6 ms·

Sounds like they worked through it, so docking was successful. I imagine that much like SpaceX in a similar situation, there are lessons learned that will be a

Sounds like they worked through it, so docking was successful. I imagine that much like SpaceX in a similar situation, there are lessons learned that will be applied to the next iteration.

colechristensen 2y ago
Generally you actually want 0 "lessons learned" in human spaceflight. Test flights especially and unmanned missions have a much higher tolerance for error. A bunch of thrusters failing with people on board could have been catastrophic and is well into the territory of "enormous problems which never should have happened". The astronauts may have to be "rescued" with a different capsule returning the departing astronauts to earth.
- justinclift 2y ago
  It'll be super ironic if they have to come back via SpaceX capsule. :)
- rootusrootus 2y ago
  I agree, zero would be ideal. Maybe I'm just too accustomed to the iteration we see all the time when SpaceX equipment fails. This is arguably more a traditional space program and less of a 'go fast and break things' process. IIRC, SpaceX also had some sort of glitch when docking with the station, on a crewed mission. Related to the thrusters, too, if memory serves.
  - colechristensen 2y ago
    Found this https://www.space.com/spacex-dragon-international-space-station-glitch https://www.space.com/spacex-dragon-international-space-stat... Which was a problem, but it was a backup in case of a leak that broke, it had 0 effect and it was a cargo mission not a people mission. Starliner's failure today did have a significant mission effect and had people. SpaceX had a bunch of rocket landing failures on real missions, but that didn't matter because landing was a bonus, basically nobody had done that before and everything was just fine if a rocket blew up trying to land. SpaceX has also done a bunch of test launches with new vehicles, this also doesn't matter if they fail, they're specifically launched to find failures instead of expending tons of engineering effort which is more expensive than blowing up test rockets. Those things are quite separate from "there's people on board and several thrusters broke requiring unexpected manual pilot intervention", this was not expecting or allowing for failure by design but a serious anomaly that has to be researched how the problems weren't caught in human certification.
    - inglor_cz 2y ago
      SpaceX destroys a lot of hardware, but during almost quarter of a century of operations, they had AFAIK just one work fatality - a worker was blown off his truck in McGregor in 2014. No doubt they were lucky several times, but so were others, and most space institutions have much longer fatality lists, especially when adjusted to total flight volume.
      - SrslyJosh 2y ago
        Fatalities aren't the whole story. > The 2023 records, newly disclosed by the Occupational Safety and Health Administration, also show that injury rates at some SpaceX facilities grew worse than those the company had reported in 2022. At its manufacturing-and-launch facility in Brownsville, Texas, for instance, SpaceX reported 5.9 injuries per 100 workers, surpassing its rate of 4.8 injuries in 2022 and topping a space industry average of 0.8. That's 7x the industry average. Ref: https://www.usnews.com/news/top-news/articles/2024-04-22/exclusive-injury-rates-for-musks-spacex-exceed-industry-average-for-second-year https://www.usnews.com/news/top-news/articles/2024-04-22/exc...
        inglor_cz 2y ago
        Could it be a reporting issue? Fatalities certainly aren't the whole story, but they do have one "advantage": they cannot be hushed away, at least not in the Western world. (In China or Russia, well...) That is why, in an analogical situation, crime and public safety of individual countries is often compared using murder levels. It is expected that most developed countries cannot really suppress/fudge murder statistics, even though they can do a lot of shenanigans around registration of lesser crime. I can see a similar situation with individual corporations and their diligence when it comes to reporting injuries.
        HPsquared 2y ago
        Even murder statistics aren't foolproof when comparing crime rates across different countries and time periods, because of differences in medical care standards. Someone can be shot several times in the modern West and recover, but 200 years ago or in a poor country (or maybe in a more remote location) the same person would not survive and it would count as a homicide.
bandyaboot 2y ago
Ideally you want the iterative lesson learning to wrap up before you put the humans inside.
- rootusrootus 2y ago
  Yeah, that would be ideal. But it seems uncommon. Lots of complexity in a spacecraft.
- wannacboatmovie 2y ago
  That's why they're called test pilots. They're not putting the general public into spacecraft. FWIW, like many previous astronauts, both crew members are US Navy test pilots. They've literally made a career out of solving complex problems and navigating ultra-sticky situations.
  - perihelions 2y ago
    The function of test pilots is to validate things that can only be validated by manual flight testing, like aircraft handling characteristics—not to risk their lives to do QA.
    - latentsea 2y ago
      Sometimes the tests aren't green.
      - fragmede 2y ago
        sometimes in the messy real world, things go wrong even when the tests are green.
  - b33j0r 2y ago
    Hmm. I hear ya, but this isn’t what I want test pilots spending effort on. Test pilots are closer to an integration test in software. You expect the unit tests to pass green before you expect the related integration tests to work. Multiple failures of maneuvering thrusters is technically a mission failure (in the flight plan sense), despite having nothing to do with the parameters of the mission, nor anything to do with adding humans. Maneuverability is the difference between a craft and junk. I wouldn’t find this acceptable in a project I was managing or funding. Just sayin! Still impressed as hell.
fabian2k 2y ago
The reports while it was happening suggested that they were almost down to no redundancies for the thrusters. Which would have been bad enough to disallow the capsule to approach the ISS. They managed to get enough thrusters working to still dock, but it seems that this was rather close. So even though it worked in the end, I would assume that this is still a quite serious incident and will require some examination. Also it seems that these problems are not well understood, which is not the kind of problem you want to have on manned spacecraft: > Some notes from the post-docking news conference on Starliner: Engineers found a fourth helium leak after capture. The thruster issues were similar to those observed on OFT-2, and are still not fully understood. https://x.com/SciGuySpace/status/1798827111454827003 https://x.com/SciGuySpace/status/1798827111454827003
leptons 2y ago
>there are lessons learned that will be applied to the next iteration. Maybe that's why there's 20 thrusters instead of 5.
ricardobeat 2y ago
SpaceX has a little advantage in that their next iteration doesn’t cost a billion dollars…

Sounds like they worked through it, so docking was successful. I imagine that much like SpaceX in a similar situation, there are lessons learned that will be applied to the next iteration.

colechristensen 2y ago

Generally you actually want 0 "lessons learned" in human spaceflight. Test flights especially and unmanned missions have a much higher tolerance for error. A bunch of thrusters failing with people on board could have been catastrophic and is well into the territory of "enormous problems which never should have happened". The astronauts may have to be "rescued" with a different capsule returning the departing astronauts to earth.

justinclift 2y ago
It'll be super ironic if they have to come back via SpaceX capsule. :)
rootusrootus 2y ago
I agree, zero would be ideal. Maybe I'm just too accustomed to the iteration we see all the time when SpaceX equipment fails. This is arguably more a traditional space program and less of a 'go fast and break things' process. IIRC, SpaceX also had some sort of glitch when docking with the station, on a crewed mission. Related to the thrusters, too, if memory serves.
- colechristensen 2y ago
  Found this https://www.space.com/spacex-dragon-international-space-station-glitch https://www.space.com/spacex-dragon-international-space-stat... Which was a problem, but it was a backup in case of a leak that broke, it had 0 effect and it was a cargo mission not a people mission. Starliner's failure today did have a significant mission effect and had people. SpaceX had a bunch of rocket landing failures on real missions, but that didn't matter because landing was a bonus, basically nobody had done that before and everything was just fine if a rocket blew up trying to land. SpaceX has also done a bunch of test launches with new vehicles, this also doesn't matter if they fail, they're specifically launched to find failures instead of expending tons of engineering effort which is more expensive than blowing up test rockets. Those things are quite separate from "there's people on board and several thrusters broke requiring unexpected manual pilot intervention", this was not expecting or allowing for failure by design but a serious anomaly that has to be researched how the problems weren't caught in human certification.
  - inglor_cz 2y ago
    SpaceX destroys a lot of hardware, but during almost quarter of a century of operations, they had AFAIK just one work fatality - a worker was blown off his truck in McGregor in 2014. No doubt they were lucky several times, but so were others, and most space institutions have much longer fatality lists, especially when adjusted to total flight volume.
    - SrslyJosh 2y ago
      Fatalities aren't the whole story. > The 2023 records, newly disclosed by the Occupational Safety and Health Administration, also show that injury rates at some SpaceX facilities grew worse than those the company had reported in 2022. At its manufacturing-and-launch facility in Brownsville, Texas, for instance, SpaceX reported 5.9 injuries per 100 workers, surpassing its rate of 4.8 injuries in 2022 and topping a space industry average of 0.8. That's 7x the industry average. Ref: https://www.usnews.com/news/top-news/articles/2024-04-22/exclusive-injury-rates-for-musks-spacex-exceed-industry-average-for-second-year https://www.usnews.com/news/top-news/articles/2024-04-22/exc...
      - inglor_cz 2y ago
        Could it be a reporting issue? Fatalities certainly aren't the whole story, but they do have one "advantage": they cannot be hushed away, at least not in the Western world. (In China or Russia, well...) That is why, in an analogical situation, crime and public safety of individual countries is often compared using murder levels. It is expected that most developed countries cannot really suppress/fudge murder statistics, even though they can do a lot of shenanigans around registration of lesser crime. I can see a similar situation with individual corporations and their diligence when it comes to reporting injuries.
        HPsquared 2y ago
        Even murder statistics aren't foolproof when comparing crime rates across different countries and time periods, because of differences in medical care standards. Someone can be shot several times in the modern West and recover, but 200 years ago or in a poor country (or maybe in a more remote location) the same person would not survive and it would count as a homicide.

bandyaboot 2y ago

Ideally you want the iterative lesson learning to wrap up before you put the humans inside.

rootusrootus 2y ago
Yeah, that would be ideal. But it seems uncommon. Lots of complexity in a spacecraft.
wannacboatmovie 2y ago
That's why they're called test pilots. They're not putting the general public into spacecraft. FWIW, like many previous astronauts, both crew members are US Navy test pilots. They've literally made a career out of solving complex problems and navigating ultra-sticky situations.
- perihelions 2y ago
  The function of test pilots is to validate things that can only be validated by manual flight testing, like aircraft handling characteristics—not to risk their lives to do QA.
  - latentsea 2y ago
    Sometimes the tests aren't green.
    - fragmede 2y ago
      sometimes in the messy real world, things go wrong even when the tests are green.
- b33j0r 2y ago
  Hmm. I hear ya, but this isn’t what I want test pilots spending effort on. Test pilots are closer to an integration test in software. You expect the unit tests to pass green before you expect the related integration tests to work. Multiple failures of maneuvering thrusters is technically a mission failure (in the flight plan sense), despite having nothing to do with the parameters of the mission, nor anything to do with adding humans. Maneuverability is the difference between a craft and junk. I wouldn’t find this acceptable in a project I was managing or funding. Just sayin! Still impressed as hell.

fabian2k 2y ago

The reports while it was happening suggested that they were almost down to no redundancies for the thrusters. Which would have been bad enough to disallow the capsule to approach the ISS. They managed to get enough thrusters working to still dock, but it seems that this was rather close. So even though it worked in the end, I would assume that this is still a quite serious incident and will require some examination. Also it seems that these problems are not well understood, which is not the kind of problem you want to have on manned spacecraft: > Some notes from the post-docking news conference on Starliner: Engineers found a fourth helium leak after capture. The thruster issues were similar to those observed on OFT-2, and are still not fully understood. https://x.com/SciGuySpace/status/1798827111454827003 https://x.com/SciGuySpace/status/1798827111454827003

leptons 2y ago

>there are lessons learned that will be applied to the next iteration. Maybe that's why there's 20 thrusters instead of 5.

ricardobeat 2y ago

SpaceX has a little advantage in that their next iteration doesn’t cost a billion dollars…