NASA’s Kepler Spacecraft Nearing the End as Fuel Runs Low

This artist’s concept depicts select planetary discoveries made to date by NASA’s Kepler space telescope. (Credit: NASA/W. Stenzel)

MOFFETT FIELD, Calif. (NASA PR) — Trailing Earth’s orbit at 94 million miles away, the Kepler space telescope has survived many potential knock-outs during its nine years in flight, from mechanical failures to being blasted by cosmic rays. At this rate, the hardy spacecraft may reach its finish line in a manner we will consider a wonderful success. With nary a gas station to be found in deep space, the spacecraft is going to run out of fuel. We expect to reach that moment within several months.

In 2013, Kepler’s primary mission ended when a second reaction wheel broke, rendering it unable to hold its gaze steady at the original field of view. The spacecraft was given a new lease on life by using the pressure of sunlight to maintain its pointing, like a kayak steering into the current. Reborn as “K2,” this extended mission requires the spacecraft to shift its field of view to new portions of the sky roughly every three months in what we refer to as a “campaign.” Initially, the Kepler team estimated that the K2 mission could conduct 10 campaigns with the remaining fuel. It turns out we were overly conservative. The mission has already completed 16 campaigns, and this month entered its 17th.

Our current estimates are that Kepler’s tank will run dry within several months – but we’ve been surprised by its performance before! So, while we anticipate flight operations ending soon, we are prepared to continue as long as the fuel allows.

The Kepler team is planning to collect as much science data as possible in its remaining time and beam it back to Earth before the loss of the fuel-powered thrusters means that we can’t aim the spacecraft for data transfer. We even have plans to take some final calibration data with the last bit of fuel, if the opportunity presents itself.

Without a gas gauge, we have been monitoring the spacecraft for warning signs of low fuel— such as a drop in the fuel tank’s pressure and changes in the performance of the thrusters. But in the end, we only have an estimate – not precise knowledge. Taking these measurements helps us decide how long we can comfortably keep collecting scientific data.

It’s like trying to decide when to gas up your car.  Do you stop now?  Or try to make it to the next station?  In our case, there is no next station, so we want to stop collecting data while we’re still comfortable that we can aim the spacecraft to bring it back to Earth. We will continue to provide updates on the science and the spacecraft, which has yet to show warning signs.

Many NASA missions must set a course for a clear-cut ending and reserve enough fuel for one last maneuver. For example, Earth-orbiting spacecraft must avoid collisions with other satellites or an uncontrolled fall to the ground, while planetary missions like Cassini have to reserve fuel to avoid contamination of a potentially life-bearing environment. In Cassini’s case, NASA sent the spacecraft into Saturn rather than risk it falling into one of the planet’s moons. Deep space missions like Kepler are nowhere near Earth or sensitive environments, which means we can afford to squeeze every last drop of data from the spacecraft — and ultimately that means bringing home even more data for science. Who knows what surprises about our universe will be in that final downlink to Earth?

While Kepler continues to bring us exciting data as it draws close the finish line, the Transiting Exoplanet Survey Satellite (TESS) will be launching on April 16 from Cape Canaveral, Florida. TESS will search nearly the entire sky for planets outside our solar system, focusing on the brightest stars less than 300 light-years away, and adding to Kepler’s treasure trove of planet discoveries.

Written by Charlie Sobeck, system engineer for the Kepler space telescope mission

NASA’s Ames Research Center in California’s Silicon Valley manages the Kepler and K2 missions for NASA’s Science Mission Directorate. NASA’s Jet Propulsion Laboratory in Pasadena, California, managed Kepler mission development. Ball Aerospace & Technologies Corporation operates the flight system with support from the Laboratory for Atmospheric and Space Physics at the University of Colorado in Boulder.

For more information, read the FAQs about the final phase of Kepler.

  • ThomasLMatula

    One of the revolutions that will result from the BFR will be the ability to service spacecraft like this one and keep them working for decades. And yes, it will require the designers of spacecraft to rethink their designs to enable it.

  • JS Initials

    Billions and trillions of planets out there, and we will never know more than a tiny fraction of that number.

  • Michael Halpern

    They can also get bigger,a lot bigger and with telescopes size matters

  • patb2009

    spacecraft servicing was always possible, it was more no PM was required to put it in, so it never happened.

  • ThomasLMatula

    Yes, and imagine the benefits of assemblying spacecraft in orbit so they are no longer limited by the size of the faring on the launcher.

    Indeed, imagine the performance of a 30 meter telescope in the EM L4.

  • Michael Vaicaitis

    …or even a “very large” array of very large telescopes

  • Michael Halpern

    Even assembled with bfr, an 8m mirror

  • ThomasLMatula

    Which is what NASA should be looking to spend money on with the new low cost launchers instead of wasting $40 billion on their planned international commune in lunar orbit. One that will only be used a few weeks a year.

  • Michael Halpern

    A few weeks every other year if they rely on sls

  • Andrew Tubbiolo

    It’s not really a function of launch vehicle, it’s a function of spacecraft design. We’ve been talking about spacecraft tenders for decades, but have not done it. An architecture has yet to be established. What BFR will give is options for what a tender will look like. Will they look like this or more like this? Aerospace has made deliberate decisions not to go maritime for a long time. It’s a world wide issue.

  • ThomasLMatula

    True. Meanwhile the tourists on the BFR will get to take pictures of it on their way to Robert Bigelow’s Lunar Resort and Spa 🙂

  • ThomasLMatula

    Or does the BFR just return it to Earth for servicing and then returns it to orbit on a future flight eliminating the need for a tender.? At $5 to $6 million a flight it might be the cheapest option.

    BTW the BFR could also be the solution of what to do with the ISS after it’s mission is over. Its big enough, and cheap enough to fly, it could return the modules to Earth to put in a museum somewhere. You might be able to return the entire ISS to Earth for under $100 million. Ok, a $ Billion since NASA would need to be involved 🙂

  • Andrew Tubbiolo

    That will be an option once it’s operational. Again, a maritime type option. Such an operations environment changes the way money is moved about. That’s why I expect the capability to be there staring us in the face for some years before we take full advantage of it. The jump to BFR, should it work, would be like handing a 727 to 1920’s aviators. They would have no idea what to do with it, and not much of a market to keep it alive past the initial pulse of thrill seekers.

  • Vladislaw

    Hey now .. settle down here .. my district counts on that space probe costing 2 billion dollars and wearing out in a couple years.. don’t be goring my ox…

  • Vladislaw

    Not a really good analogy .. they would just add 300 passengers instead of one that the biplane allowed..

    It would be more like handing them the the giant jet engine and tell them to use that on planes.. they wouldn’t know what to do with it…

  • Vladislaw

    Or a mirror that is made up of 8m panels.. somewhere in the neighborhood of a 180′ mirror ..

    https://uploads.disquscdn.com/images/0acb8e058b550d14edffea6613dd92ca902a4b4de961d118386128eaf025eb64.jpg

  • Vladislaw
  • Michael Halpern

    That would be a monster of a collector

  • Michael Halpern

    Still with bfr they can do the same things as with any other vehicle just in higher quantity if they can’t think of anything else

  • Andrew Tubbiolo

    I’ll defend my analogy. The 727 assumes there are 5000 foot long paved runways. It depends on VOR and ADS navigation systems. It also depends on advanced weather monitoring and forecasting. PAPI and marker beacons for precise approaches and landings. Since it moves so fast you need clear, controlled and coordinated airspace for it to fly in. It’s speed also demands a robust time service like WWV which was going in 1920, so okay, they had that. That said, I kinda like your analogy better, mine looks more at differences in the operating environment and market demands. Yours is more base technological mismatch. And in my opinion BFR changes operating environment and market dynamics.

  • Michael Vaicaitis

    Thanks for that, good article, packed with data.
    By “array” I was thinking quite a few more decades down the line, and the potential of space based astronomical interferometry. Combining two/several/many telescopes together we could have an effective collector of millions of metres or even millions of kilometres. Would be a technically challenging, a big investment and a long installation period, but it would be the ultimate telescope.
    Curiously enough, on the subject of interferometry, since my earlier post and this one, I notice that China are considering outlier dishes to increase the resolution of FAST, though I suspect it wasn’t my post that gave them the idea.

  • Vladislaw

    You’re right I didn’t take all that into consideration .. I was just thinking of how today the pilot presses a button and the plane does the rest ..

    Hell in 1920 they wouldn’t even have a hanger for it ..

  • Vladislaw

    like this
    “The Allen Telescope Array (ATA) is a “Large Number of Small Dishes” (LNSD) array designed to be highly effective for simultaneous surveys undertaken for SETI projects (Search for Extraterrestrial Intelligence) at centimeter wavelengths.”

    https://www.seti.org/ata

  • Michael Vaicaitis

    there it is – was probably the aliens that beamed the idea into my head

  • Michael Vaicaitis

    I do wonder, and the proposed LEO constellations are a prime example, if low launch prices won’t actually make (most) spacecraft smaller and more numerous. Having many small spacecraft allows for production line like manufacturing, more rapid technological evolution, increased overall performance and functionality, diluted financial risk, diluted technical risk, diluted operational risk. Basically, more functionality/performance for lower costs and less risk. With low cost reliable launch, the science community can benefit from the same potential advantages as the comms guys.

  • Andrew Tubbiolo

    There’s 100’s of thousands of man years that has to be unlearned and re-learned for those dynamics to come to the fore. Look at the age distro of Space X workers. Look at the same thing in the primes. For what you want to experiment with, don’t trust anyone over 30.

  • Michael Halpern

    Nah you trust them, under the condition they are not allowed to justify things based on “how it’s always been done” make them understand the utterance of that phrase is cause for disciplinary consideration.

  • Andrew Tubbiolo

    But sometimes you need that experience to draw on. I imagine it will be much like the last time you “Could not trust anyone over 30.” you throw out the good with the bad. It’s the human way of change. In large groups humans can’t nit pick out the bad and keep the good. Only individuals can do that.

  • Michael Halpern

    That’s what thoroughly vetting them out is for

  • Andrew Tubbiolo

    Is there really a way to do that?

  • Michael Halpern

    Interviews and internet

  • Andrew Tubbiolo

    Ha! No. 🙂 Does not work. The main feature I’d look for in an interview of someone (face to face) if I needed them to have humility and the ability to learn would be a self deprecating sense of humor, and a good through knowledge and ability to articulate design mishaps they had made in the past. I’m not saying that’s a sure fire way, but when you’ve only got a short period of time, it’s the best I can think of. You never really know until you’re in the trenches with them. You only know on the job, and as they get older they’ll become crotchety just like any other human.

  • Michael Vaicaitis

    The pace of change can be much faster than most might anticipate – the bulk of the effort and the bulk of the difficulty is in the software. Throw enough coders and servers at the problem and it won’t take all that long to simulate 100’s of thousands or even 100’s of millions of man years.

  • Michael Halpern

    I’d ask specifically if they had control of mission architecture how they would do it

  • Andrew Tubbiolo

    What real world mechanical system has been developed FASTER or CHEAPER because of computers? Our historical record says that CAD, CAM, and simulation has made development cycles much much longer and more expensive. It allows you do do things you could never do with human draftsmen, and slide rule and tables based computation, but it does not save time or money.

  • Michael Vaicaitis

    “What real world mechanical system has been developed FASTER or CHEAPER because of computers?”
    Did you really mean to ask that question?. First off, how about the example actually in the wording of the question – computers : CPUs/GPUs/DSPs/Modems/etc, (and by extension every modern device that has any computational and/or connectivity capability). Then there’s rockets capable of going beyond Earth’s atmosphere and returning to land vertically, every car in the world made in the last 10+ years, every smartphone ever made, LEO constellation satellites, satellites in general, rockets in general….

    All examples of cheaper faster development leading to some combination of: more performance, more functionality, low power demands, lower manufacturing costs, lower end product price.

    “Our historical record…”
    who is “our”

  • Andrew Tubbiolo

    Michael, you’re changing the subject to electronics. That’s not the mechanical world. Go educate yourself by reading old issues of Aviation Week from the 50’s and 60’s then the 70’s and 80’s on the development of the 707 and then 767 and compare that to the 787 development cycle. Also compare the development of the 747 vs the 777. I think the 747 vs 777 argument is the strongest argument against me. Also look at the development history of the C-17 vs the C-141, and the development cycle of the F-22 and F-35 vs the cycles of the F-15 and F-16. All of those examples where newer systems are more capable than their progenitors but their development cycles were much much longer and more expensive. Also take the A-380 it’s not just an American issue. Cars show an even more stark price increase even with coolie un-unionized foreign labor a modern car is more expensive to develop and manufacture (of course the automakers are making more complex cars.) than the old all union all American model from before the 80’s.

  • Michael Vaicaitis

    I cannot claim expertise on those specific aircraft, but even there, as well as in a more general sense, I think you are putting the blame in the wrong place. Surely much of the increase in development time and cost of those sorts of “mechanical” systems arises from a case of diminishing returns. Incremental improvements become more and more difficult to tease out of the laws of physics. Then on top of that, as you point out, there’s a continuous drive to add more and more functionality – either because of commercial or “battlefield” demands – it’s a technological arms race. Basically it’s gets ever more difficult to make the next design iteration better than the previous one. Increased and improved computation is the enabler of technological progress, not the root cause of the cost increases.