Wednesday, May 14, 2008

Geoffrey Raoul de Havilland Jr OBE 1910-1946

Mosquito prototype

Geoffrey Raoul de Havilland Jr.( young D.H. as he was universally called) took over the de Havilland’s chief test pilot's position in October, 1937,when R. J. Waight unfortunately lost his life on the T.K.4. Being, however, the son of an illustrious father, Sir Geoffrey de Havilland, who designed, built and test flew his own aircraft from 1911 onwards, young Geoffrey can be said to have been '' in the industry'' from the very cradle. It is not generally known that Sir Geoffrey took his R.Ae.C. ticket No. 53 in February, 1911, on the second machine of his own design and construction, and that he has made many of the first flights on new D.H.types right up to the Moth Minor in 1938. Geoffrey's first flight is lost in the dim past, but certain it is that at the tender age of six he was flying with father at Hendon in a D.H. 6 (also known as the Clutching Hand).
When 18 years of age he left school and came to de Havillands as a premium apprentice for 4 years and learnt to fly on Moths at the firm's reserve training school. After spending two years in the drawing office—much of the time being spent looking out of the windows envying the pilots—he joined the Air Operating Company, who were doing a lot of air survey work in South Africa. This, however, gave him but very little flying, and at the end of six months, he came back to England to become a flying instructor to the D.H. Technical School. The aircraft were wooden Moths built by the students. In 1929 he took his B licence ; a very simple business in those days. Some 20 or 30 hours' solo flying, a little cross-country work, a simple "Met" exam, and about one hour's night flying at Croydon was sufficient to qualify. In 1934 Capt. Hubert Broad was chief test pilot of de Havillands, and Bob Waight looked after the production side. There was so much work, however, that Geoffrey was given the opportunity to lend a hand testing Tiger
Moths, Dragons, Rapides, Express Air Liners, and Hornet Moths.

Broad left the company in 1935 and Waight took over, starting, with the Dragonfly and later the Albatross. It was during the period when the prototype Albatross was going through its development flying that Waight lost his life, and de Havilland took over as chief test pilot.
Nobody could have taken on a more interesting or more complex job because the Albatross was completely experimental from tip to tail. Engines were new, construction was new, and the layout was extremely advanced. He had a curious experience on the Albatross. While its strength was ample for all flying loads, some unfortunate drilling had weakened the fuselage under ground loads, and shortly after landing from a test flight the machine broke in halves on the ground.When war broke out he was busy testing Oxfords and Flamingoes, but when things became desperate at the time of the Battle of Britain, de Havillands did a big job doing emergency repairs to shot-up Hurricanes.
Dick Reynell of Hawkers came over and gave- Geoffrey the "know how" on Hurricanes. A little later Dick went out on operations with his old squadron (No. 43) and was, unfortunately, shot down.' He was an excellent test pilot and a gallant gentleman.

Geoffrey flew the first Mosquito at Hatfield on November 21st, 1940, but he is more proud of the first flight of the prototype Mosquito fighter. This was built at a dispersal factory with no airfield. To save some six week's wasted time in transport and re-erection at Hatfield, Geoffrey used local fields by having bridges built over ditches to give him a 450yd run for take-off, and then flew the fighter to Hatfield.

He was the 2nd man in Britain to have made a first flight on a jet aircraft,the Vampire, which he flew for the first time on September 21st, 1943, but Geoffrey had already flown the Gloster E.28 at Farnborough.

The first airing of the Vampire proved it to be a tribute to the D.H. design and aerodynamics staff, as it behaved almost exactly as they had forecast. There was, however, somewhat of an aileron overbalance which limited the speed to 250 m.p.h. and a rather severe tip stall. Geoffrey de Havilland had made a number of investigation- flights on Mosquitoes for compressibility effects, but on the Vampire he has done extensive work. The Vampire, under the effects of compressibility, executes a series of sudden high-speed stalls: The path of the machine is similar to an artist's conception of a streak of lightning, and unless the pilot is strapped-in tightly he is likely to be knocked out by hitting the cockpit roof. Geoffrey, with another pilot, has flown the Vampire in tight formation at over 500 m.p.h., and to investigate snaking, which is causing considerable trouble on most jet aircraft, he has flown the Vampire with rudder locked. Like most of the test pilots of that era, he was living on borrowed time, they having at some stage of their careers had close shaves. Strangely enough, Geoffrey's nearest go was on about the mildest type he ever flew. It was the first production Moth Minor. The prototype had completed its spinning tests, and the same tests on the production model appeared to be only a matter of form. He was flying with John Cunningham (now Group Capt., D.S.O., D.F.C., and test pilot for the D.H. engine division) at the time. The Minor was put into a spin at 5-6,oooft, but after it had failed to come out in five turns and the engine had stopped, a panic decision was made to abandon ship.Test-flying a Hurricane, too, almost saw him off. This particular aircraft had had a gruelling time in the Battle of Britain, and the whole canopy came off at 4,000ft.hitting him in the face as it blew backwards. At first blind through the amount of blood in his eyes, he flew more by instinct than anything else until he found he could get a little relief by holding his face close to the instrument board. The blood dispersed a little and he was able to land through what appeared to be a thick yellow haze. He wears the scars across his nose to this day, and there was a terrible moment during that flight when he thought he was really blind. On another occasion the oxygen bottle contained only compressed air, and the effects from this were at first blamed on the previous night's party.
In the days of peace before the war Geoffrey de Havilland was to be seen at all the air meetings and twice finished 4th in the King's Cup Race flying the TKi and TK2. He was awarded the OBE in the King's birthday honours in 1945.
Geoffrey flew the first DH 108 prototype, TG283, utilising the Vampire fuselage and a 43° swept wing on 15 May 1946. Designed to investigate low-speed handling, it was capable of only 280 mph (451 km/h). Geoffrey de Havilland, gave a display flight in the DH 108 during the 1946 Society of British Aircraft Constructors (SBAC) airshow at Radlett.

The second, high-speed prototype, TG306, with a 45° swept wing incorporating automatic leading-edge Handley Page slats and powered by a de Havilland Goblin 3 turbojet, flew soon after in June 1946. Modifications to the design included a more streamlined, longer nose and a smaller canopy (framed by a strengthened metal fairing) facilitated by lowering the pilot's seat. While being used to evaluate handling characteristics at high-speed, on 27 September 1946 TG306 suffered a catastrophic structural failure that occurred in a dive from 10,000 ft (3,050 m) at Mach 0.9 and crashed in the Thames Estuary killing Geoffrey de Havilland Jr.