16 January 2011

The Short and Deadly Career of the De Havilland Swallow

Evolution of the Comet from mailplane to jetliner
The Allies reaped a technological windfall with the unconditional surrender of Nazi Germany in May 1945. Both the United States and Great Britain in particular found that their wartime studies of future jet aircraft experienced quantum leaps in performance with the incorporation of the aeronautical knowledge of the personnel technical documents, and aircraft of the Third Reich. During the final half of the Second World War, Britain grappled with the future of passenger flight with the Brabazon Committee's deliberations on the future of commercial aviation. Of the five designs put forth by the Committee, the most advanced was the Type IV design for a jet-powered 100-passenger design. This aircraft would become the De Havilland DH.106 Comet- but the DH.106 started out as a very small, modest adaptation of the Vampire jet fighter as jet-powered mailplane with a six-seat passenger compartment. But with the input of the Brabazon Committee, the DH.106 evolved into a substantially larger aircraft that at one point was a tailless design before taking on the shape now familiar as the Comet. 

TG283, the low-speed DH.108 aircraft
The British were particularly interested in the Messerschmitt Me 163 Komet rocket-powered point interceptor- it was fast, had modestly swept wings, was tailless and was known to have good flying characteristics. As a result of their evaluation of the Komet, the government issued the E.18/45 requirement which was awarded to De Havilland for the construction of two small jet-powered research aircraft to study swept wings and the tailless configuration. Designed by De Havilland engineer Ronald Bishop, what was designated the DH.108 used the fuselage of the De Havilland Vampire jet fighter which was stretched and streamlined combined with a new swept vertical fin and new swept wings. This served to save time and effort and like the Vampire jet fighter, the DH.108 had wing root intakes. Unofficially named "Swallow", the two prototypes received the RAF serials TG283 and TG306. TG283 had a 43-degree swept wing and was intended for low speed testing while TG306 had a 45-degree wing and was assigned to high-speed transonic testing. Initial wind tunnel studies suggested that the Swallow would have lousy stall characteristics, so the first of the two to fly, the low-speed assigned TG283, had fixed wing slats and anti-spin parachutes in fairings on the wingtips. It made its first flight on 15 May 1946 and TG306, the high-speed Swallow, made its first flight a month later and had automatic wing slats but no anti-spin parachutes.

VW120, the third DH.108 Swallow that broke the sound barrier in 1948
The test pilot for the DH.108 was the chief test pilot for the company, Geoffrey De Havilland, Jr, the son of the company's founder. After several problem-free test flights it was found that both aircraft lacked the predicted poor stall characteristics and the high speed aircraft, TG306, joined the formal research program on 23 August 1946. By the time De Havilland had safely taken it to 630 mph at altitude with no problems, it was decided he would take TG306 to break the world speed record. On a practice run over the Thames Estuary on 27 September 1946, the aircraft exceeded its structural limits at high speed and broke up, killing Geoffrey De Havilland, Jr, as the aircraft had no ejection seat. John Cunningham succeeded De Havilland as chief test pilot and took over flying the DH.108. To replace the lost aircraft, a third DH.108 was ordered and received the RAF serial VW120. It featured a more pointed nose, revised canopy and cockpit including a Martin-Baker ejection seat, and a higher thrust Goblin engine than what had powered the first two aircraft. Cunningham took VW120 up on its maiden flight  on 24 July 1947. With Cunningham and fellow test pilot John Derry flying VW120, a series of new speed records were set in 1948 and on 9 September of that year, Derry took VW120 past Mach 1 in a dive from 40,000 feet, making the DH.108 Swallow the first British aircraft to break the sound barrier- though it must be pointed out that Derry had for the most part lost control of VW120 during the supersonic portion of the flight but had safely recovered and landed. 

After flying demonstrations at the 1948 Farnborough SBAC air show, VW120 was handed over to the Royal Aircraft Establishment to join TG283 (the first Swallow) in the research program. On 15 February 1950 RAE test pilot Stuart Muller-Rowland was killed when VW120 broke up due to structural failure during a high speed test run. Three months later TG283 was lost and killed its test pilot, RAF Squadron Leader G.E. Genders, when it stalled at low speed and low altitude. 

Despite the loss of all three DH.108s with the loss of life, the data from the Swallow flight test program and the research program at the Royal Aircraft Establishment benefited not just the design of the DH.106 Comet jetliner, but also that of the DH.110 Sea Vixen naval fighter as well.

Source: Military Aircraft Monthly International, Volume 9, Issue 12. "A Deadly Swallow: The short sharp story of De Havilland's DH.108" by Nico Braas, p28-32.


  1. did the dh 108, like the dh 110 have no front spar, relying on a thickened d section?

  2. I thought one was lost because of a bad oxygen system to the pilot?

    1. VW120. Muller Rowland was a friend of my parents. Dad told me that he blacked-out due to oxygen starvation, lost control and the plane broke-up in flight.

  3. I believe that Muller-Rowland's death was originally mooted as the result of oxygen starvation although the severe pitching characteristics of the aircraft were well known within DH. John Derry, who broke the sound barrier in VW120 managed to recover the aircraft on one occasion following extremely violent oscillations. Muller-Rowland's official cause of death was a broken neck although the aircraft actually broke up in flight and scattered wreckage over a wide area, including my home village. The cockpit section coming down in the churchyard. MM

  4. Eric "Winkle" Brown also retrieved himself from the violent oscillation. He had deliberately lowered his seat before the flight so he wouldn't bang his head on the canopy and break his neck. The broken neck and breakup were common to both Geoffrey DH Junior's and Mueller-Rowland's deaths. Why did M-R not learn from the experiences of the other two? Perhaps because he blacked out and slumped over the stick, initiating the oscillation.

    De Havilland joined the official Tailless Aircraft Advisory Committee in September of 1945, around the time that Vampire fuselage TG283 was being filched from the production line, saying that it "would be very helpful to us in our immediate programme".[TAAC correspondence, National Archive] Geoffrey de Havilland Sr. wrote to his old friend and tailless aeroplane pioneer J.W. Dunne in 1946, saying that he was constantly teasing his younger staff that Dunne had done it thirty years ago and had got it to work.[JW Dunne archive, Science Museum]

    In this context, the lethal oscillation was not a defect of the tailless aerodynamics but was due to the then utterly mysterious aileron flutter at transonic speeds and many a tailed American jet joined the DH.108s on the scrap heap for the same reason.