The rise of the jet fighters

For some years before the War an event was taking shape that was destined to revolutionize the aeroplane and, above all, to render completely obsolete the piston-engined fighter whose development has just been traced. Like most great inventions, jet propulsion did not happen suddenly, but was a process of painstaking research and, also like so many other discoveries, engineers were at work on it unknown to each other in different lands. In Great Britain a young R.A.F. officer had for years been experimenting with little financial help or encouragement. Today we know just how much we owe to Air Commodore

Sir Frank Whittle. Actually, work on the gas turbine as a replacement for the piston engine had also been going ahead for some years at the R.A.E., but it was Whittle who thought of using the gas turbine to drive a compressor and so produce a propulsive jet—a jet such as the Frenchman Rene Lorin had foreseen in 1913, but had no means of putting into practice. In Germany and France work was in hand with turbo-jet engines, to be interrupted in the latter country by the Occupation. Junkers, BMW and Hirth, the latter financed by Heinkel, were hard at work on jet propulsion for three years before the War, and the world's first jet flight was made by the He 178 experimental single-seater with a He S3 engine of 1,000 lb. thrust on August 27th, 1939. It is certain that if he had had similar financial backing, Whittle would have had his engine the W.1 flying before May 15th, 1941, when it first topk the air in the Gloster E28/39.

The advent of the turbo-jet power unit brought with it the immediate prospect of hitherto undreamed-of power, with the engine actually increasing its propulsive efficiency the faster it flew. Furthermore, these new engines were light and simple and dispensed with the airscrew, itself a weighty and complex accessory. There were, of course, new problems of which the greatest was probably the inclusion of a giant blow-lamp within the aeroplane. Fire has always been the deadly foe of aircraft and it was a new conception to be burning liquid fuel in thin-walled "cans" as opposed to relatively robust cylinders

The Germans maintained their lead in jet flight spurred on by their desperate need for defence against mass Allied bomber and fighter attacks. To them goes the credit of the first operational jet fighter the Me 262, which saw action in 1944. The Jumo 004 axial-compressor jet engines gave only some 1,750 lb. thrust, so the designer had, perforce, to use two. Because of the fire risk and also the short overhaul life of these engines, they were mounted in underslung nacelles. The designer realized earlier than most that a thin aerofoil section was essential for high speed flight and also the assistance to be gained from sweepback. The Achilles heel of the jet engine, its high fuel consumption, resulted in a fuselage of very large volume. The shape chosen was a triangular cross section so that the pilot could look down at an angle on each side. As in the Me 11o the cockpit hood was hinged to open sideways, instead of slide, and the four 30 mm. cannon were in the fuselage top decking. As in other Messerschmitt aeroplanes, automatic slots were fitted, in this case over the whole span, as a palliative to the sharp stalling characteristics of the swept wing. Because of the underslung nacelles a rather tall tricycle undercarriage was essential. The thin wing and the jet nacelles posed the problem of today, where to stow the main wheels after retraction. The solution chosen was to mount the undercarriage legs inboard of the nacelles and to retract them inward so that the wheels lay in the bottom of the fuselage—a now classic arrangement. The Me 262 was a good aeroplane and had a far higher performance than any other fighter of its day.

The first British jet fighter, the Gloster F9/40, which became the Meteor, was also twin-engined and deceptively conventional at first sight. A thin laminar-flow aerofoil was used and the low aspect ratio that is now so familiar, but no sweepback. The wing held no fuel and once again the fuselage size was largely dictated by the tankage required. The British designer, W. G. Carter, took advantage of the twin engines to put the pilot well forward where he had an outstanding view. The four 20 mm. guns, too, were well placed, on each side of * the pressure cockpit, low enough for the pilot to be unaffected by flash and sufficiently high to prevent nosing down. The original Rolls-Royce Welland engines (and the Derwents later fitted as standard) were short centrifugal units and so dropped neatly between the wing spars. True, the jet pipe caused something of a structural problem, since the rear spar had to be , enlarged to a "banjo" to let the pipe pass through, but it was easy to lead air under and over the front spar.

The early Meteor nacelles were short and did not match the fineness ratio of the fuselage and wing. Quite soon, however, long nacelles were introduced which, at some cost in weight, greatly improved the streamlining. Flight tests with tufting had shown that where the short, highly curved nacelle met the wing the air flow broke away causing buffeting which limited the speed of the early Meteor. Because the engines were mounted in symmetrical nacelles, a very short undercarriage was possible and it was designed to retract into the centre section between the spars.

Because of their desperate battle against Allied attacks the Luftwaffe put the Me 262 into action when the engines would only last some 25 hours between overhaul. In fact, the pilots were reluctant to fly the heavily loaded aircraft because of the number of engine failures which occurred and which often resulted in fatal crashes. In Britain a small number of Meteors had been completed by 1944, but the R.A.F. would not put them into service until a reasonable engine overhaul life was achieved. In May 1944 the Welland was approved for 180 hours between overhauls and on August 14th the first flying bomb was destroyed by a Meteor. The first Meteor unit went to France, but never saw action with enemy fighters.

Viewing these first two jet fighters as dispassionately as possible, some interesting comparisons can be made, each showing some features better than the other. The Me was true to tradition in having the higher wing loading and its dimensions were not much greater than those of the Hurricane, span 41 ft., length 34 ft. 10 in. The Meteor originally had a span of 43 ft. (the wings were later clipped to improve roll) and the length was 41 ft. 3 in.—since increased to give more fuel space. With similar powers the Me 262 had a maximum speed of 525 m.p.h. and the Meteor, with short nacelles, 480 m.p.h., while both had a service ceiling of around 40,000 ft. Some idea of the jet engine's thirst may be had from the internal fuel capacity of the Me 262, which was over 500 gallons, weighing about two tons, and was some five times that of a single piston-engined fighter with a similar endurance of about one hour.

These, then, were the first operational jet-propelled fighters and it is right that they should have pride of place in any description of development. They were not the only designs to fly before the end of the War and they certainly did not exhaust the layout possibilities. In fact, it is likely that the Me 262 and the Meteor reached maturity first just because they were so

conventional. By contrast, the original experimental jet aeroplanes, the He 178 and E28/39, were of the form that most obviously took advantage of the new power plant.

These two aeroplanes were each little more than a tiny wing attached to a well-streamlined fuselage with an air intake in the nose and the minimum of excrescences. The limited output of the early engines made it essential to use such small aeroplanes if the required speeds of over 400 m.p.h. were to be achieved. Since they had no guns or other operational equipment to increase their weight it was no very difficult task to keep down the size. What did transpire was that the ducting of the air from the nose intake posed many problems; valuable fuselage space occupied, friction losses in the duct, excessive cooling of the cockpit by the draught, these were some of the troubles. Altogether the nose intake (or pitot entry as it is often called) proved sufficiently discouraging for designers, particularly in Britain, to avoid it for many years.

When the Whittle bench tests proved the potential success of jet propulsion, the Ministry of Aircraft Production put each British engine company to work on designs, so widening the possibilities of variation on the original basic patents. Major F. B. Halford undertook the design of a centrifugal jet engine for The de Havilland Engine Co. Ltd. Work on this engine, the H.1, later to become famous as the Goblin, started in April 1941, and the engine was bench running within a year, flew in an F9/40 on March 5th, 1943, and on September 20th of that year took the air in the D.H. 100, the prototype Vampire singleseater fighter.

In this aeroplane the designer had adopted a novel approach to the jet installation. To obtain the best power with the minimum of friction losses, the intake duct and the tail pipe should be as short as possible. In the Vampire this was achieved by using a "pusher" layout with tail booms, placing the engine at the rear of a short nacelle and bringing air in through ducts in the wing root. Excellent in itself, this arrangement brought certain of its own problems with it. The sluggish envelope of the fuselage boundary layer would have slowed the intake air if scooped up with it, so it had to be spilled across the wing by special shoulders. Making the centre section sufficiently strong with the two large tunnels in it posed an intricate problem in stress distribution. The old tail boom difficulties of rigidity during manoeuvres were, of course, revived, but in a much simplified form because of the absence of piston vibrations and slipstream buffeting. Further advantages occurred from exceptional engine accessibility and close, compact, grouping of engine, guns and pilot.

Originally the D.H. 100 was known as the "Spider Crab", but the name Vampire was adopted as a tribute to its bat-like wing shape. Whatever the name, although a rather slow production start prevented active participation in the War, outstanding manoeuvrability and general good flying qualities resulted in its being a pillar of strength in the R.A.F. and the N.A.T.O. air forces for over ten years—although its guns were never fired in anger!

As late as 1941, the U.S.A. had done nothing whatever about jet propulsion, although it was the supply of their special heat-resisting nickel alloys—developed for the G.E.C. exhaust turbo-superchargers—that made possible longer life turbines for the Whittle engines. In October 1941 a Whittle W 1 engine, complete sets of blueprints, and several engineers from Power Jets Ltd.—the government-owned company which was by then sponsoring jet research—were flown to America. Within a year the G.E.C. had made several engines and the Bell XP-59A fighter prototype had flown.

A few of these aeroplanes were built, but by the time they reached the U.S.A.A.F. in August 1944 it had been realized that they were only fit for fighter training duties. The Airacomet, as the company called the P-59A, was a bulky mid-wing monoplane of undistinguished aerodynamic shape. The main claim to attention was that the engine nacelles were tucked up in the angle between wing and fuselage. At first sight this seems a logical position; good accessibility, no ducting or long tail pipe, and no appreciable asymmetric forces after engine failure. However , aerodynamic interference was considerable and so drag was high and the Airacomet was no faster than the better contemporary piston-engined types.

In 1943 another present was given to the Americans, the de Havilland H.1, which was handed over to the U.S.A.A.F. at Wright Field in July 1943. Once again with characteristic hustle a fighter was designed round the engine, but this time although "pencil to flight" time was only 143 days the result was a winner, the Lockheed XP-80, later to achieve fame as the Shooting Star.

The Shooting Star was, in effect, simply a very beautifully streamlined aeroplane incorporating all the knowledge then existing among the Allies on high-speed flight—a thin laminarflow wing with a sharp leading edge, an almost perfect streamline for a fuselage, with six -5in. guns tucked almost unnoticed in the nose, a pure "teardrop" cockpit cover, and an elegant tail unit. The fuselage, as on the other jet aeroplanes, was broad and, just ahead of the wing roots, it bulged into two scoops, while instead of a pointed tail end it had a circular orifice—sole evidences of the buried engine. The undercarriage was even shorter than that of the Meteor and the main wheels folded into the bottom of the fuselage, the front wheel under the nose. It is important to note how the new jet-engined aeroplanes brought about an immediate general adoption of the tricycle undercarriage. The main reasons for this are twofold. To take advantage of the low ground clearance without an airscrew to shorten and lighten the undercarriage the ground angle is reduced and the braking propensities of the old type landing gear are largely nullified. On the other hand, the jet engine has a relatively poor build-up of thrust at take-off and it is well to have the aeroplane initially in a low-drag attitude.

The Germans did not put all their faith in the Me 262, other designs were built and flown, but only one the Heinkel He 162 Volksjdger, was put into production—and was actually delivered in quantity before Germany's final collapse. This aeroplane , the "people's fighter", was ordered in the summer of 1944 as a desperate remedy for the shortage of fighters caused by casualties in action and the bombing of factories. It was designed in a hurry to be made largely of wood and to use one of the 1,750 lb. thrust BMW 003 axial-flow turbo-jets. The idea was to make a small, simple machine so that output could be increased as a result of the saving in labour and materials. The military load was restricted to two 30 mm. cannon, the minimum of armour and instruments and fuel for about one hour.

The engine was put on top of the fuselage to facilitate changing and this, in turn, led to the need for fins and rudders at the tail plane tips. Ground clearance for the fins meant a dihedral tail plane—so one feature affects another. With a small wing the wheels had to retract into the fuselage, which meant accepting a very narrow undercarriage. Fortunately, the lack of torque reaction with jet engines greatly reduces the tendency to swing at take-off, and a wide track is not so essential as with pistonengined fighters. Despite its low power the He 162 could keep pace with the Me 262 and the Meteor, since it could do 522 m.p.h. at 20,000 ft. The curious turned-down wingtips were an attempt to overcome tip-stalling, a trouble accentuated by high wing loading and the thin laminar-flow aerofoil used. The Volksjdger was far from being the full extent of German work on the development of a new breed of fighters intended

to take full advantage of the possibilities of flight at 1,000 km.p.h. (620 m.p.h.) presented by jet propulsion. Even before the War Busemann and Ackerit had followed up the work of the 19th century Dr. Mach by intensive research into air flow behaviour at the speed of sound. This work, at first done in tiny wind tunnels, because of the expense of pumping large quantities of air at very high speed, was extended and larger tunnels were built. There appears too, to have been considerable duplication of effort and a lack of co-ordination in German aeronautical research and design that can only be considered fortunate from the Allied viewpoint. When the enemy engineers and scientists were interrogated and their latest designs studied, it was discovered that many unrelated developments in hand were years ahead of British

and American thought. Projects ranged from the fantastic V-4 winged rocket intended for attacking New York, to light rocket engines, ultra-simple ram-jet engines, complicated 7,500 lb. thrust axial turbo jets, swept-wing fighters and bombers, and even a jet-driven 600 m.p.h. helicopter! Much of this material was brought to Great Britain, but far larger quantities (including leading personnel) went to the U.S.A. and the U.S.S.R. —where they have, apparently, remained a constant source of inspiration for nearly ten years.

In this chapter it is aerodynamic and engine knowledge as applied specifically to fighters that is of interest, so that one must regretfully dismiss these enthralling side issues. The evolution of jet bombers and rocket fighters'is dealt with in Chapters VIII and XIII, respectively. The reason for this is that the special needs of the bombers might well cloud the picture of fighter development, while the Me 163 rocket fighter fits more neatly into the jigsaw of the tailless aeroplane's evolution.

The most important feature appreciated by the Germans was the advantage of sweepback for high-speed flight. To understand this it is necessary to know something of air flow behaviour at the speed of sound. Normally when a body passes through the air it not only parts the air molecules, which flow along its sides, but it sends ahead of it a series of pressure pulses that warn the air of the body's coming. Some of these pulses are the sound waves we hear, others are of different frequency, but they all travel at the speed of sound, i.e. about 762 m.p.h. at sea level reducing with temperature to 660 m.p.h. at 36,000 ft.

When a body's speed approaches that of sound the pulses build up on each other and travel ahead in the form of a shockwave cone, and when sonic speed is reached the cone is, of

course, co-incident with the nose of the body. This shock wave is a physical change in the air, it is only a minute fraction of an inch thick but at this point the pressure, temperature and density of the air change abruptly. Moreover, air moving above the speed of sound behaves according to the laws governing compressible fluids and not those of gases, which means that quite different streamlines and other characteristics are required—details of which are too involved for discussion here. Naturally, in the boundary region where the air molecules are, as it were, being tossed between one set of laws and another there is considerable disturbance and it is this which forms the turbulent trans-sonic zone—the "sound barrier" of the Press and films.

As a body moves through air there are parts of it over which the air moves faster and actually reaches the troublesome speed locally before the body as a whole is travelling at that speed. To understand this, first imagine a streamline body with a hole down its centre moving at near-sonic speed, then an air molecule will travel the straight distance A,B at the same speed as the body. However, the molecules going round the outside have to travel farther (A,C,B or A,D,B) and

must go faster if they are all to arrive together at B. The molecules going by the more highly curved surface (A,C,B) will obviously have to go fastest of all. Nature, we are told, abhors a vacuum and it is because of this that the molecules travelling the curved long "route" have to increase speed. As

increases. The "sonic barrier" is shown shaded they reach the sharpest curves, shock waves are formed, and in their formation energy and power are absorbed. The change in temperature and pressure across the wave indicate an addition of energy and this energy is derived from the aircraft's engine. Or, to put it another way, there must be enough power available to provide the energy to propagate the shock waves and break through the sharp drag rise of the "barrier".

As well as increasing drag, the formation of these local shock waves has a bad effect on the lift of the wing. As the air flow over the wing nears the speed of sound the centre of pressure moves aft to mid-chord because of the effect of the local shock waves. It will be recalled that in normal flow conditions, as speed increases the centre of pressure moves forward. This means that on accelerating into the sonic zone there is first a nose-up and then a nose-down trim change. Air flow over the tail is affected like the wing, so that the elevator becomes ineffective and unless assisted the pilot cannot maintain control. If he trims out the nose-down tendency and goes on diving into warmer air he reaches a point where the speed of sound is considerably higher, the shock waves cease abruptly and the resumption of normal air flow results in a violent nose-up moment. Before these troubles were fully understood not a few aeroplanes lost their wings until the value of dive-recovery brakes was appreciated. These brakes take many forms, but the essential feature is a means of greatly increasing drag.

This outline of trans-sonic troubles gives an idea of the problems on which the Germans were working. One most important solution which they hit upon was the effect of sweeping the wing back. First of all, it is obvious that to prevent sudden curves causing shock waves, the thinnest possible wing is essential, that is thin in the sense of the thickness/chord ratio. Instead of thinking in terms of 12 to 15 per cent, the idea now was to go down to 10 per cent or even less. This meant very little depth in which to build strength against the bending forces of the wing lift. The German solution was to sweep the wing to the air flow, either backward or forward, and so present a lower thickness/chord ratio in the line of flight for the same physical thickness. It was a form of geometrical cheating which has proved such a success that it is now almost universal. As always, there is a price to pay, and sweepback introduces its own problems, of which wing twisting when the ailerons are deflected is one of the most severe.

Before returning to the new jet aeroplanes themselves, a word about Mach number is necessary. The foregoing shows that the crucial factor in aeroplane design had become not the actual speed, but the relationship of this speed to the local speed of sound—and the latter varies with altitude. The 19th century physicist Mach suggested the use of this ratio in his gas flow experiments and it has now become a criterion of high-speed aircraft design. Furthermore, when manoeuvring a jet aeroplane the pilot uses his Machmeter as frequently as his air speed indicator.

The Germans used limited sweepback in the Me 262, and in several partly completed fighters, notably the Messerschmitt P-1110, the Junkers EF 128, and the Focke-Wulf Ta 183. The latter is particularly interesting, since the method of installing the engine in the fuselage with only a moderate-length intake duct and a short tail pipe was unusual. The fuselage was very short, with a nose intake dividing round the pressure cockpit, as in the E28/39, and a short tail pipe behind the wing. In order to place the rudder and elevator far enough behind the c.g. for satisfactory balance and control the fin was raked sharply back and the tail plane mounted on top of it. If this description strikes a familiar note it is because of the MiG-15. Kurt Tank, designer of the Ta 183 went to the Argentine, where he applied his formula to the Pulqiii II, but his factory and many of his staff remained in the Eastern zone of Germany. There is no doubt at all that the best fighter of the Korean war is largely of German origin.

The Ta 183 layout was also taken up by SAAB in Sweden, where the J-29 was produced. In this case, instead of the swept fin a stepped fuselage was used and the tail plane was placed only high enough to keep it clear of the wing wake. Despite its large, fat fuselage, the J-29 can exceed 650 m.p.h. in level flight with the 5,000 lb. thrust de Havilland Ghost centrifugal. This brings us to the centrifugal versus axial controversy— a technical engine question, but one vitally affecting aeroplane design. The centrifugal compressor gives a rugged, heavy-duty engine, relatively simple to make, short and rather fat in appearance. The axial compressor is much more refined, it is sensitive to operating conditions and is more easily damaged by foreign bodies, it has more finely machined parts and, although slender, it rather runs to length.

The sketches show, diagrammatically, the essential features of the two types. In each case, the cycle consists of a turbine wheel connected by a shaft to a compressor that provides air to the combustion chamber where, burning with added fuel, it provides the gas stream to drive the turbine and propel the

aeroplane. Compression and combustion expand and accelerate the air, so increasing its energy and enabling it to drive the turbine and provide the jet.

The centrifugal (Whittle) engine has an impeller developed from the centrifugal supercharger. In this type of compressor it is the centrifugal force throwing the air out which compresses it and a considerable diameter is necessary. In the Whittle engine an ingenious double-sided impeller keeps the actual casing diameter to a minimum, but to feed the rear impeller it is necessary to enclose the engine in a plenum (pressure) chamber. Because of this plenum chamber the installed, or nacelle, diameter of the Whittle (Rolls-Royce) engine is little, if any, less than that of the de Havilland single-sided impeller engines. Compression by a single-stage centrifugal impeller is limited to about 4.5 to 1.

The axial compressor consists of a series of multi-bladed fans, or airscrews turning in an annular duct, each fan being separated by a row of fixed blades called stators. Compression is obtained by speeding the air backward into a duct of diminishing cross-section. Through flow is obviously less obstructed than within a centrifugal, where the air makes two right-angled turns, but this is a virtue that can cause trouble. Compression ratio depends on the number of blade rows and can be as high as 10 to I, which greatly increases power while reducing fuel consumption. Any blade damage tends to run right through the engine and during unstable flow conditions, as when starting or changing speed, there is little resistance to surging—that is uneven flow and back pressure caused by choking when the compressor is overloaded with air.

The blades of the compressor and its stator present formidable problems in design and manufacture. In the first place each blade is a very accurately shaped, thin, high-speed aerofoil and each row is a different length, so that the engine virtually contains a large series of tuning forks, excitable over a large range of frequencies—and resonance can, therefore, occur and break the blades. Making the blades, between two and three thousand per engine, is an enormous problem in mass production , because of the exceptionally high accuracy required.

Very briefly, that digression sets out the main pros and cons of the two basic types of turbo-jet, which are necessary to the understanding of the numerous queer shapes adopted for fighters since the War. I say fighters advisedly, because it is in the smaller airframes that the engine exerts the more obvious influence on form.

During the War the Germans concentrated on axialcompressor turbines, and got them into service, but with poor reliability. In Britain we worked first on the less temperamental centrifugal and achieved surprisingly good reliability, a feature which is still outstanding in the Derwents, Goblins and Ghosts of the Meteors, Vampires and Venoms in the N.A.T.O. air forces. Despite the success of the centrifugals, there were changes at the Ministry of Supply and because it seemed likely that the development of the centrifugal would be limited to about 8,000 lb. thrust it was decided to concentrate all resources on a new series of high-thrust axials. Whether this sweeping decision was right or wrong it is still too early (in 1953) to say.

What is certain is that it has taken a very long time to get any quantities of British axials into service, and that even the vast resources of the U.S.A. were hard put to it to maintain supplies of axials for the small-scale Korean War. To add poignancy to the argument, the enemy MiG-15 has an engine developed by the Russians from the Nenes given to them at the express instance of the late Stafford Cripps in 1947. A peculiar case of politics affecting aerodynamics—a feature of modern aeronautical development that is becoming more common as costs become astronomical and financing may influence not only departmental policies, but also the votes of electors. Just to turn the sword in the wound, the Swedes had in service by 1951 an exceptionally good sonic fighter in the J-29 with the Ghost centrifugal, while the Americans developed our 6,000 lb.-thrust Rolls-Royce Tay to give 7,500 lb. and put it into their newest all-weather fighter the Starfire.

So the history of the jet fighter rests on the choice of engine and where to put it. At the end of the War, four single-engined British jets were developed, each with a 5,000 lb. Nene buried in the fuselage and straight wings—a combination that meant about 600 m.p.h. maximum speed, or just about the same result as the Shooting Star.

The Supermarine Attacker was simply a Spiteful wing to which a new fuselage was wedded. The wing was continuous across the bottom of the fuselage and the pilot was put into a streamlined nacelle in front of the fat main fuselage with its large air intakes. Piston-engined origin was betrayed by the tail-wheel undercarriage. The downward inclination of the jet, about 10 deg., proved troublesome and to avoid grass and tarmac conflagrations the lower edge of the orifice was extended to act as a deflector. The Attacker was taken over by the Royal Navy, which called for so many modifications that it did not go into service until eight years after the design was started. Glosters made the E1/44 on a similar theme, or one might almost call it a fattened operational version of the E28/39, but it was not successful.

Hawkers then followed with a completely fresh design, the P-1040, which looked more like a racer than a fighter its lines were so sleek. Advantage of the jet was taken to get the pilot well forward and to put the guns under the nose. Already during the War trouble had been encountered with wing guns going out of alignment during tight, high g turns and the Attacker had met it in an acute form. So in the P-1040 (now known as the Sea Hawk naval fighter) the lesser evil of guns under the nose was chosen. The method of handling the jet problem was most unusual: the engine was put into the middle of the fuselage and air was ducted into it through triangular openings in the wing root and then the jet pipe was divided to exhaust at the trailing-edge roots. Surprisingly, this division of thrust causes negligible loss and the pipe does not overheat. Advantages of the layout include a short, efficient jet pipe, room in the rear fuselage for a comfortable fuel tank, and a much simpler tail structure without tricky frames round the jet pipe.

The de Havilland Company in 1949 brought out a redesigned version of the Vampire, the Venom. Using a Vampire fuselage and tail with small modifications a 5,000 lb. thrust Ghost engine was fitted and a new thin-section wing was designed . This new wing also had all the taper on the leading edge, so giving a modicum of sweepback. Unfortunately, making the wing thinner over all meant introducing a number of blisters for the wheels and other items, which led to rather poor behaviour at high Mach numbers and high altitude. As more powerful engines became available in the U.S.A., the Americans built a number of single-engined jet fighters. Several of these were designed along lines that were to lead nowhere, but two were of particular interest, the Republic F-84 Thunderjet and the F-86 Sabre.

The Thunderj et was again one of those simple, well-proportioned aeroplanes and was very like a slender, scaled-up version of the original E28/39. The fuselage was long and slender because it housed an axial engine, the G.E.C. J-35, and there was no fancy treatment of the propelling air—it was scooped

A nose intake is better suited to the airflow into an axial engine than it is to that into the centrifugal, particularly the Rolls-Royce types requiring a plenum chamber. The reason for this is that since an axial impeller achieves compression by accelerating air, the flow into it is faster than the forward speed of the aeroplane and it is always sucking. On the other hand, a centrifugal impeller compresses the air as it finds it, so to speak, and at over 500 m.p.h. the ram effect of forward speed actually "supercharges" the engine. To get the best results from ram it is essential to have a short intake duct, since skin friction and turbulence soon absorb much of the energy. With the axial engine sucking air through the duct there is no turbulence and therefore little loss. Naturally, at very high speeds, ram air may be travelling faster than that in the compressor and then the axial will benefit just as much by a short duct as the centrifugal.

As in the Shooting Star, the power plant was bolted to a frame attached to the rear spar of the wing. To reach the whole engine for maintenance or removal the entire rear fuselage and tail unit was unbolted and slid away on a trolley. Special quickrelease bolts and control connexions enabled removal and attachment to be done rapidly.

The F-84 had its wing raised a little above the bottom of the almost circular fuselage so as to improve the juncture and avoid the use of fillets. The tail too, is raised above the fuselage. The reason for this is that, at high speed, the wing leaves a turbulent wake and the tail plane needs to be mounted above or below it to avoid trouble. The low position is usually avoided because of the greater risk of damage.

The F-84, like the Shooting Star, is habitually flown with wingtip tanks. American designers seem to have taken a liking to these appendages and it is true that they have certain virtues. If of good shape the drag is probably little more than if the fuselage were made that degree fatter to house the fuel. The tanks act as "end-plates" on the wing, so containing the air flow at the stall, thereby reducing the tendency for a wing to drop and improving control. There is, however, no truth whatsoever in the American publicity story that the curved sides of tank and fuselage act like a venturi and accelerate the air so that speed is increased.

The Thunder jet has proved to be a sound aeroplane with the all-round performance to be expected with a 5,000 lb.

thrust engine and straight wings—that is to say a maximum speed of 600 m.p.h., or a little more, and a ceiling of about 45,000 ft. However, when, in 1950, a swept-wing version was flown with the British Armstrong Siddeley Sapphire of nearly double the original power the improvement was phenomenal.

It should be explained here that the "maximum speed" of straight-winged jet fighters is usually the Mach number at which compressibility effects become very violent and that if it were not for their rendering control impossible higher speeds could be obtained. The nose-down change of trim described earlier is usually followed by snaking, then violent lateral oscillations with, in some cases, a wing stalling completely and the aeroplane flicking over on to its back. The snaking is an uncontrollable waving motion of the nose caused by breakdown of the air flow over the fin and rudder and is one reason for the larger tails now fitted to jets in comparison with the early ones, in which the absence of torque led to designers making optimistic reductions of tail areas.

Fitting the swept wing to the Thunderjet F-84F removed all speed restrictions, as well as lowering the drag, and it became a very fast aeroplane which is now in full production with the Sapphire engine, made under licence as the Wright J-65. Two versions have been built, for fighting and for reconnaissance. The first has the original fuselage, apart from a larger intake duct and jet pipe to which the new swept wing and tail are attached, while the other has the air intakes moved to a forward extension of the wing roots so as to leave the long nose clear for the mounting of cameras. Later, to improve interchangability , both types will have wing root intakes—which will also ensure the assistance of ram pressure at top speed.

Before the advent of the F-84F there was a most curious experimental development of the Thunderjet known as the XF-91. This freak proved to be very fast and was the first service type to exceed the speed of sound in level flight. Since it had a J-35 engine boosted by both water injection and afterburning and with rockets to help it through the barriers' drag rise this is not exactly surprising—though one is left wondering what space there was left for guns! The XF-91 has an extraordinary swept wing which tapers in plan and thickness toward the roots. The idea is obviously to improve the lift at the tips and so make landing easier, but it also concentrates all the wing loads into a very thin root structure. The Thunderjet undercarriage was mounted anyway almost at mid-span, so it was fairly simple to turn it round and, by using extra thin wheels, retract it into the wingtips. The swept fin and parallel-chord, all-moving tail of the XF-91 were adopted for the F-84F.

On October 1st, 1947, the Americans flew the first fighter incorporating design features culled directly from the Germans, the fully swept North American F-86 Sabre. This was a heavy aeroplane, with a large all-embracing fuselage longer than the span, thin swept wings fitted with slats, and a thin swept tail. Despite the thinness of the wing, the designer managed to squeeze three self-sealing bag tanks into the torsion-box portion to supplement two larger tanks in the fuselage. The latter was required to carry the nose duct, pressure cockpit box, six .5 in. guns, ammunition boxes and empty case and link box, so the reason for the size of the fuselage is obvious. Experience with 600 m.p.h. fighters had shown that, while empty brass cartridge cases were blown away without difficulty the steel belt-links tended to drag along the fuselage and damage it. So valuable space has to be devoted to this enforced salvage in the latest jet fighters.

Much of the Sabre's performance, it was the first service aeroplane to exceed Mach 1 in a dive, comes from the unobstructed run of the intake duct, which passes, undivided, beneath the cockpit. The little snout and dark nose ring often excite comment. The former contains the scanner for the radar gunsight and also serves to prevent air spilling when climbing steeply. The latter was originally a non-conducting plastic part for the benefit of the gunsight, but erosion was so bad that it is now made from metal.

The use of fully automatic slats on a high-speed aeroplane perpetuates the ideas of Willy Messerschmitt. It is an effective way of overcoming tip stall and has proved a boon on active service in Korea. Another innovation on the Sabre at the time of its first flight was the "flying tail"—now considered to be indispensable supersonic equipment.

When the airflow starts to become rough in the sonic zone, at about Mach .85, the control surfaces lose effectiveness and the elevators, vitally necessary to overcome the pitching moment changes, have to be moved as much as 20 deg. in order to obtain a result that one degree of tail plane movement could achieve. From this knowledge it was a logical step to arrange that the control column should, after initially moving the elevators, alter the tail plane incidence as well. Since, on the Sabre, ailerons and elevators are irreversibly power operated in any case as a precaution against the high loads of sonic flight and flutter it was not too difficult to arrange for the stick to operate the tail for manoeuvring, allowing the trim motor to actuate it as a source of reserve power.

Fitting extra tanks to a swept wing posed a pretty problem. If they were mounted on the tips they were so far aft of the c.g. that they made the tail light when the fuel was exhausted. Incidentally, it is not often realized that the tips of a highly swept wing help to balance the aeroplane so that the tail plane is much more a control surface than a balancing one—one of the reasons why there have been more tailless designs since the War than ever before. On the F-84F the unusually wide-track undercarriage actually allows tanks to be fitted under the wing roots, but on the F-86 the only solution is to mount them close outboard of the wheels. Even here the fuel is kept as far forward as possible. The curious flat-topped shape of the tank, the nose-down carriage and the tail fins are all designed to throw it clear on release. On some American Sabres the tanks have probes for picking up fuel from aerial tankers by Sir Alan Cobham's method of Flight Refuelling, which has been enthusiastically adopted by both the U.S.A.F. and the U.S.N.—, and as equally neglected by the R.A.F. and Royal Navy.

Apart from some unusual tailless fighters, which are dealt with in Chapter XIII, the Americans had two useful thoughts in naval fighters. With its young admirals (all bound to learn to fly) supported by the success of their carrier actions in the Pacific, the U.S. Navy naturally embraced the new form of propulsion with enthusiasm albeit somewhat daunted by the fuel consumption and the difficulties of combining range with small aeroplanes. It was this naval interest that led to the McDonnell F2H Banshee and the Grumman F9F being designed , two fighters that are rather similar in appearance, though the first has two and the second a single engine.

Taking advantage of the lack of airscrew, McDonnell's designer packed two small-diameter axial turbo-jets into a thickened large-chord centre-section. This had the considerable virtue of making it easy to maintain course on one engine. In size, weight, wing-loading and power the Banshee approximates to the Meteor, but its performance is slightly higher all round. The Panther, although it looks generally similar to the Banshee, is a neatly conceived short-tailpipe installation of a centrifugal engine, actually the American-licensed Tay, the J-48. Ram intakes in an enlarged wingroot are used to advantage . An innovation was the use of leading-edge flaps to increase the camber of the thin laminar-flow wing for landing—yet another British idea dating from about 1920, examined interestedly by the Germans in 1943-44, and first used in America. The Panther is reputedly pleasant to fly, has been successful against the MiG-15 in Korea and is fast for a straight-winged aeroplane, at 625 m.p.h. Its centrifugal engine gives a handsome ceiling of over 50,000 ft. and an initial climb of 9,000 ft./min. A version with swept wing and tail, the F9F-6 Cougar is now flying and it ought to make an interesting comparison with the MiG-15, since the J-48 and the Russian engine are virtually identical.

After making a couple of undistinguished straight-winged fighters, the MiG-9 and the Yak 15, the Russians came away with a rush when they designed the MiG-15 and the rather similar La-17. It is true that without German help (forced) in the design and British assistance (unwitting) with the power plant the first truly successful operational jet fighter might never have existed—but it does and it has been built in greater numbers than have any of our Allied fighters.

The secret of the MiG-15 lies in the strictest weight economy so that the combat weight is only 11,250 lb., giving a wing loading of 44 lb./sq. ft.—approximately the same as the Vampire. To keep down weight, high-strength steel is used in the wing and tail, thereby maintaining strength in the thin structures, which are only 11 and 8 per cent thick respectively. Equipment is kept to the minimum, but not at the expense of operational efficiency. The armament of one 37 mm. and two 23 mm. cannon is harder hitting than the Sabre's .5 in. guns. Aerodynamic design is very German and is certainly a stage between the "conventional" jet fighter and the exaggerated Ta 183. Furthermore, reports came in 1952 of a fighter that is almost identical with the Focke-Wulf project—which looks like further confirmation of the origin of the MiG-15.

Two MiG-15 features are common today on swept aeroplanes . Instead of the wings being set at a dihedral angle, they have a negative anhedral. The reason for this is an unpleasant phenomenon known, rather expressively, as Dutch roll. As the swept aeroplane approaches to land and its speed falls the ailerons begin to lose effectiveness so that when the rudder is used, or something else causes a yaw, the forward-moving wing presents a larger lifting surface and therefore rises. With sluggish aileron response, the result is an unpleasant wallowing motion something like a small boat in a following sea. By giving the wing anhedral and deliberately reducing lateral stability the rolling is damped. On each wing there are two fore-andaft plates, known as boundary-layer, or air-flow, fences. These restrict the outward drift of air at low speed that is one of the causes of tip stalling in swept wings.

The French have contributed little to the development of the jet fighter. There have been a few types since the War, but only the Marcel Dassault M.D. 450 series is really worthy of serious consideration. Even so, it was as an application of the "pure" E28/39 formula that this aeroplane succeeded. Design was started in December 1947 and the first flight was on February 28, 1949, so that the results of other people's experience and (troubles) were beginning to become available. The M.D. 452, known in the Armie de I'Air as the Ouragan (Hurricane ), can be classed with the Thunderjet, Sabre and MiG-15.

Although the Ouragan had straight wings, the design target was really the fully swept M.D. 452 Mystire and, accordingly, a swept tail was fitted. With a Hispano-Suiza Nene of 5,000 lb. static thrust the Ouragan does 600 m.p.h. at sea level and can climb initially at 8,450 ft./min., which proves that it is good in its class. Like the MiG-15, the Ouragan has a low wing loading, under 45 lb./sq. ft.

Performance of the swept Mystire is secret, but it has either the 6,280 lb. Hispano-Suiza Tay or a 6,500 lb. S.N.E.C.M.A. Atar and is known to have dived supersonically. The maximum level speed is probably comparable with the very similar MiG-15, Sabre and F-84F, that is rather over 670 m.p.h. Returning to Britain's contribution in jet intercepter design; this reached its climax in two series evolved by those old rivals Hawker and Supermarine.

The P-1040 was modernized by fitting it with outer planes having 35 deg. sweep and this aeroplane, the P-1052, flew on November 19, 1948. The P-1052 behaved well and showed that given more power it could travel faster. In order to allow the fitting of an afterburner, the divided jet pipe was abandoned

and it was given a new rear end with swept surfaces—this redesign was called the P-1081. In the meantime a completely new fighter, to the latest R.A.F. intercepter requirements, was being designed and built. This was the P-1067, now called the Hunter, first British single-engined supersonic fighter.

The Hunter has a small span, but the aspect ratio is low and the wing loading is not likely to be high—certainly not after using fuel at full power, with afterburning, to reach 50,000 ft. This is one of the first British fighters designed for an axial-flow engine (the Rolls-Royce Avon in the Mark 1, the Armstrong Siddeley Sapphire in the Mark 2) and the effect can be seen in the fuselage thinness. The various little openings in the fuselage are for auxiliary cooling air. The cockpit is well forward, as on the earlier types, but is less prominent. Under it are mounted four 30 mm. cannon, apparently without empty link collectors. The tail is of simple high-speed geometry with power operation and a fully controllable tail plane. A small point to note is the "bullet" fairing behind the tail, the addition of this suppressed all compressibility vibration and enabled its test pilot, Neville Duke, to dive "through the barrier"— divergence of the various surfaces had caused a reduction of pressure and turbulence. It was as a result of experience with the P-1081 that the jet pipe fairing was taken well aft of the control surfaces. The induced air flow caused by the jet interfered with the controls. The wing of the Hunter is thin and smooth and the intakes, made by parting the top from the bottom surface, are retained. These intakes are effective and are only slightly affected by the fuselage boundary layer. Note the little flat duct scooping up the sluggish air and spilling it over the wing—this is a different way of doing the same job as the D.H. "shoulders". The "spine" running along the top of the fuselage prevents turbulent air shed by the hood from disturbing the airflow over the fin.

The way in which the Attacker grew into the Type 510, Type 535, Type 541 and, finally, the Swift, was an even more direct development. The 510 was an unarmed Attacker fuselage —retaining even the latter's tail-wheel undercarriage—with swept wing and tail surfaces. This aeroplane was very fast, about 650 m.p.h. with only the 5,000 lb. thrust Nene, but because the MoS had stopped all development of centrifugals there were none of the more powerful Tays—although, ironically , the licensees in France and America were making them in quantity. The 510 was flown intensively to provide experience in the behaviour of swept wings and numerous lessons

were learned. For instance, experience showed that if the split flaps are made strong enough and open so that there is a slight gap between their leading edges and the wing, they can be used as air brakes—so eliminating the weight and complication of fuselage "doors"- or other separate devices. The 510 was converted, by adding a nose long enough to contain cannon and a nose wheel, into the Type 535 intercepter. The first conversion was actually the identical airframe, which retained its tail wheel and even the same machine number! Into one of these developed airframes, in 1951, went an Avon engine and so the Type 541 was born and it became, after detail refinement, the Swift, first native swept fighter to reach R.A.F. Squadrons. In the Swift, the designers have made no attempt to "tailor" the fuselage to the slim Avon, but have used all surplus space for fuel tanks, so that the range is good. The "elephant's ear" intakes have been troublesome and in the latest version double boundary-layer bleeds are necessary. The rear fuselage has been fattened to take an afterburner. The very low wing and deep fuselage combination make it unnecessary to put the tail plane on the fin, a little dihedral is all that is needed to clear the wing wake.

The present jet intercepter is about half as heavy again as the Spitfire, mainly because of its huge fuel load, costs thousands of manhours and something like £100,000 each. Even so, it is largely a day fighter and has to be supplemented by even larger, and more expensive all-weather fighters. Several individuals, including W. E. W. Petter, designer of the Canberra, consider that even with today's prices and wages an effective fighter could be made for £20,000. Such a design would have about the size, weight and armament of the He 162, with a modern 4,000 lb.-thrust engine weighing about 900 lb.—with 50 per cent extra afterburning power the thrust would equal the weight and give phenomenal climb. With all the modern ideas on sweepback, streamlining and structure built in simply it ought not to be too difficult to make. It would, at any rate, make available, quickly, a large number of sonic fighters to counter mass daylight raids—which can only occur in fair weather.

The Mosquito night fighters were for some years considered adequate to deal with any possible night attacks by the TU-4 —the Russian copy of the B-29, but they would obviously require eventual replacement. At first the ranks of the R.A.F. were filled by adaptations of standard jet fighters like the Vampire and Meteor and those of the U.S.A.F. by the Shooting

Star conversion, the F-94. In each of these cases the first step was to evolve a two-seater trainer, in the Meteor and Shooting Star by lengthening the nose and in the Vampire by widening it. Having achieved two-seater aeroplanes, the next step was to load half a ton of radar into the nose, not only A.I. radar but all-weather landing aids too. Then, to increase endurance, the poor things were hung all over with extra fuel tanks. It is worth noting that similar stop-gap treatment was meted out to the MiG-15 pending the development of true all-weather patrol fighters.

Latest ideas on this type of aeroplane are represented by the Gloster Javelin and the D.H.11o, built to the same specification . Here the problem facing the designers was one of reconciling the most conflicting requirements: initially radar, armament, fuel and crew made a heavy load; a design speed approaching Mach 1 meant high power, two engines, more fuel; the aeroplane was now heavy and, to keep down its wing loading it had to be made very large; high speed meant sharp sweep and low thickness/chord ratio. Altogether a very formidable problem, so that it is scarcely surprising that the result was a pair of unusual aeroplanes.

De Havilland chose a twin-boom layout that might, at first, be considered a logical development from the Vampire and Venom. In fact, it was based much more on experience with the D.H. 108, a tailless research monoplane of 1946 (which, flown by John Deny, had dived past Mach 1) than upon the fighters. The highly swept wing was similar in plan form to that of the D.H. 108, but to it, instead of a central fin and rudder, were added short booms that were swept up to form twin fins and rudders. Between the fins was mounted a "flying tail" to cater for sonic-zone trim changes and to allow landing flaps to be used. The nacelle, with better (and bigger) radar was large (the pilot sitting in a blister, the observer in the main structure) and it was larger still where it blended with the wing roots to make the twin nacelle for the two Avons. In size the D.H. 110 is larger than the Blenheim light bomber, yet its cleanness and high power enable it to dive supersonically. The wing loading is low, so low that the aeroplane takes off in half the distance of the swept interceptors, and it is only by this strict adherence to low wing loading that good manoeuvrability can be ensured at 50,000 ft.

Gloster's approach to the all-weather specification was a completely fresh one, a delta wing, but with a tail. The virtues and vices peculiar to the delta are described in Chapter XIII, and here it suffices to say that the filling in of the trailing edge gives greater area for a given sweep and makes possible a low thickness/chord ratio, while the retention of a small tail allows the use of flaps. Like the D.H. 11o, the Javelin is as large as a light bomber of the last War, but such is the penalty for the load to be carried. In the delta form, with the bulky Sapphire engine nacelles, surface area—wetted area to the aerodynamicist—is extensive and cannot be avoided, the best solution is to suppress all possible bumps and excrescences, relying on the purest possible envelope to keep down drag. In the Javelin the radar operator is placed behind the pilot so that one long slender canopy does for both of them. The large delta wing is, of course, a wonderful natural fuel tank and is actually about three feet thick at the root.

Both the D.H. 11o and the Javelin, despite their size are capable of sea level speeds in excess of 700 m.p.h.