THE SHEER SIZE AND extra capacity of General Electric's Boeing 747 test-bed gives it an obvious advantage over its smaller predecessors. "It is five, or even ten times, as efficient as the 707," comments Phil Schultz, GE flight-test organisation (FTO) chief pilot. "We can run five or six objectives in one flight and not come down and have to change everything between tests. In addition, the modifications we've made to the cockpit mean that the pilots can see exactly what's needed in terms of flying at the right conditions, speeds and altitudes."
The quantum leap from the 707 to the 747 was driven by the under-wing clearance requirements of the 3.12m fan-diameter GE90 engine, which was first tested in December 1991. At the same time, realising the need for a more comprehensive and efficient test-bed, GE took the opportunity of developing a state-of-the art data-gathering system. To improve the quality of the data and to speed up the testing process, the system is consistent, in every respect, with those at the company's test sites in Evendale and Peebles, Ohio.
In addition, Boeing's early decision to assume responsibility, for the entire 777 propulsion system and subsequently, that of the new 737 series, meant that the 747 test-bed was also developed, to test production-standard nacelles and production-standard engine build-up hardware. The result is a "one of the kind" resource, according to Krejmas.
INSIDE GES' 747 TEST-BED
GE transformed its elderly 747-100, an ex-Pan American World Airways machine and the sixteenth off the Everett line, into the flying test-bed during an extensive D check. Structural changes and modifications included a section 41 rebuild and strengthening of the left wing and centre fuselage to accommodate the test engine mounted in the number two (left inboard) position. Wing modifications, included local stiffening of spars and ribs, while an aft bulkhead was strengthened to increase high angle-of-attack capability.
"The tail on the -100 is limited [by buffet] so tail modifications were needed to strengthen the bulkhead supporting the horizontal stabiliser," says Schultz. The GE 747 is now capable of 747-300 manoeuvres, which allows testing, at angles of attack of between 30° and 32°. The high-inlet-angle test is required for fan-stress and inlet-stability evaluation. A specially developed strut adaptor has also been made to interface between the number-two engine pick-up locations and whichever engine is being tested.
Much of the modernisation effort has been focused on the $12 million data system, which occupies most of the main deck, as well as some under-floor cargo space. The system handles three types of data: high-rate analogue/dynamic data such as vibration and stress loads; digital data of engine pressures, temperatures and strains, and information from the 747 air-data computer and full-authority digital engine-control (FADEC) system, which is carried on the aircraft's ARINC 429 digital databus system.
Analogue data are collected by four 28-track recorders, while digital data are fed into the advanced data system (ADS) which is capable of measuring 2,000 parameters and is expandable to take up to 3,000, if needed, in the future.
A third, and crucial, data-collection tool is the FADEC monitoring system. The system enables engineers to monitor the engine-control system on-line and adjust the control schedule in flight. "In terms of flexibility, it's one of our most valuable tools during the development process," says engineer team leader, Tony Bonser. "We can make adjustments to the non-volatile memory of the FADEC in flight and optimise the cycle."
Peter Thompson, CFM56-7 project leader says: "We think of the FADEC monitoring system as being like a screwdriver. We're using it to adjust and fine-tune the turbine clearances in real time by using the data to modify the control software and make it more efficient." This is achieved through minute adjustments of the ratio of bleed air from the high-pressure (HP) compressor to the temperature of the turbine shroud. The bleed air is taken in varying amounts, depending on the phase of the cycle, from the fourth and ninth HP-compressor stages, and distributed through a manifold along the outer surface of the shroud. The cooler air shrinks the casing in varying degrees to match more closely the rotating HP turbine blades within.
Thompson says, that the fine-tuning is checked, after landing by borescoping. "Notches are ground into the tips of the blade at known depths, so if we rub it, it will show up on the borescope."
All data readings are taken, in either steady state or transient conditions. Steady-state parameters are recorded at a rate of 100 sample/s averaged over 30s, while transients are recorded at sample rates varying between 1 and 100Hz. All data are processed on the aircraft. "On the older aircraft, we'd spend 16h processing the data after the flight. Here, it is done on-line and, with this system, we can plug in a datalink and send it back to Evendale within 15min in the case of the steady-state data," says Bonser.
All data are displayed throughout the cabin via the cockpit display system. The cockpit itself is fitted with several GE-designed cathode-ray-tube displays and other modifications which were deliberately limited in scope .
A second screen, forward of the throttles, is used to set N1 (fan speed), exhaust-gas temperature and control transients on the test engine. Aft of the throttle quadrant is a third screen dedicated to FADEC parameters and the ADS, while another screen at the flight-engineer's station shows the status of subsystems on the test engine, such as fuel and bleed air.
A small cathode-ray tube mounted above the glare shield displays a television picture of the engine in real time taken from a camera mounted in the cabin. "The more information we show here, and the more we can keep the intercom to a minimum, the better we are," comments Schultz.
The flight deck also retains ultimate control over power feeding the 80 racks of data equipment arranged in 13 rows on the main deck. In the case of fire, or other emergency, a switch on the flight engineer's electrical panel can be thrown to shut down the main deck system. This still leaves enough instrumentation functioning to shut down the test engine.
Source: Flight International
