significant. And in the U.S., the economics of RAM aviation are

NASA/TM-20220015470 2
improved for some routes via subsidies paid to operators by the Essential Air Service program (Ref. 20).
The RAM market is undergoing change. The growth in pushbutton e-commerce and the oversubscribed
(for now, at least) shipping network have created opportunities for RAM aviation. The recently
certificated Cessna 408 SkyCourier (Ref. 21), developed primarily for launch customer FedEx, is
expected to serve RAM freight markets where larger airplanes cannot operate competitively. A passenger
variant of the SkyCourier is the first entirely new 19-passenger airplane in decades. In the future, the
dense (and underused) network of airfields covering the U.S. and Europe could be better exploited. Of the
roughly 4900 operational public airports in the U.S. today, only 11 percent of them have commercial
operations (Ref. 22). Many of the remaining airports have surplus capacity that could be used to enhance
RAM economics. Studies (e.g., Refs. 23 and 24) have determined that future hybrid-electric RAM
airplanes can be operated profitably.
Despite the advantages that technologies may bring, fully electric aircraft are nevertheless likely to be
handicapped by the comparatively low specific energy of their batteries relative to liquid fuels. Kerosene-
based fuels have more than a 40-fold specific energy advantage compared to today’s lithium-ion batteries.
Even when the efficiencies of converting energy to useful mechanical work are considered, fuel-burning
engines still have a clear specific energy advantage over today’s batteries. It is anticipated that, at least for
the foreseeable future, fully electric aircraft will have a distinct disadvantage in range capability under
comparable conditions to fuel-burning aircraft such as the SkyCourier (Ref. 25). But, there is increasing
environmental awareness of carbon production. In the future, there may be a need for more sustainable
aircraft concepts like ORBET. Operators might relax airplane performance requirements (especially
maximum range capability), particularly if there are incentives or requirements to do so. Though some
operators might have a genuine, occasional need for long-range capability, missions flown in RAM
aviation are almost always far shorter than the maximum range capabilities of the aircraft that service
them (e.g., Ref. 26). On future RAM routes, such as those from Tucson to Tucumcari, or from Tehachapi
to Tonopah, operators might be served adequately well by fully electric transports having shorter range
than today’s fuel-burning transports.
This is an initial airplane design and concept feasibility study conducted as a part-time effort over
nine months during 2021. Due to the short time involved, this study does not assess the concept’s
economic prospects, its operating costs, or the additional airport infrastructure required to support it.
Electrical components and power management systems are not designed or analyzed in detail (NASA
already has research ongoing in these areas, e.g., Reference 27). Instead, electrical components are treated
as simple systems, with their performance and weight modeled by simple efficiency assumptions and
scalar models. This study is intended to broadly define a near-term, fully electric, 19-passenger concept
airplane and determine its battery specific energy requirements. Future work is needed to model the
concept and its economics more rigorously.

2.0 Design
ORBET has seating for up to 19 passengers and flies at airspeeds no greater than 250 kcas.1 With
regards to type classification (see section 2005 of References 14 and 15), it is an airplane in the “normal”
category, it has a “level 4” certification classification (with maximum seating varying between 10 and
19 passengers), and it is in the “low-speed” performance category. Airplanes of this type were formerly
in the so-called commuter category, but that term is deprecated following the restructuring and

1
Throughout this document, abbreviations for calibrated airspeed and true airspeed (in knots) are expressed as kcas
and ktas, respectively.

NASA/TM-20220015470 3
 Figure 1.—Solid model of the ORBET concept.

harmonization of regulations in the U.S. and in Europe in 2017. A solid model of ORBET is shown in
Figure 1. The variant having contra-rotation propellers is shown.

2.1 Performance Requirements

Performance requirements must be defined for ORBET so that it can be sized and optimized, and so
that the minimum battery specific energy can be determined. Chief among the requirements are targets for
mission range and takeoff and landing field distance, which are driven principally by market and route
demands. A statement of all performance requirements is given later in Section 7.0.

NASA/TM-20220015470 4
 Most often, mode of travel statistics, economic studies, and existing air routes are used to provide
insight into setting a range requirement for an airplane in this class. For shorter trips, there is usually a
breakpoint in trip distance where surface modes fall off and where air travel modes begin. Sources place
this breakpoint variably from 100 to 500 statute miles (87 to 434 nmi; (e.g., Refs. 28 and 29)). Economic
studies indicate that markets exist in the U.S. and in Europe to support a 19-passenger transport to serve
regional air mobility markets with ranges of 100 to 500 km (54 to 270 nmi; (e.g., Ref. 30)). And in 2016,
Cape Air (the largest commuter airline in the United States) reported that 67 percent of their routes are
shorter than 100 nmi, and all of their routes are shorter than 225 nmi (Ref. 26).
Energy to fly an additional reserve mission is also required. Operators of today’s fuel-burning airplanes
must comply with reserve requirements defined in section 167 of Reference 31. It is anticipated that
operators of electric airplanes will need to satisfy similar requirements. Therefore, ORBET is required to
have enough battery energy to cruise an additional 45 min at normal airspeed following completion of the
primary flight to its intended airport. The diversion allotment required for fuel-burning airplanes in
section 167(a)(2) of Reference 31 is assumed not to apply (with adequate ceiling and visibility conditions, at
least) since ORBET could be designed with equipment for an instrument approach.
Given these considerations, ORBET is assigned a minimum range success criterion of 250 nmi, with
additional reserve energy for 45 min of flight at optimum speed and altitude. Minimum battery specific
energy is determined from this criterion. The design mission is shown in Figure 2. Cruise airspeed and
cruise altitude are design variables subject to optimization. Reserves are modeled as a simple 45-min hold
at 8000 ft.

Figure 2.—Design mission profile.

NASA/TM-20220015470 5
 Small airfields must be accessible to RAM transports. An airplane’s takeoff and landing distances are
often used as surrogate indicators for airport accessibility. A maximum field distance appropriate for
ORBET must be defined. It should not be so large that operations to smaller airfields are constrained. In
the U.S., 96 percent of the population lives within 19 miles of an airport having a runway at least 4000 ft
long (Ref. 32): a distance that can be driven easily in a car, taxi, or a rideshare service. Of the nearly 4900
operational public airports in the U.S. today, more than half of them have runways 4000 ft in length or
longer (Ref. 22). Given these considerations, ORBET is constrained to takeoff and landing distances no
greater than 4000 ft.
Many sources fail to define precisely what is meant by takeoff field distance. Larger multiengine
airplanes like ORBET require a critical field distance calculation with one engine inoperative
(see section 2115(c) of Reference 14). This is sometimes referred to as a balanced field length calculation.
In this study, the critical field distance to a 35 ft runway obstacle from a flat, paved, dry, sea level field at
59 °F is evaluated. Takeoffs are evaluated at maximum gross weight and at maximum motor power. An
ordinary takeoff at 59 °F with both engines operating is also computed. In normal situations when both
engines are operating, the required field distance is much shorter. This type of takeoff is used to compute
reference profiles for noise certification (see section 9.0 and Chapter 10 of ICAO noise regulations
(Ref. 33)). NASA’s Flight Optimization System (FLOPS, Ref. 34) is used to compute both takeoff types.
A discussion of how critical field distance is calculated for Part 23 airplanes using FLOPS is given in
Section 6.0.

2.2 Description
ORBET has a layout, gross weight, payload, shaft power, and airspeed similar to the new SkyCourier
turboprop (Ref. 21), and it is somewhat similar to other 19-passenger turboprops with braced wings and
unpressurized cabins (Refs. 35 to 37). But it is thought that ORBET’s wing aspect ratio could be much
higher than any of these examples, since it is intended to leverage new research into braced-wing
technology (e.g., Refs. 38 to 43).
ORBET is proposed to have a maximum takeoff gross weight of 18,000 lb. This is nearly the
19,000 lb upper limit for Part 23 commuter transports. For high propulsive efficiency, ORBET has large-
diameter propellers, necessitating a high-set wing and wing-mounted powerplants. Its wing is braced with
streamlined struts. ORBET’s wing may be stiffer and less elastic than larger braced-wing concepts
(e.g., Ref. 13), so it is assumed that additional jury structural members are unnecessary. If true, then
interference drags and supervelocity flow between structures (i.e., flow at velocities higher than
freestream velocity) should be similar to other braced-wing airplanes. Unlike other electric concepts that
exploit distributed electric propulsion, ORBET’s wing is not sized for cruise conditions. Instead, it has a
necessarily larger wing that results in good climb performance and short field distance. Despite a large
wing, its low cruise airspeed enables ORBET to cruise near its maximum lift-to-drag ratio condition.
Wing size, aspect ratio, and propeller diameter are determined during the sizing and optimization process
described in Section 7.0.
Two electric motor powerplants are podded on the wing in a traditional tractor configuration. There
are advantages and disadvantages to tractor arrangements. They have downstream blockage losses, and
they could disturb any natural laminar flow on the wing that might otherwise be present. But compared to
pusher configurations, tractor propellers enjoy smoother incoming freestream flow, and they lack
additional noise generated by impinging wing wakes. A pusher variant of ORBET is not analyzed.
ORBET’s unpressurized cabin has two pilots on a flight deck and has accommodations for up to
19 passengers seated three abreast (or optional equivalent freight). All structure makes maximum use of
advanced materials and composites wherever possible. The main landing gear is retractable into fuselage

NASA/TM-20220015470 6
blister fairings. The nose gear folds forward into the fuselage. Single-slotted extensible (Fowler) trailing
edge flaps are interrupted by the nacelles. There are no leading edge high-lift devices. Actuators for all
systems are proposed to be operated by conventional hydraulics. Cruise is limited to a maximum altitude
of 10,000 ft, a popular cruising altitude for unpressurized aircraft with no supplemental oxygen (see
section 211 of Reference 31).
Stability and control of ORBET is not explicitly studied. Instead, a simple volume coefficient method
is used to size its horizontal and vertical tails. A modified tail volume sizing method is available in
NASA’s FLOPS code (Ref. 44). This method is used to size ORBET’s tails, since FLOPS is used also for
other tasks (as discussed in later sections). Regression correlations for horizontal and vertical tail volume
coefficients are based on historical aircraft data. It should be noted that the FLOPS regressions fail at the
limit for wings having zero sweep. When wing sweep nears zero (as in the case of ORBET), the
correlations for horizontal and vertical tail volume coefficients tend to infinity and to zero, respectively.
This is remedied in this study by specifying a quarter-chord wing sweep of 25 (i.e., the lowest quarter-
chord sweep considered in the FLOPS historical data) and using the resulting volume coefficients values.
All power management electronics are proposed to reside in the wing nacelles. The power
management system is conventional (i.e., not superconducting) and is cooled by air taken in via scoop
inlets on the underside of each nacelle. Batteries are proposed to lie under the cabin floor. There are two
packs: a larger pack is located forward of the main landing gear; a smaller pack is to the rear. They could
be removed from below with simple equipment; perhaps even during quick turnarounds if it could be
managed quickly enough. The volume available for batteries is estimated to be 150 ft3.

2.3 Variants
Two variants of ORBET are studied. The first is equipped with conventional single-rotation
propellers (Figure 3, left), while the second has contra-rotation propellers (Figure 3, right). The contra-
rotation propellers are expected to have greater efficiency due to swirl loss reduction. This performance
benefit, however, is offset by additional weight. Performance benefits of each variant are discussed in
Section 8.0.

Figure 3.—ORBET variants: single-rotation propellers (left), contra-rotation propellers (right).

NASA/TM-20220015470 7
 Because a single-rotation propeller is easier to analyze than a contra-rotation propeller, the single-
rotation propeller variant is studied first. Sizing and optimization are performed using a simpler set of
design variables that are appropriate for a single-rotation propeller system.
The contra-rotation propeller variant is assessed afterwards using the same airframe design as the
single-rotation propeller variant. Borrowing vernacular from the aviation industry, it can be viewed as a
“new engine option” variant of the initial airplane. The contra-rotation propeller analysis is described in
greater detail in Sections 5.0 and 8.0.

2.4 Clarification on Use of Terminology

ORBET is described as having “open rotor” propellers. A clarification is needed. The implication of
this terminology is that ORBET will benefit from nontraditional, highly efficient, contra-rotating
propellers with many, thin, highly swept, wide-chord, and often scimitar-shaped blades. But if
implemented as such, ORBET might have safety and certification concerns. As typically envisioned, an
open rotor propeller would have many blades consisting of relatively thin airfoil sections, with each blade
attached to a pitch-change mechanism in a rotor hub. In this sense, an open rotor engine is more akin to an
“unducted” turbofan than a traditional propeller engine (indeed, the GE36 demonstrator engine in the
1980s was named the Unducted Fan, or UDF (Ref. 45)). As usually proposed, a thin composite blade is
bonded to a metallic inner spar, which is secured to a pitch-change mechanism in the hub. The entire
assembly must be able to withstand maximum centrifugal loads. A system having these features is
sometimes called a “prime reliant” system (Ref. 46). But with many blade shanks crowding the hub, and
with each blade requiring the ability to change pitch, it may prove difficult to design an adequate, prime
reliant blade retention system for open rotors. For airplanes larger than ORBET, some of the regulations
regarding this issue fall under section 25.905(d) of Reference 47 and its associated advisory (Ref. 48). For
these large transports, hazards must be minimized if a propeller blade fails or if a blade is released by a
hub failure. Ensuring that an open rotor propeller is prime reliant might make the engine prohibitively
heavy. Smaller airplanes like ORBET have no equivalent requirement, but compliance with propeller
airworthiness regulations (Ref. 49) must be demonstrated. In any case, how future open rotors might
satisfy these requirements is unclear.
But the scimitar shape of open rotor propellers is driven by the need for high airspeed. ORBET’s
range performance, as will be shown later, benefits greatly from low-speed flight. Indeed, at ORBET’s
normal economy cruise speed of just 180 ktas, scimitar-shaped open rotor blades are not warranted. It
should be apparent from the figures above that ORBET is equipped with quite ordinary propellers. As
such, any conventional propeller installation for ORBET (with fewer, thicker, straight blades) should have
plenty of room in the hub to incorporate an adequate blade retention system and to satisfy ordinary
propeller requirements (Ref. 49).

3.0 Weight Model

ORBET weights are predicted using a component buildup process. Major structural and system
component weights are predicted using the statistical-empirical weight relations of FLOPS (Ref. 44) and
Torenbeek (Ref. 50). An original model for propeller weight is developed. In general, the relations given
by Torenbeek can be applied to smaller, Part 23 normal category transports (Ref. 14) as well as to larger,
Part 25 transport category transports (Ref. 47). The relations given in FLOPS, however, are most often
limited to Part 25 transports. Since methods have been developed in FLOPS for high aspect ratio, strut-
braced wings, and since FLOPS is used also to evaluate mission performance, the weight relations from
FLOPS are used for ORBET wherever they are applicable. Torenbeek’s relations, however, are used in

NASA/TM-20220015470 8
cases where those in FLOPS are deemed inappropriate for small transports. All of these instances are
noted below. Since the relations in FLOPS are generally derived from older airplane data, scalars are
frequently applied to component weights to reflect use of advanced materials and technologies. These
instances are also noted. Methods for computing the weights of relevant structures and systems are
detailed in the following sections. Note that ORBET is assumed to have battery and electrical system
redundancies that would make an onboard auxiliary power unit unnecessary.
As noted in Section 2.0, ORBET’s maximum gross weight is 18,000 lb. That is, it is set deliberately
to be near the 19,000 lb upper limit for Part 23 transports. The components of its empty weight are
predicted individually. Since the weight of payload and operating items are known, the weight of batteries
is set by the available weight remaining. Since ORBET’s mission range is dependent on the total energy
contained in the batteries, an accurate weight buildup of structural components and systems is important.
Several of the statistical-empirical weight relations used in this study are dependent upon the
maximum takeoff gross weight, Wmto. ORBET’s 18,000 lb maximum gross weight, however, is unusually
heavy for an airplane in its passenger class due to its batteries. Thus, if 18,000 lb were used in empirical
equations containing Wmto, some of the resulting component weights might be unrealistically high. The
weight relations for furnishings and equipment, for example, fall into this category. Instead, a weight of
12,500 lb is sometimes substituted for Wmto. 12,500 lb is the maximum gross weight of a 19-passenger De
Havilland DHC-6-300 Twin Otter which is somewhat in the same competitive class as ORBET. Those
instances are noted below.

3.1 Wing
The method built into FLOPS for estimating wing weight (Ref. 44) departs somewhat from the usual
statistical-empirical approach used for other structural components. There are insufficient statistical data
available to determine empirically the effects of high aspect ratio, low sweep angle, strut bracing, flutter,
and aeroelastic tailoring on the weight of advanced wings. To help determine empirically the influence of
these effects, optimum wing designs were developed using the Aeroelastic Tailoring and Structural
Optimization program (Ref. 51). The resulting trends, some of which are shown in Reference 34, were
used to develop the FLOPS wing weight estimation method. The relations are too lengthy to reproduce
here. Interested readers are referred to Reference 44 for a full explanation. This method is applicable to
advanced, braced-wing concepts such as ORBET. The maximum benefit of wing strut bracing, maximum
use of composites, and maximum use of aeroelastic tailoring are assumed for ORBET. The wing
weight estimate reacts automatically to changes in design variables during sizing and optimization
(see Section 7.0).

3.2 Fuselage
The fuselage structural weight, Wfuse, is computed using the weight relation in FLOPS (Ref. 44):

The equation from Reference 44 reduces to the above form for passenger-carrying airplanes with
wing-mounted engines. Lfuse is the overall fuselage length in feet, and Dfuse is the average fuselage
diameter in feet. Wfuse is given in pounds. Lfuse and Dfuse are obtained from the solid model of the vehicle.
Cfuse is a technology scale factor. The overall fuselage weight is reduced by 18 percent to reflect
maximum use of composites as suggested by Reference 52, and it is reduced further by 8 percent to
reflect an unpressurized cabin as suggested by Reference 50. Thus, Cfuse is equal to 0.74.

NASA/TM-20220015470 9
3.3 Landing Gear
The main and nose landing gear weights, Wmain and Wnose, respectively, are computed using the weight
relations in FLOPS for retractable gear (Ref. 44):

The equations from Reference 44 reduce to the above form for transport-category airplanes. Lmain and
Lnose are the lengths of the extended main and nose landing gear oleos in inches, respectively, and Wldg is
the airplane design landing weight in pounds. Wmain and Wnose are given in pounds. Lmain and Lnose are
obtained from the solid model of the vehicle. Cldg is a technology scale factor. The gear weight is reduced
by 15 percent to reflect maximum use of advanced materials as suggested by Reference 52. Thus, Cldg is
equal to 0.85.
Since ORBET is an all-electric transport that burns no fuel, its weight for any given mission never
changes. Thus its design landing weight must equal its design maximum takeoff weight. The landing gear
must be designed to bear the full 18,000 lb maximum takeoff weight upon landing. For FLOPS
calculations, this is achieved by setting the ratio of maximum landing weight to maximum takeoff weight
to unity.

3.4 Furnishings and Equipment

Since FLOPS is primarily a tool for Part 25 transport-category large airplanes, its weight predictions
for furnishings and equipment may be too heavy if applied small Part 23 transports. Its statistical-
empirical weight relation for these items presumably accounts for galley and water equipment, lavatories,
insulation, non-removable emergency oxygen, evacuation slides, cargo provisions, and soundproofing,
which are much lighter on (or are entirely absent from) Part 23 transports. Torenbeek (Ref. 50), however,
provides relations for Part 23 category transports.
Weight relations for flight deck accommodations (Wfda), passenger seating (Wseat), fire-detection and
extinguishing systems (Wfire), cabin floor coverings (Wfc), and miscellaneous cabin equipment (Wmisc) are:

All weights are in pounds. Delivery weight empty (Wdwe) is 7100 lb, the portion of the empty weight
accounting for structural weight, propulsion, and aircraft systems, but not batteries, furnishings, or interior
equipment. Npax is the maximum number of passengers. Wmto is the maximum takeoff gross weight in
pounds (taken as 12,500 lb as explained above). Sfloor and Vcabin are the passenger cabin floor area in
square feet and volume in cubic feet, taken from the ORBET solid model.

3.5 Anti-Ice System

Since FLOPS is primarily a tool for transport-category large airplanes, its weight relation for anti-ice
systems is based on engine customer bleed air. For an all-electric transport like ORBET, an electrically
heated system might be more simply implemented than a bleed-fed system. And unlike the electrically
heated systems that are currently available, a dedicated alternator would not be necessary. Scaling the

NASA/TM-20220015470 10
information from Reference 53 to the size of ORBET (and considering only the resistance-heat foils), the
anti-ice system for ORBET is estimated to weigh a constant 50 lb in all cases. When evaluating mission
performance, the anti-ice system is assumed to be turned off.

3.6 Avionics and Instruments

The FLOPS relation for avionics weight is a function of design range, which may not be appropriate
for a short-range, RAM transport like ORBET. Torenbeek (Ref. 50), however, provides a relation for
avionics and instruments weight, Winst, for Part 23 low-speed transports operating under instrument flight
rules:

NW is the number of wing-mounted engines and Wmto is the maximum takeoff gross weight in pounds
(taken as 12,500 lb as explained above). Winst is given in pounds.

3.7 Hydraulic Systems

The hydraulic system weight, Whyd, is computed using the weight relation in FLOPS (Ref. 44):

The equation from Reference 44 reduces to the above form for fixed-wing airplanes with wing-
mounted engines. The relation assumes control surfaces are actuated by a conventional, fully duplicated
hydraulic and pneumatic system operating at 3000 psi. Afuse is the overall fuselage planform area in square
feet, S is the reference wing area in square feet, Nw is the number of wing-mounted engines, Phyd is the
hydraulic system pressure in psi, Mmax is the maximum operating Mach number, and Chyd is a technology
scale factor. Weight is reduced by 15 percent to reflect an advanced system. Thus, Chyd is 0.85. Whyd given
in pounds.

3.8 Surface Controls

The weight of surface controls, Wsc, is computed using the weight relation in FLOPS (Ref. 44):

Where Mmax is the maximum operating Mach number, Sflap is the total movable wing surface area in
square feet, Csc is a technology scale factor, and Wmto is the maximum takeoff gross weight in pounds
(taken as 12,500 lb as explained above). Weight is reduced by 15 percent to reflect use of advanced
materials as suggested by Reference 52. Thus, Csc is 0.85. Wsc is given in pounds.

3.9 Tail
Weights of the horizontal tail (WHT) and the vertical tail (WVT) are computed using the weight
relations in FLOPS (Ref. 44):

The equations from Reference 44 reduce to the above form for single-tail airplanes. The subscripts
refer to the horizontal tail and the vertical tail. SHT and SVT are the theoretical tail areas in square feet,
determined by the modified volume coefficient method described in Section 2.0. HT and VT are the
theoretical taper ratios. CHT and CVT are technology scale factors. Wmto is the maximum takeoff gross
weight in pounds (taken as 12,500 lb as explained above). Weights are reduced by 25 percent to reflect

NASA/TM-20220015470 11
use of advanced composite materials as suggested by Reference 52. Thus, CHT and CVT are equal to 0.75.
WHT and WVT are given in pounds.

3.10 Nacelles
The weight of one nacelle, Wnac, is computed using the weight relation in FLOPS (Ref. 44):

Where Dnac is the average nacelle diameter in feet, Lnac is the length of a nacelle in feet, Cnac is a
technology scale factor, and FSLS is the rated thrust of one powerplant at the sea level static condition in
pounds. Weight is reduced by 20 percent to reflect use of advanced composite materials as suggested by
Reference 52. Thus, Cnac is 0.80. Wnac is given in pounds.

3.11 Propellers
FLOPS (Ref. 44) has no propeller weight model. Roland (Ref. 54) proposed a correlation in 1969, but
it is regressed against older propeller data that spans a rather wide range of applications. An improvement
to Roland’s correlation is sought that uses newer propeller weight data for a narrower regression space
that is more applicable to ORBET. Roland’s independent correlating variable, DPmax√𝐵, is retained,
where D is the propeller diameter in feet, Pmax is the maximum rated takeoff shaft power in bhp, and B is
the number of blades.
For the new correlation, emphasis is placed on propellers newer than those considered by Roland, and
propellers having very large values of DPmax√𝐵 (such as Dowty’s R408/6-123-F and R381/6-123-F/5
models found on the Bombardier Q400 and the Saab 2000), are excluded. Data for fifty variable-pitch,
hydraulically operated propellers with feathering and reversing features manufactured by Hartzell
Propeller, Collins Aerospace, Dowty Propellers, and MT-Propeller are taken from their Type Certificate
Datasheets. Both aluminum and composite construction is considered.
The resulting new correlation for the weight of a single propeller, Wprop, is:

Wprop accounts for the weight of the propeller blades, hub, pitch-change mechanism, and spinner.
Wprop is given in pounds. The original and revised correlation are plotted with data in Figure 4. The
propeller weight of the final, optimized single-rotation variant of ORBET is shown on the plot. The
revised correlation results in lighter weight predictions than Roland’s 1969 model.

Figure 4.—Propeller weight model.

NASA/TM-20220015470 12
3.12 Electric Motors and Power Conditioning Equipment
The weight of an electric motor powerplant, Wmotor, and its associated power management and
distribution equipment, WPMAD, are estimated using two simple linear scalar models:

The power, Pmax, is the maximum rated power of a single powerplant contained within a single
nacelle installation. For example, a Pmax of 1000 bhp would power the propeller on one side of the
airplane during takeoff, whether it is a single-rotation propeller or a set of two contra-rotation propellers
each using some portion of that power. Further complicating this accounting is that there may be more
than one electric motor per powerplant totaling 1000 bhp. That is, two electric motors per nacelle might
be employed for motor efficiency considerations, where one motor could be disengaged from the shaft
during cruise when less shaft power is required.
So to clarify, in this example for Pmax equaling 1000 bhp, all of the electric motors on the airplane
would weigh a total of (0.164 lb/bhp)(1000 bhp)(2) = 328 lb, all of the associated power conditioning
electronics would weigh a total of (0.0822 lb/bhp)(1000 bhp)(2) + (200 lb)(2) = 564 lb, and the total
power consumed by the airplane during the takeoff maximum would be (1000 bhp)(2) = 2000 bhp.
Pmax is a design variable determined by the sizing and optimization process (see Section 7.0). The
power management system is conventional, i.e., not superconducting. The coefficients in the equations
are based on high specific power electric motors under development for aviation applications
(e.g., Refs. 55 and 56). Note that 0.164 lb/bhp and 0.0822 lb/bhp equate to specific power levels of
10 kW/kg and 20 kW/kg, respectively. The 200 lb adder used in the WPMAD relation is intended to account
for the traditional electrical equipment required for conventional airplanes. Wmotor and WPMAD are given in
pounds.

3.13 Flight Crew and Payload

The weight of two crew members with their baggage and flight equipment is assumed. Although
some futuristic studies assume that increases in automation may someday enable single-pilot operations,
current requirements for flight crews require a captain and a second in command. Specifically, 14
CFR§135.99(b) (Ref. 57) states “No certificate holder may operate an aircraft without a second in
command if that aircraft has a passenger seating configuration, excluding any pilot seat, of ten seats or
more.” Single-pilot operation without a second in command is investigated as a sensitivity in a later
section. Torenbeek (Ref. 50) suggests using 205 lb per pilot with their baggage and flight equipment. The
weight per passenger assumed is 170 lb. Baggage is assumed to weigh 30 lb per passenger. Payload for a
19-passenger mission thus weighs 3800 lb. Operating items (i.e., the flight crew and their baggage) weigh
410 lb.

3.14 Batteries, Power, and Iteration to Closure

The weight of batteries is what remains after subtracting the weights of structural components,
systems, payload, and operating items (each described above) from the 18,000 lb maximum takeoff gross
weight. This accounting is similar to how fuel weight is sometimes determined during conceptual design
of conventional fuel-burning aircraft. Several of ORBET’s design variables (see Section 7.0) influence
the component weights listed above. Thus, the weight models must react properly to changes in design

NASA/TM-20220015470 13
variables during the optimization and sizing process. This requires an iteration to determine the final
weight of each component and the weight available for batteries.

4.0 Aerodynamic Model

4.1 Analysis Method
An aerodynamic model with relatively low computational cost is required to provide vehicle
aerodynamics to the sizing and optimization process described in Section 7.0. Given airfoil section
performance and a basic vehicle definition, the model must be able to compute airplane lift and drag
coefficients as functions of its angle of attack. It must react to changes in gross airplane geometry so that
airplane-related design variables can be optimized. It must provide both clean aerodynamics and flapped
aerodynamics suitable for takeoff and landing analyses.
A semiempirical method developed by Roskam (Ref. 58) is selected and programmed so that it can be
executed by a driver during sizing and optimization. Modeling begins by selecting two-dimensional
airfoil sections for the wing with known sectional lift and drag properties. Lift and drag coefficients of the
airfoil sections are represented in the model by simplified characterization parameters (i.e., primitives),
rather than by continuously real functions. These primitives (such as the lift-curve slope, maximum lift
coefficient, maximum lift angle of attack, etc.) can be used to construct continuous aerodynamic functions
at any time. The primitives are adjusted by the model when the geometry varies or when flow conditions
change, using a combination of physics-based scaling laws and empiricism. Section aerodynamics, for
example, react to changes in size and flight condition, as they are corrected for compressibility via the
Prandtl-Glauert transformation and for Reynolds number effects. Section aerodynamics are modified
further as trailing edge flaps deploy.
Clean wing aerodynamics are computed from section performance based on its planform geometry,
twist angle, and incidence angle. Stall angle, lift-curve slope, maximum lift coefficient, span efficiency,
and other parameters are computed for the three-dimensional wing. Similar calculations are made for the
flapped wing. Finally, overall three-dimensional vehicle aerodynamics are computed given fuselage, tail,
and propeller information using an empirical drag buildup method.

4.2 Validation
The method is validated against the aerodynamics of a Fokker F27 turboprop transport reported in
Reference 58. Like ORBET, the F27 is an unswept, high-wing transport driven by twin propellers.
Though it is much larger, lacks a wing brace, and has a wing aspect ratio of only 12, the F27 otherwise
bears an architectural similarity to ORBET. The aerodynamic model reacts to changes in size and
geometry via physics-based and empirical relationships. Thus, if the model predicts the aerodynamics of
the F27 closely, it is assumed that the aerodynamics of ORBET could be predicted by the same model
with some confidence.
The F27 uses NACA series 64-421 sections at the wing root and thinner 64-415 sections at the tip
(Ref. 59) and is equipped with single-slotted trailing edge flaps (Ref. 60). Flaps are estimated to occupy
the rear 25 percent of the chord and are assumed to be extensible beyond the clean trailing edge by five
percent (accuracy of these dimensions, of course, could be improved with access to more detailed data).
Primitive characteristics for two-dimensional section aerodynamics at zero Mach number are obtained
from the tool described in Reference 61. Three-dimensional vehicle-level aerodynamics are developed
using the F27’s gross geometry reported in Reference 59.

NASA/TM-20220015470 14
 Figure 5.—Model validation of Fokker F27 aerodynamics.

The drag buildup procedure comes close to – but does not quite reproduce – the F27 reported data. To
match the reported data, the zero-lift drag coefficient predicted by the model is adjusted slightly using a
scalar calibration multiplier to match the data reported in Reference 58. Final calibrated aerodynamics are
shown in Figure 5. Reported data are given by the dashed lines; the calibrated model prediction is given
by the solid lines.

4.3 Use in ORBET Analysis

The same zero-lift drag calibration factor determined for the F27 is applied to ORBET. But there are
other aerodynamic effects impacting ORBET. There is a drag penalty due to the wing brace as well as any
of its associated interference drags. And with low airspeed (the unit Reynolds number per chord foot is just
1.5×106 ft–1) and narrow wing chords, it might be reasonable to assume there is at least some natural laminar
flow (Ref. 62). Very crudely, the benefit and penalty of these two competing effects are assumed to cancel
each other exactly. Future work is needed to justify this assumption.

5.0 Propeller Model

5.1 Analysis Method
A propeller design and performance prediction model is required for the sizing and optimization
process described in Section 7.0. The open-source software XROTOR (Ref. 63) is selected for the design
and analysis of propellers. XROTOR requires just a fraction of a second to converge for any given design
or performance calculation on a modern personal computer, making it well-suited for use in airplane
sizing and optimization. As written, XROTOR is a menu-based code that relies on interactive user inputs.

NASA/TM-20220015470 15
A custom driver is written to communicate with XROTOR in batch mode. The driver is called repeatedly
during sizing and optimization to design propellers and to compute their performance data.
The XROTOR design process begins with user-defined section aerodynamics for a set of two-
dimensional airfoils. The sections are arranged into a spanwise propeller stack. Using an accounting
similar to the airplane aerodynamic bookkeeping described in the preceding section, lift and drag
coefficients of the sections are represented in the model by simplified characterization parameters, or
primitives, rather than by continuously real functions. Section drag is scaled for Reynolds number effects
as incoming flow conditions or chord lengths change. The airfoil sections are used to design a three-
dimensional propeller. As typically used, the code computes the spanwise chord distribution at the
aerodynamic design point. Chords are calculated at each station such that user-input design lift coefficient
values are satisfied. Usually, the design lift coefficients are collectively expressed as a single major
1
propeller design variable called the integrated design lift coefficient (CLi,des), defined as 4∫0 𝐶𝑙𝑖 (𝑟)𝑥 3 𝑑𝑥 ,
where Cli is the local (section) lift coefficient at the design condition as a function of spanwise position r,
and x is a dummy integration variable ranging from zero at the hub (r = 0) to unity at the tip (r = R), i.e., x
= r / R. Note that other definitions of CLi,des exist in the literature. CLi,des is an indicator of overall blade
camber and lift effectiveness.
Once determined, the chord distribution results in an activity factor per blade (AF), defined as
1
6250∫0 (𝑐(𝑟)⁄𝐷)𝑥 3 𝑑𝑥 , where c is the section chord as a function of spanwise position, r, and D is the
propeller diameter. AF is a measure of the propeller’s solidity. It is also considered to be a major propeller
design variable. As of version 7.55, XROTOR does not compute AF from the chord distribution, but since
the software is open-source, it is easily modified to do so.
Finally, most users opt to have the code compute the spanwise twist distribution. At the design point,
twist angles are set to achieve a minimum induced loss circulation distribution (Ref. 64). Once the
propeller is defined and performance is computed at its design condition, performance at any other off-
design condition can be calculated.

5.2 Validation
Before XROTOR is used in this study, it is validated against experimentally measured propeller
performance reported by United Technologies Corporation (Ref. 65). The experimental propeller has four
blades, a 39 in. diameter, and a reported AF per blade of 91.
The family of HS-1 propeller airfoils (Ref. 66) is used for the propeller airfoil stack. Their spanwise
locations are given in Reference 65. The airfoil family is shown in Figure 6 (for display purposes they are
shown having identical chord lengths). No other geometry information is given in the reference.
The data in Reference 65 were obtained in United Technologies’ Subsonic Wind Tunnel in East
Hartford, Connecticut. Data were collected for ranges of airspeed, shaft power, shaft speed, and blade
pitch angle. Results of the experiment are shown by the dashed lines in Figure 7. J is the dimensionless
propeller advance ratio V / nD, where V is true airspeed, D is diameter, and n is shaft speed in revolutions
per second. CP is the dimensionless power coefficient P / D5n3, where P is the shaft power and  is
ambient density. Efficiency is the dimensionless product of thrust and airspeed divided by shaft power.
Efficiency contours from 85 to 90 percent are shown in the plot using increments of one percent. For the
experimental data shown in the figure, airspeed was held at Mach 0.4 while shaft speed, shaft power, and
blade pitch angle varied.

NASA/TM-20220015470 16
 Figure 6.—United Technologies HS-1 propeller airfoil series.

Figure 7.—Propeller performance map validation.

NASA/TM-20220015470 17
 For the code validation, airfoil section aerodynamics are obtained from the tool described in
Reference 61. An aerodynamic match point is selected using J = 2.2 and CP = 0.30. Lacking additional
geometric information for the propeller, two assumptions are required to set the chord and twist
distributions. First, the propeller is designed iteratively by varying CLi,des until the activity factor
converges upon 91, and second, the twist distribution is set via XROTOR’s minimum induced loss
optimization. XROTOR has a native feature for creating propeller performance maps, which has a design
of experiments philosophy that differs from the wind tunnel measurements. In XROTOR, the ratio of
wheel tip Mach number and J at the design point is held constant (i.e., the ratio of shaft speed squared and
airspeed is constant everywhere on the map), while shaft power and blade pitch angle vary. In Figure 7,
XROTOR’s efficiency predictions are shown in solid lines. The efficiencies are not installed since no
information about the experimental apparatus is known.
Differences in predicted vs. measured efficiency range from one to four percent across most of the
performance map. Differences in efficiency are attributed to 1) approximate section aerodynamics, 2)
unknown chord and twist distributions, 3) unknown installation effects, and 4) differences in the design of
experiments.

5.3 Use in ORBET Analysis

In this study, XROTOR is used to design a propeller given a set of design variable values and to
predict its off-design point performance. The family of HS-1 propeller airfoils (Ref. 66) is used for the
propeller airfoil stack, and the spanwise chord and twist distributions are computed as discussed in the
previous section. Uninstalled thrust and overall blade pitch angle are calculated given the independent
variables of shaft power, shaft speed, and flight condition. Ordinarily, when a variable-pitch propeller is
connected to a fuel-burning piston or a turboshaft engine, designers will fix shaft speed at a
predetermined rate and use a constant-speed hub mechanism to maintain it. But in the case of an electric
motor, shaft speed and shaft power can be readily uncoupled. A given amount of shaft power can be
maintained over a selectable range of shaft speeds (the idea of an “electronic gearbox” is discussed more
thoroughly in Section 9.0, where it is proposed as a low-noise operating mode for electric airplanes). For
mission performance calculations, the maximum thrust occurring between 1100 rev/min and
1600 rev/min is used. Note that this technique could not be as easily replicated by a conventional system
(at least, not without a variable-ratio gearbox), or by a constant-speed propeller. For any given propeller
design, a table of thrust values as a function of flight condition and shaft power is assembled. The table is
used as input to the mission performance calculations.
Propeller performance predictions need to be adjusted to account for blockage effects of downstream
structures when installed on a vehicle. An installation method for tractor propellers adapted from FLOPS
(Refs. 34 and 52) is used. The original reference is unknown, but the model format and thrust penalties
are similar to the method reported in Chapter 6 of Torenbeek (Ref. 50) and elsewhere.
XROTOR is also able to design and analyze contra-rotation propellers by making flowfield
adjustments. In a contra-rotation configuration, the aft propeller receives additional axial and rotational
velocity components from the front propeller, while the front propeller receives an additional axial
velocity component from the aft propeller. Each propeller is modeled in its own XROTOR simulation and
flowfield information is passed from one simulation to the other. Iterations are required to converge on
the final designs of each propeller. This capability is used to assess the ORBET variant shown on the right
side of Figure 3. Implementation of this analysis is discussed in Section 8.0.

NASA/TM-20220015470 18
6.0 Vehicle Performance Model
6.1 Mission Calculations
With weight, aerodynamic, propeller, and energy models defined, airplane mission performance can
be computed. The FLOPS performance tool (Ref. 34) is used. For ORBET, however, this cannot be done
quite so straightforwardly as it is done for conventional fuel-burning airplanes.
Ordinarily, FLOPS operates under the assumption of variable airplane gross weight. As a
conventional airplane burns fuel, it becomes lighter as the mission progresses. FLOPS capitalizes on this
behavior to close the problem mathematically.
To accomplish this, FLOPS users are obliged to define a so-called “free cruise segment” somewhere
in the principal mission. The amount of fuel and the distance flown in the free segment are initially
unknown. The code computes performance from both ends of the mission, meeting at the beginning point
and at the end point of the free segment. Starting with the ramp weight, performance is computed forward
from taxi and takeoff until the beginning of the free segment is reached. Then, from the zero fuel weight,
performance is computed from the reserve mission in reverse until the end of the free segment is reached.
With the difference in fuel weight for the free segment known, its distance can be computed. A fuel-
consistent mission is the result.
But an alternate approach is required to evaluate constant-weight airplanes like ORBET. With a few
unconventional tactics, FLOPS can be persuaded into assessing fixed-weight airplanes. The basis for this
is to assign a very small (but nonzero) amount of variable weight to the airplane. This weight is neither
payload nor is it operating empty weight. Rather, it is akin to fuel weight in a traditional fuel-burning
airplane. In the case of a constant-weight airplane, it must be small enough that it does not impact the
performance of the airplane as it “burns off” during the mission, but it must be large enough to allow
FLOPS to close the performance problem mathematically as described above. This “faux fuel weight” is
on the order of 1 or 2 lb for ORBET.
To make use of the faux fuel weight, there must also be faux fuel flow rates. These fictitious rates are
assigned to the powerplant performance data. Propeller thrust is a function of airspeed, altitude, and the
shaft power consumption of the powerplants. Powerplant performance data consisting of installed thrust
levels as a function of airspeed, altitude, and power consumption are contained in a file which is read by
FLOPS in the usual manner. However, associated with every propeller thrust value must be a very small
(but nonzero) faux fuel flow rate. This flow consumes the small amount of fuel weight that is assigned to
the airplane. The mission is complete when the small weight of faux fuel is consumed and the free
segment closes. The mission range becomes known and the overall weight of the airplane changes only
inconsequently by 1 or 2 lb.
Although this enables constant-weight airplanes to be assessed, it is an arbitrary solution unless more
information is applied. To accurately assess mission range, the energy used throughout the mission must
match the total energy available from the batteries. Fortunately, FLOPS has two features that permit this
to happen.
The first is the code’s ability to track an arbitrary propulsion variable as it changes throughout a
mission. This feature was developed originally to track jet engine exhaust emissions throughout a
mission, but it can be used to track any propulsion-related variable. Data for this variable are contained in
the engine performance data file as a function of flight condition and power setting. If the propulsion
variable is propeller shaft power, then a running total of energy consumed can be tracked throughout the
mission. The total energy consumed is the sum of the energy consumed in each mission segment. The

NASA/TM-20220015470 19
second feature required is the code’s ability to apply a scale factor to engine fuel flows. This feature was
developed originally to explore sensitivities of vehicle performance with respect to engine efficiency. But
for the case of constant-weight airplanes, the fuel flow scale factor can be adjusted iteratively until the
energy used matches the energy available. A simple driver is developed for FLOPS to evaluate mission
performance for ORBET in this manner.

6.2 Special Considerations for Part 23 Requirements

As already noted, FLOPS is principally a Part 25 tool intended for analysis of large, transport-
category airplanes. As a smaller, Part 23 normal-category airplane, ORBET is required to comply with
some airworthiness requirements that FLOPS is incapable of assessing unless it is modified.
Part 23 Level 4 multiengine airplanes like ORBET require a critical field distance calculation with
one engine inoperative (also known as a balanced field length calculation), and must also meet subsequent
climb gradient requirements. While FLOPS is indeed capable of computing critical field takeoffs, the
takeoff speed requirements of Part 25 are more difficult to satisfy than those of Part 23. Part 25
requirements have additional minimum unstick-driven speeds that are not required by Part 23 (see
14 CFR§25.107 of Reference 47). Thus, using FLOPS to evaluate takeoffs without modification would be
a very conservative option. It would result in field distances several hundred feet longer than the “actual”
distance. Since (as will be shown later) the takeoff field distance is an active, binding constraint during
sizing and optimization, this would adversely impact the solution. Takeoff distance calculations in
FLOPS must be modified to accommodate ORBET.
In addition, minimum climb rate requirements after takeoff differ between Part 23 and Part 25,
requiring more FLOPS modifications. For Part 23 airplanes, the climb requirements with all engines
operating (given in section 23.2120(a) of Reference 14) as well as requirements with one engine
inoperative (given in section 23.2120(b)(3)) are custom-programmed into FLOPS for this task. For all
calculations with one engine inoperative, the engine-out yaw drag increment proposed by Torenbeek
(Ref. 50) is used. Takeoff performance for ORBET is evaluated per section 23.2115(c) of Reference 14
for Level 4 airplanes using a sea level field with a 35 ft runway obstacle at standard (59 °F) conditions.

6.3 Energy Availability and Electrical System Technologies

Much like fuel in a conventional fuel-burning airplane, stored energy determines the range
performance of an electric airplane. Stored energy in this study is modeled by a simple bookkeeping that
is a function of 1) battery pack mass, 2) battery cell specific energy (i.e., the gravimetric energy density at
the cell level, measured in W-h/kg), and 3) an energy conversion knockdown factor. The product of these
three items determines the shaft energy available to the propellers.
The definition of the knockdown factor as used here requires discussion. Battery pack performance
models often use cell performance as a starting point. In going from individual cells to the pack, losses in
specific energy come from weight gains and energy losses. Specific energy at the cell level must be
reduced to account for the weight of additional items within the battery pack beyond the battery cells.
Most of the additional weight in the battery pack comes from the thermal management system, coolant,
module housing structures, and additional electrical connectors. Cell specific energy is further degraded
by electrical losses within the pack. The total reduction in specific energy – from the battery cells to the
battery pack – is assumed to be 35 percent. Additional studies are required to justify this rough assumption.

NASA/TM-20220015470 20
 Figure 8.—Energy availability assumptions.

Between the battery pack and the propeller shafts there are additional energy conversion losses from
the power management and distribution system and the electric motors. The total of these losses is
assumed to be 5 percent. The weight of these items (i.e., WPMAD and Wmotor) is accounted for by the
relations given in Section 3.0. Thrust is computed from shaft power and propeller efficiency as discussed
in Section 5.0. This simple accounting is illustrated in Figure 8. Thus, the energy available to the
propeller shafts is

where egrav,cell is the gravimetric energy density at the cell level (converted to W-h/lb) and Wbatt is the
combined weight in pounds of the battery packs (determined as discussed in Section 3.0). Though it is a
crude estimate, an overall electrical system knockdown factor of 0.6 is thought to be reasonable and
perhaps conservative (e.g., Refs. 67 to 69).
Battery egrav,cell is treated as a technology parameter that is investigated for its influence on mission
range. Once an optimum design is determined by the sizing process (discussed in the following section),
egrav,cell is varied manually until the 250 nmi range goal is met. The value of egrav,cell required to meet the
range goal is discussed in Section 8.0. The magnitude of egrav,cell indicates the technology required for
ORBET’s battery.
This energy accounting is very crude. Further, electrical component behavior, battery current
discharge capability, and specific power are not considered. Ideally, batteries, electric motors, and the
components making up the power management and distribution system should be modeled with more
rigor. But this simple model allows quick calculations to be made, with the hope of a reasonable airplane
design resulting in a short time. Future studies are needed so that energy and power requirements can be
more accurately known. Additionally, it is possible for thermal loads to constrain airplane performance
(Ref. 70), so the behavior of each electrical component should be built into the sizing process.

7.0 Sizing and Optimization

With propeller, weight, aerodynamic, and airplane performance discipline models defined, the
airplane can be sized for optimum performance. Major design variables are identified and permuted by an
optimizer so that mission range is maximized for a given value of egrav,cell. Sizing and optimization is

NASA/TM-20220015470 21
limited to the single-rotation propeller variant. The contra-rotation propeller variant is assessed afterwards
using the same airframe design as the single-rotation propeller variant.

7.1 Drivers and Optimizer

Drivers are constructed to execute each multidisciplinary model. A framework is used to pass
information between analyses. The OpenMDAO framework and optimization software (Refs. 71 and 72)
is used to model the process. OpenMDAO is an open-source computing environment and framework tool
developed by NASA for multidisciplinary systems analysis and optimization. Components, groups and
drivers are classes available in OpenMDAO to create objects. The objects are connected to form a
sensible, multidisciplinary analysis of a problem. The models for propulsion, weight, aerodynamics, and
vehicle performance represent independent subproblems having interrelated design and response
variables. Each disciplinary model reacts to changes in design variables in the manner discussed above.
OpenMDAO supports a variety of optimization methods. Included are classical methods such as
gradient-based methods, one direct (search-strategy) method, and evolutionary and particle swarm
methods. With the number of optimization variables involved in this problem, it was hypothesized that
the objective function might be discontinuous or beset by local optima, providing opportunities for
gradient-based optimizers to become stuck. For this reason, a gradient-free particle swarm optimizer is
used during initial sizing (the Augmented Lagrange Particle Swarm Optimizer, ALPSO (Ref. 73)). The
design space is searched by ALPSO for feasible candidate designs. Simple penalty functions are used to
constrain the optimum. ALPSO is used because it is expected to provide good results, largely without
regard for its computational efficiency. The most attractive candidate design found by ALPSO serves as a
starting point for the Sequential Least Squares Quadratic Programming optimizer (Ref. 74), which
converges more precisely on the optimum.

7.2 Full Optimization Problem

The sizing and optimization problem is broken into two parts. A more complete, “full” statement of
the problem is given first. The full problem is followed by an abridged problem, which is thought to be
more tractable and practical for optimization. Design variables considered for the full optimization
problem are listed in Table 1.
Most of the design variable definitions in the table are self-evident, but some require explanation. At
the propeller design point, shaft power, Pdes, and shaft speed, Ndes, are specified at the flight conditions
Vdes and Hdes. Note that the mission design variables Vcr,des and Hcr,des need not be equal to Vdes and Hdes.
Propellers powered by gas turbine engines or by reciprocating engines can be subject to a shaft power
lapse with altitude. Propellers driven by electric motors and batteries have no such lapse unless, of course,
power is deliberately reduced by the operator or by a control system. A feature of electric motors is that
shaft speed can often be varied independently of shaft power. At least two additional design variables are
needed to account for this added flexibility. For ORBET, the design variables Pmax / Pdes and Nmax / Ndes
are introduced. These variables allow the propeller to operate with more power and/or rotational speed
during takeoff and climb. For an all-electric airplane, adding shaft power during takeoff and climb mimics
the natural thrust lapse of a naturally aspirated reciprocating engine or a gas turbine engine, but with more
flexibility. The cruise phase requires much less power than the takeoff and climb phases. If ORBET’s
electric motors are to operate efficiently during each phase of flight, it may be advantageous to have
“climb-assist motors” dedicated to takeoff and climb phases only. Each propeller could be driven by two
electric motors. All of the motors would be used during takeoff and climb to maximize thrust. At the
beginning of the cruise phase, the climb-assist motors would disengage. With such an arrangement, motor

NASA/TM-20220015470 22
efficiency during cruise could be higher than one large motor per propeller operating at reduced power.
Dual electric motors are used on some series hybrid automobiles (e.g., Ref. 75).
Note that even the “full” design problem is simplified. Design variables pertaining to the fuselage,
tail, high-lift flaps, and other systems are not identified in Table 1 since they are determined previously
and separately by other requirements. And often, propeller details (e.g., airfoil stack, chord and twist
distributions) and wing details (e.g., airfoils, chords, taper, sweep, twist, incidence) are considered in
airplane design optimization. But for ORBET, they are demoted from design variable status by fiat. This
is not to say that they are unimportant. Rather, in a conceptual design study, some variables may be
excluded so that sizing can be simplified. A reasonable overall design can oftentimes be obtained by
making reasonable simplifying assumptions.
Response variables computed from the design variables are listed in Table 2. Jdes is the design
propeller advance ratio, V / nD, where true airspeed, V, is expressed in feet per second, diameter, D, in
feet, and shaft speed, n, in revolutions per second at the propeller design condition. (Note that in the
literature, at least one other definition of J exists, where n is expressed in radians per second). Mhel,des is
the helical, or relative, propeller design tip Mach number. It is the root sum square of the tangential

TABLE 1.—DESIGN VARIABLES FOR FULL OPTIMIZATION PROBLEM

Propeller design variables
D Diameter, in.
B Number of blades per propeller
Pdes Design shaft power, bhp (per powerplant)
Ndes Design shaft speed, rev/min
Vdes Design airspeed, kcas
Hdes Design altitude, ft
Pmax / Pdes Ratio of maximum shaft power to design shaft power, > 1
Nmax / Ndes Ratio of maximum shaft speed to design shaft speed, > 1
Wing design variables
S Reference area, ft2
AR Aspect ratio
flap Takeoff flap deflection angle, deg
Mission design variables
Vcr,des Design cruise airspeed, kcas
Hcr,des Design cruise altitude, ft

TABLE 2.—RESPONSE VARIABLES

Propeller
Jdes Design advance ratio
Mhel,des Design helical tip Mach number
AF Activity factor per blade
CLi,des Integrated design lift coefficient
Pmax Maximum shaft power, bhp (per powerplant)
Nmax Maximum shaft speed, rev/min
Vehicle
R Range, nmi
Etot Total mission energy available to propellers, kW-h
dto Critical takeoff field distance, ft
dland Landing field distance, ft
400 Climb gradient at 400 ft altitude with one powerplant inoperative, percent
ROCsvc Potential rate of climb at service ceiling, ft/min

NASA/TM-20220015470 23
(wheel) tip Mach number and the forward flight Mach number at the propeller design condition. AF is the
activity factor per blade, and CLi,des is the integrated design lift coefficient as described in Section 5.0. R is
the vehicle mission range with reserves as discussed in Section 6.0. Etot is the total energy available to the
propellers for the mission. dto and dland are the critical field distance and the landing field distance,
respectively. 400 is the climb gradient at 400 ft altitude with one powerplant inoperative. A discussion of
how these values are computed for ORBET is found in Section 6.0. ROCsvc is the potential rate of climb at
the service ceiling (i.e., at Hcr,des and Vcr,des).

7.3 Design, Performance, and Behavioral Constraints

Optimization is subject to the following design, performance, and behavioral constraints:

1. dto ≤ 4000 ft and dland ≤ 4000 ft

2. Hcr,des ≤ 10,000 ft and Hdes ≤ 10,000 ft
3. ROCsvc ≥ 300 ft/min
4. 400 ≥ 2 percent
5. Vcr,des ≤ 250 kcas and Vdes ≤ 250 kcas
6. Mhel,des ≤ 0.70
7. 1.6 ≤ Jdes ≤ 2.5
8. 0.3 ≤ CLi,des ≤ 0.8
9. 80 ≤ AF ≤ 200
10. Pmax / Pdes ≥ 1
11. Nmax / Ndes ≥ 1
12. AR ≤ 19

The first five constraints are performance requirements defined for the vehicle. The limits assigned to
dto, dland, Hcr,des, and Hdes were discussed in Section 2.0. dto and dland are computed for a sea level field at
59 °F, standard day conditions. ROCsvc is evaluated at Hcr,des and Vcr,des. For Level 4 category multiengine
airplanes like ORBET, the minimum 400 allowed is 2 percent, and limits for Vcr,des and Vdes are set to
250 kcas, the upper limit of low-speed category airplanes (see 14 CFR§23.2005 and 14 CFR§23.
2120(b)(3) of Reference 14. The next six constraints are limits for typical low-speed propellers. A modest
upper limit is chosen for Mhel,des in anticipation of noise requirements (though noise is not explicitly
evaluated in this study). It is also chosen in anticipation of propeller airfoil selection (discussed in the
following section). The final constraint is the upper limit for AR. It is set to 19: the aspect ratio reported
for the Transonic Truss-Braced Wing concept developed by Boeing (Refs. 76 and 77). This constraint is
somewhat arbitrary. Higher values of AR might be possible, provided the weight increase can be properly
computed (the method in Reference 34 is valid for aspect ratio values up to 25), and if there are no
aeroelastic or structural issues. This limit is viewed as a non-mandatory, flexible constraint for an
unswept, low-speed, braced wing for an airplane like ORBET.

7.4 Abridged Optimization Problem

Simplifications are made to the full optimization problem given above to reduce the number of design
variables. The intent is to make the problem more tractable and practical for optimization. The resulting
eleven design variables are shown in Table 3.

NASA/TM-20220015470 24
 TABLE 3.—DESIGN VARIABLES FOR THE OPTIMIZATION PROBLEM
Propeller design variables
Pdes Design shaft power, bhp (per powerplant)
Ndes Design shaft speed, rev/min
Vdes Design airspeed, kcas
Hdes Design altitude, ft
CLi,des Design integrated lift coefficient
Pmax / Pdes Ratio of maximum shaft power to design shaft power
Nmax / Ndes Ratio of maximum shaft speed to design shaft speed
Wing design variables
S Wing reference area, ft2
AR Wing aspect ratio
flap Takeoff flap deflection angle, deg
Mission design variable
Vcr,des Design cruise airspeed, kcas

The abridged list is the result of the following simplifications:

1. The propeller diameter, D, is set by a disk loading assumption. Torenbeek (in Chapter 6 of
Reference 50) provides a regression model for propeller disk loading (Pmax /Aprop, or 4Pmax /D2)
for a variety of airplanes powered by reciprocating and turboprop engines. The correlating
variable is (PmaxVcr)0.5, for reasons explained in the reference. Torenbeek’s regression curve and
his data are reproduced in Figure 9, with the disk loading of the final, optimized single-rotation
ORBET variant located on the curve.
2. The number of propeller blades, B, is set to five. The relatively high blade count is chosen for its
generally positive influence on performance, noise and vibration.
3. The family of United Technologies Corporation HS-1 propeller airfoils (Ref. 66) described in
Section 5.0 is used for the propeller airfoil stack. These airfoil sections have good lift
performance during takeoff and climb, good lift-to-drag ratio in cruise, and they are popular in
low-speed applications. The propeller helical tip Mach number, Mhel, is modest to avoid
supersonic flow on the upper blade surfaces. Section aerodynamics are computed using the tool in
Reference 61. Airfoil sections are assigned to the spanwise locations suggested in Reference 65.
4. The XROTOR tool (Ref. 63) is used to set the propeller spanwise chord distribution based on the
input design integrated lift coefficient, CLi,des. Put another way, CLi,des is a design variable that
replaces the need to specify the propeller chord distribution.
5. The XROTOR tool is allowed to set the propeller spanwise twist distribution such that induced
loss is minimized at the design condition (Ref. 64).
6. The NACA 64-421 airfoil section is used from the wing root to the propulsion pod. The thinner
NACA 64-415 airfoil section is used from the pod to the tip. Section aerodynamics are computed
using the tool in Reference 61.
7. The wing is a basic trapezoid with a taper ratio of 0.42, wing twist washout angle of 1.25, and
quarter chord sweep of 1.6. These are reasonable values for a low-speed application.
8. Hcr,des is set to 10,000 ft (the maximum allowed). This is based on observations made during
preliminary optimizations which determined that airplane range is always maximized at that
condition.

NASA/TM-20220015470 25
 Figure 9.—Propeller disk loading model, adapted from Reference 50.

Using the propeller disk loading characteristic (adapted from Reference 50 and shown in Figure 9)
does not mean that propeller diameter is neglected in the abridged optimization. Rather, it is equivalent to
addressing diameter inside a nested optimization subloop under the top-level optimizer. But instead of
subjecting diameter to a formal suboptimization, it is set directly by the empirical disk loading
characteristic, where optimization has been done previously for other propeller designs. From actuator
disk theory, it can be straightforwardly shown that efficiency is a function of disk loading, and that as
design airspeed increases, higher disk loadings are generally required. So the independent variables in the
correlating variable (i.e., maximum shaft power, Pmax, and true cruise airspeed, Vcr, are sensibly chosen: it
is reasonable to assume that the optimum disk loading is proportional to PmaxVcr. The correlation of data
plotted in Figure 9 seems to justify the assumption. So stated another way, replacing diameter
optimization with a disk loading characteristic is like relying empirically on a set of historical diameter
optimizations – with respect to Pmax and Vcr – made by others for different applications in the past.
Similarly, allowing XROTOR to set the spanwise chord and twist distribution is not neglecting them
in the abridged optimization. Instead, they are addressed in a nested subloop where they are optimized for
minimum induced loss and to meet lift requirements set by CLi,des. The end result of these simplifications
is to reduce the burden on the top-level optimizer, making the problem more tractable and practical.

7.5 Formal Statement of the Abridged Optimization Problem

The abridged ORBET sizing and optimization problem consists of ten design variables. An extended
design structure matrix diagram (XDSM, (Ref. 78)) of the abridged problem is shown in Figure 10. The
design variables may pertain to one of the four local disciplinary analyses, or they may be among those
shared globally by multiple disciplines. Design variables are contained either in the vector of shared
global variables, x0 (having a zero subscript), or in one of the local discipline vectors, x1 through x4, with
the subscript denoting the disciplines of propellers, weights, aerodynamics, and performance, respectively.
The vector x (without extra notation) contains all of the design variables (i.e., x = [x0T, x1T, x2T, x3T, x4T]T).
The vector of initial values of x is denoted by x(0).

NASA/TM-20220015470 26
 Figure 10.—Extended design structure matrix (XDSM) diagram of abridged optimization problem.

TABLE 4.—SIZING AND OPTIMIZATION VARIABLES

Design variables:
Vector Discipline Contents
x0 Shared [S, AR, Vdes, Hdes]
x1 Propellers [Pdes, Ndes, CLi,des, Pmax/Pdes, Nmax/Ndes]
x2 Weights []
x3 Aerodynamics [flap]
x4 Performance [Vcr,des]
Response variables:
Vector Discipline Contents
y1 Propellers [D, Pmax, Nmax, AF, Jdes, Mhel,des, performance data]
y2 Weights [Etot]
y3 Aerodynamics [aerodynamic data]
y4 Performance [R, dto, dland, 400, ROCsvc]

Using the same subscripts, response variables computed by each discipline analysis are contained in
the vectors y1 through y4. A vector of design constraints (given in Section 7.0) is denoted by c. The
objective function, f0, is the negative of mission range, R, to be minimized. The shared objective, f0, is a
function of design variables (x) and response variables (y, which are in turn functions of x). Design
variables or response variables at their optimal value (i.e., when f0 is optimal) have asterisk superscripts.
Design variables can be under manual control or they may be under control of the optimizer. These two
analysis modes are indicated by the multidisciplinary analysis (MDA) path or the optimization path in the
diagram, respectively. Design variables and response variables corresponding to those in the diagram are
shown in Table 4.

NASA/TM-20220015470 27
 Note that the local design variable x2 is a zero vector (i.e., a vector of length zero), since the weight
discipline uses response variables and shared design variables as inputs. Also note that the responses from
the propeller analysis (y2) and the aerodynamic analysis (y3) contain tabular data files. Propeller
performance data consisting of thrust and shaft power are predicted as functions of altitude and airspeed.
Aerodynamics consisting of lift and drag coefficients are predicted as functions of angle of attack, flap
deflection angle, and Reynolds number. These sets of tabular data are required for the vehicle mission
performance analysis. New sets of tabular data files are computed and used in the performance analysis
whenever design variables change.
All design variables are optimized together at once rather than in parallel or nested groups. That is,
the optimization occurs in series, with each disciplinary subproblem executed whenever one or more of its
design variables are changed. The optimizer is relied upon to discover a feasible solution with sensible
design values for each disciplinary subproblem. The propeller design subproblem, for example, is
optimized with respect to mission range. Classical propeller design optimizations often used propeller-
centric aerodynamic metrics such as efficiency as a sole objective. Such designs were often based on the
theories of Betz (Ref. 64). But there are usually competing performance requirements such that propeller
designs for best cruise efficiency and propeller designs for best takeoff and climb performance are
different. Recommended propeller design practices for these types of problems are summarized in
Reference 79, for example. More recently, there have been multidisciplinary propeller optimizations that
account for the entire vehicle (e.g., Ref. 80). Optimization considering multiple disciplines is important
because the goal is often more complex than simply maximizing propeller efficiency or thrust. ORBET’s
sizing is a mixed-variable multidisciplinary optimization problem, where airframe and propeller variables
are optimized jointly.
A formal statement of the problem can be written as:

Minimize f0 (x, y(x))

With respect to x
Subject to c

The constraints in c are formulated so that they contribute to the simple penalty function on f0 used by
the particle swarm optimizer. The compound inequalities for Jdes, CLi,des, and AF are each broken into two
simple inequality constraints. Design variables are scaled so that their magnitudes are on the order of unity.
As discussed above in Section 6.0, battery cell specific energy, egrav,cell, is treated as a system
technology parameter. That is, it is not a continuous-real design variable that changes during a sizing
analysis. Rather, it is treated as a constant in any given sizing and optimization, but it may be changed for
other optimizations. It is varied manually after the optimization is complete until the 250 nmi range goal
is met. Alternately, the optimization problem could be reformulated to minimize egrav,cell subject to a range
constraint, but that would increase the cost of optimization.

8.0 Candidate Airplanes

8.1 Single-Rotation Propeller Variant
Optimized design variables for the single-rotation propeller variant (see Figure 3, left) are given in
Table 5. Other characteristics of the design are shown in Table 6. Batteries are predicted to weigh
6622 lb. As discussed in Section 6.0, specific energy at the cell level (egrav,cell) is a system parameter that
indicates the level of battery technology needed to meet the goal mission range. For the design case,
egrav,cell is varied so that the range (with 45 min reserves) is 250 nmi. For this range, the batteries are

NASA/TM-20220015470 28
 TABLE 5.—OPTIMIZED DESIGN VARIABLES

Pdes, bhp 587 Ndes, rev/min 1170

Vdes, ktas (kcas) 250 (216) Hdes, ft 10,000
CLi,des 0.350 Pmax / Pdes 2.00
Nmax / Ndes 1.37 S, ft2 317
AR 19.0 flap, deg 16
Vcr,des, ktas (kcas) 180 (155) −−−−−−− ---

TABLE 6.—CHARACTERISTICS OF THE DESIGN

Vehicle Power
Maximum gross weight, lb 18,000 Cruise motors, kW 210
Payload weight, lb 3800 Climb-assist motors, kW 670
Battery pack weight, lb 6622 Overall electrical system knockdown factor 0.6
Thrust loading (sea level, static) 0.508 Specific energy (cell, egrav,cell), W-h/kg 600
Wing loading, lb/ft2 56.8 Specific energy (system, 0.6 egrav,cell), W-h/kg 360
Wingspan, ft 77.6 Total mission energy, kW-h 1081

Propellers Performance
D, in. (area, Aprop, ft2) 118 (75.3) Cruise altitude, ft 10,000
AF (per blade) 85.4 Cruise airspeed, ktas (kcas) 180 (155)
Jdes 2.21 Cruise lift-to-drag ratio 20.5
Mhel,des 0.681 Critical field distance (sea level, 59 °F), ft 4000
N (sea level, static), rev/min 1600 Takeoff distance to 35 ft (sea level, 59 °F), ft 2750
Installed thrust (sea level, static), lb 4569 Range (with 45 min reserves), nmi 250

required to supply 1081 kW-h of energy. If the batteries are to occupy the 150 ft3 volume underneath the
cabin floor, they would require a volumetric energy density of 420 W-h/l at the pack level. Packaging
batteries into this volume seems practical at today’s technology levels (e.g., Ref. 69).
Cell specific energy, however, is a different matter. The requirement for egrav,cell to meet the range
goal is 600 W-h/kg. This is more than twice the specific energy of modern Type 2170 Lithium-ion
(Li-ion) cylindrical cells in use by Tesla (e.g., Ref. 69). Doubling the specific energy of Li-ion batteries is
doubtful (Ref. 81), so substantial specific energy gains for batteries using this chemistry seem unlikely.
An all-electric airplane of this type and with these performance requirements seems limited by today’s
Li-ion battery technology. But increases in battery cell specific energy may be possible with other battery
chemistries. Laboratory experiments on Lithium-Sulfur (Li-S) batteries show promise of reaching in the
near term the specific energy levels required by ORBET (e.g., Refs. 82 to 84). And battery manufacturers
Sion Power and Oxis Energy expect that future Li-S cells will have specific energy levels of up to
600 W-h/kg (Ref. 85). Provided that issues such as poor cyclability, anode corrosion, and the low
electrical conductivity of sulfur can be resolved, Li-S batteries could improve prospects for transports like
ORBET.
Three variables are at their assigned limits. The design variables, AR and Hdes, have values of 19 and
10,000 ft, respectively. The response variable, dto, has a value of 4000 ft. Solutions having optimized
design variables at their limit is a possible cause for concern. As discussed earlier, however, the limit for
AR is viewed as non-mandatory or flexible. Further study is required to determine the influence of higher
values of AR on wing weight and aerodynamics, and the influence of higher values of Hdes on propeller
performance.

NASA/TM-20220015470 29
 The optimized propeller design airspeed, Vdes, of 250 ktas (or 216 kcas when Hdes is 10,000 ft) is
seemingly odd considering that ORBET never exceeds 180 ktas in its design mission. But this is a
consequence of relying on an optimizer to design a propeller that performs well at takeoff, climb, and
during cruise. The propeller design point has a high efficiency, but it occurs at an advance ratio where the
propeller ordinarily never operates (shown in the propeller performance map in Figure 11). Nonetheless,
the propeller’s design is such that it provides good overall performance during all phases of the design
mission. Uninstalled propeller performance data are shown in Table 7. An alternate 250 ktas high-speed
cruise condition is also shown in the figure and in the table, but it is not part of the design mission.
Though the high-speed cruise condition enjoys a higher propeller efficiency, flying at 250 ktas requires
considerably more power than flying at 180 ktas, and it results in a lower cruise lift-to-drag ratio and less
range.
The active constraint for the critical field length, dto, is perhaps the most impactful. Without a field
distance requirement, ORBET’s required thrust and wing area could be much lower before other
constraints become active, resulting in better performance. This is a classic airplane sizing problem: if
wings are sized for efficient cruising at a high lift-to-drag ratio, then takeoff and climb requirements often
go unmet. In the case of ORBET, a solution for Pmax / Pdes and Nmax / Ndes is found that allows its
propellers to operate with more power and rotational speed during takeoff and climb. This behavior
mimics the natural thrust lapse of a reciprocating or a gas turbine engine. ORBET’s powerplant has
greater operational flexibility than a fuel-burning engine, since it is not limited to a natural thrust lapse.
Indeed, since shaft power and shaft speed are readily uncoupled for electric motors, propeller thrust
can be optimized somewhat for any given shaft power and flight condition. In the case of ORBET, this is
accomplished as discussed in Section 5.0. In a real airplane, this additional benefit would be implemented
using a full-authority digital system that controls propeller pitch and shaft speed.

Figure 11.—Single-rotation propeller performance map.

NASA/TM-20220015470 30
 TABLE 7.—UNINSTALLED PROPELLER PERFORMANCE AT SELECT CONDITIONS

Optimal range High-speed

Design Static Liftoff Top of climb
cruise cruise
P, bhp 587 1174 1174 822 277 576
P/Aprop, bhp/ft2 7.80 15.6 15.6 10.9 3.7 7.6
N, rev/min 1170 1600 1600 1480 1150 1210
V, ktas (kcas) 250 0 120 180 180 250
H, ft 10,000 0 0 10,000 10,000 10,000
J = V / nD 2.21 0 0.776 1.258 1.619 2.13
Mhel 0.681 0.735 0.757 0.758 0.615 0.698
CP = P / D5n3 0.276 0.159 0.159 0.191 0.137 0.242
CT = T / D4n2 0.114 0.298 0.164 0.134 0.076 0.104
 = JCT / CP 0.913 0 0.796 0.882 0.899 0.912
Thrust, lb 699 4631 2540 1313 450 686

Figure 12.—Airplane aerodynamics.

For ORBET, the cruise motors and climb-assist motors (described in Section 7.0) could be rated at
210 and 670 kW, respectively. During descent, it is possible that the motors could act as generators to
charge the batteries. This energy recuperation technique – sometimes called “energy harvesting” – is not
studied here.
In addition to the 4000 ft critical field distance, the takeoff distance to a 35 ft obstacle with all engines
operating is of interest. From a sea level field at 59 °F, this distance is 2750 ft. This allows ORBET to
operate from most runways. Airplane aerodynamics for the optimized design in clean and takeoff
configurations are shown in Figure 12. A weight statement is shown in Table 8. Sensitivity studies using
this variant are discussed in Section 8.3.

NASA/TM-20220015470 31
 TABLE 8.—WEIGHT STATEMENT (WEIGHTS IN POUNDS)

Wing 1479
Horizontal tail 140
Vertical tail 125
Fuselage 1827
Landing gear 639
Nacelles 199
Structure total 4409
Electric motors 385
Propellers 345
Battery packs 6622
Propulsion total 7352
Surface controls 163
Hydraulics 147
Power electronics 593
Avionics 235
Furnishings and equipment 840
Anti-icing 50
Systems and equipment total 2028
Weight, empty 13,789
Flight crew (2) and baggage 410
Operating weight, empty 14,199
Passengers (19) 3230
Passenger baggage 570
Maximum takeoff gross weight 18,000

8.2 Contra-Rotation Propeller Variant

A contra-rotation propeller (Figure 3, right) is assessed using the method described in Section 5.0. It
is installed on the vehicle optimized for the single-rotation propeller. No additional vehicle optimization
is performed. In a contra-rotation configuration, the rear propeller receives the flowfield created by the
front propeller. The aft row turns rotational flow back into the streamwise direction. Isolated efficiency of
the aft row decreases with increasing flowfield speed, but the overall efficiency (considering both rows)
may increase due to recovery and conversion of swirl losses into useful forward thrust. Properly designed,
swirling flow and torque reaction of a contra-rotation propeller will be near zero. Reducing swirling flow
on the vertical tail also can improve handling qualities at lower speed. A contra-rotation configuration
seems attractive for propellers driven by electric motors due to the comparative ease of implementation.
Two electric motors could straightforwardly be arranged to drive forward and aft propellers, whereas
contra-rotation propellers driven by reciprocating engines or by gas turbine engines require elaborate
geartrains or contra-rotating power turbines.
Aircraft with contra-rotation propellers have been mass-produced principally by the United Kingdom
and by Russia or the former Soviet Union. Examples are military transports such as the Tupolev Tu-95,
Antonov An-70, Avro Shackleton, and Fairey Gannet. Few civil passenger transports have been equipped
with contra-rotation propellers. Perhaps the most notable was the Tupolev Tu-114 passenger transport
(developed from the Tu-95). A Ukrainian civil version of the An-70 is a convertible cargo/passenger

NASA/TM-20220015470 32
transport that can be equipped with seats in removable modules. Cancelled civil programs include the
Bristol Type 167 Brabazon and the Douglas DC-8 piston-engine airliner.
Contra-rotation propellers have different noise characteristics than single-rotation propellers. Perhaps
chief among the differences are additional discrete interaction tones and their harmonics created by
trailing edge wakes departing the front propeller and being intercepted by the aft propeller. Thus, contra-
rotation propellers can be particularly rich in tone content. In instances where the tones occur at similar
frequencies, they can be strengthened via constructive reinforcement or be reduced via destructive
cancellation (which was recognized perhaps as early as 1948 (Ref. 86)). Contra-rotation propellers can be
noisier in the axial direction (e.g., Ref. 87). But they can be made quieter if noise mitigation strategies are
used. Unequal blade counts or shaft rotational speeds can change the strength of discrete interaction tones
(Refs. 88 and 89), which could be accomplished easily in an electric motor-driven application such as
ORBET with independent shaft speed control. Other known noise mitigation strategies for contra-rotation
propellers include aeroacoustic blade shaping, blade pitch angle and rotational speed optimization,
increased blade counts, low disk loading, blade spacing optimization, and aft blade clipping (Ref. 90).
In this study, a contra-rotation propeller is designed very simply without a great deal of rigor using
many characteristics taken from the optimized single-rotation propeller. Shaft power of a single propeller
is divided equally between two propellers (each absorbing Pmax / 2), rotating in opposite directions. For
each propeller, values for CP, J, and Mhel at the aerodynamic design point (Vdes and Hdes) are maintained,
as are the number of blades and the design variables CLi,des, Pmax / Pdes, and Nmax / Ndes (see Table 5 and
Table 7). These propellers are designed using the XROTOR tool in the presence of interdependent
flowfields calculated as described in Section 5.0. New twist and chord distributions are computed for
front and aft propellers. To maintain comparable thrust performance, blade diameters of both blade rows
are reduced equally until the uninstalled design thrust of the contra-rotation propeller is identical to the
single propeller.
This simplistic design approach results in little performance improvement. Uninstalled efficiency at
the aerodynamic design point increases by less than one point. Propeller weight increases by 25 percent
for the contra-rotation configuration. When mission performance is evaluated, there is virtually no net
range benefit. This may come as a surprise, given some of the efficiency increases reported for contra-
rotation propellers (e.g., Ref. 91). Part of the explanation for the disappointing performance is certainly
due to the simplistic design approach used here. In contra-rotation configurations, the blade rows are in
flowfields that differ from the freestream. Compared to the isolated case, the forward propeller receives
higher axial velocity, provided the blade rows are sufficiently close. More importantly, the downstream
propeller receives higher axial velocity and an additional rotational flow component. Since the resultant
velocities are higher for the aft propeller, and if given the same lift requirement as the front propeller, its
chords should be generally narrower than the front propeller. And if optimized for minimum induced loss,
the aft propeller should have less twist than the front propeller. But in this study, each propeller is
designed for minimum induced loss separately, rather than as a system. Additional benefit might be
realized if spanwise chord and twist distributions are set as a system, and if additional design variables are
introduced.
But, the disappointing performance of the contra-rotation propeller might be explained more simply:
there may not be enough swirl to justify its removal. Given ORBET’s low speed, the optimal design for
its single-rotation propeller has a rather low disk loading and a very high efficiency. Relative to more
highly loaded propellers, a propeller with lower loading and higher efficiency imparts comparatively little
rotation to its wake. The swirl losses introduced by the front propeller may not be great enough to justify
a contra-rotating aft propeller. Dramatic efficiency gains attributed to contra-rotation in the literature are
often for propellers with disk loadings three or four times higher than ORBET.

NASA/TM-20220015470 33
8.3 Sensitivity Studies and Alternative Designs Using the Single-Rotation Variant
Several sensitivity studies and alternative designs are performed using the single-rotation propeller
variant as described in this section. In all cases, the design variable values shown in Table 5 are
unchanged, though other aspects of the vehicle or the way it is used are changed.

8.3.1 Payload-Range Performance

A traditional payload versus range diagram is shown in Figure 13 (left). It is rather unlike similar
diagrams for fuel-burning airplanes, where gross weights increase as fuel and/or payload is added to their
operating empty weights. In the case of ORBET, battery weight remains constant despite its state of
charge, so the only instance when gross weight changes is when payload weight changes. Thus, there are
only two linear segments in ORBET’s diagram. The horizontal segment occurs at maximum payload
weight (3800 lb) and at maximum gross weight (18,000 lb) as the state of charge varies. The descending
segment occurs as payload is offloaded. Gross weight falls to 14,200 lb for the case of a self-ferry
mission, where range with reserves increases from 250 nmi at the design point to 301 nmi.
Other ORBET variants are possible if passengers and their baggage are replaced by additional
batteries, as shown in Figure 13 (right). For these alternative configurations, maximum gross weight is
held constant at 18,000 lb in all instances, but battery weight is allowed to increase from 6622 lb as
payload is reduced. The additional batteries (provided that interior volume can be found for them) allow
more mission energy to be stored so that range may be improved. Note that with maximum seating for
less than ten passengers, ORBET would fall into a different classification type (see 14 CFR§23.2005 of
Reference 14). Level 3 airplanes, for example, have maximum seating for between seven to nine
passengers, and they have different Part 23 airworthiness requirements. One interesting configuration
(while remaining a Level 4 type) is a ten-passenger variant, with a range of 379 nmi: a 51 percent increase
over the design configuration. A manufacturer could offer similar variants with maximum seating varying
between ten and 19 passengers to accommodate variable needs of operators. A nine-passenger variant
would fall into the Level 3 classification, where a second in command is unnecessary. The weight of the
second in command could be used instead as additional battery weight (as shown in Figure 13).

8.3.2 Influence of Reserves and Cruise Speed

Two other sensitivities are examined where the manner in which the airplane is flown is changed. It is
of interest to determine the maximum range when energy intended for reserves is reduced from the design
case, as shown in Figure 14 (left). Ordinarily, the design mission allows for 45 min of additional flying
time at 180 ktas and 8000 ft. If reduced to zero, range increases to 437 nmi. This is the maximum range at
full payload in the event there is a diversion to an alternate field.
The second sensitivity examined is the influence of cruise airspeed, as shown in Figure 14 (right).
With 19 passengers and with 45 min of reserves, flying at 250 kcas (the maximum airspeed allowed under
Part 23 for low-speed types per 14 CFR§23.2005 of Reference 14) reduces range to just 132 nmi. Higher
airspeed results in a lower lift-to-drag ratio (see Figure 12).

NASA/TM-20220015470 34
 Figure 13.—Traditional payload-range diagram (left); replacing payload with batteries (right).

Figure 14.—Reserves sensitivity (left) and cruise speed sensitivity (right).

Figure 15.—Influence of aspect ratio (left) and empty weight (right).

8.3.3 Influence of Aspect Ratio and Empty Weight

The influence of aspect ratio and wing bracing is investigated. There is a significant range
improvement associated with increasing aspect ratio (see Figure 15, left). Wing weight is predicted as
described in Section 3.0. Wing weight increases with increasing aspect ratio, but aerodynamics and
mission range improve. A brace is used for all aspect ratios greater than 13 and aspect ratio is limited to a
maximum of 19. A wing brace is considered to be an enabling technology for higher aspect ratios. Weight
is discontinuous when a brace is added due to the weight reduction possible with a brace.
The influence of underpredicting empty weight is also investigated to understand its resulting range
penalty (see Figure 15, right). Here, mission energy is reduced as batteries are displaced by ballooning

NASA/TM-20220015470 35
 Figure 16.—Influence of specific energy (left) and electrical system knockdown factor (right).

empty weight in other categories with the 18,000 lb gross weight held constant. There is a 3 percent loss
in range for every 100 lb increase in empty weight. Conversely, range can be increased if operating empty
weight can be reduced. For example, if the second in command is not necessary (see the discussion in
Section 3.0), then range improves by 6 percent.

8.3.4 Influence of Power System Technologies

The sensitivity of range to battery cell specific energy is investigated for values of egrav,cell different
than the 600 W-h/kg required for the ORBET baseline (see Figure 16, left). As discussed already in
Section 8.0, the cell specific energy required for the ORBET baseline is beyond the capability of current
Li-ion batteries. But other battery chemistries with higher cell densities may be possible in the future.
The influence of power system technologies (discussed in Section 6.0) is also of interest. Sensitivity
of range to the electrical system knockdown factor are shown in Figure 16, right.

9.0 Low-Noise Operating Mode

The noise of a propeller is a strong function of its tip speed. As a practical matter, this observation, of
course, must have been made quite early on. But ordinarily, good thrust production is at odds with low tip
speeds, and so designers are reluctant to build propellers that turn too slowly. Nevertheless, for airplanes
equipped with variable-pitch propellers, an opportunity exists for operation at low shaft speed and
maximum shaft power. At low shaft speeds, high shaft power levels and high blade pitch angles can be
combined to recover much of the thrust that would otherwise be lost. This could enable a low-noise
operating mode for propellers that are normally designed for performance rather than for noise. A pilot
could switch between low-noise and high-performance operating modes as necessary and when
practicable. Low-noise takeoffs at airports where noise is considered a problem may be possible.
Quieting propellers by means of tip speed reduction has been elusive. This has been especially true in
the case of reciprocating engines, where mechanical shaft power and shaft speed are closely coupled. The
maximum rated shaft power is typically produced at or near peak shaft speed. If an airplane equipped with
a reciprocating engine and a variable-pitch propeller attempts a low-noise takeoff by reducing the
propeller’s tip speed, propeller power and thrust are reduced. Such takeoffs are not tolerated due to
punishing performance effects, such as increased field lengths and poor climb rates. With a reciprocating
engine, a mechanical gearbox with a variable gear ratio would be necessary to deliver peak shaft power at
an arbitrary shaft speed, and it would be a means to accomplish high- and low-speed operations of a
propeller. However, shifting gears (quite literally on the fly) has not become popular, despite at least one
serious investigation unrelated to noise in 1941 (Ref. 92). A continuously variable transmission could

NASA/TM-20220015470 36
manage the job also, provided it could be made lightweight, reliable, safe, and able to handle internal
forces well enough. Even given a turboshaft engine with a free power turbine (where shaft power and
shaft speed can be uncoupled), modifying propellers to operate at a low tip speed to reduce noise has not
become popular, perhaps because it is at odds with generating thrust efficiently or because of
implementation issues.
Electric motors, and the relative ease by which their speed and power may be controlled, may be the
motivating force of change. Certain electric motors are able to deliver maximum shaft power over a range
of shaft speeds. Induction and synchronous electric motors, furnished with appropriate power
management equipment and speed controllers, are able to vary shaft speed by adjusting the frequency of
the power supplied to the motor. In application, the shaft torque can be made constant from rest to the
shaft speed defined by the so-called rated frequency. But, above the rated frequency, the motor is in the
field flux control, or constant voltage regime. Here, the torque diminishes with shaft speed, and the shaft
output power remains relatively constant. The constant-power operating regime is the focus of this
section. In practice, this behavior can be loosely described as an “electronic gearbox” that allows peak
rated shaft power output to occur at a selectable range of shaft speeds. This concept is discussed for
application to general aviation airplanes that ordinarily use piston engines in Reference 93.
Note this concept is not related to the design of a low-noise, electrically driven propeller. Indeed, that
is an interesting design and optimization problem of larger scope (e.g., Ref. 94). Designing a propeller for
low noise is usually at odds with generating thrust efficiently. The six-blade, wide-chord, low-speed
propeller designed for the Lockheed prototype QT-2PC quiet observation plane (Ref. 95) is an example of
this conflict. Instead, this study is intended to evaluate propellers already designed for thrust and
efficiency in the low-noise mode described.
An explicit noise analysis is not performed in this study. However, takeoff reference profiles
performed at reduced shaft speeds are computed. A qualitative estimate of noise reduction is given, and
comments are offered on implementation in a certification setting.

9.1 Implementation as a Selectable Noise Reduction System

For takeoff and initial climbout using maximum shaft power, two operating modes are envisioned for
ORBET: 1) a performance mode at a high shaft speed where thrust is maximized, and 2) a low-noise
mode at a lower, selectable shaft speed. In normal operations, the mode would be chosen prior to takeoff
by the flight crew, given considerations for takeoff field performance, downrange obstacles, climb rate,
and, of course, community noise.
For noise certification of new, propeller-driven light airplanes such as ORBET, the provisions of
Chapter 10 of the ICAO’s Annex 16 (Ref. 33) apply. In a noise test for airplanes of this type, the airplane
is required to take off, climb, and fly directly over a noise observation monitor on the ground, located on
the extended centerline of the runway 2500 m (8202 ft) from brake release. At the noise monitoring
station, a single microphone is flush-mounted over an acoustically hard surface at ground level. The two
operating modes could be classified as a “selectable noise reduction system,” or a SNRS, described in the
ICAO Environmental Technical Manual (Ref. 96). As envisioned, an applicant would define selectable
takeoff operating modes in its airplane flight manual or in its pilot’s operating handbook. The regulating
authority would need to approve procedures for each operating mode. The authority might certify only the
noisiest of the available SNRS options for Chapter 10 compliance, but since (to date) no applicant has
come forward with a SNRS, this remains to be seen. If the noisiest mode is the only mode certified, then
the low-noise mode would be a means to reduce noise in operational practice only.

NASA/TM-20220015470 37
9.2 Takeoff Reference Profiles
If implemented as a SNRS, maximum shaft power would be constant throughout the selectable range
of shaft speeds. Within this range, electric motor efficiency would of course vary. In practice, and at least
without any noise considerations, the shaft speed that maximizes the combined efficiency of the electric
motor, propeller system, and the entire airplane for a given flight condition and thrust requirement would
be preferred. This optimization should of course be mindful of motor and propeller operating limits, and it
should be an interesting engine control system challenge (Ref. 97). A full-authority digital engine control
system and a clever propeller hub mechanism might be developed to schedule the propeller speed and
blade pitch angle with motor system characteristics and airplane thrust requirements. Note that in this
study, electric motor efficiency is not explicitly considered (it is incorporated in the electrical system
knockdown factor).
Takeoff reference profiles using four shaft speeds are evaluated for noise certification under
Chapter 10. Referring to Table 7, ORBET’s takeoff power is 1174 bhp per motor. At this power,
maximum thrust occurs at 1600 rev/min. This condition represents the takeoff performance mode setting.
Reduced shaft speeds of 1300, 1200, and 1100 rev/min are considered for candidate low-noise modes.
Note that absorbing large amounts of power as the shaft speed falls is not perpetually sustainable. If the
shaft speed continues to fall and the blade pitch increases to compensate, eventually, the propeller blades
will stall. The ORBET propeller, expected to absorb 1174 bhp at sea level static, will begin to stall at
about 900 rev/min.
Certification procedures are used to compute the takeoff profiles. Following the guidance given in
Reference 33, calculations are made at maximum gross weight for a sea level runway at 59 °F and zero wind
using FLOPS software (Ref. 34). At brake release, maximum power is applied and flaps are deployed at
their takeoff setting. Once aloft, the landing gear is retracted and the airplane changes to its climb
configuration at a point determined by the applicant. In the climb configuration, flaps may be repositioned,
and airspeed must be maintained at a speed that maximizes the rate of climb. In the case of ORBET,
maximum climb rate over the noise monitor occurs when flaps are retracted at 200 ft and when the climbout
speed is held at 122 kcas. Takeoff reference profiles using the four shaft speeds are shown in Figure 17.
Procedures from Section 10.5.2 of Reference 33 require engine “takeoff power” to be selected and
maintained throughout the noise test to a point beyond the noise measurement location. But takeoff power
may seem a bit hazily defined when discussing airplanes having electrical power, especially when (as in this
study) maximum mechanical shaft power can occur over a selectable range of shaft speeds. To put it another
way, maximum power does not necessarily coincide with maximum thrust. In the case of light aircraft
powered by reciprocating engines or small turbine engines, the intent of the regulation is to require airplanes
to operate at their maximum takeoff-rated engine power as defined in the Airplane Flight Manual or the
Pilot’s Operating Handbook. In the case of a small electric airplane, it is possible that a manufacturer could
list more than one engine “takeoff power” rating in its manual. Hypothetically, two takeoff ratings could be
defined for ORBET: both would use maximum shaft power, but they would have different shaft speeds. For
a short field and best climb performance, the propeller would be set for maximum thrust at 1600 rev/min.
For a quieter takeoff, a second takeoff setting could be defined at a lower propeller speed.
The 1300 rev/min takeoff profile seems to be an attractive candidate for the low-noise mode. Relative to
the performance mode, field distance to the 35 ft runway obstacle increases by only 9 percent, and climb
rate decreases by only 4 percent. But tip speed is 19 percent lower relative to the performance mode. If
propeller noise varies (roughly) by tip speed to the fifth power, this would result in a generalized source
noise reduction of 4 to 5 dB. For a receiver at ground level, this noise benefit would be eroded by about
0.7 dB due to the altitude loss (see Figure 17, top). A more detailed study of takeoff performance and noise
is needed.

NASA/TM-20220015470 38
 Figure 17.—Influence of shaft speed on takeoff reference profile.

10.0 Summary
A notional, 19-passenger, fully electric transport with a braced wing and two propellers is
investigated. Parametric system weight, aerodynamic, propeller, and mission performance models are
developed to aid sizing and optimization. Battery cell specific energy is treated as a technology parameter
that is varied to determine its influence on mission range. To achieve a minimum range success criterion
of 250 nmi with 45 min of reserve power, it is found that battery cell specific energy must be at least
600 W-h/kg, more than twice the capability of today’s lithium-ion cells. Success of the ORBET design is
dependent on progress in battery technology.

NASA/TM-20220015470 39
 Single-rotation and contra-rotation propellers are studied. The contra-rotation propellers are not found
to improve performance. Swirl losses introduced by the front propeller may not be great enough to justify
a contra-rotating aft propeller. Implementation of a novel selective noise reduction system is proposed.
This study is intended to broadly define a relatively near-term, fully electric, 19-passenger airplane. It
was conducted as a part-time effort lasting just nine months. Due to the short time involved, this study
does not assess the concept’s marketability, its operating costs, or the additional airport infrastructure
required to support it. Future work is needed to model the concept and its economics more rigorously.

11.0 Future Work

This study would benefit from refined analyses in all of its disciplines. One possibility is to place it
within research focused on the Transonic Truss Braced Wing concept (Ref. 13). NASA, FAA, industry
and academia are studying the TTBW through the Sustainable Flight National Partnership (Ref. 12).
Though ORBET is much smaller and cruises at speeds considerably slower than the TTBW, their braced
wings are architecturally similar. ORBET could serve as another vehicle concept under those studies. A
smaller vehicle could act as an additional challenge to the tools under development in NASA’s model-
based systems analysis initiative (Ref. 98).

References
1. Conner, M. (ed.): “NASA Armstrong Fact Sheet: NASA X-57 Maxwell,” NASA X-57 Fact Sheet,
Sept. 2018 [URL: https://www.nasa.gov/centers/armstrong/news/FactSheets/FS-109.html, accessed
Aug. 2022].
2. Gipson, L. (ed.): “Hybrid Wing Body Goes Hybrid,” NASA N3-X Fact Sheet, Feb. 2013 [URL:
https://www.nasa.gov/content/hybrid-wing-body-goes-hybrid/, accessed Aug. 2022].
3. Dambowsky, F. (ed.): “Hybrid-Electric Propulsion Systems Enable more Climate-Friendly Air
Transport,” DLR Regional Aircraft Fact Sheet, Nov. 2021 [URL:
https://www.dlr.de/content/en/articles/news/2021/04/ 20211122_hybrid-electric-propulsion-enable-
more-climate-friendly-air-transport.html/, accessed Aug. 2022].
4. Jaggi, R.: “This Radical New eVTOL Uses 30 Tilting ‘Jet’ Fans to Get in the Air and Stay There,”
Lilium eVTOL, Robb Report, Aug. 2022 [URL: https://robbreport.com/motors/aviation/30-jet-evtol-
tested-spain-will-it-work-1234739016/, accessed Aug. 2022].
5. Stückl, S.; van Toor, J.; and Lobentanzer, H.: “VOLTAIR - The All Electric Propulsion Concept
Platform - A Vision for Atmospheric Friendly Flight,” In Proceedings of the 28th International
Congress of the Aeronautical Sciences, Brisbane, Australia, pp. 1–11, 2012.
6. Antcliff, K.; Guynn, M.; Marien, T.; Wells, D.; Schneider, S.; and Tong, M.: “Mission Analysis and
Aircraft Sizing of a Hybrid-Electric Regional Aircraft,” AIAA paper 2016-1028, 2016.
7. Marien, T.; Blaesser, N.; Frederick, Z.; Guynn, M.; Kirk, J.; Fisher, K.; Schneider, S.; Thacker, R.,
and Frederic, P.: “Results for an Electrified Aircraft Propulsion Design Exploration,” AIAA paper
2021-3280, 2021.
8. Pornet, C.; Gologan, C.; Vratny, P.C.; Seitz, A.; Schmitz, O.; Isikveren, A.; and Hornung, M.:
“Methodology for Sizing and Performance Assessment of Hybrid Energy Aircraft,” J. Aircraft,
vol. 52, no. 1, pp. 341-52, 2014.
9. Hoelzen, J.; Liu, Y.; Bensmann, B.; Winnefeld, C.; Elham, A.; Friedrichs, J.; and Hanke-
Rauschenbach, R.: “Conceptual Design of Operation Strategies for Hybrid Electric Aircraft,” MDPI
Energies vol. 11, no. 1, 217, 2018.

NASA/TM-20220015470 40
10. Mukhopadhaya, J.; and Rutherford, D.: “Performance analysis of evolutionary hydrogen-powered
aircraft,” ICCT white paper, 2022 [URL: https://theicct.org/wp-content/uploads/2022/07/global-
aviation-performance-analysis-regional-electric-aircraft-jul22-1.pdf-1.pdf, accessed Aug. 2022].
11. Kuhn, H.; Seitz, A.; Lorenz, L.; Isikveren, A.; and Sizmann, A.: “Progress and Perspectives of
Electric Air Transport,” Proceedings of the 28th Congress of the International Council of the
Aeronautical Sciences ICAS, Brisbane, Australia, vol. 6, pp. 4886-4899, 2012.
12. Gipson, L. (ed): “Sustainable Flight National Partnership,” NASA press release, Feb. 2022 [URL:
https://www.nasa.gov/aeroresearch/sustainable-aviation-np/, accessed Aug. 2022].
13. Smith, Y. (ed): “Trans-Sonic Truss-Braced Wing May Help Reduce Fuel Consumption,” NASA
press release, Sept. 2021 [URL: https://www.nasa.gov/image-feature/trans-sonic-truss-braced-wing-
may-help-reduce-fuel-consumption, accessed Aug. 2022].
14. U.S. Code of Federal Regulations, Title 14, Chap. I, Part 23, “Airworthiness Standards: Normal
Category Airplanes.”
15. European Aviation Safety Agency: “Certification Specifications for Normal, Utility, Aerobatic and
Commuter Category Aeroplanes”, CS-23 Initial Issue, 2003.
16. Greitzer, E., et al.: “N+3 Aircraft Concept Designs and Trade Studies, Final Report, Volume 1,”
NASA/CR-2010-216794, 2010.
17. Department of Transportation and National Aeronautics and Space Administration: “Civil Aviation
Research and Development Policy Study,” DOT TST-10-4, NASA SP-265, 1971.
18. Johnson, W.; and Silva, C.: “NASA Concept Vehicles and the Engineering of Advanced Air
Mobility Aircraft,” The Aeronautical Journal, vol. 126, no. 1295, pp. 59-91, 2022.
19. Harish, A.; Perron, C.; Bavaro, D.; Ahuja, J.; Ozcan, M.; Justin, C.; Briceno, S.; German, B.; and
Mavris, D.: “Economics of Advanced Thin-Haul Concepts and Operations,” AIAA paper 2016-3767,
2016.
20. U.S. Department of Transportation: “Essential Air Service,” [URL:
https://www.transportation.gov/policy/
aviation-policy/small-community-rural-air-service/essential-air-service, accessed Sept. 2022].
21. Federal Aviation Administration: “Textron Aviation, Inc., Model 408 Type Certificate Data Sheet,”
TCDS no. A00016WI, rev. 0, March 2022.
22. Federal Aviation Administration: “Airport Data and Information Portal,” Advanced Facility Search,
[URL: https://adip.faa.gov/agis/public/#/airportSearch/advanced, accessed Sept. 2022].
23. Schuh, G.; Spangenberg, M.; Zhang, Q.; Dannbeck, B.; and Stuerken, J.: “Economic Feasibility
Study of a Hybrid-Electric 19-Passenger Commuter Aircraft,” Deutscher Luft- und
Raumfahrtkongress, doc. ID 530307, 2021.
24. Spangenberg, M: “D2.1 Economic Feasibility Study for a 19 PAX Hybrid-Electric Commuter
Aircraft,” Economic Feasibility Study, Clean Sky 2 Grant no. 864551, 2020 [URL:
https://www.rolls-royce.com/~/media/Files/R/Rolls-Royce/documents/innovation/elicad2-1-
economic-feasibility-study-for-a-19-pax-hybrid-electric-commuter-aircraft.pdf, accessed Sept. 2022).
25. Textron Aviation, Inc.: “Cessna SkyCourier: Specification and Description, Serial Number 408-0003
to TBD,” SD-TBP-C408-0622, June 2022.
26. Wolf, D.: “Commuter Airline Perspective,” Joint NASA-FAA On-Demand Mobility and Emerging
Technology Workshop, Arlington, VA, March, 2016.
27. Jansen, R.; Bowman, C.; Clarke, S.; Avanesian, D.; Dempsey, P.; and Dyson, R.: “NASA Electrified
Aircraft Propulsion Efforts,” NASA report no. GRC-E-DAA-TN72947, NATO report no. STO-MP-
AVT-323, 2019.

NASA/TM-20220015470 41
28. Aultman-Hall, L.; Harvey, C.; Sullivan, J.; and LaMondia, J: “The Implications of Long-Distance
Tour Attributes for National Travel Data Collection in the United States,” Transportation, vol. 45,
pp. 875-903,2018.
29. National Academies of Sciences, Engineering, and Medicine: “Interregional Travel: A New
Perspective for Policy Making,” The National Academies Press, Washington, DC, 2016.
30. European Commission, Directorate-General for Mobility and Transport, Directorate-General for
Research and Innovation: “Flightpath 2050: Europe’s Vision for Aviation, Maintaining Global
Leadership and Serving Society’s Needs,” Publications Office, 2011 [URL:
https://data.europa.eu/doi/10.2777/50266, accessed Sept. 2022].
31. U.S. Code of Federal Regulations, Title 14, Chap. I, Part 91, “General Operating and Flight Rules.”
32. Schuh, G.; Spangenberg, M.; and Zhang, Q.: “Economically Driven Requirements for a Hybrid-
Electric 19-Passenger Commuter Aircraft,” IOP Conference Series: Material Science and
Engineering, 1024, 012079, 2021.
33. “Annex 16 to the Convention on International Civil Aviation Environmental Protection,” Vol. I,
Aircraft Noise,” International Standards and Recommended Practices—Environmental Protection,
7th ed., International Civil Aviation Org., Montreal, July 2014.
34. McCullers, L.A.: “Aircraft Configuration Optimization Including Optimized Flight Profiles,
Multidisciplinary Analysis and Optimization Part 1,” NASA CP-2327, pp. 396-412, 1984.
35. European Aviation Safety Agency: “Dornier 228 Series Type Certificate Data Sheet,” TCDS no.
EASA.A.359, issue 07, Dec. 2021.
36. European Aviation Safety Agency: “PZL M28 Type Certificate Data Sheet,” TCDS no.
EASA.A.058, issue 08, Nov. 2015.
37. Federal Aviation Administration: “Viking Air Limited, DHC-6 Twin Otter Type Certificate Data
Sheet,” TCDS no. A9EA, rev. 21, Feb. 2018.
38. Khan, K.; Mallik, W.; Kapania, R.; and Schetz, J.: “Distributed Design Optimization of Large Aspect
Ratio Wing Aircraft with Rapid Transonic Flutter Analysis in Linux,” AIAA paper 2021-1354, 2021.
39. Chakraborty, I.; Nam, T.; Gross, J.; and Mavris, D.: “Comparative Assessment of Strut-Braced and
Truss-Braced Wing Configurations Using Multidisciplinary Design Optimization,” J. of Aircraft,
vol. 52, no. 6, 2015.
40. Nguyen, N.; and Xiong, J.: “CFD-Based Frequency Domain Method for Dynamic Stability
Derivative Estimation with Application to Transonic Truss-Braced Wing.” AIAA paper 2022-3596,
2022.
41. Demasi, L.; Monegato, G; Cavallaro, R.; and Rybarczyk, R.: “Minimum Induced Drag Conditions
for Truss-Braced Wings,” AIAA Journal, vol. 56, no. 12, 2018.
42. Duggirala, R.; Roy, C.; and Schetz, J: “Analysis of Interference Drag for Strut–Strut Interaction in
Transonic Flow,” AIAA Journal, vol. 49, no. 3, 2021.
43. Secco, N.; and Martins, J.: “RANS-Based Aerodynamic Shape Optimization of a Strut-Braced Wing
with Overset Meshes,” J. of Aircraft, vol. 56, no. 1, 2019.
44. Wells, D.P.; Horvath, B.L.; and McCullers, L.A.: “The Flight Optimization System Weights
Estimation Method,” NASA-TM-2017-219627, Volume I, 2017.
45. Hager, R.; and Vrabel, D.: “Advanced Turboprop Project,” NASA SP-495, 1988.
46. Trimble, S.: “Quiet Revolution: Open-rotor engines promise a new era of fuel efficiency but their
inherently high noise has ruled them out – until now,” Flight International, 25 Feb.-3 Mar, 2014.
47. U.S. Code of Federal Regulations, Title 14, Chap. I, Part 25, “Airworthiness Standards: Transport
Category Airplanes.”

NASA/TM-20220015470 42
48. U.S. Department of Transportation: “Minimizing the Hazards from Propeller Blade and Hub
Failures,” Federal Aviation Administration, Advisory Circular, AC 25.905-1, 2000.
49. U.S. Code of Federal Regulations, Title 14, Chap. I, Part 35, “Airworthiness Standards: Propellers.”
50. Torenbeek, E.: “Synthesis of Subsonic Airplane Design,” ISBN 90-247-2724-3, Delft University
Press, Delft, Netherlands, 1982 (reprinted 1996).
51. McCullers, L.A.; and Lynch, R.W.: “Dynamic Characteristics of Advanced Filamentary Composite
Structures, Volume II - Aeroelastic Synthesis Procedure Development,” AFFDL-TR-73-111, 1974.
52. McCullers, L.A.: “Flight Optimization System User’s Manual, Version 8.29,” maintained and
released by NASA.
53. National Aeronautics and Space Administration: “NASA Spinoff: Deicing System Protects General
Aviation Aircraft,” NASA/NP-2007-10-484-HQ, 2007.
54. Roland, H.: “Advance Design Weight Analysis and Systems and Equipment Weight Prediction,”
Society of Allied Weight Engineers paper no. 790, 28th Annual Conference, San Francisco,
California, May 5-8, 1969.
55. Hinetics, LLC.: “Integrated High Frequency Electric Propulsor for Turbo-Electric Aircraft,” SBIR
18-1-A1.07-8322, contract 80NSSC18P1898, 2019 (URL:
https://www.sbir.gov/sbirsearch/detail/1559639, accessed July 2022).
56. Zhang, X.; Bowman, C.; O’Connell, T.; and Haran, K.: “Large Electric Machines for Aircraft
Electric Propulsion,” IET Electr. Power Appl., vol. 12, no. 6, 2018, pp. 767-779.
57. U.S. Code of Federal Regulations, Title 14, Chap. I, Part 135, “Operating requirements: Commuter
and on-demand operations and rules governing persons on board such aircraft.”
58. Roskam, J.: “Airplane Design, Part VI: Preliminary Calculation of Aerodynamic, Thrust and Power
Characteristics,” Roskam Aviation and Engineering Corporation, Kansas, 1990.
59. Hunter, J.: “Jane’s Aircraft Upgrades, 2009-2010”, Jane's Information Group, 2010.
60. Sadraey, M.: “Aircraft Design: A Systems Engineering Approach (Chap. 5, Wing Design),” Print
ISBN 9781119953401, John Wiley & Sons, Ltd., 2012.
61. Drela, M.: “XFOIL: An Analysis and Design System for Low Reynolds Number Airfoils,” In Low
Reynolds Number Aerodynamics, Lecture Notes in Engineering, vol. 54, Springer, Berlin,
Heidelberg, 1989.
62. Joslin, R.: “Overview of Laminar Flow Control,” NASA/TP-1998-208705, 1998.
63. Drela, M.; and Youngren, H: “XROTOR website,” (URL:
https://web.mit.edu/drela/Public/web/xrotor/, accessed Aug. 2022).
64. Betz, A. (with an appendix by Ludwig Prandtl): “Schraubenpropeller mit geringstem Energieverlust
(Screw Propellers with Minimum Energy Loss), Goettingen Reports, 1919.
65. Wainauski, H.; and Simsbury, C.: “Airfoil Blade,” United States Patent no. 4,519,746, United
Technologies Corp., 1985.
66. Black, D.; Magliozzi, B.; and Rohrbach, C.: “Small Transport Aircraft Technology Propeller Study,”
NASA CR-168 (045), 1983.
67. Chapman, J.; Schnulo, S.; and Nitzsche, M.: “Development of a Thermal Management System for
Electrified Aircraft,” AIAA paper 2020-0545, 2020.
68. National Academy of Science, Engineering, and Medicine, “Commercial Aircraft Propulsion and
Energy Systems Research: Reducing Global Carbon Emissions,” Committee on Propulsion and
Energy Systems to Reduce Commercial Aviation Carbon Emissions, Aeronautics and Space
Engineering Board Division on Engineering and Physical Sciences, The National Academies Press,
Washington D.C., 2016.

NASA/TM-20220015470 43
69. Löbberding, H.; Wessel, S.; Offermanns, C.; Kehrer, M.; Rother, J.; Heimes, H.; Kampker, A.:
“From Cell to Battery System in BEVs: Analysis of System Packing Efficiency and Cell Types,”
MDPI World Electr. Veh. J., vol. 11, no. 4, 2022.
70. Schnulo, S.; Hall, D.; Chin, J.; and Smith, A.: “Further Development of the NASA X-57 Maxwell
Mission Planning Tool for Mods II, III, and IV,” AIAA paper 2019-4491, 2019.
71. Gray, J., Moore, K., Hearn, T., and Naylor, B., “Standard Platform for Benchmarking
Multidisciplinary Design Analysis and Optimization Architectures,” AIAA Journal, vol. 51, no. 10,
Oct. 2013, pp. 2380–2394; also AIAA Paper 2012-1586, April 2012.
72. Gray, J.; Hwang, J.; Martins, J.; Moore, K.; and Naylor, B.: “OpenMDAO: an open-source
framework for multidisciplinary design, analysis, and optimization,” Struct. Multidisc. Optim.,
vol. 59, 2019, pp. 1075-1104 (doi 10.1007/s00158-019-02211-z).
73. Jansen, P.W.; and Perez, R.E.: “Constrained Structural Design Optimization via a Parallel
Augmented Lagrangian Particle Swarm Optimization Approach,” Comput. Struct., vol. 89, no. 13,
2011, pp. 1352-1366.
74. Kraft, D.: “A Software Package for Sequential Quadratic Programming,” Tech. Rep. DFVLR-FB 88-
28, DLR German Aerospace Center, Institute for Flight Mechanics, Koln, Germany, 1988.
75. Conlon, B.; et al., “The Next Generation ‘Voltec’ Extended Range EV Propulsion System,” SAE Int.
J. Alt. Power, vol. 4, no. 2, pp. 248-259, 2015.
76. Xiong, J.; Fugate, J.; and Nguyen, N.: “Investigation of Truss-Braced Wing Aircraft Transonic
Wing-Strut Interference Effects Using FUN3D,” AIAA Paper 2019-3026, 2019.
77. June, J.; Thomas, R.; and Guo, Y.: “System Noise Technology Roadmaps for a Transonic Truss-
Braced Wing and Peer Conventional Configuration,” AIAA Paper 2022-3049, 2022.
78. Lambe, A.; and Martins, J.: “Extensions to the design structure matrix for the description of
multidisciplinary design, analysis, and optimization processes,” Struct. Multidisc. Optim., vol. 46,
2012, pp. 273-284 (doi 10.1007/s00158-012-0763-y).
79. Borst, H., “Summary of Propeller Design Procedures and Data, Volume 1: Aerodynamic Design and
Installation,” U.S. Army Air Mobility Research and Development Laboratory, accession no. AD-
774-831, 1973.
80. Gur, O.; and Rosen, A.: “Optimization of Propeller Based Propulsion System,” J. of Aircraft, vol. 46,
no. 1, January–February 2009.
81. Karabelli, D.; and Birke, K.P.: “Feasible Energy Density Pushes of Li-Metal vs. Li-Ion Cells,” MDPI
Appl. Sci., vol. 11, no. 16, 2021.
82. Liu, Y.; Liu, S.; Li, G.; and Gao, X.: “Strategy of Enhancing the Volumetric Energy Density for
Lithium–Sulfur Batteries,” Adv. Materials (Weinheim), vol. 33, no. 8, p. 2003955, 2021.
83. Xue, W.; Miao, L.; Qie, L.; Wang, C.; Li, S.; Wang, J.; and Li, J.: “Gravimetric and Volumetric
Energy Densities of Lithium-Sulfur Batteries,” Current Opinion in Electrochemistry, vol. 6, no. 1,
pp. 92-99, 2017.
84. McCloskey B D: “Attainable Gravimetric and Volumetric Energy Density of Li-S and Li Ion Battery
Cells with Solid Separator-Protected Li Metal Anodes,” J. Phys. Chem. Lett., vol. 6: pp. 4581-4588,
2015.
85. Hagen, M.; Hanselmann, D.; Ahlbrecht, K.; Maça, R.; Gerber, D.’ and Tübke, J.: “Lithium–Sulfur
Cells: The Gap between the State-of-the-Art and the Requirements for High Energy Battery Cells,”
Adv. Energy Mater., vol. 5, no. 16, p. 1401986, 2015.
86. Hubbard, H.: “Sound from Dual-Rotating and Multiple Single-Rotating Propellers,” NACA TN
1654, 1948.

NASA/TM-20220015470 44
87. Block, P.: “Noise Radiation Patterns of Counter-Rotation and Unsteadily Loaded Single-Rotation
Propellers,” J. Aircraft, vol. 22, no. 9, pp. 776-783, 2012.
88. Tam, C.; Salikuddin, M.; and Hanson, D.: “Acoustic Interference of Counter-Rotation Propellers,
J. Sound & Vibr., vol. 124, no. 2, pp. 357-366, 1988.
89. Smith, D.; Filippone, A.; and Bojdo, N.: “Noise reduction of a Counter Rotating Open Rotor through
a locked blade row,” Aerospace Science and Technology, vol. 98, 2020.
90. Guynn, M.; Berton, J.; Haller, W.; Hendricks, E.; and Tong, M.: “Performance and Environmental
Assessment of an Advanced Aircraft with Open Rotor Propulsion,” NASA TM 2012-217772, 2012.
91. Strack, W.; Knip, G.; Weisbrich, A.; Godston, J.; and Bradley, E.: “Technology and Benefits of
Aircraft Counter Rotation Propellers,” NASA TM 82983, 1982.
92. Pankhurst, R. C.; Conn, J. F. C.; Fowler, R. G.; and Love, E. M.: “The Effect of Variation of Gear
Ratio on the Performance of a Variable-Pitch Airscrew for a High-Speed Aeroplane,” Ministry of
Aircraft Production, Aeronautical Research Committee R&M 2039, London, Oct. 1941.
93. Berton, J.; and Nark, D.: “Low-Noise Operating Mode for Propeller-Driven Electric Airplanes,”
J. of Aircraft, vol. 56, no. 4, pp. 1708-1714. 2019.
94. Gur, O.; and Rosen, A.: “Design of a Quiet Propeller for an Electric Mini Unmanned Air Vehicle,”
J. of Propulsion and Power, vol. 25, no. 3, pp. 717-728, May-June 2009.
95. Francillon, René J.: “Lockheed Aircraft Since 1913,” London, Putnam & Co., 1982, ISBN 0-370-
30329-6.
96. “Environmental Technical Manual, Vol. I, Procedures for the Noise Certification of Aircraft,”
International Civil Aviation Organization (ICAO), Committee on Aviation Environmental
Protection, 2nd ed., Document 9501, 2015, pp. 3.47-3.48.
97. McDonald, R.: “Modeling of Electric Motor Driven Variable Pitch Propellers for Conceptual
Aircraft Design,” AIAA Paper 2016-1025, Jan. 2016.
98. Clarke, J.: “NASA Advisory Council Meeting - Aeronautics Report,” National Advisory Council
Committee report, Aug. 2022 [URL:
https://www.nasa.gov/sites/default/files/atoms/files/nac_aeronautics_committee
_report_august_2022.pdf, accessed Sept. 2022].

NASA/TM-20220015470 45