Aircraft Structural Components

Download as pptx, pdf, or txt
Download as pptx, pdf, or txt
You are on page 1of 32

AIRCRAFT STRUCTURAL

COMPONENTS
Livil Lyle
PR15AE1004

EARLIEST
AIRCRAFT

Wooden Biplanes

The earliest aircraft of all, of course, is the Wright Flyer


(1903)

Its main structural characteristics include:

Biplane main wing, canard, rudder and fore/aft fuselage all


use similar structural arrangement of rectangular trussed
sections comprising:

Solid vertical strut members.

Wire bracing stretched from corner to corner.

Wing section formed by thin ribs in flight direction and skin


covering.

Figure 2.1 Structural layout and configuration of Wright Fly

The basic philosophy behind this structural design, and of


many similar aircraft built just afterwards, was:

The use of biplane main wings led to a lighter & stronger


structure than would result from using a single wing structure
(monoplane), as would have been suggested by observations of bird
flight.

The rectangular frame shapes were held in place by two light


diagonal wires rather than one single heavy diagonal member.

Wire bracing was used to support all the tensile loads.


The compression loads were concentrated into a small number of
compact rods.

The materials used were the cheapest available at


that time and also the most easily worked. This
included:

Wood for the vast majority of the structure - firstly


bamboo then spruce.

Piano wire for bracing.

Cotton muslin fabric for skin covering.

The major structural advantages of the layout are high


rigidity and consequently good bending and torsional
resistance.

This layout is suitable for aircraft with low wing loading


(W/S) requirements (e.g. aerobatics types with good lowspeed manoeuvrability needs).

The layout results in a very light aircraft, typically only about


10% of the weight of an equivalent sized modern metalskinned aircraft!

The majority
configuration

of

WWI

aircraft

were

based

upon

this

Figure 2.2 Biplane fighter aircraft of WWI Breguet 14 & Sopwith Camel

Wooden Monoplanes

Bleriot XI (1909)

This new concept originated in Europe and


included the use of a dedicated trussed
fuselage and monoplane wing.

The concept was largely ignored by the


Britain aircraft manufacturing community
due to conservatism and concerns over
strength & durability. Some of the most
prominent early aircraft to use this design
approach were those of Louis Bleriot, the
first to successfully undertake an
aerial English Channel crossing

The wing structure comprised two thin wooden


spars with external wire bracing to independently
support lifting & landing tensile loads

The fuselage structure incorporated a standard civil


engineering Pratt truss type of design, comprising 4
main longerons running down its length with
additional support from spacers, struts & cross-wire
bracing.

Bleriot aircraft wing structural details

Bleriot aircraft fuselage structural details

By the end of WWI (1918), certain aircraft structural design


features were virtually standard. To improve lift, wings were
now manufactured with much thicker sections and had
much less lower surface curvature. This allowed for the use
of deeper spars, thus improving wing second moment of
area values and thus the overall wing strength and stiffness.

This consequently removed the need for external wire


bracing. Examples of such aircraft include the Fokker D.VII
(1918) and Fokker DR-1 (1917) (Figure 2.7).

Sectional depth effects on structural deflection

Fokker D.VII

Fokker DR-1

The Use of Metal

There was a fairly gradual change-over from the use of wood to the use
of metal from WWI onwards, though the Junkers J-1 monoplane (1910)
was built entirely from metal (steel tubing & thin sheet iron
coverings) while the Fokker DR-1 triplane (1917) also used steel tubing
for fuselage truss members.

Junkers J-1
Fokker DR-1

A major reason for the greater use and adoption of metal for
the airframe was due to the short supply of spruce
directly after WWI.

It was a fairly gradual changeover, however, with some


manufacturers reluctant to move away from their
wood-working facilities and experience.

For instance, a typical aircraft of the 1930s, the Hawker


Fury (1931) only used metal in the form of steel tubes for
the main fuselage members, along with the usual cross-wire
bracing. Wood was still used extensively elsewhere.

Hawker Fury

The change to all or mostly all metal construction was


almost complete by the time of Hawker Hurricane (1935),
and certainly prior to start of WWII in 1939. Some still used
fabric coverings for the wing and/or fuselage, however, such
as the Hurricane and Vickers Wellington. The most notable
all-wooden aircraft by the time of WWII was the highly
successful Mosquito fighter aircraft.
Hawker Hurricane (1935)

Stressed-Skin Construction

A major universal breakthrough in aircraft structural design occurred in the 1930s, with
the advent of stressed-skin or semi-monocoque construction methods. This arose
because of the major problems caused by the standard internal cross-bracing in the
fuselage as aircraft developed requirements to carry passengers & payload internally.
Designers soon recognized that the designs applicable to flying boat fuselage construction
were also appropriate for transport aircraft use.

The main structural advantage is that the skin is then an integral load-carrying working
part of the overall structure.

The wording can be broken down into:

mono - one piece

coque - skin or egg shell

semi- - makes use of additional internal stiffening framework.

Sikorsky S-42 Clipper (1934)

Soon after the fuselages came stressed-skin wings. This was mainly driven by
the US designers, who were dominant in building large capacity transports &
bombers at that time, e.g. the B-9 and B-10 bombers; the Boeing 247, DC-1,
DC-2 and DC-3 airliners.

By this stage, the standard airframe was of aluminium alloy construction with
a structural load carrying skin riveted to frames, longerons (for fuselage) or
spars (for wing), ribs & stringers.

The European designers and manufacturers lagged behind their US


counterparts by about 5 years, mainly because they were working on smaller
aircraft and it was more difficult to scale down the stressed-skin technology.

Messerschmitt (Germany) and Supermarine (UK) were the only Europeans


making stressed-skin wings by the mid 1930s.

Douglas DC-1 (1933)

Being B-9 (1931)

Standard airframe construction by mid 1930s

Major
Aerostructures
Developments since
WWII

Douglas DC-6

Pressurized Fuselages

The use of cabin pressurization in the mid 1940s to improve


the flying environment for passengers and crew led to
considerably increased fuselage strength requirements.

Some of the earliest aircraft to use pressurization were the


Lockheed Constellation, Douglas DC-6 and Boeing 377
Stratocruiser, all of which cruised at around 20,000 ft with
cabin pressures equivalent to that experienced at 8,000 ft
altitude.

This directly led to the adoption of circular sections as the


norm, minimization of structural cut-outs & careful
consideration and alleviation of stress concentrations (by using
rounded fuselage door corners, etc.).

Lockheed Constellation

Aeroelasticity Problems

Rapid advances in aerodynamics & propulsion technology during


the 1940s led to higher speeds, more lift and increased wing
loadings. As a direct result, several previously unknown
aeroelastic phenomena became apparent for fighters during
WWII, including:

Wing and control surface flutter

Divergence

Control surface (especially aileron) reversal.

All of these problems forced the need for increased wing


torsional (twisting) stiffness. A particular problem for WWII
fighters was the lack of roll response in high speed dives
several aircraft (e.g. the Spitfire, P-51 Mustang, etc.) reduced the
wing span in order to help alleviate the problem.

P-51 Mustang (1942)

Manufacturing Techniques
Integral Construction

Increased aircraft speeds, allied with the advent of the turbojet propulsion era and
design against metal fatigue, led to the need for thicker wing skins. These were
consequently more difficult to wrap around the internal structural framework. By the
1950s, the skins were usually pre-formed by rolling or, alternatively, by stretchforming. By the late 50s integral construction was commonly being used to
form the skin and stringers as one. These would be machined from a solid slab
of aluminium alloy with perhaps 95% of original material removed.

Advantages include:

Less riveted fasteners so less wing drag.

Fewer stress concentrations so reduced fatigue issues.

Easier optimal tapering of skin thickness with weight benefits.

Easier to seal fuel tanks.

Improved fuel tank volume.

Metal Fatigue Problems

The potentially catastrophic effects of metal fatigue were


highlighted for the aerospace community in 1954 when 2
DeHavilland Comet jet airliners disintegrated in flight
within 3 months of each other.

This can be mainly attributed to the nonconsideration of low-cycle pressurisation &


depressurisation and the resultant effects on
weakening the structural strength (i.e. metal
fatigue).

The resultant cracks first appeared in areas of high


stress concentrations, in these cases at the corners of
the non-rounded windows, and soon propagated with
disastrous consequences.

DeHavilland Comet (1954)

Subsequent aircraft fuselages were better designed to withstand


such problems, through the use of:

Increased skin thicknesses,

Rounded windows,

Multiple load paths (fail-safe design),

The use of crack-resistant copper-rich aluminium alloys,

The increased use of welding, bonding, chemical etching

Integrally machined panels.

High Speed
Problems
A USAF SR-71 high-speed reconnaissance aircraft

Swept Wings

After WWII, major wing structural design changes were caused and
forced by adopting Busemanns wing sweep theory for reducing highspeed wing drag.

Saab J-29 Tunnan (1948)

F-86 Sabre (1947)

MiG-15 Fagot (1949)

Area Rule Technology

Major fuselage structure changes were forced by the


adoption of Whitcombs area ruling theory. This involved
the incorporation of a gradual change in overall sectional
area, in order to reduce the high speed wave drag
component. In particular, this requires the reduction in
fuselage area where the wing and tail is attached and
the narrowing down elsewhere, thereby producing an 14
ideal Sears-Haack body area distribution. This clearly
complicated the fuselage structural design and layout
processes.

F-102 A Delta Dagger (1956)

Refer to http://en.wikipedia.org/wiki/Whitcomb_area_rule
for more details.

B-58 Hustler (1959)

Thin Wings

Much thinner wings were needed as aircraft speeds continued to increase during the
1950s and 60s, due to aerodynamic considerations (in order to reduce the magnitude of
the wave drag due to thickness component). An example includes the Lockheed F-104
Starfighter (1956) with a cruise speed in excess of Mach 2 and a wing t/c ratio of only
3.4%. The adoption of such thin wing sections led to the common use of multi-cell &
multi-spar wing structures:

Lockheed F-104 Starfighter (1956)

A-5 Vigilante (1958)

Kinetic Heating

Ever-increasing aircraft speeds during the 1950s led to additional


previously unencountered problems in the form of large
temperature gradients due to kinetic (aerodynamic) heating.
These effects can easily lead to significant thermal stresses and
expansion/contraction
problems
and
are
big
design
considerations for supersonic aircraft, e.g. Concorde & B-58
Hustler, where steadystate temperatures of up to 128oC may
result for Mach 2 cruise.

Concorde (1976)

Design solutions include careful material selection (the increased


use of high-temperature resistant steel, titanium, etc.) and the
incorporation of cut-outs to accommodate thermal expansion
limits.
B-58 Hustler (1959)

The design solution adopted by the structural and


materials design engineers incorporated:

Titanium alloys used for the forward fuselage skin & frames and
high-strength H-11 steel used for the stiffeners.

Brazed stainless steel honeycomb sandwich panels used for the


intermediate fuselage & wing.

Titanium wing spars with sin-wave webs.

H-11 steel & titanium alloy skin for aft fuselage.

Welded joints.

Valkyrie XB-70 (1964)

Computing Advances

Computers were first used in earnest on aircraft during XB-70 Valkyrie structural
development programme. The standard structural hand calculations were heavily
supported by extensive use of matrix structural analysis methods on computers.
All of the major components were analyzed using the force method of linear
elasticity with three basic elements (rod, shear panel & built-up beam) used for
the modeling. The computer codes used by the engineers were all written in the
assembly language and were run on an IBM 7095 mainframe computer.

Finite element analysis was used first of all during the structural support and
development phases of the Boeing 747 airliner design programme in the mid
1960s. The improved analysis capabilities (all later verified through testing) led
directly to the evolution of several new design concepts. This included the
aeroelastic tailoring of the nacelle & wing/body intersection in order to reduce
potential flutter problems and also allowed for the major extensive use of
composite materials.

Advanced
Materials
Technology

Sandwich Structures

This is an ideal way of providing thin sheets with an


improved compressive buckling stability. The
Mosquito (Figure 3.10) was first to use a form of
sandwich construction; its fuselage comprised
plywood skins and a balsa core.

The most common form of sandwich material uses a


metal (usually aluminium alloy) core of honeycomb
cells with thin aluminium alloy facing skins (Figure
3.11). This type of material has been in common use
since the 1960s for many thin, secondary aircraft
sub-structures (e.g. flaps, spoilers, etc.). It has also
been used extensively in spacecraft applications.

DeHavilland Mosquito

Honeycomb sandwich panel construction

Some of the advantages associated with the use of such


honeycomb sandwich materials include:

Excellent resistance to buckling & sonic fatigue.

High values of specific strength & stiffness.

Disadvantages include:

Costly, complicated & lengthy manufacturing processes.

The thin, fragile facing skins are susceptible to damage.

Field repair is a difficult and specialized task, making it unsuitable


for many military applications.

Attachment of hardware to the panels difficult, necessitating the


use of special inserts.

Composites

Composites are simply combinations of two or more different materials so have been in general use for
many years in differing forms:

The DeHavilland Mosquito fuselage used spruce fairings on balsa wood core, as described earlier.

Fibreglass/polyester composites were used for the radomes on many WWII aircraft.

Fibreglass was used for the facing skins on aluminium honeycomb sandwich panels for the Boeing 747 control
surfaces (1969).

The modern conception of a composite material, however, is that of Carbon-fibre Reinforced Plastics
(CFRP), which is becoming increasingly dominant and of widespread use in the aerospace field, e.g. CFRP
makes up:

26% of the total weight of the AV-8B Harrier (1981) including the wing, forward fuselage, horizontal tail.

72% of the total weight of a Beech Starship (1986).

12% of the total weight of a Boeing 777 (1995).

50% of the total weight of a Boeing 787 (2008).

Beech Starship & Boeing 787

Other Advanced Materials

Many other materials have been


suggested, developed and
sometimes tried on aircraft
structures, including:

Aluminium/lithium alloys.

Aramid/aluminium laminates.

Metal matrix composites.

Thermoplastics, etc.

Thank You!

You might also like