Some say that Beech and Raytheon's turboprop failed because it tried too much, too soon.
- By Mark Huber
- Air & Space magazine, September 2004
NASM (SI Neg. #9A02243)
A FEW YEARS AGO, STARSHIP PILOT BOB BASS was flying at 31,000 feet over Kansas when he received a radio call from a U.S. Air Force jet that wanted to pull alongside. He looked out the window and saw a $2 billion B-2 bomber join up on his wing. The B-2 pilot was shooting photographs of Bass’ airplane.
Last summer Starship owner Bob Scherer and I landed in Phoenix, and as we taxied up to the terminal, a pair of excited line service workers pulled cameras out of their pockets and began snapping away. “This happens all the time,” Scherer said.
Twenty-five years after it was first conceived, the Beech model 2000 Starship is still a head-turner, its sleek, futuristic design looking as if it’s doing 400 knots standing still. Last year, parent company Raytheon Aircraft announced that it had purchased most of the fleet, wanted to buy the rest, and would destroy all it could find, calling the cost of continued product support “prohibitive.”
Most of the fleet sits on Evergreen Air Center’s heavy-maintenance ramp in Marana, Arizona, where the airplanes are being stripped, sawed up, and incinerated. Raytheon has donated a few of them to museums, but as of last May, only four Starships remain in the hands of private owners. Scherer, a southern California real estate developer, is one of them. “From my cold, dead hands will Raytheon get this airplane,” he says, echoing the National Rifle Association’s battle cry. “I would have nightmares if they sawed this plane up.”
Raytheon sees no alternative. “Many parts on that airplane are unique to the Starship and are no longer being made by suppliers,” says Raytheon spokesman Tim Travis. “From a business standpoint it was a losing proposition and it always would be.”
But the Starship was flying in turbulence from the outset. For an industry in which designs have traditionally evolved in gradual increments, the new turboprop was the loudest of thunderclaps and completely antithetical—bold, daring, and radically different. From the shape of its airframe, to the placement of its engines, to its carbon-fiber structure, to its digital avionics, almost nothing about this airplane was normal.
The Starship was one of the largest and most advanced business aircraft programs ever attempted, and it bombed. When Raytheon shut down the production line in 1995, only 53 aircraft had been produced. “I had the joyous duty of shutting down the ill-fated Starship line,” former Beechcraft president Roy Norris said in a 2002 interview published in General Aviation News. Referring to the television series “Star Trek,” Norris said, “I made myself a promise that there would be no more airplanes that look like Klingon battle cruisers.”
Bob Scherer flies serial number NC-51. (Beech serial numbers begin with two seemingly random letters, but some think NC signified “New Concept”; others think it was a play on the United Federation of Planets’ starship registration from “Star Trek.”) He thinks the Starship’s revolutionary design and price tag hurt early sales. “It wasn’t accepted at first because it was so radical,” he says. “Plus it was five million bucks, which is entry-level jet zone.” Scherer acquired his Starship from the Tyco Corporation in 1998 after seeing it listed in Trade-A-Plane, a thrice-monthly newspaper filled with ads for aircraft. A Starship on the market today would fetch about $2 million.
The Starship would have pushed the technology envelope for any aerospace company, but Beech was especially conservative. Seeking to replace its highly successful but 15-year-old King Air line, Beech decided in 1979 to develop a turboprop that would help preserve its lead in the business turboprop market, in which it held a 50 percent share. The goal was an airplane that could cruise at 400 mph, carry 10 passengers, and weigh less than 12,500 pounds. (By comparison, a King Air 350 cruises at 360 mph, seats two pilots and eight passengers, and has a basic empty weight of 9,300 pounds.)
According to a personal account of the airplane’s development by former Beech chief executive Max Bleck (Starship History; www.aviatorservices.com/starship_history_1.htm), the company studied designs that ranged from the traditional to the radical, seeking an airplane that delivered jet-like performance at cheaper turboprop operating costs. Soon one design emerged, with aft-mounted pusher engines and a wing with vertical surfaces at each wingtip. It also had a large cabin, one of the King Air’s most appealing features, but its size would make it heavy. “The decision was made early on to build using composites for [their] favorable strength-to-weight ratio,” Bleck wrote.
By the late 1970s, other companies were working on similar or competing concepts, spurred in part by spiking oil prices. In Reno, Bill Lear’s widow, Moya, was trying to finish his Learfan. It had a composite fuselage with twin turboshaft engines mounted in the tail and driving a four-blade pusher propeller. Italian airframe maker Piaggio teamed with Learjet (then owned by the Gates Rubber Company) on a design that became the P.180 Avanti. The Learfan never made it to market, and Gates bailed out of the Piaggio program.
The P.180 is in production today and has finally started selling, thanks largely to Italian government subsidies and the deep pockets and patience of Piaggio’s owners, the Ferrari family (yes, that Ferrari). During the 1970s, Cessna, Mitsubishi, Piper, Swearingen, and Rockwell fielded successful conventional twin-engine turboprops.
Beech saw its domination of the business turboprop market slipping away, and on January 30, 1980, its engineering department issued a closely held report entitled “Advance Design Comparison Studies of Several Unconventional and Conventional Corporate Turboprops and Fanjets.” Months later, Raytheon, a Massachusetts-based electronics firm, acquired Beech Aircraft, and it was not until August 1982 that the program resumed. Advanced projects engineer Bill Brown remembers the day well. “Chet [vice president of engineering Chester Rembleski] gets us all together, turns off the lights, and shows us this movie. In the movie this guy is climbing up a mountain and you hear him breathing and see him slipping on the rocks and clawing with his hands. He is really struggling, but not saying anything. Then he gets to the top of the mountain, puts on his skis, and skis down the side of the mountain. He goes like hell. Then the lights come up. We’re all sitting there puzzled, looking at each other, and Chet says, ‘Gentlemen, we are going to design a new airplane and we are going to do it in two years.’ ” The project was initially labeled 300A, perhaps to suggest it was a new King Air, and later Starship 1, and then model 2000. It was shrouded in secrecy, and most Beech employees thought Brown’s team was working on the next King Air.
By the late 1960s, composites had begun to find their way into aviation, initially in sailplanes. Beech’s own research showed that carbon fiber was at least three times stronger than aircraft aluminum. In the California desert’s Antelope Valley, next to Edwards Air Force Base, a young engineer named Burt Rutan was designing aircraft with a small forward wing, or canard, that made his aircraft more aerodynamically efficient and virtually spin-proof. (When some aircraft fail to maintain sufficient airspeed to produce lift, they have a tendency to stall and enter a spin.) His first design, the VariViggen, used wood. Then he started looking at fiberglass.
Bill Brown was a homebuilt-aircraft enthusiast who had already worked with composites—in his garage. He began making sketches of what a Rutan design would look like if it were morphed into a business aircraft. Beech then approached Rutan about joining the design effort. The company’s designers explored numerous configurations, including pushers, twins, and one like the Learfan, before finally selecting one Rutan had drawn on a napkin. According to both Rutan and Brown, it was not until Raytheon entered the picture that the project, originally conceived by Beech in 1979, really took off.
The aircraft’s systems would be as radical as its shape. The Starship would break all the rules: It was the first business aircraft with an all-glass digital cockpit—a group of 14 cathode-ray tubes in the instrument panel. Rockwell-Collins was tasked with developing this system, called EFIS, for electronic flight information system. The Starship was the first civilian aircraft with a pressurized carbon-fiber fuselage to be certificated by the Federal Aviation Administration, the first modern U.S.-built production civil aircraft with a forward wing or canard, the first without a tail, and one of the first passenger turboprops with pusher propellers (Piaggio was the other).
The program brought together some of the finest minds in aviation. They included D. Brainerd Holmes, a driving force behind NASA’s Apollo program and now president of Raytheon. Linden Blue and composites guru Ric Abbott came aboard from Learfan; Blue is credited with selecting Rutan’s design. Others involved included Rutan and a handful of brilliant aerodynamicists, including John Roncz, an airfoil expert, David Bernstorf, who led the Beech aerodynamics and loads group, Roy LoPresti, a speed merchant who had wrung the drag out of half a dozen airplane types, and Brown. At its apex, 1,000 employees were assigned to the Starship, 450 of them engineers.
Beech’s parent company, Raytheon, would spend a million man-hours and $350 million (early 1980s dollars) bringing the airplane to market and hundreds of millions more marketing and supporting it. Some estimate that Raytheon sank as much as $1 billion into the program. (Raytheon will not provide an exact number and may not even know what it is.) “The cost was a very nebulous figure to come up with,” says former Beech and Raytheon president Max Bleck. “None of our financial people tried to put a pencil to it.” But almost everyone knew that the numbers didn’t add up. An internal economic analysis by Beech in 1979 concluded that if it sold 400 Starships a year at an after-tax profit of $250,000 per unit, for a total profit of $100 million a year for 30 years, it would generate an internal rate of return of 16 percent at a time when the prime rate was almost 20 percent.
When he was president and chief executive of Beech, from 1987 to 1991, Bleck, a former Piper Aircraft president, ran the numbers and tried to bury the Starship program. “I tried to kill the airplane twice,” Bleck says, first in 1987, just months before the aircraft received type certification, and again in 1991. On both occasions, he was overruled by his bosses at Raytheon.
In 1982, the Rutan Aircraft Factory was awarded a contract to build a proof-of-concept Starship for aerodynamic testing. Working around the clock, Rutan’s crew built the airplane, which was slightly smaller than the production version, in less than a year. Beech exhibited the POC aircraft at the National Business Aircraft Association’s 1983 convention in Dallas and announced the aircraft’s target price: $2,742,500. The response was a collective gasp. This was potentially the biggest new thing since the Learjet.
Skeptics doubted that the aircraft would make its aggressive two-year certification schedule, that it would come in under 12,500 pounds gross weight, and that it would win acceptance from a conservative market. They were right. The Starship would not gain FAA certification until 1988. Its empty weight would increase by 2,400 pounds and its gross weight would balloon to 14,900 pounds. Originally designed for a pair of 750-shaft-horsepower engines, the weight gain forced designers to adopt thirstier 1,200-horsepower engines. The diameter of the propellers would grow from 94 inches to 105 inches. It also lost two passenger seats. And by the early 1990s the price would inflate to $5.3 million. Despite the excitement and the aura Rutan brought to the project, the button-down world of business aviation was not ready for an airplane that had become a moving target.
For the skeptics, including those at Raytheon, the proof-of-concept aircraft became the focal point of criticism. The POC was unpressurized, made of fiberglass rather than carbon fiber, and had a higher thrust-to-weight ratio than the production aircraft. Critics deemed it too different to be proof of anything. Bleck called the POC “virtually worthless.” Rutan and others, including Bill Brown and Beech test pilot Tom Carr, disagree. Intended to fly only 100 hours, the POC would log more than 500 between 1983 and 1986 and provide important data that affected the final design. Rutan got the POC contract on August 25, 1982. Beech started designing tooling for the production Starship six days later. Had the POC flown before tools for production aircraft were built, its impact on the Starship could have been far greater. After the POC program ended, Raytheon had the airplane destroyed in full view of those who had built it. Rutan’s staff salvaged a few mementos, including the data plate.
As tensions grew between Rutan and Beech, they also increased between Beech and Raytheon. Beech’s genteel culture buckled under the often abrupt ways of its new parent. Since 1950, Beech had been run by Walter Beech’s much younger widow, Olive Ann. Mrs. Beech would put yellow “happy face” stickers on the office doors of meritorious executives. Company picnics were courtly, civilized affairs.
It wouldn’t be long before the Raytheon clamps were tightened. The old ways were gone, Mrs. Beech stepped aside, and the door to the president’s office began revolving. Occupants were either kicked upstairs to corporate headquarters in Massachusetts or shown the door. During the Starship’s development, from 1982 to 1989, Beech had three presidents and four engineering vice presidents, creating certification delays and performance compromises.
Initially, Brainerd Holmes brought in Linden Blue from Learfan as president, largely to run the Starship program. Blue, today co-chairman of General Atomics, maker of the Predator unmanned aircraft, would last two years. Brown remembers Blue as a “real go-getter,” but others described him as a micromanager who insisted on locating the Starship’s lavatory in the front of the cabin (the toilet was in a cabinet-mounted drawer that pulled out into the aisle, a design that would later be changed), nixed an external baggage compartment door, and thought that spraying water repellent on the windshield would substitute for wipers. Brown counters that the wipers disrupted airflow, and the repellent worked fine except during taxi operations.
The Starship also stirred up cultural wars within Beech itself. From the beginning, the project was an enormous drain on company resources and capabilities, fostering resentment among those involved with other Beech programs. The Starship team was housed in its own brand-new, 150,000-square-foot building and seemed to get the best of everything.
Explains Max Bleck, “People working on the rest of the product line felt the Starship was using most of the resources.” The Starship was burning $500,000 a day. “It was insane,” says Tom Carr. “Money was going out the door at an incredible rate.” When big layoffs hit the rest of Beech in 1984, the “have nots” turned their anger on the Starship. “That caused a lot of problems for a lot of years,” says Carr.
The Starship also created a fair amount of anguish on the manufacturing side of the company, which was wise to the ways of metal airplanes but uncertain about composites. Brown recalls that the first three airframes were fabricated using a mechanized process in which carbon fiber material was automatically wound around a form, or mandrel, by a winding machine—the most advanced technology at the time. Too advanced, Brown says, though it could cut time and cost. As the material traveled through the winding machine, it picked up thick liquid resin from a reservoir and squeezed out the excess using rollers. The Utah company doing the winding was used to doing prototypes and one-off projects, not volume production. “Ed Hooper [chief of airframe design] stayed up for 32 hours straight watching that first winding operation,” Brown recalls. “Then he crashed in his motel room.” Hooper liked the winding process, but Brown says the technology “just wasn’t there”; the tooling kept failing. At airframe NC-4, Beech switched to hand “lay-up” of carbon fiber material that was already impregnated with resin, or “pre-preg.” Workers cut sheets of this material using a template and laid them in a mold, the direction and angles of the overlapping carbon fibers matching the path of loads in the part and determining its strength. A light, strong honeycomb core was sandwiched between two layers of carbon sheets and compressed to eliminate voids. After spending time in the heat and pressure of an autoclave to cure, the part was done.
Tom Carr looks back at the Starship’s complex supply chain and marvels that the aircraft ever got produced. Rutan was building the POC in Mojave. Bell Helicopter was building the canard in Fort Worth. Brunswick, the bowling ball company, had the initial contract for the control surfaces, but never delivered a usable tool or part, Brown recalls. Pratt & Whitney was making the engines in Montreal. Hercules was manufacturing composites in Utah. The propellers were coming from McCauley in Ohio. Collins was developing the avionics in Iowa. TKS was working on anti-icing technology in the United Kingdom. Precision Components was fabricating the fuselage mockup in Detroit. The list goes on. “We didn’t have the composites technology or a lot of the other technologies either,” says Carr. “We ended up contracting parts of the airplane with people all over the world.” But too many parts ended up coming back to Beech’s plant in Wichita at a time when the technologies were new and the rulebook was being written.
To keep the ballet coordinated, a fleet of King Airs flew engineers around the country several times a week. But it was inevitable that the supply chain would collapse. Internal engineering memos show major assemblies and components sometimes being delivered months late. Computers of fiendish complexity operated systems such as environmental controls, cabin pressure, and automatic deicing, and engineers struggled with all of them, even naming one Hal after the robot gone bad in the film 2001: A Space Odyssey. They were replaced with simpler systems, but delays mounted up.
Amazingly enough, the first full-scale production prototype, NC-1, made its first flight a mere 28 months after the Starship was announced. On February 15, 1986, Carr and chief test pilot Bud Francis lifted off from the snow-lined runway at Beech Field. Below the pilot’s window was a heart-shaped decal, a valentine for Olive Ann Beech—one day late. The first flight was replicated for Beech employees and the media later, and two more aircraft would join the test program. Between the airplane’s conception and June 14, 1988, the day it received its FAA type certificate, Bill Brown personally signed 50,000 change orders on more than 2,000 engineering drawings. Major changes were being made to the aircraft through the end of May 1988. Brown saw nothing wrong with that. “Change iteration is the way you maximize an airplane,” he says. Management had a different opinion. Down $350 million and counting, it wanted the airplane on the market. Now.
“If it is certifiable, certify it,” was the edict from Bob Dickerson, Beech’s vice president of engineering, in 1988. In the rush, fixes were not done optimally, often with weight gain as a result, while other fixes were not done at all, such as remedying the Starship’s light pitch sensitivity and heavy, almost truck-like roll response. “The FAA did have some issues, and we ended up with a belt-and-suspender type of approach that was an expedient solution,” says Bernstorf.
The results were disastrous for the aircraft’s early reputation. While Beech had gotten most of the difficult new technology right—from the avionics to the composite structure to the variable-sweep canard—it had let a lot of mundane, old-technology things fall through the cracks. The first production Starship, NC-4, was delivered to a Florida-based beer distributor at the summer 1989 Paris Air Show. The air conditioning on the airplane failed repeatedly. Other common problems included door seal failures, defects in bleed air valves supplying pressurized air, and bad brakes. Starships were tarred as “hangar queens,” and one exasperated operator took a Beech executive to lunch and ordered a special centerpiece—a bowl of lemons. Beech eventually fixed these problems, and airplanes with later serial numbers had few if any problems, but in the death-by-whisper world of business aviation, the damage was done.
“What killed that airplane was the reliability issue,” says Tom Carr, who flew 30 of the 53 units produced and logged hundreds of hours training customers. “Once it got that reputation, it was hard to sell airplanes.” And Beech sold its own direct competition: the King Air series, airplanes that flew almost as fast, carried as many as two more passengers, and were known for almost bulletproof reliability. At the 1992 NBAA convention, Beech’s then-president, Jack Braley, told reporters that Starship production would end at serial number NC-53 if sales didn’t pick up. “That was the kiss of death right there,” says Carr. Production ended three years later, at serial number NC-53.
Carr thinks that if Beech had held the Starship off the market for a year to address weight, reliability, and handling issues, “they would still be building Starships today.” Although considered a commercial failure, the Starship project, for those who worked on it, remains the experience of a lifetime. Prior to the Starship, Ric Abbott had worked on high-profile European programs like the Concorde and the Tornado fighter. Even today, he calls the Starship “my favorite program. It was a great time.”
Those lucky enough to fly Starships are similarly enamored. Corporate pilot Wayne Roberts has logged 2,500 hours in Starships over nine years for several owners. He has flown over 60 types of aircraft, but he says nothing handles turbulence better. Since 1995 Bob Bass has logged 1,900 hours in Starships as a corporate pilot for Vertex Aerospace in Madison, Mississippi. “It’s a wonderful airplane to fly, very maneuverable and plenty of power,” he says.
On a broiling summer day, Bob Scherer and I taxi NC-51 up to Burt Rutan’s hangar in Mojave, unannounced. A crowd gathers to greet us, and one of them calls out, “You’re not giving it back [to Raytheon], are you?” Rutan is busy but greets us warmly. To our amazement, we discover that Rutan has never flown in a production Starship, so he, his test pilots, and Scherer pile in and take off. They climb to altitude and shut down an engine, pull full aft stick, and try to spin the airplane—it won’t. After they land, Rutan pulls out a felt-tip pen and autographs the inside of NC-51’s coat closet. As we leave, someone calls after Scherer, “We’ll fix it if you break it.”
Flying back to Los Angeles, Scherer says, “I’m a convert to this design. I couldn’t fly a metal airplane with a tail in the back. Uh-uh. It just seems wrong.”