Take-off on a narrow runway was difficult. There were one or two characteristics to bear in mind. The propeller went round in a clockwise direction viewed from the cockpit. The resultant twist imparted to its slipstream also imparted a sidethrust to the upper part of the fin and rudder which turned the aircraft strongly to port — needing full and early right rudder to hold the aircraft straight. However, because of the poor view over the nose, the pilot would often be late in detecting a swing to port and would be unable to correct it once it had started, as anyone who has done a tail skid in a car will know.
However, the tail could be raised by the pilot if he shoved the stick fully forward (nose down), and he could then see over the nose and where he was going. The rudder then became effective in a straighter, faster, part of the slipstream.
The Seafire flew off the ground at about 70 knots, or it could be pulled off the ground at about 58 knots if ham use were made of the elevator.
The undercarriage was raised hydraulically. The lever which selected the position of the hydraulic valve to ‘up’ also rotated a mechanical lock to each undercarriage leg. The lock itself could not easily be rotated in its machined socket unless the weight of the undercarriage legs was held off by a momentary ‘up’ selection. This was applied automatically when selecting undercarriage ‘up’, but when selecting ‘down’ pilots would often forget. The ‘down’ selection could not then be made because the mechanical lock could not be rotated one way or the other. The leg stuck fast, with the pilot unable to move the lever. After a few aerobatics it might move, but when it did not do so, a wheels-up landing ensued. Other difficulties could also occur with the ‘up’ locks in the wheel-well fairings failing to catch. The undercarriage could then droop in flight and restrict airflow into the radiators. The engine would then overheat, necessitating a forced landing in some circumstances. One trip in ten was usually carried out with the pilot’s undercarriage warning lights on in the cockpit, denoting some fault with the adjustment of the mechanism.
Landing a Spitfire/Seafire on runways as narrow as those at Henstridge was very tricky in crosswinds. On grass airfields, the landing direction could be made so that it was, like a bird landing on a lake, dead into wind. The longest runways at Henstridge were laid in the prevailing wind direction and would usually allow a pilot sufficient stopping distance for a faster than normal crosswind landing. However, when the crosswinds came from an unaccustomed wind direction and a short runway had to be used, the skid or offset approach was necessarily made at a much slower airspeed than Pilot Officer Jack would have allowed. This often resulted in heavy landings with drift, landings on one wheel or crash landings in the undershoot area. In such cases the Seafire’s undercarriage was neither wide enough apart nor strong enough to take the sideloads which occurred. The aircraft would then drift off to one side and tip on its nose in the soft clay either side of the runway, or groundloop. With the Hurricane, the wheels were sufficiently wide apart to correct the skid by differential use of wheel brakes. Provided that they did not fade due to overheating, the aircraft could then be kept straight. With the Seafire, the brakes were insufficiently strong (designed thus to prevent the aircraft nosing over) and the wheels insufficiently wide apart, to allow differential braking effect.
Finally, there was the problem of the lack of view over the nose on the approach to land and during the completion of the landing run. This, and decklanding, was by far the most difficult part of flying the Seafire. To see the runway, it was necessary to come in on a steady, curved, approach path. Any further distractions, such as crosswind, even a dirty windscreen or a rough running engine, could so overload the human computer, that a pilot would make late corrections, and over-corrections. With the RAF’s far longer and wider runways, its pilots came in fast so that they could easily see over the nose. They made ‘wheelers’, holding the tail up high until the aircraft was running straight and true and safely on the runway. This was commonsense and to be applauded. However, the RN needed to train its pilots for the decklanding approach. This was at 1.05 x Vse, or at a speed only five per cent above the ‘engine on’ stall speed. The aircraft was then in virtually the three point attitude and the view over the nose was nil. Fleet Air Arm pilots accepted this as a normal challenge.
My PGI’s course with the RAF at Sutton Bridge started on 1 October 1943 and lasted for a month and 30 flying hours. The RAF turned the art of deflection shooting into a theoretical science. Our cameragun results were confirmed — or not — by using specially point-harmonised guns, in fours and eights. No expense was spared. Towards the end of the shooting period, I managed to get 1600 hits on a 20 foot drogue target. This was 20% of the bullets fired. Most fighters in WW II — with wing-mounted guns — used pattern harmonisation for their guns. That is, if each barrel was looked through in turn, the eight or so, guns would hit a screen about 250 yards ahead in the form of a pattern and not at a single point. The pattern was arranged to allow for wing flexing, ‘G’ drop, gun chatter and variations in incidence and yaw, so that the bullets had a maximum chance of hitting the target with 10% of the bullets fired, at ranges from 50 to 250 yards range for 0.303 ammunition, and up to 400 yards with 20mm cannon shells.
In some German and Italian fighters, the guns fired through the propeller or through the centre of the propeller shaft and along the pilot’s sight line. Point harmonisation was then a feasible proposition, less guns were needed to secure a given percentage of hits and the wings being much lighter would allow a crisper rate of roll. However, the problem of shooting through the propeller tended to slow down the firing rate, once this became a factor of high speed combat where ‘snapshooting’ was necessary to achieve a kill, and most single-engined fighters eventually used guns firing from the wings.
The RAF Course was for instructors at RAF Fighter Schools, so that I was lucky. They even allowed us pupils to take one of their Spitfire Vs back to our normal bases for each weekend. The RAF’s Mark of Spitfire used for training, the hooked version of the Spitfire Vb, was only just in front-line Fleet Air Arm squadrons.
Although the object of the Course was to give confidence to us instructors in the theory of air-to-air gunnery, one of the attractive and practically useful features of the PGI Course was that we were taught to make attacks from all angles, from above and below, as well as from each side. The ‘above and below’ attacks were called ‘Upward and Downward Jesus’ attacks. The one from below involved a loop and a roll-off-the-top while firing. The one from vertically above was, however, much easier, just a rolling pullout from a vertical dive.
Back at Henstridge I became the Station PGI. I fixed up a ‘private’ Miles Master II with all manner of cameras and gunsights, and took petrified pupils in the back to show them how upward and downward ‘Jesi’ should be done. Not only that, I was sent all over the place in England and Scotland as a travelling salesman, seeing very little of my immobile Wren as a result.
Many of the pupils joining the FAA at this time in the war found it difficult to make quarter attacks. It over-loaded the human computer to have to judge distances, angles, approach speeds, convergences, line of target flight and deflection when shooting — and flying at the same time. The first problem to solve was how to position the fighter in the right part of the sky before starting an attack. Unless a pupil pilot could do this, it was a waste of flying hours and highly dangerous to allow him to continue. It was frightening to act as a target for those who did not possess this form of judgement. For those who have not tried it, it is somewhat similar to driving a car at high speed into a crowded traffic roundabout and judging it so that no collisions occur. The difference being, that in an aircraft, the meeting takes place in three dimensions and at ten times the speed.
There were, however, two or three pupil pilots at Henstridge who were ‘naturals’. Among them were Sub-Lieutenant R. H. (Dick) Reynolds and Sub-Lieutenant Victor Lowden, both of whom met and entirely vanquished the ‘Zero’ in the Pacific war, using the ‘attacks from all angles’ techniques. (See Appendix 11 (j) — The Seafire in Combat.)
One of the PGI’s ‘perks’ at Henstridge was being first in the queue for ‘special’ aircraft. The storage section possessed a beautiful hybrid Seafire Ib — NX910 — which they had fitted with the new Merlin 55M engine. It was flush-riveted throughout and the storage section polished it so that it shone like a Greek Admiral. The engine developed 1650 horsepower at take-off and only had to cope with a very light fuselage and a cropped supercharger. It could maintain the full supercharger pressure of 16 pounds per square inch manifold pressure up to 8000 feet and shot up to this height in less than two minutes at a climb angle of 45 degrees and could maintain a true airspeed of 370 mph. I looked after it better than I did the Wolsey Hornet and there was a queue to fly it. What a pity that Tim Singleton hit a crash tender with it a few days before his wedding. He was thinking of other things, he said.
Luckily, the laws of physics dictate that lift on an aerofoil, at constant incidence and ‘efficiency’, is proportional to the square of the windspeed flowing past it — otherwise we might have swans taking off at 400 mph instead of 20 mph.
The speed range of a swan is from about 20-30 mph. His — or her — ‘centre of lift’ can be kept over his small changes in CG by sweeping his wings back a few inches at the wing tips when his CG is aft, and forward a few inches when he had just eaten and his CG is forward. He can even alter the thickness/chord ratio of his wings to suit high speed or low speed flight, and so retain a much higher degree of ‘wing efficiency’ than is possible with man-made wings. He can use his tail to correct temporary changes in fore-and-aft trim during flight, and use assymetric wing twist instead of ailerons.
The swan’s changes in maximum lift between 20 mph and 30 mph would roughly be in the ratio of 1:2. The swan merely changes his wing incidence to give a smaller incidence at the higher speed if he does not want to climb.
In the case of a Seafire, its speed changes from 60-360 knots IAS. This would give a very large theoretical lift-increase ratio of 1:36. Assuming that a mere man-made wing can only maintain a ‘wing efficiency’ of some 33% at the high speed end of the range, this would still mean that the Seafire could lift 12 times its own weight, or suffer that amount of ‘g’ without stalling on a sudden pullout of a dive, if it was strong enough to stay in one piece. In fact, the Seafire began to ‘notch’ its wingfold bolts well before 12 ‘g’. It was more like eight ‘g’ that the structure began to fail. As this loading can easily be imposed at any TAS above 300 knots without stalling, and as the stick force per ‘g’ in the Seafire was only two to three pounds per ‘g’, the pilot could easily break the aircraft. In practice, the onset of ‘g’ blindness — ‘greying out’ — usually warned him, and he would relax his pull force. Nevertheless, as we will see, the Seafire pilot was sometimes unable to react quickly enough and when the Franks ‘anti-g’ suit was worn, even this ‘seat-of-the-pants’ warning was not available. (Stick force per ‘g’ in bombers and passenger aircraft was between 10-15 pounds per ‘g’, so that very heavy pull forces usually prevented excessive ‘g’ being applied by the pilot.)
During the acceleration process after take-off from, say 60 knots – 250 knots, wing incidence has to be ‘wound off’ by rotating the whole aircraft through about ten degrees nose-down, by pushing forward on the stick. Retrimming the small tab on the trailing edge of the elevator by a handwheel in the cockpit is necessary, so that the pushforce is ‘trimmed out’. (Elevators, ailerons and rudders, plus their control circuits are, of course, mass balanced, so that ‘g’ applied to the aircraft does not affect them much.) Retrimming cannot be carried out easily in the heat of combat so that the designer of a fighter — as opposed to a bomber or passenger aircraft — designs his fighter so that elevator trim changes are very small or non-existent. This calls for careful balance between the centre of lift of the mainplane and the CG of the aircraft — like some very sensitive see-saw. Larger movements of the CG than, say, a couple of inches on a 3½ ton aircraft, cannot be allowed either way. Besides these small permitted movements of the CG itself made necessary by the carriage of stores, consumption of fuel in flight etc, the centre-of-lift position also alters during flight as IAS alters. Mitchell knew this and designed the lift characteristics of his tailplane to compensate automatically, not just during speed changes, but with ‘g’ changes, with flaps or undercarriage up or down and with engine power on or off. The original Spitfire pilot hardly had to touch his trimmer wheel from takeoff to landing and his elevator stick forces were never more than a few pounds. In spite of this almost neutral ‘stick-free’ stability and because it also had almost frictionless, well-rigged controls, the aircraft would recover on its own to straight and level flight if for any reason the pilot blacked-out or became unconscious temporarily and let go the stick.
However, such perfection did not last long. Changes in propeller weight, supercharger weight, extra fuel tanks, and various other weights added aft of the CG in the fuselage, plus the doubling of the engine power, soon upset this careful balance. When the Navy finally completed the job by adding an arrester hook some 15 feet aft of the CG and were unable to compensate for it by sufficient (more than 28 pounds) lead weight added to the engine bearers, the balance was seriously upset. The balance had to be restored by ‘fudging’. This was done by altering the original Spitfire elevator to include a device known as an ‘unshielded horn balance’, and a further clever device to the elevator control circuit which held the stick forward permanently to compensate for the tail heaviness. (Photographs of Seafires in flight will show this.)
How did they do this? Not by elastic, for the push force from this would not increase as necessary when ‘g’ was applied, and the aircraft would tighten in a steep turn or pullout of a dive in a very dangerous way. The designers therefore added a ‘positive weight’ to the elevator control so that, as ‘g’ was applied, it would help push the control column more and more forward. In the Seafire, they added a three pound lead weight to the bottom of the pilots’ control column. The weight was at the end of a six inch steel shaft, the other end of which was butt-welded at right angles to the lower end of the control column just above its bottom hinge. If any of us had cared to look down to see what the stick was actually doing as we pulled back on it out of a dive, we should have seen it move forward as ‘g’ was applied.
The designers thought they had cured the problem of the Seafire’s ‘stick-free’ instability. However, Mitchell’s original balance required such meticulous airframe manufacturing weight tolerances that some turned out to be heavier than others in the aft fuselage, and this put the CG outside the authority of the ‘positive weight’. Many of the flight conditions regularly used by 30 Wing in the Pacific and elsewhere could not have been test flown and the fatal effects were not discovered and remedied before the squadrons received these dangerous aircraft.
In 30 and 3 Wings, alone, there were four instances of wings shearing off — all put down to pilot carelessness or flak — and at least four near-misses with Seafires returning with ‘notched’ wingfold bolts, just about to shear off. Cunliffe-Owen manufactured 80 Seafires with flush-riveting in the aft fuselage, ten per cent of the total number built. All had CGs dangerously aft of the design position due to the use of thicker gauge skinning, and were prone to ‘self-tighten’ in pullouts or steep turns. Some Seafires carried F24 cameras weighing 40 pounds each. We were forbidden to pull tight turns when these were carried, a most inconvenient restriction for a fighter over an enemy airfield. However, even the bracing for the cameras made the aircraft dangerously tail heavy.
The steel shaft holding the positive weight was insufficiently strong. My rigger discovered the weight lying loose in my aircraft after the Maizuru raid. We had no idea how important it was, but luckily the workshop welded it back before the next flight, “otherwise the PO would not sign for it”. The onset of ‘g’ from self-tightening was very sudden and it had to be anticipated if there was to be much chance of avoiding its consequences. In the heat of action, pilots often forgot.
I discovered the reason for this sudden and catastrophic onset of ‘g’ after the war while I was at the Empire Test Pilots’ School. We had found in 880 and 801 Squadrons that the onset of ‘g’ was particularly sudden when we pulled out of steep dives — as against shipping. (The steeper the dive, the better the anti-flak protection.) I also remembered that the positive weight was welded at right angles to the stick, so that, during a ‘vertical dive’ — usually called ‘vertical’, but never more than 75 degrees — it would have little or no effect. However, the thought also struck me that with the aircraft approaching the earth at about 1g acceleration between 300-400 knots IAS, it would be ‘weightless’ in any case, so that with no ‘g’ in any direction, the positive weight could not function and its stabilising effects would be zero. This would only be transitory, but long enough to cause an initial snatch of very high ‘g’ at the moment of the commencement of pull out, sufficient to break the aircraft before the weight reimposed its stabilising effects and pushed the stick forward.
To make matters even worse, the makers had included in the elevator circuit another well known device. To restore the tendency for the stick to ‘droop’ forward under the influence of the weight, they had fitted a ‘negative spring’ to compensate. This amounted to a pull force of about three pounds, high by Spitfire standards, which, under ‘weightless’ conditions, would have still exerted itself. Unless the pilot made a conscious effort to trim this force out in the dive — and there was no time to do it — it would superimpose its three pounds on the pilot’s, and unexpectedly add this force to his.
Once in the dive, — which because of the cocked-up sightline, would be in a slight ‘bunt’ — concentrating on aim, the shattering vibration from the guns, the airspeed rising to 360 knots and more, the ground coming up fast with the shells yet to strike the ground to give the pilot aim correction — he would leave the pull out to the last moment and tend to grab the stick at pull out. There would be a sudden nose-up pitch from the ‘unstable’ aircraft — the spring-overload added — and the wing bolts would shear before the ‘positive weight’ could reassert its authority, and before it could help the now semi-concious pilot to reverse his pull force.
Why was this not discovered? The reason was probably because the test pilots of the day used to ‘trim into the dive’ when checking ‘stick force per ‘g’ ’, and, when carrying out their dives, they would seldom check ‘out-of-trim’ stick forces in very steep dives or carry out ‘stick-free’ pull-outs from these dives.
Later Marks of Firefly also had a ‘positive weight’. The makers did not make the same mistake as in the Seafire III, for they cocked the weight up on its shaft through about 45 degrees, so that, providing there was any positive ‘g’ in the dive, it would continue to help. Neither did they use a ‘negative spring’. They accepted the inconvenience of the stick falling forward when the aircraft was at rest on the ground.
Another fatal consequence of the Seafire’s instability (not in the early Spitfire of course, which was perfect) was the effect it had on the pilot’s ability to bale out from the recommended inverted position. In the case of the Seafire, the ‘positive weight’ acted in the reverse sense (assisted by the ‘negative spring’) when flying the aircraft with negative ‘g’ and with the pilot ‘hanging on his straps’. This aspect could only be partially discovered in our ‘practice’ bale-outs which we carried out — without actually releasing our straps of course. We found that the Seafire’s nose had to be trimmed fully nose heavy before inverting the aircraft, to overcome the negative spring (now ‘assisted’ by the positive weight which had become ‘negative’). This was essential to prevent the aircraft from completing the second half of a loop and so trapping the pilot in the aircraft with centripetal force — as in a roller coaster — and diving into the ground.
However, when Seafire pilots wanted to bale out in the Pacific, in most cases their engines had already stopped. In any case, in a real-life situation, they would have throttled back to reduce airspeed, so making it easier to fall out without risk of blowing back into the tail in the propeller slipstream. At this low speed and with the engine failed and ‘windmilling’, the nose trimmer was nowhere near powerful enough to hold the aircraft in level, inverted flight. Directly the pilot had pulled back or jettisoned the hood, trimmed fully forward, inverted the aircraft and then released his harness, he could no longer reach the stick. Whereupon, the aircraft nosed down in the second half of a loop, with the pilot half in and half out of the cockpit, and held there by ‘g’ and increasing wind pressure, until he crashed. In 30 Wing, this certainly happened to Dougy Yate, Patullo and Graham, and may also have happened to Squires and Tillet and many others.
We did not discover this for ourselves, for no one in his right mind would practice baling out near the inverted flight stall speed of about 120 knots with engine power off. No pilot ever managed to get out from the ‘engine-failed’ bale-out situation to tell us of these dangers. We put this down to bad luck or flak damage at the time. None of us had the time or the technical knowledge to hit upon any other reason.
A third important consequence of the Seafire III’s instability was its natural tendency to ‘float’ over the arrester wires. Not only do all aircraft tend to gain drag-free lift — like birds — from flying near the ground, but the Seafire ‘floated’ so serenely and for so long, that ‘Wings’ would admonish the pilots for “pulling back on the stick”, when nothing of the kind had happened. Besides the beautiful clean lines of the Seafire and the ground effect, there was another reason for the ‘Float-float-float-prang’ sequence — as in the FAA ‘A25’ song.
When the pilot cut the power and the slipstream speed reduced, the extra lift given by the positive angle on the elevator also reduced. The tail then lost a great deal of its lift and it, too, fell — without the pilot doing a thing. As this was equivalent to the pilot pulling back on the stick, in that it increased mainplane incidence and lift, and thus ‘float’, he always got the blame. It was a case of the ‘V Squared’ law making a much larger reduction to the lift on the tailplane than on the mainplane, when the slipstream over the tail surface fell from perhaps 100 mph during a power-on approach, to about 50 mph after the ‘cut’. The reason why this did not occur to other aircraft was that there is ‘negative lift’ on a stable aircraft’s tail surfaces — particularly when they are in the landing configuration — so that the ‘cut’ merely produces a nose down pitch, if any.
Although the American Naval fighters were equally light and almost neutrally balanced fore and aft, their handling characteristics were not altered dangerously by the subsequent additions of weight placed in awkward places. However, the early Mark of Corsair was well known as a ‘bouncer’ when it decklanded. The bounce occurred because, even when fully stalled and with the stick held back, a three-point landing was almost impossible. This was because the elevator power after the ‘cut’ was so reduced in the propeller ‘blanket’ effect that the pilot could not hold the tail down or the nose up. The first wheels to touch the deck were therefore the main landing wheels. The oleo design had considerable elasticity and bounced the nose up (the wheels were forward of the aircraft’s CG, so, being first to touch, they rotated the aircraft nose up), and the aircraft became airborne again in the ground effect. The problem was neatly overcome by Chance-Vought, the designers and builders of the Corsair. They lengthened the tail wheel assembly by 12 inches. This effectively allowed the pilot to do a three pointer each time he decklanded, and with new, non-rebound shock-absorber oleos on the main undercarriage, the Corsair became a very good decklander. Not only were American Naval aircraft designed for decklanding to withstand a 12 feet per second rate of descent without undercarriage failure but their greater strength and their large margin of positive stability in the landing configuration allowed a very much larger tolerance in approach airspeed and angle of descent. The batsman gave ‘advisory’ signals only — not mandatory signals as in British batting — and gave the pilot immense latitude in his approach airspeed, from 1.2 Vse instead of the Seafires’ 1.05 Vse. Lt (A) Mike Banyard describes some of the American techniques in Appendix 12.
The longer tail wheel assembly gave the Corsair much of the landing characteristics of a nosewheel aircraft. In the case of nosewheel aircraft, or aircraft with the tricycle undercarriage layout, the first wheels to touch are always the main wheels. These are aft of the aircraft’s CG. Incidence at touchdown is therefore automatically ‘wound off’ and lift is automatically ‘dumped’ — unless the pilot purposely holds the nose wheel off the ground. The nosewheel aircraft will not bounce at touchdown, unless it has a poor undercarriage design or ‘bouncy’ tyres. In fact nosewheel-designed aircraft — Vampire, Venom, Buccaneer, etc — were and are remarkably easy to deckland for they not only had a good view over the nose, but they could be approached at speeds well above the stall — up to perhaps 1.25 Vse — and still sit down immediately on coming into contact with the deck. The only limitation then, for a Phantom or Buccaneer, is a structural one, for there are usually no aircraft handling problems at such speeds so comfortably above the stall speed. Engine handling is made equally easy by maintaining high power on the approach, made possible by extending air brakes in the case of the Buccaneer.
In 1941, some preliminary trials were carried out by Lt/Cdr Peter Bramwell, RN, and others, to show that a hooked Spitfire could be landed on the deck of a carrier in just the same way as the Hurricane. Suitable or not, the Spitfire was, by 1943, the only available British single seat fighter for Fleet Air Arm use. Hawker had already transferred production from the Hurricane to the Typhoon and Tempest and the Blackburn Firebrand had another three years of testing to do before it would be safe to fly. The Fairey Firefly was a two seater and lacked the necessary performance. Folland’s intercepter fighter was not proceeded with. Neither the Implacable nor the Indefatigable had the necessary deckhead clearance for American Corsair fighters. It was the Seafire for them, or nothing.
R. J. Mitchell had designed the Spitfire purely for grass airfield operation. In these landing conditions, crosswinds need seldom be experienced and low vertical velocities at touchdown, with ‘wheeler’ landings for a good view over the nose, allowed a light, narrow-track undercarriage. Why have a heavy undercarriage when it was only used for a second or two each flight and was useless baggage at all others? The Spitfire therefore had a vertical velocity of only seven feet per second. If the vertical velocity at touchdown was a trifle greater than this the undercarriage would bust.
The liquid-cooled Merlin engine allowed the frontal area of the Spitfire to be extremely small. But, by placing the pilot behind it in a low ‘anti-g’ sitting posture, the view over the nose was restricted to four degrees below the flight line. On the airfield or flight deck, the view forward therefore consisted of treetops or ships’ masts. The RAF pilots resolved this difficulty by approaching to land in a semi-glide, at speeds well above the stall. They retained their forward view in the early part of the landing by doing ‘wheeler’ landings. This not only allowed them good forward view and good directional control after landing, but placed the loads vertically up the undercarriage legs, where Mitchell intended them to be. In the case of three-point, stalled, landings, there was a bending component, tending to wrench the attachment point free from the main spar. Naval decklandings were, perforce, always made in the three point touchdown posture, at which attitude it was at stalling incidence, about 15 degrees.
The touchdown speed for a Seafire was only three knots above the engine-on stall speed and below the engine-off stall speed — that is to say, 1.05Vse. Owing to the clean lines of the Seafire, the pilot could only lose speed, if he approached too fast, at the rate of about two knots per second if he were to close the throttle fully on a decklanding approach. As the distance over all the wires was less than 200 feet — a mere two or three seconds flying time — he had to make certain that he was at 1.05Vse by the time that he was over the round-down of the carrier. In practice, this obliged him to make a constant 1.05Vse approach all the way in. However, as it was five times easier to accelerate than to decelerate in a Seafire, speed control on the Seafire was very lopsided and pilots tended to come in too fast, particularly in gusty conditions.
The aircraft would not then be in the three point attitude when it hit the deck at its 3½ degree descent angle. The first wheels to touch the deck, the main landing wheels, being forward of the CG, pushed the nose up, while the tail was still descending. This and the ‘instability effect’ increased the aircraft’s nose up attitude, increasing wing incidence and thus, lift — even more in the ‘ground effect’ — just at the moment it was not required. The aircraft would then remain airborne after the initial bounce and float in the ground effect into the barrier two or three seconds later. A bounce would also result in the same ‘ballooning’ sequence. A pitching deck might wreck the pilot’s judgement entirely.
The Naval pilot’s solution to the poor view over the nose was to approach in a steady, left-hand turn, all the way down to the deck. With hood open, locked in position by half unlatching the small side door, with goggles on, with the head to one side and the seat raised, it was then possible to see part of the flight deck and the batsman through the haze of the port engine exhaust. If the sun was low and the carrier had been thoughtlessly pointed into sun, the sun’s reflection on the sea or on a saltladen windscreen would prevent a proper sighting of the deck or batsman. There was no windscreen wiper so that if it was raining — heavy rain completely obscured all view forward — the only way of seeing was to poke the head sideways out into the stinging rain and trust to luck your goggles stayed on and remained reasonably transparent. If the turn-in had been made too tightly or too easy, last-minute corrections could easily lose sight of the deck altogether, leading to a heavy arrival, an off-centre, or one-wheel, yawed, arrival, or a hasty ‘go round again’.
In the ‘yaw’ or ‘crabbed’ approach, the nose was ruddered crudely to starboard to allow the pilot to see to port. It might then engage the wire while still moving sideways. This would damage the hook up-lock, the hook would not lock up into its catch, and this could lead to further damage on arrestment, resulting in a ‘write-off’, as described below. This method, making use of ‘in-spin’ crossed controls so near the stall, was abominable airmanship and was forbidden in all squadrons accordingly.
The Seafire’s controls were uniformly light and responsive on the landing approach, befitting R. J. Mitchell’s masterpiece. However, because of its light wingloading — especially when landing with nearly empty tanks — the slightest gust over the round-down would balloon it into the air and above the proper descent path. If this occurred very late in the approach, the only recourse for height adjustment would be to use the elevator. With only a few ounces of push or pull required, overcorrection would produce large errors in height. If, then, the pilot hit the deck harder than usual and the aircraft bounced on its ‘grass-compatible’ undercart so that its hook missed the wires, the pilot might try to remedy this state of affairs by pushing forward on the stick to try to regain deck contact. This would raise the tailplane plus hook once more. If, then, the hook caught a wire, even if there was no yaw, the retardation force would be applied from below, initially, and the hook would not rise to lock up into its proper locked position in the fuselage. The direction of pull might distort the hook, now only hinged from its forward hinges, and the position of retardation being so near the aircraft’s CG it would not prevent the nose from ‘pecking’, the prop hitting the deck and the whole aircraft from crashing back onto its tailwheel, wrinkling the fuselage, driving the tailwheel up into the fin, shockloading the engine and breaking the propeller blades. Neither Boscombe Down, the decklanding trials unit, nor Farnborough knew how to cure any of these defects in the time available, neither had they made an extensive study of its basic causes. One of the test pilots — Lt (A). E. ‘Winkle’ Brown, DSC, RNVR — even used the ‘crab’ approach — until the inevitable result brought his efforts to a halt.
An added difficulty in speed control of the Seafire was the poor positioning and the unsuitable scale of the ASI (Air Speed Indicator). The indicator was at 90 degrees to the pilot’s forward vision as he looked at the flight deck. The needle vibrated over a scale where one eighth of an inch represented about ten knots of airspeed. Pilots tended, therefore, to fly by attitude or ‘by the seat of their pants,’ trusting in the tail-buffet stall warning to tell them when they were about to fall out of the sky. Most Seafires had loose or poorly-fitted engine cowlings or camera gun hatches near the upper surfaces of the wing roots. As this was a very critical area and where the slightest excrescence would provoke an early stall of the entire wing, the resultant rough air flow would cause an early tail-buffet, making the pilot add a few knots ‘for the sake of the wife and kids.’ In addition, following Ken Boardman’s death, pilots also feared poor throttle response, or even an engine cut following a throttle opening on the approach. This all tended to make a pilot add to his approach speed.
It is interesting to note that by 1963, the Buccaneer and Phantom were carrying out decklandings with 99.2% safety records, although their approach speeds were twice that of the Seafire. There were about 12 good reasons for this:
| (i) | The view, forward, was perfect, even in heavy rain. All jet aircraft have ‘blower’ rain clearance. |
| (ii) | The aircraft had a nosewheel configuration, three point touchdown attitudes were not essential, approach speeds some 15 knots above the stall were therefore allowed. This was a safe margin. |
| (iii) | Engine acceleration response was perfect, as was the airframe deceleration response with the use of very large airbrakes on the approach. Speed control was therefore easy. |
| (iv) | Aircraft wing incidence indicators took the place of ASIs and they even ‘talked’ to the pilot through his earphones, giving rate-of-change sound data to tell him if his speed was correct or how it was changing. The purpose of using wing incidence is that it automatically allows for changes in stall speeds due to changes in landing weight. |
| (v) | There was no risk of hitting the barrier or wrecking an expensive aircraft that way, because the deck was angled-off to port, allowing the pilot to go round again on a missed approach. There was no need for a barrier. |
| (vi) | Batsman’s errors were a thing of the past because the correct descent and approach paths were monitored by a gyro-stabilised mirror landing sight, visible to the pilot in the worst weather conditions, against sunglare and at night, and steady on a pitching deck. |
| (vii) | If the pilot’s instruments failed, his approach speed could be relayed to him from radar information in the carrier. |
| (viii) | If that failed, he had enough fuel, always, to divert ashore. |
| (ix) | If that was not possible, he could ditch, knowing his aircraft had been tank tested. |
| (x) | Or he and his observer could bale out, at any altitude, including take-off, using their hood-breaker ejector seats with or without auto hood-jettison. They would then have their seat automatically released and their dinghies inflated for them, right way up. |
| (xi) | If, when they ditched they found they could not get out of the cockpit in time before the aircraft sank, they could select ‘wet launch’ on their ejector seats. This gently eased their seats out of the sinking aircraft under water, by air pressure rams, through the hood if necessary. |
| (xii) | Finally, if the pilot or observer got a little hot under the collar thinking about it, they could select a little more ice in the cockpit fresh air supply and tune in to Radio One. |
The issue of RAF Specification Number F5/34 for an interceptor fighter in 1934, sparked off three major contributions. First, the Gladiator by H. P. Folland (F7/30), next the Hurricane by Sydney Camm and finally the Spitfire by R. J. Mitchell, which ousted all others.
The first of these was the last of the biplane fighters. I flew a Sea Gladiator at Henstridge for nearly an hour, so that I know very little about it. However, it was obvious that its main attribute was its high rate of roll (through its short wing span) its tight turning circle (through its low wingloading) and its high power to weight ratio, giving it a high rate of climb compared with its contemporaries. However, it had to make do with a Bristol Mercury VI, of 600 horsepower, not enough for future requirements and without much development potential. The Gladiator first saw service in the Chinese Air Force in 1937. The Japanese Air Force, who were trying to shoot it down, were so impressed with it, that they designed one like it for themselves, believing in the same philosophy as H. P. Folland.
The next two fighters, which went into mass production for the RAF, the Hurricane and Spitfire, made use of an in-line power plant of much greater power and development potential, the Merlin. Although the Merlin/Griffon series never produced anything like the ‘short life’ 3000 horsepower of the racing engine in the Supermarine S6B of 1934, it eventually produced more than double its original 900 horsepower by 1944 in the Merlin 55M, a much larger percentage increase in power than in its German counterpart.
Single-seat fighter designers, like racing car designers, were all after one thing, a high power/weight ratio. Folland also wanted very good turning manoeuvrability. This meant large wing areas. He also wanted a high rate of roll. This meant small wing spans. The two could then only be combined by using two wings, one above the other. He also wanted thin wings, to cut down the ‘V Squared’ frontal drag. Fabric-covered thin wings could not support themselves, so once again his choice fell upon a biplane, which could provide mutual wing support by struts and wires. As his aircraft was by now a thin-winged biplane, he could find no space in which to stow a light, retracting, undercarriage, so that he gave it a fixed one. In spite of his efforts, the Italians had a better fighter than the Gladiator in the CR 42. The latter out-climbed and out-turned the Gladiator in combat at Yeovilton in 1941 on each occasion they engaged in mock combat. The problem for Folland was the lack of power in its Mercury engine. He had hoped for the 800 hp Bristol Taurus, sleeve-valve engine, but this was given to the Bristol Beaufort and the Fairey Albacore instead.
Sydney Camm provided sufficient wing area, sufficient lift and a sufficiently small wing span from a monoplane, the Hurricane. Strength in a single unsupported metal wing was obtained by its thickness, ie the depth of the main spar, and by internal bracing. The thick wing also gave a higher lift coefficient for its small area. Although its turning circle in combat was wider than the Gladiator, it could, with its cleaner design and its much higher power, continue to climb while turning, and it always ended up above the Gladiator in a dogfight. It could then break off the fight if it wished by pointing its nose down for home. High altitude performance was also assured by the Merlin engine’s greater power at heights above 20,000 feet, due to its superior engine supercharging and its ‘constant speed’, variable pitch, three bladed airscrew.
R. J. Mitchell’s Spitfire was, however, a thing apart. He managed to produce a thin-winged fighter weighing half a ton less than the Hurricane, with superior top speed and climb, and with the same turning and roll performance. This improvement, with virtually the same engine as the Hurricane, was made possible by his invaluable experience with the S6B Schneider Trophy winner. He was able to retain his original Schneider design team, from 1931 to 1936 — when he submitted his first Spitfire design — through the gift of money by Lady Houston, O.B.E.
The huge floats of the S6B, containing fuel and oil cooling, made use of the ‘eggshell’ principle. They obtained their strength from their outside skin. Mitchell used his experience with the S6B to manufacture the wings of the Spit in a similar manner, the ‘stressed skin’ construction method, which used no internal bracing nor heavy ribs nor stringers. Spitfire wings were therefore far lighter and thinner than the Hurricane’s, but had equal strength.
The Hurricane virtually stopped in a dive at about 320 knots IAS. The Spitfire, with a third less frontal area and with its ‘ten per cent’ wing — no thinner wing was in service until the appearance of the DH Venom in the 1950s — and its semi-retracted radiators and flush-riveted wings and fuselage, could out-dive the Hurricane by about 80 knots IAS. It would also go further — and 30 knots faster — on the same fuel. It seemed as if at last an aircraft designer had got something for nothing.
Meanwhile, the Japanese, having accepted the tight turning-circle philosophy as a very important requirement in a fighter, had produced the extremely lightly constructed ‘Zero’ Zeke, or A6M single-seat monoplane fighter. They had built this fighter’s airframe half-a-ton lighter, still, than the Spitfire, for the same power and much greater internal fuel capacity. The metal skinning on its wings was so thin that the pilot could put his foot through it when stepping up to the cockpit. It had no armour plate and no self-sealing tanks. If hit, even by a few .303 inch Browning shells — to say nothing of the far more powerful American fighters’ 0.5 inch guns — it would, like the Skua, become a fire death trap to the pilot. Its early success, such as it had without radar/ADR interception aids, was entirely due to high piloting skills in combat, using its very low wing loading to out-turn its quarry — should the quarry be inexperienced enough to indulge in a normal turning-circle dogfight.
It was possible to overcome the turning deficiency of the Seafire LIII in a turning fight with the A6M, by using the ‘yo-yo’ principle, taught originally at Sutton Bridge Fighter Combat School in 1943. In this form of fight, the heavier fighter avoids a circling combat situation and uses his superior speed and perhaps climb. He attacks vertically from above or below his circling enemy. He uses aileron turns in the dive and climb to rotate his aircraft and thus his guns, in the enemy’s direction. The aspect then presented to the enemy, as he circles, would be a beam attack from vertically above or below him. The period of shooting would be extremely short, or course, as the range would close rapidly at the 90 degree angle-off. However, because the A6M would only be doing about 90 knots in the turn, the four degree deflection allowance of the gunsight — and over-nose visibility — was sufficent to hit. Very few hits would soon set it on fire.
Jet v jet (Phantom v MiG) combat in Vietnam was often similar. The MiG could out-turn the Phantom, but the latter, when light, had a better power/weight ratio and could climb vertically at lowish altitude. The combat was reduced to ‘energy conservation’, converting altitude to speed, kinetic to potential energy, and vice versa. Airspeeds varying from zero to supersonic, distances apart varying from ten miles (using radar to retain contact) to a few hundred feet, each avoiding the other’s fire in a fantastic death struggle, until one or the other made a mistake.
R. J. Mitchell’s team had designed the Spitfire ‘overload’ or ‘long range’ tanks to fit in close contour with the underbelly. The half inch gap between the ‘slipper’ tank and the underside of the fuselage was insufficient for a visual or any sort of check to be made that there was a fuel-tight fit in the pipe leading from the tank to the engine. A sliding fit was essential to allow it to be jettisoned easily, but the necessary suction-tight joint could not be guaranteed. The only check which could be made after fitting and filling up a new tank after the last one had been jettisoned, was to run the engine on the flight deck. This was not only inconvenient, it was no guarantee that the tank would still work once it got into the air. We had had three pilots unable to get their tanks to work in the air after take-off in operation ‘Goodwood’. This made it almost impossible for the squadron to operate unless there was a spare deck on which to make emergency landings, before anyone else. It once again brought the Seafire into disrepute with the Admirals. The problem was not overcome until 880 Squadron reached the Pacific in the following year and ‘invented’ the Kittyhawk ‘torpedo’ tank carried on a modified bomb pylon.
The 90-gallon version of the slipper tank was intended for ferry purposes only. In the absence of anything better, 24 Wing in Indefatigable used this on operations. Besides its unreliability, and although the Seafire was able to withstand the half-ton of extra weight in a combat situation, the use of full power for long periods overheated its fuel lines and caused serious fuel ‘aeration’ (boiling) and consequent engine failure. This engine failure also occurred to the Seafires of 30 Wing when using the Kittyhawk tank, for it used identical fuel supply lines.
In the last period of the Pacific operation, 30 Wing’s 14 casualties had been caused as follows:
Two pilots, Bedore and Dane had been killed when their wings came off following pullouts from dives (Bedore’s may have been by flak). Five had had engine failure after strafing and three of these pilots had failed to bale out or ditch satisfactorily. Seven others had crashed fatally for various reasons. Of these, one hit an Avenger in cloud; one had his engine cut out on the approach to deck land and failed to get out; one flew into the sea, got lost or got shot down in the Truk operation, having u/s r/t; one hit a radio mast when landing at Ponam; two became lost on the return to the ship and could not be found after saying they were about to bale out; one dived into the ground from probable fuel asphyxiation in Ceylon.
Of these 14 fatalities in 360 operational sorties and some 3,000 flying hours, only one pilot had certainly been shot down or had been killed by enemy action. All the remainder, except two, were probably attributable to the Seafire III itself and its unsuitable modification state to meet safety requirements for the job it was trying to do in the Pacific. The Seafire III was therefore at least six times as dangerous as the enemy. To have had but one certain enemy-action casualty (Dougy Yate) in 30 Wing instead of the assumed seven flak casualties, would have further convinced Vian that we were “failing to do our utmost”.
The cause of engine failure, certainly to Graham and Patullo (who were killed) and probably Saxe and Tucker (who ditched and were rescued), was undoubtedly due to a peculiar fuel supply problem. The Seafire/Spitfire’s main supply fuel pipe from the main tank lay close to the hot supercharger casing. With the huge overload tank in use, the fuel in this pipe would have been stationary for an hour or more while fuel was being sucked up from the Kittyhawk tank. During this long period, the main supply pipe itself would get very hot, uncooled by a flow of cool petrol from the insulated main tank further aft, and very near the over-boosted supercharger casing. During this time it was drawing fuel from a long-range tank, designed to be used only for ferry flights at economical cruise power settings, in temperate climates at high altitude. The temperature of the main delivery pipe from the main tank would be well above the boiling point of petrol (which at normal pressure is about 75 degrees Centigrade). It would boil the moment the vapour pressure in the pipe was reduced by the operation of the supply cock to ‘On’, when changing tanks from the ‘jett’ tank to partly-used main tank. Because we carried half our total fuel in the ‘jett’ tank, the change-over to main tank would always have to occur when more than half the flight had been completed and usually when the engine was very hot after the use of combat power, just after the final strafing run. Once the fuel had vaporised in the empty line the centrifugal pump would run dry and overheat and cause the engine to catch fire — as in Saxe’s case. The failure in the Seafire III’s fuel supply is, in fact, analogous to the fuel vapour lock sometimes experienced by some old cars when they are left in the sun after a hard, summer’s drive.
The first Spitfire to be flown with the new Griffon engine was flown by Jeffrey Quill in November 1941. In October 1944, he carried out the decklanding trials of the Seafire 46 and 47 in Pretoria Castle, and happened to be joined by Buster Hallett who was doing a few in the Seafire XV at the time. When I visited the Fifth Sea Lord on 8 December 1944 to ask whether we could take some Mark XVs with us to the Pacific, they were still not ready.
Thus, when we eventually received our first three XVs in Australia in November 1945, we were very proud of them. Three weeks afterwards, on 27 November, we were paid a visit at Schofields by the Australian Naval C-in-C. Not suspecting the Seafire’s XV’s potential dangers, we arranged a small flying display.
After an impressive balbo of all 36 Seafires, it was S/Lt Norton’s turn. He was to do a solo demonstration in one of our Seafire XVs. In his display he was to carry out a slow roll on take-off followed by an immediate loop, to a fast run over the airfield at about 425 knots, then another loop and a roll-off-the-top to finish with. Like all Aussies, he was an experienced Spitfire pilot and knew exactly what he was doing. Lieutenant George Willcocks, DSC, RNVR, writing from Australia in 1984 writes of this incident:
“He went in just beyond the Wardroom. The aircraft broke up in midair and the starboard wing fell off. The engine travelled on for a mile with its prop still spinning, and set fire to some trees off the airfield. It was a tragedy, for Norton’s parents were watching and so was his fiancée. He was due to marry next day with a large squadron party laid on.”
The Board of Enquiry found that it was “Pilot Error”. They said he was exceeding the limit of 425 knots IAS.
I was not at all happy about this, neither was Nat Gould or Ian Lowden of the Aussie’s Flight. Norton would not have done anything so stupid. My own experiments for the next two days revealed that both of our remaining Seafire XVs became left-wing-low at speeds above 400 knots. By 450 knots, all the aircraft had become so left-wing-low that strong aileron forces were required to keep laterally level. I looked across to the starboard wing, the one which had first come off Norton’s aircraft, and saw that the up-going aileron angle was much greater than usual, yet the down-going aileron on the port wing had barely moved at all. It was obvious that the control wires were stretching and that the aileron on the starboard wing was ‘upfloating’ a great deal. I thought that this might be a very dangerous state of affairs at high IAS, as I had not seen it happen before. Application of even harder aileron to raise the left wing had no effect and, if the speed was increased further in the dive, it showed signs of having the reverse effect. (The wing itself was twisting, the aileron acting as a tab, reversing the lift. No wonder Norton’s wing came off, for the aileron would eventually tear off due to excessive ‘upfloat’ and this, and wing twist, would precipitate wing spar failure.)
I reported my findings to the ship’s Senior Engineer Officer and he grounded all our new Seafires and those in the delivery pipeline at Brisbane. I had explained that I had thought that the large aileron forces to hold the right wing down were required at these very high speeds to offset the anti-torque twist built-in to the Griffon-powered version of the Spitfire. As 425 knots — the limit set in the Pilot’s Notes — could easily be exceeded even in a shallow dive, we should be losing wings and pilots all the time in 801, unless something was done to improve things. The technical fault — well-known to aircraft designers — is correctly described as ‘aileron reversal’.
Within two months, a party of aircraft technicians from the UK arrived. They stripped off the skin behind the main spar on the top surface of each wing root and replaced it with skin of a far thicker gauge. To be able to remedy this defect so quickly, it was obvious that the manufacturers had recently struck the snag themselves — ‘aileron reversal’ — but hoped that we in 801 would manage until they could get round to the remedy in Australia. After the modifications, we had no further trouble with wing twist, even when we exceeded 425 knots ‘accidentally’ by a further 50 knots or so.
The second lethal shortcoming of the Seafire XV was a supercharger fault. The ‘self-change’ mechanism from one blower speed to the higher speed was similar in action to the automatic clutch and gearbox of a car. When changing blower speeds on the climb, the blower speed of about 15,000 rpm had to be speeded up to about 20,000 rpm in a matter of seconds. If the clutch gripped too tightly it would strip the gears. If it gripped too loosely it would burn out. In both failures the engine would lose its supercharger and it would stop, catch fire, misfire or overheat. The only method of avoiding such failures was to engage the gear manually at reduced rpm. This was not always possible, neither did it always work, for there were several fatal engine failures.
There was a third shortcoming, much more involved and difficult for us to understand. One of the most welcome characteristics of the Seafire III had been its rapid acceleration, allowing it to take-off from the flight deck in 180-200 feet without the aid of a catapult for which it had been unnecessarily adapted. When the Australians in 801 took their first, halting, steps in our carrier training programme in January 1946, they tried to practice short take-offs — as on a flight deck — at Schofields. One just missed our flight huts as it swooped off the runway in a semi-stalled turn to the right, totally out of control. Pilots found that the extra 1000 horsepower of the Griffon VI, turning the other way to a Merlin, gave far more torque reaction. This was because of several factors, not just because of the increased engine power. First, the propeller was twisting the slipstream far more than in the Merlin engined version, for it had to absorb nearly twice the power in a propeller of the same diameter and at the same rpm. Second, the ‘three point’ unstick incidence of the Seafire XV was greater and nearer the stall, owing to the longer stroke oleos. Third, the aircraft weighed a ton-and-a-half more than the Seafire III, and therefore needed to unstick at a higher airspeed than the Seafire III. The static torque from the engine attempting to turn the aircraft in the opposite direction to the propeller was not the main reason for the trouble at take-off, for the torque on its own could be easily corrected by upfloating the port aileron a couple of degrees. It was only equivalent to a 200 lb weight on the port wing abreast the outer gun, and was not serious. The main cause of the ‘right-wing-low’ at take-off, or ‘torque stall’ as it was wrongly called, was a partial stall of the entire starboard wing. With two tons of lift on the port wing and very little indeed on the starboard, it was small wonder that the aircraft carried out a full half turn of a spin to the right, at unstick, on occasion.
At the moment of unstick, if the aircraft were ‘wrenched off at maximum incidence, both wings would be very near stalling incidence. The slightest dissimilarity in their airflow would upset the balance of lift between the port wing and the starboard — too great to correct by aileron. In the case of the Seafire XV, the starboard wing root stalled first as it was subjected to greater slipstream incidence. This stall then spread outboard, until the wing started to lose lift. The starboard wing then started to drop and in doing so, increased its incidence yet more, and fully stalled. Attempting to raise it with aileron would make matters worse.
In the case of a ‘minimum distance’ take-off on the flight deck when the pilot was pulling back on the stick to get off as soon as possible, the starboard wing might not establish a proper airflow, and therefore lift, at all.
Once the starboard wheel’s lateral steadying effect was gone as the aircraft left the deck and flew over the ship’s bow, it was inevitable that it spun to the right, into the sea, right in the path of the carrier. Further causes of this trouble might have been due to a gyroscopic ‘kick’ to the right if the pilot pulled the nose up too sharply at unstick, or it might have been that ‘Woolworth’ carriers, such as those used by 801 in their carrier work-up, had particularly bluff bows, and the upward airflow over the bow might have aggravated the situation. With insufficient left rudder available, and perhaps with too much aileron to raise the starboard wing, the aircraft would be ‘pro-spin’ at a speed at or near the stall. It was almost inevitable that the Seafire XV spun under these circumstances. It is interesting to note that the port wing root of the Blackburn Skua aircraft had a lift spoiler fixed to its leading edge — on its down-going propeller side, to equalise the stall characteristics between port and starboard wing when power was on.
Happily, a later Mark of Seafire — the 47 — was fitted with a six-bladed contra-rotating propeller. This entirely cured the aileron reversal (which killed Norton) and, of course, cured the swing and wingstalling on take-off and on going round again. It also made the Seafire 47 a beautiful ground attack and air-to-air combat aircraft, requiring no rudder trimming in the dive, and with no gunnery or rocket aiming errors due to skid caused by propeller gyroscopic effects (as in the five-bladed Seafire 17), and with slipstream yaw effects a thing of the past. In addition, the heavier engine in all Griffon-engined Seafires forward of the CG, had cancelled the need for the ‘positive weight’ and other dangerous palliatives intended to cure instability. Its extra weight forward also allowed a bigger tailplane, fin and rudder plus a sting hook and retractable tailwheel. The sting hook prevented ‘pecking’ and hung much lower than the original ‘A’ frame hook. The Griffon engine was also fitted with a Coffman starter which allowed a flight deck free of starter trollies.
Later versions of the Griffon in the Seafire developed over 2000 hp up to heights of 20,000 feet, using a two-speed, two-stage supercharger with an intercooler between each stage and a ‘ram’ air intake immediately behind the propeller. The Seafire 47 could reach 10,000 feet in 2½ minutes from a standing start carrying 200 gallons of fuel. It had geared aileron tabs which allowed a crisper rate of roll at speeds above 350 knots IAS. It had improved engine cowlings and a ‘teardrop’ fully transparent hood and a cut-away rear fuselage to allow rearward vision. The streamlined windscreen allowed full forward view in heavy rain without the need for a windscreen clearance scheme. Many of these improvements might have made it in time for the final landings in Japan scheduled for February/March 1946, but none would have made it for FAA use in ‘Olympic II’. The Seafire/Spitfire had come a long way in ten years. It had nearly trebled its engine power and rate of climb and nearly doubled its speed. The German equivalent — the Me 109 — came nowhere near doing this in its equally long life span.