Forgot your password?
typodupeerror
NASA

NASA Abandons Kepler Repairs, Looks To the Future 73

Posted by Soulskill
from the time-to-crash-it-into-planet-genesis dept.
cylonlover writes "If NASA has anything to say about it, Kepler is down, but not out. At a press teleconference on Thursday it announced that it has abandoned efforts to repair the damaged unmanned probe, which was designed to search for extrasolar planets and is no longer steady enough to continue its hunt. But the space agency is looking into alternative missions for the spacecraft based on its remaining capabilities. 'On Aug. 8, engineers conducted a system-level performance test to evaluate Kepler's current capabilities. They determined wheel 2, which failed last year, can no longer provide the precision pointing necessary for science data collection. The spacecraft was returned to its point rest state, which is a stable configuration where Kepler uses thrusters to control its pointing with minimal fuel use.'"
This discussion has been archived. No new comments can be posted.

NASA Abandons Kepler Repairs, Looks To the Future

Comments Filter:
  • A partial success (Score:3, Interesting)

    by Anonymous Coward on Friday August 16, 2013 @11:05AM (#44583745)

    It was supposed to have a 3.5 year mission. It appears that it might have come pretty close to that, but it didn't get nearly the data it was supposed to. This is pretty disappointing and they should probably hold a formal design review to determine what went wrong in the design and construction and determine if a replacement should be built to finally accomplish the mission that was planned.

    • by ackthpt (218170)

      It was supposed to have a 3.5 year mission. It appears that it might have come pretty close to that, but it didn't get nearly the data it was supposed to. This is pretty disappointing and they should probably hold a formal design review to determine what went wrong in the design and construction and determine if a replacement should be built to finally accomplish the mission that was planned.

      You're talking a lot of sense there. We learn from our mistakes. Here's an opportunity to learn.

      • let's hope so. However, much of these issues are normally tied to lack of funding. And I do not see CONgress learning from it.
        • However, much of these issues are normally tied to lack of funding. And I do not see CONgress learning from it.

          Or maybe Congress is learning. NASA has a long history of low-balling estimates, and then latter coming back to Congress with massive overruns. They weren't allowed to get away with that with Kepler. NASA needs to learn how to manage a budget.

          • by Firethorn (177587)

            At least part of the reason for the extra expenses is that Congress will often starve the projects of money over the years, forcing NASA to work slower - and as it costs money to simply 'keep the lights on', it results in overruns. Then there's the whole 'part of it has to be built in MY district', etc...

            A program that might cost $5M to build and launch a satellite over 3 years might coast $10M to do the same over 10 years.

    • Re:A partial success (Score:5, Interesting)

      by rraylion (1406761) on Friday August 16, 2013 @11:17AM (#44583853)

      as with most satellite missions gone wrong -- its was the gyroscope.... remember it was replaced on Hubble a few times... its seems to be the weakest link in a lot of missions as it has to be a moving part to induce counter rotation in the satellite. it's only feasible to put so many on board ... so maybe a redesign of this one part will save future missions ... but maybe its time to think outside the box now that we know ion drives work a kg of propellant and three exhuast ports would fix this issue with new tech

      Other than that it was an awesome mission.

      • Re:A partial success (Score:5, Informative)

        by Kjella (173770) on Friday August 16, 2013 @11:33AM (#44584019) Homepage

        as with most satellite missions gone wrong -- its was the gyroscope

        Actually the gyroscopes lasted the original 3.5 year mission, but due to more noise than anticipated they collected less data than planned - which was why the mission was extended to 7.5 years. Now they won't be able to finish that, but that was really the backup plan failing. Oh well, not everything can be a Mars Rover exceeding all design specs by leaps and bounds.

        • Didn't NASA have reaction wheels go on another probe as well? The one we're sending to explore Ceres I think had reaction wheel issues as well and had to be reconfigured to run its mission on thrusters as well.

          It certainly seems like this is probably going to be a big engineering challenge into the future since super-steady stargazing probes are hardly going out of fashion. Though I suppose the better issue is "can we make some of this stuff replaceable/repairable cost effectively?"

          • Re:A partial success (Score:5, Informative)

            by Strider- (39683) on Friday August 16, 2013 @11:53AM (#44584239)

            Didn't NASA have reaction wheels go on another probe as well? The one we're sending to explore Ceres I think had reaction wheel issues as well and had to be reconfigured to run its mission on thrusters as well.

            Reaction wheels a very well known concepts in spaceflight. The ISS uses them to point itself (The Control Moment Gyros) and pretty much any and all geosynchronous satellites also use reaction wheels to keep themselves pointed at earth. This is actually how they ended up recovering Galaxy 15. After several months of drifting while "zombie", the reaction wheels finally saturated (spinning as fast as they could go) causing the satellite to lose earth lock, and go into a safe mode.

            Anyhow, the upside and downside is that they are relatively simple devices, and allow for very precise and stable pointing without spending a lot of fuel (you don't want your exhaust condensing on your optics in a telescope now do you?), but at the same time they're mechanical devices, and thus are more fragile than something that's purely solid state.

            • Do you know what sort of bearing/contact point they use for a gyro that is supposed to perform reliably for multiple years, at alarming temperatures, in the vacuum of space? Is there some bearing material or lubricant that actually works under those conditions, or do they run the gyros in a gas-filled and temperature controlled module so that more conventional techniques can be used?

              • by tibit (1762298)

                I think that in the future the only option for gyros, barring new bearing material discoveries, would be fully non-contact operation with magnetic levitation bearings, operating in vacuum. Right now I think the gyros are at least temperature-conditioned, but yeah, the bearings are still a problem.

        • Re:A partial success (Score:5, Informative)

          by Sockatume (732728) on Friday August 16, 2013 @11:45AM (#44584139)

          To nitpick, they collected as much data as planned, but it was noisier than expected. Therefore they needed more data to bring the signal above the noise. Interestingly the source of the noise isn't Kepler, but sunspots on the stars causing fluctuations in the brightness. They'd counted on a 3.5 year mission being long enough to collect a strong enough signal by assuming sunspot noise was the same as from the Sun, but it turned out it was actually stronger.

      • Re:A partial success (Score:5, Informative)

        by Yoda222 (943886) on Friday August 16, 2013 @11:45AM (#44584141)
        You can control attitude using (ion) thruster, but reaction wheels have some advantages. One of the biggest advantage of RWs is that it's (close to) a linear actuator (outside of the zero crossing zone), where thruster are bang bang actuators. In fact they are bad bang bang actuator, because you have some transient at start and end of the pulse.
        • by NEDHead (1651195)

          I believe that what you are referring to is actually the 'Biggity Boom Boom Clunk" actuator.

          • No. That's the failure mode of a reaction wheel. A thruster is somewhere between "Ack!" and "Thbbft!"
      • by Anonymous Coward on Friday August 16, 2013 @12:00PM (#44584305)

        Given how common this mode of failure is, throwing a lot of money into the study of better reaction wheel designs would probably pay off for a slew of future missions.

        [I can't believe I just said that]

      • as with most satellite missions gone wrong -- its was the gyroscope.... remember it was replaced on Hubble a few times... its seems to be the weakest link in a lot of missions as it has to be a moving part to induce counter rotation in the satellite.

        I wonder why they aren't using gyros with magnetic bearings. Especially in micro-gravity, that could help quite a lot.

        • by dpidcoe (2606549)

          I wonder why they aren't using gyros with magnetic bearings. Especially in micro-gravity, that could help quite a lot.

          I was going to say the same thing (though for all I know maybe they are using magnetic bearings).

          There's also some new R&D being done with active magnetic bearing control, basically you monitor the position of the shaft inside the bearing and apply changes to the magnetic field in order to dampen vibration and ensure the shaft stays centered.

        • by photonic (584757)
          That was my thought too, but apparently they do exist already with magnetic bearings, for example this one [rockwellcollins.com]. But problems with gyros/reaction wheels seem an old problem, wasn't Hubble stranded once for a few months with not enough gyros?
        • by cusco (717999)
          Don't know for sure, but I'd be surprised if things like solar storms couldn't upset magnetic bearings. Also, having a really strong magnetic field nearby can perturb readings on some of the more delicate instruments.
        • by Strider- (39683)

          I wonder why they aren't using gyros with magnetic bearings. Especially in micro-gravity, that could help quite a lot.

          We aren't talking about small sensor type devices here. (Attitude sensing is done with laser ring gyros, no moving parts) Reaction wheels are rather large objects and can provide significant force to orient the spacecraft.

          • I know what they are and what they do, although your "rather large" feels like a rather relative term to me. My question still stands. In near-0g, on board of an astronomical satellite or probe, in the absence of sudden large forces and movements, why wouldn't active magnetic bearings be a viable choice if ordinary bearings seem to pose problems?
      • let's hope so. However, much of these issues are normally tied to lack of funding. And I do not see CONgress learning from it.

        Fixing CONgress is easy... stop electing idiots (ok, ok, easier said than done, but keep in mind, 2014 coming up).

      • It is not the gyroscope. The problem was with reaction wheels. Any gyroscopic effect they have is an unwanted side effect in most cases.

              Brett

      • maybe its time to think outside the box now that we know ion drives work a kg of propellant and three exhuast ports would fix this issue with new tech

        Not only will ion drives not work in this situation (insufficient thrust, insufficient fine throttle control), they have the disadvantage of potentially contaminating the optics with their exhaust. As with Hubble, the later is a huge concern and why they chose reaction wheels in the first place. (Not to mention not requiring potentially mission limiting cons

    • by cusco (717999)
      What went wrong with the design? Congress and The Office Of Management and Budget (aka TOMB), probably. That seems to be the issue with pretty much every problem we ever encounter in the space program. When lawyers and accountants get to decide that they know how to design spacecraft better than rocket scientists trouble is pretty much guaranteed. What do you want to bet that another layer of redundancy had to be eliminated because of budgetary concerns?
    • Actually it lasted for the 3.5 years it was supposed to. The mission was extended to 2016 or something like that before this failure. If this was the military no one would think twice about it and they'd just launch another as they'd probably have built two anyway (hey, why let tax payers know the military has too much money?). Unfortunately it's NASA and with their tiny budget every event (even a mission failing AFTER it's designed lifetime) has massive repercussions on their budget.

  • by ACK!! (10229) on Friday August 16, 2013 @11:19AM (#44583867) Journal
    It seems that NASA is challenging the scientific community to come up with new projects or experiments that could still work despite the Kepler's limitations now. So it is time for all the scientists with their eyes toward the sky to start dusting off the keyboards and to come up with some cool new proposals. So even though it was not a screaming success the project could still turn out to be important to the scientific community going forward. If we have some scientists who can come up with neat ideas on how to use the capabilities this device still has.
    • by steelfood (895457)

      Data collection may have ended, but the analysis has certainly not. We can't gauge the success of the mission to any degree of accuracy until all the data is analyzed. It's been enormously successful thus far in finding exoplanets, but the unanalyzed data may yet yield greater discoveries.

      And even data collection may still be possible. There is mention of using software to do the correction, rather than relying on hardware. I don't know how feasible this would be when most stars are a few pixels wide at the

    • To be fair, the project lasted as long as it's primary lifetime was scheduled (3.5 years). The only reason it wasn't 100% successful is that stars are apparently noisier than our sun, so it requires longer exposures to beat down the noise. This wasn't a failure - it was just an unforeseen issue with reality.

      • by ppanon (16583)

        The only reason it wasn't 100% successful is that stars are apparently noisier than our sun

        Hmm. I do wonder how stars are "noisier". Is it at the source or due to interference from interstellar gas and the Oort Cloud? If the former, then does it have implications for the Drake equation?

        • The have more and bigger spots, and have more variance in their total luminosity. It's not interference, the stars themselves are noisy.

          • by ppanon (16583)
            Thanks for the informative response. Then it could affect the factors in the Drake equation. If the Sun is somehow more stable than most other stars, that could have an impact on the stability of the habitable zone and the ability of life to develop to sufficient complexity to develop intelligence.
        • by slick7 (1703596)
          They found something that really scares them and if they ignore it maybe it will go away.
    • Pioneer 6 was last contacted in 2000, still operating 35 years after launch.

      I wonder if it or its companions Pioneer 7 & 8 are still functioning and if their data could still be useful.

  • by Anonymous Coward

    You can point to any angle with only two axis of rotation. Wont be as precise, but should be better than using the thruster.

    • by Sockatume (732728)

      Right, they're hoping to use it to perform new science missions because it can still steer, just with impaired precision.

    • by rubycodez (864176)

      useless for finding the very slight variations in brightness on a pixel and nearest neighbors when compared to same over month or more than a year, on which kepler's detection methods depend.

      • useless for finding the very slight variations in brightness on a pixel and nearest neighbors when compared to same over month or more than a year, on which kepler's detection methods depend.

        I imagine that that is why they are asking for proposals from scientists who have ideas for what you can do with a fairly nice telescope, already in orbit, just need to pay upkeep on the ground station, that don't require the original precision of which Kepler was formerly capable.

        As long as the optics are good, and the control is not totally shot, it's still a pretty decent telescope, and people are practically shivving each other for time on good telescopes.

    • No you can't. To have 3-axis control you must have at least 3 reaction wheels with none of them colinear. With two you can only apply torque in two dimensions, specifically, the plane containing the two wheel spin axes. You cannot apply torque out of that plane.

  • a fix (Score:5, Interesting)

    by rubycodez (864176) on Friday August 16, 2013 @11:33AM (#44584021)

    deploy another gyroscope and rpg in a package that can be attached to Kepler, a practice run for comet sampling missions

    • by thed8 (1739450)
      It's time to toss the whole mechanical gyroscope concept as long as we stay with the low bid system we doom most precision work for spacecraft. The answer, imho, is move to a laster ring or fiber ring gyroscope, then find a method of translating their output to a positioning system. Easier said than done. Or fix the procurement systems for items like this but I'm afraid a mechanical system spinning at 4 or 500 rpms is never going to last forever.
  • by Anonymous Coward on Friday August 16, 2013 @12:09PM (#44584397)

    The problem is Kepler doesn't really have the resolution for things we're interested in and may not actually be able to be pointed in certain directions due to solar panel and communications relay positioning. Best we could come up with at our brainstorming meeting this morning was parallax with our upcoming mission to Pluto and to resolve a bet on whether or not ISON will explode when it passes close to the sun later this year.

  • "The future's uncertain and the end is always near."

    Sorry, Kepler.

  • This article lead me into the world of reaction wheels and their issues. Thanks /. for the education, though some data was quite hard to find out (the brand/model installed in keppler would have been useful, but I did find out it was likely an Ithaco Space Systems unit).
          From what I can find magnetic bearing reaction wheels weren't available in the torque's required by the large keppler telescope (remember it was a while ago).

  • by Squidlips (1206004) on Friday August 16, 2013 @02:00PM (#44585625)
    It seems that Ithaco Space Systems built the control wheels according to this article in Nature. They supplied the failed control wheels for Kepler and Dawn and other missions....and they were not cheap. Great job there... http://www.nature.com/news/the-wheels-come-off-kepler-1.13032 [nature.com]
  • The gyros have been replaced on four of the Hubble servicing missions. [hubblesite.org]
  • TESS will study 2.5 million stars, an order magnitude more than Kepler.
  • Spacecraft are not designed to be 2-fault tolerant, so it's good that NASA engineers have been looking for workarounds and find other uses for this spacecraft. Having said that, what is the required pointing duration for these images? Can Kepler be put into a Drift mode (disable thrusters) for the duration of an image to ensure quiescence? I understand 2 of the momentum wheels are still functioning, which would still allow for at least 2-DOF control. Depending on the image duration and level of solar pres

10 to the minus 6th power Movie = 1 Microfilm

Working...