海角大神

Every morning, Elon Musk steels himself to once again do battle with gravity. A multimillionaire who made his fortune as cofounder of PayPal, Mr. Musk has spent six years and $100 million of his own money designing rockets for his company, Space Exploration Technologies Corp. (SpaceX). In August, he watched helplessly as a design flaw allowed Newtonian forces to triumph over his Falcon 1 鈥� the third failure in as many launches.

If Musk鈥檚 wall-mounted photo of Muham颅mad Ali鈥檚 first-round, first-minute knockout of Sonny Liston signifies anything, it鈥檚 the South African-born entrepreneur鈥檚 determination to do things quickly and efficiently. 鈥淲e did get to space on flights two and three,鈥� says Musk, 鈥渂ut flight three really made me quite sad.鈥� It took a fourth launch on Sept. 28, preceded by a family visit to Disneyland鈥檚 Space Mountain to calm Musk鈥檚 nerves, for Falcon 1 to become the first privately developed, liquid-fuel rocket to orbit Earth.

Having passed that milestone 鈥� or, more accurately, 434 vertical milestones 鈥� SpaceX is on a trajectory to revolutionize space transportation. Musk wants to make it more affordable through much cheaper launches. His larger ambition is to transport astronauts in Space X鈥檚 rocket capsule, effectively providing the National Aeronautics and Space Administration with an alternative to the space shuttle, due to be mothballed in 2010.

But his ultimate aim is Mars. 鈥淸Musk] really believes in the future of space, believes that humanity needs to be a space-faring civilization in order to survive long term,鈥� says Bruce Pittman, director of flight projects at the NASA Ames Commercial Space Team/Alliance for Commercial Enterprise and Education in Space.

Despite such grand aspirations, the company鈥檚 offices, housed in an opaque building near several airstrips, aren鈥檛 ostentatious, and the only concession to flashiness is a sports shoe-like swoosh in the Space X logo. Sitting inside a cubicle where the few personal touches include toy robots and a model of Charles Lindbergh鈥檚 鈥淪pirit of St. Louis,鈥� Musk says he didn鈥檛 set out to start a rocket company. After selling PayPal to eBay in a $1.5 billion stock deal in 2002, he first embarked on an enterprise to spur public interest in Mars.

鈥淚 had a feasibility study done on a little project called Mars Oasis, which was to put a small greenhouse with seeds in dehydrated nutrient gel [on the planet鈥檚 surface],鈥� says the soft-spoken Musk, who concentrates on interview questions the way Gary Kasparov studies a chessboard.

Musk鈥檚 grand vision of green plants against a red backdrop was dashed when he priced out the rocket launch. Too expensive. Believing he could do it more cheaply, he founded SpaceX. Few in Washington took him seriously. So he famously parked a model of a rocket prototype in front of the Federal Aviation Administration building.

In person, Musk isn鈥檛 brash or domineering 鈥� in that respect, he is neither a modern-day Howard Hughes nor a real-life Tony Stark, even though he鈥檚 renowned for his steely resolve. The one-time physics student assumed the role of designer when he was initially unable to hire away top rocket scientists.

鈥淩ockets fail all the time, and hardly ever does a rocket work the first time you鈥檝e built one,鈥� says David Livingstone, host of 鈥淭he Space Show,鈥� a weekly Internet broadcast. 鈥淔or him to have a flawless launch on the fourth try is miraculous.鈥�
Chris Thompson, SpaceX鈥檚 vice president of structures, concurs. 鈥淸Musk] has picked up on everything related to propulsion, structures, avionics over the last six years. He鈥檚 a bright guy.鈥�

Even so, there was internal skepticism when Musk proposed that the company鈥檚 flagship product, a 180-foot rocket, use nine engines, not one.

鈥淎h, the old configuration debate,鈥� says Mr. Thompson with mock grandiosity. 鈥淲hen Elon had the revelation or epiphany, or whatever you want to call it, where he said, 鈥楬ey, we鈥檙e going to put nine engines on the back of this rocket,鈥� there were a lot of comments going back and forth 鈥� most of them pretty catty.鈥�

Musk鈥檚 idea is that the design will allow the rocket to fly even if it loses an engine early in flight. SpaceX recently fired the Falcon 9鈥檚 engines for 35 seconds 鈥� enough to incinerate any lingering doubts. The rocket is vying for a NASA contract to deliver supplies to the International Space Station. If SpaceX wins the bid 鈥� and it鈥檒l have to beat rival Orbital Sciences Corp. to do so 鈥� NASA has the option of commissioning manned flights. SpaceX isn鈥檛 waiting for NASA鈥檚 green light. The Falcon 9鈥檚 free-flying and reusable Dragon cargo capsule is multiuse: It can be reconfigured to accommodate up to seven astronauts.

Shaped like a cork with a cyclopslike porthole set between its conical nose and its 18 thrusters, a full-size model Dragon sits near the metallic ribs of the real thing on SpaceX鈥檚 factory floor. The factory also manufactures the Falcon 9, its various stages lying about like pieces of an oil pipeline. (Here, too, you鈥檒l find geek touches such as a life-size model of a 鈥�Battlestar Galactica鈥� robot.)

Unlike other aerospace companies, which rely heavily on subcontractors, SpaceX builds just about everything. It鈥檚 one way for the 500-employee organization 鈥� whose nonhierarchical culture is more Silicon Valley start-up than aerospace monolith 鈥� to keep costs low. Where possible, automation and software have replaced personnel, even on the launch pad in the Marshall Islands.

鈥淸SpaceX鈥檚 competitors] are so tied to the government business and the government culture of doing things,鈥� says Lon Raines, editor in chief of Space News, a trade publication. 鈥淭he streamlined organization is where [SpaceX] can make a difference. They do make decisions faster, and they are trying to do this with far fewer people than you see on some of the traditional launch programs. That will be a big benefit but 鈥� who knows? 鈥� it could also be an Achilles heel.鈥�

Musk says SpaceX will have failed if it doesn鈥檛 produce cheaper launches. But Orbital Sciences, too, promised low-priced launches in the 1990s, and its Pegasus rocket costs about $30 million per launch. Andrew Beal, who invested $200 million in now-defunct Beal Aero颅space, wouldn鈥檛 laugh at Musk鈥檚 joke that the easiest way to become a millionaire in rocketry is to invest a billion dollars.

He wouldn鈥檛 be in it if he weren鈥檛 thinking big, and extending life beyond Earth is his goal. 鈥淚 don鈥檛 need to work 鈥� I can buy anything I want,鈥� says Musk, who is also chairman of Tesla Motors, the electric sports car manufacturer. 鈥淚鈥檓 just working because I think this is important.鈥�