Honda's 'research project' expected to fly by year-end

 - August 5, 2008, 8:59 AM

Under the cover of private funding, Honda has been secretly and very seriously developing its six- to eight-seat light turbofan twin. Though the automaker steadfastly maintains it has “no business plan” to manufacture the business jet, the project aircraft has a name, HondaJet, and the development program is well advanced. For example, Garmin is securely on board, with the soon-to-fly prototype HondaJet carrying an open-architecture version of the G1000 integrated cockpit (see box on page 107).

Other developments on the HondaJet include, to date:

• Honda’s in-house-designed and -built twin-spool 1,670-pound-thrust HF118 turbofan has been flying for more than a year on the left pylon of a CitationJet testbed.

• Honda R&D’s new-design SHM-1 natural-laminar-flow wing section has been flown (as a gloved section) on a T-33 testbed.

• Detailed structural and wind-tunnel tests have been performed on the aluminum wing (with its radical over-the-wing engine nacelles).

• The composite fuselage (with laminar-flow nose section) and aluminum empennage have also been tested in low-speed and transonic wind tunnels and structural analysis has been performed.

• The landing gear has been designed (including steer-by-wire nosewheel) and the trailing-link main gear has been drop tested.

• Following ongoing ground tests at Honda’s purpose-built facility in Greensboro, N.C., the flight-test program is expected to begin before year-end.

These developments were documented in a 17-page paper presented at an American Institute of Aeronautics and Astronautics meeting in July by Honda R&D Americas chief engineer Michimasa Fujino. Illustrated with 42 photos, graphs and line drawings, the paper also revealed other details in the development of the Japanese carmaker’s proposed business jet.

The soon-to-fly prototype is 41.14 feet long, 13.21 feet high (at its T tail) and has a wingspan of 39.87 feet. Max takeoff weight is quoted at 9,200 pounds and estimated max speed is 420 knots at 30,000 feet. Honda computes specific range as 0.8 nm/pound at 389 knots and 41,000 feet, for an IFR range of 1,100 nm. Inside, the 4.89-foot-high, flat-floor cabin is 15.09 feet long, allowing a more utilitarian eight-seat layout (including both cockpit seats) in place of the roomier six-seat floor plan envisioned as the normal configuration. Pressure differential is projected as 8.7 psi, for an 8,000-foot cabin at 44,000 feet. The anti-icing system consists of bleed-air heating for the wing leading edges and an electrically heated windshield, though no mention was made of protection for the engine inlets or the T tail.

The company’s assertions of “no business plan” notwithstanding, the paper reports that Honda has conducted market surveys in five U.S. cities to identify attributes most desired in an “entry-level, lightweight business jet.” According to Fujino, “The HondaJet is designed to satisfy these needs.” But supporting the carmaker’s assertion that it has no immediate plans for production, the paper reveals that the wing and fuselage samples tested for structural integrity are the same ones that will be assembled for the flying prototype. As such, they were tested to only 80 percent of design loads and data extrapolated.

Most interesting is the design of the over-the-wing engine-mount configuration for the airplane’s pair of HF118 engines. In its market survey, Honda identified passenger comfort at the top of the list of desired attributes. (Similarly, Raytheon designed the Premier I light jet with cabin size as a top priority.) With no carry-through structure needed in the aft fuselage for its engine pylons, Honda’s over-the-wing configuration allows a full-width cabin farther aft, maximizing interior dimensions.

Tradeoffs included the challenge of countering wave-drag interference from the pylons atop the laminar-flow wings at high speeds and aeroelasticity issues–designing the wing/pylon combination with just the right stiffness to eliminate wing flutter. Using computational fluid dynamics software combined with wind-tunnel tests, Honda R&D was able to position the engine pylons with enough sweep and at the appropriate angles to arrive at what the company says is equal amounts of aerodynamic lift and efficiency as with a clean wing. Further computer and wind-tunnel testing confirmed that, in the chosen configuration, the engine inlets would have sufficient airflow, even at high angles of attack.

Addressing flutter, Fujino’s paper further stated, “The location of the engine mass and the stiffness of the pylon relative to that of the wing are important.” He conducted his own theoretical analyses, backed by tests at Honda’s own low-speed wind tunnel and transonic wind-tunnel flutter tests conducted at the National Aeronautical Laboratory transonic flutter wind tunnel. “Based on these results,” Fujino wrote, “the wing stiffness and mass distributions were designed to satisfy the flutter-clearance requirements.”

Honda counts heavily on the benefits of natural laminar flow (NLF) to achieve the efficiency it expects from the HondaJet. In June last year, engineers (including Fujino) at the Honda R&D Wako Research Center in Japan published a separate nine-page paper at AIAA on their airfoil development research. The Honda R&D engineers cited three previous generations of NLF airfoils and their design limitations for application on a business jet.

Those airfoils and their shortcomings included the NACA 6-series designs from the early 1940s, as used on the North American P-51 Mustang. That early NLF airfoil was vulnerable to leading-edge contamination (hardly ideal for European operations during the winters of World War II) and could exhibit nasty stall characteristics in the wrong flight configurations, as many Mustang pilots learned to their dismay.

Next cited were NASA’s NLF(1)- 0215F and -0414F, airfoils designed for low-speed applications. Honda R&D concluded they suffered from too large a pitch-down moment and drag-divergence Mach numbers too low for use on a business jet. Finally, the NASA HSNLF(1)-0213 airfoil offered insufficient lift at low Reynolds numbers combined with only a 13-percent thickness-to-chord ratio, which limits fuel capacity.

The goal of designing the Honda airfoil was to achieve a high drag-divergence Mach number and slight nose-down pitching moment while retaining the NLF’s characteristic low drag for efficiency at cruise speed. Docile stall habits and tolerance for leading-edge contamination were also considered paramount, along with a stated requirement of 15-percent-chord thickness to ensure sufficient fuel capacity. Honda engineers and their computers went to work on the compromise.

The resulting airfoil was tested in both low-speed and transonic wind tunnels, as well as in flight tests using a modified Lockheed T-33. For the tests, the T-33’s entire wing was covered with polyurethane foam contoured in the desired shape, then covered with fiberglass skin. Incorporated into the “glove” were 119 static pressure holes on the upper and lower surfaces to measure pressure differential. Besides the data-collection computers plugged into the sensors, Honda mounted an infrared camera in the T-33 cockpit to record the laminar-to-turbulent boundary-layer transition throughout the flight tests. The effects of surface roughness and wing steps were also part of the flight-test regimen.

Results of tests conducted in both types of wind tunnel, as well as on the T-33 test aircraft, corresponded favorably with the predictions Honda engineers had made based on computer projections for the airfoil. They are convinced the HSM-1 airfoil, with winglets to increase aspect ratio, will provide a wing with the right combination of lift, room for fuel, docile stall characteristics, low susceptibility to contamination and favorable high-speed pitch-down tendency appropriate for a light business jet.

For low-speed performance, the HondaJet has double-slotted, 30-percent-chord flaps deployed with a mechanical linkage. Wind-tunnel tests using one-sixth- and one-third-scale models were used to refine the shape and positions of the flaps and vanes, revealing what Honda considers a flap system that satisfies its stall-speed requirement. Though no specific stall-speed target was mentioned in the paper, the general mission statement indicates that Honda’s aim is to design the HondaJet to be competitive with other light jets designed for single-pilot operation. The flap system was also tested for structural integrity and a mechanical linkage in the actuation mechanism prevents asymmetrical deployment, always a nasty surprise on final approach. The flaps have two positions, 15.7 degrees for takeoff and 50 degrees for landing.

While Honda R&D’s one-sixth-scale wing and nacelle/pylon was in the wind tunnel, engineers also ran tests on engine inlet pressure at 10-, 15-, 18- and 26-degree angles of attack. Honda also tested the inlet pressure at -18, zero and 18 degrees of sideslip angle. Due to the over-the-wing configuration, Honda recognized the engine-inlet-pressure tests to be of great importance. The tests showed that the distortion was within acceptable limits for the wing/engine/nacelle configuration. Further aerodynamic tests also assured Honda’s R&D engineers that the airplane’s stall characteristics would be straightforward and benign.

To further increase the efficiency of the design, Honda also decided on an NLF shape for the nose section of the HondaJet fuselage, as well as the airfoil. Again, Honda used wind-tunnel tests to confirm the ideal NLF shape needed to “maintain a favorable pressure gradient and minimize cross-flow instabilities.” The resulting nose configuration reduces fuselage drag by about 10 percent compared with that of a turbulent-flow nose section. Further, the rear-fuselage upsweep configuration is designed to minimize separation, lessening drag at cruise speed while providing enough clearance for takeoff and landing angles of attack.

Though its wings and empennage are made of aluminum, the HondaJet fuselage is graphite composite to save weight and money. It uses a 350-degree-Fahrenheit-cured epoxy prepreg reinforced by carbon fiber. Honeycomb sandwich-panel construction is used for the compound curves required by the laminar-flow cockpit/nose sections and tailcone while integrally stiffened panel structure was selected for the constant cross-section (“barrel”) portion of the cabin. With an apparent eye toward economy of production, Honda addressed what it described as “a technical challenge” to design both the sandwich- and stiffened-panel sections so that they could be cured in an autoclave simultaneously “to reduce weight and [production] cost.” The engineering paper also pointed out that the constant-cross-section portion of the fuselage could be stretched easily for possible expanded versions of the HondaJet.

A proof test with a total of six computer-controlled actuators was used to validate the structure of the fuselage. As with structural tests on the wing, the test load was limited to 80 percent of the limit load since the example used for the tests is the same one used for the flying prototype. According to Honda, “The strain, displacement and reaction-force data were measured and compared with those from a finite- element analysis, and the results were extrapolated to evaluate the limit-load condition.” In other words, engineers liked what they saw in the fuselage strength tests.

In contrast to the 80-percent limit on testing the wing and fuselage, Honda attached the T-tail empennage to a dummy rear fuselage and tested it to ultimate load. It endured the test without deformation. Honda also performed extensive wind-tunnel flutter testing on the T tail with satisfactory results–especially critical in a T tail and even more so since the HondaJet’s tail structure doesn’t have the added beef of an engine-pylon carrythrough.

The landing gear is fundamentally standard, with hydraulically driven retraction mechanism, full-coverage gear doors and trailing-link suspension for the mainwheels, similar to that of the Cessna Citation Bravo. Noteworthy is the HondaJet’s steer-by-wire nosewheel with a hydraulic actuator controlled by a pulse-width modulus control. The system has pilot-controlled separate modes for taxi (plus or minus 50 degrees off center) and takeoff (limited to 10 degrees).

Fuel is carried in four tanks, one in each wet wing, a carry-through tank and a rear-fuselage bladder tank located close to the c.g. Honda released no fuel-capacity specs. The system has single-point refueling capability and fuel is transferred to either wing tank via a pump in the carry-through tank. Each wing tank feeds its corresponding engine through collector tanks under their respective pylons. A fuel-transfer management unit maintains proper fuel levels in the wing tanks and an automatic crossfeed system corrects fuel imbalance.

It would be hard to estimate how much Honda has invested in the HondaJet so far. But with its purpose-built research facility in Greensboro, N.C. (a far cry from the “rented hangar space” that has been reported in the past), and the in-house low-speed wind-tunnel, just developing such infrastructure alone would require a substantial budget. There is nothing specific in the published paper to indicate that Honda has any plans to produce the HondaJet, but there is little in the report that would contradict such plans, either. It should also be noted that the investment required to develop an aircraft pales when compared with the money needed to establish a sales and service network.

Both established and startup aircraft manufacturers frequently tout new programs and create news events out of even minor milestones. Most manufacturers also rely on publicity to rally boardroom support, outside investment capital or both.  Deep-pocketed Honda, however, needs neither publicity nor investment funding to develop its HondaJet, so it’s difficult to predict where the program might be going. One thing is clear: it’s moving with more steam than most observers envisioned.