NASA Astronauts Return Home in SpaceX's Crew Dragon Spacecraft [video] | Hacker News Reader

NASA Astronauts Return Home in SpaceX's Crew Dragon Spacecraft [video]

youtube.com

405 pointsjbredeche96 comments

Not yet, not until re-entry. It's arguably the most dangerous part of the missions since launching on Falcon 9 has happened many times before.

Both Soyuz and the Space Shuttle had fatal accidents on re-entry.

I don’t believe SpaceX has ever not tried and succeeded to return a Dragon sent to ISS back to Earth.

That is true, but it has only had one successful reentry (one out of one attempts) of a Dragon Crew capsule design (which, granted, is only slightly different than Dragon Cargo).

For clarity: I think "tried and not succeeded" is the intended phrasing.

I meant “not (tried and succeeded).” English is great.

You're not wrong, but I think it's cool how much safer the generic capsule design is than something like the Shuttle. Really the Shuttle got a whole lot wrong—heat-intolerant aluminum structure and critical, sensitive machinery directly beneath a layer of tiles so fragile you can crush them in your hands (IIRC).

SpaceX's Starship will probably be a lot more tolerant to defects in the heat shield because of the relatively heat-tolerant steel underneath, but landing safely relies on the operation of massive aero surfaces and a relight of the engines with thrust vectoring. It'll be interesting to see how that goes.

This is at the stage where you cant really make assumptions about what may happen because you don't know what could go wrong. (A.K.A you don't know what you don't know).

I'm not claiming that OP or I are experts, but there's actually a pretty big body of knowledge for capsule re-entry. It's more 'known problems' than 'unknown unknowns'.

All you need to bring a traditional capsule design down to ~sea level terminal velocity safely is a blunt body with offset CG and the kind of circular ablative heat shield that's been shown over decades to be simple, safe and reliable. You don't even need active control unless you want more precision for hitting your landing target. The biggest risk, assuming your position and velocity were correct at interface, is the parachutes–which is why SpaceX took so much time testing and refining them.

Things can always go wrong, but we know enough about both classes of vehicle to be able to say that the Shuttle's design was in general terms both more complex and more risky.

This is slightly off topic, but the Shuttle was horribly unsafe on the way up as well as on the way down. Using solid rocket boosters with no means to get the crew safely off the stack until they stop burning is, for lack of a better word, demented. The Crew Dragon/Falcon 9 stack offers realistic abort modes from the moment fueling begins until the vehicle reaches its intended orbit.

The Shuttle was built to bring down soviet spy satellites and spacecraft, so in that regard, it was the best design made.

Huh interesting. Do you have sources for that information?

The main concern of the Soviets was that the Shuttle could swoop down and drop bombs without warning.

(Both Reagan and the Soviet top leadership didn't care about MAD, they cared about practical methods to stop incoming missiles. To get the arms limitations agreements signed, Reagan allowed the Soviets an exception to the anti-ABM ban to allow a ring of anti-ABM sites around Moscow, which exists until today. The "Star Wars" kinetic interception methods are also likely still active, though the laser methods probably never worked well.)

There was at least one mission to retrieve a satellite (not Soviet):

https://www.nytimes.com/1990/01/07/us/space-shuttle-poised-t...

I don't have access to classified military secrets, but you don't think they would build a vehicle that could return 15 tons of payload to earth to retrieve US satellites? US could just build and launch a new satellite easier and cheaper than bring one down to fix only to launch it again.

Obviously there is the movie about Salyut 7 that depicts the shuttle trying to capture the station. Salyut 3 had a machine gun attached to it.

Back in the 70s spy satellites used film cameras and used to return the film in re-entry capsules. There’s a _really_ good (fictional) move called Ice Station Zebra about a retrieval mission in the arctic that goes wrong.

Retrieving the satellite allows you to safely recover the film, and re-equip the satellite with new film and propellant. The latter is important because it’s useful for spy satellites to be able to change their orbits frequently to observe new targets or evade tracking.

Also bear in mind some spy satellites are huge and incredibly costly. The Hubble Space Telescope is based on a spy satellite series.

I read something about naval engagements with soviet ships. The US ships were trying to find something underwater before the soviets did while at the same time not alerting the soviet ships as to the location of what they were trying to find. It became clear they were both looking for one of these film canisters that maybe they weren't able to retrieve in mid-air. People today can't imagine the stakes during the cold war.

It was not built to retrieve Soviet satellites, despite an uncited claim in Wikipedia to the contrary. In fact, it probably couldn't. If nothing else, it'd be hard to fit a satellite designed for a different launch fairing into the payload bay of the shuttle and secure it well enough to be able to land with it. The shuttle was designed to bring cargo back, but only satellites that were specifically designed for this purpose.

What the Shuttle was designed to do in a military context is launch into a polar orbit from the coast of California, release a stealthy spy satellite, and return to Earth immediately after during the very same orbit (so about 90min from launch to landing). It therefore needed the very large wings to be able to glide back to Edwards Air Force Base in southern California, which would have rotated up to 1,500km away from the path of the orbiter. And because of the Pacific ocean, there'd be other landing sites available. This cross-range capability impacted every aspect of the design.

The idea was that big rockets which launch satellites are very easy to track, so it was hard to get a bird up without the enemy knowing its orbital path, and if the enemy knew the orbit they could evade overhead passes. So do a surprise launch on the shuttle, do a little orbit maneuvering while you are up there to maximize the enemy's uncertainty over its orbital plane, then immediately let go and land the shuttle--because otherwise they'd know where it was from tracking the shuttle.

I'm not sure if it was ever used this way, and the requirement is now obsolete anyway. You really can't hide a satellite from modern instruments, which have full global coverage, and its critical that orbits are known to avoid collisions in the much more crowded space up there.

Is the X-37b's orbit known? I dimly recall articles about observers catching glimpses of it places away from its predicted trajectory. Could it maybe be placing/observing/servicing satellites? Why's it a shuttle-style design?

I feel like it's not too outlandish to suspect that some military satellites might have a way to hide themselves from instruments, as well.

Known to the Russians and Chinese? Almost certainly.

The X-37B is a shuttle-like design because it was designed for NASA to be a crew escape vehicle for the ISS. The Air Force only took over the project when NASA cancelled it.

The idea that you could capture a satellite and return it was part of the mission requirements for the original shuttle design. If it was a 'friendly' or 'unfriendly' satellite was left unspecified.

It was also limited by the Shuttle's OMS in terms of delta-v so the orbits from which a satellite could be retrieved was constrained.

Unfriendly satellite can blow up or cause fire inside of shuttle, so it's bad idea to capture enemy satellites. It's like sending billion-worth mine capturer behind enemy lines, which can be destroyed with AK.

Don't see why you're downvoted for posting this. Certainly, satellites designed before the Space Shuttle would have to be very paranoidly designed to include a self-destruct mechanism, but it only takes a couple kilograms of explosives to strongly disincentivize kidnapping it.

If I was a military satellite designer, heaven forbid, I'd definitely include that in a world that included the possibility of stealing satellites. The capability, quietly communicated between the powers, would be enough to ensure that the satellite would never be stolen. And therefore, the billion-dollar kidnapping capability becoming useless.

Even without self-destruct, a satellite not programmed to be captured could simply fire some thrusters if it detects an orbit change, and that would likely make some big dammage inside the capturing shuttle.

As part of just normal station keeping it might do that.

That's typically ion thrusters, isn't it? That's very low mass at very high speed, and would probably take hours to even leave scorch mark. It would feel like a slow breeze.

Of course, military use could do something different, but that's more along the lines of "pack a kilogram of explosives in a pipe bomb".

Hydrazine thrusters more likely. Ion thrusters are novel.

Even without that inclusion you could still communicate it :) Save some money and more payload.

I agree with you that hostile satellite recovery would be fraught with challenges. That said, I expect there are people who 'war game' these scenarios all day and probably have various counters or techniques to prevent damage to the shuttle or its bay.

Seems like it came in handy fixing the Hubble multiple times, but what your talking about was probably overkill for what they ended up using it for.

Including features that are never used is bad design even if they don't compromise every other aspect of the design.

What mainly doomed the shuttle was the DoD requirement that the shuttle be able to pluck a satellite from a polar orbit in one orbit, making it necessary to include the massive wings so that it wouldn't be forced to land in the Pacific as the world rotated underneath it.

If they’d used those features it wouldn’t be public knowledge

Yes it would. There's no such thing as a secret satellite, it's impossible to hide them. General astronomy would notice that their's one less satellite.

> There's no such thing as a secret satellite

No known secret satellites :)

https://en.wikipedia.org/wiki/Misty_(satellite)

You absolutely could not conceal a polar shuttle launch. No such thing ever happened.

Perhaps because it existed, it never had to be used. Deterrence is a feature. An ounce of prevention, as they say.

Considering the complexity and requirements it is amazing that it was as successful as it was.

This is a bit of a partial myth. This idea was probably used when negotiating for funding in some quarters, but it was never a particularly plausible use of it, for at least two reasons. First, of course, the shuttle can only access extremely restricted orbits which most targets of interest would not be in. And second, countermeasures would be trivial if the Soviets believed such a threat to actually exist; just incorporate a few kilos of explosives into each device.

I'm reasonably sure this story was always more of a case of "how do we sell this thing" than "this is something that we can actually realistically do".

Dragon has come back before (without crew) so this isn't a total unknown.

After reentry the next most dangerous part of the entire launch/return process is likely landing at sea and being rescued by helicopters or ships.

So I will be holding my metaphoric breath until the astronauts are delivered from SpaceX responsibility to NASA responsibility, which should be on land somewhere in Florida (too lazy to look it up sorry not sorry)

The Crew Dragon is very similar to the Cargo Dragon that has already had nearly 20 returns from orbit without incident, so there is already quite a bit of experience in the systems that will be used tomorrow.

Mods, a better headline would say "returning", or that they've departed the ISS. They haven't returned to the surface yet.

I can't easily link to it because it's still live streaming but at 4:10 in they seem to be debugging a local internet/HTML error on their iPads. Next level debugging in production.

I just had to go find that for myself! Here's the link to 4:10:45 when they start talking about it - https://youtu.be/13OkD0C_TWU?t=15045

Related, I wonder if there's going to be a postmortem on this mission on the UI, because from the brief scenes when its shown updating or being clicked on, it looks to have a responsiveness on the order of half a second.

Think it’s Electron. Doesn’t surprise me. I wouldn’t want to take slack into space with me.

I'm fairly certain there's no Electron on iPad (can't take Chrome to iPad). I also don't think that a native wrapper app that loads (local) html files would display a "Safari" error message.

It was probably just a cached/saved website saved to the home screen. I'm not a rocket scientist, but as a web developer, I would rely upon that for anything earth-bound, let alone in Space.

> I'm not a rocket scientist, but as a web developer, I would rely upon that for anything earth-bound, let alone in Space.

...would?

would NOT. Yikes.

It is reportedly Chromium-based, not quite electron. UI is HTML with web components. Lots more details here: https://old.reddit.com/r/spacex/comments/gxb7j1/we_are_the_s...

Just to clarify, the error they were seeing was in an app running on an actual portable iPad they carry, not the main screens mounted in front of the seats.

I love how the ground personnel are checking that that iPad is in "airplane mode". The Dragon isn't an airplane so that might be their problem. ;-)

Undocking stream ran earlier and can be found at https://youtu.be/13OkD0C_TWU, with the undocking procedure beginning 2 hours 14 minutes in. The actual separation is at 2 hours 19 minutes.

Currently there is a live stream of the autonomous “return coast” phase while the crew sleeps: https://youtu.be/zMsxviPT2Cw

The splashdown stream is at https://youtu.be/tSJIQftoxeU and starts at 4:25 AM PDT, with deorbiting expected around 10:55 AM PDT.

The first comment I saw in the live chat was "You people have it wrong, space is flat, not earth".

Well from what I’ve read the observable universe is fairly flat, but I was wondering what if we’re just living in a basin or on a plateau that extends beyond our observable reach?

I suck at math though so I’ll just wait for theoretical astrophysicists to tell me interesting things on blogs.

I’ve always pondered a mobius strip with a river-like motion .

Would that be a mobius strip of universes? That makes me think of the central finite curve from Rick and Morty.

Then we can all say we live "in the Valley"!

We all live in the valley on this blessed day

Estimates set the lower bound at 250x the size of the observable universe, according to the limits of measurement of the "flatness" of our local space. https://arxiv.org/abs/1101.5476

It’s more like “matter tells spacetime how to curve, spacetime tells matter how to move”.

Brilliant synopsis. Kudos!

It's a famous quote from John Wheeler.

Ah. TIL.

Which means the universe is really just spacetime arguing with itself.

The SpaceX crew land in Florida at 2:48 p.m. ET on Sunday.

Or 19:48 for those of us in the UK.

NASA/SpaceX really should start using UTC for their timestamps.

I feel like it's not at all weird to say that something is arriving somewhere in that somewhere's local time zone.

But, NASA has a global audience. At least have UTC as a secondary time.

Plus, the ISS' timezone is also UTC.

ISS circles Earth every ~1.5 hours, which is ~16 times per day, so it does makes sense to have UTC. Land however is fixed, with its own timezone. I second another answer, local timezone primary and UTC secondary is best.

I agree with local and UTC, as long as both are there. Local is useful for locals, but the audience of a landing is pretty much global.

More importantly, that's the timezone Mission Control is in. There are no timezones in low Earth orbit because you're orbiting every 90 minutes, so the only timezone that's relevant to anyone involved in the Mission is the one that all the hundreds of ground personnel are using (in Mission Control and to a smaller extent on the recovery ship).

From my understanding, mission control usually doesn't use their own local time. The choice of time depends on the mission, but common choices are UTC or Mission Elapsed Time. The ISS missions use UTC, since the ISS is jointly controlled by Russia and the US, and neither would agree on using the other's local timezone(s), so they use UTC.

Stuff published to the public is not coming directly from mission control, it is being filtered through NASA PR and SpaceX PR. And PR often tries to simplify things for their audience, like converting UTC (which the average person isn't very familiar with) to local timezone. (And sometimes NASA PR also likes to turn metric into US customary, although SpaceX PR does that less.)

The primary target of NASA and SpaceX's PR is American voters, because it is American taxpayer money that pays for all this. Putting things in terms which makes it easier for an American to understand, but harder for a non-American, makes sense in that context.

Re-entry happening some time tomorrow. Cool to see the ISS separation though!

I waited all day to see it live but then I had to take the cat in from outside. And missed the live undocking. Plus I got bitten.

Really glad I got to drive out and watch this launch.

It was fantastic seeing us send people back into space again.

Perseverance also launched a few days ago: https://mars.nasa.gov/mars2020/

Hopefully the landing is just as smooth as Curiosity's.

It would be really cool to see actual, high quality footage of it, but at least there's the simulated renderings. Extremely impressive.

There are multiple landing cameras this time.

*are returning home. They haven’t made it back yet. They are on track to land 11:48am PT on Sunday

What's the link for the technical feed? Right now someone is filling time with a history of the ISS.

Hit the rewind button to see the departure. All of the "action" is pretty much done for today. They are now just floating through space waiting for the next burn. Tomorrow afternoon will be the re-entry an splash down activity

The ship departs from the station at 2 hours 19 minutes.

The video seems to be no longer live, so times have changed (?). Undocking now starts at 6m45s: https://youtu.be/13OkD0C_TWU?t=400

Physical separation is at 11m35s: https://youtu.be/13OkD0C_TWU?t=695

What happens in the event of a parachute failure? Would they be in a position where they could fire up the Super Dracos for a propulsive landing?

They originally designed Dragon 2 for propulsive landing, but NASA didn't approve it, and then development focus shifted to Starship. I don't think anyone (outside of SpaceX) knows whether the feature still exists for emergency use; perhaps they never finished the software.

There are four parachutes. It's designed to support at least a single-chute failure, I don't know about multiple chute outages.

As for propulsive landing - the latest I can see[1] is that while the craft is technically capable of propulsive landing, it's not approved by NASA, even for emergencies. So it's likely that these options are not even available for the crew, and they wouldn't have trained on how to use it/activate it.

[1] https://www.teslarati.com/spacex-crew-dragon-emergency-landi...

Not even for emergencies? Does that mean that it is officially not there, but in the unlikely unlikely event it can be used, or that it really isn't there anymore and the capsule would just drop dead on the ground?

It is there, just not certified for this use. However, if they manage to make a successful ememgency propulsive landing (even an unaccounted one), it will be investigated and then approved post-facto. The engines used for landing are the same engines that are used for post-second-stage maneuvreing.

It means it's never been tested and there's no safe way to use it - the hardware and software are not configured to have that ability for this particular flight, is how I understand it.

> Not even for emergencies?

Clearly not. NASA was already not keen about the superdraco system. The explosion that destroyed the Dragon 2 capsule that was slated for the In-flight Abort test killed any chance of it being used for human-rated flights.

> Does that mean that it is officially not there

This is life-critical hardware/software. You don't go putting functionality in there that isn't approved by NASA, and most importantly the crew who're going to fly it.

Finding out that there's undocumented features and easter-eggs, even if they might in some very remote scenario help save life is certain to be outweighed by the risk that having it in there might somehow contribute to someone's death.

Could it be used for propulsive landing? Perhaps, but they'll first have to prove to NASA that it's safe - just like they proved to NASA that reused launch vehicles were safe, and reused Dragon 1 capsules were safe.

SpaceX seems to have given up on the idea of it anyway. They deleted landing legs from the capsule, and propulsive landing at sea as an emergency backup for the failure of all four parachutes (and you'd need to include a way to detect the failure, cut the chutes and then use the superdraco thrusters) seems a shitload of work which is probably going to only increase risk.

For propulsive landing of humans/cargo, it seems Spaceship their focus.

Further reply, this question was just asked and answered on the new live-stream.

Official response from SpaceX: Superdraco thrusters are completely inhibited once the launch is sucessful.

So, no, not even as a backup.

Also, one other thing I remembered - the draco and superdraco thrusters require different pressurisation levels for their fuel systems. Before launch, the system is put into a high pressure mode, to enable launch escape. After they make orbit, the system is depressurised back to enable the draco thrusters to operate.

So they'd need to have time to re-pressurise the system, too.

Repressurisation hardware has to exist anyway to have restartability (it takes 15 minutes to switch to high pressure feed, 15 seconds to low pressure feed). Do you know which component has inhibition affected? SW only? Blown fuse? Something different? Certainly not a jettisoned part, because this vehicle is planned to be reused after minor refurbishment.

PS re repressurisation failures: If it becomes absolutely necessary, a procedure could well be devised to operate the larger thrusters at a lower flow rate, albeit with a diminished efficiency and thrust.

> Do you know which component has inhibition affected? SW only? Blown fuse? Something different? Certainly not a jettisoned part, because this vehicle is planned to be reused after minor refurbishment.

They didn't elaborate, it was a 30 second response.

We're not talking about the kind of system where you just turn on the rockets and it'll probably work okay here. Propulsive landing based on running 8 rockets is gonna take a ton of highly tested hardware and software, including speed and orientation sensors etc and all the software to balance things out perfectly.

No way they're adding in a big pile of extra untested software and hardware for an unlikely event. NASA would never approve, and correctly so, as that would be a massive number of extra things to potentially go wrong.

I believe it to be no negative impact with one parachute failure and survivable with up to two parachutes of four failing. Down to just one parachute is probably bad news, and if they all fail then you die. Just like every other capsule.

Propulsive landing was not qualified, nor likely even finished, so I very much doubt it shipped. An untested propulsive mode is a significant liability.

Reply on news.ycombinator.com