We have completed maintenance on DiscoverMagazine.com and action may be required on your account. Learn More

Private Rockets

If you're sorry you didn't buy a few shares of Federal Express about 20 years ago, read this.

By Dan Winters and Fred Guterl
Apr 1, 1999 6:00 AMNov 12, 2019 4:58 AM


Sign up for our email newsletter for the latest science news

By the time Neil Armstrong walked on the moon in July 1969, two rocket scientists--Gary Hudson, a college dropout, and George Mueller, perhaps the most respected engineer of his generation--already knew that something radical had to be done about the way we get ourselves into space.

The problem is a mean trick of physics called the rocket equation. Every rocket engineer learns about the rocket equation—first to love it, then to hate it, eventually to accept it. The equation tells you how much fuel your rocket needs to pry itself from the claws of gravity. The premise is deceptively simple: the heavier the rocket or its payload, the more fuel it takes to reach orbit. But fuel is heavy; so when you plug the weight of the fuel into the rocket equation, you find that you have to add even more fuel. Adding fuel means building a bigger tank, which makes the rocket heavier, which means more fuel. And so on.

The rocket equation first became a serious nuisance during the race to the moon. Engineers came up with a stopgap solution: rockets with several stages. Jettisoning each stage as it used up fuel reduced the total weight of the capsule that ultimately went into orbit, but some pretty expensive hardware got dumped into the ocean in the process. Hudson and Mueller wondered if it was possible to build a single-stage rocket that could be used over and over again. Would the rocket equation allow it?

The two men didn’t know each other, but they were of one mind. If somebody could build a rocket that could be refueled again and again and again, a beat-up space-going pickup truck, it would change everything. Space would no longer be the province of government bureaucracies and gold-plated aerospace contracts. Small companies, even guys like Hudson and Mueller, could scrape together enough money to get into orbit. And that would open up the solar system to manned exploration. Because once you’ve climbed beyond reach of Earth’s gravity, to paraphrase Robert Heinlein, it doesn’t take a whole lot more fuel to get just about anywhere out there.

Neither Hudson nor Mueller quite realized that they were hearing a siren’s call, or that it would be three decades before they finally got their chance to beat the rocket equation.

Hudson took his first few steps on the path to his destiny when he realized that he was bored with the countless picayune mind-numbing details that make up the discipline of engineering. And he had become so taken with the idea of a reusable rocket that he dropped out of college to build one.


Stephen Wurst learned an important lesson while trying to build the ill-fated National Aero-Space Plane. It was intended to whisk passengers from New York to Tokyo in the time it takes to drive to the airport. But it required speeds of Mach 20 or more, which turned out to be quite an expensive concept. “People in the commercial sector aren’t interested in new technology,” says Wurst. “They want to know how much it will cost.”

In 1994, shortly after the project fizzled, Wurst founded Space Access L.L.C. in Palmdale, California, with the aim of designing a less technically ambitious vehicle that can loft satellites into orbit. Rather than use rocket engines, which have to lug their own oxygen, Wurst wants to use a ramjet, a sort of super jet engine commonly used on planes that travel three or more times the speed of sound. Like jet engines, ramjets mix and compress incoming air with fuel, but they do so with a clever geometric design rather than with huge compressor blades. Because they have no moving parts, they’re more reliable than jet engines.

The problem is, without blades to suck air, they’re useless when the plane is moving slower than about 1,200 miles per hour. Wurst and his colleages, however, managed to find a way to force air and fuel into the jet at great pressure, simulating the ramjet effect even at rest. “I told my team, ‘Go make this work at low speeds,’” Wurst says. “It’s not very exotic to tell people to make something that goes fast go slow, but it’s the sort of problem a commercial endeavor can handle. The hard part was already done.”

Wurst’s idea is to mate the engine with a new airplane design, take the craft up to an altitude of about 30 miles, at which point the ramjets can no longer be used, coast to an altitude of 60 miles, then release the satellite payload with a small conventional rocket attached. The space plane will land on a runway, while the rocket, after delivering its payload, will fly back to the ground like the space shuttle. Wurst has one big problem: he has managed to raise only a tiny fraction of the “hundreds of millions” of dollars he says he’ll need to build a prototype. But he remains confident he’ll deliver satellites into orbit by 2003.—Jeffrey Winters

Ten years later he had found a wealthy Texan to bankroll a conventional multistage rocket. It wasn’t reusable, but it was cheap, and Hudson, 30 years old, had to take what he could get. It took him and his team of engineers seven months to build it and less than seven seconds to blow it up. The problem was a stuck valve. The person who was supposed to monitor the valve had been busy taking photographs.

For a long time it seemed as though that might be the end of the story. But as luck would have it, the cold war ended and the economy was on a roll. In 1990, Motorola unveiled plans to put up its Iridium network of 66 satellites to beam signals to handheld gadgets like cell phones anywhere on the globe, and several other firms quickly followed. Suddenly, thousands of satellites were sitting around in crates waiting for a ride to low Earth orbit. The bull market of the century sent venture capitalists scurrying around for upstart companies that could build cheap rockets.

Meanwhile, Hudson had taken a job as a consultant for, among other firms, Kistler Aerospace of Kirkland, Washington. Working at Kistler was bittersweet. The firm was building his dream, a single-stage reusable rocket called the K0. Hudson had been hired to work on propulsion systems.

At Kistler, Hudson met another consulting engineer, Bevan McKinney, who had his own ideas about reusable rockets. They hit it off and in 1996 formed Rotary Rocket. They managed to raise several million dollars from two individuals, one of which was novelist Tom Clancy. Before long they had enough money to throw up some tin-roofed buildings in the Mojave Desert and start building a rocket.

Mojave Airport begins just two blocks behind a strip of fast-food restaurants and gas stations along Route 14, but it seems to extend indefinitely into the broad desert valley. There among the Joshua trees and tumbleweed, old Boeing 727s sit on blocks waiting to be stripped for spare parts. According to a billboard at the edge of town, Mojave is also “home of the Voyager,” the incredibly light airplane designed by Burt Rutan that set a record by becoming the first single-engine airplane to fly around the world nonstop. Rutan’s firm, Scaled Composites, builds odd-looking contraptions like the Vari-eze that pilots find delightful to fly. Scaled Composites is building Hudson’s rocket, the Roton. Though not finished, the Roton looks a lot like a 63-foot-inverted plastic ice-cream cone: fat at the bottom, tapering slightly to a blunt nose cone.

One of the key reasons Hudson believes the time is ripe for a reusable rocket is the availability of extremely light, extremely rigid materials loosely termed composites, which means combinations of different materials, in this case not ordinary forms of foam and plastic. The plastic comes in sheets with reinforcing carbon fibers woven in a particular way that gives them properties of lightness, strength and resistance to heat. The method that Scaled Composites technicians use to make the structures doesn’t seem very techie. First, they take a huge block of ordinary Styrofoam and hollow it out into the shape they want the structure to be, like a mold. Then they lay down a sheet of plastic. Then they glue on a sheet of foam. Then they glue on a sheet of plastic. That makes a panel. Piece together a bunch of panels just so, and the resulting structure is impossibly rigid.

Hudson takes pride in his low-budget approach to rocket building, but it’s tinged with consciousness of past failures. “I was once introduced to a NASA engineer at a party as the guy who blows up rockets but takes only 25 people to do it, as opposed to 25,000,” Hudson says. “I could take off my shirt and show you where the arrows are. But it means we actually have a better chance to pull this off than probably anybody else. We’ve learned from the mistakes we’ve made, and we’ve made many.”

Unlike his earlier ventures, this project is one on which Hudson has concentrated mainly on the big picture, leaving his partner, Bevin McKinney, to the details. “At a detailed level, I wouldn’t earn my keep at an aerospace company,” he says. “At the level of an intuitive engineer, at a systems level, I’m pretty good. Bevin’s a much better engineer than I, but I’m the guy who says, ‘Why don’t we do it this way?’ And I push. I push the whole company in a direction. Fifty percent of the time I’m dead wrong, and sometimes people get frustrated. But you learn from those mistakes.”

McKinney provided the original idea for the Roton. For years he had been mulling over a design that combined the characteristics of a helicopter and a conventional rocket. Some experimental reentry vehicles were built back in the 1960s and 1970s that used rotating blades to slow them down from supersonic speeds. But McKinney added the notion of powering the blades with a tiny rocket engine mounted on the tip of each blade.

Hudson, however, insisted that the Roton carry pilots—two of them—in addition to a payload large enough for communications satellites, which will be Roton’s bread and butter. Hudson believes that people will have a big role in space, even in such mundane tasks as maintaining satellites, and that an ability to carry people is a key to Roton’s long-term usefulness.

A few months into the project, however, it became clear that helicopter blades wouldn’t be able to generate enough thrust to carry both a two-person crew and a three-ton communications satellite. “As soon as we started sizing the thing up, we were concerned,” Hudson says. “It wasn’t that the technology wasn’t there, but the development risk was growing.” He and McKinney decided to scrap the idea of lifting off with blades. Instead, the Roton would go up under rocket power with its blades folded down to the sides. Upon reentering the atmosphere, the blades would unfold and whirl around, creating drag and slowing the ship down. The little rockets on the blades’ tips would fire just as the ship neared the ground, slowing its descent for a gentle landing. The scheme has the advantage of allowing Hudson and McKinney to use conventional helicopter rotors.

There is nothing conventional about the Roton’s propulsion. It uses 72 small rocket engines arranged in a ring at the bottom of the rocket. That’s been done before on some Russian rockets, but what’s unusual in this case is that the engines will be mounted on a disk that spins at 720 revolutions per minute. Kerosene and liquid oxygen travel from the tanks above, down through a pipe at the center. When the fuel reaches the disk, centrifugal force flings it out to the perimeter and forces it into the engines’ combustion chambers. With this setup, the Roton doesn’t need fuel pumps, or turbopumps, which are famously unreliable.


Michael Kelly was building missiles at TRW in the late 1980s when he and a couple of colleagues started brainstorming ways to build a cheap, reliable commercial launcher. The project never went anywhere at TRW, but Kelly couldn’t get it out of his head. A rocket, he noticed, is most likely to blow up right after ignition, during the first 60,000 feet of launch. Why not skip that part altogether? Why not take the rocket up above 60,000 feet and then ignite it?

In 1993, Kelly quit his job, formed Kelly Space and Technology in San Bernardino, California, and promptly designed a three-stage reusable launch vehicle that is unique in the annals of spaceflight. The first stage is a secondhand Boeing 747. The airplane uses a $50,000 piece of rope, made of high-tech composites, to tow the second stage up to an altitude of 20,000 feet. The second stage, called the Astroliner, is sort of another airplane, an old Lockheed L1011 that has been gutted and retrofitted with modified wings, not to mention a Russian rocket engine. It soars clear out of the atmosphere to an altitude of more than 75 miles. At this point, pilots level off the rocket plane and open a door on its nose. Out pops the third stage: a small rocket with just enough oomph to nudge the payload into stable orbit.

Meanwhile, the pilots veer off and take the Astroliner back down to a runway, just like the space shuttle. The Astroliner, though, will have twice the wing area of the space shuttle and will reenter the atmosphere at a much shallower angle, which means it won’t heat up as much and therefore won’t need to be as rugged. “We’re going half the speed and obtaining half the altitude of the space shuttle with about a tenth of the weight,” Kelly says, “so it’s a more benign flight profile than that of the space shuttle.”

So far Kelly says he has raised “tens of millions” for the project, but says he needs another $300 million. If he gets it, the Astroliner could be ready for its first commercial launch in 2002. Motorola has already promised Kelly $89 million to launch 20 satellites for it’s Iridium telephone network.

Kelly says the idea of a tow launch never hit him as any kind of eureka moment. “As a breed,” he says, “engineers are too conservative to have such moments until a concept has been tested, over and over. But I guarantee that once we’re up and running, there’s going to be a lot of engineers who’ll be slapping themselves on the forehead and saying, ‘Jeez, why didn’t I think of that?’ ”

 —Mark Wheeler

Hudson says that he and his team have bent over backward to keep things simple and reduce risk. Whenever possible they have used off-the-shelf parts and techniques that at the very least have been used experimentally before. On the other hand, how many conventional helicopter rotors have had to slow down a 400,000 pound vehicle traveling at 25 times the speed of sound? And the spinning engine has only been tried on a very small scale by Soviet and American researchers. Still, Hudson says, “there’s nothing we’re doing that’s obviously a showstopper. Okay, we’re doing some sporty things in the engines. This is the first sort of big, rotating engine. But if that doesn’t work, we could go out and buy turbopump engines and still make the vehicle work. It would be a year or year and a half delay, and it would be traumatic, but it would work. The technical challenges can be overcome.”

So far Hudson has raised more than $30 million, and he’s spent just about all of it. “When investors come into this company,” he says, “I tell them somewhat in jest that we’re going to waste half the money they give us. Investors just freak when you say this. When you really get down to the type of decisions we have to make, you’re really doing well if you only waste half of it. Because you make mistakes.” Although his rocket is coming in under budget, he needs another $120 million just to get him through to the flight test of an experimental version of the Roton. If he gets it, he hopes to fly a prototype some time in 1999. If all goes well, a big if, he expects to achieve orbit some time in 2000. How much will it all cost in the end? “The answer is, we won’t know till we’re done,” he says. “I think it’s important to understand how revolutionary what we are doing is.”


Rocket fuel oxidizers like liquid oxygen are heavy. Why not travel light and fill up the tank as you go along? asks rocket engineer Mitchell Burnside Clapp. The former test pilot wants to launch satellites and other payloads with a rocket that takes on its liquid oxygen three to six miles up.

In 1993, Burnside Clapp was working on experimental rockets for the Air Force when he sketched his idea on an officers’ club cocktail napkin. A specially designed airplane could use conventional jet engines to take off and rendezvous with a tanker plane at between 15,000 and 30,000 feet. Once it filled its tank with liquid oxygen, it could light up the rocket engines and zoom to a height beyond the atmosphere of up to 120 miles. That would be high enough to release a satellite with an expendable upper-stage rocket. Refueling in air would reduce the weight of the plane at takeoff by up to 60 percent, cut costs, and leave plenty of room for big payloads. The Air Force studied the idea and then dropped it. Burnside Clapp went private, founding Pioneer Rocketplane in 1995.

Since then, he and a handful of Air Force veterans have been holed up at Vandenburg Air Force Base on the California coast, hammering away at technical problems. They could build a prototype and test it in a mere 36 months, he says, if investors would give them a few hundred million dollars. “Just three or four of these airplanes would provide as much launch capacity as exists in the world right now,” Burnside Clapp says.—Jeffrey Winters

George Mueller makes no claim to being revolutionary. Like just about everybody in the rocket business, his motivation is at least partly idealistic—to beat the rocket equation. But he is a pragmatist. In running NASA’s manned space program for most of the Apollo project, he amply demonstrated an ability to make tough decisions. He had taken charge of NASA’s manned space program in 1963 at the age of 45, when engineers were just starting to build the Saturn V rocket. They wanted to test each one of its five stages separately, which would have meant building and flying five experimental rockets before even starting to put together the final Saturn. Although it was standard engineering practice, Mueller saw right away that such a schedule would never do. Instead, barely a month on the job, he announced his decision to go with “all up” testing—build all five stages, stack them one on top of the other, fill them with fuel, and fly the thing all at once. It was not popular, but it happened to work. As much as anybody else, Mueller was responsible for meeting John F. Kennedy’s end-of-the-decade deadline for putting astronauts on the moon.

Even as Mueller watched the Apollo program leap to success, he knew it was winding down, and he knew what he wanted to do next. He threw his considerable reputation behind building a cheap, reusable rocket. His idea was to spend money like crazy in the design stage, use all the latest materials, build right to the edge of the envelope, and wind up with a vehicle that was dirt cheap to operate. NASA officials embraced the idea but couldn’t summon enough seed money from Congress. So they built the space shuttle instead—cheap to design, expensive to fly, partially reusable. “It was a disappointment,” Mueller says. By the time the shuttle flew, he had left for private industry.

In 1993 the founders of Kistler Aerospace sought Mueller’s advice about their new venture to build a reusable rocket. They found him tending the jojoba trees on his Arizona ranch full-time, but he was restless. “My wife tells me I flunked retirement,” he quips. At the age of 75, he surprised everybody by offering to head the project. “I told them I would sit on the board of directors, but only as the chief executive,” he says.

The Kistler venture is a product of the space race turning commercial: the company was formed with the idea of exploiting the burgeoning satellite-launch market. The end of the cold war helped, too. Two hot Russian rocket engines had recently become available, the NK-33 and NK-43. The Soviets had developed these engines during the moon race, and they are the best kerosene-burners ever made. “The Russians kept building them and improving them over a period of 15 years,” says Mueller. “They built maybe 500 of them, and they probably destroyed 200 of them in testing, just to see how they could improve the design.” As a result, the engines produce 20 percent more thrust per gallon of fuel than do the Saturn V engines developed for Apollo. Because they burn kerosene, they are far safer and cheaper to operate than the hydrogen engines used on the Saturn V. There are 100 of these engines left. Kistler has 46 of them under wraps in a Sacramento warehouse.

Mueller turned to the rocket design. He scrapped the K0 because he worried about the design and problems with the central control system, and opted instead for a more conventional two-stage design. He was willing to sacrifice the elegance of a single stage vehicle, but he was loath to give up on reusability, which he saw as a key to keeping down costs.

When the new design, the K1, is completed, it will look every bit as homely as the Roton. It will consist of a cylindrical first stage, with a slightly thinner, blunt-nosed second stage on top—119 feet in all. Upon launch, the three NK-33 engines in the first stage will power the rocket for the first 25 miles or so. Then the rocket will separate, turn itself around, and the engines will fire once more to bring it back into the atmosphere. At about 40,000 feet, six parachutes will deploy to slow the rocket. Just before it reaches the ground, air bags similar to the ones used in automobiles will open to cushion the impact. Meanwhile, the second stage’s single NK-43 will take the payload into orbit, turn around and land, too, using parachutes and air bags.

 If you had to put your money on either Kistler or Rotary Rocket, Kistler would seem to have the odds in its favor. For one thing, the K1 is further along in development than the Roton. At the end of 1998, the shell for the first rocket was almost finished, two of three fuel tanks were complete, and the software and guidance electronics were 90 percent finished. Kistler got an earlier start, and it has farmed out much of its work to subcontractors. Northrup-Grumman is building the airframe, Aerojet is modifying the engines, Irvin is making the parachutes. Of course, Kistler has spent far more money than Rotary Rocket—$400 million—and it will need a lot more, $400 million, Mueller says, before it delivers its first payload to orbit. If the money is raised, K1 could be launched from Australia late this year.

Kistler also stands a better chance of raising the money it needs than Rotary, not least because of Mueller’s reputation. He has put together a roster that includes senior managers from just about every big NASA program. Like it or not, investors are more likely to put their money on engineers with impressive résumés than on a loner with big ideas.

If Kistler does get there first, however, it will be due in no small part to Mueller’s conservative approach. Although the K1’s two-stage design would give him more leeway to squeeze a few pilots aboard than Hudson’s single-stage design would, Mueller does not entertain any notions about putting people in space. And even though Mueller has the world’s best rocket engines, he judges them to be inadequate for anything but a two-stage rocket. “Remember, a reusable rocket has got to be heavier than an expendable one, because you need some extra fuel to get back to Earth safely,” he says. “The engines do not exist that can power a single-stage reusable rocket.”

Hudson disagrees. “I do not believe that in this era of lightweight thermal protection, composite materials, lightweight avionics, et cetera, we could not find several vehicle designs that are reusable. I find it inconceivable.”

Hudson and Mueller do agree on one thing: that the biggest obstacle they each face is money. “Technical challenges can be overcome,” says Burt Rutan on a Rotary Rocket promotional video, “but engineers don’t even get to try unless they have the resources.” And resources could get scarcer: three other firms have joined the race to build reusable rockets.

1 free article left
Want More? Get unlimited access for as low as $1.99/month

Already a subscriber?

Register or Log In

1 free articleSubscribe
Discover Magazine Logo
Want more?

Keep reading for as low as $1.99!


Already a subscriber?

Register or Log In

More From Discover
Recommendations From Our Store
Shop Now
Stay Curious
Our List

Sign up for our weekly science updates.

To The Magazine

Save up to 40% off the cover price when you subscribe to Discover magazine.

Copyright © 2024 Kalmbach Media Co.