People at NASA headquarters take deep breaths when the words 'First Story' appear in their email. Late this spring, Curt Niebur, the lead scientist for flight programs, received such a message.
“You open that email right away,” Niebur said. “You read it, and then you reply, ‘Thank you for sharing,’ and then you bury your face in a pillow and you howl in terror.”
The matter prompting Niebur’s apprehension involved Europa Clipper, one of NASA’s most scientifically important missions. The agency’s science division created the “First Story” process to encourage science project staff members to communicate potentially bad news without fear of overreaction by leadership.
This news seemed exceptionally bad. If what Niebur was reading was true, Europa Clipper was cooked.
NASA had spent more than $5 billion on the planetary probe. It is the largest ever built — as big as a basketball court with its solar arrays extended. The mission’s job is to help scientists on Earth determine the potential habitability of Europa, the moon of Jupiter that is wrapped in a thick shell of ice, beneath which a warm ocean flows with twice as much salt water as on Earth. Scientists believe it has all the chemical ingredients for alien life to have emerged.
In late May, Europa Clipper was set to be shipped to Florida to prepare for its October flight opportunity. If it missed its launch window, scientists would face a long wait for another shot — if the problem could even be solved.
Across NASA and its affiliate research labs, teams would soon find their summer dominated by a terrifying question: Was Europa Clipper doomed?
Jordan Evans, the Europa Clipper project manager at NASA’s Jet Propulsion Laboratory in Pasadena, California, explained the situation to the mission’s science leader, Robert Pappalardo, on May 2.
Someone outside NASA had discovered that a type of transistor designed to survive radioactive environments was failing. Europa Clipper made extensive use of the transistors.
“How many do we use?” Pappalardo asked.
Evans didn’t know for sure.
The black, pencil-eraser-sized parts, called MOSFETs — metal-oxide-semiconductor field-effect transistors — are simple on/off switches found on circuit boards, allowing electricity to flow. No one knew where, precisely, they were on the spacecraft or whether the precise MOSFETs used for the project were even affected.
But they knew the transistors controlled virtually every critical system and scientific instrument on board.
“The parts are used everywhere,” said Geffrey Ottman, the chief engineer of civil space flight at the Johns Hopkins University Applied Physics Laboratory in Maryland.
Without the ability to survive intense radiation, there could be no Europa Clipper. The conditions around the moon are extreme, comparable to the aftermath of a nuclear war. Surviving that environment drove the design of the spacecraft and its complex flight path around Jupiter. Across four years, the spacecraft would swing past Europa 49 times, dipping briefly into the computer-frying radiation and getting as close as 16 miles to the moon’s surface.
If MOSFETs started failing, “the immediate thought is a dark spacecraft,” Ottman said.
The next day, Evans stood up a “tiger team” of specialists from his and Ottman’s labs, as well as the Goddard Space Flight Center, to work on the problem.
Soren Madsen, a 34-year Jet Propulsion Lab veteran, led the tiger team, which first had to verify that the issue was real.
That weekend, they subjected a spare MOSFET to a beam of high-speed electrons. The transistor failed, and by Monday, he and Evans knew the mission was in trouble.
Determining the number of MOSFETs in the spacecraft was like figuring out how many roofing nails were used in building a house. Initially, the tiger team estimated that there were almost 900 in the spacecraft. Two weeks later, it was about 1,500. Replacing them all could cost as much as $1 billion — money Congress might not give NASA — and take years.
Worse yet, the tiger team reported, that not only were the MOSFETs vulnerable to radiation, but there wasn’t just one universal MOSFET in the spacecraft. There are various types and families of the transistor. Confounding matters further, Madsen said, parts built at different times might not “perform and behave the same.”
Ottman was a tiger team member tasked with tracking down and testing samples of every variation of MOSFET that had been built into Europa Clipper. He got the call while picking up his daughter from college.
“I end up spending half an hour on the phone talking about this problem while stuff is not being loaded in the car,” he said.
His family got used to his absences that summer.
“We knew that we were going to be giving up nights, weekends, early mornings, and all of that,” he said.
While Madsen’s team worked through the problem, Evans assigned a second mitigation team to figure out technical solutions.
Initially, “I was kind of frozen,” said Joe Stehly, its leader.
He could not imagine a way out so late in the game. But eventually, his team came to understand that when parts like this are affected by radiation, “they’re going to heal themselves to some degree.”
The process is known as annealing. Heating a transistor allows its atoms to rearrange and redistribute in a restored condition. It doesn’t work indefinitely, but Europa Clipper doesn’t have to last forever — just the four years of its scheduled mission. His team could devise strategies for healing each MOSFET type on the spacecraft.
Stehly also worked with Pappalardo to reconsider the ways the mission conducted its experiments.
Some proposals were simple, like powering down certain instruments during certain flybys of Europa, sparing their transistors from radiation damage. Others were more complex, involving less precise key measurements, front-loaded during earlier rounds of the mission’s many passes.
With all that, plus annealing, Pappalardo felt as if a credible “compressed tour” strategy was emerging. However, he and Niebur fought against flying the mission if it could not answer the major scientific questions about Europa.
Then on July 26, representatives from JPL and the tiger and mitigation teams presented NASA headquarters with their latest findings. Although the teams felt they had made real strides, a color-coded chart depicting their progress dismayed some NASA officials. There were a handful of green cells, but it was mostly a grim sheet of yellow and red.
“They were not final assessments,” said Joan Ervin, the tiger team systems engineer at the Jet Propulsion Lab. “We just said, ‘This is where we are today.’”
Some radiation tests took weeks and were not yet completed, and they might not bring good news. “We wanted to prepare everybody,” she said.
Moreover, some MOSFETs remained unaccounted for, particularly those used in electronics built by foreign subcontractors. Without knowing how those missing chips would perform, even implementing every mitigation strategy would be insufficient.
“We were down to the last four parts,” Evans said.
Days after the meeting, however, a European subcontractor found a handful of old engineering samples. The Applied Physics Laboratory received the final parts from Germany by FedEx.
“And when we got them, boy, the pressure was on to get those into tests immediately,” Ottman said.
Back in Pasadena, Stehly was brainstorming with Evans when Jeff Srinivasan, the flight system manager, walked into the office. He had an idea.
While the spacecraft was in flight, there would be no way for the mission’s engineers to know how the MOSFETs were behaving. One day, they would simply fail.
Srinivisan proposed a solution: What if they took samples of each type of MOSFET, put them on special circuits and then installed the whole package in a box they could bolt to the spacecraft? It could send health data back to Earth throughout the mission.
It would be like a canary in a coal mine. While Europa Clipper was at Jupiter, if a certain type of MOSFET weakened or failed in the box, the team would know to anneal that type of transistor throughout the spacecraft or employ another strategy to keep the mission working.
“It was one of those that’s so-crazy-it-might-work moments,” Stehly said.
They called it the canary box.
But Srinivasan’s team would have to build it in weeks instead of years, write its software and then integrate it all with the spacecraft, which had already been shipped to Kennedy Space Center in Florida.
On Aug. 27, the teams updated their progress to senior NASA leaders, including Pam Melroy, the deputy administrator, and Jim Free, the associate administrator. This time, the cells on Ervin’s chart were green from top to bottom.
Some attendees, including Niebur, found that suspicious.
He thought maybe the presenters were succumbing to “launch fever,” the desire to fly at all costs. He wasn’t alone. A month earlier, the mission had seemed done for. How could they have made that much progress in 30 days?
Ervin laid out the case. She explained everything that had happened for every single assessment and justified every change. They had tested every type of MOSFET in the spacecraft.
“We were systematic, we were rigorous, and we were confident in our results,” Ervin said.
Especially with annealing, they were safe to fly with zero changes to the science plan. But even if they reached Jupiter and discovered unpleasant radiation surprises, the canary box, now installed on the spacecraft, would transmit warnings to Earth. They had so much data and so much resiliency in the system that the science requirements were assured.
At the end of the meeting, Nicola Fox, the head of science missions, conducted a poll of NASA leaders: “Go” or “No go.”
One by one, the agency executives voted.
Go.
Go.
Go.
Until it was unanimous around the room.
Europa Clipper will be on the launchpad on Oct. 10.
After decades of effort, a powerful observer was headed to Jupiter, capable of revealing whether Europa’s ocean is friendly for life.
Oman Observer is now on the WhatsApp channel. Click here
