The Case: Eclipse Aviation was trying to build the first twin-turbofan private jet with a price tag under $1 million. But when reliability questions surfaced around the engines being designed for Eclipse, the company had to decide whether to stick with its original supplier or find a new one – and thus push off the plane’s debut for two years. Any delay would tax the patience of the prospective buyers on the Eclipse’s waiting list – and run the risk of exhausting the company’s cash reserves.
* * *
When the wheels of Eclipse 500 jet lifted off the runway of the Albuquerque International Sunport on the morning of August 26, 2002, Ken Harness thought that the last of his problems were about to vanish like a high-altitude vapor trail.
As the New Mexico company’s director of propulsion systems, Harness had worked feverishly to get the plane ready for this inaugural flight, focusing especially on a newly designed engine that weighed just 85 pounds, but was intended to churn out an impressive 770 pounds of thrust. But after a successful first flight that day, things unraveled. As the Eclipse 500 was taxiing out for a second take-off, one of its engines quit. That failure would lead to big decision about Eclipse’s relationship with its engine supplier, Williams International.
Eclipse was founded in 1998 by Vern Raburn, a pilot and aviation enthusiast, whose personal fleet includes a 1940s era Lockheed Constellation that was actually used in the Berlin Airlift. Raburn had soaked up business experience as an early employee of Microsoft (he helped start the consumer products division), general manager of Lotus Development, and CEO of Symantec. But he’d never been involved with the world of airplane manufacturing, which is strewn with the wreckage of companies that have tried and failed to get new aircraft designs approved by the Federal Aviation Administration before running out of cash.
Raburn thought he could apply his technology savvy to airplane manufacturing, creating a streamlined, Dell-like supply chain and embracing new production techniques such as friction-stir welding, which he believed would be faster and cheaper than using rivets. (Friction-stir welding uses a rotating tool to heat up and plasticize the two sides of a joint.)
Raburn’s objective was to build a private jet that could be sold for less than $1 million – one-third the cost of the cheapest jet on the market at the time. He thought the plane would appeal to private pilots who’d always longed to upgrade from a prop plane to a jet, and that its low operating costs (primarily thanks to its fuel-efficient engines, designed to sip about 100 pounds per hour) would enable companies to establish on-demand networks of air taxis.
These networks would use Eclipse jets to fly to and from secondary airports near where people live and work, rather than forcing them to transfer through bigger hubs in metropolitan areas. Raburn often compared the plane’s operating costs to those of driving a New York City cab – it would be cheaper per mile, he claimed.
Raburn encountered little trouble raising money, by assembling a roster of well-known backers that includes his old Microsoft bosses, Bill Gates and Paul Allen, as well as retired CEOs of Ford and DaimlerChrysler. To supply the plane’s engines, Raburn turned to Williams International, a Michigan company that produced the first generation of cruise missile engines, as well as turbofans for business jets made by Cessna. (Raburn gave Williams an equity stake in the company to reward it for committing to develop a new engine, dubbed the EJ22.)
When Eclipse started taking pre-orders for the plane, 175 eager buyers showed up in person to plunk down deposits. Some aviation industry observers were skeptical, however, that there would be something to buy – at least at the Eclipse’s stated $837,500 price tag.
Before the plane’s first flight, problems with the engines – which were smaller than a beer keg (and about half the size of a typical turbofan) – began to appear. Harness says he was dispatched to Reno, Nevada for six weeks to help Williams’ engineers troubleshoot the connection between the plane’s fuel supply system and the engines. “Then we started having other, unrelated issues,” he says, such as cracks in the fan blades and surging. (When the fan blades stall, airflow can suddenly change direction, sending a flash of flame out of the inlet, known as a surge.) “Anytime you had a surge, it would damage the engine,” Harness says. That meant sending it back to Williams’ headquarters for repairs.
By August 2002, Harness and the engineering team convinced themselves that the Eclipse’s engines were reliable enough for a safe test flight – though the units weren’t yet delivering 770 pounds of thrust, so the test plane was kept extremely light. Following the first hour-long flight, however, numerous failures occurred while running the engines on the ground, according to Raburn, and so the Eclipse stayed in its hangar. “We never would have two engines that worked at the same time,” says Raburn.
Raburn says the president of Williams International, Gregg Williams, traveled to Albuquerque, assuring Eclipse that the problems would be fixed by November 1. But when that date passed with only a fraction of the fixes complete, according to Raburn, he felt it was time to hire an outside consultant to examine whether Williams would be able to deliver a workable engine.
“They told us that this engine was a minimum of 18, and perhaps as much as 36 months behind schedule,” Raburn says. “And they didn’t believe that this engine would be reliable enough.”
Raburn went to Williams chairman Sam Williams with an ultimatum: complete the development of the EJ22 engine on your dime, at the per-engine price tag your company originally promised. “Sam said he couldn’t take the risk,” Raburn recalls. (Williams’s two top executives, Sam and Gregg Williams, declined to comment for this article.)
Raburn and his team started to examine their alternatives. If they abandoned the EJ22 and sought out another engine maker, Raburn reckoned it would set back the delivery of the first FAA-certified airplane by about two years. Also, he knew that a replacement engine would entail substantial redesign work to the aircraft, and it’d probably send the price over the $1 million mark. And there were plenty of unknowns: how many customers would ask for their deposits back? Would he be able to raise the $60 million he estimated he’d need to keep the company going?
Harness recalled saying, in a presentation to Eclipse’s board, that even if Williams could have somehow delivered an engine, “we would’ve had so many service difficulties with it, it would’ve bankrupted us.”
But Raburn wasn’t convinced the company could survive if he dropped the Williams engine. “What we decided was that we wanted to take our chances,” he says. “It’s like jumping out of an airplane without a parachute.” Just before Thanksgiving 2002, Eclipse put out a press release announcing that it had severed its relationship with Williams.
“We were now selling a million-dollar glider,” Raburn says, “and those don’t sell very well.” But the company had already drawn up a list of alternative suppliers, and Eclipse eventually struck a deal with Pratt & Whitney Canada to buy a slightly heavier, slightly thirstier engine, called the PW610 F. But one benefit of the P&W engine was that it would increase the aircraft’s top speed by 20 knots. Raburn promised Eclipse’s customers that the plane would be back in the air with its new turbofans by the end of 2004.
Only about fifty customers asked for their money back after the switch was announced. At $1.5 million, the plane’s current list price is still about $500,000 cheaper than its closest rival, the A700 from Colorado-based Adam Aircraft. (Like the Eclipse, that plane has still not been FAA certified.) The increase in price was due to the higher cost of P&W’s engines and a jump in the cost of aluminum, from which most of the plane is constructed. On December 31, 2004, Raburn delivered on his promise: the Eclipse flew twice that day with its new engines.
Still, it wasn’t until this past summer that Harness, who had flown Blackhawk helicopters as an Army test pilot, got to go up in the plane for the first time. “I want to have first-hand knowledge of any aerodynamic problems we might be having,” Harness says, adding with a chuckle: “Besides, flight test belongs to me, so I get to do whatever the hell I want.” In June, he flew with one of Eclipse’s test pilots, running through a “test card” that included some touch-and-go landings, basic performance handling, and midflight engine flame-outs. These flame-outs, thankfully, were simulated.
- Total money raised: $400 million
- Total orders to date: 2,300 planes
- Aircraft’s top speed: 430 mph
- Biggest customer: DayJet (a startup air taxi service), which has ordered 239 planes
- Projected manufacturing time: 4.5 days for construction, 2 for flight test, 2 for painting
Scott Kirsner is a freelance writer based in San Francisco.
This new data poisoning tool lets artists fight back against generative AI
The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models.
Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist
An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.
Data analytics reveal real business value
Sophisticated analytics tools mine insights from data, optimizing operational processes across the enterprise.
Driving companywide efficiencies with AI
Advanced AI and ML capabilities revolutionize how administrative and operations tasks are done.
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.