Lowering The Toll On The Highway To Space

Getting into orbit around the Earth has always been an expensive proposition. Currently it costs between $2,000 and $10,000 for each pound of cargo, human or hardware, delivered to low-Earth orbit. To shrink that cost and open space to new businesses and scientific opportunities, a host of private companies, academic research teams, and government agencies are trying to develop reusable launch vehicles (RLVs).

THE REUSABLE ROTON
At the Rotary Rocket Co., San Bruno, Calif., engineers hope to build and operate the world's first fully reusable single-stage-to-orbit (SSTO) commercial launch vehicle. It will carry a crew of two, up to 7,000 lb of payload, and have a 100-flight life.

Initially, Roton was designed around three basic propositions: It would have an all-new rotary rocket engine, use an all-composite shell and structure, and it was to rely on rotors, much like an autogyro, to get back from space.

"A few years ago we were asking the financial community for funding based on these three seemingly impossible propositions," recalls Geoffrey Hughes, the company's vice president of sales and marketing. "Now composites are no longer an issue. They largely make up every new and proposed space vehicle. And although landing with rotors had a high giggle factor, we've proved the technique works and works well. But the allnew engine was a tougher sell, so we decided to change our approach."

The original concept was to build an engine that used a spinning disc with 72 combustion chambers along its out-side edge. Centrifugal force would send fuel into the chambers, letting the Roton team do away with expensive and complicated turbopumps. While the idea was rejected for the first-generation Roton, the company is holding onto the technology and hopes to use it in the future.

The current Roton uses a variation of the Fastrac engine designed at NASA's Marshall Space Flight Center in Huntsville, Ala. It was developed to be a low-cost and reusable engine. Fueled by liquid oxygen (LOX) and kerosene, it needs just one turbopump. The only electronics it carries opens and closes the fuel valves.

The turbopump in Fastrac, built by Barber-Nichols, will cost $300,000 — one-tenth the average cost of modern rocket-engine pumps. Even so, its cost should eventually drop even farther to $90,000. Engineers also saved money by using ablative cooling in which the silica-phenolic composite combustion chamber simply chars and burns away. This means the chamber must be replaced after each flight, but it eliminates hundreds of feet of meticulously welded tubing. At first, engines will cost $1.2 million apiece — about one-fifth the cost of similar engines. NASA hopes the price will soon drop to $350,000.

Roton engineers will work with NASA to modify the Fastrac to use higher chamber pressures, thus giving Roton 10% more thrust per kilo-gram of propellant burned, according to Hughes. The company will collaborate with NASA to develop a lighter, reusable and regeneratively cooled combustion chamber. "We'll pass oxygen around the engine, which will cool the chamber and warm the fuel, and then that heat value will be added back into the engine," Hughes explains.

The Roton is built to take-off and land vertically. It can also land fully loaded, so it can bring cargo from space back to Earth or land quickly after take off if there's a malfunction. Because it lands and launches vertically, there is only one load path structurally. RLVs that take off vertically and land horizontally, like the Space Shuttle, must handle vertical and horizontal loads, forcing them to use a heavier structure. The additional weight means more fuel, which carries an appreciable weight penalty. (LOX, for example, weighs 72 lb/ft ³ .)

Operating a vertical spacecraft has other advantages. Launches and landings need just a 100-ft-diameter pad. For an emergency landing, the crew needn't search for an unused runway, just an empty, flat surface.

To minimize heating on returns from orbit, Roton "flies" a ballistic trajectory, i.e., it drops like a rock. But its bucketlike shape and base-first trajectory slow it quickly. Lifting vehicles, on the other hand, must expose enough surface area to generate lift, but generating lift prolongs the high-speed portion of their reentry. Roton's fuselage walls are angled inward from the thermal flow, reducing the amount of heat they're exposed to. This gives Roton one-tenth the high-heat area as the Space Shuttle, and Roton's total heating-per-unit-area is half that of the Shuttle. But the Roton still needs a cooling system.

"We borrowed a recently declassified method that is used to cool nuclear warheads during reentry," says Hughes. Roton carries a few hundred pounds of water in a tank. Pressurizing the tank forces water through piping to the hot spots on the spacecraft's surface. There it travels through aluminum sheets made up of bonded foils with precise perforations. The perforations meter the water going onto the vehicle's skin where it forms a protective film during reentry. As the film evaporates, steam envelopes Roton and provides even more thermal protection.

Skeptics might think the Roton's seemingly unorthodox reentry via free-wheeling rotors doesn't seem possible. But back in the 1960s, the concept was studied and okayed by NASA at Mach numbers 40% higher and dynamic pressures 10 times greater than what Roton will see. The rotor blades, which will deploy in space and operate until touchdown, offer several advantages. They're lighter and have less drag during ascent than other systems. They stabilize the vehicle over the entire landing and provide smoothly modulated drag, unlike parachutes which jerk when opening. They also provide pilot-controlled flight and precision, zero-velocity touchdowns.

The four metal blades give the vehicle a 1:1 glide ratio. Roton will glide for about five miles, starting at about 28,000 ft. Glide speed is 45 knots, relatively slow compared to the Shuttle's 230 to 320 knots.

Rotary Rocket has no plans for flying Roton unmanned. "In the event of an unplanned anomaly, and these can be quite minor, having a pilot and crew puts a highly versatile self-programmed computer at the controls," says Hughes. "Our crews will have a variety of choices and options should something happen." The vehicle will carry standard avionics and off-the-shelf computers for flight control and navigation.

Hughes estimates Roton will cost about as much as a medium-sized airliner. "We should be able to reduce the cost of putting cargo into space by a factor of five initially, and there's no reason we can't drive it down by a factor of 100," he says. "After all, the theoretical energy costs for getting a person into orbit is only about $5."

While that may be true, there's still a matter of further funding. The company estimates another $100 million will make Roton operational. "After that, there's no limit to what we can do," says Hughes. "We have serious customers, both aerospace companies and world-renowned scientists, who want to use us for satellite missions and to play a part in lunar missions."

POLISHING THE CONCEPTS
At the Space Systems Design Lab at the Georgia Institute of Technology, Professor John Olds and his students are exploring different combinations of cutting edge technologies for future generations of RLVs. So far, the biggest challenges are in efficient propulsion systems, fielding lightweight structures (most likely organic and metal-matrix composites), and streamlining ground operations to minimize support costs.

One spaceship design, Hyperion, would be a horizontal takeoff and landing SSTO vehicle that uses five rocket-based combined-cycle (RBCC) engines and carries liquid hydrogen fuel. Such engines combine a ramjet operating in ejector and nonejector modes, a super-sonic ramjet (scramjet), and a rocket engine, letting the engine operate efficiently from a standstill all the way up to Mach 25.

Such engines would work in ejector mode for takeoff. In ejector mode the engine acts like a ducted ramjet or a turbofan with no moving parts. The ramjet sucks in air for combustion while the duct pulls in additional air which has fuel added and burned in an afterburnerlike arrangement. Ejector mode will take Hyperion to Mach 3, at which point the ducts close and the engine converts to a ramjet. The ramjet boosts the speed to Mach 5.5, then the scramjet kicks in. It accelerates the vehicle to Mach 10. Up to this point, Hyperion will rely on atmospheric oxygen, relieving it of the weight penalty of carrying oxygen. But it does carry LOX for use when the engine switches to a rocket and takes the craft to Mach 25 and low-Earth orbit.

"An RBCC engine has fewer parts than a turbojet, but going from ram to scramjet is going to be difficult," says Olds "There are some ramps and angles inside the engine that have to move." NASA wind-tunnel tested ejector ramjets in the late 1960s. Two companies, Aerojet and Boeing Rocket-dyne Div., are currently testing combined ram/scram engines. Olds is also heartened by the upcoming X-43 flight scheduled for this May, the first flight test of a scramjet.

There was also recent talk of fitting a D-21 drone with an RBCC engine as a quick and dirty test of the technology. The D-21 was a reconnaissance vehicle launched from SR-71s and B-52s. Unfortunately, the idea was quickly tabled over funding and disagreements over priorities.

"RBCCs probably won't get into production until about 2020," says Olds. That's when Generation 3 of the Shuttle will fly. Generation 2, scheduled for 2010, will be another rocket-based spacecraft, perhaps some variation of Lockheed Martin Corp.'s VentureStar.

To give the Hyperion some landing flexibility, Olds proposes fans that extend into the engine's flow path. Forward speed during reentry spins the fans and provides up to 5 min of flight time. This would differentiate it from the current Shuttle which makes a dead stick, or unpowered, landing.

Dead stick landings force the Shuttle to say aloft if there is bad weather or problems at the planned landing site. Similarly, the limited landing sites dictate the point at which the Shuttle can leave orbit.

Plans are for Hyperion's primary airframe structures to be metal-matrix composites such as titanium-aluminides, with graphite composite fuel tanks. NASA's slender hypervelocity aerothermodynamic research probe thermal-protection system (Sharps) will cover the surface of thin and pointed features such as the tip of the fuselage and the leading edges of the wings. Sharps uses hafnium diboride, a dense, heavy gray ceramic that can withstand temperatures up to 4,500°F. It is also strong enough to handle the stresses of takeoff and reentry despite being formed into relatively thin sections.

Toughened unified fibrous insulation (TUFI), the latest and greatest in insulating tiles, will coat Hyperion's belly and the bottom of its wings, which are exposed to the most heat during reentry. The tiles are tougher and more durable than the Shuttle's, which are fragile and can be cracked or pitted by ice, meteoroids, and maintenance accidents.

Tailorable advanced blanket insulation (TABI) will cover the top or leeward side of Hyperion. "It's a thick, white blanket made of silica-based material that can take temperatures up to 2,000°F," says Olds. "And unlike tiles, which must be inspected to ensure they're still attached, it comes in large 1.5-in.-thick sheets that are easy to install, inspect, and maintain. They're also lighter than other current methods of insulation."

Another RLV coming off the drawing boards at the Georgia Tech think tank is Argus, a slender, needlelike craft designed with a 20,000-lb pay-load. Unlike Hyperion, however, it will get to orbit using two supercharged ejector ramjet/rocket engines (SERJ) and a ground-based magnetic levitation system.

The maglev launcher will give Argus a take-off speed of 545 mph (800 fps), reducing the amount of propellant it must carry and the space-craft's size. Precisely timed magnetic fields will propel a sled carrying Argus along a track.

Several countries are already experimenting with maglev trains. A German version, for example, runs at 270 mph. The U.S. Navy is also looking at maglev to replace steam catapults for launching planes.

Engineers at NASA's Marshal Test Flight Center are experimenting with a 50-ft maglev track and hope to extend it to 400 ft by next month. They feel that a 1.5-mile track and 60,000-lb sled could help launch RLVs and other spacecraft in the 100,000 to 150,000-lb class. Besides saving weight and reducing vehicle size, maglev would give operators a flight abort option. If jet or rocket engines are malfunctioning midway down the track, there would be enough track left to stop the vehicle. And although initial costs would be high, maglev systems would use only $75 in electricity per launch.

The SERJ engine in ramjet and ejector mode uses a powered fan to pull more air into the engine, in effect supercharging it. The fan would also spin during reentry for lift-producing thrust. After the initial boost from the maglev launcher, the ramjet would take the Argus to Mach 6, at which point the craft converts to pure rocket propulsion.

Like Hyperion, the Argus fan presents a tough engineering challenge. It stays in the flow path throughout the mission, windmilling during high Mach numbers. "Once its job is finished on take-off, the fan must turn fast enough to keep pace with the incoming air, not blocking it or speeding it up," says Olds. The fan can't just stand still because that would create drag.

Both vehicles would use similar materials and thermal protection. And neither would carry a crew. This eliminates the complexity and weight of a pressurized cabin, life-support systems, windows, and the crew itself.

"Both Hyperion and Argus will be able to carry passengers, like a bus with a computer driving it," Olds says. "Aircraft can already land themselves, and the Soviets flew Buran, their space shuttle, from take-off to landing unmanned the one time it flew. The public perception now might be that they wouldn't fly without a pilot, but by 2020, pilotless flight should be more acceptable to the traveling public for ferrying staff and supplies to the space station. Besides, even with a crew, today's airplanes and spacecraft still carry a computer. One is just the backup for the other and I don't want to say which is the backup."