Successful start-ups are rare in aircraft manufacturing. Most fail to reach certification, but even if they get that far these newcomers face steep hurdles acquiring the resources and expertise to move into volume production and profitability.
"It's turned out to be really hard," says Eclipse Aviation's chief executive Vern Raburn. The company secured type certification of its Eclipse 500 very light jet in September 2006, and production certification in April, but has struggled to ramp up to the high production rate planned. "We grossly underestimated the job," he says.
Eclipse is not alone. Sino Swearingen Aircraft (SSAC) certificated the SJ30 light jet in December 2005, but delivered only its second customer aircraft in September. Adam Aircraft received initial approval for its A500 piston twin in May 2005, but has not finished certification and has delivered only a few aircraft.
The issues all start-ups face are similar: raising the financing to transition from type certification to rate production and setting up from scratch the production and quality systems essential for aircraft manufacture. For Adam and SSAC the hurdle has been financing: for Eclipse it has been building the infrastructure of a major manufacturer.
"Where are we? Nowhere near where we wanted to be," says Raburn. The company continues to work through the vendor issues that dogged certification, but its biggest challenge has been putting in place a high-rate production system. "We are not out of the woods yet, but we are seeing nothing that says we can't build at volume," he says.
Achieving the high production rates planned for the Eclipse 500 VLJ is a tough challenge
The most successful recent start-up has been US light-aircraft manufacturer Cirrus Design. Certificating its all-composite SR20 in 1998, Cirrus delivered its 1,000th aircraft within four years and is coming up on its 4,000th. Contrast this with Columbia Aircraft, which like Cirrus grew out of the kitplane industry. Columbia also certificated its first piston single in 1998, but struggled to ramp up production, only delivering its 500th aircraft in 2006. It then hit a series of problems that led to the company filing for Chapter 11 bankruptcy protection in September.
The difference between the companies has been the resources available to ramp up production. Cirrus in 2001 sold a controlling stake to a Bahrain-backed investment firm for $100 million. A controlling interest in Columbia was sold in 2003 to government-backed Composite Technology Research Malaysia for $50 million, but by 2006 CTRM was looking to sell its stake. In September Cessna entered discussions to acquire the assets of the bankrupt manufacturer.
Similar stories can be found across the industry, and the founders of Adam and SSAC have both had to relinquish control of their companies to attract the additional investment required to ramp up production.
After investing more than $700 million in SSAC, the company's Taiwanese backers have agreed to sell a controlling stake to a joint venture between US investment firm ACQ Capital and SJ30 distributor Action Aviation of the UK. A priority for the new owners is to rejig a manufacturing line that is producing only a trickle of aircraft.
New Adam president Duncan Koerbel says the Colorado-based company has secured $200 million in equity and debt financing in the last nine months and is revamping its manufacturing system with the goal of producing 150-240 A500s and A700 VLJs a year and delivering 1,000 of the all-composite aircraft within seven to 10 years.
But this ambition is eclipsed by Eclipse, which has set it sights on building at least two aircraft a day. When it handed over the first customer Eclipse 500 in December 2006, the New Mexico-based company predicted it would deliver more than 500 in 2007: so far just over 50 have entered service.
Raburn blames two problems: vendor issues carried over from certification and the challenge of setting up a high-rate production system. Both are tied to the VLJ pioneer's avowed goal of changing not only how people travel, but how aircraft are built.
"We have had significant problems with several vendors," he says. "Some we induced ourselves. We had an immature engineering organisation with a not very disciplined approach to design changes. We clearly caused problems for vendors - we still are today, but we are trying to clean up our side."
But there are stark differences in how the vendors handled the problems, says Raburn, a former Microsoft executive. "I made some bad choices of vendors and I'm still paying the price. Some say I'm enraptured with technology and do not evaluate the risks enough. But it is compounded by vendors who do not stand by their promises."
Raburn admits many established vendors no-bid when Eclipse was looking for suppliers because they did not believe the company could produce the aircraft at its price point, originally under $1 million and predicated on unheard-of production rates. "They did not believe in our business plan, so we paid hundreds of millions of dollars in non-recurring expenses to take the risk out of our different business model."
Eclipse's search for innovative and low-cost suppliers has had mixed results. "There are places where aggressiveness with technology has worked, such as the electrical power system, which has never had a failure." He also cites Pratt & Whitney Canada: "They developed a new centreline engine and it has never had an inflight failure."
Raburn says 80% of failures in the fleet are in three areas: displays, autopilot and actuators. The orginal Avidyne displays will be replaced when the new Avio NG flightdeck is introduced in November, and Meggitt "is doing everything it can with the autopilot and we are seeing reliability improve", he says
The company's problems in ramping up production are "more deeply a culture issue", says Raburn. Eclipse deliberately hired experienced aviation-industry veterans to set up its manufacturing system, "but they came out of very established companies into a start-up - a baby that had lots to learn, like, who empties the trash? They were incapable of starting with a clean sheet."
Eclipse has "learned a hell of a lot" about high-rate production, he says. "Manufacturing at rate is like squeezing a balloon - something pops out somewhere." The company has learned tolerance build-up is more important than accuracy in high-rate manufacture. "A lot of the parts are machined. We thought that was all we needed for it to go together the same every time. It turns out that tolerancing and holding the part accurately is important."
After rebuilding its production staff, and bringing in automotive industry experience, Eclipse has moved to parallel production lines. "It's no longer single-piece flow," says Raburn. There are now three independent aft/forward-fuselage mate positions to avoid bottlenecks. "By later August we were starting to see the benefits."
Eclipse has proved that every position on the production line can run at a rate of at least one aircraft a day. "Have we got all of them to run at that rate at the same time? Not yet," Raburn says. "But our analysis shows we can get to one-and-a-half to two a day."
Inevitably, Eclipse's failure to deliver on its aggressive promises has drawn a backlash of criticism from Raburn's detractors, but he remains defiant. "It turned out to be really hard, but nothing has proven to be impossible. There is nothing we set out to do that we have not finally done," he says. "We have created a product that delivers on the promise to change the way we travel, and built a company with lasting value."
Sino Swearingen certified its SJ30 light jet in 2005, but has so far delivered only two