A while ago, I wrote about the Beech Starship, a business aircraft which looks like an artefact from a future civilisation. By contrast, the DH103 Hornet fighter appears hopelessly old-fashioned – yet it reached almost 500mph in level flight, which made it the fastest piston-engined aircraft of its day. It could cruise at the speed of today’s jet airliners, and outran the first jet-powered fighters. The chances are, if you took a Hornet to the air races at Reno in Nevada today, it would out-run all the souped-up Mustangs and Sea Furies, setting a new absolute speed record.
In many senses, the Hornet was the piston-engined aircraft perfected. A few weeks ago, the BBC showed a documentary about Eric “Winkle” Brown, the Scots-born test pilot. As a naval aviator, he set a record for the number of landings on aircraft carriers which has never been beaten, and when in his unassuming way Brown describes the Hornet as the favourite from all the different aircraft he flew, that means something. Captain Brown has flown more types of aircraft than anyone else in history.
Eric Brown is a top candidate for the Most Interesting Man in the World. As a schoolboy, he attended the 1936 Olympics in Berlin and met a WW1 flying ace. During WW2, he escaped from the wreckage of a torpedoed ship, helped to liberate Belsen and took 2,000 enemy prisoners armed only with a pistol – not to mention a few close calls where he had to abort and promptly GTFO using a parachute. After the War ended, he interrogated leading Nazis including Hermann Goering, aircraft manufacturer Ernst Heinkel and designer Willie Messerschmitt. Brown was the first man to fly a jet on and off an aircraft carrier, and he set aviation records that will almost certainly never be broken.
The de Havilland Hornet was his favourite, "For the simple reason it was over-powered. This is an unusual feature in an aircraft, you could do anything on one engine, almost, that you could do on two. It was a 'hot rod Mosquito' really, I always described it as like flying a Ferrari in the sky." The Hornet was the fastest twin piston-engined operational combat aircraft in the world while in service, and the first aircraft to demonstrate a cartwheel manoeuvre.
"For aerobatics the Sea Hornet was absolute bliss. The excess of power was such that manoeuvres in the vertical plane can only be described as rocket-like. Even with one propeller feathered the Hornet could loop with the best single-engine fighter. I had felt such absolute confidence that I was mentally relaxed … Indeed, there was something about the Sea Hornet that made me feel that I had total mastery of it.”
"In my book the Sea Hornet ranks second to none for harmony of control, performance characteristics and, perhaps most important, in inspiring confidence in its pilot. For sheer exhilarating flying enjoyment, no aircraft has ever made a deeper impression on me.”
At the root of any aircraft’s design is the equation which resolves power, weight, lift, drag and trim into performance. In simplistic terms, power makes an aircraft climb whereas attitude varies its speed. The Hornet’s high rate of climb came thanks to the Rolls-Royce Merlin, arguably the engine of the 20th Century, which in this case developed more than 2000hp from 27 litres of swept volume. Just like the Mosquito, the Hornet had a pair of Merlins but in this case they were faired into streamlined “power eggs”.
The Hornet’s top speed is partly the function of a low co-efficient of drag arising from a sleek fuselage and a laminar flow wing; this thin wing was made possible by new materials. The Hornet’s long range came thanks to its light weight; both strength and light weight derived from de Havilland’s early mastery of composite construction.
During the 1930’s, aircraft structures evolved from doped fabric stretched across an ash frame, to the geodesic spaceframe of steel tubes which Barnes Wallis used in the Wellington bomber, and eventually to all-metal stressed skin structures. De Havillands went their own way, searching for a different method of achieving strength and lightness. They settled on timber, but rather than a load-bearing timber frame (like a Morgan car or a timber kit house) they developed the first composite monocoque.
Unlike the Beech Starship, a revolutionary aircraft which used carbon composites, the Hornet wasn’t a great leap into the unknown; it’s an evolutionary aircraft, albeit one at the very apex of its line of evolution. De Havillands had been working on composites for a decade before the first Hornet took flight - although their initial objective was to build stronger, lighter propellers.
In order to cope with more powerful engines, propellers had grown in diameter, gained more blades, and their tip speeds were approaching the sound barrier. As a result, the centrifugal forces at the propeller hub had increased to the point where there were many catastrophic failures. De Havilland Propellers worked with Aero Research at Duxford to overcome the drawbacks of laminated timber props, successfully using phenol-formaldehyde resin in the manufacture of propellers. The attraction of this material was that, with a density of around half that of aluminium alloy, centrifugal forces at the root were greatly reduced.
De Havillands was a rare aircraft company which made everything for itself. Piston engines were built at Stag Lane in Edgware, then jet engines and later rocket motors plus of course complete aircraft at Hatfield, Leavesden and later Hawarden. As a result, it was able to cross-fertilise materials research between propellors, wings and fuselage design.
The work on propellers “spun off” into fuselage and wing structures for the (almost) all-timber Mosquito, which the wartime Press christened the Wooden Wonder. The Mosquito was built from sandwich panels consisting of thin skins of plywood veneer bonded to a core of end-grain balsa wood. The core functions just like the web of an I-beam while the plywood skins function as the flanges. The sandwich panel's bending stiffness is proportional to the core thickness, in the same way that an I-beam becomes stiffer as the web deepens. Doubling the core thickness yields a panel roughly six times stronger and 12 times stiffer.
At a time when other WW2 combatants were desperately trying to smelt cobalt, vanadium and other rare metals into exotic alloys, it seems bizarre that de Havillands were in the market for balsa wood. You can only assume that German spies put this down to British eccentricity, if they even remarked on it at all, yet Baltek’s sawmills in America struggled to keep up with demand. Today the technology seems so accessible; hobbyists and model-makers have access to the same plywood veneers, balsa wood and epoxy glues that de Havillands used.
Plywood was a relatively new material, and also a composite, with plies of different thicknesses and orientations providing degrees of strength and stiffness. As well as de Havilland themselves, the Mosquito was built by Roe, Gloster, Phillips & Powis and even Venesta – the forerunner of Venesta Cubicles which is still in business today. In 1937 their "Venesta" plywood and "Plymax" metal-faced plywood made them an ideal choice as fabricators of ply composite aircraft such as the Mosquito.
The Mosquito was built by the furniture industry, which was mostly based around its traditional centre in Buckinghamshire: incidentally, that’s the reason High Wycombe was one of the most heavily-bombed London suburban towns. The industry had a long history (Defoe mentions it) in the town and in the 1940’s there were still many local manufacturers. The Windsor chair was its most famous product, but practically every other sort of furniture was also made. Components for the Mosquito were reportedly produced by Marples and G-Plan, and supplying the materials was a multi-national effort: the frames used Alaskan spruce and British ash, the sandwich used 3-ply Canadian birch plywood and Ecuadorian balsawood.
The DH103 Hornet evolved from the DH98 Mosquito - and as is the way of things, it became lighter, faster, more powerful and stronger. Both aircraft used variations of a pre-formed plywood monocoque shell strengthened with spruce stringers and constructed using high-strength synthetic bonding resins. This technique had been pioneered on the famous DH.88 Comet racers, and would also be used to great effect on the Dragonfly light twin and the Albatross airliner of 1938. One step forward from the Mosquito was the way de Havilland built the Hornet’s wing spars, and another was the wing surfaces themselves.
Mosquito wing spars have all-wooden tension and compression booms, but this would have been impossible for the Hornet, because of the large cross-section of wood necessary for the more highly-loaded wing. The problem was overcome by making the tension booms from aluminium extrusions, and using wood for the spar webs and compression booms. A layer of veneer was bonded to the aluminium parts then everything was assembled to form a spar of remarkably low weight and high strength.
Moulded wood veneers of a type that we’d now term cross-laminated timber were combined with more conventional parallel layered glulam to produce spars of amazing accuracy and complex geometry. Tapered and kinked spars with “L”-shaped sections were formed using this technique, which was originally developed for manufacturing Isokon furniture. Isokon is well known in architectural circles, thanks to the Lawn Road Flats designed by Wells Coates…
The Hornet’s wings comprised an aerofoil with a composite wood and metal internal structure, with a stressed birch-ply double upper skin and an under surface of reinforced “Alclad”. This was the first time that aluminium had been bonded to timber in a structural fashion. Lift acting on the Hornet’s wing meant that the metal skin on the underside of the wing went into tension, and the ply-balsa composite went into compression – so the materials’ inherent qualities were used to best advantage.
The idea of combining skins of ply and aluminium with a lightweight core was a conceptual leap born of on a new generation of synthetic adhesives. De Havillands’ composite structures relied upon a new epoxy resin developed by Aero Research. This glue, “Redux 775”, was developed in 1941 as the first modern, synthetic structural adhesive for metals - and it was first used in the Hornet Mk1 which was built at Hatfield. Hornet construction, like that of the Mosquito before it, used similar techniques as modern fibreglass wet layup. The positive mould was covered with wax, then strips of thin veneer were laid up in different directions to improve the tensile strength in all directions, just as today you would lay up glass or carbonfibre mats.
The first skin would be covered by a sandwich layer of balsa wood, followed by another layer of veneer. Metal fittings were embedded in the wooden layers and a low voltage applied to heat the resin electrically, which speeded up curing. Once everything was dry, the fuselage or wing half would be removed from the mould then after installation of some formers, cables and wiring, glued to the other half. Finally the fuselage would be covered in another layer of thin wood, covering the glued joint, then covered in aircraft linen, doped and painted to improve aerodynamic smoothness.
In 1948, de Havillands acquired an aircraft factory at Hawarden Airfield near Chester: it was used to build and assemble the Hornet Mk3, while other parts were manufactured at the firm’s factory in nearby Lostock. Incidentally, Hawarden is now called Broughton, and after de Havilland became part of Hawker Siddeley it developed sophisticated wings for their airliners: today, it builds every wing for every Airbus airliner, and is owned by GKN. The “N” in GKN stands for Nettlefolds, and when the Hornet was in production they had several huge factories in the Black Country, stamping out millions of cross-head screws an hour. Today, GKN uses carbon fibre to build composite aero-structures which owe a great deal to the principles that de Havilland developed three quarters of a century ago.
The Hornet’s fuselage was built in two halves which joined together on the centreline, so called “egg carton” construction using cold moulding to form the curves. This monocoque structure gave the fuselage a high degree of redundancy which meant that the aircraft could sustain terrible damage yet keep flying. Many Mosquitos returned home missing large chunks of wings, fins and control surfaces, shot away by enemy cannon fire. Timber composites also avoided the hidden dangers of metal fatigue, which de Havilland fell foul of with their Comet airliner during the 1950’s.
Without the work of Aero Research and de Havilland Aircraft during the 1930’s and 1940’s, it’s arguable that there would be no plywood composites or structural adhesives, hence the SIP panel and the JJI joist wouldn’t exist, either. It’s also worth noting that the Beech Starship, which was hailed as revolutionary in form and construction, isn’t as original as I implied. The Starship was also built in two halves, and epoxy resins were also used to bond its composites together. Just like de Havillands, forty years before them.
Although it marked the apex of de Havilland’s piston engine development, there’s no sense in which de Havillands developed a Pygmalion-like relationship with the Hornet. Even as it first flew in 1944, the firm was already building jet-propelled aircraft, so the Hornet’s career was cut short. After the DH98 Mosquito and DH103 Hornet, De Havilland’s plywood-balsa-plywood sandwich was later used to form the fuselage of the Vampire and Sea Venom jets.
De Havilland refined the assembly process: steel bands were latched onto heated jigs with quick release toggles, to ensure smooth fuselage cross-sections. Adhesive curing cycles were carefully instrumented and automated. Smaller glue-laminated components such as the engine intake ducts used thin timber strips which were cold-formed on jigs to tight radii. Nonetheless, the Hornet's gift to us all is composite construction, which the designers of racing cars, airliners, yachts, buildings and even fridge freezers take for granted.
The late Martin Pawley was fascinated by these technology transfers, and the crossovers between architecture and other fields. As a columnist in the AJ, BD and so forth he wrote about the design of tube trains, cars and aircraft – seeing them as complementary to architecture. He recognised the truisms that racing improves the breed, and war pushes technology forwards faster than peace. While this bandwagon was passing, I thought I'd jump onto it…
Pawley was in tune with the spirit prevailing during the late 1980’s and early 1990’s, when Richard Horden built a series of houses using off-the-shelf components from racing yachts, then Rogers and Foster completed a series of buildings which borrowed from the automotive and aerospace industries, such as neoprene gaskets and super-formed metals. This climate gave birth to a thousand architectural dissertations about Lotus sports cars, Slingsby sailplanes and McLaren's F1 operation. Unwittingly, they echoed a pattern from 50 years before, when wartime firms desperately hunted around for peacetime outlets once hostilities ended. De Havilland were fortunate, as their focus shifted readily from military to civilian aircraft. Venesta gave up flying and came to specialise in toilet cubicles and IPS systems. Others were not so lucky.
What about the legacy of Ronald Bishop, who designed the Hornet? He should be remembered for helping to win the War using pioneering materials: but today it seems that War means rousing musicals, martial style (smart uniforms never go out of fashion) and the cult of the Great Man. Our superficial treatment of that era ignores Bishop and his counterparts Barnes Wallis, RJ Mitchell and Roy Chadwick who were responsible for the Wellington, Spitfire and Lancaster respectively. They were complete designers, in the sense that they harnessed materials science, structures, aerodynamics, manufacturing techniques as well as considering damage tolerance and repairability.
They also had a sense of purpose which is difficult for us to grasp now: they were part of Churchill’s enormous enterprise which stretched from shadow factories making widgets to the invention of operational research.
The sadness is that no Hornets survive at all today, although there are rumours that an entire squadron was dismantled and buried under an airfield in Malaysia when they became surplus to requirements. It seems unlikely that anyone will disinter them, but you never know…
Some images used here are courtesy of the Hornet Project website, which has temporarily disappeared from the web.
