Abort options for Starship and Starship/SuperHeavy

[dglow]

“Will Crew Starship have abort capability?”

“Not if they design it to not have abort capability! I’m smart!”

Seriously, this isn’t like you. I hope everything is alright.

Sorry, I’m fine.

Good to hear, be well.

[Jimmy_C]

I don’t think Starship’s engines can be relied upon to work correctly during an abort. There is a lot of energy in an exploding booster. Starship’s engines are complex and big. There’s a significant chance that shrapnel causes them to break. They aren’t simple like a solid motor nor small like Crew Dragon’s or Starliner’s abort motors. I’d consider the entire bottom half of Starship unreliable in fact. And there are no blowout panels or heat shielding to prevent the exhaust from Starship from igniting the booster’s propellant. I’d suggest a different option for abort than Starship’s propulsion section.

[chopsticks]

"Blowout panels". I'm really weak on rocket history, does someone have an example of where this sort of thing has previously been used or is it just a concept at this point?

[DanClemmensen]

"Blowout panels". I'm really weak on rocket history, does someone have an example of where this sort of thing has previously been used or is it just a concept at this point?

I'm very definitely not a rocket engineer. Expendable rockets that use "hot staging" have an open-lattice interstage that lets the upper-stage exhaust vent to the sides when the second stage motors ignite. We don't want to do this for normal staging or during EDL, but we do want to do this in an abort. Solution: cover the lattice with panels that will "blow out" if the engines are lit before the stages separate. Until there is an emergency, this interstage looks and acts like the skirt on an ordinary SS.

[Robotbeat]

I don’t think Starship’s engines can be relied upon to work correctly during an abort. ...

That's why it's designed to work with engine-out.

Look, all these things are risk *mitigations*. Even a regular old launch abort system does not work 100% of the time.

[HMXHMX]

"Blowout panels". I'm really weak on rocket history, does someone have an example of where this sort of thing has previously been used or is it just a concept at this point?

[chopsticks]

"Blowout panels". I'm really weak on rocket history, does someone have an example of where this sort of thing has previously been used or is it just a concept at this point?

Thanks! Exactly what I was looking for! I guess there's a reason for calling them blowout panels, it looks really brutal.

[edzieba]

"Blowout panels". I'm really weak on rocket history, does someone have an example of where this sort of thing has previously been used or is it just a concept at this point?

Thanks! Exactly what I was looking for! I guess there's a reason for calling them blowout panels, it looks really brutal.

The blowout panels are where the 'spray' before staging comes from. The interstage violently ripping apart post-staging is not to do with the blowout panels, just not caring about its fate after its done its job being an expendable stage.

[chopsticks]

The blowout panels are where the 'spray' before staging comes from. The interstage violently ripping apart post-staging is not to do with the blowout panels, just not caring about its fate after its done its job being an expendable stage.

Makes sense, it's more like vents in this case rather than the panels blowing out upon the ignition of the second stage. (Although that obviously happened later) I think I highlighted the correct spot where we see this happening.

Found a diagram online of what I think is the same rocket - they are called blast ports here. So it basically looks like an American version of the Soviet's open lattice they used for hotstaging, just built differently.

I was under the impression that there were closed panels that popped off but apparently not.

[edzieba]

The SA-6 separation camera footage from 3:40 in this video shows one of the triangular blowout panels below the S-IV bursting shortly before staging. From 7:00 you can see the inside of the interage wit the panels visible, but the actual blowout process is obscured by engine startup (you can see before with the intact panels and after with the triangular holes after staging, though):

[alastairmayer]

The blowout panels are where the 'spray' before staging comes from. The interstage violently ripping apart post-staging is not to do with the blowout panels, just not caring about its fate after its done its job being an expendable stage.

Makes sense, it's more like vents in this case rather than the panels blowing out upon the ignition of the second stage. (Although that obviously happened later) I think I highlighted the correct spot where we see this happening.

Found a diagram online of what I think is the same rocket - they are called blast ports here. So it basically looks like an American version of the Soviet's open lattice they used for hotstaging, just built differently.

I was under the impression that there were closed panels that popped off but apparently not.

Correct, they are not.  Because the vents are rectangular in shape (between the stringers of the interstage) it isn't immediately obvious when looking at pix of Titan II that they're actually vents, rather than part of the paint scheme.

[TheRadicalModerate]

In the interest of thread reduction, I'd like to de-zombify this one and bring up a related but separate topic.

We've spent a lot of time talking about launch abort here.  I kinda came to the conclusion that it's not hopeless but it's on the border between Corvus's Case #2 and Case #3 (mod the nose vs. design an abort-safe Starship).

But we've spent almost no time looking at EDL abort.

Let's stipulate that there's nothing to be done in the hypersonic entry regime.  You either get through it or you don't.

[Update:  Stupid, offensive language removed.  Sorry.]

But there are two contingencies in almost every other vehicle's EDL regime that are eminently survivable but are funky during a Starship EDL:

1) Nav errors.  If a capsule lands hundreds of miles off target, you wind up with seasick crew or one sitting in some desolate hunk of land until somebody comes and gets them.  With a Starship, if you don't land where there are chopsticks, you have a big honkin' problem:  No legs, and terrible rough-surface stability even if you have legs.  And ditching in the water is probably about as survivable as it in a commercial aircraft.

2) The suicide burn.  I mean, it's cool and all, but it'll cause all the ASAP people to wear their brown trousers to work.  And the decision loop is so tiny that everything would have to be automated, and I'm not sure if you can enumerate all of the ways that something's going wrong for a robot to make a reliable decision.

Both of these problems can to some extent be solved with a blow-away capsule or pod, as was discussed up-thread.  However, unlike a simple shuttling Starship, which is dedicated solely to taking crews to and from LEO, and therefore can be burdened with whatever squirrelly hardware is necessary, you're going to have missions with tight enough delta-v budgets that they must use a direct EDL from cislunar or interplanetary speeds.  So there's no transferring from your Mars return Starship to a nice, safe Starship ferry.¹

Burdening every single crewed flight with a blow-away capsule is going to be a bummer for a 3-6 month trip to Mars.  At the very least, you now have a blow-away capsule with a door and a passageway that lets the crew down into a space that will keep them from losing their minds.  And the structural compromises that might allow a successful blow-away are going to bump up against the necessity of hitting the atmosphere at 11-14km/s.

I don't think this is quite a deal-breaker for crew certification, but if you want NASA to fly crews on this puppy, you're going to need some very clean answers for a lot of strange cases.

_____________
¹I guess deep aerocapture is a possibility.  It at least reduces the problem to ordinary terrifying hypersonic entry.

[Robotbeat]

So I mostly agree that EDL is probably the more dangerous part, the fact that SpaceX has done like 80 successful launch/landings of the F9 booster in a row (without the benefit of engine out, etc) is certainly encouraging and has assuaged my concerns. At that reliability level, it’s already at like better than Shuttle levels if Starship can equal it.

Which doesn’t mean they don’t have a lot of work left to do. They absolutely do. Orbital reentry requires a more engineered and largely intact heatshield system than the F9 booster. But I think they have a pretty good chance of getting there.

I also think ways to do emergency egress should probably be incorporated.

[Hog]

In the interest of thread reduction,

emphasis mine
Update:
Wow and again I am surprised and impressed with the quality of the membership here at NSF.com.  I just see this as an attempt at levity gone wrong.  You manned up, apologized and moved on- so the words typed have been wiped.  I wish more people in life would follow your lead. We ALL make mistakes.   Thank you TRM.

[TheRadicalModerate]

emphasis mine
In the interest of human decency, is that really necessary?  I'd take a zombified thread over a thread containing that crap anyday.

You're right.  It was callous and stupid.  I apologize.

[Oersted]

In my view, Starship with landing legs is a very capable machine, also when it comes to emergency aborts and landings. Unlike the Shuttle it can land anywhere reasonably flat. I can well see it landing in an unprepared field or parking lot somewhere.

[mikelepage]

But we've spent almost no time looking at EDL abort.
...
But there are two contingencies in almost every other vehicle's EDL regime that are eminently survivable but are funky during a Starship EDL:

1) Nav errors.  If a capsule lands hundreds of miles off target, you wind up with seasick crew or one sitting in some desolate hunk of land until somebody comes and gets them.  With a Starship, if you don't land where there are chopsticks, you have a big honkin' problem:  No legs, and terrible rough-surface stability even if you have legs.  And ditching in the water is probably about as survivable as it in a commercial aircraft.

2) The suicide burn.  I mean, it's cool and all, but it'll cause all the ASAP people to wear their brown trousers to work.  And the decision loop is so tiny that everything would have to be automated, and I'm not sure if you can enumerate all of the ways that something's going wrong for a robot to make a reliable decision.

Both of these problems can to some extent be solved with a blow-away capsule or pod, as was discussed up-thread.  However, unlike a simple shuttling Starship, which is dedicated solely to taking crews to and from LEO, and therefore can be burdened with whatever squirrelly hardware is necessary, you're going to have missions with tight enough delta-v budgets that they must use a direct EDL from cislunar or interplanetary speeds.  So there's no transferring from your Mars return Starship to a nice, safe Starship ferry.¹

Burdening every single crewed flight with a blow-away capsule is going to be a bummer for a 3-6 month trip to Mars.  At the very least, you now have a blow-away capsule with a door and a passageway that lets the crew down into a space that will keep them from losing their minds.  And the structural compromises that might allow a successful blow-away are going to bump up against the necessity of hitting the atmosphere at 11-14km/s.

I don't think this is quite a deal-breaker for crew certification, but if you want NASA to fly crews on this puppy, you're going to need some very clean answers for a lot of strange cases.

_____________
¹I guess deep aerocapture is a possibility.  It at least reduces the problem to ordinary terrifying hypersonic entry.

Re your footnote - I'd always understood "deep aerocapture" and transfer to a dedicated shuttle starship to be the plan B, perhaps even plan A. I vaguely remember EM saying something about not doing direct entry "probably being wise"

In any case, Totally agree with your assessment re points 1 and 2, and I wonder whether the EDL abort scenario ends up being the reason they reverse the decision to remove legs from Starship. I'd have thought they would want to get a lot of heritage around those landing legs before ever going to Mars (and then want the flexibility of legs on Mars for a long time to come). Also, we'll see if they end up needing some kind of chine arrangement in order to transfer prop in orbit, and between those and landing legs, there might be other reasons that the safe operation of a reusable second stage ends up eating into the 150 ton payload allowance.

Having seen the soft water touchdowns of F9 boosters, you could almost imagine an emergency water landing of Starship to be relatively safe, as long as they can avoid the *timber* effect of falling over and slapping into the water sideways. Makes me wonder whether they ever get to a point of designing "crumple zones" into starship. Or maybe rapidly inflatable airbags in the legs that keep Starship upright while they let the prop tanks fill with seawater, getting it to a half-submerged configuration.

[uhuznaa]

Both of these problems can to some extent be solved with a blow-away capsule or pod, as was discussed up-thread.  However, unlike a simple shuttling Starship, which is dedicated solely to taking crews to and from LEO, and therefore can be burdened with whatever squirrelly hardware is necessary, you're going to have missions with tight enough delta-v budgets that they must use a direct EDL from cislunar or interplanetary speeds.  So there's no transferring from your Mars return Starship to a nice, safe Starship ferry.¹

I think there's little point in requiring all Starships being the same. I mean even right now and without Mars missions in sight we will already get at least the Starship HLS for Artemis, a tanker Starship (for refueling it in orbit), the Starship based propellant depot and a Starlink launcher version, all of these very different.

A LEO shuttle Starship with an escape pod would at least allow crewed flights and along with cargo and tanker flights (and landings) SpaceX would be able to gather experience towards risking crewed flights without such an escape pod (like on the return leg from Mars) eventually.

Having such a dedicated crew Starship for LEO missions ideally with full rescue capabilities (from launch through orbital operations and reentry to landing) would be a great transition to that. At least it would get the ball rolling when it comes to fly people on it. NOT developing it just because it wouldn't easily work for some missions doesn't seem to be a good reason not to do it.

[Lee Jay]

Having such a dedicated crew Starship for LEO missions ...

What LEO missions?

[uhuznaa]

Having such a dedicated crew Starship for LEO missions ...

What LEO missions?

Space stations? Commercial/private space stations should be one of the things that Starship should make much more practical (and less costly). And these require both shuttling crews as well as maybe maintenance and construction work.

Also shuttling crews to outbounds ships that you don't want to launch with the crew.

F9/Dragon are limited (only very limited cargo, no airlock) and at some point it may make only little economic sense to keep them flying if you have a fleet of Starships and boosters anyway. Not in the next years necessarily.