Int. J. Engng Ed. Vol. 13, No. 3, p. 227±236, 1997 0949-149X/91 $3.00+0.

00
Printed in Great Britain. # 1997 TEMPUS Publications.

Graduate Aircraft Design Education*

J. P. FIELDING
College of Aeronautics, Cranfield University, Cranfield, Bedfordshire, England, MK43 0AL, UK
Cranfield University believes that the best way to learn about design is to do it, and that group
design projects are a very powerful means of providing practical experience. This is not unique, but
what is unique is the practical detail achieved in the MSc course. Cranfield's approach is to invest
at least 3 man-months work by staff in the preparation of each project, before it is presented to the
students. This defines the aircraft's shape, aerodynamics and mass, and is equivalent to work done
by Project Officers in industry. Students who wish to perform their own conceptual designs may
do so for their individual research theses which are the other main activities of the MSc course.
Some 25 students are allocated the responsibility for the design of a major part of the aircraft.
These responsibilities take the form of a major structural component, a flying control surface or a
mechanical system such as fuel, or the flying control system. Reliability, maintainability,
performance and cost are overall design topics also studied by students. This paper describes the
design of an entry-level executive jet, as an example of the group design project. The paper also
describes the other elements of graduate education at Masters and Doctoral level.

INTRODUCTION University Charter in 1969, under the name of

Cranfield Institute of Technology. The name was
AIRCRAFT design is a synthesis of many disci- again changed in 1993, to Cranfield University, to
plines, each of which must be understood and counter some misunderstanding in the UK, as to
correctly applied to achieve the correct balance what was an Institute of Technology. The College
essential for optimum performance. It always of Aeronautics remains in existence as one of the
involves compromises between the requirements major schools on the Cranfield Campus.
of the different disciplines involved, and the The title of the Aircraft Design course was
perspectives of the specialists concerned. changed to become `Aerospace Vehicle Design', to
Figure 1 shows sketches of aircraft designs reflect an expansion into the field of spacecraft
that might be produced by engineers from dif- design. The basic tenets of the aircraft design edu-
ferent disciplines. A competent aircraft designer cation process, however, remain those described
must know enough about those specialisations so by the Author's predecessor, Professor D. Howe
that he or she will be able to balance them to [1], namely:
arrive at a whole aircraft optimum design, rather 1. The properly equipped designer must be able to
than one that may be optimum from, say, a achieve a correct balance between synthesis and
structural or aerodynamic viewpoint. The judge- analysis and so achieve an optimum result in the
ment required to achieve this balance requires a most economic manner.
practical design approach that has been the 2. He/she must be aware of the importance of
hallmark of aircraft design teaching at Cran- working through the task to the final details.
field since the Aircraft Design course was estab- The need for a balanced outlook and almost
lished as one of the original Cranfield graduate intuitive approach to a problem has often given rise
courses. to the assertion that good designers are born, not
This occurred in 1946, when the College of made. There may be some truth in this but extensive
training is essential in the context of a complex
Aeronautics was founded on a site some 50 miles technology, and correct teaching can exploit the
north of London. The original College had it's own latent potential of design ability.
well-equipped airfield and fleet of research and A student must have a broad and deep under-
teaching aircraft. These facilities formed the bed- standing of both theory and practice before he/she is
rock of Cranfield's practical aeronautical activities capable of tackling design work effectively. The con-
and have been enhanced by the acquisition of tinuously expanding frontiers of aeronautics imply
progressively more modern aircraft and other that text books rapidly become outdated, even if
facilities. Cranfield's original objective was to they exist at all. Therefore the staff must find
provide a world-class school of post-graduate ways of keeping up to date and at the same time not
aeronautical teaching and research. The College lose sight of fundamentals and real practice. George
expanded into many other areas of engineering, Bernard Shaw once wrote that `those who can, do, but
those who can't, teach'. To instruct in design effec-
science and management studies and received it's tively one must both `do and teach' and ideally the
student should `do' as well! The course structure and
* Accepted 2 April 1997 environment at Cranfield have been built up to enable
Copyright 1997 AIAA and ICAS; reprinted with permission this to be achieved.

227
228 J. P. Fielding

Fig. 1. Aircraft that could be produced by specialists with differing design priorities.

This philosophy has been proved to be sound, COURSE STRUCTURE

and has not changed since 1946, although the
means of achieving those aims use modern tools, Figure 2 shows a summary of the Cranfield
as will be shown later. graduate aircraft design courses, the main one
 Graduate Aircraft Design Education 229

engineering diplomas. Preliminary year students

usually form the core of the design team in the
MSc year.
An MSc course option in Structural Design
also exists. This differs from the General Design
option in that students follow lectures and per-
form research most relevant to that subject.
Rather than participating in the group design
project, a more extensive individual research
topic is performed and examinations are taken
on the lecture material. A 3-year part-time version
of the aerospace vehicle design course was intro-
duced in February 1995.
An increasing number of students are following
the PhD course. Some of these are recruited after
completing the MSc course, but the majority come
directly from other Universities, or from Govern-
ment or industrial establishments. Most of the
PhD students are members of the Conceptual
Design Research group and activities include:
 The development of multivariate design synth-
esis and optimisation methods for canard delta,
agile fighter, A/STOVL, UMA, Supersonic
Transport and laminar-flow aircraft.
 Multidisciplinery preliminary design method-
ologies for conventional transport and blended
wing/body aircraft.
 Investigation of configurational aspects of
advanced airframe systems.
 The development of methods to improve the
reliability, maintainability and survivability of
civil and combat aircraft.
Other PhD students are studying topics in the
structural design areas, particularly in the use of
composite material structures.

LECTURE COURSES AND LABORATORY

WORK

The preliminary year

The preliminary year is intended as an intro-
Fig. 2. Cranfield aircraft design courses. duction or refresher in aeronautics, and is pitched
at the final year undergraduate level. There are
lecture courses in areas such as Maths, Computing,
being the 12-month MSc Course in Aerospace Structural Analysis, Aerodynamics, Aeronautical
Vehicle Design. This is an intensive course and Engineering, Electronics, etc. These are augmented
requires a high input standard in terms of pro- by laboratory work and the most important
spective students' qualifications and experience. featureÐindividual design projects. Students will
The usual entry is a good-class honours degree in complete three progressively more complex design
Aeronautical or Mechanical Engineering, prefer- projects during the year. They will perform con-
ceptual and detail designs, which are then stressed,
ably with a number of years of post-graduate
and reports produced. The first project might be a
experience. Indeed, our average students' age is relatively simple mechanism, the second a fuselage
in the late 20's. Many students have degrees in frame and the third a complete flying control
such subjects as Physics, Maths or Computing surface. Students are taught, and use, computer-
and wish to convert to an aeronautical engineer- aided design and computer structural analysis
ing course. These students can attend the pre- tools.
liminary year course (PY) and then progress
onto the 12-month MSc course. The preliminary The MSc year
year course is also used by some lower-qualified The lecture programme is carried out over two,
aeronautical graduates, or students with relevant ten-week terms, in parallel with the group design
230 J. P. Fielding

project (see below). The lecture syllabus has a large engineers; they do not actually fly the aircraft.
mandatory core, with some options, depending This deficiency has been remedied by student
on student interests. The total lecture hours vary light aircraft flight test experiments in the
between 240 and 300 hours in the following College's two-seater Beagle Pup aircraft. Each
subjects: student undertakes two flights in the aircraft. The
Cranfield pilot demonstrates and tutors the
General and project design: student in level flight, climbs, turns and descents.
Design for operationÐincluding noise, V/STOL, The student then flies the aircraft in the second
airports, reliability, maintainability and weapon flight and conducts a simple flight test experi-
systems; ment, associated either with performance or flight
Initial aircraft design; dynamics.
Design of major components; These flying activities are expensive, but are part
Computer-aided design; of the course and contribute significantly to the
Loading actions. development of a well-rounded design engineer.
Structural aspects:
Aerospace structural considerations;
Structural stability; INDIVIDUAL RESEARCH INVESTIGATIONS
Finite-element methods;
Structural optimisation; These may be theoretical and/or experimental
Fibre reinforced plastics; and are drawn from a range of topics related to
Fatigue and fracture; the course and suggestions by the staff, sponsors
Structural dynamics; or students themselves. Members of staff are
Aeroelasticity. appointed as research supervisors for each student
within a few days of the start of the academic year.
System and allied areas: There is a close relationship between student and
Aircraft systems; supervisor, reflected by the average student/staff
Control engineering; ratio of around 9:1.
Aircraft avionics systems. The experimental research is aided by the air-
Additional subjects: craft, some 20 windtunnels, an extensive struc-
Aircraft accident investigation; tural test laboratory, metalwork workshops and a
Fixed wing aircraft performance; composite component fabrication laboratory.
Theory of flight. There is a powerful computer facility with
networked PCs, workstations and mainframe
The final topic is another unique feature of computers.
Cranfield Courses. Students are given lectures in The research investigations comprise 45% of the
aircraft performance, flight mechanics and flight MSc students' final marks and are assessed by
test methods, and then complete eight flight tests in research theses.
the College's own specially-equipped Jetstream Topic areas are similar to those performed by
aircraft (Fig. 3). the conceptual design research group, mentioned
The theoretical knowledge is thus reinforced by above, but extend to conceptual design, structural
flying in the aircraft during the flight-test manoeu- design, fracture and fatigue, composite structures
vres and is then further reinforced by on-board and advanced airframe systems.
flight data acquisition, and subsequent analysis.
The latter activity has been recently enhanced
by the incorporation of a Cranfield-designed on- THE MSc GROUP DESIGN PROJECT
board computerised data acquisition and display
system. Background
Students, however, only act as flight-test The Cranfield group project is unique by virtue
of the amount of preparatory work done by staff
before work is started by the students. All other
known design projects start with the students being
given the aircraft specification. They then perform
a conceptual design, leaving little time available
for detailed design. With the Cranfield method,
this work is done by the staff, thus enabling the
students to start much further down the design
process. They thus have an opportunity to get to
grips with preliminary and detail design problems,
and become much more employable in the process.
The Cranfield project method also allows students
to use modern design tools such as CAD, finite
elements, laminate analysis and aerodynamic
Fig. 3. Jetstream flight test `classroom' aircraft. modelling. The group design project is undertaken
 Graduate Aircraft Design Education 231

by all the aerospace vehicle students and is a major

feature of the MSc course, accounting for almost
half of the final assessment. Each year the students
work in teams on the design of a project aircraft. A
substantial part of the airframe, a system, an
installation or performance aspect, is allocated to
each student as his or her own responsibility.
The aircraft chosen as the subject for the work
are representative of types of current interest to
industry. They usually incorporate some feature
which extends the bounds of existing practice, as
an applied research activity. This excites the inter-
est, enthusiasm and ingenuity of the students and
forces the staff to keep up to date. Fig. 5. T-91 TrainerÐsome maintainability features.
Civil and military aircraft are investigated in
alternate years, so that the whole of the industry
high-subsonic aircraft such as the Gulfstream IV.
is catered for. Recent examples of design projects
Cessna and Swearingen recognised the need for
included large and small business jets, a number
of medium-sized jet transports and a 500-seat entry-level executive jets by the development of
short-haul airliner. their Citationjet and SJ-30 designs. Their aim was
Figure 4 shows the project programme for the to produce new aircraft with a purchase price
last aircraft, the A-90 which is typical of all similar to that of the turbo-prop Beech King
projects. Military aircraft include basic and Air, but with greatly enhanced speed and com-
advanced trainers, close-air support aircraft, an fort. These designs were made possible by the
advanced tactical fighter, V/STOL supersonic advent of the cost-effective, quiet and fuel-efficient
strike aircraft and military transports. Williams/Rolls FJ44 engine. The Cranfield Design
Figure 5 shows a typical drawing from the T-91 was pitched between the Citationjet and SJ-30 in
trainer maintainability CAD model. The 1986/7 terms of performance, but would utilise significant
project saw the design of a two-stage to orbit space amounts of advanced composite materials in its
launcher, which is shown in Fig. 6 [2]. construction. This should lead to lower mass
The remainder of this article gives a brief despite the more generously sized cabin interior.
description of the E-92 executive jet project, as
an example of a civil aircraft project.
The design specification
Interior layout: There should be provision for 5±6
THE E-92 EXECUTIVE JET PROJECT passengers with comfort standards equivalent to
airline First Class passengers. The aircraft should
Project background be capable of single-pilot operation, but a co-pilot
It was decided to design an executive jet in seat is required. There should be beverage, baggage
1992/3, as such aircraft have an important role and toilet accommodation superior to the SJ-30.
to play in the world market for aircraft. The The fuselage should have a door capable of loading
need for executive aircraft has been satisfied by a spare engine.
designs ranging from piston aircraft, to large Performance: The following figures are based on
ISA, sea-level conditions.

Fig. 4. The A-90 programme timescale. Fig. 6. SL-86 space launcher.

232 J. P. Fielding

 High speed Mach No. ˆ 0.75;

 Economical Mach No. ˆ 0.72;
 NBAA, IFR range with 3 passengers and 1 crew
should be greater than 1800 miles (3245 km);
Max operating altitude ˆ 43,000 ft (13.1 km);
 FAR take-off balanced field length to be less
than 3300 ft (1005 m);
 FAR landing distance at max landing mass to be
less than 2600 ft (792 m).
Cost: The acquisition cost shall be no more than
$3.5m US in 1994.

Group project programme

The design process started with the conceptual
design of the aircraft by members of staff, in early
1992. This work was summarised in Reference 3 Fig. 8. Inputs into the group design project.
which was given to 25 students in October of that
year. Each structures student was given responsi- compromise was agreed between students respon-
bility for the detailed design, stressing and fatigue sible for fuselage structure, rudder pedals, nose
analysis of components such as the forward fuse- landing gear, electrical power, avionics and flight-
lage, outer wing, tail, etc. Some students designed deck layout.
airframe systems such as fuel, flying controls, The knowledge gained during lectures, project
engine installations, etc. More global design tasks meetings and discussions with members of staff
were performed by other students in the areas of was augmented by information from aircraft
flight deck layout, avionics installation, reliability manufacturers. (See Fig. 8 for project inputs).
and maintainability, aerodynamic performance Vital information on the project engine was
and cost estimation. given by Rolls-Royce and realistic information
The project was managed to a demanding eight- was received from avionics systems manufacturers.
month programme by means of weekly project An extremely useful group visit was made to Luton
meetings, where students reported progress, airport, where MAGEC's aircraft maintenance
received advice and instructions for subsequent was examined. This visit was followed by one to
work. The most important function of these meet- British Aerospace, Chester. The highlights were
ings was that of a forum where design conflicts close examinations of the BAe 800 and 1000
were resolved. production lines and those of the Airbus wing
One of the dangers of individual responsibility is assemblies.
that of parochialism. The student designing, say, a The programme ended in May, 1993 with the
portion of fuselage learns a great deal about that, submission of detailed project theses, which con-
to the exclusion of the rest of the aircraft. The tain descriptions of the designed components,
group project meetings go some way to reducing supporting analyses, drawings, CAD plots, and
this problem in that each aspect of the whole finite-element results.
aircraft design is discussed in turn in project meet- The students made a verbal presentation of
ings. There were some very lively discussions about their work to a group of external examiners and
interfaces, particularly in the forward fuselage industrialists.
area. The design was also used by some 20 flight
Figure 7 shows a computer-aided design dynamics students, who successfully simulated
(CAD) model of this very crowded area. A suitable the aircraft's handling characteristics. This activity
presages further integration of teaching activities.
It is hoped that, in the future, students will be
able to `fly' the project design in Cranfield's
flight simulator during the design evolution, so
that handling characteristics will be part of a
closed-loop design process.

DESCRIPTION OF THE FINAL DESIGN

The aircraft was designed using state-of-the-

art materials, the majority of the structure being
made from aluminium alloys, with some composite
components.
Figure 9 shows a shaded image of the computer-
Fig. 7. E-92 forward fuselage. generated surface model of the project. The
 Graduate Aircraft Design Education 233

Fig. 9. E-92 surface computer model.

surface model was generated using EDS Uni- again. The wing structure was designed by two
graphics software. teams, one with a composite, and one with metal
construction.
Wing Finite-element models were made using the
A modest sweep forward combined with NASTRAN system and showed that the modest
advanced laminar flow wing sections enable forward-sweep of the wing did not result in
Mach numbers in the region of 0.75 to be achieved. aeroelastic problems.
The aspect ratio is 8.0 and there is sufficient fuel
Fuselage
tankage in the wing and fuselage at the specified
The cross-section is generous for this class of
payload for a range of 1800 nautical miles with
aircraft, with a recessed aisle to give more head-
reserves. The high aspect ratio improves fuel burn
room to move round the cabin. The baggage/toilet
and airfield performance. Double-slotted Fowler
compartment is behind a privacy bulkhead, above
flaps, moderate wing loading, spoilers and the high
the wing carry-through structure.
aspect ratio give adequate field performance.
The interior is shown in Fig. 10. The environ-
The absence of slats, the forward sweep wing,
mental and flying control system components run
the aerofoil sections and small chord should allow
under the seat armrests and under the floor.
a significant percentage of natural laminar flow.
The toilet compartment is rather restricted for
The particular laminar-flow section used has a
large passengers, and some re-design will be
very high zero-lift pitching moment. This was
aggravated by the initial fuselage shape, giving
significant trim drag, which would negate the
effects of the drag reductions from laminar flow.
The fuselage was re-shaped to limit this effect, but
it is unlikely that Cranfield will use the section

Fig. 10. Fuselage interior. Fig. 11. Forward fuselage finite-element model.
234 J. P. Fielding

necessary. The fuselage structure is of the conven-

tional aluminium semi-monocoque type.
Figure 11 shows an exaggerated deflection
finite-element model of the forward fuselage. The
passenger door and emergency exit door cut-outs
required reinforcements to maintain structural
continuity. The rear pressure bulkhead also acted
as the wing rear-spar pick-up. The area aft of the
bulkhead was the primary equipment bay. The
environmental control system, hydraulics and elec-
trical power systems were designed in considerable
detail. These systems occupied the equipment bay,
together with a rear fuselage fuel tank and space
provision for an optional auxiliary power unit. The
main landing gear retracts under the forward part
of the equipment bay and the engine pylon front
spar passes through it. The baggage compartment
is under the rear part of the equipment bay.

Powerplant
The aircraft uses a pair of rear fuselage-mounted
Williams/Rolls-Royce FJ44 engines. They are
mounted high on the fuselage to minimise wing
interference effects.
The engine nacelles use easily-opened panels to
ease engine maintenance. The engine pylon front
spar passes through the fuselage to limit fuselage
frame bending moments, whilst the lower-loaded
rear spar is broken at the fuselage side, to facilitate
equipment-bay access.

Tail unit
The aircraft utilises a cruciform tail arrange-
ment. This takes the tailplane above the jet efflux
and increases its moment arm, due to the sweep-
back of the fin. This arrangement does not have as
severe `rolling due to sideslip' effect as does the
high T arrangement.
The fin was designed to be constructed of
carbon-fibre composite material. The component
was analysed by using Cranfield's laminate analy-
sis programs and subsequently checked using finite
elements. A simple dynamic finail analysis showed Fig. 12. E-92 general arrangement drawing.
that some redesign would be necessary to improve
dynamic structural stability. The tailplane was Predicted performance
designed in conventional aluminium alloys and The mass targets had been set using an empiri-
utilised a machined centre-box. cal mass estimation program. The correlation
The high speed of the aircraft led to the use of between the targets and predictions is very good,
mechanical assistance to the flight control system. considering the fact that the E-92 is very small,
Setback hinges and either servo or balance tabs are relative to the empirical database used in the
used on the elevator, rudder and ailerons. program.
Figure 13 shows the predicted payload range for
the aircraft. This was produced after considerable
Landing gear analysis, the production of aerodynamic computer
Single wheels are fitted to each main leg, which models and consideration of the effects of intake
retract inboard into the fuselage fairing. Several efficiency, bleed and power off-takes. It shows that
alternate retraction schemes were investigated, the aircraft could meet the range targets at a
making use of the kinematics module of the high-speed cruise Mach number of slightly less
CATIA CAD system. The nose leg uses twin 0.72 and considerably exceed it at Mach 0.58.
wheels and retracts forwards into the fuselage The calculations used pessimistic power off-takes
nose. The layout of the units can be seen in the and neglected the expected drag benefits of natural
general arrangement drawing, Fig. 12. laminar flow.
 Graduate Aircraft Design Education 235

high standard and the concentrated post-graduate

courses add significant value to the graduates'
education and design experience. The group pro-
jects and associated studies provides a realistic
environment in which students learn how to
design practical components, work as teams and
present their results orally, and in written theses.
The theses from a typical project contain some 200
engineering drawings, in total, produced by tradi-
Fig. 13. E-92 predicted payloadÐrange diagram.
tional and CAD methods. Some 30 project theses
are published, each year giving some 4000 pages of
The predicted FAR landing distance was 80 ft description and analysis, in addition to a similar
better than the target of 2,600 ft. number of individual MSc research theses.
The maximum weight take-off balanced field Students are given hands-on experience in
length was predicted to 3,750 ft, which was a computer techniques, such as CAD, finite-element
considerable degradation, relative to the target of analysis, composite materials analysis as well as a
3,300 ft. The target could be reached by improve- wide range of dedicated analysis programs. They
ments to the flap system or a slight increase in have flown as flight test engineers in the College's
engine thrust. Jetstream aircraft, and have themselves flown the
The target acquisition cost of $3.5m US should Beagle Pup aircraft. They have researched up-to-
be achieved on the basis of a production run of 300 date aeronautical technologies such as fibre-optics,
aircraft. The direct operating cost should be $4.45 all-electric aircraft, and advanced materials. These
US per aircraft nautical mile.
activities provide information of use to other
members of the aerospace community.
The students' individual research thesis work, at
E-92 PROJECT CONCLUSIONS
Masters and Doctoral level, provide significant
applied research in aircraft and structural design.
The design program fulfilled its main aim of
providing a powerful means of educating aircraft These are published, and provide new ideas for the
designers. The use of a challenging project was a industry.
means of investigating many of the problems areas The major output, however, is the output of
of executive aircraft and produced some good highly-skilled, rounded design engineers who
detailed design work. reach high positions, world-wide, in the aircraft
The aircraft that was designed showed consider- industry, airlines, academia, Air Force and Govern-
able promise but required further work to confirm ment Regulatory and Research Departments.
the performance predictions, and to evaluate its The lecturing, group project and research
operating costs more fully. activities are very demanding of faculty members'
The use of a modestly swept-forward wing is a time and require very low student/staff ratios. The
viable solution for this class of aircraft in both extensive laboratory, computing, and flying
layout and aerodynamic terms. The configura- activities are also expensive, but the positive
tion placed considerable demands on the ingenuity results of Cranfield's design education speak for
of the main landing-gear designer, but a good themselves.
solution was produced. There are many chief designers who are
Cranfield aircraft design alumni. The Cranfield
Results of Cranfield's design education process approach, started in 1946, continues and is
Student entry to the Cranfield course is of a very continually being improved.

References

1. D. Howe, An established approach to the teaching of aircraft design, AIAA, Paper AIAA-5-1046,
originally presented in 1975, but reproduced in Perspectives in Aerospace Design compiled by
Conrad F. Newberry, American Institute of Aeronautics and Astronautics Inc, 1991. ISBN 1-56347-
010-1.
2. J. P. Fielding, Project designs of alternative versions of the SL-86 2-stage horizontal take-off space
launcher, ICAS-90 Congress, Stockholm, Sweden, September 1990.
3. J. P. Fielding and H. Smith, Entry-level Executive Jet E-92 Project Specification, DAeT 9200,
College of Aeronautics, Cranfield University, July 1993.

Prof. John Fielding had 12 years industrial experience as both an engineering apprentice and
design engineer at Hawker-Siddeley Aviation. In 1975 he joined Cranfield as an ARB
Research Fellow and is now Professor of Aircraft Design and Head of the Air Vehicle
Technology Group in the College of Aeronautics. He is responsible for research and
236 J. P. Fielding

teaching in all aspects of aircraft design. He specialises in research and teaching in aircraft
conceptual design, reliability and maintainability. He has published more than 50 technical
papers at conferences and in journals. He has been visiting Professor at the University of
Texas since 1991, is external examiner at the University of Limerick, and a member of
the aircraft design technical committee of the American Institute of Aeronautics and
Astronautics.