Amos-6, revisited SpaceX pushed “sniper” theory with the feds far more than is publicly known "It came out of nowhere, and it was really violent." Eric Berger – May 5, 2025 7:00 am | 9 The Amos 6 satellite is lost atop a Falcon 9 rocket. Credit: USLaunchReport The Amos 6 satellite is lost atop a Falcon 9 rocket. Credit: USLaunchReport Story text Size Small Standard Large Width * Standard Wide Links Standard Orange * Subscribers only   Learn more The rocket was there. And then it decidedly was not. Shortly after sunrise on a late summer morning nearly nine years ago at SpaceX's sole operational launch pad, engineers neared the end of a static fire test. These were still early days for their operation of a Falcon 9 rocket that used super-chilled liquid propellants, and engineers pressed to see how quickly they could complete fueling. This was because the liquid oxygen and kerosene fuel warmed quickly in Florida's sultry air, and cold propellants were essential to maximizing the rocket's performance. On this morning, September 1, 2016, everything proceeded more or less nominally up until eight minutes before the ignition of the rocket's nine Merlin engines. It was a stable point in the countdown, so no one expected what happened next. "I saw the first explosion," John Muratore, launch director for the mission, told me. "It came out of nowhere, and it was really violent. I swear, that explosion must have taken an hour. It felt like an hour. But it was only a few seconds. The second stage exploded in this huge ball of fire, and then the payload kind of teetered on top of the transporter erector. And then it took a swan dive off the top rails, dove down, and hit the ground. And then it exploded." The dramatic loss of the Falcon 9 rocket and its Amos-6 satellite, captured on video by a commercial photographer, came at a pivotal moment for SpaceX and the broader commercial space industry. It was SpaceX's second rocket failure in a little more than a year, and it occurred as NASA was betting heavily on the company to carry its astronauts to orbit. SpaceX was not the behemoth it is today, a company valued at $350 billion. It remained vulnerable to the vicissitudes of the launch industry. This violent failure shook everyone, from the engineers in Florida to satellite launch customers to the suits at NASA headquarters in Washington, DC. As part of my book on the Falcon 9 and Dragon years at SpaceX, Reentry, I reported deeply on the loss of the Amos-6 mission. In the weeks afterward, the greatest mystery was what had precipitated the accident. It was understood that a pressurized helium tank inside the upper stage had ruptured. But why? No major parts on the rocket were moving at the time of the failure. It was, for all intents and purposes, akin to an automobile idling in a driveway with half a tank of gasoline. And then it exploded. This failure gave rise to one of the oddest—but also strangely compelling—stories of the 2010s in spaceflight. And we're still learning new things today. The “sniper” theory The lack of a concrete explanation for the failure led SpaceX engineers to pursue hundreds of theories. One was the possibility that an outside "sniper" had shot the rocket. This theory appealed to SpaceX founder Elon Musk, who was asleep at his home in California when the rocket exploded. Within hours of hearing about the failure, Musk gravitated toward the simple answer of a projectile being shot through the rocket. This is not as crazy as it sounds, and other engineers at SpaceX aside from Musk entertained the possibility, as some circumstantial evidence to support the notion of an outside actor existed. Most notably, the first rupture in the rocket occurred about 200 feet above the ground, on the side of the vehicle facing the southwest. In this direction, about one mile away, lay a building leased by SpaceX's main competitor in launch, United Launch Alliance. A separate video indicated a flash on the roof of this building, now known as the Spaceflight Processing Operations Center. The timing of this flash matched the interval it would take a projectile to travel from the building to the rocket. A sniper on the roof of a competitor's building—forget the Right Stuff, this was the stuff of a Mission: Impossible or James Bond movie. At Musk's direction, SpaceX worked this theory both internally and externally. Within the company, engineers and technicians actually took pressurized tanks that stored helium—one of these had burst, leading to the explosion—and shot at them in Texas to determine whether they would explode and what the result looked like. Externally, they sent the site director for their Florida operations, Ricky Lim, to inquire whether he might visit the roof of the United Launch Alliance building. SpaceX pursued the sniper theory for more than a month. A few SpaceX employees told me that they did not stop this line of inquiry until the Federal Aviation Administration sent the company a letter definitively saying that there was no gunman involved. It would be interesting to see this letter, so I submitted a Freedom of Information Act request to the FAA in the spring of 2023. Because the federal FOIA process moves slowly, I did not expect to receive a response in time for the book. But it was worth a try anyway. No reply came in 2023 or early 2024, when the final version of my book was due to my editor. Reentry was published last September, and still nothing. However, last week, to my great surprise and delight, I got a response from the FAA. It was the very letter I requested, sent from the FAA to Tim Hughes, the general counsel of SpaceX, on October 13, 2016. And yes, the letter says there was no gunman involved. However, there were other things I did not know—namely, that the FBI had also investigated the incident. The ULA rivalry One of the most compelling elements of this story is that it involves SpaceX's heated rival, United Launch Alliance. For a long time, ULA had the upper hand, but in recent years, it has taken a dramatic turn. Now we know that David would grow up and slay Goliath: Between the final rocket ULA launched last year (the Vulcan test flight on October 4) and the first rocket the company launched this year (Atlas V, April 28), SpaceX launched 90 rockets. Ninety. But it was a different story in the summer of 2016 in the months leading up to the Amos 6 failure. Back then, ULA was launching about 15 rockets a year, compared to SpaceX's five. And ULA was launching all of the important science missions for NASA and the critical spy satellites for the US military. They were the big dog, SpaceX the pup. In the early days of the Falcon 9 rocket, some ULA employees would drive to where SpaceX was working on the first booster and jeer at their efforts. And rivalry played out not just on the launch pad but in courtrooms and on Capitol Hill. After ULA won an $11 million block buy contract from the US Air Force to launch high-value military payloads into the early 2020s, Musk sued in April 2014. He alleged that the contract had been awarded without a fair competition and said the Falcon 9 rocket could launch the missions at a substantially lower price. Taxpayers, he argued, were being taken for a ride. Eventually, SpaceX and the Air Force resolved their claims. The Air Force agreed to open some of its previously awarded national security missions to competitive bids. Over time, SpaceX has overtaken ULA even in this arena. During the most recent round of awards, SpaceX won 60 percent of the contracts compared to ULA's 40 percent. So when SpaceX raised the possibility of a ULA sniper, it came at an incendiary moment in the rivalry, when SpaceX was finally putting forth a very serious challenge to ULA's dominance and monopoly. It is no surprise, therefore, that ULA told SpaceX's Ricky Lim to get lost when he wanted to see the roof of their building in Florida. “Hair-on-fire stuff” NASA officials were also deeply concerned by the loss of the Falcon 9 rocket in September 2016. The space agency spent much of the 2010s working with SpaceX and Boeing to develop, test, and fly spacecraft that could fly humans into space. These were difficult years for the space agency, which had to rely on Russia to get its astronauts into space. NASA also had a challenging time balancing costs with astronaut safety. Then rockets started blowing up. Consider this sequence from mid-2015 to mid-2016. In June 2015, the second stage of a Falcon 9 rocket carrying a cargo version of the Dragon spacecraft into orbit exploded. Less than two weeks later, NASA named four astronauts to its "commercial crew" cadre from which the initial pilots of Dragon and Starliner spacecraft would be selected. Finally, a little more than a year after this, a second Falcon 9 rocket upper stage exploded during flight. Video of CRS-7 launch and failure. Even as it was losing Falcon 9 rockets, SpaceX revealed that it intended to upend NASA's long-standing practice of fueling a rocket and then, when the vehicle reached a stable condition, putting crew on board. Rather, SpaceX said it would put the astronauts on board before fueling. This process became known as "load and go." NASA's safety community went nuts. "When SpaceX came to us and said we want to load the crew first and then the propellant, mushroom clouds went off in our safety community," Phil McAlister, the head of NASA's commercial programs, told me for Reentry. "I mean, hair-on-fire stuff. It was just conventional wisdom that you load the propellant first and get it thermally stable. Fueling is a very dynamic operation. The vehicle is popping and hissing. The safety community was adamantly against this." Amos-6 compounded these concerns. That's because the rocket was not shot by a sniper. After months of painful investigation and analysis, engineers determined the rocket was lost due to the propellant-loading process. In their goal of rapidly fueling the Falcon 9 rocket, the SpaceX teams had filled the pressurized helium tanks too quickly, heating the aluminum liner and causing it to buckle. In their haste to load super-chilled propellant onto the Falcon 9, SpaceX had found its speed limit. At NASA, it was not difficult to visualize astronauts in a Dragon capsule sitting atop an exploding rocket during propellant loading rather than a commercial satellite. Enter the FBI We should stop and appreciate the crucible that SpaceX engineers and technicians endured in the fall of 2016. They were simultaneously attempting to tease out the physics of a fiendishly complex failure; prove to NASA their exploding rocket was safe; convince safety officials that even though they had just blown up their rocket by fueling it too quickly, load-and-go was feasible for astronaut missions; increase the cadence of Falcon 9 missions to catch and surpass ULA; and, oh yes, gently explain to the boss that a sniper had not shot their rocket. So there had to be some relief when, on October 13, Hughes received that letter from Dr. Michael C. Romanowski, director of Commercial Space Integration at the FAA. According to this letter (see a copy here), three weeks after the launch pad explosion, SpaceX submitted "video and audio" along with its analysis of the failure to the FAA. "SpaceX suggested that in the company's view, this information and data could be indicative of sabotage or criminal activity associated with the on-pad explosion of SpaceX's Falcon 9," the letter states. This is notable because it suggests that Musk directed SpaceX to elevate the "sniper" theory to the point that the FAA should take it seriously. But there was more. According to the letter, SpaceX reported the same data and analysis to the Federal Bureau of Investigation in Florida. After this, the Tampa Field Office of the FBI and its Criminal Investigative Division in Washington, DC, looked into the matter. And what did they find? Nothing, apparently. "The FBI has informed us that based upon a thorough and coordinated review by the appropriate Federal criminal and security investigative authorities, there were no indications to suggest that sabotage or any other criminal activity played a role in the September 1 Falcon 9 explosion," Romanowski wrote. "As a result, the FAA considers this matter closed." The failure of the Amos-6 mission would turn out to be a low point for SpaceX. For a few weeks, there were non-trivial questions about the company's financial viability. But soon, SpaceX would come roaring back. In 2017, the Falcon 9 rocket launched a record 18 times, surpassing ULA for the first time. The gap would only widen. Last year, SpaceX launched 137 rockets to ULA's five. With Amos-6, therefore, SpaceX lost the battle. But it would eventually win the war—without anyone firing a shot. Eric Berger Senior Space Editor Eric Berger Senior Space Editor Eric Berger is the senior space editor at Ars Technica, covering everything from astronomy to private space to NASA policy, and author of two books: Liftoff, about the rise of SpaceX; and Reentry, on the development of the Falcon 9 rocket and Dragon. A certified meteorologist, Eric lives in Houston. 9 Comments