Aviation is progressive, iterations of design and procedure evolving to defeat limitations. Robin Evans investigates how the unique problems posed by the development of Concorde required exceptional solutions, many of which are now commonplace in modern aircraft.
Innovations gradually recede into obsolescence but Concorde is that rare thing, a benchmark that remains unequalled and unsurpassed: both past icon and future vision. It was retired in 2003 and first flew in 1969 yet its influence is still felt today on more recent aircraft designs. Plus supersonic speed still has its appeal with a crop of start-ups looking at potential civil aircraft designs. Concorde was an incredibly futuristic aircraft that still inspires.
Concorde was forged in the heady crucible of post-war engineering: a partnership between Sud Aviation (later Aérospatiale) and the British Aircraft Corporation (BAC). In November 1962 a treaty was signed that unified the intent and early designs (the Super-Caravelle and Bristol Type 223) of both parties. The importance of partnership was echoed in the formation of Airbus Industrie later in the decade. The history of Concorde is intrinsically linked to Airbus (and also to Toulouse and Filton) as the company maintained parts support for Concorde – the withdrawal of support being another catalyst for eventual retirement.
The proposal for a Mach 2.0 airliner carrying 100 passengers required extraordinary solutions. Motivated by an agreed lack of break clause (politically at least, here was a take-off that couldn’t be rejected) the Anglo-French team made unprecedented efforts on slide rules and first-generation computers. Of the advances to rise from drawing boards on both sides of the Channel, some were new, others already in development or borrowed from military applications. Some remain today, pioneering what we take now for granted.
The flight deck is accessed down a long tunnel, the port side occupied by a wall of avionics and circuit breakers. Facing this is the Flight Engineer’s huge panel on the starboard side of the flight deck, facilitating face-to-face discussion with the Captain on the port side. Supersonic thermal expansion occurred throughout the fuselage (requiring telescopic hydraulic joints and electrical cabling slack) but the effects were only visible between the Flight Engineer’s panel and the adjacent bulkhead. A standard 1970s panel of analogue instruments reveals a few subtleties, such as the machmeter and nose droop control under the glareshield. The reheat switches are white ‘piano keys’ under the thrust levers, engaged by the Flight Engineer. The hydraulics use a convention lifted from the de Havilland Comet to aid identification: two primaries and one secondary – green, blue and yellow. An identical convention is employed by Airbus today.
Centre of gravity (CG) position is essential for weight and balance purposes. It is also necessary in airliners to optimise efficiency, and critical in Concorde. Approaching Mach 1.0, the rearwards movement of the centre of pressure (the point at which the lift is considered to act through the wing) from the CG causes a nose-down pitch. Conventionally, a noseup order from the tail control surfaces or trim-tabs would counteract this, but the subsequent drag would restrict supersonic flight and push the controls towards the edge of their authority. Instead, control was maintained by pumping fuel between tanks along the fuselage, particularly into the tail. In practice, the pressure centre moved about 5ft (1.52m), requiring a corresponding fuel transfer of around 20 tonnes; the exact CG position is indicated in the flight deck.
Early research related supersonic drag to wingspan: the origin of the slender wing of the Lockheed F-104 Starfighter. This produced little lift at low speed, resulting in high approach speeds and long landing distances. The game-changing potential of the ‘slender delta’ was realised by the Royal Aircraft Establishment (RAE) in the 1950s, identifying lift-boosting vortices at high angles of attack. This ‘vortex lift’ was a function of wing length – extending the wing rearwards unlocked a balance of low-drag, supersonic flight and sufficient lift at low speed. Three final variations were assessed, favouring the iconic, curving ‘ogee.’ Looking down the leading edge and seeing the wing contour in several dimensions (sweeping, twisting and drooping) gives an idea of what an achievement it was to design and build.
So perfect was the design there was no need for slats and flaps, unthinkable on an airliner in normal operation. Despite British wing-building pedigree (Hawker-Siddeley was brought in for the inaugural Airbus A300 in 1974) the wing was French-made. A process of milling and shaping specialist aluminium alloy produced significant weight saving and finer tolerances, avoiding the weaknesses of welds or riveted joints. The constant drive for lightweight construction seems particularly prescient in our fuel price motivated climate. Externally, the leading edge of the wing is so sharp there was no room for navigation lights; these are mounted inside and fed to the leading edge by fibre-optics. In common with predominantly military deltas (and also the only direct comparison, the short-lived, Soviet Tu-144) it has only a vertical tail. Removing the horizontal stabilizer minimises drag, instead control was provided by six ‘elevons’ on the wing trailing edge. These provide partial trimming (along with fuel balancing) and via differential deflection all pitch and roll. The individual deflections of controls were indicated on an instrument called the ‘Icovol’ (French origin ‘indicateur a vol’) on the flight deck. A digital version of this instrument is the basis of the Airbus ECAM (Electronic Centralised Aircraft Monitoring) Flight Controls display page. Flight controls were fly-by-wire, a first for a commercial airliner – control inputs digitized and fed to hydraulic actuators electrically, not mechanically. In future, the removal of pulleys and cables would save much weight, but physical controls remained for this pioneering system, though unused in normal operation. With one exception, digital systems were analogue (sine wave) rather than digital (binary). The first all-digital fly-by-wire system became the heart of the A320 flight envelope protection system, first flown in 1987.
Huge forces were involved, not only in mass (maximum take-off weight was 185 tonnes) but the speeds required for take-off. As the speed and lift generated by a conventional wing increase, the undercarriage is progressively unloaded: this phenomenon can be observed on take-off in turboprops as the landing gear struts gradually extend. On Concorde, the vortex lift generated only became significant at rotation speed; up until that point there was little lift to progressively overcome the aircraft weight. Another problem specific to Concorde was that steel brake discs tended to fuse together after absorbing the energy of stopping such a mass from high take-off speeds during an aborted take-off at maximum weight. Messier-Dowty produced the undercarriage while Dunlop developed pioneering carbon brakes. These electrically commanded, hydraulically operated, analogue brake-by-wire systems were lighter, more powerful and cooled by internal fans. Innovative at the time, they are standard fit on modern airliners.
The characteristic nose-high approach attitude was a function of the wing geometry, severely reducing forward visibility from the flight deck. The only possible solution was to droop the nose to 12.5° for approach and landing, with an intermediate 5° position aiding taxiing and take-off. Furthermore, as the ground clearance getting airborne dictated the undercarriage length, this meant gear legs too long to fit into the available space. Again, an unconventional solution arose: when commanded up, the landing gear first broke a geometric lock that retracted the shock-absorber assemblies into the gear legs, thus shortening them. A tailwheel completed the ‘four greens’ necessary for landing and offered a sacrificial component; excessive pitch on touchdown would otherwise result in ground contact of the engine nozzles. An interlock isolated the undercarriage while the visor was raised, preventing inadvertent deployment at high speed.
Fuselage and Materials
At subsonic speed there is only a slight drag penalty from an increase in fuselage width, handy for widebodied aircraft. Approaching Mach 1.0 however, the physics change and a slender fuselage (denoted by ‘Fineness Ratio’ – the ratio of the length of a streamlined body to its maximum diameter) becomes critical. A specialist aluminium alloy, RR.58, patented by Rolls-Royce was used as a compromise between cost, ability to be machined and good temperature resistance. Production models typically clocked up 20,000 flight hours (about half the design life) but were also monitored by a ‘reference flight’ index that factored cycles based upon weight. Additional fatigue modelling accounted for the thermal cycles experienced: initially cooling at altitude, then heating upon supersonic acceleration, reversing upon descent and deceleration. Kinetic heating restricted the maximum speed (nose and leading edge temperatures typically peaked at 130°C and 105°C) but at Mach 2.0, heating was also by solar radiation, hence a white livery to reflect maximum heat. With increasing speed, the balance changes, requiring more heat to be radiated, hence the colour of the SR-71 Blackbird.
The term ‘powerplant’ is most appropriate for a unique combination of three elements: intake system, engine and exhaust nozzles. Olympus: even the name for the heart of this unholy trinity evokes the crackle of thunder, tracing a pedigree from the Avro Vulcan and BAC TSR2. Rolls-Royce produced the final Mk.593 using titanium and nickel-based superalloys to resist huge temperatures and compression ratios. Engines were analogue thrust-by-wire, the predecessor of Full Authority Digital Engine Control (FADEC) fitted even to light aircraft today.
The intakes are huge, containing ramps that would extend inwards beyond Mach 1.0 slowing the incoming air to a speed the engines could ingest. As a by-product, this also compressed the airflow, boosting the efficiency of the system. In a world first, ramps and their related ducts were digitally controlled (computers lifted from BAC guided weapons systems) due to the critical nature of their precise position in response to Mach shockwave onset.
Snecma Moteurs contributed the reheat and nozzle mechanisms. A variable geometry nozzle allowed reverse thrust upon landing and was synchronized with the intakes to boost system efficiency. Reheat was used for take-off and transonic acceleration above Mach 1: design elegance and the hugely efficient intake system allowed a sustained supercruise (without reheat), a feat only matched in the civilian market by the Tu-144. An unusual problem was detected very late on: vibration of engine No.4 caused by the opposing rotations of the wing vortex and the engine compressors. The port engines received a vortex rotating the other way and No.3 was inboard, further from the leading edge airflow. A modification of inlet vanes and a thrust restriction at low speeds resolved this. On take-off, the reheat flame of the No.4 engine could initially be seen to glow weaker than the other three: the relevant gauge on the flight deck had a coloured tab fitted as a reminder.
Whilst already understood prior to Concorde, actively mitigating the ‘bowing’ of the Olympus main shafts was critical, given their fine construction tolerances for optimum performance. A function of metallurgical diversity, variable rates of thermal expansion could cause a slight warping along the length of the shaft, resulting in vibration and more insidiously, metal fatigue. This effect was most apparent post-shutdown after several hours on the ground, when the effects of variable cooling were most pronounced. To combat this phenomena, engines could be started and latched at a speed below idle, allowing all components to equalise: hence the flight engineer’s de-bow function. A pre-start motoring sequence for thermal stabilisation still occurs on some jet engines, including the CFM LEAP-1A fitted to the A320neo.
There is no auxiliary power unit (APU) as the bulk of the tail contains the tail-wheel unit and fuel trim tank, impractical in the high intensity, faster rotations of modern airline operations. For the flight crew Concorde offered a full-regime autopilot and linked autothrottle, permitting a reduced workload in all phases of flight. T his was harnessed to a complex air data computer that automatically monitored parameters and fed them to related systems. Both are key in the modern, integrated flight deck.
Aerospace has changed a lot since the 1960s; progress now focuses upon metrics such as economy, turnaround times and load factor. Appropriately for the decade concerned, there are clear parallels with President Kennedy’s speech that instigated Project Apollo: “we choose to go to the moon…not because it will be easy, but because it will be hard”. Never before had the giant leaps of a terrestrial aircraft pioneered so many smaller steps along the way.