The Space Reviewin association with SpaceNews

Skylon illustration
Under development for more than 20 years, Skylon could provide low-cost access to LEO, if it can overcome engineering and other obstacles. (credit: Reaction Engines Ltd.)

Skylon: ready for takeoff?

Bookmark and Share

For decades, the ultimate goal in space transportation has been the single-stage-to-orbit reusable launch vehicle, or SSTO RLV. Such vehicles, engineers and other space advocates have argued, would make space transportation much more like air transportation: airplanes, after all, aren’t thrown away after a single flight, and don’t jettison parts en route to their destinations. A launch vehicle that could operate in the same fashion—even if it didn’t necessarily take off or land like an airplane—would, proponents claimed, be simpler, safer, and less expensive than conventional expendable rockets and thus open up new markets for space activity.

There’s just one problem with this approach: building an SSTO RLV—or any orbital RLV, for that matter—has proven to be extraordinarily difficult. Over the last few decades a number of projects, both government and commercial, have tried to develop such vehicles, only to be abandoned usually for some combination of technical and financial problems. For the last decade, work on such vehicles has been all but abandoned. Government agencies, like NASA and the Defense Department, have focused on expendable rockets to meet exploration and national security needs, while companies have been working on either suborbital RLVs (less difficult to develop than their orbital counterparts) or lower-cost expendable rockets, like SpaceX’s Falcon series, that have a long-term goal of reusability for at least some of its stages.

“[N]o impediments or critical items have been identified for either the SKYLON vehicle or the SABRE engine that are a block to further developments,” the ESA report stated.

One SSTO RLV project that has received recent attention, though, has been Skylon, a vehicle proposed by British company Reaction Engines Ltd. The long, sleek Skylon would take off from a runway like an aircraft and soar to orbit, deploy its payload, and return to Earth for a runway landing. Powered by engines that act like jet engines in the lower atmosphere and rocket engines above it, Skylon could carry over ten tonnes of cargo to low Earth orbit for potentially less than $1,000 per kilogram, a cost far less than existing expendable launch vehicles.

The new interest in Skylon came after the UK Space Agency released a report last month prepared by the European Space Agency that assessed the feasibility of Skylon. ESA engineers looked at the technical feasibility of the vehicle while economic consultants studied the vehicle’s business case. Neither group found any showstoppers that would prevent the vehicle’s development. “[N]o impediments or critical items have been identified for either the SKYLON vehicle or the SABRE engine that are a block to further developments,” it stated.

While the attention devoted to Skylon is new, the project isn’t. Work by Reaction Engines on the Skylon concept stretches back more than two decades. Skylon had its origins in another SSTO RLV project called HOTOL by British Aerospace and Rolls Royce in the 1980s. When the British government declined to further invest in HOTOL in the late 1980s, a group of engineers led by Alan Bond created Reaction Engines to further the technology that had been planned for HOTOL.

Skylon, however, is not simply a continuation of HOTOL. “It wouldn’t have worked,” Roger Longstaff, a consultant with Reaction Engines, said of HOTOL in a presentation about the company’s Skylon work at the Space Access ’11 conference in Phoenix in April. With rear-mounted engines, HOTOL had major problems with its center of gravity. HOTOL also required a complex trolley to support the vehicle for takeoff. The company has addressed those and other issues with the Skylon design.

At the heart of Skylon is its engine, called the Synergistic Air-Breathing Rocket Engine, or SABRE. Housed in curved nacelles on the tips of stubby wings, the twin SABRE engines have air intakes that allow the vehicle to use atmospheric oxygen as oxidizer, with liquid hydrogen fuel, from takeoff to an altitude of 25 kilometers, at which point the vehicle is traveling to Mach 5.5. At higher altitudes SABRE becomes a more conventional rocket engine, using onboard liquid oxygen to accelerate the rest of the way to orbit. This approach greatly reduces the amount of liquid oxygen the vehicle has to carry, decreasing its takeoff weight, which in turn has other effects on the vehicle’s design.

“The Skylon C1 design could be realized with current technology, provided the engine demonstration proceeded as planned,” Longstaff said, referring to the current iteration of the Skylon design. “The entire Skylon program rests upon the SABRE engine.”

“We bet the farm on this, so it better work,” Longstreet said of the upcoming test of the SABRE engine’s precooler.

The next major milestone for Skylon and SABRE could come as soon as this month with a planned test of a key element of the engine technology. That test will focus on a component of the engine called the precooler, designed to cool air coming into the engine at high Mach numbers to temperatures that allow other components of the engine to operate within known limits of current technology.

The test will be of a subscale version of the planned SABRE precooler, attached to a Rolls Royce Viper jet engine, but Longstaff said it will effectively test the technology that is essential to the engine. “It will be sufficient to verify the aerodynamics and thermodynamics and the frost control technology” designed to prevent water vapor in the atmosphere from freezing within the engine. “We bet the farm on this, so it better work.” (As this article was going to press, a spokesperson from Reaction Engines confirmed that the tests were scheduled to start this month, “a few months earlier than originally scheduled.”)

Even if that test is successful, though, the company has a long way to go before Skylon takes to the skies. The company plans to build a full-scale boilerplate version of the engine for ground tests by 2013 or 2014. A subscale Nacelle Test Vehicle, about nine meters long, will be flown at speeds of up to Mach 5 to verify the internal design of the engine and its aerodynamics. “By 2014 we would be ready to pass over the manufacturing drawings for the engine to a contractor, and we would have the analysis of the airframe completed,” Longstaff said. If that work goes as planned and the project is funded, he added, “we could fly about 2018 and we could be fully operational by 2020.”

Although Skylon has received some laudatory press in recent weeks associated with the ESA review, there are numerous obstacles that still lie ahead for the vehicle. The ESA review devoted much of its attention to the SABRE engine, identifying it as “critical for the successful development of the SKYLON vehicle.” However, the vehicle includes other critical technologies, such as an active cooling system to protect the vehicle from the heat of reentry, which received less attention in the review. The ESA report concluded that the overall design “does not demonstrate any areas of implausibility.” But as any engineer can attest, the path from “not implausible” to “operational” can often be long, circuitous, and expensive.

The biggest obstacle, though, could be financial. Reaction Engines has projected the Skylon development cost at roughly that of the Airbus A380 superjumbo, reported to be about €12–14 billion (US$17–20 billion).

Another challenge will be the business case for Skylon. The current Skylon C1 design can carry up to 10,275 kilograms to low Earth orbit (LEO), with a goal of 12,000 kilograms; Longstaff noted in April that it could carry about 7,000 kilograms to the orbit of the International Space Station and 3,000 kilograms to Sun-synchronous orbit. That suggests Skylon may be undersized to serve many existing portions of the commercial launch market, such as the launch of geosynchronous orbit (GEO) communications satellites that can weigh in excess of 6,000 kilograms, plus an upper stage needed for transfer from LEO to GEO.

“It’s been apparent that the goal of 12 tonnes to low Earth orbit, at 300 kilometers [altitude], is perhaps a bit small,” Longstaff said at Space Access. The company, he said, is looking at an enhanced version of the Skylon, designated the D1, using an upgraded version of the SABRE engine and an optimized trajectory that could carry 15 tonnes to LEO.

Limited payload capacity is a problem inherent with many RLV designs, though, which means such vehicles may find more traction in the market through new applications. A study by London Economics included in the ESA report identified several “downstream services” that could be enabled by Skylon, including space tourism, space solar power, space manufacturing and research, and even, curiously, e-commerce, but did not include a quantitative analysis of each in the final document.

Reaction Engines has studied the inclusion of a passenger module in the Skylon cargo bay that could carry 30–40 passengers; a revised concept published in 2008 included room for 20 people for missions to ISS orbit. (Interestingly, Skylon is designed to fly, at least for satellite deployment missions, without a crew onboard.) Skylon, however, may not be well-suited to point-to-point high-speed transport: the vehicle requires on takeoff a runway at least 5.6 kilometers long, which is longer than virtually any existing runway. In addition, the first four kilometers of that runway must be “stronger than normal”, according to the company’s web site, due to the heavy loads the vehicle puts on the runway surface during its high-speed takeoff.

The biggest obstacle, though, could be financial. Reaction Engines has projected the development cost at roughly that of the Airbus A380 superjumbo, reported to be about €12–14 billion (US$17–20 billion), although the ESA report cites a “pessimistic estimate” of the development cost, provided by a costing model developed by the company, of only $12.3 billion. In any case, that is a large sum of money, especially since company officials have indicated they plan to privately finance the vehicle’s development rather than government funding.

While company estimates put the per-flight operating costs as low as $9.5 million, that requires a flight rate of 70 missions a year; costs when the vehicle enters service will be on the order of $30–40 million a flight, according to the Reaction Engines web site. That could put the vehicle at a competitive disadvantage to some expendable vehicles. Assuming a capacity of 15 tonnes to LEO, initial Skylon costs would be $2,000 to nearly $2,700 per kilogram. By comparison, SpaceX’s Falcon Heavy, with a capacity of 53 tonnes to LEO and a projected cost of $80–125 million per launch, would come in at $1,500 to $2,350 per kilogram. Moreover, development of the Falcon Heavy is expected to cost a small fraction of the Skylon’s projected cost, and it could enter service as soon as 2013, years before Skylon will be ready in even the most optimistic scenario.

The fact that Skylon development has continued, at even a modest pace, while other RLV ventures have come and gone is a testament to the effort put into the project by Reaction Engines, and this summer’s engine tests could be a major milestone in that project. However, it’s clear Skylon has many major challenges, both technical and financial, that its backers will have to overcome before the first vehicle takes off from a runway for space.