by Damien Burke
First flight of first pre-production aircraft (No. 11) October 1966
First flight of first aircraft for RAF April 1967
First CA release date (nuclear strike role) January 1968
Full CA release date (all roles in specification) June 1969
The size of the RAF’s TSR2 force, once envisaged as a total buy of 300 (170 frontline) aircraft, had dropped to a total buy of 138 (106 front-line) by 1963, and 110 (74 front-line) by January 1965. During those last few months of desperate money-saving studies there had been suggestions that as few as fifty would be purchased, with the balance replaced by Buccaneers – or F-111s. Worse was to come. On 6 April 1965, while BAC factory workers worked on TSR2 assembly and listened to their radios to find out what effect the Budget would have on their pay packets and the cost of their beer and cigarettes, they found out instead that the TSR2 project was cancelled forthwith.
CHAPTER FIVE
Flight Test Development
Flight test programme
BAC’s initial flight test programme had the nine development-batch aircraft carrying out an intensive flying programme of around 10hr per aircraft per month, flying by day and being serviced by night, to a total of 2,000hr of flying over 33 months (increased from a very early estimate of 1,200hr). This phase was to be carried out by BAC, with the A&AEE getting some previews to enable it to get a leg-up before its own flying began. This would be undertaken on the first five pre-production aircraft, at around the twenty-four-month point, bringing the aircraft to CA release over 500hr of flying and nine months. The final six pre-production aircraft would then enter service with the RAF for intensive flying trials in the nuclear strike role and the investigation of any problems to be expected in service.
There was some criticism that this plan was unrealistic, and a BAC team visited the USA in October 1962 to study the procedures used in the development and release to service of new aircraft. Contractors associated with the production of the North American A3J Vigilante, Grumman A2F Intruder and F-105 Thunderchief were consulted. Impressed by American methods, the team suggested a complete revision of BAC’s own test plan, including closer integration of MoA and RAF personnel with the company testing team, double day shifts to enable flying and servicing to continue simultaneously, and establishment of the first RAF squadron at Boscombe Down, to make early use of BAC and A&AEE flight-test experience. While the average utilization of 10hr per aircraft per month would be kept, the overall flying time to CA release was reduced to 1,500hr, with staged clearances enabling the RAF to begin flying the type earlier. Three fewer aircraft would be needed. Around 50 per cent of the flight testing would be done by BAC, 35 per cent by the A&AEE and 15 per cent by a joint BAC/MoA team. The MoA was naturally upset by this, describing BAC’s proposals as ‘rambling and difficult to follow’. It believed it was a programme in which BAC would call the tune while the A&AEE and MoA were reduced to playing from the company’s music sheet.
The Vickers-Armstrongs plant at Weybridge, bordered by the old Brooklands racing track with the runway at the centre. Vickers flew many newly built aircraft from this relatively short runway, including Valiants and, later, VC10s, but safety precluded flying the first TSR2 from here to nearby Wisley. BAE Systems
Historically, flight testing by manufacturers had created problems, with firms claiming that various targets had been met while withholding precise details of test conditions. Consequently, different results were found when the A&AEE carried out its own trials, and much time and effort was spent duplicating portions of the flight test programme. For that reason it was not unreasonable for the MoA to want a more open and less wasteful programme; but it went further than that. It wanted the flight test programme to be a joint MoA/BAC effort with an entirely integrated flight-testing team consisting of equal measures of BAC staff and A&AEE and RAF personnel, with ‘full MoA participation both in the air and in the project office’. Given the MoA’s record of introducing delays and confusion throughout the project to date, this was a sure-fire way to turn the flight test programme into a disaster. Furthermore, the anonymous writer of an MoA note on the matter stated: ‘… whilst in no position to question BAC on aerodynamic and engineering competence, I have the strongest misgivings regarding the quality of the flight test organization [allocated to the TSR2] and, in particular, on the suitability of some of the aircrew involved …’. This was almost certainly a reference to Roland Beamont. While he was a noted Second World War fighter pilot and the test pilot of the Canberra and P.1, he had not been trained at the Empire Test Pilots’ School (ETPS), and a strong faction believed that company test pilots with no genuine test-pilot background were ‘cowboys’ and risk takers.
Preparing XR219 for an early engine run at Boscombe Down. In this view the bomb bay and engine accessories bay doors are open and several panels have been removed for additional access to various areas of the engine and reheat units. The pristine white finish of the aircraft, nicknamed ‘Joe’, did not last long. BAE Systems via Brooklands Museum
Here, XR219 is seen on the ‘peardrop’ at Boscombe Down, an area on the eastern side of the airfield set aside for TSR2 ground runs and pre-flight preparations, complete with hazard markings painted on the concrete. BAE Systems via Warton Heritage Group
The TSR2’s braking parachute was stored in a bay above and between the two jetpipes, a ‘beak’ door springing open when the parachute handle was pulled in the pilot’s cockpit. The parachute itself was attached to a railing below the jetpipes, as can be seen here. BAE Systems via Warton Heritage Group
Three photographs of the braking parachute deployment, taken at different points. 1) The drogue is fully open and in the process of pulling the main parachute open; 2) The main parachute is open but reefed and 3) The main parachute de-reefed. BAE Systems
The auxiliary intake door mechanism soon proved too weak to hold the doors open at the required 25-degree angle at higher power settings, and locks were fitted to hold the doors fully open for the first flight. BAE Systems via Warton Heritage Group
The RAF was also lukewarm about the new plan, believing it could learn more about the new aircraft by forming the first squadron at an RAF station, away from the sterile flight test environment, and it was concerned that the new plan reduced the number of aircraft on the flight test programme. Statistical analysis of previous aircraft projects was carried out to see how many prototypes could be expected to be lost on a flight test programme, and at a loss rate of one aircraft per 1,000hr or so there appeared a good chance that one or two TSR2s would be lost during testing. With so much riding on the project, BAC would have to be incredibly careful, and lucky.
The development-batch aircraft were all to be extensively instrumented to measure handling, performance and aerodynamic qualities, plus pressure distribution and various parameters to aid with the further development of the auto-stiffening and auto-stabilization systems. The weapons bay would house much of the flight-test instrumentation, wrapped in insulating blankets and provided with air conditioning, and spare space in other bays would be used where possible. On the first two aircraft, for instance, the sideways-looking camera bays below the navigator’s position were used, as no cameras were fitted. The second aircraft, XR220, was fitted with various transducers and ‘bonkers’ to try to excite flutter on the flying surfaces, as well as camera fairings on the intakes for pylon and store flutter monitoring; also for its role as a backup for XR224, carrying out external stores trials. A MIDAS crash recorder was also installed in each development-batch aircraft, housed in the upper equipment bay behind the navigator’s position and fitted with an explosive release unit so that it could be fired clear of the aircraft (through a frangible window in the bay access panel) if needs be. At least that way the conditions leading up to the loss of an aircraft would be known, and the loss would not be a total one.
The question of the location for the first TSR2 flight and the subsequent flight test programme was, predictably, one on which English Electric and Vickers could n
ot agree. Vickers, in overall charge, wanted to use one of its airfields. Development-batch airframes were undergoing final assembly at Weybridge, which had a relatively short runway (1,500yd; 1,370m) nestling within the old Brooklands motor racing track. This allowed no margin for error whatsoever, so Vickers chose Wisley. A great many Vickers types, such as the Viscount and Valiant, had already flown from there, having been built at Weybridge and either transported by road or flown to Wisley for their construction to be completed. Wisley’s runway, at over 2,200yd (2,010m), seemed long enough for TSR2 flying (it was, after all, designed to a STOL requirement), but the local airspace was congested, with London (Heathrow) Airport not very far away, and problems arising during early test flights could require an immediate landing. In that event, Wisley’s runway was too short for safety. English Electric assumed that its airfield at Warton, with nearly 2,660yd (2,430m) of runway, was the logical place for the flight-test base, pointing out the proximity to low-level and supersonic training areas, the company’s extensive supersonic flight-test experience and that long runway. Vickers resisted. The first flight had to be from Wisley, regardless, though it reluctantly admitted that Wisley’s surrounding airspace was perhaps unsuitable for development flying. By early 1961 the creation of BAC had brought the two firms a little closer together, and by May of that year the Vickers component had come round to the use of Warton for test flying. Finally presenting a united front on this important choice, BAC informed the MoA accordingly.
The MoA, naturally, promptly decided that Warton was unsuitable. Only Boscombe Down, with its even larger runway (3,500yd; 3,200m) was suitable, its additional length and width providing the necessary safety factors for early test flights. Any operation of development-batch TSR2s from Warton subsequent to this initial flying would require the aircraft to have arrester hooks installed. This was not a popular decision at either end of the TSR2 project, but, for the time being, that was how it was going to be. A year later Wisley was back on the table, with BAC struggling to get the project back on schedule and anxious to save the month or so it thought would be lost by dismantling the first aircraft and transporting it to Boscombe for reassembly and first flight. The plan was to regain the lost month by transporting the aircraft intact by road to Wisley, then flying it from there to Boscombe.
Roland Beamont, however, had signed on with English Electric for this final flight-testing job in spite of his wife’s ill health, and he was not about to let the change to Wisley go ahead without a fight, particularly as the aircraft’s weight had continued to grow and the engine continued to have development problems. In March 1963 Beamont put his objections to the Wisley plan in writing to the MoA. He said that the plan to make a first flight from Wisley, landing at Boscombe was ‘… made more on the grounds of expediency than of the operational requirements’. There were a number of practical problems. First, if anything went wrong during that first flight and an immediate emergency landing was required, the aircraft’s high approach speed meant it would need a very long runway. (As neither braking parachute nor wheelbrakes could be relied upon, a minimum length of 2,500yd (2,290m) was specified for diversion landings.) Only nearby Heathrow was suitable on that score. At that time the first flight was expected to be made in December 1963, and the question of weather was also raised. The chances of good weather all the way between Wisley and Boscombe in the winter months were slim, and the unpredictability of this would make planning impossible.
By August, after further problems had arisen with Olympus development, the MoA was in agreement that Wisley was perhaps not suitable, and the risk of a TSR2 with a problem crashing in the heavily populated surroundings of Heathrow while on approach to an emergency landing did not bear thinking about. Another study was made into the feasibility of dismantling the aircraft and taking it to Boscombe Down for the first flight. Beamont was not exactly happy. Warton was still a more logical choice, but at least Wisley was off the cards. The first-flight date had by now slipped once again, to March 1964, and the need to dismantle the aircraft, transport it to Boscombe and reassemble it would delay the first flight to April 1964.
This would turn out to be yet another optimistic estimate. Once XR219 had been transported to Boscombe its reassembly took longer than expected, and the continuing problems with both the Olympus engine and ejection seat developments meant that engine runs finally began with a single seat installed in the aircraft, and carried on in fits and starts as niggling little problems arose.
The port console in XR219’s pilot’s cockpit, taken around a month before the first flight, when the cockpit was in the final stages of being equipped. Non-standard items in this view include the instrumentation recorder controls on the block on the windscreen arch. BAE Systems via Warton Heritage Group
The main panel of XR219’s pilot’s cockpit. The major difference from a production panel is that the moving-map display is replaced by a basic TACAN/VOR/DME navigation display (hiding behind the joystick). BAE Systems via Warton Heritage Group
The starboard console of XR219’s pilot’s cockpit. The box on the windscreen arch holds a time ident/base display used by flight test crew for accurately recording times of particular events. BAE Systems via Warton Heritage Group
The port console of XR219’s navigator’s cockpit. This cockpit was considerably denuded compared with the intended production cockpit, and most of the controls visible here relate to fuel management and control of the crash and flight test instrumentation recorders (in the ‘role panel’ area that would normally be used for weapons/reconnaissance-pack control). The Y-shaped item on the sidewall is part of the air-conditioning system. BAE Systems via Warton Heritage Group
The main panel of XR219’s navigator’s cockpit. Once again this is considerably different from a production panel, being limited to a selection of flight instrument repeaters, controls for the engine instrumentation recorder, and a basic TACAN/VOR/DME navigation display taking up the space intended for the FLR display. BAE Systems via Warton Heritage Group
The starboard console of XR219’s navigator’s cockpit. At far left the central computing system control panel is replaced by a selection of engine instruments, including warning lights for the troublesome No.7 bearing. The rapid take-off panel on the sidewall to the right of the Y-shaped air conditioning tube shows blanks for many of the missing systems on this aircraft, including stable platform, Doppler, IFF, Computer, HF, ECM and SLR. This was called the rapid-start panel, as pulling any of the obvious bars on this panel upwards would simultaneously turn on all of the systems it controlled. BAE Systems via Warton Heritage Group
Flight testing programme – airframe allocation
Aircraft
Phase
Flight-test role
1 – XR219
-
Handling and performance
2 – XR220
A
Flutter, vibration, structure loads and engines
2 – XR220
B
Backup for XR224 phase A
2 – XR220
C
Handling, roll/yaw coupling
2 – XR220
D
External stores – handling, structure loads, flutter
3 – XR221
A
Nav-attack and terrain following, CWAS (Conventional Weapons Aiming System)
3 – XR221
B
Bombing with navigation
4 – XR222
A
AFCS and handling
4 – XR222
B
Auto terrain following
5 – XR223
A
Preliminary bombing
5 – XR223
B
Conditioning and air system, temperature survey
5 – XR223
C
Strike photography
5 – XR223
D
Tropical trials
6 – XR224
A
Engi
ne handling, engine performance, pressure errors, aircraft performance with and
without external stores
6 – XR224
B
Further development
7 – XR225
A
Hydraulics, fuel, electrical, miscellaneous systems
7 – XR225
B
In-flight refuelling
7 – XR225
C
Reconnaissance pack
8 – XR226
A
Nav-attack development with stores
8 – XR226
B
Reserve
9 – XR227
A
Temperature and vibration of internal stores, initial release of internal stores
9 – XR227
B
Additional temperature and vibration of internal stores, subsequent release of internal stores
9 – XR227
C
Temperature and vibration of external stores, release and fusing of external stores (nuclear)
9 – XR227
D
Release of internal and external stores
Taxying trials
After months of frustrating delays, XR219, nicknamed ‘Jim’ by the BAC testing team (after a Goon Show character who called everybody Jim), was finally ready to begin taxying trials on 2 September 1964, a hot summer’s day. An engine run in the morning was followed by the first taxy runs in the afternoon. Steering via differential braking on the main wheels was initially impossible, as the aircraft came to a standstill each time, and nosewheel steering was used instead. The first taxi run was in dry power to 40kt (46mph; 74km/h), and on the next one reheat was used. Differential braking was tried again and was successful, if jerky, at higher speeds, the run getting up to 60kt (69mph; 111km/h).