On August 9, 2012, on a sunny day in eastern Florida, NASA's Morpheus lander hovered 15 feet over the ground. One second passed, then two. Then the craft tilted to its right, kept tilting, collapsed into the ground, and thudded for a moment before fuel met light and a fireball erupted. A further explosion 100 seconds later sent sparks 25 feet out, destroying most of the remainder of the craft.
The clip of the Morpheus lander's destruction would be nearly comical -- the squat, disco ball-decked craft looks more amphibious than aeronautic -- if it didn't show the culmination of a series of failures since the project's start. The crash was but the latest in a litany of engineering and logistical failures plaguing the lander and helped to serve as something of a microcosm for recent issues surrounding the entire agency. And while no one was hurt during this crash, NASA's stated goal of returning to manned launches within the near future means that forthcoming tests won't be limited to empty machinery.
Mistakes and failures, however, can be corrected. With its myriad struggles during its Apollo and Shuttle heydays, NASA knows this well.
But a check of NASA's records shows that there never was any follow-up investigation of the August crash released. Instead, according to those who worked on the project, NASA dismissed the accident as another company's hardware problem rather than anything that the agency needed to correct.
According to NASA software engineer Kyle Clement (not his real name), NASA, which did not return calls for this story, should have known problems were amiss.
"You could tell something was wrong prior to [Morpheus] blowing up," Clement, who has been working at NASA for decades, told the Houston Press.
A former NASA contractor, Tim Cousins (also not his real name), saw the issues as the culmination of the atrophy that has slowly been smothering NASA.
"That's the thing about NASA -- they don't like hiring people who are smarter than themselves," Cousins, who recently retired, told the Press. "So you've got this transition where things just get worse and worse, and then you have shit like this."
While Cousins could attest to other projects within NASA suffering through similar issues, Clement had a chance to see the Morpheus project from its inception three years ago. Attempting to test both autonomous landing technology and a combination of liquid methane and liquid oxygen as fuel -- a potential mix for future missions to Mars -- Morpheus was meant as a relatively low-cost, low-risk project, coming in at only $8 million. At 2,800 pounds with a full fuel tank, it was intended to be far less significant than anything that, say, private space contractor SpaceX is currently approaching.
But despite Morpheus's relative simplicity, problems arose early.
The first issues came during the project's Tethered Test phase, in which Morpheus was suspended 20 feet off a concrete pad by a 120-ton crane. While the first test suffered from only one "procedural oversight," the second, coming on April 27, 2011, led to an immediate problem with the engine. According to a Morpheus review, "An H-bridge circuit controlling the throttle valve fail[ed]." Pitching asymmetrically for 13 straight seconds, the lander failed to shut itself down, saved only by the nylon straps to which it was attached. Per the document, "[p]re-test haste" led to the issues thrashing the lander.
Minor issues beset the next few tests, but on June 1, 2011, the team attempted the fifth Tethered Test, with a targeted hover time of 40 seconds. The test was largely considered an engineering success. However, as before, a procedural oversight led to a problem that should have been foreseeable. As the recap related:
Despite being the most successful test to date, TT5 resulted in a non-vehicle test range issue that caused the team to stand down from testing for over a month. Following the test, a small grassfire caused by spalling concrete was not properly contained and grew to a large grassfire covering 29 acres of JSC property. The grassfire was finally contained when local fire departments were called in.
In the "Morpheus Tether Test #5 Delta-Test Readiness Review," obtained through a federal public records request, the only form of "Fire Prevention" NASA sought prior to the Tethered Test was "[m]owing to occur Tuesday AM." In the midst of the worst drought Texas had seen in modern history, the Morpheus team believed that the only measure necessary to prevent the tinder from lighting was an early-morning mow. "The thing nearly destroyed a firehouse nearby," said Cousins.
"You got signs," says Clement. "These are the signs. Cutting corners." All of which culminated in last August's accident, clips of which have been viewed more than a million times.
However, while public records requests netted documents pertaining to the Tethered Tests in question, a request for "any findings pertaining to the follow-up investigation into the causes of the [August 2012] crash" brought only a pair of links to NASA's official Morpheus blog. The first link pointed to a post from this May in which the team announced it was "picking up where we left off -- in fact we never stopped working." The second link, from last August, points to a "hardware failure." The post notes that "at the start of ascent we lost data from the Inertial Measurement Unit (IMU) that supplies navigation updates to the flight computer." (Clement claims the NASA blog is routinely scrubbed of comments critical of the project.) Ironically, the second link also claims that the material recovered after the crash "will give us greater insight into the source of the problem," though NASA seems either unwilling or unable to share such information.
"They didn't do an investigation on this," Clement says. "Why? Because they're hacking this. They got lucky nobody got hurt."
For Keith Cowing, who helps run NASA Watch, the issues surrounding Morpheus stem from the culture. "This is kind of a problem with NASA these days," Cowing told the Press. "They like to walk around with a 'Yes, I'm A Rocket Scientist' button, but rockets blow up, and a lot of people don't necessarily want to, as they say, 'have a bad day.' That was Morpheus."
According to Cousins, the two major issues besetting the earlier tests -- an unsuccessful engine kill and a lack of fire prevention -- came to a head in August's crash. "They had no provision to 'safe' the lander, i.e. shut the engine off, in the event of the [software] failure ... while there were only two guys in fire extinguishers to attempt to put out the fire," Cousins said. Clement, meanwhile, says that NASA initially claimed the issue stemmed from Morpheus's Space Integrated Global Positioning System/Inertial Navigation System (SIGI), provided by Honeywell. Such a cause is plausible -- but SIGI was cited as an issue in the TT2 mishap, which should have alerted NASA to potential errors therein.
Even that reason, however, doesn't sit well with Clement. "That's bullshit," he says. "It wasn't a hardware problem. [NASA] tested that SIGI all the time. This was all software...Yet it was blamed on another company. It's not our fault, and then there's no review done on it."
And that is what NASA stuck to. In a follow-up message to his staff, NASA Administrator Charles Bolden noted that he considered the Morpheus project -- millions down the drain, and the prototype completely destroyed -- a success:
This small project was formulated at JSC as a way to develop, understand, and demonstrate some new technologies and to build the capabilities of our work force in a rapid engineering cycle environment - in the vernacular, learn fast, fail forward. I am sure some might think of Morpheus as a failure since a significant piece of hardware crashed and burned while under test. Contrary to this view, I regard Morpheus as a success.
"We've gotten away with this 'fail forward' crap -- what the hell is that about?" Clement says. "It's almost like, 'Hey, let's light a pack of firecrackers and see what happens.'"
Indeed, it seems Bolden considered it such a success to this date, 11 months later, that no follow-up investigation has ever come to light. That apparent lack of investigation was made all the starker after the Press received the follow-up report about an accident taking place within the Active Response Gravity Offload System (ARGOS), set to test astronauts in no- and low-gravity situations. On January 16, a "test participant...was un-intentionally dropped, approximately 12 to 18 inches (30.5 to 45.7 cm) in the vertical (-Z) direction." The participant, fortunately, was not significantly injured, though "the un-intended drop could have been as much as 4 to 5 feet."
If you like this story, consider signing up for our email newsletters.
SHOW ME HOW
You have successfully signed up for your selected newsletter(s) - please keep an eye on your mailbox, we're movin' in!
The report on the ARGOS accident was damning. "ARGOS was an engineering development project that was treated as if it were operational," the report, issued last March, read. "The ARGOS team was beset with funding issues from the outset. This drove the project to experience 'Groupthink' and seek additional non-developmental users which gave the appearance that the system was operational."
But at least there was a follow-up. It seems there's been far less -- or even nothing -- following the Morpheus crash. "You could see that this is a different way of doing business, that we should expect failures -- well, yeah, duh -- but the problem I see is the pattern," says Clement. "They hide behind the 'prototype' word, but they truly intend to deliver this Morpheus crap once they get it to 'work.' This is their method: Build a prototype without any oversight, then call it ready for production and deliver. Just like ARGOS."
This shift, this pattern, is hampering NASA moving forward. According to Clement, the trend will only continue.
"We're using a new, evolutionary approach to developing this -- used to be called faster-better-cheaper -- and it's a complete, utter failure," he adds. "And we're doing it again."