by Damien Burke
Fin
The fin was an all-moving unit of similar construction to the tailplanes, though it was a slab unit, rather than being fitted with any trailing-edge control surface. It was mounted on an inclined spigot attached to the rear fuselage, access to the outer bearing and securing nut being available through a circular access panel on the port side of the fin. A sprung mass-balance was fitted within the leading edge cover at the root end of the fin, and the leading edge also housed a ram-air intake that directed cooling air to the braking-parachute container. The complete trailing edge of the fin, like the tail plane tips, was a honeycomb unit, and the tip of the fin was a dielectric cover for the UHF aerial.
The fin structure. This illustration predates the addition of a cooling intake in the leading edge, which ducted outside air to the parachute housing in the rear fairing.
The fin of XR220. A blanking plug is inserted into the cooling intake in the leading edge. Damien Burke
Undercarriage layout. BAE Systems via Brooklands Museum
The nose undercarriage. BAE Systems via Brooklands Museum
Undercarriage, braking parachute and airbrakes
The nose undercarriage consisted of a conventional twin-wheeled steerable oleo strut retracting rearwards into the fuselage. The main oleo incorporated an extension that could be selected by the pilot to give an additional 30in (75cm) of extension to assist with nose lift on short take-offs. This extension was automatically contracted during the retraction process, but was slated to be removed from production aircraft because it proved to be of little assistance in reducing the take-off roll, the tailplane authority proving to be greater than predicted. The leg was powered by the No.2 hydraulic services, emergency operation being provided by the No.1 services. There were four bay doors, split into port/starboard pairs and forward/aft sections. The smaller forward sections remained open with the nose-gear leg dropped, while the larger aft sections sequenced closed when the leg was locked in place. These doors could also be opened on the ground to give access to the nose-gear bay, which included a refuelling control panel.
The main undercarriage was considerably more complex, the design being driven primarily by the rough-field requirements and the limited space available in which to stow it. A main shock-absorber strut, braced at the top, was connected via a swivelling ankle joint to a two-wheel bogie beam with wheels mounted in tandem. Hydraulically operated disc brakes were fitted to each wheel. A hop damper was attached to the rear of the bogie beam in an attempt to eliminate any tendency for the aircraft to bounce from one wheel on the bogie to the other. A retraction jack at the forward end pulled the bogie upwards to the same angle as the shock-absorber strut and then canted the bogie assembly sideways to match the lateral angle before retracting it forwards into the large bay. The three main undercarriage doors on each side consisted of a small aft door, linked to the leg itself and remaining open with the undercarriage down, and two larger underside and side doors that were only open during the retraction or extension cycle. As with the aft nosewheel doors, the main undercarriage doors could also be sequenced open on the ground to give access to the bay. Items requiring immediate groundcrew access on shut-down, such as the master armament break safety switch, were located in the aft section of the bay, so that access would not be hindered by any possible door problems.
XR219 with the nose gear leg in the extended position. Unexpectedly powerful taileron response meant that the extension feature was to be deleted from production airframes. BAE Systems via Warton Heritage Group
The nose undercarriage of XR220 viewed from behind, looking forward. The bay roof doubles as a fuel tank wall, with dished recesses to accommodate the wheels. Damien Burke
Nose undercarriage functioning, also showing the extended strut initially thought necessary to meet the required take-off performance. Unexpectedly powerful tailplanes rendered this unnecessary, nosewheel lifting occurring at speeds 30 to 40kt (34 to 46mph; 55 to 74km/h) lower than predicted, so the extending nose gear strut was to be omitted from later airframes. BAE Systems via Brooklands Museum
The nose undercarriage doors and operating mechanism. BAE Systems via Brooklands Museum
The main undercarriage (port side), before the addition of the tie strut to alter the bogie trail angle. BAE Systems via Brooklands Museum
The port main undercarriage of XR220. As can be seen, this aircraft was fitted with the ‘Aylesbury tie’, as was XR219 for its last few flights. Damien Burke
Main undercarriage functioning. BAE Systems via Brooklands Museum
The main undercarriage doors and operating mechanism. BAE Systems via Brooklands Museum
The starboard main undercarriage bay of XR220, looking aft. The shape of the bay was determined by the walls of the intake tunnels, weapons bay and the fuel tanks above. The three orange cables on the right lead into the weapons bay through holes in the panel here. This was a removable panel designed to provide access for winching items such as bomb suspension units into the weapons bay. Damien Burke
The brake-parachute installation. The parachute container was simply rolled into place, a considerable improvement on the Lightning’s lower-fuselage parachute installation, which could require riggers to lie on the ground and push the pack into place with their feet while swearing like navvies. BAE Systems via Brooklands Museum
The TSR2’s brake parachute in fully developed form, seen here at the conclusion of the first flight. BAE Systems via Warton Heritage Group
Initial ground-retraction tests of this complex undercarriage threw up several sequencing problems, and as the first flight was to be of very short duration, no attempt was going to be made to retract it initially. (This was also in keeping with traditional Vickers first-flight practice.) One of the most famous of the TSR2’s problems originated with this undercarriage design, as on the first landing and almost every subsequent one a shattering lateral oscillation set in at touchdown which was so pronounced at the pilot’s position that vision and any semblance of control was lost for the several seconds’ duration of the oscillation. A tie strut fitted on the first aircraft to alter the bogie trail angle and detune the whole assembly was found to be very effective in reducing this problem, and was also fitted to the second aircraft. Production aircraft would, at the very least, have incorporated this modification, but it is likely that more substantial redesign of the main undercarriage would have been required to solve the problem completely.
Braking parachute
The TSR2’s braking parachute, a two-stage, two-diameter ribbon parachute, was installed in the rear fairing between the engine exhausts. The central part of the aft end of the fairing hinged upwards forming the ‘beak door’, and the two drogues, primary and emergency (both of 6ft (1.8m) diameter), would stream out, pulling the main parachute out with them. The parachute would deploy initially in a reefed condition, to a diameter of 16ft (4.8m); full deployment created a 28ft (8.5m)-diameter parachute once airspeed had fallen below 135kt (155mph; 250km/h) (assuming there was no crosswind condition).
The choice of a two-stage parachute was dictated by the wide variety of landing speeds the aircraft could experience, from the lightweight landing with full flap at 130kt (150mph; 240km/h), to the much more exciting case of flap and ‘blowing’ failure at heavy weight; 220kt (250mph; 400km/h). Failure of the reefing mechanism in this case would test the parachute attachment structure to its ultimate design limits; in practice, some damage would no doubt have occurred.
Whereas in the Lightning the brake-parachute attachment was on top of the rear fuselage (an area designed to be strong to cope with rudder loads), on the TSR2 the fin was all-moving and the spigot on which it rotated was much further forward. Attaching the brake parachute here would have fouled the fin, so the attachment was placed below the fuselage, in line with the fin spigot. To provide clearance for the cable, a small fairing protruded below the valley between the jetpipes. A rail running from this fairing to the rear of the aircraft enabled the first section of the cabl
e to be held straight regardless of the crosswind, spreading the load and easing the stress on the attachment point.
The parachute was activated by pulling the brake-parachute handle in the pilot’s cockpit. This was to be a two-pull handle. The first pull would open the parachute beak door and allow the drogue parachute to be ejected by a spring and deploy. The next stage was to be either automatic, in which case switches on the undercarriage would trigger the removal of a pin allowing the secondary drogue and primary parachute to be pulled out by the already-deployed primary drogue; or manual, in which case a momentary release of pressure on the parachute handle, followed by a further pull, would allow the secondary drogue and primary parachute to deploy at pilot command. The first aircraft, XR219, had a lower standard of installation using a single-pull system, both steps being carried out in a single sequence, and no automatic mode.
Airbrakes
Four airbrakes were mounted on the rear-fuselage quarters, each operated by a screw-jack that was shaft-driven by a single central unit to ensure symmetrical travel, the shafts running through sealed tunnels in the rear-fuselage fuel tanks. The airbrake control unit was hydraulically powered and electrically controlled, both systems being dual-channel so that a single failure would not disable the airbrake system. An automatic blow-back system was incorporated to protect the airbrake doors and jacks from overload in the event of extension beyond their limiting airspeed. If airspeed increased with the airbrakes out, the increased loading would cause them to retract automatically, and extension commands would be ignored if extension would result in overstress. Unfortunately the choice of screwjacks was not a happy one. On the development-batch aircraft their inherent inaccuracy made it impossible to adjust the airbrakes with sufficient precision to ensure that all four were equally closed without the possibility of one or more overrunning the stops and being damaged, or damaging the fuselage structure. An interim solution was to set the ‘closed’ position so that the air-brakes were actually open by about 1.5in (3.8cm), and XR219 made all but four flights (17 to 20) with the airbrakes cracked slightly open like this.
Brake parachute anchor point and railing. This enabled the first part of the cable to be held straight regardless of crosswind, reducing side loading on the anchor point. Damien Burke
The airbrake mechanism. The choice of a centrally controlled and thus hopefully synchronized mechanism using screwjacks was an unhappy one, and production aircraft would probably have had individual hydraulic rams for each petal. BAE Systems via Brooklands Museum
Slipping-clutch units were to be introduced into the final driveshafts to allow some limited self-adjustment at final closure, but there was also a proposal to replace the screwjack-and-driveshaft system with hydraulic rams. The existing control unit would have been replaced by a purely electrical one operating push-pull rods connected to hydraulic servo valves positioned near the new airbrake rams. This would have solved the problem of unequal retraction, as the ram’s jack-body length could be adjusted to calibrate a particular airbrake door to its recess. As a bonus, time for full extension of the airbrakes would have been cut by 1.5sec to 4sec, and £12,000 would have been saved on each aircraft. While the airbrakes performed well in flight testing over the first 60 degrees of their extension range, producing no serious trim changes or buffeting, the last 5 degrees of travel produced considerable buffet, and an investigation into the use of perforated airbrakes was due to begin when the project was cancelled.
The interior of XR220’s starboard lower airbrake, showing the large screwjack and the smaller driveshaft linking this brake to the other three. Damien Burke
Airbrake, flap and taileron flap mechanisms. The inherent inaccuracy of screwjacks was not a problem with the wing flaps or taileron flaps. BAE Systems via Brooklands Museum
Arrester hook
A basic design for an arrester hook was drawn up during the overall airframe design stage, but this was not a popular addition as it introduced weight and drag penalties just when the design was having balance problems. Freddy Page said later that he ‘… had never thought the hook was a sensible suggestion’, and it was soon dropped as an easy and early economy measure to take.
As the weight of the aircraft had increased throughout the type’s development, so take-off and landing runs had also inevitably increased. The ability of the brakes and brake parachute to stop a loaded TSR2 in emergencies was judged insufficient in all required situations, so the RAF asked BAC to quote some figures on weights, and any possible production implications of adding a simple ‘one-shot’ arrester hook. The answers arrived just when the engine accessories bay situation began to complicate matters, so work on a design study was not carried out until late 1964, being completed in November and submitted to the MoA for approval. The arrester hook would not have been incorporated in most of the development-batch aircraft, but was expected to be fitted from the preproduction aircraft onwards, had the project not been cancelled. In fact the Ministry never bothered to seek funding from the Treasury for this modification owing to the uncertainty over the project’s future, and in February 1965 the Air Staff postponed the implementation of the hook as an interim money-saving measure, intending to reintroduce it later, possibly for full production aircraft only.
Crew space
The pilot’s and navigator’s compartments together made up the ‘crew space’, a term coined early in development so that discussion of the two cockpits would always treat them as a single entity, without the accidental overlooking of any aspect. English Electric’s mock-up of its P.17 design proved useful in the early stages of deciding crew space layout, along with a Scimitar at Vickers’ Wisley plant, as this aircraft had a similarly sized cockpit to that of TSR2.
Early arrester hook design proposal. The hook idea was dropped at an early stage, but the RAF was keen for the aircraft to have one, and production aircraft would probably have been fitted with a simple ‘one-shot’ hook much like that fitted to later marks of Lightning. Damien Burke
An early mockup of a proposed pilot’s cockpit for the TSR2. Some elements of the final layout can be discerned, such as the warning captions along the coaming edge, the large attitude indicator and moving map display (showing admirable attention to detail, as it is centred on a typical target area, south-eastern Poland, near the Ukrainian border). BAE Systems via Brooklands Museum
This was to be an all-weather strike aircraft, and a forward view for the navigator was therefore considered unnecessary. Any visual fixes were to be obtained by the pilot, who would be provided with a colour moving-map display. Roland Beamont did express doubts at an early stage regarding the abilities of a pilot to map-read at high speed and low level, but Vickers thought that some assistance from the navigator (who could not see where he was going), calling out landmarks in advance of their arrival, would suffice.
The choice between a central joystick and a small sidestick was made quite early. Trials carried out on a Gloster Meteor fitted with a sidestick had generally gone well, though one obvious drawback was the use of the stick if the pilot was wounded or injured in the appropriate arm. Beamont considered a sidestick preferable, basing his reasoning on arm fatigue encountered when holding a normal stick for long periods, particularly at low level. However, US experiences with sidesticks on the Convair F-106 had been mixed, and the crews considered central sticks generally preferable. Test pilot Brian Trubshaw suggested that if a central stick was chosen, it should be topped with a W-shaped wheel to decrease the amount of the centre of the instrument panel that was obscured. This, of course, ended up being used on Concorde. Some months later a U or Y shape was being proposed, and it gives some idea of the amount of time wasted in cockpit meetings that one Minute records a request from the Ministry officials to Vickers to ensure that ‘the control column movement was such as to be free from fouls against pilot or aircraft structure’. One can only wonder at the mindset that fears a firm may design a control column that cannot move freely!
Crew vision
/> A conventional windscreen was chosen, despite the disadvantages of the two struts blocking some of the view forward. At the time the optical qualities of a one-piece curved windscreen were suspect, and a two-piece V-shaped windscreen would have made the required HUD an even trickier proposition.
The layout of the TSR2 pilot’s cockpit as of February 1965. By this point the various warning captions had been moved to a central warning panel, the coaming edge being used for commonly used controls. BAE Systems via Brooklands Museum
The pilot’s cockpit under night lighting conditions. BAE Systems via Brooklands Museum
The pilot’s cockpit layout of the Type 579 (pre-production aircraft). The development-batch aircraft differed in having various items omitted; for example, XR219 had no AFCS or role panels. BAE Systems via Brooklands Museum
The pilot’s control column handle. BAE Systems via Brooklands Museum
An early mockup of the navigator’s cockpit, with TV display on the left (showing Heathrow airport!), moving map (showing southern Czechoslovakia) and fix controls on the right, with a blank area for an SLR display to the left of the moving map. The CRT on the right appears to be for the FLR display. BAE Systems via Brooklands Museum
