WHEN TEST PILOT TONY LEVIER FIRST SAW the F-104 Star-fighter, he asked, “Where are the wings?”
It is true that the wings were short, straight, and almost razor thin, but they carried quite a load. They were the result of tests on some 50 wing models fired on instrumented rockets attaining speeds of 1,500 miles an hour.
The “Missile With a Man in It” resulted from my tour of the Korean battlefields in 1952. Lt. Gen. Benjamin Chidlaw, chief of Materiel Command, and other Air Force officers wanted to discover—and wanted the aircraft designers to know—how our aircraft were performing and what our combat pilots needed in confrontation with the enemy. This was the first war in which both sides had jet aircraft. The North Koreans—in effect, Chinese—had the MiG-15. The South Koreans—in effect, USA—had the F-80, F-84, and later the F-86.
It was an education to see our pilots operating from forward bases like Taigu, their aircraft so heavily laden on takeoff with wingtip fuel tanks to fly into enemy territory that the tanks would scrape the runway. The runway was made of steel planking and there was danger of fire, but they did it day after day.
We interviewed these pilots as they climbed out of their aircraft on return from missions. What did they want in a fighter? It was unanimous. They wanted speed and altitude. In combat at fighting altitude the two sides were about equal in speed. We were much more maneuverable with power controls, and our pilots had the advantage of the latest gunsights. Overall, our margin of victory was something like ten to one.
But our pilots were insulted constantly by “High-altitude Charlie,” sitting at 50,000 feet and directing the MiGs in Chinese or Russian.
“Don’t worry about the American airplanes. Your airplane will take care of you. Just come up here. They can’t get up to you,” our pilots heard constantly.
On our tour of the Korean battlefronts, we covered more than 23,000 miles and visited 15 air bases, flying in an Air Force Constellation.
On night flights, my civilian companion, Lee Atwood, vice president of engineering for North American Aviation, and I slept on a piece of plywood placed on the floor. I could detect vibrations from number three engine coming through a rear spar. The engine ran rougher and rougher but held together as we flew to Hawaii, then Wake, up to Japan, around the Korean air bases, down to Okinawa, and on to Manila. Just as we attained cruising altitude on takeoff from Manila, there was a big bang and number three was dead. We dumped fuel and returned. The Air Force had carried enough spare parts for a minor overhaul, so within 12 hours we were on our way again.
When we returned to the U.S., I set out to design a jet fighter that would fly higher and faster than anything flying anywhere. There was no formal requirement from the Air Force yet for such a plane, but that was cleared up very quickly with a visit to the Pentagon. When I showed my proposal to Gen. Don Putt and Don Yates and Col. Bruce Holloway, they were very receptive. The only obstacle to a contract was the absence of a document specifying what the next fighter should be.
“Well, if there isn’t a requirement, I’m going to make one,” Colonel Holloway decided. “Stick around, Kelly. Come back in a couple of hours.”
He then wrote a page and a quarter on what the Air Force required for its next fighting plane. It should be lightweight, capable of certain performance at sea level and high altitude, and carry specific armament, radio, and other instrumentation. In that short span, he had obtained the approval of all the appropriate generals.
“Here are your requirements,” he told me. Generals Putt and Yates agreed. “Go see what you can do with this.”
The result was the F-104 Starfighter, which would become the SuperStarfighter, and evolve through eight different models and production in the U.S. and six other countries of the Free World over a period of more than 25 years. This was to be the largest international industrial collaboration in the world. The airplane eventually would be flown by 14 allied air forces as well as the USAF. A two-place version also was built, not only for training but for tactical use.
The F-104 became the first operational airplane to fly twice the speed of sound in level flight.
Development of those short, thin wings utilized a new test technique—out of necessity. We didn’t have a wind tunnel capable of going to Mach 2, the speed for which we were designing the airplane, so I visited Lt. Gen. Earle E. “Pat” Partridge, in charge of the Air Research and Development Command in Baltimore, and described the problems we were having in testing the thin wing as well as the tail and air intake ducts for the new fighter.
“What can I do for you?” he asked.
“Well, if we had a bunch of five-inch rockets, we could put the wing models on the rockets. If we shot enough of them we could find out how to make a wing with the thickness we want—about twice that of a razor blade. We can see if it flutters or not in supersonic flight.”
Immediately, General Partridge sent a message to Korea, “Stop shooting rockets for one morning and send them all to Kelly.”
Two weeks later when I returned to Burbank, the Skunk Works was in an uproar. There were about 460 rockets on hand and no one knew what to do with them. It wasn’t a good idea to store them in the middle of Burbank, and, of course, we couldn’t fire them from the city. We moved them all to Edwards Air Force Base, where there was plenty of open country within the security of the test base boundaries.
We fitted the rockets with automatic cameras and telemetering equipment to radio data to ground observation stations. We put wing after wing on instrumented rockets and fired them from the desert base. Wings with different stiffness, different shapes, different designs, at speeds up to 1,500 miles an hour. There had been some question within the industry about the advisability of making a wing as thin as we proposed. Not only did the final design prove solid in test flight and in service, but we later hung wing tip fuel tanks on it, a great deal of armament, and even added A-bomb carrying capability.
The unusual “flying” tail configuration also was tested initially on these rocket “wind tunnels.”
With the first F-104, we faced again the problem of starting work on an airplane before there was an engine for it. So, like the F-80, the F-104 flew with one engine in the first two prototypes before acquiring the engine for which it really was designed—the General Electric J-79.
The first airplane was built and flown in a year and a day. Tony LeVier took off from Edwards AFB on February 28, 1954. I like to schedule first flights on my birthday when I can, but I missed this one by a day.
That new GE engine gave us trouble early in the flight program. The problem was in the control to the afterburner—which injects extra fuel into the hot exhaust gas, burning it to raise temperatures and speed flow out the tailpipe. It can double the thrust when an extra burst of power is needed for takeoff or in flight.
On this engine the afterburner eyelids had a nasty habit of opening unexpectedly under certain conditions, resulting in an almost total loss of thrust. We built our own engine test tunnel specifically to run tests ourselves and try to speed redesign of the engine.
Before it was corrected in engine design, the problem cost seven planes and seven pilots. That still sticks in my craw. The fix was very slow in coming.
The gun caused problems, too, in early firing tests. The flights required all the expertise of veteran pilots LeVier and Herman “Fish” Salmon—who flew the first production model of the F-104—as distinguished from the first experimental model flown by LeVier.
Lockheed test pilot “Fish” Salmon with F-104 Starfighter.
For LeVier it meant making the first dead-stick landing in the airplane. He was airborne on the first cannon-firing test at supersonic speed when the cannon exploded and blew a hole in the fuel cell. With a cockpit full of smoke, Tony’s first thought was to bail out. But 50 miles from home and at high altitude, he looked down and thought better of it.
“Oh, my God, I’ll freeze to death before I get to the ground,” he decided.
“I stayed in,” he recalls, “and headed back to base. As I neared the base, the engine quit. And at the very last, just before my flare for landing, I discovered that my wing flaps wouldn’t work properly. Again I wanted to bail out. It would have been down over the jetstream, though, and I knew I didn’t have a chance. So I made my flare fully expecting the airplane to snap and do everything it’s not supposed to do. Lo and behold, it flared out like a Piper Cub.”
We found out what was wrong. The cannon had exploded, and the wing flaps had failed because they lost electrical power. The dead-stick landing with an F-104 is quite an adventure, because as heavy as it is, with that small wing and no power, the pilot must be precise in putting the airplane down. LeVier had studied the figures for such a landing in advance, knew exactly what he should do, did just that, and made a perfect dead-stick landing.
In his case, Herman had no choice but to leave his airplane. He was wearing a pressure suit for gun-firing tests at supersonic speed above 50,000 feet when he felt a rush of air and his faceplate froze over. He couldn’t see any instruments but knew he did not want to disconnect the pressure suit from the aircraft at that high altitude. So he started counting to himself, waiting as long as he felt he could before raising the faceplate.
“The only thing I was really interested in was the altimeter,” he explained. “I didn’t care where anything else was.”
Herman wanted to help us get all the anwers we could to the cause of that accident so he volunteered to take sodium pentothal administered by Lockheed medical director Dr. Charles I. Barron. From his descriptions we found out exactly what happened.
An explosion from accumulated gun gas had blown open one of the landing wheel well doors, admitting a rush of upper atmospheric cold. Having been an exhibition parachute jumper in his early career allowed “Fish” to take this bailout in stride. And the gun-firing problem soon was solved.
The F-104 had been designed as an interceptor-fighter, an assignment for which the plane was very, very good. But when the NATO countries decided they wanted to build and use them, they wanted additional performance—low-level ground attack as well. We were able to double the airplane weight from the original 16,500 to 33,000 pounds without any change in wing area—which is less than 200 square feet. To do this, takeoff speed had to go way up. The airplane became a hotrod of the first order. And it was being flown in some of the world’s worst weather and terrain. Some countries—Norway, Canada, and Taiwan, for example—set safety records for fighter aircraft with it.
For special reasons, West Germany had problems with the airplane. The Germans had a very sophisticated version—very high performance combining interceptor, bomber, and reconnaissance capability in the aircraft. Later models added infrared gunsight and inertial navigation.
The trouble was that for ten years after World War II, German pilots had not had modern jet experience, especially supersonic. Nor were conditions conducive to keeping trained pilots and mechanics in the air force. While there was a great deal of press attention to the initial high accident rate with the F-104s, it went virtually unnoticed that the West Germans earlier had bought a number of F-84s and lost about 40 percent of that fleet in a very short time.
It wasn’t the equipment but the nature of the operation that was the basic problem. Finally, West Germany arranged to train their pilots at Luke Air Force Base in Arizona, where weather permitted year-round flying. Properly trained in an air force that offered a stable career, the West German pilots achieved an excellent in-service record with the airplane.
In its first year of service with the U.S. Air Force, the F-104 set some significant official records. In 1958, it recaptured the world altitude record for this country at 91,243 feet. That same year it set a new speed mark of 1,404.19 miles an hour. It established seven records for time-to-climb to altitude. In 1959, it set a new altitude record of 103,395.5 feet. The F-104 won for the Air Force, GE, and Lockheed the Collier Trophy in 1959, for the previous year’s “greatest achievement in aviation.”
The F-104 later became also an aerospace trainer for the USAF, simulating re-entry and zero gravity conditions for astronauts at the Aerospace Research Pilot School at Edwards AFB.
When the MiG-21—a high-performance Russian jet—began to appear in increasing numbers with Soviet bloc air forces, replacing the earlier MiG-17 and -19, Western Europe began to look toward a new air-superiority fighter. Their principal air defense weapons were Starfighters, Phantoms, and Lightnings. Intelligence reports and news items indicated that advanced fighter aircraft in East European countries outnumbered NATO air forces by as much as six to one.
As a follow-on airplane to the F-104 for our European allies, Lockheed developed a very practical and productive proposal. Use the expensive, proven components and systems from the F-104 but add a larger wing and new tail and increase the power for an all-around fighter. The new design promised to outmaneuver all other aircraft known to be flying including the MiG-21. We called the airplane the “Lancer.”
It was proposed with a choice of engines, the familiar GE model or a new advanced-technology design from Pratt & Whitney with a Mach 2.5 speed. We proposed to conduct development and initial flight tests at the Skunk Works in cooperation with the European engineers, and that production of the plane be programmed entirely in Europe. It could have meant millions of dollars saved in production costs as well as jobs for many thousands of people in the plants where they had turned out the F-104.
The international competition among manufacturers for sale of a new airplane to the NATO countries was intense. France’s Dassault proposed an advanced version of its Mirage F-1.
Two other U.S. manufacturers were after the business. McDonnell-Douglas offered a modification of its twin-engine Phantom, to be designated F-4F. Northrop proposed a totally new design, P-530 (F-5), which would not have been available until 1976. Lockheed promised the Lancer for service in 1973. By that year, no decision had yet been reached, and we still were trying to sell the Lancer overseas. The airplane was not proposed to the USAF, which was interested in developing a new aircraft.
As a sidelight on international sales of aircraft, on the very day that we started touring Europe with our proposals for the Lancer, one of our competitors set out on the same circuit to sell theirs. Months later we discovered that both of us had retained the same overseas marketing consultant. He was being paid by both sides.
The Air Force in this country was considering development of two new fighters, the F-14 and F-15.
Back in 1969, I questioned publicly whether these aircraft actually would be competitive with the best Russian fighters. Specifically, I said that I thought the cost of the proposed F-15 would be more expensive than necessary, that a smaller, less-expensive airplane could do the job, just as well.
Stuart Symington, former Secretary of the Air Force, then Senator from Missouri, very much wanted that F-15 contract for his state. He called in Dan Haughton, then Lockheed board chairman, and me and announced that whether we liked it or not that contract for the F-15 would be awarded to McDonnell. Kelly Johnson was not to give any more argument. Haughton was under the gun and promised that he’d see that I didn’t. I did not promise in my own right.
We, Lockheed, had made an unsolicited proposal to the USAF for an advanced, highly-maneuverable lightweight fighter that we could have had flying within a year at absolute minimum cost. We had lined up a dozen of our vendors with whom we were then working on another project, including Pratt & Whitney for the jet engines, other suppliers for armament, gunsights, wheels, tires—the whole package. It was a darn good airplane. The X-27, later designated X-29, was basically a new airplane, but it utilized the nose design of the F-104 which by that time had fired millions of rounds of ammunition. We even proposed firing tests on the first flight to prove we had a fighting airplane.
David Packard, then Deputy Defense Secretary, was very much impressed with the proposal. But Robert Seamans, Jr., Air Force Secretary, did not like the idea of buying a fighter in that manner. He preferred the conventional method—an experimental model first, production plans later.
I disagreed with the USAF on procurement policy.
“This airplane is not so advanced that you cannot develop the ‘X,’ the experimental airplane, into a production prototype,” I argued. “I don’t want to draw a line on paper that does not consider production. Why go to double prototyping?”
We came close to receiving a contract for that airplane, but what eventually killed any prospect for our producing the lightweight fighter were the financial problems that Lockheed encountered in 1971—first, losses from several fixed-price contracts for the U.S. military, then the threatened loss of the company’s new L-1011 commercial transport program with the unexpected bankruptcy of Great Britain’s Rolls-Royce, manufacturer of the engine. The very future of Lockheed was in question, and the Air Force reasoned understandably that they should not risk awarding the contract to the company.
Our proposal did, I believe, result in the USAF’s eventual design competition for a lightweight fighter. The plane they got at least ten years later, after double prototyping by General Dynamics and at nearly three times the cost, was comparable in performance. That was the F-16.
If the military would spend one or two percent of the cost of developing an experimental airplane in planning production at the same time, it would come back in savings many, many times over.
The Skunk Works method of developing the F-104, for example, could save a considerable amount of money if applied to procurement of new aircraft.
On that airplane, every time we released an engineering drawing to our manufacturing director, Art Viereck, to build a part for the experimental airplane, we also released it to a group of production engineers with these instructions: “Find every alternative way of making this, ruling out adverse effect on drag, maintenance, or cost. You can affect them all favorably.”
When we finished building the prototype, we had a thick report on how to build a production model. We sat down for three days with that book to choose the best way to build the airplane from every point of view. It saved from $10,000 to $20,000 for every airplane built, by my estimate. Total production worldwide was about 2,500.
Kelly with the first JetStar, which critics declared would never attain commercial success.
Costs must be considered. Aircraft are getting to be so expensive they hardly are worth it for what they can do. With the price of fighter aircraft now running more than $30 million per plane with all the equipment, not including pilot costs, I can foresee the day when the fighter pilot will be on the ground, flying an unmanned fighter with a missile in it. With the latest electronic advances, I think this can be done remotely at a great saving in aircraft costs—and, of course, great saving in manpower, to say nothing of the greater safety for the pilot. It’s worth considering.