TSR2

by Damien Burke

  VTOL

  Although VTOL had of course been discarded early on in the design process, 1961 changed things a little. Hawker’s P.1127 had flown successfully, and the company was working on the P.1154, a supersonic VTOL strike fighter for the RAF and RN. There was some concern that, if it was successful, this aircraft could threaten the TSR2, as it was capable of carrying out a portion of the TSR2’s role and VTOL would give it vastly more operational flexibility. In response, BAC put together a rough design for a lightweight ‘baby TSR2’ as part of a costing exercise. The idea was to predict costs for a small, cheap, VTOL type (the baby TSR2) and compare them with the costs for the standard TSR2 and a heavier VTOL type with variable-geometry wings.

  The ‘baby TSR2’ was specified as being a single-seat delta-wing type, weighing around 32,700lb (14,840kg) for vertical take-off, with an overload of up to 48,000lb (20,000kg) possible for a short take-off of 1,220ft (372m). Powered by a single RB.168 with reheat and ten RB.162 lift engines, the aircraft would have a combat radius of 250nm (290 miles; 470km) at 200ft (60m) and Mach 0.92, with a weapon load of 2,000lb (900kg). Ferry range would be 1,500nm (1,725 miles; 2,775km). In terms of appearance, about the only remaining aspects of TSR2 were the nose and intake shapes. The VTOL variable-geometry type, apparently a study that was already independently under way (and for which sadly no drawings were included in the report), was to be powered by two thrust engines with provision for thrust deflection, and six lift engines carried in the forward fuselage. Throwing a variable-geometry wing into this mix seems an odd case of overkill.

  This VTOL ‘Baby TSR2’ was a purely paper exercise produced as a baseline for comparison of a minimally-sized VTOL type against the standard TSR2 or a VTOL and VG version. Damien Burke

  VTOL TSR2 General Arrangement. This was suggested as a ‘minimal changes’ solution to incorporating VTOL within the TSR2 to provide a research airframe. A tailplane was thought unnecessary as pitch control would be provided by elevons on the mainplane at high speeds, and via vectored thrust at low speeds. Damien Burke

  The basic response of BAC to the threat of a VTOL type was that at light weights, and operating over shorter ranges, the TSR2 had the capability to operate from short, rough strips and thus carry out the close-airsupport role without the need for a complex VTOL system. The larger TSR2 could also carry more weapons and hit more targets than a single smaller VTOL type, and thus fewer would be needed. Of course, the TSR2 was more expensive, and, as a larger aircraft, more vulnerable to being hit by enemy fire than a smaller type. Against this was the argument that a TSR2 was less likely to be hit in the first place by virtue of its terrain-following ability. As for cost, comments on the value of the TSR2’s additional capabilities (reconnaissance, all-weather attack) and the costs of supporting dispersed operations for VTOL types dotted about the countryside in forest clearings and the like left the report’s readers to make up their own minds about which was better value for money, even before reaching the meat of the report. The graphs showing break-even points for various ‘efficiency factors’ (the number of TSR2s you would need to replace a single VTOL type; always less than one, of course). Even the most basic VTOL type would need to be produced in thousands to ‘save money’, argued BAC, whereas one could simply buy a few more TSR2s and give them the close-airsupport role. Needless to say, BAC’s figures and common sense did not make good bedfellows, and the P.1154 programme was never seriously challenged by any prospect of more TSR2s being purchased instead.

  Variable geometry

  Variable geometry (also known as variable sweep or the ‘swing wing’), had been a pet project of Barnes Wallis at Vickers since 1948, research being funded by the company and some contribution from the government. Wallis had put together a brochure on a revolutionary variable-sweep bomber to be called the Swallow, and spent years trying to get it built, to no avail, as many of the additional features of the aircraft (such as abandoning conventional flying controls and controlling the aircraft via swivelling engines on the wingtips) were believed by most to be a step too far. Early Vickers work on the GOR.339 submission had looked at variable geometry and discarded it as being unnecessary. The requirement could be met by the fixed-wing designs being drawn up, and no weight or range advantages could be found. Similar opinions had shot down Wallis’s own submission of a Swallow variant to GOR.339, and as a result Wallis had turned to the US-funded Mutual Weapons Development Pact in April 1958, which resulted in Wallis and his team visiting the USA in late 1958.

  The USA had already carried out some development flying of experimental variable-sweep aircraft (the Bell X-5 and Grumman XF10F-1 Jaguar), and had found that stability and control were serious problems. Moreover, the weight and complexity of a translation mechanism to shift the wings forward and backward when sweep angle changed, to keep the wing’s aerodynamic centre near the airframe’s c.g., had pretty much brought variable geometry to a grinding halt there. Researchers at NASA Langley studied Wallis’s Swallow with great interest, but found that it exhibited longitudinal instability at relatively low pitch angles when fully swept, and also at moderate pitch angles when unswept. The amount of control available from the pivoting engines was insufficient, and loss of an engine could lead to disastrous loss of control authority in the unswept configuration, when the deflection angle available to the engines would be insufficient to cope with the large moments introduced by the great distance between the engines on each side. Although NASA tried replacing the engines with a pivoting tail surface mounted on the wingtips, this led to even more complexity and did not substantially help stability, as the wing’s longitudinal position still needed to be moved to deal with the changes in aerodynamic centre when the sweep angle was changed. Even more worrying was an apparent tendency for the aircraft to pitch up owing to interference between the variable sweep wing and the fixed forebody.

  NASA Langley took the Wallis Swallow design and basically discarded much of the concept, concentrating on the separate outboard pivots to produce a more practical variable-sweep design, leading eventually to the TFX/F-111. Damien Burke

  However, NASA’s researchers came up with a more successful solution to the stability problem. In a drastically simplified version of the Swallow, a larger inner swept portion of the wing or fuselage forebody was fixed, a smaller outer section pivoting about a point at the leading edge of the fixed portion. The swivelling engines were replaced by conventional engines in the fuselage and a conventional fin. No translation was needed, as the moving part of the wing was so much smaller and the resulting shift of the aerodynamic centre was of much lesser magnitude. The Americans thought it was a breakthrough, but Wallis was less than impressed by what he saw as a minor piece of tinkering with the wing and an invalidation of almost the entire remainder of the Swallow concept. But NASA’s researchers could not have cared less, soon realizing that Wallis’s laterally separated pair of pivots were the breakthrough they had been seeking in their own attempts to create a successful variable-geometry design. The most important part of their ‘new’ concept was the pair of outboard pivots, and with these the pivoting wing could equally be applied to a much more conventional fuselage layout, discarding entirely the Swallow’s large forebody and the wing-mounted engines that doubled as a means of control. This also gave NASA the excuse it needed to claim that this was all its own work, and it wasted no time in providing the results of its research to the American aviation firms. Wallis did not realize it, but his attempt to gain funding from the Americans had helped to doom the GOR.339 project.

  While Wallis returned to the UK disappointed with the Swallow’s reception, Vickers took note of the NASA findings, backed up by its own windtunnel research, and sketched out a version of the TSR2 fitted with what was now described as the ‘NASA wing’. However, with a history of years of frustrated effort behind the company in this field, it was clear that the benefits of this kind of wing could be wiped out by development problems and weight gains, and variable geometry was put aside, to
be looked at again in the future if circumstances permitted. In October 1960 the Minister of Defence, Harold Watkinson, visited Vickers and, concerned that the press would portray TSR2 as on obsolete design when the Americans were working on a tactical bomber employing variable sweep, knocked up a press release that did nothing more than draw the attention of the press to precisely that! The Ministry also asked Vickers to provide a list of its variable-sweep patents to give to the Americans, so that the Americans would have ‘no difficulties’ in paying for design rights if they did end up using a variable-sweep design for their new bomber.

  By late 1962 the American TFX programme was well under way, and Vickers was a little surprised to find it had a variable-geometry wing, in fact the NASA wing, and that no offers of payments had been made by the Americans. It appeared to Vickers that the 1958–1959 joint research programme had been nothing less than a successful attempt by the Americans to grab a load of useful information for their own purposes. Vickers instructed its American lawyers to find out how successful it might be in pursuing a legal case against General Dynamics, but there was no support forthcoming from the UK government, and NASA’s researchers had successfully applied for patents on ‘their’ variable-sweep wings (US patents Nos 3,087,692 and 3,053,484). After months of work Vickers eventually decided that these patents were couched in such general terms that it would be difficult to prove a direct infringement of Wallis’s own patents or invalidate the patents on the basis of prior publication of relevant work, because much of it was only in print in brochures of a ‘Secret’ classification, which could not count as prior publication. Vickers reluctantly drew back from a legal battle, keeping a wary eye on TFX and its incarnation of variable-geometry nonetheless in case any royalties could be squeezed out of the Americans.

  Using the ‘NASA wing’, Vickers-Armstrongs drew up a VG version of the TSR2, but with the airframe having been designed for a challenging combination of short strips and supersonic performance already, the advantages offered by the VG wing were not thought to be worth the extra development effort. Damien Burke

  In late 1963 the subject of variable geometry was raised again. Since the early work on incorporating the ‘NASA wing’ in 1960, both OR.343 and design of the airframe had undergone various changes, and Ministry attention had now also turned to the American TFX and its variable-geometry wing. Accordingly, it was felt worthwhile to revisit variable geometry, particularly as Vickers was also still working on the OR.346 naval requirement for a variable-geometry fighter (Types 581, 583 and 589, all of which would come to naught in the end). During January 1964 a brief study was carried out into the effect of fitting a variable-geometry wing to the TSR2, this time using the wing planform designed for the Type 583. This had a maximum sweep of 74 degrees and was calculated to incur a weight penalty of some 5,000lb (2,300kg). However, with an accompanying reduction in fuel capacity, the take-off weight was actually only going to be some 3,000lb (1,400kg) greater than that of the standard TSR2 and there would be some serious performance benefits. The take-off ground roll was reduced by 23 per cent and landing approach speed by 14 per cent; subsonic endurance was increased by 45 per cent, and range by 18 per cent. The only downer was that supersonic range would be reduced by 8 per cent. The indications were that fitting the basic airframe with a variable-geometry wing could be a useful exercise, and could possibly be carried out as a major modification to existing airframes, as changes could be limited to fitting a redeveloped centre-fuselage area. However, the fact remained that the aircraft as already designed looked like it was going to meet the existing operational requirement, and changing things would only delay its entry into service. The conclusion was obvious. Variable geometry was a distraction, and no further efforts were made to gain support for a variable-geometry TSR2.

  As for the TFX, General Dynamics, unaware that it had come very close to being taken to court by BAC, welcomed a delegation from the British company in October 1964. The members had a good look at the aircraft and gleaned a lot of information about its development, and the decisions taken along the way, which was all of particular interest to BAC for future projects. The president of General Dynamics, Frank Davis, later claimed that the TFX’s variable-sweep wing owed nothing to Wallis’s work, and NASA published a working paper in 1966 entitled Summary of NACA/NASA Variable-Sweep Research and Development leading to the F-111 (TFX), in which it was claimed that: ‘Late in 1958 a research breakthrough at Langley provided the technology for designing a variable-sweep wing having satisfactory stability through a wide sweep angle range without the necessity for fore and aft translation of the wing’, and that ‘… the variable-sweep concept [was] born at NACA/NASA (Langley) …’. A thorough read of the paper reveals plenty of references to Vickers and the Swallow, but little hint that the Swallow was actually the research breakthrough on which all of NASA’s work was based.

  Another study into applying VG to the TSR2 was carried out in January 1964, this time using the wing planform drawn up for the Vickers Type 583, a naval strike fighter project. Damien Burke

  TSR2 in the strategic role

  In May 1960 the Minister of Defence, Harold Watkinson, had expressed the opinion that ‘we could give the TSR2 an increased strategic capacity by fitting it with some sort of missile’. Not only would this make the aircraft better value, but it could also interest the Americans. This kick-started a sequence of events that led to TSR2 moving away from its original purely tactical role to a wider-ranging role that would include the delivery of larger nuclear weapons at distances far removed from the original battlefield scenarios. The MoS (soon to become the Ministry of Aviation (MoA)) was tasked with advising what was possible. George Edwards at Vickers was asked for his opinion, and was, of course, keen to assist; adding a capability to keep the customer happy is not something you normally refuse. The MoA and the Air Ministry started off by outlining some possibilities to arm the aircraft with a strategic weapon.

  Blue Steel was out: there was insufficient ground clearance to mount it underneath, and costly structural modifications would be needed to carry it on top of the aircraft. Performance and range would suffer badly. A pair of Skybolts could possibly be carried on top of the wings without so much aerodynamic penalty, but structural modifications and changes to the aircraft’s fin would be needed, and costs would be substantial. There was also little point, as the combination would be inferior in most ways to the planned Vulcan and Skybolt combination, with low-level use of the missile leaving it unable to use its star tracker for navigation (because of likely cloud cover), and thus reducing its accuracy substantially. Development of a suitable ballistic missile to be carried in the TSR2’s weapons bay, and capable of navigational autonomy over a long distance, would be a task approaching the size and cost of the TSR2 itself. A simpler propelled bomb with guidance system and 50 miles (80km) range when flown at Mach 2.5 at 200ft (60m) would be easier but was still a technical challenge, and would cost somewhere between £60 million and £150 million.

  The TSR2 already had some strategic capability with overload fuel and twin WE177 carriage, though vulnerability over strategic targets would have been an issue. BAE Systems via Brooklands Museum

  In September 1960 BAC produced a report entitled A Study of the Use of the TSR2 in the Overload Condition to Fulfil Other Roles, covering the various possibilities as it saw them, and summed up all the suggestions it had been putting forward up to this point. The aircraft could be safely overloaded to a weight of 120,000lb (54,500kg), and this would enable a significant increase in armament and/or fuel load when it was operated from large airfields. A long-distance sortie was postulated, carrying a strategic nuclear weapon (a bomb fitted with the Red Snow warhead used in Blue Steel and, in modified form, slated for use in Skybolt) with high-level weapon delivery in which targets up to 1,800nm (2,070 miles; 3,330km) from base could be hit. If the 200nm (230 miles; 370km) entry into and exit from the immediate target area was at low level, the combat radius would be reduced to 1,600nm (1,
840 miles; 2,960km); still a respectable figure. This would require the carriage of underwing and ventral fuel tanks, all jettisoned before the attack.

  English Electric was already developing a surface-to-surface nuclear missile for the Army, Blue Water, and the study found that this could be modified for TSR2 carriage, half-buried in the bomb bay (along with a bomb-bay fuel tank), and would have a range of 70 to 100 miles (110 to 160km), either low level all the way at Mach 1.5 or taking a more ballistic trajectory, in which case it could reach Mach 3 at 70,000ft (21,000m). An alternative was to create a new missile more suited to TSR2 carriage, using Blue Water components repackaged as appropriate. This would have larger wings and a pair of rocket motors side by side, giving improved lift/drag characteristics and longer range; up to 200nm (230 miles; 370km) from a low-level launch, with the TSR2 itself able to take the missile up to 1,340nm (1,540 miles; 2,480km) from base before launch. Development, however, would probably require as much time and effort as the entire existing Blue Water programme.

  The third missile suggestion from BAC was to carry a single-stage ballistic missile carrying a Polaris re-entry head. The largest possible such missile that could still fit in the TSR2 weapons bay would have a range of 350nm (400 miles; 640km), but this sort of distance risked unacceptable navigational inaccuracy, so it was suggested that a smaller missile with a range of 200nm would be more viable. This would weigh 5,500lb (2,500kg), have a diameter of 26in (66cm) and climb to a maximum altitude of 300,000ft (90,000m) before its trajectory took it back down towards the target. From a high-low-high sortie this could mean a total combat radius of 1,740nm (2,000 miles; 3,220km). However, if the TSR2 stayed at altitude and launched the missile from there, the missile would gain an extra 100nm (115 miles; 185km) and the total combat radius would be a whopping 2,000nm (2,300 miles; 3,700km).