Deltas and tailless aeroplanes
The delta-wing and the tailess aeroplane are very far from being synonymous, but certain of their troubles are shared, troubles that have led to the emergence of the one almost at the expense of the other. And yet it is scarcely true to say that the delta was evolved from the tailless aeroplane—a point emphasized by the fact that many delta designs have tails. The reasons for these contradictions are connected with the objectives underlying the adoption of the different unconventional forms.
As long ago as 1913, after eight years of patient experiment, that versatile inventor J. W. Dunne built and flew a tailless aeroplane. Dunne—whose interests ranged from fly fishing to experiments with time—wanted to make an inherently stable aeroplane. He appreciated that this could be done by sweeping back the wings so that the centre of pressure lay along a Vshaped line about the c.g. instead of on a line at right angles to the centreline of the aeroplane. Once this was correctly done, the wing became self-balancing over the whole speed range and the conventional stabilizing tail was superfluous. Dunne also realized the dangers of tip stalling with such a heavily swept wing and so he washed out the incidence toward the tip and increased it at the centreline. This arrangement ensured a nosedown pitching moment at the stall, leading to an automatic increase of speed to aid recovery.
The Dunne 8 biplane had a sweepback of 30 deg. and directional stability was achieved by vertical panels between the wingtips—which probably helped to inhibit tip stalling as well since they would act as end-plates. There were only two control surfaces, one on the outer part of each upper wing, which could be moved together or independently (elevons in today's nomenclature) by two levers one on each side of the pilot. In this way Dunne had achieved control in three planes by the simplest method ever devised, since moving both levers back and forward together obtained a pitching moment, one back and one forward caused both yawing and banking.
Because of the very simplicity of these controls, co-ordination on the part of the pilot must have had to be very precise and only the inherent stability of the aeroplane made them practicable. There were several Dunne biplanes, which were fairly competitive with other aeroplanes of their day. Structural knowledge , and the materials available, were scarcely adequate for the rather complex geometry and the experiments did not get very far. Dunne himself seemed to lose interest and the outbreak of the First World War turned people's minds toward the most easily and rapidly achieved goals.
About ten years after Dunne's experiments were abandoned G. T. R. Hill, in collaboration with Westlands, evolved a series of experimental tailless monoplanes. These, the Pterodactyls, so-called because their sharply tapered wings resembled those of the prehistoric flying reptile, proved to have good flying qualities. Hill's contributions to tailless evolution included rotating wingtips for control and independent vertical rudders mounted well along the span. The wingtips, which were elevons, were designed to avoid tip stalling. Because their incidence when they were used as elevators was less than that of the wing, there was always an unstalled air flow to make them effective as ailerons. The rudders did not move quite in unison, since the inner one in a turn had to provide drag—so giving yaw— while the outer one had to be turned only enough to suppress drag. Wingtip rudders have been used quite often on tailless aeroplanes, but Hill had few imitators of his rotating wingtips.
In Germany during the twenties there was another tailless enthusiast raising his voice in the wilderness, Dr. Alexander Lippisch. Starting with gliders, one of Lippisch's most significant early design was a braced high-wing monoplane, the Storch IV. Here there were the elevons and wingtip rudders, while sweepback was slight. The elevons on all these early aeroplanes were at the wingtips so as to give the longest possible lever arm about the c.g. for the elevator effect, but this position also introduced the unpleasant phenomenon of wing twisting, with its attendant danger of control reversal— a risk avoided by Hill's rotating wingtips. The small degree of sweepback on Lippisch's designs very much restricted the c.g. range that could be used.
Dissatisfied with the limitations of swept tailless aeroplanes, Lippisch hit on the idea of filling in the V-shaped gap at the trailing edge. That was how the A 1 was born in 1931. One of the main reasons for the revival of the tailless aeroplane had been the designer's dream of the flying wing— the idealized conception of a pure aerodynamic body providing lift but devoid of all excrescences. Unfortunately, although swept-wing designs had eliminated the rear fuselage and tail plane, their overall cleanness had not been remarkable. Lippisch hit on the idea of cleaning up his aeroplane by using the triangular planform. Although this first aeroplane had a fuselage, it was really a flying model for large all-wing transport projects. The large root chord of the delta made possible a greater depth of wing than any swept-wing could ever achieve without an uneconomical thickness/chord ratio. At this stage Lippisch retained his wingtip fins and rudders—probably because of the help the end plates gave to tip stalling—but used separate ailerons and elevators, since the whole trailing edge now had the same value for elevator leverage.
Although Lippisch, assisted at first by B. S. Shenstone, continued to work on delta aeroplanes and sailplanes until the outbreak of war, results showed that new powerplant was needed before success could be achieved. Changes in the slipstream
from the airscrew had a much greater effect on airflow over the wing than with a conventional aeroplane. To avoid these disturbances a pusher installation was used on the A 1. Again, a pusher engine meant concentrating more weight aft than was desirable from the c.g. viewpoint. Altogether, although the tailless delta might be a step toward the ideal flying wing, it is scarcely surprising that it received scant interest and no finance from the severely practical minds of industry and the cautious heads of air ministries.
Other German experiments of the period with a significance for the future were the flying-wing sailplanes of the Horten brothers. The underlying reason for these aircraft was the search for the ultimate low-drag airframe by two gliding enthusiasts, the elimination of every excrescence, every source of parasite drag. In a sailplane, the designer can neglect the problem of c.g. range, since he has no changing load to cater for and he can concentrate upon obtaining the lowest value of induced drag from his wing. For this reason the Horten sail
planes had the minimum of sweep, were tapered to the maximum degree, and had the highest possible aspect ratio consistent with structural strength. The fuselage was reduced to the smallest fairing that would enclose a man in a kneeling, semi-prone position. Although unpowered, these designs were later to be the basis of some remarkable jet projects, partly built, but unfortunately never flown.
In the U.S.A. another man of original ideas and remarkable inventive ability, John K. Northrop, started to take an interest in the tailless type in the late thirties. Starting with a small swept-wing aeroplane having a precautionary tail on booms, he achieved successful tailless flight in 1940. The N.1 M was a flying-wing model for the huge XB-35 and YB-49 bombers. Northrop, since he was aiming at a very fast, very long range flying wing, followed closely the Horten conception, but he added typically original ideas of his own. The turned-down wingtips on the model were an attempt to add fin area. Action of the control surfaces was arranged so that they acted as drag rudders as well as ailerons.
For his tailless aeroplanes Northrop experimented with several original controls. One idea was to have split surfaces at the tips, inside each of which was a bellows connected to a duct. When the controls were neutral, the duct was open and venturi suction kept the control surfaces flat and they were operated differentially as ailerons. When rudder action was required the appropriate duct was closed and air filled the bellows and opened the flaps, so making a drag rudder. This curious system achieved a good yawing moment without appreciable wingtip drag when not in use.
One of the difficulties encountered with the clean tailless monoplanes was that landing flaps acted like elevators and caused the nose to go down when they were lowered. Several efforts have been made to overcome this—including double flaps, the front halves of which were raised as the rear went down—but none was really successful.
Northrop's bomber was first flown with piston engines and when suitable turbo jets became available these were installed, the side area being adjusted by adding four fins. The fences extending forward of the fins were intended to prevent the jets from sucking the air off too large an area of wing. Note how the huge wing is otherwise clear of excrescences—save for the small blisters for the crew and armament. The centre of the wing is deep enough to house a large bomb bay, while the rest of the wing is potentially a huge fuel tank. Like most high aspect ratio flying wings the Northrop is rather unstable longitudinally.
Completely different were two of Northrop's attempts at tailless fighters, the XP-56 and XP-79. The former had a pusher piston engine and fairly large vertical fins. The twin-jet XP-79 was interesting because the pilot was put in a prone position to reduce head resistance. Both these aeroplanes had the bellows-operated, split-surface controls. After the War, in Britain and France there was an effort to pursue the tailless type. After a series of glider experiments the beautiful Armstrong Whitworth A.W.52 was flown, while the three D.H.108 monoplanes were a source of much valuable information.
The D.H.108 was a research aeroplane built in 1946 to investigate the behaviour of a highly swept wing. The three examples put in many hours of valuable work and it was in one of them that John Deny was the first British pilot to exceed the speed of sound in a dive—which proved how very clean it was, since the engine-thrust was only about 3,500 lb. Although often demonstrated in public to be a practical aeroplane , the D.H.108 convinced the de Havilland designer, R. E. Bishop, that his D.H.11o fighter should have a tail, so that flaps could be fitted. Without flaps, landing has to be made at high incidence and with heavy sweepback the wingtips are in danger of fouling the ground.
When War came Lippisch had been working on a rocketdriven delta fighter which he was told to stop because the Nazis thought they would win in a short time. When Allied bombing raids were disorganizing German industry Lippisch was told to co-operate with Messerschmitt on a rocket-driven interceptor. The two men could not agree and Lippisch went off to his native Vienna in 1944, but only after he had laid the foundation of the design of the Me 163—the only tailless aeroplane ever to be used in quantity.
The concept of the Me 163 was to have a very light fighter capable of climbing very fast and of being highly manoeuvrable at its operational altitude. Endurance was drastically sacrificed to attain this goal, because the chosen engine, the Walter 109-509 bi-fuel rocket had a consumption of nearly half a ton a minute. Rocket aircraft are extravagant on propellant because they carry their own oxygen for combustion, which is about fourteen times the weight of the actual fuel. However, the motor is very light and, far from power falling with altitude, it increases as the outside air pressure falls. The main trouble with the Walter rocket was that it could not be throttled. Since, however, the two fuels were self-igniting the endurance of the Me 163 could be extended by proceeding with short bursts of power followed by coasting.
The tailless layout was chosen for the Me 163 in order to get a compact airframe with the lowest drag and weight for the largest wing area. The sweepback necessary for stability also fitted well with the new theories for high speed flight. The wing had slots and washout to improve control at the stall. The washout was found to lead to serious instability at high speed because it caused boundary-layer breakaway. The fuselage was short and fat, being mainly a fuel tank, and there was a large fin above and below to give the damping in roll which all tailless configurations lack. To save weight the Me 163 had no undercarriage, only a retractable landing skid and tail wheel, take-off was made from a trolley. Armament was two 30 mm. cannon, mounted in the wing roots.
As a weapon the Me 163 proved to be a curious mixture. In climb, speed and manoeuvrability it was far superior to anything else in the air and it could dive through a Fortress formation, shoot, pull up and be away before the gunners could fire. Endurance was only eight minutes on power, with perhaps fifteen minutes gliding from 30,000 ft. This was effective for last-minute interception in clear daylight only. The trolley takeoff technique was not a success, because the returned aeroplanes had to be retrieved from all over the aerodrome like gliders. In a later form, this fighter was given a tricycle undercarriage, which simplified recovery, and a small-thrust cruising nozzle that increased endurance under power to 15 minutes.
That really concludes the history of the tailless aeroplane. It may possibly have a revival, but it is difficult to see where its virtues cannot be bettered by the delta—with or without a tail. The delta as originally conceived by Lippisch was of moderate aspect ratio. Other experimenters in the thirties, approaching the flying wing from other directions, discovered the curious properties of very low aspect ratios. It appeared that, like a child's circular kite, where the aspect ratio was less than 3 to 1, lift and stability remained at very large angles of incidence. The Hoffman Arup monoplane and the work of Carl Zimmerman proved the theories. Zimmerman's work came to fruition during the War in the weird Chance-Vought V-173. This aeroplane was the flying model for a "convertiplane"
fighter, the XF5U, in which special large airscrews would take over from the wing at low speed to give lift. The sketch shows such of the theory of this aeroplane as has been released. Concentration on high speed flight killed this "circle plane" despite promising experimental results—although it may well be revived. Although when Zimmerman left Chance-Vought the company abandoned the circle plane, they had obviously been attracted by the characteristics of low aspect ratio aerodynamics , for they made the even more extraordinary XF7U
Cutlass. This aeroplane is definitely tailless, but otherwise it is in a class of its own. It has less taper and less sweep than usual and its long nose is unique. The characteristically large fins extend under the wing to form fairings for the short mainwheel legs. The nosewheel leg is very long because high incidence is used for take-off and landing. The cockpit being high and well forward, the pilot has an excellent view in any attitude . The two engines have wing root intakes, giving a compact power installation. Flying qualities of this weird aeroplane must be good, since it is in general use on aircraft carriers. After his quarrel with Messerschmitt, Lippisch worked on his delta ideas at the Vienna Aeronautical Research Institute and he now combined them with low aspect ratio. After extensive wind-tunnel work he evolved a series of designs, starting with a very advanced delta glider and culminating in a supersonic aeroplane with a revolutionary coal-burning ram-jet.
These projects were essentially pure deltas with a huge triangular fin, the root of which formed the cockpit canopy. This shape, in theory, takes most advantage of the delta form, since it leaves the wing free to develop its unique high-incidence lift characteristics; but it is a difficult shape to achieve in practice.
The virtues of the delta are brought out to best advantage when used to achieve a large speed range—that is to say moderate stalling speed and high top speed. Flight characteristics are not yet by any means fully understood—neither are those for any but the simpler conventional aeroplanes judging by the troubles accepted as normal with prototypes—and much of the data obtained are unpublished because of security. It is, nevertheless, a striking fact that after forty years of practical r flying, a radically new shape of aeroplane should suddenly come to the fore and threaten the very existence of the conventional layout.
The reason for the suddenness of the popularity of the delta-wing layout lies in two complementary factors. The jet engine made possible the elimination of the slipstream that had hitherto hindered development. The vast increase of power that came with the jet engine lifted aeroplane speeds into the
troublesome sonic zone where leading-edge sweep is essential. Practical delta flight was made possible by the turbo-jet, the delta wing made possible a geometry ideal for using jet power to full advantage. The delta planform gives the largest area for a given span and the root chord is so great that for a low relative thickness a large physical depth is available. With a low aspect ratio a sweepback of as much as 60 deg. is possible. The triangular structure is inherently stiff, and the extra working depth makes it relatively easy to construct. The large wing volume is excellent for holding fuel and undercarriage—the engines as well in the large sizes. There are two approaches to the delta design, the "envelope"
and the "barrel". The envelope is, of course, more easily achieved on the large aeroplanes. It is the flying wing conception in modern form, with every possible excrescence removed . The Avro Vulcan comes very near to this ideal and only the nose, fin and jet pipes protrude. In the barrel—the Convair XF-92A and the Fairey FD-1 are examples—a very small, thin delta wing is married to a fat fuselage containing pilot, engine and fuel. Some compromise can be accepted for medium-sized aeroplanes, and even small ones. The Javelin has a wing containing fuel and undercarriage, but it cannot house the crew and two Sapphire engines, so that a plump central nacelle is necessary. The S.A.A.B. S-210, on the other hand, which is a model for a twin-engined fighter of similar size, is an ingenious adaptation of the envelope wherein a compound sweep gives a deeper centre-section to contain the engines and equipment.
The flying qualities of the delta form are excellent. In the , first place the drag rise at sonic speed is low because of the sweep and relative thinness. Secondly, the changes in trim are less than with conventional types and are small enough not to require a tail plane. Thirdly the large wing area, with consequent low wing loading, compensates for the severe loss of lift at Mach numbers above -y. This last factor is of particular importance in the thin air of the stratosphere. Because of a „ phenomenon known as wave suction, a delta wing does not require the usual sharp, high-speed leading edge—and the rounded leading edge improves the stalling characteristics.
In its general handling the delta-wing aeroplane is normal. It will roll rapidly because the span is less than that of a comparable conventional type. In fact, damping in roll is one of the problems, but this is usually achieved by the large fin. So long as the fin area is large enough the delta will be directionally stable and yet will respond well to its controls. The chord is much greater than with other tailless types and because of this the elevators are powerful, for the same reason stability in pitch is good and the c.g. range is ample.
It is at the low-speed end of the speed-range that the delta wing exhibits its most curious characteristics—or to be more exact it is to the low aspect ratio that the credit should be given. The aeroplane will maintain lift at angles of as much as . 35 deg. and can be controlled in this attitude. The reason for this appears to be connected with a breakaway of the airflow down the centreline, which curls out to meet the air flowing round the leading edge to the suction area above the wing. At first, some trouble was experienced with elevator reversal at high s incidence and also with lateral instability, but flight tests have cleared up these points.
The earlier deltas usually had pointed wingtips, but now the general practice is to clip the tips, since they provide little lift while adding drag. They also increase the aspect ratio out of all proportion to the small amount of extra area they provide.
Flying the large Avro Vulcan is, according to its test pilot, . > easier than flying an Anson. Furthermore, if the aeroplane is trimmed into its correct gliding angle it automatically levels t out as it gets into its own ground cushion—that is the air deflected downward by the huge wing. It may appear curious, after this catalogue of virtues, to find some delta aeroplanes fitted with tail planes. The principal reason for fitting these adjuncts, which add both weight and
drag, is a desire to use flaps. On a night fighter like the Javelin, for instance, the high-incidence landing attitude might well prove troublesome. Similarly, a delta air liner might be expected to have a tail plane to save the passengers from a startling, even though perfectly safe, experience. The tail plane, once fitted, becomes part of the design and is used also to trim out Mach effects and aid manoeuvrability. It is, though, curious to note that the first two delta aeroplanes to be designed with tail planes ran into a trouble impossible with the other deltas—tail nutter!
So much for a general picture of the theory behind the delta wing, which must be supplemented by an outline of its rapid evolution over the past five years. After Germany collapsed in 1945 the technical investigating teams from the U.S.A. were impressed by Lippisch's work and carried him off, accompanied by his DM-1 glider, to the N.A.C.A. The glider was subjected to tests in the full-scale wind tunnel with such promising results that Convair was asked to make a full-scale jet-engined delta in a hurry. This aeroplane, the XF-92A, was built so rapidly that it was concocted from parts of several other aircraft . The delta wing was only 6 per cent thick, the thinnest ever flown. The XF-92A flew in 1948, was later crash-landed after an engine cut on take-off, eventually rebuilt and fitted with an afterburner. Official interest appeared to languish, and at one time it looked as if the U.S. authorities were going to drop the project because it was not immediately successful. However, the Convair company were encouraged by what they learned and succeeded in persuading first the U.S. Navy and then the U.S.A.F. to order fighter prototypes—the XF2Y-1 Skate and the XF-102.
The U.S. Navy also ordered, independently, the XF4D-1, Skyray, from Douglas. This is a fairly conventional delta configuration with a thickened air intake blending the wing into the fuselage. It is remarkable for a much smaller fin and rudder than is usual on a delta-wing aeroplane. Although the Skyray has a J-40 turbojet, which is in the 10,000 lb. static thrust class, and must be very fast since it is quite small, the cockpit hood is large and angular. Certain photographs suggest that, instead of an ejector seat, this aeroplane has a "survival capsule", that is to say the whole nose and cockpit can be jettisoned, slowed by drogue and then let down by parachute—this is only conjecture . The Skyray first flew on January 23rd, 1951, but it is still completely "Restricted" at the time of writing.
British delta development has already taken its place in history as part of the post-War re-equipment programme for the R.A.F., with production of the Vulcan and Javelin ordered and new fighters on the way. It is, however, worth recapitu• lating the stages of evolution since for several years British constructors remained the sole proponents of the revolutionary layout.
First in the line were the Avro scale models for the Vulcan. The first of these, the 707, flew on September 4th, 1949, the second, the 707B, on September 6th, 1950, and the third, the 707A, in July 1951. Each is about one-third full size and is powered with a Rolls-Royce Derwent of only 3,500 lb. s.t. The v first two were for research into delta-wing behaviour and the air for the engine was scooped from above the fuselage so as to leave the wing unimpeded. Although there was no particular call for very high speed, this intake was so inefficient that it was modified by fitting two long fences ahead of it to part the , boundary layer. The Avro 707A is a more exact model of the Vulcan, and the wing root and intake shape served to test those for the bomber. The blunt shoulders for disposing of the fuselage boundary layer on both aeroplanes are of a unique shape.
The Avro Vulcan, which first flew on August 30th, 1952, is remarkable for having no excrescences other than the wingtip pitot heads. The fuselage, such as it is, is absolutely smooth 'and the smoothness of the wing is unbroken save for the four jet pipe fairings. Like the 707s, the Vulcan has a tail fairing containing a braking parachute for reducing the landing run. In addition to Avros and Glosters, Fairey and Boulton Paul have carried out research programmes on the delta configuration . The Fairey F.D.1, which flew on March 10th, 1951, was fitted with a Derwent engine and was to have been used for experimenting with control by variation of jet thrust from four • nozzles. This was the first delta-wing aeroplane to be seen with a tail plane, but in this case it was a temporary precaution for early flights and not a true control surface.
The Boulton Paul P.111 flew before the F.D.1, on October 10th, 1950, and proved to be an agile little aeroplane. Later, in the summer of 1952, a second delta, the P.120 was built. In this the tail fin and rudder were cut off short and an adjustable tail plane added for the purpose of comparing the characteristics of these two styles of delta. Unfortunately, tail flutter developed and this prototype was lost. Both the Boulton Paul designs had Nene engines, so that their thrust was too low to obtain very high speed.
In its short active life of five years the delta wing can be considered to have proved itself as one of the most important configurations of the immediate future. It is equally obvious that there are going to be quite as many variations as there are of classic wing plans. Even though the designer of a delta wing is committed to casting in his lot with moderate wing loading, he may well vary the proportions for different needs, such as equipment and engine installations. It is likely that military types will generally be made without tails, while any possible air liner will have a tail to enable napped, low-incidence landings to be made. It may be that the tail plane will be used for military types, where a level landing attitude is required, or where extra stability is needed in the transonic zone.