Tail Blazer

When Lockheed Martin submitted its successful design for the JSF it looked to the UK to put its revolutionary propulsion concept – where thrust is directed from the aircraft’s back end via a nozzle – into practice.

It’s worth an estimated £140bn – maybe double that – and everyone wants a piece of it. Joint Strike Fighter, better known as JSF, is the biggest defence deal in history. But though it was an American idea, it isn’t your average all-American plane.

Since the Lockheed Martin, Northrop Grumman and BAE Systems team’s X-35 demonstrator beat Boeing’s X-32 version in a winner-takes-all competition last October, Canada, Denmark, Norway and the Netherlands have signed up for a share in the Systems Development and Demonstration (SDD) phase.

Many more nations are set to follow in the biggest international project ever seen. ‘Nobody comes on board unless they’re the best athlete,’ stresses Lockheed Martin’s UK JSF Business Development Manager, Ivor Evans.

The UK, however, has been aboard almost since the start and has made a fundamental contribution to the plane’s cutting-edge technology. JSF is now heading for ‘lines freeze’ – the time when the exact shape is set and avionics refinement, flight tests and production begin – and just as before UK engineers are at the heart of the process.

Interoperable

What’s so special about JSF? At the moment the US Air Force, Navy and Marine Corps, plus the RAF and Royal Navy, have about 10 specialised types of fighter-bomber aircraft between them, with all the separate logistic and infrastructure costs that follow. The aim of JSF is not only to perform 50-100 per cent better in combat but also to cut costs by having a jack-of-all-trades aircraft – joint for all services, and internationally interoperable for any nation that can afford it.

There are three versions: a conventional take-off and landing (CTOL) plane for the USAF that, like the F-16, works only from normal runways; a carrier (CV) model with larger folding wings and stronger undercarriage for operations at sea; and an all-singing, all-dancing Short Take-Off and Vertical Landing (STOVL) aircraft. Each will sell for between £25 and £30m.

During the current 10-year phase the next prototype should fly in 2005. Deliveries to the US forces begin three years later and the UK receives its first units in 2011. Nine years may seem a long time, says Evans, but for the defence industry this is very fast. Whereas only 600 Eurofighters are to be made in total, even JSF’s low-rate initial production phase will roll out 450 aircraft. To produce the planes at the required rate and quantity, said Evans, they are relying on new technologies.

Why Lockheed won

Boeing’s Harrier-like design played safe. It was a vast improvement on theconfiguration of the world’s most famous STOVL aircraft, but was still a design rooted in the past. So when Lockheed Martin performed a short take-off, supersonic dash and vertical landing in the same ‘Mission-X’ test flight, Boeing probably realised it was in trouble.

Neither competitor actually had to perform the routine, but Lockheed’s aggressive display of the X-35B’s extreme capabilities advertised its capacity to match and exceed expectations.

The Pentagon never explicitly stated its detailed reasons for Boeing’s loss, though ‘programme risk’ was cited in some quarters – despite the comparative gamble of Lockheed’s STOVL approach compared to Boeing’s tried-and-tested idea. Another issue was stealth, Lockheed’s promised high-production standards preferred to a Boeing idea that involved fencing off the inlet to shield the engine from radar.

Confusion had also arisen when Boeing changed from a delta-wing (like the X-32 demonstrators themselves) to a conventional wing-and-tailplane layout. Boeing’s design would also have meant a smaller radar and a ‘drop-down’ infrared sensor as opposed to Lockheed’s fixed flush mounting (which is perhaps more reliable during combat manoeuvres).

Despite these factors, and cost overruns, Lockheed’s ‘Mission-X’ flight assured many analysts that it was on the right track.

Boeing chairman Phil Condit was clearly disappointed at the decision, but the company may be able to salvage some of the technology. The JSF team came up with new methods of manufacturing, new ways of designing, said Condit in October.

‘We will use these in every programme inside Boeing as we go forward. We are going to stay focused on the tasks ahead.’

The lift fan

Like the Mini, the Harrier dates from the 1960s and while nippy and convenient it is not very fast. For its STOVL and in-flight manoeuvring properties it uses four swivelling thrust-directing nozzles which can’t quite push it past the sound barrier.

Lockheed’s success was due to a new concept. The most revolutionary aspect of the X-35B STOVL demonstrator design was the shaft-driven lift fan – a departure from conventional thinking on vertical landing systems and was perhaps the single most important factor why it beat Boeing’s X-32B. And as with the Harrier before it, the engineering expertise behind it was British.

Though the idea originated at Lockheed Martin’s Skunk Works facility, where the company conducts all its secret testing, it was down to Rolls-Royce to put it into practice. The lift fan is behind the cockpit and during hover mode sucks in air from above the aircraft. This is blasted back down from the belly in conjunction with thrust directed from the JSF’s back end via a nozzle with three swivelling ducts. Small additional ‘roll nozzles’ in the wings provide extra manoeuvrability.

Consisting of two sets of counter-rotating blades, the lift fan is driven by the main engine via a clutch and gear system. The engine produces 25,000lb of ‘dry power’ – with afterburner this shifts up to 40,000lb in normal straight-and-level flight. Though the lift fan and afterburner can’t be used at the same time, using it with the main engine also produces 40,000lb.

Designing the lift fan presented a number of challenges, says Rolls-Royce JSF STOVL director Chris Cholerton. As no other engine has a similar horizontal-to-vertical shaft arrangement, the team needed to ‘leverage aircraft brake technology’ to engage the clutch and lock the vertical to the horizontal shaft. There are understood to have been some difficult moments before the trial phase.

Issues of wear and tear and vibration in the clutch are still being solved, adds Cholerton, and additional system production issues include further reducing weight and cost. ‘The design is reducing complexity and risk,’ he says. ‘We are taking complexity out.’ This is no mean feat as the software controlling the clutch alone contains 155,000 lines of computer code.

One way of achieving this simplicity will be to integrate the lift fan and roll nozzles straight into the aircraft as it is made, plus make the three-bearing swivel duct at the rear a part of the engine itself. This is another fiendish structural issue: the nozzle must be able to redirect enormous thrust through 105deg in 2.5 seconds. It is also difficult to control the swivel action at speeds of 290mph due to pressure distortions.

The fans themselves will also be constructed as bladed discs or ‘bliscs’, borrowing a civil engine method not previously used in military aircraft. Rather than attaching different parts to a central hub, the JSF ‘blisc’ exploits lighter hollow blades that are linear friction welded to a base, making a seamless one-part component. The first of these – no blisc has ever been made with hollow blades – will be produced in 2003 and should save 20 per cent of the lift fan’s weight.

Below the lift fan Rolls-Royce plans to save another 70kg by redesigning the belly nozzle to a ‘Venetian blind’ configuration. Again, this will be a structural feature rather than an add-on and actuators will allow the pilot to control the airflow thrust better.

Under a £1bn contract with Pratt & Whitney, the Rolls-Royce parts work in conjunction with Pratt & Whitney’s F-135 engine. The preliminary design review for fixing the final configuration took place in May. The UK company also has a 40 per cent share (General Electric holding the rest) in an alternative engine under development, the F-136. This near-identical ‘spare’ engine design should run for the first time in 2004. Such is the engine’s importance to JSF that, like the F-135, it is government funded rather than subcontracted to Lockheed Martin.

Manufacturing

Figuring out how the aircraft would be made was a critical part of the Concept Development phase, to plan a six to nine-month ‘make-through’ build, says Evans. Following the lines freeze, manufacture of the 14 flight test aircraft and eight static test aircraft starts at the end of this year.

BAE’s expertise in laying composites was one of the main reasons for its partnership with the US companies in the programme. Lockheed Martin and Northrop Grumman want to exploit the UK techniques developed from Airbus and Eurofighter at their facilities in Fort Worth, Texas, and El Segundo, California.

These methods mean that JSF will be made at unprecedented levels of speed and efficiency, says BAE Systems’ JSF chief engineer Phil Cronshaw. With older fighters like Tornado and Harrier, fitting different sections together took days; once the parts are made JSF can be assembled in minutes by wheeling them together on trolleys. ‘The aircraft literally goes together like pieces of Lego,’ says Cronshaw.

The CTOL, CV and STOVL versions all have minor differences, with the CV having larger wings and the STOVL needing accommodation for the nozzles and intakes. ‘But the process should be such that you press the STOVL button and it cuts the right shape for STOVL,’ says Evans. This will save having three expensive production lines.

The secret is a laser alignment technique controlled directly from digital data held on computers networked between the UK and US. Reference points on the factory jigs are controlled by a system of tiny mirrors that the ‘intelligent laser head’ tracks to ensure every part is identical to sub-millimetre levels. Though the exact accuracy is classified, the improvement is understood to be ‘an order of magnitude’ greater than that on Eurofighter.

Large ‘knitting machines’ put together the layers of composite materials, carbon fibres and resin into complex 3D shapes (for example air inlets) that otherwise would have to be hand formed. The digital manufacturing suite is completed by a drilling machine that again is integrated with design data, saving days of manual work. These automated processes will speed up manufacturing while needing perhaps 10-20 per cent fewer operators on the shop floor.

Cronshaw estimates that the £40m facility being built at Samlesbury (commissioning is scheduled for late-2003) will be able to make 20 titanium, aluminium and composite rear fuselages per month compared to around four for older fighters. These will be shipped to the US and stuck straight on to the aircraft bodies.

As well as saving costs, precision production has numerous operational advantages. For example, low-observable aircraft like the B-2 stealth bomber have historically been vulnerable to the elements, so it requires constant ‘fixing and blending’ of the surface to keep its low-observability characteristics. A tiny blemish could make the B-2 almost as obvious to radar as any other aircraft.

Not only is JSF designed to operate ‘without the bubble wrap’, says Evans, but the precision manufacturing technique means that when panels are damaged they can be whipped off and replaced exactly with no need for time-consuming taping or welding. ‘The accuracies built into JSF from the start are a major plus point,’ adds Cronshaw.

Combined with Rolls-Royce’s STOVL lift fan, which even now is being further improved, and the countless other UK contributions, JSF is proof that UK aeronautical engineering is still alive and well.

Sidebar: The UK’s contribution

The JSF pilot’s helmet-mounted display allows him to look down and see a video of what’s right below the plane. According to Lockheed Martin’s Ivor Evans, a former RAF Tornado pilot, it’s a frightening experience.

The cockpit comprises an 8in x 20in display more like a computer desktop than a traditional layout. Along with stick and throttle parts developed by BAE Systems Rochester, other UK technology will help the plane fly itself.

Qinetiq’s VAAC (Vectored-thrust Aircraft Advanced flight Control) Harrier test plane aims to make flying STOVL aircraft, with their complex 3D capabilities, as easy to fly as traditional planes. In one trial an officer from the carrier Invincible, who was a trained pilot but had no STOVL experience, successfully landed the aircraft on the ship.

Building on previous working in conjunction with NASA, the VAAC Harrier achieves its aims with sophisticated software that allows digital controls to take over the piloting aspects a human finds most difficult. The technology is likely to be inserted into JSF, and now Qinetiq is working on automatic recovery techniques, a means of easily landing the Harrier at a predefined point like a carrier deck.

A further 11 UK companies are providing smaller but no less important components for JSF. Martin Baker is working on a version of the Mk16 ejector seat as used in Eurofighter, while Dorset-based Flight Refuelling is to provide the interfaces for air-to-air refuelling. MBDA is integrating UK armaments, along with TRW-Lucas and Ultra Electronics which are contributing to weapons bay systems. Smiths Aerospace will provide electrics.

However, one UK company earmarked for JSF work seems to have lost out. GKN’s Aerospace Composite Technologies division had been named to provide de-icing equipment for the lift fan plus stealth coatings for the canopy, but is understood to have lost out to Pilkington in the US, which one GKN employee felt was a ‘betrayal’. Other GKN companies are still involved with composite moulding processes.