See also:-The Royal Aero Club Year Books (1911-9); Flight (Jan. 1909 to Dec. 1920, the Official Organ of the Royal Aero Club); Captain McCudden, Five Years in the Royal Flying Corps (1918). (R. M. H.)

II. DEVELOPMENT OF AEROPLANE DESIGN Design of Lifting Surfaces.-The determination of the forces acting upon a body moving through a viscous fluid, such as the atmosphere, is a problem so far not amenable to mathematical solution, and design must therefore be based upon experiment. A vast mass of experimental data has been obtained by testing models in wind tunnels (by Eiffel in Paris, by Prandtl at Göttingen, at the National Physical and other laboratories) and by experiments upon aeroplanes in flight, principally in England at the Royal Aircraft Establishment, Farnborough. A very useful amount of information had been acquired before the war, but this has been greatly extended during the war period.

Lifting-surfaces of various shapes have been used in the design of aeroplanes, disposed in a variety of ways. It was immediately evident that the span or spread of the wing across the line of flight should be large in comparison with the "chord" or dimension along the flight path. The ratio of the span to the chord has been termed the " aspect ratio." Aerodynamic efficiency increases with increasing aspect ratio; but it is desirable to limit the aspect ratio for constructional reasons and in order to reduce the room required for housing. The greater aerodynamic efficiency, moreover, becomes neutralized after a point by the head resistance due to the additional external bracing required. A compromise must be made, and the average figure used is in the region of six to one. It was also evident that the wings should be cambered along the line of flight. The early aeroplane wings had approximately the same curvature of upper and lower surfaces. Wind-tunnel experiments, however, showed that the curvature of the under surface had but small influence compared with that of the upper surface, a result which enabled the designer to increase the thickness and internal strength of the wings and reduce external bracing. Extensive wind-tunnel research has been carried out to find the best cross-section shape of wings. Greater lift can be obtained from highly cambered wings, but thinner wings offer less resistance to motion at small angles. An aeroplane should have as large a speed range as possible. While a wing of high lifting-capacity is required to fly slow, small resistance is required for fast flying, that is at fine angles of attack. A greater speed range is obtained by the use of wings of small curvature (about 1 in 15), the same lower limit being attained by the use of a larger area to carry a given weight. Wind-tunnel experiments further determined the extent to which the curvature should be greater towards the leading edge of the wing.

Early writers sometimes stated the requirements of a wing as consisting purely of a high ratio of lift to resistance at some angle of attack. The requirements are in reality more complex. To secure a wide range of speed a high ratio of lift to resistance is required at fine angles (fine in comparison with the angle at which the wing attains its greatest lift at a given speed) and in addition a high value of this ratio is required at the intermediate angle at which the aeroplane climbs. This is not all. For longitudinal stability the travel of the centre of pressure when the angle of attack varies should be small, as this travel on a curved surface produces instability. The wing section best meeting all these requirements is probably the British Royal Aircraft Factory's No. 15, designed early in 1916.

99 66

The term "wing" is commonly used of the half of a liftingsurface on one side of the aeroplane, the whole surface constituting a "plane." Thus a monoplane has one pair of wings. A tandem aeroplane has two or more pairs of wings arranged as the name implies. The terms “biplane," triplane," "quadruplane" denote that two, three, or four planes are superposed. Langley's "aerodrome" is an early example of the tandem aeroplane. This type is inconvenient structurally and aerodynamically very inefficient. The rear plane acts upon air to which a downward trend has been imparted by the plane in front. The reaction upon the rear plane is therefore inclined backward by the angle through which the air has been "downwashed by the leading plane. In multiplane systems in which the planes are placed one above the other, each plane operates in air whose motion is influenced by the others, and the ratio of resistance to lift is less than the ratio which each would experience if acting alone. If, however, the planes are placed at a sufficient distance apart, so that the gap between is roughly equal to the chord of the planes, the mutual interference produces an effect comparable with that due to a reduction in aspect ratio such as is found necessary in the design of a monoplane. Using the same aspect ratio a given area is disposed in a biplane in half the span required in a monoplane. The biplane forms a good structure, the planes forming the flanges of a box girder. In the monoplane the bracing wires make small angles with the planes, with consequent high tension in the wires and high compression in the spars of the wing. In the biplane the wires make obtuser angles with the planes. In reviewing the examples of the two types, it is found that the monoplanes are relatively of heavy wing loading and low aspect ratio. In the triplane the upper and lower planes may form the flanges of the girder, or the structure may consist of two girders superposed. This does not possess the same structural superiority over the biplane, as does the latter over the monoplane. The triplane arrangement provides a means of reducing span by increasing height. An early example of the triplane is that designed and flown by A. V. Roe in 1909. A Sopwith triplane was used by the British army during the war. The type may be suitable to large aeroplanes, in which reduction of the weight of the structure and of bulk is especially needed.

The resistance of a wing must, however, be considered in relation to the resistance of the external bracing attendant upon

The great majority of aeroplanes have been of the monoplane and the biplane types, the latter predominating since 1912. The first aeroplanes to fly were biplanes and by far the larger number of aeroplanes in use to-day are of this type. The monoplane appeared about the opening date of the period under

FIG. 7.-Early Wright Aeroplane. (Propeller Biplane.) (Elevators in Front; Rudder in Rear.)

discussion, and on an aeroplane of this type Blériot crossed the Channel in July 1909. It was more cleanly designed than the biplane of that date and was regarded as the faster type. It was largely used for trick flying, and figured ever more widely in aeronautical exhibitions. At the outset of the war it had still a reputation for speed, but had found a rival in the better de

signed tractor biplanes. During the war the monoplane was more largely used by the French and the Germans than by the British. The names most associated with the monoplane are French: Blériot, Morane, Nieuport. The "Fokker " monoplanes used by the Germans take their name from a Dutch designer probably inspired by the French designs. During the years 1914-8, the biplane was in the ascendant, but the monoplane was afterwards revived in the form of the aeroplane with thick cantilever wings without external bracing. The monoplane appears to be a type convenient in small sizes, but unsuited for the larger aeroplanes.

66

In the tractor, monoplane or biplane, the order of disposition of the component parts is generally from front to rear:-airscrew, engine, crew; and the body is prolonged to carry stabilizing and controlling surfaces at the rear. In the pusher the order is reversed and the controlling surfaces are carried on an open frame ("outriggers ") in front, at the rear, or in both positions. On a "pusher" the field of view forward is superior, and great stress was laid upon this by the British War Office after the military trials in 1912. The necessity of aerial fighting was proved in 1914, and the tractor was found unsuitable owing to the obstruction in the most effective direction for firing. Pushers were therefore ordered for fighting, at first carrying pilot and gunner, and later carrying only one man with a machine-gun fixed in the aeroplane. The situation was completely altered by the device of firing through the airscrew disc. The blades were at first protected by deflector plates, but shortly after mechanism was used to time the fire between them, the invention of Constantinescu, a Rumanian. The aeroplane was directed bodily at the target. The " tractor" then replaced the "pusher" fighting aeroplane; but "propeller" airscrews continued to be

in Front; Rudder in Rear.)

and on large twin-engine aeroplanes.

The "tractor" is the more convenient design, slightly better aerodynamically and reputed safer in a "crash.'"

FIG. 9.-Propeller Biplane of 1914-16.

Weight and Head Resistance. The aeroplane designer is continually interested in the relative importance of weight and head resistance. At the start attention was naturally concentrated upon the production of a light structure. Knowledge of the resistance to motion of bodies of various shapes was meagre and was most probably gauged in the mind of the designer by the frontal area exposed, irrespective of shape. It was not realized that a strut of circular section offers twelve times the resistance of a strut of the best "streamline or "fair" shape of the same frontal area. The light biplane structure of the Wrights and the Farmans contained a network of struts and wires offering a very high resistance. To reduce resistance, exposed parts may be "faired," which involves adding weight; and the number of external parts may be reduced, which again increases the weight of the structure. Wrights and Farmans may be contrasted with the fast monoplanes and biplanes, the latter employing only a single bay of struts on either side, and finally with the unbraced monoplanes of Junker and Fokker.

"Streamline" wires were first designed for the British army dirigible "Beta" in 1912, and fairshaped wires were in 1914 fitted to aeroplanes designed at the Royal Aircraft Factory. They have since become the most usual bracing of British aeroplanes. They offer approximately one-eighth of the resistance of cable of the same tensile strength. Their metallurgy required careful study, and hence in other countries cable has continued to be used, frequently duplicated, the cables lying one behind the other with a wood "fairing" between them. Struts of streamline shape were in use at an earlier date, The bodies of aeroplanes have improved in form, the crew has been protected from wind pressure, and the spokes of wheels have been covered in with fabric.

The drag of a biplane of moderate speed is made up roughly as follows:

These figures show the importance of careful design of all parts. Much of the resistance of the wing-bracing occurs at the joints of wires and struts to the planes, and the resistance of the body is largely due to the necessity of cooling the engine, either by water radiator or by flow of air over the cylinders.

The weight of the complete structure, excluding the power unit, fuel, crew and other load borne, is about one-third of the whole weight of the aeroplane, but varies with the total weight, with the weight carried per unit of area of lifting surface, and with the strength of the structure. The following figures are

Total weight

Area of lifting surface

.

Load borne per unit area
Load factor
Structure weight of % of
total weight.

6

ft.

ΙΟ

4 8 4 8

ft.

ft.

28 %35 %27 %34 % 31 %40%29 % 37 %!

The "load factor" is the number of times the weight of the craft which the wings will support; a measure of the strength. Using one of the light engines now available, the power unit to give a speed of 100 m. an hour will weigh about onequarter of the total, leaving 40 to 45% for fuel, crew and cargo.

Wing Loading and Horse-Power.-The lift of a wing is proportional to its surface, the atmospheric density, the square of the speed and the angle at which it meets the air measured from the angle giving no lift and up to an angle near that known as the "critical angle." At this angle the lift is a maximum (if the other factors be supposed constant) and above it the lift decreases. The wing in passing through this angle is said to be "stalled." Stalling occurs when flying as slowly as possible. After stalling it is no longer possible to increase the lift by depressing the tail of the aeroplane and it is necessary to dive in order to recover flying speed. This has been a frequent cause of accidents when flying too low to have room for a dive. Moreover, the wings when stalled have lost their normal tendency to oppose rotation about the line of flight and now tend to "auto-rotate or act as a windmill. The aeroplane may therefore drop one wing and pass into a steep spiral glide known as a "spinning nose-dive" from which it may be brought to normal flight by the same diving process reducing the angle of attack of the wings. There is no danger in the stall or the spin so long as there is space for the recovery and knowledge of the action required.

100 M.P.H.

50

15 lbs./sq. ft. FIG. 10. Curve showing lowest speed of flight possible with given wing-loading and the usual thin wings.

Wing-loading, the weight borne per unit area of sustaining surface, determines the speed at which the wings become stalled and therefore the slowest alighting speed. With constant loading, as the speed of aeroplanes increases, wings attack the air at ever finer angles, very soon passing the angle of lowest resistance for a given lift. To increase speed it therefore becomes desirable to increase the loading, or in other words to reduce the area of the wings. This reduction has also the merit that it reduces the bulk of the craft, the resistance of external bracing and the weight of the wing structure. To attain the greatest height heavy wing-loading is not required, and the best loading for a high ceiling would to-day be considered a light loading. For fighting, power of rapid manoeuvre is essential. The aeroplane of light loading can be turned in a smaller circle. The total weight is, however, approximately fixed by military considerations, and light loading implies large wing area and consequent greater resistance to angular acceleration, so that the lightly loaded aeroplane cannot so quickly be "banked" to the correct angle for the turn. Given the wing area, the aeroplane having the lighter loading is the more manoeuvrable; given the weight, the heavier loaded aeroplane is at least the equal of the other. Aeroplanes carry a larger area of sustaining

difficulty of bringing the aeroplane to land at high speeds which prevents the increase of loading beyond 10 lb. to the square foot.

In commercial use, economy dictates an increase of loading; safety demands that the aeroplane may alight at speeds and in a space impossible with high loading. Attempts have been made to make the wing area or the wing shape variable in order to reduce the lowest speed of flight, while retaining the other advantages of heavy loading. None has so far been successful.

purpose-the reconnaissance two-seater aeroplane-and the speed is more than half as great again. Aerodynamically there is little difference between the two aeroplanes. As the power of engines grew their weight per horse-power was reduced. To save two pounds in every four on an engine weighing onethird of the whole aeroplane was important.

The largest engines developed were insufficient for the larger aeroplanes, into which two engines were commonly built, and in some cases four or more.

150 M.P.H.

100

50.

FIG. 13.-Large Twin-Engine Aeroplane.

Two separate power units have been regarded as conducive to safety. Experience has so far not confirmed this. It is essential that the power of one engine alone should be sufficient to fly the aeroplane, and the " twin-engine" aeroplanes used during the war were not all provided with so large a total power. Again, the engines were carried on either side of the centre and the line of thrust of each offset by a considerable amount. This introduced difficulties of control, because rudders were unable to balance the offset line of thrust at the low speed at which the aeroplane could be flown level on one engine only, and there was danger in the event of sudden failure of one engine near the ground.

The table gives some particulars of a few typical aeroplanes through the period under review. The figures are approximate:

lb. face power lb. per 1000 m.p.h. sq.ft. 1908 1,000 540 25 1.8 25 1908 1,150 560 40 1909 670 168

sq. ft.

lb.

2.I

25

4

Dunne. Cody

Roe biplane.

1911

750 280 30 2.7

40

35 30-40

40

25

30

40

35

40

40

40

The horse-power and speed given above are uncertain.

1916-7 14,000 1,640 1918-9 4,220 490 400 1918-9 2,290 330 300

8.5 95 125

7 130 145

The Large Aeroplane. For the same aerodynamic performance, the lifting-surface of an aeroplane must be propor

tional to the weight. If aeroplanes of all sizes were constructed of the same materials and geometrically similar in all parts, the weight of the structure would increase with increasing size as the cube of the linear dimensions, that is, as the 3/2 power of the total weight. This does not in fact obtain, because geometric similarity would give greater strength to the larger aeroplane; also the design may be elaborated and materials worked to relatively finer dimensions; and moreover, large aeroplanes are not designed to have the same strength as smaller craft, as they are less sharply manoeuvred. Nevertheless, the weight of the structure is to be expected and is in fact found to become a larger proportion of the total weight as the size increases. It is therefore disadvantageous to increase size indefinitely and there is in fact a best size depending upon the duty to be done.

To carry an indivisible unit of cargo, such as a large bomb, an aeroplane of at least a certain size is required; hence we find size increasing. Sometimes it is preferable to carry a total load in a smaller number of larger aeroplanes, because the weight of the crew becomes less in proportion to the cargo carried, so that every square foot of wing and every unit of engine power of a fleet carries more useful load. Initial outlay and fuel consumption are reduced and there is further an economy of pilots. At some point the larger aeroplane requires a larger crew, and for war the larger "bomber" must carry a number of gunners and offensive armament for defence against more mobile attackers. The optimum size for a commercial service with a sufficient volume of traffic is what would be termed to-day a large aeroplane (say 7,000 lb. at least). The actual size depends to some extent upon the speed of the service, which governs the relative costs of fuel and personnel, and also upon the distances.

The first large aeroplane flown was the Russian Sykorsky in 1913. Large aeroplanes were demanded in 1915 for bombing and were increasingly used during the war. The Handley Page (13,000 lb. gross) was extensively used by the British. The "Gotha" and others were used for raids on London. The same Handley Page aeroplanes and a subsequent design were employed on a passenger service between London and Paris throughout 1919 and 1920. The " Vimy " (12,500 lb. gross) crossed the Atlantic, flew from Cairo to the Cape, and from Europe to Australia, and has been used on a London-Paris commercial service.

46

clear field of view forwards. The early biplanes with an elevator in front and rudder at the rear disappeared about 1914; the monoplanes conformed to the modern usage. Both elevators and rudders are usually hinged portions of fixed surfaces, but in some cases the entire surface has been movable and constituted the elevator or rudder. The latter arrangement has not provided stability if the controls were abandoned. Later the fixed horizontal surface was made adjustable trimming tail plane," by the pilot during flight and known as a a device much used by the British from 1916 onwards. It enabled the flier to vary the speed of flight at which no pressure upon the controlling lever was required, and effectively increased the range of control resulting from the application of a definite force. became standardized in 1915-6, and consists of a The arrangement of control levers or wheels, at first very diverse, "rudder bar operated by the feet and a hand lever whose fore-and-aft movement operates the elevators and whose lateral movement provides latera control. The rudder bar and the lever are moved in the direction in which it is desired to move the aeroplane. In larger aeroplanes rotation of a wheel mounted on the fore-and-aft lever actuates the ailerons, the fore-and-aft control remaining as before. The lever or wheel is generally connected to the control surfaces by steel cables, although shafts in torsion and tension or compression members have also been used.

Balanced control surfaces, although in use from an early date, only became necessary as the size of aeroplanes increased. A part of the surface to be balanced is carried in front of the hinge and this surface is most frequently the rear portion of a fixed element, the part brought forward of the hinge being extended beyond the end of the fixed element. This so-called "horn "balance proved unsatisfactory. If a large "horn" were used (adequate to give ease in normal flight), there was overbalance at low speeds, or when the aeroplane sideslipped, and the controls would then tend to take charge.' A more uniform effort results if the balancing projection is run the full span of the hinge, which must then be set back behind the fixed element. The front edge of the balanced surface is sharp and its movement takes place behind the bluff end of the fixed element. Alternatively separate balancing surfaces in advance of the hinge have been rigidly attached to the moving element and placed above the fixed element.

HINGE

EXAMPLE OF HORN BALANCE FIG. 14.

HINGE

EXAMPLE OF SET-BACK HINGE BALANCE
FIG. 14a.

Two Methods of Balancing Ailerons.

The imperfection of balancing obtained has led to the development of relay motors to reduce the effort. In these, power derived from the air by a small windmill is brought into play whenever the flier attempts to move the controls. Relay motors had been but little used up to 1921.

Chassis or Undercarriage.-The Wright aeroplane alighted upon skids. It was launched by a catapult. The French pioneers took the air under their own power, and the Farman and Blériot used wheels. From 1909-14 combined wheels and skids were used. The wheels were commonly sprung by means of rubber cord. The skids might be brought into action if the alighting were imperfectly executed, and were carried well forward to prevent the aeroplane from turning over forwards when landing. Sometimes additional wheels were fitted in a forward position in place of the skids for this purpose. Under the tail a wheel was often fitted, but a small skid was used alternatively. Wing-tip wheels or more commonly light skids were used to protect the wing tips from contact with the ground. In Blériot's