How NASA harnessed sunlight to revive its planet-hunting telescope

The inside story of the Kepler rescue


Are we alone in the universe?

That's a big question, and there have been many attempts to answer it, whether it’s beaming radio waves at specific star systems or sending recordings of multilingual greetings out of our solar system. But for the last few years, NASA has been using the Kepler Space Telescope to hunt for life in a more direct way — by looking for planets.

It wasn’t until the Kepler began collecting data in 2009 that we knew just how many planetary neighbors we have. Since then, the telescope has found over 4,000 planetary candidates within 3,000 light years from us, almost 1,000 of which have been confirmed by the observations of other telescopes.

It also helped discover that multi-planet solar systems — like our own — are also more abundant than scientists had imagined. Many of these planet candidates are even orbiting stars similar to ours, and some of those are even near the vaunted "goldilocks zone" — the distance from a star where a planet is cool enough to allow water to form but not so cold that it freezes.

Kepler was changing how we viewed our place in the galaxy

For over three years Kepler painted this increasingly diverse picture of the nearby galaxy. The section of the Milky Way that it studied — referred to as the Kepler Field — contains around 100,000 stars, some as far as 3,000 light years away and some as close as 600. Finding so many planets so relatively close was changing how we viewed our place in the galaxy, and the results kept pouring in. The mission was succeeding beyond most everyone’s expectations.

Then, in July of 2012 one of the telescope’s four gyroscopic reaction wheels failed. That’s a problem because the wheels — which are spun at varying speeds to make precise targeting movements — are crucial for aiming. Since Kepler only needed three to operate, this wasn’t an immediate problem; in fact it was a strange, temporary blessing. "It turned out that using the three wheels gave us very, very slightly improved pointing than when we were using four," says Dr. Steve Howell, a project scientist for Kepler.

But the long-term implications were grim. By studying the amount of friction the wheels were encountering at different speeds, the team saw a problem. The typically smooth data curves representing the wheel’s performance were distorted by both an increase in friction and an increase in the amount of power needed to overcome that friction. Worse, the same anomalies were also showing up in the data from a second reaction wheel — meaning Kepler was facing another mechanical failure.

Kepler data 2

A visual representation from 2013 of all the multiple-planet candidate systems discovered by Kepler in orbit around their stars. Hot to cool colors represent big planets to smaller planets, relative to the other planets in the system. Video is available here.

Howell said the team knew that meant they were dealing with a potentially mission-threatening problem. "When the first reaction wheel failed, a few of us here and a few people at Ball Aerospace (who built Kepler) already started in the back of our minds kind of thinking ‘What if we lose another reaction wheel?’ You don't want to just throw away a relatively good, expensive telescope that can still do great science."

"What if we lose another reaction wheel?"

The teams called on scientists and engineers in the community at large to help come up with a solution. Before they arrived at one, the second wheel failed in May of 2013. And while Kepler loosely follows the Earth’s orbit around the Sun, it does so at 45 million miles away from us — nowhere close enough to attempt a repair.

The Kepler team, about 40 people strong, had already completed the main mission objectives, so they declared in August along with NASA that they were ending their attempts to fully recover the telescope. But behind the scenes, everyone got to work on an idea that allowed the mission to survive by evolution. The solicitation had stirred up an ingenious solution that the team at Ball Aerospace ran with: use the pressure exerted on the telescope’s solar panels by the Sun as a stand-in for the third wheel.

In space, even a tiny force can move an object as big as a telescope. Something negligible on Earth, like the small amount of pressure that the Sun’s light exerts on things it touches, is still a force that has to be calculated and corrected for in zero-gravity. That’s why the wheels were there: to create forces that worked in opposition to allow for precise adjustments. But that seemed impossible after the second wheel’s failure. Luckily, the remaining wheels were aligned in such a way that pressure from the Sun could virtually replace the crucial third wheel. That means that the scientists can still use the other two wheels to maneuver, a development so lucky that Howell called it "almost magic."

Kepler K2 mission

If any other combination of wheels had failed, Kepler might have been left adrift, Howell says. "I can't say we would’ve ever found a solution in that case, but you know we might have," he says. "We're creative people." Though the team obviously can’t vary the Sun’s weak force like they could control a wheel, that force is still enough to give the mission a new life.

With a feasible solution in hand, the Kepler team worked on refining the specifics of this new mission, dubbed K2. By November of 2013, they sorted out the details and announced their plans. The team began performing engineering runs as early as February of 2014 to help show NASA that the solution was possible. In fact, data from this original nine-day test run led to the discovery of the first K2 exoplanet, which was announced two weeks ago alongside the news of Kepler's resurrection. Now the only thing standing in their way was the Astrophysics Senior Review.

The Senior Review decides the fate of operating missions. Every two years, scientists behind projects like Kepler write proposals to pitch for the extension — or survival — of their missions to a peer review group from NASA’s Astrophysics Division. The review group then makes recommendations to NASA headquarters on whether to increase funding, slash it, or scrap a mission altogether — recommendations which are followed "fairly closely" by NASA, according to Howell.

In early 2014, the Senior Review endorsed the K2 mission’s extension. "This is an outstanding mission, and we look forward to the results from the program," the recommendation reads, acknowledging the wider scope of new science the team can perform as the reason for the extension.

Then, in May, one line on the third page of NASA’s response to the Senior Review announced the telescope’s official resurrection: "Kepler Space Telescope: extension approved."

"The summer of wheels failing and K2 being born, that six-month period was quite the emotional ride," said Howell. "We all had a great attachment to Kepler and the [original] Kepler mission, and we would’ve loved to continue to look at that field for the next 20 years."

Reviving the telescope means that the team has to adjust the types of science it expects to do. The years-spanning observation methods it used to find (roughly) Earth-sized planets are impossible now. But that mission was fulfilled in grand scale, so the changes are welcome.

This time around, the team is focusing on more specific targets, which Howell describes with obvious excitement. The new mission will study stars of various ages much younger than the Sun — eschewing more adult stars like the 4.5 billion-year-old Sun for toddlers that are only 100 million years old. The team wants to know what types of planets orbit these young stars, in the hopes that observations can tell us more about how and when planets form. They’ll also study variations among galaxies, and try to learn more about stars that explode as supernovae. And K2 will look closely at sections of our own solar system, particularly the objects in the Kuiper belt. There they can study asteroids and other small bodies in deep orbit around the Sun (like Pluto) as old as our solar system, which can provide clues to how our system formed in the first place. It’s another type of observation the original mission didn’t perform, and it might be able to tell us where our planet’s water came from.

Kepler K2 field

This image of the Milky Way shows the approximate locations of the nine target fields of Kepler's K2 mission. Data will be collected from each target field will for about 80 days, and then will then be studied in search of exoplanets that exist about 100 light-years away from us. A hi-res version can be seen here.

Howell was clear, though, that planets are still the focus. "K2 has this great breadth of science, but it also continues the legacy of planet detection," he says. K2 will study stars in hopes of finding planetary candidates closer than those in the original Kepler field, perhaps only 100 light years away, which the team refers to as "high-value planets."

That creates an opening for other telescopes, which can take these candidates and make more refined measurements about their composition. Instead of just knowing if an exoplanet exists, K2’s data will eventually help us know what they are like, and one day even capture an image of one — something that scientists hope the James Webb Space Telescope, the Hubble’s successor, will be able to do when it launches in 2018.

There's still one big obstacle

Kepler’s K2 mission does have one unavoidable obstacle, though, and it’s one that has stopped many missions before it: fuel. "Barring anything else breaking or any other hardware failures, fuel will limit the lifetime of K2," Howell told me. The fuel aboard Kepler is used for thruster movements, which are important for when the team has to reorient the telescope to point its antenna towards the Earth for downloading data and uploading commands.

Though the fuel will run out, the good news is that the amount left should sustain Kepler for about three more years. In addition, the two remaining reaction wheels show none of the anomalies that signaled the death of the others. But this also means the telescope will be around long enough that it has to survive one more Senior Review in 2016.

Until then, the Kepler team will be scouring for new planets by studying the new target star fields for 80 days at a time. Howell hopes that Kepler will make worthwhile observations until its fuel runs out. He remains in awe of all the mission has accomplished. "It's certainly one of these history-changing events," he said. "Not only is there this enormous number of planets in the Kepler field, but that's one tiny little sample of the galaxy and one tiny, tiny little sample of the universe."