by Orr Kelly
Ideally, the Hornet should have had a roll rate of at least 180 degrees a second. In other words, it should have been able to do a complete 360-degree roll in two seconds. What Field found was that this strike-fighter had a ponderous roll rate of only about eighty degrees a second. It took the plane all of four and a half seconds to do what it should have done in two.
As Field’s experience demonstrated, the new plane’s roll rate was too slow for it to be an effective bomber. But the agility with which a plane can roll is also vital in a dogfight. If an enemy gets on a pilot’s tail, and the pilot can roll more quickly than his pursuer, even by a fraction of a second, he can spoil his opponent’s aim and perhaps even get in a shooting position himself. A pilot flying a plane as sluggish as the F/A-18 against a skilled adversary would probably have only moments to live.
If the Hornet’s lethargic roll rate could not be fixed, it wouldn’t be able to do either job for which it was intended. And even if it could be fixed, the fix might be so expensive that the Hornet would be doomed politically.
When the engineers got back to their computers and wind tunnels, they found something very weird indeed. There was a basic flaw in the design of the wing that permitted it to bend during abrupt maneuvers in the dense air at low altitudes. This was partly due to one of the compromises involved in attempting to build a plane that would be both a fighter and a bomber, and partly due to a failure to understand how some of the new materials used in the wing would perform in flight.
The original design called for a wing that would be as thin as possible to make it a good fighter, but also as strong as possible so it could carry a heavy bomb load. In their effort to build a very thin, very strong wing, the designers used far more composite materials, rather than the traditional aluminum and titanium, than had been used in previous planes. Much of the wing is made up of a lightweight honeycomb material covered by graphite epoxy panels. The result is a wing four inches thick at its thickest point. As one pilot says, “We fly on razor blades.”
At the time the design work was done in the early and mid-1970s, industry was still learning about composite materials and how they work. One thing that was not understood clearly is the flexibility of these new materials. The unexpected bending of the wing of the F/A-18 taught an important lesson. The manufacturers of golf clubs learned the same lesson when they first used these new materials. The graphite was plenty strong enough for the shaft of a golf club. But when the head of the club hit the ball, the shaft twisted almost like the wing of the Hornet.
Normally, if a pilot wants to roll his plane, he moves his control stick in the desired direction. That activates two flaps, known as ailerons, attached by hinges to the rear of the wings at the outer ends. One aileron is pressed down and that tends to raise the wing. The other aileron is pressed up and that lowers the wing, causing the plane to roll in that direction.
In the F/A-18, the wings proved so flexible that, when the ailerons were activated, the wings bent in the opposite direction, in effect tending to make the plane try to roll the wrong way.
Fixing the roll rate proved to be a difficult and expensive affair. It involved making the ailerons bigger, adding several hundred pounds of composite fabric material to stiffen the wings, and modifying the computer instructions to involve other adjustable parts of the wings in helping the ailerons to roll the plane quickly and crisply.
The result of all this effort was a roll rate so high that a pilot could actually damage the plane by snapping it around too fast. To keep that from happening, the computers were programmed to prevent the plane from rolling faster than 220 degrees a second—a 360-degree roll in a second and a half.
Serious as the problem was, it was not as bad as it might have been. Almost as soon as Field reported his brush with disaster, word spread in the navy and on Capitol Hill that McDonnell Douglas had recommended the plane be given an entirely new wing of a different design. That was not true. If it had been, the entire F/A-18 program would very likely have been killed.
Like all very big Pentagon weapons programs, the Hornet was threatened repeatedly with a cutoff of funds. Almost every year, such an effort came to a vote in one or both houses of Congress. All of those efforts failed. But, in contrast to other big ticket items, the F/A-18 suffered more death threats more often, and it really did hover on the brink of death on a number of occasions.
One of the most serious threats came from the very highest levels of the navy itself late in 1977. As final decisions were being made on the budget that would go into effect the following fall—the budget that would provide funds to move from development into production of the F/A-18—Navy Undersecretary R. James Woolsey wrote an extraordinary five-page memo to Defense Secretary Harold Brown, urging him to cancel the F/A-18 program outright and fill the carrier decks with F-14 fighters and A-7 attack planes.
“Although the F/A-18 provides some capabilities that we would like to have, in light of the fiscal constraints we surely face, it is far preferable to terminate it than to suffer the disproportionate loss in other aviation and non-aviation programs which otherwise seem inevitable,” Woolsey wrote.
The navy’s top admirals were careful not to take a public stance on Woolsey’s proposal. If the F/A-18 remained in the budget, they would have to defend it on the Hill and support its use in the fleet. But many of the admirals had favored a fighter force made up of F-14s all along, so Woolsey’s recommendation was not entirely unwelcome.
Although Woolsey’s memo had some harsh words for the F/A-18 in comparison with the F-14, he later insisted that the recommendation to kill the program stemmed from the money crunch rather than any vendetta against the Hornet.
Extraordinary as Woolsey’s proposal was, the outcome was perhaps even more extraordinary. Instead of grasping the navy’s offer to sacrifice the biggest item in its budget, Brown accepted the advice of Russ Murray, his assistant for program analysis and evaluation. Although once a sharp critic of the F/A-18, Murray had gradually become convinced of its value. In a four-page memo of his own, he convinced Brown the plane was worth building. Brown okayed production of the first five Hornets—and cut F-14 production nearly in half.
One can only wonder whether Brown, trying to make ends meet on a tight budget, would have made the same decision to go from development to production if he had been able to see a few years into the future.
Almost as soon as production began, both McDonnell Douglas and Northrop found it was taking longer to assemble a plane than had been expected, and the cost started to go up. The problem at Northrop was truly alarming. According to the original plan, it should have required 67,500 hours of labor for Northrop to assemble its rear section of the plane. On the assembly line, however, the work actually required as much as 147,000 hours.
Thomas Burger, manager of the F/A-18 program for Northrop, blames much of the trouble on the fact that the central portion of the fuselage—a heavy, dense structure full of complicated plumbing for the fuel system—is difficult and time-consuming to assemble.
“We’ve got some of the hard stuff,” he says. “The central fuselage is the worst part of the plane because it has all the heavy structural stuff, the critical fasteners, the tightest density of stuffing things in.”
At the heart of the central section of the fuselage, right behind the cockpit, are three large metal bulkheads. Instead of a single wing extending right through the fuselage, there is a separate wing on each side and they are attached to each of the three bulkheads with two-and-a-half-inch diameter titanium pins, one at the top, one at the bottom, for a total of six pins for each wing. The three bulkheads and those dozen pins must take all the stress of gravity during the plane’s violent maneuvers.
Production had barely begun when cracks developed in one of the bulkheads in a test rig. This meant work on the plane was delayed while the design was reworked and the bulkheads strengthened. Burger says a conscious decision was also made to incorporate about 200 improvements into the first plane on the asse
mbly line and this caused further delay.
As reports of the rapidly soaring labor hours at Northrop reached St. Louis and Washington, teams of experts were sent to help out. John Capellupo, who had worked with Northrop earlier on development of the plane and who was later to head McDonnell Douglas’s F/A-18 operation, was dispatched to Hawthorne with a large group of McDonnell Douglas engineers.
To both the navy and McDonnell Douglas, the prime contractor, there were two critical issues: First, if planes could not be produced on schedule, the flight test program would be delayed. Second, if the time and labor involved could not be reduced, there could be a major increase in the cost of the plane. The possibility that the entire program might be cancelled was very real.
To the experts from St. Louis, it seemed clear that Northrop’s problem was not only the difficulty of assembling the dense central section of the fuselage, but also a lack of recent experience in setting up a major production program. Northrop had last gone into production in a big way on the F-5, and that had been many years before. McDonnell Douglas, on the other hand, had had a steady stream of major production contracts, including the F-4, the F-15, and the AV-8.
Lenox took a worried look at the situation and concluded that Northrop’s managers had failed to realize the F/A-18 was much more dense, much more complex than the F-5, and that they “were too damn proud to admit it.”
The influx of outside experts, insisting that Northrop change its way of doing business, was a direct affront to the professional pride of the Northrop managers, and the cause of a good deal of resentment. The ill-feeling was intensified by a bitter legal dispute pitting Sandy McDonnell and Tom Jones in a fight over foreign sales.
Even after losing the air force competition for the lightweight fighter and ending up playing second fiddle to McDonnell Douglas on the navy plane, Jones still pursued his dream of selling a new fighter to the many countries whose F-5s and other planes were in need of replacement. The two companies agreed that Northrop would be free to develop and sell a lighter, less expensive, land-based version of the Hornet called the F-18L.
Jones went vigorously to work trying to market the F-18L. He even obtained the blessing of the Joint Chiefs of Staff for a sale to Iran, but the White House blocked the deal because the land-based version of the plane was not in use by an American service.
Tom Jones tried to get around this problem. In a “Dear Davy” letter to Gen. David C. Jones, then chief of staff of the air force, on 22 February 1977, he offered to sell the F-18L to the air force at a remarkably low fixed price: $4.77 million apiece for any quantity between 600 and 1,000 aircraft, and $3.8 million apiece between 1,000 and 1,500 planes. The air force, already committed to the F-15 and F-16, passed up the offer.
The F-18L was, of course, in competition with the F/A-18 in the foreign market, and Northrop suspected McDonnell Douglas of working with the navy to sell the Hornet at the expense of its land-based sibling. Any foreign sales would help reduce the cost of the plane to the navy. On 26 October 1979, Northrop filed a suit against McDonnell Douglas, accusing its partner of fraud, economic coercion, unfair competition, industrial espionage, and antitrust violations. It asked for more than $400 million in damages. McDonnell Douglas filed a counter suit accusing its partner of similar offenses. It upped the ante to more than $2.3 billion in damages. The dispute dragged on for nearly six years in an atmosphere of suspicion and animosity. A series of navy officers tried to push Sandy McDonnell and Tom Jones to agree on a settlement, but they resisted. At one point, they were forced to sit down together with an admiral acting as arbitrator, and they actually reached an agreement. But it fell apart almost as soon as they left the room, apparently because of a lack of trust between the two men.
Finally, a settlement was reached on 8 April 1985, after Navy Secretary Lehman let the companies know the government was not about to pay their rapidly mounting legal bills. McDonnell Douglas agreed to pay Northrop $50 million in return for the uncontested right to sell the F/A-18 wherever it could. Northrop would, of course, share in these sales, but that was not the same as producing its very own plane for the export market.
While the case was being contested, it was obvious to those in the know that production problems, plus the rise in inflation, were pushing the cost of the plane upward, although no one seems to have realized—or dared to imagine—how high costs threatened to go. Adm. Wes McDonald, who was DCNO for air warfare in the late 1970s, says he didn’t sense any dramatic increase, “just continuing soft growth.”
Late in 1979, as required by law, the navy had to calculate the cost of its major programs and report the numbers to Congress. That forced the navy to face up to a startling increase in the projected cost of the entire program. Early in 1980, the nervous admirals appeared on Capitol Hill to report the bad news: the program was projected to balloon by $5.1 billion.
Lenox broke it down this way: The total cost of purchasing 1,366 aircraft, with deliveries through 1991, would be $26.9 billion. Of the $5.1 billion jump from the previous year’s estimate, he blamed $3.5 billion on inflation and $1.6 billion on increases caused by the troubles on the assembly lines.
Congressman Chappell, one of the most persistent critics of the F/A-18, seized on the report: “There is no question, of course, that the program cost has just gone out of sight.”
“No question,” a glum Admiral McDonald agreed.
One of the interesting characteristics of the F/A-18 program is that each of the program managers seems to have left at least one big, and often unexpected, problem to his successor. In this case, Lenox was reassigned early in 1980 to take over as commander of the shore facilities of the Pacific Fleet’s attack squadrons at Lemoore NAS in California. In that position, he would oversee the introduction of the Hornet into the fleet. He passed on the program manager’s job to Capt. John C. Weaver, and, for the next three years, one of Weaver’s biggest jobs would be to deal with questions about the cost of the plane and try to satisfy Congress that he was not playing a shell game with appropriated funds.
The appointment of Weaver was part of the navy’s effort to correct the problem so quickly identified by Willoughby, the expert on reliability: that program managers learned on the job and then were reassigned. Weaver was a fighter pilot, but he also held advanced degrees in aeronautical engineering and electronics, and he had spent a decade and a half working on some of the navy’s most important development programs, including the F-14 and the Phoenix missile. He was not just a fighter pilot brought in from the fleet for on-the-job training as a program manager.
One of Weaver’s most urgent tasks was to try to reduce production hours. Although it took longer than everyone had hoped, Northrop did manage to bring about a dramatic decrease in the time needed to build its part of the plane. Northrop’s Burger says proudly: “The way we went down the learning curve will probably go down in the annals of aircraft assembly as unbelievable.”
As the Navy scrambled to complete development and go into production of the Hornet in the face of the high inflation rates of the late 1970s and early 1980s, it moved more than 300 million dollars from the weapons and spare parts accounts into the accounts needed to pay the bills coming due from the aircraft factories. Later, Congress had to appropriate more money to pay for the spare parts and weapons. To the critics, it seemed a clear case of paying twice for the same thing. The House Defense Appropriations Subcommittee and the Pentagon inspector general’s office both issued sharply worded reports questioning the legality of these transfers.
The navy’s actions had been strongly influenced by Vice Adm. Richard Seymour, who was commander of NAVAIR from April 1980 until the summer of 1983. A veteran attack pilot, Seymour had served in Washington since the mid-1970s, learning his way around the political no-man’s land between the Pentagon and Capitol Hill.
“A big thing I learned is that if Congress says something in a report, it does not bind you to do it,” Seymour says. “If they put it in a law, you have to do it. I us
ed that a number of times later in my career. They used to write all this stuff in reports. I would sometimes ignore it at my own risk. If the sensible thing to do was not to do what they said to do, we would not do it. They clearly did not want me to do something stupid.”
Seymour’s philosophy guided the navy as it shuffled around the money to keep the F/A-18 on track. Weaver was held in his program manager’s job for an extra year to answer questions from the Hill about the funds transfers. In the end, he convinced skeptical members of Congress that what had been done was legal and that the navy had avoided doing “something stupid.”
During this difficult period, every technical problem with the plane cost money and only added to the task Weaver faced in keeping the budget under control.
The year he took over, 1980, when a number of bad things happened to the F/A-18, was a particularly dicey period. There was the roll rate issue and the engine problem revealed by the crash in the English countryside, both matters of deep concern. And then, on 14 November, Lt. Travis Brannon, a pilot brought in from the fleet to conduct operational testing of the F/A-18 at Patuxent, accidentally discovered that the Hornet would do something it was not supposed to do.
Brannon and another pilot were engaged in a simulated dogfight when Brannon racked his plane around in a sharp turn with the nose pointing upward at a steep angle. Suddenly, it felt as though someone had jammed on the brakes. The plane went into a flat spin and began to drop toward the Chesapeake Bay, four miles below, at 500 feet a second.
Under the rules, Brannon should have ejected at 10,000 feet, but he fought to control the plane for thirty-four seconds before finally abandoning the craft at 5,000 feet. He descended safely and was quickly rescued. But the plane pancaked into the bay.
