As designers of future airliners look increasingly beyond traditional tube-and-wing configurations to meet the high efficiency goals of the 2030s and beyond, new territory is being carved out in the critical area of airframe-engine integration.

Unusual features ranging from recessed inlets to pylon-mounted upper-surface engines have become familiar sights in wind tunnels, but even seasoned researchers are surprised by a new engine architecture proposed by Pratt & Whitney. The concept not only physically separates the propulsor from the gas generator, but also mounts the core backward and at an angle. This novel arrangement is aimed at overcoming installation challenges in new configurations like the D8 double-bubble airliner concept under study by NASA and the Massachusetts Institute of Technology (MIT).

Aimed at NASA’s N+3 performance goals for an airliner that could enter service around 2035, the D8 is designed to burn at least 60% less fuel than the current generation of narrowbody airliners. The secret behind this leap in performance is a configuration that clusters the engines together atop the wide tail of a flattened fuselage. Besides providing a clean high-aspect-ratio wing for low drag, this enables the engines to reenergize to slow-moving boundary layer flow over the fuselage, increasing efficiency.

But such a configuration creates several issues. The engines lie so close to the upper surface of the fuselage their fans must be sufficiently robust to cope with flow distortion from ingesting the boundary layer. Fan size will also be large because the engines envisioned for the D8 will have a bypass ratio of at least 20:1, and be targeted at extremely low noise levels of -52 EPNdb below current Stage 4 limits. Scale tests conducted at NASA of a distortion-tolerant fan developed by United Technologies Research Center show the boundary-layer challenge has been met, but other key questions remain.

Because engine cores are becoming more efficient and operating at higher pressure ratios, they are also shrinking and becoming disproportionately small compared to the propulsor section as bypass ratios increase. This leads to blade heights of 0.5 in. or less at the exit of the high-pressure compressor. At this small scale, tip clearances not only become harder to maintain, but there is little space within the core through which to run the driveshaft connecting the fan to the low-pressure turbine. Additionally, because the core is proportionately longer and thinner, designers face the issue of backbone bending which further affects clearance control.

“So that’s when we had the breakthrough idea of turning the core backward,” says Pratt & Whitney Technology and Environment Vice President Alan Epstein. Air enters the engine through the fan as normal, but instead of continuing directly into the compressor, it is ducted around the side and back of the core to enter from the opposite direction. In an arrangement similar to Pratt & Whitney Canada’s PT6, in which air flows forward through the engine, hot gas will be discharged forward through a power (low-pressure) turbine connected to the fan via a gear system. The turbine, gearbox and fan will be connected via “a really short shaft, and because the core is not connected to the power side, you can take the core off easily for maintenance,” Epstein explains.

The concept also overcomes another challenge. The idea of nested engines, as in the D8, does not meet current FAA certification criteria under the “1 in 20” rule. This states that there should be only a 1 in 20 chance of debris from an uncontained engine failure causing a second engine to fail. However, because the core and propulsor are no longer mechanically linked, “the designers have come up with an extraordinarily clever arrangement in which the cores are angled relative to each other,” Epstein says.

“We cant them at around 50 deg. and the exit from the core turns via a 50-deg. duct to go into the power turbine.  So now they are more than 90 deg. off from each other. It’s simple geometry,” he says. “It enables you to have a large bypass ratio, and you are not turning much of the airflow if you are turning just the core flow, so pressure losses are low.”

Pratt hopes to lay out a road map for future studies, possibly with NASA, to further define the architecture and evaluate related elements such as shorter-length inlets that would also help development of the next-generation geared turbofan. Other focus areas could include studies of the ducting to evaluate weight and temperature requirements and whether materials such as ceramic matrix composites might be suitable. “Then there’s the question of how do you convince the FAA it can be certified,” says Epstein.

For the MIT-led team, Pratt’s novel engine design is a key enabler of the D8 configuration. Another crucial element has been validating the efficiency benefit from boundary layer ingestion (BLI) by the clustered aft-mounted engines through large-scale wind-tunnel testing with NASA. This has quantified the power-saving from BLI at around 8% “in a realistic configuration,” says Ed Greitzer, MIT professor and D8 principal investigator. “This is the proof of concept for BLI in civil transports,” he told the American Institute of Aeronautics and Astronautics’ SciTech conference in Orlando in January.

In a conventional aircraft design, a significant amount of kinetic energy is lost in the slow-speed wake behind the fuselage and wasted in the high-velocity jet exhaust from the engines. This increases the power required. By ingesting and reenergizing the boundary layer flow, “BLI reduces the wasted kinetic energy in the combined jet and wake,” says Greitzer. With its engines mounted atop the aft fuselage, the D8 ingests about 40% of the kinetic-energy deficit.

By conducting back-to-back tunnel tests of BLI and conventional podded-engine “non-BLI” versions of the D8, the MIT team set out to quantify the benefit by measuring the mechanical power transmitted to the flow by the propulsors to maintain the same conditions. The measured power reduction with BLI was 8.4% when propulsor nozzle area was held constant, increasing to 10.4% when mass flow was kept equal. “That’s a significant benefit,” Greitzer says.

“The dominant effect is an increase in propulsive efficiency through the reduction in jet velocity,” as a result of starting with slower flow into the propulsor, says Alejandra Uranga, MIT technical lead. “The D8 engine has similar specific thrust but higher propulsive efficiency than current-generation engines like the CFM56-7. That’s why we think the results are applicable to full scale,” she notes.

The D8 model was tested at angles of attack up to 8 deg. and slideslip angles up to 15 deg. and the flow behaved well, Uranga says. Engine-out testing showed no adverse impact on the running engine and the observed fan efficiency loss from ingesting distorted boundary-layer flow was “an order of magnitude less” than the BLI benefit, says Greitzer, adding  that "there are no show-stoppers to the D8 configuration.”