Initial CFM Leap engine runs start the clock on huge development/certification plan
It is ironic that when the first fired into life in 1974, barely anyone outside of or its French partner noticed or even seemed to care.
It was a low-key debut for an engine that would become an industry game-changer, yet 39 years later the contrast could not be greater for the first run of its 21st-century successor, the all-new Leap. The reason is simple. Compared to the first CFM56, which at the time was an unproven orphan with no applications or orders, the Leap is already a pivotal component of the plans of, and Comac. Remarkably, more than 5,000 Leap engines were on order before a drop of fuel was ignited.
The pressure on CFM is therefore intense. Not only does the Leap inherit the vast medium-thrust, high-bypass market dynasty created by the CFM56, it must also maintain from Day 1 the CFM56's mature reliability rate while concurrently improving fuel burn by a step-changing 15%. Just as challenging to GE and Snecma, the large backlog means production is set to start at record rates from the get-go. Delivery rates are set to exceed 1,700 Leap engines per year within just three years.
For the first time since its formation in 1974, CFM is also facing a competitor from the start: Pratt & Whitney'sgeared turbofan. This powerplant not only challenges across a broader thrust range than the International Aero Engines which it succeeds, but will enter service on the Airbus ahead of the Leap. Although the Leap and PW1000G only compete head-to-head for the Neo, the engine contest is as cutthroat as ever.
The air of expectancy is therefore almost palpable at GE's test site here in rural Ohio where the firstengine is in the early phases of evaluation following its start-up on Sept. 4. The Leap-1A is destined for the , slated to enter service in 2016, and will be followed by the aero-mechanically identical Leap-1C for the , as well as the architecturally common Leap-1B for Boeing's 737 MAX. The C919, the Chinese airliner project that officially launched the Leap in late 2009, is scheduled to enter service in 2017, although CFM is maintaining its original engine development schedule, which calls for the -1C to be certificated in mid-2015. The Leap-1B is on track to enter service on the MAX in 2017.
“The engine is running fantastically,” says Ted Ingling, engineering lead for the Leap-1A/C. Initial work is focused on aero-mechanical engineering tests to look at the high-order attributes of the engine and establish that the “fundamental architecture is working as expected,” Ingling says. “We are making sure the thrust in the rotor architecture is where we want it to be, and to make sure we understand the fundamentals of the engine. We do those kind of tests while we're developing initial software loads for follow-on runs. We use the first engine to tune that software,” he adds.
Staying on track will be critical to the success of the busy test effort that quickly ramps-up before the end of this year and into 2014 when, at one point, up to 15 development engines will be running in parallel. “That is unprecedented and that's only part of the overall test plan,” says CFM Executive Vice President Chaker Chahrour. The overall test and certification program will eventually include 60 engines. Some 28 will be development engines for the Neo, MAX and C919, while the balance is made up of compliance engines which will power the three new airliners during their certification campaigns.
From a schedule perspective, the monumental effort got off to a good start when the first engine entered testing three days earlier than the target date established when the program was laid out in 2010. “We also got combustor light-off first time, and the first day we did mechanical check out,” says Ingling. Full power was achieved five days later with a run up to 33,100 lb. thrust to cover the highest rating required for the A321. The first engine was then pushed further, demonstrating more than 35,000 lb. thrust and with it the availability of fundamental growth margin, should it be required, adds Ingling.
“In early testing, we're looking at the initial performance of the engine, vibration and the aero-mechanics,” he says. “We do runs to check the fan and compressor blades, the engine dynamics and the bearings. There's also a lot of starting to map out the start envelope, figure out the fuel schedules and see how the starter interacts with the engine.”
The Leap's basic architecture is based on a scaled version of the same eCore design at the heart of the, with an advanced low-pressure (LP) system from Snecma. However, while the Leap core is designed to operate at a higher pressure ratio than the latest CFM56, it is deliberately tuned to lower pressures than the bigger GE engine that powers Boeing's 787 and . “We're well inside GE's experience for these kind of pressure ratios,” says Ingling. “The GEnx is the high watermark in terms of pressure ratio and temperature and, although we had the option to mimic or even exceed the GEnx, we didn't because we wanted durability for this marketplace. We want it to be a copy of the CFM56 in terms of time on wing.”
The third version of the eCore ran earlier this year in the build-up to Leap testing. “Using this, we have done aerodynamic, stall, performance-mapping and so on, but what we don't get effectively is interaction with the LP system and we don't get transients [rapid changes between thrust demand],” says Leap program manager Gareth Richards. The full-up engine is also therefore the first to check the full operability characteristics of the integrated LP and high-pressure (HP) systems. Testing includes “bodies”—or throttle bursts and chops—that check the related response of the compressor.
The engine is also fully configured with standard systems such as the eductor-based surface oil-cooling mechanism that, like the same system on the GEnx, consists of surface coolers mounted around the inner lining of the fan duct. The eductor device produces a venturi effect, which ensures a positive pressure to keep oil in the lower internal sump.
Engineers are also verifying the performance and behavior of the composite fan using “clearance-ometers” developed by Snecma to sense the exact motion and vibration of individual blades. The sensors are mounted in the fan case at the leading-edge, mid-chord and trailing-edge positions to measure variations in the local magnetic field as each blade passes by. Using careful “per-rev” calculations, engineers can deduce from the measurements whether the blades—and specific parts of each blade—are passing the sensors early, late or on time. The results indicate whether the flexible blade, made by Snecma using a resin transfer molding (RTM) process, is untwisting to the correct degree with increasing rotational speed.
“The untwist we're looking for is right on the prediction,” says Richards. “That calibrates the engine models, which say as the engine speed goes up, forces act on the fan, and we use those sensors to be witness to the fan mechanical characteristics. All of this is confirmation of our analysis. We are validating what we expect the engine to be doing.”
In the case of the fan, CFM also continues to run parallel validation tests of the final blade configuration in a demonstrator, even as the first Leap engine starts to run. The final phase of the Mascot 2 program, which involves testing an RTM fan on a CM56-5C, is now underway with a series of crosswind evaluations at Peebles. “This is building confidence ahead of further runs,” adds Ingling.
The first engine, designated 598-001, is festooned with 1,300 pieces of instrumentation. These include thermocouples in rotor cavities “to understand the secondary systems, feed and purge cavities, loads and temperatures of the bearings,” says Richards. “We also have performance rakes to understand how the fan is pumping and how the core is pumping,” he adds. Following its first phase of testing, the first engine will go through a second build cycle in readiness for early icing tests at GE's dedicated facility in Winnipeg, Manitoba.
“This will include a full demonstration of the certification test profile [a year before the actual certification test], plus an engineering test for icing. It is another risk burn-down, and we have taken lessons learned from GEnx icing and then some,” says Richards, referring to the ongoing development work at GE to tackle the core icing issues encountered particularly on the GEnx-2B. The engine will undergo a third and final rebuild in readiness for use in the destructive blade-off test scheduled to take place in Villaroche, near Paris, next September,
The start of engine tests comes as CFM continues to rack-up record orders for the current CFM56-5/7 as well as the Leap models. The GE-Snecma joint venture has taken orders for 2,196 engines this year, outpacing the 1,972 booked over the whole of 2012. The 2013 orderbook is split almost evenly between current and next-generation engines, with booking taken for 1,094 CFM56s and 1,102 Leap units so far.
“The two CFM product lines are doing very well so far,” says CFM president Jean-Paul Ebanga. “At this stage of the Leap program, we have more than 5,000 engines already on order. In terms of backlog, we are in good shape, and on the CFM56, our backlog is also above 5,000.”
The sales success continues to put greater onus on preparations for meeting delivery demands and ensuring a smooth production ramp-up. Cedric Goubet, CFM executive vice president, says the goal is to achieve an accelerated delivery curve that will see Leap production rise from zero to 1,700 engines “within less than three years.” Overall, the plan calls for “1,700 Leaps by the end of 2018 and start of 2019, and maybe going up to 1,800 by the end of the decade,” he says. All these exceed the recently achieved historic maximum annual delivery rate of 1,500 engines for the CFM56.