GuyNorris/LOS ANGELES
Mojave Airport, in California's high desert, is well known for unusual sights. Among stored airliners, experimental prototypes and rare warbirds, General Electric's Boeing 747 testbed rarely merits more than a second glance. It is only in recent weeks that sharp-eyed observers may have caught a glimpse of something out of the ordinary - a fifth engine nestled between the 747's fuselage and number two engine.
The engine is the relatively diminutive General Electric CF34-8C1 - the latest member of the company's increasingly successful regional jet engine family. The engine may be little, but GE is under no illusions about the growing size of the regional business at which it is aimed. "I'm right in the middle of a booming market," says Frank Klaus, general manager of small commercial engines for GE. The -8C1 is in development for Bombardier's Canadair Regional Jet Series 700 (CRJ-700) regional jet which, by the time of its roll-out at the end of May, had accumulated 96 firm orders and 137 options.
Added to the existing orders for the CF34-3-powered CRJ200, and the recently launched -8D-powered Fairchild Aerospace 728JET, total orders for the regional jet engine family have soared to almost 700 shipsets. This does not include engines operating on almost 300 aircraft in service and a further 450 CRJ 100/200/700 and 728JET options.
This is a far cry from the original market aims of the TF34, the CF34's predecessor, which began flight tests not far from Mojave, at Edwards AFB, on a leased US Air Force Boeing B-47 in 1972. Aimed at the US Navy's S-3A Viking, and later the USAF's A-10, the high bypass TF34 was first proposed to the US Navy in January 1968. It went on to enter service six years later. Its low fuel consumption and noise provided a perfect foundation for the first commercial derivative, the CF34-1A, which entered service on the Canadair Challenger in 1983. By the 1990s, the relationship with Bombardier and the business jet unexpectedly flourished into the current regional jet combination.
"We were either lucky or as smart as hell, but either way when we launched the -8C it gave us a big advantage, particularly since the -3 has been so reliable," says Klaus. Whereas the engine had experienced relatively modest thrust growth over the first 20 years of its operational life, the launch of the -8C1 derivative committed GE to a more dramatic thrust boost of around 50% to 14,000lb (60kN). The extra power, which is required by the bigger CRJ700, is largely achieved by increasing the size of the fan by 56mm (2.2in) to 1.12m and moving to a wide chord fan blade. The bigger fan pumps more airflow through the engine and increases pressure ratio.
Downstream changes include a 10-stage high pressure (HP) compressor, based on the F414 developed by GE for the Boeing F/A-18E/F, in place of the original 14-stage HP compressor in the -3B1. The first three stages of the new compressor are blisks, which minimise inter-stage leakage and reduce weight. The jump to F414 technology cuts compressor part count by 50% and stages by 23%. It also introduces a machined ring combustor based on a mixture of F414, CF6 and earlier CF34 designs. It also introduces a dual channel full authority digital engine control (FADEC). This forms a crucial element of the new engine as it is designed to improve operability, reduce maintenance and increase life in the harsh, rapid cycle environment of short-haul regional transport.
The HP turbine is externally indistinguishable from the -3B1, but on closer inspection reveals new three-dimensional aerodynamic design refinements and directionally solidified second-stage vanes. The second-stage turbine blades also feature cooling to cope with the higher operating temperatures of the larger engine. It also has boltless blade retainers, derived from similar technology adopted for the growth CF6 models, which increase component life. The low pressure (LP) turbine is similar to the -3A/3B1 rotor structure, but is otherwise different on the -8C1. The turbine consists of four stages, each designed using three-dimensional aerodynamic techniques. The oil cooling for the local bearing sump, is also redesigned to avoid coking, which troubled earlier engines.
Development time
GE began a protracted propulsion system development effort in early 1996, with engine certification targeted for November 1999. The company opted to reduce the risk by running up-front demonstrations of redesigned modules and the all-important fan containment system in rig tests. Core engines were separately developed to map the HP compressor, combustor and HP turbine performance.
The value of this approach was proved, regrettably for GE, with the failure of the containment system. Although this was strengthened with extra wraps of Kevlar composite to cope with the higher energy of the heavier fan blades, the rig tests revealed basic weaknesses in the design which led to fragments penetrating the casing. Several redesigns were subsequently tested, before GE settled on a strengthened containment system consisting of thicker Kevlar and steel.
Windtunnel tests were also performed on the nacelle, which was designed in consultation with several key regionals including Air Littoral, Britair, Comair, Lufthansa CityLine, Lauda Air and SkyWest. With maintainability and service readiness identified by the group as the major requirements, GE adapted the design for easy access to line-replaceable units and even engine change-out in one shift. The company took responsibility for the entire propulsion system while the UK's Shorts designed and manufactured the inlet, fan cowl, reverser, core cowl and engine build-up units.
To make the engine more accessible, the inlet, fan cowl doors, reverser halves and aft core cowl were designed to split wide open along a hinge line at the pylon. The nacelle, which has 30% fewer parts than previous designs, is of a "jam-tolerant" design and can be opened by one mechanic without using tools. Although tests on the nacelle, LP turbine and HP compressor rigs went well, the problems with fan containment dogged the development programme, forcing a delay of several months to the first flight on the testbed. This took place on 10 March instead of early October last year as originally planned. The test team has, nonetheless, worked hard to make up for lost time and is expected to have completed the basic flight test effort by the time this article appears. First flight of the prototype CRJ-700, expected around late March, took place in late May, although little of this postponement was attributable to the earlier engine delays.
The flying testbed programme was expected to be completed after around 27 flights and 150 flight hours. Thanks to the endurance of the 747 and its large capacity for test equipment, test flights were averaging 6-7h. "We're getting a fire hose of data," says Klaus, who adds that the sophisticated tools on board have allowed the FADEC schedules to be changed in flight on some sorties.
"We have done lean blow-outs, wind-up turns, operability transients, inlet stability, starting and icing tests, the latter of which we did in the Aleutian Islands near Alaska. We also travelled up to Casper, Wyoming, for high-altitude lapse rate take-offs," Klaus says.
Flying the engine
Unusually for this stage in an engine development effort, guest test pilots from the airframe manufacturer have also been invited to "fly the engine" on the 747. "We want to make sure they [the Bombardier test pilots] have the right production configuration, so they use the same throttle system and gauges as in the CRJ-700. Obviously they won't feel the power of the engine in the seat of their pants, but they will get a feel for how it responds. It will be like a simulator," says Klaus. For the tests, the engine is mounted on a slaved pylon attached to the inboard, under-wing skin of the 747. The engine itself is cantilevered away from the strut on a Bombardier designed and built CRJ-700 pylon.
Two flight-compliant engines, numbers 101 and 102, are mounted on the initial CRJ-700 and will be modified with "things we've learned from the flying testbed", says Klaus. This mostly involves FADEC software changes. "You learn things when you put an engine on an aircraft, such as the stator scheduling on the compressor, which is optimised in flight for stall margin and specific fuel consumption."
Another, perhaps more unexpected, lesson was learned during the hail ingestion test. "This dinged up the noise suppression panels in the inner flow path," says Klaus, who likens the event to "having your car in a hailstorm". The resulting damage was surprising, but not bad enough to warrant an immediate redesign. "The question is-is that acceptable, or do we strengthen it? We have Shorts' people looking at the hardware," he says.
Some discoveries have led to pleasant surprises, he adds. "One of the things we were concerned about was the air start envelope because the core is somewhat hidden." Although the deliberately "hidden" splitter affords better core protection from foreign object damage, dirt and hail, it also threatened to make the engine harder to re-start after an in-flight shut down. "It has not been an issue," says Klaus. "We have also been pleasantly surprised at the lapse rate of the engine," he adds. Based on initial results, cruise specific fuel consumption is estimated to be between 1% and 1.5% better than expected.
Other tests are also under way with another five ground-based engines. Crosswind tests on one engine were recently completed at GE's Peebles test site in Ohio. A 150h block endurance test, due to begin within a few weeks, has also been preceded by an "unofficial" 150h test. "We did it to see what we needed to change, and we were fairly happy with the durability of the hot section," Klaus says. There were, however, minor cracks on the shroud, which GE says will be avoided on production engines, with additional cooling added as a result of the test. "Overall we are very happy, and we are contemplating running at an even higher redline temperature as a consequence."
Major hurdles still standing in the way of certification include the full engine fan blade-off test, as well as the large bird strike and water ingestion tests. Although Klaus admits that the re-run of the containment test, due around late August, is "still risky", he believes that, with the many modifications already verified in earlier rig tests, the actual milestone should pass "like a dress rehearsal".
Family planning
In the midst of flight tests on the CF34-8C1, GE was making plans for the next family member, the -8D, following the official launch order from Lufthansa CityLine for the Fairchild Aerospace 728JET fleet. The engine will be rated at the same 14,000lb thrust level as the -8C1 and shares 100% common turbomachinery and 87% overall parts commonality with the CRJ-700 engine.
As it is wing mounted however, the -8D will have a different mounting and rearranged accessories. In the case of the -8D, the thrust reverser, nacelle and engine build-up units are being designed and built by a newly formed partnership between Hurel-Dubois of France and Aermacchi of Italy. The first -8D goes to test in June 2000, followed by first flight of the 728JET in the second quarter of 2001. Engine certification by the US Federal Aviation Administration is targeted for September 2001, with 728JET entry into service scheduled for May 2002.
A more ambitious, but closely related development, is the proposed follow-on 17,000- 18,000lb thrust -8XX derivative for the 928JET. GE's plan calls for the engine to begin flight tests in mid-2003, with FAA certification as early as the fourth quarter of 2003. "It will have a similar core to the other -8C/D derivatives, but with a new fan and low pressure system," says Klaus. The new LP system, with three stages, will be driven by a 1.34m diameter fan and a scaled-up, four-stage LP turbine section.
GE should also know by mid-June if its bid has been successful to power Embraer's proposed ERJ-170/190 family with yet another derivative, the CF34-8E. Broadly similar in size and configuration to the -8XX, the -8E faces stiff competition from Snecma and Pratt & Whitney Canada with their SPW16 and SPW18, as well as from a proposed hybrid engine from BMW Rolls-Royce called the BR715-50.
With or without the Brazilian regional jet, the CF34 is already destined for the record books. Production of all -8 series engines is ramping up at a new line at the company's Durham site in North Carolina, while the -3 line remains at Lynn, Massachusetts.
Source: Flight International
