Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
NASA is about to make its most important safety decision in nearly a generation (arstechnica.com)
46 points by rntn on Aug 12, 2024 | hide | past | favorite | 64 comments


This should be an easy choice for NASA. The whole point of having different vehicles available is to make this kind of choice easy. With Dragon available, there should be no pressure to use Starliner. That this is apparently controversial shows that NASA has lost the plot.


Yes. But it's quite possible that NASA's deciders made the real "Sorry, Boeing!" decision a while back - and are letting things seem to drag out for the sake of appearances. (No "uninformed/rushed/secret decision", nor "anti-Boeing bias", nor whatever.)


I wonder at what point the astronauts can veto NASA on this. In airplanes, the pilot in command has the last word about whether to take an aircraft into the air.


They're up there. NASA isn't. Technically, they can veto them at any point.


Didn’t that happen with Skylab?


Given all that we know, that system shouldn't have flown to the ISS.

Sincerely hoping NASA plays it safe this time.

A tragedy would have far-reaching consequences for both NASA and the US, well beyond the horrifying loss of life.


It appears to me there's probably a "best of both worlds" compromise:

1. Send the astronauts home on Dragon

2. Certify Starliner for human flight if it can make it home safely (uncrewed)

If the capsule can make it down, it doesn't matter if there were people in it or not, right?


They cannot do it at the moment, as Boeing has removed the software support for flying Starliner uncrewed from this specific craft, even though it has already completed a mission without humans onboard already. And we’ve only learned about it after Starliner was stuck docked to ISS


NASA already clarified this is not correct, only software parameters need to be modified not the software


Interesting, I haven’t seen this news. Would you happen to have a link?


This is in the story:

* It will also take time for NASA and Boeing to update parameters in Starliner's flight software to enable an unpiloted undocking and reentry.*

It’s apparently a file called the “Mission Data Load” (MDL).


Thank you for pointing this out


Is there a mechanism for updating the Starliner software while it's docked at the ISS?


Put a microSD with the firmware update in the slot for the navigation system card. Then restart the system and hold down the "Play" button while restarting until the screen turns blue. System may reboot several times afterwards, when everything is done, the system will startup as usual.


Yes, and in the story it talks about how that updating is what would cause the longest delays in getting Starliner off ISS if they go that route. So yes, it’s possible.


Presumably, yes. But making such major alterations to the flight software might be considered an unacceptable risk to the ISS.


The point of certification isn't and shouldn't be "happened to be successful once"


For context on cost, SpaceX only charged NASA $3.1 billion for developing the Dragon 2 manned capsule. Boeing charged NASA $6.7 billion for Starliner and has spend at least 1.6 billion more than that so far.


If NASA decided that Butch & Suni would be flying SpaceX on their return trip, that would probably allow more aggressive testing of the Starliner's thruster systems after it departs the ISS.

In theory, the ISS crew could do a spacewalk or two before departure, to get a closer look at the problematic hardware. In practice - no time to plan nor train for that, improvised-at-best tools, and all the myriad ways that a flaky NTO/hydrazine fuel system could suddenly go very wrong - I'd be really surprised if they attempted anything.


Closely related story -

https://arstechnica.com/space/2024/08/with-starliner-stuck-i...

- by a journalist who seems very well connected with NASA folks. And who seems to be dropping a fair number of "they're fly home on SpaceX" hints.


Can someone spell out the consequences of a failed thruster? Eg miss the reentry cone and skip off the atmosphere into space?


> Can someone spell out the consequences of a failed thruster? Eg miss the reentry cone and skip off the atmosphere into space?

With no thrusters Starliner can’t orient itself during reëntry, which almost guarantees burning up. I’m not sure what the minimum number of thrusters Starliner needs to safely deörbit.


> deörbit

Do you have this set as autocorrect or do you type the ö each time?


isn't this a "new yorker styleguide" sort of idiosyncrasy?

EDIT: "Languages such as Dutch, Afrikaans, Catalan, French, Galician, and Spanish make regular use of the diaeresis" according to wikipedia.


Only for repeated vowels, which deoörbit is not


Excessive diaeresis is usually caused by a vowel problem.


Explosive diaeresis is the woooorst.


It's a diphthong though. Same as the difference between George (roughly dzo:rdz) and Geörge (roughly dze:ordz).


> We apologise for the fault in the subtitles. Those responsible have been sacked.


> or do you type the ö each time?

It’s easy on mobile and less so on my Mac, so I do it more on mobile. I grew up writing British English (and French), in which reorbit and cooperate are a thing, but they get systematically misread in American English, and re-orbit and co-operate look messy.

I’ve also noticed when I add a diaresis my iPhone has less of a tendency to fuck with the word after it’s been typed.


> miss the reentry cone and skip off the atmosphere into space?

My layman understanding is that this was never an option even for the Apollo trips from the moon - the problem was that a wrong re-entry angle wouldn't slow the capsule down enough to bring it down to the surface, and it would have to orbit several more times before finally slowing down enough to return.

This was a problem for two reasons: (1) the landing site would be unpredictable, and it wouldn't really do to land on the Kremlin, or in the middle of the Sahara, or in the middle of the Pacific. (2) The capsules didn't have enough supplies to keep the astronauts alive for too long, and too many orbits would mean that they'd be dead on arrival.


Incorrect angle also leads to higher temperatures that may exceed design capabilities during re-entry


In Carrying the Fire, in pre Apollo days when they did manual burns for re-entry, this was described as a big worry (burn sending you off into space). But maybe that was mitigated


There are several issues. One is that the craft doesn't properly separate from the station. If it backs away from the station, then dies, it will risk colliding with the ISS after an orbit. (Lower perigee but same apogee = collision potential.) The craft needs to back away from the station sufficiently to reenter or at least conduct a second maneuver to ensure its apogee is lower than the ISS's orbit.


My understanding (which may be wrong) is that they don't really understand how it's failed at this point so it could be anything from "no thrust" to "explodes and sends debris into the ISS".


Or come in too steep and burn up. Or miss the target drop zone and possibly come down in the wilderness or ocean (it's meant to land on flat and level land.)


That's one possibility.

Unfortunately, failure modes are not limited to "the thruster(s) shut down before the desired burn was completed".

Worse case looks kinda like this, but in space: https://www.cbsnews.com/news/spacex-explosion-destroyed-crew...


In God we trust. All others must bring data.


why take the risk.


> ...maintain a clearer future for Boeing's Starliner program, which NASA would like to become operational for regular crew rotation flights to the station.

Why do people in the government bend over so deeply backwards for this shitty company?


With Soyuz out of the picture, if Boeing can’t get it together aren’t they basically 100% dependent on SpaceX?

And the head of SpaceX is, shall we say, not giving off strong vibes of calm consistency right now.

Single sourcing is never great. The article even mentions the problem of when the astronauts got stuck in space on ISS for a very long time after Columbia.

While Boeing is strategically important, some of this may simply be there are no other good options out there right now for a second source.

We don’t want to depend on China. We’re not on good terms with Russia. Is there anything else available?


The European Space Agency has a space program as well: it has the Ariane rocket. I don't think it's capable to bring astronauts to the ISS but maybe it would be worth starting to cooperate to enable this in the future.


There's several issues involved there. For starters a "human rated" rocket needs to operate within certain performance envelopes and have certain features. For instance the rocket carrying a human crew will have different vibration needs than one carrying cargo. The rocket also shouldn't vent cryogenic gases to the atmosphere while on the launch pad which creates a danger for astronauts and a ground crew when the rocket is fueled.

The Ariane hasn't ever really gone through a human rating process and doing so would take several years. Even then, the ESA does not have a crew launch vehicle available. The Automated Launch Vehicle was used to resupply the ISS but can't carry a crew despite having a habitable volume when it was docked to the station.

The ATV doesn't land but burns up in the atmosphere at the end of its mission. That's not a vehicle astronauts would really want to ride home in. While I don't doubt the ESA could develop a manned vehicle, it would just end up being so different from the existing ATV that it would take longer to develop than the remaining life of the ISS.


That makes sense. I don’t follow this stuff closely but o didn’t think they were moving people yet.

That would be a great alternative but they could end up in a situation where I’m once it’s ready it takes a while to shake out too.

So the US may not have that option for quite a while. Or maybe they can’t launch frequently enough when they’re ready.


What's the worst-case scenario they're trying to avoid from being 100% dependent on SpaceX? Perhaps it's something like the present scenario. If the vaccine is as bad as the disease, then skip the vaccine and just chance it.


Huh? Isn't the right interpretation that they are 100% dependent on SpaceX right now, in the present scenario, but simply pretending otherwise? The facts are what they are.

The reason one spaceship is better than another isn't a management or strategy problem. It is what it is. There's no decision here that changes whether or not Starliner is a shitty spaceship. It might look like it does.

Is NASA management is personally going to be assembling and running the plans? One interpretation is that they will now (?). Another is that is already happening and they are sucking at it, and that is SpaceX's opportunity in a nutshell.

If the government wants better spaceships, it has to occasionally buy them in politically unideal circumstances. It is that simple. Sometimes it will take longer to turn some senator's state into a manufacturing powerhouse than we will have political appetite to sustain all the failures on that journey. The fact that it's not an economic problem, that no amount of money can turn around Boeing, is bad, and that's why I am surprised. Any honest interpretation of the facts makes NASA, a bunch of civil service people, look like they are delaying the inevitable to benefit no one.


> And the head of SpaceX is, shall we say, not giving off strong vibes of calm consistency right now.

What is your logic?


> What is your logic?

That is precisely the question many people would like to ask the head of SpaceX.


Why?


Pretty clear, I thought.

Having your entire access to space controlled by one company is a gamble. If that one company is controlled by one man, that's more of a gamble. If that one man is showing signs of mental instability, that's an even bigger gamble.


You would know based on your physical interaction with said man?


I believe that this page answers your question: https://www.boeing.com/defense#products


NASA's charter and total function is to promote the commercialization of space.


F*ck NASA and their "Starliner."

Soyuz is the most reliable and most successful vehicle at this point.

NASA simply does not want to use it because they will lose face.


You think they should use the launch vehicle of the country in a war where we’re literally arming the other side?

They would 100% use Soyuz if Russia hadn’t changed the relationship by starting a war to take over a sovereign country.


You're very confused, I wonder where you're getting your "facts" from. NASA hasn't stopped flying American astronauts on Soyuz. The most recent Soyuz launched March of this year with Tracy Caldwell Dyson, an American astronaut, on board. It comes back in September, with her. Another is scheduled to launch in September, with Americans in the primary and backup crews.


Thanks for the insult. You could have decent instead of assuming I’m some nutcase.

I thought we weren’t using Soyuz anymore. I didn’t remember hearing about it and given the geopolitical situation that didn’t surprise me.

I’m not a major space buff who keeps up with all the launches and schedules. I thought I knew the current situation. I guess I was wrong.

A simple note that I was incorrect would have easily sufficed. I’d have thanked you for the info.


Throughout this thread you're trying to intuit knowledge by projecting your own biases, instead of simply looking things up. NASA still using Soyuz is something you easily could have known, you have access Wikipedia and it takes less than a minute to pull up the list of Soyuz missions, but instead you made up some story about NASA being motivated to use Starliner by a supposed inability to use Soyuz, something you incorrectly assumed, about a topic you aren't even familiar with. It's little better than your other theory that the government is scared of relying on SpaceX because they think Elon Musk is unstable. That's your bias which you are projecting onto the government. The reality is the government is already committed to and planning more missions which require the use of SpaceX hardware. They don't care that he talks shit on twitter, the services that SpaceX provide to the government are consistently world class.

As for insults, I didn't call you a malicious idiot, I said that you were confused. You were confused. That wasn't an insult.


The SpaceX vehicle may look like the safest choice. But is it? A handful of manned flights does not prove that a vehicle is 'safe'. Likewise, some anomalies in 1 flight do not prove 'unsafe'.

Perhaps the knowledge of Starliner issues makes it more accurate to assess the risk.

Either vehicle could lose the crew on the return. "Astronauting" is a risky business.

I do not envy anyone on the decision chain.


One vehicle has demonstrated it can return crew safely, the other has never done that and has exhibited problems on what should have been a demonstration flight (small issues, perhaps, but nothing as critical or questionable as they've found).


space shuttle looked safe for the first 24 launches. not shilling for Boeing. just pointing out that space is dangerous. and a couple of successes does not make the risk go away.


Nobody claims that the Dragon capsule is guarantee to be perfectly safe. The claim is that given the available evidence and track records, no rational assessment of risk would rank Dragon as more risky than Starliner.


Some anomalies in the first half of the first flight - some of which are directly relevant to its ability to successfully conduct the second half of the first flight.

No, that doesn't prove 'unsafe'. It gives reason to seriously question it, though.


This is ridiculous boeing apologism. the starliner craft is broken, and only semi-operable on it's maiden voyage. the spacex craft has flown succesfully multiple times without any incident, and they have data from multiple trips which was used to analyze and improve the flight. boeing does not have that.

There is a very clear safer option, and to pretend there isn't is enlightened centrism.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: