Abort options for Starship and Starship/SuperHeavy

[Stan-1967]

the bottom dome is very odd indeed

In answer to the original question, it's too big and heavy for an escape system.  It also doesn't have any canards, which would be required for an EDL-capable escape system.

What do you base this on?   What do you think the mass of this section is likely to be.  ISTM that if thrust ring shown in the concept for HLS was beneath this section, the thrust that would need to land a full SS with payload & prop on Luna would be more than sufficient for also serving as a abort motor if this single section was above the thrust ring & the separation plane in the event of an abort  left everything below ( workshop & prop tanks, raptors) behind. 

What do canards do that active thrust control cannot do in the abort scenarios likely needed if this was some path to human rating SS?

[dwheeler]

the bottom dome is very odd indeed

It's a pressure dome.  Looks like a full-up prototype of the LSS crew module.

I didn't think the LSS crew module would have tiles or the reentry flap thingies.

[TheRadicalModerate]

the bottom dome is very odd indeed

It's a pressure dome.  Looks like a full-up prototype of the LSS crew module.

I haven't pixel-counted: is that the full ogive section, or just a portion of it?  I'd guess that it would fit onto the cylindrical portion of the payload bay, likely with a tunnel (or two) into the airlocks on the "garage" deck.

In answer to the original question, it's too big and heavy for an escape system.  It also doesn't have any canards, which would be required for an EDL-capable escape system.

Is the door how crew ingress for launch?  Seems pretty unlikely to be the hatch/elevator on the garage deck.

Does SS need a pressure dome for 1 bar differential?

I'd think that a dome would be more mass-efficient than what you'd have to do to reinforce a flat deck.

BTW, after the fact I realized that my speculation on the hatch being for crew ingress is stupid, because the LSS never has crew entering on the ground.  So maybe the whole thing is a prototype for a launch/EDL-capable crew system.  Or maybe it's both, with a common design for most of the innards.

[TheRadicalModerate]

What do you base this on?   What do you think the mass of this section is likely to be.  ISTM that if thrust ring shown in the concept for HLS was beneath this section, the thrust that would need to land a full SS with payload & prop on Luna would be more than sufficient for also serving as a abort motor if this single section was above the thrust ring & the separation plane in the event of an abort  left everything below ( workshop & prop tanks, raptors) behind. 

Well, it certainly weighs a lot more than a D2.  I fiddled with a model for how heavy a 12-person abort system would be if the mass scaled like a D2.  I got about 21t.  Figure 300m/s of delta-v @ Isp=300s, and that's a wet mass of 24t.  NB:  I'd expect Starship to be considerably heavier than this, since the nose is made out of stainless steel.

If you figure you need at least 5G of acceleration for pad aborts (T/W=6), that's 1410kN of thrust.

In contrast, an LSS, when landing on the Moon, needs about 275t of wet mass to get back to NRHO, and the delta-v for the last 10m of landing and the first 10m of ascent is close to negligible.  To hover, that requires 446kN.

FWIW, I expect Starship to have two different kinds of thrusters:  cold gas for fine control, and combusting methox for orbital maneuvers and lunar landing.  If you figure there are 8 of them in the landing waist, they're maybe 60kN each, which would make them sort of a methox-driven SuperDraco.

What do canards do that active thrust control cannot do in the abort scenarios likely needed if this was some path to human rating SS?

The whole system has to be dynamically stable during an abort at max q.  For that matter, it has to be dynamically stable during a pad abort.

I guess you could wind up with a control system that was tight enough to avoid everything tumbling with the canards, but the center of pressure is going to be way out in front of the center of mass.  Waist engines can't really gimbal, so control would have to be through throttling, which introduces more lag into the control loop.

Control theory is an area where I'm especially bad (Laplace transforms!  Bah!), but my intuition is that this would be an extremely difficult problem.

If you could find a reliable way to blow the canards off during the abort process, that might work.  But if you only want to develop one variant of a 60kN combusting methox pressure-fed engine, you'd need about 25 of them.

[JayWee]

The whole system has to be dynamically stable during an abort at max q.  For that matter, it has to be dynamically stable during a pad abort.

Tbh, I was thinking more about a re-entry backup capsule, if something goes severly wrong with the heatshield (ie, if you'd put a backup PICA-X under the dome). Not exactly ascent.

[TheRadicalModerate]

The whole system has to be dynamically stable during an abort at max q.  For that matter, it has to be dynamically stable during a pad abort.

Tbh, I was thinking more about a re-entry backup capsule, if something goes severly wrong with the heatshield (ie, if you'd put a backup PICA-X under the dome). Not exactly ascent.

Separating in the high hypersonic sounds like it has even worse stability problems than max-q.

Everybody's pretty used to the idea that a failure during hypersonic entry is a sure-fire loss of crew.  But the only crew-rated launcher that had more than minimal black zones on ascent was the Shuttle, and nobody was particularly happy with results of the design decisions there.

[Twark_Main]

Not seeing it, but is this a case of TapAtalk not showing strike-throughs?

I do see the strike-throughs. The wording is just unnecessarily tortuous and difficult-to-follow.



Edit: f**k it, I'll just Cunningham it... 

Ahem!  Fellow Internet-ers, your attention please!!  I am 100% sure that TRM definitely meant this:

"Best I can tell, relative escape acceleration of a minimally-loaded Starship, with just enough prop to reach LEO and do EDL in an emergency, is going to be about 3G.  If you figure a 5-second¹ warning time (the "uh-oh-to-boom" interval),  the difference between 0.3s and 2.9s to full power is the difference between 325m of separation and 65m:  260m.  Figure the center of the explosion is 40m from the separation plane."

Looks right to me.

Thanks, good to know.

Hey, it was nothing a sed script couldn't fix...

Nope. Using a "dumb" automated process like a sed script, it would not have been possible to fix it unambiguously. That was precisely my issue.

Seriously, go back to the original post and try writing the script. I'll wait......       It's not as simple as merely deleting the crossed-out sections.

Instead I had to use my (always fallible!) intuition and judgement to guess at which non-crossed-out number was supposed to replace which crossed-out number.  That's what happens when the non-crossed-out numbers are just thrown in a big pile after a large crossed-out section. Miraculously, I got it right on the first try!

The fact that we don't notice this ambiguity (when we're speaking) isn't surprising. In fact, it's just human nature. In practice it's almost impossible to recognize the possibility for confusion, since of course you know what you meant — you wrote it! 




This isn't "sniping" btw, just friendly help so we all can achieve better communication in the future. Back to our regularly scheduled program...

[meekGee]

Nope. Using a "dumb" automated process like a sed script, it would not have been possible to fix it unambiguously. That was precisely my issue.

Seriously, go back to the original post and try writing the script. I'll wait......       It's not as simple as merely deleting the crossed-out sections.

Instead I had to use my (always fallible!) intuition and judgement to guess at which non-crossed-out number was supposed to replace which crossed-out number.  That's what happens when the non-crossed-out numbers are just thrown in a big pile after a large crossed-out section. Miraculously, I got it right on the first try!

The fact that we don't notice this ambiguity (when we're speaking) isn't surprising. In fact, it's just human nature. In practice it's almost impossible to recognize the possibility for confusion, since of course you know what you meant — you wrote it! 




This isn't "sniping" btw, just friendly help so we all can achieve better communication in the future. Back to our regularly scheduled program...

What's happening rn?

[TheRadicalModerate]

A little tidbit from the NASA-SpaceX Collaboration for Commercial Space Capabilities 2 SAA, attached.

Note that this is a concept review for crewed launch/EDL, not a test, and it's scheduled almost four years from now.

[CorvusCorax]

A little tidbit from the NASA-SpaceX Collaboration for Commercial Space Capabilities 2 SAA, attached.

Note that this is a concept review for crewed launch/EDL, not a test, and it's scheduled almost four years from now.

What is this then:

Milestone #5 Countermeasure Feasibility Review
SpaceX will conduct a formal feasibility assessment review to
accelerate a human health countermeasure tech demonstration
Success Criteria:
Feasibility assessment conducted and reviewed to the satisfaction of
SpaceX management.
Q4 2024

What is being countermeasured? Threats in space such as van allen belt and deep space radiation? Or threats during transfer operations including ascent and descent? If the latter, countermeasure could mean crew rescue/abort system.

Might be both, considering Milestone 9 is the corresponding concept review for long term ECLSS. So by 2024 they do the "feasability assessment" aka can it be done at all (more or less what this thread is doing?), and by 2027 - if the how would actually work. - just, you know, more informed than we are.

The timing does make sense. You definitely want some flight data, to assess what the risks are in the first place, before you can assess countermeasures against those risks. Without that, SpaceX couldn't do much more than this very forum thread did - speculate wildly in all directions (Which is on topic, so not a bad thing)

[Robotbeat]

Could be short arm centrifuge or bolo type artificial gravity. Radiation shielding also possible.

[Corey Mandler]

poking at Murphy's law is fun
https://docs.google.com/document/d/1mPqgt9etYl3QP6GwQS8r4-K_BgsG8uEeluWn5N4Rpkk/edit
a story of unfortunate events

[InterestedEngineer]

Someone up-thread kept insisting that shutting off the engines was a requirement for AFSS.

Kinda hard to do when the control lines get cut like on IFT-1.  Which is why AFSS needs to be completely independent of the control system of the rocket

https://www.spacex.com/updates/index.html

During ascent, the vehicle sustained fires from leaking propellant in the aft end of the Super Heavy booster, which eventually severed connection with the vehicle’s primary flight computer. This led to a loss of communications to the majority of booster engines and, ultimately, control of the vehicle.

[TheRadicalModerate]

Someone up-thread kept insisting that shutting off the engines was a requirement for AFSS.

Kinda hard to do when the control lines get cut like on IFT-1.  Which is why AFSS needs to be completely independent of the control system of the rocket

https://www.spacex.com/updates/index.html

During ascent, the vehicle sustained fires from leaking propellant in the aft end of the Super Heavy booster, which eventually severed connection with the vehicle’s primary flight computer. This led to a loss of communications to the majority of booster engines and, ultimately, control of the vehicle.

Several things:

1) Is "AFSS" just "AFTS" after the marketing people got done with it, or are you using this in the context of a crew abort system?

2) There's nothing to say that the individual engine controllers can't do a safe shutdown if they lose contact with the primary flight computers.  It's not a great solution, but odds are that things have already gone pear-shaped, and the shutdown is probably the least bad solution, or close to it.

3) You also have a set of manifold valves somewhere farther upstream, which are nominally opened at the start of chilldown.  Not sure what would happen if you closed them with 21t/s of prop flowing through them, but in the abort case, you're only making things marginally worse, and termination of thrust probably is more important than whatever reduction in the uh-oh-to-boom time is incurred.

4) Which finally brings us to the question of whether this impacts the ability of a Starship to carry itself away in an abort:

Let's use... 240t for SuperHeavy mass near burnout?  So 33 engines @ 2400kN each would be more than 33G of acceleration, and escape is hopeless.  So if SpaceX can't terminate SuperHeavy thrust reliably, then a Starship-mediated escape has a pretty hefty black zone close to staging.

No matter.  I'm pretty sure that comment #2 means that thrust termination works fine.

[Barley]

2) There's nothing to say that the individual engine controllers can't do a safe shutdown if they lose contact with the primary flight computers.  It's not a great solution, but odds are that things have already gone pear-shaped, and the shutdown is probably the least bad solution, or close to it.

But that might cause the engine controller to be considered to be part of the AFSS and add another couple of 9's to the reliability requirement.  Probably easier to hardwire a relay in series with a valve solenoid or hardwire explosives to the inlet manifold or hardwire something.  I really don't want to have to discuss byzantine errors in a bus controller with a regulator.

[TheRadicalModerate]

2) There's nothing to say that the individual engine controllers can't do a safe shutdown if they lose contact with the primary flight computers.  It's not a great solution, but odds are that things have already gone pear-shaped, and the shutdown is probably the least bad solution, or close to it.

But that might cause the engine controller to be considered to be part of the AFSS and add another couple of 9's to the reliability requirement.  Probably easier to hardwire a relay in series with a valve solenoid or hardwire explosives to the inlet manifold or hardwire something.  I really don't want to have to discuss byzantine errors in a bus controller with a regulator.

What regulator?  All the FAA cares about is that the thing goes boom when told to.  Thrust termination has nothing to do with that.  But thrust termination is incredibly important for any kind of launch escape, especially full Starship escape, which has pretty wimpy acceleration. The FAA doesn't care about that, at least until the human spaceflight moratorium expires (which could be next month, I guess).

[joek]

But that might cause the engine controller to be considered to be part of the AFSS and add another couple of 9's to the reliability requirement.  Probably easier to hardwire a relay in series with a valve solenoid or hardwire explosives to the inlet manifold or hardwire something.  I really don't want to have to discuss byzantine errors in a bus controller with a regulator.

No it would not. The AFSS/AFTS specs are pretty clear. It's job is to tell everyone to shut down (or whatever) when it detects an anomolous condition; it is a simple binary signal. If it does that reliably, then done.

That said, if we consider the entire vehicle as a system, and the "safety" of that system, then we're in a much deeper and broader discussion.  "Byzantine errors in a bus controller" discussions with a regulartor will be the least of your worries.

If you don't have accurate trajectorym IIP and debis modeling on which to base AFSS/AFTS decisions, you're not going to make it to first base.

[joek]

What regulator?  All the FAA cares about is that the thing goes boom when told to.  Thrust termination has nothing to do with that.  But thrust termination is incredibly important for any kind of launch escape, especially full Starship escape, which has pretty wimpy acceleration. The FAA doesn't care about that, at least until the human spaceflight moratorium expires (which could be next month, I guess).

They care about a bit more than that. E.g., For years all flights to the ISS required individual FAA waivers--every single one of them --due to their proximity to the US East Coast. Because even if the flight was told go boom when commanded, it violated Ec limits (based on accepted models).

[TheRadicalModerate]

What regulator?  All the FAA cares about is that the thing goes boom when told to.  Thrust termination has nothing to do with that.  But thrust termination is incredibly important for any kind of launch escape, especially full Starship escape, which has pretty wimpy acceleration. The FAA doesn't care about that, at least until the human spaceflight moratorium expires (which could be next month, I guess).

They care about a bit more than that. E.g., For years all flights to the ISS required individual FAA waivers--every single one of them --due to their proximity to the US East Coast. Because even if the flight was told go boom when commanded, it violated Ec limits (based on accepted models).

Yeah, but if the tanks go boom, the engines stop in short order--at least the liquid-fueled ones.  They're not going to cause big flight path deviations.

Acronym fault:  Ec limits?

[joek]

... Acronym fault:  Ec limits? ...

Ec = Expected Casuality; see, e.g., FAA Advisory Cirlcular Expected Casualty Calculationsb for Commercial Space Launch and Reentry Missions.