Skunk Works: A Personal Memoir of My Years at Lockheed

2 ENGINES BY GE, BODY BY HOUDINI

Kelly Johnson was not impressed. He took one look at Dick Scherrer’s sketch of the Hopeless Diamond and charged into my office. Unfortunately, he caught me leaning over a work table studying a blueprint, and I never heard him coming. Kelly kicked me in the butt — hard too. Then he crumpled up the stealth proposal and threw it at my feet. “Ben Rich, you dumb shit,” he stormed, “have you lost your goddam mind? This crap will never get off the ground.”

Frankly, I had the feeling that there were a lot of old-timers around the Skunk Works who wanted badly to do what Kelly had just done. Instead they did it verbally and behind my back. These were some of our most senior aerodynamicists, thermodynamicists, propulsion specialists, stress and structures and weight engineers, who had been building airplanes from the time I was in college. They had at least twenty airplanes under their belts and were walking aviation encyclopedias and living parts catalogs. Over the years they had solved every conceivable problem in their specialty areas and damned well knew what worked and what didn’t. They were crusty and stiff-necked at times, but they were all dedicated, can-do guys who worked fourteen-hour days seven days a week for months on end to make a deadline. Self-assurance came from experiencing many more victories than defeats. At the Skunk Works we designed practical, used off-the-shelf parts whenever possible, and did things right the first time. My wing man, for example, had designed twenty-seven wings on previous Skunk Works’ airplanes before tackling the Hopeless Diamond. All of us had been trained by Kelly Johnson and believed fanatically in his insistence that an airplane that looked beautiful would fly the same way. No one would dare to claim that the Hopeless Diamond would be a beautiful airplane. As a flying machine it looked alien.

Dave Robertson, one of Kelly’s original recruits and aerospace’s most intuitively smart hydraulic specialist, ridiculed our design by calling it “a flying engagement ring.” Dave seldom minced words; he kept a fourteen-inch blowgun he had fashioned out of a jet’s tailpipe on his desk and would fire clay pellets at the necks of any other designers in the big drafting room who got on his nerves. Robertson hated having anyone look over his shoulder at his drawing and reacted by grabbing a culprit’s tie and cutting it off with scissors. Another opponent was Ed Martin, who thought that anyone who hadn’t been building airplanes since the propeller-driven days wasn’t worth talking to, much less listening to. He called the Hopeless Diamond “Rich’s Folly.” Some said that Ed’s bark was worse than his bite, but those were guys who didn’t know him.

Most of our veterans used slide rules that were older than Denys Overholser, and they wondered why in hell this young whippersnapper was suddenly perched on a throne as my guru, seemingly calling the shots on the first major project under my new and untested administration. I tried to explain that stealth technology was in an embryonic state and barely understood until Denys unearthed the Ufimtsev theory for us; they remained unconvinced even when I reminded them that until Denys had come along with his revelation, we had known only two possibilities to reduce an airplane’s radar detection. One way was to coat the fuselage, tail, and wing surfaces with special composite materials that would absorb incoming electromagnetic energy from radar waves instead of bouncing it back to the sender. The other method was to construct an airplane out of transparent materials so that the radar signals would pass through it. We tried an experimental transparent airplane back in the early 1960s and to our dismay discovered that the engine loomed ten times bigger on radar than the airplane because there was no way to hide it.

So all of us, myself especially, had to trust that Denys Overholser, with his boyish grin and quiet self-confidence, really knew what in hell he was talking about and could produce big-time results. Dick Cantrell, head of our aerodynamics group, suggested burning Denys at the stake as a heretic and then going on to conventional projects. Cantrell, normally as soft-spoken and calm as Gregory Peck, whom he vaguely resembled, nevertheless had the temperament of a fiery Savonarola when, as in this instance, basics of fundamental aerodynamics were tossed aside in deference to a new technology understood only by witches and mathematical gnomes. But after a couple of hours of listening to Overholser’s explanations of stealth, Dick dropped his lanky frame onto the chair across from my desk and heaved a big sigh. “Okay, Ben,” he muttered, “I surrender. If that flat plate concept is really as revolutionary as that kid claims in terms of radar cross section, I don’t care what in hell it looks like, I’ll get that ugly son-of-a-bitch to fly.”

We could get the Statue of Liberty to do barrel rolls with the onboard computers that achieved aerodynamic capability by executing thousands of tiny electrohydraulic adjustments every second to an airplane’s control surfaces. This computerized enhanced flight stability gave us latitude in designing small, stealthy wings and short tails and mini-wing flaps, and left the awesome problems of unstable pitch and yaw to the computers to straighten out. Without those onboard computers, which the pilots called “fly-by-wire,” since electric wiring now replaced conventional mechanical control rods, our diamond would have been hopeless indeed. But even with the powerful onboard computers, getting into the sky, as Kelly’s boot to my butt suggested, would be far from a cakewalk.

We had a very strong and innovative design organization of about a dozen truly brilliant engineers, working at their drawing boards in a big barnlike room on the second floor of our headquarters building, who simply could not be conned or browbeaten into doing anything they knew would not work. One day, Kelly called upstairs for an engineer named Bob Allen. “Bob Allen there?” he asked. Whoever answered the phone replied, “Yeah, he is.” And hung up. Kelly was livid, but deep down he appreciated the feisty independence of his best people. The designers were either structural specialists who planned the airframe or systems designers who detailed the fuel, hydraulics, electrical, avionics, and weapons systems. In many ways they comprised the heart and soul of the Skunk Works and also were the most challenged by the structural demands of the new stealth technology. Thanks to Ufimtsev’s breakthrough formula, they were being told to shape an airplane entirely with flat surfaces and then tilt the individual panels so that radar energy scattered away and not back to the source. The airplane would be so deficient in lift-drag ratio that it would probably need a computer the size of Delaware to get it stable and keep it flying.

Several of our aerodynamics experts, including Dick Cantrell, seriously thought that maybe we would do better trying to build an actual flying saucer. The shape itself was the ultimate in low observability. The problem was finding ways to make a saucer fly. Unlike our plates, it would have to be rotated and spun. But how? The Martians wouldn’t tell us.

During those early months of the Hopeless Diamond, I dug in my heels. I forced our in-house doubters to sit down with Denys and receive a crash course on Stealth 101. That helped to improve their confidence quotient somewhat, and although I acted as square-shouldered as Harry Truman challenging the Republican Congress, deep down I was suffering bouts of angst myself, wondering if Kelly and some of the other skeptics had it right while I was being delusional. I kept telling myself that the financial and personal risks in pursuing this project were minimal compared to its enormous military and financial potential. But the politics of the situation had me worried: stealth would have been a perfect third project for me, after two reassuring successes under my belt.

But if stealth failed, I could hear several of my corporate bosses grousing: “What’s with Rich? Is he some sort of flake? Kelly would never have undertaken such a dubious project. We need to take charge of that damned Skunk Works and make it practical and profitable again.”

Kelly Johnson would never double-cross me by bad-mouthing the stealth project in the corridors of the Skunk Works, but all of us knew Kelly too well not to be able to read his mood and mind. If he didn’t like something or someone, it was as obvious as a purple pimple on the tip of his nose. So I had him in for lunch and said, “Look, Kelly, I know you find this design aesthetically offensive, but I want you to do me one favor. Sit down with this guy, Overholser, and let him answer your questions about stealth. He’s convinced me that we are onto something enormously important. Kelly, this diamond is somewhere between ten thousand and one hundred thousand times lower in radar cross section than any U.S. military airplane or any new Russian MiG. Ten thousand to one hundred thousand times, Kelly. Think of it!”

Kelly remained unmoved. “Theoretical claptrap, Ben. I’ll bet you a quarter that our old D-21 drone has a lower cross section than that goddam diamond.”

We had a ten-foot wooden model of the diamond, and we took it and the original wooden model for the manta ray — shaped D-21 drone and put them side by side into an electromagnetic chamber and cranked up the juice.

That date was September 14, 1975, a date etched forever in my memory because it was about the only time I ever won a quarter from Kelly Johnson. I had lost about ten bucks’ worth of quarters to him over the years betting on technical matters. Like me, my colleagues collected quarters from Kelly just about as often as they beat him at arm wrestling. He had been a hod carrier as a kid and had arms like ship’s cables. He once sprained the wrist of one of our test pilots so badly he put the poor guy out of action for a month. So winning a quarter was a very big deal, in some ways even more satisfying than winning the Irish Sweepstakes. (Depending on the size of the purse, of course.)

I really wanted a photographer around for historical purposes to capture the expression on Kelly’s big, brooding moon-shaped mug when I showed him the electromagnetic chamber results. Hopeless Diamond was exactly as Denys had predicted: a thousand times stealthier than the twelve-year-old drone. The fact that the test results matched Denys’s computer calculations was the first proof that we actually knew what in hell we were doing. Still, Kelly reacted about as graciously as a cop realizing he had collared the wrong suspect. He grudgingly flipped me the quarter and said, “Don’t spend it until you see the damned thing fly.”

But then he sent for Denys Overholser and grilled the poor guy past the point of well-done on the whys and hows of stealth technology. He told me later that he was surprised to learn that with flat surfaces the amount of radar energy returning to the sender is independent of the target’s size. A small airplane, a bomber, an aircraft carrier, all with the same shape, will have identical radar cross sections. “By God, I never would have believed that,” he confessed. I had the feeling that maybe he still didn’t.

Our next big hurdle was to test a ten-foot wooden model of the Hopeless Diamond on an outdoor radar test range near Palmdale, on the Mojave desert. The range belonged to McDonnell Douglas, which was like Buick borrowing Ford’s test track to road test an advanced new sports car design, but I had no choice since Lockheed didn’t own a radar range. Our model was mounted on a 12-foot-high pole, and the radar dish zeroed in from about 1,500 feet away. I was standing next to the radar operator in the control room. “Mr. Rich, please check on your model. It must’ve fallen off the pole,” he said. I looked. “You’re nuts,” I replied. “The model is up there.” Just then a black bird landed right on top of the Hopeless Diamond. The radar operator smiled and nodded. “Right, I’ve got it now.” I wasn’t about to tell him he was zapping a crow. His radar wasn’t picking up our model at all.

For the first time, I felt reassured that we had caught the perfect wave at the crest and were in for one terrifically exciting ride. I saw firsthand how invisible that diamond shape really was. So I crossed my fingers and said a silent prayer for success in the tests to follow.

In early April 1976, I got the word that we had officially won the competition with Northrop and would go on to build two experimental airplanes based on our Hopeless Diamond design. The program was now designated under the code name Have Blue. We knew we could produce a model with spectacularly low radar signatures, but the big question was whether we could actually build an airplane that would enjoy the same degree of stealthiness. A real airplane was not only much larger, but also loaded with all kinds of anti-stealth features — a cockpit, engines, air scoops and exhausts, wing and tail flaps, and landing gear doors. In any airplane project the design structures people, the aerodynamics group, and the propulsion and weight specialists all argue and vie for their points of view. In this case, however, I served notice that Denys Overholser’s radar cross section group had top priority. I didn’t give a damn about the airplane’s performance characteristics because its only purpose was to demonstrate the lowest radar signature ever recorded. I joked that if we couldn’t get her airborne, maybe we could sell her as a piece of modern art sculpture.

I assigned the design project to Ed Baldwin, who was our best and most experienced structural engineer. “Baldy” had started out with Kelly designing the P-80, America’s first jet fighter, in 1945, and had designed the configuration of the U-2 spy plane. His task was to take the preliminary design concept of the Hopeless Diamond and make it practical so that it could actually fly. Dick Scherrer had done the preliminary design, laying out the basic shape, and Baldwin had to make certain that the shape’s structure was sound and practical; he would determine its radius, its thicknesses, its ability to withstand certain loads, the number of parts. “Baldy” would put the rubber on the ramp.

All of our structure and wing guys worked for him, and Baldwin enjoyed badgering aerodynamicists, especially in meetings where he could score points with his fellow designers by making aerodynamicists squirm or turn beet red in fury. One on one, Baldy was a pleasant chap — at least moderately so for a crusty Skunk Works veteran — but in meetings we were all fair game and he was a bad-tempered grizzly. Early on, for example, he got into a heated exchange with a very proper Britisher named Alan Brown, our propulsion and stealth expert, about some aspect of the structure he was designing. Baldwin turned crimson. “Goddam it, Brown,” he said, “I’ll design this friggin’ airplane and you can put on the friggin’ stealth afterwards.”

The airplane Baldy designed was a single-seat, twin-engined aircraft, 38 feet long, with a wingspan of 22 feet and a height of slightly more than seven feet. Its gross weight was 12,000 pounds. The leading edge of the delta wing was razor-sharp and swept back more than 70 degrees. To maintain low infrared signatures, the airplane could not go supersonic or have an afterburner because speed produced surface heating — acting like a spotlight for infrared detection. Nor did we want the airplane to be aurally detected from the ground. For acoustical reasons we had to make sure we had minimized engine and exhaust noise by using absorbers and shields. To keep it from being spotted in the sky, we decided to use special additives to avoid creating exhaust contrails. And to eliminate telltale electromagnetic emissions, there was no internal radar system on board.

Our airplane wasn’t totally invisible, but it held the promise of being so hard to detect that even the best Soviet defenses could not accomplish a fatal lock-on missile cycle in time to thwart its mission. If they could detect a fighter from a hundred miles out, that airplane was heading for the loss column. The long-range radar had plenty of time to hand off the incoming intruder to surface-to-air missile batteries, which, in turn, would fire the missiles and destroy it. Early-warning radar systems could certainly see us, but not in time to hand us over to missile defenses. If the first detection of our airplane was at fifteen miles from target, rather than at fifty miles, there simply would be no time to nail us before we hit the target. And because we were so difficult to detect, even at fifteen miles, radar operators would also be thwarted while trying to detect us through a confusing maze of ground clutter.

I had asked Kelly to estimate the cost of building these two experimental Have Blue airplanes. He came back with the figure of $28 million, which turned out to be almost exactly right. I asked the Air Force for $30 million, but they had only $20 million to spend in discretionary funds for secret projects by which they bypassed congressional appropriations procedures. So, in the late spring of 1976, I was forced to go begging for the missing $10 million to our CEO, Bob Haack, who was sympathetic but not particularly enthusiastic. He said, “Look, Ben, we’re in tough straits right now. I don’t think we can really afford this.” I pushed a little harder and got him to agree to let me present the proposal to the full board of directors. Bob set up the meeting, and I just laid it all out. Larry Kitchen, Lockheed’s president, and Roy Anderson, the vice chairman, spoke up enthusiastically in support. I told the board I thought we were dealing with a project that had the potential for $2 to $3 billion in future sales. I predicted we would be building stealth fighters, stealth missiles, stealth ships, the works. I was accused of hyperbole by one or two directors, but in the end I got my funding, and as time went on my sales predictions proved to be conservatively low.

Even worse, I began picking up rumors that certain officials at the Pentagon were accusing me of rigging the test results of the radar range competition against Northrop. An Air Force general called me, snarling like a pit bull. “Rich, I’m told you guys are pulling a fast one on us with phony data.” I was so enraged that I hung up on that son of a bitch. No one would have ever dared to accuse Kelly Johnson’s Skunk Works of rigging any data, and by God, no one was going to make that accusation against Ben Rich’s operation either. Our integrity was as important to all of us as our inventiveness. The accusation, I discovered, was made by a civilian radar expert advising the Air Force, who had close ties to leading manufacturers of electronic jamming devices installed in all Air Force planes to fool or thwart enemy radar and missiles. If stealth was as good as we claimed, those companies might be looking for new work.

His motivation for bad-mouthing us was obvious; but it was equally apparent that we were unfairly being attacked without any effective way for me to defend the Skunk Works’ integrity from three thousand miles away. So I invited one of the nation’s most respected radar experts to Burbank to personally test and evaluate our stealth data. MIT’s Lindsay Anderson accepted my invitation in the late summer of 1976 and arrived at my doorstep carrying a bag of ball bearings in his briefcase. The ball bearings ranged in size from a golf ball to an eighth of an inch in diameter. Professor Anderson requested that we glue each of these balls onto the nose of the Hopeless Diamond and then zap them with radar. This way he could determine whether our diamond had a lower cross section than the ball bearings. If the diamond in the background proved to be brighter than the ball in the foreground, then the ball could not be measured at all. That got me a little nervous because nothing should measure less than an eighth-of-an-inch ball bearing, but we went ahead anyway. As it turned out, we measured all the balls easily — even the eighth-of-an-inch one — and when Professor Anderson saw that the data matched the theoretical calculated value of the ball bearings, he was satisfied that all our claims were true.

That was the turning point for the entire stealth adventure, which could have ended right there if Lindsay Anderson had reinforced the accusation that we were being unscrupulous and presenting bogus data. But once he corroborated our achievement back in Washington, I was informed by a telegram from the Air Force chief of staff that Have Blue was now classified “Top Secret — Special Access Required.” That security classification was rare — clamped only on such sensitive programs as the Manhattan Project, which created the first atomic bomb during World War II. My first reaction was “Hooray, they finally realize how significant this technology really is,” but Kelly set me straight and with a scowl urged me to cancel the whole damned project right then and there.

“Ben,” Kelly warned me, “the security they’re sticking onto this thing will kill you. It will increase your costs twenty-five percent and lower your efficiency to the point where you won’t get any work done. The restrictions will eat you alive. Make them reclassify this thing or drop it.” On matters like that, Kelly was seldom wrong.

Before the government would sign a contract with me I had to submit for approval a security plan, detailing how we would tighten all the hatches of what was already one of the most secure operations in the defense industry. Hell, we already operated without windows and behind thick, eavesdrop-proof walls. We had special bank-vault conference rooms, lined with lead and steel, for very sensitive discussions about very secret matters. Still, the Air Force required me to change our entire security system, imposing the kinds of strictures and regulations that would drive us all nuts in either the short or long run. Every piece of paper dealing with the project had to be stamped top secret, indexed in a special security filing system, and locked away. Full field investigations were demanded of every worker having access to the airplane. They imposed a strictly enforced two-man rule: no engineer or shop worker could be left alone in a room with a blueprint. If one machinist had to go to the toilet, the co-worker had to lock up the blueprint until his colleague returned.

Only five of us were cleared for top secret and above, and over the years we had worked on tremendously sensitive projects without ever suffering a leak or any known losses to espionage. In fact, Kelly evolved his own unorthodox security methods, which worked beautifully in the early days of the 1950s. We never stamped a security classification on any paperwork. That way, nobody was curious to read it. We just made damned sure that all sensitive papers stayed inside the Skunk Works.

My biggest worry was clearing our workers for this project. They needed Special Access clearances, and I had to make the case for their Need to Know on an individual basis. But the government, not the employer, was the final arbiter of who was granted or denied access. The Air Force security people made the decision and offered no explanation about why certain of my employees were denied access to the program. No one in Washington conferred with me or asked my opinion or sought my advice. I knew my people very well. Some were horse players, several were skirt chasers, a few were not always prompt about paying their bills. For all I knew some of my best people might be part-time transvestites. I had no doubt that some of the younger ones may have indulged in “recreational drugs,” like toking marijuana at rock shows. Any of these “sins” could sink a valuable worker. I did win a couple of important concessions: the Air Force agreed that only those few technicians with a need to know the airplane’s radar cross section would require the complete full field investigation, which took around nine months, and I was granted temporary clearances for twenty specialists working on particular sensitive aspects of Have Blue. Most important, I raised so much cain that Air Force security finally granted me a “grandfather clause” for many of our old-timers who had been working on all our secret projects since the days of the U-2. They were granted waivers to work on Have Blue.

But security’s dragnet poked and prodded into every nook and cranny of our operation. Keith Beswick, head of our flight test operations, designed a coffee mug for his crew with a clever logo showing the nose of Have Blue peeking from one end of a big cloud with a skunk’s tail sticking out the back end. Because of the picture of the airplane’s nose, security classified the mugs as top secret. Beswick and his people had to lock them away in a safe between coffee breaks. The airplane itself had to be stamped SECRET on the inside cockpit door. I was named its official custodian and had to sign for it whenever it left its hangar area and was test-flown. If it crashed, I was personally responsible for collecting every single piece of it and turning all of it over to the proper authorities.

These draconian measures hobbled us severely at times, tested my patience beyond endurance, and gave Kelly every right to scold, “Goddam it, Rich, I told you so.” At one point I had to memorize the combinations to three different security safes just to get work done on a daily basis. A few guys with lousy memories tried to cheat and carried the combination numbers in their wallets. If security caught them, they could be fired. Security would snoop in our desks at night to search for classified documents not locked away. It was like working at KGB headquarters in Moscow.

The Air Force wanted the two test planes in only fourteen months. Over the years we had developed the concept of using existing hardware developed and paid for by other programs to save time and money and reduce the risks of failures in prototype projects. I worked an agreement with the Air Force to supply me with the airplane engines. They assigned an expediter named Jack Twigg, a major in the Tactical Air Command, who was cunning and smart. Jack requisitioned six engines from the Navy. He went to General Electric’s jet engine division, did some fast talking to the president and plant manager, got some key people to look the other way while he carted away the six J-85 engines we needed right off their assembly line, and had them shipped in roundabout ways, so that nobody knew the Skunk Works was the final destination. We put two engines in each experimental airplane and had a couple of spares. Jack was a natural at playing James Bond: he ordered parts in different batches and had them shipped using false return addresses and drop boxes.

We begged and borrowed whatever parts we could get our hands on. Since this was just an experimental stealth test vehicle destined to be junked at the end, it was put together with avionics right off the aviation version of the Kmart shelf: we took our flight control actuators from the F-111 tactical bomber, our flight control computer from the F-16 fighter, and the inertial navigation system from the B-52 bomber. We took the servomechanisms from the F-15 and F-111 and modified them, and the pilot’s seat from the F-16. The heads-up display was designed for the F-18 fighter and adapted for our airplane. In all we got about $3 million worth of equipment from the Air Force. That was how we could build two airplanes and test them for two years at a cost of only $30 million. Normally, a prototype for an advanced technology airplane would cost the government three or four times as much.

Only the flight control system was specially designed for Have Blue, since our biggest sweat was aerodynamics. We decided to use the onboard computer system of General Dynamics’s small-wing lightweight fighter, the F-16, which was designed unstable in pitch; our airplane would be unstable in all three axes — a dubious first that brought us plenty of sleepless nights. But we had our very own Bob Loschke, acknowledged as one of the very best onboard computer experts in aerospace, to adapt the F-16’s computer program to our needs. We flew the airplane avionically on the simulator flight control system and kept modifying the system to increase stability. It was amazing what Loschke could accomplish artificially by preempting the airplane’s unstable responses and correcting them through high-powered computers.

The pilot tells the flight control system what he wants it to do just by normal flying: maneuvering the throttle and foot pedals directing the control surfaces. The electronics will move the surfaces the way the pilot commands, but often the system will automatically override him and do whatever it has to do to keep the system on track and stable without the pilot even being aware of it. Our airplane was a triumph of computer technology. Without it, we could not even taxi straight.

In July 1976, we began building the first of two Have Blue prototypes in Building 82, one of our big assembly hangars, the size of three football fields. We had our ownunique method for building an airplane. Our organizational chart consisted of an engineering branch, a manufacturing branch, an inspection and quality assurance branch, and a flight testing branch. Engineering designed and developed the Have Blue aircraft and turned it over to the shop to build. Our engineers were expected on the shop floor the moment their blueprints were approved. Designers lived with their designs through fabrication, assembly, and testing. Engineers couldn’t just throw their drawings at the shop people on a take-it-or-leave-it basis and walk away.

Our senior shop people were tough, experienced SOBs and not shy about confronting a designer on a particular drawing and letting him know why it wouldn’t work. Our designers spent at least a third of their day right on the shop floor; at the same time, there were usually two or three shop workers up in the design room conferring on a particular problem. That was how we kept everybody involved and integrated on a project. My weights man talked to my structures man, and my structures man talked to my designer, and my designer conferred with my flight test guy, and they all sat two feet apart, conferring and kibitzing every step of the way. We trusted our people and gave them the kind of authority that was unique in aerospace manufacturing. Above all, I didn’t second-guess them.

Our manufacturing group consisted of the machine shop people, sheet metal fabrication and assemblers, planners, tool designers, and builders. Each airplane required its own special tools and parts, and in projects like Have Blue, where only two prototypes were involved, we designed and used wooden tools to save time and money. When the project ended, we just threw them away.

The shop manufactured and assembled the airplane, and the inspection and quality assurance branch checked the product at all stages of development. That was also unique with us, I think. In most companies quality control reported to the head of the shop. At the Skunk Works quality control reported directly to me. They were a check and balance on the work of the shop. Our inspectors stayed right on the floor with the machinists and fabricators, and quality control inspections occurred almost daily, instead of once, at the end of a procedure. Constant inspection forced our workers to be supercritical of their work before passing it on. Self-checking was a Skunk Works concept now in wide use in Japanese industry and called by them Total Quality Management.

Our workers were all specialists in specific sections of the airplane: fuselage, tail, wings, control surfaces, and power plant. Each section was built separately then brought together and assembled like a giant Tinkertoy. We used about eighty shop people on this project, and because we were in a rush and the airplane was small, we stood it on its tail and assembled it vertically. That way, the assemblers could work on the flat, plated structural frame, front and back, asses to elbows, simultaneously. I kept Alan Brown, our stealth engineer, on the floor all the time to answer workers’ questions.

Flat plates, we discovered, were much harder to tool than the usual rounded surfaces. The plates had to be absolutely perfect to fit precisely. We also had nagging technical headaches applying the special radar-absorbing coatings to the surfaces. Each workday the problems piled higher and I sat behind Kelly’s old desk reaching for my industrial-size bottle of headache tablets. Meanwhile, the Navy came to us to test the feasibility for a stealthy weapons system and set up their own top secret security system that was twice as stringent as the Air Force’s. We had to install special alarm systems that cost us a fortune in the section of our headquarters building devoted to the naval work. And we were also doing some prototype work for the Army on stealthy munitions.

In the midst of all this interservice rivalry, security, and hustle and bustle, Major General Bobby Bond, who was in charge of tactical air warfare, came thundering into the Skunk Works with blood in his eye on a boiling September morning. The Santa Ana winds were howling and half of L.A. was under a thick pall of smoke from giant brush fires, mostly started by maniacs with matches. My asthma was acting up and I had a lousy headache and I was in no mood for a visit from the good general, even though I had a special regard for the guy. But General Bond was a brooder and a worrier, who drove me and everyone else absolutely bonkers at times. He always thought he was being shortchanged or victimized in some way. He pounded on my desk and accused me of taking some of my best workers off his Have Blue airplane to work on some rumored secret Navy project. I did my best to look hurt and appeased Bobby by even raising my right hand in a solemn oath. I told myself, So, it’s a little white lie. What else can I do? The Navy project is top secret and Bond has no need to know. We could both go to jail if I told him what was really up.

Unfortunately, on the way out to lunch, the general spotted a special lock and alarm system above an unmarked door which he knew from prowling the rings of the Pentagon was used only by the Navy on its top secret projects. Bond squeezed my arm. “What’s going on inside that door?” he demanded to know. Before I could think up another lie, he commanded me to open up that door. I told him I couldn’t; he wasn’t cleared to peek inside. “Rich, you devious bastard, I’m giving you a direct order, open up that goddam door this instant or I’ll smash it down myself with a goddam fire ax.” The guy meant every word. He began pounding on the door until it finally opened a crack, and he forced his way in. There sat a few startled Navy commanders.

“Bobby, it isn’t what you think,” I lied in vain.

“The hell it isn’t you lying SOB,” he fumed.

I surrendered, but not gracefully. I said, “Okay, you got me. But before we go to lunch you’re going to have to sign an inadvertent disclosure form or security will have both our asses.” The Navy, of course, was outraged at both of us. An Air Force general seeing their secret project was as bad as giving a blueprint to the Russians.

Bobby[2] didn’t worry about the Navy very long, because we gave him far bigger worries than that: four months before we were supposed to test-fly Have Blue our shop mechanics went out on strike.

The International Association of Machinists’ negotiations with the Lockheed corporation on a new two-year contract failed in late August 1977. Our workers hit the bricks just as Have Blue was going into final assembly, perched on its jig with no hydraulic system, no fuel system, no electronics or landing gear. There seemed to be no way we would be ready to fly by December 1, our target date, and our bean counters wanted to inform the Air Force brass that we would be delayed one day for each day of the strike. But Bob Murphy, our veteran shop superintendent, insisted that he could get the job done on time and meet our commitment for first flight. To Murphy, it was a matter of stubborn Skunk Works pride.

Bob put together a shop crew of thirty-five managers and engineers who worked twelve hours a day, seven days a week, over the next two months. Fortunately, most of our designers were all great tinkerers, which is probably why they were drawn to engineering in the first place. Murphy had Beswick, our flight test head, working with a shop supervisor named Dick Madison assembling the landing gear. Murphy himself put in the ejection seat and flight controls; another shop supervisor named John Stanley worked alone on the fuel system. Gradually, the airplane began coming together, so that by early November Have Blue underwent strain gauge calibrations and fuel system checkout. Because Have Blue was about the most classified project in the free world, it couldn’t be rolled outdoors, so the guys defied rules and regulations and ran fuel lines underneath the hangar doors to tank up the airplane and test for leaks. But how could we run engine tests?

Murphy figured out a way. He rolled out the plane after dark to a nearby blast fence about three hundred yards from the Burbank Airport main runway. On either side he placed two tractor trailer vans and hung off one end a large sheet of canvas. It was a jerry-built open-ended hangar that shielded Have Blue from view; security approved provided we had the airplane in the hangar before dawn.

Meanwhile an independent engineering review team, composed entirely of civil servants from Wright Field in Ohio, flew to Burbank to inspect and evaluate our entire program. They had nothing but praise for our effort and progress, but I was extremely put out by their visit. Never before in the entire history of the Skunk Works had we been so closely supervised and directed by the customer. “Why in hell do we have to prove to a government team that we knew what we were doing?” I argued in vain to Jack Twigg, our assigned Air Force program manager. This was an insult to our cherished way of doing things. But all of us sensed that the old Skunk Works valued independence was doomed to become a nostalgic memory of yesteryear, like a dime cup of coffee.

We had lived and died by fourteen basic operating rules that Kelly had written forty years earlier, one night while half in the bag. They had worked for him and they worked for me:

Although most of our cherished rules were now in tatters, my guys managed to finish their work on Have Blue in mid-November, nearly three weeks before the flight test target date of December 1, 1977. “Rich,” Bob Murphy teased, “you’d never have made your deadline by using regular workers. You had the cream of the crop in management delivering the goods for you.” The airplane was loaded onto a C-5 cargo plane at two in the morning and roared away to our remote test site, leaving behind several complaints to the FAA from irate citizens whose sleep was disturbed by this violation of late-night takeoffs from the Burbank Airport. Frankly, it was such a relief to get Have Blue out of assembly that I would have gladly paid a fine.

The plane was now in the hands of our flight test crews, who would spend the next couple of weeks performing flight control, engine, and taxi tests. Even though the test site was in a remote location, our airplane was kept under wraps inside its hangar most of the time. Soviet satellites made regular passes, and every time our airplane was rolled out everyone on the base who wasn’t cleared for Have Blue had to go into the windowless mess hall and have a cup of coffee until we took off.

Seventy-two hours before the first test flight, the airplane began to seriously overheat near the tail during engine test runs. The engine was removed, and Bob Murphy and a helper decided to improvise by building a heat shield. They noticed a six-foot steel shop tool cabinet. “Steel is steel,” Murphy said to his assistant. “We’ll send Ben Rich the bill for a new cabinet.” They began cutting up the cabinet to make the heat shield panels between Have Blue’s surface and its engine. And it worked perfectly. Only in the Skunk Works…

It’s the first of December, 1977, just after sunup, the best time for test pilots to take off. Winds are usually calmest then, but this morning the wind chill blasts through my topcoat like it’s tissue paper. I’m wondering how I can be so damned cold while I’m sweating bullets over this test flight — probably the most critical test of my career. This flight will be every bit as important to the nation’s future and the future of the Skunk Works as the first test flight of the U-2 spy plane, which took place at this very same highly secret sand pile more than a quarter century ago.

Back then, I was a Skunk Works rookie and this base, which we built for the CIA, was just a tiny outpost of windswept quonset huts and trailers, guarded by rookie CIA agents with tommy guns. Kelly had jokingly nicknamed this godforsaken place Paradise Ranch, hoping to lure young and innocent flight crews to work on a dry lake bed where quarter-inch rocks blew around most afternoons. It is now a sprawling facility, bigger than some municipal airports, a test range for sensitive aviation projects. No one nowadays gains access without special clearances that include a polygraph test. Such paranoia has kept our most guarded national defense secrets secret.

I’ve been here many times over the years on many Skunk Works test flights, usually accompanying Kelly Johnson. Today, the Have Blue prototype that will soon be rolling down this runway is the first built under my regime after Johnson’s retirement three years earlier. But we really aren’t one hundred percent certain that this sucker can actually get off the ground. It is the most unstable and weirdest-looking airplane since Northrop’s Flying Wing, built on a whim back in the late 1940s.

I watch nervously as Have Blue emerges from the guarded cavity inside its hangar and is rolled out. It is a flying black wedge, carved out of flat, two-dimensional angles. Head on, with its black paint and highly swept wings, it looks like a giant Darth Vader — the first airplane that has not one rounded surface.

Bill Park, our chief test pilot, complained that it was the ugliest airplane he’d ever strapped himself into. Bill claimed that flying such a mess earned him the right to double hazard pay. I agreed. He’s getting a $25,000 bonus for this series of Have Blue test flights. To Bill, even the opaque triangular cockpit is ominous, especially if he has to punch out. But the specially coated glass will keep radar beams from picking up his helmeted head. The real beauty of Have Blue is that Bill’s head is a hundred times more observable on radar than the airplane he will be flying.

The sharp edges and extreme angular shape of our small prototype create whirling tornadoes and make the airplane a flying vortex generator. To be able to fly at all, the airplane’s fly-by-wire system must operate perfectly, otherwise Have Blue will tumble out of control.

I check my watch. Nearly 0700. I give the thumbs-up sign to Bill Park in the cockpit, who’s preoccupied with last-minute preflight checks. Kelly Johnson is standing at my side, looking stoic. He’s still skeptical about whether or not this prototype will prove way too draggy to get off the ground. But Kelly brought along a case of champagne on the Jetstar from Burbank to celebrate after Park’s flight. Over the years at the Skunk Works we’ve never failed to celebrate a successful maiden test flight of anything we’ve ever built. We always polished off a hard-earned success with a boisterous party where Kelly challenged all comers to an arm-wrestling contest. He’s an old man now, ailing, but I still wouldn’t take him on.

We’ve had our share of crashes during long weeks and months of test-flying new airplanes, but they didn’t really upset us too much as long as no one got hurt, because we always learned important lessons from mistakes. But we never had a mishap on a first test flight — a catastrophe that would send us back to the drawing board with our tails between our legs.

Adding to the tensions of this day, the White House Situation Room is monitoring this flight. So is the Tactical Air Command at Langley Air Force Base in Virginia. But my anxieties are closer to home: I’ve got ten million bucks of Lockheed’s money riding on this flight and the success of this program. I’m the one who talked our board into going along with me. So I didn’t need any black coffee this morning. I am wired.

Bill Park fires the twin engines. The airplane has a muffled sound because its engines are hidden behind special radar-absorbing grids. Bill has been practicing using these flight controls under all conditions in a simulator for five weeks and I know he’s ready for any emergency. He and I have been through tight spots on other test programs. Once he ejected from an SR-71 that began to flip over on takeoff. I was sure Park was about to become a grease spot on the tarmac, but his chute opened just as his feet hit the ground, yanking him upward as he was impacting. He left three-inch-deep heel imprints in the sand, but was unhurt. Bill is damned thorough and damned lucky, a great combination for someone in his line of work.

Kelly Johnson is watching intently as the prototype taxis past us heading for the end of the runway, where it will turn and take off into the stiff wind. Suddenly, in a blast of loud noise, the medevac chopper, with two paramedics on board, takes off and heads down range to be in position if Park augers in. It is followed by a T-38 jet trainer carrying one of Bill’s test pilot colleagues, who will fly chase, visually monitoring his airplane and supplying any help or advice in an emergency.

Bill pushes on the throttle, and Have Blue slowly begins to accelerate. To stay stealthy Have Blue has no afterburner, and it will need almost as much runway to get airborne as a 727 loaded with fuel, baggage, and passengers bound for Chicago.

Bill goes full throttle. He’s chewing up a lot of runway as he sweeps past us. I’m thinking, Damn it, with all that wind he should be up by now. He’s far down the runway and I’m no longer breathing. Uh-oh. He’s damn near off the end of the goddam runway. Then I see him lift off. Slow as a jumbo jet a hundred times its weight, but he’s up. His nose is high. But just hanging there. Get up. Up, up, up. The little airplane hears me. It’s heading toward the snow-powdered mountains. Ken Perko of the Pentagon’s Advanced Research Projects Agency, who is among the half dozen outsiders cleared to witness this flight test, reaches out to shake my hand. “By God, Ben,” he says, “the Skunk Works has done it again.”

Kelly slaps me on the back and shouts, “Well, Ben, you got your first airplane.”

Not so fast. It’s standard procedure to leave the landing gear down on maiden test flights checking out airworthiness, but even so it seems to me the airplane is way too sluggish gaining altitude. There are some significant foothills looming in Bill’s flight path and I try to do some quick mental calculating to get him safely over the hump. I raise my binoculars and quickly try to adjust the focus. By the time the mountains come clear, our airplane is across the other side.