With its X-65 demonstrator, DARPA aims to test whether today’s control surfaces can be replaced. Paul Marks spoke to the program leads about the design and forthcoming flight tests.
For over a century, aircraft pitch, yaw and roll control has been provided by a familiar trio of control surfaces: ailerons, rudders and elevators. But have these heavy, draggy, difficult-to-actuate mechanisms had their day? Could the aerodynamic maneuvering forces these legacy control surfaces generate be supplied by other means?
The answer to those questions could begin to emerge in late 2027. That’s when the X-65, a 3-metric-ton demonstrator commissioned by DARPA and built by Aurora Flight Sciences, is to take to the skies to test a new method of steering: active flow control, or AFC.
This remotely piloted aircraft is to maneuver by firing small jets of pressurized air, tapped from its single jet engine, through banks of nozzles that are strategically distributed around the surfaces of the wings and tail structure, to give full control of the airplane in three dimensions.
If it works, AFC could bring a swath of advantages. Getting rid of the need for heavy, hinge-jointed control surfaces and their complex hydraulic or electromechanical actuators could make airplanes lighter, less complex and more maneuverable, says Christopher Kent, who manages X-65 within DARPA’s Control of Revolutionary Aircraft with Novel Effectors (CRANE) program.
“There’s lots of good reasons to do this, some military, and some very interesting ones that are more to do with imagining a world where aircraft no longer need control surfaces, which allows you to really simplify parts of the aircraft,” Kent told me in a mid-February interview.
“You could potentially even 3D print the whole aircraft wing with no hydraulics in it and just have all the air distribution surfaces designed in.”
The idea of blowing pressurized air from aircraft surfaces isn’t a new one, he notes. “Blowing-based lift augmentation has been around for decades — the British Buccaneer fighter had it, and the Japanese US-2 amphibian plane still has it,” says Kent.
Both of those designs had pressurized air bled off the engine or auxiliary power unit and blown over the wing surfaces, plus a horizontal stabilizer to enhance lift and stall suppression in short takeoff-or-landing applications. Called boundary layer control, this technique helped the Royal Navy’s Blackburn Buccaneer in carrier landings in the 1950s, and the ShinMaywa US-2 today in search-and-rescue scenarios on water. By forcing air through a vent to blow over a flight surface, the separation of airflow from that surface can be prevented — seriously enhancing lift at low speeds.
Blown air has also been used to successfully steer aircraft, says Kent, but only on small tailless, flying-wing-style drones. That research was led by another Pentagon lab, the Office of Naval Research, alongside the Illinois Institute of Technology and NASA.
“At DARPA, we’re particularly interested in proving that you can do this at a much larger scale than previously demonstrated with those drones, which weigh just hundreds of pounds,” he says. “We’re going to do it at a very large size, with our 7,000-pound [3-ton] aircraft, because we need to fully understand if, and how, active flow control scales.”
Building X-65
As of late February, Aurora Flight Sciences was preparing to mate the aircraft’s diamond-shaped swept wings to the fuselage, working toward “a complete aircraft rollout sometime later this year,” says Kent. With a 9-meter wingspan, X-65 will have close to the wingspan and mass of a six-seater Beechcraft Baron 58 twinprop.
An aircraft that size cannot take to the skies maneuvered solely by experimental vectoring technology. So as part of Aurora’s FAA experimental certification requirements, X-65 will also have regular control surfaces as a safety backup.
“We’ll have both systems, with the hydraulics and the AFC air jets, so we have the ability to fall back to the conventional control surfaces,” says Kent. “It’s a crawl, walk, run approach.”
Central to the design is a bleed air distribution system, which taps pressurized air off an auxiliary power unit above the turbofan engine that runs the length of the airplane. That bleed air will be routed through the aircraft via pipes, regulated by flight computer-controlled valves, that direct air volume flow to 14 differently sized banks of nozzle-like air effectors embedded in the wing and tail surfaces.
Kent describes each effector as “basically a long tube with a number of small holes in it that lead out to nozzles on the wing surface.”
One advantage of distributing relatively low-pressure air this way is eliminating the need for the highly pressurized hydraulics that today’s aircraft require to move their control surfaces. Leaks in such systems can do maintenance engineers serious harm through burns, severe cuts or fluid injection.
A big unknown with AFC acting as a control surface replacement is whether the sweep of the wings affects the operation of the air jet banks. Wings are swept one way or the other to reduce drag, but how will that air-pressure-based effect be modified when jets of air are providing the control? To find out, Aurora has designed a reconfigurable diamond-shaped wing with removable sections so that multiple wing sweep shapes — forward, backward and somewhere in between — can be tested.
“The two forward-most wings have a backward sweep angle, and the aft ones have a forward sweep angle. The outboard ones are different again. All these different sweep angles let us check what each one does best under different blowing configurations,” says Kent.
These removable wings largely account for X-65’s stumpy appearance, because they require Aurora to place the fuel tanks in the fuselage.
“The volume of the fuselage is driven by the amount of stuff we have to fit in there,” says Kent. “We have to power both the conventional hydraulic system and the air distribution system. And because the wings are modular and removable, we’re not storing fuel in the wings.”
Based on the results of model-based wind tunnel tests, DARPA anticipates some novel maneuverability possibilities from the aircraft’s air-driven vectoring system.
“In a conventional aircraft, it’s very hard to turn it in the air relative to its flight angle, what’s known as ‘crabbing’ the aircraft through the air,” says Kent, referring to a phenomenon mainly seen on high crosswind landings, when a plane approaches the runway threshold at quite an angle to the runway center line, only to be straightened up dramatically on landing with a sharp rudder input.
“We think AFC may allow us to actually potentially point the nose of the X-65 a little bit off of that and crab sideways” in any wind, he says — a maneuver that could be tactically useful for some missions.
And because AFC vectoring is based on fast, reactive firings of that pressurized bleed air, rather than slower aileron, rudder and elevator movements, designers suspect it will also be easier to keep the airplane straight and level in turbulence.
“In theory, this type of AFC control could potentially be much quicker for responding to gusts, or changes in the air environment,” says Kent. If so, that would be a boon for aircraft carrier landings: “You can turn these air jets on and off so quickly, and so fractionally over the surface, they might be able to do that even better.”
Those fast, fractional bursts of vectoring air emitted on wings and tail are going to need a hyper-accurate control system to regulate the air volume flow out of the effectors. Kent says the flight computer and autopilot Aurora is developing to do this is a “key piece of core technology that will convert your intent into how the AFC jets perform the maneuver that you so desire.”
Preparing for first flight
X-65’s flight computer is slated to go into ground-based bench tests later this year, incorporating learnings from “a billion computer hours” of computational fluid dynamics simulations, Kent says. If all goes as planned, taxi tests will commence in early 2027, and first flight — with an Aurora remote pilot in control — will take place in late 2027.
Aurora is encouraged by the results of the testing done to date, says Larry Wirsing, the company’s vice president of aircraft development: “The most surprising finding we’ve made while developing the X-65 has been the versatility and scalability of Active Flow Control as an aerodynamic tool,” he told me via email. “While AFC was initially considered a ‘localized’ solution to manage the boundary layer, through multiple wind tunnel tests, subsystem validations, and full-scale bench testing we’ve now demonstrated that integrating AFC into the control loop not only works – but also enables entirely new design trades that can improve airplane performance.”
And despite AFC’s novel way of vectoring an airplane in three dimensions, Aurora is following a familiar process to train its remote pilots to fly the X-65, says Wirsing. “We’ll start with training on concepts of operations to learn how the aircraft works, and the behaviors that are expected, and then we’ll move into system reviews to get more detail on the individual systems and how they will all work together.”
He adds: “Once that ‘book’ learning is done, each crew member will participate in sessions on our hardware-in-the-loop simulator to run through normal flights using the same ground station hardware that will be used in flight. And when the team is comfortable with normal operations, we’ll inject abnormal and failure situations into flight scenarios to ensure that the entire crew knows how to diagnose and resolve any problems that may come up during flight.”
But AFC won’t be working from the moment the airplane first leaves the ground. Instead, “We’ll first fly under the conventional control surfaces, and then convert over to active flow control as we work through the test program,” Kent says.
“Our first flight under AFC is where DARPA’s responsibility ends — and after that, the program will live on through Aurora’s efforts.”
