On the morning of June 23, 2014, an F-35 burst into flames just moments before its pilot was set to take off on a routine training mission. He heard a loud bang and felt the engine slow as warning indicators began flashing “fire” and other alerts signaled that systems in the plane were shutting down. Witnesses at Eglin Air Force Base near Pensacola, Fla., reported seeing the pilot escape from the cockpit and run away from the fighter jet, which was engulfed in thick plumes of black smoke. It was the first major mishap involving a F-35 Joint Strike Fighter, and it couldn’t have happened at a worse time.

In less than a month, the F-35, America’s high-profile next-generation fighter jet, was poised to make its international debut in Britain at Farnborough Airshow, the second-largest event of its kind in the world. Officials from the Pentagon and the aircraft’s manufacturer, Lockheed Martin, had eagerly anticipated the opportunity to show off a working, flying F-35 after a decade of delays and spiraling cost overruns.

The F-35 initiative is the Defense Department’s most expensive weapons program ever, expected to cost taxpayers more than $1 trillion over its 60-year lifespan. It’s also the United States military’s most ambitious international partnership, with eight other nations investing in the aircraft’s development. Its advocates promised that the jet would be a game-changing force in the future of war — so much was riding on its success that a program cancellation was not an option. And yet for years it seemed as if the F-35 might never make it beyond its development phase.

Christopher Bogdan, the Air Force lieutenant general in charge of the program at the time of the fire, received a call about the incident within the hour. His first reaction was relief that it had been detected before takeoff, a stroke of good fortune that allowed the pilot to escape uninjured. “If that engine problem would have occurred 30 seconds, 60 seconds, two minutes later, that airplane would have been airborne,” Bogdan said in a recent interview. “Heaven knows what could have happened then.”

An investigation of the incident determined that a fan blade in the jet’s engine had overheated from friction and cracked, throwing off fragments of metal that punched through the fuselage, severed hydraulic and fuel lines and ignited a spray of jet fuel. Officials couldn’t guarantee that other F-35s wouldn’t have the same problem, and they didn’t want to risk a potentially catastrophic fire during a trans-Atlantic flight. The F-35 never made it to Farnborough that year, and the public-relations coup that Pentagon and Lockheed officials had hoped for turned into another round of ammunition for the plane’s critics.

It was one more bad news story for a controversial program that had been dogged by bad news.

Slowly, though, the program and its reputation have improved over the ensuing five years. Lockheed has now delivered more than 400 planes to American and foreign militaries, and the unit cost per aircraft has dropped significantly. In 2018, the F-35 completed its first combat operation for the Marine Corps in Afghanistan. The Air Force used it for airstrikes in Iraq about six months later. Later this year or in early 2020, the F-35 will go into full-rate production, with Lockheed expected to churn out 130 to 160 or more planes per year, a huge step up from the 91 planes delivered in 2018. That production milestone will be a symbolic turning point for the program, evidence that major problems that plagued the Joint Strike Fighter in the past are now history.

Yet even as the program plows forward, unresolved technical issues have continued to emerge. In June, my colleagues and I at Defense News reported that the plane still faced at least 13 severe technical deficiencies during operational testing, including spikes in cabin pressure, some rare instances of structural damage at supersonic speeds and unpredictability while conducting extreme maneuvers — all problems that could affect the pilot’s safety or jeopardize a mission’s success. At the same time, the F-35s already delivered to squadrons have introduced new complications: On military bases around the United States, the high cost of operating the aircraft, a shortage of spare parts and a challenging new approach to updating the jet’s crucial software code have program officials and military leaders urgently looking for solutions. Still, they assure the public that nothing will prevent the program from moving forward. It’s a stance that breaks with the advice of the Government Accountability Office, which advised that all serious problems should be resolved before transitioning to full-rate production.

With the critical problems that once dominated both headlines and congressional hearings seemingly resolved, the F-35 may not be the high-profile problem child it once was. But the Pentagon’s efforts to play down new complications raise questions about whether America’s most controversial warplane is actually ready to move into its next phase and what kind of new problems might surface in that transition.

The Joint Strike Fighter program was conceived in the 1990s as the most ambitious aircraft development effort in the Defense Department’s history. One company would oversee design and production of three different versions of an aircraft that could be operated by the United States Air Force, Navy and Marine Corps as well as America’s allies, who would help offset the development costs. The project would result in a technologically superior plane that would be manufactured in such large quantities that the jets would cost no more than the older planes it would replace.

The jet was to replace a wide variety of the American military’s combat aircraft, including the Air Force’s F-16 Fighting Falcon and A-10 Thunderbolt II, the Marine Corps’s AV-8B Harrier and the F/A-18 Hornet operated by the Navy and Marine Corps. All of the services’ requirements for new fighter jets were combined into one program, which would be awarded to a single contractor. Like the Air Force’s F-22, this new fighter jet needed to be stealthy and able to fly at supersonic speeds. To meet the needs of the Marine Corps, it needed to land vertically on ships, while the Navy version would need larger wings and different landing gear so it could take off from and land on aircraft carriers.

“If you were to go back to the year 2000 and somebody said, ‘I can build an airplane that is stealthy and has vertical takeoff and landing capabilities and can go supersonic,’ most people in the industry would have said that’s impossible,” said Tom Burbage, Lockheed’s general manager for the program from 2000 to 2013. “The technology to bring all of that together into a single platform was beyond the reach of industry at that time.”

Lockheed Martin thought otherwise. In 2001, the defense company’s model — then called the X-35 — won against Boeing’s X-32 after both companies demonstrated working prototypes of a stealth fighter capable of hovering and vertical landings.

There soon turned out to be an essential flaw in the grand plan for a single plane that could do everything. Design specifications demanded by one branch of the military would adversely impact the F-35’s performance in another area. “It turns out when you combine the requirements of the three services, what you end up with is the F-35, which is an aircraft that is in many ways suboptimal for what each of the services really want,” said Todd Harrison, an aerospace expert with the Center for Strategic and International Studies. “It is much more expensive than originally envisioned, and the three versions of the plane actually don’t have that much in common.”

But early in the program, Lockheed Martin began construction with glowing optimism. The company decided to build the Air Force’s F-35A first because it was considered the simplest model, then move on to the difficulties of the F-35B short-takeoff and vertical-landing version and then the F-35C, which can land on an aircraft carrier — a decision that turned out to be a mistake. Once Lockheed’s engineers proceeded with the more demanding design of the F-35B, they found that their initial weight estimates were no longer accurate and the B model was on track to be 3,000 pounds too heavy to meet specifications. The company was forced to begin an extensive redesign project that added an 18-month delay to the program.

Later, serious problems resulted from starting production while the aircraft was still under development, a process the Pentagon calls concurrency. The strategy was meant to allow the services to begin flying their F-35s sooner. Instead, F-35s started rolling off the production line with unresolved technical problems, forcing the Pentagon to continually retrofit even newly built jets.

In 2010, the ballooning costs — which put the cost per plane more than 89 percent over the baseline estimate — triggered a breach of the Nunn-McCurdy Act, a law that forces the Pentagon and Congress to evaluate whether to cancel a troubled program. But because the F-35 was intended to replace so many legacy fighter jets, military leaders essentially had no choice but to keep going.

One factor that kept sending the F-35 program off course was the level of control Lockheed exerted over the program. The company produces not only the F-35 itself but also the training gear for pilots and maintainence technicians, the aircraft’s logistics system and its support equipment, like carts and rigs. Lockheed also manages the supply chain and is responsible for much of the maintenance for the plane. This gave Lockheed significant power over almost every part of the F-35 enterprise. “I had a sense, after my first 90 days, that the government was not in charge of the program,” said Bogdan, who assumed oversight as the program’s executive officer in December 2012. It seemed “that all of the major decisions, whether they be technical, whether they be schedule, whether they be contractual, were really all being made by Lockheed Martin, and the program office was just kind of watching.”

Bogdan particularly worried that Lockheed had too much control of the government’s test flights. The company was allowed to manage the test program and had the power, for example, to defer more challenging tests until later. In past programs, the government had controlled testing and had aimed to find any difficult, high-priority problems early, so they could be addressed as soon as possible. Bogdan also argued that the Pentagon’s program office was not transparent enough in letting the military services know how their money was being spent. Because Lockheed was not required to report its financials in detail, the program office itself did not have a clear picture of exactly how much an F-35 truly cost and how the money was being used.

Costs and complications were spiraling. Someone needed to intervene before the Defense Department lost control entirely, Bogdan thought. In September 2012, when he was the program’s second-in-command, he took to the lectern at the Air Force’s largest conference and said something that had not been publicly acknowledged before: The relationship with Lockheed was the worst arrangement he had ever seen between the Pentagon and a defense contractor. The audience was shocked. At military conferences and trade shows, Defense Department officials and their contractors typically boast about their collaborative efforts. Instead, Bogdan publicly shamed the defense behemoth, criticizing the lagging production time and skyrocketing costs.

At that point, there was a lot to rebuke. The Pentagon had restricted the F-35 from flying near thunderstorms after flight tests revealed that its lightning-protection system was deficient. That became easy fodder for skeptics, given the plane’s designation as the F-35 Lightning II. The jet’s cutting-edge helmet display, which melds imagery from the F-35’s multiple cameras and sensors into a single picture, didn’t work properly, with pilots experiencing a jittery, delayed video feed. And the jet’s software development had lagged behind schedule, leaving pilots stuck with an interim version that allowed only for basic training. When Bogdan was promoted to the top post as the program’s executive officer, a rule change empowered him to stay on past the previous two-year limit, and he let Lockheed know that he wasn’t going anywhere until the F-35’s major difficulties were behind it. The Pentagon and the contractor got to work on correcting the jet’s technical issues, one by one.

Within a year, Bogdan was saying publicly that the relationship with Lockheed was improving and that the contractor was making progress in addressing the F-35’s problems, albeit more slowly than he would have liked. The program office and Lockheed had figured out some ways to cut the cost of manufacturing the fighter. More flight testing was happening. The lightning-protection system was redesigned in 2014, and the F-35 can now fly in bad weather. A series of hardware and software changes to the helmet have solved the image-quality problems.

Still, as problems were fixed, new ones surfaced. In 2015, the program office found that the plane’s ejection seat could cause serious neck injuries to lightweight pilots, prompting the Air Force to ban pilots under 136 pounds from flight operations until a fix was implemented in 2016. Perhaps most damning was a 2015 report by War Is Boring, a well-read military blog, citing a document by an Air Force test pilot that asserted that the F-35 could not defeat the 1970s-era F-16 in aerial combat. If that was true, the staggeringly expensive high-tech jet was already obsolete before it would ever take to the skies in wartime. It emerged that the assessment was provisional and incomplete. The pilot had based his judgment on a single day’s mock aerial battle between an F-16 and an F-35 that temporarily had limited maneuverability and restricted performance because it had an early version of its software package. Nevertheless, the characterization of the F-35 as an overpriced but mediocre dogfighter has haunted it ever since.

Though the F-35’s bad reputation among the public has persisted, the military has grown increasingly confident in the jet’s capabilities as problems have been eliminated and additional weapons and software have made the aircraft more capable in a fight. As more pilots log more flight time, they have raved about the F-35’s performance and technological advances.

The Marine Corps, which began normal flying operations with the jet in 2015, became the first military branch to fly the F-35 in combat when it used the jets for airstrikes in Afghanistan last year. Both the Air Force and Navy are also now operating their own F-35s. In 2017, during the F-35A’s first outing in Red Flag, the Air Force’s largest training exercise for aerial warfare, the jet killed 20 aircraft for each F-35 shot down in simulated combat. In April, Air Force pilots took that training and put it into practice for the first time, using the F-35 in an airstrike against ISIS in Iraq.

Pressure from Congress also helped shape the F-35 into a more successful program. In 2016, Senator John McCain of Arizona labeled the program a “scandal and a tragedy with respect to cost, schedule and performance.” Recognizing that it was nearly impossible to cancel the program, McCain nevertheless aimed to hold it accountable for its repeated setbacks. He moved forward with a carrot-and-stick approach, approving additional funding for the F-35 as chairman of the Senate Armed Services Committee while at the same time regularly grilling Defense Department officials during congressional hearings.

Since McCain’s death last August, no other lawmaker has exacted that level of scrutiny. That’s by design, said Dan Grazier of the Project on Government Oversight, a watchdog organization that has repeatedly criticized the aircraft. “It’s no accident that there are more than 1,500 suppliers for the F-35 program, and they’re spread out to almost every state,” he said. “That means that there’s basically a veto-proof constituency bloc on Capitol Hill for the F-35 program, so it becomes very difficult for members of Congress to really criticize this program.” Legislators looking to relieve pressure on the program — and argue for increasing the number of aircraft purchased annually — have been able to point to measurable progress in the F-35’s performance and management. But that leniency comes at the expense of taxpayers’ dollars, say critics like Grazier, who want to see lawmakers do more to push for cost savings.

Within the last few years, the Defense Department has corrected hundreds of Category 1 deficiencies, the label the Pentagon gives to serious technical problems including ones that could affect safety or mission effectiveness, and many of the F-35’s most visible problems have been resolved. However, as recently as June, at least 13 Category 1 deficiencies were still on the books. “Each is well understood, already resolved or on a near-term path to resolution,” Lockheed said in a statement. Vice Adm. Mat Winter, who led the Pentagon’s F-35 program office until July, maintained that the 13 deficiencies were not ones so serious that they could cause loss of life or aircraft. The department has a plan to fix all but two of the issues — the two have occurred only once during flight tests and are considered anomalous — and according to Winter, the problems will not affect the Pentagon’s plan to move to full-rate production.

The F-35 was in the news again in July when the White House decided to expel Turkey from the program when that contentious ally refused to give up its plan to simultaneously acquire an advanced air defense system from Russia. That new partnership with Moscow presented a risk that technological secrets from F-35s in Turkey could make their way to Russia.

The program’s international reach, meanwhile, is only becoming bigger. Ten international partners and customers have committed to buying the jet, and eight of them have received their first F-35s. Israel was the first country to use the fighter in combat, announcing in May 2018 that it had used the F-35 in two separate airstrikes on undisclosed targets in the Middle East.

The cost of the plane continues to decrease as the growth in sales reduces the unit cost per jet, with the price of a conventional F-35A — the variant purchased by most international customers — falling to $89.2 million in 2018. Back in 2006, the first batch cost $241.2 million per plane. In June, Lockheed and the Pentagon announced a handshake deal that would see the price of the F-35A drop to the long-awaited level of $80 million, about equal to older planes like the F/A-18 Super Hornet. “This is going to be the first fighter jet produced in the thousands for a very long time,” said Richard Aboulafia, an aerospace analyst with the Teal Group. “None of this is stoppable. It will be remembered, as the smoke clears, as something that worked far better than critics thought it would, but something you’d never, ever want to do again.”

Hill Air Force Base in Utah, home to the Air Force’s first operational F-35s, is tasked with preparing pilots for combat and has some of the highest availability rates among all installations that fly the aircraft — a key measure that the service uses to track the proportion of aircraft that are operational and ready to fly. But long turnaround times for some maintenance tasks has meant that about 30 percent of the squadron’s aircraft are grounded at any given time. At some bases that fly the older models, the availability rate is far lower: Sometimes more than 60 percent of their F-35s are not operable. In 2017 and 2018, only about half of the F-35 fleet was available to fly at a given time, with the rest down for maintenance.

A major cause of F-35s sitting idle on the ground is a shortage of replacement parts. Lockheed and Defense Department officials have blamed each other for the problem, and there probably is plenty of fault to go around. The planes require a dizzying number of components sourced from different suppliers, and replacement parts are not getting to the flight line when they are needed. For instance, F-35s have encountered problems with the canopy, the glass enclosure that protects the cockpit, and jets can sometimes wait for a year to receive the necessary repair.